Article ID: CJ-19-0415
Background: Atrial fibrillation (AF) is the most common arrhythmia with serious complications and a high rate of recurrence after catheter ablation. Recently, mutation of MYL4 was reported as responsible for familial atrial cardiomyopathy and AF. This study aimed to determine the association between polymorphism in MYL4 with the onset and recurrence of AF.
Methods and Results: A total of 7 single-nucleotide polymorphisms were selected by linkage disequilibrium and genotyped in 510 consecutive AF patients and 192 controls without structural heart disease. A total of 246 AF patients who underwent cryoballoon ablation had a 1-year scheduled follow-up study for AF recurrence. C allele and CC genotype of rs4968309 and A allele of rs1515751were associated with AF onset both before and after adjustment of covariation (age, sex, hypertension, and diabetes). AF type and left atrial size were different among the genotypes of rs4968309. Moreover, CC genotype of rs4968309 increased susceptibly of AF recurrence after cryoballoon ablation. The prevalence of hypertension was associated with rs1515752, and left atrial size was associated with the genotype of rs2071438.
Conclusions: C allele and CC genotype of rs4968309 in MYL4 were associated with AF onset and recurrence. Moreover, the A allele of rs1515751 had a significant association with AF onset. The polymorphisms of MYL4 can predict AF onset and prognosis after ablation in AF patients without structural heart disease.
Atrial fibrillation (AF) is the most common arrhythmia affecting an estimated 1–2% of the population and is associated with serious outcomes such as stroke and heart failure.1,2 Although catheter ablation is an effective method of improving life span and quality of life for many AF patients, there are still 20–40% of patients who require further management.3 AF is associated with several clinical risk factors, such as diabetes mellitus, obesity, and hypertension, and its development is affected by genetic background. To date, rare variants of more than 30 genes have been identified as correlating with AF.4–6 Most of these studies have focused on ion channels, including cardiac sodium, calcium, and potassium.6–12 Linkage and candidate gene studies have also identified rare genetic variants of non-ion channel-related genes in individuals with familial or lone AF, such as a frameshift mutation in the gap junction protein a 1 (GJA1) gene,13 a homozygous mutation in the nucleoporin 155 (NUP155) gene,13 and the natriuretic peptide A (NPPA) gene.14
Recently, using exome sequencing and systematic bioinformatics analyses, we identified a rare missense variant of the gene encoding essential myosin light chain type-4 (MYL4 c.31G>A. p.E11K), causing missense-induced loss-of-function in a large multiplex atrial cardiomyopathy family pedigree.15 Two recent studies have also discussed the significance of variants in MYL4 in familial atrial cardiomyopathy and AF.16,17 The MYL4 protein forms a collar around the lever of the myosin heavy chain, close to the site forming actin-myosin cross-bridges, and regulates the interaction.18 This 21.6-kDa protein consisting of 197 amino acids virtually disappears from the human ventricle after birth, becomes essentially atrial-specific, and reappears at the ventricular level in response to ventricular pathology.18 The corresponding ventricular protein, MYL3, is essential for normal ventricular function. MYL3 mutations cause hypertrophic cardiomyopathy.19 Our further study indicated that loss-of-function MYL4 variants cause progressive atrial cardiomyopathy in humans and rats, and MYL4 is a key gene required for atrial contractile, electrical, and structural integrity.15
Monogenic mutations in lone and familial AF, although rare, have been recognized as both gain- and loss-of-function mutations in the same gene and can cause AF.20 Genome-wide association studies have indicated that common single-nucleotide polymorphisms (SNPs) have a role in the development of AF. Common genetic variants associated with AF may be adopted in the clinical setting, such as risk stratification of the development of AF, estimation of the efficacy of pulmonary vein isolation, and indication of anticoagulant therapy. Excepting the rare gene variants, SNPs associated with AF may identify individuals at an increased risk and enable stratification of ablation therapy or peri-interventional management. There has been a study based on large-scale gene sequencing of Icelanders that identified SNPs (rs185183057, rs28588212, rs117626672, rs117626672) in MYL4 that predisposed to AF.21,22 In this study, we scanned all sites of MYL4 by gene sequencing in small Chinese cohorts with and without AF and found a new SNP rs4968309 differing significantly in the AF and non-AF cohorts. Next, we selected 6 SNPs in MYL4 according to linkage disequilibrium via Haploview and genotyping for all 7 SNPs to expand the participation. Through baseline data analysis and follow-up of patients after catheter ablation, we investigated whether those SNPs were associated with AF onset or recurrence after ablation in AF patients without structural heart disease.
