PLANT SECONDARY METABOLITES AS A STRATEGY TO REDUCE METHANE EMISSIONS AND IMPROVE FEED EFFICIENCY IN RUMINANTS

The livestock supply chains account for about 14.5% of anthropogenic greenhouse gas emissions. Enteric fermentation represents the second largest source of these emissions, being responsible for approximately 40% of total livestock emissions, mainly in the form of methane (CH4). It is crucial to reduce enteric CH4 emissions from ruminant production in order to limit the increase in global temperature to 1.5°C by 2050. For these reasons, the general aim of this PhD project was to evaluate the effects of several nutritional strategies to reduce CH4 production from ruminants. The present thesis includes the results of five experiments, three conducted inside the CASCO Project (Evaluation of an integrated system for the control of gastrointestinal parasitosis of goat farms: environmental, agricultural-zootechnical and health implications), while the other two were performed during my year abroad at the Department of Animal and Veterinary Sciences of Aarhus University (DK), being part of the ECOCO2W project (Tannins of willow and hemp as organic feed additives for methane reduction in dairy cows). All experiments aimed to evaluate different plant secondary metabolites (PSM), included in ruminant diets, in order to reduce rumen methanogenesis. 1 - In the first paper, originated from the ECOCO2W project, eight PSM (gallic and salicylic acids, tannic acid, catechin, epicatechin, quercetin, rutin, and salicin), that had previously been found in willow and hemp, have been evaluated in an in vitro system (simulating rumen fermentation) for their anti-methanogenic potential when included to maize silage at a concentration of 12% of the feed dry matter. After 48 h of incubation in buffered rumen fluid, CH4 production was significantly reduced by 43% with addition of quercetin (but not the other flavonoids catechin, epicatechin, or rutin), 39% by tannic acid (but not its subunit gallic acid) and 34% by salicylic acid compared to control. The observed differences in the anti-methanogenic effects of the PSM tested were related to lipophilicity, the presence of double bond and carbonyl group, sugar moieties, and polymerization of the compounds. The reduction in CH4 production by quercetin, tannic acid and salicylic acid was not associated with changes in total volatile fatty acids (VFA) concentration compared to control but was associated with changes in the composition of the individual VFA. Moreover, the flavonols quercetin and rutin, the flavanols catechin and epicatechin, and gallic acid showed an in vitro degradability higher than 90%. 2 - Due to the interesting results observed in the first paper, quercetin, salicylic acid, tannic acid, and catechin (that in the first experiment showed a numerical reduction in CH4 production), were added to two different basal feeds (maize and grass silages) at three inclusion doses of 1.5, 3 and 6% of the feed DM in order to study a possible dose-response effect and an interaction among the PSM and the substrate. Quercetin was the only tested PSM that significantly reduced CH4 production. The extent of the reduction in CH4 production induced by quercetin was different among the substrates and increased linearly with increasing dose, i.e., by 51 and 43%, respectively for maize and grass silages, at a dose of 3% of feed DM, and by 86 and 58%, respectively for maize and grass silages, at a dose of 6% of feed DM. Positively, the reduction in CH4 production was not associated with a reduction in total VFA concentration. Moreover, quercetin significantly reduced NH3 concentration by >12% at doses of 3 and 6% in feed DM irrespective of the basal feed used as compared to when the basal feeds were incubated alone. Although none of the other additives affected CH4 formation, several additives had significant impacts on concentrations of NH3 and VFA in the incubated fluid after fermentation. Therefore, from the results of these two first experiments, it emerged that quercetin is a promising candidate for in vivo trials as CH4-reducing compound. Further studies should investigate the precise mechanism of action and the possible effects on milk production, also in long-term studies. The third, fourth, and fifth papers were born from experiments carried out within the CASCO project, and therefore were focused on the use of tannins in the feeding of ruminants and mainly goats. 3 - In the third paper, an in vitro experiment was conducted using two commercial feed additives: one was a quebracho condensed tannin (CT) extract (70% based on DM) and the other was a chestnut hydrolysable tannin (HT) extract (75% based on DM). The two commercial extracts have been tested at four levels of inclusion (2, 4, 6, 8% on an as-fed basis) against a negative control. Both types of tannins significantly reduced total gas and CH4 production during the 48 h of incubation, although quebracho CT was more effective than chestnut HT. The lower total gas and CH4 production levels were linked to the reduction in dry matter degradability caused by CT and HT, although no difference was observed for total VFA concentration. Furthermore, no significant differences were observed for the protozoan and archaeal populations, suggesting a low direct effect of tannins on these rumen microorganisms in vitro. As expected, both type of tannins linearly reduced the NH3 content. Therefore, due to the more promising results, quebracho CT extract was chosen to be tested in an in vivo experiment with lactating dairy goats that resulted in two papers (the fourth and the fifth paper of this thesis). In this in vivo trial with eight lactating goats, the effects of four levels of inclusion of the quebracho commercial CT extract (0, 2, 4, 6% on DM basis; C, Q2, Q4, Q6, respectively) were investigated. The experimental design was a repeated 4 × 4 Latin square with 28-d periods (24 d of diet adaptation and 4 d of sample collection) using metabolic cages and four open circuit respiration chambers. 