Copyright is owned by the Author of the thesis. Permission is given for a copy to be downloaded by an individual for the purpose of research and private study only. The thesis may not be reproduced elsewhere without the permission of the Author. Population studies of ultraviolet-B radiation responses in white clover (Trifolium repens L.) A thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy In Plant B iology at Massey University, Palmerston North, New Zealand. by Rainer W. Hofmann 2000 To Angelika and Stephan III Abstract White clover growing in New Zealand is experiencing increasing levels of ultraviolet-B (UV-B) radiation as a result of ozone depletion . This thesis reports a series of investigations on morphological and physiological responses to UV -B in various white clover populations. In addition, these studies examined UV -B responsiveness in combination with drought and consequences for insect herbivores. Plants were grown in controlled environment rooms with and without supplemental UV -B radiation at a dose of 1 3 .3 kJ m-2 d-I , corresponding to a 25% mid-summer ozone depletion above Palmerston North, New Zealand. Morphological measurements included numerous attributes of plant growth and morphogenesis as well as several aspects of leaf structure. Physiological studies investigated both primary and secondary metabolic functions. In generaL UV -B reduced components of white clover growth. The white clover populations showed a number of constitutive and UV -B-induced differences in many morphological and physiological attributes. UV -B sensitivity was mitigated by drought and was less pronounced with increasing duration of UV -B exposure. Bioassays revealed differential dietary effects of UV -B-treated foliage on the performance of lepidopteran insects. UV -B effects were also apparent under frequent defoliation. Comparisons of morphological and physiological features showed that white clover UV -B responsiveness was mainly l inked to inherent differences in morphology and growth among the populations, while on a physiological level it was more the UV -B­ induced differences that conferred UV -B tolerance. In particular, UV -B tolerance of the white clover populations was related to lower constitutive productivity. This was further l inked to several constitutive leaf attributes, including small leaf size, small and lens­ shaped epidermal cells as well as low leaf water potential. UV -B-induced features l inked to UV -B tolerance included high percentage of leaf dry mass, accumulation of UV -absorbing compounds and of total flavonols (particularly quercetin glycosides) as well as increases in leaf water potential. UV -B tolerance was greatest in white clover populations adapted to multiple and severe stresses in the habitat of origin. This series of investigations was used to propose a functional framework, linking UV-B responsiveness to underlying specialisation of the white clover populations. IV This thesis has been written with a view to publish much of the presented material. Accordingly, each Chapter represents a self-contained unit, forming the basis of papers published, submitted or prepared for submission. v Acknowledgements Profound gratitude goes to my wife Angelika and my son Stephan. Your love and patience was vital in supporting me throughout this project. Thanks are also due to my family overseas, especially my mother who patiently accepted the fact that happiness and scientific enjoyment for one of her children happened to coincide with a study career on the other side of the world. And thanks to Keith Cederman and his family i t was possible to call New Zealand our new home, resulting in the commencement of my studies and of this PhD project. Greatest thankfulness is due to my supervisors. Associate Professor David Fountain was instrumental in the initiation of this proj ect and a constant source of encouragement and advice throughout. From the outset of the work Dr Bruce Campbell provided invaluable intellectual, material and personal SUppOlt in all aspects of the work, making the proj ect in i t s entirety possible. Furthermore, thanks for helpful advice to Associate Professor Brian Jordan and to Dr Dennis Greer who accompanied the fluorescence studies. Special acknowledgements go to Chris Hunt for design and operation of the UV-B enrichment facilities and to Derryn Hunt and The 3 Herbageladies Yvonne Gray, Christine van Meer and Margaret Greig for technical support. Thanks also to Greig Cousins, Sheryl and Paul Doyle, Laurie Kennedy, Joce Tilbrook, Dave Scammel, Helen Little, Joanne Morris , Bruce Jackson, Linda Robinson, Charlotte Madie, Lyn Watson and Mush Williamson for technical advice and help. Greatly appreciated is helpful advice and discussion provided by Drs Grant Abernethy, Dave Barker, Richard Biggs, Janet Bornman, John Brock, John Caradus, Harry Clark, Sandra Diaz, John Hunt, Geoff Lane, William Laing, Soheila Mackerness, Cory Matthew, Chris Mercer, Tessa Mills, Paul Newton, Tony Parsons, Jel te Rozema, Derek Woodfield, Warren Williams and Sang Dong Y 00. Inspiring debate was provided by Professor Rod Thomas who taught me and participated in the dissection of white clover stolon tips. Thanks also to all staff at AgResearch Grasslands, who were extremely friendly and accommodating to their students, treating Todd White and me completely as their own. And Todd, it was a pleasure to share this intense time of learning with you. - - - - ------ �------ VI The PhD work also allowed me to experience a number of fruitful, productive collaborations. This included extensive work together with Professor Richard Lindroth (University of Wisconsin) during the insect bioassay studies, resulting in a published paper written jointly with Rick (Chapter 7). Great appreciation also goes towards the plant chemistry group at IRL in Lower Hutt, including Drs Ken Markham, Stephen Bloor, Ewald Swinny and Ken Ryan with whom it was not only a pleasure and privilege to extract flavonoids and conduct HPLC analyses but who also provided many inspirational discussions. Further acknowledgement is due to Ewald for NMR studies identifying the structure of the flavonoid compounds detected in HPLC. I am also particularly grateful to Wilhelmina Martin of AgResearch and Doug Hopcroft and 'Crunch' of HortResearch for showing me the techniques involved in the preparation of samples and the conducting of analyses for plant nutrient levels and electron microscopy. Valuable statistical advice is acknowledged from Or Sam Beckett and Duncan Heddedey at Massey University and from the site statisticians at AgResearch Grasslands, Dr Oavid Baird, Robert Fletcher and Fred Potter. The following institutions are gratefully acknowledged for financial support: PhD Fel lowship: • The Agricultural and Marketing Research and Development Trust (AGMARDT). Operational funding: • AgResearch Grasslands. • Massey University. • New Zealand Foundation for Research, Science and Technology (contract number C I0632). vu Table of Contents Chapter 1. Introduction .................................................................................... 1 The solar UV environment of New Zealand . . . . . . . ... . . . . . . . . . . . .. ... ............ .... ..... ....... .... .. .... . 1 Solar radiation and ozone ..................................................................................................... 1 Hemisphere comparisons and New Zealand-related factors ................................................ 2 Past and future trends ...................................................................................................... .... . 4 The New Zealand pasture ecosystem . .. ....... .... ............... ........... .......... ........ .... .......... .... 5 UV -B radiation and plants . .. .. . ..... .................................. ........... .......... ...... ...... ....... ....... . 6 History .................................................................................................................................. 6 Legulnes and white clover ..................................................................................................... 7 Population studies ............................... .................................................................................. 9 Underlying llzechanislns ...................................................................................................... 10 Interaction with otherfactors ....................................................................................... ...... 12 General description of the experimental approaches ... .. ...... ... ... . ................. ... . . . . .... . . . .. 17 Aims ...... .... ......... . .... . . . ... . . . . . ...... ..... .................. ..... ............ ..... ............ ........... ..... . ........ . 1 8 Chapter 2. Trifolium repens responses to u ltraviolet-B radiation: integrated analysis of 17 traits and 26 populations ..................................... 20 Abstract. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 Materials and Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 Experimental design ............ ................................................................................................ 24 Plant cultivation and UV-B irradiation .............................................................................. 24 Plant material .............................................................................................................. ....... 25 Whole plant growth measurements ..................................................................................... 28 Leaf llleasurements ....................................................................................................... ....... 28 Epidennal studies .......................................................................................................... ..... 29 Statistical analysis ......................................................................................... ........ .... ....... .. 29 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 Univariate UV-B responses ................................................................................................ 30 Plant trait responses in the first principal component ........................................................ 31 Population responses in the first principal component ...................................................... 33 Plant trait responses in the second principal component ................................................... 34 Population responses in the second principal component .................................................. 34 Correlation analyses ........................................................................................................... 35 -- ----------------- Vlll Discussion ................................................................................................................... 37 Plant trait responses in the first principal component ........................................................ 37 Population responses in the first principal component ...................................................... 38 Plant trait responses in the second principal component .................................................. .40 Population responses in the second principal component ................................................. .41 Relationship between PC I and plant features .................................................................... 42 Conclusions ................................................................................................................. 44 Chapter 3. Growth responses to u ltraviolet-B radiation in Trifolium repens populations depend on water availabil ity, productivity, and duration of stress ............................................................................................................... 45 Abstract. ....................................................................................................................... 46 Introduction ................................................................................................................. 46 Materials and Methods ................................................................................................ 49 Experimental design. ........................................................................................................... 49 Plant cultivation and UV-B irradiation .............................................................................. 49 Harvesting of plant material and drought application ....................................................... 52 Statistical analysis .............................................................................................................. 53 Results ......................................................................................................................... 53 Influence of drought on UV-B sensitivity ............................................................................ 53 Influence (?fpopuiation differences and time on UV-B sensitivity ...................................... 57 Discussion ................................................................................................................... 60 Influence of drought on UV-B sensitivity ............................................................................ 60 Influence of population differences on UV-B sensitivity ..................................................... 61 Influence of time on UV-B sensitivity .................................................................................. 63 Chapter 4. Effects of u ltraviolet-B radiation on Trifolium repens morphology: mediation by drought and relationships to UV-B sensitivity ........................................................................................................ 65 Abstract. ....................................................................................................................... 66 Introduction ............................................................. .................................................... 66 Materials and Methods ................................................................................................ 69 Biomass production and leaf attributes .............................................................................. 69 Stolon growth and apical bud dissection ............................................................................ 70 Statistical analysis .............................................................................................................. 70 Results ......................................................................................................................... 7 1 Between-population comparisons ....................................................................................... 73 IX Discussion ............................................................................................ ....................... 77 Leaf and stolon fnorphology ............................................................................................... 77 Relative leaf dry mass allocation ........................................................................................ 79 Between-population comparisons ....................................................................................... 79 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 Chapter 5. Effects of u ltraviolet-B radiation on Trifolium repens physiology: mediation by drought and relationships to UV-B sensitivity ........................................................................................................ 82 Abstract. . . .. . . . .. . . .. .... . . ........ . ..... . . ...... . ..... ...... ......... . . . . . . . . .. . . .. . . . ........ . .. ... . . . . . . . . . . . . .... . . . ....... 83 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 Materials and Methods . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 UV-absorbing cOfnpounds ................................ .................................................................. 8 7 Flavonoid analysis .............................................................................................................. 8 7 Chlorophyll concentration .................................................................................................. 8 8 Chlorophyll fluorescence .................................................................................................... 88 Leaf water potential and proline ......................................................................................... 89 Statistical analysis .............................................................................................................. 90 Results ......................................................................................................................... 90 UV-absorbing compounds and flavonoids .......................................................................... 90 Photosynthetic pigmentation and photochemistry .............................................................. 92 Leaf water potential and proline ......................................................................................... 93 IntraspeciJic relationships to biomass production and UV-B sensitivity ............................ 96 Discussion .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 UV-absorbing compounds and flavonoids .......................................................................... 99 Photosynthetic pigmentation and photochemistry ............................................................ 101 Leaf water potential and proline ....................................................................................... 102 1ntraspecific relationships to biomass production and UV-B sensitivity .......................... 104 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106 Chapter 6. Responses of nine Trifolium repens L. populations to u ltraviolet-B radiation: differential flavonol g lycoside accumulation and biomass production ..................................................................................... 108 Abstract. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 09 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 09 Materials and methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112 Flavonoid isolation and identification .............................................................................. 112 x Results ....................................................................................................................... 112 Flavonoid ident!fication ......................................................................... .................. ......... 112 Totalflavonols ..... .................................................... . ................ ................... ................. .... 115 Spec(tlc flavonols .............. ........ . ........ .................................................................... ........... 115 Q:K ratio ..................... ... ........................................................... ........................................ 117 Flavonol relationships to biomass production. ......................................... ................... ..... 118 Discussion ......... ... . .................................................. . ................................................. 121 Total flavonols .................................................................................................................. 121 D(tferential UV-B effects on quercetin and kaempferol glycosides .................................. 122 Population d(tfe rences in the accumulation of total flavonols .......................................... 123 Population differences in the accumulation of quercetin and kaempferol ....................... 125 Relationship ()fflavonol levels to UV-B tolerance ...... . .......... ...... ..................................... 126 Conclusions ............................................................................................................... 127 Chapter 7. Differences in Trifolium repens L. responses to ultraviolet-B radiation: foliar chemistry and consequences for two lepidopteran herbivores ..................................................................................................... 128 Abstract. ..................................................................................................................... 129 Introduction ............................................................................................................... 129 Materials and lnethods ............................................................................................... 131 Experimental design .......................................................................................................... 13 J Plant cultivation and UV irradiation ................................................................................ 131 Sampling procedure for plant nutrient analyses ............................................................... 132 Nitrogen ..... ............. . ...................................................... . .. ................................................ 132 Available carbohydrates ................................................................................................... 133 Neutral detergent fibre ...................................................................................................... 133 Cyanogenesis .................................................................................................................... 134 UV-absorbing cOfnpounds ................................................................................................ 134 Spodoptera litura bioassays .............................................................................................. 135 Graphania mutans bioassays ............................................................................................. 136 Statistical analysis ............................................................................................................ 136 Results ....................................................................................................................... 137 FoliaI' chernistry ................................................................................................................ 137 Spodoptera litura performance ......................................................................................... 139 Graphania mutans peiformance ........................................................................................ 145 Discussion .......................................................... ......... ... ... ........................................ 145 FoliaI' chemistry related to primary metabolism .............................................................. 145 Xl Foliar chemistry related to secondary metabolism ......................................................... .. 146 Insect Responses ............................................................................................................... 147 Chapter 8. Responses to u ltraviolet-B radiation in frequently defoliated Trifolium repens populations: differential growth, photochemistry and anatomy ............................ 11 •••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 150 Abstract. . . . . ............. ................ . . . ................... .......... . ........ . .................... . ............. ........ 15 1 Introduction ............................................................................................................... 15 1 Materials and Methods .............................................................................................. 153 Leaf fnaterial ..................................................................................................................... 153 Chlorophyll fluorescence ................................................................................................ .. 153 Structural studies .............................................................................................................. 154 Results ........ . . .............. . ................ . . . . .... . . . ...................................... . . . ......... . ................ 155 Discussion ................................................................................................................. 160 Leaf growth and photochemistry ...................................................................................... 160 Structural studies ........................................................................................................ ...... 161 Chapter 9. Discussion .................................................................................. 164 A functional framework to synthesise results ........................................................... 164 Comparisons between experiments ........................................................................... 174 Context and l imitations of the findings ..................................................................... 178 Perspectives and future studies .................................................................................. 18 1 Organ-5pec�fic studies ...................................................................................................... 184 Population-specific studies ............................................................................................... 186 Interaction of UV-B with additional stress, disturbance and competition ........................ 187 Chapter 10. References ................................................................................ 189 Xll List of Figures Fig. 1-1. Cyanide production in the white clover cultivar 'Huia' .................................... 17 Fig. 1-2. Scheme of the research project based on plant attributes ................................. 19 Fig. 2-l. Principal components analysis ......................................................................... .32 Fig. 2-2. Relationships between UV -B sensitivity and plant attributes .......................... 36 Fig. 3-1. Biomass production under UV-B and drought ..................... ............................ 55 Fig. 3-2. Relative changes in the rate of biomass production in response to UV -B. ...... 56 Fig. 3-3. Relationship between constitutive productivity and relative growth . . . . . . . . . . . . . . . . 5 8 Fig. 3-4. Time-course of the UV -B response .................................................................. 59 Fig. 4-1. Leaf initiation in apical buds under UV -B and drought . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 Fig. 4-2. Leaf size under UV -B and drought . . ................................................................. 74 Fig. 4-3. Percentage leaf dry mass under UV -B and drought. ........................................ 74 Fig. 4-4. Stolon elongation rate under UV-B and drought . . ............................................ 75 Fig. 4-5. Relationships of constitutive productivity to UV -B sensitivity and leaf attributes .................................................................................................................. 76 Fig. 5-1. UV-absorbing compound levels under UV-B and drought .............................. 91 Fig. 5-2. Flavonol glycoside accumulation under UV-B and drought ............................ 91 Fig. 5-3. Leaf water potential under UV -B and drought ................................................. 94 Fig. 5 -4. Proline concentration under DV -B and drought . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 Fig. 5-5. Relationships of constitutive productivity to quercetin accumulation and leaf water potential ......................................................................................................... 97 Fig. 5-6. Relationship between DV -B-induced relative changes in growth and in DV- absorbing compounds . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 X 111 Fig. 6-1. HPLC profiles of 'Tienshan' grown with and without UV-B ......................... 114 Fig. 6-2. Total flavonol concentrations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116 Fig. 6-3. Quercetin and kaempferol concentrations ................................................. .. . .. 116 Fig. 6-4. Ratio of quercetin to kaempferol (Q:K) ................................................... . . ..... 118 Fig. 6-5 . Relationship between quercetin accumulation and productivity ........ ....... . .... 119 Fig. 6-6. Relationship between quercetin accumulation and relative growth . ............ .. 120 Fig. 7-1. Foliar chemical composition ................................................................ ........... 13 8 Fig. 7-2. Growth of Spodoptera fitura and Graphania mutans larvae .......................... 140 Fig. 7-3 . Development times and pupal weights of S. litura ............................. ............ 141 Fig. 7-4. Stadium duration and growth rates of S. litura and G. mutans . . . . . . . . . . . . . . . . . . . . . . . 142 Fig. 7-5. Consumption rates and food utilisation of S. litura and G. mutans . . . . . . . . . . . . . . . . 143 Fig. 7-6. Development times and pupal weights of G. mutans . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144 Fig. 8-1. Light micrographs of transverse white clover leaflet sections ....................... 158 Fig. 8-2. TEM micrographs of transverse white clover leaflet sections ........................ 159 Fig. 9-1. Model .............................................................................................................. 165 Fig. 9-2. Scheme of the research project based on attributes ........................................ 166 Fig. 9-3. Relationship of growth responses ................................................................... 176 Fig. 9-4. Relationship of constitutive growth ................................................. . .... .. ....... . 177 Fig. 9-5 . Relationship of UV -absorbing compound responses ...................................... 177 Fig. 9-6. Morpho-physiological attributes in 'Huia' and 'Red' ...................................... . 180 XIV List of Tables Table 1 - 1 . Comparison of UV -B irradiances in New Zealand and Germany ................... 3 Table 2- 1 . Characteristics of 26 white clover populations .............................................. 26 Table 2-2. UV -B-induced univariate percent changes ................................................... . 3 1 Table 2-3. Correlations of plant attributes to the principal components ......................... 33 Table 3- 1 . Description of nine white clover populations ................................................ 50 Table 3-2. REML analysis of aerial biomass production ................................................ 54 Table 4- 1 . Attributes of growth and morphology under UV-B and drought. . ................. 72 Table 5- 1 . B iochemical and photochemical attributes under UV -B and drought . . . . . . . . . . . 93 Table 7- 1 . Effects on cyanogenesis and UV -absorbing compounds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 39 Table 8- 1 . Morphological attributes in frequently defoliated plants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 56 Table 8-2. Photochemical parameters in frequently defoliated plants .......................... 1 57 Table 9- 1 . Overall features of the series of investigations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 67 Table 9-2. Key aspects from large-scale screening .......................................... ......... . ... 1 68 Table 9-3 . Key aspects from examinations of UV -B susceptibility in interaction with drought and flavonoid responses ....................................................................... ' " . 1 69 Table 9-4. Key aspects from bioassays and studies of frequently defoliated plants . . . . . 1 7 1 Table 9-5. General comparisons between experiments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 75 AC ACR AD AGR M/Fm DM ECI FvlFm HCN LAR LM LWR NDF NGP NPQ PC PCA PDM PPF PSII qP RGR RSR SER SLM TEM UV-A UV-B UV-C lI'L Abbreviations Available carbohydrates Average consumption rate Approximate digestibility A verage growth rate Photochemical yield Dry mass, dry matter Efficiency of conversion of ingested food Intrinsic efficiency of PSI! Hydrogen cyanide Leaf appearance rate Light microscopy Leaf weight ratio Neutral detergent fibre Number of growing points Non-photochemical quenching Principal component Principal components analysis Percent leaf dry mass Photosynthetic photon flux Photosystem II Photochemical quenching Relative growth rate Root:shoot ratio Stolon elongation rate Specific leaf mass Transmission electron microscopy Ultraviolet-A radiation Ultraviolet-B radiation Ultraviolet-C radiation Leaf water potential xv 1 . Introduction 1 Chapter 1. Introduction This Chapter has been published in part in CampbeU et al. (1999). The solar UV environment of New Zealand Solar radiation and ozone Recently published results from long-term studies provide strong evidence that depletion of the stratospheric ozone layer leads to a net increase in biologically active ultraviolet (UV)-B radiation (290-315 nm) in New Zealand (McKenzie et al., 1999). First reports on the reduction of stratospheric ozone by anthropogenic emissions appeared nearly thirty years ago (Johnston, 1971). The destructive processes in the stratosphere caused by halogen-containing compounds were elucidated by Molina and Rowland (1974), based on earlier studies by Crutzen (1972). This achievement earned those three scientists the 1995 Nobel Prize for chemistry. The catalytic losses of stratospheric ozone in the southern hemisphere ('ozone hole') were first reported a decade later (Farman et al., 1985). The long and narrow north-to-south orientation of New Zealand covers a relatively wide range of latitude: 34°S to 47°S, equivalent to extending from parts of middle Europe, through southern Europe and to northern Africa. Across this gradient, UV levels differ more between north and south than levels of total solar radiation, the latter ranging around an annual mean daily dose of 14-15 MJ m-2 (McKenzie et al. , 1996). Most of the country receives about 2000 sunshine hours per year. The peak mean hourly photosynthetic photon flux (PPF) in central New Zealand during summer is approximately 2100 �mol m-2 so l , with a monthly summer average of up to 1700 �mol m-2 so l (Greer and Laing, 1992). On average, northern New Zealand receives 25% more erythemally-weighted (or 'sunburning') UV than the south and thi s is mainly due to a smaller solar zenith angle (SZA, determining optical path length) in the north (McKenzie et al., 1996). Regulated by the seasonal changes in the SZA, typical clear-sky daily erythemally-weighted UV J. Introduction �- - --- --�-- 2 doses range from 0.5 kJ m-2 (winter) to 7 kJ m-2 (summer) (McKenzie et ai., 1996). While the mean annual erythemally-weighted UV dose (ca. 3 kJ m-2) is about half that experienced at the equator, the summer values in the north of New Zealand are similar to equatorial levels (McKenzie et al., 1996). The daily clear-sky mid-summer plant­ weighted UV dose above Palmerston North, normalised to 300 nm is approximately 8 .4 kJ m-2. Ozone levels in New Zealand display a maximum In late winter/early spring (September/October, ca. 350-360 Dobson units, DU) and a minimum in late summer/early autumn (FebruarylMarch, ca. 270 DU) (McKenzie et al., 1996). Thus the maximum ozone levels over New Zealand can be found around the time of greatest depletion over Antarctica, prior to the annual break-up of the polar vortex and spread of this ozone-depleted atmosphere. Hemisphere comparisons and New Zealand-related factors Although stratospheric ozone reductions have been measured in both hemispheres, erythemally-weighted UV levels in New Zealand are on average about 40-50% higher compared to similar northern hemisphere latitudes (Table 1-1) (Seckmeyer et al., 1995). This also takes into account differing conditions in cloudiness, aerosol loading and surface albedo. The higher UV levels have been attributed mainly to decreased ozone levels and are seen to result both from lower ozone amounts in the stratosphere over New Zealand as well as from higher levels of tropospheric ozone at the similar northern hemisphere latitudes (Seckmeyer and McKenzie, 1992). Other contributing factors responsible for elevated UV levels in New Zealand compared to similar northern mid­ latitudes are seen to be the closer summer Sun-Earth separation and lower aerosol loading of the atmosphere (Seckmeyer and McKenzie, 1 992). i. introduction Table 1 -1 . Comparisons of mean daily erythemally-weighted UV-B irradiances (kJ m-2 d- I m-2) in summer at similar latitudes in New Zealand and Germany during 1993-1994 (Seckmeyer et al. , 1995). Site Lauder, New Zealand Garmisch­ Partenkirchen, Germany Location Noy 3.4 May 2.5 Dec 4.7 Jun 2.9 Jan Feb 4.8 4.2 Jul Aug 3.3 2.7 3 Cloud cover is one of the most important factors influencing UV levels (Bodeker and McKenzie, 1996). New Zealand is called 'Aotearoa' in the Maori language, meaning 'land of the long white cloud' . While partly cloudy conditions can either enhance (by up to 25%, e.g. cumulus type) or reduce UV levels (Bodeker and McKenzie, 1996; McKenzie et al., 1997), cloud cover in general reduces UV levels in New Zealand by about 25-30% per year (McKenzie et al., 1996). Low amounts of aerosols and relatively small aerosol disturbances (Rosen et al. , 1997) result from the comparatively clean air in New Zealand. Apart from shifting the balance between direct and diffuse UV irradiance during summer (when aerosol concentrations are highest), in New Zealand effects of aerosols on UV levels (contrasting to the northern hemisphere) can in general be regarded as of minor importance (Forgan and Liley, 1998). Also contrary to the northern hemisphere, aerosols from the Pinatubo eruption showed no clear impact on ozone profiles over New Zealand (Forgan and Liley, 1998). During the Austral spring of 1998, the l argest-ever area of ozone depletion was recorded over Antarctica (27 .3 million km2), with the 'ozone hole' developing two weeks earlier than usual (NASA, 1998). Incidents of polar vortex ozone-depleted air passing over New Zealand can occasionally be observed (Brinksma et al., 1998). Together with the decreasing background ozone levels and other factors, this could help explain the record low ozone levels measured over New Zealand during winter/early spring 1997 (Brinksma et al. , 1998). To date, however, these direct, vortex-related episodes are seen - ------- I. Introduction 4 to contribute only to a small degree to the overall reduction of ozone levels, when compared e.g. to the natural day-by-day variation in ozone of ca. 10% over New Zealand (McKenzie et al., 1996). Interestingly, the levels of the key ozone destroying species chlorine nitrate (ClONOz), show a similar seasonal cycle to the ozone dynamics over New Zealand, also with the maximum and minimum levels occurring in early spring and autumn, respectively (Rei singer et al., 1995). An annual trend of a 3-4% stratospheric increase in chlorine­ containing reservoir species like HCI and ClONOz is considered possible at New Zealand latitudes (Rei singer et al., 1995). Past and future trends During the last 25 years, ozone levels over New Zealand showed a decrease of ca. 3% per decade (Matthews, 1998). Erythemal UV levels during the 1980s increased by ca. 3 - 5% and plant-weighted UV at New Zealand southern latitudes increased by ca. 6-9% (Madronich, 1992). Much of the ozone reduction in New Zealand seems to have occurred during the mid 1980s, with smaller changes since. One explanation for the latter could be a delay in the break-up of the polar vortex due to decreased stratospheric temperatures, preventing the transport of ozone-rich air away from lower latitudes to the polar regions and thus even leading to a small increase in ozone over New Zealand during late winter/early spring (Matthews, 1998) . Nevertheless, recent studies demonstrated conclusively that long-term decreases in summertime ozone over New Zealand have led to substantial increases in peak UV radiation intensities (McKenzie et al., 1999). Catalytic stratospheric ozone destruction due to increased halogen loading is predicted to continue well into the next century, but is perceived to decrease over the next decades as a result of international restrictions on halogen-containing source chemicals (McKenzie et al., 1996). Estimates of the maximum ozone depletion vary depending on assumptions about compliance. If there was lasting compliance to the Montreal Protocol and its amendments, it is predicted that ozone depletion over latitudes covering New Zealand could peak early next century, with an 11 % decline compared to 1960 levels 1. Introduction 5 (Madronich et al., 1995). This would result in a corresponding peak in erythemally­ weighted UV of ca. 15%. The New Zealand pasture ecosystem From the above it is evident that the temperate pasture species in New Zealand experience much more pronounced UV -B irradiation compared to their origin at similar latitudes in the northern hemisphere. Today the predominant pasture species in New Zealand lowland areas are white clover (Trifolium rep ens L.) and ryegrass (Lolium perenne L.), grown in close association in mixtures (Harris, 1990). White clover is a herbaceous perennial legume, spreading by stolons with roots at nodes. The annual production of dry matter (DM) by ryegrass-clover based pastures ranges from 10-20 t DM/ha depending on summer rainfall and fertility (Coop, 1986), whereas the lower fertility pastures produce 5 -10 t DM/ha. Pasture ecosystems in New Zealand can experience pronounced periods of summer drought, particularly in dry east cost regions (Campbell et al., 1999). In these pasture communities, white clover provides an important source of biologically-fixed nitrogen which drives ecosystem productivity. The white clover component of these pastures is a key component of the New Zealand pastoral industries which strongly rely on this low-cost feed source. In addition, white clover benefits pastoral industries by seasonal complementarity with grasses and its ability to increase livestock feed intake and utilisation rates (Caradus et al., 1996). White clover fulfi ls these roles particularly well in countries with a moist, temperate climate such as New Zealand, where climatic conditions are especially suited for white clover growth (Brock et al., 1989). The annual values for white clover nitrogen fixation in New Zealand are extremely variable depending on white clover content and availability of mineral nitrogen (Crush, 1987). However, it is estimated that the symbiosis with Rhizobium nitrogen fixing bacteria can provide around 200 kg N/ha/year, but has been measured in the range 17-400 kg N/ha/year. Nitrogen fixation by legumes is the primary source of nitrogen for these pastures and l ittle fertiliser nitrogen is applied. Nitrogen deficiency still l imits agricultural production in New Zealand (Caradus et al. , 1996) and there are ongoing efforts to increase the i . introduction 6 contribution of white clover to pasture through new breeding, pest management and grazing management strategies. These systems are grazed i ntensively by ruminant herbivores. This defoliation interrupts the pattern of plant development by regularly removing leaves before they are fully aged, altering source-sink relationships, changing canopy structure and influencing nutrient cycling. The grazing management generally removes a large propoltion of the leaf area of the pasture plants, either under a continuous stocking system or with intermittent grazing allowing rest periods for re­ growth. The predominant l ivestock enterprises on the lowland areas are dairying and lamb finishing, whereas the hill areas are predominantly used for sheep and beef production. UV-8 radiation and plants History Medical research on the effects of ultraviolet radiation on human skin 100 years ago inspired a range of studies on plant responses to UV over the last century. Early reviews of these findings were conducted by Popp and Brown (1936), and later by Lockhart (1961). Much of the UV -supplementation experiments during the first three quarters of that period were based on research with lamp systems emitting high levels of UV-C radiation « 280 nm). These irradiation protocols were therefore more typical for extraterrestrial conditions. Nevertheless some of the earlier results were remarkably reflective of findings obtained from more current research using UV -B. For example a study on 67 plant species showed that Austrian Pine (Pinus nigra) needles sustained only slight injury after 44 consecutive days exposure to Martian-Ievel UV (Cline and Salisbury, 1966). However, a number of herbaceous species, including red clover and other legumes were UV sensitive (Chne and Salisbury, 1966). It had been recognised for some time that numerous plant macromolecules (e.g. proteins and nucleic acids) absorb in the UV-B region (Giese, 1964). A range of adaptations and mechanisms against UV damage were suggested, i ncluding UV absorption by outer cell layers and formation of polyphenolics (Cline and Sal isbury, 1966). UV -B-protective functions had been proposed for flavonoids in epidermal plant layers (Jagger, 1967). I . Introduction 7 A major methodological problem of the early studies was that measurements of UV radiation intensities were taken without consideration for the biological effectiveness of the wavelengths applied. Thilty years ago, however, Martyn CaldwelI suggested the use of the generalised plant action spectrum, which defined a single curve for the relative photon effectiveness of UV-B for plant responses (Caldwell, 1968; Caldwell, 1971). This function was subsequently used in numerous examinations of UV -B-induced plant responses, strongly improving interpretation of measurements between studies. S ince then there was an exponential increase in UV -B plant research, leading to the publication of more than 600 papers (Caldwell et al., 1998). More than 60 books and reVIeWS have summarised this research, ranging from molecular to ecosystem levels (e.g. recently in Allen et al., 1998; Britt, 1999; Caldwell et al. , 1998 ; Hader et al., 1998 ; Huttunen et al., 1998 ; Jansen et al., 1998 ; Krupa et al. , 1 998b; Laakso and Huttunen, 1 998; Mackemess and Jordan, 1 999; Mackemess and Thomas, 1999; Short and Neckles, 1999; Tevini, 1999; Tevini, 2000; Zepp et al. , 1998). A review book recently summarised UV -B effects on terrestrial ecosystems (Rozema, 1999, and chapters therein) . UV-B effects on agriculture and agro-ecosystems have also been reviewed extensively (Campbell and Hofmann, 1998 ; Corlett et al. , 1997; Krupa and Kickert, 1 993; Krupa et aI. , 1 998a; Krupa et al., 1 998b; 01szyk et al., 1 996; Rozema et al., 1997c; Zinser et al., 1997) . Individual findings will be reviewed in more detail during the introductions for the individual Chapters of this thesis. Legumes and white clover It has been predicted that a primary consequence of increased UV -B in New Zealand would be a reduction in white clover growth rates to a greater extent than observed in grasses (Campbell et al., 1999). Previous findings (Krupa et al. , 1998b; Teramura, 1990) identify legumes as one of the most UV -B-sensitive plant families. Both temperate and tropical legumes appear to be affected (Singh, 1997). Various effects of UV-B on legumes have been described, from molecular studies (Jordan et al., 1994; Mackemess et al., 1998 ; Strid et al., 1996) to levels of growth (AntoneIIi et al. , 1997; S ingh, 1 997). However, other research suggests less distinctive effects on legumes (Deckmyn and Impens, 1995 ; Miller et al., 1994). Sensitivity appears to be dependent on the particular species involved and environmental conditions (Antonelli et aI., 1997; 1 . Introduction 8 Cen and Bornman, 1990). Developmental stage of the tissue (Jordan et al. , 1994) and duration of UV-B exposure also seem to be important (Bjorn et al., 1997 ; Deckmyn and Impens, 1995; Stephanou and Manetas, 1997). At higher UV -B levels, a white clover population in Jamaica showed decreases in growth (SLM) and UV -B-absorbing compounds, contrasting with other simultaneously investigated plant species (Rozema et al., 1997 a). Studies on species of clover and ryegrass showed variable results regarding their UV-B sensitivity (Deckmyn and Impens, 1998 ; Kaiser, 1995 ; Krupa et al., 1998b; Matthew et al., 1996; Nakayama et al., 1996; Newsham et al., 1998 ; Papadopoulos et al., 1999; Rozema et al., 1997a). While there is some information available on UV-B effects on native New Zealand vegetation (Hunt, 1997; Hunt et al., 1996), only one published study has so far described UV -B effects on the New Zealand pasture association (Matthew et al., 1996). This will therefore be described in more detail in the following. Using filters to reduce current UV-B radiation levels, the experiment tested effects of ambient UV -B levels on white clover and New Zealand pasture growth under sheep grazing (Matthew et al., 1996). A UV-B-transmitting perspex sunbed filter and a UV-B­ absorbing glasshouse polythene film were used to create contrasting UV-B environments. The UV -B-opaque canopy reduced ambient UV -B levels by about 87%, while the perspex filter absorbed ca . 