This article is mentioned in:

Introduction

Previous genomic studies identified several common genetic variants that could play a role in large-vessel ischemic stroke (LVIS) (1). Precise type of their effect on brain ischemia remains unclear, and it is assumed that common genetic variants could explain between 39–66% of variation in ischemic stroke incidence (2). It has also been postulated that the accumulated effect of the remaining uncommon or rare damaging variants (called genetic burden) might explain a significant portion of the genetic predisposition to many common diseases or phenotypes (3,4). A small number of re-sequencing reports of European patients with LVIS were published so far. These studies have demonstrated that infrequent coding variants in numerous genes might be linked to stroke (5–8).

Platelets have an important role in the pathogenesis of LVIS based on their activation and adherence to the endothelium within cerebral arteries, as well as progression of thrombus (9,10). Most research strategies to date revealed the effect of genetic variation on reactivity of platelets and were obtained by analyzing common variants within candidate genes and/or genome-wide association studies (GWAS), followed by in vitro studies to assess platelet functions (11). Overall, previous studies on the genetic background of platelet reactivity indicated that many different genes contribute to platelet function. Thus far, the potential contribution of genetic variants within genes encoding proteins essential to thrombus formation in LVIS have been analyzed in a small number of studies and the majority of them focused on the common single nucleotide polymorphisms (SNPs) within a few genes associated with glycoproteins on the platelet surface (11). The results of these studies suggested that additional loci associated with platelet functions are yet to be found and that the known loci may contain high effect rare risk variants that have thus far gone undetected by GWAS.

Rare coding variants appear to be restricted to small populations and that was shown in only one previous study, which concentrated on the re-sequencing of the common variants in the platelet genes associated with membrane function (12). In addition, another recent investigation revealed that the rare variants in receptors commonly associated with platelet functions (e.g. purinergic receptors) could be associated with the occurrence of LVIS in the Polish population (13).

Thus, the objective of our current study was to explain the contribution played by the another set of uncommon and damaging genetic variants within selected genes associated with changes in platelet functions in LVIS. We have chosen 54 genes associated with less known or investigated biochemical processes associated with platelet functions as the focus for the re-sequencing study (14–21).

Materials and methods

Clinical material

The study was permitted by both local ethics committee of the Institute of Psychiatry and Neurology, Warsaw, Poland, and Warsaw Medical University, Warsaw, Poland. The study conduct conformed to the most recent version of the Declaration of Helsinki. All participants in the study signed the informed consent. Full description of the study cohort, including the inclusion and exclusion criteria were published previously (12,13,22–24). Based on the Trial of Org 10172 in Acute Stroke Treatment (TOAST) classification we included: i) All patients classified as having ischemic stroke due to large-vessel atherosclerosis and ii) a subset of patients classified as having ischemic stroke of unknown etiology, provided they had at least 50% stenosis of the carotid artery ipsilateral to the infarct side and no evidence or no history of atrial fibrillation. Patient with the history of hemorrhagic or embolic stroke were excluded from the study. As a control group we used samples from age- and sex-matched 500 patients without history of stroke coming from the same geographical area as patients with LVIS (13) and collected for unrelated studies performed at Warsaw Medical University in Poland. Both cohorts of patients (study and control) were white Caucasian of Polish ethnicity and originated from central Poland. DNA extraction from collected blood samples was done as described before (9).

Pooled sequencing

The list of 54 re-sequenced genes is shown in Table I. These targets contain 846 exons and 10 adjoining bases beyond each exon on both sides. The genetic loci were selected using the human database (H. sapiens, hg19, GRCh37, February 2009). Pooled targeted enrichment of DNA, from LVIS patients (five polls with 100 subjects per pool) and 500 age-, sex-matched control patients, without stroke history (five pools with 100 subjects per pool), was done as described previously. Further explanation of re-sequencing and analysis of data is provided in the Supplementary material (Tables SI–SIV).

Table I. List of 54 platelet genes with exons sequenced in the present study.

Table I.

List of 54 platelet genes with exons sequenced in the present study.

