January 01, 2018. doi:10.12123/npcd201801002
BIOWED January 01, 2018
Metrics: PDF views | HTML views
Liu Liu, Lingli Zhang*
School of Public Health, Guangdong Pharmaceutical University, Guangzhou, Guangdong, China.
Communication: firstname.lastname@example.org (Liu Liu)
Keywords: Hypertension; Leptin receptor; Gene polymorphisms; Risk; Meta-analysis
Received: September 01,2017 Accept: November 01,2017 published:January 01,2018
Studies have shown that the single nucleotide polymorphisms of the leptin receptor gene may increase the risk of essential hypertension. However, there still no consistent conclusion has been drawn. In this study, the meta-analysis was used to summarized the association in LEPR Gln223Arg, Lys109Arg polymorphisms and the risk of essential hypertension based on 16 case–control studies retrieved from Pubmed, Web of Science, CNKI, VIP, CBM. Statistical analyses were carried out with the Stata12.0 and Revman5.3 software. In the 13 articles that covered the Gln223Arg polymorphism, a significant association was found between Gln223Arg gene polymorphism and essential hypertension in allelic model (OR=1.36; 95% CI:1.10-1.69), dominant genetic model (OR=1.48; 95% CI:1.14-1.93) and recessive genetic model (OR=1.27; 95%CI: 1.04-1.55). In the 10 articles which were related to Lys109Arg polymorphism, no significant association was found between the Lys109Arg polymorphism and essential hypertension risk under allelic model (OR= 1.02; 95% CI: 0.86-1.21), dominant genetic model (OR=1.00; 95% CI:0.81-1.23) and recessive genetic model (OR=1.11; 95% CI:0.86-1.44). In summary, the variation of Gln223Arg locus can increase the risk of essential hypertension significantly, but there is no evidence shows association between Lys109Arg polymorphism and essential hypertension.
Essential hypertension is a worldwide health problem. It is not only a risk factor for stroke, myocardial infarction, and kidney disease, but also often comes with other metabolic diseases like high cholesterol, obesity, insulin resistance and so on. Hypertension is currently considered a disease caused by varies factors, including the environment and lifestyle, such as diet, exercise, smoking and drinking, as well as the impact of polygenes.
In recent years, more and more people believe that hypertension is associated with leptin and its receptor. Previous studies have shown that leptin and leptin receptor can regulate blood pressure and adipose tissue metabolism directly, or indirectly lead to obesity and hypertension(1). In addition, Leptin can increase the sympathetic nerve activity through the leptin receptor and further affect the blood pressure. The human leptin receptor gene (LEPR) is localized at 1P31, which consists of 20 exons and 19 introns. Studies have shown that mutation of LEPR gene may have a direct impact on the biological function of leptin(2). As for Gln223Arg (rs1137101) and Lys109Arg (rs1137100), which are located on the 6th and 4th exons of leptin receptor gene, many studies were done on the relationship between hypertension and these two loci. However, no conclusion has been drawn. To further clarify this relationship, we performed a meta-analysis on the published studies based on the principles of evidence-based medicine and explored whether the mutation of Gln223Arg and Lys109Arg locus can affect the incidence risk of essential hypertension preliminarily.
Materials and Methods
Computer-based retrieval along with manual retrieval were done, and all the published articles related to the association between essential hypertension and leptin receptor gene were collected from databases including: CNKI, VIP, CBM, Pubmed, Web of Science. Using keywords: "polymorphism", "variant", "variation", "mutation", "SNP", "hypertension", "hypertensive", "blood pressure", "BP", "Leptin Receptor", "LEPR Protein", "CD295 Antigens", "LEPR". The searching date is updated to August 1st, 2017.
Data Extraction and Quality
The inclusion criteria were as follows:(1) Articles included case-control studies which reported the relationship between the polymorphism of leptin receptor gene and hypertension; (2) Articles that covered LEPR Gln223Arg (rs1137101) polymorphism or Lys109Arg (rs1137100) polymorphism; (3) All the cases have been included were essential hypertension (systolic blood pressure ≥ 140mm Hg or diastolic blood pressure ≥ 90mm Hg; (4) The frequency distribution of genotypes in both case group and control group were provided either directly or indirectly; (5) For studies with repetitive data, include the one which provided the largest amount of information.
The exclusion criteria were as follows: (1) Studies were experimental research or review; (2) Studies were designed as cohort studies or cross-sectional studies; (3) Subjects in the studies were secondary hypertension patients; (4) Studies did not report the comparison of hypertension and non-hypertension; (5) There was a defect in the study design, or the data provided in the results were incomplete or incorrect, or an inappropriate statistical method was used.
