Figures
Abstract
Background
Chronic kidney disease (CKD) is associated with accelerated cardiovascular disease and heart failure. Endothelial nitric oxide synthase (eNOS) Glu298Asp single nucleotide polymorphism (SNP) genotype has been associated with a worse phenotype amongst patients with established heart failure and in patients with progression of their renal disease. The association of a cardiac functional difference in non-dialysis CKD patients with no known previous heart failure, and eNOS gene variant is investigated.
Methods
140 non-dialysis CKD patients, who had cardiac magnetic resonance (CMR) imaging and tissue doppler echocardiography as part of two clinical trials, were genotyped for eNOS Glu298Asp SNP retrospectively.
Results
The median estimated glomerular filtration rate (eGFR) was 50mls/min and left ventricular ejection fraction (LVEF) was 74% with no overt diastolic dysfunction in this cohort. There were significant differences in LVEF across eNOS genotypes with GG genotype being associated with a worse LVEF compared to other genotypes (LVEF: GG 71%, TG 76%, TT 73%, p = 0.006). After multivariate analysis, (adjusting for age, eGFR, baseline mean arterial pressure, contemporary CMR heart rate, total cholesterol, high sensitive C-reactive protein, body mass index and gender) GG genotype was associated with a worse LVEF, and increased LV end-diastolic and systolic index (p = 0.004, 0.049 and 0.009 respectively).
Conclusions
eNOS Glu298Asp rs1799983 polymorphism in CKD patients is associated with relevant sub-clinical cardiac remodelling as detected by CMR. This gene variant may therefore represent an important genetic biomarker, and possibly highlight pathways for intervention, in these patients who are at particular risk of worsening cardiac disease as their renal dysfunction progresses.
Citation: Chand S, Chue CD, Edwards NC, Hodson J, Simmonds MJ, Hamilton A, et al. (2015) Endothelial Nitric Oxide Synthase Single Nucleotide Polymorphism and Left Ventricular Function in Early Chronic Kidney Disease. PLoS ONE 10(1): e0116160. https://doi.org/10.1371/journal.pone.0116160
Academic Editor: Jaap A. Joles, University Medical Center Utrecht, NETHERLANDS
Received: June 23, 2014; Accepted: December 2, 2014; Published: January 22, 2015
Copyright: © 2015 Chand et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Data Availability: All relevant data are within the paper and its Supporting Information files.
Funding: SC is in receipt of Queen Elizabeth Hospital Charity Funding 12-30-46: http://www.qehb.org. MJS is in receipt of a Diabetes, Research and Wellness Foundation (DRWF) Non-Clinical Fellowship: https://drwf.org.uk. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript
Competing interests: The authors have declared that no competing interests exist.
Introduction
Chronic kidney disease (CKD) is a major public health issue, mainly due to accelerated cardiovascular disease, affecting an estimated 10–16% of the population in developed countries[1, 2]. Non-traditional risk factors and early cardiovascular changes in CKD have been increasingly recognised to lead to heart failure and sudden cardiac death related cardiovascular mortality, implicating left ventricular disease [3, 4]. The determinants of the severity of myocardial disease are poorly characterised though hypertension, oxidative stress and activation of the renal angiotensin system are all thought to be relevant. Research into the genetic predisposition to the development of heart failure in CKD has been limited [5].
In the general population, there has been interest in the association between the Glu298Asp polymorphism within endothelial nitric oxide synthase (eNOS) and heart failure [6, 7]. Although this polymorphism has been associated with endothelial dysfunction and progression of CKD through nitric oxide effects [8], it is not known if this polymorphism is associated with early cardiac structural changes that occur in non-dialysis CKD. In light of this, we investigated if this gene variant is associated with changes in systolic and diastolic function, based on detailed cardiac magnetic resonance imaging (CMR) in non-dialysis CKD patients with no known history of heart failure.