4 - In the fourth paper it was showed that the inclusion of CT in the diets did not affect the dry matter intake (DMI), but caused a linear decrease in diet digestibility, with reductions up to -10.7% for DM, -21.1% for crude protein (CP), -23.1% for neutral detergent fiber (aNDFom), and -12.9% for gross energy (GE), when comparing the Q6 and C diets. Ruminal total VFA concentration was not affected by quebracho CT extract, although there were changes in VFA proportions. Milk yield (g/d) was highest for Q4 (3,371) and lowest for Q6 (3,066). In terms of milk composition, quebracho CT extract induced a linear reduction of fat and CP concentrations. The reduction in CP digestibility resulted in a linear reduction in the milk urea level, up to -36.8% with Q6. Positively, quebracho CT extract linearly reduced the somatic cells count expressed as linear score. The feed efficiency was linearly decreased by quebracho CT extract inclusion. Furthermore, a shift from urinary to faecal nitrogen excretion was observed with quebracho CT extract. The CH4 yield (on average 19.2 g CH4/kg DMI) was linearly reduced by quebracho CT extract inclusion, up to -17.8% with Q6. Regarding the CH4 intensity, quebracho CT extract induced a linear reduction when expressed per kg of milk, but not per kg of fat and protein corrected milk. Moreover, the CH4 production per kg of digestible aNDFom was linearly increased by quebracho CT extract. The metabolizable energy (ME) intake was not affected by the treatments, but the metabolizability (q = ME intake/GE intake) was reduced as quebracho CT extract inclusion increased. 5 - In the fifth paper the ruminal and faecal microbiota of the goats in the trial were evaluated. Regarding the ruminal microbiota, the prokaryotic alpha- and beta-diversities were influenced by quebracho CT extract. The archaeal genus Methanobrevibacter, together with the corresponding phylum (Euryarcheota) and family (Methanobacteriaceae) were reduced by Q6. Moreover, the phylum Bacteroidota was enriched by all three CT diets, leading to an inversion of the ruminal Bacteroidota/Firmicutes ratio when compared to the C diet. The family Ruminococcaceae was reduced by all three diets, while Prevotellaceae and Acidaminococcaceae were enriched by Q6. Of interest was also the reduction in Synergistaceae and Flexilinea and Fretibacterium, as well as the enrichment in Succiniclasticum and Fibrobacter. The quebracho CT extract did not induce significant changes in the overall fungal biodiversity of the rumen. The ruminal protozoan microbiota was dominated by Entodinium, followed by unclassified members of Trichostomatia and Ciliophora. No changes in abundance were observed for Entodinium; on the other hand, Diplodinium abundance decreased, while Polyplastron and Isotrichia abundance increased at Q6. Considering the faecal microbiota, no significant effects on prokaryotic alpha- or beta-diversities were observed. Notably, the faecal microbiota showed an opposite behaviour for some relevant prokaryotic taxa, as the genus Methanobrevibacter and the corresponding phylum and family (Euryarcheota and Methanobacteriaceae, respectively) were significantly enriched in the faeces by dietary integration with quebracho CT extract. In conclusion, increasing levels of dietary integration with quebracho CT extract induced increasingly relevant changes in the rumen microbiota. Quebracho CT extract successfully reduced the abundance of microbial taxa responsible for CH4 production, with a possible shift towards a lower CH4-producing microbial community state. On the other hand, the faecal microbiota showed less relevant changes and followed an opposite trend concerning CH4-producing taxa. These differences could be explained by two different causes. Firstly, quebracho CT could be at least partially degraded in the rumen, as evidenced for their subunits in the first paper, thus losing part of their antimicrobial activity. Secondly, in the fourth paper was evidenced that quebracho CT extract reduced feed digestibility, thus increasing the amount of potentially degradable substrate that has reached the last part of the digestive tract and favouring the proliferation of overmentioned microorganisms. Therefore, merging the results of the fourth and fifth papers, it is possible to conclude that quebracho CT extract has an anti-methanogenic activity that is partially explained by its antimicrobial activity but mostly due to the reduction in ruminal feed degradability. Moreover, quebracho CT extract reduced feed efficiency and did not improve N balance. According to these findings, the positive environmental impacts of quebracho CT extract are mainly related to the shift from urine to faecal N excretion. In conclusion, PSM can represent a strategy for CH4 mitigation. However, due to their chemical variety and the different effects that PSM can induce in ruminants, the identification of pure compounds with a clear anti-methanogenic effect is recommended.