Author, year	Gene	Protein encoded	Chromosome and regions	(Ref.)
Janicki et al, 2017	FCER1G	Fc fragment of IgE, high affinity I, receptor for; γ polypeptide	1q23.3region	(13)
Janicki et al, 2017	VAV3	vav 3 guanine nucleotide exchange factor	1p13.3	(13)
Jones et al, 2009	RAF1	v-raf-1 murine leukemia viral oncogene homolog 1	3p25.1	(14)
Jones et al, 2009	MAPK14	Mitogen-activated protein kinase 14	6p21.31	(14)
Jones et al, 2009	JAK2	Janus kinase 2	9p24.1	(14)
Jones et al, 2009	MAP2K4	Mitogen-activated protein kinase kinase 4 17. p12	17p12	(14)
Jones et al, 2009	AKT2	v-akt murine thymoma viral oncogene homolog 2	19q13.1-q13.2	(14)
Jones et al, 2009	MAP2K2	Mitogen-activated protein kinase kinase 2	19p13.3	(14)
Jones et al, 2009	GNAZ	Guanine nucleotide binding protein (G protein), α z polypeptide	22q11.22	(14)
Jones et al, 2009; Goodall et al, 2010	TRIM27	Tripartite motif containing 27	6p22.1	(14,15)
Goodall et al, 2010	LRRFIP1	Leucine rich repeat (in FLII) interacting protein 1	2q37.3	(15)
Goodall et al, 2010	COMMD7	COMM domain containing 7	20q11.21	(15)
Postula et al, 2013	RGS7	Regulator of G-protein signaling 7	1q23.1	(16)
Guerrero et al, 2011	LPAR1	Lysophosphatidic acid receptor 1	9q31.3	(17)
Guerrero et al, 2011	MYO5B	Myosin VB	18q21.1	(17)
Mathias et al, 2010	LDHAL6A	Lactate dehydrogenase A-like 6A	11p15.1	(18)
Mathias et al, 2010	ANKS1B	Ankyrin repeat and sterile α motif domain containing 1B	12q23.1	(18)
Johnson et al, 2010	PIK3CG	Phosphoinositide-3-kinase, catalytic, γ polypeptide	7q22.3	(19)
Johnson et al, 2010	SHH	Sonic hedgehog homolog	7q36.3	(19)
Johnson et al, 2010	JMJD1C	Jumonji domain containing 1C	10q21.2	(19)
Johnson et al, 2010	MRVI1	Murine retrovirus integration site 1 homolog	11p15.4	(19)
Johnson et al, 2010; Johnson, 2011	RGS18	Regulator of G-protein signaling 18	1q31.2	(19,20)
Johnson et al, 2010; Johnson, 2011	ST3GAL3	ST3 β-galactoside α-2,3-sialyltransferase 3	1p34.1	(19,20)
Johnson et al, 2010; Johnson, 2011	UGT1A10	UDP glucuronosyltransferase 1 family, polypeptide A10	2q37.1	(19,20)
Johnson et al, 2010; Johnson, 2011	NUP210	Nucleoporin 210 kDa	3p25.1	(19,20)
Johnson et al, 2010; Johnson, 2011	RAPGEF2	Rap guanine nucleotide exchange factor (GEF) 2	4q32.1	(19,20)
Johnson et al, 2010; Johnson, 2011	ADAMTS2	ADAM metallopeptidase with thrombospondin type 1 motif, 2	5q35.3	(19,20)
Johnson et al, 2010; Johnson, 2011	FBXL7	F-box and leucine-rich repeat protein 7	5p15.1	(19,20)
Johnson et al, 2010; Johnson, 2011	KLHL31	Kelch-like family member 31	6p12.1	(19,20)
Johnson et al, 2010; Johnson, 2011	GMDS	GDP-mannose 4,6-dehydtase	6p25.3	(19,20)
Johnson et al, 2010; Johnson, 2011	WBSCR17	Williams-Beuren syndrome chromosome region 17	7q11.22	(19,20)
Johnson et al, 2010; Johnson, 2011	ATP6V1F	ATPase, H+ transporting, lysosomal 14 kDa, V1 subunit F	7q32.1	(19,20)
Johnson et al, 2010; Johnson, 2011	SGCZ	Sarcoglycan, ζ	8p22	(19,20)
Johnson et al, 2010; Johnson, 2011	STMN4	Stathmin-like 4	8p21.2	(19,20)
Johnson et al, 2010; Johnson, 2011	PSKH2	Protein serine kinase H2	8q21.3	(19,20)
Johnson et al, 2010; Johnson, 2011	PIP5K1B	Phosphatidylinositol-4-phosphate 5-kinase, type I, β	9q21.11	(19,20)
Johnson et al, 2010; Johnson, 2011	PTPRD	Protein tyrosine phosphatase, receptor type, D	9p24.1	(19,20)
Johnson et al, 2010; Johnson, 2011	MIPOL1	Mirror-image polydactyly 1	14q13.3-q21.1	(19,20)
Johnson et al, 2010; Johnson, 2011	SVIL	Supervillin	10p11.23	(19,20)
Johnson et al, 2010; Johnson, 2011	CUBN	Cubilin (intrinsic factor-cobalamin receptor)	10p13	(19,20)
Johnson et al, 2010; Johnson, 2011	ST3GAL4	ST3 β-galactoside α-2,3-sialyltransferase 4	11q24.2	(19,20)
Johnson, 2010 Johnson, 2011	KCNQ1	Potassium voltage-gated channel, KQT-like subfamily, member 1	11p15.5-p15.4	(19,20)
Johnson et al, 2010; Johnson, 2011	HSD17B6	Hydroxysteroid (17-β) dehydrogenase 6	12q13.3	(19,20)
Johnson et al, 2010; Johnson, 2011	RAP1B	RAP1B, member of RAS oncogene family	12q15	(19,20)
Johnson et al, 2010; Johnson, 2011	PTPN11	Protein tyrosine phosphatase, non-receptor type 11	12q24	(19,20)
Johnson et al, 2010; Johnson, 2011	THSD4	Thrombospondin, type I, domain containing 4	15q23	(19,20)
Johnson et al, 2010; Johnson, 2011	TAOK1	TAO kinase 1	17q11.2	(19,20)
Johnson et al, 2010; Johnson, 2011	SETBP1	SET binding protein 1	18q12.3	(19,20)
Johnson et al, 2010; Johnson, 2011	KIAA0802	SOGA family member 2	18p11.22	(19,20)
Johnson et al, 2010; Johnson, 2011	CTCFL	CCCTC-binding factor (zinc finger protein)-like	20q13.31	(19,20)
Johnson et al, 2010; Johnson, 2011	PCK1	Phosphoenolpyruvate carboxykinase 1	20q13.31	(19,20)
Johnson et al, 2010; Johnson, 2011	PRNP	Prion protein	20p 13	(19,20)
Shiffman et al, 2006***	VAMP8	Vesicle-associated membrane protein 8	2p12-p11.2	(21)
Lee et al, 2014	GLIS3	GLIS family zinc finger 3	9p24.2	(26)

