Cellular mechanisms of mutations in Kv7.1: auditory functions in Jervell and Lange-Nielsen syndrome vs. Romano-Ward syndrome

Abstract

As a result of cell-specific functions of voltage-activated K(+) channels, such as Kv7.1, mutations in this channel produce profound cardiac and auditory defects. At the same time, the massive diversity of K(+) channels allows for compensatory substitution of mutant channels by other functional channels of their type to minimize defective phenotypes. Kv7.1 represents a clear example of such functional dichotomy. While several point mutations in the channel result in a cardio-auditory syndrome called Jervell and Lange-Nielsen syndrome (JLNS), about 100-fold mutations result in long QT syndrome (LQTS) denoted as Romano-Ward syndrome (RWS), which has an intact auditory phenotype. To determine whether the cellular mechanisms for the diverse phenotypic outcome of Kv7.1 mutations, are dependent on the tissue-specific function of the channel and/or specialized functions of the channel, we made series of point mutations in hKv7.1 ascribed to JLNS and RWS. For JLNS mutations, all except W248F yielded non-functional channels when expressed alone. Although W248F at the end of the S4 domain yielded a functional current, it underwent marked inactivation at positive voltages, rendering the channel non-functional. We demonstrate that by definition, none of the JLNS mutants operated in a dominant negative (DN) fashion. Instead, the JLNS mutants have impaired membrane trafficking, trapped in the endoplasmic reticulum (ER) and Cis-Golgi. The RWS mutants exhibited varied functional phenotypes. However, they can be summed up as exhibiting DN effects. Phenotypic differences between JLNS and RWS may stem from tissue-specific functional requirements of cardiac vs. inner ear non-sensory cells.

Keywords: genetic diseases; hearing loss; membrane trafficking; mutant; potassium channels.

Figure 1

Schematic diagram of hKv7.1. JLNS mutations are indicated in blue, RWS, mutations are indicated in red. Amino acid numbers in transmembrane domains and P-loop regions were labeled based on UniProtKB/Swiss-Prot P51787. S4-S5 linker, channel gating; P-loop, permeation; C-terminal, regulation and tetramerization.

Figure 2

Variations in current phenotypes of JLNS mutation in the hKv7.1 channel expressed in CHO cells A. EGFP-alone transfected cells did not yield measurable current. Membrane K⁺ currents traces recorded from CHO cells after 24 h of transfection with EGFP, at a holding potential of −80 mV and step voltages ranging from −110 to 50 mV, with Δ V = 10 mV. (B) In contrast, robust outward K⁺ current traces ensued after wild-type (WT) hKv7.1 was transfected after 24 h. The current traces were generated using the voltage protocol described in (A). Tail currents were recorded at −40 mV. (C) The corresponding current density-voltage relation shown was generated from n = 11 cells. The activation voltage of the expressed hKv7.1 current was ~-50 mV. (D) Voltage-dependence of steady-state activation generated from tail currents (I) as a ratio of the maximum tail current (I_max). Each symbol represents mean current ratio obtained from 11 cells. Continuous line represents a single Boltzmann function fit to the data points. The half-activation voltage (V_1/2) was −16.2 ± 0.9 mV, and the slope was e-fold for 11.4 ± 0.8 mV (n = 11). (E) Using similar activation voltage steps as described in (A), CHO cells transfected with hKv7.1 channel mutants (MT), linked to JLNS, namely T311I, T322M, A336fs+16X, R518X, Q530X, E543fs+107X, and G589D, were evaluated. The MTs and resulting current traces are indicated. None of the seven mutants yielded detectable current, even after 48 h post-transfection. (F) Another recognized JLNS mutation is W248F. Unlike the other seven mutants, W248F yielded reduced current with pronounced inward rectification at positive voltage steps. Shown is a family of current traces recorded from transfected CHO cells (24-h-old). (G) Current density (in pA/pF)-voltage relations of the WT (in gray and dashed lines) and MT channels. Each symbol represents the mean of 11 cells.

