Abstract
Objective: Atherosclerosis plays a key role in the inducibility and persistence of coronary heart disease. Clinical evidence, in vitro and in vivo studies have implicated Urotensin II (U-II/UTS2) in the development of atherosclerosis and coronary artery disease, contributing to the (patho) physiological regulation of cardiovascular homeostasis in humans. Increased U-II plasma levels have been reported in patients with atherosclerosis and coronary heart disease. Considering these, our objective was to evaluate possible role of the UTS2 gene polymorphisms (Thr21Met and Ser89Asn) in the genetic susceptibility to coronary heart disease and myocardial infarction in a Chinese population. Methods: A case-control study was designed to compare the distribution of alleles and genotypes between case group (subjects with myocardial infarction, n=409) and control group (subjects with coronary heart disease, n=830). The detection of UTS2 gene polymorphisms was achieved with PCR-RFLP technique. Results: We did not identify statistically significant differences between the myocardial infarction and coronary heart disease groups, neither with regard to the frequency of genotype/variant at the Ser89Asn locus nor at the Thr21Met locus. When stratified by sex, differences in genotype distribution of polymorphism Ser89Asn were only seen in female subjects in both additive tested inheritance model (OR=0.257, 95% CI: 0.074-0.896, P=0.033) and recessive tested inheritance model (OR=0.280, 95% CI: 0.082-0.955, P=0.042). For subjects with myocardial infarction, we identified statistically significant differences between the ST-segment elevation myocardial infarction and non ST-segment elevation myocardial infarction groups. Differences in genotype distribution of polymorphism Ser89Asn not Thr21Met were seen in both additive tested inheritance model (OR=0.202, 95% CI: 0.049-0.833, P=0.027) and recessive tested inheritance model (OR=0.208, 95% CI: 0.052-0.835, P=0.027). When stratified by sex, differences in genotype distribution of polymorphism Ser89Asn were only seen in male subjects in both additive tested inheritance model (OR=0.208, 95% CI: 0.049-0.890, P=0.034) and recessive tested inheritance model (OR=0.197, 95% CI: 0.047-0.824, P=0.026). Conclusions: Ser89Asn (S89N) polymorphisms of the UTS2 gene were significantly associated with coronary heart disease and myocardial infarction in Chinese population. Additionally, we demonstrated that Genotype Asn89Asn may imply a potential benefit role for myocardial infarction.
Keywords: Coronary heart disease (CAD), myocardial infarction, urotensin-II (U-II/UTS-II), single nucleotide polymorphisms (SNP)
Introduction
Urotensin II (U-II) is a cyclic peptide synthesized through proteolytic cleavage of a precursor molecule, prepro-U-II, and is a potent vasoconstrictor [1]. U-II has been identified in the heart [2], which is also known to display an abundant expression of U-II receptors.
It is now well established that U-II levels are significantly increased in several cardiovascular diseases such as coronary artery disease (CAD). Hassan et al. demonstrated up regulation of U-II in endothelial and sub endothelial cells of atherosclerotic human coronary arteries, and suggested a possible role for U-II in the pathogenesis of coronary atherosclerosis [3]. The above findings are in agreement with that of Maguire et al. showing the presence of U-II immunoreactivity in the plaque of atherosclerotic coronary arteries [4]. It is therefore reasonable to suggest that the increased expression of U-II in coronary atherosclerosis may contribute to the pathophysiology of the coronary atherosclerosis disease through an endocrine or paracrine fashion.
Acute myocardial infarction is the leading cause of morbidity and mortality. The most common cause of MI is coronary atherosclerotic plaque rupture or erosion, resulting in exposure to thrombus formation [5]. As vasoconstriction is involved in unfavorable myocardial and vascular remodeling, changes of U-II secretion after myocardial infarction may deteriorate or counterbalance its clinical course. In patients with myocardial infarction both increased and decreased plasma U-II concentrations were demonstrated. Several authors have investigated plasma levels of U-II in patients with acute coronary syndrome (ACS). Joyal et al. found lower plasma U-II concentration in patients with ACS compared with patients with stable angina and healthy controls [6]. Khan et al. found raised plasma U-II concentration in patients with acute myocardial infarction as compared to healthy controls [7]. Magdalena et al. aimed to analyze whether plasma concentration reflects the severity of ACS and found that decreased plasma U-II concentration in patients with ACS could be associated with more severe injury of myocardium [8]. As the results of the performed studies are ambiguous, it may imply that U-II concentration is not a good marker of myocardial necrosis and the relationship between myocardial necrosis and plasma U-II concentration may be indirect. On the other hand, genetic factors are important determinants.
The purpose of the present study was to assess the impact of the UTS2 Ser89Asn (S89N) and Thr21Met (T21M) polymorphism on the incidence and types of myocardial infarction in a case-control study in Chinese population. To the best of our knowledge, no data are available on the effect of the UTS2 polymorphisms on myocardial infarction compared with coronary artery disease.
Materials and methods
The investigation was conducted according to principles outlined in the Declaration of Helsinki. For research involving human participants, informed consent has been obtained from the patients or the guardian of the patients. The research has been approved by the Ethics Committee of the Peking University First Hospital.
Subjects
Study subjects were enrolled consecutively from the Department of Cardiology, Peking University First Hospital between October 2013 and June 2015. The study included 409 patients admitted to our division of cardiology for acute myocardial infarction and 830 control subjects which were admitted in the same period diagnosed as coronary heart disease without history of myocardial infarction. The Study subjects all underwent coronary angiography. The diagnosis of MI was based on typical electrocardiographic changes as well as on increases in the serum activities of enzymes such as creatine kinase, aspartate aminotransferase, and lactate dehydrogenase and in the serum concentration of troponin T/I. The diagnosis was confirmed by the presence of a wall-motion abnormality on left ventriculography as well as by identification of the responsible stenosis in any of the major coronary arteries or in the left main trunk by coronary angiography.
All study individuals were genetically unrelated ethnic Han Chinese and all came from North China.