Verification of selected variants by individual genotyping

Individual genotyping for selected markers in individual DNA samples was performed using a custom Sequenom iPLEX assay in conjunction with the Mass ARRAY platform (Sequenom Inc., La Jolla, CA, USA). Panels of SNP markers were designed using Sequenom Assay Design 3.2 software (Sequenom Inc.), in a similar fashion to the previously described methodology from our laboratory (9,16,17).

Statistical analysis

A cumulative minor allele frequency (cMAF) was utilized to show the allelic frequency of the investigated variants, which encompasses all rare damaging variants individually genotyped in the investigated cohorts, as well as within each of the analyzed loci. Pearson Chi-square test was used in order to analyze differences in cMAF for all individually genotyped variants between the study and control cohorts (VassarStats: Website for Statistical Computation on http://vassarstats.net/). The pooled minor allele test (CMAT) (10,000 × permutations) was used for comparison of all variants within investigated loci to estimate the statistical significance of the observed differences in the accumulation of variants. CMAT is a pooling method proposed by Zawistowski et al (25) and works by comparing weighted minor-allele counts (for cases and controls) against the weighted major-allele counts (for cases and controls). Although the CMAT test statistic is based on a chi-square statistic, it does not follow a known distribution and its significance has to be determined by a permutation procedure. The calculations of CMAT were performed using AssotesR package (0.1-10) from CRAN repository (cran.r-project.org/package=AssotesteR) and written by Gaston Sanchez (gastonsanchez.com/) as documented at www.rdocumentation.org/packages/AssotesteR/versions/0.1-10. The significance threshold was adjusted to the number of re-sequenced loci, when needed (13,25).

