Modulation of Kv4.2 channel expression and gating by dipeptidyl peptidase 10 (DPP10)

Abstract

The dipeptidyl aminopeptidase-like protein DPPX (DPP6) associates with Kv4 potassium channels, increasing surface trafficking and reconstituting native neuronal ISA-like properties. Dipeptidyl peptidase 10 (DPP10) shares with DPP6 a high amino acid identity, lack of enzymatic activity, and expression predominantly in the brain. We used a two-electrode voltage-clamp and oocyte expression system to determine if DPP10 also interacts with Kv4 channels and modulates their expression and function. Kv4.2 coimmunoprecipitated with HA/DPP10 from extracts of oocytes heterologously expressing both proteins. Coexpression with DPP10 and HA/DPP10 enhanced Kv4.2 current by approximately fivefold without increasing protein level. DPP10 also remodeled Kv4.2 kinetic and steady-state properties by accelerating time courses of inactivation and recovery (taurec: WT = 200 ms, +DPP10 = 78 ms). Furthermore, DPP10 introduced hyperpolarizing shifts in the conductance-voltage relationship (approximately 19 mV) as well as steady-state inactivation (approximately 7 mV). The effects of DPP10 on Kv4.1 were similar to Kv4.2; however, distinct biophysical differences were observed. Additional experiments suggested that the cytoplasmic N-terminal domain of DPP10 determines the acceleration of inactivation. In summary, DPP10 is a potent modulator of Kv4 expression and biophysical properties and may be a critical component of somatodendritic ISA channels in the brain.

FIGURE 1

Sequence comparison between DPP10 and DPP6 in human. Based on the structure of DPP4/CD26, DPP6 and DPP10 proteins consist of a cytoplasmic N-terminal domain, a transmembrane domain (TM, with underline), and a long C-terminal domain. The C-terminal domain in DPP4/CD26 has been partitioned into a glycosylation-rich, a cysteine-rich, and a catalytic region (Mentlein, 1999). Putative potential N-glycosylation sites in DPP6 and DPP10 are indicated in bold and underline, and the conserved catalytic residues are indicated by asterisks. Note that the DPP4 catalytic serine (GKSYGG) in DPP10 and DPP10 is respectively replaced by aspartate (D) and glycine (G). Residues identical between DPP6 and DPP10 are highlighted by shading. The S9B dipeptidyl peptidase consensus sequence DW(V/L)YEEE is indicated with underline and S9B. Transcripts of DPP10 and the two isoforms of DPP6, DPP6-S (short) and DPP6-L (long), are present predominantly in the brain, as detected by Northern hybridization (Qi et al., 2003; Allen et al., 2003).

FIGURE 2

Physical association between DPP10 and Kv4.2 and determination of glycosylation level of DPP10. (A) Coimmunoprecipitation of DPP10 with Kv4.2 in Xenopus oocytes. Membrane preps of oocytes expressing Kv4.2 alone (left blots, upper and lower), HA-tagged DPP10 (HA/DPP10) alone (middle blots, upper and lower), and both (right blots, upper and lower) were immunoprecipitated using goat IgG, goat monoclonal anti-Kv4.2 antibody, or goat monoclonal anti-HA antibody (as indicated). The immunoprecipitated proteins were separated out on 6% SDS-PAGE gels and transferred onto Immobilon membranes. The immunoblots were probed using rabbit anti-Kv4.2 antibody (lower blots) or rat anti-HA antibody (upper blots), followed by secondary anti-rabbit and anti-rat antibodies conjugated to horseradish peroxidase (HRP). The Kv4.2 and anti-HA antibodies were highly specific in our assays, as seen in upper left and lower middle blots. HA/DPP10 ran at ∼97 kDa. Kv4.2 bands represents monomeric and aggregate forms. (B) DPP10 is less glycosylated when compared to DPP6-S (lower blots). A subset of oocytes injected with Kv4.2, Kv4.2 + HA/DPP6-S, and Kv4.2 + HA/DPP10 cRNAs for expression were treated with tunicamycin (tu: 2.5 mM) by injection and incubation in tunicamycin-containing ND96. Whole-oocyte homogenates were separated on SDS-PAGE, transferred onto Immobilon, and probed with anti-Kv4.2 and anti-HA antibodies. Coexpression with HA/DPP10 or HA/DPP6-S does not significantly affect Kv4.2 protein levels, but tunicamycin treatment produces marked reduction (upper blots). Xo = uninjected oocyte.

FIGURE 3

Effects of DPP10 on Kv4.2 current in Xenopus oocytes. (A) Whole-oocyte currents of Kv4.2 and Kv4.2 coexpressed with DPP10. Families of current traces were elicited by depolarizing steps from −100 mV to +70 mV at 10 mV steps for 1 s, but only the initial 250 ms is shown. Inset shows the current waveform at +50 mV during the initial 25 ms. (B) Comparison of Kv4.2 current expression with and without DPP10 coexpression. Each data point set (connected by line) represents results from single-day sessions (n = 5). As indicated in Table 1, DPP10 increased Kv4.2 current expression on average by ∼5.9-fold. (C) Voltage-dependence of time-to-peak. Measurements taken from traces from respective voltages and plotted against the membrane potential. (D) Voltage-dependence of peak conductance and steady-state inactivation. The peak conductance-voltage relationship was calculated from current-voltage relationship according to the method described in Materials and Methods. The steady-state inactivation protocol consists of a 10-s prepulse that varied between −100 and −30 mV in 10-mV increments, followed by a test pulse at +50 mV. The time between episode starts was 30 s. (E) The time course of recovery from inactivation at −100 mV, measured using a double-pulse protocol. From a holding potential of −100 mV, a +50 mV depolarization of 1 s (prepulse) activated and inactivated the channels. A second pulse to +50 mV (test pulse) was delivered after a variable interpulse interval (shown on panel) at −100 mV. Non-inactivating current at the end of the prepulse was subtracted from peak values of prepulse and test pulse, and subsequently the adjusted test pulse value was divided by that of the prepulse to produce the fractional recovery. The inter-episode interval for the recovery protocol was 15 s. •, Kv4.2; ○, Kv4.2 + DPP10. Data points shown as mean ± standard deviation.

