Nutritional factors affecting greenhouse gas production from ruminants: implications for enteric and manure emissions
Animal agriculture has been identified as one of the major sources of greenhouse gases (GHGs), accounting for approximately 40% of total agricultural-related emissions (IPCC, 2006) (Fig. 1). Animal production and manure management comprises 26.8% and 31.0%, respectively, of the 7.1 Gt of C02-eq that the livestock sector is estimated to produce annually (Gerber et al., 2013). The two central GHGs emitted directly from animal agriculture include methane (CH4) and nitrous oxide (N,0) which have 28 and 298 times the global warming potential of CC>2, respectively (Gerber et al., 2013). Livestock CH4 and N20 emissions have been estimated to contribute 40% and 48% of livestock sector emissions and ruminants account for 80% of the total livestock sector's emissions (Opio et al., 2013). Enteric fermentation and manure-related CH4 contribute 82% and 18% of CH4 related to livestock production, respectively. The main sources of N20 emissions arise from chemical fertilisers, applied
http://dx.doi.org/10.19103/AS.2020.0067.16 Published by Burleigh Dodds Science Publishing Limited. 2020.
Figure 1 Greenhouse gases associated with animal agriculture.
manure and N deposition from housed animals and manure storage (Adler etal.,2015).
Greenhouse gas production
Ruminants produce CH4 as a natural by-product of microbial fermentation, with the biochemical pathways being well documented (Huws et al., 2018). Starches, cell wall polymers and proteins are fermented by a consortium of rumen microbiota into simple sugars and carbon skeletons. Both primary and secondary fermenters convert these products, under anaerobic conditions, into volatile fatty acids (VFA), CO, and metabolic hydrogen [Н]. Rumen ciliates and anaerobic fungi are two groups of eukaryotes which produce large volumes of [H] and share a commensal relationship with archaea (Guyader et al., 2014). Both protozoa and fungi contain hydrogenosomes which are specialised organelles that are responsible for the conversion of the intermediates of monosaccharide fermentation into [Н]. Methanogens play an important role in maintaining a low partial pressure of [H], favouring hydrogenase activity within hydrogenosomes. During fermentation, the reduced co-factors NADH, NADPH and FADH are oxidised and the released [H] is transferred to methanogenic archaea through a series of biochemical pathways to reduce C02 to CH4 (Ungerfeld, 2015b).
Manure CH4, like enteric CH4, is produced during anaerobic decomposition of organic matter (OM). Manure is also a significant emitter of N,0, ammonia
Figure 2 Process of nitrification and denitrification via nitrite pathway. N,0 is a greenhouse gas with a global warming potential that is 298 times that of CO,.
(NH3) and nitrous oxides (NOx). These gases may act as direct or indirect sources of GHGs and environmental pollutants. Factors which influence the concentration of N20, NH3 and NOx include (i) the type of feed, (ii) manure nutrient profile and (iii)the handling and storage of manure. The conversion of N into gases occurs through the simultaneous nitrification and denitrification process (Fig. 2). Nitrification occurs by both NPI3 oxidising bacteria (i.e. Betaproteobacteria or Thaumarchaeota NPI3-oxidizing archaea) and nitrite (N02~) oxidising bacteria (i.e. Alphaproteobacteria Nitrobacter and Nitrospira). Denitrifying bacteria are phylogenetically diverse and have specific genes coding for their catalytic enzymes (Maeda et al., 2011).
Although not a direct source of GHG, NH3 emissions from manure should also be considered when assessing the impact of feed additives on air quality. Ammonia arises from the rapid hydrolysis of urea in urine and can also be a precursorto N20. Ammonia is highly volatile and can have serious implications for human health when threshold limits of 25-35 ppm are exceeded (National Research Council, 2008). Additionally, dry or wet NH3 deposition may contribute to soil acidity and eutrophication of surface water (Hunerberg et al., 2013a). Shifting the excretion of N from urine to faeces may be more environmentally beneficial as faecal N is considered a slow-release form of N that is more likely to be captured by soil flora.
