Surface expression and subunit specific control of steady protein levels by the Kv7.2 helix A-B linker

Abstract

Kv7.2 and Kv7.3 are the main components of the neuronal voltage-dependent M-current, which is a subthreshold potassium conductance that exerts an important control on neuronal excitability. Despite their predominantly intracellular distribution, these channels must reach the plasma membrane in order to control neuronal activity. Thus, we analyzed the amino acid sequence of Kv7.2 to identify intrinsic signals that may control its surface expression. Removal of the interlinker connecting helix A and helix B of the intracellular C-terminus produces a large increase in the number of functional channels at the plasma membrane. Moreover, elimination of this linker increased the steady-state amount of protein, which was not associated with a decrease of protein degradation. The magnitude of this increase was inversely correlated with the number of helix A - helix B linkers present in the tetrameric channel assemblies. In contrast to the remarkable effect on the amount of Kv7.2 protein, removal of the Kv7.2 linker had no detectable impact on the steady-state levels of Kv7.3 protein.

Figure 1. Identification of C-terminal regions that influence Kv7.2 surface expression. A.- and C.-

Left, schematic representation of serial deletions introduced into the intracellular C-terminal Kv7.2 sequence. The boxes in the expanded intracellular C-terminal scheme indicate regions with a high probability of adopting an alpha helix configuration, and ovals (numbered 1 to 4) indicate regions with a significant PESTfind score. The PESTfind score (www.biu.icnet.uk/projects/pest) was 19.1, 12.6, 6.1, and 8.6 for PEST1 (₃₉₇KDPPPEPSPSQK₄₀₈), PEST2 (₄₃₇RSPSADQSLEDSPS₄₅₁), PEST3 (₄₇₆RQNSEEASLPGEDIVDD₄₉₃), and PEST4 (₈₀₃RPYIAEGESDTDSDLCTPCGPPPR₈₂₆), respectively. Right, normalized surface expression in Xenopus oocytes of the Kv7.2 subunits indicated on the left tagged with HA at the extracellular S1–S2 loop (n ≥12, two batches). The amount of Kv7.2-HA containing channels in the oocyte membrane was quantified using a single whole-oocyte chemiluminescence assay (see Materials and Methods). The background of uninjected oocytes was subtracted and the values given are the means (± SEM) normalized to the values obtained from WT-Kv7.2-HA channels from the same batch. * P≤0.05; *** P≤0.001; unpaired Student’s t test. B.- and D.- Steady-state protein levels differed among the various mutant subunits. Proteins from Xenopus oocytes injected with the same amount of mRNA expressing the indicated HA-tagged constructs were separated by SDS-PAGE and analyzed in Western blots probed with anti-HA antibodies (n = 4).

Figure 2. Characterization of the interaction of Del6 with CaM. A.

Emission spectra of 12.5 nM D-CaM in the absence (left panel; 10 mM EGTA and no added Ca²⁺), and in the presence of 2 µM free Ca²⁺ (right panel), as well as in the presence of molar excess of the indicated GST-Kv7.2 fusion proteins (200 nM). Note the difference in the fluorescence emission axis, as well as the shift in the peak emission to the left in the presence of Ca²⁺. B. Relative concentration-dependent enhancement of D-CaM fluorescence emission in the absence (left) and presence of 2 µM free Ca²⁺ (right). The parameters used to fit a Hill equation to the data for WT in absence of Ca²⁺ were: Max = 111±1.4, EC₅₀ = 11±0.5 nM, h = 1.8±0.2 (n = 9). For Del6 in absence of Ca²⁺: Max = 83±1.3, EC₅₀ = 13±0.7 nM, h = 1.5±0.1 (n = 5). For WT in the presence of Ca²⁺: Max = 115±2.2, EC₅₀ = 27±1.2 nM, h = 3.1±0.4 (n = 9). For Del6 in presence of Ca²⁺: Max = 95±1.1, EC₅₀ = 31±0.8 nM, h = 4.2±0.3 (n = 5). The data represent the means ± standard error from five or more independent experiments. The error bars are smaller than the symbols.

