Kv4.2 is the pore-forming component of A-type potassium channels expressed prominently in dendrites, where the channels constrain excitability and influence the integration of incoming signals. These channels operate within larger complexes that include regulatory proteins such as the KChIP family and the membrane-associated subunits DPP6 and DPP10. DPP6 is particularly important in hippocampal CA1 pyramidal neurons: it supports channel surface expression, alters conductance and inactivation, and helps establish the increase in Kv4.2 abundance from the cell body toward distal dendrites. The new study used wild-type mice, mice lacking maternal UBE3A expression to model Angelman syndrome, and separate DPP6 knockout mice to distinguish the effects of the two proteins.

The first set of experiments asked where the UBE3A ubiquitin ligase encounters Kv4.2. Researchers tagged Kv4.2 with a binding site for alpha-bungarotoxin, allowing them to label channels that were present on the surface of living cells. In HEK293T cells expressing the tagged channel and UBE3A, surface-labeled Kv4.2 showed little colocalization with UBE3A when membrane trafficking was restricted at 4 degrees Celsius. When the labeled cells were returned to 37 degrees Celsius, permitting internalization and intracellular movement, Kv4.2–UBE3A colocalization increased markedly. A biochemical version of the experiment produced the same pattern: recovery of UBE3A with surface-tagged Kv4.2 was barely detectable under restricted trafficking but rose after incubation allowed endocytosis, with more UBE3A recovered at 50 minutes than at 20 minutes.

These observations place UBE3A downstream of channel internalization rather than showing that it constitutively binds channels at the plasma membrane. That fits earlier work in which neuronal activity promotes phosphorylation of Kv4.2 at threonine 607, recruitment of the prolyl isomerase Pin1, dissociation from DPP6 and subsequent UBE3A-dependent degradation. The current results refine the sequence: activity-associated remodeling is followed by endocytosis, after which UBE3A preferentially processes the internalized channel pool. The study did not directly image the accumulation of such a pool in Angelman syndrome neurons, so that part of the proposed mechanism remains a model that requires further testing.

In hippocampal tissue from UBE3A-deficient “AS” mice, the amount of DPP6 and its distribution among synaptosomal, postsynaptic-density and non-postsynaptic-density fractions were unchanged relative to wild-type littermates. Co-immunoprecipitation also found no detectable association between UBE3A and DPP6 in transfected cells, arguing against a simple model in which UBE3A regulates DPP6 by binding it directly. By contrast, the amount of DPP6 associated with Kv4.2 was significantly reduced in AS mice. The interaction between Kv4.2 and KChIP proteins was unaffected, indicating that loss of UBE3A did not broadly dismantle every interaction between Kv4.2 and its auxiliary subunits.

Seizure induction with pentylenetetrazole provided an activity-dependent test of the complex. In wild-type mice, pentylenetetrazole at 60 mg/kg reduced Kv4.2–DPP6 association after 30 minutes, as expected from the established model of activity-induced channel remodeling. The same treatment did not produce a significant additional reduction in AS mice, whose association was already reduced. Basal phosphorylation of Kv4.2 at threonines 602 and 607 was normal in AS mice, and seizure-induced phosphorylation at threonine 607 remained intact. The defect therefore appears to lie after the principal phosphorylation response rather than in the basal phosphorylation signal itself, although AS mice showed an additional increase in threonine-602 phosphorylation after seizure induction that was not seen in wild-type animals.

DPP6 knockout mice displayed a different molecular phenotype, helping show that the auxiliary subunit is not merely a passive partner. Basal phosphorylation at both threonine 602 and threonine 607 was elevated in DPP6-deficient tissue, while the phosphorylation increase normally triggered by pentylenetetrazole was absent. Phosphorylation of p38 MAPK, the kinase associated with the two threonine sites, remained normal at baseline and after treatment, suggesting that upstream kinase activation was intact. The authors interpret the findings as evidence of reciprocal regulation: activity-dependent phosphorylation promotes dissociation from DPP6, while DPP6 helps organize a channel complex capable of appropriate phosphorylation dynamics. This distinction is important because the AS phenotype involves altered complex remodeling with largely preserved basal phosphorylation, whereas removing DPP6 alters the phosphorylation pattern itself.

Immunostaining then connected the biochemical changes to dendritic structure. Kv4.2 protein normally forms a proximal-to-distal gradient along both apical and basal dendrites of CA1 pyramidal neurons. The researchers quantified staining about 10 micrometers from the stratum pyramidale and at more distant sites, including 130 micrometers in stratum radiatum and 70 micrometers in stratum oriens. That gradient was significantly impaired in DPP6 knockout mice on both apical and basal dendrites, consistent with an earlier electrophysiological finding that DPP6 is required for the corresponding increase in A-type potassium current toward distal dendrites. The imaging results therefore show that the electrical gradient has a matching protein-distribution component.

UBE3A-deficient mice also had impaired Kv4.2 gradients along both sets of CA1 dendrites. Kainic-acid-induced seizures altered the distribution in wild-type mice, whereas AS mice showed a loss of the gradient in stratum oriens but a significantly smaller seizure-associated change than wild-type littermates. This reduced responsiveness suggests that UBE3A-deficient neurons have less capacity to redistribute Kv4.2 in response to increased activity. Their basal distribution resembled the remodeled pattern seen after seizure induction in wild-type animals, raising the possibility that loss of UBE3A leaves neurons in a persistently shifted trafficking state. The findings do not establish that this mechanism directly causes seizures, cognitive impairment or other features of Angelman syndrome, and the experiments were conducted in animals and cultured cells rather than patients.

Kv4.2 normally limits dendritic excitability and helps shape the propagation and integration of excitatory signals. Its disrupted organization could therefore have effects beyond the biochemical status of the channel complex, especially given the hyperexcitability and seizure susceptibility associated with Angelman syndrome. The results also reconcile, at least partly, the previously reported increase in somatic A-type current in AS mice with the impaired dendritic gradient found here: greater overall Kv4.2 abundance could raise somatic currents, while weakened DPP6 association could disturb the channel’s normal spatial distribution. The study identifies post-endocytic Kv4.2 processing as a new mechanistic link between UBE3A deficiency and dendritic ion-channel dysfunction, but whether restoring complex assembly, trafficking or distribution would improve symptoms remains untested.

Subject of Research: Kv4.2–DPP6 channel-complex dynamics and dendritic organization in an Angelman syndrome mouse model

Article Title: Impaired Kv4.2-DPP6 complex dynamics in a mouse model of Angelman syndrome

Article References: Hu, J.-H., Liu, Y., Malloy, C., & Hoffman, D. A. (2026). Impaired Kv4.2-DPP6 complex dynamics in a mouse model of Angelman syndrome. Frontiers in Cellular Neuroscience, 20, Article 1953686. https://doi.org/10.3389/fncel.2026.1953686

Image Credits: AI Generated

DOI: 10.3389/fncel.2026.1953686

Keywords: Angelman syndrome, UBE3A, Kv4.2, DPP6, dendritic excitability, hippocampus, ion channels, membrane trafficking, mouse model, neuronal hyperexcitability, Impaired, complex