Figure 3

Properties of heteromeric hKv7.1 and JLNS mutant channel currents. (A) Macroscopic outward currents were recorded after expressing homomeric wildtype (WT:MT (4:0) channel, left panel) and upon co-expressing heteromeric hKv7.1 (WT) and MT T322M at different ratios (WT:MT) of 3:1 (left middle panel), 2:2 (right middle panel), and 1:3 (right panel). Representative current traces for a family of K⁺ currents obtained from a holding potential of −80 mV and stepped from −100 to 50 mV using Δ V = 10 mV are shown. The tail currents were elicited at −40 mV. (B) Plots of current density-voltage relation of currents derived from WT hKv7.1 alone (▾) and a combination of WT Kv7.1 and MT T322M (3:1, ◾ 2:2, ⚫, and 1:3, ▴). At a step potential of 0 mV, the current density (pA/pF) for WT hKv7.1 was 29.9 ± 1.9, as compared with the combined WT:MT (3:1) was 8.4 ± 2.8, WT:MT T322M (2:2) was 5.2 + 3.1 and WT: T322M (1:3) was 3.3 ± 1.3 (n = 11). (C) Co-expression of the MT T322M with WT-hKv7.1 channels did not alter the gating properties. Steady-state activation curves of WT Kv7.1 homomeric (▾) and heteromeric (3:1, ◾ 2:2, ⚫ and 1:3, ▴) currents are shown. Tail currents were measured immediately after pulsing to −40 mV, normalized to the largest tail current recorded, and plotted against the preceding pre-pulse voltages. Neither the mid-point (V_1/2, in mV) nor the slope factors (k, in mV) were statistically different among the WT and cocktails of MT at different ratios. The V_1/2 (in mV) and k (in mV) for WT Kv7.1 homomeric channel, and heteromeric WT:T322M (3:1; 2:2; 1:3) were: WT = −15.9 ± 1.1, 11.5 ± 0.5; WT:MT 3:1 = −16.0 ± 0.8, 11.7 ± 0.8; WT:MT 2:2 = 2:2 = −14.8 ± 1.8, 10.8 + 1.1; WT:MT 1:3 = 15.3 ± 0.6, 11.5 ± 0.8 (n = 11; p = 0.3, 0.4, 0.4, 0.5), respectively. respectively. (D) The currents derived from the WT hKv7.1 homomeric (green trace) and heteromeric channels (WT:T322M, 2:2; red trace) were normalized and superimposed, as shown in the inset. The time constants (τ) of mutations (MT, T311I, T322M, A336fs+16X, R518X, Q530X, E543fs+107X, and G589D) expressed jointly with the WT at a ratio of 2:2 were comparable, ranging from ~65 to 170-ms except W248F, which ranged from 27 to 52-ms, depending on the step voltage. Data were assembled from n = 11 Data were assembled from n = 11 cells.

Figure 4

Summary of experimental and predicted mechanisms of JLNS mutations. The four α-subunits of K⁺ channels assemble to form a functional channel. The dotted lines denote the expected current reduction if the channels form monomers (black), dimers (red), trimers (green), tetramers (blue), and pentamers (magenta). We generated a dominant negative version of hKv7.1 in which the pore amino acids GYG (WT) were mutated to AAA (MT). The experimental data (in blue, star symbol) are in accord with the predicted relationship of tetramers, demonstrating that the WT_GYG and MT_AAA subunits co-assemble equally well with each other to form tetrameric functional channels. Current reduction of WT hKv7.1 by MT channels, T311I (⚫), W248F (○), T322M (♦), A336fs (▵) R518X (▾), E543fs (◻), G589D (◊) vs. WT/(WT + MT) ratio of DNA transfected. Reduction of WT hKv7.1 current was enhanced with decreasing WT:MT ratio. Whereas co-expression of WT hKv7.1 and the pore mutant channel AAA (dominant negative mutant) at different ratios yielded current magnitudes which confirmed the tetrameric structure of functional K⁺ channels, the experimental data from JLNS mutants deviated from the predicted relationship of tetrameric K⁺ channel.