Our study complied with the Declaration of Helsinki and was approved by the hospital’s ethical review board (Shanghai Tenth People’s Hospital, Tongji University School of Medicine, Shanghai, China). Written informed consent was given by all subjects before their participation in the study. This was an observational study of 510 consecutive AF patients from the Department of Cardiology, Shanghai Tenth People’s Hospital who were diagnosed with non-valvular AF according to the diagnostic criteria of the AF guidelines23 between March 2016 and May 2018. The 192 non-AF controls were patients who were admitted to the Department of Cardiology at the same time and met the exclusion criteria. Exclusion criteria included: (1) acute or chronic infections or inflammatory diseases, severe liver or renal dysfunction, malignant tumor, hematologic disorders, and history of cerebral infarction or transient cerebral ischemia within 6 months; (2) myocarditis or cardiomyopathies; (3) reversible AF caused by acute alcoholism, cardiopulmonary surgery, endocrine abnormalities etc.; (4) recent interventional treatment of acute coronary events; (5) undergone left atrial (LA) ablation or cardiac surgery; (6) LA diameter (LAd) >60 mm; and (7) heart failure.
Paroxysmal AF was defined as AF that terminated spontaneously or with intervention within 7 days of onset. Persistent AF was defined as continuous AF sustained ≥7 days, or episodes beyond >7 days that were terminated by cardioversion, either with drugs or by direct current cardioversion. Coronary artery disease was defined as the occurrence of exercise- and stress-related chest symptoms caused by ≥70% narrowing of 1 or more of the major coronary arteries and ≥50% in the left main coronary artery.24 The LAd and left ventricular ejection fraction (LVEF) were measured by transthoracic ultrasonic testing. Heart failure patients were New York Heart Association class IV25 or LVEF <40%.26 Body mass index (BMI) was calculated as weight divided by height squared. Hypertension was defined as blood pressure >140/90 mmHg or the use of antihypertensive medication. Diabetes was diagnosed according to the WHO criteria. Cryoballoon ablation for 246 AF patients was performed by qualified physicians using a standard ablation procedure that has been previously reported in detail.27
SNPs Selection and DNA GenotypingDNA was extracted from peripheral blood using standard phenol chloroform extraction techniques. The 192 control cohort and 288 AF patients were randomly selected for direct sequencing of MYL4. Genotyping was performed in the remaining the samples with TaqMan allelic discrimination by means of an ABI 7900HT (Applied Biosystems, Foster City, CA, USA). The TaqMan assay kits and probes were obtained from Applied Biosystems. All PCR data were analyzed using ABI Prism SDS software version 2.3. The samples that PCR failed to detect or distinguish production were excluded. Based on the results of sequencing, we found that rs4968309 differed significantly in the AF and non-AF cohorts. Next, we selected 6 SNPs in MYL4 according to linkage disequilibrium via Haploview.
Follow-up StudyAll participants were monitored by qualified physicians during their hospital stay and the procedure (i.e., the first 24–48 h) with ECGs and Holter studies. For patients who underwent cryoballoon ablation there were scheduled visits at 1, 3, 6, and 12 months after discharge, or earlier if symptoms were consistent. 24-h Holter and 12-lead ECG were necessary to determine the rapid recurrence of atrial arrhythmias. Recurrence of AF was defined as any recording of AF on ECG or an episode >30 s on 24-h Holter after a blanking period of 3 months or a repeat ablation at any time. Atrial tachycardia was defined as a regular, organized supraventricular rhythm at a constant atrial rate >100 beats/min. Based on the morphology and regularity of the rhythm, early and late recurrences were classified as AF or AT.