Blood samples, DNA isolation and polymerase chain reaction restriction fragment length polymorphism
Blood samples from cases and controls were collected and stored at -80°C, Genomic DNA was extracted from EDTA-anticoagulated peripheral venous blood from each individual using the QIAamp DNA blood kit (Qiagen). The typing was performed using polymerase chain reaction restriction fragment length polymorphism (PCR-RFLP) technique by means of an Applied Biosystems® 2720 Thermal Cycler. PCR optimization for each primer set was validated by temperature gradient and primers are listed in Table 1. The thermo cycling procedure consists of initial denaturation step at 95°C for 4 min, denaturation at 95°C for 30 s, annealing temperatures (58°C for 30 s for Thr21Met) and (60°C for 30 s for Ser89Asn) with 35 cycles, extension at 72°C for 30 s), and final extension (72°C for 7 min). Thr21Met polymorphism was identified by TaiI restriction enzyme and restricted on this region. The wild-type DNA (21T) yielded two bands with 83 and 58 bp, while the mutated DNA (21M) showed three bands with 141, 83, and 58 bp in size. Ser89Asn polymorphism was identified by RsaI restriction enzyme and restricted on this region. The 89S generated three bands with 161, 84 and 18 bp, while 89N showed two bands with 245 and 18 bp. Digested PCR products were resolved by 3% agarose gel electrophoresis, stained by using ethidium bromide and photographed under UV illumination. Genotyping was conducted in a blinded fashion.
Table 1.
Urotensin II gene polymorphisms Thr21Met and Ser89Asn with location, primer sequences, sizes of the amplicons and annealing temperatures
| Reference SNP Number | Aminoacid changes | Enzyme | Nucleotide composition (5’→3’) | Expected size of PCR product (bp) | Annealing temperature (°C) |
|---|---|---|---|---|---|
| rs228648 | Thr21Met | TaiI | GGAAACCAACGTATTTCATC | 141 | 60 |
| GCAAAAGAGGCAACTTACAGC | |||||
| rs2890565 | Ser89Asn | RsaI | GTGCCTGTCTGTCTGCATTCA | 263 | 58 |
| GAGTCCTGTAAAACCAGTACAG |
We re-genotyped approximately 20% of the samples to verify the initial results. The check confirmed the previous genotyping results by 100%. Ten percent of individuals from each study group were randomly chosen, directly sequenced and the results were 100% concordant.
All of the UTS2 gene polymorphisms (Thr21Met and Ser89Asn) were in HWE (P>0.05) in population controls.
Statistical analysis
Statistical analysis was carried out using the SPSS 19.0. Independent t-test was used to examine the statistical significance of the difference of age and sex in different groups. Differences of genotype distribution and allele frequency were tested by chi-square analysis. Fisher’s exact test was used when the expected number of samples in a group was less than five. Two sided testing was used to evaluate statistical significance. Multivariate logistic regression analysis was used when age and sex were adjusted. Odds ratios (OR) with 95% confidence intervals (95% CI) were calculated to identify the relationship between SNPs and case groups. Linkage disequilibrium was analyzed with SHEsis software (http://analysis.bio-x.cn/myAnalysis.php).
Results
Cases and controls
There were no statistically significant differences between the control and case groups for age. Male were more common in case group (76.8% in case group vs. 64.3% in control group; P<0.001) (Table 2).
Table 2.
Clinical characteristics of patients with myocardial infarction compared with controls
| Demographic characteristic | Cases | Controls | t | χ2 | P |
|---|---|---|---|---|---|
| Sex | |||||
| Male | 314 (0.768) | 534 (0.643) | |||
| Female | 95 (0.232) | 296 (0.357) | 19.616 | <0.001 | |
| Mean age (years) | 63.05±12.50 | 63.88±10.27 | 1.164 | 0.245 | |
| Hypertension | |||||
| Yes | 275 (0.672) | 568 (0.684) | |||
| No | 134 (0.328) | 262 (0.316) | 0.18 | 0.671 | |
| Diabetes | |||||
| Yes | 170 (0.416) | 348 (0.419) | |||
| No | 239 (0.584) | 482 (0.581) | 0.015 | 0.903 | |
| Hyperlipidaemia | |||||
| Yes | 218 (0.533) | 477 (0.575) | |||
| No | 191 (0.467) | 353 (0.425) | 1.934 | 0.164 | |
| Hyperhomocysteinemia | |||||
| Yes | 56 (0.137) | 119 (0.143) | |||
| No | 353 (0.863) | 711 (0.857) | 0.094 | 0.759 | |
| Smoker | |||||
| Yes | 257 (0.628) | 415 (0.500) | |||
| No | 152 (0.372) | 415 (0.500) | 18.188 | <0.001 |
Analysis of HWE and the associations between SNPs of the UTS2 gene with myocardial infarction
The genotypes of control were in Hardy-Weinberg equilibrium (HWE) (P>0.05).
The genotype and allele frequencies of UTS2 polymorphisms in case and control groups were summarized in Table 3. The frequencies of genotype and allele were adjusted for age, sex and concomitant diseases.
Table 3.
Allele frequency and genotype distribution of urotensin II gene polymorphisms Thr21Met and Ser89Asn in cases and controls
| SNP | Cases | Controls | OR (95% CI) | P | aOR (95% CI) | P |
|---|---|---|---|---|---|---|
| rs2890565 | ||||||
| Additive | ||||||
| GG | 215 (0.526) | 432 (0.520) | ||||
| AG | 167 (0.408) | 339 (0.408) | 0.990 (0.773-1.267) | 0.935 | 0.973 (0.757-1.251) | 0.832 |
| AA | 27 (0.066) | 59 (0.071) | 0.920 (0.567-1.492) | 0.734 | 0.961 (0.588-1.571) | 0.873 |
| Dominant | ||||||
| GG | 215 (0.526) | 432 (0.520) | ||||
| AG+AA | 194 (0.474) | 398 (0.480) | 0.979 (0.773-1.241) | 0.863 | 0.971 (0.764-1.236) | 0.813 |
| Recessive | ||||||
| GG+AG | 382 (0.934) | 771 (0.929) | ||||
| AA | 27 (0.066) | 59 (0.071) | 0.924 (0.576-1.480) | 0.972 (0.602-1.570) | 0.908 | |
| Alleles | ||||||
| G | 597 (0.730) | 1203 (0.725) | ||||
| A | 221 (0.270) | 457 (0.275) | 0.974 (0.807-1.176) | 0.788 | ||
| rs228648 | ||||||
| Additive | ||||||
| GG | 190 (0.466) | 366 (0.443) | ||||
| AG | 172 (0.422) | 379 (0.458) | 0.874 (0.680-1.124) | 1.294 | 0.870 (0.675-1.123) | 0.285 |
| AA | 46 (0.113) | 82 (0.099) | 1.081 (0.723-1.614) | 0.705 | 1.140 (0.758-1.714) | 0.53 |
| Dominant | ||||||
| GG | 190 (0.466) | 366 (0.443) | ||||
| AG+AA | 218 (0.534) | 461 (0.557) | 0.911 (0.718-1.156) | 0.442 | 0.961 (0.685-1.349) | 0.819 |
| Recessive | ||||||
| GG+AG | 362 (0.887) | 745 (0.901) | ||||
| AA | 46 (0.113) | 82 (0.099) | 1.154 (0.788-1.692) | 0.461 | 1.220 (0.827-1.800) | 0.316 |
| Alleles | ||||||
| G | 552 (0.676) | 1111 (0.672) | ||||
| A | 264 (0.324) | 543 (0.328) | 0.979 (0.818-1.170) | 0.812 |
Adjusted for clinical characteristics.