Figure 5

Co-expression of JLNS mutant W248F and wild-type (WT) hKv7.1 channels. (A) An example of outward current traces recorded from a holding potential of −80 to 50 mV using a voltage increment of 10 mV in depolarizing and hyperpolarizing step voltages. Current traces recorded CHO cells, which were transfected with WT-hKv7.1:MT-W248F (3:1, left panel), (2:2, middle panel), and (1:3, right panel). (B) Group data of current density-voltage curves (mean ± SD) for co-expression of WT-hKv7.1 alone () with WT-hKv7.1:MT-W248F (3:1, ◻), (2:2, ⚫), (1:3, ▵) and (0:4, ▾). In contrast to the WT channel-alone outward currents, co-expression of the WT and MT, as well as the MT channel by itself, yielded currents which showed robust inward rectification at depolarizing voltages greater than 0 mV step (data were generated from 14 cells for each group (3:1), (2:2), (1:3) and (0:4). (C) Summary data of the steady-state voltage-dependent activation of WT-hKv7.1 alone (), WT-hKv7.1:MT-W248F (3:1, ◻), (2:2, ⚫), (1:3, ▵) and (0:4, ▾). The V_1/2 of the steady-state activation curves of the five combinatorial expressions and the resulting currents were (in mV): WT-hKv7.1 alone (1:0), (3:1), (2:2), (1:3) and (0:4) −16.2 ± 0.9, −16.1 ± 0.8, −21.9 ± 1.6, −32.0 ± 0.7, −31.9 ± 1.2 (n = 9), respectively. The slope factors (k) of the resulting Boltzmann function curves were also not statistically different. The k values (in mV) for WT-hKv7.1 alone (),WT-hKv7.1:MT-W248F (3:1, ◻), (2:2, ⚫), (1:3, ▵) and (0:4, ▴) were 11.4 ± 0.8, 12.1 ± 0.5, 11.6 ± 1.1, 12.1 ± 0.7 and 12.2 ± 0.6 (n = 11; p = 0.7), respectively.

Figure 6

Activation properties of homomeric and heteromeric hKv7.1 channel and RWS mutants, D242N, N586D and L619M currents. (A) Whole-cell outward currents were recorded upon expressing hKv7.1 RWS mutant (MT-D242N) channel singly (left panel). Current traces recorded after co-expressing WT hKv7.1 and MT D242N at a ratio of 1:1 (middle panel). Representative current traces for a family of K⁺ currents obtained from a holding potential of −80 mV and stepped from −100 to 60 mV using Δ V = 10 mV are shown. The tail currents were elicited at −40 mV. Current traces recorded from CHO cells transfected with WT:MT-D242N ratio of 3:1 (right panel). (B) Plots of current density-voltage relation of currents derived from WT hKv7.1 alone () and a combination (ratio 2:2) of WT Kv7.1 and mutants (MT-D242N, ⚫, MT-N586D, ▴, and MT-L619M, ◻). (C) Heteromeric association between the WT hKv7.1 and MT-D242N, N586D and L619M at a ratio (2:2) altered the voltage-dependent activation of the ensuing currents. Steady-state activation curves of hKv7.1 alone () and after suppression by co-joint expression with the MT channels are shown. Tail currents were measured immediately after pulsing to −40 mV, normalized to the largest tail recorded, and plotted against the preceding pre-pulse voltages. The midpoint (V_1/2, in mV) and the slope factors (k, in mV) were as follows: V_1/2 and k for WT-hKv7.1 alone and in combination WT-hKv7.1:D242N, WT-hKv7.1:N586D, and WT-hKv7.1:L619M were −16.3 ± 0.8, 11.4 ± 0.9; −1.0 ± 1.5, 13.1 ± 1.1; −12.8 ± 1.2, 11.7 ± 0.6; −13.2 ± 0.9, 11.8 ± 0.9 mV (n = 10), respectively. (D) Homomeric MT channels and combined expression of the MT channels (D242N, N586D, and L619M) and WT hKv7.1 produced ~3-8-fold increase in the time constant of activation (WT-hKv7.1 (), D242N (○), N586D (▵), L619M (◾), WT-hKv7.1:D242N (2:2) (⚫), WT-hKv7.1:N586D (▴), WT-hKv7.1:L619M (◻) (n = 11 cell for each group).

