Galvanin (TMEM154) is an electric-field sensor for directed cell migration
Nathan M. BelliveauiD, Matthew J. FooteriD, Amy Platenkamp, Celeste RodrigueziD, Heonsu Kim, Christopher K. PrinziD, Aaron Paul van LooniD, Yubin LiniD, Tara E. Eustis, Michelle M. ChaniD, Daniel J. CoheniD, Julie A. TheriotiD
Abstract
Directed cell migration of immune and epithelial cells is critical for their rapid response to tissue injury or infection. Endogenous electric fields generated by disruption of the transepithelial potential across the skin have been postulated to play an important role in guiding cells to wound sites, though how individual cells sense these tissue-scale physical cues remains largely unknown. We have identified Galvanin (TMEM154), a previously uncharacterized single-pass transmembrane protein, as being required for electric-field-guided migration of individual rapidly moving cells. Galvanin functions in both immune and epithelial cell types. Upon exposure of cells to an electric field, Galvanin rapidly relocalizes to the anodal side of a cell, and the net charge on its extracellular domain is necessary and sufficient to drive this spatial relocalization. Furthermore, expression of Galvanin is sufficient to confer electric field-guided migration on otherwise non-responsive epithelial cells. In human neutrophils, we show that Galvanin relocalization is immediately followed by changes in the spatial pattern of cellular protrusion and retraction. The strong directional response of these cells is lost upon truncation of Galvanins intracellular domain, suggesting that Galvanin acts as a direct sensor of the electric field, transducing spatial information about a cells electrical environment to the intracellular migratory apparatus. This sensor relocalization mechanism of cell steering defines a new paradigm for directed cell migration.
02 /Specification & Classification
Click any tag to browse matching papers- Citations
- 25
- FWCI
- 34.53
- Field percentile
- top 1%
- Organism
- Assay
- Research Focus
- Approach
In-depth summary
Claude Sonnet 4.6 · xhigh effort · 8 Nov 2026 · from the full text of version 2 (17 Aug 2025) · CC BY 4.0
A CRISPRi migration screen in neutrophil-like HL-60 cells identified Galvanin (TMEM154), a single-pass membrane protein whose negatively charged ectodomain electrophoreses anodally to steer cathodal electrotaxis.
Fast-moving single cells such as immune and skin cells follow wound-induced electric fields, but no dedicated field-sensing protein had been identified for them. This work supplies one, with necessity, sufficiency and a charge-based relocalization mechanism.
Key results
ShownA two-stage CRISPRi screen of electric-field-guided versus undirected migration found a subset of genes specific to the field condition. Galvanin was the transmembrane protein with the strongest field-specific knockdown phenotype.
Secondary screen of 1,070 genes (from a library targeting 18,901 genes): 473 genes affected field-directed and 544 affected undirected migration (adjusted p cutoff 0.05); r = 0.8 between conditions; 111 genes were field-specific. · CRISPR-interference screen of electrotaxis in…; Figure legends
ShownGalvanin knockout in HL-60 neutrophil-like cells strongly reduced cathodal directionality in collagen without changing cell speed, and re-expressing Galvanin-GFP fully restored it.
Knockout vs wild-type at 300 mV/mm: speed along the field p=0.006; six-minute compass autocorrelation p=0.003 (100 mV/mm) and p<0.0001 (300 mV/mm); Mann-Whitney U; 400-1,000 cells per condition. · Galvanin is important for the directional…; Figure legends
ShownGalvanin knockout also reduced cathodal bias in mouse EL4 T cells and in primary zebrafish keratocytes, though keratocytes retained some cathodal directionality.
EL4 T cells: p < 0.0001, 295-956 cells per line and condition. Zebrafish keratocytes: p < 0.0001, 126-373 trajectories from 4 sample preparations. Both by two-sided Mann-Whitney U test. · Galvanin contributes to electrotaxis in other…; Figure legends
ShownExpressing human Galvanin-GFP in MDCK cells, which respond poorly as single cells, produced a dose-dependent cathodal switch. Parental single cells showed an anodal bias that Galvanin reversed.
Two clonal MDCK lines at 300 mV/mm for three hours; 971-1760 cells per condition in 40 fields of view across two sample preparations; p < 0.0001 for 300 vs 0 mV/mm (Mann-Whitney U). · Expression of Galvanin is sufficient to enable…; Figure legends
ShownGalvanin relocalizes anodally within about a minute, in step with front-rear protrusion and retraction changes. Ectodomain charge is sufficient: a strongly negative engineered ectodomain rescued, a weak positive one did not, and intracellular truncation blocked rescue.
