Multigenerational non-genetic adaptation to environmental change in C. elegans.

Abstract

Multigenerational non-genetic adaptation to environmental change in C. elegans. The discovery of Mendelian genetics and the identification of DNA as the carrier of heritable information largely resolved the question of how parental traits are transmitted to progeny, nearly closing the door on the leading 19th century theory of inheritance of acquired characteristics. Yet some traits, including paramutation and silencing associated with RNA interference (RNAi), are transmitted across generations without any change to DNA sequence, leaving open the question of whether what a parent experiences can influence the biology of its descendants. This gap gave rise to the field of non-genetic inheritance, which seeks to understand how environmentally acquired information is encoded and passed to offspring independently of DNA sequence. It also represents a meaningful challenge to the modern synthesis, which treated the germline as insulated from environmental influence. We now recognize that the germline is better understood as a dynamic interface between an organism and its surroundings, capable of registering experience and relaying it forward. This realization has revealed a layer of biological complexity that traditional genetics cannot fully account for. The specific mechanisms that encode and transmit this information across generations, however, remain poorly understood. The stakes are not purely theoretical: large-scale genomic studies in human populations consistently fall short of identifying DNA sequence variation that fully accounts for heritable disease risk, suggesting that non-genetic mechanisms contribute to human heredity as well. Reasoning that the ability to transmit acquired adaptations to offspring would confer a fitness advantage, we used the fast-growing, self-fertile nematode Caenorhabditis elegans to identify ecologically relevant environmental variables with large effects on reproductive fitness and to ask whether adaptation to those variables could be inherited. We found that moving animals from one monoculture bacterial diet to another produced large, reproducible reductions in self-progeny brood size. Populations maintained on the new diet recovered gradually, reaching normal brood sizes over five to nine generations through a non-genetic process governed by cumulative time on the new food. When animals were returned to their original diet, fitness dropped again before recovering, repeating the full cycle. This shows that non-genetic dietary adaptation is both rapid and reversible. To identify the cellular source of reduced fertility, we examined germ cell production in animals undergoing dietary transitions. The effects were diet-specific: animals transitioned to Pseudomonas berkeleyensis showed preferential disruption of sperm function, while those transitioned to Sphingobacterium multivorum showed preferential disruption of oocyte viability. Crosses between strains adapted to different diets revealed that heritable adaptation behaves as a dominant trait, and that the gamete type affected by a given diet is also the primary vehicle through which adaptation to that diet is transmitted. Sperm carry paternal dietary adaptation and oocytes carry maternal dietary adaptation. This pattern points to diet-specific molecular states being established and maintained independently within each gamete type. Sequencing of mRNA and small RNA populations showed that dietary transitions produce gene expression changes whose dynamics correlate tightly with antisense siRNA levels across the maladaptation and recovery cycle, supporting a role for endogenous RNAi pathways in adaptation and the inheritance of adapted states. To test this directly, we surveyed genes required for environmental RNAi and identified heritable adaptation defects in animals carrying mutations in sid-1, sid-2, and rde-1, three genes previously characterized in the context of RNA interference. Each mutant affected a distinct aspect of the adaptive response, and their phenotypes did not overlap in ways that would be expected if these genes were acting through their canonical RNAi functions. This argues against a straightforward environmental RNAi model and suggests instead that maladaptation reflects an active prior regulatory commitment rather than a simple failure to maintain homeostasis. Finally, preliminary multi-omic profiling of animals undergoing reciprocal dietary transitions reveals that these transitions drive asymmetric transcriptomic responses across generations, with the CSR-1a/22G-siRNA pathway emerging as a candidate mediator. These findings are preliminary and full interpretation will require deeper sequencing and analysis. Taken together, these results support a role for non-genetic mechanisms in heritable adaptation to environmental change. The inheritance of acquired characteristics, once dismissed, demands a second look, leaving the open question of whether parental experience can heritably influence the biology of its descendants