SNP, single nucleotide polymorphism.
The wild-type genotype GG at the Thr21Met locus was observed in 190 (46.6%) of the myocardial infarction patients, whereas the frequencies of variant genotypes GA and AA were 172 (42.2%) and 46 (11.3%) respectively. In the control group, the frequencies of genotypes were 366 (44.3%) for GG, 379 (45.8%) for GA and 82 (9.9%) for AA. However, there were no statistical differences in genotype or allele frequency at the Thr21Met locus between case and control group.
The wild-type genotype GG at the Ser89Asn locus was observed in 215 (52.6%) of the myocardial infarction patients, whereas the frequencies of variant genotypes GA and AA were 167 (40.8%) and 27 (6.6%) respectively. In the control group, the frequencies of genotypes were 432 (52%) for GG, 339 (40.8%) for GA and 59 (7.1%) for AA. Concerning the UTS2 Ser89Asn gene polymorphisms, there were no statistical differences in genotype or allele frequency between case and control group.
We did not identify statistically significant differences between the myocardial infarction and coronary heart disease groups, neither with regard to the frequency of genotype variant at the Ser89Asn locus nor at the Thr21Met locus.
Association of UTS2 gene polymorphisms with myocardial infarction stratified by sex
UTS2 Ser89Asn genotype distribution was analyzed with stratification by sex. When case group was compared with control group, the significant association regarding to the frequency of variant genotypes at the Ser89Asn locus was only seen in female subjects in both additive tested inheritance model (OR=0.257, 95% CI: 0.074-0.896, P=0.033) and recessive tested inheritance model (OR=0.280, 95% CI: 0.082-0.955, P=0.042) (Table 4).
Table 4.
Genotype distribution of Ser89Asn in the urotensin II gene in cases and controls stratified by sex
| rs2890565 | Cases | Controls | OR (95% CI) | P | aOR (95% CI) | P |
|---|---|---|---|---|---|---|
| Male | ||||||
| Additive | ||||||
| GG | 162 (0.516) | 284 (0.532) | ||||
| AG | 128 (0.408) | 221 (0.414) | 1.015 (0.759-1.358) | 0.918 | 1.011 (0.755-1.355) | 0.94 |
| AA | 24 (0.076) | 29 (0.054) | 1.451 (0.817-2.576) | 0.202 | 1.453 (0.817-2.586) | 0.204 |
| Dominant | ||||||
| GG | 162 (0.516) | 284 (0.532) | ||||
| AG+AA | 152 (0.484) | 250 (0.468) | 1.066 (0.806-1.409) | 0.654 | 1.063 (0.803-1.407) | 0.67 |
| Recessive | ||||||
| GG+AG | 290 (0.924) | 505 (0.946) | ||||
| AA | 24 (0.076) | 29 (0.054) | 1.441 (0.823-2.522) | 0.199 | 1.446 (0.824-2.537) | 0.198 |
| Alleles | ||||||
| G | 452 (0.720) | 789 (0.739) | ||||
| A | 176 (0.280) | 279 (0.261) | 1.101 (0.883-1.374) | 0.393 | ||
| Female | ||||||
| Additive | ||||||
| GG | 53 (0.558) | 148 (0.50) | ||||
| AG | 39 (0.411) | 118 (0.399) | 0.923 (0.572-1.490) | 0.743 | 0.829 (0.507-1.357) | 0.455 |
| AA | 3 (0.032) | 30 (0.101) | 0.279 (0.082-0.953) | 0.031 | 0.257 (0.074-0.896) | 0.033 |
| Dominant | ||||||
| GG | 53 (0.558) | 148 (0.50) | ||||
| AG+AA | 42 (0.442) | 148 (0.50) | 0.792 (0.498-1.261) | 0.326 | 0.716 (0.444-1.155) | 0.171 |
| Recessive | ||||||
| GG+AG | 92 (0.968) | 266 (0.899) | ||||
| AA | 3 (0.032) | 30 (0.101) | 0.289 (0.086-0.970) | 0.033 | 0.280 (0.082-0.955) | 0.042 |
| Alleles | ||||||
| G | 145 (0.763) | 414 (0.699) | ||||
| A | 45 (0.237) | 178 (0.301) | 0.722 (0.495-1.053) | 0.09 |
Adjusted for clinical characteristics.
However, there was no statistical difference in allele frequency at the Ser89Asn locus between case and control groups among female subjects. We did not find statistical differences in genotype or allele frequencies at the Ser89Asn locus between case and control groups among male subjects.
UTS2 Thr21Met genotype distribution was analyzed with stratification by sex, we did not find any statistical difference (data not shown).
Association of UTS2 gene polymorphisms with types of myocardial infarction
For subjects with myocardial infarction, we identified statistically significant differences between the ST-segment elevation myocardial infarction and non ST-segment elevation myocardial infarction groups. Differences in genotype distribution of polymorphism Ser89Asn not Thr21Met were seen in both additive tested inheritance model (OR=0.202, 95% CI: 0.049-0.833, P=0.027) and recessive tested inheritance model (OR=0.208, 95% CI: 0.052-0.835, P=0.027) (Table 5). When stratified by sex, differences in genotype distribution of polymorphism Ser89Asn were only seen in male subjects in both additive tested inheritance model (OR=0.208, 95% CI: 0.049-0.890, P=0.034) and recessive tested inheritance model (OR=0.197, 95% CI: 0.047-0.824, P=0.026) (Table 6).
Table 5.