Inclusion and Exclusion Criteria
Based on the pre-design data extraction table, one of the two researchers responsible for extracting and entering data, the other make sure everything is on the right track. In case of disagreement, it can be discussed by the two researchers or turn to the third party.
Data that were extracted from each study included: name of the first author, year of publication, race of the subjects, age, gender, blood pressure, the matching variable, research method, source of the case group and the control group, as well as the frequency distribution of each genotype and loci in both case group and control group. In the meantime, Chi-square was used to test whether the frequency of genotype distribution in case group and control group were corresponding to the Hardy-Weinberg Equilibrium. The quality of the included case-control studies was evaluated using the Newcastle-Ottawa Scale (NOS) document.
Stata12.0 and Review Manager5.3 were used to analyze the included data which were input and analyzed independently by two reviewers. In this study, continuous variables, such as age, were expressed in terms of "mean ± standard deviation'. The incidence risk of hypertension in different genotype populations was expressed as Odds Ratio (OR) and its 95% Confidence Interval (CI). P-value which was less than 0.05 indicated statistical significance.
I2 and Q tests were used to perform heterogeneity testing of the included literature. If I2<50% and P>0.10, then it was considered as heterogeneity and fixed effect model (the Mantele–Haenszel method) should be used for the pooled OR. Otherwise, a random-effects model would be applied (the DerSimonian and Laird method). Moreover, the source of heterogeneity should be found. If there was no way to find out the source of heterogeneity, then no meta-analysis should be performed but using description analysis instead. The funnel plot was used to assess the publication bias.
At the same time, the Hardy-Weinberg Equilibrium was tested for the control group using Chi-square test., if P>0.05, then it indicated the chosen population in this study is consistent with the genetic balance status. According to the matching condition of the case and control groups or to the HWE of the control group, a sub-group analysis was performed to find out the source of heterogeneity. By sequentially omitting one article at a time, the influence of a single article on pooled effect estimation was evaluated in the sensitivity analysis. The publication bias was estimated by the funnel plot, in which the funnel plot asymmetry was assessed by the method of Egger's linear regression test, a linear regression approach for measuring funnel plot asymmetry on the natural logarithm scale of OR.
Study Identification and Selection
According to the retrieval strategy, 614 articles were included initially after retrieving from online database and endnote auto duplicate checking, as well as manual duplicate checking. Next, applied the exclusion and inclusion criteria mentioned in 2.2 on the titles and abstract, 490 articles were excluded, including reviews, animal experimental studies, clinical randomized controlled trials, conference reports, letters and so on. Finally, read through the full text carefully, excluding the studies with repetitive data, not case-control or with subjects that are not essential hypertension patients. Eventually, 16 studies were included into this meta-analysis(3-18). A flow diagram is provided in Figure 1 for more details of the articles screening process.
Figure.1 A flow diagram of the study selection process.
All included articles were published between 2007 and 2017, have covered 5037 cases in the case group and 2987 in the control group. Among these articles, 13 of them were related to LEPR Gln223Arg polymorphism(4-7,10-14). Within these 13 articles, the frequency distribution of the control group in 3 of the articles is not consistent with the HWE(8,14,15); There are 11 articles talked about leptin receptor Lys109Arg locus polymorphism(3,7-12,14,15,17,18). Within these 11 articles, the frequency distribution of the control group in 3 of the articles is not consistent with the HWE(3,7,15). Refer to Table.1 for more basic characteristics of the included articles.
Table.1 Characters of included studies.
Note: Abbreviating Words: NA, not available; FBG, fasting blood-glucose; BG, blood-glucose; TC, total cholesterol; TG, triglyceride; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol.
Beside the samples in study by Wu, JH2017(14) were from Chinese residents nutrition and health status survey, the cases in the rest of the studies were collected continuously and all from the hospitals. The control groups were chosen from the same hospitals with non-essential hypertension. The diagnose criteria was the same and the process and outcome were in fine detail. (Blood pressure was measured after a 10 min rest at a sitting position. Hypertension was defined at a mean systolic blood pressure ≥140 mmHg and/or a mean diastolic blood pressure ≥90 mmHg. Patients taking any antihypertensive medication were defined as hypertensive.(17)).In addition, there is no explanation on the matching scale of the case group and the control group in the study by Shi, YW2007. However, based on the gender and age distribution, Chi-square test showed there was no statistical difference between the case group and the control group (P>0.05).
By calculating the pooled OR and its 95% CI of the allele genetic model (A allele vs G allele), recessive genetic model (AA vs AG+GG) and dominant genetic model (AA+AG vs GG), a meta-analysis on the LEPR Gln223Arg and Lys109Arg polymorphisms was performed(19).