Figure 7

Current phenotype of homomeric and heteromeric hKv7.1 channel and RWS mutants, R243P, L250H and G306V channels. (A) CHO cells were transfected with RWS mutant R243P DNA. Whole-cell outward currents were recorded after 48 h. Cells were held at −80 mV and stepped to voltages ranging from −100 to 60 mV using Δ V = 10 mV. Homomeric MT channel R243P did not yield outward currents (left panel). Similar currents were obtained for homomeric L250H and G306V MTs. However, co-expression WT hKv7.1 and MT R243P at a ratio of 2:2 (middle panel) yielded reduced current magnitude (middle panel), which was enhanced further after increasing the ratio of the WT-hKv7.1:R243P (3:1, right panel). (B) Plots of representative current density-voltage relation of currents derived from WT hKv7.1 alone () and a combination (ratio 2:2) of WT Kv7.1 and MTs (MT-R243P ◻, MT-L250H ◾, and MT-G306V ○). (C) Co-expression of WT hKv7.1 and MT-R243P, L250H and G306V at a ratio (2:2) produced a right-ward shift in the voltage-dependent activation of the resulting currents. Steady-state activation curves of hKv7.1 alone () and after suppression by co-expression with the MT channels are shown. Tail currents were measured at −40 mV, and normalized to the largest tail current magnitude, and plotted against the preceding pre-pulse voltages. The midpoint (V_1/2, in mV) and the slope factors (k, in mV) were as follows: V_1/2 and k for WT-hKv7.1 alone and in combination WT-hKv7.1:R243P, WT-hKv7.1:L250H, and WT-hKv7.1:G306V were −15.8 ± 1.2, 11.8 ± 0.9; 2.5 ± 1.1, 12.4 ± 0.6; 7.6 ± 2.2, 12.7 ± 1.0; 15.7 ± 2.4, 12.8 ± 1.6 mV (n = 11), respectively.

Figure 8

RWS MT D317N and L374fs+43X exhibit strong dominant negative effect on WT hKv7.1 channel functions. (A) Characteristic current traces recorded from a holding potential of −80 mV and activated from −100 to 40 mV with Δ V of 10 mV. The MT subunits (D317N and L374fs+43X) when expressed alone did not produce measurable currents. Co-expression of the WT-hKv7.1 and MT subunits yielded little or no measurable currents. (B) Current density (in pA/pF)-voltage relations of the WT and MT channels and after co-expression of both subunits as indicated. Each symbol represents the mean of 11 cells. (C) Using similar analyses as described in Figure 4, we show that the RWS MTs indicated suppressed the WT-hKv7.1 current in a DN fashion. We used the peak current magnitude at 40 mV step voltage to perform the analyses.

Figure 9

Detection of cell surface expression using epitope tagged hKv7.11. HA- or c-Myc-epitope tag was inserted with an extension of ClC-5 chloride channel D1–D2 loop sequences into the extracellular loop between S1 and S2-transmembrane domains. Farnesylated GFP was used as a plasma membrane binding protein. (A) Wild-type (WT); Anti-HA antibody stained wild-type hKv7.1 protein expressed on the cell surface in non-permeabilized condition (NP), and stained both the cell surface and the cytoplasm expressed proteins in permeabilized condition (P). (B) JLNS mutants (MT); One MT out of 8, W248F, was detected on the cell surface, while T311I was not detected on the cell surface (left panel). When the MT subunit was co-transfected with WT subunit, T311I MT failed to localize on the plasma membrane while WT subunits were detected in NP conditions. Permeabilized cells expressing both WT and MT subunits (right panel). (C) RWS MTs; All of the RWS MTs were detected on the cell surface with different levels of intensities. Left panels: Fluorescent intensities of GFP (green) and hKv7.1 channel (red) were plotted against the distance, which was marked in the merged image with the white arrow. DAPI (blue; 4′, 6-diamidino-2-phenylino-2-phenylindole) is a fluorescent stain that binds strongly to A-T rich regions in DNA, and was used as nuclear stain. Right panels: HA-MT subunit (red) and c-Myc-WT subunit (cyan) were plotted against the distance.

Figure 10

Detection of subcellular localization of mutant hKv7.1 channels. HA-tagged MT Kv7.1 channels and subcellular organelles were double-stained in permeabilized cells (P450/ER, anti-cytochrome P450 as endoplasmic reticulum marker; P230/tranG, anti-golgin A4 (P230) as trans-Golgi marker; GM130/cisG, anti-golgin A2 (Golgi matrix protein of 130 kDa) as cis-Golgi marker; EEA1/e-Endo, anti-early endosome antigen 1 as endosome marker). Fluorescent intensities of MT hKv7.1 (red) and subcellular organelles (cyan) were plotted against the distance, which was marked in a merged image with a white arrow. The overlap pattern of red and cyan signals showed different levels of co-localization between organelles.