Introduction
More than 5,800 animal species to date are endangered, as the Earth experiences a mass extinction event (Ceballos et al., 2015). Conservationists have multiple factors to consider in order to arrest population declines. Extrinsic of population declines include habitat loss and degradation, over-exploitation, emerging infectious diseases, invasive species, and climate change (Butchart et al. 2010). Intrinsic drivers of extinction, such as genetic factors, also play a key role for population viability, especially when species are reduced to small, isolated populations that can be negatively affected by genetic load (Hedrick, 2001). In this scenario, finding a suitable mate is challenging and reproduction with related individuals can occur, leading to inbreeding depression. Inbreeding has largely been documented in small populations in the wild (Hedrick, 2001; reviewed in Spurgin & Gage, 2019), impacting individual and population fitness through the fixation of detrimental alleles. An increase of detrimental alleles in endangered species increases their susceptibility to extrinsic ecological drivers of extinction (Frankham, 2005). One ex situconservation tool used to mitigate the decrease of genetic diversity in endangered species are Captive Breeding Programs (CBPs). Captive populations of endangered species have the difficult goal of ensuring the survival of stable, self-sustaining populations for later reintroduction into the native habitat (Mallinson, 1995). A key challenge of CBPs is to maintain genetic diversity and avoid inbreeding depression with a small number of founders (Bouman, 1977; Ralls et al., 1979).
The majority of captive breeding schemes rely on studbooks that document pedigree information within the CBPs. While studbooks can be useful to minimize inbreeding effects (Pelletier et al., 2009; Witzenberger & Hochkirch, 2011), information from pedigrees can be flawed in some captive populations (Bowling et al., 2003; Marshall et al., 1999; Signer et al., 1994). Molecular genetic analyses can provide more insights into the relationships within captive populations and their genetic structure. Recently, genetic studies of endangered species have increased, using highly variable loci non-coding for fitness traits such as microsatellite markers (Witzenberger & Hochkirch, 2011). Microsatellites are known to be highly informative for small populations that have reached bottleneck events and are considered a tool to measure neutral genetic variation, and generally represent the extent and pattern of molecular variation within a population (Selkoe & Toonen, 2006). However, both empirical and simulated data indicate that patterns of variation and divergence in adaptive traits are not always associated with concomitant variation in neutral markers (Hedrick, 2001; Larson, 2012; Reed & Frankham, 2001), and some conservation biologists advocate for genetic diversity analysis for adaptive variation in CBPs (Hughes, 1991; Sommer, 2005). One targeted adaptive region is the Major Histocompatibility Complex (MHC) because its genes play a crucial role in the adaptive immune system. Historical events such as bottlenecks and founder effects, but also constraints of the mating system, such as limited sexual selection in CBPs (Schulte-Hostedde and Mastromonaco 2015), can be reflected in low numbers of MHC alleles (Schad et al., 2004; Hapke et al., 2004). However, in some free-ranging populations, genetic variation at the MHC might persist due to balancing selection, through heterozygote and/or rare allele fitness advantage, despite low levels of variability shown by neutral markers (Jarvi et al., 2004; Rico et al., 2016). These studies support the importance of balancing selection as a mechanism to maintain variation in small populations, and the difficulty of using neutral markers as surrogates for variation in fitness-related loci. Furthermore, MHC genes, particularly of class I, contribute to participate in pheromone recognition, playing a role in sexual selection as well (Penn, 2002).
MHC genes have a crucial role in adaptive immunity in jawed vertebrates, by encoding proteins that bind peptide antigens and present them at the cell surface to lymphocytes for their activation (Ujvari & Belov, 2011). T and B lymphocytes are known to interact directly with gut microbial communities to prevent invasion, pathogenesis, and detrimental immune response towards commensals (Ost & Round, 2018). T cells react to foreign molecules via cell co-receptors: glycoproteins CD8 and CD4 encoded by MHC class II and class I genes, respectively (Penn & Potts 1999). MHC-I molecules are present on all host nucleated cells and antigen presenting cells, and are known to act at the intracellular level, while MHC-II molecules are strictly found on antigen presenting cells and target extracellular microbes (Ost & Round, 2018). MHC presentation of antigens to T cells is thus the basis for initiating antigen specificity and, immune responses are considered specific to the organisms they target.
