KCNG4

KCNG4 (Kv6.4) is an electrically silent voltage-gated potassium channel subunit that obligately heteroassembles with Kv2 family α-subunits to form functional heterotetrameric channels with modified gating properties in neurons and other cell types. Assembly with Kv2.1 requires conserved CDD aspartates in the T1 tetramerization domain and Kv2.1-His105 (PMID:19717558, PMID:19074135), while atypical residues in the Kv6.4 S6 activation gate restrict stoichiometry to a predominant 3:1 (Kv2:Kv6.4) ratio (PMID:30322883); the incorporated Kv6.4 subunit confers distinct closed-state inactivation properties, shifts the voltage dependence of inactivation, and localizes to Kv2-dependent ER–plasma membrane junction clusters in motoneurons (PMID:26505474, PMID:40919874). Kv6.4 modulates neuronal excitability across multiple circuits—tuning fast motor neuron identity downstream of Dlk1 signaling (PMID:24626931), shaping action potential waveform and GABAergic synaptic output of cortical parvalbumin interneurons (PMID:41632839), and regulating nociceptor firing threshold in DRG and uterine sensory neurons (PMID:32697988, PMID:40423692)—and its loss in mice causes male sterility through defective spermiogenesis (PMID:27677211).

1. 2008 High

   Identifying the molecular determinants of Kv2.1–Kv6.4 heteromerization resolved how a silent subunit selectively engages its partner: Kv2.1-His105 in the T1 domain is required for T1–T1 interaction with Kv6.4, while being dispensable for Kv2.1 homomerization.

   Evidence Yeast two-hybrid, FRET, co-immunoprecipitation, and two-electrode voltage clamp with H105 point mutants in heterologous cells

   PMID:19074135

   Open questions at the time

   • Structural basis of the H105-mediated interface not resolved at atomic level
   • Whether H105 dependency generalizes to all KvS subunits not tested
2. 2009 High

   Conserved negatively charged aspartates (CDD motif) in the T1 A/B linker were shown to be essential for both Kv2.1 homotetrameric and Kv2.1/Kv6.4 heterotetrameric assembly, establishing that electrostatic interactions in the T1 domain are a shared assembly requirement.

   Evidence Site-directed charge-reversal mutagenesis with FRET, co-IP, and immunocytochemistry in HEK cells

   PMID:19717558

   Open questions at the time

   • Role of the CDD motif in heteromers with other KvS members not systematically tested
   • No structural data for the T1 heteromeric interface
3. 2014 High

   Linking Kv6.4 to a physiological circuit role, Dlk1 was shown to activate Kcng4 expression in fast motor neurons, establishing Kv6.4 as an effector that tunes delayed-rectifier currents and fast biophysical identity downstream of a Notch-inhibitory pathway.

   Evidence Transgenic mouse and chick in ovo electroporation with gain/loss-of-function, electrophysiology, and transcriptome analysis

   PMID:24626931

   Open questions at the time

   • Whether Kv6.4 is the sole mediator of the Dlk1-dependent fast signature is not definitively isolated
   • Transcriptional mechanism linking Dlk1 to Kcng4 promoter not defined
4. 2015 High

   The lower S6 segment of Kv6.4 was identified as the structural determinant of closed-state inactivation in Kv2.1/Kv6.4 heteromers, explaining the unique pharmacological sensitivity of these channels to 4-AP-mediated potentiation.

   Evidence Chimeric subunit substitutions between Kv6.4 and Kv9.3 with electrophysiology in Xenopus oocytes

   PMID:26505474

   Open questions at the time

   • No structural model of the Kv6.4 S6 gate in the heteromeric context
   • Physiological relevance of 4-AP potentiation in native neurons not addressed
5. 2015 High

   KCNE5 was shown to form a tripartite complex with Kv2.1/Kv6.4, differentially modulating the heteromeric channel's kinetics versus the homomeric channel's current density, revealing an additional layer of heteromeric channel regulation by accessory subunits.

   Evidence FRET and electrophysiology with systematic co-expression of KCNE5 with Kv2.1 ± Kv6.4 in HEK293 cells

   PMID:26242757

   Open questions at the time

   • Physiological context of the tripartite complex not established in native tissue
   • Other KCNE family members not tested with Kv2.1/Kv6.4
6. 2017 High

   Kcng4 knockout mice revealed a non-redundant in vivo role for Kv6.4 in spermiogenesis, as loss caused male sterility with severely abnormal sperm morphology and absent motility.

   Evidence Kcng4−/− knockout mouse with semen analysis and testicular histology

   PMID:27677211

   Open questions at the time

   • The specific potassium current disrupted in spermatogenic cells not identified
   • Whether the phenotype is Kv2-dependent or involves a non-canonical Kv6.4 function is unknown
7. 2018 High

   Atypical S6 gate residues in Kv6.4, rather than T1 domain incompatibility, were identified as the dominant constraint enforcing 3:1 (Kv2:Kv6.4) stoichiometry, resolving a long-standing question about why silent subunits occupy only one position per tetramer.