Table. 2 The distribution of genotypes and alleles of LEPR Gln223Arg in the case and control group, and the results
of meta-analysis for essential hypertension risk under the three genetic models.
Note: OR: odds ratio; 95% CI: 95% confidence interval; P: P values of test for pooled OR; I2, inconsistency index; Phet: P values of heterogeneity test; R: Random model; F: Fixed model.
Association between LEPR Gln223Arg Polymorphism and the Incidence Risk of Hypertension
As shown in Table 2, for the site of LEPR Gln223Arg, heterogeneity test showed that: heterogeneity existed in the allele genetic model and the dominant genetic model, but not in the recessive genetic model. Therefore, the random effects model should be applied to the first two models and for the recessive genetic model, fixed effects model should be used (P<0.00001, I2=81%; P<0.00001, I2=82%; P=0.05, I2=43%). In the meantime, the allele genetic pattern showed that the risk of developing essential hypertension in the population carrying A allele (mutant gene) was 1.36 times higher than those carrying the G gene (wild-type gene) (OR = 1.36; 95% CI: 1.10-1.69); The results of the dominant model showed that the risk of developing essential hypertension in population with (AA+AG) genotype was 1.48 times higher than those with GG genotype (OR=1.48; 95% CI: 1.14, 1.93); In the recessive model, the risk of developing essential hypertension in the AA genotype was 1.27 times higher than that in the (AG+GG) genotype (OR = 1.27; 95% CI:1.04, 1.55).
Association between LEPR Lys109Arg Polymorphism and the Incidence Risk of Hypertension
As shown in Table 3, for the leptin receptor Lys109Arg locus, heterogeneity test showed that there was a moderate heterogeneity both in the allele genetic model and dominant genetic model (P=0.006, I2=59%; P=0.004, I2=61%), and there was no heterogeneity in the recessive genetic model (P=0.55, I2=0%). Therefore, random effects modal was applied to the first two genetic model and fixed effects model was applied to the recessive genetic model. After performing the accumulated analysis on the OR value of included studies, results showed that the effect of LEPR Lys109Arg polymorphism under the 3 genetic models were all not statistical significant.
Table.3 The distribution of genotypes and alleles of LEPR Lys109Arg in the case and control group, and results of
meta-analysis for essential hypertension risk under the three genetic models.
Publication Bias of the Literature
A bias analysis on the results computed from the 3 genetic models of GLn223Arg locus and Lys109Arg locus were performed respectively, where the outcomes were shown in Figure 2 and Figure 3. The funnel plot of each of the genetic models all showed symmetry, means no obvious bias.
Figure. 2 Funnel plot of allele genetic model (A), dominant genetic model (B) and reccesive genetic model (C) for LEPR Gln223Arg polymorphism.
Figure.3 Funnel plot of allele genetic model (A), dominant genetic model (B) and reccesive genetic model (C) for LEPR Lys109Arg polymorphim.
There were 16 references included in this meta-analysis, 13 of which refer to LEPR Gln223Arg polymorphism, including 4,587 patients with essential hypertension and 2,771 patients with non-essential hypertension, all the population were Chinese. Among those, the results of 10 references showed that the incidence of essential hypertension was closely associated with the mutation of LEPR Gln223Arg polymorphism(4-6,10-13,15,16,18), while other 3 references reported no relationship existed, which had no statistical significance. Meanwhile, 11 references refer to the LEPR Lys109Arg polymorphism, including 3,317 patients with essential hypertension and 1,952 patients with non-essential hypertension. During the process of literature screening, the disease type of all the patients were secondary hypertension in the researches of JIANG LY2006(20), Gao L2009(21), Wiedemann, A2010(22), Tennekoon, KH2012(23), therefore, these references were not included in this paper.
For the LEPR Gln223Arg polymorphism, the meta-analysis results of these 13 references showed that there was no heterogeneity in recessive genetic model, while a large heterogeneity in allelic genetic model and dominant genetic model were found. However, all these 3 genetic models had statistical significance, and had a small P value (allelic genetic model: P=0.004; dominant genetic model: P=0.003; recessive genetic model: P=0.02), so we supposed that the mutation of LEPR Gln223Arg would increase the risk of essential hypertension. Besides, HWE was assessed only in controls, since cases may not be in HWE if there was indeed an association between genotype and disease outcome. So the references which are not satisfy with HWE were excluded, and the residual references were analyzed for the sources of heterogeneity, and the result of which showed as below: there were statistical significance in the meta-analysis results of allelic genetic model and dominant genetic model, which indicate that the SNPs associate with the increasing risk of hypertension. But there still were some heterogeneity among these researches, which implied the heterogeneity was not from these 3 references(5,7,15). To specified the sources of heterogeneity, we used the method of sensitivity analysis via excluding the references which from the 13 references one by one, and compared the pooled odds ratio (OR) before excluded and after, and finally found the heterogeneity came from Liu, Y2014(18) and Liu, Y2017(13). In these two researches, Liu, Y2014 and Liu, Y2017 used TaqMan genotyping technique while others used polymerase chain reaction-restricted fragment length polymorphisms PCR-RLFP), which might be the source of heterogeneity.