Allele frequency and genotype distribution of urotensin II gene polymorphisms Thr21Met and Ser89Asn with types of myocardial infarction
| SNP | STEMI | Non-STEMI | OR (95% CI) | P | aOR (95% CI) | P |
|---|---|---|---|---|---|---|
| rs2890565 | ||||||
| Additive | ||||||
| GG | 59 (0.541) | 53 (0.491) | ||||
| AG | 47 (0.431) | 46 (0.426) | 0.918 (0.529-1.591) | 0.76 | 0.940 (0.515-1.715) | 0.839 |
| AA | 3 (0.028) | 9 (0.083) | 0.299 (0.077-1.165) | 0.068 | 0.202 (0.049-0.833) | 0.027 |
| Dominant | ||||||
| GG | 59 (0.541) | 53 (0.491) | ||||
| AG+AA | 50 (0.459) | 55 (0.509) | 0.817 (0.479-1.392) | 0.456 | 0.782 (0.439-1.390) | 0.402 |
| Recessive | ||||||
| GG+AG | 106 (0.972) | 99 (0.917) | ||||
| AA | 3 (0.028) | 9 (0.083) | 0.311 (0.082-1.183) | 0.072 | 0.208 (0.052-0.835) | 0.027 |
| Alleles | ||||||
| G | 165 (0.757) | 152 (0.704) | ||||
| A | 53 (0.243) | 64 (0.296) | 0.763 (0.498-1.167) | 0.212 | ||
| rs228648 | ||||||
| Additive | ||||||
| GG | 47 (0.431) | 44 (0.415) | ||||
| AG | 48 (0.440) | 44 (0.415) | 1.021 (0.572-1.824) | 0.943 | 0.974 (0.519-1.828) | 0.936 |
| AA | 14 (0.128) | 18 (0.170) | 0.728 (0.324-1.638) | 0.442 | 1.014 (0.417-2.466) | 0.975 |
| Dominant | ||||||
| GG | 47 (0.431) | 44 (0.415) | ||||
| AG+AA | 62 (0.569) | 62 (0.585) | 0.936 (0.545-1.608) | 0.811 | 0.985 (0.549-1.767) | 0.958 |
| Recessive | ||||||
| GG+AG | 95 (0.872) | 88 (0.830) | ||||
| AA | 14 (0.128) | 18 (0.170) | 0.720 (0.338-1.535) | 0.394 | 1.027 (0.445-2.371) | 0.95 |
| Alleles | ||||||
| G | 142 (0.651) | 132 (0.623) | ||||
| A | 76 (0.349) | 80 (0.377) | 0.883 (0.596-1.309) | 0.536 |
Adjusted for clinical characteristics.
SNP, single nucleotide polymorphism.
Table 6.
Association of UTS2 gene polymorphisms Ser89Asn with types of myocardial infarction stratified by sex
| rs2890565 | STEMI | Non-STEMI | OR (95% CI) | P | aOR (95% CI) | P |
|---|---|---|---|---|---|---|
| Male | ||||||
| Additive | ||||||
| GG | 50 (0.532) | 37 (0.493) | ||||
| AG | 41 (0.436) | 29 (0.387) | 1.046 (0.553-1.980) | 0.89 | 1.141 (0.572-2.275) | 0.708 |
| AA | 3 (0.032) | 9 (0.120) | 0.247 (0.062-0.975) | 0.034 | 0.208 (0.049-0.890) | 0.034 |
| Dominant | ||||||
| GG | 50 (0.532) | 37 (0.493) | ||||
| AG+AA | 44 (0.468) | 38 (0.507) | 0.857 (0.467-1.573) | 0.618 | 0.879 (0.460-1.680) | 0.696 |
| Recessive | ||||||
| GG+AG | 91 (0.968) | 66 (0.880) | ||||
| AA | 3 (0.032) | 9 (0.120) | 0.242 (0.063-0.927) | 0.027 | 0.197 (0.047-0.824) | 0.026 |
| Alleles | ||||||
| G | 141 (0.750) | 103 (0.687) | ||||
| A | 47 (0.250) | 47 (0.313) | 0.730 (0.453-1.178) | 0.197 | ||
| Female | ||||||
| Additive | ||||||
| GG | 9 (0.60) | 16 (0.485) | ||||
| AG | 6 (0.40) | 17 (0.515) | 0.627 (0.182-2.164) | 0.459 | 0.579 (0.132-2.541) | 0.469 |
| AA | 0 (0.00) | 0 (0.000) | ||||
| Dominant | ||||||
| GG | - | - | - | - | - | - |
| AG+AA | - | - | - | - | - | - |
| Recessive | ||||||
| GG+AG | - | - | - | - | - | - |
| AA | - | - | - | - | - | - |
| Alleles | ||||||
| G | 24 (0.80) | 49 (0.742) | ||||
| A | 6 (0.20) | 17 (0.258) | 0.721 (0.252-2.061) | 0.54 |
Adjusted for clinical characteristics.
Linkage disequilibrium was analyzed with SHEsis software
We used SHEsis software to analyze linkage disequilibrium between Thr21Met locus and Ser89Asn locus based on R2 value. The R2 value was 0.12, which indicated that Thr21Met locus and Ser89Asn locus were not in linkage disequilibrium.
Discussion
The present study did not identify statistically significant differences between the case and control groups, neither with regard to the frequency of genotype variant at the Ser89Asn locus nor at the Thr21Met locus. When stratified by sex, differences in genotype distribution of polymorphism Ser89Asn were only seen in female subjects. For subjects with myocardial infarction, we identified statistically significant differences between the ST-segment elevation myocardial infarction and non ST-segment elevation myocardial infarction groups. Differences in genotype distribution of polymorphism Ser89Asn not Thr21Met were seen in both additive tested inheritance model and recessive tested inheritance model. When stratified by sex, differences in genotype distribution of polymorphism Ser89Asn were only seen in male subjects. Our study suggests that Ser89Asn (S89N) polymorphisms of the UTS2 gene were significantly associated with coronary heart disease and myocardial infarction in Chinese population. Additionally, we demonstrated that Genotype Asn89Asn may imply a potential benefit role for myocardial infarction.
