Further evidence for shared genetic susceptibility in both sporadic and Thyrotoxic periodic paralysis

Periodic paralysis is a disorder of skeletal muscle that presents with transient weakness lasting hours to a day or more, followed by spontaneous recovery [1]. Several variants of periodic paralysis have been clinically delineated and have correspondingly varied pathogenic mechanisms. Familial periodic paralysis, also called primary periodic paralysis, is caused by mutation of an ion channel gene (SCN4A, CACNA1S, or KCNJ2) that renders the fiber susceptible to episodes of sustained depolarization and thereby inactivates sodium channels, reduces fiber excitability, and causes weakness. Familial periodic paralysis is inherited as a dominant Mendelian trait, although symptomatic expression may be reduced for some individuals. Secondary periodic paralysis may occur with severe derangements of ion homeostasis (e.g. hypokalemia in aldosteronism or diuretic overuse). Somewhere between primary and secondary periodic paralysis are forms with established genetic risk but do not follow a Mendelian monogenic inheritance pattern. This latter group includes thyrotoxic periodic paralysis (TPP, the most commonly occurring variant of periodic paralysis) and sporadic periodic paralysis (SPP). In this issue of JNS, Nakaza and colleagues report that nine previously identified single nucleotide variants (SNVs) variably associated with TPP and/or SPP [2–4] were all confirmed to show susceptibility to SPP in a Japanese cohort of 43 individuals [10]. These data add to the growing body of evidence that SNVs on chromosome 17 downstream of KCNJ2 confer risk for periodic paralysis, and imply that SPP and TPP may share a common genetic etiology.

Sporadic periodic paralysis has been somewhat of an enigma. The first challenge is how to define the syndrome. Generally agreed upon criteria [2,3] include periodic paralysis in the absence of a family history of muscle disease, no secondary disturbance of serum electrolytes, normal thyroid studies, and no pathogenic variants detected in exon screening of associated ion channel genes (SCN4A, CACNA1S, or KCNJ2). Nakaza et al. [10] added the additional more stringent criteria of two or more episodes of weakness, or one episode plus an abnormal decrement of the compound muscle action potential (CMAP) on the long exercise test, and the absence of EKG changes consistent with the Andersen-Tawil Syndrome ATS). Exclusion of ATS by EKG is important because only 2/3 of patients who meet clinical criteria of the disorder will have a mutation identified in KCNJ2 coding for the Kir2.1 inward rectifying potassium channel [5].

A possible connection between SPP and TPP comes from genome-wide association studies (GWAS). While mutations of KCNJ18 encoding Kir2.6 have been associated with TPP [6], this is a rare cause of the disorder (< 2%) in Chinese and Thai populations. Consequently, GWAS was performed to identify additional genetic risk factors for TPP, and the first identified SNV (rs623011) was reported in a Thai population [7]. Curiously, rs623011 is about 75 kb downstream of KCNJ2 on 17q24.3 (i.e. the gene associated with ATS), which is quite distant from the TPP susceptibility gene KCNJ18 on 17p11.2. A subsequent study in a Chinese cohort confirmed rs623011 in association with TPP and also SPP [3]. This joint susceptibility to both syndromes was further expanded in a Taiwanese cohort where an association was determined for five additional SNVs, all near rs623011 on 17q [4]. Moreover one SNV, rs312736, was mapped to a lincRNA, CTD-2378E21.1, with the risk allele (A) in exon 3. Two additional SNVs (rs723498 and rs312707) in this same study were associated with TPP but did not confer risk for SPP [4]. In this first study of a Japanese cohort, Nakaza et al. [10] report that all nine previously reported SNVs were associated with susceptibility to SPP. Importantly, this included the “TPP-only” variants rs723498, rs312707 and a known TPP variant that was not previously assessed in SPP (rs312691). This study thereby shows conservation of these SNVs for SPP in Japan, as in other East Asian populations, and also strengthens the possibility that SPP and some forms of TPP share a common genetic basis.

Even with these genetic insights, the mechanism by which contractility is impaired during an episode of SPP remains to be established. For all types of familial periodic paralysis (hypokalemic, hyperkalemic, and ATS), the ictal weakness is caused by anomalous sustained depolarization of the resting potential, which renders the fiber chronically refractory and inexcitable [1]. A similar final common pathway is likely for episodes of SPP and TPP, which is consistent with the decrease in fiber excitability demonstrated by the CMAP exercise test. However, anomalous depolarization of the resting potential has not yet been tested for TPP or SPP. One attractive hypothesis for disease mechanism is impairment of the inward rectifying potassium conductance of skeletal muscle. Transcription of KCNJ18 is thyroxine responsive and the Kir2.6 subunit promotes retention of Kir2.1 (the major contributor for Kir conductance) in the endoplasmic reticulum [8]. Similarly, ATS is caused by loss of function mutations of Kir2.1 [9]. In other words, evidence of reduced Kir2.1 is an established mechanism for episodic weakness in both ATS and KCNJ18-associated TPP. The GWAS discovery of a risk allele (A) for the lincRNA CTD-2378E21.1 in SPP/TPP provided an opportunity to test for a connection to Kir2.1 expression [4]. Overexpression of the normal allele for CTD-2378E21.1 in a skeletal muscle line (C2C12) significantly decreased Kir2.1 expression, whereas the risk allele (A) produced no change. While this experiment demonstrates impaired regulation of Kir2.1 by the risk allele (A), the change is in the opposite direction (i.e. a gain of function caused by loss of inhibition) from the canonical loss of Kir2.1 associated with periodic paralysis in ATS or TPP [8,9]. Theoretically, increased expression of Kir2.1 at the membrane could reduce fiber excitability by stabilizing a normally hyperpolarized resting potential. Indeed, a much larger stimulus current was required to elicit an action potential after over-expression of Kir2.1 in mouse flexor digitorum brevis muscle (unpublished data, Prof. DiFranco, UCLA). More studies are needed, however, to determine if these changes are detectable in fibers from SPP and 17q-linked TPP patients or some other Kir-dependent mechanism is responsible.

Abbreviations:

ATS

Andersen-Tawil Syndrome

CMAP

compound muscle action potential

GWAS

genome-wide association study

SNV

single nucleotide variant

SPP

sporadic periodic paralysis

TPP

thyrotoxic periodic paralysis