25% of ambient UV -B . Both canopies reduced PPF levels by around 10-13%. PPF was ca. 3% higher and temperature was increased by O.4°C under the transmitting canopy, compared to the absorbing filter. Differences in growth parameters were not significant during the early harvest periods in spring and early summer but by mid-summer (when UV -B levels were highest) the UV-B-transmitting canopy decreased sward height and herbage mass by 14% and 19%, respectively. This was accompanied by a decline in rye grass leaf extension rate under the UV -B-transmitting filter. During summer, ryegrass leaf elongation decreased over time, showing a non-significant trend two weeks after grazing and a significant decrease three to four weeks later. This was more pronounced in young tillers (-25%), compared to adult plants (-5%). A similar effect was observed on white clover leaf area and petiole elongation, showing decreases of 14% and 22%, respectively. J. Introduction 9 These observations are consistent with several other studies using methods of solar UV­ B exclusion. For example, a reduction of ambient UV -B levels by 8-10% resulted in greater height and growth in bush bean in Portugal (Saile-Mark and Tevini , 1997). Using the ozone filter technique at a similar location, 10-35% lower growth was found in sunflower and maize seedlings at near ambient UV-B levels compared to a 30% reduced UV -B environment (Mark and Tevini, 1997). However, in other studies using differentially UV -transmitting filters in the Netherlands, no solar UV -B effects on biomass production, morphology and photosynthesis could be detected in four coastal grassland species (Tosserams et al., 1996) or in Vicia faba seedlings (Visser et al., 1997b). In the New Zealand study, there was a trend suggesting that clover percentage in the sward was reduced in the UV -B-transmitting treatment but the generality of this finding was limited due to inherent constraints in the use of UV -B-transmitting canopies for this type of research. In particular, pronounced effects of the covering canopies were observed in terms of temperature, possibly leading to higher nitrogen levels, and thus increasing the longevity and content of rye grass in the sward. This is supported by the simultaneously measured decrease in rye grass leaf senescence under both covers. Nevertheless, the results of that study suggest that the present solar UV -B levels in New Zealand can have an effect on regularly defoliated New Zealand field swards, impacting on both white clover and ryegrass and possibly also on the interaction between the two speCIes. Population studies White clover i s a plant species that can conveniently be used to investigate within­ species differences in stress responses due to extensive plasticity, on both the genotype and the population levels (Caradus, 1994; Caradus et al. , 1993a) . Population comparisons are not only useful in examining the consistency of the UV -B response within a species but also allow the assessment of relationships between plant attributes and UV -B responses in plant types not confounded by species-specific differences. In the context of the present work, the term 'population' will include cultivated varieties (cultivars) and breeding lines as well as ecotypes collected in the wild. I. Introduction 1 0 I f selection pressure as a result o f increased UV -B radiation i s severe, i t i s possible that the population structure of pastures may shift towards genotypes which are better adapted to higher UV -B levels. It has been suggested that a shift in population genetics may be most rapid in the clover populations (Campbell et al., 1 999). Due to the l arge intrinsic genetic plasticity within many developmental aspects of white clover it i s reasonable to expect differential responses to UV-B among white clover populations. White clover has shown population differences in response to a number of other environmental variables, including drought (Wang et al., 1 996; Woodfield and Caradus, 1 987), mineral supply (MacKay et aI. , 1 990), frost (Caradus et al., 1989a) and aluminium stress (Caradus et al., 1 987). Population-specific responses to UV-B have been demonstrated in a wide variety of plant species, for instance in birch (Lavola, 1 998), Brassica (Allen et al., 1 997; Olsson et al. , 1 998), Coleus (Burger and Edwards, 1 996), cucumber (Middleton et aI., 1 996), maize (Correia et al., 1 998; Correia et al., 1 999a), pea (Gonzalez et al. , 1 996; Mepsted et al., 1 996), Pin us (Pukacki and Modrzynski, 1 998), Phaseolus (Saile-Mark and Tevini , 1 997), rice (Barnes et al., 1 993 ; Dai et al. , 1 994; Kumagai et al., 1 999), Silene (van de Staaj i et al., 1 997), soybeans (Murali et al., 1 988) and wheat (Hader, 1 996). In soybeans, for example, studies suggest UV-B sensitivity for about two thirds of the examined populations (Biggs et al., 198 1 ; Teramura and Murali , 1986). Underlying mechanisms There are suggestions that origin from habitats high in ambient UV -B irradiation confers UV -B tolerance (Barnes et al. , 1 987). Other studies, however, do not support this hypothesis. Rice cultivars from equatorial origin in general were not more tolerant to UV-B than those from higher latitudes (Dai et al. , 1 994). Tropical plant species also can exhibit considerable sensitivity and it has been shown that current levels of UV-B already impact on low-latitude species (Searles et al., 1 995). Furthermore, it has been proposed that tolerance to other forms of stress may confer UV-B tolerance (AI-Oudat et al. , 1 998; Gwynn-Jones et al., 1 999b). Stress in the context of this thesis is defined as the extent to which external constraints l imit dry matter production (Grime, 1 979; Poorter and Gamier, 1 999). A recent discussion of 1 . Introduction 1 1 potential UV -B effects on plant population genetics suggests that elite germplasm producing high yields may be more UV -B-sensitive, while UV -B tolerance in natural plant populations may be l inked to smaller size (Groth, 1 998). This would also be supported by ecological theory on stress tolerance, establishing a link between plant productivity (growth rates), habitat productivity and stress tolerance (Lambers and Poorter, 1 992; Poorter and Gamier, 1 999). According to this, a primary axis of adaptation separates 'competitors' (specialised primarily towards high rates of biomass acquisition in productive environments, but vulnerable to high rates of resource loss under stress) from 'stress-tolerators' showing lower productivity and higher resistance to biomass reduction in unproductive habitats (Grime, 1 979; Grime et al., 1 997). This is also of particular relevance for plant functional types. These are sets of plants that exhibit similar attributes in response to environmental conditions (Diaz and Cabido, 1 997). To date no studies are available that investigate the relationship between UV-B responsiveness and functional plant types (Gwynn-Jones et al., 1 999b). This i s compounded by the fact that usually l ittle environmental background information i s available in most plant studies on UV-B responsiveness (Gwynn-Jones e t al., 1 999b). It is surprising that to date these relationships have not been investigated in detail for UV-B responsiveness. It is timely to test whether stress-tolerant populations are better protected against UV -B , and if so, how this could arise. In particular it can be proposed that slow-growing stress-tolerant plant types would be more tolerant to UV-B than faster-growing, productive plants. For the New Zealand situation, it is of importance to determine how the commonly used white clover cultivars bred for high productivity compare with ecotypes from natural, particularly stress-exposed, environments. Common cultivars are often relatively sensitive during periods of water stress and there are intraspecific differences i n this response (Wang et al., 1 996). In assessing genetic variation in UV-B tolerance there is a need to identify: ( 1 ) which attributes appear to make plants more susceptible or more tolerant/resistant to UV-B damage, and (2) which attributes appear to be the most responsive to enhanced UV-B exposure. Attributes often related to UV -B tolerance in experimental studies include content of UV -B absorbing compounds and anti-oxidants, leaf thickness, specific leaf area, architecture, and habitat from origins with naturally high UV -B background levels. Attributes most likely to be affected by elevated UV -B radiation in experimental studies 1. Introduction 1 2 include growth, specific leaf area, content of DV -B absorbing compounds, palatability, stomatal conductance, leaf water content, and architectural parameters (Campbell and Diaz, 1 998; Krupa et al., 1 998b; Rozema et al. , 1 997b). The use of multivariate statistical approaches, particularly principal components analysis (PCA), has been recommended for studies of numerous plant attributes across various plant species or populations (Grime et al. , 1 997). Such multivariate analytical techniques are readily available in most statistical packages and allow combined analysis into a single data set which normally would require numerous univariate analyses of variance (Matthew et al., 1 994). In particular the usefulness of these approaches has been pointed out for summarising multivariate data sets in ecological investigations, e.g. of consequences due to c limate change (Diaz and Cabido, 1 997). PCA was used in the examination of species tolerance to extreme events (MacGillivray and Grime, 1 995), showing that nutrient stress-adapted plant species also display greater resistance to drought and frost. Other studies used PCA to illustrate that the trade-off between high rates of resource acquisition and stress tolerance is linked to differential mineral nutrient accumulation (Grime et al., 1 997). A study of environmental adaptation within three populations in Avena barbata utilised PCA and discriminant analysis to show that the populations mainly differ in their adaptation to reduced water availability and to different light conditions (Somersalo et al., 1 998). PCA demonstrated greater tolerance to drought and frost in stress-adapted plant species (MacGillivray and Grime, 1 995). PCA and cluster analysis were also used to classify white clover cultivars on the basis of leaf size and cyanogenesis (Caradus et al. , 1 989b). However, despite their proven usefulness in ecological literature, there is a general lack of multivariate investigations in the examination of DV -B responses in plants. Interaction with other factors Drought A key issue to be determined is whether effectiveness of increased DV-B radiation on pasture species would be altered by other environmental factors, such as nutrient or water stress. The necessity to investigate DV -B effects in conjunction with other stress factors including drought has been pointed out in a number of reviews (e.g. Bornman 1 . Introduction 1 3 and Teramura, 1 993; Caldwell et al., 1 995 ; Jordan, 1 996; Tevini, 1 993). Findings from earlier studies (Murali and Teramura, 1 985b; Murali and Teramura, 1 986b; Sullivan and Teramura, 1 990; Teramura et al. , 1 984a; Teramura et al., 1 984b; Teramura et al., 1 983 ; Tevini et al. , 1 983a) strongly suggested the need to study this interaction for species of the New Zealand pasture system, where the phase of high UV -B levels often coincides with periods of summer drought. More than 20 studies have investigated the UV -B effects on plants in combination with differential water supply. This includes field-based research (Allen et al., 1 999; Barnes et al., 1 995 ; Bjorn et al., 1 997; Chaves et al. , 1 997; Conner and Zangori, 1 997; Drilias et al., 1 997; Manetas et al., 1 997; Murali and Teramura, 1 986a; Murali and Teramura, 1 986b; Nikolopoulos et al., 1 995; Petropoulou et al., 1 995; Stephanou and Manetas, 1 997; Sullivan and Teramura, 1 990; Teramura et al., 1 990) and indoor studies using glasshouses or controlled environment rooms (Balakumar et al., 1 993 ; Chaves et al., 1 997; Conner and Zangori, 1998; Eswaran et al. , 1 993 ; Nogues et al. , 1 998; Premkumar et al. , 1 993 ; Schmidt et al., 2000; Teramura et al., 1 984a; Teramura et al., 1 984b; Teramura et al., 1 983 ; Tevini et al., 1 983a). UV -B has been shown to aggravate drought effects, e.g. for photosynthetic parameters or growth (Nikolopoulos et al., 1 995 ; Teramura et al., 1 984b; Tevini et aI. , 1 983a) . Paradoxically, however, water stress may also reduce the damage caused by UV-B exposure, suggesting mutual amelioration of the stress forms by each other. For example, beneficial effects of the UV -B x drought stress combination have been described for anatomical and biochemical responses in cowpea (Balakumar et al. , 1 993). In pea, UV -B was found to ameliorate the drought impact via reductions in leaf area, stomatal conductance and plant water-loss (Nogues et al., 1 998). Similar effects have been observed in several studies of Pinus and can be related to UV -B-induced morphological changes such as increased leaf cuticle thickness (Bjorn et al., 1 997; Manetas et al., 1 997) . Higher leaf water content under the UV -B x drought combination in Arabidopsis has been related to biosynthesis of stress proteins and osmolytes (Schmidt et aI., 2000). In Cistus, not only a UV -B-induced stimulation in flavonoid production was found, but also a synergistic increase of this response under drought i. introduction 1 4 (Chaves e t al., 1 997). The production o f flavonoids under these conditions is seen to contribute to stress protection. Interaction of UV-B and drought might also alter within-species differences in UV-B sensitivity. Contrasting morpho-physiological responses to UV -B were observed in two soybean populations in dependency to moisture supply (Murali and Teramura, 1 986b). While the cultivar 'Williams' remained unaffected, 'Essex' displayed decreases in yield, leaf area, CO2 assimilation and stomatal resistance under adequate precipitation, but not in seasonal drought (Murali and Teramura, 1 986b; Teramura et al., 1 990). These findings highlight the need to investigate mechanisms that could alter effectiveness of one stress form by another, e.g. changes in the levels of UV -B-absorbing compounds or osmoprotectants. Duration of UV-B exposure Furthermore, duration of UV-B exposure can also modify UV-B effectiveness. Studies in cucumber showed that increased time of exposure to UV -B can lead to detrimental effects for growth and photosynthesis (Tevini et al., 1983a). Similarly cumulative effects have also been observed in heathland species of Calluna and Vaccinium (Bjorn et al., 1 997). Other findings, however, showed beneficial effects of time and plant development for UV-B sensitivity (Teramura et al., 1 984b; Visser et al., 1 997b). Plant-herbivore relationships It has been predicted that UV-B-induced reduction of white c lover content in pasture ecosystems and resulting reduced nitrogen input would impact negatively on the forage production and forage quality of New Zealand pastures (Campbell et aI., 1 999). Lower overall productivity as a result of lower nitrogen inputs would increase feed deficits for livestock, and result in a reduced stock carrying capacity. Potentially, a lower proportion of clover in the pasture can also modify grazing behaviour. In cases where there is a high clover content, higher utilisation of the grass component of the community can be achieved by grazing l ivestock. Changes in the biochemistry of plant tissues might also alter the nutritional value of the herbage per unit of dry matter. From other work it could be anticipated that protein, carbohydrate, fibre contents and I . Introduction 1 5 secondary plant chemical compounds may be altered in white clover as a result of higher UV-B radiation levels (Grant-Petersson and Renwick, 1 996; Hatcher and Paul, 1 994; Yazawa et al., 1 992). Direct and plant-mediated indirect effects of UV-B on consumer organisms have been suggested for some time (Caldwell et al. , 1 989). To date some studies have investigated UV -B effects on the plant-herbivore relationship (e.g. Grant-Petersson and Renwick, 1 996; Hatcher and Paul, 1 994; Lavola et al., 1 998; Rousseaux et al., 1 998; Salt et al., 1 998). Considering the impact of UV -B on secondary metabolism (Grant-Peters son and Renwick, 1 996; McCloud and Berenbaum, 1994; Meijkamp et al., 1 999) such investigations seem warranted, especially for pasture plants like white clover where changes could influence secondary metabolism-mediated plant interactions with several trophic levels. This includes nitrogen fixation, attraction to pollinators, plant-herbivore interaction and forage quality (Dakora, 1 995). Production of flavonoids and other phenolics in response to UV -B treatment has frequently been described and reviewed (Bornman et al., 1 997 ; Jordan, 1 996; Meijkamp et al. , 1 999). Such changes can also relate to protection of plants from increased levels of UV-B (Bornman et al., 1997 ; Middleton and Teramura, 1 993). UV -B-screening effects as well as radical-scavenging antioxidant functions and more effective energy dissipation have been proposed (Born man et al. , 1 997; Meijkamp et al. , 1 999; Smith and Markham, 1 998). There is evidence that UV -B-induced phenolics such as flavonoids can act as feeding deterrents (Dakora, 1 995 ; Grant-Petersson and Renwick, 1 996), although other findings demonstrate that the involvement of plant phenolics in host plant resistance is less than conclusive (Bi et al., 1 997 ; Lavola et al., 1 998). UV -B may influence herbivory also through other changes in nutritive quality, e.g. alteration in levels of amino acids (Salt et al., 1998), carbohydrates or nitrogen (Hatcher and Paul, 1 994). In pea, it was shown that increases in leaf nitrogen concentration, rather than phenolics, determined herbivore response (Hatcher and Paul, 1 994). UV-B-mediated changes in nitrogen production could be of particular importance for white clover populations depending on nitrogen­ fixation for their nitrogen intake and producing nitrogenous compounds such as cyanogenic glucosides. Until now, no information has been available on the alteration of cyanogenesis in response to UV -B treatment. In some white clover populations, the presence of the cyanogenic glucosides linamarin and lotaustralin, and the structural gene i . introduction 1 6 for the enzyme linamarase, can result in cyanogenesis in the leaf tissue. This production of cyanide in response to stress can deter feeding insects such as molluscs and possibly insects. Plants that possess only the cyanoglucosides, but not the enzyme, are protected against herbivores that have beta-glucosidases in their gut, e.g. molluscs (Kakes, 1989). In June 1 998, a field-based DV -B supplementation experiment was begun in Palmerston North, New Zealand to test some of the predicted effects of increased DV -B radiation on an established rye grass and white clover (cultivar 'Huia') pasture. Three supplemental DV -B radiation treatments are being used, representing 0, 1 5 and 30% reductions in annual ozone concentration. These correspond to predicted c lear-sky summer maximum increments in plant-weighted DV -B doses at the Palmerston North site of 0, 30 and 72% respectively, according to model predictions (McKenzie, 1 99 1 ). Preliminary results from this experiment illustrate that detectable changes are occurring in the growth and development of species within the pasture. For example, measurements of cyanogenesis conducted by the thesis author in spring 1 998 using a picrate paper test (Williams et al. , 1 998) indicate a rise in cyanogenesis in leaves of field-grown 'Huia' with increasing exposure to elevated DV -B radiation (Fig. 1 - 1 ) . This response is consistent with higher stress on the plant population, and provides an early indication of changes which might impact on herbivory by invertebrates. Shifts in plant­ herbivore interactions may in turn directly or indirectly (via trophic cascades) alter community and ecosystem dynamics. Very few studies, however, have addressed effects of DV -B radiation on trophic interactions, and none have been conducted with species of importance to New Zealand agriculture. The need for research activities in a number of areas outlined above is in several aspects also reflected by earlier recommendations, urging research on plant species introduced into New Zealand agro-ecosystems (Laing, 1 99 1 ; Laing, 1 993; McKenzie, 1 992), including: ( 1 ) investigations on New Zealand pasture species, (2) estimates of the degree of DV -B sensitivity in these species, (3) detennination of the degree of variability in DV -B sensitivity, (4) examinations of mechanisms of DV -B action, (5) determination of the metabolic costs of DV -B protection, and (6) detennination of DV­ B effects on product quality (e.g. flavour). i. introduction 17 - :lE 1 .2 0 -ctI ..!!! ... . C) E 0.8 - � 0 (.) (/) Z 0.4 (,) :r: o t---'--_ .......... _-,-- UVBO UVB15 UVB30 Fig. 1 - 1 . Intensity of cyanide (HCN) production in the white clover cultivar 'Huia' growing at three DV -B levels in a grazed pasture community in Palmerston North, New Zealand (September 1 998). The bars are the standard errors of the means. DVB 0, DVB 1 5 and DVB 30 correspond to UV-B levels at 0, 1 5% and 30% ozone depletion, respectively. DM = dry mass. Figure published in Campbell et al. ( 1 999). General description of the experimental approaches All experiments used growth rooms at the National Climate Laboratory in Palmerston North, New Zealand with average PPF levels of 420 to 435 /-lmol m-2 S- I and plant­ weighted UV-B levels of 1 3 .3 kJ m-2 d- 1 , mimicking 25% summer ozone depletion above Palmerston North. All experiments were conducted on a population level to ( 1 ) determine the generality of possible UV -B responses within white clover, (2) expose genetic differences among white clover populations in response to UV-B, and (3) to discern possible functional mechanisms underlying differential UV-B responsiveness. A number of white clover populations with a wide range of genetic and geographical backgrounds were used in the experiments. This comprised populations collected in the wild (ecotypes), including high latitude populations from Scandinavia, low latitude populations from South America as well as mid l atitude (temperate) populations from Asia, Southern Europe and New Zealand. Other populations used were productive agricultural cultivars selected for high agronomic value, breeding l ines with ornamental i . introduction 1 8 leaf characteristics and an octoploid line with double the usual tetraploid white clover chromosome set. Detailed description of the white clover populations used can be found in Chapter 2 (see Table 2- 1 ) . These investigations included examinations of other biotic and of abiotic interactions. This involved interaction with drought, time and insect species-specific herbivory as well as UV-B susceptibility of frequently defoliated plants. A number of morphological and physiological investigations were conducted, including attributes l inked to photosynthetic function, nutritive value and insect performance as well as morphogenetic features of bud and leaf structure (Fig. 1 -2). A subset of these attributes represented those that could be particularly important for UV -B protection (Fig. 1 -2). It was anticipated that UV -B effects on these features would eventually be reflected in morphological changes and altered productivity (Fig. 1 -2). Aims In particular these studies intended to test the following hypotheses: 1 . White clover i s sensitive to UV-B under controlled environmental conditions, based on total plant growth, morphology and physiology. 2 . These responses differ among white clover populations. 3 . Differences in UV-B sensitivity at the population level can be l inked to plant productivity. 4. UV -B responsiveness is related to other constitutive and UV -B-induced features of white clover morphology or physiology. 5 . UV-B tolerance can be linked to habitat adaptation, particularly tolerance to other forms of stress. 6. UV -B responsiveness i s modified by drought. 7 . UV -B effectiveness changes over time. 8. UV -B responsiveness can be detected in frequently defoliated white clover leaves. 9. The performance of insect herbivores differs when reared on UV -B-treated white clover foliage. >­CJ o .....I o U5 >­:::r: 0. o :::r: 0. a: a � z a i= () LU I­a a: 0. / Photosynthesis & water relations � • PSI! function & chlorophylls • Chloroplast ultrastructure • Mesophyll structure • Stomatal density/distribution " • Leaf water potential � UV-B absor)2tion & dissi)2ation • UV -absorbing compounds • Flavonoids I I Nutritive value • Nitrogen • Carbohydrates • NDF • Cyanogenesis • Insect performance � I I Antioxidant function • Proline • Flavonoids I / Structure & t mor:Qhogenesis Mor:Qhology • Primordia & growth formation • Total mass • Epidermal • RGR cell size • Organ mass " • Cell number � • RSR • LWR • Leaf size Structure & • Petiole length mor)2hogenesis • LAR • Cell geometry • SER • SLM & PDM • NGPs t Fig. 1 -2. Scheme of the current research project, based on morphological and physiological attributes i n white clover. LAR = leaf appearance rate; L WR = leaf weight ratio; NDF = neutral detergent fibre; NGPs = number of growing points ; PDM = percent leaf dry mass; PSII = photosystem II; RGR = relative growth rate; RSR = root:shoot ratio; SER = stolon elongation rate; SLM = specific leaf mass. 2. peA of UV-8 responses in 26 white clover popu/ations 20 Chapter 2 . Trifolium repens responses to u ltraviolet-8 radiation : i ntegrated analysis of 1 7 traits and 26 populations DV -B screening facilities: computer feedback system and walk-in growth chamber An abbreviated version of this Chapter has been submitted to Science. 2. peA of UV-B responses in 26 white clover populations 2 1 Abstract The pasture legume white clover (Trifolium repens L.) is experiencing increased levels of ultraviolet-B radiation (UV-B) in New Zealand due to depletion of the stratospheric ozone layer. This study analysed the consequences of UV -B supplementation on 1 7 morphological, morphogenetic and physiological attributes of T. repens to identify traits l inked to the UV -B response. Principal components analysis (PCA) was conducted on 26 T. repens populations to explore within-species variation in this response. After 1 8 days of exposure to 1 3 .3 kJ m-2 d- 1 UV-B in controlled environments, UV-B significantly decreased total plant dry mass (-26%) with reductions in leaves, stolons and roots. A less pronounced UV -B effect on roots led to a 7% increase in the root:shoot ratio, concomitant with a 2% decrease in the leaf weight ratio. Reductions were also observed in relative growth rate (- 1 4%), number of growing points (-9%) and petiole length (-22%). Differential decreases in leaf size and leaf dry mass resulted in an 8% increase of the specific leaf mass. Adaxial epidermal cell size was reduced by 1 0% and stomatal density was increased by 1 1 %. UV -absorbing compound levels increased slightly by 3% and intrinsic efficiency of photosystem II displayed a small overall reduction by 2% in response to UV -B . Aspects of cell division and cell expansion were negatively affected by UV-B . The main PCA dimension of UV-B sensitivity in T. repens (PC 1 ) was characterised by changes in whole plant growth attributes, young leaf growth and, to a lesser degree, UV -absorption. UV -B-induced alterations in allocation and partitioning within and between plant organs constituted a secondary tier of UV-B responsiveness (PC2). Overall , UV-B tolerance was more easily predicted by traits linked to features of plant and habitat productivity, rather than by the l ikely UV -B level in the habitat of origin. Plant attributes related to UV -B tolerance included lower relative growth rate and smaller epidermal cell size as well as higher UV -B-induced accumulation of UV -absorbing compounds. Ecotypes from environments exposed to other forms of stress ranked highly for UV -B tolerance in the multivariate assessment, while several agricultural cultivars exhibited greater overall UV -B sensitivity. These results suggest differences in UV -B sensitivity depend on ecological specialisation with particular advantages for stress-adapted populations that have not been selected for high yield. 2. peA of UV-B responses in 26 white clover populations 22 I ntroduction Depletion of the stratospheric ozone layer is leading to increased levels of ultraviolet-B radiation (UV -B, 290-3 1 5 nm) reaching tropospheric levels in New Zealand (McKenzie et al., 1 999). Numerous responses to UV -B have been observed for plants, affecting molecular, physiological, morphological and ecosystem levels (e.g. reviewed more recently in Caldwell et al., 1 998; Mackerness and Jordan, 1 999; Rozema, 1 999; Tevini, 2000). Morphological changes frequently observed in response to UV -B irradiation include decreases in productivity, plant biomass and leaf area (Correia et al., 1 999b; Li et al., 1 998), increases in leaf thickness and specific leaf mass (Cen and Bornman, 1 990; Cen and Bornman, 1 993) as well as changes in biomass allocation (Gwynn-Jones and Johanson, 1 996; Li et al., 1 998). UV -B-induced differences in leaf surface morphology have also been reported, including changes in the number and size of epidermal and stomatal cells (Staxen and Bornman, 1 994; Stewart and Hoddinott, 1 993; Tevini et al. , 1 983b; Visser et al., 1 997a). There are also reports on UV -B-elicited decrea')es in the intrinsic efficiency (FJFm) of photosystem II (PS II), indicating inhibition of photosynthetic function in PS II (Zeuthen et al., 1 997). Frequently, UV­ absorbing compound levels are increased in response to UV-B (Day and Demchik, 1 996; Jordan et al. , 1 998). This is considered to be a potential mechanism of protection as these compounds can act as selective optical filters (Murali and Teramura, 1 986c). White clover (Tr(folium rep ens L. ) is an important forage legume in temperate regions of the world, constituting high quality feed while efficiently providing nitrogen to pastures (Brock et al. , 1989). T. rep ens has shown a wide variety of plasticity, on both the genotype and the population levels (Caradus, 1 994). It could be expected that variation might exist among T. rep ens populations with respect to sensitivity to UV -B. Several previous studies have reported differences in morphological and physiological UV -B responses among populations of a species, (e.g. Barnes et al., 1 993; Correia et al. , 1 999a; Dai et al., 1 994; Murali and Teramura, 1 986c). These differences might also be used to suggest possible underlying mechanisms of UV -B sensitivity. Some reports examining the ambient UV -B radiation environment in which plants grow indicate higher susceptibility for plants from lower natural UV-B backgrounds (Barnes et al., 1 987; Sullivan et al., 1 992). There are other indications that UV-B sensitivity 2. peA of UV-8 responses in 26 white clover populations 23 could be a function of intrinsic susceptibility to other forms of stress, for instance to drought (AI-Oudat et al., 1 998). Tolerance to one form of stress due to adaptation to another has also been demonstrated in other studies. For example, principal component analysis of 26 traits in 2 1 British plant species revealed greater tolerance to drought and frost for nutrient stress-adapted species (MacGillivray and Grime, 1 995). Similar multivariate analytical approaches have also been used in other investigations on stress tolerance (Grime et al., 1 997; Somersalo et al. , 1 998). However, despite their proven usefulness in ecological literature, there is a general lack of multivariate investigations in the examination of UV -B responses in plants. This appears particularly relevant for investigations of UV-B sensitivity based on numerous traits and several populations within a species. This study analysed 1 7 morpho-physiological characteristics of 26 T. repens populations grown with and without supplementation of UV -B radiation. The UV-B response for each trait and each population was established from the difference between treatment and control values and the resulting population x response matrix was subjected to principal components analysis. Attributes included aspects of plant size and growth, measurements of biomass allocation, shoot and leaf morphology, leaf surface characteristics, intrinsic photochemical efficiency of PS I I and concentration of UV­ absorbing compounds. The objective was to determine the extent of changes for these attributes in T. repens and, as a consequence, to obtain a summary of the multivariate effects characterising the overall UV -B response. Numerous populations of T. repens with a variety of original habitats were used to explore whether they might appear dissimilar in their overall UV -B sensitivity. From this it was intended to analyse which plant features would distinguish UV -B-tolerant from sensitive T. repens populations. It was hypothesised that stress-adapted populations or those from an origin of higher background UV -B levels would be more tolerant to UV-B. 2. peA of UV-B responses in 26 white clover papulations 24 Materials and Methods Experimental design The experimental design was a 2 x 26 factorial, with two levels of UV -B radiation and 26 populations of T. repens. There were ten replicated pots per population in each UV­ B treatment. Pots were arranged randomly on trolleys which were rotated in each room every second day to reduce effects from differential shading under the UV-B rigs. Plant cultivation and UV-B irradiation T. repens seeds were obtained from the Margot Forde Germplasm Centre in Palmerston North, New Zealand. After scarification, seeds were placed in petri dishes on moist filter paper and incubated at 25°C for 3-5 days. The emerging seedlings were transferred to trays containing watered potting mix and placed in a glasshouse at a temperature of 20°e. After four weeks, the seedlings were transplanted to individual 1 .5 L pots in the same medium. Eight weeks later, four stolon cuttings were taken from each pot and planted pair-wise in steril ised sand/mineral mix at opposite corners of square 3 L pots. Pots were placed in growth rooms of the National Climate Laboratory (Palmerston North) and supplied three times daily for 2 min with an automatic drip irrigation system using half-strength Hoagland's nutrient solution. The average photosynthetic photon flux (PPF) during the 1 0 h light period was 435 Jlmol m-2 S- I , supplied by a water-screened array of four 1 kW Sylvania 'metal-arc' high pressure discharge lamps, together with four 1 kW Philips tungsten iodide lamps. PPF was measured using a LI-COR 1 85 meter with attached quantum sensor. Day/night room temperatures were 1 8°C and 1 3°C. The day/night values for relative humidity were 80% and 73% and those for mean CO2 levels 370 ppm and 390 ppm. After eight weeks, pots were thinned down to one plant per pot and 1 0 representative stolons from each population were used to determine initial plant dry mass. The remaining plants were allowed to grow without supplemental UV -B for a further 1 0 days before onset o f the UV -B treatment. The level o f irradiation was 1 3 .3 kJ m-2 d- 1 of plant-weighted, biologically effective UV-B (Caldwell , 1 97 1 ) , normalised to 300 nm. 2. peA of UV-B responses in 26 white clover populatiofls 25 This dose corresponds to a 25% depletion of summer ozone levels above Palmerston North (latitude: 40.2 S, longitude: 1 75 .4 E). DV -B radiation was supplied 8 hours per day for 2 1/2 weeks in the middle of the photo-period from Phil ips TL 40WJ l 2 RS DV fluorescent tubes, enclosed in cellulose diacetate filters. The tubes were suspended 500 mm above plant level. A Solar Light Company model 50 l A UV biometer sensor measured the DV output which was controlled by a feedback system to maintain the constant, pre-set UV-B irradiation doses in response to tube ageing and cellulose acetate filter degradation. UV-B levels were monitored and recorded digital ly with an IBM-PC computer that adjusted an analogue voltage supplied to Osram high-frequency dimming ballasts to achieve the required UV -B output from the fluorescent lamps. Cellulose diacetate filters were changed when the feedback system output approached maximum voltage. A Bentham DM1 50 DV spectroradiometer was used to calibrate the sensor at regular intervals against the generalised plant action spectrum (Caldwell , 197 1) . Plant material Twenty-six T. repens populations with a wide range of genetic and geographical backgrounds were included in the experiments (Table 2- 1 ) . S ixteen populations were ecotypes, i .e. populations collected in the wild. They included high latitude populations from Scandinavia, low latitude populations from South America as well as mid latitude (temperate) populations from Asia, Southern Europe and New Zealand. A further eight populations were agricultural cultivars, selected for high agronomic value in breeding programmes. Two further populations, 'Green' (higher chlorophyll levels) and 'Lime' (lower chlorophyll levels) represented unmultiplied selections (breeding lines) of the cultivar 'Huia'. Table 2-1a. Site characteristics of 1 6 T. repens populations collected in the wild (ecotypes), arranged on a regional basis by latitude. Accession numbers are from the Margot Forde Germplasm Centre in Palmerston North, New Zealand. N° I 2 3 4 5 6 7 S 9 1 0 I 1 1 2 1 3 1 4 1 5 1 6 Site characteristics Annual Altitude rainfall Accession Name Country Latitude (m) (mm) Other descriptors C 1 5 ! 3 3 Kangos a S weden 67 .3°N 270 500 Sand with clay, abandoned cultivation, north of arctic circle. C 1 5 1 2 1 Koskenkyla a Finland 66.3°N 1 00 500 Sand with clay, cultivated land, near arctic circle. C 1 5 1 30 Haggas a S weden 64.2°N 500 700 Southwest clay slope, abandoned cultivation. C76 l S Sacile b I taly 45.5°N 1 00 1 000 Open road verge, likely ladino type; with relative climatic tolerance but lower adaptation to sustained drought. C9507 Veneto b Italy 4 5 . 5 °N 250 1 000 Shady road verge in forest, likely ladino type! with relative climatic tolerance but lower adaptation to sustained drought. C 8749 Tienshan C China 43.0oN 2000 600 Hay meadow, heavy winter snow, very cold. C I 054 1 Gudauri d Georgia 42.3°N 2390 1 200 Alpine meadow, western slope, moderate grazing pressure, cold. C I 0523 Schrosha d Georgia 42. 1 oN 275 500 Fallow with shrubs, southeast slope, light grazing pressure. C 1 0525 Sviri d Georgia 42. 1 oN 1 55 900 Pasture and trees on 11at, light grazing pressure. CS734 Sarikamis d Turkey 40.2°N 2300 5S0 Basaltic, extended snow cover and clouds, very cold. C7986- Kahumingi e New 40,SO S 200 6 1 5 North slope, dry hilI country pasture, relatively high soil cation levels and lower organic matter %, plant CS022 Zealand genetic background most likely from cv. 'Huia'\ relati vely tolerant to nematode damage compared 10 'Kavala' (but lower than 'Huia'). CS240- Kavala e New 40,SO S 375 1 400 South slope, wet hill country pasture, relatively low soil cation levels and higher organic matter %, plant CS270 Zealand genetic background most likely from cv. 'Huia'\ prone to nematode damage compared to 'Kahumingi'. C 1 3286 Irazu f Costa Rica 9 , SON ca. 2000 ca. 1 750 Roadside, tall type. C6840 Colombian g Colombia ca. 5 .00N nJag nla n/a. C l 6028 Tulcan h Ecuador 0.5°N 3200 >2000 Highland pasture with low clover content ( 1 %), acidic peat soil with high A I (pH 4.5), mean annual temp C l 6029 Sanana h Ecuador O. I °S approx. 1 1 °C, low grazing pressure. 3600 > 1 500 Sloping open highlands pasture with sporadic clover content, acidic volcanic soil (pH 5), mean annual temp approx. 1 0°C, moderate grazing pressure. a (Nordiska Genbanken, 1 993); b (Forde and Easton, 1 986); C Field notes Margot Forde China field collection expedition 1 98 7 ; d (Caradus and Forde, 1 996); e (Caradus et aI., 1 990b); f Field notes from Costa Rica field collection (pers. comm. Paul Mueller); g 'Colombian' site description data not available (nJa); h Field notes from Ecuador field collection (pers. comm. Alan Stewart); i Derek Woodfield (pers. comm. ) Table 2-1b. Characteristics of 1 0 T. repens populations selected or bred for particular traits (cultivars and breeding l ines). Accession numbers are from the Margot Forde Germplasm Centre in Palmerston North, New Zealand. N° Accession Name Country Type Description 1 7 C5067 Aran j Ireland Cultivar B red from French and Israel germplasm, high yielding, performs well in hay sward, less tolerant to grazing, some disease resistance. 1 8 C7545 Feathermark i New Cultivar Similar genetic background and agronomic performance to 'Huia', red (anthocyanin-richk) midrib leaf mark. Zealand 1 9 C I 3 1 73 Haifa i I sraeV Cultivar B red from Israel germplasm, high yielding, heat- and salt-tolerant, more sensitive to frost and drought, decreased persistence Australia under grazing, relatively poor disease resistance, good performance i n lowland conditions. 20 C7544 H u i a i New Cultivar Long-established New Zealand general purpose cultivar, selected for wide adaptation and yield, high phenotypic plasticity, Zealand persistence to grazing, high amino acid content, sensitive to P-deficiency, water stress and frost, relatively poor disease resistance, best suited for lowland conditions. 2 1 C I 3 1 7 6 Kopu i New Cultivar Incl udes germplasm from ladino cultivars, 'Pitau' (and thus 'Huia') and a Spanish strain, bred for lowland farming, nematode Zealand resistance and yield, more productive and leaves more upright habit than 'Huia', recommended for grazing, l arge roots but no persistence in dry hill country, frost sensitive. 22 C876 1 Luclair i France Cultivar Leaf mark discolouration and patch of anthocyaninl, medium frost tolerance and forage yield, better drought resistance than 'Huia'. 23 Cl 1 67 8 Prestige i New Cultivar Bred from ecotypes in Northland (northern New Zealand), higher productivity than 'Huia', improved heat and drought tolerance, Zealand grazing-tolerant, suited for hil l country climate and dry sunny slopes, relatively good disease resistance. 24 C 1 1 697 S u stain j New Cultivar Includes Mediterranean and US germplasm, highly stoloniferous and persistent, well suited for grazing. Zealand 2 5 C 1 5 7 4 1 Green m New Breeding Genetic background from 'Huia', d ark green phenotype selected for high chlorophyll content. Zealand line 2 6 C 1 5740 Lime m New Breeding Genetic background from 'Huia', l ime green phenotype selected for low chlorophyll content. j (Caradus, 1 986) and (Caradus and Woodfield, 1 997); k (Corkill, 1 963); I (Lenoble and Papineau, 1 970); m Description of 'Lime' and 'Green' from Derek Woodfield (pers. comm.) N -..) 2. peA of UV-B responses in 26 white clover populations 28 Whole plant growth measurements At harvest, aerial plant parts were collected and stored in plastic bags under darkness at 4°C before separation into stolons and leaves. After removal of the pots, roots were washed and separated from sand. All plant parts were dried at 80°C for 48 h. Allocation parameters calculated were root: shoot ratio (RSR, the ratio of root mass over shoot mass) and leaf weight ratio (L WR, the ratio of leaf mass over total biomass). Relative growth rate (RGR) per day was calculated from initial and final total plant dry mass. The number of growing points was obtained by counting the number of stolon tips present on each plant at final harvest. Leaf measurements Leaf size (area per leaf) of four fully open distal leaves per plant was measured at the end of the experiment with a U-COR model 3 100 area meter. Laminae were subsequently dried at 80°C for 48 h for dry mass determination. Specific leaf mass (SLM) of the laminae was calculated as the ratio of dry mass over leaf size. In addition, petiole length on the fully open distal leaf was measured at the end of the UV-B treatment period. Chlorophyll fluorescence measurements were conducted at room temperature on fully open distal leaflets of five replicate samples for each population under each UV -B level (Greer, 1 995b). Immediately after collection, samples were stored on moist filter paper in cuvettes sealed with a rubber stopper and dark-adapted for 30 min. Measurements were carried out using a P AM fluorometer (Pam 1 0 1 ; Walz, Effeltrich, Germany). Initial fluorescence in darkness Fo was measured using a weak measuring light. Maximum chlorophyll fluorescence in darkness Fm and the fluorescence ratio Fv/F m (Fy = Fm - Fo), indicating intrinsic efficiency of photosystem I I (PS II) were determined using a flash intensity of 10 000 Jlmol m-z S- I (KL 1 500; Walz). Methanol-extractable UV -absorbing compounds were estimated following standard procedures (Mirecki and Teramura, 1 984). Four fully open distal trifoliate laminae were oven-dried and 1 5-20 mg of ground material was weighed and extracted in 1 .4 ml MeOH:HzO:HCI (79 :20: 1 ) . Extraction in darkness for 24 h under occasional shaking was followed by centrifugation and spectrophotometric analysis of the supernatant in 3 ml quartz cuvettes. Absorbance readings at 300 nm were calculated on the basis of leaf 2. peA of UV-B responses in 26 white clover populations 29 dry weight. This procedure was repeated on five replicates for each T. rep ens population under each UV -B treatment. Epidermal studies Samples were collected using a modified leaf replica technique (Beerling and Chaloner, 1 992; Hilu and Randall , 1 984). Using a fine brush, a solution of 4% formvar in chloroform (w/v) was applied to the middle leaflet of fully open distal T. repens leaf laminae. Celluloid tape was used to remove the dried solution, representing a cast replica of the epidermal surface. After placement on a glass slide, the replica was viewed through a Nikon Diaphot inverted microscope with a phase contrast Nomarski filter attachment. The image was then captured with a Sony black and white camera, connected to a Zeiss Kontron-VIDAS image analysis system, giving a resolution of 760 x 568 pixels. Digital image processing used VIDAS and MA TLAB software. Two frames were obtained from each of 1 0 l eaf samples per population under each UV­ B leveL Stomata were identified manuall y in each frame and the number of stomata was determined in order to calculate stomatal density. The remaining epidermal cells in the frame were segmented by binary thresholding, fol lowed by dilation until cells touched. After removal of stomata, the pixel area of epidermal cell s inside a guard frame of defined area was detennined. This value was divided by the number of epidennal cells examined to obtain mean epidermal cell size and the result was converted from pixels to J..lm2• Stomatal density was calculated as the number of stomata per mm2. Statistical analysis Univariate analyses and regression studies were conducted with the General Analysis of Variance and Regression Analysis procedures in GENSTAT (Genstat, 1 993). Principal components analysis (PCA) (Kershaw and Looney, 1 985) was used as an exploratory tool in order to indicate an overall measure of UV-B sensitivity for the 26 T. repens populations on the basis of the simultaneous analysis of 1 7 responses to UV-B. The analysis was performed with the SAS statistical package (SAS, 1 996). PCA was chosen as this multivariate approach is considered to be particularly suitable for datasets with possible correlation among attributes (Manly, 1 994). Correlations among variables are 2. peA of UV-B responses in 26 white clover populations 30 removed by PCA and converted into sets of independent attributes. Thus, information from the individual responses is brought together to create new, combined variables representing uncorrelated vectors in 17 dimensions. The resulting principal components (PCs) are ranked in the PCA according to their importance in explaining the variation in the data set. Data were log-transformed and the differences between means (representing measures of DV -B responsiveness) were subjected to PCA. Following PCA, these differences were regressed against the PC scores. The resulting probability values and correlation coefficients facilitated distinction of key traits characterising the DV -B response in each dimension. T. repens populations placed close to either end of a particular axis could then be related to responses positively or negatively correlating with a particular PC axis. A further aim was to identify constitutive plant attributes (i.e. measured under no DV-B supplementation) l inked to the principal DV -B sensitivity measure (PC 1 ). To this purpose, PC 1 scores were regressed against the constitutive means of the attributes. Finally, this was compared to regressions of PC 1 with the means of the attributes measured under DV -B supplementation. Resu lts Univariate UV-B responses Across populations, supplemental DV -B decreased plant dry mass in leaves, stolons and roots (Table 2-2). A slightly smaller relative reduction in root dry mass led to an increase in the rootshoot ratio and a small decrease in the leaf weight ratio. Relative growth rate, number of growing points and petiole length were also reduced under DV­ B. Leaf size and leaf dry mass decreased differently under DV -B, leading to an increase in the specific leaf mass (Table 2-2). Leaf surface characteristics were also affected, including reduced epidermal cell size and increased stomatal density. There was a small increase i n DV -absorbing compound levels and a slight decrease in dark-adapted chlorophyll fluorescence ratios (FvlFm) (Table 2-2). 2. peA of UV-8 responses in 26 white clover populations Table 2-2. UV-B-induced univariate percent changes (calculated from the ratio UV +/UV -) in 17 plant attributes for T. repens, ranked by the degree of change. P < 0.001 = * * *; P < 0.01 = ** ; P < 0.05 = * ; P < 0. 1 0 = +. % Trait description Change P Adaxial stomatal density + 10.7 * * * Specific leaf mass +7 .6 * * * Root:shoot ratio +6.5 * UV -absorbing compounds +3.2 + Ratio adaxial : abaxial stomata + 1 . 1 Chlorophyll fluorescence - 1 .8 * * * Leaf weight ratio -2.3 * * * Number of growing points -9. 1 * * Adaxial epidermal cell size - 1 0.4 * * * Relative growth rate - 1 4.2 * * * Plant root dry mass -20.7 * * * Dry mass of first open leaf -2 1 .3 * * * Petiole length -22.4 * * * Plant stolon dry mass -23 .5 * * * Total plant dry mass -25.8 *** Size of first open leaf -27.0 * * * Plant leaf dry mass -27.9 * * * Plant trait responses in the first principal component 3 1 The first principal component (PC I ) explained 33% of the variance in the dataset, more than double to the next principal component. There were ten correlations, all of which could be found on the positive PC l axis (Fig. 2- 1 and Table 2-3). High scores on PC l were characterised by l arge decreases in all whole plant biomass attributes, fol lowed by young leaf lamina growth traits. This was accompanied by reductions in the number of growing points and adaxial epidermal cell size. In addition, the changes in PC 1 tended to be associated with changes in the concentration of UV -absorbing compounds. I::,. 22 RSR Root OM � 2 r SLM () I Stomatal density Q. I 0 5 Ratio stomata I I::,. 1 8 4 0 I::,.Il25 1 3 0 1 1 1 [' Total OM • • 7 A 2 I::,. 2 6 Leaf OM � 1 5 0 1 7 0 20 :��on OM I 0 12 Ro� OM ---- 1�--.-- 1 - 6 -7-;;----� 21 2 0 ��� size Cel l size OM 1st leaf I (UVabs) 2301 14 1 �PC--1 �?�� A 3 LWR Cell size (Petiole) 0 241 2 l A 1 Fig. 2-1. PCA of 1 7 DV -B responses in 26 T. repens populations with PC I scores plotted against PC2 scores. Associated plant responses are grouped at the end of each principal component. Abbreviations for plant responses: DM = dry mass; DM 1 sI leaf = DM of fully open distal leaf; LWR = leaf weight ratio; NGP = number of growing points; PC = principal component; Petiole = petiole length; Ratio stomata = ratio adaxial : abaxial stomata; RGR = relative growth rate; RSR = root:shoot ratio; SLM = specific l eaf mass; DVabs = DV -absorbing compounds. Dark-adapted photochemical efficiency of photosystem I I (FvlFm) remained the only response not associated with either principal component. Numbers identifying the i ndividual populations are given in Table 2- 1 . Symbols for populations represent Asian (_), tropical (e), Scandinavian (.), MediterraneanlNew Zealand ecotypes (0), common cultivars (0) and populations with particular features of pigmentation (.6). 2. peA (�f UV-B responses in 26 white clover populations Table 2-3. Correlation coefficients and probabilities for plant attributes l inked to the first two principal components, ranked by probabilities. P < 0.001 = *** ; P < 0.01 = ** ; P < 0.05 = * ; P < 0. 10 = +. Correlation Correlation with PCl with PC2 Trait description r P r P Total plant dry mass 0.952 * * * 0. 1 27 Plant leaf dry mass 0.925 * * * -0.07 1 Relative growth rate 0.879 * ** 0.096 Plant st010n dry mass 0.778 * * * 0.288 Plant root dry mass 0.7 1 7 *** 0.575 * * Leaf size 0.69 1 * * * -0. 1 74 Number of growing points 0.626 * * -0.0 1 5 Adaxial epidermal cell size 0.6 10 * * -0.4 1 6 * Dry mass first open leaf 0.590 * * 0. 1 2 1 UV -absorbing compounds 0.332 + 0. 1 89 Rootshoot ratio -0.046 0.798 * * * Specific leaf mass -0. 1 25 0.622 * * Leaf weight ratio 0. 1 63 -0.620 * * Adaxial stomatal density -0.294 0.48 1 * Ratio adaxial : abaxial stomata -0. 