Statistical AnalysisStatistical Package for Social Sciences (SPSS) for Windows 10 was used for statistical analysis. Unpaired t-test and Chi-square test were used for continuous traits and categorical traits respectively. For the allelic analysis, Chi-square test and Fisher’s exact test were used to compare the frequency of SNP variation between AF and non-AF patients and then confirmed in 3 genetic models. To further evaluate the relationship between alleles and AF, alleles with P<0.05 in the Chi-square test or Fisher’s exact test were selected and multiple logistic regression analysis was performed with risk factors of AF (age, sex, hypertension, and diabetes mellitus) as adjusted covariates. For the predictors of AF recurrence after cryoballoon ablation, survival analysis was performed to test the effect of the risk allele using Kaplan-Meier test. Odds ratio (OR) and 95% confidence interval (CI) are reported. Numerical variables with a normal distribution are presented as the mean±SE, while numerical variables with a skewed distribution are presented as the median, and categorical variables are presented as percentages. Two-tailed P<0.05 was considered statistically significant for all analyses. In addition to multiple comparisons, the P-value was corrected to <0.01 based on Bonferroni adjustment in post hoc analysis. According to the minor allele frequency (MAF) ranging from 0.112 for rs4968309 to 0.465 for rs16941662 reported in HapMap of the East Asian population, OR of the study was 1.5, promising >80% statistical power to detect an association (at P<0.05) under the dominant model.
There were 510 AF patients and 192 non-AF controls in our study, including 246 AF patients who underwent cryoballoon ablation. In the AF cohort, 25.40% was persistent AF and 74.60% was paroxysmal AF. With regard to general demographic characteristics (age, sex and BMI) and basic clinical factors (hypertension, diabetes mellitus, LAd and pro-BNP), there were statistical differences between the AF and control groups (Table 1). AF patients had larger LAd and more prevalence of hypertension and diabetes mellitus compared with non-AF patients (P<0.001 for both). In addition, AF patients had high BMI and increased pro-BNP. However, thyroid dysfunction, LVEF, and serum level of low-density lipoprotein (LDL) were not significantly different between the AF and non-AF groups.
| Characteristic | Non-AF (n=192) |
AF (n=510) |
P value | Non-recurrence (n=197) |
Recurrence (n=49) |
P value |
|---|---|---|---|---|---|---|
| Age (M±SD, years) | 60.15±7.53 | 65.66±9.10 | <0.001 | 65.35±9.79 | 63.70±7.24 | 0.278 |
| Sex (male, %) | 79 (41.80) | 250 (56.43) | 0.001 | 116 (58.59) | 26 (55.32) | 0.683 |
| BMI (M±SD, kg/m2) | 24.06±3.08 | 25.52±3.61 | 0.012 | 25.07±3.19 | 26.45±2.85 | 0.268 |
| Hypertension (n, %) | 6 (3.16) | 265 (60.64) | <0.001 | 105 (54.40) | 27 (58.70) | 0.599 |
| DM (n, %) | 2 (1.04) | 70 (16.02) | <0.001 | 25 (12.95) | 9 (19.57) | 0.249 |
| Thyroid dysfunction (n, %) | 1 (0.53) | 5 (1.15) | 0.464 | 4 (2.07) | 0 (0.00) | 0.325 |
| pro-BNP (M±SD, pg/mL) | 90.21±142.40 | 791.56±1,524.57 | <0.001 | 572.32±835.80 | 562.85±480.33 | 0.943 |
| LDL (M±SD, mg/dL) | 107.12±31.71 | 106.73±274.94 | 0.977 | 88.94±31.71 | 90.10±29.78 | 0.841 |
| AF type (persistent/paroxysmal) | NA | 111/326 | NA | 23/170 | 17/29 | <0.001 |
| LVEF (M±SD, %) | 60.81±18.41 | 58.69±13.82 | 0.106 | 59.03±13.42 | 59.80±14.10 | 0.730 |
| LAd (M±SD, mm) | 35.58±4.55 | 42. 43±6.47 | <0.001 | 40.52±5.40 | 43.15±5.99 | 0.004 |
Data are mean±standard deviation (M±SD), unless otherwise shown. AF, atrial fibrillation; BMI, body mass index; BNP, B-type natriuretic peptide; DM, diabetes mellitus; LAd, left atrial diameter; LDL, low-density lipoprotein; LVEF, left ventricular ejection fraction; NA, not applicable.
