Pathophysiology/Complications ORIGINALDiabetes ARTICLE Care Publish Ahead of Print, published online April 4, 2011

Impact of Natriuretic Peptide Clearance (NPR3) Variants on Blood Pressure in Type 2 Diabetes

1,2 2,10 PIERRE-JEAN SAULNIER, MD RICHARD MARECHAUD, MD ypertension and type 2 diabetes are 3,4,5 1,2 RONAN ROUSSEL, MD, PHD STEPHANIE RAGOT, PHARMD, PHD 6 3,4,5 two chronic conditions associated JEAN MICHEL HALIMI, MD, PHD MICHEL MARRE, MD, PHD H 7 1,2,10,11 epidemiologically and pathophy- JEREMIE LEBREC, PHD SAMY HADJADJ, MD, PHD 8 siologically. Hypertension is a well- DURED DARDARI, MD FOR THE SURDIAGENE, 3,4 known risk factor for diabetic vascular SULYIA MAIMAITIMING, PHD DIAB2NEPHROGENE, AND 9 GÉRARD GUILLOTEAU, MD DIABHYCAR STUDY GROUPS complications in type 1 diabetes and type 10 XAVIER PRUGNARD 2 diabetes, and is a complex trait with an established heritability (1). Genome-wide association studies are complementary OBJECTIVE—Hypertension in diabetes is characterized by abnormal sodium homeostasis, to a candidate gene approach and have suggesting a particular role of natriuretic peptide pathway. Natriuretic peptides can affect blood helped to identify new genetic pathways pressure (BP) through their plasma concentrations, which are dependent on their receptor ac- involved in blood pressure (BP). How- tivities. We thus assessed the association between nine NPR3 gene polymorphisms and BP levels ever, only a few genetic associations in patients with type 2 diabetes. have been clearly identified and replicated RESEARCH DESIGN AND METHODS— to date, leaving some impetus to search Nine single nucleotide polymorphisms (SNPs) for new candidate . tagging the haplotype structure of the NPR3 gene were genotyped in the 3,126 French Non- insulin-dependent Diabetes, Hypertension, Microalbuminuria or Proteinuria, Cardiovascular Diabetic patients differ from nondia- Events, and Ramipril (DIABHYCAR) trial participants. We then used a second population (Dia- betic patients by having an increase in bete de type 2, Nephropathie et Genetique [DIAB2NEPHROGENE]/Survie, Diabete de type 2 et total body sodium (2,3), renal tubular so- Genetique [SURDIAGENE] study) of 2,452 patients for the purpose of replication. Finally, we dium reabsorption, and an impaired abil- separately investigated subjects selected according to their rs 2270915SNP genotypes for their BP ity to excrete a sodium load (4). These response to salt restriction. factors suggest that natriuretic peptides RESULTS— (NPs) may play a key role in the patho- In DIABHYCAR patients, three SNPs (rs6889608, rs1173773, and rs2270915) physiology of hypertension in the diabetic were significantly associated with systolic BP (SBP). The effect of the rs2270915 was replicated in 6 population. the second step population: AA homozygotes had a lower SBP than G carriers (137.4 19.1 vs. fl 140.0 6 20.2 mmHg, P=0.004). The rs2270915 influenced the response of SBP to salt re- The NP system regulates BP and uid duction, with AA homozygous patients showing greater reductions after restriction of salt intake homeostasis by modifying glomerular compared with G carriers: 220 mmHg (243 to 28) vs. 23(220 to +7); P=0.006. filtration rate (GFR) and sodium urinary excretion. This system consists of a family CONCLUSIONS—We found a consistent and significant association between the rs2270915 NPR3 of three peptidic hormones (A-type NP, polymorphism of the gene and SBP in diabetic patients. This genetic variation may affect B-type NP [BNP], and C-type NP) that pressure response to changes in dietary sodium. interact with three receptors (NP receptor A, NP receptor B, and NP clearance re- ceptor [NPRC]). The NPRC is encoded by the NPR3 gene and is a determinant of NP plasma concentration. Local expression of NPRC is linked to NP activity in the kidney (5,6). The intracellular domain ccccccccccccccccccccccccccccccccccccccccccccccccc of NPRC modulates the contraction/re-