1 79 0.458 * Petiole length 0.320 -0.338 + Chlorophyll fluorescence 0.09 1 0.220 Population responses in the first principal component 33 T. rep ens populations sensitive to UV -B (displaying stronger UV -B-induced responses in PC 1 ) were located towards the positive end of the first principal axis while the negative end represented less responsive populations (Fig. 2- 1 ) . The most pronounced separation along this sensitivity vector was between Asian ecotypes, which could consistently be found in the negative, UV -B-tolerant part of the vector, and several common cultivars towards the UV-B-sensitive, positive end of the axis, with 'Kopu' and 'Huia' at the top. Within the cultivars, the populations 'Prestige' and 'Sustain ' were placed towards the central and more negative region of PC 1 (Fig. 2- 1 ). Low latitude ecotypes tended to distribute around the central area of the axis and there was no c lear demarcation to the high latitude ecotypes from Scandinavia. T. repens populations with 2. peA (�f UV-B responses in 26 white clover populations 34 conspicuous differences in pigmentation were heterogeneous in their UV -B response (Fig. 2- 1 ). Pronounced UV-B tolerance could be observed for 'Luclair', intermediate sensitivity for 'Green' and 'Feathennark', and relatively high UV -B sensitivity for 'Lime'. Plant trait responses in the second principal component The second principal component (PC2) accounted for 1 6% of the variance in the mean differences. The trait combination detelmining the overall UV-B response in PC2 was largely composed of the remaining traits which did not correlate with PC I , namely effects on plant dry matter allocation and epidermal characteristics (Table 2-3). There were eight correlations, most of which could be found towards the positive end of the axis (Fig. 2- 1 ). This end represented higher UV-B sensitivity mainly due to combined effects on below-ground parameters, SLM and stomatal distribution. The most pronounced feature of UV -B tolerance (high negative PC2 scores) was preferential biomass allocation towards growth of roots, rather than shoots (higher RSR), concomitant with a decrease in relative leaf mass (lower L WR). Consequently, UV-B sensitivity in PC2 (high positive PC2 scores) was characterised by no preferential pattern of biomass allocation towards root growth and accompanied by larger absolute decreases in root biomass (Fig. 2- 1 ). Other allocation changes linked to PC2 were those that occurred within the T. repens leaf. High positive PC2 scores correlated with a relative shift of stomata towards the abaxial leaf surface while high scores at the opposite end of the axis were l inked to increases in specific leaf mass. This was also associated with increases in adaxial stomatal density, paired with UV -B-induced reductions in adaxial epidermal cell size. Finally, there was a marginal relationship of PC2 with changes in petiole length. Change in intrinsic chlorophyll fluorescence yield was the only plant attribute that showed no association with either multivariate sensitivity axi s (Table 2-3). Population responses in the second principal component A characteristic feature in the second PCA dimension was that several populations which were neutral in PC 1 appeared towards the positive (UV -B-sensitive) end of the axis in PC2, including ecotypes from Italy and New Zealand and the breeding lines 'Feathermark' and 'Green' (Fig. 2- 1 ). Populations less responsive to UV-B based on the 2. peA of UV-B responses in 26 white clover popuiations 35 PC2 trait combination included 'Sustain' , 'Sarikamis', Tienshan' and the two Swedish ecotypes. The focus of this paper i s on the first two dimensions as no new plant attributes were linked to the subsequent lower-ranked PC3 and PC4 dimensions, showing fewer and lower correlations for changes in plant attributes and no mechanisms of UV-B sensitivity for populations which displayed UV-B-tolerance in the earlier dimensions (data not shown). Moreover, intrinsic yield of chlorophyll fluorescence (Fv/F m) continued to make no contribution to the characterisation of the UV -B response with a low cumulative squared correlation coefficient (R2) of 0. 10, even after four principal components. Correlation analyses A further aim was to examine possible l inks between plant characteristics of the T. repens populations and PCI , the dimension identified by PCA as the global (overall) measure of their UV -B-responsiveness. Regression analyses revealed that UV-B tolerance (negative PC l score) was related to slow intrinsic growth rate (Fig. 2-2a) and small constitutive epidermal cell size (Fig. 2-2b). Rather than linked to constitutive UV­ absorbing compound levels (Fig. 2-2c), UV-B tolerance was correlated to accumulation of these compounds under UV-B (Fig. 2-2d). Additional correlations related to the peA results revealed that fast-growing T. repens populations in general had constitutively larger adaxial epidermal cells (r = 0.473, P < 0.05) and that in turn large-celled populations had l arger leaves (r = 0.660, P < 0.00 1 ). Furthermore, the most productive populations displayed lower relative root proportions (low RSR, r = -0.483, P < 0.05), concomitant with higher relative leaf production (high LWR, r = 0.583, P < 0.01 ). y =: 52.2D7x � 4.0646 r = 0.3560 <3 1.5 p := 0.074 !!:. � � 0.5 � > � �.5 c � ID :> ·1.5 ::l -2.5 0.06 0.07 y = ·1.7056x • 0.0648 r = ..(),1141 U 1 .5 p = 0.579 !!:. f 0 � 0.5 . 1 � > � ..().S c � ID :> ·1 .5 ::l -2.5 .Q.3 .Q.2 . 21 . 26 . 3 . 19 . 22 0.08 0.09 RGR (UV-) .20 . 21 . 26 . 3 . 19 5 . 16 25- t12 . 18 24',1115 2 .,4 23 14 . 7 . 9 . 13 . 10 . 6 . 8 .22 .Q.l 0 UV-absorbing compounds (UV-) . 20 (a) 0.10 (C) . 17 0.1 _ r =: O.4484 � y � 3.2366x - 22.457 o 1.5 P = 0.022 !!:. f 0 Ii: 0.5 � > � .Q.5 c � � - 1.5 t -2.5 6.5 6.7 � r � .Q.4185 f , . . M", . • _ o 1 .5 P = O.033 !!:. f 0 � 0.5 � > :� -{).5 !j -a.3 -a.2 . 21 (b) . 20 . 3 . 26 . 19 . 8 . 22 6.9 7.1 7.3 Cell size (UV-) . 21 (d) . 20 26 ·.3 . 19 . 17 23 . 13 e 9 . 6 . 10 . 8 .22 .Q.1 0 0.1 UV-absorbing compounds (UV+) Fig. 2-2. Relationship between overall UV-B sensitivity (PC I scores) and plant attributes (log-transformed means from PCA) of 26 T. repens populations. Numbers identifying the individual populations are given in Table 2- 1 . PC I relationship to (a) constitutive (i.e. measured under no UV-B supplementation) relative growth rates; (b) constitutive adaxial epidermal cell sizes; (c) constitutive levels of UV-absorbing compounds; and (d) UV-B-induced levels of UV-absorbing compounds. w 0\ 2. peA of UV-B responses in 26 white clover popu/afions 37 Discussion This study provides for the first time an overall summary of various population responses to UV -B for a plant species based on multivariate analysis of numerous plant attributes. PCA proved useful in exploring the main response patterns among the 26 populations in the 1 7-dimensional structure. In univariate analyses, T. repens was significantly affected by UV -B in most traits measured here (Table 2-2). Whole plant growth characteristics and young leaf growth displayed the strongest decline, while allocation parameters and physiological responses showed more subtle responses. Plant trait responses in the first principal component PCl , the most important dimension in describing the overall UV -B response, contained all whole plant growth responses, indicating the integrating nature of these traits as an overall summary for the various individual effects caused by UV -B (Fig. 2- 1 ). Reduced total plant growth is considered the end point of DV -B damage (Stapleton et al., 1 997) and the present findings are consistent with other reports showing that growth reduction is due to simultaneous decreases in aerial as well as in below-ground biomass (Krizek et al., 1 998 ; Krizek et al. , 1 997b). This association also reflects the usual relationship between leaf and root growth (Grime et al. , 1 997). Decreases in whole plant growth were associated with reductions in attributes l inked to cell division (number of growing points) as well as cell expansion (adaxial epidermal cell size), pointing at both growth-conferring mechanisms underlying the PC 1 response. The overall decline in leaf size by 27% (Table 2-2) was accompanied by reduced cell expansion (- 1 0%, Table 2-2) (Fig. 2- 1 ) . Thus the remaining portion of leaf size reduction was due to a UV -B-induced decrease in cell division, with epidermal and stomatal cell numbers per leaf (cell density multiplied by leaf size) reduced by 1 7% . This could suggest a larger contribution o f cell division than o f cell expansion for the UV -B-induced reduction of leaf size. Similar conclusions have been drawn in pea, with UV-B-generated reductions of 1 3% for epidermal cell size and of 26% for epidermal cell number contributing to a 39% decrease in leaf size (Nogues et al., 1 998) . Lower UV -B-induced abaxial cell density, indicating larger cell size, was observed for Vicia jaba (Visser et al. , 1 997a) and stimulation of cell division as a consequence of UV-B 2. peA of UV-B responses in 26 white clover populations 38 i rradiation was found in Petunia hybrida (Staxen and Bornman, 1 994). In conjunction with the current findings this indicates i nterspecific specificity in the meristematic UV­ B response. The marginal association of PC 1 with changes in the concentration of UV -absorbing compounds (Table 2-3) suggests that biochemical deficiencies (no increase in UV­ absorbing compound levels) could also bring about high PC 1 scores (i .e. pronounced growth reductions). This is consistent with other reports of a relationship between UV-B sensitivity and changes in UV -absorbing compound accumulation (Day and Demchik, 1 996; Murali and Teramura, 1 986c). UV-absorbing compounds usually include UV-B­ screening pigments such as flavonoids and hydroxycinnamic acids (Lavola et al., 1 997) and could therefore perform a variety of UV -B-protective functions, including screening and antioxidant activities. Population responses in the first principal component PC 1 scores for the Asian ecotypes were consistently higher compared to most other populations used in this study, including all T. repens cultivars and temperate ecotypes from other geographical regions (Fig. 2- 1 ) . Compared to the latter, climatic conditions for the Asian ecotypes are much more continental and the growing conditions generally unproductive (Caradus and Forde, 1 996). Their usually small-leafed, low productive phenotype is typical of T. repens populations originating in harsh environments (Williams, 1 987a). Several of the Asian ecotypes originate from habitats exposed to one or more stress form, including low fertility, low temperature, high altitude and precipitation levels which would be considered marginal for T. repens growth (Jones and Lowe, 1 993). Predisposition to other forms of stress contributing towards adaptation to enhanced UV-B has been suggested (AI-Oudat et aI., 1 998) and would be supported by ecological theory on stress tolerance (Diaz et aL. , 1 999; Grime, 1 979; Poorter and Gamier, 1 999). According to such models of differential plant strategies, populations from stress-prone habitats ( 'stress tolerators') would more likely experience less severe relative biomass reduction under stress than 'actively foraging' populations from more productive backgrounds (,competitors'), This notion finds additional support from the correlation studies showing lower relative investment into leaf biomass (high LWR) and higher RSR in less productive populations, which would again characterise 2. peA of UV-B responses in 26 white cloverpopulations 39 them as 'stress-tolerators'. In other studies the Asian ecotypes also were distinct from T. repens cultivars such as 'Huia' on the basis of lower productivity, smaller leaves and lower cyanogenesis (Caradus and Forde, 1 996). A link to ecological specialisation is fUlther supported by the observation that many of the more productive T. repens cultivars (Table 2- 1 b) can be found towards the DV -B­ sensitive region of PC 1 . The pronounced UV -B sensitivity in PC 1 for 'Huia' is less expected as this general-purpose cultivar has displayed a wide range of adaptability to various environments (Caradus and Woodfield, 1 997). Most of the cultivars however are bred for high productivity and it is possible that such specialisation places them at a disadvantage for adaptation to DV -B. The positioning of 'Prestige' and 'Sustain ' towards the central and slightly negative, more tolerant part of the sensitivity vector distinguished them from the other cultivars (Fig. 2- 1 ) . 'Prestige' has displayed relative tolerance to other forms of stress (e.g. drought) and grows in regions in northern New Zealand with higher background UV -B (Table 2- 1 b). Relative DV -B adaptability for 'Sustain' might also be related to plasticity resulting from a diverse genetic background including New Zealand, US and Mediterranean germplasm. From the distribution of Scandinavian and tropical ecotypes i t appears that there is no general distinction in DV -B-sensitivity based on difference in latitudinal origin (Fig. 2- 1 ). DV -B susceptibility in plants from higher latitudes has been reported (Barnes et al., 1 987). Other studies, however, point at lesser importance for latitudinal origin in explaining differential DV -B adaptation (Robakowski and Laitat, 1 999). There was also no relationship between latitudinal origin and UV-B sensitivity in 1 88 rice cultivars (Dai et al., 1 994) . The difference in DV-B sensitivity between Scandinavian and Asian ecotypes could suggest that latitudinal differences might still play a role among populations sharing similar features of plant and habitat productivity. Furthermore, the Asian T. repens populations are considered to be particularly adapted to extreme changes in environmental conditions (Forde, 1 989) and the possibility cannot be excluded that adaptation for the Swedish ecotypes would have required more extended periods of UV -B exposure. Heterogeneity in the UV -B response was apparent for T. repens populations displaying prominent differences in leaf pigmentation (Fig. 2- 1 ) . Lower constitutive chlorophyll 2. peA of UV-B responses in 26 white clover populations 40 levels in the breeding line 'Lime' did not confer additional UV -B sensitivity compared to the mother population 'Huia'. 'Luclair' and 'Feathermark' are known to contain higher anthocyanin levels (Table 2- 1 b). In particular, high anthocyanin levels have been reported in the leaf mark region for 'Luclair' (Lenoble and Papineau, 1 970) and other reports suggest a relationship between higher anthocyanin accumulation and UV-B tolerance (Burger and Edwards, 1 996; Stapleton and Walbot, 1 994) . A recent review points at several possible mechanisms of protective anthocyanin function, including antioxidant activities (Chalker-Scott, 1 999). Moreover, it could be argued that UV -B i s filtered in the clear portions of the 'Luclair' lamina, which in addition to anthocyanins contains UV -absorbing compounds, while photosynthetically active radiation would be transmitted at least to some degree in these regions due to the lack of chlorophyll. This would effectively provide a UV -B filter for canopy leaves growing underneath such translucent regions. Furthermore, cell shape can be expected to be altered to a large degree in extended leaf mark regions such as in 'Luclair' (Williams, 1 987b), which could also lead to differences in dispersion of DV -B within the leaf. Taken together, the first principal component suggests a UV -B-induced separation among the T. rep ens populations that can largely be explained by ecological specialisation. In particular, morphological stress response and deficiencies i n biochemical adaptation appear more pronounced in UV-B-sensitive 'competitor' - rather than UV -B-resistant 'stress-tolerator' populations. Plant trait responses in the second principal component Rather than representing decreases in morphological and growth attributes to UV -B (as in PC 1 ) , PC2 indicated several UV-B-induced morphogenetic shifts between plant characteristics. Thus this vector appears to constitute a more subtle tier of UV-B responsiveness. A predominant feature in PC2 was association with changes in aerial and below-ground biomass allocation (Table 2-3). A relative percentage increase of root mass leading to a higher root:shoot ratio under UV -B could suggest an adaptive mechanism redistributing carbon towards plant parts not exposed to UV -B . While some studies report decreases or no changes in the rootshoot ratio under UV -B (Gwynn-Jones and Johanson, 1 996; 2. peA of UV-B responses in 26 white clover populations 4 1 Sullivan and Teramura, 1 988), other investigators also observed increases in relative biomass allocation towards roots in response to DV -B (Barnes et al., 1 993; Tosserams et al., 1 996). DV-B sensitivity in PC2 was also characterised by smaller increases in leaf dry mass production, relative to leaf size (specific leaf mass, SLM) (Table 2-3) . Increased SLM can be due to either increased leaf density or thicker leaves and has been interpreted as an adaptive feature in response to DV -B , changing absorption properties of the leaf (Britz and Adamse, 1 994). Further responses associated with high PC2 scores included changes in epidermal characteristics (Table 2-3). Increases in stomatal density were most l ikely due to the simultaneously occurring decreases in epidermal cell size, thus leading to a closer, more compact arrangement of the stomata. Stomatal density has displayed a diversity of responses to DV-B, including decreases (Dai et al., 1 995 ; Tevini et al. , 1 983b), no change (Staxen and Bornman, 1 994; Visser et al. , I 997a) and increases (Stewart and Hoddinott, 1 993). A relative shift of stomata towards the abaxial leaf surface at high positive PC2 scores could indicate a morphogenetic response that differentiates stomata away from direct DV-B exposure. Such a direct effect of DV-B on stomata is suggested in rice, showing greater loss in stomatal density from the adaxial, rather than abaxial surface (Dai et al. , 1 995). However, decreased stomatal numbers due to DV -B have been observed for the abaxial, but not the adaxial , leaf surface in Vicia faba (Visser et al., 1 997 a). It was of additional interest to note that intrinsic efficiency of PS I I (FJF m) showed no contribution to the overall DV -B responses represented by the main principal components. This is supported by recently reviewed evidence questioning the role of intrinsic PS II efficiency as a key response of DV -B sensitivity (Allen et al . , 1 998). Population responses in the second principal component Several populations displaying moderate DV -B responsiveness in PC I appeared to be affected in thi s second dimension of DV -B sensitivity, including temperate ecotypes from Italy and New Zealand (Fig. 2- 1 ). While not selected for yield, these populations generally derive from relatively productive environments, which could help explain generally higher principal component scores in both dimensions compared to the 2. peA of UV-B responses in 26 white clover populations 42 continental Asian ecotypes originating from lower productive habitats. Furthermore, while unselected, the two New Zealand ecotypes have their genetic origin in the cultivar 'Huia' (Table 2- 1 a), which already proved to be UV -B-sensitive in PC 1 . The PC2 trait combination as a measure of UV-B responsiveness was also found for two other populations of intermediate sensitivity in PC l , namely the pigmented l ines 'Feathermark' and 'Green', both also of 'Huia' origin (Table 2- 1 b). It thus appears that cultivars such as 'Huia' which experience long-term ecotypic adaptation or selection for particular pigmentation (higher chlorophyll or anthocyanin levels) can show less sensitivity in direct UV -B-related growth reductions (PC 1 ) but are still UV -B­ responsive in several aspects of biomass allocation (PC2). Populations on the opposite end of PC2 included 'Sarikamis ' and 'Tienshan', which also distributed in the negative, more tolerant part of PC 1 (Fig. 2- 1 ) . These populations derive from high altitude habitats, with adaptation to a number of stress forms including low fertility, low temperatures and low precipitation (Table 2- I a) . Similar to PCI , 'Sustain' and - to a lesser degree - 'Prestige' again performed best among the cultivars. In addition, the two Swedish ecotypes also appeared in this spectmm, suggesting lesser overall importance of the PC2 trait combination for the DV -response in these populations. PC2 thus offers possible explanations for the nature of UV -B sensitivity in populations of intermediate overall (PC l) UV -B responsiveness, largely by affecting allocation attributes between and within plant organs. Moreover, it further points at particularly DV-B-tolerant populations, which not only do relatively well on the main separation axis (PC I ), but also rank highly in aspects of the next important dimension CSarikamis' and 'Tienshan'). Relationship between PC] and plant features From the regression studies it appears that T. repens populations which are fast-growing (Fig. 2-2a) as well as those with larger adaxial epidermal cells (Fig. 2-2b) are more sensitive in their overall UV -B response. In addition, while constitutive UV -absorbing compound levels showed no relationship to PC l (Fig. 2-2c), populations with the capacity to accumulate higher levels of these compounds under UV-B were less 2. peA of UV-B responses in 26 white clover populations 43 sensitive in the first principal dimension (Fig. 2-2d). Thus in addition to the results from PC 1 showing a relationship to UV -B responsiveness of relative changes (i .e. treatment values compared to control values) in these plant attributes, the overall reaction to UV­ B also appears to be l inked to absolute values in these features. Absolute, rather than relative accumulation of UV-absorbing compounds has also been l inked to UV-B protection, for instance in rice (Sato and Kumagai, 1 997). The positive relationship between UV -B sensitivity and high constitutive growth rate is further supported by ecological theory of stress tolerance, suggesting higher stress susceptibility for populations investing in high dry matter production (Diaz et al. , 1 999; Grime, 1 979). The link between larger constitutive epidermal cell size and UV -B-sensitivity could also be related to plant productivity. The correlation studies show that the fast-growing populations generally had larger cells and that large-celled populations also had larger leaves . Large epidermal cells could therefore also be seen as providing a morphological basis for enhanced resource acquisition and - as outlined above - this is l inked to UV-B sensitivity. From a plant strategy perspective (Grime, 1 979; Poorter and Gamier, 1 999) it is also possible that populations developing smaller epidermal cells in response to environmental stress could have maintained - or even increased in compensation - the capacity to produce compounds of relevance for stress protection. Higher cellular concentrations of epidermis-specific pigments such as flavonoids would contribute to UV -B protection and other experiments demonstrated that smaller-celled T. repens populations described here (Fig. 2-2b) contained the highest flavonoid levels (Chapter 6). Finally, there could also be differences in leaf optical properties depending on epidermal cell size. Differences in constitutive epidermal cell size and shape have been related to differences in light focus sing and scattering (Vogelmann et al., 1 996). Microscopy studies conducted on 'Huia' (sensitive in PCI ) and 'Tienshan' (tolerant in PC 1 ) revealed that the smaller adaxial epidermis cel ls in the latter popUlation had a more convex appearance (see Chapter 8 , Fig. 8-1). 2. peA of UV-B responses in 26 white clover populatiolls 44 Conclusions T. repens displayed a number of morphological and physiological responses to UV-B under controlled environmental conditions. Reduced cell division and cell expansion lead to smaller leaves, fewer stolons and ultimately to decreases in whole plant growth. PCA and other multivariate assessments could provide useful tools for future examinations of UV -B-induced intraspecific plant responses. The relationship of UV-B sensitivity to habitat productivity and in turn to aspects of plant productivity and physiology (relative growth rate, epidermal cell size, accumulation of UV -absorbing compounds) suggests the existence of plant functional types within T. repens. Summary representation of the morpho-physiological responses in PCA indicates higher sensitivity to UV -B for productive populations, mainly reflected in decreases in plant growth. Thus trade-offs appear to exist between morphological and physiological plant features with strongest UV -B-induced relative biomass decreases in populations that invest most metabolic energy in gaining that biomass. 3. Modifications of white clover growth responses to UV-8 45 Chapter 3. Growth responses to u ltraviolet-B radiation in Trifolium repens populations depend on water availability, productivity, and duration of stress White clover growth under controlled environmental conditions This chapter has been submitted to New Phytologist. 3. Modifications (){white clover growth responses to UV-B 46 Abstract This study examined how drought stress might modify the response of white clover to ultraviolet-B radiation (UV -B). Nine white clover populations from a variety of different habitats were used to investigate intraspecific differences in the stress response. Several harvests were used to examine the effect of time. Plants were grown under controlled environmental conditions for 1 2 weeks with and without supplemental application of 1 3. 3 kJ m-2 d- I UV-B and drought was applied during the last four weeks of the experiment. At the end of the experimental period, white clover growth was decreased by about 20% in response to UV -B under well-watered conditions. In drought conditions however, growth reductions due to UV -B were less than half as pronounced, indicating a mitigating effect of moisture deficit for UV -B sensitivity. The extent of the UV -B effect differed among white clover populations under well-watered conditions, with greatest UV -B tolerance observed in slow-growing ecotypes characterised by low original habitat productivity. White clover cultivars selected and bred for yield and agricultural perfomlance displayed higher sensitivity to UV -B. However, drought appeared to increase UV -B tolerance for these cultivars as no pronounced population differences in UV -B responsiveness could be detected under l imited water supply. Analysis of the time course of UV -B responsiveness revealed that UV -B sensitivity decreased both with increasing exposure to drought and with longer duration of UV -B irradiation. In conclusion, UV -B effects in white clover can be modified by drought and the extent of this depends on the duration of stress. Furthermore, slower growing populations from environments exposed to multiple stresses are likely to display higher UV -B tolerance. Introduction Stratospheric ozone loss leads to increased summertime ultraviolet-B radiation (UV-B , 290-3 1 5 nm) (McKenzie et aI., 1 999). This highlights the need for examinations of UV­ B responses in conjunction with other forms of stress that can simultaneously affect plant growth during high UV -B episodes. White clover plants in New Zealand pasture ecosystems often experience pronounced periods of drought during summertime when UV -B levels are highest (Campbell et al., 1 999). Drought is one of the most l imiting environmental factors to plant productivity, for instance by decreasing plant growth rates, leaf expansion, stem elongation and stomatal conductance (Daie, 1 988; Hsaio, 3. Modifications o.fwhite clover growth responses to UV-B 47 1 973). Similar plant responses have been observed under elevated UV -B radiation and numerous findings also suggest reduced plant productivity as a possible result of elevated UV-B irradiation (Drilias et al., 1 997; Murali and Teramura, 1 986a; Nogues et al., 1 998 ; Sullivan and Teramura, 1 990) . Over the last two decades more than 20 studies have investigated the combined effects of UV -B and drought on plant attributes. While such changes can be additive (Teramura et al. , 1 984b), several reports show modifications of the UV -B effect either by antagonistic or by synergistic interactions. The latter response, represented by enhanced UV -B sensitivity under drought has been observed for growth in the moss Hylocomium splendens (Bjorn et al., 1 997), the shrub Nerium oleander (Dril ias et aI., 1 997) and radish (Tevini et al., 1 983a). Frequently however, ameliorating antagonistic effects of drought for DV -B sensitivity have also been reported, e.g. for soybeans (Murali and Teramura, 1 986a; Sullivan and Teramura, 1 990; Teramura et al., 1 990), Pinus (Man etas et al., 1 997; Petropoulou et al., 1 995) and cowpea (Balakumar et al. , 1 993). Several studies have revealed between-species differences in plant responses to the combined application of DV -B and drought, for example in Brassica (Conner and Zangori, 1 998), Pinus (Petropoulou et al., 1 995), cucumber and radish (Teramura et al., 1 983 ; Tevini et al., 1 983a) , wheat and wild oat (Barnes et al., 1 995) and European heathland species (Bjorn et al., 1 997). There are also i ndications of within-species differences for UV -B responses in combination with drought from comparisons of two soybean cultivars in the field (Murali and Teramura, 1 986b; Teramura et al., 1 990). A number of other reports have shown that within-species differences can influence plant growth responses to DV-B alone (Barnes et al., 1 993 ; B iggs et al., 1 98 1 ; Correia et al., 1 998 ; Dai et al., 1 994; Murali and Teramura, 1 986c ; Pukacki and Modrzynski, 1 998; Teramura et al., 1 990). Furthermore, there are also indications of population differences for drought tolerance in white clover (Woodfield and Caradus, 1 987). It has been suggested that adaptation to one stress can faci litate tolerance to other forms of stress (Grime, 1 979; Manetas et al. , 1 997 ; Petropoulou et aI. , 1 995). Some studies also indicate UV -B tolerance for plants originating from habitats of higher ambient UV­ B levels (Barnes et al., 1 987; Pukacki and Modrzynski, 1 998; Sullivan et ai. , 1 992) as well as for drought -tolerant species (AI-Oudat et al. , 1 998). There is a need for more 3. Modifications ()fwhite clover growth responses to UV-B 48 comprehensive investigations on possible l inkages between UV -B responsiveness and adaptation to environmental conditions other than background UV -B levels . This warrants inclusion of a variety of populations of the same species. Population comparisons strongly facilitate comparability between plant types because they are not affected by species-specific differences. The majority of studies investigating the combined effects of UV -B and drought have been conducted in the field. Such studies were often based on altering water supply due to seasonal differences in precipitation, but only a few specifically included differential water treatments. While field-based research provides natural environmental conditions (e.g. temperature, humidity and herbivory), plant responses to these conditions can potentially mask or interact with effects caused by drought and by UV -B (Allen et al. , 1 999; Teramura et al., 1 990). Investigations of UV -B responses in combination with drought under controlled environmental conditions allow direct investigation of processes inherent to this important stress interaction without other confounding environmental influences. Nine white clover populations from a wide variety of environmental backgrounds were exposed to a combination of 1 2 weeks elevated UV-B and four weeks drought under controlled environmental conditions. The objective was to determine overall UV-B sensitivity for white clover, based on measurements of aboveground biomass accumulation as a measure of plant productivity. It was hypothesised that this response would be modified by water stress and differ among white clover populations. A further aim was to examine possible relationships of UV-B sensitivity with the habitat and productivity of the white clover populations. Additionally, several harvests were used to examine whether UV -B responsiveness would change over time. 3. Modifications ()fwhite clover growth responses to UV-B 49 Materials and Methods Experimental design A 2 x 2 x 9 factorial design included two levels of UV -B radiation, two watering regimes and nine populations of white clover. There were five replicated pots per population, each pot containing 1 0 plants. To minimise effects from differential shading under the UV -B rigs, pots were arranged randomly on six trolleys and the trolleys were rotated daily in the two rooms using a standardised pattern allowing different positions for 24 consecutive days before the starting position was resumed. Plant biomass in droughted plants was harvested the day after rewatering in order to prevent collection of wilted material . Plant cultivation and UV-B irradiation Seeds of nine different white clover populations were obtained from the Margot Forde Germplasm Centre in Palmerston North, New Zealand. These populations were chosen to provide a range of plant types from differential environmental backgrounds (Table 3 - 1 ) . The study included three slow-growing ecotypes collected in the wild from habitats exposed to multiple stresses CSarikamis', 'Tienshan' and 'Haggas'), four cultivars multiplied i n breeding programmes for agricultural yield CHuia', 'Kopu', 'Haifa' and 'Prestige'), and two breeding lines (un-multiplied selections). The latter included the high-yielding population 'Octoploid' and the less productive 'Syrian', selected for tolerance to leaf rust and drought (Table 3- 1 ). Following scarification, seeds were placed in petri dishes on moist filter paper and incubated at 25°C for 3-5 days. Table 3-1. Description of Trifolium repens ecotypes (populations collected in the wild), cultivars (populations multiplied for high agronomic value) and breeding l ines (un-multiplied selections). Accession numbers are from the Margot Forde Germplasm Centre in Palmerston North, New Zealand. References for ecotypes: 'Haggas' (Nordiska Genbanken, 1 993), 'Sarikamis' (Caradus and Forde, 1 996), 'Tienshan' (Forde, 1 987). References for cultivars from Caradus ( 1986) and Caradus and Woodfield ( 1 997), and for breeding l ines : 'Syrian' (Caradus et al., 1 990a), 'Octoploid' (Woodfield and Cousins, 1 999). Accession Type Name Country Description C l 5 1 30 Ecotype Haggas Sweden Latitude 64.2°N, 500m altitude, 700mm annual rainfall , south-west clay slope, abandoned cultivation. C8734 Ecotype Sarikamis Turkey Latitude 40.2°N, 2300m altitude, 580mm annual rainfall, basaltic, extended snow cover and clouds, very cold. C8749 Ecotype Tienshan China Latitude 43.0oN, 2000m altitude, 600mm annual rainfall, hay meadow, heavy winter snow, very cold. C 1 3 1 73 Cultivar Haifa Israeli Bred from Israel germplasm originating near 33°N latitude, high yielding, heat- and salt-tolerant, more Australia sensitive to frost and drought, decreased persistence under grazing, relatively poor disease resistance, good performance in lowland conditions. C7544 Cultivar Huia New Long-established New Zealand general purpose cultivar, selected for yield and persistence to grazing, high Zealand phenotypic plasticity, high amino acid content, sensitive to P-deficiency, water stress and frost, relatively poor d isease resistance, best suited for lowland conditions. C l 3 1 76 Cultivar Kopu New Includes germplasm from ladino cultivars, cv. 'Pitau' and a Spanish strain, bred for lowland farming, Zealand nematode resistance and yield, more productive and leaves more upright habit than 'Huia', recommended for grazing, large roots but no persistence in dry hill country, frost sensitive. C l 1 678 Cultivar Prestige New Bred from ecotypes in Northland (northern New Zealand), improved heat and drought tolerance, grazing- Zealand tolerant, suited for hil l country climate and dry sunny slopes, relatively good disease resistance. C l879 1 Breeding Octoploid New Higher ploidy level (8x), parentage from cultivar 'Dusi ' with tall, large-leafed productive phenotype, large line Zealand root system with persistence in soils with drought incidence, more frost-sensitive than 'Huia'. C 1 0575 Breeding Syrian Syria Selected from Syrian germplasm for disease tolerance (leaf rust), improved persistence in drought line conditions under grazing. 3. Modifications (�fwhite clover growth responses to UV-B 5 1 The emerging seedlings were subsequently transplanted into square gallon pots contain ing a sterilised sand/mineral mix and transferred into large growth chambers at the National Climate Laboratory in Palmerston North. The pots were supplied with an automatic drip irrigation system with half-strength Hoagland's nutrient solution at regular intervals during the 1 2 h day and 1 2 h night cycle and flushed with water weekly. The experiment was conducted for 1 2 weeks in growth rooms of identical design and under continuous monitoring of environmental conditions. In other studies these rooms demonstrated excellent environmental control with insignificant differences between rooms (Warrington et al., 1 999; Warrington and Kanemasu, 1 983). The average photosynthetic photon flux (PPF) during the light period was 425 /lmol m-2 S- l , supplied by a water-screened array of four 1 kW Sylvania 'metal-arc' high pressure discharge lamps, together with four 1 kW Philips tungsten iodide lamps. PPF was measured using a Ll-COR 1 85 meter with a LI- 1 90S quantum sensor. Temperatures were 22°C during the day and 1 8°C at night, with the respective mean CO2 levels at 367ppm and 383ppm. Relative humidity was 70-80%. One hour after onset of the l ight period until one hour before darkness, DV -B radiation was supplied by Philips TL 40W/12 RS UV fluorescent tubes, enclosed in cellulose diacetate filters. The UV-B irradiance levels were gradually increased during the first three weeks of UV -B supplementation to faci litate acclimation of the plants to UV -B . After this, plant-weighted (Caldwell , 197 1 ) biologicall y effective UV-B (UV-BBE) levels (normalised to 300 nm) were 1 3 .3 kJ m-2 d- l , mimicking 25% summer ozone depletion above Palmerston North, New Zealand. The tubes were suspended 500 mm above plant level. A Solar Light Company model 50l A UV biometer sensor measured the UV output which was controlled by a feedback system to maintain the constant, pre­ set DV -B irradiation doses in response to tube ageing and cellulose diacetate filter degradation. These filters were changed when the feedback system output approached maximum voltage. A Bentham DM 1 50 DV spectroradiometer was used to calibrate the sensor at regular intervals against the generalised plant action spectrum (Caldwell, 1 97 1 ). 3. Modifications (if white clover growth responses to UV-B 52 Harvesting of plant material and drought application During the 1 2-week experimental period, plants were clipped at four harvests near pot height to collect plant biomass. This also minimised effects of shading and prevented the plants from exceeding pot size. The first harvest was conducted four weeks after establishment. The second harvest was performed three weeks later, followed by the onset of the drought period one week thereafter. A third harvest was conducted midway between this and the fourth harvest which concluded the drought treatment after four weeks. The third harvest included growth in non-drought and developing drought conditions and thus represented a less severe water stress treatment compared to the final harvest. The following treatments were applied during the four-week drought period: ( 1 ) UV- W+ No UV-B supplementation and well-watered conditions. (2) UV+ W+ UV-B supplementation and well-watered conditions. (3) UV- W- No UV-B supplementation and water l imitation. (4) UV+ W- UV-B supplementation and water l imitation. Drought treatment was applied by withholding water and monitored gravimetrically using techniques employed in other studies of water stress in white clover (Barbour, 1 996; Barbour et al. , 1 996). Plants in droughted pots were rehydrated when half or more of their petioles showed signs of incipient wilting. At this point each pot was weighed and watered to 4% soil moisture content, with no loss from drainage after rewatering. In the well-watered treatment, the normal watering and nutrient solution regime was continued. Aerial biomass from each harvest was dried for 48 h at 80°C and weighed. The accumulated biomass from these harvests was used to calculate the rate of dry matter production. Sensitivity to UV -B was expressed as the percentage change calculated from the comparison of dry matter production between UV-B-treated and control plants. 3. Modifications of white clover growth responses to UV-B 53 Statistical analysis The General Analysis of Variance procedure in GENST A T (Genstat, 1 993) was used as the principal statistical method for analysis of main and interaction effects on aerial biomass accumulation transformed to a logarithmic scale. Repeated measurements of biomass collected in four harvests allowed testing for the effect of time on main and interaction terms. This was examined with the method of residual maximum likelihood (REML) in GENSTAT (Genstat, 1 993). The Arepmeasures procedure (ANOVA for repeated measurements) was used to characterise the type of correlation between observations of the four harvests (Genstat, 1 993). The resulting correlation matrix revealed a typical auto-regressive stmcture (ARI ) for the covariance model and the REML analysis was specified accordingly. For conciseness, the overall REML analysis incorporating all treatments is presented here. Individual REML analyses distinguishing between droughted and well-watered samples gave similar results. The relationship between UV -B sensitivity and constitutive productivity (i.e. biomass production without UV -B supplementation) of the white clover populations was examined with the GENSTA T Regression Analysis procedure. Results Influence of drought on UV-B sensitivity At the end of the experiment (final harvest, Fig. 3- 1 a), UV -B significantly decreased biomass accumulation by about 20% across white clover populations in well-watered conditions, compared to a marginal UV-B-induced reduction by about 8% under moisture deficit. Examination of the UV -B x drought interaction at final harvest showed that this represented a significant mitigating effect of drought for UV -B sensitivity (P < 0.05). UV-B-induced growth decreases were almost equal at around 30% under both water regimes following drought establishment (at harvest 3, Fig. 3- 1 b). Mediation of the UV -B effect depended on the duration of drought (significant harvest x UV -B x drought term in REML analysis of biomass accumulation during drought, Table 3 -2). 3. Mod{ficGtions of white clover growth responses to UV-B Table 3-2. Residual maximum likelihood (REML) analysis of log-transformed aerial biomass production of nine white clover populations collected in four harvests during 1 2 weeks growth under supplementation with or without UV -B of 13 .3 kJ m -2 d- l . Half of the samples collected during the final two harvests also experienced four weeks of water l imitation. Plant Response Harvest UV-B Water Population Harvest x UV-B Harvest x Water UV-B x Water Harvest x Population UV -B x Population Water x Population Harvest x UV-B x Water Harvest x UV-B x Population Harvest x Water x Population UV-B x Water x Population Harvest x UV-B x Water x Population p < 0.00 1 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 0.584 < 0.00 1 0.0 1 1 0.033 0.003 0.808 < 0.001 0.463 0.935 54 At final harvest (Fig. 3 - 1 a) the extent of growth reduction after 12 weeks of UV-B i rradiation was less than half that of four weeks of drought treatment across the white clover populations. At final harvest, the average drought-induced productivity decrease across populations was 7 1 % in non-irradiated plants, compared to a drought-elicited growth reduction of 65% under UV-B (Fig. 3 - 1 a) . Examination of the UV-B x drought interaction revealed that this difference in the drought response was non-significant between the two UV -B treatments. Drought effects were even more comparable at the first harvest of the droughted samples (harvest 3), showing near equal growth reductions by about 57% in both UV-B regimes (Fig. 3 - 1b) . A significant harvest x drought term (Table 3-2) revealed that water stress-elicited biomass reduction became greater with increasing duration of drought. 3. Mod!fications of white clover growth responses to UV-B 600 -.... � 500 0) E - C 400 o '';:; (J :l 300 " o .. 0.200 :E C 1 00 o 500 -.... � 0)400 E - c o ;::; 300 (J :l " o 0.. 200 :E C 1 00 o (a) (b) 55 Fig. 3-1. Aboveground biomass production in nine white clover populations grown with (UV+) and without (UV-) supplementation of 1 3 .3 kJ m-2 d- UV-B at (a) final harvest after four weeks drought, compared to (b) the penultimate harvest, 10 days after onset of the drought treatment. The populations are ranked by the productivity of unstressed plants (UV - W +) at final harvest. Error bars are ± SE. W + = well watered; W - = droughted. 3, Modifications of white clover growth responses to UV-B 56 - (a) I (5 ... 1 00 -s::: 0 (.) - 80 0 :E Cl � 60 0 -Q) V) s::: 0 40 C-V) Q) ... m 20 I > :::> 0 ,�'tt .� :::> 0 .*,'tt e ·i> � ,s.'tt .,'tt� '!tr'l1 'b:� ;-tlfI �� �<$ :),.0 *,oq � C::>� -i:-��o; � Il> .,...oq .*� � ... 0 � <,0 � .. oc.; C::>� Fig. 3-2. Relative changes in the rate of biomass production for nine white clover populations in response to supplementation of 13 . 3 kJ m-2 d- I UV -B . (a) Accumulated yield of plants grown for 1 2 weeks under well-watered conditions and (b) plants harvested at the end of the experiment after exposure to drought for four weeks. The populations are ranked by the degree of UV -B sensitivity in (a). Values are the ratios of back-transformed means of logarithm-transformed data. The bars represent the percentage changes required for significant differences (P < 0.05) in UV -B sensitivity for the nine white clover populations calculated from the least significant ratio. 3. Mod(fications (if white clover growth responses to UV-B 57 Influence of population d�fferences and time on UV-B sensitivity In addition to significant (P < 0.00 1 ) within-species differences in constitutive productivity under no UV -B supplementation, there were significant population-specific differences in the degree of the UV-B response in plants growing for 12 weeks under UV -B in well-watered conditions (P < 0.05) (Fig. 3-2a). This was l inked to the productivity of the white clover populations (Fig. 3-3). Greater UV -B tolerance was found in less productive white clover ecotypes (Table 3- 1 , Fig. 3-3). 'Sarikamis' performed best among the ecotypes with no UV-B-induced growth reduction, while the more productive white clover cultivars were generally more sensitive to UV -B, with 'Haifa' showing the strongest biomass reduction by more than 40% (Fig. 3-2a). At final harvest the analysis of biomass accumulation under well-watered conditions largely confirmed these population differences in UV -B sensitivity, but productivity of the droughted samples (four weeks of soil moisture deficit) revealed no significant population differences in the UV -B response (Fig. 3-2b). In addition to an effect of time on UV -B responses under drought, duration of DV-B exposure also affected relative dry matter production under well-watered conditions (Fig. 3-4). On average, the rate of white clover biomass production decreased by 30% during 1 2 weeks of DV-B exposure under well-watered conditions (P < 0.00 1 ), but this was more pronounced during initial harvests with a 40% growth reduction, compared to a decrease of about 20% eight weeks later. A significant harvest x UV -B interaction (Table 3-2) underlines that the white clover populations became less UV -B-sensitive with progressing exposure to DV-B . The ranking in DV-B sensitivity among the white clover populations generally remained similar, irrespective of time (non-significant interaction term for harvest x UV -B x population, Table 3-2). 3. Modifications afwhite clover growth re.lpanses to UV-B 58 y = -0.0219x + 104.28 - r = -0.7017 _arikamis "0 1 00 ... P < 0.05 -C 0 Q - 80 0 :E 0 � 60 0 - _restige Q) "aifa I/) c 0 40 Cl.. I/) � m , 20 > :::> 0 0 500 1000 1 500 2000 OM production (mg week-1) Fig. 3-3. Relationship between constitutive productivity (Le. biomass production without UV -B supplementation) during 1 2 weeks growth in controlled environmental conditions and relative growth (dry matter production under UV-B expressed as a percentage of that of the control) in response to elevated UV -B of 1 3 .3 kJ m-2 d- I for nine white clover populations. DM = dry matter. 3. Modifications of white clover growth responses to UV-B .-.. E E 80 L I GI III C o a. III GI ... 60 CD :> :> 50 �------------�------------�--------------�------------� 59 Fig. 3-4. Time course of the mean UV-B response under well-watered conditions measured as relative growth (dry matter production under UV-B expressed as a percentage of that of the control) across nine white clover populations during four consecutive harvests. Plants were grown for 1 2 weeks with and without supplementation of 1 3 .3 kJ m-2 d- 1 UV-B. The graph shows the ratios of the back­ transformed means, constituting an overall summary of UV -B sensitivity for each harvest. The error bar represents the percentage change calculated from the least significant ratio that is required for a significant harvest effect (P < 0.05). 3. Modifications o.fwhite clover growth responses to UV-8 60 Discussion This study revealed significant reductions in biomass production for white clover after 1 2 weeks of exposure to UV -B under controlled environmental conditions. Most importantly, it became evident that the main UV-B effect was significantly modified through interactions with drought, white clover population and duration of stress exposure. Influence of drought on UV-B sensitivity The results from biomass production after four weeks of drought indicate an antagonistic, ameliorating influence of water stress on UV -B sensitivity in white clover (Fig. 3- 1 a). Other studies have shown additive (Teramura et al., 1984b), antagonistic (Balakumar et al., 1 993 ; Manetas et al. , 1 997; Murali and Teramura, 1 986a; Petropoulou et al., 1 995; Sullivan and Teramura, 1 990; Teramura et al., 1 990) and synergistic (Bjorn et al., 1 997 ; Drilias et al., 1 997; Tevini et al., 1 983a) interactions between UV-B and drought. Studies reporting beneficial effects of drought for UV-B sensitivity have shown that detrimental UV -B effects under well-watered conditions can either be removed (Murali and Teramura, 1 986a) or even reversed by drought (Balakumar et al. , 1 993 ; Teramura et al. , 1 990). This has been related to a number of drought-induced morphological and physiological changes of relevance for UV-B protection, including growth delay and radical scavenging (Beggs et al. , 1 986). From Fig. 3- 1 it appears that with increasing duration and severity of drought, the resulting water deficit assumes the main l imiting function for productivity, masking more subtle effects caused by UV-B . Thus the extent of the UV-B x water stress interaction and even its direction can be influenced by the extent of drought. Examination of the UV -B x drought interaction at the first harvest during drought application would have resulted in concluding an additive response, rather than the antagonistic interaction observed at final harvest (Fig. 3- 1 a,b). Drought alone reduced productivity more than double compared to UV -B . Such pronounced differences i n responses to the two stress forms have also been observed in other studies, e.g. i n glasshouse studies of Brassica (Conner and Zangori, 1 998). A masking effect of drought for UV -B sensitivity i s in accord with reports from field-grown soybean 3. Modifications of white clover growth responses to UV-B 6 1 (Sullivan and Teramura, 1 990), where UV-B effects appeared predominant under wel 1 - watered conditions while UV -B-generated growth reductions were obscured when growth was already strongly decreased by drought. Beneficial interaction between UV -B and drought has also been observed as an improvement of the drought response through preconditioning with UV -B (Nogues et al. , 1 998). Such effects have been linked to UV -B-induced cuticle thickening, resulting in decreased water loss and improved plant performance under drought (Manetas et al., 1 997). In the present study, there was no significant interaction of UV -B with water stress responses, indicating that UV -B did not play a major role in the improvement of drought sensitivity. It cannot be ruled out, however, that such interaction would occur over longer drought periods. In addition, there may be beneficial effects on subsequent recovery from drought stress. Influence ofpopulation differences on UV-B sensitivity To date, only limited research on UV -B responses in combination with drought has been conducted for populations within the same species. Such intraspecific information is largely based on earlier comparisons between two soybean cultivars, suggesting that population differences in UV -B sensitivity can be altered or removed under drought (Murali and Teramura, 1 986b; Teramura et al., 1 990). In the present comparison of nine white clover populations, significant differences in the UV -B response were found among white clover populations under well-watered conditions, ranging from no decrease in biomass accumulation in the ecotype 'Sarikamis' to more than 40% growth reduction in the agricultural cultivar 'Haifa' (Fig. 3 -2a) . However, no significant population differences could be recorded in the UV -B response of droughted plants (Fig. 3-2b) . A levelling effect by drought on within-species differences in DV-B responsiveness could therefore be of particular advantage for higher yielding agricultural white clover cultivars. Population-specific differences in UV -B sensitivity have been demonstrated in a number of plant species, including Pinus (Pukacki and Modrzynski, 1 998), maize (Correia et al. , 1 998), rice (Barnes et al. , 1 993 ; Dai et al. , 1 994), cucumber (Murali and Teramura, 1 986c) and soybeans (Biggs et al., 198 1 ; Murali and Teramura, 1 986b; 3. Modifications ()fwhite clover growth responses to UV-B 62 Teramura et al., 1 990). No DV -B effect on yield was measured within and between a number of forage species under ambient northern hemisphere DV -B levels in the field, including white clover (Papadopoulos et al., 1 999). However, that study did not specify the type and environmental background of the cultivars used. A similar study under southern hemisphere DV -B levels showed reduced growth for a New Zealand white clover cultivar (Matthew et al., 1 996). The significant within-species differences for DV -B sensitivity observed here allowed closer examination of possible relationships to the environmental background and productivity of the white clover populations. While some reports suggest a relationship between DV-B responsiveness and natural DV-B backgrounds (Barnes et al., 1 987 ; Pukacki and Modrzynski, 1 998; Sullivan et al., 1 992), others show no such relationship (Dai et al., 1 994; van de Staaij et al., 1 995). Comparisons between DV-B-induced growth responses of the white clover populations also suggested no clear relationship to likely DV -B levels at the habitat of origin. The Scandinavian ecotype 'Haggas' (low ambient DV -B) displayed similar DV -B tolerance to the ecotypes from lower latitudes and higher altitudes (Fig. 3-2a). Nevertheless, another relationship between DV -B responsiveness and habitat of origin became apparent, suggesting a link to habitat productivity. Plant populations exposed to multiple forms of stress in their habitat of origin displayed greater tolerance to DV-B (Table 3- 1 ). Stress factors operating in the natural habitat of the ecotypes 'Sarikamis' , 'Tienshan' and 'Haggas' include reduced habitat fertility, cold temperatures and low precipitation (Table 3- 1 ). These factors are all known to elicit defense mechanisms that could also be of potential advantage for DV -B protection, including higher antioxidant levels (Grace et al., 1 998; Pizzi and Cameron, 1 986). A related study found high accumulation of specific flavonoids and a link to DV -B protection in these ecotypes (see later in Chapter 6, Fig. 6-6). Similarly, the 'Syrian' breeding line derives from drought-prone habitats and was selected for tolerance to leaf rust, a stress form reported to result in higher polyamine (Bharti and Sawhney, 1 996) and oxidase activity (Sharma and Sharma, 1 997). Thus it appears these other stresses may have resulted in specialisations for these popuIations that could better adapt them to DV-B . 