We followed the 246 AF patients who underwent cryoballoon ablation, including 21 patients who relapsed within 3 months (early recurrence of AF, ERAF) and 28 patients with late recurrence of AF (LRAF, between 3 and 12 months). In total, 19.51% of the AF patients relapsed in our follow-up study. As shown in Table 1, compared with the non-recurrence cohort, there were more patients diagnosed as persistent AF among the cases of relapses (11.92% vs. 42.5%, P<0.001), and the recurrence group had larger LAd compared with the non-recurrence cohort (43.47±5.61 vs. 40.51±5.46 mm, P=0.02).
Rs4968309 and rs1515751 Associated With the Risk of AF OnsetAccording to the results of direct sequencing, the frequency of the minor C allele in rs4968309 was significantly higher in AF cases. For the selected 6 SNPs (rs1515751, rs2071438, rs16491971, rs16491662, rs151572, and rs9894365) with R2 of linkage disequilibrium exceeding 0.8 compared with rs4968309 by Haploview (Figure 1), the MAF of all SNPs exceeded 0.05 and was similar to the frequencies in the Chinese population reported by the Hapmap Project. Table 2 shows the genotype and allele frequencies of the MYL4 polymorphisms, with the ancestral allele of each SNP used as the referent for minor allelic homozygote and heterozygote analyses. Rs4968309 and rs1515751 had different frequencies in AF patients and non-AF controls (P<0.001 and P=0.003, respectively). After Bonferroni correction, the difference was still significant for P<0.01.
Positions of the single-nucleotide polymorphisms (SNPs) in the MYL4 gene and linkage disequilibrium plot (LD) calculated with Haploview software and shown in the red diamonds.
| SNPs / Genotype |
Control (n=192) |
AF (n=510) |
P value | OR (95% CI) |
Allele | Control (n=192) |
AF (n=510) |
P value | OR (95% CI) |
|---|---|---|---|---|---|---|---|---|---|
| rs4968309 | |||||||||
| TT | 134 (74.44) |
383 (80.63) |
Ref. | T | 309 (85.83) |
797 (83.89) |
0.388 | 1.163 (0.825, 1.639) |
|
| CT | 41 (22.78) |
31 (6.53) |
<0.001 | 3.780 (2.278, 6.272) |
C | 51 (14.17) |
153 (16.11) |
||
| CC | 5 (2.78) |
61 (12.84) |
0.001 | 4.268 (1.680, 10.848) |
|||||
| rs2071438 | |||||||||
| AA | 133 (78.24) |
369 (82.00) |
Ref. | A | 303 (89.12) |
813 (90.33) |
0.524 | 0.876 (0.584, 1.316) |
|
| AG | 37 (21.76) |
75 (16.67) |
0.162 | 0.731 (0.470, 1.135) |
G | 37 (10.88) |
87 (9.67) |
||
| GG | 0 (0.00) |
6 (1.33) |