Methods
Patients
The study cohort consisted of patients who were initially enrolled into two completed, single centre, randomised controlled trials (CRIB-II [9] and CRIB-PHOS [10]) based at the Queen Elizabeth Hospital Birmingham. To limit the confounding effect of population stratification only white patients (self-reported ethnicity) were included in this genetic substudy (n = 140). Inclusion criteria were: baseline CMR and echocardiographic investigations, age 18–80 with non-dialysis CKD, total cholesterol <5.5mmol/L, resting blood pressure controlled to <140/90 for at least 6months. Exclusion criteria were: diabetes mellitus, peripheral vascular disease, previous myocardial infarction, known heart failure, valvular heart disease and atrial fibrillation. The study was approved by East Midlands Nottingham 1 Research Ethics Committee and adhered to the Declaration of Helsinki. Study participants provided written informed consent.
Genotyping
Whole blood (8.5ml) was collected in PAXgene Blood DNA Tubes (Qiagen, Manchester, UK) which were then frozen and stored at -80°C. Samples were defrosted at room temperature for 2 hours before starting the DNA extraction process. DNA extraction was undertaken using the PAXgene Blood DNA Kit (Qiagen Manchester, UK) as previously described [11]. All DNA samples were then all diluted to a working stock of 4ng/µl with 2.25µl of DNA added into each reaction. eNOS (G894T) SNP rs1799983 genotyping was performed using Taqman technology as previously described [12]. 384-well plates were read using a 7900HT Fast Real-Time PCR System (Applied Biosystems, Foster City, USA).
Cardiovascular Magnetic Resonance Imaging
CMR was performed on a 1.5-T scanner (Sonata Symphony, Siemens, Erlangen, Germany). Serial contiguous short axis cines were piloted from the vertical long axis and horizontal long axis of the right and left ventricle (electrocardiogram gated, steady-state free precession imaging [True-Fisp]; temporal resolution 40–50ms, repetition time 3.2ms, echo time 1.6ms, flip angle 60°, slice thickness 7mm) in accordance with previously validated methodologies [13]. Analysis was performed off-line (Argus Software, Siemens) by two blinded observers (NE and CDC) for the assessment of ventricular volumes (end-diastole, end-systole, stoke volume) and ejection fraction. Heart rate and baseline brachial blood pressure was measured at the time of CMR.
Echocardiography
Transthoracic echocardiography (Vivid 7; GE Vingmed Ultrasound, Horten, Norway) was performed by an experienced echocardiographer. All parameters were measured in triplicate according to American Society of Echocardiography recommendations [14] and analysed offline (EchoPAC; GE Vingmed Ultrasound, Horten, Norway) by two blinded observers (NE and CDC ). Resting left ventricular (LV) diastolic function was assessed using standard techniques [15].
Arterial Stiffness and Distensibility
Pulse wave velocity, augmentation index and ascending aortic distensibility were common measures of arterial stiffness and distensibility in the two randomised control trials, they have been included here. Pulse wave analysis was performed on the radial artery using a high-fidelity micromanometer (SPC-301; Miller Instruments, Houston, TX). The peripheral arterial waveform was used to generate a central arterial waveform using a validated transfer function (SphygmoCor; AtCor Medical, Sydney, Australia). The same system was used to determine aortic pulse wave velocity by sequentially recording ECG-gated carotid and femoral waveforms as previously described [16]. CMR data was used for the ascending aortic distensibility measurement data.
Outcome Measures
The primary aim was to evaluate the association between LV ejection fraction (LVEF) with eNOS genotype. Secondary outcomes included: LV end-diastolic volume indexed to body surface area (LVEDVI), LV end-systolic volume indexed to body surface area (LVESVI) and S’ lat (systolic velocity of the basal anterolateral LV wall). As there is much debate over which parameters of diastolic function should be measured as well as their variability in assessment [17], six diastolic parameters were assessed in a cluster analysis (see statistical analysis below) to separate the cohort into two risk categories of diastolic dysfunction; these were then assessed by eNOS genotype. The “diastolic” parameters included: e’ lat (early diastolic velocity of the basal anterolateral LV wall on tissue doppler), E/e’ lat (maximum velocity of the E-wave of mitral valve inflow by the maximal anterolateral LV wall velocity of e’), mitral valve E/A (ratio of early to late mitral inflow velocities), LVMI (left ventricular mass index), mitral valve propagation velocity (MV VP) and LAVI (left atrial volume index).