3. Modifications of white clover growth responses to UV-B 63 In addition, a significant relationship was demonstrated between constitutive biomass production and UV -B sensitivity (Fig. 3-3). White clover cultivars selected and bred for agricultural productivity were more affected by UV -B than less productive, unselected ecotypes (Fig. 3-3, Table 3- 1 ) . This is in accord with similar findings from large-scale screening of white clover populations using short-term UV -B radiation (Chapter 2). The highly productive breeding line 'Octoploid' contains double the usual tetraploid white clover chromosome set (Woodfield and Cousins, 1 999), which could also represent an additional target for UV-B-induced nuclear DNA damage (Campbell and Grime, 1 993). Considering relative drought tolerance and origin at latitudes with potentially higher UV-B i rradiation (Table 3 - 1 ), the observed UV-B sensitivity for 'Prestige' was less expected. However, in other studies 'Prestige' was Iow in the accumulation of flavonol glycosides of relevance for UV -B protection (Chapter 6). The UV -B-sensitive high­ yielding populations 'Haifa' and 'Kopu' have displayed poor performance under a variety of stress forms, for instance in dry conditions and frost (Table 3- 1 ). Similar has also been observed in studies on the general-purpose cultivar 'Huia' (Table 3- 1 ). Taken together, these results are in accord with ecological models of the evolution of stress tolerance, predicting higher tolerance (less severe relative biomass reduction) for species and populations from low habitat and plant productivity (Grime, 1 979; Poorter and Gamier, 1 999). Influence of time on UV-B sensitivity UV -B sensitivity in white clover also appeared to be influenced by time, suggesting improvements for aerial biomass reductions with increasing duration of UV -B exposure (Fig. 3-4). Some studies report cumulative detrimental effects for plant growth in response to increased duration of UV-B exposure (Bjoro et al. , 1 997; Tevini et aI. , 1983a). Other findings, however, point at beneficial effects of increasing time for UV-B sensitivity. In soybean, growth reductions have been observed in early, but not advanced stages of plant development (Teramura et al. , 1 984b). In faba beans, decreasing UV-B sensitivity over time has been related to improved photorepair functions, changes in plant architecture and increasing absorbance by UV -absorbing compounds (Visser et al. , 1 997b). It remains to be tested whether time dependency of 3. Modifications (?fwhite clover growth responses to UV-B 64 the UV-B response under continuous UV-B exposure would experience beneficial or detrimental modifications during the daily and seasonal UV -B fluctuations in the field. It is concluded that white clover growth could be less affected by UV -B during periods of drought, compared to moist seasonal conditions. Within-species comparisons show that plant and habitat productivity can serve as predictors for UV-B responsiveness. 4. White clover morphology under UV-8 and mediation by drought 65 Chapter 4. Effects of u ltraviolet-B radiation on Trifolium repens morphology: mediation by drought and relationsh ips to UV-B sensitivity Dissected white c lover bud with apical dome and leaf primordium This Chapter has been submitted to Plant, Cell and Environment. 4. White clover morphology under UV-B and mediation by drought 66 Abstract This study examined effects of UV -B on several aspects of white clover growth and morphology. Comparisons were made with and without drought to determine whether limited moisture availability would alter such responses. Nine white clover populations were examined to investigate possible relationships between morphology and UV-B sensitivity. A factorial design included supplementation with and without 1 3 .3 kJ m-2 d- I UV-B for 1 2 weeks, with and without additional exposure to four weeks of drought. Changes in total biomass production were accompanied by changes in morphological attributes contributing to productivity. In general, UV -B in well-watered conditions reduced leaf mass and leaf size by about 1 5%, rates of leaf appearance by 5% and stolon elongation by 17%. Under drought conditions, these UV-B-induced changes were only marginal or non-significant. Drought treatment resulted in much greater decreases in these attributes than UV -B. The number of leaf primordia formed in the apical bud was decreased by 4% under UV -B , and by 10% in response to drought. Leaf dry matter al1ocation attributes (specific leaf mass, SLM and percentage leaf dry mass, PDM) were strongly affected by moisture deficit, rather than by UV -B. Drought increased SLM by 50% and PDM by 40%. UV-B sensitivity in the white clover populations was associated with higher plant productivity, larger leaf size as well as higher PDM under UV-B. Thus morphological attributes conferring fast potential growth under productive conditions carry a cost in the form of higher UV -B sensitivity. The observed amelioration of UV -B sensitivity under drought could be related to pronounced increases in SLM and PDM. These findings propose the existence of distinct functional types within white clover that might be used to predict UV-B sensitivity_ Introduction Stratospheric ozone loss has been demonstrated to result in increased ultraviolet-B irradiation (UV-B, 290-3 1 5 nm) (McKenzie et al. , 1 999). The temperate pasture legume white clover experiences UV -B levels in New Zealand that are up to 50% higher than at comparable temperate latitudes in the northern hemisphere (Seckmeyer et al. , 1 995). Several publications have reviewed UV -B effects on agro-ecosystems (e.g. Krupa et al. , 1 998b; Rozema et al. , 1 997c), but detailed information on UV -B responses in pasture species is l imited, particularly in conjunction with other forms of stress . In the New Zealand pasture ecosystem and many other habitats there is an increased likelihood of 4. White clover mOfphology under UV-8 and mediation by drought 67 drought coinciding with higher DV-B levels during prolonged clear-sky periods. Drought is one of the most common forms of stress encountered by plants in the field (Hsaio, 1 973). Water availability has been ranked a primary factor impacting on DV-B responsiveness among environmental change processes (Gwynn-Jones et al., 1 999a) . The combination of DV -B and drought has been investigated in several studies (e.g. reviewed in Bomman and Teramura, 1 993; Caldwell et al. , 1 995). Some examinations point at a synergistic, worsening effect of drought stress for DV -B sensitivity (Conner and Zangori, 1 997 ; Feldheim and Conner, 1 996; Gwynn-Jones et al., 1 999a) while others found no drought-induced modifications of the DV -B effect (All en et al. , 1 999). Reports of additivity for DV -B-induced changes under differential moisture availability also suggest that DV -B responses can act independently of water availabi lity (Conner and Zangori, 1 998 ; Teramura et al., 1984b). Many studies, however, point at ameliorating effects of drought exposure for DV -B responses (Balakumar et al. , 1 993 ; Manetas et al., 1 997; Murali and Teramura, 1 986a; Sullivan and Teramura, 1 990; Teramura et al., 1 990). DV -B regulates plant development from molecular to ecosystem levels (Rozema et al. , 1 997b). It has been suggested that morphological changes are a primary outcome of elevated DV -B radiation (Rozema et al., 1 997b). Examinations of intemode elongation revealed increases (Bjom et al., 1 997), no change (Murali and Teramura, 1 986a) and decreases in response to DV-B (Gonzalez et al., 1 998; Hoffmann, 1 999). Leaf area in Nerium oleander decreased in response to DV -B under drought, but not in well-watered seasonal conditions (Drilias et aL. , 1 997), while the opposite response has been found for soybean (Murali and Teramura, 1986a; Sullivan and Teramura, 1 990). The number of leaves was not affected by DV -B in pea, but showed decreases under drought (Allen et al., 1 999; Nogues et al. , 1 998). Few studies on DV-B-induced leaf area responses have separated effects on leaf size (area per leaf) from those on leaf number (Nogues et al., 1 998). Furthermore, l ittle is known about DV -B effects on leaf primordia formation in the apical bud, a process that is affected by other environmental conditions, e.g. temperature and daylength (Thomas, 1 987b). 4. White clover morphology under UV-B and mediation by drought 68 There are also repolts on alterations in relative leaf dry matter allocation under UV -B. Findings of UV -B-elicited decreases in fresh mass without simultaneous reductions in dry mass indicate disruptions in plant water relations for a number of plant species, including cucumber, radish, maize and barley (Krizek, 1 975; Tevini et al., 1 98 1 ) . Leaf dry mass allocation to each unit of leaf area (specific leaf mass, SLM) has been shown to increase in response to higher UV -B levels (Filella and Penuelas, 1 999), in drought conditions (Murali and Teramura, 1 986a) or under the combination of UV -B and drought (Balakumar et al., 1 993). Drought-mediated changes in SLM also have been related to UV -B protection (Balakumar et al. , 1 993; Murali and Teramura, 1 986a). A number of drought-tolerant plant species are adapted to other forms of stress such as elevated temperatures and high photosynthetic ally active radiation. This has led to suggestions that drought (or general stress-) tolerance might also confer resistance to UV -B (Gwynn-Jones et al. , 1 999b). To date this hypothesis has not been tested comprehensively. A study of two distant plant species (broad bean and wheat) suggested that growth advantages for wheat under UV -B were l inked to higher drought tolerance in the latter species (AI-Oudat et al., 1 998). Functional types are sets of plants displaying similar responses to environmental conditions (Diaz and Cabido, 1 997). However, constraints due to species specificity do not allow ready generali sation of plant functional types based on UV-B responsiveness (Bjorn et al., 1 997) and it appears that more conclusive inferences can only be drawn from large-scale comparisons between species. Alternatively, it is possible to use comparisons among several populations of the same species, which are not compounded by between-species differences. There are indications for UV -B tolerance in populations from habitats with higher ambient UV-B levels (Pukacki and Modrzynski, 1 998). However, at present there is l ittle information avail able on other relationships between UV -B responsiveness and features inherent to plants and their habitat, both on the species and on the population level. This precludes attempts to classify UV -B responsiveness according to functional types and is further complicated by the fact that most studies do not provide information about the environmental and physiological background of the investigated plant material (e.g. tolerance/sensitivity to drought, low temperature, shade) (Gwynn-Jones et 4. White clover mOfphology under UV-B and mediation by drought 69 al. , 1 999b). To predict effects of UV-B on vegetation it would be valuable to have a framework that recognised underlying plant specialisations determining UV -B responsiveness. This work is a step towards such a framework by examining between­ population differences in plant attributes and habitat adaptation for white clover and relating these to the sensitivity of those populations to DV-B. The study examined genotypic differences among nine white clover populations to test several morphological and growth responses to DV -B. Detailed background information of plant and habitat features for these populations have been described in Chapter 3 (Table 3- 1 ). Plants were exposed to a combination of 1 2 weeks UV-B and four weeks drought. In addition to aboveground productivity, particular emphasis was placed on leaf morphology and leaf initiation in the apical bud, complemented by measurements of stolon elongation rates. Two different water regimes were included to test the hypothesis that DV -B responses would be less pronounced under drought. It was further hypothesised that morphological differences among the white clover populations would be linked to their DV-B sensitivity. Materials and Methods Experimental design, plant material and cultivation, DV -B radiation treatment and application of drought have been described in Chapter 3. All plant measurements reported here were conducted at the conclusion of the 1 2 week UV-B exposure period (final harvest). Droughted plants were harvested the day after rewatering in order to prevent collection of wilted material . Biomass production and leaf attributes Aerial biomass was dried for 48 h at 80°C and subsequently weighed to calculate the rate of dry matter production. Sensitivity to UV -B was expressed as the percentage change calculated from the comparison of dry matter production between UV -B-treated and control plants. Leaf size was measured at the end of the experiment with a Ll-COR Model 3 100 area meter on four fully open distal leaves per plant. Laminae were subsequently dried at 80°C for 48 h for dry mass determination. The ratio of dry mass 4. White clover morphology under UV-B and mediation by drought 70 over leaf area was calculated to give specific leaf mass (SLM), while leaf dry mass percentage (PDM) was calculated from (dry mass/fresh mass)* 1 00. Stolon growth and apical bud dissection Shortly before onset of the drought period, ten stolons per population and treatment condition were marked at the node of a fully open distal leaf (leaf developmental stage 1 .0 (Carlson, 1 966)). At final harvest, the marked stolons were collected and stolon length and number of nodes determined in order to calculate stolon elongation rate (SER) and leaf appearance rate (LAR) per day. A subset of white clover populations CKopu', 'Huia' and 'Tienshan') were investigated in detail during dissection studies, examining leaf primordia formation in the apical bud. Examinations were conducted using a Nikon SMZ-2T stereoscopic dissecting microscope with lighting supplied by a Schott KL 1 500 electronic cold-light source through fibre optic cables. Leaf primordia and their stipules were sequentially removed with a scalpel until the apical meristem was exposed, surrounded by the youngest primordia. The number of leaf primordia in the apical bud was determined for five replicate samples per population under each treatment condition. Statistical analysis Statistical analysis was performed with the General Analysis of Variance procedure in GENSTA T (Genstat, 1 993), using either raw values or values transformed to a square­ root or logarithmic scale. The GENSTAT Regression Analysis and Correlation procedures were used to examine relationships of plant attributes and UV -B sensitivity at final harvest. In an earlier study significant differences were established among the nine white clover populations under well-watered conditions for constitutive productivity (i.e. biomass production without UV -B supplementation) as well as for UV-B sensitivity (Chapter 3 , see Fig. 3-2a). However, these earlier examinations revealed no population differences for UV-B sensitivity in plants exposed to drought. Correlations to aspects of constitutive biomass production and UV -B sensitivity were thus conducted on well-watered plants. 4. White clover morphology under UV-B and mediation by drought 7 1 Results In well-watered conditions, UV -B significantly reduced leaf size and leaf mass by about 1 5% across white clover populations, while plant productivity was reduced approximately by 20% (Table 4- 1 ). This was different (P < 0.05) from observations under drought where the productivity decrease was marginal at about 8% while no significant UV -B-generated changes could be found for leaf size and leaf mass (Table 4- 1 ) . Leaf appearance rate (LAR, -5%) and stolon elongation rate (SER, - 1 7%) also revealed significant overall reductions in well-watered conditions, but no change under drought (Table 4- 1 ). Across white clover populations, specific leaf mass (SLM) did not change significantly under UV -B in either moisture treatment, and UV -B-elicited percent leaf dry mass (PDM) reduction by about 2% was marginally significant (P = 0.078) under drought (Table 4- 1 ). Compared to the UV -B-induced changes, drought affected morphology to a stronger degree than UV -B in all attributes measured (Table 4- 1 ). Across UV -B treatments and white clover populations, significant (P < 0.00 1 ) decreases were recorded for productivity (-70%), leaf size (-50%), leaf mass (-25%), LAR (-25%) and SER (-60%), while SLM and PDM similarly significantly (P < 0.00 1 ) increased by 50% and 40%, respectively, in response to moisture deficit. Dissection of the stolon tip in a subset of white clover populations revealed decreases for leaf initiation in the bud by an average 4% under UV-B and by 1 0% in response to drought (P < 0.05, Fig. 4- 1 ) . 4. White clover morphology under UV-B and mediation by drought 72 Table 4-1 . Attributes of growth and morphology (mean ± 1 SE) in white clover grown with (UV+) and without (UV-) supplementation of 1 3 .3 kJ m- 2 d- l UV-B , under well­ watered (W +) or drought conditions (W -). Probabil ities reflect significance of the UV-B effect under the two water regimes. P < 0.001 = *** ; P < 0.01 = **; P < 0.05 = *; P < 0. 1 0 = +, P � 0. 10 = n/s. LAR = leaf appearance rate; PDM = percent leaf dry mass; SER = stolon elongation rate; SLM = specific leaf mass. Attribute Productivity (mg dail ) Leaf size (cm2) Leaf mass (mg) SLM (mg cm-2) PDM (%) LAR (dail ) SER (mm dail ) "0 :J � 6 I\'l (.) '0.. I\'l c I/) Q) 4 > I\'l .E! '0 ... Q) .0 2 E :J Z Kopu Water W+ W - W+ W - W+ W W+ W - W+ W - W+ W - W+ W - UV- 353 . 1 ± 22.04 1 03 .6 ± 4. 14 8 . 1 2 ± 0.798 3 .68 ± 0.267 1 7 .6 ± 1 .93 1 1 .8 ± 0.95 2. 1 2 ± 0.034 3 . 1 7 ± 0.041 1 2. 8 ± 0. 1 7 1 8. 1 ± 0.20 0.256 ± 0.0043 0 . 1 83 ± 0.0044 3 . 32 ± 0. 1 65 1 .22 ± 0.073 Huia UV+ P 269.0 ± 1 3 .37 *** 94.3 ± 2.87 + 6.90 ± 0.677 *** 3 .5 1 ± 0.292 nls 1 5 .4 ± 1 .69 ** 1 1 .2 ± 1 .0 1 n/s 2. 1 8 ± 0.027 nls 3 . 1 5 ± 0.036 nls 1 2.6 ± 0. 1 8 nls 1 7 .7 ± 0. 1 9 + 0.242 ± 0.0050 * 0. 1 83 ± 0.0045 nls 2.75 ± 0. 1 54 * 1 . 1 5 ± 0.067 n/s Tienshan Fig. 4-1 . Leaf initiation in apical buds of three white clover populations after 1 2 weeks supplementation with (UV+) or without (UV-) 1 3 , 3 kJ m-2 d- l UV-B, under well­ watered (W +) or drought conditions (W -), Error bars are ± SE. 4. White clover morphology under UV-B and mediation by drought 73 Between-population comparisons Previously, significant differences were reported in the productivity and UV-B sensitivity of the nine white clover populations under investigation, with UV-B tolerance for slower-growing, stress-tolerant ecotypes such as Tienshan', 'Sarikamis' and 'Hag gas' (Chapter 3, Fig. 3-2a). In the present study there were significant intrinsic differences among the white clover populations for all leaf attributes (P < 0.0 1 ) . For example, 'Octoploid', Haifa' and 'Kopu' were characterised by large leaf sizes (Fig. 4-2) and low PDM (Fig. 4-3), when compared to less productive white c lover populations. However, response functions among white clover populations were more similar than within populations for the leaf measurements, thus preventing detection of significant within-species differences in foliar UV-B responses. Tienshan', 'Sarikamis' and 'Haggas' were more tolerant to UV -B in their stolon elongation rate under well-watered conditions, and together with 'Kopu' and 'Octoploid' this was significantly (P < 0.05) different from the SER reductions in the remaining populations (Fig. 4-4). At the conclusion of the experiment, UV -B sensiti vity in the nine white clover populations was related to higher constitutive productivity (Fig. 4-5a). Morphological attributes that were associated with both aspects of white clover growth in this study included leaf size and PDM. Large leaf size was associated with higher constitutive productivity (Fig. 4-5b) and also tended to relate to UV-B sensitivity (r = 0.588, P = 0.096). Significance for the latter correlations was also maintained under UV -B . PDM showed inverse relationships under UV-B to plant productivity (Fig. 4-5c) and to UV-B sensitivity (r = -0.636, P = 0.066). Productivity under drought was directly related to subtle increases in SLM (r = 0.690, P < 0.05). 4. White clover morphology under UV-B and mediation by drought 74 Fig. 4-2. Leaf size in nine white clover populations grown with (UV +) or without (UV -) supplementation of 1 3 .3 kJ m-2 do l UV-B, under well-watered (W+) or drought conditions (W -). The populations are ranked by the leaf size of unstressed plants (UV­ W+). Error bars are ± SE. 20 :iE o Q. 10 5 o Fig. 4-3. Percentage leaf dry mass (PDM) in nine white clover populations grown with (UV+) or without (UV-) supplementation of 1 3 .3 kJ m-2 d- 1 UV-B, under well-watered (W +) or drought conditions (W -). The populations are ranked by the PDM of unstressed plants (UV- W+). Error bars are ± SE. 4. White clover morphology under UV-B and mediation by drought 5 .:::'"' 4 '>. (IJ "0 e 3 e - a: w 2 (J) o 75 Fig. 4-4. Sto)on elongation rate (SER) in nine white clover populations grown with (UV+) or without (UV-) supplementation of 1 3.3 kJ m-2 d- 1 UV-B, under well-watered (W +) or drought conditions (W -). The populations are ranked by the SER of unstressed plants (UV - W +). Error bars are ± SE. 4. White clover morphology under UV-B and mediation by drought - e c 100 o (.) '0 :E o t 80 5l c: 8. ! 11) 60 :> :::I (a) Sarikamis. y = ,0.0968x + 114.58 . = ,0.739 P < 0.05 • Haifa � L-____ � ____ � ____ � ______ � ____ � ____ � ____ � 20 � E 15 .!:. Cl .!::! III 1ii 10 .s 5 0 14 13 C :E 0 Q. 12 1 1 10 0 (b) (c) 100 y = 0.0378x ' 5.2468 r = 0.954 P < 0.001 y = -O.0069x + 15.034 r = -O.930 • Haggas P < 0.001 200 300 400 DM production (mg dol) eOctoploid eOctoplOid 500 600 76 Fig. 4-5. Relationships of the constitutive productivity of nine white clover populations, measured as biomass production after 1 2 weeks growth under well-watered conditions without UV -B supplementation, to UV -B sensitivity and leaf attributes, (a) Relative growth, i .e, dry matter production after supplementation for 1 2 weeks with 1 3 .3 kJ m-2 do l UV -B, expressed as a percentage of that of the control ; (b) constitutive leaf size; (c) percentage leaf dry mass (PDM) under UV-B . DM = dry matter. 4. Whife clover morphology under UV-B and mediation by drought 77 Discussion Leaf and stolon morphology This study revealed morphological changes that could affect plant and ecosystem processes (Rozema et al., 1 997c). Significant UV-B-induced reductions in overall aboveground productivity under well-watered conditions were accompanied by decreases in all measured morphological features contributing to biomass production (leaf size, leaf mass, leaf number and stolon elongation) (Table 4- 1 ) . Simultaneous decreases in total biomass accumulation and leaf area have been reported in a number of other studies (Correia et aL., 1 999b; Li et al., 1 998; Nogues et al., 1 998). A field study of white clover also revealed similar UV -B-induced decreases in leaf size (Matthew et al., 1 996). In addition, UV -B-generated reductions in leaf appearance (LAR) were observed under well-watered conditions (Table 4- 1 ) . In a study on pea, leaf size was decreased by UV -B, while leaf number only was significantly reduced by drought (Nogues et al., 1 998). Leaf number reductions due to UV -B have been observed in Phlomis fruticosa (Nikolopoulos et al., 1 995). The observed reductions in both leaf size and leaf number suggest that reduced biomass accumulation results from reduced net carbon gain in both aspects contributing to total plant leaf area. Apical bud dissections in three white clover populations show for the first time decreases in leaf primordia formation under UV -B (Fig. 4- 1 ). This represents a developmental stage largely characterised by cell division (Thomas, 1 987b). Previously it was demonstrated that UV -B-induced growth reductions were accompanied by decreased initiation in the number of apical buds (Chapter 2, Fig. 2- 1 ) and the present findings suggest that detrimental UV -B effects on growth processes also occur inside the remaining buds. UV-B-induced changes in regions concealed from direct exposure have also been reported in pea, with reductions of leaf area and cell division in young leaves shielded against UV -B by folded bracts (Nogues et al. , 1 998). Rather than direct effects, these findings suggest indirect modes of UV -B action resulting in reduced cell division in well-shielded plant parts, for instance decreased photosynthate supply or systemic responses through signal transduction and plant growth substances (Meijkamp et al. , 1 999; Perennes et al., 1 999). 4. White clover morphology under UV-B and mediation by drought 78 The present study also revealed detrimental UV -B effects for the elongation of white clover stolons under well-watered conditions (SER, Table 4- 1 and Fig. 4-4). Reduced internode elongation in response to UV-B has also been observed in other studies (Ballare et al., 1 995; Gonza1ez et al., 1 998; Hoffmann, 1 999). Plant height was reduced due to UV-B in Brassica under well-watered, but not droughted conditions (Conner and Zangori, 1 998). White clover stolon growth predominantly occurs horizontally and partially shaded by the leaf canopy and the observed decreases in elongation could further suggest systemic UV -B effects for plant growth. S imilar to the findings from apical bud dissection, the observed reduction in stolon elongation could be related to reduced cell division. Reduced internode elongation in pea has been related to a decrease in cell numbers (Gonzalez et al., 1 998). Other studies also point at UV -B­ induced destruction of IAA, again resulting in reduced elongation growth (Huang et al., 1 997). Drought appeared to ameliorate or remove UV -B sensitivity III attributes of plant growth and morphology (Table 4- 1 ). It has been argued that UV-B-irradiated plants might be advantaged under drought due to reduced water uptake as a consequence of DV -B-induced area reductions (Nogues et al., 1 998). However, in drought conditions, leaf size and leaf number did not differ significantly between UV -B treatments (Table 4- 1 ), suggesting that such a mechanism was not operative here. Other studies also found DV -B-generated decreases in total plant biomass accumulation and leaf growth under well-watered conditions but a masking effect for such responses under drought (Murali and Teramura, 1 986a; Sullivan and Teramura, 1 990; Teramura et al., 1 990). It i s possible, that mechanisms linked to reduced biomass accumulation under moisture limitation already provide grounds for UV-B protection. Such protective function could be mediated by drought via delays in cell division and expansion, thus reducing the chance of damage by UV -B on these processes. Growth delay has been described as an important UV-B-protective mechanism, particularly when growth of the particular organ is to a significant extent due to cell division (Beggs et al. , 1 986). This would be supported by observations of significant drought effects on processes linked to cell division in the apical bud (Fig. 4- 1 ). 4. White clover morphology under UV-B and mediation by drought 79 Relative leaf dry mass allocation A different response pattern was observed In attributes of relative leaf dry mass allocation (specific leaf mass, SLM and percent leaf dry mass, PDM) (Table 4- 1 ). UV-B had a similar effect on leaf size and leaf mass and thus no significant UV-B-induced change could be found for SLM under either water regime. Other studies found increases for SLM in response to UV-B (Li et al. , 1998; Tosserams and Rozema, 1 995), while - similar to the present findings - SLM was unaffected by UV -B irrespective of water supply in studies on soybeans (Murali and Teramura, 1986b; Sullivan and Teramura, 1 990) . Long-term studies in Phlomis fruticosa (Nikolopoulos et al., 1 995) and investigations of 38 plant species in Mediterranean-type ecosystems (Wand, 1 995) also detected no change in SLM under elevated UV -B, while PDM was slightly decreased in the latter study (Wand, 1 995). Nevertheless, strong increases for SLM and PDM were observed under drought (Table 4- 1 and Fig. 4-3), concomitant with a decrease in overall UV -B sensitivity when moisture supply was l imited (Table 4- 1 ) . This has also been reported in other studies, suggesting involvement of drought-generated SLM increases for UV -B tolerance (Balakumar et al. , 1 993; Murali and Teramura, 1 986a). Thus, in addition to a passive response to drought via reduced cell division, active drought-induced morphogenetic functions could contribute to UV -B protection, e.g. via increases in relative leaf dry mass allocation. Increased SLM can represent increased thickness or higher density of the leaf and protective functions for higher SLM have been related to improved leaf UV-B absorption properties (Britz and Adamse, 1 994). While small changes in SLM under UV -B did not appear to relate to UV -B tolerance, it is possible that only a marked increase in SLM is of relevance for UV -B protection. Between-population comparisons Differential UV-B sensitivity of the white clover populations was related to their higher constitutive productivity (Fig. 4-5a). Populations bred for agricultural yield were more affected by UV -B than unselected, wild ecotypes of lower productivity and from stress­ exposed habitats (Chapter 3, Fig. 3-2a). It was therefore of interest whether such constitutive and UV -B-induced differences in growth would be related to morphological 4. White clover morphology under UV-B and mediation by drought 80 features. It was found that more productive white clover populations had larger leaves with higher relative moisture content (lower PDM, especially under UV -B) (Fig. 4-5b,c) and these attributes in turn tended to correlate with UV -B sensitivity. Thus plant features contributing towards resource acquisition also showed some direct relationships to UV -B sensitivity. Previously, a relationship has been reported between differential leaf size and sensitivity to drought among white clover populations (Barbour et al. , 1 996). Smaller constitutive leaf size and higher PDM in the DV -B-tolerant white clover ecotypes can also be seen as an adaptive characteristic to multiple stress conditions in their habitat of origin. These ecotypes derive from habitats with low annual precipitation of 700mm or below, which would be considered marginal levels for white clover growth (Hoglund and White, 1 985 ; lones and Lowe, 1 993). In other studies, the Asian and Scandinavian ecotypes were distinct from white clover cultivars originating in warmer, moister climates such as 'Huia' on the basis of lower yield, smaller leaves and low cyanogenesis (Caradus and Forde, 1 996; Caradus et al. , 1989b). Lower PDM is reflective of higher relative leaf moisture content which in turn is related to leaf protein content and ultimately higher growth rates (Wilson et al. , 1 999). This is also confirmed by the relationship between PDM and productivity observed in the present study (Fig. 4-5c). No such relationship could be found for SLM, and other studies have also pointed at higher relative importance for leaf dry matter content in the prediction of stress tolerance, rather than SLM which can be influenced by a number of leaf-internal variables (Wilson et al. , 1 999). Nevertheless it was of interest to note that productive populations showed small UV -B-induced increases in SLM under drought. While not related to UV -B tolerance, this could still reflect some morphological adaptation in higher yielding white clover populations. UV -B sensitivity in the higher­ yielding cultivars could be also related to increased genome size or activity (Campbell and Grime, 1 993). This seems supported by UV -B sensitivity observed in 'Octoploid', containing double the usual tetraploid white clover chromosome set. 4. White clover morphology under UV-8 and mediation by drought 8 1 Conclusions From these results taken overall, the observed responses can be divided into two distinct categories. The first type of response is represented by decreased morphological development under UV-B in well-watered conditions, accompanied by further drought­ mediated reductions without an additional UV -B effect (Fig. 4-2 and Fig. 4-4). The second response pattern is that of a pronounced drought-induced increase in morphogenetic aspects of relative leaf dry matter allocation, largely independent of UV­ B (Fig. 4-3). These two response classes are similar to main means of UV-B protection described for plants (Beggs et al., 1 986), namely minimising damage by growth delay and by structural attenuation. The present work addresses a recently expressed need to classify UV -B responsiveness according to functional types (Gwynn-Jones et al. , 1 999b). It has been stated that UV-B sensitivity in crop plants is related to the plant breeding process, selecting (often in glasshouses eliminating UV -B radiation) for attributes that potentially could confer UV­ B sensitivity such as high growth rates, thin leaves, reduced secondary compounds and elimination of structural features affecting palatability (Rozema et al., 1 997 c). Elite germplasm producing high yields is usually adapted to a narrow set of environmental conditions and does not perform well under more marginal environmental conditions (Groth, 1 998). The studies in white clover go beyond the species level, showing for the first time that within-species differences in UV -B responsiveness can be related to different functional population types, separated by attributes linked to plant and habitat productivity. A trade-off between investment in competitive strategy and UV-B susceptibility proposes ecological relationships under UV -B that have previously been established for other forms of stress (Grime and CampbeU, 1 99 1 ; Poorter and Garnier, 1 999). Comprehensive testing will be required to examine such relationships on a between-species level. 5. White clover physiology under UV-8 and mediation by drought 82 Chapter 5. Effects of u ltraviolet-B radiation on Trifolium repens physiology: mediation by drought and relationsh ips to UV-B sensitivity Water potential measurements An abbreviated verSIOn of this Chapter has been submitted to Plant, Cell and Environment. 5. White clover physiology under UV-B and mediation by drought 83 Abstract This study investigated functional responses of white clover to UV -B in interaction with drought. A particular aim was to relate these responses to previously established differences in white clover sensitivity to UV -B . Nine white clover populations were used, ranging from highly productive cultivars to slow-growing ecotypes from stress­ exposed habitats. Using controlled environmental conditions, plants were exposed in a factorial design to supplementation with or without ultraviolet-B radiation (UV-B) of 1 3 .3 kJ m-2 do l for 1 2 weeks, accompanied by four weeks of drought. Examinations included accumulation of UV -absorbing compounds, flavonoids and chlorophylls as well as several measures of photosystem II function. Measurements linked to plant water status included leaf water potential and free proline accumulation. Flavonoids were identified as glycosides of the flavonols quercetin and kaempferol. UV-B increased the levels of UV -absorbing compounds and flavonol glycosides and this was synergistically enhanced in combination with drought. These responses were more pronounced for quercetin than for kaempferol glycosides. Drought alone did not increase flavonol glycoside accumulation. Chlorophyll levels and photochemical parameters were not negatively affected by UV -B . UV -B increased leaf water potential by 1 6% under drought, and proline levels by 23% in well-watered conditions. Differences in these responses due to water treatment and white clover population revealed that increases in UV -absorbing compounds, higher quercetin glycoside accumulation and increased water potential under UV-B can confer UV-B tolerance in drought and are characteristic of slow-growing, UV -B-tolerant populations adapted to multiple stress. I ntroduction New Zealand pasture ecosystems are experiencing elevated ultraviolet-B radiation (UV­ B, 290-3 1 5 nm) as a result of stratospheric ozone loss (McKenzie et al., 1 999). White clover serves as the principal nitrogen-donating component in these ecosystems and has shown susceptibility both to UV-B and drought (Barbour et al. , 1 996; Matthew et al. , 1 996), stress forms that frequently coincide in the natural environment. Drought­ induced water stress influences most physiological processes via turgor reductions and decreases in cell division as well as cell expansion (Hsaio, 1 973) . UV -B effects on 5. White clover physiology under UV-B and mediation by drought 84 various aspects of plant function have also been demonstrated (Rozema, 1 999; Tevini, 2000). Furthermore, drought modifies physiological responses to UV -B (Nikolopoulos et ai. , 1 995; Teramura et al., 1 983 ; Tevini et al. , 1 983a). While UV -B can act as a stress affecting primary plant metabolism, it is also seen as a regulatory factor affecting secondary plant processes, for example via effects on phenolic compounds (Rozema et al., 1 999). Several studies suggest UV -B-protective functions for phenylpropanoid derivatives such as hydroxycinnamic acids and flavonoids (Markham et al. , 1 998b; Meijkamp et al. , 1 999; Olsson et al. , 1 998). Some studies attribute the primary role for flavonoid induction to UV -B (Allen et al., 1 999; Balakumar et aL., 1 993), while other reports suggest an interaction with drought (Nogues et al., 1 998). Comparisons between and within species show that higher UV­ absorbing compound accumulation can be associated with plants from habitats with elevated natural UV -B background levels (Rozema et al. , 1 997 a; Wand, 1 995). A number of studies have also investigated UV -B effects on photosynthesis and photochemistry in conjunction with drought. UV -B-elicited changes on chlorophyll levels and on the chlorophyll a:b ratio can be independent of water supply (Eswaran et al., 1 993 ; Teramura et al. , 1984b), but synergistic interactions have also been demonstrated (Nikolopoulos et al. , 1 995). Other reports show population-specific differences in chlorophyll content under UV -B (Murali and Teramura, 1 986b). Chlorophyll fluorescence studies have frequently been used to study UV -B effects on photosystem I! (PSI!) function. Intrinsic efficiency of PS I! (FvlF m) in dark-adapted leaves is proportional to apparent photochemical quantum yield and often decreases when plants are stressed. Investigations of UV -B effects on the photochemical efficiency of PSI! in combination with drought showed UV-B-induced decreases (Nikolopoulos et aI. , 1 995), no change (AlIen et al. , 1 999), or even increases of FvlFm (Manetas et al. , 1 997; Petropoulou et al. , 1 995). UV-B stress in PSI! after light adaptation has been observed in the form of reduced photochemical yield (MIF m) (Reuber et al., 1 996a). Reduced photochemical efficiency under UV -B has also been measured as decreased photochemical quenching (qp) (Reuber et al. , 1 996b). Photochemical UV -B stress can also be reflected by changes in non-photochemical quenching (NPQ), giving an indication of energy dissipation as heat (Reuber et al., 5. White clover physiology under UV-B and mediation by drought 85 1 996a). Studies in rye showed UV -B-generated decreases in both photochemical and non-photochemical quenching (Reuber et al., 1 996b). There are also reports of population-specific effects of UV -B on PS I I activity (Reuber et al., 1 996a). Compared to the tolerant mother variety, a UV -B sensitive barley mutant displayed reduced FJF m and M/F'm as well as increases in qp under UV-B (Reuber et al., 1 996a), while NPQ showed UV-B-induced decreases in both populations (Reuber et aI., 1 996a). Several studies have investigated UV -B effects on plant water relations. UV -B­ generated improvements have been observed for relative water content and leaf water potential (lfId under drought in pea (Nogues et al. , 1 998). Studies in soybean also showed that UV -B effects on leaf water relations can depend on moisture availability (Teramura et al. , 1 984a). Accumulation of free proline has been related to protection against a number of environmental factors, including drought, salt, temperature and heavy metal stress (Naidu, 1 998; Saradhi et al., 1 995). However, few studies have investigated proline responses to UV -B . A study in Phaseolus mungo reported proline increases under UV -B (De Britto, 1 995). Proline levels have shown insensitivity to UV­ B in well-watered cow pea plants but were decreased by UV -B in drought conditions (Balakumar et al., 1 993). More than 20 studies have investigated UV -B effects in combination with drought, but information about physiological mechanisms underlying the resulting growth responses is less than comprehensive (Nogues et al. , 1 998). There is a need for studies examining constitutive and UV -B-induced functional attributes that can potentially interact with both forms of stress . A beneficial effect of the UV -B x drought combination for growth responses has been related to drought-induced stimulation of UV-absorbing compound accumulation (Murali and Teramura, 1 986a) and to increased leaf water content via increases in leaf cuticle thickness under DV -B (Manetas et aI., 1 997). Very little functional information is available about stress interactions in conjunction with populations of the same species (Krupa et al., 1 998a). In particular it has yet to be tested comprehensively whether adaptation to other forms of stress can be linked to UV-B tolerance (Gwynn-Jones et al., 1 999b). This could be expected from other studies on stress interactions using between-species comparisons 5. White clover physiology under UV-B and mediation by drought 86 (Grime et al. , 1997). In Chapter 3 and Chapter 4 it was demonstrated that detrimental UV -B effects on white clover growth occurred in well-watered conditions, and that these were mitigated under concurrent drought. These studies also showed that UV-B sensitivity under well-watered conditions was related to the productivity of the white clover populations and of their habitat of origin. The present study aims to investigate whether these differences in productivity and UV -B sensitivity can be related to physiological differences among the white clover populations. This would contribute to the establishment of a predictive framework that relates plant specialisations to UV-B responsiveness. To this effect these studies utilised the significant potential for population differences within white clover. Nine white clover populations from different genetic and geographical backgrounds were exposed to 1 2 weeks of UV -B and four weeks of drought. This was done in growth rooms to exclude variability in environmental conditions that would normally interfere in the investigation of processes underlying plant responses to UV -B (Corlett et al. , 1 997). Characteristics of the white clover populations have been outlined in more detail in Chapter 3 (Table 3- 1 ). The study examined accumulation of UV -absorbing compounds, flavonoids and chlorophylls as well as several chlorophyll fluorescence parameters. The application of moisture deficit also warranted examination of aspects related to plant water status, including measurements of VJL and of free proline levels. The drought treatment further allowed testing the hypothesis that functional UV-B responses in white clover would be altered by moisture deficit. This was done to discern possible mechanisms underlying amelioration of the UV -B growth response established previously. A further aim was to compare functional aspects in the white clover populations to the previously observed within-species differences in plant productivity and UV -B-induced growth responses. It was hypothesised that white clover populations tolerant to conditions of stress in their habitats of origin would display a higher degree of biochemical or physiological adaptation to UV -B . Materials and Methods Experimental design, plant material and cultivation, UV -B radiation treatment and application of drought have been described in Chapter 3. All plant attributes reported 5. White clover physiology under UV-B and mediation by drought 87 here were measured at the conclusion of the 12 week UV -B exposure period (final harvest). Droughted plants were harvested the day after rewatering in order to prevent collection of wilted material . UV-absorbing compounds UV -absorbing compound levels were estimated by established procedures (Mirecki and Teramura, 1 984). After oven-drying, samples of four fully open distal trifoliate laminae were ground and weighed ( 1 5-20 mg) into 1 .5 ml centrifuge tubes. Extraction in 1 .2 ml MeOH:H20:HCI (79:20: 1 ) was performed in darkness for 24 h under occasional shaking. This was followed by centrifugation and spectrophotometric analysis of the supernatant in 3 ml quartz cuvettes. Absorbance readings at 300 nm were calculated on the basis of leaf dry weight. This procedure was repeated on five replicates for each white clover population under each treatment condition. Flavonoid analysis For flavonoid analysis, 1 .5-2 g of fully open trifoliate laminae were collected from each pot. This yielded five replicate samples for each of the nine populations at the two treatment levels . Immediately after collection, samples were frozen in liquid nitrogen and stored at -80°C. Each sample was subsequently ground in liquid nitrogen using a mortar and pestle. The resulting fine leaf powder was freeze-dried and stored at -80°C until analysed. The dried, ground plant material (50-60 mg) was extracted with occasional vortex shaking for 24 h in 3 ml MeOH:H20:HOAc (89: 1 0: 1 ), followed by centrifugation and de-fatting of the supernatant with 1 ml petroleum ether. After solvent evaporation under nitrogen followed by vacuum drying, the residue was taken up into 700 �l CH3CN:HOAc:H20 (32:3 :65). The solution was centrifuged and a 25�1 sample was used for HPLC analysis. Analytical HPLC was conducted using a Jasco PU-980 Intelligent HPLC solvent delivery system, Waters 994 programmable photodiode array detector and a Gilson 234 autosampler. Chromatography was carried out on a Merck Supersphere Lichrocart 1 25- 4 RP- I 8 endcapped column (4 �m, 4 mm x 1 19 mm) with a gradient solvent system 5. White clover physiology under UV-B and mediation by drought 88 comprising solvent A [ 1 .5% H3P04] and solvent B [HOAc:CH3CN:H3P04:H20 (20:24: 1.5 :54.5)], mixed using a l inear gradient starting with 80% A, decreasing to 33% A at 30 min, 1 0% A at 33 min and 0% at 39.3 min. Peaks of quercetin and kaempferol derivatives were identified on the basis of the on-line spectra. The integrated areas of all flavonol peaks (measured at 352 nm) were added to calculate flavonol glycoside levels from a standard curve prepared using rutin (quercetin 3-rutinoside). Chlorophyll concentration Extraction of chlorophyll from fully open white clover laminae was performed in N, N­ dimethylformamide (DMF) (Moran and Porath, 1 980). Three samples for each population under each treatment condition were ground, freeze-dried and weighed ( l 0 mg) into centrifuge tubes. Samples were extracted with 1 .2 ml DMF i n darkness at 4°C for 24 h under occasional vortex shaking. This was followed by centrifugation at 20 000 g and 4°C for 3 min. Absorbance of the supematant was read at 664.5 nm and 647 nm and chlorophyll a and b content was calculated using the following equations (Inskeep and Bloom, 1 985): Chlorophyll a = 1 2.7 A664.5 - 2.79 A647 Chlorophyll b = 20.7 A647 - 4.62 A664.5 Total Chlorophyll = 1 7 .9 A647 + 8.08 A664.5 Chlorophyll fluorescence Chlorophyll fluorescence measurements were performed on fully open distal leaflets of 4-5 replicate samples for each population under each treatment condition (Greer, 1 995a; Greer, 1 995b). Samples were stored on moist filter paper immediately after collection in cuvettes sealed with a rubber stopper. Dark-adapted samples were collected before onset of the light period and had been in darkness for at least 1 0 h. Using a PAM fluorometer at room temperature (Pam 10 1 ; Walz, Effeltrich, Germany), initial fluorescence in darkness Fo was measured using a measuring light. Following this, maximum chlorophyll fluorescence in darkness Fm and the fluorescence ratio FvlFm (Fv = Fm - Fo), indicating intrinsic efficiency of photosystem I I (PS II) were determined using a flash 5. White clover physiology under UV-B and mediation by drought 89 intensity of 1 0 000 J.lmol m-2 S- 1 (KL 1500; Walz). For light-adapted fluorescence measurements, samples were collected in the i lluminated growth room and allowed to equilibrate in the cuvette for 4 min at the PPF level equivalent to that in the growth room before measurements were conducted for l ight-adapted initial (Ft) and maximum (Fm') fluorescence and actual efficiency of PS II during i llumination (photochemical yield = M/F'm). Photochemical (qp) and non-photochemical (NPQ) quenching were calculated, where qP = (Fm' - Ft)/(Fm' - Fo) and NPQ = (Fm - Fm')/Fm'. Leaf water potential and proline Leaf water potential ( I{fd was determined using the pressure chamber technique (Scholander et al. , 1 965 ; Turner, 1 98 1 ). Fully open distal leaves were enclosed in a pressure chamber and the cut surface of the protruding petiole was monitored while pressure in the chamber was gradually increased. The pressure required to bring xylem water to the surface (equivalent to the water tension inside the xylem) was recorded. Measurements were conducted on five replicate samples per population and treatment condition. Levels of free proline in fully open white clover laminae were determined using established methods (Magne and Larher, 1 992). For each population under each treatment condition, three samples were ground in liquid nitrogen, freeze-dried and 1 0 mg weighed into centrifuge tubes. Precipitation of protein was done i n 1 .2 ml of 3% (w/v) sulphosalicylic acid under vortex shaking, followed by centrifugation at 1 2 000 g for 7 min. The supernatant was removed and after renewed centrifugation 500 J.ll were collected and made up to 1 ml with water. This was followed by addition of 2.0 ml ninhydrin reagent ( 1 % [w/v] ninhydrin in 60% [v/v] glacial acetic acid), vortex shaking and reaction of the solution for 1 h at 98°C. The reaction was stopped in an ice-water bath, followed by addition of 2 Illl toluene and vOltex shaking for 20 sec. Phases were allowed to separate for at least 5 min and the products extracted in toluene were examined spectrophotometrically in a 1 ml glass cuvette at 520 nm. Free proline content in the white clover samples was calculated from a standard curve of known proline concentrations (0 to 25 J.lg/ml) prepared in an identical manner alongside each batch of samples. 