0.317 | NA | |||||
| rs1515751 | |||||||||
| GG | 151 (94.97) |
204 (84.65) |
Ref. | G | 310 (97.48) |
442 (91.70) |
0.001 | 3.507 (1.619, 7.595) |
|
| AG | 8 (5.03) |
34 (14.11) |
0.003 | 3.146 (1.416, 6.990) |
A | 8 (2.52) |
40 (8.30) |
||
| AA | 0 (0.00) |
3 (1.24) |
0.369 | NA | |||||
| rs1515752 | |||||||||
| TT | 84 (56.38) |
349 (78.25) |
Ref. | T | 229 (76.85) |
787 (88.23) |
<0.001 | 0.443 (0.316, 0.621) |
|
| TG | 61 (40.94) |
89 (19.96) |
<0.001 | 0.351 (0.235, 0.526) |
G | 69 (23.15) |
105 (11.77) |
||
| GG | 4 (2.68) |
8 (1.79) |
0.408 | 0.481 (0.142, 1.636) |
|||||
| rs9894365 | |||||||||
| GG | 80 (52.98) |
226 (48.92) |
Ref. | G | 209 (69.21) |
634 (68.61) |
0.848 | 0.973 (0.734, 1.289) |
|
| GA | 49 (32.45) |
182 (39.39) |
0.185 | 0.761 (0.507, 1.141) |
A | 93 (30.79) |
290 (31.39) |
||
| AA | 22 (14.57) |
54 (11.69) |
0.621 | 1.151 (0.659, 2.010) |
|||||
| rs16941662 | |||||||||
| TT | 42 (26.92) |
128 (27.29) |
Ref. | T | 164 (52.56) |
486 (51.81) |
0.818 | 0.970 (0.751, 1.254) |
|
| CT | 80 (51.28) |
230 (49.04) |
0.791 | 1.060 (0.689, 1.632) |
C | 148 (47.44) |
452 (48.19) |
||
| CC | 34 (21.80) |
111 (23.67) |
0.795 | 0.934 (0.556, 1.568) |
|||||
| rs16941671 | |||||||||
| CC | 66 (47.83) |
323 (69.31) |
Ref. | C | 199 (72.10) |
765 (82.08) |
<0.001 | 0.564 (0.413, 0.771) |
|
| GC | 67 (48.55) |
119 (25.54) |
<0.001 | 0.363 (0.243, 0.541) |
G | 77 (27.90) |
167 (17.92) |
||
| GG | 5 (3.62) |
24 (5.15) |
0.970 | 0.981 (0.361, 2.664) |
|||||
Genotype distribution of SNPs between groups: rs4968309, P<0.001; rs2071438, P=0.119; rs1515751, P=0.005; rs1515752, P<0.001; rs9894365, P=0.273; rs16941662, P=0.860; rs16941671, P<0.001. CI, confidence interval; OR, odds ratio; SNP, single-nucleotide polymorphism.
As shown in Table 3, under additive, dominant, and recessive models, the C allele CC genotype of rs4968309 (P=0.001, OR=4.268 under additive model and P=0.001, OR=5.157 under recessive model) was associated with AF onset. Multivariable analysis with covariations (age, sex, hypertension, and diabetes) showed that the CC genotype of rs4968309 was still associated with AF onset under both the additive and recessive models (P=0.002, OR=5.397 and P=0.001, OR=6.368, respectively). An allele of rs1515751 was associated with AF onset under the dominant model (P=0.001, OR=4.441), and the association remained significant after covariation adjustment (P=0.001, OR=6.022).