Power and sample size considerations

For the power calculations, instead of using individual rare variants, we decided to use predicted cMAF for all deleterious rare variants in the sequenced loci. We have followed a self-sufficient, closed-form, maximum-likelihood estimator for allele frequencies that accounts for errors associated with sequencing, and a likelihood-ratio test statistic that provides a simple means for evaluating the null hypothesis of monomorphism (13,26,27). Unbiased estimates of allele frequencies 10/N (where N is the number of individuals sampled) appear to be achievable, and near-certain identification of a single nucleotide polymorphism (SNP) requires a cMAF of at least 0.01 (i.e., 10 variants per cohort). In addition, because the power to detect significant allele-frequency differences between the two populations is limited, we set both the number of sampled individuals (500 in the cohort) and depth of sequencing coverage in excess of 100.

Fluorescence-based (FLIPR) functional assay for KCNQ1 variants

Chinese hamster ovaries cells (CHO) cultures stably transfected with human muscarinic type 1 receptor (M1) were purchased from cDNA Resource Center, Bloomberg, PA. The CHO-M1 cells were then transiently co-transfected with KCNQ1 cDNA constructs. Wild-type KCNQ1 cDNA constructs (in pcDNA3.1) were prepared by Watson Bio Sciences (Houston, TX, USA). The fidelity of the mutations for each variant was confirmed by Sanger sequencing. Details about the transfection techniques and cell culture conditions used in our laboratory were published before (13).

The heterologous expressed potassium KCNQ1 channels were inhibited by the M1 receptor agonist oxotremorine (OxoM, 10 nM), resulting in significant changes in membrane polarization (fully antagonized by muscarinic receptor antagonist atropine at 1 µM-not shown).

The Fluorescence Imaging Plate Reader (FLIPR) on Flex Station 3 (Molecular Devices, San Jose, CA, USA) was employed to measure the fluorescence changes. After the cells were loaded with membrane dye, the plate with the cells and plate containing OxoM (10 nM) were inserted into the plate reader. The raw fluorescence readings were converted and plotted as the change in relative fluorescence before (F0 baseline=100%) and after OxoM application F), according to the formula ((F/F0)*100%). For all experimental conditions, minimum fluorescence (Fmin) and maximum fluorescence (Fmax) were recorded in triplicates after administration. The summary data are presented as summary trend lines for WT and mutants, as well as averages (with standard deviations) for Fmax and Fmin and for each variant and WT cells used in FLIPR experiments. The final data represent all measurements done in triplicates and in three independent experiments (cell populations). Data and statistical analysis were performed using analysis of variance, followed by the Student's t-test. P<0.05 was considered to indicate a statistically significant difference.

Results

Study design and sequencing coverage

The design of the study is presented in Fig. 1. The enrichment of the target loci resulted in coverage of 99.6% and produced 36.1 (22.7–45.9 range) million reads which is 5.3 (3–7 range) Gbp per re-sequenced sample. It relates to an average coverage of 12,000× per pool and an average coverage of 120× per patient sample (range 21–369).

Figure 1. Study-flow diagram. TOAST, Trial of Org 10172 in Acute Stroke Treatment; hx, history; KCNQ1, potassium voltage-gated channel subfamily Q member 1; FLIPR, fluorescence imaging plate reader.

Selection of rare variants (based on MAF<1%)

The step-wise analysis of the observed variants is presented in Fig. 2. In total, 1018 unique variants, irrespective of MAF and with adequate quality were observed in both investigated cohorts. The complete list of all observed variants is provided in the Supplementary materials, Tables SII for all non-coding variants and SIII for all non-synonymous variants. Seventy two percent of all variants were previously listed in the available databases, and 28% were new. Out of all observed variants, 327 (32.1%) were located in the coding segments of the sequenced genes, and the remainder was located in untranslated regions. Out of all coding variants, 120 variants (Supplementary material, Table SIV-list of all rare non-synonymous variants) were selected based on MAF <1% (i.e. rare variants) (28), which consisted of 52 known (by dbSNP149 November 2016) and 68 unlisted, novel variants.

Figure 2. Number and characteristics of variants identified at each stage of the step-wise variant analysis. MAF, minor allele frequency.