Statistical Analysis
Baseline demographics and cardiac investigations were compared across the genotype groups, using Kruskal-Wallis (continuous data) and Fisher’s Exact tests (categorical data) as appropriate. The relationship between these was then investigated using regression analysis. Initially, univariate linear regression models were produced, with variables being log-transformed where there was evidence of a non-linearity. Multivariate regression models were used (including all factors simultaneously) in order to adjust for potentially confounding factors.
A cluster analysis was then performed, to divide patients into groups based on the values of a range of diastolic parameters. The Two-Step cluster analysis in IBM SPSS 19 was used, with the number of clusters determined automatically (in this study, two main clusters). Demographic factors were then compared between the resulting clusters, using t-tests or Fisher’s Exact test, as applicable. All analyses were performed using IBM SPSS 19 (IBM Corp. Armonk, NY), with p<0.05 deemed to be indicative of statistical significance.
Results
Genomic DNA was successfully genotyped in 132 (>94%) patients. The eNOS SNP rs1799983 patient genotype frequency was GG in 47% (62/132), TG in 39% (51/132), and TT in 14% (19/132). This distribution was within Hardy-Weinberg equilibrium bounds (p>0.05). Patient demographics are presented in Table 1 for the cohort as a whole and stratified by genotype (GG, TG and TT).There were no significant demographic differences across the three genotypes. Median age was 57 years, eGFR was 50.5 mls/min/1.73m2, with 85% of patients prescribed an angiotensin converting enzyme inhibitor or angiotensin II receptor blocker. The most common renal disease group was glomerular disease, with 16% of the cohort patients been diagnosed with IgA nephropathy on renal histology.
Univariate analysis revealed a significant difference across genotypes for LVEF and LVESVI (p = 0.006 and p = 0.024 respectively, Table 2). Post hoc Mann-Whitney U testing demonstrated significant differences between GG genotype and both TG and TT genotypes (p<0.05 for both). Because of this, and to align with existing literature, further analysis compared the GG group with the non-GG (TG+TT) group, i.e. a “dominant” model. Linear regression analysis revealed an absolute 4% lower LVEF (p = 0.005) and a 21% relative increased LV end-systolic volume index (p = 0.011) in GG versus non-GG (TG+TT) genotyped patients as shown in Table 3. In multivariate analysis, (adjusting for age, gender, estimated glomerular filtration rate (eGFR), CMR heart rate (HR), total cholesterol, high sensitive C-reactive protein (hsCRP), body mass index (BMI—if variables were indexed accounting for body surface area, BMI was excluded in the analysis) and brachial mean arterial pressure (BMAP)) the GG genotype and male sex were associated with significantly lower LVEF (p = 0.004 and p = 0.017 respectively; Table 3). Male gender, GG genotype, eGFR, and heart rate were independently associated with a higher end systolic and diastolic volume index. A trend for GG genotype association with lower S’ lat was seen, but this failed to reach statistical significance (p = 0.106). As shown in S1 Table, increased aortic stiffness was associated age, male gender and BMAP but not GG genotype; ascending aortic distensibility decreased with age.
A cluster analysis was performed based on diastolic parameters, in order to group patients based on their risk of diastolic function (Table 4). The most important factor in defining the clusters was found to be e’ lat, with mitral valve E/A (MV E/A) and E/e’ lat being moderate contributors, and the remaining parameters being minimally discriminative. The first cluster had lower levels of e’ lat, MV E/A and MV VP and higher levels of E/e’ lat and LAVI, hence represented those patients at most risk of diastolic dysfunction. Comparisons between the clusters found that increased age (p<0.001) and reduced eGFR (p = 0.014) were significantly associated with the risk of diastolic dysfunction clusters on univariate analysis (Table 5).