1 2 laxation properties of smooth muscle cells From the Centre d’investigation clinique, Inserm CIC0802, CHU de Poitiers, Poitiers, France; the Université NPR3 de Poitiers, Poitiers, France; the 3Inserm U695, Paris, France; the 4Universite Paris 7 Denis Diderot, Paris, (5). Genetic variation in the may France; the 5Service d’Endocrinologie, diabétologie, nutrition, Groupe Hospitalier Bichat Claude Bernard, therefore affect the activity of NP through Assistance Public-Hopitaux de Paris, Paris, France; the 6Centre d’investigation clinique, Inserm CIC0202, affecting plasma levels or NPRC-mediated Service de Néphrologie, CHU de Tours, Tours, France; the 7Centre d’investigation clinique, Inserm vascular effects. The NPR3 is located on 8 ’ CIC0502, CHU de Brest, Brest, France; the Service d Endocrinologie, CH Sud Francilien, Corbeil Es- 5p14-p13. It spans ;70 kb sonnes, France; the 9Service d’Endocrinologie, diabétologie, nutrition, CHU Angers, Angers, France; the 10Service de Médecine interne, endocrinologie et maladies métaboliques, CHU de Poitiers, Poitiers, France; and contains eight exons and seven in- and the 11Inserm U927, Poitiers, France. trons. Within the general population, al- Corresponding author: Pierre-Jean Saulnier, [email protected]. lelic variants of the NPR3 gene have been Received 30 October 2010 and accepted 18 February 2011. reported to be associated with hyperten- DOI: 10.2337/dc10-2057 – This article contains Supplementary Data online at http://care.diabetesjournals.org/lookup/suppl/doi:10. sion (6 8). 2337/dc10-2057/-/DC1. This study aimed to assess the in- © 2011 by the American Diabetes Association. Readers may use this article as long as the work is properly fluence of genetic polymorphisms within cited, the use is educational and not for profit, and the work is not altered. See http://creativecommons.org/ NPR3 on BP in patients with type 2 licenses/by-nc-nd/3.0/ for details.

