Abstract
Context:
Experimental data support a role for vasopressin in metabolic disorders.
Objective:
We investigated associations of plasma copeptin, a surrogate of vasopressin, and of allelic variations in the arginine vasopressin-neurophysin II gene with insulin secretion, insulin sensitivity, and the risk for impaired fasting glucose (IFG) and type 2 diabetes mellitus (T2DM).
Design, Setting, and Participants:
We studied 5110 unrelated French men and women from a prospective cohort of the general population (Data from Epidemiological Study on the Insulin Resistance Syndrome cohort, 9-y follow-up). Six single nucleotide polymorphisms were genotyped.
Main Outcome Measure:
Incidence of IFG or T2DM during follow-up.
Results:
The incidence of hyperglycemia (IFG/T2DM) during follow-up by quartiles of baseline plasma copeptin was 11.0% (Q1), 14.5% (Q2), 17.0% (Q3), and 23.5% (Q4), log-rank test P = .003. Participants in the upper quartile of plasma copeptin had significantly lower insulin sensitivity (homeostasis model assessment index) at baseline and during follow-up, as compared with other participants. Cox proportional hazards regression analyses showed significant associations of the CC genotype of rs6084264, the TT genotype of rs2282018, the C-allele of rs2770381, and the CC genotype of rs1410713 with the incidence of hyperglycemia. The genotypes associated with an increased risk of hyperglycemia were also associated with increased plasma copeptin in men but not in women.
Conclusions:
High plasma copeptin was associated with reduced insulin sensitivity and an increased risk for IFG/T2DM diabetes in this community-based cohort. Moreover, in men, allelic associations support a causal role for vasopressin in these disorders.
“High plasma copeptin was associated with reduced insulin sensitivity and with hyperglycemia in a community-based cohort. Allelic associations support a causal role for vasopressin in these disorders.”
Vasopressin or antidiuretic hormone has many physiological actions in addition to its well-defined role in the control of fluid homeostasis and urine concentration (1, 2). An increasing body of data suggests that the vasopressin-hydration axis plays a role in glucose homeostasis and that high vasopressin levels might be a risk factor for hyperglycemia and diabetes. Vasopressin stimulates hepatic gluconeogenesis and glycogenolysis (3–6) and the release either of glucagon or insulin, depending on the extracellular glucose level (7). Mice with deletion of V1a and/or V1b vasopressin receptors present with metabolic disorders ranging from insulin hypersensitivity and enhanced glucose tolerance to insulin resistance, obesity, and impaired glucose tolerance (5, 8–10). Infusion of vasopressin induces a transient rise in blood glucose in rats (11) and in healthy volunteers (12). Increased plasma vasopressin concentration was observed in people with type 1 or with type 2 diabetes mellitus (T2DM) (13) and in animal models with spontaneous or streptozotocin-induced diabetes (14, 15).
Vasopressin is not easily measurable because of its very low concentration in the blood (10−12M), its small-size (a nonapeptide) and poor stability in blood samples. Copeptin, the stable C-terminal portion of the preprovasopressin peptide, is easier to assay (16, 17) and was shown to be an adequate surrogate marker of vasopressin and of the hydration status (18). Circulating levels of copeptin and vasopressin correlate significantly over a wide range of osmolalities (18–20). Associations of plasma copeptin with insulin resistance, metabolic syndrome, obesity, and type 2 diabetes have been reported (21–25).
Increased plasma osmolality is the main stimulus for vasopressin and copeptin secretion, which is thus, strongly dependent on the hydration status. We have previously reported that water intake was inversely and independently associated with the risk of developing hyperglycemia in a prospective cohort of the French general population, the Data from Epidemiological Study on the Insulin Resistance Syndrome (D.E.S.I.R.) (26). In a recent investigation, we observed a causal relationship between the vasopressin-hydration axis and metabolic risk in obese Zucker rats (6). Chronic ip infusion of vasopressin promoted hyperinsulinemia and glucose intolerance, whereas treatment with a selective V1a receptor antagonist improved glucose tolerance. In the present investigation, we assessed associations of plasma copeptin concentration at baseline with insulin secretion and insulin sensitivity indices, and with the incidence of hyperglycemia and T2DM during follow-up in the D.E.S.I.R. Study. In addition, to address causality between vasopressin and hyperglycemia, we looked at polymorphisms in the arginine vasopressin-neurophysin II gene (AVP) locus and their relationship with circulating levels of copeptin and clinical outcomes.
Materials and Methods
Study population
D.E.S.I.R. was a 9-year prospective study conducted in 5212 men and women from the French general population (27, 28). Participants were recruited in 10 healthcare centers in the Western central part of France on the occasion of a periodic health check-up offered by the Social Security. The study protocol included extensive clinical, nutritional, lifestyle, familial, socioeconomic, and biological evaluations at inclusion (between 1994 and 1998) and at visits after 3, 6, and 9 years of follow-up. Blood (total blood for DNA extraction, plasma, and serum) and urine samples were obtained at all time points, baseline and follow-up visits. Participants answered a self-administered questionnaire including their mean daily intake of water, sweet beverages, beer/cider, wine, and spirits. Daily water and sweet beverage intakes were each categorized into 6 mutually exclusive levels: none, less than 0.5 L, 1 L, 1.5 L, 2 L, or more than 2 L. In the present analyses, they were grouped into 3 classes: less than 0.5 L, 0.5–1 L, and more than 1 L. Three classes of glycemic status were defined on the basis of fasting plasma glucose (FPG) and glucose-lowering medication (29). Normal fasting glucose (NFG) was defined as FPG less than 6.1 mmol/L and impaired fasting glucose (IFG) as FPG between 6.1 and 6.9 mmol/L, in the absence of treatment with a glucose-lowering agent for both classes. Diabetes was defined as FPG more than or equal to 7.0 mmol/L or as FPG less than 7.0 mmol/L in the presence of a treatment by a glucose-lowering drug and previous diagnosis of diabetes. The research protocol was approved by the ethics committee of Bicêtre Hospital, and all participants signed an informed consent. At baseline, 4634 individuals (88.9%) presented with NFG, 342 (6.6%) with IFG, and 134 (2.6%) with T2DM. The glycemic status at baseline of 102 participants (1.9%) could not be ascertained, and they were excluded from the analyses. Clinical characteristics by glycemic status at baseline were published previously (28, 30) and are summarized in Table 1. NFG individuals lost during follow-up and whose final glycemic status could not be determined (n = 1003) were excluded from the analyses of incidence.