The gene for urotensin II (UTS2) is located at chromosome 1p36-p32 and encodes a somatostatin-like peptide identified as a potent vasoconstrictor. The vasoconstrictive effect of U-II is mediated via GPR14, also known as UT, which is the specific receptor of U-II. U-II and its receptor UT are widely distributed in vascular and cardiac tissues [2] and they play an important role in cardiovascular diseases such as hypertension [9,10], coronary artery disease [3], and heart failure [11]. In addition to clinical evidence, in vitro and in vivo studies have further implicated U-II in the development of atherosclerosis and coronary artery disease, contributing to the (patho) physiological regulation of cardiovascular homeostasis in humans. Migration and proliferation of vascular smooth muscle cells are important features of coronary atherosclerosis. U-II has been shown to induce proliferation of arterial smooth muscle cells through the activation of RhoA/Rho-kinase signal transduction pathway, which is one of the hallmark pathological features of atherosclerosis [12]. In addition to its mitogenic effects, U-II has been shown to stimulate hyperlipidemia by enhancing depot lipase activity in Coho salmon [13]. There is also emerging evidence for the U-II’s ability to regulate pancreatic function by inhibiting insulin release and thus predisposing to diabetes [14], a potential risk factor to atherosclerosis. Moreover, U-II was also shown to act synergistically with mildly oxidized LDL, a factor known to be associated with atherosclerosis, in inducing DNA synthesis in vascular smooth muscle cells via the cSrc/PKC/MAPK pathway [15]. Another characteristic of coronary atherosclerosis is the deposition of extracellular matrix proteins, and in particular collagen type I in the sub endothelial layer [16]. Other data have demonstrated that U-II increases precollege production by cardiac fibroblasts [17] and endothelial cells [18]. In the previous study, we examined the direct effects of U-II on collagen synthesis in cultured rat aortic smooth muscle cells (VSMCs) [19]. We demonstrated that U-II stimulates collagen secretion and increases the mRNA expression of collagen I in a concentration-dependent manner. TGF-β1 is a potent profibrotic factor that is implicated in pathological fibrosis and that potentially acts in cardiovascular fibrosis. We explored the role of TGF-β1 in the effects of U-II on collagen synthesis in VSMCs. We found that U-II promotes collagen I production in VSMCs through mechanisms that include TGF-β1 expression and secretion [20]. In another study we provided evidence for the first time that U-II could induce VEGF expression in adventitial fibroblasts by concentration-dependent manner and that VEGF was involved in UII-induced collagen synthesis [21]. Ames et al. demonstrated increased U-II and UT in atheroma of coronary arteries [1]. Maguire et al. have demonstrated the increased expression of U-II in atherosclerotic human coronary artery. They found that, in atherosclerotic coronary artery, U-II immunoreactivity localized to regions of macrophage infiltration [4]. Hassan et al. investigated the expression of U-II in coronary arteries of normal control subjects and patients with coronary atherosclerosis and demonstrated up regulation of U-II in endothelial, myointimal and medial smooth cells of atherosclerotic human coronary arteries. In general, U-II mRNA expression shown by in situ hybridization matched that of U-II immunoreactivity shown by immunohistochemistry. The expression of endothelial U-II appears to be associated with the presence of inflammation and lipid deposition. Greater expression of U-II was noted in endothelial cells of lesions with sub endothelial inflammation or fibro fatty lesion compared with that of endothelial cells underlined by dense fibrosis or minimal intimal thickening [3].
In summary, these observations demonstrate abundant expression of the vasoactive peptide U-II in coronary atherosclerosis. The expression of endothelial U-II appears to be associated with the presence of inflammation and lipid deposition suggesting that increased expression of U-II may contribute to arterial remodeling associated with coronary atherosclerosis, by increasing cellular proliferation and extracellular matrix deposition. These findings demonstrate up regulation of U-II in endothelial and sub endothelial cells of atherosclerotic human coronary arteries, and suggest a possible role for U-II in the pathogenesis of coronary atherosclerosis.
The effects of U-II on cardiovascular function in pathological conditions are still controversial and remain to be explained. It has been reported that U-II can protect the heart against ischemia-reperfusion (I/R) through enhancing coronary flow and reducing cardiac contractility [22]. A recent study found that U-II receptor antagonist SB-710411 protects rat heart against ischemia-reperfusion injury via RhoA/ROCK pathway [23]. Plasma U-II has been observed to be raised in patients with acute myocardial infarction and a lower U-II response is associated with more severe injury of myocardium [7], indicating a possible cardio protective role for this peptide. Reactive oxygen species (ROS) and antioxidant pathway are involved in the protective effect of U-II [24]. To explore the potential molecular mechanisms, Gong et al. treated cultured cardiomyocytes with H2O2 to induce oxidative stress and found that U-II pretreatment significantly reduced the number of apoptotic cardiomyocytes induced by H2O2 by partly abolishing the increase of pro-apoptotic protein Bax and the decrease of anti-apoptotic protein Bcl-2 in cardiomyocytes induced by H2O2. U-II rapidly promoted the phosphorylation of ERK and upregulated CSE level and H22S production, which in turn activated ERK signaling to protect cardiomyocytes from apoptosis under oxidative stress. These results suggest that increased plasma U-II level may protect cardiomyocytes at the early-phase of acute myocardial infarction in patients.
