Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Open Access

Peer-reviewed

Research Article

Hepatocyte Nuclear Factor 4 Alpha Polymorphisms and the Metabolic Syndrome in French-Canadian Youth

  • Valérie Marcil,

    Affiliation Research Institute, McGill University, Montreal, Quebec, Canada, H3G 1A4

    ⨯
  • Devendra Amre,

    Affiliation Department of Pediatrics, CHU-Sainte-Justine, Université de Montréal, Montreal, Quebec, Canada, H3T 1C5

    ⨯
  • Ernest G. Seidman,

    Affiliations Research Institute, McGill University, Montreal, Quebec, Canada, H3G 1A4, Canadian Institutes for Health Research Team on the Digestive Epithelium, Department of Anatomy and Cellular Biology, Faculty of Medicine and Health Sciences, Université de Sherbrooke, Sherbrooke, Quebec, Canada, J1H 5N4

    ⨯
  • François Boudreau,

    Affiliation Canadian Institutes for Health Research Team on the Digestive Epithelium, Department of Anatomy and Cellular Biology, Faculty of Medicine and Health Sciences, Université de Sherbrooke, Sherbrooke, Quebec, Canada, J1H 5N4

    ⨯
  • Fernand P. Gendron,

    Affiliation Canadian Institutes for Health Research Team on the Digestive Epithelium, Department of Anatomy and Cellular Biology, Faculty of Medicine and Health Sciences, Université de Sherbrooke, Sherbrooke, Quebec, Canada, J1H 5N4

    ⨯
  • Daniel Ménard,

    Affiliation Canadian Institutes for Health Research Team on the Digestive Epithelium, Department of Anatomy and Cellular Biology, Faculty of Medicine and Health Sciences, Université de Sherbrooke, Sherbrooke, Quebec, Canada, J1H 5N4

    ⨯
  • Jean François Beaulieu,

    Affiliation Canadian Institutes for Health Research Team on the Digestive Epithelium, Department of Anatomy and Cellular Biology, Faculty of Medicine and Health Sciences, Université de Sherbrooke, Sherbrooke, Quebec, Canada, J1H 5N4

    ⨯
  • Daniel Sinnett,

    Affiliation Department of Pediatrics, CHU-Sainte-Justine, Université de Montréal, Montreal, Quebec, Canada, H3T 1C5

    ⨯
  • Marie Lambert,

    Affiliation Department of Pediatrics, CHU-Sainte-Justine, Université de Montréal, Montreal, Quebec, Canada, H3T 1C5

    ⨯
  • Emile Levy

    emile.levy@recherche-ste-Justine.qc.ca

    Affiliations Canadian Institutes for Health Research Team on the Digestive Epithelium, Department of Anatomy and Cellular Biology, Faculty of Medicine and Health Sciences, Université de Sherbrooke, Sherbrooke, Quebec, Canada, J1H 5N4, Department of Nutrition, CHU-Sainte-Justine, Université de Montréal, Montreal, Quebec, Canada, H3T 1C5

    ⨯

Abstract

Objectives

Hepatocyte nuclear factor 4 alpha (HNF4α) is a transcription factor involved in the regulation of serum glucose and lipid levels. Several HNF4A gene variants have been associated with the risk of developing type 2 diabetes mellitus. However, no study has yet explored its association with insulin resistance and the cardiometabolic risk in children. We aimed to investigate the relationship between HNF4A genetic variants and the presence of metabolic syndrome (MetS) and metabolic parameters in a pediatric population.

Design and Methods

Our study included 1,749 French-Canadians aged 9, 13 and 16 years and evaluated 24 HNF4A polymorphisms that were previously identified by sequencing.

Results

Analyses revealed that, after correction for multiple testing, one SNP (rs736824; P<0.022) and two haplotypes (P1 promoter haplotype rs6130608-rs2425637; P<0.032 and intronic haplotype rs736824-rs745975-rs3212183; P<0.025) were associated with the risk of MetS. Additionally, a significant association was found between rs3212172 and apolipoprotein B levels (coefficient: -0.14 ± 0.05; P<0.022). These polymorphisms are located in HNF4A P1 promoter or in intronic regions.

Conclusions

Our study demonstrates that HNF4α genetic variants are associated with the MetS and metabolic parameters in French Canadian children and adolescents. This study, the first exploring the relation between HNF4A genetic variants and MetS and metabolic variables in a pediatric cohort, suggests that HNF4α could represent an early marker for the risk of developing type 2 diabetes mellitus.

Citation: Marcil V, Amre D, Seidman EG, Boudreau F, Gendron FP, Ménard D, et al. (2015) Hepatocyte Nuclear Factor 4 Alpha Polymorphisms and the Metabolic Syndrome in French-Canadian Youth. PLoS ONE 10(2): e0117238. https://doi.org/10.1371/journal.pone.0117238

Academic Editor: Frances M. Sladek, Univeristy of California Riverside, UNITED STATES

Received: April 13, 2014; Accepted: December 19, 2014; Published: February 11, 2015

Copyright: © 2015 Marcil 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: Due to the sensitive nature of the data, a dataset cannot be made publicly available. However, these data (1999 Quebec Child and Adolescent Health and Social Survey) can be made available by the IRB to other researchers whom they deem appropriate, upon request. The authors may be contacted for the data at Research Centre, Ste-Jusine Hospital, 3175 Ste-Catherine Road # 5731A, Montreal, H3T 1C5 or emile.levy@recherche-ste-justine.qc.ca.