Table 1.
Clinical Characteristics at Baseline by Glycemic Status at Baseline and by Incidence of IFG or T2DM During Follow-Up. The D.E.S.I.R. Study
| Glycemic Status at Baseline |
P Value | Incidence of Hyperglycemia During Follow-Upa |
P Value | |||||
|---|---|---|---|---|---|---|---|---|
| NFG | IFG | T2DM | NFG | Incident Cases of IFG | Incident Cases of T2DM | |||
| n | 4634 | 342 | 134 | 3065 | 460 | 106 | ||
| Age (y) | 47 ± 10 | 52 ± 9 | 55 ± 8 | <.0001 | 47 ± 10 | 49 ± 9 | 49 ± 9 | <.0001 |
| Sex: men/women (%) | 47/53 | 73/27 | 71/29 | <.0001 | 44/56 | 60/40 | 69/31 | <.0001 |
| BMI (kg/m2) | 24.5 ± 3.7 | 26.9 ± 4.1 | 28.9 ± 4.3 | <.0001 | 24.1 ± 3.4 | 25.9 ± 3.7 | 27.8 ± 4.7 | <.0001 |
| FPG (mmol/L) | 5.2 ± 0.4 | 6.4 ± 0.2 | 8.4 ± 2.7 | <.0001 | 5.1 ± 0.4 | 5.5 ± 0.4 | 5.6 ± 0.4 | <.0001 |
| HbA1c (%) | 5.4 ± 0.4 | 5.8 ± 0.5 | 6.9 ± 1.6 | <.0001 | 5.4 ± 0.4 | 5.5 ± 0.4 | 5.6 ± 0.4 | <.0001 |
| HbA1c (mmol/mol) | 35 ± 4 | 40 ± 5 | 52 ± 18 | <.0001 | 35 ± 4 | 37 ± 4 | 38 ± 5 | <.0001 |
| Fasting serum insulin (pmol/L) | 45 ± 27 | 65 ± 35 | 87 ± 66 | <.0001 | 42 ± 22 | 51 ± 31 | 72 ± 53 | <.0001 |
| HOMA%B (%) | 87 ± 28 | 73 ± 26 | 62 ± 32 | <.0001 | 87 ± 27 | 82 ± 27 | 101 ± 50 | <.0001 |
| HOMA%S (%) | 118 ± 47 | 84 ± 40 | 67 ± 40 | <.0001 | 122 ± 46 | 104 ± 43 | 86 ± 45 | <.0001 |
All data were obtained at baseline and are expressed as mean ± SD. P values are from Pearson's χ2 test (for sex) or ANOVA with log-transformed data.
Glycemic status during follow-up of participants who had NFG at baseline.
New cases of IFG and T2DM were observed in 460 (12.7%) and 106 (2.9%) people, respectively, out of the 3631 participants with NFG at baseline. Clinical characteristics at baseline of incident cases of IFG or T2DM and of participants who remained with NFG during follow-up are summarized in Table 1. Glycemic status at the end of follow-up (n = 4107) was: NFG 74.6%, IFG 17.2%, and T2DM 8.2%. For the purpose of this investigation, to increase statistical power we have considered the incidence of hyperglycemia (IFG and T2DM, combined) during follow-up. For the measurement of copeptin, 1800 participants were randomly selected after stratification on sex, age, body mass index (BMI), and class of water intake at baseline (18, 31). Characteristics of participants selected for copeptin measurement and of other participants are shown in Supplemental Table 1. The flow chart and exclusion criteria of participants are shown in Supplemental Figure 1.
Procedures
Copeptin concentration was measured recently in fasting plasma-EDTA samples, collected at baseline and kept frozen at −80°C. An automated immunofluorescent sandwich immunoassay was used (B·R·A·H·M·S Copeptin US KRYPTOR CT-proAVP; Thermo Fisher Scientific) (16, 17). The lower limit of detection was 0.9 pmol/L. Intraassay and interassay coefficients of variation reported by the manufacturer are less than 8% and less than 10%, respectively, for a concentrations range of 4.0–15.0 pmol/L, and less than 15% and less than 18%, respectively, for a concentrations range of 2.0–4.0 pmol/L. Plasma glucose was assessed by an enzymatic method (modified glucose oxidase peroxidase) with automated analyzers (Technicon RA1000; Bayer Diagnostics or Specific or Delta, Konelab). Serum insulin was quantified by micro particle enzyme immunoassay with an automated analyzer (IMX, Abbott; cross-reactivity with proinsulin <0.02%).
