Background
Long QT syndrome (LQTS) is a cardiac disease characterized by an abnormal repolarization in cardiac cells causing a prolonged QT interval on the electrocardiogram (ECG). Ventricular tachycardia known as “torsades de pointes” is one of the main features of LQTS and leads to seizures, syncopes, cardiac arrest and sudden death [
1].
LQTS has two inherited forms. The extensive form is the autosomal dominant condition referred to as Romano-Ward syndrome (RWS), which most commonly presents itself in the form of cardiac associated symptoms [
2,
3]. The less common form is Jervell and Lange-Nielsen Syndrome (JLNS), inherited as a recessive manner and presents a severe cardiac phenotype in addition to profound bilateral hearing-loss [
4]. Most of the causative genes identified for LQTSs encode the ion channels that are involved in the regulation of the cardiac action potential period, these genes include the Na
+and K
+ channel genes;
SCN5A, HERG,
KCNE2,
KCNQ1 and
KCNE1. JLNS usually results from recessive bi-allelic mutations leading loss of function in either the
KCNQ1 or
KCNE1 genes [
5,
6].
These voltage-gated channels characteristically contain four subunits that surround a pore which is centrally localized. Each subunit is composed of six transmembrane segments (S1–S6) including the NH2 and COOH termini which is located on the intracellular membrane. The voltage sensor of the channel is S4 segment. A P-loop domain connects the S5-S6 segment to create the pore zone. Also, the transmembrane domains of S2–S3 and S4–S5 are connected by Cytoplasmic loops (C-loops) [
7]. It was reported that missense mutations and mutations localized at the transmembrane part were related to high risk for cardiolocigal events than the mutations localizing at the C-terminal region. Furthermore, it was suggested that C-loops influence the regulation of adrenergic channel by Protein Kinase A (PKA) [
8].
Over the past decade some isolated cases of JLNS presenting with only cardiac manifestations without hearing loss (so called AR LQT1) has been reported. It has been suggested that hearing preservation in these KCNQ1 homozygotes or compound heterozygotes is derived from the presence of mildly effective mutations that can not totally remove the Kv7.1 (KvLQT1) function and maintain the normal K
+ cycle with the occurence of sufficient residual IKs current in the inner ear, however the normal electrical activity of the heart is disrupted [
9].
Long QT syndromes may often be misdiagnosed as epilepsy due to the associated seizures that occur [
10,
11]. In the current report, we identify both clinical and genetic findings from three proband’s with JLNS, and discuss the phenotype-genotype correlation with each case.
Discussion & conclusion
Long QT syndrome may often be a fatal disease, and symptomatic cases who do not receive treatment have a elevated mortality ratio, 21% within a year from the initial attack [
13]. JLNS is featured by congenital bilateral sensorineural deafness and long QTc distance that is generally higher than 500 ms [
14].
KCNQ1 and
KCNE1 gene mutations have been demonstrated to be associated with JLNS [
15].
KCNQ1 encodes a subunit of voltage-gated K
+ channel protein which have ainteraction with the β subunit encoded by the
KCNE1 gene to generate a cardiac K
+ channel. The molecular genetic analysis JLNS families demonstrated that the homozygous/compound heterozygous mutations in the
KCNQ1 gene are the main cause of the syndrome.
The c.728G > A (p. Arg243His) mutation of case 1 (II-1 of family I)-was reported previously in two non-consanguineous French families in a compound heterozygous manner who displayed a typical JLNS and in two consanguineous Turkish JLNS families [
12,
15‐
17]. This missense mutation caused the change of the 243th basic charged polar arginine to histidine that has similar physicochemical properties. It is positioned in S4 domain of the protein (Fig.
2e). Mutations striking to S4/S5 domains are predicted to disrupt the gating function of the ion channels and alter
KCNQ1 interactions with the minK sub-units [
18]. Chouabe et al. demonstrated that this mutation is related to a considerably shorter prolongation of the QTc distance, though it may cause sudden deaths due to full loss of function [
17]. It seems clearly that the cardiac functions is affected by the
KCNQ1 mutations, based on either the voltage shift and loss of function in the ion channel. The voltage alteration mutations may also end up with inconstant phenotypes. The homozygous carriers of p.Arg243His mutation have the the maximum shift of the voltage current in the cardiac cells, however the heterozygous carriers are asymptomatic with no major longer OTc interval. The cardiac symptoms of case 1 (II-1 of family I) became progressively severe in an age dependent fashion and ICD implantation and cervical sympathectomy was performed due to recurrent ventricular tachycardia attacks. Due to the progressive and malignant disease course, further targeted next generation sequencing of cardiac panel containing 68 functional genes were examined and the patient had additional heterozygous alterations in the
RYR2 and
NKX2–5 genes. The
RYR2 gene had been previously described to be involved in catecholaminergic polymorphic ventricular tachycardia type 1 (CPVT1) and arrhythmogenic right ventricular dysplasia type 2 (ARVD2) (OMIM#
600996 and
604772) but the patient phenotype was consistent with neither CPVT1 nor ARVD2. Although these two genes were not associated with LQTS or JLNS alone, they could aggravate the clinical phenotype of the patient additive to the homozygous
KCNQ1 mutation.