Funding: This study was supported by the Canadian Institutes of Health Research Team Grant (CTP-82942), the J.A. DeSève Research Chair in Nutrition (EL), the Canada Research Chair in Immune Mediated Gastrointestinal Disorders (ES), the Canadian Institutes of Health Research Fellowship Award and The Richard and Edith Strauss Postdoctoral Fellowships Award in Medicine, McGill University (VM). 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

The constantly rising prevalence of childhood obesity is becoming one of the most alarming public health problems worldwide. In parallel, there has been a significant increase in the number of children and adolescents with clinical signs of insulin resistance (IR) and prediabetes [1], which are likely to progress to type 2 diabetes mellitus (T2D) and early atherosclerosis [2]. Development of T2D in young people is of particular concern because complications are common and appear early in the disease [3,4]. Consequently, the identification of early markers and genetic risk factors for IR and T2D are becoming important tools for the management and the prevention of long-term cardiometabolic consequences.

The hepatocyte nuclear factor 4 alpha (HNF4α; HNF4A) is a member of the nuclear receptor superfamily of ligand-dependant transcription factors [5] and is mainly expressed in liver, intestine, pancreatic islets and kidney [5]. It influences lipid transport and metabolism [6–8] and is essential to hepatocyte differentiation and liver function [9]. Moreover, HNF4α maintains glucose homeostasis by regulating gene expression in pancreatic β cells [10–12] and gluconeogenesis in the liver [13,14].

The HNF4A gene is composed of thirteen exons and two promoters that drive the expression of many splice variants (isoforms) [15], for which 6 of the 9 splice variants appear to yield to full length transcripts [16,17]. The transcription of three of these isoforms is driven by an alternate promoter known as P2, which is located 45.6 kb upstream P1 promoter [18,19]. P2-driven transcripts have been described as the predominant splice variant in pancreatic β-cells [18–21], while the P1 promoter appears to be mainly active in liver cells [19,22,23].

Mutations in both the coding and regulatory regions of HNF4α have been associated with maturity-onset diabetes of the young (MODY)-1, a dominantly inherited, atypical form of T2D for which IR is absent [24,25]. Additionally, several whole-genome scan studies for T2D susceptibility loci have identified linkage on chromosome 20q12–13 in a region that encompasses the HNF4A locus [26–28]. The association between HNF4A and T2D has been extensively studied [29]. HNF4A genetic variants have been shown to contribute to the risk of T2D in Finnish [30] and Ashkinazi Jewish subjects [31]. These results have been partially replicated in the UK population [32], American Caucasians [33], Amish [34], Danish [35], and French Caucasians [36]. However, other studies did not find associations between HNF4A variants and T2D [27,37–39]. Besides, HNF4A polymorphisms were found associated with lipid traits, namely levels of high density lipoprotein (HDL) [40–42].

The present study aimed to investigate the relationship between HNF4A genetic variants and the presence of metabolic syndrome (MetS) in a pediatric French Canadian population, and to explore their association with metabolic parameters, for instance levels of blood glucose, insulin and lipids.

METHODS

Population study

The design and methods of the 1999 Quebec Child and Adolescent Health and Social Survey, a school-based survey of youth aged 9, 13, and 16 years, have previously been reported in detail [43]. On a total of 2,244 DNA samples available [44], we restricted the current analysis to 1,749 children and adolescents of French Canadian origin to reduce the confounding of genetic analyses by population stratification. The study was approved by the Institutional Review Board of Sainte-Justine Hospital and investigations were carried out in accordance with the principles of the Declaration of Helsinki. Written informed consent was obtained from parents/guardians, and written informed assent was obtained from study participants.

Anthropometry, blood pressure and lipids

Height, weight and blood pressure (BP) were measured according to standardized protocols [43]. Body mass index (BMI) was computed as weight in kilograms divided by height in meters squared. Values of percentile cut-off points used to identify subjects with metabolic risk factors were estimated from the study distributions. Cut-off points were age and sex specific, and BP cut-off points were also height specific, according to the National High Blood Pressure Education Program Working Group on High Blood Pressure in Children and Adolescents [45]. Subjects with BMI ≥ 85th percentile values were categorized as overweight/obese. High triglycerides (TG), insulin, and systolic BP were defined as values ≥ 75th percentile, and low HDL-cholesterol (HDL-C) was defined as values ≤ 25th percentile. Impaired fasting glucose (IFG) was defined as concentrations ≥ 6.1 and < 7.0 mmol/L. No study participant had fasting plasma glucose ≥ 7.0 mmol/L.

Currently, estimating the prevalence of childhood MetS continues to be challenging and controversial and there is no internationally accepted definition of childhood MetS. More than 40 definitions for childhood MetS have so far been proposed and most of them were based on adaptations of adult criteria [1]. Therefore, in the present work, we have based our definition on previously published work from our group, which was useful to assess the clustering of metabolic risk factors and to estimate the prevalence of MetS in a representative sample of youth in the province of Quebec in Canada [46,47]. MetS in our analyses required the presence of obesity and at least two other risk factors among high systolic or diastolic BP, high TG, low HDL-C and IFG [43,48]. In our MetS definition, general obesity was used instead of central obesity since waist circumference data were not available for this study.