Discussion
In this cohort of white patients with non-dialysis dependent CKD, and without heart failure, GG genotype for eNOS SNP rs1799983 was associated with a significant lower LVEF, greater LVESVI and greater LVEDVI than those found in non-GG genotypes. The burden of myocardial disease in CKD suggests the investigation of stratification by genetic risk in this setting to be a worthwhile endeavour, and this study represents the first such attempt with this eNOS polymorphism.
Previous data from the general population suggest this gene variant represents an attractive candidate SNP, and support the findings of the current study. For instance Velloso et al studied a multi-ethnic Brazilian population and demonstrated increased frequency of GG genotype in patients with systolic heart failure compared with healthy controls [7]. Another Brazilian study showed GG genotype was associated with a near 5% reduction in LVEF compared with TT genotype patients, findings very similar to those of the current study [18]. Also noteworthy is the higher all-cause mortality associated with the GG genotype in hypertensive patients [19]. An important aspect of the current study is the inclusion of white patients only, in an attempt to reduce confounding by population stratification. Indeed this is highlighted by the study of Velloso et al which did indeed show differences in genotype frequency at this locus between White and Afro-Brazilian individuals [7]. It should be acknowledged, however, that further validation of these findings in diverse populations are required to confirm the robustness of our (and other’s) findings.
The functional change associated with this gene variant also supports the clinical data. This polymorphism results from the nucleotide guanine substituting thiamine at position 894 of exon 7 on chromosome 7, and results in different cleavage of the eNOS enzyme depending on genotype [20]. The GG genotype of the studied SNP is associated with increased eNOS activity and nitric oxide levels [21, 22] and experimental overexpression of eNOS (which is present within ventricular myocytes) results in reduced ventricular function [21, 23]. This is particularly the case in conditions of oxidative stress such as CKD[24], since “uncoupling” of eNOS may lead to generation of superoxide anion radicals that further exacerbate cardiac dysfunction [25].
The influence of genotype on cardiac function and outcome may be context-specific. Of note, McNamara et al suggested a beneficial effect of GG genotype outcome in patients with established, clinically evident heart failure [6]. Whilst at first sight this data conflicts with the current study, and with that of other reports [7, 8], it should be noted that 84% of patients displayed an ejection fraction ≤35% (97.2% of participants were in NYHA stage II or greater, with >50% in stages III or IV). Furthermore there were differences in age and aetiology (ischaemic versus non-ischaemic) between genotype groups which may have influenced the results as well as variation in the technique used in measuring ejection fraction. Thus, it is certainly possible that this eNOS SNP influences outcome differentially depending on the stage of heart failure studied. Although the present study’s exclusion criteria (diabetes mellitus, peripheral vascular disease, myocardial infarction, known heart failure, valvular heart disease, atrial fibrillation, dialysis-dependence and uncontrolled hypertension) limits the generalizability of its findings, the exclusion criteria does allow removal of these potential external factors that affect both eNOS activity and left ventricular function, allowing a more ‘pure’ analysis of eNOS polymorphism association with LVEF in early CKD. Long-term follow-up of the present study population is also desirable to monitor how these patients’ LVEFs and heart failure symptoms develop as their CKD progresses, in relation to their eNOS genotype.
This study benefits from a uniform technique of detailed CMR assessment of cardiac volumes and systolic function, and very careful clinical phenotyping. Although no association with “diastolic dysfunction” parameters derived from echocardiography and genotype was evident, the size of the cohort means that such an effect cannot be excluded, and further study in larger cohorts is required.