5. White clover physiology under UV-8 and mediation by drought 90 Statistical analysis The same statistical approaches were used as described in Chapter 4. Results UV-absorbing compounds and flavonoids DV -absorbing compound levels displayed an overall increase in response to DV -B (P < 0.00 1 ). This effect was most pronounced in drought-stressed plants, with a significantly (P < 0.05 for the DV-B x drought interaction) greater average change (+1 2%) than well­ watered plants (+3%) (Fig. 5 - 1 ) . Changes in DV-absorbing compound levels under DV­ B showed differences among the white clover populations (P < 0.05), with particular increases for the ecotypes 'Tienshan' and 'Sarikamis' across water treatments (Fig. 5- 1 ) . Flavonoids in the white clover leaf samples were identified by HPLC analysis as glycosides of the flavonols quercetin and kaempferol . S ignificant DV -B-induced increases in flavonol glycosides (P < 0.00 1 ) differed between water treatments (P < 0.00 1 for the DV-B x drought interaction). Drought-induced changes alone showed no consistent pattern and flavonol glycoside levels on average were similar to unstressed conditions. Flavonol glycoside accumulation in the white clover populations in response to DV -B was consistently higher than under unstressed conditions or under drought alone (Fig. 5-2). Compared to unstressed conditions, a more than double increase of flavonol glycoside levels was observed under DV-B, and the increase was more than triple in the DV -B x drought combination across populations (Fig. 5-2). 5. White clover physiology under UV-B and mediation by drought ::lE o ';'tn 1 .3 E 1/1 'C I: :::s 1 .1 o 0.. E o (.) tn 0.9 I: :c .... o .3 0.7 as I > ::::> 0.5 9 1 Fig. 5-1 . UV-absorbing compounds (at 300 nm) per mg leaf dry mass (DM) i n nine white clover populations grown with (UV+) and without (UV-) supplementation of 1 3 .3 kJ m-2 d- I UV-B, under well-watered (W+) or drought (W-) conditions . The populations are ranked by the levels of unstressed plants (UV - W +). Error bars are ± SE 3 .. - 2.5 'tn tn E 2 - 1/1 '0 I: 1 .5 o > as u:: 0.5 o I DUV- W+ D UV+ W+ . UV- W- . UV+ w- I Fig. 5-2. Flavonol glycoside levels per g leaf dry mass in nine white clover populations grown with (UV+) and without (UV-) supplementation of 1 3 .3 kJ m-2 d- I UV-B, under well-watered (W +) or drought conditions (W -). The populations are ranked by the flavonol glycoside accumulation of unstressed plants (UV - W +). Error bars are ± SE. 5. White clover physiology under UV-B and mediation by drought 92 Furthermore, there were significant differences in the UV -B-induced flavonol glycoside responses among white clover populations (P < 0.00 1 ) . Particularly high flavonol glycoside accumulation was observed for the white clover ecotypes 'Tienshan', 'Sarikamis' and 'Haggas' (Fig. 5-2). These responses were also observed for individual flavonols, especially quercetin glycosides, which under UV -B showed higher absolute levels and larger increases than kaempferoJ compounds (Table 5- 1 ). Accumulation of individual flavonol glycosides in the nine white clover populations has been further detailed in Chapter 6, but see also Fig. 5-5a here. The UV -B-generated changes in UV­ absorbing compound levels were related to the UV -B responses in flavonol glycoside accumulation (r = 0.698, P < 0.05). Photosynthetic pigmentation and photochemistry Photosynthetic pigmentation experienced little change in response to UV-B (Table 5- 1 ). UV -B led to a marginal increase in chlorophyll levels by 4% under well-watered conditions and to no change in the chlorophyll a:b ratio. Similarly, no marked UV-B effects could be measured for chlorophyll fluorescence and related quenching factors, with a small 3% UV-B-generated increase in photochemical yield (MIF m) under both watering regimes the only significant change (Table 5- 1 ). Drought had a stronger effect with an average 17% decrease in chlorophyll levels (P < 0.01 ) and a 35% increase for non-photochemical quenching (NPQ, P < 0.00 1 ). No significant interactions between UV -B and drought or white clover populations could be found for chlorophyll levels or any of the photochemical measurements. 5. White clover ph,vsiology under UV-B and mediation by drought 93 Table 5-1. Biochemical and photochemical attributes (mean ± 1 SE) in white clover grown with (UV+) and without (UV-) supplementation of 1 3 .3 kJ m-2 d- l UV-B, under well-watered (W +) or drought conditions (W -). Probabilities reflect significance of the UV-B effect under the two water regimes. P < 0.00 1 = *** ; P < 0.05 = *; P < 0. 10 = +, P 2 0. 1 0 = n/s. /�FIFm = photochemical yield; FvfFm = intrinsic efficiency of photosystem II; NPQ = non-photochemical quenching; qP = photochemical quenching. Attribute Water Quercetin (mg go ! ) W+ W - Kaempferol (mg go ! ) W+ W - Chlorophyll (mg go ! ) W+ W - Chlorophyll a:b ratio W+ W - FvfFm W+ W - MIFm W+ W - qp W+ W - NPQ W+ W - Leaf water potential and proline UV- 0.26 ± 0.03 1 0. 1 3 ± 0.0 19 0.30 ± 0.028 0.35 ± 0.034 1 5 .36 ± 0.225 1 3 .04 ± 0.242 3 . 1 3 ± 0.014 3 . 1 7 ± 0.0 1 6 0.842 ± 0.00 1 7 0 .847 ± 0.00 1 6 0.626 ± 0.0042 0.620 ± 0.0035 0.905 ± 0.0083 0.966 ± 0.0094 0.938 ± 0.0307 1 .332 ± 0.0387 UV+ 0.77 ± 0.049 1 .05 ± 0.048 0.47 ± 0.035 0.65 ± 0.040 1 5 .98 ± 0.306 1 3.32 ± 0. 1 9 1 3 . 1 2 ± 0.014 3 . 1 4 ± 0.02 1 0.843 ± 0.0014 0.848 ± 0.0014 0.639 ± 0.0027 0.639 ± 0.0025 0.922 ± 0.0084 0.978 ± 0.0088 0.970 ± 0.03 1 3 1 .274 ± 0.0387 P *** * * * * * * * * * + nls nls nls nls nls * * * * nls nls nls nls There was a significant (P < 0.00 1 ) UV-B effect on leaf water potential (ljId. The greater overall increase occurred under drought ( 1 6%), which was significantly (P < 0.00 1 ) different from the overall 3% increase for IjIL under well-watered conditions (Fig. 5-3). This compared to a pronounced average decrease of about 50% for If/L under drought (P < 0.00 1 ) (Fig. 5-3) . There were significant differences among well-watered white clover populations in the constitutive (i .e. UV-B-independent) IjIL (P < 0.0 1 ) and in the degree of the UV -B-induced changes for IjIL (P < 0.05). In particular, increases in IjIL could be observed for 'Tienshan' and 'Sarikamis' and decreases for 'Haifa' and 'Octoploid' (Fig. 5-3). A direct relationship between constitutive IjIL and UV -B­ generated IjIL changes (r = 0.729, P < 0.05) showed that populations with low intrinsic IjIL were able to increase their o/L under UV -B, while less negative constitutive values for IjIL were l inked to UV -B-elicited decreases in IjIL (Fig. 5-3). 5. White clover physiology under UV-B and mediation by drought -as Do :E • - .J � 2 1 .5 0.5 o I D UV- W+ D UV+ W+ . UV- W- .UV+ w- I 94 Fig. 5-3. Leaf water potential (l/fd in nine white clover populations grown with (UV +) and without (UV-) supplementation of 1 3 .3 kJ m-2 d- l UV-B, under well-watered (W+) or drought conditions (W -). The populations are ranked by the l/fL of unstressed plants (UV - W +). Error bars are ± SE. Free proline levels were consistently enhanced by UV -B under well-watered conditions, leading to an overall increase by 23% (P < 0.01 ) (Fig. 5-4a). Compared to the well­ watered plants, drought led to a 30-fold increase in proline accumulation independent of UV -B treatment. Averaged across white clover populations, there was no significant overall difference for free proline accumulation in response to UV -B under drought. Nevertheless there were differences in this response among the white clover populations (P < 0.05), in particular with decreases in proline accumulation for 'Syrian' and 'Sarikamis' (Fig. 5-4b). Significant inverse relationships were found between intrinsic proline concentration (i .e. without UV -B supplementation) and increases in these levels under UV-B (r = -0.867, P < 0.01 for well-watered and r = -0.708, P < 0.05 for droughted plants). This showed that in both water regimes free proline concentrations were more strongly increased by UV -B in white clover populations that had low intrinsic proline levels (Fig. 5-4a,b). Populations with low l/fL contained high proline levels under UV -B (r = -0.694, P < 0.05) in drought conditions and a similar relationship could also be found for well-watered plants (r = -0.6 10, P = 0.08 1 ). Under 5. White clover physiology under UV-B and mediation by drought 95 well-watered conditions, there was also an inverse constitutive relationship between ljfL and quercetin glycoside levels (r = -0.730, P < 0.05) . -..... 'm 0.3 m E - 0.2 Q) c:: (5 ... Il. 0.1 o 1 2 1 0 - '701 � 8 - Q) .5 6 (5 ... Il. 4 2 o (a) (b) Fig. 5-4. Proline accumulation per g leaf dry mass in nine white clover populations grown with (UV+, closed columns) and without (UV-, open columns) supplementation of 1 3 .3 kJ m-2 d- I UV -B in (a) well-watered conditions, and (b) under drought. Under both watering regimes the populations are ranked by the proline concentration of plants not exposed to UV -B supplementation (UV -). Error bars are ± SE. 5. White clover physiology under UV-B and mediation by drought 96 Intraspecific relationships to biomass production and UV-B sensitivity Earlier studies showed that UV -B sensitivity in the nine white clover populations was related to higher constitutive productivity (Chapter 3, Fig. 3-3 and Chapter 4, Fig. 4-5a). Physiological factors that were associated with both aspects of white clover growth in this study included quercetin glycoside accumulation and If/L. Productivity of the white clover populations under well-watered conditions showed an inverse relationship to constitutive (r = -0.826, P < 0.01 ) and UV-B-induced accumulation of quercetin glycosides (Fig. 5-5a) and this in turn was linked to UV-B tolerance (r = 0.629, P = 0.069). No relationships to productivity or UV -B tolerance could be found for kaempferol glycosides. S imilarly under well-watered conditions, productive populations had higher (less negative) constitutive If/L (Fig. 5-5b), and this in turn tended to be inversely related to UV -B sensitivity (r = -0.605, P = 0.085). Productive populations decreased their If/L most under UV-B (Fig. 5-5c) and this decrease again showed an inverse trend towards UV-B sensitivity (r = -0.637, P = 0.065). Furthermore, UV-B tolerance in the white clover populations correlated with the capacity to increase accumulation of UV­ absorbing compounds under UV -B (Fig . 5-6), and a similar relationship was found for increases in flavonol glycoside levels (r = 0.588 , P = 0.096). In drought conditions there was a significant inverse correlation of proline concentration with plant productivity (r = -0.728, P < 0.05), showing that the less productive ecotypes contained the highest proline levels (Fig. 5-4b) . 5. White clover physiology under UV-8 and mediation by drought 1 .6 - 1.2 0> 0> .§. c: � 0 0.8 ... as iL 0.5 o Fig. 6-2. Total flavonol concentration in the leaves of nine Trifolium repens populations grown with and without DV-B supplementation of 1 3 .3 kJ m-2 day' l . The bars are the standard errors of the means. DM = dry matter. 1 .4 i" 1 .2 C .... '0') 0') E _ 0.8 III '0 c 0.6 o > as iL 0.4 0.2 o DQ UV- .Q UV+ D K UV- .K UV+ � r r- � rf � � � d � d rf ! -I 11 � Fig. 6-3. Quercetin and kaempferol concentration in the leaves of nine Trifolium repens populations grown with and without DV -B supplementation of 1 3 .3 kJ m-2 day' l . The bars are the standard errors of the means. DM = dry matter. 6. Flavonol responses to UV-B 1 17 Population differences were also maintained under UV-B (P < 0.00 1 ) (Fig. 6-3). High concentrations of both flavonol derivatives and pronounced UV -B-induced increases were found in the ecotypes, particularly Tienshan', reflecting the observations made for total flavonols. Similarly marked responses were observed for the quercetin and kaempferol glycosides in 'Huia'. Compared to the ecotypes, quercetin levels under UV­ B were about half or less in 'Octoploid' and 'Haifa', while kaempferol glycoside levels were more similar (Fig. 6-3). 'Haifa', together with 'Kopu', displayed no measurable increase in kaempferol glycosides under UV -B. 'Kopu', however, displayed a pronounced increase in quercetin derivatives under UV-B. In the populations 'Prestige' and 'Syrian' the low constitutive kaempferol glycoside levels also remained minimal under UV -B at about one third or less of the ecotypes. Furthermore, there was a significant relationship between constitutive and UV -B-induced levels for quercetin (r = 0.785, P < 0.05), but not for kaempferol glycosides. Q:K ratio Significant increases (P < 0.00 1 ) in the quercetin :kaempferol glycoside (Q:K) ratio for most populations further il lustrate the tendency for UV -B-generated quercetin accumulation, rather than kaempferol (Fig. 6-4). Low specific constitutive flavonol levels strongly affected Q:K ratios. Especially low constitutive Q:K ratios were found in the quercetin-poor populations 'Kopu' and 'Octoploid'. The popUlation lowest in kaempferol glycosides, cultivar 'Prestige', displayed the highest Q:K value. Between these extremes there was a significant trend towards higher relative constitutive quercetin glycoside amounts in the T. rep ens ecotypes and cultivar 'Huia' (P < 0.00 1 ). In contrast to the constitutive Q:K ratios, UV -B-irradiated T. repens populations consistently contained more quercetin than kaempferol glycosides (Q:K > 1 ), (P < 0.00 1 ) (Fig. 6-4). UV -B-induced Q:K ratios were again highest in the cultivar 'Prestige', followed by 'Syrian' and 'Kopu' . Q :K ratios were intermediate in the T. repens ecotypes and lowest in cultivar 'Haifa' and in 'Octoploid' (Q:K near 1 ). 6. Flavonol responses to UV-B 5 (5 "- oS! 4 Q. E Q) tU � 3 I: .� � 2 Q) ::::I o o 1 1 8 Fig. 6-4. Ratio of quercetin to kaempferol (Q:K) derivatives in nine Trifolium repens populations grown with and without UV-B supplementation of 1 3 .3 kJ m-2 dai l . The bars are the standard errors of the means. In addition to significant differences among the T. repens populations in altering the Q:K ratio under UV -B, there was also an inverse relationship between such changes and constitutive Q:K ratios (r = -0.894, P < 0.01 ) . 'Kopu', the population with the lowest initial Q:K ratio displayed the highest Q:K increase, while populations with notably more constitutive quercetin than kaempferol glycosides CHaggas' and 'Prestige') displayed the opposite Q:K response. S imilar to quercetin glycoside accumulation per unit leaf dry matter, the quercetin amounts relative to kaempferol also displayed a relationship between initial and UV -B-elicited levels (r = 0.663, P = 0.05). Flavonol relationships to biomass production Constitutive plant productivity and UV -B tolerance correlated with several aspects of quercetin - but not kaempferol - glycoside accumulation. There was a significant inverse correlation between productivity of the T. repens populations to constitutive quercetin glycoside levels (r = -0.866, P < 0.01 ) and Q:K ratios (r = -0.692, P < 0.05). Quercetin glycoside accumulation under UV-B was also inversely l inked to constitutive 6. Flavonol responses to UV-B 1 19 productivity (Fig. 6-5) and in turn tended to relate to UV -B tolerance of the T. repens populations (Fig. 6-6). Examination of the growth responses for the nine T. repens populations suggests significant UV -B tolerance for the lower-yielding ecotypes and sensitivity to UV -B for more productive cultivars (P < 0.05) . The UV -B-mediated growth response also showed some relationship to increases in total flavonols (r = 0.639, P = 0.064) and quercetin glycosides (r = 0.595, P = 0.09 1 ). Y = -0.0004x + 1 .3974 1 .4 r = -0.7752 P < 0.05 - 1 .2 eTienshan :lE 0 '7 en en E 0.8 - s:: .. Q) 0.6 0 .... Q) eOctoploid ;:, 0 0.4 0.2 0 0 500 1 000 1 500 2000 OM production (mg week-1) Fig. 6-5. Relationship between UV -B-induced quercetin accumulation and constitutive dry matter (DM) production for nine Trifolium repens populations. 6. Flavonol responses to UV-B y :: 0.0108x + 0.0001 1 .4 r = 0.6274 P = 0.070 SE 1 .2 o ,... '0) 0) E _ 0.8 c: i (.) 0.6 "-Cl) :::l o 0.4 0.2 eTienshan • Sarikamis o �------------�------------�--------------�----------� 40 60 80 1 00 UV-8 response (%OM of control) 1 20 Fig. 6-6. Relationship between UV -B-induced quercetin accumulation and relative growth (dry matter production under UV-B expressed as a percentage of that of the control) in response to elevated UV -B for nine Tr�folium repens populations. DM = dry matter. 6. Flavonol responses to UV-B 1 2 1 Discussion Total flavonols This study investigated Trifolium repens, a species previously not described with respect to its flavonoid UV -B response. After 1 2 weeks of UV -B exposure in controlled environment conditions there was evidence for treatment-, population- and compound­ specific differential accumulation of flavonols which are likely to contribute towards protection from UV -B radiation. The involvement of flavonols in the UV -B response has been reported for several plant species, including other legumes such as soybeans, Glycine max (Middleton and Teramura, 1 993). The levels of flavonols in T. repens and the pronounced UV -B­ induced change in their amounts by more than 1 00% (Fig. 6-2) are similar to those observed elsewhere, for instance in Brassica (Olsson et al. , 1 998) where total flavonols, composed of kaempferol and quercetin glycosides, exhibited up to 1 50% enhancement in response to UV -B . Similarly, the increment for flavonoid compounds (mainly the flavone saponarin) in Hordeum vulgare (barley) was about 1 00% (Liu et al. , 1 995). Accumulation of flavonoids is a frequently observed response to UV -B radiation (Meijkamp et al., 1 999) and has been linked to DV -B absorption characteristics displayed by flavonoids (Hoque and Remus, 1 999) as well as to the potential for these compounds to act as radical-scavenging antioxidants (Cooper-Driver and Bhattacharya, 1 998). However, the antioxidant function of flavonoids is complex and depends on a variety of factors, including compartmentalisation, redox potential , presence of double bonds, glycosylation and hydroxylation (Bors et al. , 1 995 ; Cooper-Driver and Bhattacharya, 1 998; Rice Evans et al., 1 996). This complexity therefore also needs to be taken into account in the consideration of possible antioxidant functions for increased flavonoid levels under UV-B (Bomman et al., 1 997; Olsson et al. , 1 998). Recent studies on keto-enol tautomerism in the ground- and excited states suggest a mechanism by which kaempferol and quercetin glycosides might provide protection against UV-B damage by (harmless) energy dissipation (Smith and Markham, 1 998). 6. Flavonol responses to UV-B 1 22 D�fferential UV-B e.ffects on quercetin and kaempferol glycosides There was a clear trend among the flavonols for UV -B-induced accumulation of quercetin glycosides (Fig. 6-3). On average, the T. rep ens populations double their Q:K ratio under UV -B, and i t appears this increase is only l imited for populations already high in intrinsic quercetin glycosides relative to kaempferol (Fig. 6-4). Studies in Anethum graveolens (dill) provided early indications towards selective formation of quercetin glycosides as a response to UV-B (Moh le et al., 1 985 ; Mohle and Wellmann, 1 982). However, only recently an increasing number of studies have pointed at differential UV -B effects favouring B-ring-dihydroxylated flavonoids over their monohydroxylated counterparts, for instance in Marchantia (Markham et al. , 1 998a), Hordeum (barley) (Uu et al., 1 995; Reuber et al., 1 996a), Oryza (rice) (Markham et al., 1 998b), Petunia (Ryan et al., 1 998) and Brassica (Olsson et al., 1 998). The present studies on T. repens not only confirm such differential flavonol effects under long-term UV -B treatment for a leguminous plant species, but also show that the attainment of a higher Q:K ratio under UV -B is well-conserved for several genetical ly diverse populations within the same species. It is unlikely that preferential quercetin accumulation would lead to higher UV -B­ absorption, compared to kaempferol (Lavola et al. , 1 997; Mabry et aI., 1 970). UV-B radiation generates free radicals (Hideg and Vass, 1 996) and higher free radical antioxidant activity relative to their monohydroxylated equivalents has been reported for ortho-dihydroxylated flavonoids (Montesinos et al. , 1 995). A possible additional relative advantage for quercetin compounds has also been indicated by photochemical studies (Smith and Markham, 1 998). These suggest that quercetin glycosides may be better able to dissipate potentially harmful UV -B radiation through tautomerisation than the equivalent kaempferol glycosides. Specific quercetin accumulation has been observed as a response to a number of other forms of stress, ranging from heavy metal pollution (Loponen et al. , 1 998) and nitrogen deficiency (Bongue Bartelsman and Phillips, 1 995) to electron-donating paraquat application (Steger-Hartmann et al. , 1 994). In combination with the findings presented here, this underscores a prominent role for ortho-dihydroxylated flavonol accumulation in the general biochemical plant strategy against environmental stress. 6. Flavonol re,lponses to UV-B 1 23 Population differences in the accumulation oftotalflavonols UV -B-generated flavonoid accumulation differed markedly among the nine genetical ly diverse T. repens populations. There were important population-specific differences for the accumulation of quercetin, kaempferol and total flavonols and the capacity to increase these compounds under UV -B. This is in accordance with more general investigations reporting population-specific effects on UV -B-generated flavonoid accumulation (Dai et ai. , 1 992; Dumpert and Knacker, 1 985 ; Murali and Teramura, 1984). The T. rep ens ecotypes stood out among the other populations in that they usually were highest in specific and total flavonols under both treatment conditions, and were also able to increase these levels most, compared to control conditions. Overall the best performing population was Tienshan' which had high constitutive and UV -B­ induced total flavonols and also displayed pronounced UV -B-induced increases. This high performance was usually also seen in the other two ecotypes, 'Sarikamis' and 'Haggas'. The origin of Tienshan' and 'Sarikamis' at high altitudes could suggest adaptation to higher UV -B levels resulting in increased flavonol accumulation. While some repolts indicate a relationship between natural UV -B background levels and the production of UV -B-attenuating pigments, including flavonoids (Rozema et al. , 1 997a; Ziska et al. , 1 992), other studies show no such pattern (Pilon et al., 1 999). Similarly, the Swedish ecotype 'Haggas', originating at high latitudes (Nordiska Genbanken, 1 993), and thus representing a habitat with lower UV -B levels, displays higher flavonol accumulation. This suggests that no overall association exists between UV-B habitat and flavonol production in T. repens. There are several other features shared by the T. rep ens ecotypes. In addition to lower productivity, another common denominator for the ecotypes is adaptation to low temperatures. Low-temperature stress can increase antioxidant capabilities due to enhanced phenylpropanoid accumulation, including flavonoid production (Grace et al. , 1 998). Furthermore, these populations all derive from environments with low annual rainfall , at precipitation levels considered marginal for T. repens growth (Jones and Lowe, 1 993). It can therefore be suggested that the T. rep ens ecotypes in this investigation are also particularly well adapted to extended periods of drought. Tolerance to other forms of stress such as drought has been related to higher UV-B tolerance (AI-Oudat et al. , 1998). More specifically, drought tolerance has been l inked 6. Flavonol responses to UV-B 1 24 to higher flavonoid content (Pizzi and Cameron, 1 986) and could further explain the higher constitutive and UV -B-induced flavonoid levels for these ecotypes. The high flavonol-accumulating capabil ities of the ecotypes were closely followed by the cultivar 'Huia', which displayed similarly high levels and increases. This general-purpose cultivar has been used extensively in temperate pasture ecosystems, and exhibits relatively high productivity and an abil ity to adapt to a variety of environments with high degree of plasticity (Caradus et al., 1 993b; Caradus and Woodfield, 1 997). It appears that biochemical adaptation in 'Huia' was characterised by elevated flavonol production under UV -B during the 1 2 week growth period. The remaining populations with low to intermediate flavonol accumulation generally represented the higher-yielding cultivars and breeding lines . This included the cultivars 'Haifa' and 'Kopu' which are also sensitive to other forms of stress such as drought (Macfarlane et al. , 1 990) and frost (Caradus et al. , 1989a). Flavonol accumulation in 'Prestige' and 'Syrian' could be related to selection for specific stress tolerance. Low kaempferol but high quercetin accumulation in 'Prestige' could be due to some drought adaptation (Caradus and Woodfield, 1 997). There could also be flavonol-independent defence mechanisms in 'Syrian' which has been selected for leaf rust tolerance (Caradus et al. , 1 990a). Flavonol accumulation was also low in the highly productive population 'Octoploid' . The general picture emerging from these observations is that predominant carbon flow in favour of biomass accumulation could come at some cost for sequestration towards other carbon-requiring processes, such as production of specific flavonoids. There seem to be general advantages for populations from natural environments not selected or bred for productivity but adapted to a variety of stress forms. An explanation for this can be found in ecological models of stress tolerance, proposing distinct plant types depending on primary plant strategies (Grime, 1 979; Poorter and Gamier, 1 999). Such models of plant special isation predict that less productive populations originating in stress-prone habitats would display a higher degree of cellular biochemical stress acclimation, compared to more morphology-based stress responses expected in populations largely selected and bred for high productivity (Grime and Campbell , 1 99 1 ) . Thus the lower plant yield and reduced habitat productivity of the T. repens ecotypes could place them 6. Flavonol responses to UV-B 125 at a particular advantage in regard to the accumulation of specific compounds required in the protection against increased DV -B irradiation. Population differences in the accumulation of quercetin and kaempferol In this study, an inverse relationship was observed between plant productivity and DV­ B-generated accumulation of quercetin (Fig. 6-5), but not kaempferol glycosides. A l imited number of published studies have reported population differences in specific flavonoid responses. Differential increases have been documented for quercetin glycosides, but also for myricetin and for kaempferol glycosides in several birch populations (Lavola, 1 998). Furthermore, population-specific efficiency in DV -B­ generated ortho-dihydroxylated flavonoid accumulation, relative to their monohydroxylated counterparts, has been observed for two cultivars in rice (Markham et al., 1 998b) and in Brassica (Olsson et aI., 1 998). The present study established differential effects of DV -B and increases in flavonol glycosides for an extended range of populations of the leguminous species T. repens. The degree of specific flavonol accumulation in the natural, unselected populations and Q:K ratios between 1 and 2 under both treatment conditions show that they can maintain efficiency in the accumulation of quercetin relative to kaempferol during the accumulation of high flavonol amounts under DV -B. Similar properties could again be observed in the general-purpose cultivar 'Huia'. However, there were differences in these features among the remammg, more specialised and selected populations, accumulating lower amounts of flavonols . The maximal Q:K ratios and minimal kaempferol levels for 'Prestige' , 'Syrian' and 'Kopu' under DV -B indicate that these populations are to some degree able to compensate for the relatively low available flavonol levels with more efficient quercetin glycoside accumulation. While this could be related to selection for specific stress tolerance in 'Syrian' and 'Prestige', it could for 'Kopu' be indicative of the high plasticity it shares with its 3/8-progenitor 'Huia'. Low constitutive quercetin glycoside accumulation similar to 'Kopu' and strong increases in these levels and in the Q:K ratio under DV-B have been reported in Brassica (Olsson et al., 1 998). Particularly low quercetin glycoside levels were found in the highly productive populations 'Octoploid' and 'Haifa', which - in addition to low constitutive Q:K levels - also had minimal Q:K ratios under 6. Flavonol responses to UV-B 1 26 UV -B. It thus appears that in addition to l imited flavonol availability these two populations are further disadvantaged in their quercetin production due to low efficiency of UV -B-generated quercetin glycoside accumulation, relative to kaempferol. Taken together these findings suggest a central role for the dihydroxylated:monohydroxylated flavonoid ratio especially in circumstances of l imited flavonoid accumulation. This conclusion can also be drawn from similar observations in other studies (Markham et al. , 1 998a; Ryan et al. , 1 998). Relationship offlavonol levels to UV-B tolerance In addition to an inverse relationship between quercetin levels and plant productivity (Fig. 6-5), UV -B-sensitivity in the T. repens populations also appears to be l inked to UV -B-induced accumulation of quercetin, rather than kaempferol glycosides (Fig. 6-6). Other reports also link higher B-ring hydroxylation to UV -B tolerance, for instance in rice (Markham et al., 1 998b) and Petunia (Ryan et al. , 1 998). In addition, indirect advantages were noted for high quercetin-containing populations such as the T. rep ens ecotypes 'Tienshan ' , 'Sarikamis' and 'Haggas'. T. repens populations with an ability to accumulate high levels of constitutive quercetin glycosides also displayed higher UV -B­ induced quercetin concentrations. This also suggests a constitutive advantage for quercetin-rich populations, as high final quercetin amounts showed correlation with UV -B tolerance (Fig. 6-6). No such relationships could be observed for kaempferol accumulation. Stress-induced stimulation of specific phenylpropanoid branch pathways has been documented (Dixon and Paiva, 1 995) and the present findings lend support to views attributing quercetin a specific role in stress acclimation (Olsson et al., 1 998). Additional trends observed here - linking UV -B growth response with the capacity to change flavonoid levels - are consistent with other reports of a general advantage for plants able to increase flavonoid levels under UV -B (Lois, 1 994; Murali and Teramura, 1 984; Olsson et al., 1 998). 6. Flavonol responses to UV-B 1 27 Conclusions This study demonstrated flavonoid-specific responses to long-term UV -B irradiation with particular advantages for quercetin glycoside accumulation. Population comparisons show marked intraspecific differences in leaf flavonol accumulation and support the concept that the capacity to specifically accumulate higher levels of quercetin glycosides under UV-B confers a higher degree of UV-B tolerance. The l inkage between ( 1 ) relatively high flavonol (and more specifically quercetin) accumulation, (2) stress-adapted populations of lower constitutive plant productivity, and (3) UV -B tolerance is in accord with current ecological theory on stress tolerance. Stress-adapted ecotypic populations appear particularly advantaged while the inverse correlation between plant productivity and quercetin levels as well as UV -B sensitivity clearly points at disadvantages for the most productive T repens populations. These populations have largely been selected and bred for high productivity and it is possible that this affects other attributes such as the capacity to accumulate specific secondary compounds. An increase in dihydroxylated flavonoid compounds is a well-conserved UV-B response for markedly different plant species. The present study demonstrates that this observation also applies to distinctly different populations within a species. These findings, in conjunction with similar observations in studies of other forms of plant stress further suggest a particular role for ortho-dihydroxylated flavonol glycosides in the biochemical plant strategy as a response to stress . For future UV -B studies this highlights a need for investigations of the chain of events from UV -B treatment via specific flavonol accumulation to possible protective functions. The present findings suggest a requirement to include a wider range of populations in such studies as the quality and quantity of the UV -B response could differ considerably within a species. Field testing could provide further confirmation of the role of ortho-dihydroxylated flavonoids as potential predictors for UV-B sensitivity. 7. UV-8 effects an/aliar chemistry and insect peifarmance 1 28 Chapter 7. Differences in Trifolium repens L. responses to u ltraviolet-B radiation: fol iar chemistry and consequences for two lepidopteran herbivores Bioassay studies with lepidopteran larvae This Chapter has been accepted for publication in Oecoiogia. 7. UV-B effects onfoliar chemistry and insect peiformance 1 29 Abstract White clover growing in New Zealand is experiencing increasing levels of ultraviolet-B (UV -B) radiation as a result of ozone depletion. The study evaluated the effects of UV­ B radiation on foliar chemistry of two populations of white clover (Trifolium repens L.), 'Huia' and 'Tienshan' , and consequences for the performance of armywonns (Spodoptera fitura) and cutworms (Graphania mutans). Plants were grown in controlled environment rooms with and without supplemental UV -B radiation at a dose of 1 3 . 3 kJ m·2 d- 1 , corresponding to a 25% mid-summer ozone depletion above Palmerston North, New Zealand. In both white clover populations UV -B radiation elicited changes in foHar chemistry, including slight increases in nitrogen concentrations and decreases in carbohydrate concentrations. In addition, the popUlation 'Huia' showed decreases in fibre concentrations and marked increases in cyanogenic activity. No change in UV­ absorbing compounds was detected in either population. Long- and short -term feeding trials were conducted to assess dietary effects on insect growth, consumption and food utilisation. Changes in the performance of both insect species were generally small . The most pronounced effect was a 36% reduction in weight of S. litura after two weeks of feeding on 'Huia' grown at high UV, but larval development times were only slightly prolonged and pupal weights were unaffected. S. litura short-term performance was affected by differences in white clover population. Long-term performance of G. mutans was not affected and short-term performance (stadium duration and consumption rate) was only marginally affected by the high UV treatment. It is concluded that the effects of elevated UV -B radiation on white clover plant chemistry can be specific to certain plant populations. The differences in sensitivity of the two generalist insect species suggest that effects may also be specific to certain plant­ herbivore associations . These results indicate that future UV -B herbivory studies should examine genotypic effects in both plants and animals . I ntroduction Recent changes m DV-B radiation due to atmospheric ozone depletion are widely recognised as having significant consequences for plant biochemistry, physiology, and productivity (Caldwell et al. , 1 998). DV -B-mediated changes in the quantity and quality of plant components are also likely to affect heterotrophs, via trophic interactions and 7. UV-B effects on foliar chemistry and insect peiformance 1 30 cascades (Mazza et al., 1 999). Exceedingly few studies, however, have addressed the impact of UV -B on such interactions (Paul et al., 1 997). The consequences of high UV -B irradiances for consumer organisms are of particular interest in New Zealand, where UV -B doses are markedly higher than at comparable latitudes in the northern hemisphere (Seckmeyer and McKenzie, 1 992). This is due to hemispheric differences in stratospheric and tropospheric (pollutant) ozone, differential aerosol loading, and because the distance between the Sun and Earth is minimal during the Austral summer (Seckmeyer and McKenzie, 1 992). Furthermore, while studies have demonstrated UV-B sensitivity for several legumes (e.g. Krupa et al. , 1 998b; Teramura, 1 990) only limited research has been conducted on UV -B sensitivity of white clover (Trifolium rep ens L.) (Matthew et al. , 1 996; Rozema et al. , 1 997a). White clover is the dominant legume of the pasture ecosystems in New Zealand, performing a vital role for ecosystem processes as a primary source of soil nitrogen and also as a highly nutritious component of the diet for grazing ruminants (Caradus et al. , 1 996). The study investigated the impact of UV -B radiation on the chemical composition of two populations of white clover, 'Huia' and 'Tienshan', and subsequent effects on the performance of two phytophagous insects, Spodoptera litura (Lepidoptera: Noctuidae) and Graphania mutans (Lepidoptera: Noctuidae). The white clover population 'Huia' is a commonly used New Zealand cultivar, whereas the 'Tienshan' population is a high­ altitude ecotype from northwest China. The generalist tropical armyworm, S. litura, became established in northern New Zealand in the mid- 1970s. It has been known to cause severe, though localised, defoliation of clover (Chapman, 1 984). The generalist cutworm, G. mutans, is native to New Zealand and causes damage to pasture, vegetable and fruit crops (McGee, 1 987; McGregor et aL. , 1 987). A small but growing number of studies have assessed the effects of UV-B-mediated changes in plant chemistry on insects. Some studies suggest that UV-B-induced effects may alter herbivory through changes in plant primary metabolism, e.g. increased nitrogen levels (Hatcher and Paul, 1 994) and decreased sugar concentrations (Yazawa et aL., 1 992). Other research points to the involvement of secondary metabolites such as flavonoids and related compounds of the shikimic acid pathway, (e.g. Grant-Petersson and Renwick, 1 996). 7. UV-B effects onfoljar chemistry and insect performance 1 3 1 Plant chemistry linked to primary metabolism and of relevance to insect feeding was analysed. The study also examined aspects of plant secondary metabolism, including levels of UV -absorbing compounds and cyanogenesis, a mechanism present in several populations of white clover and more than 2000 other angiosperm species (Kakes, 1 990). Hydrogen cyanide (HCN) production eventuates when cyanide-containing glucosides (linamarin and lotaustralin) and the hydrolytic �-glucosidase linamarase, normally separated by compartmentalisation, are brought together during lesion of plant tissues. White clover is known to show increases in cyanogenesis as a common response to several forms of stress, including herbivory, frost and drought (Caradus et al. , 1 990b; Williams, 1 987a). UV -B could possibly alter this potential defence pathway by a general stimulatory effect on secondary metabolism (Rozema et al. , 1 997b). In this study, it was sought to determine whether UV -B would affect key attributes of white clover chemical composition, and whether the two white clover populations would be differentially affected. It was also predicted that performance of S. litura and G. mutans would decline when reared on foliage of white clover grown under high levels of UV-B, and that the magnitude of this effect would vary between white clover populations. Materials and methods Experimental design The overall design was a 2 x 2 factorial, with two levels of UV -B radiation and two populations of white clover. Trolleys containing the plant material were rotated regularly within and between the two experimental rooms in order to equalise potential location effects. In a separate room maintained with no UV supplementation, feeding experiments were first conducted with S. fitura, then repeated with G. mutans. Plant cultivation and UV irradiation Stolon cuttings of 'Huia' and 'Tienshan' white clover populations were planted into separate 3 litre pots containing 3 kg of soil (Karapoti brown sandy loam). The soil was 7. UV-8 effects on foliar chemistry and insect peiformance 1 32 supplemented with 2 g of l ime and 10 g of slow-release fertiliser ( 1 5-4- 1 0 N-P-K plus micronutrients). Plants were watered daily by an automatic drip irrigation system. Relative humidity in the growth rooms was 70% and day/night temperatures were 24°C and 1 8°C, respectively. Daylength was 14 hours and mean photosynthetic photon flux (PPF) was 420 !lmol m-2 S- I , supplied by four I kW Sylvania 'metal-arc' high pressure discharge lamps, together with four 1 kW Phi lips tungsten iodide lamps. PPF was measured using a U-COR 1 85 meter with a U- 1 90S quantum sensor. Elevated UV-B radiation was supplied by Philips TL 40W/1 2 RS fluorescent UV tubes, enclosed in cellulose diacetate filters. B iologically effective UV -B levels were 1 3 .3 kJ m -2 d- I , calculated from the generalised plant action spectrum (Caldwell, 197 1 ) and normalised to 300 nm. The UV treatment started 1 h after onset of the l ight period and ended 1 h before darkness. A feedback control system continuously monitored and adjusted the output of the UV lamps in response to degradation of the filters and ageing of the tubes in order to maintain the set UV -B level. The system incorporated a Solar Light Company UV -B biometer to measure incident radiation. This sensor was regularly calibrated against the generalised plant action spectrum (Caldwell, 1 97 1 ) using a Bentham DM1 50 UV spectroradiometer. In the control room, Mylar filter sheets were used to screen out UV-B but maintain UV-A levels simHar to those in the UV-B treatment. The plants were subjected to the UV treatments for 29 days prior to initiation of insect bioassays. Sampling procedure for plant nutrient analyses Leaf samples for chemical analyses were collected from five pairs of pots within each treatment. Fully open trifoliate laminae were frozen in liquid nitrogen and freeze-dried. Samples were then finely ground in a coffee grinder and stored at -20 °C until chemically analysed. Nitrogen Samples containing 1 0 mg freeze-dried material were automatical ly loaded into a Carlo­ Erba Instruments NA 1 500 Series II Analyser to determine nitrogen levels (Carlo-Erba, 1 988). The instrument detects N2 by thermal conductivity after flash combustion 7. UV-B effects onfo/iar chemistry and insect performance 1 33 (around 1000°C) of the plant material. The resulting electrical signal is then converted to % N2 present in the sample using internal calibration standards (atropine, phenantrene ). A vailable carbohydrates A vailable carbohydrates (AC) in the white clover samples were determined as reducing sugars using the p-hydroxybenzoic acid hydrazide (PAHBAH) procedure, yielding soluble carbohydrates and starch (Blakeney and Mutton, 1 980; Southgate, 1 99 1 ). Freeze-dried samples ( 100 mg) were solubilised with 0.4 ml dimethyl sulfoxide (DMSO). After boiling in DMSO for 30 min, the carbohydrates in the solution were exposed to enzyme digestion, converting polysaccharides into simpler sugars. 1 .6 ml amylase solution (Southgate, 1 99 1 ) was added and the solution boiled for another 1 5 min. This was followed by 30 min incubation (50°C) in 1 00 �l amyloglucosidase solution (Southgate, 199 1 ). A glucose standard solution (5 mg mr ! ) was prepared and assayed alongside the white clover sample batches to determine the percentage of AC in the freeze-dried material . The sample digests were prepared for colorimetric determination using standard procedures (Blakeney and Mutton, 1 980), using double precipitation in 400 �l barium hydroxide and 1 00 III zinc sulphate solution, fol lowed by precipitation for 30 min in 7 .5 ml cold ethanol (absolute). After centrifugation, 1 00 �l of the supernatant was subjected to acid hydrolysis (200 �l of 0.5 M HCI) in a boiling water bath for 30 min. This was followed by addition of 2 ml PAHBAH solution (0.5 g 1 00 mr ! ), vortex mixing and further boiling for 6 min. The tubes were then cooled and 4 ml water was added. Absorbance of the solution was measured as reducing sugars at 420 nm (Blakeney and Mutton, 1980) . Neutral detergent fibre Fibre content was measured with the neutral detergent fibre (NDF) procedure (Southgate, 1 99 1 ; van Soest et al. , 1 99 1 ). Samples of 100 mg freeze-dried material were boiled in 1 0 ml neutral detergent solution (Southgate, 1 99 1 ) , centrifuged at 2000 g and the residue digested at 37°C for 1 8 h with 5 ml enzyme solution (2.5% a-amylase [w/v] in 0. 1 M phosphate buffer, pH 7) (van Soest et al. , 1 99 1 ). These procedures remove 7. UV-8 effects on f()/iar chemistry and insect performance 1 34 pectins and plant cell contents (proteins, sugars and lipids), leaving a fibrous residue (a NDF), consisting primarily of plant cell wall components (hemicellulose, cellulose and lignin). Centrifugation was followed by several washing steps of the residue in water ( lO ml), followed by washing in 96% ethanol and lastly in acetone. Following evaporation and desiccation, the resulting final mass of the residue corresponds to the fibre content (%NDF) in the sample (Southgate, 1 99 1 ). Cyanogenesis Cyanide (HCN) production of fully unfolded young trifoliate laminae was examined in a two-step procedure using a modified picrate paper test (Williams et al. , 1 998). The picrate paper technique is a semi-quantitative method involving scoring of colour intensity by eye, which gives satisfactory correlation with quantitative procedures (Melville et al. , 1 940) . In the first assay, a leaflet from each pot was tested for HCN production with picric acid as indicator. After addition of 10 f..l\ toluene and overnight incubation at 37°C, the colour change in picric acid was examined by comparison with a standard colour scale, ranging from 0 (unchanged, yellow) to 5 (dark orange) (Corkill, 1 940). Cyanogenesis intensity per leaf was expressed on a dry weight basis. Plants that did not prove cyanogenic in this first test were then examined in a second assay for the presence of substrates (linamarin and lotaustralin). The same picric acid procedure was used, but included addition of 20 f..ll of 0.5% l inamarase (a glucosidase enzyme) to liberate HCN (Williams et al. , 1 998). This amount was used so as to assume that enzyme availability did not hmit the reaction. UV-absorbing compounds Standard protocols were used to estimate levels of methanol-extractable DV -absorbing compounds (Mirecki and Teramura, 1 984). For each sample, four fully unfolded young trifoliate laminae were oven-dried and 1 5-20 mg of ground material was weighed into Eppendorf tubes. The extraction was performed with 1 .2 ml MeOH:H20:HCl (79:20: 1 ), followed by spectrophotometric analysis. Absorbance readings at 300 nm were calculated on the basis of leaf dry weight. This procedure was repeated on seven replicates for each white clover population under each DV-B treatment. 7. UV-8 effects on foliar chemistry and insect performance 1 35 Spodoptera l i tura hioassays S. filum egg masses were provided by Biodiscovery of Auckland, New Zealand. All rearing was conducted in a controlled environment room similar to those used for the UV treatments, at 24: 1 8°C, with a 14 h: 10 h l ight-dark cycle . A long-term feeding trial was used to assess the effects of UV -B-mediated changes in white clover composition on insect growth, development, and pupal weight. Groups of 1 2- 1 4 neonate larvae were apportioned to 9 cm plastic petri dishes, one dish for each of seven pots per white clover population per UV treatment. Trifoliate leaves were clipped from potted plants and petioles inserted into 1 .5 ml microcentrifuge tubes containing water. Leaves were placed into the petri dishes with larvae and replaced every 1 -3 days. At one week of age, larvae showed signs of cannibalism, so seven surviving larvae were randomly selected from each petri dish and placed individually into dishes. Larval weights were recorded weekly until the onset of pupation. Upon pupation, pupal weight, sex, and development time (egg hatch to pupation) were recorded. A short-term feeding trial was conducted with penultimate (fourth) instars to determine the effects of experimental treatments on larval feeding, growth rates, and food processing efficiencies. Penultimate instars were used because the insects were large enough to accurately measure consumption rates, and because the beginning and end of the penultimate stadium is demarcated by clearly defined moults. Insects were reared on white clover population 'Kopu' (grown in the low UV room) from egg hatch through the third stadium. Upon moulting into the fourth stadium, larvae were weighed and placed individually into 9 cm plastic petri dishes containing a weighed white clover trifoliate. Leaves were kept hydrated and replaced as described previously. Two insects were assayed for each of seven pots per white clover popUlation and UV treatment. Upon moulting into the fifth stadium, larvae were frozen, then larvae, frass, and uneaten l eaf tissue were freeze-dried and weighed. Calculation of average growth rate (AGR), average consumption rate (ACR), approximate digestibility (AD), and efficiency of conversion of ingested food (Eel) were done according to standard formulas (Waldbauer, 1 968). Initial dry weights of larvae were estimated from a wet:dry conversion factor derived from a sample of 1 5 newly moulted fourth instars. Simi lar 7. UV-B effects onfoliar chemistry and insect performance 1 36 conversion factors for leaves were obtained from leaf samples collected for chemical analyses. Graphania mutans bioassays G. mutans egg masses were obtained from gravid females light-trapped in the vicinity of Palmerston North, New Zealand. Bioassays were conducted following completion of the S. litura feeding assays, using the same procedures except as follows. For the long­ term feeding trial , groups of 1 2- 1 5 larvae were apportioned to 9 cm petri dishes. Larvae were transferred to 1 5 cm petri dishes as third instars. When larvae reached the fifth stadium the number was reduced to six per container and transferred to 1 L ventilated plastic boxes. For the short-term feeding assay, penultimate instar larvae were in the fifth stadium. One insect was assayed for each of nine pots per white clover population and UV treatment. Statistical analysis Effects of DV -B irradiation and white clover popUlation were examined usmg the General Linear Model procedure in SAS, release 6. 