For more than a century, biologists knew that cells could sense and follow electric fields. The phenomenon was called galvanotaxis, named after Luigi Galvani, the 18th-century physicist who discovered bioelectricity. But the question of how cells actually detect these fields remained one of the most stubborn mysteries in cell biology. Now, a team led by Nathan Belliveau and Julie Theriot at the University of Washington and HHMI has identified the molecular sensor: a single-pass transmembrane protein called TMEM154, renamed Galvanin, that acts as a cellular antenna for electrical gradients.
The discovery, published in Cell on May 12, 2026, settles a question that has persisted since 1889, when German physiologist Max Verworn first observed bacteria swimming directionally in an electric field. Two years later, Belgian microscopist E. Dineur documented the same behavior in vertebrate cells, frog leukocytes migrating toward the cathode (the negative pole). But no one could find the molecule responsible.
A sensor that moves
Galvanin’s mechanism is unlike anything seen in chemotaxis, the far better-understood process by which cells follow chemical gradients. In chemotaxis, a signaling molecule binds to a receptor on the cell surface, triggering a cascade of intracellular signals. Galvanin does something fundamentally different: the electric field physically pushes the protein itself.
Galvanin is a small protein, just 161 amino acids long, with a negatively charged extracellular domain carrying an estimated net charge of roughly -18 electron equivalents. When a DC electric field is applied across the cell, that negative charge experiences a Coulombic force, the same force that makes electrons flow in a wire, driving the protein laterally within the plasma membrane toward the anode (the positive pole). This redistribution happens fast: live-cell imaging showed clear anodal accumulation within about one minute.
The anodal side of the cell, where Galvanin accumulates, becomes the rear. The cathodal side, which lacks Galvanin, becomes the front, the site of forward-directed protrusion. Critically, Galvanin’s intracellular tail is required for this directional response. When the team truncated the tail (a mutant called Δ108), the protein still redistributed normally in the membrane but the cells could no longer migrate directionally. The sensor relocalization itself is the directional cue; the intracellular domain transduces that spatial information to the actomyosin machinery.
“We think of this as a ‘sensor relocalization’ paradigm,” the authors write, a mechanism in which the physical movement of the receptor within the membrane is the signal, not a trigger for a separate signaling cascade.
From CRISPR to zebrafish
Belliveau and colleagues identified Galvanin through an exhaustive two-stage CRISPRi screen. First, they tested 18,901 genes, essentially the entire genome, in human neutrophil-like HL-60 cells placed in a custom-built electric field device. A secondary screen narrowed the candidates to 1,070 genes specifically affecting electrotaxis (but not undirected migration). Of those, 111 genes were uniquely required for electric-field-guided migration. TMEM154 was the top transmembrane hit with the strongest electric-field-specific phenotype.
The team validated the finding across four species: human neutrophil-like cells, mouse T cells, zebrafish keratocytes (skin cells), and dog-derived MDCK epithelial cells. In each case, knocking out Galvanin reduced or eliminated directed electrotaxis while leaving other forms of migration, including chemotaxis, intact.
“That allowed them to distinguish electric-field-specific genes from the general migration machinery,” note Michael Riedl (TU Dresden) and Michael Sixt (ISTA) in a commentary accompanying the paper. “Of 1,070 candidates, only 111 were electrotaxis-specific, a remarkably sharp filter.”
The strongest proof came from a gain-of-function experiment: MDCK epithelial cells, which normally show weak electrotaxis, acquired robust cathodal-directed migration when engineered to express Galvanin-GFP. The effect was dose-dependent, more Galvanin meant stronger bias.
The charge is the message
To confirm that the charge itself drives the mechanism, the team replaced Galvanin’s native ectodomain with synthetic alternatives. A supercharged -42e construct (green fluorescent protein with extra negative charges, linked by flexible XTEN spacers) restored directional migration. A weakly positive +9e construct did not. The results were unambiguous: net negative charge on the extracellular domain is both necessary and sufficient for Galvanin’s relocalization.
Biophysical measurements placed Galvanin’s net charge at -15 to -22e with a diffusion coefficient of approximately 0.53 square micrometers per second, consistent with electrophoretic drift driving the redistribution.
Why it matters
Endogenous electric fields exist throughout the body. Wounds generate fields of 50 to 500 mV/mm when the transepithelial potential is disrupted, comparable to the 300 mV/mm used in the lab experiments. Neutrophils, the first immune cells to arrive at a wound, express Galvanin, and the rapid one-minute relocalization time is compatible with the timescale of wound-induced fields. Keratinocytes, the skin cells that close the wound, do too.
The discovery has implications beyond wound healing. Electric fields guide collective cell migration during embryonic development, neural crest cells, limb bud formation, and organ morphogenesis all involve endogenous bioelectric signals. In cancer, tumor cells have been shown to follow electric fields during invasion, and the identification of a dedicated molecular sensor provides a potential drug target for blocking metastasis.
“This is the first molecular receptor proven to act as a direct electric-field sensor for single-cell migration,” Riedl and Sixt write, calling Galvanin a “cellular antenna” that fills a major gap in the understanding of bioelectricity.
The caveats
The exact signaling pathway downstream of Galvanin’s intracellular domain remains unknown. The paper suggests links to GIT1/2, alpha/beta-PIX, or phosphoinositide signaling (PI3K/PTEN), but the binding partners have not been identified. For wound healing and cancer applications, the relevance of Galvanin in complex in vivo environments, where electric fields coexist with chemical gradients, mechanical forces, and cell-cell signaling, remains to be demonstrated. And the zebrafish experiments showed reduced but not eliminated cathodal bias, suggesting some redundancy or alternative sensing mechanisms may exist in certain cell types.
What’s next
The Theriot and Belliveau labs are now probing how different cell types interpret electrical cues, including cells in tumor microenvironments. The ability to engineer synthetic sensors by modulating Galvanin’s charge, as demonstrated with the -42e construct, raises the possibility of designing cells with programmable electrotactic responses for applications in immunotherapy (steering immune cells to tumors) and regenerative medicine (guiding repair cells to wounds).
After 130 years, cell biology finally has its electrical sense.
Source:
Belliveau NM, Footer MJ, Platenkamp A, Rodriguez C, Kim H, Prinz CK, van Loon AP, Lin Y, Eustis TE, Chan MM, Cohen DJ, Theriot JA. “Galvanin (TMEM154) is an electric-field sensor for directed cell migration.” Cell, Vol. 189, Issue 13, pp. 4107–4121.e22. DOI: 10.1016/j.cell.2026.04.026