Biochemical analyses

Blood samples were collected in the morning, after an overnight fast. Plasma total cholesterol (TC), HDL-C, TG and glucose concentrations were determined on a Beckman Synchron Cx7 instrument as previously described [43,44]. Apolipoprotein (apo) A-I and apo B were measured by nephelometry (Array Protein System; Beckman). The Friedewald equation was used to calculate low-density lipoprotein-cholesterol (LDL-C). Plasma insulin concentration was determined with the ultrasensitive Access® immunoassay system (Beckman Coulter, Inc.), which has no cross-reactivity with proinsulin or C-peptide. Plasma free fatty acids (FFA) concentrations were quantified by an enzymatic colorimetric method (Wako Chemicals).

Genotyping

Genomic DNA was prepared from white blood cells using the Puregene® DNA Isolation kit (Gentra Systems, Inc.). The genotyping was carried out as part of a previous study performed on this precise cohort [49]. 24 SNPs with a minor allele frequency > 5% were identified by sequencing the HNF4A gene in a French Canadian sample population [49]. The fragments were genotyped using the Luminex xMAP/Autoplex Analyser CS1000 system (Perkin Elmer, Waltham, MA). They were amplified in a single multiplex assay and hybridized to Luminex MicroPlex®—xTAG Microspheres [50] for genotyping using allele-specific primer extension. Amplification and reaction conditions are available upon request. Allele calls were assessed and compiled using the Automatic Luminex Genotyping software [51].

Statistical Analysis

Statistical analyses were performed with STATA v.10 statistical software (StataCorp LP). Potential genotyping errors were assessed using Chi-square tests, which evaluate the deviation of each SNP from Hardy-Weinberg equilibrium. Subjects were categorized according to their MetS status (yes/no). Between-group allele and genotype frequency distributions were compared by a Chi-square test. Allelic association for individual SNPs was carried out using logistic regression by fitting an additive model. To take the design effect into account, mixed models were used for all analyses of variance and regressions, with genetic markers and other independent variables treated as fixed effects and with clustering between subjects in the same school treated as a random effect. We used mixed logistic regression to examine the association between MetS status and HNF4A genotypes. We performed Fisher’s exact test to study the associations for the polymorphism without rare variant. We used mixed ANOVA and mixed linear regression to study the associations between genotypes and metabolic variables. Scheffe’s contrasts were used for posthoc pair comparisons. Insulin, TG, FFA and BMI values were loge transformed for statistical analyses to improve the normality of their distributions. Because we pooled age and sex groups, age- and sex-specific Z scores for BMI, insulin, glucose, TG, LDL-C, HDL-C, apo B and apo A-I were used in linear regression analyses. To standardize a value (i.e., compute its Z score), we subtracted the mean of the corresponding study distribution and divided by the SD. Haplotype analysis was carried out using HAPLOVIEW Software, version 3.11 [52] on the 9 SNPs for which the allelic association was significant or close to significant. Haplotype blocks created using the confidence interval feature. For each block, the haplotype association for each haplotype with MetS was examined by logistic regression. The association with the metabolic markers was evaluated using linear regression and P values were estimated.

RESULTS

Population Characteristics

The clinical and biochemical characteristics of participants are shown in Table 1. The prevalence of MetS was 11.03%. As expected, youth with MetS displayed significantly higher BMI, systolic and diastolic BP, TC, LDL-C, apo B, TG, FFA, insulin and glucose as well as lower levels of HDL-C and apo A-I than youth without MetS. No differences were detected in age and gender between the two groups. S2 Table indicates the cut-off values according to sex and age for BMI, TG, HDL-C, BP and insulin.

thumbnail
Download:
Table 1. Characteristics of study participants according to metabolic syndrome status.

https://doi.org/10.1371/journal.pone.0117238.t001

Effect of Polymorphisms on the Risk of Metabolic Syndrome

A total of 1,749 subjects were included for genotyping. Among the 24 SNPs genotyped, two deviated from Hardy-Weinberg equilibrium and were excluded from subsequent analyses (S1 Table). Since rs1884614 was found to be monomorphic, it was also excluded from association analyses. The remaining 21 SNPs were analyzed for association with MetS and results are presented in Table 2. Before correction for multiple testing, there was a significant difference in allele frequencies between MetS- and MetS+ subjects for seven polymorphisms. The minor alleles of two SNPs were associated with an increased risk of MetS (rs1800963, OR: 1.29, P<0.025; and rs3212183, OR: 1.38, P<0.003), while the minor alleles of five SNPs were associated with a reduced risk of MetS (rs6130608, OR: 0.73, P<0.019; rs2425637, OR: 0.74, P<0.007; rs3212172, OR: 0.68, P<0.030; rs736824, OR: 0.68; P<0.001; rs745975, OR: 0.63; P<0.003). For two other polymorphisms, the association between the minor allele and the MetS was close to significant (rs6031543, OR: 1.28, P<0.092; rs2425639, OR: 0.81, P<0.059). Fisher’s exact test was used to analyze the association between rs1884614 and MetS and did not reveal any significant association. However, after correction for multiple comparisons, only the association for rs736824 remained significant (P<0.021).

thumbnail
Download:
Table 2. Association between HNF4A polymorphisms and metabolic syndrome: odds ratio.