AVP spans approximately 3 kb on chromosome 20p13 and is contained in a single haplotype block. Six single nucleotide polymorphisms (SNP) were chosen in HapMap (public release number 28) on the basis of giving information on 80% of the allelic variation of SNPs with minor allele frequency more than 5% at r2 > 0.8 in the haplotype block containing AVP: rs6084265 (∼5.5 kb at 5′ from start of exon 1), rs6084264 (∼5.2 kb at 5′ from start of exon 1), rs3761249 (∼1 kb at 5′ from start of exon 1), rs2282018 (intron 1), rs2770381 (∼1.2 kb from the end of 3′ untranslated region), and rs1410713 (∼2.9 kb from the end of 3′ untranslated region). The structure of AVP, location of the SNPs, and linkage disequilibrium between SNPs are shown in Supplemental Figure 2. Genotypes were determined by competitive allele-specific PCR genotyping system assay (KASP; LGC Genomics). Genotyping success rate was 93%–98%.
Computations and statistical analyses
Homeostasis model assessment index of β-cell function (HOMA%B) and homeostasis model assessment index of insulin sensitivity (HOMA%S), based on fasting circulating levels of glucose and insulin were computed for each time point (baseline and follow-up visits) with the Microsoft Excel spreadsheet implementation of the HOMA Calculator (v2.2) available at http://www.dtu.ox.ac.uk/homacalculator/download.php (accessed in March 2006). Differences between groups were assessed by ANOVA, analysis of covariance (ANCOVA), contingency table χ2 test, and Fisher's exact test. Comparison of HOMA%B and HOMA%S throughout the study (baseline and follow-up combined) between groups used mixed regression models with random effects. This computation method maximizes the data by allowing for different sample sizes at different time points to be used and takes into account that samples from the same individual are not independent. Results are expressed as mean ± SEM, unless otherwise stated. Interaction between sex and genotype in the comparisons of copeptin level was assessed by including in the regression model a “crossed” compound covariate (sex/genotype). Stratification by sex of genotype-related comparisons was then performed by nesting the genotype variable within the sex variable in the regression analysis. This results in the computation of statistical effects for men and women separately, and adjusted for multiple comparisons due to the stratification by sex. Cox proportional hazards survival regression analyses examined the effect of explanatory variables on time-related incident rates in prospective analyses. Adjustments for clinical and biological parameters were carried out by including these parameters as covariates in the regressive model. Kaplan-Meier curves were used to plot cumulated incidences over time. For all analyses, data were log-transformed for the analyses when the normality of the distribution was rejected by the Shapiro-Wilk W test.
Correction for multiple comparisons due to multiple SNP testing took into account the effective number of independent tests (Meff) based on the degree of linkage disequilibrium between SNPs (32). Thus, for genotype-related comparisons, P < .02 was considered significant; P < .05 was considered significant for other comparisons. The power to detect associations of the SNPs with the incidence of IFG/T2DM was more than 80% for hazard ratio (HR) more than or equal to 1.2 and α = 0.05. The power to detect a 20% difference of plasma copeptin concentration in sex and age-adjusted comparisons by glucose status was 91%. Statistics used JMP (SAS Institute, Inc) and Stata (StataCorp) softwares.
Results
Plasma copeptin and glycemic status
Plasma copeptin at baseline was higher in men than in women: 6.48 ± 5.41 vs 4.18 ± 3.86 (mean ± SD, ANOVA; P < .0001). Plasma copeptin by the glycemic status during the study is shown in Table 2. Plasma copeptin at baseline was significantly higher in people with T2DM than in people with IFG or NFG at baseline. It was also significantly higher at baseline in incident cases of IFG or T2DM during follow-up than in participants who remained normoglycemic during the study. The cumulated incidence of IFG/T2DM during follow-up by quartiles of plasma copeptin levels was 11.0% (Q1), 14.5% (Q2), 17.0% (Q3), and 23.5% (Q4); log-rank test following Kaplan-Meier incidence curve P = .003 (Figure 1). Cox proportional hazards survival regression analysis confirmed the association of plasma copeptin with the incidence of IFG/T2DM: HR per unit of log[copeptin], 1.55, 95% confidence interval (CI), 1.21–2.00; P = .0006, adjusted for sex, age, and BMI. The association was observed both in men (HR, 1.42; 95% CI, 1.05–1.95; P = .03) and in women (HR, 1.86; 95% CI, 1.19–2.89; P = .007). It remained significant when water intake, FPG, and fasting serum insulin at baseline were included in the regression model: HR, 1.54; 95% CI, 1.18–1.99; P = .001.
Table 2.
Plasma Copeptin Concentration at Baseline by Glycemic Status During the Study. The D.E.S.I.R. Study
| Glycemic Status | n | Copeptin (pmol/L) | P Value | P Value* |
|---|---|---|---|---|
| Baseline | ||||
| NFG | 1601 | 5.76 ± 0.20 | ||
| IFG | 115 | 5.74 ± 0.46 | ||
| T2DM | 47 | 9.05 ± 0.71 | .03 | .01 |
| Follow-upa | ||||
| NFG | 1076 | 5.32 ± 0.16 | ||
| Incident cases of IFG | 181 | 6.04 ± 0.31 | ||
| Incident cases of T2DM | 30 | 6.42 ± 0.46 | .002 | .002 |
Results expressed as mean ± SEM. P values are from ANCOVA adjusted for sex and age (P value) and with further adjustment for water intake (P value*). Analyses performed with log-transformed data.
Glycemic status during follow-up of participants who had NFG at baseline.
Figure 1.
Cumulated incidence of hyperglycemia (IFG or T2DM) during follow-up by quartiles of baseline plasma copeptin concentrations: Q1 (dotted line), Q2 (small dash line), Q3 (large dash line), and Q4 (solid line). Log-rank test for incidence of IFG/T2DM by quartiles: χ2 = 19.2, P = .0003. The D.E.S.I.R. Study.
