Emerging research indicates that the manner in which the microbiome assembles in the gut can affect the early-life development vertebrates, including their neurodevelopment [30, 31, 74–78]. We hypothesized that exposure to the environmental pollutant BaP could impact gut microbiome assembly to influence behavior development in zebrafish. To determine the validity of this hypothesis, we evaluated how embryonic exposure to BaP impacts the assembly of the zebrafish gut microbiome and whether the gut microbiome associates with the effect of BaP on embryonic and larval zebrafish photomotor function. Overall, the present study supports this hypothesis by finding that (1) BaP exposure perturbs gut microbiome assembly in a dose-dependent fashion, (2) the composition of the microbiome that assembles in the gut explains larval zebrafish behavior, and (3) the gut microbiome impacts how embryonic BaP exposure affects behavioral development in zebrafish. These observations indicate that environmental chemical exposure may influence vertebrate neurodevelopment by impacting gut microbiome assembly and hold important implications for our understanding of the underlying mechanisms of neurodevelopmental disorders.
Prior work observed that the gut microbiomes of mice are sensitive to BaP, as are specific gut isolates: BaP is toxic to some gut microbiota, whereas others are able to resist or even degrade the chemical [52]. However, to date, no study has evaluated how early-life BaP exposure impacts the assembly of the gut microbiome through early development in zebrafish. We found that embryonic BaP exposure perturbs microbiome assembly (in terms of alpha- and beta-diversity) in the zebrafish gut in a dose-dependent manner. In both cases, this dose-dependent effect was only observed using unweighted metrics (i.e., metrics that do not consider the abundance of taxa), indicating that the effect of BaP on the microbiome principally manifests among the rarer taxa in the community. Past studies have found that gut microbiota diversity metrics, composition, functional capacity, and metabolites can have a dose-dependent response to certain chemical toxicants, for example, tetracycline [79], nanoplastics [80], and BPA/BPA alternatives [81]. While the effects we measured on gut microbiome diversity are significant, the impact of BaP exposure concentration appears to be somewhat stochastic, evidenced in the way that we observed extensive variation among microbiome samples collected from fish exposed to the same concentration of BaP. This is consistent with research that characterized host response to BaP exposure; BaP is a precursor metabolite that is biotransformed into toxic compounds by the liver, so the host response itself is also highly variable [16, 39]. That said, we do find that the ASVs observed in the gut are able to predict exposure concentration much better than random chance, which supports the notion that BaP exposure results in ecological selection for specific functional groups of microbiota in the gut. Collectively, these observations indicate that assembly of the zebrafish gut is a relatively stochastic process, likely influenced by random sampling of the meta-community in the embryonic media and priority effects [76, 82, 83], and that BaP exposure increases selective pressure for particular community assemblages with high interindividual variability, as is common in other xenobiotic studies [23].
Given that the gut microbiome can impact brain functioning through the gut-brain axis [29–37], we sought to determine if BaP-induced variation in the gut microbiome associates with larval photomotor response (LPR). Prior work demonstrated that the gut microbiome contributes to vertebrate behavior development [30, 31, 74–78], and that specific gut isolates from the zebrafish gut can modulate larval photomotor response [34]. However, it is unclear whether exposure-related changes to gut microbiome assembly associate with alterations to behavioral development. Several lines of evidence in our study support the hypothesis that, at least in the case of embryonic BaP exposure, environmental chemicals induce perturbations to the gut microbiome that link to exposure-induced variation in behavior development. First, we found that both the alpha- and beta-diversity of the gut microbiome explains variation in LPR, in terms of light and dark cycle swimming activity. Additionally, a random forest analysis found that the relative abundance distributions of specific subsets of taxa in the gut predict LPR measures in zebrafish exposed to BaP. These taxa notably include members of the Shewanellaceae, members of which have been shown to metabolize BaP and influence behavior in mice [66–68]. Collectively, these results indicate that the larval zebrafish gut microbiome links to zebrafish behavior. We then sought to determine if the effect of BaP exposure on the microbiome explains exposure-associated variation in fish behavior. Our analyses find that the combined interaction between microbiome Shannon entropy and BaP exposure concentration explains variation in dark cycle LPR activity. Additionally, while we observed no interaction effects between BaP exposure concentration and behavior on microbiome beta-diversity, we resolved several taxa that manifest different associations with behavior measures as a function of BaP exposure concentration. These observations included members of the Lachnospiraceae, which have been linked to BaP degradation as well as mammalian behavior [71]. Taken together, these results indicate that the gut microbiome not only associates with larval behavior, but also does so in a dose-dependent manner and suggests that exposure-induced perturbations to the gut microbiome may be a mechanism through which BaP elicits its neuroactive effects.
