Divergence in larval jaw gene expression reflects differential trophic adaptation in haplochromine cichlids prior to foraging

Abstract

Background Understanding how variation in gene expression contributes to morphological diversity is a major goal in evolutionary biology. Cichlid fishes from the East African Great lakes exhibit striking diversity in trophic adaptations predicated on the functional modularity of their two sets of jaws (oral and pharyngeal). However, the transcriptional basis of this modularity is not so well understood, as no studies thus far have directly compared the expression of genes in the oral and pharyngeal jaws. Nor is it well understood how gene expression may have contributed to the parallel evolution of trophic morphologies across the replicate cichlid adaptive radiations in Lake Tanganyika, Malawi and Victoria. Results We set out to investigate the role of gene expression divergence in cichlid fishes from these three lakes adapted to herbivorous and carnivorous trophic niches. We focused on the development stage prior to the onset of exogenous feeding that is critical for understanding patterns of gene expression after oral and pharyngeal jaw skeletogenesis, anticipating environmental cues. This framework permitted us for the first time to test for signatures of gene expression underlying jaw modularity in convergent eco-morphologies across three independent adaptive radiations. We validated a set of reference genes, with stable expression between the two jaw types and across species, which can be important for future studies of gene expression in cichlid jaws. Next we found evidence of modular and non-modular gene expression between the two jaws, across different trophic niches and lakes. For instance, prdm1a , a skeletogenic gene with modular anterior-posterior expression, displayed higher pharyngeal jaw expression and modular expression pattern only in carnivorous species. Furthermore, we found the expression of genes in cichlids jaws from the youngest Lake Victoria to exhibit low modularity compared to the older lakes. Conclusion Overall, our results provide cross-species transcriptional comparisons of modularly-regulated skeletogenic genes in the two jaw types, implicating expression differences which might contribute to the formation of divergent trophic morphologies at the stage of larval independence prior to foraging. Electronic supplementary material The online version of this article (10.1186/s12862-019-1483-3) contains supplementary material, which is available to authorized users.

Gene coexpression networks reveal molecular interactions underlying cichlid jaw modularity

“…This might suggest the possibility of sub-functionalisation of prdm1 paralogues in a jaw-specific and modular manner during cichlid development. We previously found prdm1a to exhibit modular gene expression in jaws of these carnivorous Tropheini cichlids as well as other haplochromines cichlids using qPCR [ 55 ]…”

“…This could potentially be an indication for sub-functionalisation of the paralogues as well as potential changes in modular function of prdm1a along the anterior–posterior axis in viscerocranial skeleton of cichlids compared to zebrafish. Using qPCR, it was shown that cichlids from all major East African radiations express prdm1a modularly and differentially in oral and pharyngeal jaws [ 55 ]. It should be noted that prdm1a/b were enriched in all the three Blue sub-modules (Fig.…”

Transcriptomics unravels molecular players shaping dorsal lip hypertrophy in the vacuum cleaner cichlid, Gnathochromis permaxillaris

“…In order to select candidate reference genes, we used the transcriptome data and followed an approach that we have already used in our previous studies [ 11 , 29 , 127 , 128 ]. In brief, we first identified the genes showing no expression difference (FDR = 1) between the lip regions for each species and ranked them according to their level of expression to attain those with highest expression.…”

Transcriptional study of appetite regulating genes in the brain of zebrafish (Danio rerio) with impaired leptin signalling

“…Finally, geNorm calculates mean pairwise variations in a stepwise manner between each gene and the other candidates ( M value) 101 . The ranking outcomes, from each of these alone, cannot be reliable unless consistency of the results are observed between the software for the top ranked genes 102,103 . In this study, we used the Cq value of the most stable reference gene across both genotypes and different feeding conditions, Cq reference , to normalize Cq values of target genes for each sample (ΔCq target = Cq target − Cq reference ).…”