Back

Spotted hyena gut cross-talks:Symbionts modulate mucosal immunity

Veloso Soares, S. P.; JarquinDiaz, V. H.; Veiga, M. M.; Karl, S.; Czirjak, G. A.; Weyrich, A.; Metzger, S.; East, M. L.; Hofer, H.; Heitlinger, E.; Benhaiem, S.; Martins Ferreira, S. C.

2024-07-25 ecology
10.1101/2024.07.25.605121 bioRxiv
Show abstract

The intestinal mucosa is at the front line of host-microbiome interactions, but little is known about these interactions within natural populations. Here, we non-invasively investigated associations between the gut microbiome and mucosal immune measures while controlling for host, social, and ecological factors in 199 samples of 158 wild spotted hyenas (Crocuta crocuta) in the Serengeti National Park, Tanzania. We profiled the microbiome composition, including bacteria, fungi and parasites, using a multi-amplicon approach, and measured faecal immunoglobulin A and mucin. Probabilistic models indicated that both immune measures predict microbiome similarity among individuals in an age-dependent manner. The strength of the association effect varied, being strongest within bacteria, intermediate within parasites, and weakest within fungi communities. Machine learning regression accurately predicted both measures and identified the taxa driving these associations: symbiotic bacteria reported in humans and laboratory mice, unclassified bacteria, a hookworm, host DNA likely reflecting inflammation, and diet. Our findings indicate a complex interplay between the host, its environment and symbionts. These findings increase our knowledge of the gut microbiome in natural populations, which harbour highly dynamic and diverse eukaryotes under the influence of unpredictable environmental factors and where selection is not artificially biased.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.