Affinity for phosphatidylinositol 4,5-bisphosphate determines muscarinic agonist sensitivity of Kv7 K+ channels

Abstract

Kv7 K(+)-channel subunits differ in their apparent affinity for PIP(2) and are differentially expressed in nerve, muscle, and epithelia in accord with their physiological roles in those tissues. To investigate how PIP(2) affinity affects the response to physiological stimuli such as receptor stimulation, we exposed homomeric and heteromeric Kv7.2, 7.3, and 7.4 channels to a range of concentrations of the muscarinic receptor agonist oxotremorine-M (oxo-M) in a heterologous expression system. Activation of M(1) receptors by oxo-M leads to PIP(2) depletion through G(q) and phospholipase C (PLC). Chinese hamster ovary cells were transiently transfected with Kv7 subunits and M(1) receptors and studied under perforated-patch voltage clamp. For Kv7.2/7.3 heteromers, the EC(50) for current suppression was 0.44 +/- 0.08 microM, and the maximal inhibition (Inhib(max)) was 74 +/- 3% (n = 5-7). When tonic PIP(2) abundance was increased by overexpression of PIP 5-kinase, the EC(50) was shifted threefold to the right (1.2 +/- 0.1 microM), but without a significant change in Inhib(max) (73 +/- 4%, n = 5). To investigate the muscarinic sensitivity of Kv7.3 homomers, we used the A315T pore mutant (Kv7.3(T)) that increases whole-cell currents by 30-fold without any change in apparent PIP(2) affinity. Kv7.3(T) currents had a slightly right-shifted EC(50) as compared with Kv7.2/7.3 heteromers (1.0 +/- 0.8 microM) and a strongly reduced Inhib(max) (39 +/- 3%). In contrast, the dose-response curve of homomeric Kv7.4 channels was shifted considerably to the left (66 +/- 8 nM), and Inhib(max) was slightly increased (81 +/- 6%, n = 3-4). We then studied several Kv7.2 mutants with altered apparent affinities for PIP(2) by coexpressing them with Kv7.3(T) subunits to boost current amplitudes. For the lower affinity (Kv7.2 (R463Q)/Kv7.3(T)) or higher affinity (Kv7.2 (R463E)/Kv7.3(T)) channels, the EC(50) and Inhib(max) were similar to Kv7.4 or Kv7.3(T) homomers (0.12 +/- 0.08 microM and 79 +/- 6% [n = 3-4] and 0.58 +/- 0.07 microM and 27 +/- 3% [n = 3-4], respectively). The very low-affinity Kv7.2 (R452E, R459E, and R461E) triple mutant was also coexpressed with Kv7.3(T). The resulting heteromer displayed a very low EC(50) for inhibition (32 +/- 8 nM) and a slightly increased Inhib(max) (83 +/- 3%, n = 3-4). We then constructed a cellular model that incorporates PLC activation by oxo-M, PIP(2) hydrolysis, PIP(2) binding to Kv7-channel subunits, and K(+) current through Kv7 tetramers. We were able to fully reproduce our data and extract a consistent set of PIP(2) affinities.

Figure 1.

Agonist-induced Kv7.2/7.3-current suppression is reduced by overexpression of PIP 5-kinase. (A) Averaged time course of normalized current amplitude during sequential application of a range of concentrations of oxo-M to CHO cells coexpressing Kv7.2/7.3 channels and M₁ receptors, without (open circles) or together with PIP 5-kinase (closed circles). Times of applications of oxo-M at the given concentrations are indicated by the arrows. (B) Concentration dependence of current inhibition by oxo-M from cells expressing Kv7.2/7.3 alone (open circles) or together with PIP 5-kinase (closed circles). The lines represent the fits of experimental data by a Hill equation, with the values given in the text. Each point represents the mean ± SEM from n = 5–7 experiments. (C) Current waveforms before and after the application of a range of concentrations of oxo-M to CHO cells not coexpressing (left) or coexpressing PIP 5-kinase (right). The dashed line in the current traces is the zero current level, and the pulse protocol used is shown at the bottom in the left panel.

