In vivo and in vitro functional characterization of Andersen's syndrome mutations

Abstract

The inward rectifier K(+) channel Kir2.1 carries all Andersen's syndrome mutations identified to date. Patients exhibit symptoms of periodic paralysis, cardiac dysrhythmia and multiple dysmorphic features. Here, we report the clinical manifestations found in three families with Andersen's syndrome. Molecular genetics analysis identified two novel missense mutations in the KCNJ2 gene leading to amino acid changes C154F and T309I of the Kir2.1 open reading frame. Patch clamp experiments showed that the two mutations produced a loss of channel function. When co-expressed with Kir2.1 wild-type (WT) channels, both mutations exerted a dominant-negative effect leading to a loss of the inward rectifying K(+) current. Confocal microscopy imaging in HEK293 cells is consistent with a co-assembly of the EGFP-fused mutant proteins with WT channels and proper traffick to the plasma membrane to produce silent channels alone or as hetero-tetramers with WT. Functional expression in C2C12 muscle cell line of newly as well as previously reported Andersen's syndrome mutations confirmed that these mutations act through a dominant-negative effect by altering channel gating or trafficking. Finally, in vivo electromyographic evaluation showed a decrease in muscle excitability in Andersen's syndrome patients. We hypothesize that Andersen's syndrome-associated mutations and hypokalaemic periodic paralysis-associated calcium channel mutations may lead to muscle membrane hypoexcitability via a common mechanism.

Figure 1. Pedigrees of families with AS mutations

Pedigrees of 3 families (A, B and C) with Andersen's syndrome phenotypes and carrying mutations in KCNJ2 gene. Circles represent females, squares represent males, affected members are shown with filled symbols. Arrows indicate probands. Genotyped individuals are shown with (+) for mutation carriers and with (–) for non-carriers. D, alignment of amino acid sequences around the residues that have been modified in Andersen's syndrome patients. The modified residues are shown in bold in Kir2.1 and in the other Kir channels when they are conserved.

Figure 2. Electromyographic tests of AS patients

The compound muscle action potentials (CMAPs) of the abductor digiti minimi (ADM) were recorded with skin electrodes following the ulnar nerve stimulation at wrist before and after exercise. A and B, short exercise test of the left ADM in an unaffected control and in patient B5 (T309I substitution), respectively, D and E, long exercise test of the right ADM. Pre-exercise (top trace) and post-exercise recordings (below) at different times following the trial (Ex) as indicated by an arrow. Scale between 2 dots: 5 ms, 5 mV. C and F, changes in CMAP amplitude of ADM muscle following short and long exercises, respectively. The amplitude of the CMAPs expressed as a percentage of its pre-exercise value is plotted against the time elapsed after the exercise trial (noted by pairs of vertical bars). ○: changes of CMAP amplitude in 30 control subjects (mean ± s.e.m.). • and ▪: changes of CMAP amplitude in patients A6 (C154F substitution) and B5 (T309I substitution), respectively.

Figure 3. Loss-of-function mutations in AS

A, whole-cell currents in COS-7 cells after transfection with KCNJ2-WT, KCNJ2-C154F and KCNJ2-T309I. Cells were held at −80 mV, then depolarized to various test potentials (−100 to +50 mV) for 200 ms duration in 10 mV increments. Both mutants failed to produce any current when expressed alone. Each value represents the average of 8 cells for KCNJ2-WT, KCNJ2-C154F and KCNJ2-T309I, respectively. Values are mean ± s.e.m. B, current density of KCNJ2-DsRed, KCNJ2C154F-EGFP, KCNJ2T309I-EGFP, KCNJ2-DsRed/KCNJ2-EGFP, KCNJ2-DsRed/KCNJ2C154F-EGFP, and KCNJ2-DsRed/KCNJ2T309I-EGFP. COS-7 cells were held at −80 mV and depolarized for 500 ms to −100 mV. Each value represents mean ± s.e.m. for n cells tested. Inset, examples of current recordings for KCNJ2 WT and KCNJ2-T309I.

Figure 4. Cellular localization of AS mutant channels

A, confocal microscopy image of KCNJ2 clones expressed in HEK293 cells. Cells were transiently transfected with either of the clones and images were taken 48 h later. Each KCNJ2 clone was fused to either DsRed (WT) or EGFP. Both EGFP and DsRed show a cytoplasmic diffuse fluorescent pattern. However, KCNJ2-EGFP, KCNJ2-DsRed, KCNJ2C154F-EGFP, and KCNJ2T309I-EGFP are localized to the plasma membrane. Magnification 100 × using a Leica microscope. B, cellular co-localization of WT and AS mutant channels. HEK293 cells were transfected with pBud plasmid containing KCNJ2-DsRed under CMV promoter and KCNJ2-EGFP, KCNJ2C154F-EGFP or KCNJ2T309I-EGFP under the EF-1α promoter. Images were taken as described above. Yellow colour indicates overlapping localization.

Figure 5. Expression of AS mutations in a muscle cell line

EGFP and EGFP-tagged Kir2.1 constructs were transiently introduced into a mouse muscle C2C12 cell line. Confocal microscopy images were taken 24 h or 48 h post-transfection with a 100 × magnification lens.