care.diabetesjournals.org DIABETES CARE 1 Copyright American Diabetes Association, Inc., 2011 NPR3 polymorphisms and blood pressure diabetes. We first used a candidate gene rs2270915 genotypes were invited to was measured on a Hitachi911 automatic approach in two independent large popu- participate in a “sodium-restriction func- analyzer (Roche Diagnostics, Meylan, lations. We then completed our investiga- tional” study (study design detailed in France). Glycated hemoglobin (A1C) tion with a sodium restriction functional Supplementary Data and Supplementary was determined by using a high-perfor- study. Fig. 1). We included seven AA-genotype mance liquid chromatography method and seven AG- or GG-genotype patients, with an ADAMS A1C HA-8160 analyzer RESEARCH DESIGN AND matched for age (65years),urinaryalbu- (normal values 4.0–6.0%; Menarini, Flor- METHODS min status (micro- or macroalbuminuria), ence, Italy). Plasma N-terminal pro B-type and antihypertensive therapy: renin-an- natriuretic peptide (NT-proBNP) concen- Study protocols and subjects giotensin-aldosterone system blocker tration was determined using Elecsy- First step population: DIABHYCAR use (angiotensin receptor blocker or sproBNP sandwich immunoassay on a study. The design and results of the Non- ACE inhibitor) and diuretics. Exclusion Modular System E (Roche Diagnostics, insulin-dependent Diabetes, Hyperten- criteria included uncontrolled hyperten- Meylan, France). Measures were dupli- sion, Microalbuminuria or Proteinuria, sion, estimated GFR ,30 mL/min (Mod- cated on each sample, and mean was con- Cardiovascular Events, and Ramipril (DI- ification of Diet in Renal Disease formula), sidered for statistical analyses. ABHYCAR) study have been reported (9). and prior low sodium diets. The study Briefly, the DIABHYCAR study is a clini- design was approved by the Poitiers Uni- Statistical analysis cal trial comparing the effect of a low dose versity Hospital Ethics Committee. All pa- Statistical analyses were performed with of ramipril with placebo on cardiovascu- tients gave informed consent before StatView 5.0 (SAS Institute, Cary, NC). lar and renal complications of patients starting the study. Patient characteristics were expressed as with type 2 diabetes with micro- or mac- means 6 SD or medians (interquartile roalbuminuria. The main selection crite- Polymorphism determination range) for skewed distributions. Groups riawereasfollows:patientswithtype2 On the basis of the Caucasian European were compared using the x2 test for cate- diabetes undergoing treatment with oral population data available in the HapMap goric variables or parametric (ANOVA) or antidiabetic drugs, high urinary albumin phase II panel (http://www.hapmap.org), nonparametric (if not normally distrib- excretion, age $50 years, and serum cre- at least one SNP was selected in each hap- uted, Kruskal–Wallis or Mann–Whitney atinine #150 mmol/L. This first step pop- lotype block of this large gene, focusing U tests) tests for continuous variables. ulation included 3,413 patients recruited on SNPs associated with relevant end Allele frequencies were estimated by in France. From the first step cohort data, points in the literature. Thus, we selected the gene-counting method, and Hardy- we selected the single nucleotide poly- one SNP (rs9716700) in the promoter re- Weinberg equilibrium (HWE) was tested morphisms (SNPs) that were most signif- gion (6,7,11), three in haplotype block 1 by the x2 test. We used THESIAS software icantly associated with differences in (rs1421811, rs1252246, rs6889608) (11), for pairwise linkage disequilibrium (LD) systolic BP (SBP). two in haplotype block 2 (rs700923, between NPR3 polymorphisms (ex- Second step population: DIAB-2- rs16890196) (12), one in haplotype block pressed in terms of the D’ and r2 statistics) NEPHRO-GENE and SURDIAGENE 3 (rs1173773) (11), and two in haplotype and haplotype analysis (http://genecanvas. studies. This cohort, dedicated to repli- block 4 (rs1173743, rs2270915) (11,13,14) ecgene.net/downloads.php?cat_id=1). As- cation, was previously described (10). (Supplementary Fig. 2). sociations between BP and genotypes Briefly, patients were recruited for the Di- The full set of nine SNPs was geno- were evaluated by linear regression analysis abete de type 2, Nephropathie et Geneti- typed in the first step cohort analysis. and ANOVA. que (DIAB2-NEPHRO-GENE; D2NG) or SNPs were determined by quantitative Meta-analysis of our two cohorts was the Survie, Diabete de type 2 et Genetique PCR performed using the Taqman performed using the METAL software (SURDIAGENE; SDG) study. The D2NG method (ABI/PRISM 7700; Applied Bio- (http://www.sph.umich.edu/csg/abecasis/ is a French multicenter case-control study systems, Foster City, CA) or allele-specific Metal/index.html) based on the weighted of patients with type 2 diabetes, designed PCR-based KASPar SNP genotyping sys- z score method. to assess the genetic determinants of di- tem (KBiosciences, Hoddesdon, U.K.). The Bayesian method implemented abetic nephropathy in type 2 diabetes. ThethreeofthenineSNPswith in BIMBAM (http://quartus.uchicago. The SDG study is an inception cohort nominal association with SBP were se- edu/~yguan/bimbam/index.html) was with ongoing recruitment of patients lected for the replication step and de- used with the HapMap CEU data to im- with type 2 diabetes regularly attending termined with the same methods. pute the genotype of other HapMap SNPs the diabetes department at Poitiers Uni- Genotyping errors were checked on 96 in the NPR3 gene. The imputed SNPs versity Hospital, France. Altogether, the randomly selected samples with both were subsequently subjected to associa- second step population comprised TaqMan and KASPAR methods with a tion testing with BP as well. 2,635 unrelated patients (10). The design 100% concordance rate. P values # 0.05 were considered sta- of these studies was approved by Univer- tistically significant. sity Hospital Ethics Committees (Angers Laboratory methods and Poitiers, respectively). All partici- Serum creatinine and urinary albumin RESULTS pants in the study gave their written in- were measured by nephelometry on a formed consent. Modular System P (Roche Diagnostics First step population: DIABHYCAR GmbH, Mannheim, Germany). Renal study—univariate analysis Sodium restriction functional study function was estimated by GFR using DNA samples were available from 3,126 A subset of SDG micro- and macroalbu- the four-variable Modification of Diet in French patients in the DIABHYCAR study. minuric patients selected for their Renal Disease formula. Urinary creatinine Clinical and biological characteristics of the

