This study demonstrates the remarkable power of an animal's environment to modulate the signaling network that promotes LTP in the hippocampus and to improve contextual fear memory formation across generations. In particular, we have dramatically expanded our previous finding that exposing young animals to an EE enhances the magnitude of LTP induction in the CA1 hippocampus (Li et al., 2006b
), by showing that it also enhances LTP in their future offspring through adolescence, even if the offspring are not exposed to EE. In both generations, this phenomenon involves the addition of cAMP and p38 MAP kinase dependence to LTP induction. We also demonstrated that juvenile enrichment can rescue defective LTP and improve contextual fear memory, normally associated with mutant mice lacking Ras-GRF2 protein, across generations.
Transgenerational epigenetic inheritance is a newly appreciated form of transfer of gene control from one generation to the next. The first example in mammals was detected in “viable yellow” (AVY
) and “Axin-fused” (Axin Fu
) mice where variably penetrant phenotypes are linked to DNA methylation of a retrotransposon driving expression of the AVY
and Axin Fu
genes (Morgan et al., 1999
; Rakyan et al., 2003
). This phenomenon can be affected by the environment since maternal diet during pregnancy influences the epigenetic status of F1 offspring through the alteration of methylation of the AVY
locus in utero
(Waterland and Jirtle, 2003
; Dolinoy et al., 2006
). Another, more clinically relevant, example is the epigenetic transgenerational effect of endocrine disruptors that alter the methylation state of early germ cells in the developing embryo in utero
and generate reproductive disease across multiple generations (Anway and Skinner, 2006
Multigenerational transmission of epigenetic changes has also been demonstrated to occur through somatic tissues. The best-understood example is the transmission of natural variations in maternal behavior, where the degree of licking/grooming behavior of mothers is transmitted to their offspring in the first week of life (Weaver et al., 2004
) (for review, see Champagne, 2008
). Pups who are exposed to strong nurturing mothers display life-long alterations in the patterns of promoter DNA methylation and changes in expression of specific genes in the hypothalamus that regulate the stress response. These offspring then become strong nurturing mothers and pass the phenotype on to the next generation by their behavior. Interestingly, this effect can be modified by the degree of enrichment to which animals are exposed after weaning (Champagne and Meaney, 2007
The phenomenon described here is clearly distinct from the studies described above. For example, epigenetic transmission from mothers to newborn pups through licking/grooming occurs after the birth of affected offspring. In contrast, in our study it occurs before birth because the effect is maintained even if offspring are raised by nonenriched foster mothers.
We also showed that the effect of juvenile enrichment on LTP lasts for >3 months, long enough for the effects to be present when the enriched female becomes fertile and pregnant. The idea that the effect of enrichment in the mother can be passed on to offspring during embryogenesis is consistent with a behavior study from >20 years ago, which showed that exposure of pregnant rats to an enriched environment enhances the maze learning abilities of their offspring, even if the offspring are raised by non-enriched foster mothers (Kiyono et al., 1985
). Our findings may also explain, at least in part, findings from another behavior study from >20 years ago, where offspring of female rats exposed to an enriched environment before pregnancy “inherit” their parent's enhanced exploratory behavior and learning skills (Dell and Rose, 1987
Another distinct feature of our findings is that juvenile enrichment affects LTP in the next generation but not in subsequent ones, possibly because the effects of this EE wears off faster in the offspring, such that it is not present when they become fertile. Finally, our work is the first to demonstrate the inheritance of a change in a signaling pathway that promotes LTP, and enhancement of contextual memory formation. It is also the first to demonstrate that defects in synaptic plasticity and memory formation caused by a genetic mutation can be reversed, at least in part, in an animal by the environment to which its mother was exposed during her youth.
How juvenile enrichment unlocks an otherwise latent cAMP/p38-dependent signaling cascade to enhance LTP in both enriched mice and in their future offspring remains to be revealed. For example, which component of EE, enhanced social interaction, exposure to novel stimuli or exercise is mostly responsible? Moreover, while it is clear that the EE effect must be indirect in the offspring of enriched mice, possibly through the endocrine system of their mother, is this also true for mice exposed directly to EE? Nevertheless, it is likely that in both generations a long-lasting change in the transcription of a rate-limiting gene(s) product that promotes this signaling cascade is involved.
It is highly likely that an enriched environment influences many signaling cascades in the brain and a variety of memory functions in young mice, not just the ones we detected here. Thus, EE likely influences many aspects of animal behavior that could be passed on to the next generation. We could clearly detect EE-induced enhancement of contextual fear memory across generations of ras-grf knock-out mice that correlated tightly with EE induced enhancement of LTP. However, we found that the 2 week EE paradigm used in these experiments had only a small, if any, effect on contextual fear memory in wild-type mice (data not shown). Thus, we could not confirm that EE effects on this particular behavior could be transmitted transgenerationally in normal mice. This could be a reflection of the fact that the enhancement of contextual fear memory by juvenile enrichment is most important when learning capabilities are suppressed, such as in ras-grf knock-out mice or possibly in learning-disabled humans. Another possibility is that contextual fear memory is not the behavior that is most sensitive to the novel cAMP/p38/LTP pathway that we have identified. Contextual fear memory can be demonstrated in all animal ages, but EE can only activate the cAMP/ p38/LTP pathway in young mice and in their offspring through adolescence. Future experiments are aimed at specifically blocking this newly identified NMDAR/p38/LTP signaling cascade in mice, without blocking other cascades that influence synaptic plasticity. In this way, we hope to reveal its unique contribution to adolescent behavior. The fact that a mechanism has evolved to insure that this signaling cascade is functional through early adolescence in the offspring of enriched mice, even if the offspring are not exposed to EE, strongly argues that its role in adolescent behavior is a very important one.
The enriched environment used as an experimental paradigm in these studies may actually be more natural than a conventional laboratory environment that may border on sensory deprivation. Thus, the transgenerational inheritance of this new LTP-inducing signaling pathway may be a mechanism that has evolved to protect one's offspring from deleterious effects of sensory deprivation, which may be particularly potent in the young and exacerbated by the presence of mutations in signaling molecules like GRF proteins that contribute to synaptic plasticity. Overall, this study highlights the power of environmental stimulation during youth to influence the composition of signaling networks that influence synaptic plasticity and memory formation, not only in the enriched animal, but also in their future offspring during their youth.