Introduction: Single-Cell Transcriptomics in Weakly Electric Fish
Last updated on 2026-05-08 | Edit this page
Estimated time: 40 minutes
Overview
Questions
- What genomic resources have been developed for weakly electric fish, and why?
- How does single-cell RNA sequencing differ from bulk RNA-seq, and why does that matter for studying the electric organ system?
- What is the integrated story linking hormones, the electric organ, sensory receptor tuning, and central corollary discharge in mormyrid fish?
- How can changes at the cellular and molecular level reshape behavior across seasonal, developmental, and evolutionary timescales?
Objectives
- Describe the genomic resources developed for weakly electric fish at the MSU Electric Fish Lab.
- Explain why single-cell transcriptomics is a transformative technique for biology and what kinds of questions it enables.
- Summarize how testosterone coordinately reshapes the electric organ, knollenorgan electroreceptor tuning, and central corollary discharge timing.
- Identify the molecular and cellular changes implicated in testosterone-induced EOD elongation in mormyrids.
- Connect the single-cell approach to open questions that motivate this module.
Introduction and Overview
The MSU Electric Fish Lab (https://efish.integrativebiology.msu.edu) has been leading efforts to develop genomic resources for weakly electric fish for the last 13 years. We started with the electric eel genome (Gallant et al. 2014, Science), and from there we’ve developed genomes for many of the major lineages of weakly electric fish — both South American (Gymnotiformes) and African (Mormyroidea). We’ve also developed techniques for genome manipulation using CRISPR/Cas9 and Tol2 transposases, though that’s out of the scope of this module!
Ask Vicky and Jason
If you’re curious about CRISPR/Cas9 and Tol2 transposase work in electric fish, ask Vicky Salazar and Jason about their efforts to develop these techniques.
A major innovation since the original genome and transcriptome sequencing has been the ability to sequence transcriptomes from individual cells. This technique was named Science magazine’s Breakthrough of the Year in 2018 (Pennisi 2018, Science 362:1344), and it has already led to incredible insights across biology:
- Neuroscience — comprehensive cell-type atlases of the mammalian brain (e.g., Tasic et al. 2018, Nature; the BRAIN Initiative Cell Census Network’s 2021 mouse primary motor cortex atlas, Nature) have replaced classical morphological taxonomies with molecular ones.
- Developmental and evolutionary biology — whole-embryo single-cell atlases of zebrafish (Wagner et al. 2018, Science; Briggs et al. 2018, Science) and Xenopus have reconstructed cellular lineages from zygote to organogenesis, while comparative cell-type atlases (reviewed in Tanay & Sebé-Pedrós 2021, Trends in Genetics) are letting us trace cell-type homology across millions of years of evolution.
- Genomics — single-cell approaches have revealed regulatory heterogeneity invisible to bulk RNA-seq, including rare cell states that drive disease progression and tissue regeneration.
We’ll be exploring this technique in this course module end-to-end — including sample preparation, bioinformatics, and data analysis.
The Story So Far
One of the great things about electric fish is that they “wear their behavior on their sleeves” — the behavior of interest, of course, is the electric organ discharge (EOD), which is used for both communication and electrolocation. The temporal properties of the EOD waveform, as well as the rate at which the electric organ fires, have considerable bearing on the day-to-day lives of these fish, as you have doubtlessly learned. What we’ll be working on for this module is a rather interesting example of integration across the nervous system, coordinated by hormones.
The peripheral story: hormones, the electric organ, and the EOD
The story begins with Bass & Hopkins (1983, Science 220:971–974), who demonstrated that mormyrids show seasonal sexual dimorphism in EOD waveform: males elongate their EODs during the breeding season — a change associated with the drop in conductivity that comes with the rainy season — while females do not. Bass and Hopkins showed that this change can be induced in the laboratory by exposure to 17α-methyltestosterone (17αMT), and that hormone treatment elicits a male-like response in females and juveniles as well.
A series of follow-up studies established what testosterone is actually doing to the electric organ:
- Structural changes (Bass, Denizot, & Marchaterre 1986; Freedman et al. 1989). The wafer-shaped electrocytes that make up the electric organ undergo dramatic remodeling under T treatment. Surface area of both the anterior and posterior faces of the electrocyte increases, but the increase is more pronounced on the anterior face, driven by the elaboration of deep membrane invaginations. Overall electrocyte width also increases.