2 DIABETES CARE care.diabetesjournals.org Saulnier and Associates patients are presented in Supplementary Multivariate and conjunct analysis to +7) (P=0.006) in AA versus G carriers, Table 1. Genotyping success ranged from Multiple linear regression analysis respectively. 99.1 to 98.3%. The genotype distribution showed that both rs1173773 and No significant modification of body was in HWE for all SNPs except for rs2270915, age, and BMI were indepen- weight, heart rate, and serum creatinine rs6889608 (P=0.046) and rs1173743 dent predictive factors of SBP in each was observed in either of the two geno- (P=0.049). Pairwise LD is shown in cohort (Table 3). type groups (Supplementary Table 7). Supplementary Table 2. In addition, we combined the geno- Of the nine studied SNPs, three type analysis for SBP in a meta-analysis of CONCLUSIONS—We performed a (rs6889608, rs1173773, rs2270915) were our two cohorts, confirming the significant large-scale analysis of several genetic var- significantly associated with SBP (Table 1) association between SBP and rs2270915 iants within the NPR3 gene in more than and six were not (Supplementary Table 3). (z score = 23.109, P = 0.002) but not 5,500 patients with type 2 diabetes. We AA homozygotes for the rs2270915 SNP rs1173773 and rs6889608 (P =0.111 found that the rs2270915 SNP was asso- had an SBP of 143.9 6 13.0 vs. 144.9 6 and 0.519, respectively). ciated with differences in SBP in patients 13.6 mmHg for G carriers (P=0.03). Strat- The haplotype analysis on conjunct with type 2 diabetes; in the DIABHYCAR ification on presence or absence of obesity cohorts evidenced that one haplotype and D2NG/SDG studies, the G allele was (defined as BMI .30 kg/m2) showed an containing the rs2270915 G allele was sig- associated with a higher adjusted SBP of association between SBP and rs1173773 nificantly associated with SBP (Supplementary 1.1 and 1.7 mmHg, respectively. This G in obese patients (P = 0.003) but not Table 6). allele was also present in a haplotype sig- in nonobese patients (Supplementary nificantly associated with higher SBP. We Table 4). Sodium reduction study found that salt intake strongly influenced No differences were observed regarding the relationship between BP and this ge- Second step population: D2NG and duration of diabetes, medical history, or netic polymorphism, with a decreased SDG studies—univariate analysis baseline BP between rs2270915AA geno- salt sensitivity in G carriers compared Baseline characteristics of replication type and G-carrier participants. with AA homozygotes. study participants are presented in The clinical and biological responses The choice of the different SNPs was Supplementary Table 1. DNA samples to the sodium restriction functional study based on the haplotype structure of the were available for 2,452 patients. Geno- are presented in Table 4. Urinary sodium NPR3, as detailed in RESEARCH DESIGN AND typing success ranged from 97.02 to excretions did not differ between groups METHODS. We tested the association be- 99.71%. There were 2,043 Europid and at baseline or in response to sodium re- tween imputed genotypes and SBP in 409 non-Europid patients, including 88 striction. the DIABHYCAR population. Graphically subjects of African origin. We restricted SBP, urinary albumin excretion, and this analysis further supported the choice our analyses to Europid patients. In this NT-proBNP significantly decreased un- of SNPs rs6889608, rs1173773, and group, all the selected SNPs (rs6889608, der restricted salt diet. The decrease was rs2270915 (Supplementary Fig. 4). rs1173773, rs2270915) were in HWE. more marked in AA than in G carriers for All the studied SNPs were in HWE, SBP was significantly associated with SBP and urinary albumin, suggesting a suggesting no selection bias in the repli- rs2270915 but not with rs6889608 and significant gene–environment interac- cation study for which ethnic background rs1173773 (Table 2). In agreement with tion. However, no such association was was available. The current study mainly the DIABHYCAR study, AA homozygotes found for NT-proBNP. The responses found that the rs2270915 was associated for the rs2270915 SNP had a significantly to the salt restriction of SBP, according with SBP in the DIABHYCAR population, lower SBP than G carriers (137.4 6 19.1 vs. to genotype group, are plotted in with positive replication in the D2NG/ 140.0 6 20.2 mmHg, P=0.004). Stratifi- Supplementary Fig. 3. The absolute dif- SDG studies. Other reports from the gen- cation on obesity did not replicate in this ference in SBP after salt restriction was eral population were not in accordance population (Supplementary Table 5). 220 mmHg (243 to 28) vs. 23(220 with our finding in type 2 diabetes. Iemitsu

Table 1—Relationship between SBP and SNPs of NPR3 in patients from the DIABHYCAR study

P value P value A1 P value A2 Genotype codominant dominant dominant † ‡ SNP A1 A1 A1 A2 A2 A2 model*/§ A1 carriers model A2 carriers model CC CT TT C carriers T carriers rs6889608 n =231 n =1,151 n =1,715 n =1,382 n =2,866 145.2 6 12.3 144.9 6 13.0 143.7 6 13.4 0.034/0.012 145.0 6 12.9 0.010 144.2 6 13.3 0.26 GG AG AA G carriers A carriers rs1173773 n =332 n =1,397 n =1,369 n =1,729 n =2,766 142.4 6 13.2 144.4 6 13.5 144.7 6 12.9 0.018/0.019 144.1 6 13.5 0.16 144.5 6 13.2 0.005 GG AG AA G carriers A carriers rs2207915 n =140 n = 987 n =1,971 n =1,127 n =2,958 145.8 6 13.9 144.8 6 13.5 143.9 6 13.0 0.08/0.024 144.9 6 13.6 0.035 144.2 6 13.2 0.17 6 P P Data are means SD. */§ Upper values correspond to ANOVA statistics: A1 A1 vs. A1 A2 vs. A2 A2./lower values correspond to simple regression analysis: 0, 1, or 2 minor alleles. †A1 carriers vs. A2 A2. ‡A2 carriers vs. A1A1. care.diabetesjournals.org DIABETES CARE 3 NPR3 polymorphisms and blood pressure