Verification of selected variants by individual genotyping

In total, 28 SNVs with the highest predicted damaging score calculated by Combined Annotation Dependent Depletion (CADD) score were chosen for individual genotyping. The minimum CADD score of 10 served as a threshold for predicted deleteriousness. The individual genotyping was performed in patients from the original cohort and revealed that the identity of all 28 initially selected variants could be confirmed by individual genotyping, which indicates no sequencing errors for these variants.

The initial statistical analysis utilized Pearsons Chi-square test for the assessment of the differences in the cumulative frequency of all 28 individually genotyped variants between the investigated cohorts. There was a highly statistically significant (P=0.00045) difference (cMAF control=1.2% vs. cMAF stroke=3.6%) in cMAF for all damaging variants in the LVIS group when compared with controls.

The statistical analysis of the number of variants within the single gene loci was based on combined minor allele association test (CMAT). The region-based, Bonferroni corrected, significance threshold was P=0.00092 (nominal P=0.05/54 sequenced genes). It demonstrated a statistically significant difference (P=0.0009) between control and IS cohorts for the KCNQ1 location. The KCNQ1 exons locus contained 3 novel and 3 known rare and deleterious (CADD score range 13.22-39) coding variants (Table II).

Table II. cMAF for 28 rare and most damaging variants observed in the individually genotyped patients from control and LVIS cohorts.

Table II.

cMAF for 28 rare and most damaging variants observed in the individually genotyped patients from control and LVIS cohorts.

Gene locus	Number of carriers in control cohort, n=500	Number of carriers in LVIS cohort, n=500	P-value
ADAMT2	2	3	0.4
CUBN	0	4	0.06
GLIS3	2	0	0.1
KCNQ1	0	6	0.0009a
LDHAL6A	3	2	0.4
MAP2K2	0	2	0.2
MIPOL1	0	2	0.8
MTCL1	4	3	0.9
MYO5B	0	4	0.06
PTPRJ	0	2	0.1
SHH	0	2	0.1
SVIL	1	3	0.15
UGT1A1	0	2	0.1
Total number of carriers and cMAF for all loci	12	36	0.00045b
Odds ratio and 95% confidence intervals (in brackets)		3.07 (1.59–5.94)

a Statistical significance calculated using burden combined minor allele test for differences in cMAF within genetic locus

b statistical significance calculated using Pearson's χ² test for differences in cMAF for all loci. cMAF, cumulative minor allele frequency; LVIS, large-vessel ischemic stroke.

Functional analysis of selected rare variants within KNCQ1 gene

To evaluate whether the observed variants exert a damaging effect on KCNQ1 function, we selected 3 novel and most damaging variants from the KCNQ1 locus to examine the coupling between the heterologous expressed human M1 receptor and KCNQ1 channels in CHO-M1 cells (Table III). Fig. 3 shows the summary of changes in relative fluorescence for CHO-M1 cells expressing wild-type KCNQ1, and variants KCNQ1 c. G855T p. K285N, KCNQ1 c. G1545T p. K515N, and KCNQ1 c. 1637A p. S546X. Following oxoM application (final concentration 10 nM), the fluorescence decreased rapidly, indicating corresponding changes in cell membrane potential. We expressed the changes in fluorescence both as the differences between baseline fluorescence and its changes (in % of baseline=Fo=100%) over 250 sec after administration of oxoM (shown as trend line for all observations) and average values of minimum (Fmin) and maximum (Fmax) of relative fluorescence during the recording period (separately for WT and variants). The provided trend lines represent summary values for all observations (2–3 independent cell cultures and all measurements in triplicates of plate wells). A significant decrease in fluorescence signal for each variant was observed after administration of oxoM when compared with the KCNQ1 wild-type-expressing CHO-M1 cells. Correspondingly, the statistically significant decrease (P<0.05 ANOVA, followed by t-test) in the average Fmin and Fmax was observed for all 3 investigated variants, indicating at least partial loss-of-function characteristics for variant proteins, when compared to WT.