- Physiological changes (Bass & Volman 1987, PNAS). Intracellular recordings from individual electrocytes showed that T elongates the duration of the action potentials generated by both faces of the electrocyte by 2–3×, with little effect on AP amplitude. The current model is that increased anterior-face surface area raises membrane capacitance, which delays spike initiation in the anterior face relative to the posterior face — preventing the two APs from canceling and producing a longer overall EOD.
- Sensory tuning (Bass & Hopkins 1984). Knollenorgan electroreceptors — the receptors specifically dedicated to detecting incoming EODs — also retune under hormone treatment. And wouldn’t you know it: their tuning shifts in exactly the same direction, becoming more sensitive to the lower peak frequencies that characterize long-duration EODs. So the periphery isn’t just changing the signal — it’s also retuning the receivers.
Stop and predict
Before reading the next section: if testosterone elongates a male’s EOD, and elongates the spike that knollenorgan receptors fire in response to that EOD, what would have to happen centrally in the brain to keep the fish from confusing its own (now-longer) EOD with an EOD from another fish?
The molecular layer: what we recently published
Last year, Mauricio Losilla and I published a paper (Losilla & Gallant 2025, J. Exp. Biol. 228:jeb249548) that took the next mechanistic step: what genes change expression as the EOD elongates? We used a clean experimental design — Brienomyrus brachyistius exposed to 17αMT for 1 day or 8 days versus ethanol-vehicle controls — and performed RNA-seq plus differential expression and functional enrichment analyses on caudal peduncle tissue.
By day 8, EOD duration had nearly doubled, and we identified 44 genes of highest interest whose expression tracked the elongation. They cluster into four functional themes:
- Cytoskeletal & sarcomeric — strong upregulation of actin and actin-related genes (actn1, actc1c, myh6, acta1b, xirp1, myo1d) and microtubule-related genes (eml6, tubb2, cavin4b, mid1ip1l). These are consistent with the cytoskeletal scaffolding required to support the expanded, more elaborately invaginated membrane.
- Extracellular matrix — upregulation of a collagen gene (si:dkey-61l1.4), thrombospondin thbs4b, and cadherin cdh11; downregulation of epdl2, an ependymin-related gene that we previously implicated in EOD waveform variation across Paramormyrops.
- Lipid metabolism — strong upregulation of elovl7a and selenoi (membrane-lipid biosynthesis) and fam126a (a gene with a known role in cell types with expanded plasma membrane). Increased membrane area presumably requires more phospholipid.
- Cation homeostasis — a coordinated shift in voltage-gated ion channel expression. We found one Na⁺ channel β-subunit gene (scn4b, downregulated) and five voltage-gated K⁺ channels. Four are delayed-rectifiers — kcna1 (Kv1.1), kcna7a1 (Kv1.7), kcnc1 (Kv3.1), and kcnc2 (Kv3.2) — all downregulated in long-EOD fish, which would slow K⁺ efflux and prolong APs. The fifth, kcng4a (Kv6.4), is a “modifier/silencer” subunit and is upregulated; its known effect on Kv2.1 is to shift inactivation to more negative potentials, again favoring AP elongation. Notably, the most highly expressed K⁺ channel in mormyrid electrocytes — kcna7a2 — did not change significantly, suggesting EOD elongation is modulated through coordinated regulation of several less-abundantly-expressed channels rather than the dominant one.
So the molecular picture aligns beautifully with the morphological and physiological work: the same cells are simultaneously remodeling their cytoskeleton and ECM contacts, expanding their membrane via lipid biosynthesis, and retuning their ionic currents.
Why this matters for single-cell
The Losilla & Gallant (2025) study used bulk RNA-seq on whole caudal peduncle tissue — a mixture of electrocytes, connective tissue, nerves, and skin. The signals we recovered are real, but they’re averaged across cell types. Single-cell transcriptomics promises to tell us which cells are doing what — and that’s a key motivation for this module.
The central layer: corollary discharge
Another aspect of this system that we haven’t yet discussed is the corollary discharge — something mormyrids are a famous model system for. When a mormyrid produces an EOD, the command originates in the command nucleus (CN) in the hindbrain. CN drives the medullary relay nucleus (MRN), which in turn drives spinal electromotor neurons that activate the electric organ.
Critically, CN also sends an efference copy of that command down a parallel corollary discharge (CD) pathway: CN → bulbar command-associated nucleus (BCA) → mesencephalic command-associated nucleus (MCA) → sublemniscal nucleus (slem). The slem is GABAergic and provides precisely-timed inhibition onto the adendritic neurons of the nucleus of the electrosensory lateral line lobe (nELL) — the same neurons that receive excitatory input from knollenorgan primary afferents. Because the inhibition arrives just as the knollenorgan reafferent spikes do, the fish is effectively “deaf” to its own EODs, and the downstream knollenorgan pathway is dedicated to processing EODs from other fish (Bell & Grant 1989; reviewed in Fukutomi & Carlson 2020).