Plasma copeptin, insulin sensitivity, and insulin secretion
HOMA%S and HOMA%B at baseline and at years 3, 6, and 9 of follow-up by quartiles of plasma copeptin at baseline are shown in Table 3. Participants in the upper quartile of plasma copeptin had significantly lower insulin sensitivity (HOMA%S) at baseline and during follow-up than other participants. These results were confirmed with more robust statistical significance in pooled analyses of baseline plus follow-up data. Insulin secretion (HOMA%B) at baseline or during follow-up was not significantly different across quartiles of baseline plasma copeptin.
Table 3.
HOMA%S and HOMA%B During the Study by Baseline Plasma Copeptin Quartiles. The D.E.S.I.R. Study
| Baseline Plasma Copeptin (Quartiles) |
P Value | P Value* | ||||
|---|---|---|---|---|---|---|
| Q1 | Q2 | Q3 | Q4 | |||
| Baseline (n) | 413 | 401 | 408 | 400 | ||
| HOMA%S (%) | 119 ± 3 | 120 ± 3 | 117 ± 3 | 113 ± 3 | .06 | .01 |
| HOMA%B (%) | 85 ± 1 | 85 ± 1 | 85 ± 1 | 83 ± 1 | .42 | |
| Follow-up year 3 (n) | 361 | 358 | 348 | 348 | ||
| HOMA%S (%) | 115 ± 3 | 118 ± 3 | 114 ± 3 | 107 ± 3 | .007 | .001 |
| HOMA%B (%) | 84 ± 1 | 85 ± 1 | 85 ± 1 | 82 ± 1 | .25 | |
| Follow-up year 6 (n) | 327 | 312 | 303 | 290 | ||
| HOMA%S (%) | 108 ± 3 | 112 ± 3 | 109 ± 3 | 101 ± 3 | .01 | .002 |
| HOMA%B (%) | 86 ± 1 | 86 ± 1 | 87 ± 1 | 87 ± 1 | .88 | |
| Follow-up year 9 (n) | 325 | 324 | 312 | 290 | ||
| HOMA%S (%) | 112 ± 3 | 115 ± 3 | 113 ± 3 | 107 ± 3 | .19 | .03 |
| HOMA%B (%) | 84 ± 1 | 84 ± 1 | 87 ± 1 | 84 ± 1 | .49 | |
| Baseline and follow-up (n)a | 1426 | 1395 | 1371 | 1328 | ||
| HOMA%S (%) | 116 ± 2 | 117 ± 2 | 115 ± 2 | 108 ± 2 | .002 | .0001 |
| HOMA%B (%) | 86 ± 1 | 86 ± 1 | 87 ± 1 | 85 ± 1 | .39 | |
Results expressed as mean ± SEM. Quartiles of baseline plasma copeptin (median; range): Q1 (2.19 pmol/L; 0.91–2.92 pmol/L), Q2 (3.46 pmol/L; 2.93–4.05 pmol/L), Q3 (5.15 pmol/L; 4.06–6.57 pmol/L), and Q4 (8.74 pmol/L; 6.58–115 pmol/L). P values are from ANCOVA, with log-transformed data. P values*, Q4 vs Q1-Q2-Q3 pooled data. HOMA%S comparisons are adjusted for sex, age, and BMI. HOMA%B comparisons are adjusted for sex, age, BMI, and HOMA%S.
Baseline and follow-up values were compared between groups in mixed regression models with random effects, including as additional covariates duration of follow-up and subject ID.
Plasma copeptin concentration by AVP genotype
We observed an interaction between sex and genotype in the comparisons of plasma copeptin by AVP genotypes (P < .02 for all interactions) (Table 4). In men, the CC genotype of rs6084264, the TT genotype of rs3761249 and rs2282018, and the CC genotype of rs2770381 and rs1410713 were significantly associated with higher plasma copeptin. No significant association was observed in women.
Table 4.
Plasma Copeptin Concentration by AVP Genotypes. The D.E.S.I.R. Study
| SNP | Women |
Men |
*P Value (Interaction) | ||||
|---|---|---|---|---|---|---|---|
| n | Copeptin (pmol/L) | P Value | n | Copeptin (pmol/L) | P Value | ||
| rs6084265 | |||||||
| CC | 303 | 5.02 ± 0.59 | 304 | 8.46 ± 0.55 | |||
| TC | 355 | 4.84 ± 0.58 | 368 | 8.46 ± 0.52 | |||
| TT | 143 | 5.02 ± 0.63 | .71 | 150 | 8.25 ± 0.66 | .88 | .89 |
| rs6084264 | |||||||
| CC | 348 | 4.99 ± 0.60 | 368 | 9.12 ± 0.53 | |||
| TC | 336 | 5.23 ± 0.61 | 342 | 8.04 ± 0.52 | |||
| TT | 112 | 5.17 ± 0.69 | .51 | 117 | 7.58 ± 0.66 | .007 | .02 |
| rs3761249 | |||||||
| TT | 644 | 5.00 ± 0.55 | 682 | 8.62 ± 0.47 | |||
| TG/GG | 184/20 | 5.01 ± 0.59 | .98 | 161/19 | 7.38 ± 0.59 | .0007 | .002 |
| rs2282018 | |||||||
| TT | 358 | 4.78 ± 0.54 | 380 | 9.14 ± 0.62 | |||
| TC | 382 | 5.05 ± 0.55 | 355 | 8.01 ± 0.49 | |||
| CC | 117 | 4.94 ± 0.62 | .71 | 134 | 7.57 ± 0.62 | .002 | .008 |
| rs2770381 | |||||||
| AA | 298 | 4.94 ± 0.56 | 314 | 8.30 ± 0.52 | |||
| AC | 384 | 4.36 ± 0.57 | 389 | 8.27 ± 0.50 | |||
| CC | 161 | 4.65 ± 0.61 | .34 | 157 | 9.48 ± 0.61 | .002 | .004 |
| rs1410713 | |||||||
| CC | 363 | 4.88 ± 0.59 | 380 | 9.25 ± 0.52 | |||
| CA | 353 | 5.11 ± 0.60 | 353 | 7.75 ± 0.52 | |||
| AA | 94 | 5.51 ± 0.59 | .05 | 105 | 7.80 ± 0.66 | .0002 | <.0001 |
SNPs are sorted in 5′ to 3′ order. Results expressed as mean ± SEM. P values are from ANCOVA, with log-transformed data and adjusted for age, water intake and blood glucose status (NFG, IFG, or diabetes).
