The effect of condensed tannins upon protein degradation in the rumen and on animal production in sheep fed fresh Lotus corniculatus : a thesis in partial fulfillment of the requirements for the degree of Doctor of Philosophy in the Institute of Food, Nutrition and Human Health, College of Sciences at Massey University
A series of in vitro, in sacco and in vivo indoor and grazing experiments were conducted at Massey University and AgResearch Grasslands, Palmerston North, New Zealand to study the effect of condensed tannins (CT) in Lotus corniculatus (Birdfoot trefoil; CV. Grasslands Goldie) upon protein digestion in the rumen and on animal production. Aspects studied included effects of CT upon proteolytic bacterial activity, protein solubilization and degradation in the rumen and wool production and reproduction in grazing sheep. The studies also investigated the potential of L. corniculatus compared with perennial ryegrass/white clover pasture (hence referred as to pasture). The nutritional effects of CT in L. corniculatus were assessed by administrating polyethylene glycol (PEG; MW 3500) into the rumen of one group of sheep (PEG sheep; CT-inactivated), whilst a separate group of sheep received water (control sheep; CT-acting). PEG selectively binds with CT, preventing the CT from binding plant proteins in the rumen, so that effects of CT can be determined by comparing CT-acting sheep with PEG sheep. The productivity of mixed age ewes in grazing trials was measured in two experiments in the summer of 1995/1996 (Chapter 2) and the summer/autumn of 1997 (Chapter 3), to evaluate the effects of CT in L. corniculatus upon efficiency of animal production. A rotational grazing system with restricted feed allowance was used in both experiments. 1. During 1995/1996 (Chapter 2), a grazing trial was conducted to evaluate the effect of CT in L. corniculatus on wool growth and on wool processing characteristics in sheep fed close to maintenance for 125 days during summer and autumn (20 December 1995 until 25 April 1996). Half the ewes received twice daily supplements of PEG. The Lotus corniculatus contained 32 g total nitrogen (N) and 28 g total CT/kg dry matter (DM) and had an in vitro organic matter digestibility of 0.70. Action of CT reduced rumen ammonia concentration (P < 0.05) and reduced blood plasma urea concentration (P < 0.01) but increased blood plasma cysteine concentration (P < 0.05) compared to their counterparts receiving PEG supplementation. The concentration of blood plasma methionine was unaffected by CT. The CT had no effect on voluntary feed intake (VFI) and average liveweight gain (P > 0.05) but increased both clean fleece weight (P < 0.05) and staple length (P < 0.001). The CT also reduced dag percentage (P < 0.05) and tended to reduce wool yellowness (P = 0.07) relative to sheep receiving PEG. There were no significant effects of CT on fiber diameter (µm), staple breaking force (Newtons), bulk density (cm3/g) or wool resilience (cm3/g). It was concluded that the action of CT in sheep fed L. corniculatus increased the efficiency of wool production, with more wool being produced at the same feed intake. 2. Another grazing trial (Chapter 3) was conducted to study the effects of CT in L. corniculatus upon VFI, concentration of plasma metabolites, reproductive efficiency and wool production in ewes during two synchronised oestrous cycles in autumn 1997. The ewes were restricted to maintenance feeding for the first 12 days of each oestrous cycle and then increased to ad-libitum for the last five days before ovulation. The experiment was of 2 x 2 factorial design, using two types of forage (L. corniculatus vs. pasture), with half the ewes grazing each forage being given twice daily oral PEG supplementation. A rotational grazing system with 200 mixed aged dry ewes (52±0.88 kg/ewe) was used. The Lotus corniculatus contained 17 g total CT/kg DM in the diet selected, with only trace amounts of total CT present in pasture. Ewes grazing L. corniculatus had higher plasma concentrations of branched chain amino acids (BCAA; 57 %) and essential amino acids (EAA; 52 %) than sheep grazing pasture. Again CT in L. corniculatus had no effect on mean VFI. The PEG supplementation had no effect upon ovulation rate (OR; 1.33 vs. 1.35) and lambing percentage (1.36 vs. 1.36 %) of the ewes grazing pasture. The CT increased both OR (1.78 vs. 1.56) and