3.3 The pelagic environment drives species-specific differences
in gene expression and reveals signatures of genetic assimilation.
We have shown previously that foraging conditions can have a marked
impact on the genotype-phenotype map. Specifically, quantitative trait
loci (QTL) for the same trait map to largely distinct regions of the
genome when animals are reared under alternate benthic/pelagic foraging
conditions (Parson et al., 2016; Zogbaum et al., 2021). We therefore
examined expression differences between species within each environment,
and documented a marked imbalance in DEGs. Specifically, when only
considering animals exposed to pelagic conditions, we found over 3500
DEGs between species, whereas fewer than 1000 DEGs were detected between
species when only comparing animals reared under benthic conditions
(Table 1, Figure 4A-C). Additionally, when comparing
environment-specific DEGs to the total dataset (i.e., combining both
environments), we found that over 2500 genes from pelagic animals were
represented in the global comparison, whereas only 155 genes from
benthic animals overlapped between datasets (Table 1, Figure 4C). These
data underscore the importance of environmental context in determining
the genetic basis of species-specific bone shapes (Parsons et al 2016;
Zogbaum et al, 2021). More specifically, they suggest that the pelagic
foraging environment is driving species differences in gene expression
within the IOP-RA functional complex.
This trend is drawn out when comparing genes from an additive model
(S+E), whereby DEGs were detected at the level of both species and
foraging environment (Table S2; Figure 5A). When illustrated in a
heatmap, these data support the assertion that species differences in
gene expression are driven by the pelagic environment, and reveal
patterns consistent with either genetic accommodation or assimilation.
Genetic assimilation is a mechanism by which plasticity is lost over
evolutionary time as genetic variation that facilitated plasticity in an
ancestral population becomes fixed as descendent populations adapt to a
specific environment (reviewed by Pigliucci et al., 2006). If we assume
that plasticity is ancestral, evidence for genetic assimilation is
apparent in several gene clusters (denoted by pink dots, Figure 5A),
whereby TRC expression levels are indistinguishable between foraging
environments and match those of benthic MZ. Consistent with previous
data many of the DEGs identified by this model contribute to cell cycle
regulation – e.g., cdc20 , cdca5 , cdca8 ,ccne2 , ccnb1 , ccnb2 , ccnf . A list of all the
DEGs in this model can be found in Table S2.
Alternative to genetic assimilation is genetic accommodation, or an
increase in genetic plasticity over evolutionary time. We cannot rule
out that this is the case, as it is possible that plasticity has been
enhanced beyond the ancestral condition in MZ. Regardless, the main
conclusion to be drawn from these data is that the evolution of
plasticity in this system may be traced to divergent patterns of gene
expression associated with cell cycle regulation.
We next performed GO analyses for DEGs between species in each foraging
environment. When considering animals reared in the pelagic foraging
environment, GO analysis revealed a diversity of biological processes;
however, those associated with cell division were among the most
enriched in MZ, whereas translation and cell differentiation were among
the most enriched processes in TRC (Figure 4D). For animals reared in
the benthic environment, comparatively fewer biological processes were
enriched in general, consistent with fewer DEGs being identified.
Similar to pelagic fishes, this analysis found enrichment of cell cycle
genes in MZ, and cell differentiation in TRC (Figure 4E). While a
greater number of DEGs, contributing to a larger number of biological
processes, underlie species-specific differences in the pelagic
environment, there are notable consistencies between environments.
Specifically, an increase in cell number seems to be important for
skeletal growth in MZ, whereas cell differentiation may shape growth in
TRC at the time points when tissues were collected.
Unsurprisingly, enriched GO terms for the additive model are similar to
those for MZ in the pelagic environment, and include cell cycle, cell
division, and chromosome segregation (Figure 5B). In addition, this
analysis found enrichment of cytoskeleton organization, which is
critical to many cellular functions relevant to bone formation and
plasticity, including mechanotransduction (Gunst & Zhang, 2008), and
primary cilia formation (Mirvis et al., 2018), which we and others have
found to be necessary for load-induced bone formation (Chen et al.,
2016; Moore et al., 2019; Gilbert & Tetrault et al., 2021).