| SNPs | P value | OR (95% CI) |
P-adj | Exp(b) (95% CI) |
|
|---|---|---|---|---|---|
| rs4968309 | |||||
| Additive | CC | 0.001 | 4.268 (1.680, 10.848) | 0.002 | 5.397 (1.845, 15.785) |
| TT | |||||
| Dominant | CC+CT | 0.083 | 0.700 (0.467, 1.049) | 0.197 | 0.698 (0.404, 1.206) |
| TT | |||||
| Recessive | CC | <0.001 | 5.157 (2.037, 13.054) | 0.001 | 6.368 (2.192, 18.502) |
| TT+CT | |||||
| rs1515751 | |||||
| Additive | AA | 0.369 | NA | NA | NA |
| GG | |||||
| Dominant | AA+AG | <0.001 | 4.441 (1.825, 10.810) | 0.001 | 6.022 (1.995, 18.176) |
| GG | |||||
| Recessive | AA | 0.412 | NA | NA | NA |
| GG+AG | |||||
Covariations: age, sex, hypertension and diabetes. Exp(b), adjusted OR; P-adj, adjusted P value. Other abbreviations as in Tables 1,2.
Clinical data for participants with different genotypes are shown in Table 4. Both LV function and LAd were within normal ranges among the AF and non-AF patients. Genotypes of rs4968309 were significantly associated with LAd and AF type distribution (P=0.002 and P=0.028). The CC genotype of rs4968309 was persistent with AF (P=0.019, OR=2.889 (1.190, 7.012)). LAd was also associated with rs2071438 under the dominant model (P=0.011, Figure 2). The prevalence of hypertension was associated with the G allele of rs1515752 under the dominant model (P=0.001).
| Characteristic | rs4968309 (n=655) |
rs2071438 (n=620) |
rs1515751 (n=400) |
rs1515752 (n=598) |
rs9894365 (n=613) |
rs16941662 (n=625) |
rs16941671 (n=604) |
|---|---|---|---|---|---|---|---|
| CC/CT/TT (66/72/517) |
GG/AG/AA (6/112/502) |
AA/AG/GG (3/42/355) |
GG/TG/TT (12/150/436) |
AA/GA/GG (76/231/306) |
CC/TC/TT (145/310/170) |
GG/GC/CC (29/186/389) |
|
| Sex (male, n) | 29/34/246 | 1/52/230 | 0/19/151 | 7/68/207 | 35/119/136 | 66/152/78 | 15/89/184 |
| Age (M±SD, years) | 63.87±8.97 | 64.17±8.80 | 63.66±9.55 | 63.86±8.94 | 64.06±9.03 | 63.96±9.07 | 64.20±8.99 |
| BMI (M±SD, kg/m2) | 25.10±3.53 | 25.12±3.49 | 24.62±3.23 | 25.09±3.52 | 25.12±3.55 | 25.04±3.55 | 25.11±3.58 |
| DM | 9/7/48 | 0/17/43 | 0/5/18 | 2/13/48 | 10/25/29 | 21/30/15 | 5/19/41 |
| Hypertension | 29/22/202 | 2/41/194 | 0/16/84 | 3/49/189* | 33/103/111 | 65/121/65 | 16/73/158 |
| AF type | |||||||
| Non-AF | 5/41/134 | 0/37/133 | 0/8/151 | 4/61/84 | 22/49/80 | 34/80/42 | 5/67/66 |
| Paroxysmal AF | 44/20/231* | 2/56/224 | 1/24/129 | 3/62/213 | 36/117/138 | 73/147/78 | 14/73/203 |
| Persistent AF | 6/11/91* | 2/11/84 | 0/3/13 | 3/19/80 | 13/45/46 | 29/52/23 | 10/32/63 |
| LAd (M±SD, mm) | |||||||
