Seasonal composition of tropical C4 grasses, and its influence on rumen prokaryotic diversity in relation to methane production from beef cattle in the northern Australian rangelands

Abstract

In Australia, enteric methane production accounts for 68% of national greenhouse gas (GHG) emissions. Methane production (MP) is affected by diet quality. Approximately 50% of Australian beef cattle are grazed in the northern rangelands (Queensland, the Northern Territory, and part of Western Australia). In this region, cattle consume predominantly native C4 grasses, which are susceptible to substantial seasonal changes in nutritional value. The MP from Australian C4 grasses and associated changes in methane and methanogenic archaeal populations within the rumen are not well described. This thesis aimed to quantify methane from C4 tropical pastures and associated rumen microbiota during seasonal changes in forage quality. Four experiments were conducted to achieve these objectives. Experiment 1 (Chapter 3) measured cumulative gas production in vitro by Ankom Gas Production System batch culture method. Bottles containing ground forages and buffered rumen fluid were incubated for up to 48 h and pH, volatile fatty acids (VFAs); total gas production (TGP) methane (CH4), dry matter degradability (DMd) and organic matter degradability (OMd) were measured. Substrates were a variety of tropical C4 grasses (obtained in Experiment 4) ranging in CP from 29 g/ kg DM to 120 g/ kg DM and digestibility from 38 - 60%. A legume was included for comparison. A general linear model determined effect of forage type and time on variables. Regression relationships were determined between nutritive characteristics and gas variables. More gas, methane and total VFA (mmol/L) were produced when forages were more degradable. Positive linear relationships were observed between CP and both TGP and methane. Negative relationships were observed between fibre and methane. Experiment 2, (Chapter 4) determined contribution of inoculum to fermentative characteristics when donor cattle were fed forage diets representative of wet or dry season in northern Australia. The experimental design was the same as for experiment 1. The substrates included the Mitchell grass, and the Lucerne from experiment 1 and two other forages a pasture (PAS) Urochloa mosambicensis (CP 90 g/ kg DM’ DMD 63%) and a low quality hay (LQH) sample of Chloris gayana (CP 31 g/kg DM; DMD 41%). Inoculum was collected from cattle consuming PAS (PASi) and LQH (LQHi) diets. A general linear model was used with forage type, time and inoculum source as fixed effects. Substrates incubated in PASi had greater MP and lower degradability than the same substrates incubated in LQHi. The in vitro experiments confirm that forage quality is the primary factor effecting gas production but under closed batch system conditions, specific effects of inoculum are observed. Two in vivo experiments (Chapter 5 – 8) were conducted to quantify MP and microbial communities associated with seasonal change in forage quality. In Experiment 3, Bos indicus steers were fed a low quality C. gayana hay and then switched to either moderate quality U. mosambicensis pasture (PAS), a high quality C. gayana hay (HQH) or remained on low quality hay (LQH). Individual measurements of DMI and rumen outflow rates were taken and rumen fluid was sampled for analysis of pH, VFAs, NH3-N and microbial populations (using comparison of the 16S rRNA gene). Open circuit respiration chambers measured daily MP. Higher DMI, MP, and rumen outflow rates accompanied higher forage quality in PAS and HQH compared with LQH steers, although all produced approximately 19.8 g methane /kg DMI. Low quality hay produced more methane /kg digestible DMI and none of the diets reached predicted daily emissions compared with the equations in use for the national greenhouse gas inventory. Analysis of the prokaryotic community revealed predominance of the genus Prevotella and an unassigned genus in the family Ruminococcaceae in bacteria, and of Methanobrevibacter and Thermoplasmatales in the archaea. Finally, Experiment 4, a longitudinal design was conducted at Brunette Downs Station (Barkly Tableland, Northern Territory) in which rumen fluid was collected from 10 heifers grazing Mitchell grass in two wet seasons (May 2012, March 2013) and the intervening wet season (August, November 2012). Regardless of sampling month during the year, dominant genera of bacteria and methanogenic archaea were similar, although differentiation in relative abundance between months was observed. Prevotella and Clostridial genera were predominant in the bacteria and Methanobrevibacter and Thermoplasmatales associated lineages in the methanogens. Methylotrophic methanogens appeared to comprise a greater proportion of the population in both in vivo experiments when forage quality was improved. However, one species identifying with Thermoplasmatales was negatively affected by increasing rumen outflow rates. It can be concluded that methane emissions from cattle consuming tropical grasses was lower than predicted. The experimental analysis confirmed Methanobrevibacter to be the dominant methanogen in northern beef cattle, however, Thermoplasmatales affiliated methanogens make up a significant proportion of the methanogenic population in cattle consuming tropical C4 grasses in Australia. Culturing a representative of this clade would greatly improve understanding of methanogenesis in Australian tropically adapted beef cattle.