Table 2—Relationship between SBP and SNPs of NPR3 in Europid patients from the D2NG/SDG studies

P value P value A1 P value A2 Genotype codominant dominant dominant SNP A1A1 A1 A2 A2A2 model*/† A1 carriers model‡ A2 carriers model§ CC CT TT C carriers T carriers rs6889608 n =116 n = 685 n =1,160 n = 801 n =2,230 135.5 6 20.1 139.0 6 19.2 138.4 6 19.7 0.20/0.56 138.5 6 19.3 0.93 138.6 6 19.5 0.09 GG AG AA G carriers A carriers rs1173773 n =234 n = 864 n =860 n =1,098 n =2,190 140.0 6 19.1 138.9 6 20.1 137.6 6 18.9 0.16/0.056 139.1 6 19.9 0.08 138.2 6 19.5 0.19 GG AG AA G carriers A carriers rs2207915 n =91 n = 659 n =1,271 n = 750 n =2,440 137.0 6 19.8 140.4 6 20.2 137.4 6 19.1 0.004/0.031 140.0 6 20.2 0.004 138.4 6 19.5 0.48

Data are means 6 SD. */†Upper P values correspond to ANOVA statistics: A1 A1 vs. A1 A2 vs. A2 A2./lower P values correspond to simple regression analysis: 0, 1, or 2 ‡ minor alleles. A1 carriers vs. A2 A2.§A2 carriers vs. A1A1. et al. (14) found that the G allele was not compared with the reference haplotype. than salt-resistant patients (G carriers). related to SBP in the general population. In multivariate analysis, the rs2270915 However, it has been reported that salt These Japanese patients had a similar geno- dose effect had an impact on SBP ranging supplementation downregulated NPRC type frequency as our patients, but only from 1.1 to 1.7 mmHg in these two expression in the kidney (20). The gene 291 nondiabetic subjects were studied different populations. This effect seems interaction found in the current study with a focus on arterial stiffness. Lanfear to be modest, but it could be clinically therefore leads us to hypothesize that et al. (13) did not find a relationship be- relevant. We can speculate from data of the influence of NPR3 variants on BP tween this SNP and BNP concentrations on clinical trials on BP that the rs2270915 could be masked in the presence of high the 100 genotyped Caucasian patients. The effect on SBP could translate into a lower sodium intake. findings from Sarzani et al. (6), suggesting cardiovascular mortality risk by 2.5% The speculations about the biological that the rs9716700 was associated with BP, over 4 years (16). impact of the rs2270915 SNP are excit- were not replicated. In our study, the Our data from the functional salt ing. It is a nonsynonymous polymor- rs9716700 was in HWE, at variance with restriction study, although generated phism resulting in the replacement of an the data from Sarzani et al. (6). Genotypic on a small group, help to better under- asparagine by an aspartate residue in the frequencies were different in the two stud- stand the effect of this genetic polymor- COOH-terminal region of the NPRC, ies, suggesting different recruitments. We phism on BP in patients with type 2 N521D. In silico analysis using Poly- also evidenced a relationship between SBP diabetes. We found that salt restriction Phen2 (http://genetics.bwh.harvard.edu/ and rs1173773 in obese patients (not re- was associated with a decrease in SBP and pph2) predicted the consequence of this ported by Sarzani et al. [6]) but not for albuminuria, as demonstrated by others rs2270915 to be “probably damaging” to rs9716700 as documented by others (17). As previously reported, the salt-sen- the protein. This residue is conserved in (6,8). However, we focused on patients sitive AA genotype responded more various species, including humans, ro- with type 2 diabetes, who are well known strongly to salt restriction on albumin ex- dents, and cows (21). This intracellular to be particularly salt sensitive (15), possi- cretion than the salt-resistant G carriers truncated domain has affinity for pertus- bly explaining the discrepancy for these (18). This result seems counterintuitive sis toxin-sensitive G proteins and is able results. because it is generally accepted that salt- to activate/inhibit various enzymes in the Theconjunctanalysisshowedthata sensitivity is associated with hypertension cell, such as adenylyl cyclase and isoforms haplotype containing the rs2270915 G (19), and we found that salt-sensitive of phospholipase C (21,22). Synthetic allele was associated with an increased BP patients (AA genotype) had a lower SBP polypeptides generated by mutational study of this 37-amino acid intracellu- lar domain were used to identify the Table 3—Impact of NPR3 genetic variants and clinical variables on SPB (mmHg) homologous 17-amino acid sequence, from the DIABHYCAR and D2NG/SDG studies (multivariate analysis) R513RNHQEESN521IGKHRELR529,respon- sible for these binding capacities (21) and fi DIABHYCAR D2NG/SDG were necessary and suf cient for the acti- vation of G proteins and effector enzymes Estimate 6 SEP value Estimate 6 SE P value (23). Recent studies suggest that changes Sex (compared with male) 1.9 6 0.5 0.0004 21.5 6 0.9 0.1058 to this sequence, such as that introduced Age (for each increment of 1 year) 0.2 6 0.1 , 0.0001 0.3 6 0.04 , 0.0001 by the missense mutation in the current BMI (for each increment of 1 kg/m2)0.36 0.1 , 0.0001 0.3 6 0.1 0.0003 study, are likely to modify the tone of rs6889608* 1.2 6 0.5 0.0087 20.5 6 0.7 0.5189 smooth muscle cells, as suggested by re- rs1173773* 20.9 6 0.4 0.0141 1.4 6 0.7 0.0296 cent studies (23). NPRC is widely ex- rs2270915* 1.1 6 0.5 0.0163 1.7 6 0.8 0.0306 pressed in vascular smooth muscle cells *For each addition of a minor allele (i.e., C, G, and G for rs6889608, rs1173773, rs2270915, respectively). and mesangial cells (23,24). Changes in

