DPP10

DPP10 is an enzymatically inactive type II transmembrane glycoprotein of the S9B serine protease family that functions as an auxiliary subunit of voltage-gated potassium (Kv4) and sodium (Nav1.5) channels, shaping neuronal and cardiac excitability. DPP10 physically associates with Kv4.2 and Kv4.3 channels through its transmembrane and short cytoplasmic domains, accelerating channel inactivation and recovery from inactivation, shifting voltage dependence in the hyperpolarizing direction, and increasing surface current density; together with KChIP proteins it forms native ternary Kv4/KChIP/DPP10 complexes that reconstitute somatodendritic subthreshold A-type currents (ISA) in specific neuronal populations (PMID:15454437, PMID:16123112, PMID:25355692). N-linked glycosylation at specific extracellular asparagine residues is required for DPP10 dimerization, plasma membrane trafficking, and functional modulation of Kv4 channels, while the crystal structure reveals a β-propeller/α/β-hydrolase two-domain architecture with a glycine substitution at the catalytic serine position that accounts for the complete absence of dipeptidyl peptidase activity (PMID:22387313, PMID:25740212, PMID:16290253). DPP10 also interacts with cardiac Nav1.5 channels, modulating sodium current kinetics and voltage dependence in cardiomyocytes (PMID:30638748).

1. 2004 High

   Establishing DPP10 as a Kv4.2 auxiliary subunit resolved how a catalytically dead dipeptidyl peptidase-family protein could have a physiological role — it directly binds and modulates Kv4 channel gating, with the cytoplasmic N-terminus controlling inactivation acceleration.

   Evidence Co-immunoprecipitation and two-electrode voltage clamp with N-terminal truncation in Xenopus oocytes

   PMID:15454437

   Open questions at the time

   • Mechanism by which DPP10 promotes channel surface expression was unknown
   • Whether DPP10 participates in native neuronal channel complexes was unresolved
   • How DPP10 cooperates with KChIP co-subunits was not addressed
2. 2005 High

   Demonstration that DPP10 co-immunoprecipitates with Kv4.2 from native rat brain and forms ternary complexes with KChIP3 that uniquely reconstitute the rapid recovery kinetics of native ISA established DPP10 as an essential component of neuronal A-type channel macromolecular complexes.

   Evidence Co-IP from rat brain membranes, chimera domain mapping, ternary reconstitution electrophysiology in oocytes and CHO cells

   PMID:15671030 PMID:16123112

   Open questions at the time

   • Stoichiometry of the ternary complex was unknown
   • Structural basis for interaction was not resolved
   • Identity of specific neuronal populations expressing the complex was not mapped
3. 2005 High

   Mutagenesis of the catalytic triad residues confirmed that DPP10 is enzymatically inactive even when key catalytic residues are restored, establishing that its physiological function is non-enzymatic.

   Evidence Site-directed mutagenesis with enzymatic activity assay in 293T cells

   PMID:16290253

   Open questions at the time

   • Structural basis for lack of activity was not yet determined
   • Whether ancestral DPP10 orthologs possessed enzymatic activity was unknown
4. 2006 Medium

   Identification that the transmembrane plus 58-amino-acid cytoplasmic domain of DPP10 is the minimal unit sufficient for Kv4 gating modulation, and that multiple splice variants with alternative N-termini similarly modulate Kv4.3, refined understanding of which protein regions are functionally essential versus modulatory.

   Evidence Truncation mutant electrophysiology and splice variant RT-PCR/expression in Xenopus oocytes

   PMID:16738002 PMID:16899223

   Open questions at the time

   • Whether splice-variant-specific N-termini confer distinct kinetics in ternary complexes was untested
   • In vivo relevance of splice variant expression patterns was not demonstrated
5. 2007 Medium

   Discovery that the DPP10a splice variant produces uniquely fast, voltage-dependent inactivation that dominates in ternary complexes revealed that N-terminal splice variation does tune channel kinetics in an isoform-specific manner, with DPP10a enriched in cortex suggesting region-specific ISA tuning.

   Evidence Two-electrode voltage clamp in oocytes with isoform combinations, qRT-PCR and in situ hybridization

   PMID:17475505

   Open questions at the time

   • Mechanism by which the DPP10a N-terminus produces faster inactivation was not resolved
   • Functional impact in cortical neurons in vivo was not tested
6. 2010 Medium

   Pharmacological disruption of N-glycosylation abolished DPP10 surface expression and its functional effects on Kv4.3 in both heterologous cells and native human atrial myocytes, establishing glycosylation as a prerequisite for DPP10 trafficking and channel modulation.