Figure 3. The Del6 mutant subunit presents increased Kv7.2 total protein. A.-

Western blot of increasing sample load from HEK293T cells extracts expressing the same amount of DNA coding WT- or Del6-Kv7.2 Myc-tagged subunits. The optical densities (OD) of the bands were analyzed using ImageJ software. A linear regression fit was obtained from the OD vs load relation. From this regression, the amount required to obtain the same OD’s was found to be 4.7±1.22 fold larger (n = 5) for WT subunits than for Del6 mutant Kv7.2 subunits. Top, the cartoons are schematic representations of the tetrameric assemblies. Each square represents a subunit of the tetrameric channel, the ovals represent the A-B loops, and the black circles highlights the presence of a Myc tag. B.- The steady-state protein levels are inversely proportional to the number of A–B loops present in the tetrameric Kv7.2 assemblies. Western blot of protein extracts from HEK293T cells expressing Del6- (right panel) or WT- Myc-Kv7.2 subunits (left panel) when co-expressed with a six fold larger amount of plasmid DNA coding the indicated YFP-tagged protein (n = 4). Myc tagged WT and Del6 Kv7.2 channels were detected using anti-Myc antibody. The Del6 protein levels decreased when co-expressed with WT subunits, whereas increased protein levels of WT Kv7.2 were detected when co-expressed with Del6. A schematic representation of the theoretical assemblies detected with the anti-Myc antibody is represented at the top. The subunits that were overexpressed and that lack a Myc tag, so they are not detected in the Western blot, are indicated at the bottom of each lane. C.- Normalized quantification of Western blot signals as a function of the theoretical number of A–B loops present on the detected channel assemblies. On each Western blot, the OD’s were normalized to the signal obtained for tubulin. The values were subsequently normalized to those obtained with Myc-Del6 or Myc-WT in the same Western blot. Data points for 0 (Del6) and 4 (WT) loops were derived from data obtained as in panel A, data points for 1 and 3 loops were obtained from Western blots equivalent to the left and right columns in panel B, respectively. The points represent the means ± S.E.M. (n ≥4). D.- Pulse-chase analysis of WT-, Del2- and Del6-Kv7.2 subunit stability. Densitometric quantification of the bands normalized to the value at time 0 (no chase). Each data point is the mean ± SEM calculated from three separate experiments. Inset: representative images of autoradiographic films of experiments in HEK293T cells transfected with the indicated plasmids. Metabolic labeling was performed for 1 h, 36 h post-transfection, followed by chase times of 1, 2 and 4 h.

Figure 4. Removal of the A-B linker resulted in functional Kv7.2 channels. A.-

Representative currents recording from HEK293T cells transfected with WT-, Del6- or Del2-Kv7.2, activated from a holding potential (V_h) = −30 mV after 1,500 ms steps to the indicated voltages. B.- Current density-voltage relationship from tail currents of WT (n = 13) or Del6 (n = 15) channels. Each point represents the mean ± SEM. A Boltzmann equation D = Dmax/(1+e^((V-V _1/2 ^)/S)) was fitted to the data. The averaged Boltzmann parameters were: WT: V_1/2 =  −34.8±1.9 mV, Slope = 11.6±1.7, Dmax = 30.7±0.9 pA/pF; Del6: V_1/2 =  −30.6±5.1 mV, Slope = 11.2±4.5, Dmax = 34.4±2.6 pA/pF; Del2: V_1/2 =  −29.6±5.5 mV, Slope = 14.2±4.5, Dmax = 34.1±2.6 pA/pF.

Figure 5. Removal of the A-B linker did not decrease PIP₂ affinity. A.-

Current recorded in cells transfected with WT and Danio rerio voltage dependent phosphatase (VSP, red) and Del6+ VSP (blue) activated from a holding potential (V_h) = −60 mV. The initial pulse to −20 mV opens the channels without activating VSP. The second pulse to +100 mV opens additional channels faster than VSP is activated, giving rise to an initial current increase, followed by a decline. The decline phase is governed by the reduction on PIP₂ levels at a rate that depends on the PIP₂ affinity of the channels. After a variable period (2.560 ms in this example), voltage is returned to −20 mV, and the relative inhibition can be measured. This protocol is described in detail in . B.- Averaged time course of the current decline during 2.560 ms VSP activation at +100 mV in cells transfected with WT + VSP (red; n = 6) and Del6+ VSP (blue; n = 11) subunits. The shadows represent the mean ± SEM. C.- Normalized current at −20 mV (after/before step to +100 mV) for different durations at +100 mV. Each point represents the mean ± SEM for 6–11 cells. The data from cells expressing Del6 subunits presented a large scatter, and the differences with the data from WT expressing cells did not reach statistical significance. D.- Time-course during recovery at −20 mV. The shadows represent the mean ± SEM for data from cells expressing WT (red) or Del6 (blue) channels. The rate of recovery after PIP₂ depletion was indistinguishable.