https://doi.org/10.1371/journal.pone.0117238.t002

Effect of Polymorphisms on Metabolic Variables

We studied the effect of HNF4A polymorphisms on mean LDL-C, HDL-C, apo A-I, apo B, FFA, TG, glucose and insulin levels. Because we did not detect heterogeneity of effect of HNF4A polymorphisms by sex or age, sex and age groups were pooled in subsequent analyses. Regression coefficients were calculated for the interactions between HNF4A genotypes and Z-score for blood glucose, insulin, TG, HDL-C, LDL-C, apo A-I and apo B after correction for age, sex and BMI. After correction for multiple testing, one SNP (rs3212172) had minor allele associated with decreased levels of apo B (coefficient: -0.14; P<0.001) (Table 3). A negative coefficient suggests decreasing value of the marker for every additional copy of the haplotype. Concomitantly, this allele was also associated with reduced TC (coefficient: -0.09; P<0.008) and LDL-C (coefficient: -0.13; P<0.008) although these associations did not remain significant after correction for multiple testing.

thumbnail
Download:
Table 3. Association between rs3212172 and metabolic variables.

https://doi.org/10.1371/journal.pone.0117238.t003

Haplotype Analyses

We then performed linkage disequilibrium (LD) analysis on the 9 SNPs for which the allelic association was significant or close to significant (Fig. 1). This analysis revealed that the SNPs were distributed within two major haplotype blocks: a first block including two SNPs in the region between both promoters (rs6130608, rs2425637) and a second block of three intronic SNPs (rs736824, rs745975, rs3212183). Table 4 shows the frequencies of the identified haplotypes. Haplotype analyses were performed on the SNPs within each block of LD (Table 4). Two haplotypes in the first block (TT and CG) and in the second block (TCC and CTT) were significantly associated with MetS after adjustment for age, sex, BMI, alcohol and cigarette consumption. The association for only one haplotype in each block remained significant after correction for multiple testing (TT, P<0.032; CTT, P<0.025). On the other hand, these haplotypes were not associated with significant variations in metabolic parameters.

thumbnail
Download:
Fig 1. Two major haplotype blocks found in the HNF4A gene.

Linkage disequilibrium plot in the HNF4A region is displayed. Haplotype analysis was carried out on the 9 SNPs for which the single SNP allelic association was significant or close to significant using HAPLOVIEW Software version 3.11.

https://doi.org/10.1371/journal.pone.0117238.g001

thumbnail
Download:
Table 4. Association between HNF4α haplotypes and metabolic syndrome.

https://doi.org/10.1371/journal.pone.0117238.t004

Study Power

We post-priori estimated the power of the study to detect important associations for the main outcome variable “MetS” that had a frequency of 11.03% in the sample. With the noted range of allele frequencies (15% to 50%), the sample of n = 1,749 participants provided sufficient power (≥ 80%) to detect odds ratio (OR) ≥ 1.55 (or ≤ 0.65) after correcting for multiple comparisons (alpha set at 0.002 for correcting for ~25 SNP associations). For lower OR (in the range 1.2 to 1.5), the study did not have adequate power. For OR of 1.3, the study power ranged from 11% to 28%, while for OR of 1.4 it ranged from 26% to 53% and for OR of 1.5 it ranged from 46% to 77% for an allele frequency range between 15% and 45%. As for the metabolic variables, based on the distribution of the mean (SD) of the levels in the population, estimated β-coefficient and correcting for multiple comparisons (alpha = 0.002), power of the study ranged from 5% to 100% depending on the metabolite and the allele frequency of the SNP.

DISCUSSION

In the present study, we aimed to evaluate the association between HNF4A genetic variants and MetS in French Canadian children and adolescents. Our analyses revealed that, after correction for multiple testing, one SNP (rs736824) and two haplotypes (P1 promoter haplotype rs6130608-rs2425637 and intronic haplotype rs736824-rs745975-rs3212183) were significantly associated with the risk of MetS (Fig. 2). Additionally, another significant association was found between rs3212172 and apo B levels (Fig. 2). To our knowledge, this is the first study exploring the relation between HNF4A genetic variants, MetS and metabolic variables in a pediatric population. Single SNP analysis revealed that the presence of the minor allele C of the intronic SNP rs736824 (intron 1A/1B-2) was associated with a 0.68 fold reduced risk of MetS. While this SNP was not found associated with T2D in American Caucasians [33] or with the conversion to T2D in the STOP-NIDDM trial [53], it was independently associated with fasting glucose levels in North Indians of Indo-European control subjects [54]. Moreover, an intronic haplotype (rs736824-rs745975-rs3212183) containing the rs736824 C allele was also protective for MetS. Accordingly, rs3212183 was modestly associated with T2D (OR: 1.34) in Pima Indians [55] and the protective effect of the T allele was confirmed in a meta-analysis (OR: 0.843) carried out on 4 studies [56]. Also, a haplotype containing the polymorphism rs745975 (rs745975-s2425640) has been associated with TG and glucose levels in Mexicans [57] but was never found independently associated with T2D or metabolic parameters in the literature.

thumbnail
Download:
Fig 2. Schematic illustration of the HNF4A SNPs found associated with metabolic syndrome or metabolic parameters.

Relative position of SNPs within the HNF4A locus. aSNP associated with the metabolic syndrome; bhaplotype associated with metabolic syndrome; cSNP associated with apo B levels.

https://doi.org/10.1371/journal.pone.0117238.g002

Among the SNPs identified in our study, rs2425637, which is part of the P1 promoter haplotype associated with risk of MetS, has been the most reported in the literature. It was found associated with T2D in Finnish, Ashkenazi and French Caucasian populations [58,59], but not with conversion to T2D in the STOP-NIDDM trial [53]. In a meta-analysis, a haplotype containing rs2425637 was not significantly associated with T2D, except for a marginal effect in Scandinavians [56].