   Evidence Tunicamycin treatment, flow cytometry, and patch clamp in CHO cells and human atrial myocytes

   PMID:20354865

   Open questions at the time

   • Which specific glycosylation sites were essential was not determined
   • Whether glycosylation affects DPP10 dimerization was unknown
7. 2012 High

   Site-directed mutagenesis of six N-glycosylation sites identified four asparagines required for surface trafficking and pinpointed N257 as additionally essential for DPP10 dimerization and interaction with the Kv4.3/KChIP2a complex, linking a specific post-translational modification to complex assembly.

   Evidence N→Q mutagenesis at six sites, flow cytometry, co-IP, and electrophysiology in CHO cells

   PMID:22387313

   Open questions at the time

   • Structural mechanism by which N257 glycosylation enables dimerization was not resolved
   • Whether glycosylation requirements differ across DPP10 splice variants was untested
8. 2014 Medium

   Immunohistochemical mapping in rat brain localized DPP10 to neuronal somata and confirmed in vivo co-localization with Kv4.2, Kv4.3, KChIP1, and KChIP3 in defined neuronal populations (parvalbumin/somatostatin interneurons, layer 5 pyramidal neurons, olfactory bulb mitral cells), validating the ternary complex model in native circuits.

   Evidence Immunohistochemistry with custom DPP10 antibody and co-localization analysis in rat brain sections

   PMID:25355692

   Open questions at the time

   • Functional consequences of DPP10 loss in these specific neuron types were not assessed
   • Co-localization does not prove physical interaction in every identified neuron type
9. 2015 High

   The crystal structure of human DPP10 revealed the β-propeller/α/β-hydrolase two-domain fold with glycine replacing the catalytic serine and a remodeled active-site entrance, providing the structural explanation for enzymatic inactivity; single-molecule imaging separately established that the Kv4.2/DPP10 complex preferentially adopts a 4:2 stoichiometry with DPP10 forming constitutive dimers.

   Evidence X-ray crystallography (molecular replacement from DPP6); single-molecule subunit counting and voltage clamp in Xenopus oocytes

   PMID:25740212 PMID:26209633

   Open questions at the time

   • No structure of the Kv4/DPP10 or Kv4/KChIP/DPP10 complex existed
   • How the transmembrane and cytoplasmic domains engage the channel pore domain was structurally unresolved
   • Whether 4:4 versus 4:2 stoichiometry exists in native neurons was not determined
10. 2016 Medium

    The Drosophila DPP10 ortholog retained both channel modulatory and dipeptidyl peptidase enzymatic activities, indicating that loss of catalytic function in mammalian DPP10 is a derived evolutionary event rather than an ancestral feature of the family.

    Evidence Co-IP, electrophysiology, and fluorometric enzyme assay for Drosophila ortholog with rat Kv4.3

    PMID:27198182

    Open questions at the time

    • Physiological significance of enzymatic activity in Drosophila was not explored
    • Evolutionary timeline of catalytic loss was not traced
11. 2019 Medium

    Discovery that DPP10 co-immunoprecipitates with Nav1.5 from human ventricle and modulates cardiac sodium current gating extended DPP10's role beyond Kv4 channels to cardiac sodium channels, with implications for cardiac conduction.

    Evidence Co-IP from human ventricular tissue, adenoviral DPP10 expression in rat cardiomyocytes, patch-clamp electrophysiology

    PMID:30638748

    Open questions at the time

    • Domain determinants of DPP10-Nav1.5 interaction were not mapped
    • Whether KChIPs participate in the Nav1.5/DPP10 complex is unknown
    • In vivo cardiac phenotype of DPP10 loss or gain has not been studied

• No high-resolution structure of the full Kv4/KChIP/DPP10 ternary complex exists, and the in vivo physiological consequences of DPP10 deletion in specific neuronal or cardiac cell types remain uncharacterized.
• No cryo-EM or crystal structure of the full ternary complex has been reported
• No DPP10 knockout or conditional knockout phenotyping in neurons or heart has been published in the timeline
• Mechanism of DPP10 specificity for Kv4 versus other Kv or Nav channels is structurally unresolved