Balancing enteric methane production and manure emissions
Significant interrelationships exist between enteric CH4 production and both CH4 and N,0 manure emissions (Knapp et al., 2014). Balancing net emissions produced directly from the rumen or indirectly from manure is challenging.
Published by Burleigh Dodds Science Publishing Limited. 2020.
Figure 3 Consequences of dietary manipulation on enteric production and greenhouse gas emissions. Symbols indicate: f = increase, J, = decrease, - = no change, NA = not applicable, ~ = variable/unknown.
Another important consideration when evaluating dietary manipulation strategies is accounting forthe difference in GHG production and GHG intensity. Although global GHG emissions from ruminants have decreased as a factor of animal product (intensity), the total production of GHG has increased, and will continue to do so as the world's domestic ruminant population is projected to increase from 3.2 to 5.3 billion by 2050 (Turk, 2016).
A standard feedlot diet fed to cattle may result in a higher CH4 production (g/d) and a reduced CH4 intensity (g/kg consumable product) than those grazing on pasture (low CH4 production and high CH4 intensity). However, due to low dietary energy content, pasture-raised ruminants produce manure with half the CH4 yield potential of those raised in feedlots (Koneswaran and Nierenberg, 2008), as pasture-raised animals have much lower starch levels in manure (Hales et al., 2013). Additionally, dietary alterations which result in a shift of fermentation from the rumen to the hindgut may decrease enteric CH4 production, but not change overall net GHG emissions. This concept is also known as pollution swapping, in which an alteration in the production of one GHG results in an upstream or downstream change in the emissions of the same or another GHG (Hristov et al., 2013). Nutritional strategies that alter diet digestibility through increasing dietary fermentable carbohydrates, N and fat content can all result in pollution swapping (Fig. 3).
Published by Burleigh Dodds Science Publishing Limited, 2020.
Diet digestibility and fermentable carbohydrates
Diet digestibility is intrinsically linked to enteric and manure GHG production. The more readily a diet is fermented, the lower the nutrient wastage and GHG emission intensity. Factors such as forage quality, forage-to-concentrate ratio and type of concentrate/forage can all contribute to the microbial efficiency of feed digestion. For example, increasing the quantity of concentrates in the diet can reduce enteric CH4 production through a greater proportion of easily fermentable carbohydrates. This can shift fermentation from acetate which produces [H], towards propionate which utilises [H] in its synthesis and consequently, decreases the availability of [H] for methanogenesis. Increasing fermentable carbohydrates within the diet can also increase digestibility and passage rate. This can both improve productive performance and decrease the amount of OM excreted in the manure. Less OM in manure reduces the amount of substrate available for decomposition and thus the supply of [H] to methanogens, decreasing manure CH4. Alternatively, increasing the digestibility of the diet can also increase enteric CH4 on a g/d basis as more substrates are fermented in the rumen and the production of reducing equivalents increases. For example, lactating dairy cows grazing pasture have shown to have an increase in CH4 production (g/d) and milk yield (van Wyngaard et al., 2018; Munoz et al., 2015) when supplemented with increasing concentrates (0-8 kg/d).
Increasing concentrate in the diet can increase the protozoa, Entodinium in the rumen, a known non-fibrous carbohydrate degrader. Whereas microbes associated with cellulolytic degradation including Fibrobacter, Polyplastron and Ostracodinium decrease as the level of concentrate increases in the diet (Zhang et al., 2017). The diversity and richness of fungal communities were similar in high-forage versus high-concentrate diets; however, the relative abundance of the fungal phyla Ascomycota, Basidiomycota, Cercozoa and Chytridiomycota increased, and Neocallimastigomycota decreased with increasing proportions of concentrate (Zhao et al., 2018). Other studies also report that fungal communities have been enriched as the forage proportion of the diet increases (Kumar et al., 2015), reflecting their role in the degradation of complex fibre. Changes in eukaryote abundance and diversity are likely to impact methanogen abundance and diversity as eukaryotes produce the [H] required for methanogenesis.