FIGURE 4

Coexpression of DPP10 changes the subcellular localization of Kv4.2 expressed in COS-7 cells by promoting surface trafficking. COS-7 cells were transiently transfected with EGFP-Kv4.2 cDNA alone (A) or along with DPP6-S (B), DPP6-L (C), KChIP3 (D), or DPP10 (E) cDNA. The EGFP signals were imaged by confocal microscopy, showing the distribution of total Kv4.2 proteins. When expressed alone, Kv4.2 proteins accumulate in the perinuclear ER/Golgi compartments (A). Coexpression of DPP10 (E) redistributes Kv4.2 proteins, decreasing ER/Golgi accumulation and increasing cell surface expression. Similar Kv4.2 redistribution is observed with coexpression of DPP6-S (B), DPP6-L (C), and KChIP3 (D). Scale bars, 20 μm (as indicated).

FIGURE 5

DPP10 affects Kv4.1 kinetics and steady-state properties. (A) TEVC recordings of currents from oocytes expressing Kv4.1 and Kv4.1 + DPP10. Current traces were elicited by depolarizing steps from −100 to +60 mV at 10 mV steps for 1-s duration, with the first 500 ms shown here. (Inset) Current waveforms at +50 mV during the initial 25 ms. (B) Normalized and superimposed traces of Kv4.1 (thick) and Kv4.1 + DPP10 (thin) currents at +50 mV, showing the dramatic acceleration in the time course of inactivation in the presence of DPP10 (C) Time-to-peak measurements for voltages ranging from −40 mV to +80 mV. (D) Voltage-dependence of peak conductance and steady-state inactivation. (E) Recovery from inactivation at −100 mV as determined by double-pulse protocol. The experiments were conducted and analyzed as described in Fig. 3. • = Kv4.1, ○ = Kv4.1 + DPP10. Data points shown as mean ± standard deviation.

FIGURE 6

Quantitative analysis of Kv4.2 inactivation at high and low voltages. (A) Normalized and superimposed traces of Kv4.2, Kv4.2 + DPP10, and Kv4.2 + DPP6-S currents at high voltage (+50 mV). (B) The effects of DPP10 and DPP6-S on Kv4.2 inactivation at +50 mV. Sum of three exponential terms was necessary to properly describe the current decays over 1 s, generating time constants and fractional amplitudes for fast (fast), intermediate (inter), and slow (slow) components. The fractional amplitude of the non-inactivating, steady-state (S-S) current is also shown. • = Kv4.2, ○ = Kv4.2 + DPP10, ▵ = Kv4.2 + DPP6-S. (C) The time course of pre-open, low-voltage inactivation of Kv4.2 and Kv4.2 + DPP10 measured at V_0.5 and V_0.5 ± 5 mV of steady-state inactivation using the protocol described in the inset. In brief, from the holding potential of −100 mV, a 100-ms pulse to +50 mV was delivered to obtain the control current. This pulse was followed by a return to −100 mV for 1 s to allow full channel recovery. The membrane was then stepped to a prepulse potential (around V_0.5 of steady-state inactivation) for a variable duration before depolarizing again to +50 mV for 500 ms to test amount of available, activatable channels. Single exponential function was used to describe the data. (D) The voltage-dependence of pre-open inactivation at ∼V_0.5.

FIGURE 7

Effects of HA-tagging DPP10 and DPP6-S at the N-terminus. (A, left panel) Kv4.2, Kv4.2 + DPP10, and Kv4.2 + HA/DPP10 currents at +50 mV, after normalization. (B, left panel) Kv4.2, Kv4.2 + DPP6-S, and Kv4.2 + HA/DPP6-S current traces at +50 mV after normalization. (Middle panels) Comparison of peak conductance-voltage relationships. (Right panels) Comparison of recovery from inactivation at −100 mV.

FIGURE 8

Probing the role of the DPP10 cytoplasmic N-terminus. DPP10/6 chimera consists of DPP10 cytoplasmic N-terminal domain fused to the transmembrane and extracellular C-terminal domain of DPP6-S. In DPP6/10, the N-terminal domain of DPP10 is replaced by that of DPP6. (A and B, left and middle panels) Families of current traces from oocytes expressing Kv4.2 + DPP10, Kv4.2 + DPP6-S, Kv4.2 + DPP10/6, and Kv4.2 + DPP6/10 provoked by voltage steps from −100 mV to +80 mV at 10 mV steps over 100 ms. The holding potential was −100 mV. Capacitance and leak were subtracted online by using P/4 protocol. (Right panels) Normalized traces at +50 mV superimposed for comparison.

FIGURE 9

Kinetic state diagram describing the allosteric model for Kv4 inactivation. In this model (Beck and Covarrubias, 2001; Bähring et al., 2001a), Kv4 channels can inactivate either from the open state or preferentially from closed states along the activation pathway. The coupling between closed-state inactivation and voltage-dependent activation (transitions from C₀ to C₄) is defined by the allosteric factor (f), resulting in increased likelihood of inactivation occurring from later closed states. Transitions between states are represented by forward and reverse arrows.