4 DIABETES CARE care.diabetesjournals.org Saulnier and Associates

Table 4—Clinical and biological response to salt reduction in the functional study

Global genotype Variable Salt diet AA G carriers Genotype effect effect (n =7) (n =87) (P value) (P value) UNa (mmol/24 h) Usual 215 6 91 232 6 85 0.749 0.51 Low 103 6 65 121 6 69 0.565 Treatment effect (P value) 0.018 0.018 Global treatment effect (P value) 0.001 0.949* SBP (mmHg) Usual 144.6 6 14.3 137.4 6 7.7 0.338 0.51 Low 124.2 6 11.0 134.2 6 7.7 0.096 Treatment effect (P value) 0.018 0.40 Global treatment effect (P value) 0.006 0.006* Ualb (mg/24 h) Usual 233 (419) 282 (405) 0.655 0.97 Low 178 (239) 239 (456) 0.655 Treatment effect (P value) 0.018 0.40 Global treatment effect (P value) 0.013 0.048* NT-proBNP (pg/mL) Usual 97 (141) 147 (651) 0.338 0.07 Low 46 (149) 95 (181) 0.227 Treatment effect (P value) 0.128 0.0425 Global treatment effect (P value) 0.013 0.338* Data are mean 6 SD or median (interquartile). Treatment effect corresponds to the effect of the dietary salt restriction. UNa, urinary sodium excretion; Ualb, urinary albumin excretion. *P values for estimated genotype-treatment interaction. the NPR3 primary sequence might there- overcorrect the multiplicity of testing. The by grants from PHRC-Poitiers 2004 and AFD fore affect systemic and renal hemo- borderline P values nevertheless deserved (Research Grant 2003). The Association Dia- dynamics (natriuresis, vascular and additional testing in a replication popula- bète Risque Vasculaire (Paris, France) and ’ mesangial tone) via at least two different tion. In the D2NG/SDG study, the signif- Groupe d Etude des Maladies Métaboliques et mechanisms: changes in NP clearance by icant association of rs2270915 with SBP Systémiques (Poitiers, France) supported the present studies. No other potential conflicts of internalization and changes in the activa- was maintained after Bonferroni correc- interest relevant to this article were reported. tion/inhibition capacities of the intracellu- tion. The consistency of the results across P.-J.S. and R.R. wrote the article, partici- lar domain. Measuring atrial natriuretic the studies, including the functional pated in data collecting and genotyping, con- peptide response in kidney cells of pa- study, strongly suggests that the associa- tributed to data analysis, and contributed to tients with type 2 diabetes according to tion is not a pure chance finding. Further discussion. J.M.H. participated in data col- rs2270915 genotype was beyond of the studies, particularly in patients from other lecting and genotyping. J.L. contributed to scope of this article. ethnic origins, are warranted. data analysis and contributed to discussion. Some limitations must be acknowl- The effect of this NPR3 variant ob- D.D., S.M., G.G., X.P., and R.M. participated edged in this study. We had no informa- served in two independent populations in data collecting and genotyping. S.R. par- tion regarding salt intake in the was substantiated in a functional study. ticipated in data collecting and genotyping and contributed to data analysis. M.M. wrote the DIABHYCAR and D2NG/SDG studies. Homozygous AA patients for the article, participated in data collecting and Salt intake may have affected the relation- rs2270915 SNP could be identified as a “ genotyping, and contributed to discussion. ship between BP and the genetic variants low-risk genotype and salt restriction S.H. wrote the article, participated in data of NPR3. Our functional study on salt re- responders.” Conversely, G carrier pa- collecting and genotyping, contributed to data striction is exciting but must be regarded tients could be classified as “high-risk” be- analysis, and contributed to discussion. All as a pilot study. Although not performed cause they seem to have higher SBP and authors reviewed the article. on patients with type 2 diabetes, ade- are less sensitive to salt reduction. Our The authors thank all of the patients and quately powered studies on the genetic data suggesting a gene–environment in- physicians who took part in the D2NG study component of salt sensitivity, such as teraction support the speculation of an in- (see list in Supplementary Data); Cecile Demer the recently released DASH trial, are man- dividual tailoring of antihypertensive (Diabetology Department, CHU de Poitiers, Poitiers, France) for secretarial help; Elodie datory to replicate our findings (25). therapeutics according to its NPR3 geno- fi Rogeon (Inserm CIC0802, Poitiers, France) Finally, the question of multiple test- type. This point deserves further con r- and the staff of the Diabetes Department at ing is an important issue. We found a mation before routine clinical use. Poitiers for help with data collection and replicated effect of the rs2270915 SNP in monitoring; S. Leroux, I. Petit, and G. Mauco two independent cohorts. Associations (Inserm U927 and Biochemistry Department, were of borderline signification after Bon- Acknowledgments—The DIABHYCAR trial Poitiers, France) for biological determinations; ferroni correction in the DIABHYCAR was supported by grants from sanofi-aventis N. Bellili, F. Travert, and F. Fumeron (Inserm population for any of the selected SNPs. (Paris), Programme Hospitalier de Recherche U695 and University of Paris 7, Paris France) However, the LD between the SNPs sug- Clinique (PHRC)-Angers 1996 (French Min- for assistance; E. Nowak (Inserm CIC0502, gests that the tests were not indepen- istry of Health), and the Association Francaise IFR148 and CHU de la Cavale Blanche, Brest, des diabétiques (AFD) (Research Grant 2004). France) for statistical help; and Alex Edelman dent, and thus Bonferroni method may The D2NG and SDG studies were supported