1 2 (SAS, 1 996) . For short-term feeding trials such as those employed, differences in insect performance variables may result from differences in initial larval weight. Analysis of variance revealed no significant differences in initial larval weight for insects in various treatments. Moreover, analysis of covariance (SAS, 1 996) revealed that initial larval weight was not a significant covariate for the insect performance variables reported here as showing statisticall y significant effects. 7. UV-B effects onfo/iar chemistry and insect performance 1 37 Results F oliar chemistry White clover chemical composition varied in response to UV treatment and between populations. Concentrations of available carbohydrates declined an average of 22% in high-UV clover, but did not differ between populations (Fig. 7- 1 ). Levels of neutral detergent fibre (NDF) declined 1 4% in high-UV 'Huia', but were unaffected in Tienshan' (significant UV x population interaction) (Fig. 7- 1 ). Levels of nitrogen tended to increase in UV -B-treated plants and the amount of nitrogen was slightly, but significantly, higher in 'Huia' than in Tienshan' (Fig. 7- 1 ) . While UV-B did not change the proportion of acyanogenic to cyanogenic 'Huia' plants ( 1 5 plants each), cyanogenic activity showed a strong, 50% increase in response to UV-B treatment in cyanogenic 'Huia' (Table 7- 1 ). The first cyanogenesis assay indicated that the Tienshan' population was acyanogenic. Application of linamarase to the acyanogenic plant tissues indicated that acyanogenesis could be due to deficiencies in the enzymatic steps of cyanide production, rather than to absence of substrates. The low levels of linamarase-catalysed cyanide production in Tienshan' increased under high UV, but this effect was only half as pronounced as that exhibited by acyanogenic 'Huia' plants supplied with l inamarase (Table 7- 1 ). The UV-B-induced cyanide levels in Tienshan' were the same as the constitutive, linamarase-catalysed cyanide levels in 'Huia'. Levels of UV -absorbing methanol-extractable compounds did not change in response to UV treatment (Table 7- 1 ) . 7. UV-B effects onfoliar chemistry and insect peiformance -..... 6 .t:. Cl 5 '0> � > 4 ... "0 '#- 3 - c: � 2 o ... ;t::: Z 1 o 1 6 -..... .t:. ,21 1 2 4) � � "0 8 '#-- o et 4 o 1 2 -..... .t:. Cl '0> � 8 > ... "0 � 0 - LL 4 a z o T � Huia Tienshan pvalues: UV·B .012 Popul. .001 UV·B x Popul. . 144 pvalues: UV·B <.001 Popul. .561 UV·B x Popul. .406 Pvalues: UV·B .066 Popul. .006 UV·B x Popul . . 048 1 3 8 Fig. 7-1. Foliar chemical composition (mean + 1 SE) of two populations of white clover grown under low and high DV radiation. AC = available carbohydrates; NDF = neutral detergent fibre. 7. UV-8 effects onfoliar chemistry and insect performance 1 39 Table 7-1. Effects of UV treatment on cyanogenesis (HCN score mg- 1 leaf dry matter) and UV -absorbing compounds (A30o mg- I leaf dry matter) in white clover (mean ± 1 SE) Treatment Clover population UV - Huia UV - Tienshan UV + Huia UV + Tienshan a I HCN score mg- 0.565 ± 0.053 ndb 0.850 ± 0.097 nd Unamarase-catalysed c I HCN score mg- 0.350 ± 0.050 0.256 ± 0.030 0.6 1 9 ± 0.047 0.358 ± 0.03 1 A d - I 300 mg 1 .088 ± 0.037 1 .089 ± 0.042 1 .090 ± 0.05 1 1 .089 ± 0.042 a Hest revealed a significant effect of UV treatment on 'Huia' foliage, P = 0.0 1 7. b nd = not detectable. C ANOVA revealed effects of treatment and population (both P < 0.00 1 ) and of treatment x population (P = 0.043), on the linamarase-catalysed HCN score. d ANOV A revealed no significant effects of UV or population on UV -absorbing compounds. Spodoptera l itura performance The long-term bioassays showed that growth of larvae reared on high UV -treated foliage was reduced for insects fed the white clover population 'Huia' (Fig. 7-2). At two weeks of age, body weights of larvae in the high UV 'Huia' treatment were only 64% of those of larvae in the low UV 'Huia' treatment. In contrast, growth of larvae fed population 'Tienshan' over the same period was unaffected by DV treatment. No significant UV -B effect was observed on pupal weights, and only a slight prolongation of developmental time, especially in females (Fig. 7-3). White clover population only slightly affected larval development times, with insects fed 'Huia' requiring an additional day (relative to insects fed 'Tienshan') before the onset of pupation. Final pupal weights of males and females did not differ between white clover populations. Short-term (fourth instar) bioassays revealed few if any effects of UV treatment on S. litura performance (Fig. 7-4 and Fig. 7-5). Approximate digestibility (AD) tended to improve in insects fed high-UV foliage, but the difference was slight and only marginally significant. Effects of white clover population were more pronounced. Insects fed 'Tienshan' had 20% higher consumption rates (ACR) but 1 0% lower food conversion efficiencies (ECI), compared with insects fed 'Huia'. Consequently, these insects exhibited a marginally significant, 1 0% increase in growth rates. 7. U\I-B effects on foliar chemistr}' and insect pelj(Jrlnance 400 - en 300 E - .., '§, 200 .- Cl) � 1 00 400 - en 300 E - .., .r:. en 200 .- Cl) � 1 00 o S. litura G. mutans 5 • -o- uv- Huia -l::r-UV- Tienshan · · . · · UV+ Huia . ... . . UV+ Tienshan 1 0 1 5 Age (days) 20 1 40 Fig. 7-2. Growth of Spodoptera fitura and Graphania mutans l arvae between egg hatch and pupation. Data collection at weekly intervals was terminated when the first larva in a treatment pupated. Vertical lines indicate ± 1 SE. (j) >-� 20 -- .5 1 5 - C � 10 C. o � 5 � o +I---"-----LJ D> E 300 ...... -.s:::. Cl '(ij 200 � ii g. 100 Cl. o t Huia Female D UV· . UV+ Tienshan P values: UV-B .039 Popul. .054 UV-B x Popul. .21 0 P values: UV-B .1 1 5 Popul. .584 UV-B x Popul. .91 6 20 15 10 5 o + __ , _ _ L--J __ 300 200 100 O j��� --L--_.L Huia Male Tienshan P values: UV-B . 179 PopuL .055 UV-B x Popul. .250 P values: UV-B .703 Popul. .331 UV·B x Popul. .474 Fig. 7-3. Development times and pupal weights (mean + 1 SE) of Spodoptera litura reared on two populations of white clover grown under low and high UV radiation. ,...... .j::;. ...... 3 en >- ca � 2 c 0 ;::: ca .... :::J 1 c S. litura ouv- I -UV' I 1 - 1 - ! , ! UL LJL! i : o L- ,- - i ..-;- 1.5 1 n. 1 . I. >- , ca I -c Cl .§. 1 .0 ex: ­+>- 7. UV-B effects on fo/iar chemistr}' and insect petiormance 1 45 Graphania mu tans peiforrnance Over the course of the three week period spanning egg hatch to onset of pupation, growth of G. mutans larvae was nearly identical on the four experimental diets (Fig. 7-2). Moreover, neither UV treatment nor white clover population influenced larval developmental periods or final pupal weights (Fig. 7-6). Relative to S. fitura, G. mutans required an additional week for larval development and attained smaller pupal weights. The short-term (fifth instal') feeding trials provided results similar to those of the long­ term trials . Neither UV treatment nor white clover population affected insect growth rates or food utilisation efficiencies (Fig. 7-4 and Fig. 7-5). Trends toward longer development times and reduced consumption rates for insects on high-DV white clover were only marginally significant. Discussion Foliar chemistry related to primary rnetabolism In this experiment, UV -B radiation elicited alterations in white clover composition that could be expected to affect the performance of herbivores. The largest effect was a decrease in available carbohydrate levels by more than 20% in both white clover populations (Fig. 7- 1 ) . This change was accompanied by decreased fibre levels in 'Huia'. To date, few studies have evaluated the effects of UV -B radiation on plant carbohydrate or fibre content. Decreased available carbohydrate concentrations at elevated UV-B levels have been reported for glucose in pea (Mackerness et al., 1 997), sucrose in mulberry (Yazawa et al., 1 992), and soluble carbohydrates in Pinus (Katzel et al. , 1 996). In leaf l itter of Vaccinium uliginosum, no effect of UV -B dose was observed on soluble carbohydrates, but a decrease in cellulose concentrations (Gehrke et al. , 1 995). However, no effect of UV -B radiation was observed on hemicellulose content of Gunnera magellanica (Rousseaux et aI. , 1 998). Overall levels of nitrogen in both white clover populations were quite high, well above the limiting threshold for most insects (Mattson, 1 980; Slansky and Scriber, 1 985). The slight increase in nitrogen concentrations in 'Huia' could therefore be regarded as less 7. UV-B effects on f()liar chemistry and insect performance 1 46 important for herbivory than the aforementioned chemical changes, and the effect in Tienshan' was negligible (Fig. 7- 1 ) . Shifts of comparable magnitude have been reported for other legume species, such as soybean (Murali and Teramura, 1985a) and pea (Hatcher and Paul, 1 994). FoliaI' chemistry related to secondary metabolism Together with the findings presented in Chapter 1 (Fig. 1 - 1 ), this is the first study to assess the impact of UV-B radiation on cyanogenesis. A clear UV-B-induced increase in cyanogenesis levels was found in cyanogenic plants (Table 7- 1 ). The findings from linamarase supplementation indicate that UV -B-induced increases in cyanogenesis could primarily be due to increased substrate biosynthesis and are less dependent on increases in �-glucosidase concentration/activity. The possibility of UV -induced increases in white clover cyanogenic glycosides is notable given that these compounds are not derived from aromatic amino acids. The latter are products of the shikimic acid pathway, well known to be induced by UV radiation (Rozema et al., 1 997b). Linamarin and lotaustralin, however, are derived from the aliphatic amino acids valine and isoleucine (Seigler, 1 998). Thus, their accumulation is mediated by the effects of UV-B on biochemical mechanisms other than those of the shikimic acid pathway. Interestingly, these findings suggest that in acyanogenic white clover genotypes, cyanoglucoside levels may rise under enhanced UV -B as well (Table 7- 1 ). The cyanogenesis levels in 'Huia' can be classified as moderate-high (Caradus and Woodfield, 1 997; Crush and Caradus, 1 995). The apparent acyanogenic nature of Tienshan' is consistent with its origin in a cold habitat, where frequent frost-related lesions of the tissue would disadvantage cyanogenic plants. The low linamarase­ catalysed cyanide levels in Tienshan' could suggest lower constitutive levels of available cyanogenic substrates , which would be consistent with findings from other cyanogenesis studies on white clover populations originating in colder climates (Caradus and Eerens, 1 992; Daday, 1 965). Foliar concentrations of flavonoids and related UV -absorbing compounds typically increase in response to UV -B radiation (Bornman et al., 1 997; Rozema et al., 1 997b). In contrast, no change was found in the UV -absorbing capacity of white clover grown 7. UV-B effects onfoliar chemistry and insect performance 1 47 under elevated UV -8 radiation (Table 7- 1 ). This result is not without precedent, as several other recent studies (e.g. Rousseaux et al., 1 998; Salt et al. , 1 998) have also failed to detect such a response. Moreover, the possibility cannot be ruled out that concentrations of some individual tlavonoids or other phenolics changed, while overall UV -absorbing capacity remained unchanged. Another study in white clover revealed strongly increased tlavonol levels under UV -B, concomitant with much less pronounced changes in UV -absorbing compounds (Chapter 5, Fig. 5 - 1 and Fig. 5 -2). In summary, results from the chemical analyses lend support to the pattern observed by others (e.g. Lavola et al., 1 998; Paul et al., 1 997), that phytochemical responses to UV­ B radiation are compound-, species-, and population-specific. Such variability can be expected to influence the interactions of plants with higher trophic levels. Insect Responses Results from this research demonstrate that insect responses to UV -B-mediated changes in white clover chemical composition are specific to the particular plant-herbivore interaction investigated. The most pronounced effect observed was for growth of S. lilura during the first two weeks of the larval developmental period (Fig. 7-2). Ultimately, however, UV -B-treated white clover had only a slight effect on development time of female S. litura and no effect on pupal weights of either insect species (Fig. 7-3 and Fig. 7-6). Because the fecundity of female Lepidoptera is generally correlated with pupal weight, the results suggest that consumption of UV -B-irradiated white clover is unlikely to affect reproductive success in S. filura or G. mutans. However, these experiments focussed on the first generation and thus possible cumulative effects over several generations cannot be ruled out. The cause of decreased growth in young S. filura larvae reared on high-UV 'Huia' is unclear, but may be related to high cyanogenic activity. Cyanogenesis was elevated in high-UV 'Huia' (Table 7- 1 ) and only for this treatment was larval performance reduced. Although cyanide is a broadly toxic constituent, some insects exhibit metabolic adaptations (e.g., B-cyanoalanine synthase activity) against it (Lindroth, 1 99 1 ). In a related noctuid species, S. eridania, it was found that cyanide stimulated larval feeding 7. UV-B e[feets onfoliar chemistrv and insect performance 1 48 and growth (Brattsen et al. , 1 983). Other findings suggest that cyanogenic glycosides in white clover can act as feeding deterrents to insect pests (Ellsbury et al. , 1 992). In contrast to insects feeding on cyanogenic 'Huia' plants, a cyanogenic effect for insects feeding on Tienshan' would only be possible in the presence of both appropriate amounts of cyanogenic plant substrates and of suitable p-glucosidases in the insect gut. While the presence of the latter has been reported for several lepidopteran species (Ferreira et al. , 1 998), the low levels of l inamarase-catalysed cyanide production in Tienshan' under high UV (Table 7- ] ) could suggest that the lack of insect response was due to insufficient amounts of available cyanogenic substrates. For the population 'Huia' it is proposed that S. filura required several weeks to habituate to higher levels of cyanogenic activity in high UV, after which growth accelerated and compensated for earlier growth suppression. G. mutans may not have exhibited reduced growth because it is better adapted than S. litura to cyanogenic compounds in its diet. G. mutans larvae are noted apple pests in New Zealand (Burnip et al., 1 995) and apple (Malus pumilus) produces cyanogenic glycosides such as amygdalin (Conn, 1 979). The lack of pronounced effects on insect performance in these studies could also be attributed to the overall excellent nutritional quality of the white clover diet, which exhibited relatively high nutrient concentrations and low fibre levels across all treatments, compared with levels for this species under field conditions (Rattray and Joyce, 1 974; Wanjaiya et al., 1 993). It remains to be investigated whether significant UV -B-induced changes persist in the field, where compounds of relevance to nutrition are l ikely to occur at more l imiting levels for herbivore growth. In addition, several white clover populations used in pasture ecosystems have higher constitutive levels of cyanogenesis compared to 'Huia' (ClUsh and Cm'adus, 1 995) and a further UV-B­ induced increase could be of relevance for insect feeding as well as mammalian herbivory. Consideration of ruminant sensitivity to cyanogenesis (e.g. Lehmann et al., 1 99 1 ) indicates a need for field experimentation under grazing conditions and concomitant inclusion of other stress forms (e.g., drought) which are l ikely to influence cyanogenesis and other aspects of white clover nutritive composition, and in turn nutrient utilisation by animals (Wanjaiya et al. , 1 993). 7. UV-B effects on foliar chemistrv and insect performance 149 Relatively few experiments have investigated the plant-mediated effects of UV radiation on insects, and results have varied among species and studies. With respect to consumption, no effect was found on feeding rates of S. litura, and only a slight trend toward reduced feeding in G. mutans. (Fig. 7-4 and Fig. 7-5) Other studies have documented the complete range of potential feeding responses to high-UV radiation, including decreases (Grant-Petersson and Renwick, 1 996; Hatcher and Paul, 1 994; Rousseaux et al., 1 998), no change (Grant-Petersson and Renwick, 1 996), and increases (Lavola et al., 1998). In terms of insect growth performance, little if any effect was found of high-UV radiation. Other researchers have reported reduced growth (Grant­ Petersson and Renwick, 1996; McCloud and Berenbaum, 1 994), no change (Grant­ Petersson and Renwick, 1996; Stout et al., 1 998) and increased growth (Hatcher and Paul, 1 994) . It is concluded that generalisations about the implications of higher UV -B radiation levels for plant biochemistry and herbivory must take account of genotypic differences in both plant and animal populations. The study shows that important differences may occur between populations of the same plant species in the biochemical response to UV­ B radiation and in the subsequent effects on herbivore performance. Further research is required to understand the underlying functional basis for differential responses in both plants and animals given that UV-B effects on plant-insect interactions are species- and population-specific. These data suggest that a scheme for classifying the functional bases of plant-herbivore interactions should, amongst other attributes, take into account cyanogenic capacity, differential climatic adaptation and stress tolerance of plants. 8. UV-8 responses in frequently defoliated white clover 1 50 Chapter 8. Responses to u ltraviolet-B radiation in frequently defol iated Trifolium repens populations : differential growth, photochemistry and anatomy White clover leaf structure: adaxial epidermal cell with underl ying palisade mesophyl l This Chapter has been submitted to Physiologia Plantarum. 8. UV-B responses in frequently defoliated white clover 1 5 1 Abstract This study examined responses to DV -B in two white clover populations exposed to frequent defoliation and 1 6 weeks supplementation with or without ultraviolet-B radiation (DV -B) of 1 3 .3 kJ m-2 do l under controlled environmental conditions. Investigations included examinations of a number of leaf morphological characteristics, several chlorophyll fluorescence parameters as well as aspects of internal leaf structure and ultrastructure. In particular, it was aimed to explore whether possible differences in these features between the two white clover populations could be related to their DV-B sensitivity. Leaf growth decreased by 1 6% in the DV -B-sensitive cultivar 'Huia' under DV -B, while the tolerant ecotype 'Tienshan' was unaffected. DV -B effects on leaf dry mass and leaf size were similar, leading to no change of specific leaf mass (SLM). Percentage leaf dry mass (PDM) was reduced by 8% across populations in response to DV -B . Chlorophyll fluorescence measurements showed no photochemical sensitivity to DV -B. Results from light- and electron microscopy suggest some regulatory, rather than destructive DV -B effects on leaf ultrastructure. While the integrity of epidermal and mesophyll cells and of chloroplasts was maintained, it appeared that there was a decrease in overall starch content. There were also structural differences between the two white clover populations, suggesting higher, more densely packed dome-shaped epidermal cells for 'Tienshan'. It is concluded that differential UV -B-sensitivity in white clover populations can also occur under frequent defoliation. UV -B tolerance could be related to structural epidermal characteristics inherent to the slower-growing stress­ tolerant population. Introduction Increased ultraviolet-B radiation (DV-B, 290-3 1 5 nm) resulting from depletion of the stratospheric ozone layer can influence plant processes either through direct damage or via various regulatory effects (Rozema et al., 1 999). The latter includes UV -B­ generated morphological changes such as smaller leaf size (Matthew et al. , 1 996) decreased leaf weight (Krizek et al. , 1 998) as well as differential DV-B effects on these two attributes, yielding higher specific leaf mass (leaf dry mass based on leaf size, SLM) (Filella and Penuelas, 1 999). There are also indications that leaf dry mass based on leaf fresh mass (percentage leaf dry matter, PDM) could be affected by DV-B (Wand, 1 995). 8. UV-B responses infrequent!.v deloliated white clover 1 52 Direct UV -B-induced plant damage has often been observed in alterations of plant photochemistry, particularly photosystem II (PSlI) function, measured by chlorophyll fluorescence. This includes UV -B-generated decreases in the intrinsic efficiency of PS II (FvlFm) (Nikolopoulos et al., 1 995), reduced photochemical yield (I1F/F�m) (Reuber et al., 1 996a) and decreased photochemical quenching (qp) (Reuber et al. , 1 996b ). There are also reports of altered non-photochemical quenching (NPQ) in response to UV-B (Reuber et al. , 1 996a). Regulatory functions of UV -B for plant structure include increased number of cell layers (Nagel et al., 1 998), epidermal thickening (Hunt and McNeil , 1 998) as well as palisade cell elongation (Cen and BOI-nman , 1 990; Nagel et al., 1 998). Other studies found a general UV -B-induced decrease in epidermal cell surface size and that small constitutive cell size (i .e. measured under no UV -B supplementation) was l inked to UV­ B tolerance (Chapter 2, Fig. 2-2b). DV -B-induced damage symptoms on a structural level include collapsed epidermal cells (Cen and Bornman, 1 990; Santos et al., 1 993) and intracellular disintegration (Santos et al., 1 993). Disruption of chloroplast ultrastructure has also been described as a main target of DV -B irradiation in several studies (e.g. Caasi-Lit et al. , 1 997; He et al., 1 994). The pasture legume white clover is frequently exposed to defoliation pressure from grazing animals . Little is known on the effectiveness of DV -B on plants exposed to continuous defoliation. Frequent defoliation stress can affect various plant functions, for instance l ight interception, stolon and root development or carbohydrate accumulation (Hart, 1 987). A field study in a New Zealand sward including white clover showed DV­ B-induced reductions in pasture growth under sheep grazing. In other controlled environment studies (Chapter 3, Fig. 3-2a) differential UV -B effects were observed in white populations subjected to infrequent defoliation. These findings provided a framework suggesting a fundamental relationship between DV -B responsiveness one the one hand and the productivity of white clover populations and their habitat on the other. 8. UV-B responses infrequentlv defoliated while clover 1 53 The present study aims at further expanding this framework by examining the two white clover populations 'Huia' and 'Tienshan' after frequent defoliation during 16 weeks growth under UV -B. 'Huia' is a commonly used white clover cultivar bred for agricultural productivity, while 'Tienshan' is a slow-growing ecotype adapted to multiple stresses. Morphological, photochemical and structural attributes and their responses to UV -B were examined. In particular, this study intended to explore whether possible between-population differences in these features could be l inked to differential UV -B sensitivity. Materials and Methods Experimental design, plant cultivation and UV irradiation treatments were identical to those described in Chapter 7. The plants were subjected to the UV treatments for 29 days prior to initiation of defoliation pressure. From then on, all plants experienced frequent periods of defoliation during the collection of l eaf material used for bioassays as described in Chapter 7. Plants were allowed to recover from defoliation stress for 1 0 days before leaves were harvested for morphological , photochemical and structural examinations. Leaf material Leaf size was measured at the end of the experiment with a LI-COR Model 3 100 area meter on four fully open distal leaves per plant. Laminae were subsequently dried at 80°C for 48 h for dry mass determination. The ratio of dry mass over leaf area was calculated to give specific leaf mass (SLM), while leaf dry mass percentage (PDM) was calculated from (dry mass/fresh mass)* 100. These measurements were conducted on ten plants (replicates) for each white clover population under each UV treatment. Chlorophyll fluorescence Chlorophyll fluorescence measurements were performed on fully open distal leaflets of 5 replicate samples for each population under each treatment condition using the same methods as described in Chapter 5 . Effects of UV -B irradiation and white clover 8. UV-B responses in frequently defoliated white clover 1 54 population on leaf attributes and chlorophyll fluorescence parameters were examined with the General Analysis of Variance procedure in GENSTA T (Genstat, 1 993). Structural studies In order to explore possible structural differences in response to UV -B, duplicate leaf samples per population under each treatment condition were prepared as follows. Strips ( 1 mm wide) were cut perpendicular to the main leaf vein from the central part of the middle leaflet of fully open distal white clover leaves. Sections were fixed in 3% glutaraldehyde + 2% formaldehyde in O. lM phosphate buffer (pH 7.2) (Karnovsky, 1 965). After vacuum infi ltration, the sections were kept in the fixative for 24 hours at 4°C. This was followed by three washes in fresh buffer for 30 min. The samples were subsequently post-fixed for 1 hour at room temperature in 1 % osmium tetroxide in the same buffer. Following three further buffer washes the specimens were dehydrated in an acetone series (25 , 50, 75, 95, 1 00%), infiltrated and embedded in Procure 8 1 2 epoxy resin (cured at 60°C for 48 hours). For light microscopy, sections of 1 .0 /lm thickness were cut with a diamond knife in an ultra-microtome (Reichert-lung Ultracut E). Samples were stained with toluidine blue and mounted in immersion oil on glass slides. Light microscopy was done using a Zeiss Axiophot photomicroscope (Zeiss, Oberkochen, Germany) with 35 mm colour film. For transmission electron microscopy, sections of 0. 1 /lm thickness were cut using the ultra­ microtome. Grid-mounted samples were stained for 4 min with saturated uranyl acetate in 50% ethanol, followed by washing in 50% ethanol and subsequently in distilled water. Staining with lead citrate (4 min) (V enable and CoggleshaU, 1 965) was followed by a final washing step in distilled water. Transmission electron microscopy was conducted with a Philips 20IC electron microscope operating at 60kV. Figures presented here are typical representations of findings from light- and electron mlcroscopy. 8. UV-B responses in frequently defoliated white clover 1 55 Results After 1 6 week exposure, the most notable UV -B-induced plant morphological change was a decrease in leaf dry matter (Table 8- 1 ) . Least significant difference analysis (LSD, at P < 0.05) showed that this decrease was significant in the cultivar 'Huia' (- 1 6%), while Tienshan' was unaffected by UV-B (Table 8- 1 ). In the other leaf growth attributes measured here the white clover populations did not display significant sensitivity to UV -B, even though Tienshan' consistently appeared less affected by UV­ B than 'Huia' (Table 8- 1 ). However, there were significant inherent leaf morphological differences between the two white clover populations, with 'Huia' displaying larger leaves than Tienshan' (Table 8- 1 ) . Percent leaf dry mass (PDM) decreased by about 8% across populations under UV-B (Table 8- 1 ). No differences between UV-B treatments or white clover populations were found for specific leaf mass (SLM). Chlorophyll fluorescence parameters generally showed no population-specific or UV -B­ induced differences (Table 8-2). Apart from a small rise in intrinsic efficiency of photosystem II (FvIFm) by about 2%, changes in photochemical yield (MIF m) , and photochemical (qp) as well as non-photochemical quenching (NPQ) were not significant (Table 8-2). Adaxial epidermal cells were intact, with lens-shaped (domed) cells that were particularly convex in Tienshan' (Fig. 8- 1 ) . Mesophyll cells were also undamaged and there were 1 -2 layers of palisade tissue and about 4 layers of spongy mesophyll in all samples (Fig. 8- 1 ). Chloroplast integrity was maintained with intact external (chloroplast envelope) and internal membranes, showing a regular pattern of the grana and stroma thylakoid network (Fig. 8-2). Starch accumulation could be observed in all samples (Fig. 8-2), but examination of the leaf cross-section suggested a decrease in starch content under UV -B (Fig. 8- 1 ) . Table 8-1. Morphological attributes (mean ± 1 SE) in white clover grown with (UV+) and without (UV-) supplementation of 1 3 .3 kJ m-2 d- 1 UV-B . Statistical information includes significances for main and interaction effects as well as the least significant differences (LSD) for the UV -B x population term. DM = dry mass; PDM = percent leaf dry mass; SLM = specific leaf mass. Treatment UV ­ UV - UV + UV + UV Population UV x population LSD (P < 0.05) Clover population Huia Tienshan Huia Tienshan Leaf DM Leaf size (mg) (cm2) 1 9. 1 ± 1 .24 7 .01 ± 0.503 1 1 .8 ± 0.56 4.36 ± 0.320 1 6.0 ± 0.78 6.20 ± 0.25 1 1 1 .4 ± 0.79 4.4 1 ± 0.450 0.050 0.349 < 0.00 1 < 0.00 1 0. 1 16 0.282 2.48 1 . 1 29 Petiole SLM PDM (%) length (mm) (mg cm-2) 98.6 ± 7. 1 9 2.75 ± 0.063 1 9.7 ± 0.26 65.9 ± 4.95 2.74 ± 0.080 20.4 ± 0.60 92.0 ± 4. 1 3 2.57 ± 0.059 17 .7 ± 0.39 70.3 ± 5 .28 2.67 ± 0. 109 19 . 1 ± 0.59 0.839 0. 1 30 0.002 < 0.00 1 0.522 0.038 0. 322 0.5 1 1 0.4 1 8 1 5 .78 0.230 1 .3 8 Table 8-2. Photochemical parameters (mean ± 1 SE) in white clover grown with (UV +) and without (UV -) supplementation of 1 3 .3 kJ m-2 d- I UV -B. Statistical information includes significances for main and interaction effects as well as the least significant differences (LSD) for the UV-B x population term. AF/F m = photochemical yield; Fy/ Fm = intrinsic efficiency of photosystem II; NPQ = non-photochemical quenching ; qp = photochemical quenching. Treatment UV ­ UV - UV + UV + UV Population UV x population LSD (P < 0.05) Clover population Huia Tienshan Huia Tienshan FvlFm 0.8 14 ± 0.0027 0.803 ± 0.0069 0.825 ± 0.0028 0.827 ± 0.00 16 < 0.00 1 0.274 0. 1 1 1 0.0 1 2 1 M/Fm qp NPQ 0.620 ± 0.007 1 0.879 ± 0.0089 0.59 1 ± 0.0332 0.625 ± 0.0096 0.897 ± 0.0067 0.547 ± 0.09 1 3 0.626 ± 0.0 1 69 0.87 1 ± 0.0204 0.609 ± 0.0340 0.638 ± 0.0070 0.878 ± 0.0 1 59 0.582 ± 0.0475 0.402 0 .342 0.645 0.4 16 0.377 0.54 1 0.76 1 0.73 1 0 .88 1 0.0328 0.0422 0. 1700 Fig. 8- 1 . Light micrographs of transverse white clover leaflet sections stained with toluidine blue. Starch grain appear as smal l granules inside the mesophyl l chloroplasts. (a) 'Huia' UV-; (b) 'Huia' UV+; (c) 'Tien han' UV-; (d) 'Tienshan' UV+. AB = abaxial epidermal cells; AD = adaxial epidermal cells; PM = palisade mesophyl l ; SM = spongy mesophyl l . ...... Vl 00 8. UV-8 responses in frequel1lly defoliated white clover 1 59 b Fig. 8-2. Transmission electron micrographs of transverse sections of white clover leaflets exposed to supplementat ion with (UV+) or without (UV-) 1 3 .3 kJ m-2 d- I UV-B. Intact chloroplasts. (a) 'Huia' UV-; Cb) 'Huia' UV+; Cc) 'Tienshan' UV-; Cd) 'Tienshan' UV+. EN = chloroplast envelope; GR = granum; SG = starch grain. x 1 6 000. 8. UV-B responses infrequently defoliated white clover 1 60 Discussion Leaf growth and photochemistry Constitutive and UV -B-generated differences in leaf growth were observed between the two white clover populations (Table 8- 1 ). In particular, leaf mass in the larger-leafed cultivar 'Huia' was susceptible to UV -8, while the less productive ecotype Tienshan' showed no such response after 16 weeks UV -B exposure (Table 8- 1 ) . Population­ specific differences in leaf growth responses under UV -B have also been observed in other studies (Dai et al., 1 994; Fumess et al. , 1999). Contrary to 'Huia', Tienshan' also shows tolerance to a number of other stress forms, including low temperature, low fertility and drought (Chapter 3, Table 3- 1 ). The direct link between higher constitutive leaf mass in 'Huia' and UV -B-induced decreases in leaf dry matter production suggests that previously established l inks of plant and habitat productivity to UV -B sensitivity (Chapter 3) can also be observed under strong defoliation pressure. While no UV -B-induced change was observed in specific leaf mass (SLM), leaf dry mass percentage (PDM) was decreased under UV -B (Table 8- 1 ). Similarly, examinations of 38 South African plant species show no change in SLM, but decreases in PDM under increased UV -B levels (Wand, 1 995). A number of studies reported increases in SLM under UV -B (Britz and Adamse, 1 994; Li et al. , 1 998), but decreases have also been observed (Balakumar et al., 1 993 ; Krizek et al., 1997a). A lack of consistent changes in SLM has led to suggestions that this attribute may be a less reliable indicator of UV -B stress (Fiscus et al., 1 996). Other findings point at higher relative importance for leaf dry matter content, rather than SLM for the prediction of stress tolerance (Wilson et al., 1999). The observed decreases in PDM under UV-B could be indicative of improved water status. Other studies in white clover found UV-B­ induced increases in leaf water potential (Chapter 5 , Fig. 5-3). Improved water relations under UV -B could be due to a number of factors, including improved rootshoot ratios, stomatal limitations or biosynthesis of stress proteins and osmolytes (Schmidt et al., 2000; Teramura et al., 1 984a). Measurements of various chlorophyll fluorescence parameters showed that photosystem II (PSII) photochemistry did not differ between the white clover populations and was 8. UV-B responses in frequently defoliated white clover 1 6 1 not negatively affected by UV -B (Table 8-2). Inhibition of growth without negative effects on PSII photochemistry has also been observed in other studies (Gonzalez et al. , 1 996; Nogues et al. , 1 998). The present findings suggest adequate protection of PSII function, which could be mediated by photoprotective mechanisms such as the presence of UV -B-protective pigmentation (AlIen et al. , 1 998). Structural studies Results from light microscopy (Fig. 8- 1 ) and transmission electron microscopy (Fig. 8-2) suggest no marked structural damage in the UV -B-treated white clover samples. Epidermal cells were intact and displayed the distinctive lens shape on the adaxial surface described for this cell type in white clover (Vogelmann et al. , 1 996) (Fig. 8- 1 ) . In particular, 'Tienshan' displayed a more convex curvature at the adaxial surface than 'Huia' (Fig. 8- 1 ). This could be a function of the previously established differences in cell sizes between the two populations, with smaller size (lower surface area) and thus more dense packing of cells in 'Tienshan' (Chapter 2, Fig. 2-2b). Accordingly, the microscopic investigations suggest smaller periclinal cell diameter for 'Tienshan' epidermal cells, and in addition these cells also appear to be higher than in 'Huia' (Fig. 8- 1 ). Previously, an inverse correlation was found between UV -B sensitivity and constitutive cell size (Chapter 2, Fig. 2-2b) and the present findings may indicate that this could also be due to differential cell height (and thus epidermal thickness). A longer vertical pathlength in 'Tienshan' for incoming UV -B radiation could constitute a UV -B­ protective feature (Gorton and Vogelmann, 1 996). The pronounced lens shape particularly in 'Tienshan ' could further affect leaf optical properties of relevance for UV -B transmission. Differences in epidermal cell size and shape have been related to differences in l ight focussing and scattering (Gorton and Vogelmann, 1 996; Vogelmann et al. , 1 996). In particular, epidermal cells with a small radius of curvature generate a focal plane that l ies in the epidermis or in the upper mesophyll (Vogelmann et al. , 1 996). A survey of 47 plant species showed that white clover had one of the smallest radius of curvature (highly convex outer cell surface), focussing light just below the epidermal layer (Vogelmann et al. , 1 996). Higher convexity of 'Tienshan' epidermal cel ls could suggest increased focussing of l ight and of the accompanying UV -B radiation. This would further increase optical pathlength and 8. UV-B responses in frequently def(lliated white clover 1 62 concentrate UV -B towards the epidermal cells which usually contain specific UV -B­ attenuating pigmentation such as flavonoids (GOl'ton and Vogelmann, 1 996). Another study revealed higher UV -B-induced levels of specific flavonoids in 'Tienshan', compared to 'Huia' (Chapter 6, Fig. 6-2). Epidermal focussing has been noted in early publications (Haberlandt, 1 9 14), but to date little is known about the adaptive and physiological significance of epidermal lens effects, on a species or population level and between habitats (Vogelmann et al., 1 996). The studies in white clover suggest that epidermal differences of relevance for l ight penetration can occur between plant populations and that this could have adaptive ecological significance for UV-B responsiveness. In particular, epidermal cell geometry in the less productive, stress-tolerant white clover ecotype 'Tienshan' displayed properties which could contribute towards the protection from UV -B. Further studies are required to measure whether UV-B penetration through the white clover epidermis, particularly directly underneath central cell regions, differs between white clover populations. This, in conjunction with measurements of UV -attenuating pigmentation in the epidermal layer would assist in the verification of differential epidermal cell shape contributing to UV -B protection. Palisade and spongy mesophyll cells also were intact and showed a similar number of layers in all samples (Fig. 8- 1 ) . This is in contrast to other findings, reporting disruption of intracellular structure and of cell walls (Caasi-Lit et al. , 1 997). The 1 -2 layers of palisade tissue and approximately 4 layers of spongy mesophyll cells are in accordance with previous observations for white clover (Thomas, 1 987a). Integrity of the photosynthetic tissue as well as of chloroplasts (Fig. 8-2) conforms with the findings from chlorophyll fluorescence studies, showing no damaging UV -B effects on PSII function (Table 8-2) . Other reports link chloroplast membrane damage to inhibition of PSII function (Brandle et al., 1 977; He et al. , 1 994). Integrity of granal stacks is of particular importance in this regard, as they contain most of the PSII complexes. A study in rice reported UV -B-generated ultrastructural disruption of granal stacks and of the chloroplast envelope (Caasi-Lit et al., 1 997). In the present study there were no indications of differential vesicle formation or thylakoid dilation between UV-B treatments, as observed elsewhere (Brandle et al., 1 977; He et al., 1 994). Structural chloroplast damage in Pisum (thylakoid dilation, disruption in the envelope) (He et al., 8. UV-B responses infrequently defoliated white clover 1 63 1 994) was accompanied by the accumulation of large starch grains, also observed in maize (Santos et al., 1 993). Contrasting to this, examinations of the white clover samples suggest a decrease in starch content under UV -B (Fig. 8- 1 ) . UV -B-induced decreases in starch volumes were also recorded in Brassica (Fagerberg and Bornman, 1 997). Such decreases were compared to a shade-type response and it was suggested that this could be a result of starch mobilisation as well as of decreased starch synthesis (Fagerberg and Bornman, 1 997). The UV-B-induced decreases in starch content observed here are retlected by decreased levels of available leaf carbohydrates (starch and soluble sugars, see Fig. 7- 1 in Chapter 7). Decreases in soluble sugars could point towards reduced substrate availability for starch synthesis under UV -B. Reduced starch accumulation in the absence of chloroplast damage can be considered a regulatory, rather than destructive consequence of UV -B irradiation. In conclusion, exploration of leaf internal leaf structures suggests structural integrity and inherent differences in epidermal cell structure between the two white clover populations. This warrants further structural examinations on a larger scale, also investigating quantitatively the observations made here on a qualitative basis. The present study adds further information to a framework linking plant specialisation to UV -B responsiveness. In particular, this suggests a relationship between population­ specific epidermal cell geometry and UV -B sensitivity. Between-population variation in leaf mass and UV -B sensitivity under frequent defoliation could suggest that UV-B effects may also occur under grazing pressure. 9. Discussion 1 64 Chapter 9. Discussion This Chapter will be submitted in part to Trends in Ecolog}' and Evolution. These studies address a number of issues that provide new information to the field of ultraviolet-B plant research. Thus far there had been a lack of comprehensive investigations on pasture species, particularly on a population level. Inclusion of a diverse range of wbite clover populations added weight and generality to the findings and facilitated comparability among functional plant types, independent from confounding species-specific differences. This also allowed testing whether stress­ tolerant plants would be UV -B resistant, a hypothesis previously not examined in detai l . Investigations ranged from specific biochemical responses to trophic interactions under controlled environmental conditions. A number of morphological and physiological UV -B effects were detected. There were significant population differences in several attributes, both constitutively and in response to UV -B . Differences in UV -B sensitivity of the white clover populations were linked to constitutive features of plant productivity, morphology and tolerance to other forms of stress as well as to DV -B­ induced physiological attributes. UV -B responsiveness was mitigated by drought and by duration of DV -B exposure. Population-specific DV -B responsiveness could also be detected in frequently defoliated white clover leaves and differentially affected the performance of insect herbivores. In the following section, a framework is proposed, based on constitutive and UV -B-induced attributes of the white clover populations in relation to their UV -B responsiveness. A functional framework to synthesise resu lts The results from this series of investigations can be drawn together in a conceptual model, suggesting a fundamental relationship of UV -B responsiveness to the productivity of white clover populations and of their habitat (Fig. 9- 1 ). This is further l inked to a number of morphological-physiological attributes (Fig. 9- J) . The key steps contributing to this overall result are summarised in Table 9- 1 , using the plant attributes listed in Fig. 9-2 and the individual experimental approaches in Table 9-2 to Table 9-4. 9. Discussion CONSTITUTIVE: • Habitat productivity r-------/- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -/---------, • Leaf size • Epidermal cell size • Leaf water potential PLANT PRODUCTIVITY I LOW I+--� -1 '-----.--� t- UV-B- INDUCED: • UV-absorbing compounds H IGH I .... • Total flavonols • Quercetin r UV-B­ TOLERANT • Leaf water potential • Leaf dry mass percentage .. HIGH -. LOW ,,. r UV-B­ SENSITIVE 1 65 Fig. 9-1. Model l inking plant productivity and related morphological and physiological features in white clover to UV -B responsiveness. >­ <.9 g Q Cl) >­I a.. 6 I a.. er: o � z o i= o w � o er: a.. r • • \. • J Photosynthesis & water relations '" PS 1 1 fu nct ion & ch lorophy l l .� Chl ()rop la�l u l t ra�lruct u re • Me. oph)' l l �truct u re Stomatal densi ty/d i:-. tr ibut i oll • Leaf water potent ial ./ I I UV -B absorI!tion & dissiI!ation • UV -absorbing compounds • Flavolloids I I Nutritive value" • Nitrogen • Carbohydrates • NDF • Cyanogenesis • I n sect , performance ./ J I Antioxidant function • Proline • Flavonoids I I � r Structure & "" r " mOrJ2ho2enesis MOrJ�hology • Primordia & 2rowth format ion • Total mas:-. • Epidermal • RGR cel l � i l.c • Organ ma:-.s , . Cel l n umber ./ • RSR I • LWR J • Leaf s i ze Structure & • Pet iole l engt h morI!hogenesis • LAR • Cel l geo met ry • SER • SLM & PDM " • NGP:-. ./ I t Fig. 9-2. Scheme of the current research project, based on morphological and physiological attributes in white c lover. Changes in plant attributes due to UV -B are marked. Red font: increases; g reen font: decreases; g rey font : no change. LAR = leaf appearance rate; LWR = leaf weight ratio; NDF = neutral detergent fibre; NGPs = number of growing points; PDM = percent leaf dry mass; PSI I = photosystem I I ; RGR = relative growth rate; RSR = rootshoot ratio; SER = stolon elongation rate; SLM = specific leaf mass. ....... 0\ 0\ 9. Discussion 1 67 Table 9-1. Overall features of the series of investigations. '"""' N I 0\ on ii '-...../ (,() V ...... :::l ...0 ·c .... .... . Of) 0 � VJ :>. ..c CL. VJ c .8 ..... C'Cl :3 0.. 0 CL. PDM, leaf primordia formation in the bud. stolon growth. • Confirmation of the overall picture obtained from large-scale screening. • Reduction of all growth attributes by UV-B, including productivity, leaf growth, rates of stolon elongation and leaf appearance. leaf primordia formation. • Both leaf size and leaf number affected by UV -B . • UV -B effects on aboveground structures shielded from UV -B, suggesting indirect impacts on cell division. • Dry matter allocation within the leaf (PDM, SLM) generally not affected. • • • • • • • • • • Identification of the flavonoid structures involved in the white clover UV­ B response. Flavonols, particularly quercetin glycosides involved. Total flavonol responses and quercetin accumulation well-conserved on a population level: increased in all populations. General increase in the quercetin :kaempferol ratio. Flavonols much more increased than UV -absorbing compounds. No UV -B effect on fluorescence quenching, suggesting well-protected PSII function. Consistent UV -B-induced increases for If/L in drought-stressed plants and for higher proline levels in well-watered conditions. No overall association between UV-B habitat and flavonol production, but between habitat productivity and accumulation of specific flavonoid compounds: highest flavonol and quercetin levels in stress-tolerant ecotypes, suggesting particular ecophysiological roles for dihydroxylated flavonoids in stress protection. Similarly, UV -B-induced increases in UV -absorbing compounds for stress­ tolerant Asian ecotypes. Increases of If/L in these ecotypes, indicating UV -B-elicited advantages in aspects of water retention. 9. Discussion 1 70 � "Ci ...... c: 0 u � '-" c: 0 . ..... ...... C\l '3 0., 0 0... ...... ...c: /;)[J ::l 0 l-; "0 "0 c: C\l o:l I > ;.::J 4-< 0 c: . S ...... C\l c: :.0 El 0 u • • • • Increased proline levels and decreased If/L under UV -B in white clover populations with intrinsically low proline and high If/L levels . High intrinsic If/L in populations with low PDM under well-watered conditions, also related to lower proline levels. Highest proline levels (under drought) and lowest intrinsic l/fL (in well­ watered conditions) for stress-tolerant populations of low productivity. Inverse constitutive relationship between quercetin glycoside accumulation and l/fL. • Much stronger influence of drought for primary productivity: decreases in total growth and leaf growth, in chlorophyll levels and l/fL, increases in leaf dry matter allocation (SLM and PDM), NPQ and proline. Little effect on UV -absorbing compounds and flavonols . • • • • • • Amelioration of UV -B responses by drought: only marginal or non­ significant UV -B effects on morphology and growth, concomitant with no population differences in the UV -B response. Particularly beneficial effect of drought for UV-B sensitivity in productive white clover cultivars. Primary role of UV -B for flavonol induction, but pronounced synergistic interaction with drought. Synergistic interaction also for UV -absorbing compounds and If/L. UV -B-induced improvements of water relations under drought: PDM decreased, while l/fL was consistently increased across populations . Possible amelioration of UV -B sensitivity mediated by ( 1 ) drought alone via (a) delays in a number of growth attributes, (b) increases in SLM, PDM, proline, as well as by (2) synergistic UV -B x drought interaction via increases in UV -absorbing compounds, flavonols (especially quercetin gl ycosides) and l/fL. • Decreased UV -B sensitivity with increasing duration of drought (additive effect in the early drought phase, antagonistic after four weeks) and with increasing exposure to UV -B. • Similar UV -B sensitivity ranking among the white clover populations over time. • Links to UV -B growth tolerance via ( 1 ) constitutive features, including low productivity, small leaf size, low l/fL, and (2) via UV -B-induced features, including increases in UV -absorbing compounds, in flavonols, and in l/fL, as well as higher quercetin (but not kaempferol) glycoside accumulation and higher PDM (rather than SLM) under UV-B . 9. Discussion 1 7 1 Table 9-4. Key aspects from bioassay studies and examinations of UV -B susceptibility of frequently defoliated white clover populations. r/; 1) ...... ;:l ..0 ·c ...... ...... -< • Investigation of species differences in pasture insect pests feeding on UV­ B-treated leaves of two different white clover populations. • • • • UV -B effects on primary metabolic and morphological processes : increased nitrogen levels ; decreases in fibre, PDM, carbohydrate, starch levels, reduced leaf growth; PSIl function unaffected. Regulatory, rather than destructive UV -B effects on leaf ultrastructure. Secondary metabolism: cyanogenesis strongly increased by UV -B in cyanogenic cuItivar 'Huia'. Increased substrate biosynthesis for cyanogenesis in both cyanogenic and acyanogenic genotypes; no UV -B effect on UV -absorbing compounds. • Differential insect responses to UV -B-mediated foliar changes in 'Huia', temporary sensitivity in Spodoptera litura may be related to increased cyanide production. [/J � • Other UV -B-induced changes in the performance of both insect species :J generally small . • UV-B tolerance in white clover may be related to structural epidermal characteristics inherent to the slower-growing stress-tolerant white clover population. 