Given that the effects of single polymorphisms on the development of MI are likely to be small, the association between a given polymorphism and the prevalence of MI might be influenced by the absence or presence of conventional coronary risk factors. Furthermore, conventional risk factors, such as hypertension, hypercholesterolemia, and diabetes mellitus, may themselves have genetic components and these components may interact with gene polymorphisms associated with MI. Many studies have focused on U-II polymorphisms as risk factors for these health outcomes, it has been suggested that the SNPs in UTS2 gene (Thr21Met and/or Ser89Asn) may affect susceptibility to hypertension, diabetes, pre-eclampsia, systemic sclerosis and Behcet’s disease. In this study, we aimed to investigate the significance of UTS2 gene polymorphism in pathogenesis of myocardial infarction in Chinese population. Our study found that Ser89Asn (S89N) polymorphisms of the UTS2 gene were significantly associated with myocardial infarction in Chinese population. Additionally, we demonstrated that Genotype Asn89Asn may imply a potential benefit role for myocardial infarction. Several studies on the SNPs in UTS2 gene have suggested a role of U-II in the development of diabetes. It was reported that there were statistical differences in the allele frequency and genotype distribution of Thr21Met and Ser89Asn polymorphisms in UTS2 gene between case (subjects of Type 2 diabetes) and control groups suggesting Thr21Met may be associated with type 2 diabetes in Han people of China [25-28]. Ser89Asn, but not Thr21Met, was reported to be significantly associated with susceptibility of developing type 2 diabetes, increased plasma insulin level, and higher homeostasis model assessment of insulin resistance index (HOMA-IR) in Japanese subjects [29-31]. Ong et al. [32] also showed that plasma U-II level was related to Ser89Asn, but not to Thr21Met, in Hong Kong Chinese (29 hypertensive and 54 normotensive unrelated subjects). Ser89Asn was also significantly associated with higher fasting plasma insulin levels HOMA-IR and HOMA-β% in that study. However, neither Thr21Met nor Ser89Asn was associated with hypertension in this study. Yi et al. assessed the significance of polymorphism of the gene for UTS2 as risk factors for essential hypertension in two populations (Han and Dongxiang Populations) from north-western China and found that in the Han population there were significant differences in Ser89Asn genotype and allele frequencies between patients and controls [33]. Recently, Liu et al. explored the potential association between the UTS2 gene polymorphisms Ser89Asn and essential hypertension (EH) risk in a larger population of Northern Han Chinese. Significant associations were found in genotype distribution and allele frequencies between hypertensive and normotensive groups. However, when the data were subjected to logistic regression analysis after adjusting for confounding risk variables, no significant associations were found [34]. Yoshiji et al. [35] aimed to assess the genetic risk for myocardial infarction (MI) in individuals with or without conventional coronary risk factors. To a certain extent, our results on Ser89Asn are inconsistent with that of Yoshiji et al. We did not identify statistically significant differences between the myocardial infarction and coronary heart disease groups, neither with regard to the frequency of genotype variant at the Ser89Asn locus nor at the Ser89Asn locus. Yoshiji et al. have shown that the G→A (Ser89Asn) polymorphism of UTS2 was associated with the prevalence of MI only in hypertensive individuals but not other subjects (with or without hypercholesterolemia/diabetes mellitus). The possible explanations may be that the control subjects in our study are all diagnosed with coronary heart disease with or without conventional coronary risk factors, while the control subjects in Yoshiji et al. study only comprised individuals who visited the outpatient clinics of participating hospitals for an annual health checkup. They had no history of CHD, peripheral arterial occlusive disease, or other atherosclerotic diseases. In another study by Yoshiji et al. [36], they revealed that after multivariable logistic regression analysis with adjustment for age, sex, and the prevalence of hypertension, diabetes mellitus, and hypercholesterolemia, the G→A (Ser89Asn) polymorphism of UTS2 (all genetic models) were significantly associated with MI in individuals with metabolic syndrome. The control subjects comprised individuals who visited the outpatient clinics of participating hospitals for an annual health checkup or who were community-dwelling individuals recruited to the prospective cohort study. They had no history of coronary heart disease, which is different from our control subjects who were diagnosed with coronary heart disease. In our study, when stratified by sex, differences in genotype distribution of polymorphism Ser89Asn was only seen in female subjects in both additive tested inheritance model (OR=0.257, 95% CI: 0.074-0.896, P=0.033) and recessive tested inheritance model (OR=0.280, 95% CI: 0.082-0.955, P=0.042). For subjects with myocardial infarction, we identified statistically significant differences between the ST-segment elevation myocardial infarction and non ST-segment elevation myocardial infarction groups. Differences in genotype distribution of polymorphism Ser89Asn not Thr21Met were seen in both additive tested inheritance model (OR=0.202, 95% CI: 0.049-0.833, P=0.027) and recessive tested inheritance model (OR=0.208, 95% CI: 0.052-0.835, P=0.027). When stratified by sex, differences in genotype distribution of polymorphism Ser89Asn was only seen in male subjects in both additive tested inheritance model (OR=0.208, 95% CI: 0.049-0.890, P=0.034) and recessive tested inheritance model (OR=0.197, 95% CI: 0.047-0.824, P=0.026). Ser89Asn (S89N) polymorphisms of the UTS2 gene were significantly associated with coronary heart disease and myocardial infarction in Chinese population. Additionally, we demonstrated that Genotype Asn89Asn may imply a potential benefit role for myocardial infarction.
In conclusion, our results suggest that subjects with Ser89Asn in the UTS2 gene may modify individual susceptibility to myocardial infarction in the Chinese population. It is hoped that the results presented in this report will lead to additional basic and clinical studies and further studies with large-scale are required to verify these findings in different ethnic groups.
Limitations
There are some limitations of this study. One of them is that the number of study subjects was relatively small and from single ethnic group. Therefore, large population studies are necessary to confirm the role of the UTS2 gene as a risk factor for myocardial infarction. Another limitation of this study is that although we applied a rigorous epidemiological design in selecting study subjects and adjusted for confounding factors in further statistical analysis to minimize the potential biases, inherent selection bias cannot be completely excluded.
Conclusion
In this case-control study, we have shown for the first time that Ser89Asn polymorphisms of the UTS2 gene were significantly associated with myocardial infarction in Chinese population. Therefore, blockade of the Urotensin-II system may become a promising therapeutic strategy for myocardial infarctions with specific UTS2 polymorphism. Additionally, we demonstrated that Genotype Asn89Asn polymorphisms may imply a potential benefit role for myocardial infarction. This knowledge and proposed mechanism may open up new possibilities for the discovery of the way of pathogenesis and treatment of myocardial infarction.
Acknowledgements
This project was supported by the “12th Five-Year National Science and Technology Support Program” (No.2011BAI11B06). Conflict of interest: There were no conflicts of interest for this study. Ethical approval: For research involving human participants, informed consent has been obtained from the patients or the guardian of the patients. The research has been approved by the Ethics Committee of the Peking University First Hospital.
Disclosure of conflict of interest
None.