The c.477 + 1G > A splice site and the c.520C > T (p. Arg174Cys) missense mutations detected in family II were both described previously in North America and Europe, [
19,
20]. As far as we know, this is the first report of compound heterozygote Turkish family of JLNS carrying these mutations. The splice site mutation detected in case2 (III-5 of family II) is located in intron 2. Recently, Zehelein J. and his colleagues reported that this mutation alters the evolutionally conserved splice donor site resulting to splice error. Predicted exon skipping introduces a premature translational termination resulting in truncated KCNQ1 potassium channel subunits [
21].
In a functional study examining a p. Arg174Cys mutation demonstrated that PKA, protein kinase C (PKC) and Phosphatidylinositol-4, 5-bisphosphate (PIP2) regulation decreased [
22]. This study also revealed that a decline in the activation of either PKA or PKC may promote to the disease phenotype by reducing the activation of normal ion channel via receptor stimulus.
The c.1097G > A (p. Arg366Gln) mutation detected in family III was first reported in RWS and recently in one Turkish JLNS family [
23‐
25]. The mutation was found at the C-terminal of
KCNQ1. It was reported that the mutant p. Arg366Gln channel presents a weaker interference with PIP2 in contrast with the wild type channel. An increasement in the channel activation was observed for the two C-terminal
KCNQ1 mutants of p. Arg366Gln and p. Arg555Cys. These C terminal mutants might be protective and reduce the severity of the mutation [
23]. Cases with pathogenic variants at the C terminal part of the
KCNQ1 subunit was found to be less prone to cardiological events than cases with pathogenic variants in the transmembrane parts [
26]. In family III, all affected subjects had a milder QTc prolongation and milder cardiac phenotype. Interestingly one of the subjects (IV: 4 of family III) had no hearing loss, although she had the same homozygous mutation as the index case (IV-5 of family III) and the other affected cousin (IV-6 of family III). It remains largely elusive that what determines this great variability in the severity of disorder, where even a relative carrying the same pathogenic variant demonstrate significant differences in phenotypic/clinical manifestations. Giudicessi JR et al. suggested that a great majority of unrelated KCNQ1 homozygotes and compound heterozygotes with a possible autosomal recessive model of inheritance (64%, 7 out of 11) show no sensorineural deafness. Their study showed that the full breakdown of KvLQT1 mediated K
+ secretion in the cochlea linked to the biallelic inheritance of 2 truncating/haploid insufficient pathogenic variants (complete loss-of-function) is the major mechanism that affects the homeostasis of inner ear fluid, causing the downfall of endolymph in the inner ear and sensorineural hearing loss in JLNS cases and murine models of JLNS [
9]. Missense variations with dominant-negative effects on the tetrameric KCNQ1 channel results in an altered molecular function but not complete loss of function. Recent electrophysiological studies showed that homozygous and compound heterozygous pathogenic variants in either
KCNQ1 or
KCNE1 gene involved in JNLS resulted in a broad range of spectrum of effects from loss of function to dominant negative behavior [
27]. Recently Bhuiyan ZA et al., reported a homozygous novel splice site c.387-5 T > A mutation in the first intron of the
KCNQ1 gene, in the two Saudi Arabian families. RNA analysis demonsrated that c.387-5 T > A mutation leads incomplete transcriptional error of the
KCNQ1 gene, leaving 10% of the normal transcript intact, which restores the auditory function. An other intronic pathogenic variant leading to incomplete exon 2 skipping in
KCNQ1 saves hearing function in JLNS [
28]. Apparently, as in our family III, the homozygous c.1097G > A (p. Arg366Gln) mutation does not cause a complete loss of KvLQT1 function in endolymph of the inner ear in all of the homozygous carriers. The amount of residual KvLQT1 function necessary for auditory preservation in JLNS is not known. The other mechanism proposed to explain the development of intact hearing in this case could be the mosaicism. It can be speculated that JLNS mutation mosaicism results in functional channels in the marginal cells of the inner ear with ordinary membrane transmission, so the phenotypic outcomes are not monitored in the inner ear. Although the possibility seemed unlikely in familial cases, it still cannot be ruled out. Another speculation is that there may be other potential genetic factors in normal functioning of cochlear hair cells leading to intact hearing.