care.diabetesjournals.org DIABETES CARE 5 NPR3 polymorphisms and blood pressure and Associates (Malakoff, France) for editing on cardiovascular and renal outcomes in randomized controlled trials. Am Heart J the article. patients with type 2 diabetes and raised 1999;138:211–219 excretion of urinary albumin: rando- 17. Vedovato M, Lepore G, Coracina A, et al. mised, double blind, placebo controlled Effect of sodium intake on blood pressure References trial (the DIABHYCAR study). BMJ 2004; and albuminuria in type 2 diabetic pa- 1. Newton-Cheh C, Larson MG, Vasan RS, 328:495 tients: the role of insulin resistance. Dia- et al. Association of common variants in 10.HadjadjS,FumeronF,RousselR, betologia 2004;47:300–303 NPPA and NPPB with circulating natri- et al.; DIABHYCAR Study Group; 18. Jelic S, Sumarac-Dumanovic M, Caparevic uretic peptides and blood pressure. Nat DIAB2NEPHROGENE Study Group; Z. Does salt sensitivity influence the Genet 2009;41:348–353 SURDIAGENE Study Group. Prognostic magnitude of microalbuminuria in hy- 2. Beretta-Piccoli C, Weidmann P. Body so- value of the insertion/deletion poly- pertensive type 2 diabetes patients (Ab- dium-blood volume state in nonazotemic morphism of the ACE gene in type 2 stract). European Diabetic Nephropathy diabetes mellitus. Miner Electrolyte Metab diabetic subjects: results from the Non- Study Group 21st annual meeting. Ab- 1982;7:36–47 insulin-dependent Diabetes, Hyperten- stract book 2008;▪▪▪:57–57 3. O’Hare JA, Ferriss JB, Brady D, Twomey B, sion, Microalbuminuria or Proteinuria, 19. Sanders PW. Dietary salt intake, salt sen- O’Sullivan DJ. Exchangeable sodium and Cardiovascular Events, and Ramipril sitivity, and cardiovascular health. Hy- renin in hypertensive diabetic patients (DIABHYCAR), Diabete de type 2, pertension 2009;53:442–445 with and without nephropathy. Hyper- Nephropathie et Genetique (DIAB2- 20. Sun JZ, Chen SJ, Majid-Hasan E, Oparil S, tension 1985;7:II43–II48 NEPHROGENE), and Survie, Diabete de Chen YF. Dietary salt supplementation 4. DeFronzo RA. The effect of insulin on type 2 et Genetique (SURDIAGENE) stud- selectively downregulates NPR-C receptor renal sodium metabolism. A review with ies. Diabetes Care 2008;31:1847–1852 expression in kidney independently of clinical implications. Diabetologia 1981; 11. Roussel R, Bellili N, Hadjadj S, Péan F, ANP. Am J Physiol Renal Physiol 2002; 21:165–171 Jeunemaitre X, Marre M. A novel variant 282:F220–F227 5. Matsukawa N, Grzesik WJ, Takahashi N, in exon 8 of natriuretic peptide clearance 21. Murthy KS, Makhlouf GM. Identification et al. The natriuretic peptide clearance receptor (NPRC) gene is associated with of the G protein-activating domain of receptor locally modulates the physio- progression of nephropathy in type 1 di- the natriuretic peptide clearance receptor logical effects of the natriuretic peptide abetic patients. Diabetologia 2004;47 (NPR-C). J Biol Chem 1999;274:17587– system. Proc Natl Acad Sci U S A 1999;96: (Suppl. 