A high-concentrate diet decreased the overall abundance (richness) of the archaeal population, but did not change the range of microbial species (diversity) (Zhang et al., 2017; Mao et al., 2016). Despite their role in methanogenesis, archaeal communities have been reported to show less variation and diversity as a ruminant adapts from a high-forage to a high-concentrate diet (Henderson et al., 2015; Kumar et al., 2015). This may be due to their low density and their less-diverse metabolic capabilities (Kumar et al., 2015; Henderson et al., 2015). However, Methanomicrobium (Methanomicrobiales order) and Methanomicrococcus (Methanosarcinales order) are reported to be sensitive to dietary changes with both these taxa being absent in high-grain diets and only detected in forage diets (Friedman et al., 2017a). The absence of Methanomicrococcus (Methanosarcinales order) in high-grain diets is likely related to an increase in redox potential associated with a lower rumen pH (Friedman et al., 2017a).
Diets composed of highly fermentable substrates can result in conditions where organic acid production by the microbial population exceeds the buffering capacity of the rumen, leading to a prolonged reduction in rumen pH. Ruminal acidosis is characterised by a reduction in microbial diversity and rumen malfunction including decreased feed intake and feed digestibility. Abundance and diversity of bacteria were decreased in sheep ruminal fluid (Li et al„ 2017) and ruminal fluid and faeces in dairy cows (Plaizier et al., 2017) with induced sub-acute ruminal acidosis. High-grain diets usually result in an increase in starch and lactate utilisers as well as propionate producers (Prevotella, Selenomonas, Megasphaera, Streptococcus) (Plaizier et al., 2017; Zhu et al., 2018). Specifically, Prevotella and Succinivibrionaceae dominate the rumen of ruminants-fed high-grain diets (Henderson et al., 2015). In contrast, fibrolytic bacteria including Butyrivibrio, Ruminococcus and Fibrobacter are vulnerable to low pH and usually decrease in abundance in high-grain diets (Zhu et al., 2018). Whilst other studies report a decrease in fungal diversity (Kumar et al., 2015; Tapio et al., 2017a), Ishaq et al. (2017) found that abundance and diversity of rare fungal taxa was increased with diet-induced sub-acute acidosis, including those associated with lactic acid utilisation (Pichia and Candida). The abundance of the archaeal population has been shown to not change with increasing concentrate in the diet and decreased ruminal pH (Hook et al., 2011). This suggests that they are resilient to changes in pH and only their functional activity is suppressed under conditions of low pH. This may explain why there is a poor relationship between methanogen abundance and CH4 production (Firkins and Yu, 2015).
Altering the ability of ruminal microbes to degrade feed can increase nutrient loss in the faeces, increasing the amount of OM available for CH4 production from manure. Although increasing the concentration of ruminal escape starch in the diet can modulate fermentation and potentially reduce methanogenesis, starch digestion can also be limited (<60%) in the lower digestive tract (Haque, 2018). This results in more starch in the faeces, potentially increasing CH4 emissions during decomposition of the manure. Ruminococcaceae was more abundant in manure of cattle-fed processed grain and forage-fed diets, whereas Prevotella dominated in manure from cattle- fed unprocessed grain (Shanks et al., 2011). Bacteroidetes, as a reflection of their role in the digestion of complex carbohydrates, increased and Firmicutes decreased with increased levels of starch in the faeces. The concentration of starch in faeces from cattle-fed unprocessed grains and processed grains was 98.4% and 66.9% higher, respectively, than in those fed forage, suggesting that inadequate grain processing could increase CH4 emissions from manure (Shanks et al., 2011).
Improving diet digestibility and thereby the potential of the ruminant to obtain nutrients from feed improves overall growth efficiency. This may lead to an increase or decrease in total GHG emissions depending on the types of gases produced and the balance between enteric and manure emissions. Regardless, if there is an actual decrease in GHG when production from the animal is improved (i.e. less days on feed), GHG emission intensity as a proportion of usable product (i.e. meat, milk, wool, etc.) is reduced (Hristov et al., 2013).