MHC genes are considered one of the most diverse loci in jawed vertebrates and good candidates for genetic diversity analysis in endangered species (Hughes, 1991). High genetic diversity in these loci could allow targeting more combinations of gut microbes, reflected in variable immunity or tolerance among individuals through rare allele and heterozygous advantage in balancing selection. Bolnick et al. (2014) examined the role of MHC-II motifs (amino acid sequences) in gut microbial community variation in sticklebacks (Gasterosteus aculeatus ) and found that common MHC motifs were linked to increases in microbial abundance and diversity, and rare motifs had the opposite impacts. Similarly, the microbiota was less phylogenetically diverse in individuals with high MHC-II diversity in the plumage of blue petrels (Halobaena caerulea , Leclaire et al., 2018), the gut of laboratory mouse strains (BALB/c, Khan et al., 2019), and the fur microbiota of fur seals (Arctocephalus gazella , Grosser et al., 2019). However, no study to date has investigated the MHC-gut microbiota relationships in endangered species under CBPs.
While the host environment has a strong impact on its gut microbial community, the genetics and biology of the host should also be taken into account to fully understand the complex dynamics that occur in this system (Koskella et al., 2017; Spor et al., 2011). This is especially true when considering the gut microbiota of endangered species, where a small number of founders in CBPs are likely to experience overall low genetic variation and diversity. Transition from natural to captive settings for breeding can become more challenging in these conditions when considering host-associated microbes (Trevelline et al., 2019; West et al., 2019). Within this context, Foster et al. (2017) proposed a theoretical framework known as the leash model, which posits that the host is under strong selection to evolve mechanisms to keep the microbiota under control, or “on a leash”. The presence of a genetically diverse microbiota leads to the dominance of the fastest growing microbes instead of the microbes that are most beneficial to the host (Foster et al., 2017). The targeting of microbial taxa to either promote or limit their proliferation could thus be beneficial to the host, through its adaptive immune response. We therefore hypothesize that less host control, expressed by more genetically diverse gut microbes, should happen in individuals with reduced genetic diversity in both neutral and adaptive markers. To test this hypothesis, we investigated the genetic diversity and gut microbial community assemblages in the critically endangered European mink (Mustela lutreola ).
The European mink (E-mink) is a semi-aquatic carnivore from theMustelidae family. Once widespread throughout Europe, it was evaluated as “critically endangered” in 2011 (Maran et al., 2016). There have been drastic declines in population and range, historically due to overexploitation and still today driven by habitat loss, degradation and fragmentation, road collisions, and the impacts of the alien American mink (Mustela vison ). E-mink populations are now restricted to enclaves in western France and northern Spain (referred as the western population), the delta of the Danube in Romania, and Ukraine and Russia (referred as the eastern population, Maran et al., 2016), the latter being the focal origin of a captive breeding effort in Estonia with successful reintroduced populations on Hiiumaa Island.
Two major studies have documented the genetic diversity of the free-ranging E-mink populations (Michaux et al., 2005; Cabria et al., 2015). The western population was characterized by a single mitochondrial DNA haplotype against 17 haplotypes in the overall eastern population. Western E-mink had also a much lower microsatellite genetic diversity and allelic richness compared to the eastern population. The authors concluded that the western free-ranging population reached a recent bottleneck, and potentially inbreeding depression due to geographic isolation. However, no proof of fitness reduction in this population through inbreeding has been reported as of yet (Carbonell et al., 2015). The antigen-binding site, encoded by exon 2 of the DRB MHC class II gene, was also investigated in the eastern captive population by Becker et al. (2009). They detected nine alleles within the 20 individuals investigated, estimating low to moderate variability when comparing to other endangered species in similar situations to the E-mink. However, no comparison is yet available for the captive western population.
Both populations are currently in a CBP. The majority of the eastern breeding stock is held at the Tallinn Zoo and extending to other zoos in Europe such as Zoodyssée in France and is only composed of captive-born individuals for over thirty of generations (Maran, pers. comm., 2021). This stock is managed under an EAZA Ex situ Program (EEP). On the other hand, the western breeding stock is held in facilities in Spain. Due to the recent implementation of a CBP in Spain, most of the breeding stock originates from the free-ranging western population captures in Spain within the last seven years (i.e. seven generations), and wild-born individuals from Spain are still being introduced as founders to this date. Those populations are considered as two distinct stocks and are bred separately, although few cross breeds are currently being conducted in Spanish facilities and at the Tallinn Zoo.
Both captive E-mink populations therefore offer a range of variation in neutral and adaptive genetic diversity. Due to extreme population variation over time and the emergence of small and isolated populations, the E-mink provides a unique framework to study the relationship between host genetics and gut microbial communities. Following the ecosystem on a leash model, the aims of this study were to (i) characterize the genetic diversity in the two captive E-mink populations with neutral and adaptive genetic markers as well as their gut microbial communities, (ii) examine the relationship between gut microbial diversity and genetic diversity, and (iii) investigate if gut microbial community structure and composition is linked to specific MHC motifs.