Zero-lag maximum negative cross-correlation (population average of 41 cells; individual-cell lags from 46 cells); net charge estimate averaged across field strengths ±1.1 SEM; D 0.53 μm/s (±0.1 SEM); -42e construct measured 37e (±2.5 SEM) and rescued; +9e and Galvanin(Δ108) did not rescue. · Relocalization of Galvanin on the plasma…; Relocalization of Galvanin depends specifically…; Figure legends
Methods, data and reproducibility
Pooled CRISPRi screening sorted HL-60 cells by electric-field-driven migration through a porous membrane. Hits were validated by knockout, rescue and gain-of-function in several cell types. Nuclear tracking in collagen, live imaging of Galvanin-GFP and a diffusion/electrophoresis model estimated charge.
- Pooled CRISPRi screens in differentiated HL-60 neutrophil-like cells, genome-wide then focused library, with and without electric field; CRISPRi sequencing data · human (HL-60 cell line) · n = 18,901 genes then 1,070 genes; 11 replicates (field), 10 (undirected) · PRJNA1147270
- HL-60 clonal Galvanin knockout, rescue and domain-engineered lines tracked in collagen gels under electric fields · human (HL-60 cell line) · n = two clonal knockouts; 400-1,000 cells per condition (wild-type comparison)
- EL4 T cell wild-type and Galvanin knockout migration in collagen at 300 mV/mm · mouse (EL4 cell line) · n = 295-956 cells per line and condition
- Primary keratocytes from wild-type and homozygous Galvanin knockout embryos at two days after fertilization · zebrafish (Danio rerio) · n = 126-373 trajectories per condition; 4 sample preparations
- MDCK parental and two clonal human Galvanin-GFP expressing lines, 2D tracking at 300 mV/mm · dog (MDCK cell line) · n = 971-1760 cells per condition; 40 fields of view
- Under-agarose live imaging of Galvanin-GFP and protrusion/retraction in HL-60 cells · human (HL-60 cell line) · n = 46 cells
- Latrunculin A-immobilized HL-60 cells used for vE/D and charge estimation · human (HL-60 cell line) · n = 13, 23 and 26 cells at 150, 300 and 500 mV/mm
- Methods & tools
- CRISPR interference (CRISPRi) pooled screening, CRISPR-Cas9 knockout, NetNGlyc 1.0, NetOGlyc 4.0, Adapt package (protrusion/retraction quantification), Latrunculin A, Under-agarose assay, Nuclear tracking in 3D collagen gels, Compass autocorrelation metric, Flow cytometry, Supercharged GFP ectodomain variants, matplotlib v. 3.5.2
- Data & accessions
- https://www.ncbi.nlm.nih.gov/bioproject/PRJNA1147270
- Reagents & environment
- Figure generation scripts use the Python package matplotlib (v. 3.5.2).
- Reproducibility gaps
- Raw cell-tracking images are available only on request. Code and figure data are in a GitHub repository (nbellive/CRISPRi_galvanotaxis_pub); no full URL is printed.
Limitations (4)
- The signaling pathway downstream of Galvanin relocalization is unknown, as is how it regulates actomyosin and whether pathways are shared across cell types and between MDCK and immune cells. · stated by the authors
- Galvanin is structurally poorly characterized, so the authors note it was unclear whether protein features beyond net ectodomain charge contribute to its sensory function. · stated by the authors
- Glycosylation as the source of the high ectodomain charge is inferred from sequence-based predictions and charge estimates; glycan occupancy and sialylation were not measured in the reported data. · noted in this summary
- Several conclusions rest on single clonal knockout lines (EL4) or two clonal MDCK overexpression lines, so clone-specific effects are not excluded; keratocytes kept residual cathodal bias, implying other mechanisms. · noted in this summary
Who should read:Read this if you study cell migration, wound healing or membrane biophysics and want a screen-derived electric-field sensor backed by charge-engineering rescue experiments.
Reproducibility ledger
Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.
Sign in to file a reproducibility report.
Related papers
- Protein diffusion controls how single cells respond to electric fields
Ifunanya Nwogbaga, Nathan M. Belliveau, … Brian A. Camley·bioRxiv·Apr 9, 2025
- Competing signaling pathways controls electrotaxis
S. Kulkarni, Francesc Tebar, … Pablo Sáez·iSciencePeer-reviewed·Apr 3, 2025
- Cortical waves mediate the cellular response to electric fields
- Connexin 43 mediated collective cell migration is independent of Golgi orientation
M. Sharma, S. Mukherjee, … J. DasSarma·bioRxiv·May 3, 2023
- SCHEEPDOG: programming electric cues to dynamically herd large-scale cell migration
T. J. Zajdel, G. Shim, … D. J. Cohen·bioRxiv·Dec 20, 2019
- Computational insights on the competition between electrotaxis and durotaxis
Pablo Sáez, Shardool Kulkarni, … Elías H. Barriga·bioRxiv·Feb 4, 2026