Figure 3. Effect of Oxo-M on relative fluorescence in CHO-M1 cells heterologously expressing KCNQ1 wild-type and KCNQ1 variant channels. Each panel shows % change (from baseline Fo=100%) of F in CHO-M1 transfected with wild-type K285N, K515N and S546X variants (where X is random amino acid substituted). Fluorescence signals were acquired before and during Oxo-M (10 nM, solid line) application. The data are presented trend lines for all the experimental points (2–3 independent experiments, triplicate measurements). For each variant and the calculated Fmin and Fmax relative fluorescence change, including standard deviation was calculated. Fmax and Fmin averages were compared with WT using analysis of variance, followed by a t-test. *P<0.05. Oxo-M, oxotremorine; CHO-M1, human muscarinic type 1 receptor; KCNQ1, potassium voltage-gated channel subfamily Q member 1; F, fluorescence; Fmin, minimum fluorescence; Fmax, maximum fluorescence; Fo, baseline fluorescence; WT, wild-type.

Table III. Rare coding variants for KCNQ1 gene observed in the study patients, as identified by exome sequencing and verified by individual genotyping.

Table III.

Rare coding variants for KCNQ1 gene observed in the study patients, as identified by exome sequencing and verified by individual genotyping.

dbSNP	MAFctrl pooled	MAFstroke pooled	MAF individual genotyping in control	MAF in individual genotyping in stroke	DNA change	Amino acid change	CADD
rs12720457	0.80%	ND	0.01%	ND	c.G1179T	p.K393N	6.560
novel	ND	0.75%	ND	0.1%	c.G4T	p.D2Y	5.53
rs199472712	ND	0.95%	ND	0.1%	c.G724T	p.D242Y	19.35
Novel	ND	0.74%	ND	0.1%	c.G855T	p.K285N	15.48
Novel	ND	0.65%	ND	0.1%	c.G1545T	p.K515N	13.22
rs199472793	ND	0.71%	ND	0.1%	c.C1597T	p.R533W	15.28
Novel	ND	0.85%	ND	0.1%	c.C1637A	p.S546X	39

[i] Novel variants marked in bold were additionally investigated in the in vitro functional test. cMAF=0.01% for MAF individual genotyping in control. cMAF=0.06% MAF in individual genotyping in stroke, P=0.0009 using combined minor allele test for comparison of coding rare variants in the KCNQ1 gene. cMAF, cumulative minor allele frequency; ND, not detected; CADD, Combined Annotation Dependent Depletion score; KCNQ1, potassium voltage-gated channel subfamily Q member 1.

Discussion

We present the results of the analysis of the genetic burden of the infrequent coding variants in 54 genes associated with platelet function and its possible association with LVIS. The assessment of MAF for the investigated uncommon coding variants established that there was a significant accumulation of those variants in the LVIS group when compared to the control group. By grouping these variants by sequenced loci instead of analyzing them individually, we were able to observe associations which could be underpowered when applied to single variants, as shown in previous studies of other traits (12,13). In particular, we found an association between the increased accumulation of rare variants in KCNQ1 locus and LVIS. It is important to note that, with the exception of 3 already known variants, the remaining 3 observed damaging variants within KCNQ1 locus were novel. It is therefore likely that at least some of the observed variants might be restricted to the Polish cohort. So far detailed genotypes in the Polish population have been rather poorly characterized in the available genomic databases. This in turn might suggest that the verification of the obtained results in the independent cohorts could be challenging. For example, it was previously demonstrated that, at least in case of rare damaging variants associated with ulcerative colitis, the rare variants observed in the Dutch population could not be replicated in a German cohort (29). Another study on inflammatory bowel disease, that included several thousands of European individuals and individuals of other ancestry, showed that although the majority of the loci with MAF>5% were shared between different ancestry groups (30), no such similarities were observed for uncommon alleles. In fact, rare variants were even more likely to be specific to a particular population, as was confirmed by a recent sequencing study (31). What is more, rare variants might differ significantly among even closely related populations (32).

We were able to observe only few (out of several hundreds) of previously listed rare coding variants in the KCNQ1 locus, which might indicate either limited power of the study or population-specific distribution of these variants. Further studies in other populations will be helpful to verify if the rare damaging variants in the KCNQ1 coding locus are indeed associated with large-vessel IS, or also with other types of stroke (small-vessel and embolic).

The KCNQ channels, members of the voltage-gated (KV7.1) K+-selective channel subfamily, play a major role in K+ ion transport. For instance, previous studies have shown that both KV channels KCNA3 and G protein-gated inwardly rectifying K+ channels (GIRK) regulate platelet activation (33,34). The presence of Kv7.1 channel in blood cells, including platelets, suggests that they play a role in agonist-mediated regulation of platelet-driven thrombotic pathways that is crucial to hemostasis during IS (35,36).