References
- 1.Ames RS, Sarau HM, Chambers JK, Willette RN, Aiyar NV, Romanic AM, Louden CS, Foley JJ, Sauermelch CF, Coatney RW, Ao Z, Disa J, Holmes SD, Stadel JM, Martin JD, Liu WS, Glover GI, Wilson S, McNulty DE, Ellis CE, Elshourbagy NA, Shabon U, Trill JJ, Hay DW, Ohlstein EH, Bergsma DJ, Douglas SA. Human urotensin-II is a potent vasoconstrictor and agonist for the orphan receptor GPR14. Nature. 1999;401:282–286. doi: 10.1038/45809. [DOI] [PubMed] [Google Scholar]
- 2.Matsushita M, Shichiri M, Imai T, Iwashina M, Tanaka H, Takasu N, Hirata Y. Co-expression of urotensin II and its receptor (GPR14) in human cardiovascular and renal tissues. J Hypertens. 2001;19:2185–2190. doi: 10.1097/00004872-200112000-00011. [DOI] [PubMed] [Google Scholar]
- 3.Hassan GS, Douglas SA, Ohlstein EH, Giaid A. Expression of urotensin-II in human coronary atherosclerosis. Peptides. 2005;26:2464–2472. doi: 10.1016/j.peptides.2005.05.028. [DOI] [PubMed] [Google Scholar]
- 4.Maguire JJ, Kuc RE, Wiley KE, Kleinz MJ, Davenport AP. Cellular distribution of immunoreactive urotensin-II in human tissues with evidence of increased expression in atherosclerosis and a greater constrictor response of small compared to large coronary arteries. Peptides. 2004;25:1767–1774. doi: 10.1016/j.peptides.2004.01.028. [DOI] [PubMed] [Google Scholar]
- 5.Cozzolino M, Biondi ML, Galassi A, Turri O, Brancaccio D, Gallieni M. Matrix metalloproteinase-1 and matrix metalloproteinase-3 gene promoter polymorphisms are associated with mortality in haemodialysis patients. Nephrol Dial Transplant. 2009;24:2207–2212. doi: 10.1093/ndt/gfp061. [DOI] [PubMed] [Google Scholar]
- 6.Joyal D, Huynh T, Aiyar N, Guida B, Douglas S, Giaid A. Urotensin-II levels in acute coronary syndromes. Int J Cardiol. 2006;108:31–35. doi: 10.1016/j.ijcard.2005.04.001. [DOI] [PubMed] [Google Scholar]
- 7.Khan SQ, Bhandari SS, Quinn P, Davies JE, Ng LL. Urotensin II is raised in acute myocardial infarction and low levels predict risk of adverse clinical outcome in humans. Int J Cardiol. 2007;117:323–328. doi: 10.1016/j.ijcard.2006.05.016. [DOI] [PubMed] [Google Scholar]
- 8.Babinska M, Holecki M, Prochaczek F, Owczarek A, Kokocinska D, Chudek J, Wiecek A. Is plasma urotensin II concentration an indicator of myocardial damage in patients with acute coronary syndrome? Arch Med Sci. 2012;8:449–454. doi: 10.5114/aoms.2012.29400. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Cheung BM, Leung R, Man YB, Wong LY. Plasma concentration of urotensin II is raised in hypertension. J Hypertens. 2004;22:1341–1344. doi: 10.1097/01.hjh.0000125452.28861.f1. [DOI] [PubMed] [Google Scholar]
- 10.Mori N, Hirose T, Nakayama T, Ito O, Kanazawa M, Imai Y, Kohzuki M, Takahashi K, Totsune K. Increased expression of urotensin II-related peptide and its receptor in kidney with hypertension or renal failure. Peptides. 2009;30:400–408. doi: 10.1016/j.peptides.2008.09.021. [DOI] [PubMed] [Google Scholar]
- 11.Russell FD, Meyers D, Galbraith AJ, Bett N, Toth I, Kearns P, Molenaar P. Elevated plasma levels of human urotensin-II immunoreactivity in congestive heart failure. Am J Physiol Heart Circ Physiol. 2003;285:H1576–1581. doi: 10.1152/ajpheart.00217.2003. [DOI] [PubMed] [Google Scholar]
- 12.Sauzeau V, Le Mellionnec E, Bertoglio J, Scalbert E, Pacaud P, Loirand G. Human urotensin II-induced contraction and arterial smooth muscle cell proliferation are mediated by RhoA and Rho-kinase. Circ Res. 2001;88:1102–1104. doi: 10.1161/hh1101.092034. [DOI] [PubMed] [Google Scholar]
- 13.Sheridan MA, Plisetskaya EM, Bern HA, Gorbman A. Effects of somatostatin-25 and urotensin II on lipid and carbohydrate metabolism of coho salmon, Oncorhynchus kisutch. Gen Comp Endocrinol. 1987;66:405–414. doi: 10.1016/0016-6480(87)90251-6. [DOI] [PubMed] [Google Scholar]
- 14.Silvestre RA, Rodriguez-Gallardo J, Egido EM, Marco J. Inhibition of insulin release by urotensin II--a study on the perfused rat pancreas. Horm Metab Res. 2001;33:379–381. doi: 10.1055/s-2001-15414. [DOI] [PubMed] [Google Scholar]
- 15.Watanabe T, Pakala R, Katagiri T, Benedict CR. Synergistic effect of urotensin II with mildly oxidized LDL on DNA synthesis in vascular smooth muscle cells. Circulation. 2001;104:16–18. doi: 10.1161/hc2601.092848. [DOI] [PubMed] [Google Scholar]
- 16.Ross R. Atherosclerosis--an inflammatory disease. N Engl J Med. 1999;340:115–126. doi: 10.1056/NEJM199901143400207. [DOI] [PubMed] [Google Scholar]
- 17.Tzanidis A, Hannan RD, Thomas WG, Onan D, Autelitano DJ, See F, Kelly DJ, Gilbert RE, Krum H. Direct actions of urotensin II on the heart: implications for cardiac fibrosis and hypertrophy. Circ Res. 2003;93:246–253. doi: 10.1161/01.RES.0000084382.64418.BC. [DOI] [PubMed] [Google Scholar]
- 18.Wang H, Mehta JL, Chen K, Zhang X, Li D. Human urotensin II modulates collagen synthesis and the expression of MMP-1 in human endothelial cells. J Cardiovasc Pharmacol. 2004;44:577–581. doi: 10.1097/00005344-200411000-00010. [DOI] [PubMed] [Google Scholar]