9. Discussion l 72 Characteristic features of the model in Fig. 9- 1 first emerged from the application of multivariate statistical techniques, integrating 17 UV -B responses across 26 white clover populations after 2 1 /2 weeks of exposure to UV-B (Table 9-2). Further investigations tested these findings under longer-term UV -B irradiation ( 1 2 weeks) and in combination with drought stress (Table 9-3). Thus far, very little functional information had been available on ecological interactions between plant stress forms in conjunction with populations of the same species (Krupa et al. , 1 998a). Very few studies had previously examined the combination of UV -B with drought on a population level. The observations from this work (Table 9-3) suggest major physiological UV -B action occurs via indirect effects on processes l inked to secondary metabolism, rather than processes linked to primary photosynthetic function. This supports current views attributing UV -B an indirect, regulatory role in physiological and ecological processes, mediated by polyphenolics (Rozema et al., 1 999). Compared to productive cultivars, slow-growing, UV -B-tolerant populations from stress-exposed habitats displayed greater ability for such biochemical mechanisms of UV -B protection. In addition to deficiencies in biochemical defense mechanisms, stress susceptibility in the cultivars was also related to plant features contributing towards resource acquisition, e .g. larger leaves (Table 9-3). While less pronounced, this could also be observed under frequent defoliation (Table 9-4). Sensitivity of one generalist lepidopteran species, but not of another to UV -B-treated 'Huia' foliage points at the possibility of highly specific plant-herbivore interactions (Table 9-4). These findings contributed to a conceptual framework for UV-B responsiveness, suggesting ( 1 ) a general advantage of lower habitat and plant productivity for increased capacity in biochemical UV-B protection, while (2) specialisation towards biomass gain leads to morphological change resulting in pronounced reductions of plant productivity (Fig. 9- 1 ). Comparisons of morphological and physiological mechanisms showed that UV -B responsiveness is mainly linked to constitutive differences (i .e. measured under no UV -B supplementation) in morphology and growth among the populations, while it is more the UV -B-induced physiological differences that confer UV -B tolerance (Fig. 9- 1 ) . A metabolic cost of a plant strategy towards constitutive and UV -B-induced protection against UV -B may be lower investment of carbon allocation towards primary constitutive productivity. The results from these studies thus propose functional 9. Discussion 173 population types for UV -B responsiveness in white clover, separating stress-tolerant, slow-growing ecotypes with greater ability for biochemical mechanisms of UV-B protection from agricultural white clover cultivars showing the opposite characteristics. This model is consistent with ecological models of stress tolerance, proposing distinct plant types depending on primary plant strategies (Diaz and Cabido, 1 997 ; Diaz et al., 1 999; Grime, 1 979; Grime et al. , 1 997; Poorter and Gamier, 1 999). Plant stress responses can be classified into two main categories: morphological and physiological (Bradshaw, 1 965). The latter category is seen of particular relevance for stress-tolerant plant types (Grime et al., 1 997). Stress tolerators are less likely to use morphological changes as a strategy against stress, because of generally slower growth and biomass turnover. Instead, cellular acclimation through biochemical adjustments would be the expected primary strategy (Grime and Campbell , 1 99 1 ). On the other hand, plant types specialised towards higher resource capture would be expected to display a larger degree of morphological plasticity in response to stress, mainly due to changes in the meristematic regions of the plant (Grime and Campbell, 1 99 1 ). A common route for expression of the latter would be decreased growth (Grime and Campbell , 1 99 1 ). Analysis of 24 vegetative and regenerative traits in 1 00 Argentinean plant species over a wide range of climatic conditions demonstrated trade-offs between investment in growth on the one hand and allocation to stress protection on the other (Diaz and Cabido, 1 997). Attribute-environment linkages allow particular traits or trait combinations to be used for predictions of responses within and between plant species under changing climatic conditions (Diaz et al., 1 999). Compared to their less productive counterparts, fast growing plant types usually have higher leaf area and less relative root mass, contributing to larger carbon gains per unit plant mass (Lambers and Poorter, 1 992). They often have higher photosynthetic rates , higher levels of organic nitrogen and minerals as well as lower carbon levels (Lambers and Poorter, 1 992; Poorter and Gamier, 1 999). Slow-growing plant types on the other hand often contain higher levels of cell wall compounds, starch and secondary metabolites, display increased leaf longevity, higher leaf density as well as higher dry matter content, relative to fresh mass (Lambers and Poorter, 1 992; Poorter and Gamier, 1 999). These features are seen to confer general tolerance to abiotic and biotic stress factors (Lambers and Poorter, 1 992; Poorter and Gamier, 1 999). 9. DisclIssion l 74 Many of these characteristics were found in the slower-growing white clover populations originating in stress-exposed habitats. The relationships between morpho­ physiological plant attributes, productivity and UV -B responsiveness observed here propose linkages previously recognised in other fields of stress research. The relationship to habitat productivity confirms for UV -B responsiveness suggestions that resistances to a number of stress forms can be predicted from each other (MacGillivray and Grime, 1 995) . Furthermore, it appears that all three classes of UV-B protection (Beggs et al. , 1 986) may be present in the white clover UV -B response, including repair (e.g. antioxidant functions by flavonols), attenuation (e.g. UV -absorbing compounds) and growth delay (e.g . reduced cell division). Comparisons between experiments Seven white clover populations from the initial screening programme (Chapter 2) were also included in the drought interaction experiments (Chapters 3-5) and in investigations of flavonoid responses (Chapter 6). These included the ecotypes 'Sarikamis', Tienshan' and 'Haggas' as well as the cultivars 'Huia' , 'Kopu', 'Prestige' and 'Haifa'. Chapters 7 and 8 further investigated Tienshan' and 'Huia' attributes during the bioassay experiments. With each experiment, the plants experienced progressively increased duration of UV-B exposure, from 2 1h weeks to 16 weeks (Table 9-5) . The 1 2-week trials (Chapters 3-6) also included three-week preconditioning with gradual increases in UV -B. Nevertheless, the general pattern of UV -B sensitivity was consistent, irrespective of duration in UV-B application, developmental stage of the original plant material, soil type or defoliation treatment (Table 9-5). UV -B-generated dry matter reduction for the populations in the screening experiment ranked similarly to that in well-watered conditions during the stress interaction trials (Fig. 9-3). This was even more pronounced when comparisons were made across water treatments (r = 0.759, P < 0.05). There was also no apparent overall change in effect when ultraviolet-A radiation (UV-A) was included in the control treatment of the bioassay studies (Table 9-5). However, the previously established time-dependency of UV-B responsiveness (Fig. 3-4) seemed further confirmed by the observation that after the longest period of UV -B exposure ( 1 6 weeks, Chapter 8), only some aspects of leaf growth were affected by UV -B (Table 8- 1 ). Table 9-5. General comparison between UV -B application, plant conditions and UV -B sensitivity in three main experimental approaches studying UV -B effects on white clover. Daily UV -B levels in all experiments were 1 3 . 3 kJ m -2 d- 1 . Feature Large-scale screening Stress interaction studies and Bioassay studies flavonoid responses Duration of UV -B treatment 21 /2 weeks 1 2 weeks (with preconditioning) 1 6 weeks UV-A included No No Yes Plant material Stolon cuttings Seedlings Stolon cuttings Soil material Sand Sand Soil Defoliation No Infrequent Frequent UV-B sensitivity yes yes yes Differences m UV-B sensitivity yes yes yes among populations UV-B tolerance for ecotypes yes yes yes 9. Discussion 1 76 Constitutive plant growth attributes correlated well between experiments, exemplified by the productivity measurements in Fig. 9-4. Similarly close relationships could be found for leaf dry mass (r ::: 0.977, P < 0.00 1 ) and leaf size (r ::: 0.975, P < 0.001 ). Population-specific differences in leaf growth were also maintained after 1 6 weeks growth under controlled environmental conditions (Chapter 8, Table 8- 1 ). Plant attributes not affected after long-term UV -B irradiation (SLM and Fv/F m,) were only marginally or not significantly related between experiments. This further indicates little importance of these attributes in white clover UV -B responsiveness. However, UV -B­ induced accumulation of UV-absorbing compounds (r ::: 0.72 1 , p ::: 0.068) and relative increases in these levels were significantly related between experiments (Fig. 9-5). It was of interest to note that UV -absorbing compounds did not show UV -B-induced increases in 'Tienshan' in Chapter 7 (Table 7- 1 ) . It is possible that the frequent defoliation stress led to alterations in leaf composition away from UV -absorbing compounds. � e - c: o (.) '0 1 00 e:, Q) III C o g. 80 Q) ... .r:. � ... tn E 60 ... � en c: o ..J e Haifa y = 1 . 1027x · 2.4393 r = 0.685 P = 0.089 • Sarikamis • Prestige 40 � ____________ � ____________ � ______________ L-__________ � 50 60 70 80 Short-term growth response (% of control) Fig. 9-3. Comparison of growth responses under well-watered conditions of seven white clover populations exposed to supplementation with 1 3 .3 kJ m -2 d- I UV -B for 1 2 weeks (ordinate) versus a 2 1/2 week screening experiment (abscissa). The growth response is expressed as relative growth (dry matter production under UV -B expressed as a percentage of that of the control). 9. Discussion --. '>. <'Cl "0 Cl .§.. 300 >. "0 :::I "iii E � 200 Cl c: ..2 c: .2 "0 :::I 100 "0 E Co :: o Tienshan . y == 0.5851 x + 31 .59 r == 0.946 P < 0.01 Kopu o � __________ L-________ �� ________ �� ________ � __________ � o 1 00 200 300 400 OM production short-term study (mg day·l) 1 77 Fig. 9-4. Comparison of constitutive growth (dry mass production per day under well­ watered conditions without UV -B supplementation) of seven white clover populations grown under controlled environmental conditions for 1 2 weeks ( ordinate) versus a 2 [/2 week screening experiment (abscissa). DM = dry mass . ...... y ::: 1 .1 01x - 8.4826 E r ::: 0.776 C P < 0.05 • Tienshan 0 () - 1 20 0 � 0 -- • Sarikamis Cl> I/) c: 0 g- 1 10 Cl> .... I/) .c <'Cl . > ::::l E 1 00 .... Prestige .!!! • Haifa . Cl • Haggas c: 0 ...J 90 90 100 1 1 0 1 20 Short-term UV-abs response (% of control) Fig. 9-S. Comparison of UV-absorbing compound (UV-abs, measured at 300 nm) responses under well-watered conditions of seven white clover populations ex�osed to supplementation with 1 3 .3 kJ m-2 do l UV-B for 1 2 weeks (ordinate) versus a 2 /2 week screening experiment (abscissa). 9. Discussion 1 78 Context and l imitations of the findings The daily indoor photosynthetic phc 'n flux (PPF) levels used in these experiments were about half of the average daily mid-summer PPF (Greer and Laing, 1 992) and the plant-weighted UV -B dose was 58% higher, compared to the daily clear-sky mid­ summer UV-B dose above Palmerston North (McKenzie, 1 99 1 ). These experiments used a 'square-wave' approach, subjecting plants to constant UV -B exposure, rather than to the gradual increase in the field. The control was a 'zero' UV -B treatment which in most cases (Chapters 2-6) had no UV -A supplementation. As a consequence, these conditions are l ikely to lead to larger UV -B effects on plants, compared to outdoor conditions. For example, higher PPF and UV -A levels contribute to protective functions such as photorepair and flavonoid production (Nogues et al., 1 998). Nevertheless, such artificial indoor conditions are often used to study mechanisms involved in UV -B responses (All en et al., 1 997; Delong and Steffen, 1 997 ; Lavola et al., 1998; Nogues et al., 1 998; Olsson et al., 2000; Olsson et al., 1 999). This approach was chosen ( 1 ) to clearly expose genetic differences among white clover populations in response to UV -B, (2) to discern possible functional mechanisms underlying differential UV -B responsiveness, and (3) to examine specific stress interactions without confounding effects by other environmental variables. Screening experiments examining UV -B responses in a number of populations are seen to be particularly practicable under controlled conditions (Corlett et al. , 1 997). This is due to a number of factors, e.g. control of population-specific environmental interactions but also in consideration of the time and expense otherwise involved for field experimentation. Furthermore, similar UV -B responsiveness was observed among the white clover populations after preconditioning with UV -B by gradually increasing UV -B levels over three weeks (Chapter 3). Differential UV -B sensitivity within white clover was also maintained when plant attributes under UV -B were compared to a UV -A control (Table 9-5). The zero UV-B control can be used to discern UV-B effects per se, which would not be possible with two levels of UV -B irradiation but no zero controL There are merits as well as limitations in the approach used here and caution will be necessary to extrapolate the findings from these studies to the field. The usefulness of laboratory data in predicting plant stress responses has been highlighted in a number of 9. Discussion 1 79 studies on plant functional types (MacGil livray and Grime, 1 995; Poorter and Gamier, 1 999). One major function of the present experiments was to generate hypotheses leading to the development of a model for UV-B responsiveness, which can in consequence be tested in the field. Such investigations are presently being conducted. A further limitation in these studies is that much of the evidence concerning linkages to UV -B responsiveness is correlative. However, there are also other indications for similar relationships from comparative studies between the cultivar 'Huia' and the mutant breeding line 'Red' (Fig. 9-6). The latter population was selected from 'Huia' phenotypes with high anthocyanin content. The comparison within a very similar genotypic range strengthens the evidence that there may be metabolic trade-offs between biochemical defense mechanisms and plant productivity. The results showed 25% higher constitutive levels of UV -absorbing compounds in 'Red' (P < 0.01 ) and a similar difference in growth rates, with higher productivity for 'Huia' (P < 0.00 1 ) Fig. 9-6a,c . Differences of cell size could not be assessed as these were similar between the two populations (Fig. 9-6d), which may be related to their common ancestry. However, constitutive leaf sizes were markedly smaller in 'Red' (P < 0.00 1 ) (Fig. 9-6b). Given the similarity in cell sizes, this therefore is a consequence of reduced cell numbers in 'Red', suggesting that potential morpho-physiological trade-offs between the two populations act at the level of cell division, rather than cell expansion. 'Red' displayed higher UV -absorbing compound levels than 'Huia', irrespective of UV-B treatment (Fig. 9-6c). Thus 'Red' did not increase levels of UV -absorbing compounds in response to UV-B, which could be due to the high constitutive values, already providing sufficient UV -B protection. Comparisons of UV -B effects on the productivity of the two populations clearly showed differential UV -B susceptibility with tolerance for 'Red' (P < 0.01 ) (Fig. 9-6a). The general picture emerging from the comparisons between the 'Red' mutant and its mother population 'Huia' is a reflection of the findings from the other studies in this project, suggesting UV -B tolerance for the slow-growing, small ­ leafed population with high UV -absorbing compound production. This is further supported by examinations of a green (anthocyanin-poor) and a red (anthocyanin-rich) Coleus population (Burger and Edwards, 1 996). UV -B tolerance for the red population was related to protection by anthocyanins, e.g. UV -B absorption due to acylation with cinnamic acids (Burger and Edwards, 1 996). (a) 0.1 :;: I (b) C 1.2 ., a> E '" 0.0Il '0 <: ::> a: (!) 0.06 a: &. 0.8 E 0 (.) 0.6 a> 0.04 <: :e 0.4 0 '" 0.02 .0 '1' 0.2 > :::l o L- (c) [(d) 12 1400 1200 10 NE 1000 .§. "e .3- ., 800 .� '" .� '" 4i 600 () ,. � 400 ,00 . I - Huia Red Huia Red Fig. 9-6. Morpho-physiological attributes in the white c lover populations 'Huia' and 'Red', grown with (closed columns, .) or without (open columns, D) supplementation of 1 3 .3 kJ m-2 d- 1 DV-B; (a) relative growth rate (log-transformed increment of total plant biomass per day), (b) leaf size, (c) DV-absorbing compounds (d) cell size. DM = dry matter. 00 o 9. Discussion 1 8 1 Perspectives and future studies This work portends a range of future studies, from sub-cellular to ecosystem levels. UV­ B-induced growth reductions occurred in the absence of damaging effects on chlorophyll accumulation or of marked inhibition of light conversion in the photo system II reaction centre. While it is possible that other photosynthetic processes such as CO2 fixation could still be affected by UV -B, growth reductions in the absence of any primary photosynthetic changes have been reported (Gonzalez et al., 1 996). The observed growth reductions were l inked to UV -B-elicited decreases in cell division and cell expansion, resulting in decreases in leaf size and leaf number, thus reducing total leaf area and thereby photosynthesis at the whole plant level . Together with fewer and shorter stolons this ultimately resulted in lower white clover yield. The effects on cell division and expansion could be due to UV -B effects on plant growth substances. For example, the auxin indole acetic acid (IAA) can undergo photodestruction by UV-B (Huang et al., 1 997). Furthermore, flavonol compounds such as quercetin can decrease IAA activity or concentration, and increased flavonol levels under UV -B could thus affect IAA function also indirectly (Meijkamp et al. , 1 999). While higher constitutive levels of quercetin glycosides could be related to slower growth in the stress-tolerant white clover ecotypes, these could in turn be better adapted to increased accumulation of these pigments under UV -B. A UV -B-induced drop in available carbohydrate levels was observed both in 'Huia' and 'Tienshan'. However, significant UV -B-generated decreases in growth only occurred in 'Huia'. It also became obvious that decreased sugar levels did not lead to increased starch accumulation in the leaves or to higher foliar fibre content. This suggests that the available sugars could have at least partly been converted in other carbon-requiring metabolic functions. In addition to flavonoids resulting from aromatic amino acids originating in the shikimic acid pathway, UV -B stimulated accumulation of other amino acid types, including the heterocyclic proline and cyanogenic compounds based on the aliphatic valine and isoleucine. It became apparent that UV -B effects on such processes differ among white clover populations. From the correlation studies with UV-B responsiveness it can be suggested that white clover populations capable of specifically directing carbon flow towards flavonol, and particularly quercetin glycoside 9. Discussion 1 82 accumulation, are better protected against DV -B than those directing these resources towards cyanogenesis or free proline accumulation. This suggests merit for carbon balance analyses in future UV -B studies. Both proline and the cyanogenic glycosides present in white clover are not likely to contribute to UV-B absorbance (Nahrstedt, 198 1 ). Specific UV -B-induced shifts of carbon-containing substrates from primary metabolism to UV -B-protective secondary product formation have recently been described in parsley (Logemann et al., 2000). The present work indicates a particular role for metabolic steps linking sugar and flavonoid metabolism in the protection of white clover against UV -B. Future studies should thus especially focus on the shikimate pathway, linking these two metabolic processes. Within the flavonols, however, other issues than carbon balances also appear involved due to the highly specific increase in quercetin glycosides, relative to kaempferol compounds. In particular this indicates a cmcial role for the enzyme flavonol 3 ' hydroxylase (F3'H), catalysing the hydroxylation of kaempferol to quercetin . Such pivotal involvement of F3'H has been suggested previously (Ryan et al. , 1 998) and is further supported by the present work as no other possible alternative flavonoid compounds indirectly resulting from F3'H activity (e.g. myricetin via flavonol 3 ' ,5 '­ hydroxylase) were detected in the present studies. Specificity for UV-B-elicited ortho­ dihydroxylated flavonoid accumulation has also been highlighted by other observations demonstrating that biosynthetically related compounds possessing 3 ',4'-dioxygenation, but excluding dihydroxylation, do not increase under UV -B (Markham et al., 1 998b). It would appear that quercetin accumulation under UV -B in populations with moderate flavonol accumulation but highly efficient Q:K ratios ( 'Syrian', 'Prestige' and 'Kopu' - see Fig. 6-4) could largely depend on substrate availability. Assuming a biosynthetic relationship to flavonol levels, quercetin glycoside accumulation in these populations seems more limited by inefficiencies in earlier steps of flavonol synthesis, rather than also by inefficient quercetin conversion (as in 'Octoploid' and 'Haifa', Fig. 6-4). Other studies also report a central role for the dihydroxylated:monohydroxylated flavonoid ratio in circumstances of limited flavonoid accumulation under long-term DV-B irradiation (Markham et al., 1 998a; Ryan et al., 1 998). From the Q:K ratios it appears that the high UV -B-induced quercetin glycoside levels in ecotypic populations are a result of relatively efficient quercetin conversion from high flavonol levels while the 9. Discussion 1 83 more productivity-oriented populations all are disadvantaged in their quercetin compound accumulation due to deficiencies in early, and some additional ly in late, steps of quercetin synthesis. This suggests that these populations could be useful for detailed molecular examinations in several stages of flavonoid synthesis. As mentioned briefly (Chapter 6), in addition to UV -B screemng and potential antioxidant activities these f1avonols could perform their functions also by harmless energy dissipation. Studies on keto-enol tautomerism point at UV -B protection by f1avonols via excited state intramolecular proton transfer (ESIPT) (Smith and Markham, 1 998). Absorbed UV energy could, via flavonol excitation, facili tate ESIPT between flavonol tautomers, for instance proton transfer from the excited state of an enolic tautomer to the keto form of the ground state, thus harmlessly dissipating potentially damaging UV -B radiation. Both quercetin and kaempferol glucosides have shown such keto-enol tautomerism, which is seen to be faci litated by the formation of flavonol dimers, especially at higher aqueous concentrations (Smith and Markham, 1 998). This would be especially relevant in regard to the often very pronounced UV -B-induced increases of flavonoid levels, particularly flavonols, as observed in the present study. Furthermore, it has been shown that kaempferol-7 -O-glucoside can also differ from quercetin-7 -O-glucoside in aqueous solution (Smith and Markham, 1998). Depending on concentration, the kaempferol glucoside keto tautomer can be present at lower quantities in the ground state, with the majority of tautomers in the enol form. The enol tautomer absorbs in the blue spectral region while the absorption spectrum of the keto form is located in the UV region. Thus it is possible that quercetin glucoside, having relatively higher availability of the keto tautomer, could absorb and protect from UV-B more effectively compared to kaempferol glucoside (Smith and Markham, 1 998) . This suggests closer examination of energy dissipation roles for such compounds in future UV -B studies. Such studies should also explore the population-dependent accumulation of UV -B­ induced flavonoids, especially quercetin and kaempferol glycoside production, steady state and degradation. Further work needs to establish whether marked increases in these flavonol compounds are systemic. Pronounced systemic flavonol increases could, in addition to effects on plant morphology and physiology, also impact on ecosystem processes. The observed links between quercetin glycoside accumulation and biomass 9. Discussion 1 84 production warrant detailed investigations of possible linkages to growth regulatory mechanisms, e.g. effects on auxin function . The synergistic interaction of UV-B and drought for the initiation of these compounds clearly warrant such investigations in combination with other forms of stress. Optical studies could determine whether UV -B penetration through the white clover epidermis, particularly directly underneath central cell regions, differs among white clover populations. This, in conjunction with measurements of UV -attenuating pigmentation in the epidermal layer would assist in the verification of differential epidermal cell shape contributing to UV -B protection. The findings of this work also suggest detailed investigations on UV -B effects on plant growth substances in white clover. Decreases in processes linked to cell division in plant parts shielded against UV -B indicate indirect, rather than direct UV -B effects on cell division. The results could suggest a particular focus on IAA function, but from the observations on reduced cell division also on cytokinins, gibbere11ins and other growth regulators. The observed UV -B effect on young leaf and bud growth could affect both auxins and gibberellins, which are both produced in these organs. This could also help explain the observed decreases in shoot elongation. Cytokinins can decrease under drought stress (Shashidhar et al., 1 996) and have been shown to counteract negative UV -B effects (Jain et al. , 1 988). Abscisic acid levels increase in drought (Shashidhar et al. , 1 996) and under UV -B (Rakitina et al., 1 994). Combined investigations of UV-B with drought on this growth inhibitor could be of particular interest in consideration of the marked improvements in leaf water status under UV -B observed here. Decreases in elongation growth could also be related to increases in ethylene production, a hormonal response that has been observed under UV-B (Predieri et al., 1 993). From other studies it is possible that the observed population-specific UV -B effects in white clover are related to differential efficiency of jasmonic acid and ethylene signal pathways (Mackerness et al., 1 999). Organ-specific studies The main focus of this research was on white clover leaf attributes. This is both due to the primary role for UV -B reception by these organs and because of their predominant contribution to white clover yield. However, findings from this work also suggest merit 9. Discussion 1 85 for future studies focussing on other plant organs in white clover. These experiments did not aim at a comprehensive investigation of other polyphenolics, e.g. lignin or tannins. Lignin is part of NDF (neutral detergent fibre) which showed decreased levels in 'Huia' and no change in 'Tienshan' (Fig. 7- 1 ). Condensed tannins only occur at negligible quantities in white clover leaves (Caradus et al. , 1 995). It would however be tempting to investigate whether UV -B affects levels in plant organs that contain higher amounts of these polyphenolics, i .e. in stolons for lignin and in flowers for tannins. The latter would also be of particular relevance in field studies as condensed tannins increase protein utilisation and reduce bloat in ruminants (Dakora, 1 995). Visitation by pollinators can be changed by UV -B-induced changes to plant characteristics of importance for attraction of beneficial insects (Conner and Zangori , 1 997). Reproductive parts of the plant like pollen and ovules are relatively well protected from UV-B. Anther walls, for instance can absorb more than 98% UV-B (Flint and Caldwell, 1 984). A susceptible target to UV-B, however, is the germinating pollen after transfer to the stigma. UV -B damage can alter in vitro pollen germination (Flint and Caldwell , 1 984). This would be of particular relevance in consideration of the observed changes in flavonoid compounds which are involved in the processes of pollinator attraction and germination (Dakora, 1995). Taking into account the lower soluble sugar and starch levels in white clover leaves under UV -B (Fig. 7- 1 ) it would be interesting to examine whether starch levels would increase in the roots. Given higher root :shoot ratios under UV -B (Table 2-2), it is possible that some carbon was preferentially exported towards roots. This could again suggest mediation of the UV -B response via plant growth substances. However, heterogeneity of damage between aerial and belowground plant organs can also be due to organ-specific differences at the molecular level. In maize, less DNA damage was observed in roots compared to leaf tissues and this was related to either physical separation from UV -B exposure or higher rates of DNA repair (Stapleton et al., 1 997). Increased relative root mass also suggests examination of other important below ground white clover functions, including nodulation and interaction with beneficial microbes. A relative improvement of root development could in the longer-term benefit the shoot due to possible increases in total nodulation of the larger root surface. 9. DisclIssion 1 86 This is also of particular relevance in regard to the observed increases in flavonols as they are involved in beneficial interactions with soil microbes (Dakora, 1 995). Production of flavonoids which are released by legume roots and which are involved in the inoculation of the plant by symbiotic nitrogen-fixing bacteria can be affected by UV -B (Sallaud et al., 1 995). Treatment of three tropical leguminous crops with elevated UV -B resulted in reduced nitrogen fixation via decreased nodule activity, accompanied by alterations in growth and net photosynthetic rates (Singh, 1 997). Alterations of symbiotic interactions through the influence of UV -B radiation could have wide­ ranging implications for the New Zealand pasture ecosystem, largely reliant on such relationships. The cultivar 'Huia' belongs to a group of white clover populations that is particularly dependent on its nitrogen supply through symbiosis with nitrogen fixing bacteria, especially under conditions of low soil nitrogen. Effects on this interaction could again lead to effects of UV -B on population dynamics within New Zealand pasture ecosystems. Population-specific studies As mentioned above, these studies revealed a number of constitutive and UV -B-induced differences among the white clover populations. The background information on the environmental and breeding background of all white clover populations provided here will also be useful for future classifications of UV -B responsiveness based on plant functional types. Apart from the general distinctions in UV -B responsiveness between stress-tolerant ecotypes and productive cultivars, specific issues were also raised from observations on individual white clover populations. For example, productivity in some high-yielding white clover populations (e.g. in 'Octoploid' and 'Kopu', Fig. 3- 1 a,b) still exceeded that of the UV-B-tolerant ecotypes, even when reduced by UV-B exposure. Thus the absolute yield in some white clover population could still be sufficient even though they are more UV -B sensitive. Long-term studies on higher-yielding white clover populations should examine whether this would be maintained during growth conditions in the field. Such studies should also take into consideration possible shifts in UV -B sensitivity for some populations. Compared to the short-term screening experiment, the cultivar 'Prestige' in particular appeared more UV -B-sensitive during growth in the longer-term UV -B study (Fig. 9-3). Thus longer-term UV -B exposure 9. DisclIssion 1 87 could have overcome initial UV -B tolerance in this cultivar which might be related to some UV -B adaptation in its habitat of origin (Table 3- 1 ). UV -B tolerance in 'Luclair' should be further explored, also in regard to the optical measurements mentioned above. It would be tempting to investigate the possible sunscreen role for translucent 'Luclair' leaves by measuring UV -B transmission below the clear leaf regions. UV -B tolerance in 'Syrian' was accompanied by relatively low flavonol accumulation. 'Syrian' was selected for tolerance against leaf rust and such tolerance has been related to higher constitutive polyamine levels (Bharti and Sawhney, 1996) as well as enhanced oxidase activity, including peroxidases (Sharma and Sharma, 1 997). Peroxidases and polyamines have been implicated in the possible alleviation of UV -B-induced oxidative stress (Kramer et al . , 1 99 1 ; Tekchandani and Guruprasad, 1 998). Specific selection of 'Syrian' could have led to the preference for one UV -B-related antioxidant system at the cost of another. Interaction of UV-B with additional stress, disturbance and competition From the stress interaction studies in this thesis it can be suggested that white clover growth may be less affected by UV -B during periods of drought, compared to moist seasonal conditions. However, it appears that a possible beneficial interaction may only come to effect during more extended drought periods. Climate change projections of enhanced UV -B irradiation effects will increasingly need to take into account alterations in patterns and amounts of moisture supply. The observed UV -B-induced decreases in leaf area as well as increases in proline levels , leaf water status, rootshoot ratio and stomatal density could have beneficial effects for subsequent stages of recovery from drought. Considering its usual association with ryegrass in the New Zealand pasture ecosystems - and the relative insensitivity of ryegrass to UV-B observed in concomitant studies (Campbell and Hofmann, 1 998) - competitive relationships could also be affected by UV -B, and these might in turn be further modified in dependency to water availability (Chapter 1 ) . The findings also indicate a need for future studies examining interaction of UV-B with defoliation, particularly in pasture plants such as white clover that are regularly subjected to grazing under field conditions. Furthermore the results from bioassay studies indicate that UV -B herbivory studies should examine genotypic effects 9. Discussion 1 88 in both plants and animals. Close scrutiny should be gIven to potential impacts on herbivory via UV -B-induced increases in secondary metabolites, including cyanogenesis. Long-term studies would be of value to test possible cumulative effects on herbivory and reproductive success over several insect generations. As concerns the observed time dependency of the UV -B response under continuous UV -B exposure i t would be important to test whether this interaction undergoes modifications during the daily and seasonal UV-B t1uctuations in the field. Should marked UV-B-induced biochemical changes on major plant nutrients also occur in the field, then this could potentially impact on mammalian herbivores such as l arger ruminants normally feeding on white clover. Decreased carbohydrate levels could affect white clover palatability and suitability as silage. The high levels of nitrogen present naturally in white clover are known to lead to inefficiencies in the utilisation of herbage protein in ruminants (Wanjaiya et al., 1 993) and a further UV -B-induced increase could enhance this effect. Furthermore, an increase in nitrogen levels could lead to lower utilisation of NDF by ruminants (Wanjaiya et aI., 1 993), potentially intensifying effects from UV -B-induced decreases in 'Huia' NDF levels observed here. The cyanogenesis levels in 'Huia' can be classified as moderate-high but stil l ranging below levels considered as critical for ruminant grazing (Caradus and Woodfield, 1 997 ; Crush and Caradus, 1 995). Marked UV -B-induced increases of cyanogenesis will have to be investigated as there is a potential that sufficiently high levels in white clover diets could cause cyanide poisoning and nutrient deficiency in ruminants. Possible consequences could include decreased wool production in sheep, goitrogeny and nutritional myopathy (Caradus et al., 1 995 ; Gutzwiller, 1 993; Lehmann et al., 1 99 1 ). Should potential increases in cyanogenesis under UV -B remain below the sensitivity thresholds for mammalian herbivores, there may however still be beneficial consequences via possible insecticidal cyanogenic effects. Taken together these findings point to a number of avenues for future research, from metabolic processes l inked to specific t1avonoid synthesis to the population biology of UV -B effects and resulting impacts on pasture ecosystems. l O. References 1 89 Chapter 10. References Alien DJ, McKee IF, Farage PK, Baker NR. 1997. Analysis of l imitations to CO2 assimilation on exposure of leaves of two Brassfca napus cultivars to UV-B. Plant, Cell and Environment 20: 633-640. AHen DJ, Nogues S, Baker NR. 1998. Ozone depletion and increased UV-B radiation: is there a real threat to photosynthesis? Journal of Experirnental Botany 49: 1 775- 1 788 . AlIen DJ, Nogues S, Morison JIL, Greenslade PD, McLeod AR, Baker NR. 1999. A thirty percent increase in UV -B has no impact on photosynthesis in well-watered and droughted pea plants in the field. Global Change Biology 5: 235-244. AI-Oudat M, Baydoun SA, Mohammad A. 1998. Effects of enhanced UV-B on growth and yield of two Syrian crops wheat (Triticum durum var. Horani) and broad beans (Vicia faba) under field conditions. Environmental and Experimental Botany 40: 1 1 - 1 6. Antonelli F, Grifoni D, Sabatini F, Zipoli G. 1997. Morphological and physiological responses of bean plants to supplemental UV radiation in a Mediterranean climate. Plant Ecology 128: 1 27 - 1 36. Balakumar T, Vincent VHB, PaliwaI K. 1993. On the interaction of UV-B radiation (280-3 1 5 nm) with water stress in crop plants. Physiologia Plantarum 87: 2 1 7- 222. Ballare CL, Barnes PW, Flint SD. 1995. Inhibition of hypocotyl elongation by ultraviolet-B radiation in de-etiolating tomato seedlings. 1 . The photoreceptor. Physioiogia Plantarum 93: 584-592. Barabas KN, Szegletes Z, Pestenacz A, Fulop K, Erdei L. 1998. Effects of excess UV -B irradiation on the antioxidant defence mechanisms in wheat (Triticum aestivum L.) seedlings. Journal (�f Plant Physiology 153: 1 46- 1 53 . Barbour M. 1996. Effects �f water stress on the growth and performance of white clover. MSc Thesis. Hamilton : The University of Waikato. Barbour M, Caradus JR, Woodfield DR, Silvester WB. 1996. Water stress and water use efficiency of ten white clover cultivars. In : Woodfield DR, ed. White clover: New Zealand's competitive edge. Christchurch: Agronomy Society of New Zealand, 1 59- 162. Barker DJ, Caradus JR, McManus MT. 1 993. Physiological responses of white clover genotypes to water deficit. Proceedings of the XVII International Grassland Congress 67-68. / O. References 1 90 Barnes PW, Flint SD, Caldwell MM. 1987. Photosynthesis damage and protective pigments in plants from a latitudinal arctic/alpine gradient exposed to supplemental UV -B radiation in the field. Arctic and Alpine Research 19 : 2 1 -27. Barnes PW, Flint SD, Caldwell MM. 1995. Early-season effects of supplemented solar UV -B radiation on seedling emergence, canopy structure, simulated stand photosynthesis and competition for light. Global Change Biology 1 : 43-53 . Barnes PW, Maggard S, Holman SR, Vergara BS. 1993. lntraspecific variation in sensitivity to UV -B radiation in rice. Crop Science 33: 104 1 - 1 046. Beerling D, Chaloner W. 1992. Stomatal density as an indicator of atmospheric CO2 concentration. The Holocene 2: 7 1 -78 . Beggs CJ, Schneider-Ziebert U, Wellmann E. 1986. UV-B radiation and adaptive mechanisms in plants . In : Worrest RC, Caldwell CC, eds. Stratosperic ozone reduction, solar ultraviolet radiation and plant life. Berlin : Springer, 235-250. Bharti MVR, Sawhney RN. 1996. Involvement of polyamines in resistance of wheat to Puccinia recondita. Phytochemistry 43: 1 009- 10 1 3 . Bi JL, Felton GW, Murphy JB, Howles PA, R.A. D, Lamb CJ. 1997. Do plant phenolics confer resistance to special ist and generalist insect herbivores. journal of Agricultural and Food Chemistry 45: 4500-4504. Biggs RH, Kossuth SV, Teramura AH. 1981. Response of 1 9 cultivars of soybeans to ultraviolet-B irradiance. Physi% gia Plantarum 53: 1 9-26. Bjorn LO, Callaghan TV, Johnsen I, Lee JA, Manetas Y, Paul ND, Sonesson M, Wellburn AR, Coop D, Heide Jorgensen HS, Gehrke C, Gwynn Jones D, Johanson U, Kyparissis A, Levizou E, Nikolopoulos D, Petropoulou Y, Stephanou M. 1997. The effects of UV-B radiation on European heathland species. Plant Ecology 128: 253-264. BJakeney AB, Mutton LL. 1980. A simple colorimetric method for the determination of sugars in fruit and vegetables. journal of the Science of Food and Agriculture 3 1 : 888-897 . Bodeker GE, McKenzie RL. 1996. An algorithm for inferring surface UV irradiance including cloud effects . journal of Applied Meteorology 35: 1 860- 1 877. Bongue Bartelsman M, Phillips DA. 1995. Nitrogen stress regulates gene expression of enzymes in the flavonoid biosynthetic pathway of tomato. Plant Physiology and Biochemistry 33: 539-546. Bornman JF, Reuber S, Cen YP, Weissenbock G. 1997. Ultraviolet radiation as a stress factor and the role of protective pigments. In : Lumsden PJ, ed. Plants and UV-B - responses to environmental change. Cambridge: Cambridge University Press, 1 57- 1 68 . /0. References 1 9 1 Bornman JF, Teramura AH. 1993. Effects of ultraviolet-B radiation on terrestrial plants. In : Young AR, Bjorn LO, Moan J, Nultsch W, eds. Environmental UV Photobiology. New York: Plenum Press, 427-47 1 . Bors W, Michel C, Schikora S. 1995. Interaction of tlavonoids with ascorbate and determination of their univalent redox potentials : A pulse radiolysis study. Free Radical Biology and Medicine 19: 45-52. Bradshaw AD. 1965. Evolutionary significance of phenotypic plasticity m plants. Advances in Genetics 13: 1 1 5 - 1 55 . Brandle JR, Campbell WF, Sisson WB, CaldweH MM. 1977. Net photosynthesis, electron transport capacity, and ultrastructure of Pisum sativum L. exposed to ultraviolet-B radiation. Plant Physiology 60: 1 65- 1 69. Brattsen LB, Samuelian JH, Long KY, Kincaid SA, Evans CK. 1983. Cyanide as a feeding stimulant for the southern armyworm, Spodoptera eridiana. Ecological Entomology 8: 1 25- 1 32. Brinksma EJ, Meijer YJ, Connor BJ, Manney GL, Bergwerff JB, Bodeker GE, Boyd IS, Liley JB, Hogervorst W, Hovenier JW, Livesey NJ, Swart DPJ. 1998. Analysis of record-low ozone values during the 1997 winter over Lauder, New Zealand. Geophysical Research Letters 25: 2785-2788. Brito G, Lopes VC, Caldeira GN. 1997. Morphological and biochemical responses of maize calli to UV -C rays: interaction of UV -C radiation with ion content and peroxidase and alpha-amylase activity. Phyton Buenos Aires 6 1 : 37-44. Britt AB. 1999. Molecular genetics of DNA repair in higher plants. Trends in Plant Science 4: 20-25 . Britz SJ, Adamse P. 1994. UV -B-induced increase in specific leaf weight of cucumber as a consequence of increased starch content. Photochemistry and Photobiology 60: 1 1 6- 1 19 . Brock JL, Caradus JR, Hay MJM. 1989. Fifty years of white clover research in New Zealand. Proceedings of the New Zealand Grassland Association 50: 25-39. Burger J, Edwards GE. 1996. Photosynthetic efficiency, and photodamage by UV and visible radiation, in red versus green leaf Coleus varieties. Plant and Cell Physiology 37: 395-399. Burnip GM, Suckling DM, Shaw PW, White V, Walker JTS. 1995. Monitoring Graphania mutans (Noctuidae) in apple orchards. In : Proceedings of the 48th New Zealand Plant Protection Conference 1995. 1 25- 1 29. Caasi-Lit M, Whitecross MI, Nayudu M, Tanner GJ. 1997. UV-B irradiation induces differential leaf damage, ultrastructural changes and accumulation of specific phenolic compounds in rice cultivars. Australian Journal of Plant Physiology 24: 26 1 -274. 10. References 1 92 Caldwell M, Teramura AH, Tevini M, Bornman JF, Bjorn LO, Kulandaivelu G. 1995. Effects of increased solar ultraviolet radiation on terrestrial plants. Ambio 24: 1 66- l 73 . Caldwell MM. 1968. Solar ultraviolet radiation a s an ecological factor for alpine plants . Ecological Monographs 38: 234-268. Caldwell MM. 1971 . Solar UV radiation and the growth and development of higher plants. In : Giese AG, ed. Phytophysiology. New York: Academic Press, 1 3 1 - 1 77. Caldwell MM, Bjorn LO, Bornman JF, Flint SD, Kulandaivelu G, Teramura AH, Tevini M. 1998. Effects of increased solar ultraviolet radiation on terrestrial ecosystems. Journal (�f Photochemistry and Photobiology B - Biology 46: 40-52. Caldwell MM, Teramura AH, Tevini M. 1989. The changing solar ultraviolet climate and the ecological consequences for higher plants. Trends in Ecology and Evolution 4: 363-366. Campbell BD, Diaz SM. 1998. Testing southern hemisphere grasses for sensitivity to UV-B radiation. Report to the New Zealand Ministry for Research, Science and Technology, Contract Number 98-BRAP-J3-CAMP. Palmerston North: AgResearch. Campbell BD, Grime JP. 1993. Prediction of grassland plant responses to global change. In : Baker MJ , ed. Grasslands for our world. Wellington : SIR Publishing, 406-4 1 5 . Campbell BD, Hofmann RW. 1998. UV-B effects on pastures. In: UV radiation and its effects - an update. Wellington: Royal Society of New Zealand, 3 1 -33 . Campbell BD, Hofmann RW, Hunt CL. 1999. UV-B effects on New Zealand pasture ecosystems. In: Rozema J, ed. Strato�]Jheric ozone depletion. The effects of enhanced UV-B radiation on terrestrial ecosystems. Leiden: Backhuys Publishers, 227-249. Caradus JR. 1986. World checklist of white clover varieties. New Zealand Journal of Experimental Agriculture 14: 1 19- 1 64. Caradus JR. 1994. Genetic diversity within white clover (Trifolium repens L.) . Proceedings Annual Conference Agronomy Society of New Zealand 24: 1 -7 . Caradus JR, Cooper B, Widdup K, Ryan D. 1990a. Breeding and selection for improved white clover production and persistence in New Zealand. Proceedings Agronomy Society of New Zealand 20: 1 1 - 1 5 . Caradus JR, Eerens JPJ. 1992. Genetic adaptation to frost tolerance in white clover. Proceedings Annual Conference Agronomy Society �f New Zealand 22: 1 03- 109. Caradus JR, Forde MB. 1996. Characterisation of white clover populations collected from the Caucasus and high altitude regions of eastern Turkey. Genetic Resources and Crop Evolution 43: 143- 1 55 . 10. References 1 93 Caradus JR, Hay MJM, MacKay AD, Thomas VJ, Ounlop J, Lambert MG, Hart AL, van den Bosch J, Wewa)a S. 1993a. Variation within white clover (Tr�folium repens L.) for phenotypic plasticity of morphological and yield related characters, induced by phosphorus supply. New Phytologist 123: 1 75- 1 84. Caradus JR, MacKay AC, Charlton JFL, Chapman OF. 1990b. Genecology of white clover (Tr(folium repens L.) from wet and dry hill country pastures. New Zealand Journal (�fAgricultural Research 33: 377-384. Caradus JR, MacKay AC, van den Bosch J, Greer OH, Wewala GS. 1989a. Intraspecific variation for frost hardiness in white clover. Journal of Agricultural Science 1 12 : 1 5 1 - 1 57 . Caradus JR, MacKay AC, Woodfield DR, van den Bosch J, Wewala S . 1 989b. Classification of a world collection of white clover cultivars. Euphytica 42: 1 83- 1 96. Caradus JR, MacKay AO, Pritchard MW. 1987. Towards improving the aluminium tolerance of white clover. Proceedings of the New Zealand Grassland Association 48: 1 63- 1 69. Caradus JR, McNabb W, Woodfield OR, Waghorn GC, Keogh R. 1995. Improving quality characteristics of white clover. Proceedings Annual Conference Agronomy Society of New Zealand 25 : 7- 12. Caradus JR, Pinxterhuis JB, Hay RJM, Lyons T, Hog)und JH. 1993b. Response of white clover cultivars to fertiliser nitrogen. New Zealand Journal �f Agricultural Research 36: 285-295. Caradus JR, Wood field OR. 1997. World checklist of white clover varieties n. New Zealand Journal of Agricultural Research 40: 1 1 5-206. Caradus JR, Woodfield OR, Stewart AV. 1996. Overview and vision for white clover. In : Woodfield DR, ed. White clover: New Zealand's competitive edge. Christchurch: Agronomy Society of New Zealand, 1 -6. Carlo-Erba. 1988. Nitrogen Analyser 1500 Series II Instruction Manual. Milan : Carlo­ Erba. Carlson GE. 1 966. Growth of clover leaves - developmental morphology and parameters at ten stages. Crop Science 6: 293-294. Cen YP, Bornman JF. 1990. The response of bean plants to UV-B radiation under different irradiances of background visible l ight. Journal of Experimental Botany 41 : 1 489- 1 495. Cen YP, Bornman JF. 1993. The effect of exposure to enhanced UV -B radiation on the penetration of monochromatic and polychromatic UV -B radiation in leaves of Brassica /lapus. Physiologia Plantarum 87: 249-255 . J O. References 1 94 Chalker-Scott L. 1999. Environmental significance of anthocyanins in plant stress responses. Photochemistry and Photobiology 70: 1 -9 . Chapman RB. 1984. Pasture pests. In: Scott RR, ed. New Zealand pest and beneficial insects. Canterbury: Lincoln University College of Agriculture, 1 1 9- 1 67. Chaves N, Escudero JC, Gutierrezmerino C. 1997. Role of ecological variables in the seasonal variation of flavonoid content of Cistus ladan�fer exudate. Journal of Chemical Ecology 23: 579-603. Cline MG, Salisbury FB. 1966. Effects of ultraviolet radiation on the leaves of higher plants. Radiation Botany 6: 1 5 1 - 1 63 . Conn EE. 1979. Cyanide and cyanogenic glycosides. I n : Rosenthal GA, lanzen DH, eds. Herbivores: Their interactions with secondary plant metabolites. New York: Academic Press, 387-4 1 2. Conner JK, Zangori LA. 1997. A garden study of the effects of ultraviolet-B radiation on pollination success and lifetime female fitness in Brassica. Oecologia 1 1 1 : 388-395. Conner JK, Zangori LA. 1998. Combined effects of water, nutrient, and UV-B stress on female fitness in Brassica (Brassicaceae). American Journal of Botany 85: 925-93 1 . Coop lE. 1 986. Pasture and crop production. In: McCutcheon SN, DcDonald MF, Wickham GA, eds. Sheep production, Volume I/, feeding, growth and health. Wellington: New Zealand Institute of Agricultural Science, 1 10- 1 36. Cooper-Driver GA, Bhattacharya M. 1998. Role of phenolics in plant evolution. Phytochemistry 49: 1 1 65 - 1 1 74. Corkill L. 1940. Cyanogenesis in white clover (Trifolium repens L.). I . Cyanogenesis in single plants. New Zealand lournal of Science and Technology 22B: 65-67. Corkill L. 1963. A white clover strain with a distinguishing leaf mark. New Zealand lournal of Agricultural Research 6: 457-459. Corlett JE, Stephen J, Jones HG, Woodfin R, Mepsted R, Paul ND. 1997. Assessing the impact of UV -B radiation on the growth and yield of field crops. In: Lumsden Pl, ed. Plants and UV-B - responses to environmental change. Cambridge: Cambridge University Press, 1 95-2 1 1 . Correia CM, Area) EL V, Torres-Pereira MS, Torres-Pereira JMG. 1998. Intraspecific variation in sensitivity to ultraviolet-B radiation in maize grown under field conditions. I. Growth and morphological aspects. Field Crops Research 59: 8 1 -89. lO. References 1 95 Correia CM, Areal EL V, Torres-Pereira MS, Torres-Pereira JMG. 1999a. lntraspecific variation in sensitivity to ultraviolet-B radiation in maize grown under field conditions - n. Physiological and biochemical aspects. Field Crops Research 62: 97- 1 05 . Correia CM, Torres-Pereira MS, Torres-Pereira JMG. 1999b. Growth, photosynthesis and UV-B absorbing compounds of Portuguese Barbela wheat exposed to ultraviolet-B radiation. Environrnental Pollution 104: 383-388. Crush JR. 1987. Nitrogen fixation. In: Williams WM, Baker MJ, eds. White clover. Wallingford: CAB, 1 86-20 1 . Crush JR, Caradus JR. 1995. Cyanogenesis potential and iodine concentration in white clover (Trifolium repens L.) cultivars. New Zealand Journal of Agricultural Research 38: 309-3 1 6. Crutzen PJ. 1972. SST's - a threat to the earth's ozone shield. Ambio 1 : 4 1 -5 1 . Daday H. 1965. Gene frequencies in wild populations of Tr(folium repens L. IV. Mechanism of natural selection. Heredity 20: 355-365. Dai Q, Peng A, Chavez AQ, Vergara BS. 1994. Intraspecific responses of 1 88 rice cultivars to enhanced UVB radiation. Environmental and Experimental Botany 34: 433-442. Dai Q, Peng S, Chavez AQ, Vergara BS. 1995. Effects of UVB radiation on stomatal density and opening in rice (Oryza sativa L.). Annals (�f Botany 76: 65-70. Dai QJ, Coronel VP, Vergara BS, Barnes PW, Quintos AT. 1992. Ultraviolet-B radiation effects on growth and physiology of four rice cultivars. Crop Science 32: 1 269- 1 274. Daie J. 1988. Mechanism of drought induced alterations in assimilate partitioning and transport in crops. Critical Reviews in Plant Sciences 7: 1 1 7 - 1 37 . Dakora FD. 1995. Plant flavonoids: biological molecules for useful exploitation. Australian Journal qf Plant Physiology 22: 87-99. Dawar S, Vani T, Singhal GS. 1998. Stimulation of antioxidant enzymes and lipid peroxidation by UV-B irradation in the thylakoid membranes of wheat. Biologia Plantarum 4 1 : 65-73. Day TA, Demchik SM. 1996. Influence of enhanced UV -B radiation on biomass allocation and pigment concentrations in leaves and reproductive structures of greenhouse-grown Brassica rapa. Vegetatio 127 : 1 09- 1 1 6. De Britto AJ. 1995. Impact of ultraviolet radiation on the pulse Phaseolus mungo. Journal qf Ecotoxicology and Environmental Monitoring 5: 1 1 3- 1 17 . Deckmyn G, Impens I . 