In summary, eNOS Glu298Asp polymorphism in non-dialysis CKD patients is associated with relevant sub-clinical cardiac remodelling as detected by CMR. This gene variant may therefore represent an important genetic biomarker, and possibly highlight pathways for intervention, in these patients who are at particular risk of worsening cardiac disease as their renal dysfunction progresses.
Supporting Information
S1 Table. Univariate and multivariate analysis of arterial stiffness and arterial distensibility as compared to GG vs non-GG.
https://doi.org/10.1371/journal.pone.0116160.s001
(DOCX)
Author Contributions
Conceived and designed the experiments: SC CDC NCE JH LH RPS JNT CJF RB. Performed the experiments: SC CDC NCE MJS AH SCLG. Analyzed the data: SC CDC NCE JH MJS SCLG LH RPS JNT CJF RB. Contributed reagents/materials/analysis tools: SC CDC NCE JH MJS AH SCLG LH RPS JNT CJF RB. Wrote the paper: SC CDC NCE JH MJS AH SCLG LH RPS JNT CJF RB.
References
- 1. Coresh J, Selvin E, Stevens LA, Manzi J, Kusek JW, et al. (2007) Prevalence of chronic kidney disease in the United States. JAMA 298: 2038–2047. pmid:17986697
- 2. Matsushita K, van der Velde M, Astor BC, Woodward M, Levey AS, et al. (2010) Association of estimated glomerular filtration rate and albuminuria with all-cause and cardiovascular mortality in general population cohorts: a collaborative meta-analysis. Lancet 375: 2073–2081. pmid:20483451
- 3. Chue CD, Townend JN, Steeds RP, Ferro CJ (2010) Arterial stiffness in chronic kidney disease: causes and consequences. Heart 96: 817–823. pmid:20406771
- 4. Go AS, Chertow GM, Fan D, McCulloch CE, Hsu CY (2004) Chronic kidney disease and the risks of death, cardiovascular events, and hospitalization. N Engl J Med 351: 1296–1305. pmid:15385656
- 5. Ferguson JF, Matthews GJ, Townsend RR, Raj DS, Kanetsky PA, et al. (2013) Candidate gene association study of coronary artery calcification in chronic kidney disease: findings from the CRIC study (Chronic Renal Insufficiency Cohort). J Am Coll Cardiol 62: 789–798. pmid:23727086
- 6. McNamara DM, Holubkov R, Postava L, Ramani R, Janosko K, et al. (2003) Effect of the Asp298 variant of endothelial nitric oxide synthase on survival for patients with congestive heart failure. Circulation 107: 1598–1602. pmid:12668492
- 7. Velloso MW, Pereira SB, Gouveia L, Chermont S, Tardin OM, et al. (2010) Endothelial nitric oxide synthase Glu298Asp gene polymorphism in a multi-ethnical population with heart failure and controls. Nitric oxide 22: 220–225. pmid:20079452
- 8. Page A, Reich H, Zhou J, Lai V, Cattran DC, et al. (2005) Endothelial nitric oxide synthase gene/gender interactions and the renal hemodynamic response to angiotensin II. J Am Soc Nephrol 16: 3053–3060. pmid:16093452
- 9. Edwards NC, Ferro CJ, Kirkwood H, Chue CD, Young AA, et al. (2010) Effect of spironolactone on left ventricular systolic and diastolic function in patients with early stage chronic kidney disease. Am J Cardiol 106: 1505–1511. pmid:21059444
- 10. Chue CD, Townend JN, Moody WE, Zehnder D, Wall NA, et al. (2013) Cardiovascular effects of sevelamer in stage 3 CKD. J Am Soc Nephrol 24: 842–852. pmid:23599381
- 11. Stringer S, Sharma P, Dutton M, Jesky M, Ng K, et al. (2013) The natural history of, and risk factors for, progressive chronic kidney disease (CKD): the Renal Impairment in Secondary care (RIISC) study; rationale and protocol. BMC Nephrol 14: 95. pmid:23617441