1):A393 17592 7403–7408 12. Fox AA, Collard CD, Shernan SK, et al. 22. Murthy KS, Teng B, Jin J, Makhlouf GM. G 6. Sarzani R, Dessì-Fulgheri P, Salvi F, et al. Natriuretic peptide system gene variants protein-dependent activation of smooth A novel promoter variant of the natriuretic are associated with ventricular dysfunc- muscle eNOS via natriuretic peptide clear- peptide clearance receptor gene is associ- tion after coronary artery bypass grafting. ance receptor. Am J Physiol 1998;275: ated with lower atrial natriuretic peptide Anesthesiology 2009;110:738–747 C1409–C1416 and higher blood pressure in obese hy- 13. Lanfear DE, Stolker JM, Marsh S, Rich 23. Zhou H, Murthy KS. Identification of pertensives. J Hypertens 1999;17:1301– MW, McLeod HL. Genetic variation in the the G protein-activating sequence of the 1305 B-type natriuretic peptide pathway affects single-transmembrane natriuretic peptide 7. Pitzalis MV, Sarzani R, Dessì-Fulgheri P, BNP levels. Cardiovasc Drugs Ther 2007; receptor C (NPR-C). Am J Physiol Cell et al. Allelic variants of natriuretic peptide 21:55–62 Physiol 2003;284:C1255–C1261 receptor genes are associated with family 14. Iemitsu M, Maeda S, Otsuki T, et al. Ar- 24. Ardaillou N, Placier S, Striker L, Striker G, history of hypertension and cardiovascu- terial stiffness, physical activity, and atrial Ardaillou R. Mesangial cells from diabetic lar phenotype. J Hypertens 2003;21:1491– natriuretic peptide gene polymorphism in NOD mice constitutively express in- 1496 older subjects. Hypertens Res 2008;31: creased density of atrial natriuretic pep- 8. Aoi N, Soma M, Nakayama T, et al. Vari- 767–774 tide C receptors. Kidney Int 1999;55: able number of tandem repeat of the 15. Trevisan R, Fioretto P, Semplicini A, et al. 1293–1302 59-flanking region of type-C human na- Role of insulin and atrial natriuretic pep- 25. Sun B, Williams JS, Svetkey LP, Kolatkar triuretic peptide receptor gene influences tide in sodium retention in insulin-treated NS, Conlin PR. Beta2-adrenergic receptor blood pressure levels in obesity-associated IDDM patients during isotonic volume genotype affects the renin-angiotensin- hypertension. Hypertens Res 2004;27: expansion. Diabetes 1990;39:289–298 aldosterone system response to the Di- 711–716 16. He J, Whelton PK. Elevated systolic blood etary Approaches to Stop Hypertension 9. Marre M, Lievre M, Chatellier G, Mann JF, pressure and risk of cardiovascular and (DASH) dietary pattern. Am J Clin Nutr Passa P, Ménard J; DIABHYCAR Study renal disease: overview of evidence from 2010;92:444–449 Investigators. Effects of low dose ramipril observational epidemiologic studies and

6 DIABETES CARE care.diabetesjournals.org