Ruminants are a significant contributor to the global N budget. In ruminants, nitrogen cycles through a series of complex biogeochemical interactions involving inorganic- and organic-N in feed, manure and soils (Fig. 4; Robertson and Vitousek, 2009). Plants and animals utilise N throughout this cycle; however, they are limited in their ability to deposit it as a product. For example, the conversion of dietary N into consumable protein (i.e. milk, meat) by ruminants is very low (20-30%) and fertiliser-N recovery by cereal crops seldom exceeds 50% of applied N (Fageria and Baligar, 2005). Excess N from animal agriculture results in the release of a large surplus of reactive N into the environment, mainly via NH3 and N20 emissions and/or nitrate (N03-N) leaching (Galloway et al., 2004; Powell et al., 2011). Pollution of groundwater by N03-N, widespread eutrophication and global warming through NzO emissions
Figure 4 Nitrogen and carbon cycle within ruminant agriculture.
are some well-documented contributions of agricultural-anthropogenic N (Erisman et al., 2013).
The cycle of N within ruminant systems is mainly defined by the transformation of feed-N into milk or meat products, with the remaining N excreted in urine or manure.The concentration of N in urine and faeces depends on the crude protein (CP) content of the diet (Dijkstra et al., 2013). Feeding ruminants to the level of metabolisable protein requirements ensures the best utilisation and the least loss of nutrients (Broderick, 2003). The amount and type of N fed in ruminant diets also has several implications for how it is utilised and excreted by ruminants. Dietary protein supplies both rumen-degradable and -undegradable protein. Rumen-degradable protein is composed of true protein and non-protein N, which when broken down can be utilised for microbial protein synthesis and growth (Bach et al., 2005). Requirements for dietary protein and energy are intrinsically linked, as high-energy diets will stimulate microbial synthesis, enhancing the need for rumen-degradable protein (Broderick, 2003). Whilst changing the CP content in the diet has no obvious direct effect on enteric CH4, its replacement by carbohydrates can influence emissions.
Replacing protein supply with fermentable carbohydrates is an effective way to reduce urinary N excretion, increase microbial N capture and decrease NH3 production (Dijkstra et al., 2013). However, replacing dietary CP with fermentable carbohydrates can exacerbate enteric CH4 production (Sauvant et al., 2011) as increased substrates are available for ruminal methanogenesis. Dijkstra et al. (2013) estimated that higher enteric CH4 fluxes from increased carbohydrates in the diet are frequently offset by a decrease in N20 emissions from manure.
Prevotella is a predominant genus within ruminants around the globe (Henderson et al., 2015) and participates in both carbohydrate and N metabolism. Specifically, Prevotella ruminicola strain 23 can efficiently degrade hemicellulose and pectin, utilising both NH3 and peptides as a N source for growth as opposed to amino acids (Kim et al., 2017). Their importance in N metabolism was supported by an observed decline in Prevotella bryantii in dairy cattle fed a low-protein diet (Belanche et al., 2012). In this study, the relative abundance of Ruminococcus albus, Ruminococcus flavefaciens, Fibrobacter succinogenes and Butyrivibrio fibrisolvens all declined, potentially highlighting the vulnerability of cellulolytic bacteria to N shortages. In contrast, non- cellulolytic bacteria including Prevotella ruminicola, Selenomonas ruminantium, Streptococcus bovis, Megasphaera elsdenii and Aliiglaciecola lipolytica were not affected by a N shortage which may reflect their low NH3 requirements for growth (Belanche et al., 2012). Niu et al. (2016) evaluated the effect of two different concentrations of CP (15.2% vs 18.5%) in dairy cows and demonstrated that the low-protein diet decreased total tract digestibility of OM, N and starch compared to the high protein diet (18.5%). Belanche et al. (2012) indicated that ruminal concentrations of protozoa and methanogens declined with the low- protein diet, whereas Niu et al. (2016) observed no differences in CH4 emissions as a result of differing protein content in the diet.
Degradation of dietary protein and its assimilation into microbial protein can also contribute to a decrease in available [H] for methanogenesis as this process can both utilise and produce reducing equivalents (Knapp et al., 2014). For example, the synthesis of amino acids can increase with a decrease in methanogenesis, as they act as a [H] sink (Ungerfeld, 2015b). An increase in amino acid production could be related to an increase in the relative abundance of Bacteroidetes and Prevotella species, both associated with increased proteolytic activity (Martinez-Fernandez et al., 2016). Increasing the proportion of soluble carbohydrates in diets has been related to decreases in branched-chain fatty acids, which are needed for de novo synthesis of amino acids by ruminal microbes. Decreasing availability of branched-chain fatty acids could decrease microbial protein synthesis and microbial growth (Hall and Huntington, 2008) and thereby reduce the extent to which this process acts as an alternative [H] sink to CH4 production.