The minor allele G of the P1 promoter SNP rs3212172 showed overall cardioprotective effects since it was linked to decreased risk of MetS and lower levels of TC, LDL-C and apo B, with only the latest remaining significant after correction for multiple testing. To our knowledge, no association has previously reported in the literature for that polymorphism. Since HNF4α is known to regulate apo B gene expression [60,61], the functional impact of this particular SNP would be particularly interesting to explore. In fact, it has been demonstrated that MODYI patients have lower levels of very-low density lipoprotein-C and LDL-C than controls, which was attributed, at least in part, to a reduced transactivation activity of HNF4α for the acyl-coenzyme A: cholesterol acyltransferases 2 promoter [62]. Although the association between TG levels and MODY1 (HNF4A Q268X mutation) has previously been demonstrated [63], the correlation does not hold true when assessed with HNF4A common SNPs in our French Canadian population.

Interestingly, the HNF4A genetic variants identified in our study are located in P1 promoter and intronic regions; none of the P2 promoter SNPs was found associated with MetS or with metabolic parameters. Conversely, in the literature, attention has been mostly paid to P2 promoter SNPs. Evidence for association between SNPs in the beta-cell P2 promoter region of HNF4A was recognized in Finnish [30] and Ashkenazi [31,64] populations, with data suggesting that HNF4A P2 SNPs (or variants in strong linkage disequilibrium with them) contribute to the linkage signal on chromosome 20q [30,31]. Yet, association with HNF4A promoter SNPs has been replicated in some [65] but not all [66,67] populations tested. Hence, there was evidence for association with SNPs or haplotypes in the HNF4A region other than the P2 SNPs [68,69]. Moreover, a meta-analysis showed that P2 promoter SNPs were associated with T2D only in Scandinavians [56]. Recently, P2 promoter SNPs have been associated with insulin resistance and BMI in adult subjects [70], but the study was performed on a small sample size of 160 subjects. Data obtained in our study support the lack of association between P2 promoter variants and metabolic parameters in children.

Functional studies have initially reported that the P2 promoter drives transcription in β-cells and that the P1 promoter drives transcription in extra-pancreatic cells, such as liver cells [19,21]. However, studies have previously linked P1 promoter polymorphisms to T2D and, along with our study, suggest important contribution for P1-driven genes in insulin resistance, glucose tolerance and MetS development.

The fact that different SNPs in the HNF4A region are associated with diabetes in different populations suggests that none of these alleles themselves are causative functional variants but that they may be in linkage disequilibrium with a nearby functioning allele. Alternatively, some of these alleles may be causative, but allelic heterogeneity across populations may make their identification difficult. Moreover, it has been suggested that HNF4α can be constitutively bound to fatty acids [71] and it can bind to linoleic acid in a reversible fashion [72]. HNF4α was revealed as important for hepatic response to changes in nutritional status [73]. Hence, diverge results in association studies might be explained by the dietary influence that might play a role and dilute the genetic impact to a variable extent depending on the study population [29].

As mentioned before, the definition of overweight/obesity in children (BMI ≥ 85th percentile) used herein was based on our previous publications performed on a representative Canadian population [46,47]. This definition also corresponds to the one proposed by the Center for Disease Control and Prevention (CDC). According to their charts, the CDC defines overweight as a BMI above the 85th percentile of the reference population and obesity as a BMI above the 95th percentile [74]. Moreover, the World Health Organization (WHO) system defines overweight as a BMI > 1 SD and obesity as a BMI > 2 SD from the mean of the WHO reference population [75]. The WHO reference BMI-for-age curves at 19 years closely coincides with adult overweight (BMI = 25.0 kg/m2) at +1 SD and adult obesity (BMI = 30.0 kg/m2) at +2 SD. It was found that these obesity and overweight cut-off values identified children with higher metabolic and vascular risk [76]. According to our reference population, the 85th percentile corresponds in a BMI of 20.04, 23.85 and 26.45 kg/m2 for boys of 9, 13 and 16 years old, respectively, and of 20.51, 26.01 and 26.25 kg/m2 for girls of 9, 13 and 16 years old, respectively (S2 Table). According to the WHO charts, these values correspond in boys to +2 SD for the 9-year-old age group and +1.5 SD for the 13- and the 16-year-old groups, and in girls to +2 SD for the 9- and 13-year-old age groups and +1.5 SD for the 16-year-old group.

Despite the continued use of the MetS concept, a number of ongoing issues surround the MetS notion and its application to children. Children, unlike the adults for whom the MetS concept was originally developed, reside in vastly different stages of growth, development and pubertal status, thereby questioning whether such variability can be accommodated by a single MetS definition [77]. Another difficulty is the fact that reference values for some MetS components, such as waist circumference, exist for only some populations and that there remains disagreement over how to measure waist circumference in children [77]. Also, the lack of reference values in some populations for blood pressure or HDL-C level render cross-cultural comparisons problematic [78]. On the other hand, the use of dichotomous (normal vs. abnormal) variable categories is also debated. Strict cut-off points are difficult to apply in the pediatric population given the well-known fluctuations associated with growth and puberty [77]. For these reasons and based on previous studies from our group [46,47] we have decided to identify a sub-group of children in our population who are more at risk of cardiovascular complications and we have identified that group as MetS+. Importantly, the definition used in this manuscript identified 11.03% of the population with MetS, which corresponds to what we have found in a previous investigation [46]. As a matter of fact, several definitions of the MetS have been compared using this specific population and the overall prevalence of MetS was ranging between 11.5% and 14.0% according to the stringency of the definition [46].

