Uncovering the adaptive function of group-living in a facultatively social rodent, the highland tuco-tuco (Ctenomys opimus)

An individual’s social environment can profoundly affect many aspects of their biology, including their behavior, reproductive success, physiology, and survival. Facultatively social species—those in which some individuals live in social groups while other individuals live solitarily—provide an important opportunity to explore the impact of variable social environments on conspecifics that experience similar ecological conditions. Uncovering the relative biological differences between individuals across alternative social environments is critical to the advancement of our understanding of the adaptive benefits and evolution of sociality.Tuco-tucos (Rodentia:Ctenomyidae) are subterranean rodents endemic to South America, ranging from southern Peru to southern Argentina. There are over 50 described species within the genus Ctenomys. Of the species whose behavior has been described, the majority are thought to be solitary meaning that a single adult occupies its own underground burrow, which is spatially distinct from other such burrows. However, recent field surveys are revealing greater complexity of social structure than previously realized, warranting targeted research on the social behaviors of these species. For my dissertation, I aimed to describe the behavior of a population of a previously undescribed tuco-tuco from the highlands in northwestern Argentina (Ctenomys opimus), commonly referred to as the highland tuco-tuco. I targeted this species due to anecdotal reports that suggested this population of highland tuco-tucos may be social. For my first chapter, I used visual observations and radiotelemetry to quantify the spatial movements and consequent social structure of adult highland tuco-tucos located at Laguna de los Pozuelos, Jujuy Province, Argentina (hereafter referred to as Pozuelos). This study revealed that this population consisted of both lone and group-living individuals, and that the number of individuals per group as well as the sex ratio within groups varied markedly. Further, I compared the spatial and social structure of individuals across ecological contexts (i.e., above- versus below-ground) as well as during the daytime and nighttime. I found that social relationships were robust regarding ecological context (above- versus below-ground), but that some groups identified during the daytime fissioned during the nighttime. Collectively, the findings from this chapter suggested that this population may be facultatively social. For my second chapter, I aimed to confirm the possibility that the population of highland tuco-tucos located at Pozuelos are indeed facultatively social using spatial data collected over five consecutive years. From these data, I sought to (1) confirm the regular occurrence of both lone and group-living individuals within the population and (2) characterize the temporal consistency of individual social relationships. I found that while the study population consistently contained a mixture of both lone and group-living animals, individual spatial and social relationships varied markedly across time. Specifically, the extent to which individuals remained resident in the same location across years varied, as did the number of conspecifics with which an animal lived, with an overall tendency for individuals to live in larger groups over successive years. Collectively, this chapter indicated that population-level patterns of behavior in this population of C. opimus are consistent with facultative sociality but that this variation does not arise due to persistent differences in individual behavior. For my third chapter, I sought to provide the first characterization of the glucocorticoid physiology in C. opimus and investigate how the observed variation in social behavior within the population of C. opimus at Pozuelos may impact individual glucocorticoid physiology. Earlier work in another known social tuco-tuco (Ctenomys sociabilis) found that yearling females that dispersed from their natal burrow to live alone had higher baseline glucocorticoid levels relative to females that remained in their natal burrow with conspecifics. Thus, for my third chapter, I aimed to determine if a similar pattern was also found in the population of highland tuco-tucos at Pozuelos. I collected fecal samples from all individuals captured on the field site during two consecutive years to assess the relationship between baseline glucocorticoid levels and multiple metrics of social behavior (i.e., group size, sex ratio of group, and metrics measured via social network analysis). Additionally, I conducted a biochemical validation study to confirm that fecal glucocorticoid metabolites provide robust measures of glucocorticoid levels in C. opimus. The results from the enzyme-linked immunosorbent assays revealed that corticosterone is the primary glucocorticoid metabolite produced by C. opimus. Despite marked variability in social relationships among the animals sampled, differences in social behavior did not appear to predict variation in fecal glucocorticoid metabolites. Rather, individual variability in fecal glucocorticoid metabolites was best explained by sex, with males having higher corticosterone levels than females. This pattern was also observed for individuals in the biochemical validation study. Collectively, this chapter underscores the importance of intrinsic factors (i.e., sex) in shaping glucocorticoid variation in wild populations of mammals.For my fourth chapter, I sought to provide the first characterization of the gut microbiome in C. opimus and investigate how the observed variation in social behavior within the population of C. opimus at Pozuelos may impact diversity of gut microbiome composition both within and between individuals. Studies in other mammalian taxa have shown a strong link between gut microbiome diversity and sociality, such that individuals connected with more conspecifics had great microbial diversity. Additionally, these studies have shown that individuals within groups tend to have more similar gut microbiome compositions that individuals between groups, further demonstrating the effect of sociality on the gut microbiome. Thus, for my final chapter, I aimed to determine if a similar pattern was also found in the highland tuco-tucos at Pozuelos. I collected fecal samples from all individuals captured on the field site during two consecutive years to assess the relationship between gut microbiome diversity and multiple metrics of social behavior (i.e., group size, sex ratio of group, and metrics measured via social network analysis). I found that gut microbiome alpha diversity (diversity within an individual) was best predicted by eigenvector centrality and clustering coefficient, relative to other social network metrics. Further, I found that while gut microbiome beta diversity (similarity between individuals) was not correlated with social network metrics, it was correlated with degree of home range overlap between individuals, highlighting the importance of contact between conspecifics outside of an individual’s immediate social group. Additionally, I found that beta diversity clustered by year, likely due to differing preservation methods between field seasons. Sex did not explain variation in gut microbiome alpha or beta diversity. Collectively, this chapter provides the first description of the gut microbiome in highland tuco-tucos and suggests that horizontal transmission plays an important role in maintaining gut microbiome diversity in C. opimus