1995. UV -B increases the harvest index of bean (Phaseolus vulgaris L.). Plant, Cell and Environment 18 : 1 426- 1433. /0. R(�ferences 1 96 Deckmyn G, Impens I. 1998. UV-B and PAR in a grass (Lolium perenne L.) canopy. Plant Ecology 137: 1 3- 1 9. Delong JM, Steffen KL. 1997. Photosynthetic function, l ipid peroxidation, and alpha­ tocopherol content in spinach leaves during exposure to UV-B radiation. Canadian Journal of Plant Science 77: 453-459. Diaz S, Cabido M. 1997. Plant functional types and ecosystem function in relation to global change. Journal qf Vegetation Science 8: 463-474. Diaz S, Cabido M, Casanoves F. 1999. Functional implications of trait-environment linkages in plant communities. In: Weiher E, Keddy P, eds. Ecological assembly rules. Cambridge: Cambridge University Press, 338-362. Dixon RA, Paiva NL. 1995. Stress-induced phenylpropanoid metabolism. Plant Cell 7 : 1 085- 1097. Drilias P, Karabourniotis G, Levizou E, Nikolopoulos D, Petropoulou Y, Manetas Y. 1997. The effects of enhanced UV -B radiation on the Mediterranean evergreen sclerophyll Nerium oleander depend on the extent of summer precipitation. Australian Journal of Plant Physiology 24: 301 -306. Dumpert K, Knacker T. 1985. A comparison of the effects of enhanced UV-B radiation on some crop plants exposed to greenhouse and field conditions. Biochemie und Physiologie der Pflanzen 1 80: 599-6 1 2. Ellsbury MM, Pederson GA, Fairbrother TE. 1992. Resistance to fol iar-feeding hypergine weevils (Coleoptera: Curculionidae) in cyanogenic white clover. Journal qfEconomic Entomology 85: 2467-2472. Eswaran K, Premkumar A, Kulandaivelu G. 1993. Impact of enhanced UV -B on photosynthetic and biochemical characteristics of maize under water stress. Plant Physiology and Biochemistry New Delhi 20: 36-40. Fagerberg WR, Bornman J);.... 1997. Ultraviolet-B radiation causes shade-type ultrastructural changes in Brassica napus. Physiologia Plantarum 101 : 833-844. Farman JC, Gardiner BG, Shankli JD. 1985. Large losses of total ozone in Antarctica reveal seasonal CIOx - NOx interaction. Nature 315: 207-2 10. Feldheim K, Conner JK. 1996. The effects of increased UV-B radiation on growth, pollination success, and lifetime female fitness in two Brassica species. Oecologia 106: 284-297. Ferreira C, Torres BB, Terra WR. 1998. Substrate specificities of midgut beta­ glycosidases from insects of different orders. Comparative Biochemistry and Physiology B-Biochemistry and Molecular Biology 119 : 2 1 9-225. Filella I, Penuelas J. 1999. Altitudinal differences in UV absorbance, UV reflectance and related morphological traits of Quercus ilex and Rhododendron ferrugineum in the Mediterranean region. Plant Ecology 145 : 1 57- 1 65 . 10. References 1 97 Fischbach RJ, Kossmann B, Panten H, Steinbrecher R, Helier W, Seidlitz HK, Sandermann H, Hertkorn N, Schnitzler JP. 1999. Seasonal accumulation of ultraviolet-B screening pigments in needles of Norway spruce (Picea abies (L.) Karst.) . Plant, Cell and Environment 22: 27-37. ]j'iscus EL, Booker FL, Miller JE. 1996. Response of soybean bulk leaf water relations to ultraviolet-B irradiation. Journal (�lPlant Physiology. 148: 63-68. Flint SD, Caldwell MM. 1984. Partial inhibition of in vitro pollen germination by simulated solar ultraviolet-B radiation. Ecology 65 : 792-795 . Forde MB. 1987. Report (�f China field collection expedition 1 987. Palmerston North: Grasslands Research Centre. Forde MB. 1989. Report of Caucasus expedition 1 989. Palmerston North: DSIR. Forde MB, Easton HS. 1986. Report of the New Zealand - France - IBPGR forage germplasm collecting expedition to southwest Europe 1 986. Palmerston North: Grasslands Research Centre. Forgan BW, Liley JB. 1998. Aerosols and UV in New Zealand. In : UV radiation and its effects - an update. Christchurch: The Royal Society of New Zealand, 5 1 -54. Furness MH, Upadhyaya MK, Ormrod DP. 1999. Seedling growth and leaf surface morphological responses of three rangeland weeds to ultraviolet-B radiation. Weed Science 47: 427-434. Gehrke C, Johanson U, CaUaghan TV, Chadwick D, Robinson CH. 1995. The impact of enhanced ultraviolet-B radiation on litter quality and decomposition processes in Vaccinium leaves from the subarctic. Oikos 72: 2 1 3-222. Genstat. 1993. Genstat 5, Release 3 Reference Manual. Oxford: Clarendon Press. Giese AC. 1964. Studies on ultraviolet radiation action upon animal cells . In: Giese AC, ed. Photophysiology. New York: Academic Press, 203-245. Gonzalez R, Paul ND, Percy K, Ambrose M, McLaughlin CK, Barnes JD, Areses M, Wellburn AR. 1996. Responses to ultraviolet-B radiation (280-3 1 5 nm) of pea (Pisum sativum) lines differing in leaf surface wax. Physiologia Plantarum 98: 852-860. Gonzalez R, Well burn AR, Paul ND. 1998. Dose responses of two pea lines to ultraviolet-B radiation (280-3 1 5 nm). Physiologia Plantarum 104: 373-378 . Gorton HL, Vogelmann TC. 1996. Effects of epidermal cell shape and pigmentation on optical properties of Antirrhinum petals at visible and ultraviolet wavelengths. Plant Physiology 1 12 : 879-888. Grace SC, Logan BA, Adams WW, 111. 1 998. Seasonal differences in foliar content of chlorogenic acid, a phenylpropanoid antioxidant, in Mahonia repens. Plant, Cell and Environment 21 : 5 1 3-52 1 . JO. References 1 98 Grant-Petersson J, Renwick JAA. 1996. Effects of ultraviolet-B exposure of Arahidopsis thaliana on herbivory by two crucifer-feeding insects (Lepidoptera). Environmental Entomology 25 : 1 35 - 142. Greer DH. 1995a. Effect of canopy position on the susceptibility of kiwifruit (Actinidia deliciosa) leaves on vines in an orchard environment to photoinhibition throughout the growing season. Australian Journal of Plant Physiology 22 : 299- 309. Greer DH. 1995b. Effect of daily photon receipt on the susceptibility of dwarf bean (Phaseolus vulgaris L. ) leaves to photoinhibition of photosynthesis. Planta 197: 3 1 -38 . Greer DH, Laing WA. 1992. Photo inhibition of photosynthesis in intact kiwifruit (Actinidia deliciosa) leaves : changes in susceptibility to photoinhibition and recovery during the growth season. Planta 186 : 4 1 8-425. Grime JP. 1 979. Plant strategies and vegetation processes. Chichester: Wiley. Grime JP, Campbell BD. 1991. Growth rate, habitat productivity and plant strategy as predictors of stress response. In: Mooney HA, Winner WE, Pell El, eds. Response of plants to multiple stresses. Orlando: Academic Press, 1 43- 1 59. Grime JP, Thompson K, Hunt R, Hodgson JG, Cornelissen JHC, Rorison IH, Hendry GAF, Ashenden TW, Askew AP, Band SR, Booth RE, Bossard CC, Campbell BD, Cooper JEL, Davison A W, Gupta PL, Hall W, Hand DW, Hannah MA, Hillier SH, Hodkinson DJ, Jalili A, Liu Z, Mackey JML, Matthews N, Mowforth MA, Neal AM, Reader R.I, Reiling K, Ross-Fraser W, Spencer RE, Sutton F, Tasker DE, Thorpe PC, Whitehouse J. 1 997. Integrated screening validates primary axes of specialisation in plants. Oikos 79: 259-28 1 . Groth JV. 1998. Plant population genetics. In : Krupa SV, Kickert RN, lager Hl, eds. Elevated ultraviolet (UV)-B radiation and agriculture. Berlin : Springer, 247-262. GutzwilJer A. 1993. The effect of a diet containing cyanogenetic glycosides on the selenium status and the thyroid function of sheep. Animal Production 57: 4 1 5- 4 1 9. Gwynn-Jones D, Johanson U. 1996. Growth and pigment production in two subarctic grass species under four different UV -B irradiation levels. Physiologia Plantarum 97: 701 -707. Gwynn-Jones D, Johanson U, Phoenix GK, Gehrke C, Callaghan TV, Bjorn LO, Sonesson M, Lee JA. 1999a. UV -B impacts and interactions with other co­ occurring variables of environmental change: an arctic perspective. In : Rozema J , ed . Stratosperic ozone depletion. The effects of enhanced UV-B radiation on terrestrial ecosystems. Leiden : Backhuys Publishers, 1 87-20 l . 10. References 1 99 Gwynn-Jones D, Lee JA, Johanson U, Phoenix GK, Callaghan TV, Sonesson M. 1999b. The responses of plant functional types to enhanced UV-B radiation. In : Rozema J, ed. Stratosperic ozone depletion. The eilects (�l enhanced UV-B radiation on terrestrial ecosystems. Leiden : Backhuys Publishers, 1 73- 1 85 . Haberlandt G. 1914. Physiological plant anatomy. London: Mac-Millan . Hader DP. 1996. Effects of solar radiation on local and German wheat seedlings in a Chilean high mountain station. Journal of Photochemistry and Photobiology. B - Biology 35: 1 8 1 - 1 87 . Hader DP, Kumar HD, Smith RC, Worrest RC. 1998. Effects on aquatic ecosystems. Journal of Photochemistry and Photobiology. B - Biology 46: 53-68. Harris W. 1990. Pasture as an ecosystem. In: Langer RHM, ed. Pastures: their ecology and management. Oxford: Oxford University Press, 75- 1 3 l . Hart AL. 1987. Physiology. In : Williams WM, Baker MJ, eds. White clover. Wallingford: CAB, 1 25- 1 52. Hatcher PE, Paul ND. 1994. The effect of elevated UV-B radiation on herbivory of pea by Autographa gamma. Entomologia Experimentalis et Applicata 7 1 : 227- 233 . He J , Huang LK, Whitecross MI. 1994. Chloroplast ultrastructure changes in Pisum sativum associated with supplementary ultraviolet (UV -B) radiation. Plant, Cell and Environment 17: 77 1 -775. Hideg E, Vass I. 1996. UV -B induced free radical production In plant leaves and isolated thylakoid membranes. Plant Science 115 : 25 1 -260. Hilu K, Randall J. 1984. Convenient method for studying grass leaf epidermis. Taxon 33: 4 1 3-4 1 5 . Hoffmann S . 1999. The effect of photoselective cladding materials o n the growth of ornamental plants H. Effect of UV range on stem elongation. Gartenbauwissenschaft 64: 1 83- 1 89. Hoglund JH, White JGH. 1985. Environmental and agronomic constraints in dryland pasture and choice of species. In: Byrgess RE, Brock JL, eds. Using herbage cultivars. Palmerston North: New Zealand Grasslands Organisation, 39-43. Hoque E, Remus G. 1999. Natural UV-screening mechanisms of Norway spruce (Picea abies L. Karst) needles. Photochemistry and Photobiology 69: 1 77- 1 92. Hsaio TC. 1973. Plant responses to water stress. Annual Review of Plant Physiology 24: 5 19-570. Huang S, Dai Q, Peng S, Chavez AQ, Miranda MLL, Visperas RM, Vergara BS. 1997. Influence of supplemental ultraviolet-B on indoleacetic acid and calmodulin in the leaves of rice (Oryza sativa L.). Plant Growth Regulation 21 : 59-64. 10. References 200 Hunt JE. 1997. Ultraviolet-B radiation and its effects on New Zealand trees. PhD Thesis. Lincoln : Lincoln University. Hunt JE, Kelliher FM, McNeil DL. 1996. Response in chlorophyll a fluorescence of six New Zealand tree species to a step-wise increase in ultraviolet-B i rradiance. New Zealand Journal of Botany 34: 401 -4 1 0. Hunt JE, McNeil DL. 1998. Nitrogen status affects UV -B sensitivity of cucumber. Australian Journal (�lPlant Physiology 25: 79-86. Huttunen S, Kinnunen H, Laakso K. 1998. Impact of increased UV -B on plant ecosystems. Chemosphere 36: 829-833 . Inskeep WP, Bloom RP. 1985. Extinction coefficients of chlorophyll a and b in N,N­ dimethylformamide and 80% acetone. Plant Physiology 77: 483-485. Jagger J. 1967. Introduction to research in ultraviolet photobiology. New Jersey: Prentice Hal l . Jain VK, Goyal AK, Rakesh K, Sharma MM, Kumar R. 1 988. An interaction of UV -B radiation and growth regulators on chlorophyll development in Cucumis utilissimus Roxb. cv. Jaunpuri . Indian Journal of Plant Physiology 31 : 1 74- 1 77. Jansen MAK, Gaba V, Greenberg BM. 1998. Higher plants and UV -B radiation: balancing damage, repair and acclimation. Trends in Plant Science 3: 1 3 1 - 1 35 . Johnston HS. 1971. Reduction of stratospheric ozone by nitrogen oxide catalysts from supersonic transport exhaust. Science 173 : 5 17-522. Jones R, Lowe K. 1993. Managing white clover in subtropical pastures. In : Mason W, ed. White clover: a key to increasing milk yields. Glen Iris : DRDC, 6 1 -64. Jordan BR. 1 996. The effects of ultraviolet-B radiation on plants: a molecular perspective. Advances in Botanical Research 22: 98- 1 62. Jordan BR, James PE, Mackerness SAn. 1998. Factors affecting UV -B-induced changes in Arabidopsis thaliana L. gene expression : The role of development, protective pigments and the chloroplast signal. Plant and Cell Physiology 39: 769- 778. Jordan BR, James PE, Strid A, Anthony RG. 1994. The effect of ultraviolet-B radiation on gene expression and pigment composition in etiolated and green pea leaf tissue: UV-B-induced changes are gene-specific and dependent upon the developmental stage. Plant, Cell and Environment 17: 45-54. Kaiser F. 1995. Auswirkungen erh(jhter Kohlendioxid- und Ozonkonzentrationen, und unterschiedlicher UV-B Strahlung auf den Photosyntheseapparat von Lol ium perenne und Trifolium pratense. PhD Thesis. MUnchen: Ludwig-Maximilians­ U n i versi Hit. Kakes P. 1989. An analysis of the costs and benefits of the cyanogenic system in Trifolium repens. Theoretical and Applied Genetics 77 : 1 1 1 - 1 1 8. / O. References 201 Kakes P. 1 990. Properties and functions of the cyanogenic system in higher plants. Euphytica 48: 25-43 . Karnovsky MJ. 1965. A formaldehyde-glutaraldehyde fixative of high osmolality for use in electron microscopy. Journal o.fCellular Biology 27: 1 37- 1 38 . Katzel R, Loffler S, Grabo G. 1 996. The influence of higher UV-B radiation on chlorophyll fluorescence and selected needle compounds of Scots pine (Pinus sylvestris). Beitrdge fur Forstwirtschaft und Landschaftsokologie 30: 1 32- 1 36. Kershaw KA, Looney JHH. 1985. Quantitative and dynamic plant ecology. London: Edward Arnold. Kozak RG, Ricco RA, Gurni AA, Boveris AD, Puntarulo S. 1999. Antioxidant response of soybean cotyledons (Glycine max) to ultraviolet irradiation. Canadian Journal (�f Plant Science 79: 1 8 1 - 1 89. Kramer GF, Norman HA, Krizek DT, Mirecki RM. 1991. Influence of UV-B radiation on polyamines, l ipid peroxidation and membrane lipids in cucumber. Phytochemistry 30: 2 10 1 -2 108. Krizek DT. 1975. Intluence of ultraviolet radiation on germination and early seedling growth. Physioiogia Plantarum 34: 1 82- 1 86. Krizek DT, Britz SJ, Mirecki RM. 1998. Inhibitory effects of ambient levels of solar UV-A and UV-B radiation on growth of cv. New Red Fire lettuce. Physiologia Plantarum 103: 1 -7 . Krizek DT, Kramer GF, Mirecki RM. 1997a. Influence of UV-B radiation and putrescine on shoot and root growth of cucumber seedlings grown in nutrient solution. Journal o.f Plant Nutrition 20: 6 1 3-623 . Krizek DT, Mirecki RM, Britz SJ. 1997b. Inhibitory effects of ambient levels of solar UV-A and UV-B radiation on growth of cucumber. Physiologia Plantarum 100 : 886-893. Krupa SV, Kickert RN. 1993. The greenhouse effect : the impacts of carbon dioxide (C02), ultraviolet-B (UV-B) radiation and ozone (03) on vegetation (crops). Vegetatio 105 : 223-238 . Krupa SV, Kickert RN, Jager HJ. 1 998a. Integrated view of environment-crop interactions. In : Krupa SV, Kickert RN, Jager HJ, eds. Elevated ultraviolet (UV)­ B radiation and agriculture. Berlin : Springer, 1 8 1 -2 1 8. Krupa SV, Kickert RN, Jager HJ. 1998b. Elevated ultraviolet (UV)-B radiation and agriculture. Berlin : Springer. Kumagai J, Katoh H, Miyazaki T, Hidema J, Kumagai T. 1999. Differences in the sensitivity to UVB radiation of two cultivars of rice (Oryza sativa L.) based on observation of long-lived radicals. Journal of Radiation Research 40: 303-3 10. 10. References 202 Laakso K, Huttunen S. 1998. Effects of the ultraviolet-B radiation (UV -B) on conifers : a review. Environmental Pollution 99: 3 19-328. Laing WA. 1991. The consequences (�f increased ultraviolet-B radiation jor plants. Palmerston North: DSIR. Laing WA. 1993. UVB radiation and New Zealand agriculture. In: RSNZ, ed. UV radiation and its effects. Wellington: The Royal Society of New Zealand, 1 6- 17 . Lambers H, Poorter H. 1992. Inherent variation in growth rate between higher plants : a search for physiological causes and ecological consequences. Advances in Ecological Research 23: 1 87-26 1 . Larsen LM, Nielsen JK, Sorensen H . 1982. Identification of 3-0-[2-0-(beta-D­ xylopyranosyl)-beta-D-galactopyranosyl] f1avonoids in horseradish leaves acting as feeding stimulants for a flea beetle. Phytochemistry 2 1 : 1 029- 1034. Larson RA, Garrison WJ, Carlson RW. 1990. Differential responses of alpine and non-alpine Aqui/egia species to increased ultraviolet-B radiation. Plant, Cell and Environment 13: 983-987. Lavola A. 1 998. Accumulation of flavonoids and related compounds in birch induced by UV -B irradiance. Tree Physiology 18 : 53-58 . Lavola A, Julkunen Tiitto R, Aphalo P, de la Rosa T, Lehto T. 1 997. The effect of u.v.-B radiation on u.v.-absorbing secondary metabolites in birch seedlings grown under simulated forest soil conditions. New Phytologist 137: 6 17-62 1 . Lavola A, Julkunentiitto R, Roininen H, Aphalo P. 1998. Host-plant preference of an insect herbivore mediated by UV -B and CO2 in relation to plant secondary metabolites. Biochemical Systematics and Ecology 26: 1 - 1 2. Lehmann J, Meister E, Gutzwiller A, Jans F, Charles JP, Blum J. 1991. Should one use white clover (Trifolium rep ens L.) varieties rich in hydrogen cyanide? Revue Suisse d 'Agriculture 23: 107- 1 1 2 . Lenoble M, Papineau J. 1970. Note sur une nouvelle marque foliaire chez Trifolium repens. Annales de [ 'amelioration des plantes 20: 485-487. Li Y, Vue M, Wang X. 1998. Effects of enhanced ultraviolet-B radiation on crop structure, growth and yield components of spring wheat under field conditions. Field Crops Research 57: 253-263 . Lindroth RL. 1991. Differential toxicity of plant allelochemicals to insects : roles of enzymatic detoxication systems. In : Bernays EA, ed. Insect-plant interactions. Boca Raton : CRC Press, 1 -33 . Liu L, Gitz DC, Ill, McClure JW. 1995. Effects of UV-B on f1avonoids, ferulic acid, growth and photosynthesis in barley primary leaves. Physiologia Plantarum 93: 725-733. 10. References 203 Lockhart JA. 1961. The effect of ultraviolet radiation on plants. In: Ruhland W, ed. Encyclopedia (�f Plant Physiology XVI. Berlin : Springer, 532-554. Logemann E, Tavernaro A, Schulz W, Somssich lE, Hahlbrock K. 2000. UV light selectively coinduces supply pathways from primary metabolism and flavonoid secondary product formation in parsley. Proceedings �f the National Academy of Sciences 97: 1 903- 1 907. Lois R. 1994. Accumulation of UV -absorbing flavonoids induced by UV -B radiation in Arabidopsis thaliana L. I . Mechanisms of UV -resistance in Arabidopsis. Planta 194: 498-503. Lois R, Buchanan BB. 1994. Severe sensitivity to ultraviolet radiation in an Arabidopsis mutant deficient in flavonoid accumulation. n. Mechanisms of UV­ resistance in Arabidopsis. Planta 194: 504-509. Loponen J, Ossipov V, Lempa K, Haukioja E, Pihlaja K. 1998. Concentrations and among-compound correlations of individual phenolics in white birch leaves under air pollution stress. Chemosphere 37 : 1445 - 1456. Mabry TJ, Markham KR, Thomas MB. 1970. The systematic identification of flavonoids. New York: Springer. Macfarlane MJ, Sheath GW, McGowan AW. 1990. Evaluation of clovers in dry hill country. 5 . White clover at Whatawhata, New Zealand. New Zealand Journal of Agricultural Research 33: 549-556. MacGillivray CW, Grime JP. 1995. Testing predictions of the resistance and resilience of vegetation subjected to extreme events. Functional Ecology 9 : 640- 649. MacKay AD, Caradus JR, Hart AL, Wewala GS, Dunlop J, Lambert MG, van den Bosch J, Mouat MCH. 1990. Phosphorus uptake characteristics of a world collection of white clover (Tr�folium repens) cultivars. Developments in Plant and Soil Sciences 41 : 655-658 . Mackerness SAH, Jordan BR. 1 999. Changes in gene expression lI1 response to ultraviolet-B-induced stress. In: Pessarakli M , cd. Handbook �f plant and crop stress. New York: Dekker, 749-768. Mackerness SAH, Surplus SL, Blake P, John CF, Buchanan-Wollaston V, Jordan BR, Thomas B. 1999. Ultraviolet-B-induced stress and changes in gene expression in Arabidopsis thaliana: role of signalling pathways controlled by jasmonic acid, ethylene and reactive oxygen species. Plant, Cell and Environment 22: 1 4 1 3- 1 423. Mackerness SAH, Surplus SL, Jordan BR, Thomas B. 1997. Ultraviolet-B effects on transcript levels for photosynthetic genes are not mediated through carbohydrate metabolism. Plant, Cell and Environment 20: 143 1 - 1437. 10. References 204 Mackerness SAH, Surplus SL, Jordan BR, Thomas B. 1998. Effects of supplementary ultraviolet-B radiation on photosynthetic transcripts af different stages of leaf development and light levels in pea (Pisum sativum L.): role of active oxygen species and antioxidant enzymes. Photochemistry and Photobiology 68: 88-96. Mackerness SAH, Thomas B. 1999. Effects of UV-B radiation on plants : gene expression and signal transduction pathways. In: Smallwood MF, Calvert CM, Bowles DJ, eds. Plant reS1}()flses to environmental stress. Oxford: Bios, 1 7-24. Madronich S. 1992. Implications of recent total atmospheric ozone measurements for biologically active ultraviolet radiation reaching the earth's surface. Geophysical Research Letters 19 : 37-40. Madronich S, McKenzie RL, Caldwell M, Bjorn LO. 1995. Changes in ultraviolet radiation reaching the earth's surface. Ambio 24: 143- 1 52. Magne C, Larher F. 1992. High sugar content of extracts interferes with colorimetric determination of amino acids and free proline. Analytical Biochemistry 200: l 1 5- 1 1 8 . Manetas Y , Petropoulou Y, Stamatakis K , Nikolopoulos D , Levizou E , Psaras G, Karabourniotis G. 1997. Beneficial effects of enhanced UV -B radiation under field conditions: improvement of needle water relations and survival capacity of Pinus pinea L. seedlings during the dry Mediterranean summer. Plant Ecology 128: 1 00- 1 08 . Manly BFJ. 1994. Multivariate statistical methods. London : Chapman and Hal l . Mark U, Tevini M. 1997. Effects of solar ultraviolet-B radiation, temperature and CO2 on growth and physiology of sunflower and maize seedlings. Plant Ecology 128: 224-234. Markham KR. 1982. Techniques offlavonoid identification. London: Academic Press. Markham KR, Ryan KG, Bloor SJ, Mitchell KA. 1998a. An increase in the luteolin : apigenin ratio in Marchantia polymorpha on UV -B enhancement. Phytochemistry 48: 79 1 -794. Markham KR, Tanner GJ, Caasi Lit M, Whitecross MI, Nayudu M, Mitchell KA. 1998b. Possible protective role for 3 ' ,4'-dihydroxyflavones induced by enhanced UV-B in a UV-tolerant rice cultivar. Phytochemistry 49: 1 9 1 3- 1 9 19 . Matthew C, Hofmann A, Rapson GL, McKenzie RL, Kemp PD, Osborne MA. 1996. Growth of ryegrass and white clover under canopies with contrasting transmission of ultraviolet-B radiation. Proceedings Agronomy Society of New Zealand 26: 23-30. 10. Relerences 205 Matthew C, Lawoko CRO, Korte CJ, Smith D. 1994. Application of canonical discriminant analysis, principal component analysis, and canonical correlation analysis as tools for evaluating differences in pasture botanical composition. New Zealand Journal of Agricultural Research 37: 509-520. Matthews WA. 1998. Ozone changes, causes and expected future changes. In: UV radiation and its e.ffects - an update. Christchurch: The Royal Society of New Zealand, 49-50. Mattson WJ. 1980. Herbivory in relation to plant nitrogen content. Annual Review of Ecology and Systematics 1 1 : 1 1 9- 1 6 1 . Mazza CA, Zavala J, Scopel AL, Ballare CL. 1999. Perception of solar UVB radiation by phytophagous insects : Behavioral responses and ecosystem implications. Proceedings (�f the National Academy of Sciences of the United States of America 96: 980-985. McCloud ES, Berenbaum MR. 1994. Stratospheric ozone depletion and plant-insect interactions: effects of UVB radiation on foliage quality of Citrus jambhiri for Trichoplusia ni. Journal of Chemical Ecology 20: 525-539. Mc Gee fR. 1987. Graphania mutans (Walker) and Acremonium lolii (Latch). The relationship between an insect herbivore and a fungal endophyte of perennial ryegrass. MSc Thesis. Palmerston North: Massey University. McGregor PG, Watts PJ, Esson MJ. 1987. Light trap records from southern North Island hill country. New Zealand Entomologist 10: 104- 1 2 1 . McKenzie R, Conner B , Bodeker G. 1999. Increased summertime UV radiation in New Zealand in response to ozone loss. Science 285: 1 709- 1 7 1 l . McKenzie RL. 1991 . Application of a simple model to calculate latitudinal and hemispheral differences in ultraviolet radiation. Weather and Climate 1 1 : 3- 1 4. McKenzie RL. 1992. Current Status �f UVB Research in New Zealand. Well ington: S IR Publishing. McKenzie RL, Bodeker GE, Keep DJ, Kotkamp M. 1996. UV radiation in New Zealand: north-to-south differences between two sites, and relationship to other latitudes. Weather and Climate 16: 1 7-26. McKenzie RL, Paulin KJ, Kotkamp M. 1997. Erythemal UV irradiances at Lauder, New Zealand - relationship between horizontal and normal incidence. Photochemistry and Photobiology 66: 683-689. Meijkamp B, Aerts R, van de Staaij J, Tosserams M, Ernst WHO, Rozema J. 1 999. Effects of UV-B on secondary metabolites. In: Rozema J, ed. Stratosieric ozone depletion. The effects of enhanced UV-B radiation on terrestrial ecosystems. Leiden: Backhuys Publishers, 7 1 -99. ID. References 206 Melville J, Coop lE, Doak BW, Reifer I. 1940. Cyanogenesis in white clover (Tr�folium repens L.). IV . Methods of determination and general considerations. New Zealand Journal of Science and Technology 22B: 144- 1 54. Mepsted R, Paul ND, Stephen J, Corlett JE, Nogues S, Baker NR, Jones HG, Ayres PG. 1996. Effects of enhanced UV-B radiation on pea (Pisum sativum L) grown under field conditions in the UK. Global Change Biology 2: 325-334. Middleton EM, Chappelle EW, Cannon TA, Adarnse P, Britz SJ. 1996. Initial assessment of physiological response to UV -B irradiation using fluorescence measurements. Journal «f Plant Physiology 148: 69-77 . Middleton EM, Terarnura AH. 1993. The role of flavonol glycosides and carotenoids in protecting soybean from ultraviolet-B damage. Plant Physiology 103 : 741 -752. Miller JE, Booker FL, Fiscus EL, Heagle AS, Pursley WA, Vozzo SF, Heck WW. 1994. Ultraviolet-B radiation and ozone effects on growth, yield, and photosynthesis of soybean. Journal of Environmental Quality 23: 83-9 1 . Mirecki RM, Terarnura AH. 1984. Effects of ultraviolet-B irradiance on soybean. V. The dependence of plant sensitivity on the photosynthetic photon flux density during and after leaf expansion. Plant Physiology 74: 475-480. Mohle B, Helier W, Wellrnann E. 1985. UV-induced biosynthesis of quercetin 3-0- beta-D-glucuronide in dill cell cultures. Phytochemistry 24: 465-467. MohJe B, Wellrnann E. 1982. Induction of phenylpropanoid compounds by UV-B irradiation in roots of seedlings and cell cultures from dill (Anethum graveolens L.). Plant Cell Reports 1 : 1 83- 1 85 . Molina MJ, Rowland FS. 1974. Stratospheric sink for chlorofluoromethanes : chlorine atom-catalysed destruction of ozone. Nature 249: 8 1 0-8 1 2. Montesinos MC, Ubeda A, Terencio MC, Paya M, Alcaraz MJ. 1995. Antioxidant profile on mono- and dihydroxylated flavone derivative in free radical generating systems. Zeitschrift fur Natut:forschung Section C Biosciences 50: 552-560. Moran R, Porath D. 1980. Chlorophyll determination in intact tissues using N,N­ dimethylformamide. Plant Physiology 65: 478-479. Murali NS, Terarnura AH. 1984. Differential flavonoid induction in UV -B irradiated Cucumis sativus cultivars. Plant Physiology 75 : 1 92. Murali NS, Terarnura AH. 1985a. Effects of ultraviolet-B irradiance on soybean VII. Biomass and concentration and uptake of nutrients at varying P supply. Journal of Plant Nutrition 8: 1 77- 1 92. Murali NS, Terarnura AH. 1985b. Ultraviolet-B radiation effects on field grown soybean. Bulletin of the Ecological Society of America 66: 237-238 . 10. References 207 Murali NS, Teramura AH. 1986a. Effectiveness of UV -B radiation on the growth and physiology of field-grown soybean modified by water stress. Photochemistry and Photobiology 44: 2 1 5-220. Murali NS, Teramura AH. 1986b. Effects of supplemental ultraviolet-B radiation on the growth and physiology of field-grown soybean. Environmental and Experimental Botany 26: 233-242. Murali NS, Teramura AH. 1986c. Intraspecific differences in Cucumis sativus sensitivity to ultraviolet-B radiation. Physiologia Plantarum 68: 673-677 . Murali NS, Teramura AH, Randall SK. 1988. Response differences between two soybean cultivars with contrasting UV -B radiation sensitivities. Photochemistry and Photobiology 48: 653-658 . Nagel LM, Bassman JH, Edwards GE, Robberecht R, Franceshi VR. 1998. Leaf anatomical changes in Populus trichocmpa, Quercus rubra, Pseudotsuga menziesii and Pinus ponderosa exposed to enhanced ultraviolet-B radiation. Physiologia Plantarum 104 : 385-396. Nahrstedt A. 1981. Isolation and stlUcture elucidation of cyanogenic glycosides. In : Vennesland B , Conn EE, Knowles CJ, Westley J, Wissing F, eds. Cyanide in biology. London: Academic Press, 1 45- 1 8 1 . Naidu BP. 1998. Separation of sugars, polyols, proline analogues, and betaines in stressed plant extracts by high performance l iquid chromatography and quantification by ultra violet detection. Australian Journal of Plant Physiology 25: 793-800. Nakayama H, Murayama S, Kosaka SI. 1996. Influence of ultraviolet-B (UV-B) on the early growth of grasses. Journal of Rakuno Oakuen University, Natural Science 20: 1 59- 1 63 . NASA. 1998. Antarctic ozone depletion sets new size record. NASA Internet press release number 98-1 78. Newsham KK, Lewis GC, Greenslade PD, McLeod AR. 1998. Neotyphodium loW, a fungal leaf endophyte, reduces fertility of Lolium perenne exposed to elevated UV-B radiation. Annals of Botany 8 1 : 397-403 . Nikolopoulos D, Petropoulou Y, Kyparissis A, Manetas Y. 1 995. Effects of enhanced UV -B radiation on the drought semi-deciduous Mediterranean shlUb Phlomis fruticosa under field conditions are season-specific. Australian Journal of Plant Physiology 22: 737-745. Nogues S, Alien DJ, Morison JIL, Baker NR. 1998. Ultraviolet-B radiation effects on water relations, leaf development, and photosynthesis in droughted pea plants . Plant Physiology 1 17: 1 73- 1 8 1 . Nogues S, Alien DJ, Morison JIL, Baker NR. 1999. Characterization of stomatal closure caused by ultraviolet-B radiation. Plant Physiology 121 : 489-496. 10. References 208 Nordiska Genbanken. 1993. The nordic forage catalogue. Alnarp: Nordiska Genbanken. Olsson LC, Fraysse L, Bornman JF. 2000. Influence of high light and UV -B radiation on photosynthesis and D 1 turnover in atrazine-tolerant and -sensitive cultivars of Brassica napus. Journal of Experimental Botany 51 : 265-274. Olsson LC, Veit M, Bornman JF. 1999. Epidermal transmittance and phenolic composition in leaves of atrazine-tolerant and atrazine-sensitive cultivars of Brassica napus grown under enhanced UV -B radiation. Physiologia Plantarum 107 : 259-266. Olsson LC, Veit M, Weissenbock G, Bornman JF. 1998. Differential flavonoid response to enhanced UV -B radiation in Brassica napus. Phytochemistry 49: 1 02 1 - 1 028. Olszyk D, Dai Q, Teng P, Leung H, Luo Y, Peng S. 1996. UV-B effects on crops: Response of the i rrigated rice ecosystem. Journal of Plant PhysioLogy 148: 26-34. Papadopoulos YA, Gordon RJ, McRae KB, Bush RS, BeJanger G, Butler EA, Fill more SAE, Morrison M. 1999. Current and elevated levels of UV-B radiation have few impacts on yields of perennial forage crops. GLobal Change BioLogy 5 : 847-856. Paul ND, Rasanayagam S, Moody SA, Hatcher PE, Ayres PG. 1997. The role of interactions between trophic levels in determining the effects of UV -B on terrestrial ecosystems. Plant Ecology 128 : 297-308. Perennes C, Glab N, Guglieni B, Doutriaux MP, Phan TH, Planchais S, Bergounioux C. 1999. Is arcA3 a possible mediator in the signal transduction pathway during agonist cell cycle arrest by salicylic acid and UV irradiation? Journal of Cell Science 1 12 : 1 1 8 1 - 1 1 90. Petropoulou Y, Kyparissis A, Nikolopoulos D, Manetas Y. 1995. Enhanced UV-B radiation alleviates the adverse effects of summer drought in two Mediterranean pines under field conditions. Physiologia Plantarum 94: 37-44. Pilon JJ, Lambers H, Baas W, Tosserams M, Rozema J, Atkin OK. 1999. Leaf waxes of slow-growing alpine and fast-growing lowland Poa species : inherent differences and responses to UV -B radiation. Phytochemistry 50: 57 1 -580. Pizzi A, Cameron FA. 1986. Flavonoid tannins - structural wood components for drought-resistance mechanisms of plants. Wood Science and Technology 20: 1 1 9- 1 24. Poorter H, Garnier E. 1999. Ecological significance of inherent variation in relative growth rate and its components. In: Pugnaire FI, Valladares F, eds. Handbook of functional plant ecology. New York: Marcel Dekker, 82- 1 20. JO. References 209 Popp HW, Brown F. 1936. The effect of ultraviolet radiation upon seed plants. In : Dugger BM, ed. Biological effects of radiation Il. New York: McGraw-Hill, 853- 888. Predieri S, Krizek DT, Wang CV, Mirecki RM, Zimmerman RH. 1993. Influence of UV -B radiation on developmental changes, ethylene, C02 flux and polyamines in cv. Doyenne d'Hiver pear shoots grown in vitro. Physiologia Plantarum 87: 109- 1 17 . Premkumar A, Eswaran K, Kulandaivelu G. 1993. Growth, biochemical and photosynthetic responses of green gram seedlings under water stress to ultravio1et­ B radiation. Plant Physiology and Biochemistry New Delhi 20: 3 1 -35 . Pukacki PM, Modrzynski .I. 1998. The influence of ultraviolet-B radiation on the growth, pigment production and chlorophyll fluorescence of Norway spruce seedlings. Acta Physiologiae Plantarum 20: 245-250. Rakitina TV, Vlasov PV, Jalilova FK, Kefeli VI, DzhaliIova FK. 1994. Abscisic acid and ethylene in mutants of Arabidopsis thaliana differing in their resistance to ultraviolet (UV -B) radiation stress. Russian Journal of Plant Physiology 41 : 599- 603. Rattray PV, Joyce JP. 1974. Nutritive value of white clover and perennial ryegrass. 4. Utilization of dietary energy. New Zealand Journal of Agricultural Research 17 : 401 -406. Reisinger AR, Jones NB, Matthews W A, Rinsland CP. 1995. Southern hemisphere midlatitude ground-based measurements of CION02 - method of analysis , seasonal cycle and long-term trend. Journal of Geophysical Research­ Atmospheres 100: 23 1 83-23 1 93 . Reuber S, Bornman .IF, Weissenbock G. 1996a. A flavonoid mutant of barley (Hordeum vulgare L.) exhibits increased sensitivity to UV-B radiation in the primary leaf. Plant, Cell and Environment 19: 593-60 l . Reuber S , Bornman .IF, Weissenbock G. 1996b. Phenylpropanoid compounds i n primary leaf tissues of rye (Secale cereale). Light response of their metabolism and the possible role in UV -B protection. Physiologia Plantarum 97: 1 60- 1 68 . Rice Evans CA, Miller NJ, Paganga G. 1996. Structure-antioxidant activity relationships of flavonoids and phenolic acids. Free Radical Biology and Medicine 20: 933-956. Robakowski P, Laitat E. 1999. Effects of an enhanced ultraviolet-B irradiation on photosynthetic apparatus of several forest coniferous tree species from different locations. Acta Physiologiae Plantarum 2 1 : 283-296. Robberecht R, Caldwell MM, Billings WD. 1980. Leaf UV optical properties along a latitudinal gradient in the arctic-alpine life zone. Ecology 61 : 6 1 2-6 1 9. 10. Relerences 2 1 0 Rosen JM, Kjome NT, Liley JB. 1997. Tropospheric aerosol back scatter at a midlatitude site in the northern and southern hemispheres. Journal (�f Geophysical Research-Atmospheres 102 : 2 1 329-2 1 339. Rousseaux MC, Ballare CL, Scopel AL, Searles PS, Caldwell MM. 1998. Solar ultraviolet-B radiation affects plant-insect interactions in a natural ecosystem of Tierra del Fuego (southern Argentina). Oecologia 1 16: 528-535 . Rozema J. (Ed.) 1999. Stratosperic ozone depletion. The effects �f enhanced UV-B radiation on terrestrial ecosystems. Leiden: Backhuys Publishers. Rozema J, Chardonnens A, Tosserams M, Hafkenscheid R, Bruijnzeel S. 1 997a. Leaf thickness and UV -B absorbing pigments of plants in relation to an elevational gradient along the Blue Mountains, Jamaica. Plant Ecology 128: 1 50- 1 59. Rozema J, van de Staaij J, Bjorn LO, Caldwell M. 1997b. UV-B as an environmental factor in plant l ife : stress and regulation. Trends in Ecology and Evolution 12: 22-28. Rozema J, van de Staaij J, Bjorn LO, de Bakker N. 1999. Depletion of stratospheric ozone and solar UV -B radiation : evolution of land plants, UV -screens and function of polyphenolics. In: Rozema J, ed. Stratosperic ozone depletion. The effects (�f enhanced UV-B radiation on terrestrial ecosystems. Leiden: Backhuys Publishers, 1 - 1 9. 4 Rozema J, van de Staij JWM, Tosserams M. 1997c. Effects of UV-B radiation on plants from agro- and natural ecosystems. In: Lumsden PJ, ed. Plants and UV-B, Re!;ponses and Environmental Change. Cambridge: Cambridge University Press, 2 1 3-232. Ryan KG, Markham KR, BIoor SJ, Bradley JM, Mitchell KA, Jordan BR. 1998. UVB radiation induced increase in quercetin : kaempferol ratio in wild-type and transgenic l ines of Petunia. Photochemistry and Photobiology 68: 323-330. Saile-Mark M, Tevini M. 1997. Effects of solar UV-B radiation on growth, flowering and yield of central and southern European bush bean cultivars (Phaseolus vulgaris L.). Plant Ecology 128: 1 1 5- 1 25 . Sallaud C , E l Turk J, Breda C , Buffard D, Kozak Id, Esnault R , Kondorosi A . 1995. Differential expression of cDNA coding for chalcone reductase, a key enzyme of the 5-deoxyflavonoid pathway, under various stress conditions in Medicago sativa. Plant Science 109: 1 79- 1 90. Salt DT, Moody SA, Whittaker JB, Paul ND. 1998. Effects of enhanced UVB on populations of the phloem feeding insect Strophingia ericae (Homoptera: PsyUoidea) on heather (Calluna vulgaris). Global Change Biology 4: 9 1 -96. 10. References 2 1 1 Santos I, Almeida JM, Salema R. 1993. Plants of Zea mays L. developed under enhanced UV -B radiation. 1. Some ultrastructural and biochemical aspects. Journal (?f Plant Physiology 141 : 450-456. Saradhi PP, Alia, Arora S, Prasad KVSK. 1995. Proline accumulates in plants exposed to UV radiation and protects them against UV induced peroxidation. Biochemical and Biophysical Research Communications 209: 1 -5 . SAS. 1996. The SAS system for windows, release 6. 12 . Cary: SAS Institute. Sato T, Kumagai T. 1997. Role of UV-absorbing compounds in genetic differences in the resistance to UV -B radiation in rice plants. Breeding Science 47: 2 1 -26. Schmidt AM, Ormrod DP, Livingston NJ, Misra S. 2000. The interaction of ultraviolet-B radiation and water deficit in two Arabidopsis thaliana genotypes. Annals of Botany 85: 57 1 -575. Schnitzler JP, Junghlut TP, Feicht C, Kofferlein M, Langebartels C, HelIer W, Sandermann H, Jr. 1997. UV-B induction of flavonoid biosynthesis in Scots pine (Pinus sylvestris L.) seedlings. Trees: Structure and Function 1 1 : 1 62- 1 68 . Scholander PF, Hammel HT, Bradstreet ED, Hemmingsten E. 1965. Sap pressure in vascular plants. Science 148 : 339-346. Searles PS, Caldwell MM, Winter K. 1 995. The response of five tropical dicotyledon species to solar ultraviolet-B radiation. American Journal of Botany 82: 445-453. Seckmeyer G, Mayer B, Bernhard G, McKenzie RL, Johnston PV, Kotkamp M, Booth eR, Lucas T, Mestechkina T, Roy CR, Gies HP, Tomlinson D. 1 995. Geographical differences 1I1 the UV measured by intercompared spectroradiometers. Geophysical Research Letters 22: 1 889- 1 892. Seckmeyer G, McKenzie RL. 1992. Increased ultraviolet radiation in New Zealand (45°S) relative to Germany (48°N). Nature 359: 1 35- 1 37 . Seigler DS. 1998. Plant secondary metabolism. Boston: Kluwer Academic Publishers. Sharma AK, Sharma SK. 1997. Peroxidase and polyphenol oxidase activity changes in relation to leaf rust of wheat. Journal (�f Maha rash tra Agricultural Universities 22: 286-29 1 . Shashidhar VR, Prasad TG, Sudharshan L. 1996. Hormone signals from roots to shoots of sunflower (Helianthus annuus L.). Moderate soil drying increases delivery of abscisic acid and depresses delivery of cytokinins in xylem sap. Annals of Botany 78: 1 5 1 - 1 55 . Sheahan JJ. 1996. Sinapate esters provide greater UV -B attenuation than flavonoids i n Arabidopsis thaliana (Brassicaceae). American Journal C?f Botany 83: 679-686. Short FT, NeckJes HA. 1999. The effects of global climate change on seagrasses. Aquatic Botany 63: 169- 1 96. 10. References 2 1 2 Singh A . 1997. Increased UV -B radiation reduces N2-fixation i n tropical leguminous crops. Environmental Pollution 95: 289-29 1 . Slansky F, Scriber JM. 1985. Food consumption and utilisation. In: Kerkut GA, Gilbert LI, eds. Comprehensive insect physiology, biochemistry, and pharmacology. Volume 4: digestion, nutrition, excretion. New York: Pergamon Press, 87- 1 63 . Smith GJ, Markham KR. 1998. Tautomerism of flavonol glucosides - relevance to plant UV protection and flower colour. Journal of Photochemistry and Photobiology A - Chemistry 1 18 : 99- 1 05 . Somersalo S , Makela P, Rajala A , Nevo E , Peltonensainio P. 1998. Morpho­ physiological traits characterizing environmental adaptation of A vena barbata. Euphytica 99: 2 1 3-220. Southgate DAT. 1991. Determination offood carbohydrates. Barking: Elsevier. Stapleton AE, Thornber CS, Walbot V. 1997. UV-B component of sunlight causes measurable damage in field-grown maize (Zea mays L.) : developmental and cellular heterogeneity of damage and repair. Plant, Cell and Environment 20: 279- 290. Stapleton AE, Walbot V. 1994. Flavonoids can protect maize DNA from the induction of ultraviolet radiation damage. Plant Physiology 105 : 88 1 -889. Staxen I, Bornman JF. 1994. A morphological and cytological study of Petunia hybrida exposed to UV-B radiation. Physiologia Plantarum 9 1 : 735-740. Steger-Hartmann T, Koch U, Dunz T, Wagner E. 1994. Induced accumulation and potential antioxidative function of rutin in two cultivars of Nicotiana tabacum L. ZeitschriJt fur Natuiforschung Section C Bioscience,�' 49: 57-62. Stephanou M, Manetas Y. 1997. The effects of seasons, exposure, enhanced UV-B radiation, and water stress on leaf epicuticular and internal UV -B absorbing capacity of Cistus creticus: A Mediterranean field study. Journal of Experimental Botany 48: 1 977- 1 985 . Stewart JD, Hoddinott J. 1993. Photosynthetic accIimation to elevated atmospheric carbon dioxide and UV irradiation in Pinus banksiana. Physioiogia Plantarum 88: 493-500. Stout MJ, Workman KV, Bostock RM, Duffey SS. 1 998. Stimulation and attenuation of induced resistance by elicitors and inhibitors of chemical induction in tomato (Lycopersicon esculentum) fol iage. Entomologia Experimentalis et Applicata 86: 267-279. 10. References 2 1 3 Strid A , Chow WS, Anderson JM. 1996. Changes i n the relaxation of electrochromic shifts of photosynthetic pigments and in the levels of mRNA transcripts in leaves of Pisum sativum as a result of exposure to supplementary UV -B radiation. The dependency on the intensity of the photosynthetic ally active radiation. Plant and Cell Physiology 37: 6 1 -67. Sullivan JH, Teramura AH. 1988. Effects of ultraviolet-B irradiation on seedling growth in the Pinaceae. American Journal of Botany 75: 225-230. Sullivan JH, Teramura AH. 1990. Field study of the interaction between solar ultraviolet radiation and drought on photosynthesis and growth of soybean. Plant Physiology 92: 1 4 1 - 146. Sullivan JH, Teramura AH, Ziska LH. 1992. Variation in UV -B sensitivity in plants from a 3 ,000-m elevational gradient in Hawaii . American Journal of Botany 79: 737-743. Tekchandani S, Guruprasad KN. 1998. Modulation of a guaiacol peroxidase inhibitor by UV -B in cucumber cotyledons. Plant Science 136: 1 3 1 - 1 37 . Teramura AH. 1990. Implications of stratospheric ozone depletion upon plant production. HortScience 25: 1 557- 1 560. Teramura AH, Forseth IN, Lydon J. 1984a. Effects of ultraviolet-B radiation on plants during mild water stress. IV. The insensitivity of soybean internal water relations to ultraviolet-B radiation. Physiologia Plantarum 62: 384-389. Teramura AH, Murali NS. 1986. Intraspecific differences in growth and yield of soybean exposed to ultraviolet-B radiation under greenhouse and field conditions. Environmental and Experimental Botany 26: 89-95. Teramura AH, Perry MC, Lydon J, McIntosh MS, Summers EG. 1984b. Effects of ultraviolet-B radiation on plants during mild water stress. Ill. Effects on photosynthetic recovery and growth in soybean. Physiologia Plantarum 60: 484- 492. Teramura AH, Sullivan JH, Lydon J. 1 990. Effects of UV-B radiation on soybean yield and seed quality: a 6-year field study. Physiologia Plantarum 80: 5- 1 1 . Teramura AH, Tevini M, Iwanzik W. 1983. Effects of ultraviolet-B irradiation on plants during mild water stress. I . Effects on diurnal stomatal resistance. Physiologia Plantarum 57: 175- 1 80. Tevini M. 1 993. Effects of enhanced UV-B radiation on terrestrial plants. In : Tevini M, ed. UV-B Radiation and Ozone Depletion: Effects on Humans, Animals, Plants, Microorganisms and Materials. Boca Raton: Lewis Publishers, 1 25- 1 53 . Tevini M. 1994. UV -B effects on terrestrial plants and aquatic organisms. Progress in Botany 55 : 1 74- 1 90. lO. References 2 14 Tevini M. 1999. UV -effects on plants. In: Singhal GS, Renger G, Soporay SK, Irrgang KD, eds. Concepts in photobiology: photosynthesis and photomorphogenesis. Dordrecht: Kluwer, 588-6 1 3 . Tevini M. 2000. UV -8 effects on plants. In: Agrawal SB, Agrawal M, eds. Environmental pollution and plant responses. Boca Raton: Lewis Publishers, 83- 97 . Tevini M, Iwanzik W, Teramura AH. 1983a. Effects of UV-B radiation on plants during mild water stress. n. Effects on growth, protein and flavonoid content. Zeitschrift fur Pflanzenphysiologie 1 10: 459-467 . Tevini M, Iwanzik W, Thoma U. 1981. Some effects of enhanced UV-B irradiation on the growth and composition of plants. Planta 153: 388-394. Tevini M, Thoma U, Iwanzik W. 1983b. Effects of enhanced UV -B radiation on germination, seedling growth, leaf anatomy and pigments of some crop plants. Zeitschriftfur Pflanzenphysiologie 109 : 435-448 . Thomas RG. 1987a. The structure of the mature plant. In : Williams WM, Baker MJ, eds. White clover. Wallingford: CAB, 1 -30. Thomas RG. 1987b. Vegetative growth and development. In: Williams WM, Baker MJ, eds. White clover. Wallingford: CAB, 3 1 -62. Tosserams M, Pais de Sa A, Rozema J. 1996. The effect of solar UV radiation on four plant species occurring in a coastal grassland vegetation in the Netherlands. Physiologia Plantarum 97 : 73 1 -739. Tosserams M, Rozema J. 1995. Effects of ultraviolet-B radiation (UV -B) on growth and physiology of the dune grassland species Calamagrostis epigeios. Environmental Pollution 89: 209-2 14 . Turner Ne. 1981. Techniques and experimental approaches for the measurement of plant water status . Plant and Soil 58: 339-366. van de Staaij JWM, Huijsmans R, Ernst WHO, Rozema J. 1995. The effect of elevated UV-B (280-320 nm) radiation levels on Silene vulgaris: a comparison between a highland and a lowland popUlation. Environmental Pollution 90: 357- 362. van de Staaij JWM, Bolink E, Rozema J, Ernst WHO. 1997. The impact of elevated UV -B (280-320 nm) radiation levels on the reproduction biology of a highland and a lowland population of Silene vuLgaris. Plant Ecology 128: 1 72- 1 79. van Soest PJ, Robertson JB, Lewis BA. 1991. Methods for dietary fiber, neutral detergent fiber, and nonstarch polysaccharides in relation to animal nutrition. JournaL of Dairy Science 74: 3583-3597. Venable JH, Coggleshall R. 1965. A simplified lead citrate stain for electron microscopy use. Journal of Cellular Biology 25: 207-208. 10. References 2 1 5 Visser AJ, Tosserams M , Groen MW, Kalis G, Kwant R, Magendans GWH, Rozema J. 1997a. The combined effects of CO2 concentration and enhanced UV­ B radiation on faba bean. 3. Leaf optical properties, pigments, stomatal index and epidermal cell density. Plant Ecology 128: 208-222. Visser AJ, Tosserams M, Groen MW, Magendans GWH, Rozema J. 1997b. The combined effects of C02 concentration and solar UV-B radiation on faba bean grown in open-top chambers. Plant, Cell and Environment 20: 1 89- 1 99. Vogelmann TC, Bornman JF, Yates DJ. 1996. Focussing of light by leaf epidermal cells. Physiologia Plantarum 98: 43-56. Waldbauer GP. 1968. The consumption and utilisation of food by insects. Advances in Insect Physiology 5: 229-288 . Wand SJE. 1995. Concentration of ultraviolet-B radiation absorbing compounds m leaves of a range of fynbos species. Vegetatio 1 16 : 5 1 -6 1 . Wang X, Caradus JR, Chu ACP. 1996. Effect of summer moisture deficit on growth of five white clover cultivars. In: Proceedings of the New Zealand Grassland Association. Waitangi : New Zealand Grassland Association, 73-76. Wanjaiya GW, Wales WJ, Dellow DW. 1993. Utilisation of white clover pasture and maize silage by the lactating dairy cow. Proceedings of the New Zealand Society of Animal Production 53: 73-75. Warrington IJ, Fulton TA, Halligan EA, de Silva HN. 1999. Apple fruit growth and maturity are affected by early season temperatures. Journal of the American Society for Horticultural Science 124 : 468-477. Warrington IJ, Kanemasu ET. 1983. Corn growth response to temperature and photoperiod. I. Seedling emergence, tassel initiation, and anthesis. Agronomy Journal 75 : 749-754. Williams WM. 1987a. Adaptive variation. In : Williams WM, Baker MJ, eds. White clover. Wallingford: CAB, 299-32 1 . WilIiams WM. 1987b. Genetics and breeding. In: Williams WM, Baker MJ, eds. White clover. Wal lingford: CAB, 343-4 1 9. Williams WM, Mason KM, Williamson ML. 1 998. Genetic analysis of shikimate dehydrogenase allozymes in Trifolium repens L. Theoretical and Applied Genetics 96: 859-868. Wilson KE, Wilson MI, Greenberg BM. 1998. Identification of the flavonoid glycosides that accumulate in Brassica nap us L. cv. Topas specifically in response to ultraviolet B radiation. Photochemistry and Photobiology 67: 547- 553 . Wilson PJ, Thompson K, Hodgson JG. 1999. Specific leaf area and leaf dry matter content as alternative predictors of plant strategies. New Phytologist 143: 1 55- 1 62 . 10. Relerences 2 16 Woodfield DR, Caradus JR. 1987. Adaptation o f white clover to moisture stress. Proceedings (�f the New Zealand Grassland Association 48: 143- 149. Woodfield DR, Cousins GR. 1999. Development and comparative growth of a white clover ploidy series. In: Langridge P, ed. 1 1 th Australian Plant Breeding Conference. Adelaide: Australian Plant Breeding Conference, 272-273. Yamasaki H, Sakihama Y, Ikehara N. 1997. Flavonoid-peroxidase reaction as a detoxification mechanism of plant cells against H202. Plant Physiology 1 15 : 1405- 1 4 1 2. Yazawa M, Shimizu T, Hirao T. 1992. Feeding response of the silkworm, Bombyx mori, to UV irradiation of mulberry leaves. Journal (�f Chemical Ecology 18 : 561 - 569. Zepp RG, Callaghan TV, Erickson DJ. 1998. Effects of enhanced solar ultraviolet radiation on biogeochemical cycles. Journal of Photochemistry and Photobiology. B - Biology 46: 69-82. Zeuthen J, Mikkelsen TN, Paludan Muller G, Ro Poulsen H. 1997. Effects of increased UV -B radiation and elevated levels of tropospheric ozone on physiological processes in European beech (Fagus sylvatica). Physiologia Plantarum 100: 28 1 -290. Zinser C, Helier W, Rau W, Sandermann H, Jr. 1997. Is enhanced UV-B radiation likely to be a problem to agriculture? Agrarforschung 4: 320-323. Ziska LH, Teramura AH, Sullivan AH, McCoy A. 1993. Influence of ultraviolet-B (UV -B) radiation on photosynthetic and growth characteristics in field-grown cassava (Manihot esculentum Crantz) . Plant. Cell and Environment 16 : 73-79. Ziska LH, Teramura AH, Sullivan JH. 1 992. Physiological sensitivity of plants along an elevational gradient to UV -B radiation. American Journal of Botany 79: 863- 87 1 .