This study presents a certain number of limitations. First, because waist circumference values were not available for this study, the International Diabetes Federation diagnostic criteria for MetS in children and adolescent could not have been used. Also, data available in this study did not make possible the analysis between HNF4A polymorphisms and previously reported lipid abnormalities in MODY1 such as apo A-II, apo C-III and lipoprotein (a)

In conclusion, this study, the first exploring the relation between HNF4A genetic variants, MetS and metabolic variables in a pediatric cohort, supports the hypothesis that HNF4A P1 promoter and intronic polymorphisms play a role in predisposing to T2D and could represent an early marker for the risk of developing the disease.

Supporting Information

S1 Table. Genotyped SNPs and Hardy-Weinberg equilibrium test. Among the 24 SNPs genotyped, two deviated from Hardy-Weinberg equilibrium and were excluded from subsequent analyses.

HWE, Hardy-Weinberg equilibrium. *SNPs with a significant HWE test were excluded for further analyses. **SNP with no rare homozygote was excluded for association analyses.

https://doi.org/10.1371/journal.pone.0117238.s001

(PDF)

S2 Table. Cut points used to define risk factors by age and sex. The cut points correspond to the 85th percentile of the study population for BMI, the 75th percentile for triglycerides, insulin, systolic BP and diastolic BP and the 25th percentile for HDL-cholesterol.

BMI, body mass index; BP, blood pressure.

https://doi.org/10.1371/journal.pone.0117238.s002

(PDF)

Acknowledgments

The authors thank Mrs Schohraya Spahis for her technical assistance and acknowledge the assistance of Dr Marie Lambert (deceased) in allowing us to access the QCAHSS cohort.

Author Contributions

Conceived and designed the experiments: VM EL. Performed the experiments: DA ES FB FPG DM JFB DS ML. Analyzed the data: EL DA. Contributed reagents/materials/analysis tools: VM EL. Wrote the paper: VM EL.