Therefore, whole genome exome sequencing may assist with the investigation of the underlying genetic causes of normal hearing. Furthermore, to explain this variability in the clinical phenotype; expression studies of wild-type and mutant type
KCNQ1 using cardiac muscle cells [
29] and cochlear hair cells obtained from patient specific induced pluripotent stem cells need to be performed.
Children with JLNS are frequently misdiagnosed with epilepsy at the beginning of therapy, causing both delayed care for JLNS and improper anti-epileptic medication [
10]. In our families; Case 1 (II-1 of family I) and the cousin of case 3 (IV-6 of family III) were also diagnosed as refractory epilepsy and treated with anti-epileptic drugs in their previous hospitalizations. In the current study the two cases underlined the importance of careful examination of the patient having seizures and primarily to exclude the cardiac cause for the differential diagnosis. This would not only save the lives of the patients but will also assist with proper treatments and specific genetic counseling sessions for other high-risk individuals in their families.
However, a group of these cases are presumably to have a neurocardiac syndrome through ion channels, as recent evidence demonstrated that epileptiform activity determined with electroencephalography (EEG) in 15% of LQTS cases who showed seizures or seizure like crisis [
30]. Therefore, because of this possible coexistence of both conditions we routinely began to do detailed cardiologic and neurologic screening of patients at the University of Uludag.
Management of JLNS consists of β-blockers, ICD and/or left cardiac sympathetic denervation (LCSD) for those with β-blocker resistive symptoms for the cardiological events, and cochlear implants for the recovery of hearing loss [
4]. Beta-blocker therapy is the fundamental medication for JLNS; potential ICD and/or LCSD for those with β-blocker resistive symptoms, insufficiency to take β-blockers, and/or cardiac arrest history [
31]. Beta-blockers significantly decrease the sudden death risk.
Beta blocker theraphy is clinically indicated in all symptom-free cases meeting diagnostic criteria, consisting of those who possess a mutation on genetic screening and a normal QTc interval [
4]. Usually, implantation of ICD is not indicated for asymptomatic cases of JLNS. Prophylactic ICD treatment might be conceived for cases with JLNS who show no symptoms but suspected to be at very high-risk (e.g., those with at least 2 pathogenic mutations on genetic testing and/or family history of young sudden unexplained death) [
32].
Therefore, we advise that all carriers of an identified pathogenic variant in either the KCNQ1 or the KCNE1 gene be thoroughly examined at a minimum by favour of ECG and clinical family history, before initiating prophylactic therapy like beta-blockers in these cases. Cardiac screening is recommended in more complex cases. Because the mutation carriers of JLNS might have a slight deafness sign, a hearing test should also be considered. Medications and conditions that stimulate the prolongation on the QT interval such as dynamic sport activities, horror films, loud environments are to be avoided. Training and educating of all family members for cardiac arrest resuscitation might be life saving.
After all, here we present three families of JLNS who display long QT and deafness.
This genotype-phenotype study suggests that the patients with trans-membrane versus C-terminus mutations and these with mutations having dominant negative versus haploinsufficiency demonstrate the variable degree of the failure of ion channels as well as the clinical course of the disorder.
It is well known that protein function is affected by the C terminal mutations in KCNQ1 gene far less degree. Also, it is precise that there is no absolute lack of K+ inflow into the endolymph of the inner ear in our JLNS patient IV: 4 of family III with intact hearing who is carrying a homozygous C-terminal c.1097G > A (p. Arg366Gln) missense mutation. Our study suggests a milder effect of the mutation on the clinical phenotype for family III as well as hearing preservation of the IV: 4 patient of family III. The degree of the remaining Kv7.1 function necessary for auditory preservation in KCNQ1 homozygosity and/or compound heterozygosity is not known.
It was also emphasized that broad targeted cardiac panels or whole exome analysis may be useful to predict the outcome especially in patients with unexplained phenotype-genotype correlation according to the mutation detected and the progressive and malignant course.
Our research group screened approximately 200 cases with our custom made 68 gene cardiac panel diagnosed as hereditary arrhythmias like recessive JLNS and/or LQT syndromes, we have seen that this disease group is highly heterogeneous in terms of genetics and caused by mutations of more than one variant and/or multiple variants [
33]. Due to the genetic complexity of hereditary arrhythmias like LQTS, JLNS, it currently makes it necessary to screen the other cardiac genes in addition to the most common genes related to the complex phenotypes before the molecular-genetic workup is completed. Families, clinicians and researchers together will have a essential role in the advancement of biomedicine towards personalized treatments.