Can We Select for an Improved Microbiome?

Feed efficiency (FE) is an important trait in the pig industry, as feed costs are responsible for the major part of production costs. Availability in the market and cost of feed ingredients dictate changes in feed composition. As a result, fibre level and composition can vary between pig diets.

Results of a retrospective study on digestibility trials of feed ingredients performed at Schothorst Feed Research, suggests that crude protein and crude fibre digestion are heritable traits (Ouweltjes et al., 2017). Pigs don’t digest crude fibre themselves. Microbiota in the gastrointestinal tract play an important role in fibre digestion, because they produce enzymes that break down fibre structures and deliver volatile fatty acids to the pig. These volatile fatty acids can be used as metabolic energy source.

To investigate the association between FE and faecal microbiome in commercial grower-finisher pigs we set up an experiment (Verschuren et al., 2017); three-way crossbreed grower-finisher pigs (154) were either fed a diet based on corn/soybean meal (CS) or on wheat/barley (WB). Faecal samples of the day before slaughter (mean bodyweight 122 kg) were sequenced for the V3V4 16S ribosomal DNA regions and clustered according to operational taxonomic units (OTU) for each individual, forming a table of OTU abundancy. A partial least square regression was applied to the dataset, together with a discriminant analysis using principal components of FE extreme groups (10 high and 10 low FE animals for each diet x sex-combination). Pigs on different diets and males vs. females had a very distinct microbiome, needing only two OTUs for diet (P=0.018) and 18 OTUs for sex (P=0.002) to separate the groups. Faecal microbiome was not related to FE groups fed the CS diet, but was related in the WB diet, be it sex specific. In conclusion, our results show a diet and sex dependent relationship between the faecal microbial composition and FE in grower-finisher pigs at slaughter weight.

A different approach to quantify the effect of GIT microbiota in feed efficiency traits is given by Difford et al. (2016). He replaced the traditional G-matrix based on the genotypes of the animals by those of the microbiota and called this the M-matrix. The fraction of the variance explained by the M-matrix was termed microbiability (analog to the heritability applying the G-matrix). Carmarinha-Silva et al. (2017) applied this technique in a study on 207 pigs. The fraction of the phenotypic variance explained by the microbial variance was always higher compared to the heritability for the traits daily gain, feed intake and feed conversion ratio.

All three studies showed the importance of GIT microbiota for FE traits but are inconclusive in answering the question: Can we select for an improved microbiome? To be able to, we need to disentangle the effect of the genotype of the animal (G-matrix) and the microbiota (M-matrix) and be able to estimate the covariance between them. For that, large numbers are required which currently are collected.