REFERENCES

  1. 1. Cook S, Auinger P, Li C, Ford ES (2008) Metabolic syndrome rates in United States adolescents, from the National Health and Nutrition Examination Survey, 1999–2002. J Pediatr 152: 165–170. pmid:18206683
  2. 2. Weiss R, Taksali SE, Tamborlane WV, Burgert TS, Savoye M, et al. (2005) Predictors of changes in glucose tolerance status in obese youth. Diabetes Care 28: 902–909. pmid:15793193
  3. 3. Eppens MC, Craig ME, Jones TW, Silink M, Ong S, et al. (2006) Type 2 diabetes in youth from the Western Pacific region: glycaemic control, diabetes care and complications. Curr Med Res Opin 22: 1013–1020. pmid:16709323
  4. 4. Ettinger LM, Freeman K, DiMartino-Nardi JR, Flynn JT (2005) Microalbuminuria and abnormal ambulatory blood pressure in adolescents with type 2 diabetes mellitus. J Pediatr 147: 67–73. pmid:16027698
  5. 5. Sladek FM, Zhong WM, Lai E, Darnell JE Jr (1990) Liver-enriched transcription factor HNF-4 is a novel member of the steroid hormone receptor superfamily. Genes Dev 4: 2353–2365. pmid:2279702
  6. 6. Marcil V, Delvin E, Sane AT, Tremblay A, Levy E (2006) Oxidative stress influences cholesterol efflux in THP-1 macrophages: role of ATP-binding cassette A1 and nuclear factors. Cardiovasc Res 72: 473–482. pmid:17070507
  7. 7. Iwayanagi Y, Takada T, Suzuki H (2008) HNF4alpha is a crucial modulator of the cholesterol-dependent regulation of NPC1L1. Pharm Res 25: 1134–1141. pmid:18080173
  8. 8. Leng S, Lu S, Yao Y, Kan Z, Morris GS, et al. (2007) Hepatocyte nuclear factor-4 mediates apolipoprotein A-IV transcriptional regulation by fatty acid in newborn swine enterocytes. Am J Physiol Gastrointest Liver Physiol 293: G475–G483. pmid:17556588
  9. 9. Parviz F, Matullo C, Garrison WD, Savatski L, Adamson JW, et al. (2003) Hepatocyte nuclear factor 4alpha controls the development of a hepatic epithelium and liver morphogenesis. Nat Genet 34: 292–296. pmid:12808453
  10. 10. Miura A, Yamagata K, Kakei M, Hatakeyama H, Takahashi N, et al. (2006) Hepatocyte nuclear factor-4alpha is essential for glucose-stimulated insulin secretion by pancreatic beta-cells. J Biol Chem 281: 5246–5257. pmid:16377800
  11. 11. Stoffel M, Duncan SA (1997) The maturity-onset diabetes of the young (MODY1) transcription factor HNF4alpha regulates expression of genes required for glucose transport and metabolism. Proc Natl Acad Sci U S A 94: 13209–13214. pmid:9371825
  12. 12. Wang H, Maechler P, Antinozzi PA, Hagenfeldt KA, Wollheim CB (2000) Hepatocyte nuclear factor 4alpha regulates the expression of pancreatic beta-cell genes implicated in glucose metabolism and nutrient-induced insulin secretion. J Biol Chem 275: 35953–35959. pmid:10967120
  13. 13. Rhee J, Inoue Y, Yoon JC, Puigserver P, Fan M, et al. (2003) Regulation of hepatic fasting response by PPARgamma coactivator-1alpha (PGC-1): requirement for hepatocyte nuclear factor 4alpha in gluconeogenesis. Proc Natl Acad Sci U S A 100: 4012–4017. pmid:12651943
  14. 14. Hall RK, Sladek FM, Granner DK (1995) The orphan receptors COUP-TF and HNF-4 serve as accessory factors required for induction of phosphoenolpyruvate carboxykinase gene transcription by glucocorticoids. Proc Natl Acad Sci U S A 92: 412–416. pmid:7831301
  15. 15. Jiang G, Sladek FM (1997) The DNA binding domain of hepatocyte nuclear factor 4 mediates cooperative, specific binding to DNA and heterodimerization with the retinoid X receptor alpha. J Biol Chem 272: 1218–1225. pmid:8995424
  16. 16. Huang W, Wang P, Liu Z, Zhang L (2009) Identifying disease associations via genome-wide association studies. BMC Bioinformatics 10 Suppl 1: S68. pmid:19208172
  17. 17. Harries LW, Locke JM, Shields B, Hanley NA, Hanley KP, et al. (2008) The diabetic phenotype in HNF4A mutation carriers is moderated by the expression of HNF4A isoforms from the P1 promoter during fetal development. Diabetes 57: 1745–1752. pmid:18356407
  18. 18. Thomas H, Jaschkowitz K, Bulman M, Frayling TM, Mitchell SM, et al. (2001) A distant upstream promoter of the HNF-4alpha gene connects the transcription factors involved in maturity-onset diabetes of the young. Hum Mol Genet 10: 2089–2097. pmid:11590126
  19. 19. Boj SF, Parrizas M, Maestro MA, Ferrer J (2001) A transcription factor regulatory circuit in differentiated pancreatic cells. Proc Natl Acad Sci U S A 98: 14481–14486. pmid:11717395
  20. 20. Eeckhoute J, Moerman E, Bouckenooghe T, Lukoviak B, Pattou F, et al. (2003) Hepatocyte nuclear factor 4 alpha isoforms originated from the P1 promoter are expressed in human pancreatic beta-cells and exhibit stronger transcriptional potentials than P2 promoter-driven isoforms. Endocrinology 144: 1686–1694. pmid:12697672
  21. 21. Hansen SK, Parrizas M, Jensen ML, Pruhova S, Ek J, et al. (2002) Genetic evidence that HNF-1alpha-dependent transcriptional control of HNF-4alpha is essential for human pancreatic beta cell function. J Clin Invest 110: 827–833. pmid:12235114
  22. 22. Briancon N, Weiss MC (2006) In vivo role of the HNF4alpha AF-1 activation domain revealed by exon swapping. EMBO J 25: 1253–1262. pmid:16498401
  23. 23. Nakhei H, Lingott A, Lemm I, Ryffel GU (1998) An alternative splice variant of the tissue specific transcription factor HNF4alpha predominates in undifferentiated murine cell types. Nucleic Acids Res 26: 497–504. pmid:9421506
  24. 24. Ryffel GU (2001) Mutations in the human genes encoding the transcription factors of the hepatocyte nuclear factor (HNF)1 and HNF4 families: functional and pathological consequences. J Mol Endocrinol 27: 11–29. pmid:11463573
  25. 25. Yamagata K, Furuta H, Oda N, Kaisaki PJ, Menzel S, et al. (1996) Mutations in the hepatocyte nuclear factor-4alpha gene in maturity-onset diabetes of the young (MODY1). Nature 384: 458–460. pmid:8945471
  26. 26. Bowden DW, Sale M, Howard TD, Qadri A, Spray BJ, et al. (1997) Linkage of genetic markers on human chromosomes 20 and 12 to NIDDM in Caucasian sib pairs with a history of diabetic nephropathy. Diabetes 46: 882–886. pmid:9133559