| Referent | 40.87±6.68* | 40.93±6.83* | 38.84±5.87 | 41.22±6.71 | 40.80±6.70 | 41.06±6.71 | 41.20±6.67 |
| Alternation | 41.13±6.55 | 41.00±8.28 | 40.33±11.59 | 42.25±6.54 | 41.47±7.75 | 41.55±7.22 | 44.69±7.86 |
| Heterozygote | 38.12±7.47* | 39.04±6.30* | 39.90±5.73 | 39.65±7.24 | 40.78±6.68 | 40.25±6.61 | 39.99±6.60 |
| LVEF (M±SD, %) | |||||||
| Referent | 59.07±15.45 | 58.87±15.60 | 60.70±13.40 | 58.79±15.68 | 59.12±15.99 | 60.22±15.50 | 58.69±15.47 |
| Alternation | 57.05±15.67 | 49.50±24.36 | 60.67±1.16 | 66.50±7.68 | 58.38±15.91 | 59.11±14.86 | 53.45±20.65 |
| Heterozygote | 60.35±17.98 | 60.24±14.32 | 60.03±10.37 | 57.92±17.82 | 59.37±14.55 | 58.59±15.70 | 60.14±14.54 |
| pro-BNP (M±SD, pg/mL) | |||||||
| Referent | 618.30± 1,467.93 |
635.06± 1,498.71 |
380.31± 835.72 |
663.21± 1,543.15 |
625.61± 1,725.48 |
421.08± 693.71 |
713.07± 1,675.26 |
| Alternation | 723.51± 1,120.84 |
773.83± 505.75 |
406.198± 878.46 |
434.55± 713.84 |
425.66± 642.12 |
620.55± 1,057.04 |
676.84± 845.65 |
| Heterozygote | 284.27± 404.37 |
471.35± 859.17 |
630.28± 1,179.41 |
482.29± 932.30 |
639.14± 1,042.68 |
658.60± 1,649.56 |
466.03± 799.54 |
Data are mean±standard deviation (M±SD), unless otherwise shown. *P<0.01. Abbreviations as in Tables 1,2.
Association of rs4968309 (A) and rs2071438 (B) with left atrial size in the AF and non-AF patients. AF, atrial fibrillation.
In the univariate analysis, rs4968309 was associated with AF recurrence (P=0.040). Indeed, both the univariable analysis and multiple regression analysis with adjustment for sex, age, diabetes, hypertension, LVEF, LAd, LDL, and pro-BNP demonstrated that patients with the CC genotype of rs4968309 had increased susceptibly to AF recurrence (P=0.044 and P-adj=0.017 for the recessive model, Table 5).
| Variations | P value | Exp(b) | (95% CI) |
|---|---|---|---|
| Sex | 0.132 | 1.850 | (0.832, 4.116) |
| Age | 0.127 | 0.963 | (0.917, 1.011) |
| DM | 0.093 | 2.548 | (0.857, 7.576) |
| Hypertension | 0.734 | 1.154 | (0.506, 2.633) |
| LAd | 0.007 | 1.116 | (1.030, 1.209) |
| LVEF | 0.380 | 1.020 | (0.976, 1.066) |
| LDL | 0.683 | 1.115 | (0.660, 1.884) |
| pro-BNP | 0.863 | 1.000 | (0.999, 1.001) |
| rs4968309 recessive | 0.017 | 6.709 | (1.411, 31.904) |
Exp(b), adjusted OR. Abbreviations as in Tables 1,2.
From this case-control study, we first report that a novel SNP rs4968309 in MYL4 was associated with AF onset and LAd in AF patients without structural heart disease. Moreover, our study results extend the possibility of using MYL4 polymorphisms in risk prediction for AF recurrence after cryoballoon ablation.