lambing percentage (1.70 vs. 1.42%) in the ewes grazing L. corniculatus relative to sheep supplemented with PEG. Increases in OR and lambing % of ewes grazing L. corniculatus were due to increases in fecundity (more multiple ovulations and less single ovulations), with no effect on ewes cycling/ewes mated. Compared to ewes grazing pasture, ewes grazing L. corniculatus had increased clean fleece weight (19 %). It was concluded that action of CT in the lotus diet was partly responsible for the increased efficiency of reproduction, with more lambs being produced at the same VFI. 3. In situ and in vitro rumen incubations (Chapter 4) were used to determine the effect of CT on both the solubilization and degradation of Rubisco (ribulose-1,5-bisphosphate carboxylase/oxygenase; EC 126.96.36.199; fraction 1 leaf protein) from white clover (Trifolium repens; 0.3 g CT/kg DM) and Lotus corniculatus (22.1 g CT/kg DM). The sheep used for the experiments were fed either white clover or L. corniculatus. The loss of DM and neutral detergent fibre (NDF), total N and Rubisco from polyester bags suspended in the rumen of sheep was used as a measurement of solubilisation. The effect of CT extracted from L. corniculatus on the degradation of Rubisco from white clover was measured by in vitro incubations with rumen fluid obtained from the same fistulated sheep fed either white clover or L. corniculatus. In the absence of PEG, the solubilisation of Rubisco from L. corniculatus was less rapid than the solubilisation of this protein from white clover when each forage was incubated in the rumen of sheep fed the same diet. Addition of PEG tended to increase the solubilisation of Rubisco from L. corniculatus, suggesting that CT slowed the rates of solubilization of Rubisco from this forage. The action of CT did not inhibit the in situ loss of NDF from either white clover or L. corniculatus. In the absence of PEG, the in vitro degradation of Rubisco from L. corniculatus was slower when compared to the degradation of this protein from white clover; PEG addition increased the degradation of Rubisco from L. corniculatus, but not from white clover, showing that CT was the causal agent. The addition of CT extracted from L. corniculatus markedly depressed the degradation of Rubisco from white clover, with the effect being completely reversible by PEG. The large subunit (LSU) of Rubisco was consistently degraded at a faster rate than the small subunit (SSU) and added CT had a greater effect in slowing the degradation of the LSU compared to the SSU. It was concluded that the action of CT from L. corniculatus reduces the digestion of protein in the rumen of sheep through a minor effect on solubilization and a major effect on degradation. The main effects of CT on protein solubilization and degradation seemed to be produced locally by CT present in plant tissue. 4. Eleven strains of proteolytic rumen bacteria (Chapter 5) were used to determine the effect of CT extracted from Lotus corniculatus on the in vitro proteolysis of Rubisco protein, bacterial specific growth rate and maximum optical density (ODmax). Effects of CT on the rate of Rubisco proteolysis (%/h) were determined through making measurements in the presence and absence of PEG. Streptococcus bovis strain NCFB 2476 and B315, Butyrivibrio fibrisolvens strain WV1 and C211a, Prevotella ruminicola strain 23 and C21a, Clostridium proteoclasticum B316T, Ruminococcus albus 8, Fibrobacter succinogenes S-85, Eubacterium sp. strain C12b and C124b were tested against 1.5 mg CT/ml for Rubisco proteolysis and were examined with 0, 50, 100, 200, 400, and 600 µg CT/ml for bacterial growth measurements. In general, the presence of CT markedly depressed the degradation of both the LSU and SSU of Rubisco, with the effect being completely reversible by PEG. However, the rates of proteolysis per hour for both sub-units of Rubisco varied considerably between individual bacterial species and subunits of Rubisco. In the absence of CT, S. bovis strain NCFB 2476 and B315 and P. ruminicola like-strain C21a appeared to be most active in both LSU and SSU degradation, while P. ruminicola 23, Eubacterium sp. strain C12b and C124b, C. proteoclasticum B316T, B. fibrisolvens strain WV1 and C211a had moderate to lower rates of LSU and SSU degradation. In the presence of CT, S. bovis strain NCFB 2476 and B315 and P. ruminicola-like strain