Our in vitro results indicate that oxoM-mediated changes in the membrane potential of cells expressing M1 receptors and KCNQ1 channel variants were attenuated (loss-of-function characteristics) when compared to cells expressing the wild-type receptors. These findings suggest that the signaling might be diminished in cells expressing the mutant KCNQ1 channels. The coding variants in KCNQ1 were previously evaluated in different cardiovascular diseases and DM (37–41). It has been also reported that obesity along with IS may modify methylation of KCNQ1 gene and plasma KCNQ1 protein concentration (38,39). The coding variants in KCNQ1 have been reported to lead to congenital long QT syndrome (42), an autosomal dominant disorder. Because of the demonstrated deleterious properties of the investigated variants, it should be also considered that the stroke patients in the study may have suffered from an ischemic stroke secondary to a congenital disorder. Whether KCNQ1 variants are disease-associated with ischemic stroke (possibly via platelet function) or disease-causing for congenital QT syndrome (with reportedly higher incidence of ischemic stroke is one of the questions which should be clarified in the future investigations (41).

Moreover, in one of the largest to date GWAS on platelet function, KCNQ1 locus was discovered to contribute to platelet function variability (19,20). The exact mechanism of these interactions remains unknown, as KCNQ1 is mostly co-assembled with the product of the KCNE1 (minimal K+-channel protein) gene in the heart to form a cardiac-delayed rectifier-like K+ current and the effect of KCNQ1 channels on platelet function has not been not directly investigated so far. However, Gallego-Fabrega et al (43) reported recently that the methylation pattern of KCNQ1 locus might be associated with vascular recurrence in aspirin-treated stroke patients.

The study is limited by the absence of independent verification of accumulation of deleterious KCNQ1 rare variants. However, it was demonstrated previously that the occurrence of rare variants, because of their private character, is often limited to very restricted cohorts and has been difficult to repeat in other cohorts, unless the confirmation cohorts are truly large (in this case several tens of thousands of patients). The added drawback of the presented research is that the direct effect of the observed genomic variants on the platelet function (e.g. aggregation) was not evaluated. This might raise an issue if the observed change in the frequency of variants were entirely related to the platelet function or, perhaps, some other mechanisms related to biochemical pathways. Moreover, it should be noted that only a limited number of all known genes related to platelet function were re-sequenced in this study. We would like to stress that the results of re-sequencing of the more frequently investigated 26 genes related to platelet function were published by our group in the past (13).

The outcome of this study indicates that the increased accumulation of rare damaging variants in the exons of the sequenced 54 platelet genes (and in particular for variants located in the region of potassium channel KCNQ1 gene) could be associated with LVIS. The mechanism of the interaction of these variants with LVIS currently appears unclear and therefore requires further investigations. It is also uncertain if our results could be directly translated to other populations, as the variants responsible for the observed associations appear to be limited to the investigated cohort. Further studies in different, as well as much larger cohorts, are required to address this problem.

Supplementary Material

Supporting Data

Supporting Data

Acknowledgements

The authors would like to thank Dr Michal Karlinski (Institute of Psychiatry and Neurology, Warsaw, Poland) and Dr Agnieszka Cudna (Center for Preclinical Research and Technology CEPT, Warsaw, Poland) for preparing the samples and database for further analysis.

Funding

Research subject was implemented with CEPT infrastructure financed by the European Union-the European Regional Development Fund within the Operational Program ‘Innovative economy’ for 2007–2013. The study was supported financially as part of the research grant from the National Science Center OPUS research grant (grant no. 2013/11/B/NZ7/01541).

Availability of data and materials

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Authors' contributions

PKJ and MP conceived the concept and design for the study, were involved in data collection and analysis, and supervised the work. CE and VRV contributed to the design of the research, and were involved in data collection and analysis. SS and YIK verified the analytical methods. JP, AC, IKJ and DMG were involved in data collection and analysis. All authors read and approved the final manuscript.

Ethics approval and consent to participate

All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in 2000. Informed consent was obtained from all patients for being included in the study.

Patient consent for publication

The consent for publication was obtained from all patients included in the study.

Competing interests

The authors declare that they have no competing interests.

Glossary

Abbreviations

Abbreviations:

References