Essential myosin light chains, MYL4 (also known by the abbreviations ALC-1 (for atrial light chain-1) and ELC a (for essential light chain, atrial)), are essential myosin light chain proteins that are expressed in cardiac and skeletal muscles.28 The essential myosin light chains wrap around the lever arm domain of myosin, and together with regulatory myosin light chains are necessary for the normal interactions between myosin heavy chains and actin that are responsible for cross-bridge movement and associated contraction. The N-terminal domain is particularly involved in maintaining isometric tension.29 MYL4 is expressed in both atria and ventricles in embryonic life; however, after birth, ventricular MYL4 expression plummets and the protein becomes highly atrial selective.28 Forced ventricular overexpression of MYL4 in transgenic rats enhances the muscle shortening velocity, rate of tension development, and isometric force production,30 possibly explaining the teleological role of MYL4 re-expression in diseased human ventricles.29–31 MYL4 overexpression protects against contractile impairment in volume-overloaded rats.31
Atrial standstill, a complete loss of atrial electrical and mechanical function, eventually develops in many patients with MYL4 mutations; in some, mechanical function is greatly reduced before atrial electrical activity is lost.15 Several lines of evidence suggest that the MYL4 E11K mutation is not directly responsible for atrial standstill, but rather that it results from the delayed consequences of atrial remodeling induced by the mutation. First, the hypocontractile effects of the mutation are generally expressed at birth, whereas atrial standstill develops much later in life. Second, MYL4 dysfunction reduces atrial contractility but does not necessarily eliminate it. Third, knockout of the MYL4 ortholog in zebrafish causes extensive structural disruption of sarcomere structure, particularly at Z-lines,16 where many macromolecular structures involved in cell signaling are located.18 Disruption of macromolecular complexes might cause changes in regulatory processes that lead to cell death and fibrosis. The extensive atrial remodeling caused by MYL4 disruption that we have already observed in our previous study15 likely contributes to the loss of electrical and mechanical responses.
In contrast to rare variants that typically have large effects and can be identified in studies of familial AF, such as MYL4 E11K mutation15 and MYL4 (c.234delC) frameshift deletion,17 a previous study based on the large-scale gene sequencing of Icelanders identified SNPs (rs185183057, rs28588212, rs117626672, rs117626672) in MYL4 that predispose to AF.21,22 It suggested that common variants with low or modest effects and their association with AF also should be identified. In the present study, 7 SNPs on MYL4 were screened and polymorphisms of this gene could affect the development of AF. By comparing the allele and genotype frequencies of specific SNPs between individuals with AF and controls, we first report that patients who carry the CC genotype of rs4968309 and carry the A allele of rs1515751 tend towards AF onset and there was an association between rs4968309 and LAd in the enrolled patients. This result indicated that these site polymorphisms may also affect the function of MYL4 and result in atrial remodeling and AF.
Although pulmonary vein isolation is an established rhythm control therapy for AF, recurrence may occur in a proportion of individuals with AF. Thus, it would be beneficial to predict which patients are more likely to benefit from the procedure. There are established risk factors of AF recurrence after catheter ablation such as hypertension, obesity, metabolic syndrome, dilation of the left atrium, and persistent AF. Because of increasing data highlighting the important role of genetics in AF, genetic status has the potential to guide therapeutic strategies in the treatment of AF. In this study, both the univariable analysis and multiple regression analysis with adjustment for traditional risk factors demonstrated that patients with the CC genotype of rs4968309 had an increased susceptibility to AF recurrence, which suggests that rs4968309 may help in preprocedural consideration of efficacy and evaluation. Rs4968309 is a non-coding SNP, and non-coding SNPs may play an important role in disease development as enhancer elements, DNase hypersensitivity regions, and chromatin markers, indirectly influencing the function of encode regions.32 These SNPs are considered the signals for the causative genes. The genes in closest proximity to these SNPs have, therefore, been investigated and significant information has been obtained. However, the exact biological pathway between these non-coding SNPs and the emergence of AF still remains unsolved.
Study LimitationsFirst, some patients in this study were at high risk for AF because hypertension, diabetes, and obesity are the most commonly reported risk factors. Second, our findings were based on a relatively small number of participants, and would be more credible in larger cohorts. The polymorphisms of MYL4 may differ in different races. Third, the study only focused on observation of MYL4, which may not be relevant to AF development with other genes.
The C allele and CC genotype of rs4968309 on MYL4 were associated with AF onset and recurrence. The A allele of rs1515751 was also associated with AF onset. The polymorphism of MYL4 might be helpful in predicting AF onset and prognosis after ablation in AF patients without structural heart disease.
This work was supported by the Chinese National Natural Science Foundation grants no. 81670230, 81700291, and 81770226.
None.