- 19.Saleh D, Furukawa K, Tsao MS, Maghazachi A, Corrin B, Yanagisawa M, Barnes PJ, Giaid A. Elevated expression of endothelin-1 and endothelin-converting enzyme-1 in idiopathic pulmonary fibrosis: possible involvement of proinflammatory cytokines. Am J Respir Cell Mol Biol. 1997;16:187–193. doi: 10.1165/ajrcmb.16.2.9032126. [DOI] [PubMed] [Google Scholar]
- 20.Zhao J, Ding W, Song N, Dong X, Di B, Peng F, Tang C. Urotensin II-induced collagen synthesis in cultured smooth muscle cells from rat aortic media and a possible involvement of transforming growth factor-beta1/Smad2/3 signaling pathway. Regul Pept. 2013;182:53–58. doi: 10.1016/j.regpep.2012.12.006. [DOI] [PubMed] [Google Scholar]
- 21.Song N, Ding W, Chu S, Zhao J, Dong X, Di B, Tang C. Urotensin II stimulates vascular endothelial growth factor secretion from adventitial fibroblasts in synergy with angiotensin II. Circ J. 2012;76:1267–1273. doi: 10.1253/circj.cj-11-0870. [DOI] [PubMed] [Google Scholar]
- 22.Prosser HC, Forster ME, Richards AM, Pemberton CJ. Urotensin II and urotensin II-related peptide (URP) in cardiac ischemia-reperfusion injury. Peptides. 2008;29:770–777. doi: 10.1016/j.peptides.2007.08.013. [DOI] [PubMed] [Google Scholar]
- 23.Luo SY, Chen S, Qin YD, Chen ZW. Urotensin-Receptor Antagonist SB-710411 Protects Rat Heart against Ischemia-Reperfusion Injury via RhoA/ROCK Pathway. PLoS One. 2016;11:e0146094. doi: 10.1371/journal.pone.0146094. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Gong H, Chen Z, Zhang X, Li Y, Zhang J, Chen Y, Ding Y, Zhang G, Yang C, Zhu Y, Zou Y. Urotensin II Protects Cardiomyocytes from Apoptosis Induced by Oxidative Stress through the CSE/H2S Pathway. Int J Mol Sci. 2015;16:12482–12498. doi: 10.3390/ijms160612482. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Sun HX, Du WN, Zuo J, Wu GD, Shi GB, Shen Y, Qiang BQ, Yao ZJ, Hang JM, Wang H, Huang W, Chen Z, Fang FD. [The association of two single nucleotide polymorphisms in PRKCZ and UTS2 respectively with type 2 diabetes in Han people of northern China] . Zhongguo Yi Xue Ke Xue Yuan Xue Bao. 2002;24:223–227. [PubMed] [Google Scholar]
- 26.Zhu F, Ji L, Luo B. [The role of urotensin II gene in the genetic susceptibility to type 2 diabetes in Chinese population] . Zhonghua Yi Xue Za Zhi. 2002;82:1473–1475. [PubMed] [Google Scholar]
- 27.Wu XJ, Qin L, Zhou L, Zhou JY, Cheng JL, Shen MY, Chang JS, Du YF, Hu JH. Research on polymorphism and haplotype of UTS2 gene and type 2 diabetes. Chin J Public Health. 2008:1081–1083. [Google Scholar]
- 28.W HQ, Z WJ, Z JL. A study of the association between polymorphism of urotensin II gene and the onset of type 2 diabetes mellitus in Chinese subjects. Chin J Diabetes. 2007:26–28. [Google Scholar]
- 29.Wenyi Z, Suzuki S, Hirai M, Hinokio Y, Tanizawa Y, Matsutani A, Satoh J, Oka Y. Role of urotensin II gene in genetic susceptibility to Type 2 diabetes mellitus in Japanese subjects. Diabetologia. 2003;46:972–976. doi: 10.1007/s00125-003-1145-1. [DOI] [PubMed] [Google Scholar]
- 30.Suzuki S, Wenyi Z, Hirai M, Hinokio Y, Suzuki C, Yamada T, Yoshizumi S, Suzuki M, Tanizawa Y, Matsutani A, Oka Y. Genetic variations at urotensin II and urotensin II receptor genes and risk of type 2 diabetes mellitus in Japanese. Peptides. 2004;25:1803–1808. doi: 10.1016/j.peptides.2004.03.030. [DOI] [PubMed] [Google Scholar]
- 31.Saez ME, Smani T, Ramirez-Lorca R, Diaz I, Serrano-Rios M, Ruiz A, Ordonez A. Association analysis of urotensin II gene (UTS2) and flanking regions with biochemical parameters related to insulin resistance. PLoS One. 2011;6:e19327. doi: 10.1371/journal.pone.0019327. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Ong KL, Wong LY, Man YB, Leung RY, Song YQ, Lam KS, Cheung BM. Haplotypes in the urotensin II gene and urotensin II receptor gene are associated with insulin resistance and impaired glucose tolerance. Peptides. 2006;27:1659–1667. doi: 10.1016/j.peptides.2006.02.008. [DOI] [PubMed] [Google Scholar]
- 33.Yi L, Gu YH, Wang XL, An LZ, Xie XD, Shao W, Ma LY, Fang JR, An YD, Wang F, Zhang DL. Association of ACE, ACE2 and UTS2 polymorphisms with essential hypertension in Han and Dongxiang populations from north-western China. J Int Med Res. 2006;34:272–283. doi: 10.1177/147323000603400306. [DOI] [PubMed] [Google Scholar]
- 34.Liu Y, Zhang B, Liu JL, Wang ZG, Wen J, Li M, Wang H, Wang LJ, Wen SJ. Role of Urotensin II gene polymorphisms in susceptibility to the development of essential hypertension in a Northern Han Chinese population. Journal of Cardiovascular & Pulmonary Diseases. 2013:239–242. [Google Scholar]
- 35.Nishihama K, Yamada Y, Matsuo H, Segawa T, Watanabe S, Kato K, Yajima K, Hibino T, Yokoi K, Ichihara S, Metoki N, Yoshida H, Satoh K, Nozawa Y. Association of gene polymorphisms with myocardial infarction in individuals with or without conventional coronary risk factors. Int J Mol Med. 2007;19:129–141. [PubMed] [Google Scholar]
- 36.Oguri M, Kato K, Yokoi K, Itoh T, Yoshida T, Watanabe S, Metoki N, Yoshida H, Satoh K, Aoyagi Y, Nishigaki Y, Tanaka M, Nozawa Y, Yamada Y. Association of genetic variants with myocardial infarction in Japanese individuals with metabolic syndrome. Atherosclerosis. 2009;206:486–493. doi: 10.1016/j.atherosclerosis.2009.02.037. [DOI] [PubMed] [Google Scholar]