   Evidence Single-molecule imaging, chimeric T1/S6 swaps, electrophysiology, and evolutionary analysis across cnidarian orthologs

   PMID:30322883

   Open questions at the time

   • Structural basis of S6-gate incompatibility not resolved at atomic level
   • Whether all KvS subunits share this S6-based stoichiometry mechanism not demonstrated
8. 2020 High

   A rare human variant (KV6.4-Met419) was shown to act as a dominant-negative by failing to traffic to the plasma membrane while sequestering Kv2.1, linking Kv6.4 function to nociceptor excitability and uterine sensory neuron firing thresholds.

   Evidence Human SNP association, heterologous expression trafficking assays, electrophysiology in transfected DRG neurons, retrograde labeling of uterine afferents

   PMID:32697988

   Open questions at the time

   • The full pain phenotype of carriers not characterized beyond association data
   • Mechanism of ER retention of the Met419 variant not defined
9. 2021 Medium

   Kcng4-expressing GPe neurons were defined as a functionally distinct PV+ subclass with unique electrophysiological and behavioral output properties, establishing Kcng4 as a marker and functional modulator in basal ganglia circuits.

   Evidence Kcng4-Cre transgenic mice with optogenetics, electrophysiology, and machine learning behavioral analysis

   PMID:33731450

   Open questions at the time

   • Causal contribution of Kv6.4 channel activity versus Kcng4 as a mere genetic marker not disentangled
   • Downstream circuit effects on other basal ganglia nuclei not mapped
10. 2024 Medium

    A migraine-associated L360P mutation in the Kv6.4 S4–S5 linker was shown to drastically suppress Kv2.1/Kv6.4 channel function, providing a molecular mechanism for a channelopathy-based migraine pathway through trigeminal Kv2/Kv6.4 dysfunction.

    Evidence Electrophysiology with monomeric and fixed-stoichiometry tandem dimer constructs in heterologous cells

    PMID:39159549 PMID:39201645

    Open questions at the time

    • Variant effect not validated in native trigeminal neurons
    • Only heterologous overexpression systems used
    • Human genetic evidence linking L360P to migraine not detailed in these studies
11. 2024 Medium

    Zebrafish kcng4b mutations demonstrated that Kv6.4 modulates Kv2.1 activity during inner ear development, with loss-of-function causing ectopic otolith formation and gain-of-function blocking otolith development.

    Evidence Zebrafish LOF and GOF mutant alleles with developmental phenotyping and structural modeling

    PMID:38492873

    Open questions at the time

    • Direct electrophysiological measurement of Kv2/Kv6.4 currents in zebrafish otic vesicle not performed
    • Relevance to mammalian ear development untested
12. 2025 High

    In spinal motoneurons, Kv6.4 was shown to co-cluster with Kv2.1 and Kv2.2 at ER–PM junctions beneath C-bouton synapses, with clustering entirely dependent on Kv2 subunits and their ER-VAP binding motif, establishing the subcellular nanoarchitecture of native Kv2/Kv6.4 channels.

    Evidence Immunofluorescence in Kv2.1 KO, Kv2.2 KO, and Kv2.1-S590A knock-in mouse spinal cord sections

    PMID:40919874

    Open questions at the time

    • Functional consequence of disrupted ER-PM clustering on motoneuron firing not tested
    • Whether Kv6.4 itself contributes to ER-PM junction formation or is passively recruited is unresolved
13. 2025 High

    Pharmacological dissection using differential sensitivities to RY785 and GxTX established that Kv2/Kv6.4 heteromers constitute the predominant Kv2-containing conductance in mouse and human DRG nociceptors, providing tools to distinguish heteromeric from homomeric Kv2 currents in native neurons.

    Evidence Patch-clamp pharmacology in heterologous expression and native mouse/human DRG neurons

    PMID:40423692

    Open questions at the time

    • Whether Kv6.4 predominance extends to all DRG neuron subtypes not resolved
    • RY785/GxTX selectivity for other KvS heteromers not fully profiled
14. 2025 High

    Conditional Kv6.4 deletion in cortical PV interneurons revealed that Kv6.4 shapes action potential height, width, and threshold during high-frequency firing and modulates GABAergic synaptic transmission at PV-to-pyramidal synapses, establishing a cell-autonomous role for the silent subunit in cortical inhibitory circuit function.

    Evidence Conditional knockout with patch-clamp, optogenetic activation, and paired PV→pyramidal recordings in cortical slices

    PMID:41632839

    Open questions at the time

    • Network-level consequences of altered PV neuron output on cortical computation not assessed
    • Whether compensatory changes in other KvS subunits occur in KO not examined

• Key unresolved questions include the atomic structure of the Kv2/Kv6.4 heterotetramer, the identity of the Kv2-dependent potassium current disrupted in spermiogenesis, and whether Kv6.4 contributes structurally to ER–PM junction formation beyond being passively recruited by Kv2.
• No cryo-EM or X-ray structure of any Kv2/KvS heterotetramer
• Mechanism linking Kv6.4 loss to spermiogenesis failure undefined
• Kv6.4 role in ER-PM junction biogenesis versus passive localization not distinguished