Abstract
Pathogen-mediated balancing selection shapes host ecology and evolution across the tree of life, fueling a co-evolutionary arms race based on frequency-dependent adaptations of hosts and counter-adaptations of pathogens. While rare immune genotypes are predicted to be more resistant to pathogens, evidence for this rare allele-advantage and negative frequency-dependent feedbacks has only been observed in model species or inferred from short-term field observations. Evidence from long-term data and wild populations is missing. Here, we leverage two decades of immune genetic and disease surveillance data from over 1,500 wild meerkats (Suricata suricatta) to reveal co-evolutionary dynamics between the Major Histocompatibility Complex (MHC) and Mycobacterium suricattae, causing tuberculosis (TB). We uncover fluctuating adaptive MHC allelic, functional and haplotypic diversity over time compared with stable neutral genetic diversity. Crucially, we show that meerkats carrying the MHC allele Susu-DRB*13 faced initially higher TB infection probability, with the effect reversing over the course of the study, followed again by an increase in frequency of Susu-DRB*13. Similarly, TB progression first accelerated in meerkats carrying Susu-DRB*13, but decelerated thereafter. Susu-DRB*13 (and its supertype/haplotype) also prolonged survival in individuals with clinical TB. Our results present strong evidence that the rare allele-advantage propels negative frequency-dependent selection in a wild mammal. We discuss meerkats’ social structure as possible reasons for these comparably rapid co-evolutionary dynamics.