P Value (interaction) is the statistical significance observed in the regression model for a crossed compound covariate “sex/genotype.”
Incidence of IFG/T2DM by AVP genotype
The cumulated incidence of IFG/T2DM during follow-up by genotype was as follows: 17.0% (CC), 14.7% (TC) and 14.2% (TT) for rs6084264; 16.9% (TT), 14.9% (TC) and 13.8% (CC) for rs2282018; 14.0% (AA), 16.1% (AC) and 17.6% (CC) for rs2770381; 17.5% (CC) and 14.5% (CA/AA) for rs1410713. Cox proportional hazards survival regression analyses showed significant associations of the CC genotype of rs6084264, the TT genotype of rs2282018, the C-allele of rs2770381 and the CC genotype of rs1410713 with the incidence of IFG/T2DM (Table 5). We observed no significant sex/genotype interaction in these associations. Supplemental Table 2 summarizes the associations of AVP genotypes with plasma copeptin levels and with the incidence of IFG/T2DM during follow-up.
Table 5.
Genotype Frequencies of AVP Polymorphisms by the Incidence of IFG/Type 2 Diabetes During Follow-Up. The D.E.S.I.R. Study
| NFG During Follow-Up |
Incident Cases of IFG/T2DM |
HR (95% CI) | P Value | Model | |||
|---|---|---|---|---|---|---|---|
| n | Genotype Frequency | n | Genotype Frequency | ||||
| rs6084265 | |||||||
| CC | 1050 | 0.373 | 183 | 0.347 | |||
| TC | 1266 | 0.449 | 237 | 0.450 | |||
| TT | 502 | 0.178 | 107 | 0.203 | 1.17 (0.94–1.44) | .15 | TT vs XC |
| rs6084264 | |||||||
| CC | 1253 | 0.442 | 257 | 0.488 | |||
| TC | 1229 | 0.434 | 212 | 0.402 | |||
| TT | 352 | 0.124 | 58 | 0.110 | 1.22 (1.03–1.45) | .02 | CC vs TX |
| rs3761249 | |||||||
| TT | 2314 | 0.778 | 435 | 0.789 | |||
| TG | 604 | 0.203 | 105 | 0.191 | |||
| GG | 55 | 0.019 | 11 | 0.020 | 1.03 (0.78–1.27) | .79 | TT vs XG |
| rs2282018 | |||||||
| TT | 1272 | 0.425 | 259 | 0.468 | |||
| TC | 1340 | 0.448 | 234 | 0.422 | |||
| CC | 382 | 0.127 | 61 | 0.110 | 1.21 (1.03–1.43) | .02 | TT vs XC |
| rs2770381 | |||||||
| AA | 1102 | 0.371 | 180 | 0.328 | |||
| AC | 1356 | 0.457 | 260 | 0.474 | |||
| CC | 511 | 0.172 | 109 | 0.198 | 1.31 (1.04–1.66) | .02 | Codominant C |
| rs1410713 | |||||||
| CC | 1312 | 0.460 | 279 | 0.516 | |||
| CA | 1222 | 0.429 | 204 | 0.377 | |||
| AA | 317 | 0.111 | 58 | 0.107 | 1.28 (1.08–1.52) | .004 | CC vs XA |
SNPs are sorted in 5′ to 3′ order. P values for the test for HWE were P = .0004 and P = .06 (rs6084265), P = .006 and P = .16 (rs6084264), P = .04 and P = .12 (rs3761249), P = .32 and P = .46 (rs2282018), P = .008 and P = .39 (rs2770381), and P = .20 and P = .03 (rs1410713) for NFG and IFG/T2DM cases during follow-up, respectively. Hazards ratio determined in Cox proportional hazards survival regression analyses, adjusted for sex, age, BMI, and water intake.
Discussion
In this study, we demonstrated that high plasma copeptin at baseline is associated with decreased insulin sensitivity and with increased cumulated incidence of IFG or T2DM during a 9-year follow-up in a cohort of French people from the general population. These results complement our previous observation in this cohort that water intake was inversely and independently associated with the risk of developing hyperglycemia (26) and that circulating levels of copeptin reflected the hydration status (water intake and urine osmolarity) (18). The present study is the first to investigate and report associations between a set of tagSNPs in AVP, plasma copeptin and incidence of hyperglycemia during a long-term follow-up.
The allelic associations observed in our study suggest a pattern of Mendelian randomization (33), at least in men, and support the causality of the association between vasopressin and hyperglycemia. However, the genetic basis of the association of these common variants with the phenotypes is unclear. None of the SNPs has obvious functional properties that could predict deleterious effects on AVP expression or function. The genotypes associated with an increased risk of hyperglycemia were associated with higher plasma copeptin in men but not in women. Possible nonexclusive explanations for this sexual dimorphism include lower statistical power in women than in men, as well as the uncertainty regarding the functional variant. Lower statistical power might be related to the lower plasma copeptin concentration in women than in men, observed in our cohort and in other populations (22, 24, 25, 34–37) and the smaller range of the distribution of plasma copeptin in women than in men (interquartile range 2.33 vs 4.38 pmol/L).