Figure 2.

Kv7.3^T currents display reduced sensitivity to muscarinic agonist. (A) Averaged time course of normalized Kv7.3^T-current amplitude during sequential application of a range of concentrations of oxo-M to CHO cells expressing Kv7.3^T channels and M₁ receptors. Times of applications of oxo-M at the given concentrations are indicated by the arrows. (B) Concentration dependence of inhibition by oxo-M of Kv7.3^T (open circles) and Kv7.2/7.3 (dashed line) current. The line represents the fit of the experimental data by a Hill equation with the values given in the text. Each point represents the mean ± SEM from n = 5–7 experiments. (C) Kv7.3^T-current waveforms before and after the application of oxo-M at the concentrations indicated. The dashed lines in the current traces are the zero current levels, and the pulse protocol is shown at the bottom in the right panel.

Figure 3.

Kv7.4 channels are more sensitive to suppression by muscarinic stimulation. (A) Averaged time course of normalized Kv7.4-current amplitude during sequential application of a range of concentrations of oxo-M to CHO cells expressing Kv7.4 channels and M₁ receptors. Times of applications of oxo-M at the given concentrations are indicated by the arrows. (B) Concentration dependence of inhibition by oxo-M of the Kv7.4 (closed circles) and Kv7.2/7.3 (dashed line) currents. The line represents the fit of experimental data by a Hill equation, with the values given in the text. Each point represents the mean ± SEM from n = 3–4 experiments. (C) Kv7.4 current waveforms before and after the application of oxo-M at the concentrations indicated. The dashed lines in the current traces are the zero current levels, and the pulse protocol is shown at the bottom in the right panel.

Figure 4.

Altered PIP₂ sensitivity of Kv7.2 mutant subunits confers differential sensitivity to muscarinic stimulation. (A) Averaged time course of normalized current amplitude during an application of a single concentration of oxo-M to CHO cells expressing M₁ receptors together with Kv7.2 (R463E)/7.3^T channels (closed triangles), Kv7.2 (R463Q)/7.3^T channels (open squares), or Kv7.2 (EEE)/7.3^T channels (closed circles). The time of application of oxo-M at the given concentration is indicated by the arrow. (B) Concentration dependence of inhibition by oxo-M of Kv7.2 (R463E)/7.3^T (closed triangles), Kv7.2 (R463Q)/7.3^T (open squares), Kv7.2 (EEE)/7.3^T (closed circles), or Kv7.2/7.3 (dashed line) current. The line represents the fit of experimental data by a Hill equation, with the values given in the text. Each point represents the mean ± SEM from n = 3–4 experiments. (C) Kv7.2 mutant current waveforms before and after the application of oxo-M at the concentrations indicated. The dashed lines in the current traces are the zero current levels, and the pulse protocol is shown at the bottom in the right panel.

Figure 5.

A kinetic model reproduces the experimental findings. (A) Model reactions. Oxo-M activates PLC, which hydrolyzes PIP₂. Kv7 subunits bind to PIP₂, which allows Kv7 current. Each Kv7 channel is formed by a set of four Kv7 subunits, which can differ in their PIP₂ affinity in the case of heteromers or are identical in the case of homomers. K_A values for Kv7 subunits are 500 µm⁻² Kv7.2 (WT), 75 µm⁻² Kv7.3 (assumed to be the same as Kv7.3^T), 2,500 µm⁻² Kv7.4, 10⁶ µm⁻² Kv7.2 (EEE), 75 µm⁻² Kv7.2 (R463E), and 5,000 µm⁻² Kv7.2 (R463Q). Also see Tables S1–S3. DAG, diacylglycerol. (B–F) Model outputs. (B) Steady-state levels of PIP₂ reached under a range of concentrations of oxo-M. (C and E) Maximum current inhibition over a range of concentrations of oxo-M. (D and F) Dependence of current amplitudes (relative to baseline) on PIP₂ levels.