Given these observations, we next sought to determine whether the gut microbiome dictates the larval photomotor response to BaP exposure through a germ-free zebrafish experiment. Prior work has demonstrated that germ-free fish are hyperactive relative to conventionally reared zebrafish [48, 56], a finding that is replicated in our study. We subsequently sought to determine whether the colonization state of the gut microbiome determines BaP-driven developmental neurotoxicity. Our results indicate that the gut microbiome plays a role in determining the effect of embryonic BaP exposure on behavioral development in zebrafish. Specifically, the presence of the gut microbiome modulated the impact of BaP on light cycle LPR. Additionally, microbiome state elicited an additive effect alongside BaP exposure concentration on dark cycle LPR phenotype. These observations hold implications for our understanding of how BaP elicits its neurotoxic effects and suggests that the presence of the microbiome, if not also its compositional state, is at least one major factor defining toxicity. However, it remains unclear how these effects manifest. Recent research suggests that vertebrate receptors for environmental chemical toxicants like BaP that have an effect on neurodevelopment are also associated with gut microbial composition [19, 26, 27] and can bind microbial metabolites [26], supporting the idea that gut microbiota and their hosts have evolved methods of responding to toxicant exposure in concert. The present study and others have also established a strong connection between xenobiotic exposure and dysbiosis [14, 16, 20, 22, 47, 81], and gut dysbiosis is often associated with atypical behavior via the gut-brain axis [33, 36, 37, 84–86], providing yet another link between host neurophysiology, toxicant exposure, and gut microbiota. The results presented here reveal new factors to consider in the effort to define the cellular mechanisms through which BaP impacts neurodevelopment, namely microbiome metabolism of BaP as well as BaP-induced gut dysbiosis.
Overall, observations derived from this study are consistent with the hypothesis that environmental neurotoxicants interact with and impact the gut microbiome in ways that influence developmental neurotoxicity. Future research can build upon these observations to improve the strength of this hypothesis in several ways. We identified taxa perturbed by BaP, and future work should seek to clarify the specific role these taxa may play in defining zebrafish neurodevelopment. Studies in mammal models as well as human populations should also consider whether BaP exposure links to perturbed gut microbiome assembly and whether these effects drive neurodevelopmental disorders. Additionally, while our investigation reveals the impact of BaP on microbiome assembly through early life, less is known about how embryonic exposure to BaP impacts the successional development of the microbiome throughout lifespan. Given that embryonic BaP exposure drives neurodegenerative phenotypes in adult zebrafish, which is a phenotype that is transgenerationally inherited [3, 56, 87], it would be valuable to determine if the entanglement our study has uncovered with respect to exposure, the gut microbiome, and behavior also manifests later in life and even across generations. Finally, it is critical that we ultimately define the molecular mechanisms the microbiome utilizes to resist the effects of BaP as well as those used to mediate BaP neurotoxicity in the effort to combat the toxic developmental effects of this ubiquitous environmental pollutant.
BaP is a ubiquitous environmental chemical toxicant that is impossible to escape in modern, industrialized countries [38, 39]. Past studies have focused on the effect of differential exposures to host physiology, in model organisms and humans, including but not limited to behavioral responses [38, 39, 42, 43, 46–49, 56, 75]. Furthermore, BaP is a precursor molecule to toxic compounds with myriad endpoints in vertebrate hosts [16, 39], so defining a mechanism for how BaP affects the host is challenging. This range in response could be in part attributed to the gut microbiome, which in recent years has been implicated as a possible mechanism in both chemical toxicant exposure responses and behavior (via the gut-brain axis) [4, 13, 14, 16–18, 22, 23, 29–34, 34–37, 75–77, 81, 84–86, 88], as well as a reason behind the high inter-individual variability in xenobiotic response [23]. We found evidence not only that BaP is associated with measurable shifts in microbial assembly in early life, but also that the gut microbiome impacts the developmental neurotoxicity of BaP in zebrafish. Our work clarifies growing concerns surrounding BaP-induced neurotoxicity and suggests that continued exploration of the gut microbiome’s role in this toxicity may reveal novel strategies of mitigating adverse outcomes of exposure.