Other investigations have shown circulating copeptin to be positively correlated with insulin resistance and/or with the metabolic syndrome in African-American (21), non-Hispanic White American (21), and North European cohorts (35, 38), and with the incidence or the prevalence of T2DM in North European cohorts (22, 24, 25, 35, 38). An earlier study showed that plasma vasopressin concentration was markedly elevated in patients with poorly controlled diabetes mellitus (13).
The mechanisms of vasopressin-induced hyperglycemia are only partially understood. Vasopressin binds to 3 different G protein-coupled receptors. V1a receptors are widely expressed, particularly in vascular smooth muscle cells, hepatocytes, platelets, and the central nervous system. V1b receptors are expressed in the endocrine pancreas, in cells of the anterior pituitary and throughout the brain. V2 receptors are predominantly expressed in the kidney collecting ducts, but possibly also in the liver. Vasopressin stimulates gluconeogenesis and glycogenolysis through the activation of hepatic V1a receptors (3–6), and the release either of glucagon or insulin, depending on concomitant extracellular glucose levels, through the activation of V1b receptors in pancreatic islets (7). It also stimulates the release of adrenocorticotropic hormone and cortisol through activation of pituitary V1b receptors (39, 40). Acute infusion of vasopressin in rodents and in healthy people was shown to induce a transient rise in blood glucose levels (11, 12). Recently, we showed that chronic ip infusion of vasopressin in obese Zucker rats and lean controls induces an increase in FPG in lean animals, and promotes hyperinsulinemia and glucose intolerance in obese animals (6). These effects were blunted by the coadministration of a V1a receptor antagonist, whereas the coadministration of a V1b receptor antagonist had no protective effect. The role of the liver in these vasopressin effects is likely to be central, because V1a receptors are expressed in hepatocytes, and because vasopressin markedly increased the plasma glucose response to a pyruvate test that assesses hepatic gluconeogenic capacity. Moreover, low circulating levels of vasopressin obtained by increasing daily water intake were associated with a drastic reduction in liver steatosis in obese rats (6). However, studies in vasopressin-deficient mice suggest that vasopressin-induced insulin resistance might be mediated through V1b receptors. Mice that lack V1b receptors have increased insulin sensitivity and low blood glucose (9), whereas V1a receptor-deficient mice have high AVP levels, insulin resistance, and impaired glucose tolerance (8). Thus, the vasopressin system seems critical to glucose homeostasis in several species, but the differences in the signaling pathways observed in different species strengthen the need for human investigations, notably tracer studies to assess the impact of vasopressin action on hepatic and peripheral insulin sensitivities.
The design of our study had some intrinsic limitations. Glucose tolerance status of participants was based on FPG only and not on an oral glucose tolerance test. Secondly, insulin secretion and insulin sensitivity were assessed by HOMA indices that provide only an estimate of these parameters. Finally, plasma copeptin was assessed only in a subset of people in our cohort. There was a slight excess of rs1410713 AA homozygotes among incident cases of IFG/T2DM, and of rs2770381 CC, rs3761249 GG and rs6084265 TT homozygotes among NFG individuals, resulting in a deviation from Hardy-Weinberg equilibrium (HWE) in these groups (Table 5). The lack of HWE could be due to chance or may be a sign of depletion of a specific haplotype due to unknown causes. However, the associations of these SNPs with the incidence of IFG/T2DM could not be attributed to bias induced by HWE deviation, because the genotypes in excess were the protective genotype of rs1410713 in incident cases of IFG/T2DM and the risk genotype of rs2770381 in NFG individuals. On the contrary, the association might be more robust in the presence of HWE. The main strengths of our study are the detailed phenotype assessment during the 9-year follow-up of the D.E.S.I.R. Study and the genotyping of tagSNPs covering the haplotype block containing AVP.
In conclusion, the present investigation confirmed in a community-based cohort the association between plasma copeptin and reduced insulin sensitivity and increased risk for IFG and diabetes. Moreover, it extends these observations by showing allelic associations in men between AVP variants and plasma copeptin and the risk of IFG and diabetes. The results support a causal role for vasopressin in these disorders. They provide a rationale for intervention studies aiming to evaluate if a reduction of vasopressin secretion or action, achieved by an increase in water intake or by treatment with vasopressin receptor antagonists (vaptans), could improve the metabolic status. If confirmed, these interventions could then be indicated in people at high risk of diabetes and presenting high copeptin levels, who, presumably, would benefit the most from the reduction of vasopressin effects.
Acknowledgments
Measurement of plasma copeptin was performed by Thermofisher Scientific (Hennigsdorf, Germany), in anonymized tubes, blinded to characteristics and outcomes of patients.
This work was supported by a Conventions Industrielles de Formation par la Recherche grant from Inserm and Danone Research (R.E.B.). The D.E.S.I.R. Study has been supported by Inserm contracts with Caisse Nationale de l'Assurance Maladie des Travailleurs Salariés, Lilly, Novartis Pharma, and Sanofi-Aventis and by Inserm (Réseaux en Santé Publique, Interactions entre les Déterminants de la Santé), Cohortes Santé TGIR, the Association Diabète Risque Vasculaire, the Fédération Française de Cardiologie, La Fondation de France, Association de Langue Française pour l'Etude du Diabète et des Maladies Métaboliques, Société Francophone du Diabète, Office National Interprofessionnel des Vins, Abbott, Ardix Medical, Bayer Diagnostics, Becton Dickinson, Cardionics, Merck Santé, Novo Nordisk, Pierre Fabre, Roche, and Topcon.