Older literature (and Losilla & Gallant 2025) uses “pacemaker nucleus” for the small cells ventral to the relay nucleus that initiate each EOD command. Modern Carlson-lab papers (Fukutomi & Carlson 2023, Jarzyna & Carlson 2026) use “command nucleus (CN)” for the same population. They refer to the same structure — flag this for students if they see both terms across the reading list.
What’s more, Bruce Carlson’s lab has shown that this central pathway is also hormonally sensitive — and adjusts in a way that beautifully complements the peripheral changes. Three findings are especially relevant:
- Fukutomi & Carlson (2023, Current Biology) showed that T treatment in B. brachyistius not only elongates the EOD (peripherally) but also delays and elongates the time window of CD inhibition in the brain. Critically, the timing of this shifted inhibition matches the shifted timing of knollenorgan reafferent input almost exactly — so the system continues to cleanly cancel responses to the fish’s own EOD even as the EOD changes shape.
- The shift is not driven by sensory feedback. Fukutomi & Carlson surgically silenced fish (spinal cord transection) so that T elongated electrocyte properties without producing any actual EODs, and then measured CD timing. The CD shift still occurred. So testosterone is acting directly on the central CD circuit, not adjusting it via altered reafference.
- Jarzyna & Carlson (2026, preprint) then localized the central effect: simultaneous field-potential recordings throughout the CD pathway show that the shift in onset and duration of CD activity originates at the mesencephalic command-associated nucleus (MCA), propagating through slem to nELL. Even more elegantly, the same MCA locus accounts for evolutionary differences between short-EOD (C. compressirostris) and long-EOD (C. numenius) species, and for developmental plasticity in extreme long-EOD adult C. numenius. So one neural substrate handles seasonal hormonal shifts (days), developmental shifts (years), and evolutionary divergence (millions of years).
That last point is what makes this such a satisfying integration story: from a single hormonal cue, the periphery (electric organ) and the central nervous system (corollary discharge circuitry) are co-regulated to keep the entire sensorimotor loop coherent. And the central locus appears to be reused across timescales of plasticity. This is the system we’ll be probing at single-cell resolution in this module.
Putting it together
Sketch the loop linking testosterone, the electric organ, knollenorgan electroreceptors, and the corollary discharge pathway. For each component, note one thing that changes under T treatment and one outstanding question that single-cell transcriptomics could address.
Possible answers (yours may differ — the goal is to make the integration explicit):
- Electric organ / electrocytes — change: anterior-face surface area expands, K⁺/Na⁺ channel expression shifts, cytoskeleton remodels. Open question: are these changes uniform across all electrocytes, or are there electrocyte sub-populations responding differently?
- Knollenorgan electroreceptors — change: tuning shifts to lower peak frequency. Open question: is this driven by a single cell-type-specific transcriptional program, and is the same program triggered by T as in the electric organ?
- CD pathway (especially MCA) — change: onset and duration of CD activity is delayed and elongated to match shifted reafferent input. Open question: which cell types in MCA express androgen / estrogen receptors, and is the same molecular toolkit used here as in the periphery?
- Loop integration — change: the entire sensorimotor loop stays internally consistent across seasons, development, and species. Open question: what shared molecular signatures define cell types that are “T-responsive” across this whole circuit?
Key references
Pennisi, E. (2018). Development cell by cell. Science 362:1344 [Breakthrough of the Year].
- The MSU Electric Fish Lab has built genomic resources spanning the major weakly electric fish lineages, beginning with the electric eel genome in 2014.
- Single-cell RNA-seq (named Science’s Breakthrough of the Year in 2018) reveals cell-type-specific biology that bulk RNA-seq cannot resolve.
- In mormyrids, testosterone elongates the EOD by remodeling electrocyte morphology and physiology, retunes knollenorgan electroreceptors, and shifts central corollary discharge timing — all coordinated to keep the sensorimotor loop coherent.
- The Losilla & Gallant (2025) paper identified 44 genes of interest, including cytoskeletal, ECM, lipid-metabolism, and ion channel genes — but bulk RNA-seq cannot tell us which cell types drive these signals.
- The same neural locus (MCA) appears to handle seasonal, developmental, and evolutionary shifts in CD timing — a striking example of substrate reuse across timescales.
- Single-cell transcriptomics is the natural next step to ask which cells in this integrated system carry which molecular signatures of plasticity.