C21a appeared to be most active in both LSU and SSU breakdown. Most bacterial strains showed significantly (P< 0.05-0.01) decreased specific growth rate and ODmax with increasing CT concentrations. However, some of the strains, C. proteoclasticum B316T and R. albus 8 showed transient increases in specific growth rate at low concentrations of CT (between 50 to 100 µg CT/ml), but not at high concentrations of CT. In terms of specific growth rate, addition of CT at low concentrations (50-200 µg CT/ml), S. bovis NCFB 2476, Eubacterium sp. C124b and F. succinogenes S-85 were most affected compared to the minus CT controls, while P. ruminicola sp. C21a and C. proteoclasticum B316T were not greatly inhibited at the highest concentrations of CT. The degree of inhibition of both bacterial growth and Rubisco degradation in the presence of CT varied considerably between individual bacterial species and will be discussed in Chapter 5. It was concluded that action of CT from L. corniculatus reduces both the rate of Rubisco proteolysis and the growth rate of proteolytic rumen bacteria, but the magnitude of the CT effect differed between strains used. 5. Twelve six month old Romney sheep were fistulated in the rumen and abomasum and fed Lotus corniculatus (32 g CT/kg DM), to examine the effects of CT on proteolytic rumen bacterial populations and on quantitative N digestion in the rumen. Half the animals were given continuous intraruminal infusions of PEG. In the first part of the experiment, the populations of four proteolytic rumen bacteria were enumerated directly from rumen samples using a competitive polymerase chain reaction (cPCR) technique. During pre-feeding on a perennial ryegrass/white clover pasture diet, populations of C. proteoclasticum B316, Eubacterium sp. C12b, S. bovis B315 and B. fibrisolvens C211a were 1.6 x 10 8, 2.7 x 10 8, 7.1 x 10 6 and 1.2 x 10 6 per ml respectively. When the diet was changed from pasture to L. corniculatus (average of 8 h to 120 h), the average populations of C. proteoclasticum B316, Eubacterium sp. C12b, S. bovis B315 and B. fibrisolvens C211a from the same animals were decreased significantly (P < 0.001) to 5.1 x 10 7, 1.5 x 10 8, 2.6 x 10 6 and 1.0 x 10 6 per ml, respectively. When the PEG was infused into the rumen of sheep fed L. corniculatus, the populations of proteolytic bacteria were significantly increased (P < 0.01-0.001) compared to the CT-acting group. Rumen proteinase activity, concentrations of rumen ammonia and soluble N were decreased significantly (P < 0.05-0.001) in the CT-acting compared to the PEG treatment group. In the quantitative N studies, the principal effects of CT were to reduce rumen N digestibility (P < 0.05) and ammonia pool size, and to increase the flow of non-ammonia nitrogen (NAN) to the abomasum. Dry matter intake and DM digestibility were unaffected. The N intake, rumen NAN and microbial NAN pool sizes were similar in both CT-acting and PEG sheep. Non-microbial NAN fluxes to the abomasum were significantly higher (P < 0.01) in the CT-acting sheep than in the PEG sheep, but microbial NAN flux to the abomasum was unaffected by treatment. It was concluded that L. corniculatus CT reduced forage protein degradation in the rumen, and increased the flow of undegraded feed NAN to the abomasum. Proteolytic bacterial populations seemed to be reduced by CT, but these changes did not effect the total rumen microbial NAN pool or abomasal microbial NAN flux. Therefore, more protein was potentially available for absorption from the small intestine. 6. This study is the first to report that action of CT increased reproductive efficiency in grazing ewes. It is also the first study to show that action of CT decreased proteolytic bacterial populations measured directly from rumen samples using cPCR techniques. Feeding forages containing CT such as L. corniculatus has been shown to reduce proteolysis in the rumen, with the mechanisms being to slightly reduce protein solubilization, to markedly reduce protein degradation and to reduce the populations of proteolytic bacteria. CT increased NAN flux into the abomasum (in indoor studies) and increased animal production in grazing ewes without affecting VFI, thus improving the efficiency of animal production. It is concluded that forage CT can be used to increase the efficiency and sustainability of livestock production from grazed forages.