The D.E.S.I.R. Study Group. INSERM U1018: B. Balkau, P. Ducimetière, E. Eschwège; INSERM U367: F. Alhenc-Gelas; CHU D'Angers: A. Girault; Bichat Hospital: F. Fumeron, M. Marre, R Roussel; CHU de Rennes: F. Bonnet; CNRS UMR8090, Lille: S. Cauchi, P. Froguel; Centres d'Examens de Santé: Alençon, Angers, Blois, Caen, Chateauroux, Chartres, Cholet, Le Mans, Orléans, Tours; Institute de Recherche Médecine Générale: J. Cogneau; General practitioners of the region; Institute inter-Regional pour la Santé: C. Born, E. Caces, M. Cailleau, O Lantieri, J.G. Moreau, F. Rakotozafy, J. Tichet, S. Vol.
Disclosure Summary: The authors have nothing to disclose.
Footnotes
- ANCOVA
- analysis of covariance
- AVP
- arginine vasopressin-neurophysin II gene
- BMI
- body mass index
- CI
- confidence interval
- D.E.S.I.R.
- Data from Epidemiological Study on the Insulin Resistance Syndrome
- FPG
- fasting plasma glucose
- HOMA%B
- homeostasis model assessment index of β-cell function
- HOMA%S
- homeostasis model assessment index of insulin sensitivity
- HR
- hazard ratio
- HWE
- Hardy-Weinberg equilibrium
- IFG
- impaired fasting glucose
- NFG
- normal fasting glucose
- SNP
- single nucleotide polymorphism
- T2DM
- type 2 diabetes mellitus.
References
- 1. Laycock JF. Vasopressin and its interactions with other hormones and control systems. In: Laycock JF, ed. Perspectives on Vasopressin. London, UK: Imperial College Press; 2010:202–229. [Google Scholar]
- 2. Mavani GP, DeVita MV, Michelis MF. A review of the nonpressor and nonantidiuretic actions of the hormone vasopressin. Front Med. 2015;2:19. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Hems DA, Rodrigues LM, Whitton PD. Rapid stimulation by vasopressin, oxytocin and angiotensin II of glycogen degradation in hepatocyte suspensions. Biochem J. 1978;172:311–317. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Whitton PD, Rodrigues LM, Hems DA. Stimulation by vasopressin, angiotensin and oxytocin of gluconeogenesis in hepatocyte suspensions. Biochem J. 1978;176:893–898. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Koshimizu TA, Nakamura K, Egashira N, Hiroyama M, Nonoguchi H, Tanoue A. Vasopressin V1a and V1b receptors: from molecules to physiological systems. Physiol Rev. 2012;92:1813–1864. [DOI] [PubMed] [Google Scholar]
- 6. Taveau C, Chollet C, Waeckel L, et al. Vasopressin and hydration play a major role in the development of glucose intolerance and hepatic steatosis in obese rats. Diabetologia. 2015;58:1081–1090. [DOI] [PubMed] [Google Scholar]
- 7. Abu-Basha EA, Yibchok-Anun S, Hsu WH. Glucose dependency of arginine vasopressin-induced insulin and glucagon release from the perfused rat pancreas. Metabolism. 2002;51:1184–1190. [DOI] [PubMed] [Google Scholar]
- 8. Aoyagi T, Birumachi J, Hiroyama M, et al. Alteration of glucose homeostasis in V1a vasopressin receptor-deficient mice. Endocrinology. 2007;148:2075–2084. [DOI] [PubMed] [Google Scholar]
- 9. Fujiwara Y, Hiroyama M, Sanbe A, et al. Insulin hypersensitivity in mice lacking the V1b vasopressin receptor. J Physiol. 2007;584:235–244. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Nakamura K, Aoyagi T, Hiroyama M, et al. Both V(1A) and V(1B) vasopressin receptors deficiency result in impaired glucose tolerance. Eur J Pharmacol. 2009;613:182–188. [DOI] [PubMed] [Google Scholar]
- 11. Hems DA, Whitton PD, Ma GY. Metabolic actions of vasopressin, glucagon and adrenalin in the intact rat. Biochim Biophys Acta. 1975;411:155–164. [DOI] [PubMed] [Google Scholar]
- 12. Spruce BA, McCulloch AJ, Burd J, et al. The effect of vasopressin infusion on glucose metabolism in man. Clin Endocrinol. 1985;22:463–468. [DOI] [PubMed] [Google Scholar]
- 13. Zerbe RL, Vinicor F, Robertson GL. Plasma vasopressin in uncontrolled diabetes mellitus. Diabetes. 1979;28:503–508. [DOI] [PubMed] [Google Scholar]
- 14. Brooks DP, Nutting DF, Crofton JT, Share L. Vasopressin in rats with genetic and streptozocin-induced diabetes. Diabetes. 1989;38:54–57. [DOI] [PubMed] [Google Scholar]
- 15. Yi SS, Hwang IK, Kim YN, et al. Enhanced expressions of arginine vasopressin (Avp) in the hypothalamic paraventricular and supraoptic nuclei of type 2 diabetic rats. Neurochem Res. 2008;33:833–841. [DOI] [PubMed] [Google Scholar]
- 16. Morgenthaler NG, Struck J, Alonso C, Bergmann A. Assay for the measurement of copeptin, a stable peptide derived from the precursor of vasopressin. Clin Chem. 2006;52:112–119. [DOI] [PubMed] [Google Scholar]
- 17. Morgenthaler NG. Copeptin: a biomarker of cardiovascular and renal function. Congest Heart Fail. 2010;16(suppl 1):S37–S44. [DOI] [PubMed] [Google Scholar]
- 18. Roussel R, Fezeu L, Marre M, et al. Comparison between copeptin and vasopressin in a population from the community and in people with chronic kidney disease. J Clin Endocrinol Metab. 2014;99:4656–4663. [DOI] [PubMed] [Google Scholar]