Figure 6. Surface expression of WT and Del6 Kv7.2 subunits in human HEK293T cells.

Analysis of confocal images of non-permeabilized cells expressing the indicated constructs. The subunits have a mCFP tag at the N-terminus (rendered in green) and an extracellular 2×HA tag, allowing simultaneously monitoring total (green) and surface expression (red). The proportion of the cells with surface staining in confocal images was determined in >40 mCFP positive cells for each construct in three independent experiments. A. - Grey bars represent mean ± SEM of the percentage of cells expressing the channel at the surface. *** P≤0.001; unpaired Student’s t test. B. - Representative images of cells expressing the indicated subunit. C.- Ratio of surface/total expression (red fluorescence/cyan fluorescence) from wide field epifluorescence images of cells expressing the indicated Kv7.2 subunits. *** P≤0.001; unpaired Student’s t test. D.- Plot of the cyan fluorescence vs red fluorescence intensity from wide field epifluorescence images of cells expressing the subunits indicated. There was no correlation between total and surface expression (>90 cells from more than 5 independent experiments).

Figure 7. The Kv7.2 A–B linker is not critical for Kv7.2 mediated reduction on Kv7.3 protein levels. A.-

Western blot of protein extracts from HEK293T cells transfected with YFP-Kv7.3 and the constructs indicated at the bottom of each column in a 1∶5 ratio. The constructs indicated at the bottom did not have a fluorescent protein tag, and were not detected by the anti-GFP antibody. B. - Densitometric quantification of the band intensity relative to the tubulin signal, normalized to the value obtained for cells overexpressing Kv7.3. Overexpression of WT- or Del6-Kv7.2 caused a significant reduction in the signal of YFP-Kv7.3, and the extent reduction was not statistically different. Bars represent mean ± SEM from 5 experiments. ** P≤0.01; * P≤0.05; paired Student’s t test. C.- Western blot of protein extracts from cells expressing WT or Del6 subunits tagged with YFP (detected with the anti GFP antibody) in conjunction with Kv7.3 (tagged with HA) in a 1∶5 ratio. D.- Densitometric quantification as in B from the data obtained as in C from three independent experiments. No significant differences were observed on the relative abundance of WT or Del6 subunits upon Kv7.3 overexpression. E.- Representative current recording from HEK293T cells transfected with Kv7.3 and WT- or Del6-Kv7.2, activated from a holding potential (V_h) = −30 mV after 1,500 ms steps to the indicated voltages. F.- Current density-voltage relationship from tail currents of Kv7.3 co-expressed with WT or Del6 (n = 7) Kv7.2 subunits. Each point represents the mean ± SEM. The averaged Boltzmann parameters were: Kv7.2/3: V_1/2 =  −38.2±8.5 mV, Slope = 12.6±5.1, Dmax = 73.0±3.4 pA/pF; Del6/3: V_1/2 =  - ±3.5 mV, Slope = 12.0±3.1, Dmax = 94.3±8.4 pA/pF. Inset: Maximal current density from cells expressing the indicated subunits (data from Fig. 4D and 6D). * P≤0.05 for maximal density; unpaired Student’s t test. G. Influence of Kv7.3 in surface expression of Kv7.2 WT and Del6 subunits. On top, cartoon representing the scheme of the experiment, indicating that only Kv7.2 subunits are detected at the surface due to the presence of an extracellular tag. Middle, bar graph of the surface expression index normalized to the same index obtained without Kv7.3 (see inset and Fig. 5C). The relative increase in surface expression caused by Kv7.3 was significantly larger for Del6 subunits. * P≤0.05; unpaired Student’s t test.