- 12. Testa A, Spoto B, Sanguedolce MC, Parlongo RM, Pisano A, et al. (2012) eNOS and caveolin-1 gene polymorphisms interaction and intima media thickness: a proof of concept study in ESRD patients. Am J Hypertens 25: 103–108. pmid:21976276
- 13. Maceira AM, Prasad SK, Khan M, Pennell DJ (2006) Normalized left ventricular systolic and diastolic function by steady state free precession cardiovascular magnetic resonance. J Cardiovasc Magn Reson 8: 417–426. pmid:16755827
- 14. Lang RM, Bierig M, Devereux RB, Flachskampf FA, Foster E, et al. (2005) Recommendations for chamber quantification: a report from the American Society of Echocardiography’s Guidelines and Standards Committee and the Chamber Quantification Writing Group, developed in conjunction with the European Association of Echocardiography, a branch of the European Society of Cardiology. J Am Soc Echocardiogr 18: 1440–1463. pmid:16376782
- 15. Nagueh SF, Appleton CP, Gillebert TC, Marino PN, Oh JK, et al. (2009) Recommendations for the evaluation of left ventricular diastolic function by echocardiography. J Am Soc Echocardiogr 22: 107–133. pmid:19187853
- 16. Wilkinson IB, Fuchs SA, Jansen IM, Spratt JC, Murray GD, et al. (1998) Reproducibility of pulse wave velocity and augmentation index measured by pulse wave analysis. J Hypertens 16: 2079–2084. pmid:9886900
- 17. Chapman CB, Ewer SM, Kelly AF, Jacobson KM, Leal MA, et al. (2013) Classification of left ventricular diastolic function using American Society of Echocardiography Guidelines: agreement among echocardiographers. Echocardiography 30: 1022–1031. pmid:23551740
- 18. Tardin OM, Pereira SB, Velloso MW, Balieiro HM, Costa B, et al. (2013) Genetic polymorphism G894T and the prognosis of heart failure outpatients. Arq Bras Cardiol 101: 352–358. pmid:23949326
- 19. Zhang X, Lynch AI, Davis BR, Ford CE, Boerwinkle E, et al. (2012) Pharmacogenetic association of NOS3 variants with cardiovascular disease in patients with hypertension: the GenHAT study. PloS One 7: e34217. pmid:22470539
- 20. Tesauro M, Thompson WC, Rogliani P, Qi L, Chaudhary PP, et al. (2000) Intracellular processing of endothelial nitric oxide synthase isoforms associated with differences in severity of cardiopulmonary diseases: cleavage of proteins with aspartate vs. glutamate at position 298. Proc Natl Acad Sci U S A 97: 2832–2835. pmid:10717002
- 21. Massion PB, Dessy C, Desjardins F, Pelat M, Havaux X, et al. (2004) Cardiomyocyte-restricted overexpression of endothelial nitric oxide synthase (NOS3) attenuates beta-adrenergic stimulation and reinforces vagal inhibition of cardiac contraction. Circulation 110: 2666–2672. pmid:15492314
- 22. Philip I, Plantefeve G, Vuillaumier-Barrot S, Vicaut E, LeMarie C, et al. (1999) G894T polymorphism in the endothelial nitric oxide synthase gene is associated with an enhanced vascular responsiveness to phenylephrine. Circulation 99: 3096–3098. pmid:10377070
- 23. Brunner F, Andrew P, Wolkart G, Zechner R, Mayer B (2001) Myocardial contractile function and heart rate in mice with myocyte-specific overexpression of endothelial nitric oxide synthase. Circulation 104: 3097–3102. pmid:11748107
- 24. Massy ZA, Stenvinkel P, Drueke TB (2009) The role of oxidative stress in chronic kidney disease. Semin Dial 22: 405–408. pmid:19708991
- 25. Tang L, Wang H, Ziolo MT (2014) Targeting NOS as a therapeutic approach for heart failure. Pharmacol Ther Jun;142(3):306–15. pmid:24380841