Electric organs in fish have evolved independently in six lineages and are an interesting example of convergent evolution. However, the genetic basis underlying the convergence of this trait is poorly understood. By sequencing and assembling the transcriptomes from the electric organ (EO) and skeletal muscle of three of those lineages of electric fish, Galant et al. showed the presence of shared patters of gene expression in pathways related to differentiation from muscle cell, increased cell size, reduced contractility and increased excitability.
Paper summary
Electric organs allow fish to communicate, navigate and cope with predators and preys. They have evolved rather recently: less than 100 million years ago in the Cenozoic (as shown in Figure 1A). Importantly they have evolved independently in at least six taxonomically diverse lineages, constituting a clear example of convergent evolution.
Electrocytes are thought to be developmentally derived from myogenic precursors and are morphologically very different among fish lineages. This is illustrated in Figure 1B, where the authors show micrographs of electrocytes in two lineages of electric fish: gymnotiformes, such as Electrophorus electricus and Sternopygus macrurus, present electrocytes devoid of sarcomere, the contractile unit of muscle cells. In contrast, in mormiroids like Paramormyrops kingsleyae a disorganized and non-functional sarcomere can be found in electrocytes.
In order to understand the genetic programme that led to the common function of electrocytes in such morphologically different cells among lineages, Gallant et al. assembled the genome of the gymnotiform E. electricus and use RNA-seq reads from eight tissues for gene prediction, giving rise to 29,363 gene models. Genes co-expressed between tissues in E. electricus were subjected to k-means clustering analysis to reveal groups of genes that are either up-regulated (211 genes) or down-regulated (186 genes) in the EO as compared to skeletal muscle (Figure S1).
They sequenced and assembled the transcriptomes from EO and skeletal muscle of two other gymnotiforms (S. macrurus and E. virescens) and two species with an independently-evolved electric organ: M. electricus and the mormyroid B. brachyistius. In this four species they looked for the orthologs of those transcripts found up/donw-regulated in the cluster analysis of E. electricus to detect shared patterns of gene expression. Finally they focused on genes that might explain the convergent features of electrocytes versus muscle cells by selecting pathways related to down-regulation of muscle differentiation, increased excitability and insulation, decreased contractility and larger cell size.
A summary of the results is presented in figure 2A. Consistent with their idea that electrocytes derived from muscle cells, they found that transcription factors typically down-regulated in mature muscle cells are highly expressed in electrocytes (e.g. six2a, hey1), together with the down-regulation of transcription factors specifically involved in muscle cell differentiation (e.g. six4b, myogenin), except in S. macrurus.
They also showed that the increased excitability of electrocytes compared to muscle cells could be explained by the enhanced expression of certain genes involved in ion pumps and transporters (e.g. atp1a2a, scn4aa), with the notable exception of atp1a3a in E.electricus.
Similarly, they found a general down-regulation of genes related to the assembly of sarcomeres (e.g. smyd1a. cacna1sa), again with milder results in the EO of S. macrurus., that would account for the lack of functional sarcomeres and thus reduced contractility of these cells as opposed to muscle cells.
Finally, they found a general enhancement of the insulin-like growth factor (IGF) signalling pathway (e.g. igf2b, net-37, further illustrated in figure 2B), which would contribute to the larger cell size of electrocytes. An overview of the combined contribution of the studied pathways to the characteristic phenotype of electrocytes is proposed in figure 2C.
Personal comment
In the present study, Gallant et al. use transcriptome sequencing to elucidate the genetic basis of the convergent function of independently-evolved and morphologically diverse EO. However, in spite of this transcription-wide approach, the authors focused on a very selected and relatively small number of genes and transcription factors, potentially ignoring other genetic contributions that could be provided with the rich and large dataset generated.
Based on the notion that electrocytes are derived from muscle cells, they first selected genes that are up/down-regulated in the EO compared to skeletal muscle solely in E. electricus, disregarding the implications of genes that do not show differential expression, or genes up/down-regulated in EO of the other species but not in E. electricus.
Secondly, they selected certain pathways that they considered a priori to be likely responsible for the distinct phenotype of electrocytes, and for each of those pathway they presented in the main text results for only 5 genes that strongly supports their hypothesis. However in the supplementary figures S2-4 they extended their findings with other genes within the selected pathways that show more variable and unconserved patterns among species and that are nor further discussed. One particular case, atp1a3a shown in figure 2A, do not follow the pattern of enhanced expression in EO of E. electricus as they claim in the text for ion transporters responsible for increased excitability, but the unexpected result is not justified.
The presence of vestiges of disarrayed and non-functional sarcomeres in the electrocytes of the mormiroid S. macrurus that nevertheless might be energetically expensive to keep could suggest a more recent evolution of electric organs in this species. Consistent with this idea would be the milder up/down-regulation reported for some of the studied genes in S. macrurus compared to the other electric fish.
However it is important to notice the lack of biological and/or technical replicates in the study, a caveat that weakens its conclusions and questions its presence in a high-impact journal like Science. Given that the used animals were not wild but commercially obtained, increasing the sample size to a minimum of 3 animals per lineage studied could be feasible and would eliminate potential undesired technical or biological variability and provide more robust and conclusive results.
Reference
Gallant, J., Traeger, L., Volkening, J., Moffett, H., Chen, P., Novina, C., Phillips, G., Anand, R., Wells, G., Pinch, M., Guth, R., Unguez, G., Albert, J., Zakon, H., Samanta, M., & Sussman, M. (2014). Genomic basis for the convergent evolution of electric organs Science, 344 (6191), 1522-1525 DOI: 10.1126/science.1254432