Potential interactions between dietary protein content and ruminal methanogenesis remain unclear. Addition of protein to high-fibre diets could increase the efficiency of microbial protein synthesis and reduce the intensity of enteric CH4 emissions, while increasing N excretion and N20 missions from manure. Increasing the efficiency of microbial protein synthesis could redirect [H] away from methanogenesis towards the formation of microbial cells and increase the productivity of the ruminants.
Dietary additives which target microbes involved in methanogenesis may be superior at decreasing enteric CH4 production without affecting manure emissions. For example, dietary fats may decrease enteric CH4 production by (i) having a toxic effect on methanogens and protozoa, (ii) replacing fermentable carbohydrates or (iii) providing an alternative [H] sink via biohydrogenation (Beauchemin et al., 2008; Knapp et al., 2014).
The effects of dietary lipids on the rumen are largely dependent on fat composition, concentration and source. Consequently, the impact on the rumen microbial populations varies depending on the nature of the oil. Methanobrevibacter ruminantium is the most abundant species of rumen methanogenand was found to be reduced by saturated fatty acids and oleic acid (Henderson etal.,2015; Enjalbert et al., 2017). Addition of linseed and coconut oil decreased CH4 production, but this was not correlated with changes in the abundance or diversity of the archaeal population (Patra and Yu, 2013; Martin et al., 2016). Similarly, lambs fed linseed oil had a higher relative abundance of Succinivibrionaceae (succinate producers) and Veillonellaceae (propionate producers) and a decreased abundance of Ruminococcaceae (Lyons et al., 2017). An increased abundance of Succinivibrionaceae and Ruminococcaceae has been associated with low and high CH4 emitters, respectively (Wallace et al., 2015). Succinivibrionaceae produce succinate via utilisation of [H], whereas Ruminococcaceae are known hydrogen producers (Wallace et al., 2015). Lambs supplemented with linseed oil had a 19.5% decrease in the relative abundance of Methanobrevibacter and a 34.7% increase in Methanosphaera, although CH4 emissions were not measured in this study.
The growth of the rumen fungus, Neocallimastix frontalis, was impeded by soybean oil (Boots et al., 2013) and others have found that the fibrolytic bacteria, Fibrobacter and Ruminococcus are also inhibited by lipids (Enjalbert et al., 2017). Increasing the degree of unsaturated fatty acids in the diet may correlate with decreased protozoal counts (Oldick and Firkins, 2000), and as a result of disrupting the close relationship between protozoa and archaea, change the diversity of archaeal communities (Hristov et al., 2012).
Alteration of the ruminal microbiome due to the type of fat may prompt a high variation in physiological responses including an inhibition of fibre digestion which may decrease CH4 production. It is well documented that increasing fat content in ruminant diet above 6-7% dry matter (DM) can reduce the digestibility of fibre (Johnson and Johnson, 1995). In continuous culture, an oil (Tucuma) high in oleic acid inhibited CH4 at 1 % (v/v) through suppression of Fibrobacter with no alteration in methanogens (Ramos et al., 2018). Decreasing fibre digestibility can result in an increase in manure C, providing substrate for CH4 emissions. Although, Gautam et al. (2016) found that varying sources of dried distillers grain plus solubles (DDGS) with corn oil (dietary fat of 3-5.5% dietary DM) had no effect on nutrient composition or GHG emissions from manure.
Dietary fat supplementation is an effective enteric CH4 mitigation strategy. Depending on their fatty acid composition, dietary fats may decrease both daily and the intensity of CH4 production. However, the main constraint with dietary fat is that the amount that can be supplemented without inhibiting fibre digestion is restricted to =6% of dietary DM, so the CH4 mitigation potential is limited to 10-15%.