For the LEPR Lys109Arg polymorphism, the meta-analysis results of these 11 references showed that there was no heterogeneity in recessive genetic model (P=0.55, I2=0%), while there was a medium heterogeneity both in allele genetic model and dominant genetic model (P=0.006, I2=59%; P=0.004, I2=61%). But in the meta-analysis results of all these 3 genetic models, the difference of pooled OR between two groups had no statistical significance. For allelic genetic model, OR=1.02, P=0.83; for dominant genetic model, OR=1.00, P=0.99; for recessive genetic model, OR=1.11, P=0.42. Therefore, meta-analysis results indicated there was no association between the mutation of LEPR Lys109Arg and the incidence risk of essential hypertension.
However, in this meta-analysis, the subjects in the control group were the healthy people in a same period in the hospitals, rather than in communities, which might be lack of representation; in addition, for the lack of the severity state of illness, there might be an influence on the results of research. And our analyzing only selected two SNPs of LERP for the incidence of essential hypertension which was influenced by multi-genes. But in our study, we strictly restrained the patients as essential hypertension, excluded the sources of heterogeneity by sensitivity analysis, and got a stable result, which objectively and fairly reflected the relationship between the SNPs of LEPR Gln223Arg/Lys109Arg and the incidence risk of essential hypertension.
Finally, evidence-based medical analysis is a kind of research that needs to develop continuously. Even though there are deficiency, this study has systematically evaluated this issue based on the existing data resources. Considered the included references are retrospective case-control research with various potentially mixed factors, the result of this study also requires a large sample to deal with, and also need the study with more representative population sample to verify.
In this study, the result of the meta-analysis showed that no significant association between the LEPR Lys109Arg polymorphism and essential hypertension is observed. However, LEPR Gln223Arg polymorphism is highly related to the risk of developing essential hypertension. The risk of essential hypertension increases significantly in the population with A allele gene (mutant gene).
All authors sincerely acknowledge the support given by School of Public Health, Guangdong Pharmaceutical University.
1. M. Muy-Rivera, Y. Ning, I. O. Frederic, S. Vadachkoria, D. A. Luthy, and M. A. Williams. Leptin, soluble leptin receptor and leptin gene polymorphism in relation to preeclampsia risk. Physiol. Res. (2005) 54(2): 167-174.
2. T. Murakami, T. Yamashita, M. Iida, M. Kuwajima, and K. Shima. A short form of leptin receptor performs signal transduction. Biochem. Biophys. Res. Commun. (1997) 231(1): 26-29.
3. Y. W. Shi, Y. B. Zhang, Z. F. Luo, J. L. Zhang, J. C. Liu, J. Yang, Q. Lou, X. H. Zhang, L. Yin, Qiong., and M. M. Fu. Hypertension with insulin resistance and hypertension and leptin receptor gene mutation. Chin. J. Intern. Med. (2007) 46(3): 238-239.
4. N. Wang, Y. L. Li, and B. Cheng. Relationship of Leptin Receptor Gene With Left Ventricular Function and Renal Hemodynamics in Hypertension. Chin. Circ. J (2007) 22(4): 260-262.
5. L. S. Zhao, G. D. Xiang, and Y. Tang. Association of Gln 223 Arg Variant in Leptin Receptor Gene with Hypertension Complicated with Obesity in Wuhan Population. Chin. J. Prev. Control Chronic Non-Communicable Dis (2007) 15(2): 91-94.
6. R. Pan, N. Shi, and E. Zhang. Study on the relationship between Gln223Arg variation in leptin receptor gene and hypertension complicated with obesity in Wenzhou population. Med. J. Chin. PLA. (2008) 33(10): 1192-1194.
7. P. Gu, W. M. Jiang, J. Wang, J. Q. Shao, H. Du, J. D. Zou, Y. Y. Wang, B. Lu, and D. J. Zou. Association of Gln223Arg an d Lysl09Arg variant in leptin receptor gene with hypertension and metabolic characteristics. J. Clin. Intern. Med. (2009) 11): 762-765.
