Gene expression correlates and mechanistic insights into electric organ discharge duration changes in mormyrid electric fish

Abstract

Electric organ discharge (EOD) duration in African weakly electric fish (Mormyridae) is the most variable waveform component between species and the basis for distinguishing species-specific signals. EOD duration is thought to be influenced by morphological and physiological features of electrocytes (the cells that comprise the electric organ), but the mechanistic details are poorly understood. It has long been known that EOD duration is modulated by androgen hormones, affording an opportunity to identify gene expression correlates of EOD duration differences. We induced EOD elongation in the mormyrid Brienomyrus brachyistius by administering 17α-methyltestosterone (17αMT) to three treatment groups: control (no 17αMT exposure), T1day and T8day (samples taken 1 and 8 days after a single exposure to 17αMT, respectively). We then performed RNAseq, differential gene expression and functional enrichment analysis to detect gene expression changes during EOD duration change. Our analyses indicate 44 genes whose expression changed in tandem with EOD elongation and include genes responsible for actin filaments and microtubules, extracellular matrix organization and membrane lipid metabolism. Additionally, we found expression changes in one Na+ channel β-subunit, and five voltage-gated K+ channels. Together, these genes point toward specific cellular processes that contribute to morphological and physiological changes that contribute to EOD duration changes.

Keywords: Androgen hormones; Electrocyte; K+ channel; Na+ channel; Testosterone.

Fig. 1.

Experimental fish tank configuration. The housing arrangement was attached to the bottom of the tank, positioned in the middle. The fixed electrode locations (positive, negative and ground) are indicated.

Fig. 2.

Electric organ discharge duration in the experimental treatments and classification of gene expression changes. (A) Electric organ discharge (EOD) duration at the onset (Day 0) and at the end (Last day) of the experiment for each of the treatments: T1day [17α-methyltestosterone (17αMT) treated, sampled on day 1], T8day (17αMT treated, sampled on day 8) and control (not treated with 17αMT, sampled on day 8). The filled circles and vertical lines represent the mean±s.d. EOD duration per treatment; open triangles are individual fish measurements. A small horizontal jitter was added to better visualize overlapping observations. Asterisks represents significant differences (P<0.001) between the treatments connected by the horizontal bar. (B) Each graph shows the traces of the averaged EODs from one representative fish per treatment, taken on their last day of recording. The arrows indicate the three pairwise treatment comparisons, and the gene expression changes detected in each comparison are classified as ‘broad’ or ‘granular’, based on each treatment's time of exposure to 17αMT. Granular changes are further divided into ‘early’ and ‘late’.

Fig. 3.

Heatmap of mitch enrichment scores (S) for all significantly enriched gene sets in the three pairwise treatment contrasts studied. n=96 gene sets, higher dimensional enrichment score >0.1, false discovery rate (FDR) P<0.01. Gene sets were further classified into broad categories (curly braces, left).