- 19. Balanescu S, Kopp P, Gaskill MB, Morgenthaler NG, Schindler C, Rutishauser J. Correlation of plasma copeptin and vasopressin concentrations in hypo-, iso-, and hyperosmolar States. J Clin Endocrinol Metab. 2011;96:1046–1052. [DOI] [PubMed] [Google Scholar]
- 20. Szinnai G, Morgenthaler NG, Berneis K, et al. Changes in plasma copeptin, the c-terminal portion of arginine vasopressin during water deprivation and excess in healthy subjects. J Clin Endocrinol Metab. 2007;92:3973–3978. [DOI] [PubMed] [Google Scholar]
- 21. Saleem U, Khaleghi M, Morgenthaler NG, et al. Plasma carboxy-terminal provasopressin (copeptin): a novel marker of insulin resistance and metabolic syndrome. J Clin Endocrinol Metab. 2009;94:2558–2564. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Enhörning S, Wang TJ, Nilsson PM, et al. Plasma copeptin and the risk of diabetes mellitus. Circulation. 2010;121:2102–2108. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Enhörning S, Struck J, Wirfält E, Hedblad B, Morgenthaler NG, Melander O. Plasma copeptin, a unifying factor behind the metabolic syndrome. J Clin Endocrinol Metab. 2011;96:E1065–E1072. [DOI] [PubMed] [Google Scholar]
- 24. Abbasi A, Corpeleijn E, Meijer E, et al. Sex differences in the association between plasma copeptin and incident type 2 diabetes: the Prevention of Renal and Vascular Endstage Disease (PREVEND) study. Diabetologia. 2012;55:1963–1970. [DOI] [PubMed] [Google Scholar]
- 25. Then C, Kowall B, Lechner A, et al. Plasma copeptin is associated with type 2 diabetes in men but not in women in the population-based KORA F4 study. Acta Diabetol. 2015;52:103–112. [DOI] [PubMed] [Google Scholar]
- 26. Roussel R, Fezeu L, Bouby N, et al. Low water intake and risk for new-onset hyperglycemia. Diabetes Care. 2011;34:2551–2554. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. The D.E.S.I.R. Study Group. An epidemiologic survey from a network of French Health Examination Centres. D.E.S.I.R. (Epidemiologic Data on the Insulin Resistance Syndrome). Rev Epidem et Sante Publ. 1996;44:373–375. [PubMed] [Google Scholar]
- 28. Balkau B, Lange C, Fezeu L, et al. Predicting diabetes: clinical, biological, and genetic approaches: data from the Epidemiological Study on the Insulin Resistance Syndrome (DESIR). Diabetes Care. 2008;31:2056–2061. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. World Health Organization, International Diabetes Federation. Definition and diagnosis of diabetes mellitus and intermediate hyperglycaemia. Report of a WHO/IDF consultation. Geneva, Switzerland: WHO Press; 2006. [Google Scholar]
- 30. Cheurfa N, Brenner GM, Reis AF, et al. Decreased insulin secretion and increased risk of type 2 diabetes associated with allelic variations of the WFS1 gene: the Data from Epidemiological Study on the Insulin Resistance Syndrome (DESIR) prospective study. Diabetologia. 2011;54:554–562. [DOI] [PubMed] [Google Scholar]
- 31. Roussel R, Matallah N, Bouby N, et al. Plasma copeptin and decline in renal function in a cohort from the community: the prospective D.E.S.I.R. Study. Am J Nephrol. 2015;42:107–114. [DOI] [PubMed] [Google Scholar]
- 32. Nyholt DR. A simple correction for multiple testing for single-nucleotide polymorphisms in linkage disequilibrium with each other. Am J Hum Genet. 2004;74:765–769. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Thomas DC, Conti DV. Commentary: the concept of 'Mendelian randomization'. Int J Epidemiol. 2004;33:21–25. [DOI] [PubMed] [Google Scholar]
- 34. Bhandari SS, Loke I, Davies JE, Squire IB, Struck J, Ng LL. Gender and renal function influence plasma levels of copeptin in healthy individuals. Clin Sci (Lond). 2009;116:257–263. [DOI] [PubMed] [Google Scholar]
- 35. Enhorning S, Bankir L, Bouby N, et al. Copeptin, a marker of vasopressin, in abdominal obesity, diabetes and microalbuminuria: the prospective Malmo Diet and Cancer Study cardiovascular cohort. Int J Obes (Lond). 2013;37:598–603. [DOI] [PubMed] [Google Scholar]
- 36. Meijer E, Bakker SJ, Halbesma N, de Jong PE, Struck J, Gansevoort RT. Copeptin, a surrogate marker of vasopressin, is associated with microalbuminuria in a large population cohort. Kidney Int. 2010;77:29–36. [DOI] [PubMed] [Google Scholar]
- 37. Velho G, Bouby N, Hadjadj S, et al. Plasma copeptin and renal outcomes in patients with type 2 diabetes and albuminuria. Diabetes Care. 2013;36:3639–3645. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Wannamethee SG, Welsh P, Papacosta O, Lennon L, Whincup PH, Sattar N. Copeptin, insulin resistance, and risk of incident diabetes in older men. J Clin Endocrinol Metab. 2015;100:3332–3339. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Gillies GE, Linton EA, Lowry PJ. Corticotropin releasing activity of the new CRF is potentiated several times by vasopressin. Nature. 1982;299:355–357. [DOI] [PubMed] [Google Scholar]
- 40. Holmes CL, Landry DW, Granton JT. Science review: vasopressin and the cardiovascular system part 1–receptor physiology. Crit Care. 2003;7:427–434. [DOI] [PMC free article] [PubMed] [Google Scholar]