Physiological and biochemical response of endophyte infected Lolium perenne to water stress and to plant hormone treatment when water sufficient : a thesis presented in partial fulfilment of the requirements for the degree of Master of Science in Agricultural Science at Massey University, Palmerston North, New Zealand

Abstract

This focus of this thesis was centred on the water stress-induced interactions between the Lolium perenne and Neotyphodium lolii symbiotum and the in vivo biomass relationship between the two species. The aim was to test the hypotheses that endophyte bio-protective metabolites are synthesized at higher levels in water-stressed endophyte-colonised plants than in water-sufficient plants; to carry out ELISA experiments to measure tiller levels of ABA and JA in perennial ryegrass during the application of a controlled water stress and apply these levels, by dipping water-sufficient plants, thereby mimicking the abiotic stress, to test whether the levels of the endophyte-produced bio-protective metabolites, ergovaline, lolitrem B and peramine, increased. The evapotranspiration rate of endophyte hosted and endophyte free perennial ryegrass was measured and no significant difference was found. The sodium borohydride reduction of ABA and JA, to yield protonated ABA and CA was developed using small quantities of the reducing agent, the products verified by nuclear magnetic resonance (NMR) spectroscopy and high performance liquid chromatography (HPLC). Radioactive sodium borohydride was then used to produce tritiated ABA and CA, the reduction products purified and used as spiking controls in the solid state extraction of ABA and JA from perennial ryegrass. The efficiency of DEAE Sephadex solid state columns at removing ABA and JA from perennial ryegrass leaf and sheath tissue was tested. The spiking methods, the results obtained and their use in adjusting analysed ABA and JA tissue levels are outlined. Hormone dipping regimes, obtained from the literature, were used as the basis for testing hormone uptake, after dipping, by measuring tissue levels by ELISA. A trial was carried out, that measured by competitive ELISA, the ABA and JA leaf tissue levels in the perennial ryegrass genotypes after they had been progressively water stressed, by re-applying one third of the water lost in the previous 48 h. After 8 days of applied water stress the ABA levels increased substantially, peaking at 2.5 x 10 4 picomoles/g fresh weight while the JA increase was less dramatic and more sustained, peaking at 1.68 x 10 3 picomoles/g fresh weight over the same period. This information was used to test the hypothesis that dipping water sufficient perennial ryegrass in known water stress levels of ABA and JA would increase alkaloid output by N. lolii. The main statistically analysed split plot experiment, involved the dipping of water sufficient potted L. perenne genotypes in solutions (tissue mass adjusted) of increasing ABA, JA and ABA/JA concentrations from day 8 of an applied water stress, every 48 h for 16 days. The progression of water stress in the control treatments was measured independently. After this period of water stress the leaf and sheath tissue was harvested separately and the tissue levels of lolitrem B, ergovaline and peramine were analysed by HPLC. These dipped replicates were then compared with un-dipped water stressed and water sufficient controls. The data was analysed in three parts. Leaf and sheath combined, leaf only and sheath only data. There was a positive correlation coefficient (r) between the three alkaloids ergovaline, lolitrem B and peramine for the water-sufficient, hormone-treated and water-deficit-treated plants. Considering ABA, JA and ABA/JA treatments on ergovaline levels in perennial ryegrass for the combined data; ergovaline levels in the sheath were significantly higher than the controls. During water stress where there was a highly significant increase in ergovaline levels in the sheath tissue. There was a significant increase in ergovaline levels in water stressed leaves when using leaf data. A significant increase in ergovaline level occurred in the leaves, but not the sheath, when water sufficient plants were dipped in JA and ABA separately. These levels dropped significantly in leaves when water sufficient plants were dipped in ABA/JA mixed solutions. Lolitrem B was 3-fold higher in the sheath than the leaf. There was no response to hormone treatments. When sheath data was used, a significant decrease in sheath lolitrem B levels occurred in perennial ryegrass treatments that were water stressed. Peramine levels were significantly higher in the sheath than the leaves when the combined data was analysed. When leaf data was considered peramine levels fell significantly during water stress while the analysed sheath data showed a significant 1.4-fold increase. Treatments that involved dipping in JA gave significant peramine increases (1.35 fold). Other hormone treatments increased peramine levels but they were not significant. This study gave strong indications that there is a plant stress hormone communication between L. perenne and the mutualistic endophyte N. lolii and that water stress does increase endophyte ergovaline and peramine output and decreases lolitrem B levels. During a confocal microscopic study, L. perenne meristems were successfully dissected out, re-hydrated, fixed, stained and 1.3 micron longitudinal sections, viewed. The sections were then digitalized, enhanced and re-assembled into a 3 dimensional rotating image; using computer based confocal microscope software. A stereoscopic anaglyph was made of ryegrass tiller leaf sections, highlighting N. lolii colonization patterns. A non-invasive, means of measuring endophyte to ryegrass biomass ratios was developed using glutaraldehyde based fixing and staining protocols and used to threshold the endophyte mycelium in confocal microscopic images, using off line draw fill software. Once this was done the biomass was estimated for each section, totaled and percentage endophyte occupancy estimated.