8. B. K. Bao, J. Ying, H. Q. Zhang, and H. X. Su. Study on the relationship between Lys109Arg variation of leptin receptor gene and hypertension complicated with obesity in Wenzhou population. Chin. J. Birth Health Heredity (2010) 12): 20-22.
9. N. Wang, B. Cheng, and L. Y. Ma. Relationship between the mutation of the exon 4 of leptin receptor gene and left ventrtcular function and renal hemodynamics in patients with hypertension. Chin. J. Postgrad. Med. (2010) 33(1): 13-15.
10. Z. Cai, F. Zhao, T. Liu, and Y. Mao. Association of Leptin Receptor Gene Polymorphisms with Essential Hypertension and Plasma Lipid Levels. Tianjin Med. J. (2011) 39(3): 196-198.
11. C. C. Li. Study on the relationship between leptin receptor gene polymorphism and Hypertension and obesity in Han population. Kunming Medical University (2012).
12. X. M. Wang, C. C. Li, and L. X. AY. Study on the Relationship Between Leptin Receptor Gene Polymorphism and Patients of Hypertension with Obesity in Yunnan Area. Chin. J. Arterioscler (2012) 20(11): 1007-1012.
13. Y. Liu, J. L. Liu, Z. G. Wang, B. Zhang, M. Li, C. Li, L. J. Wang, J. Wen, and S. J. Wen. Association of leptin receptor gene polymorphisms and essential hypertension. J. Cardiov. Plum. Dis (2017) 36(3): 155-158.
14. J. H. Wu, Q. Zhuo, Y. Tian, Y. N. Ma, L. Cui, J. H. Piao, and X. G. Yang. Study on the relationship of hypertension in older Han adults with leptin receptor gene rs1137100 and rs1137101 polymorphism. J. Capit. Med. Univ. (2017) 38(3): 439-445.
15. P. Gu, W. Jiang, M. Chen, B. Lu, J. Shao, H. Du, and S. Jiang. Association of leptin receptor gene polymorphisms and essential hypertension in a Chinese population. J. Endocrinol. Invest. (2012) 35(9): 859-865.
16. H. Zheng, N. Xie, H. Xu, J. Huang, X. Xie, and M. Luo. Association of Gln223Arg polymorphism of the leptin receptor with hypertensive left ventricular hypertrophy. Folia Biol. (Praha) (2013) 59(6): 246-252.
17. B. Jiang, Y. Liu, Y. Liu, F. Fang, X. Wang, and B. Li. Association of four insulin resistance genes with type 2 diabetes mellitus and hypertension in the Chinese Han population. Mol. Biol. Rep. (2014) 41(2): 925-933.
18. Y. Liu, Y. Q. Lou, K. Liu, J. L. Liu, Z. G. Wang, J. Wen, Q. Zhao, S. J. Wen, and L. Xiao. Role of leptin receptor gene polymorphisms in susceptibility to the development of essential hypertension: a case-control association study in a Northern Han Chinese population. J. Hum. Hypertens. (2014) 28(9): 551-556.
19. A. Thakkinstian, P. McElduff, C. D'Este, D. Duffy, and J. Attia. A method for meta-analysis of molecular association studies. Stat. Med. (2005) 24(9): 1291-1306.
20. L. Y. Jiang. The relationship between leptin receptor polymorphism Gln223Arg, leptin and hypertensive disorder complicating pregnancy. Huazhong University of Science and Technology (2006).
21. L. Gao, L. X. Guan, H. B. Li, X. Y. Du, and Z. C. Liu. Relationship between the variation 3057 nucleotide G→A of exon 20 of leptin receptor gene and the pathogenesis pregnancy induced hypertension syndrome in Chinese population. Chin. J. Birth Health Heredity (2009) 10): 23-24.
22. A. Wiedemann, F. Vocke, J. S. Fitzgerald, U. R. Markert, U. Jeschke, P. Lohse, and B. Toth. Leptin gene (TTTC)(n) microsatellite polymorphism as well as leptin receptor R223Q and PPARgamma2 P12A substitutions are not associated with hypertensive disorders in pregnancy. Am. J. Reprod. Immunol. (2010) 63(4): 310-317.
23. K. H. Tennekoon, W. L. Indika, R. Sugathadasa, E. H. Karunanayake, J. Kumarasiri, and A. Wijesundera. LEPR c.668A>G polymorphism in a cohort of Sri Lankan women with pre-eclampsia / pregnancy induced hypertension: a case control study. BMC Res. Notes (2012) 5(308): 1-4.