  27. 27. Ghosh S, Watanabe RM, Hauser ER, Valle T, Magnuson VL, et al. (1999) Type 2 diabetes: evidence for linkage on chromosome 20 in 716 Finnish affected sib pairs. Proc Natl Acad Sci U S A 96: 2198–2203. pmid:10051618
  28. 28. Permutt MA, Wasson JC, Suarez BK, Lin J, Thomas J, et al. (2001) A genome scan for type 2 diabetes susceptibility loci in a genetically isolated population. Diabetes 50: 681–685. pmid:11246891
  29. 29. Love-Gregory L, Permutt MA (2007) HNF4A genetic variants: role in diabetes. Curr Opin Clin Nutr Metab Care 10: 397–402. pmid:17563455
  30. 30. Silander K, Mohlke KL, Scott LJ, Peck EC, Hollstein P, et al. (2004) Genetic variation near the hepatocyte nuclear factor-4 alpha gene predicts susceptibility to type 2 diabetes. Diabetes 53: 1141–1149. pmid:15047633
  31. 31. Love-Gregory LD, Wasson J, Ma J, Jin CH, Glaser B, et al. (2004) A common polymorphism in the upstream promoter region of the hepatocyte nuclear factor-4 alpha gene on chromosome 20q is associated with type 2 diabetes and appears to contribute to the evidence for linkage in an ashkenazi jewish population. Diabetes 53: 1134–1140. pmid:15047632
  32. 32. Weedon MN, Owen KR, Shields B, Hitman G, Walker M, et al. (2004) Common variants of the hepatocyte nuclear factor-4alpha P2 promoter are associated with type 2 diabetes in the U.K. population. Diabetes 53: 3002–3006. pmid:15504983
  33. 33. Bagwell AM, Bento JL, Mychaleckyj JC, Freedman BI, Langefeld CD, et al. (2005) Genetic analysis of HNF4A polymorphisms in Caucasian-American type 2 diabetes. Diabetes 54: 1185–1190. pmid:15793260
  34. 34. Damcott CM, Hoppman N, Ott SH, Reinhart LJ, Wang J, et al. (2004) Polymorphisms in both promoters of hepatocyte nuclear factor 4-alpha are associated with type 2 diabetes in the Amish. Diabetes 53: 3337–3341. pmid:15561969
  35. 35. Hansen SK, Rose CS, Glumer C, Drivsholm T, Borch-Johnsen K, et al. (2005) Variation near the hepatocyte nuclear factor (HNF)-4alpha gene associates with type 2 diabetes in the Danish population. Diabetologia 48: 452–458. pmid:15735891
  36. 36. Vaxillaire M, Dina C, Lobbens S, Dechaume A, Vasseur-Delannoy V, et al. (2005) Effect of common polymorphisms in the HNF4alpha promoter on susceptibility to type 2 diabetes in the French Caucasian population. Diabetologia 48: 440–444. pmid:15735892
  37. 37. Malecki MT, Antonellis A, Casey P, Ji L, Wantman M, et al. (1998) Exclusion of the hepatocyte nuclear factor 4alpha as a candidate gene for late-onset NIDDM linked with chromosome 20q. Diabetes 47: 970–972. pmid:9604877
  38. 38. Manning AK, Hivert MF, Scott RA, Grimsby JL, Bouatia-Naji N, et al. (2012) A genome-wide approach accounting for body mass index identifies genetic variants influencing fasting glycemic traits and insulin resistance. Nat Genet 44: 659–669. pmid:22581228
  39. 39. Winckler W, Graham RR, de Bakker PI, Sun M, Almgren P, et al. (2005) Association testing of variants in the hepatocyte nuclear factor 4alpha gene with risk of type 2 diabetes in 7,883 people. Diabetes 54: 886–892. pmid:15734869
  40. 40. Kathiresan S, Willer CJ, Peloso GM, Demissie S, Musunuru K, et al. (2009) Common variants at 30 loci contribute to polygenic dyslipidemia. Nat Genet 41: 56–65. pmid:19060906
  41. 41. Teslovich TM, Musunuru K, Smith AV, Edmondson AC, Stylianou IM, et al. (2010) Biological, clinical and population relevance of 95 loci for blood lipids. Nature 466: 707–713. pmid:20686565
  42. 42. Lu Y, Dolle ME, Imholz S, van’t SR, Verschuren WM, et al. (2008) Multiple genetic variants along candidate pathways influence plasma high-density lipoprotein cholesterol concentrations. J Lipid Res 49: 2582–2589. pmid:18660489
  43. 43. Paradis G, Lambert M, O’Loughlin J, Lavallee C, Aubin J, et al. (2003) The Quebec Child and Adolescent Health and Social Survey: design and methods of a cardiovascular risk factor survey for youth. Can J Cardiol 19: 523–531. pmid:12717488
  44. 44. Allard P, Delvin EE, Paradis G, Hanley JA, O’Loughlin J, et al. (2003) Distribution of fasting plasma insulin, free fatty acids, and glucose concentrations and of homeostasis model assessment of insulin resistance in a representative sample of Quebec children and adolescents. Clin Chem 49: 644–649. pmid:12651818
  45. 45. (2004) The fourth report on the diagnosis, evaluation, and treatment of high blood pressure in children and adolescents. Pediatrics 114: 555–576. pmid:15286277
  46. 46. Lambert M, Paradis G, O’Loughlin J, Delvin EE, Hanley JA, et al. (2004) Insulin resistance syndrome in a representative sample of children and adolescents from Quebec, Canada. Int J Obes Relat Metab Disord 28: 833–841. pmid:15170466
  47. 47. Stan S, Levy E, Delvin EE, Hanley JA, Lamarche B, et al. (2005) Distribution of LDL particle size in a population-based sample of children and adolescents and relationship with other cardiovascular risk factors. Clin Chem 51: 1192–1200. pmid:15890892
  48. 48. Alberti KG, Zimmet P, Shaw J (2006) Metabolic syndrome—a new world-wide definition. A Consensus Statement from the International Diabetes Federation. Diabet Med 23: 469–480. pmid:16681555
  49. 49. Marcil V, Sinnett D, Seidman E, Boudreau F, Gendron FP, et al. (2012) Association between genetic variants in the HNF4A gene and childhood-onset Crohn’s disease. Genes Immun 13: 556–565. pmid:22914433
  50. 50. Koo SH, Ong TC, Chong KT, Lee CG, Chew FT, et al. (2007) Multiplexed genotyping of ABC transporter polymorphisms with the Bioplex suspension array. Biol Proced Online 9: 27–42. pmid:18213362
  51. 51. Bourgey M, Lariviere M, Richer C, Sinnett D (2011) ALG: Automated Genotype Calling of Luminex Assays. PLoS One 6: e19368. pmid:21573116
  52. 52. Barrett JC, Fry B, Maller J, Daly MJ (2005) Haploview: analysis and visualization of LD and haplotype maps. Bioinformatics 21: 263–265. pmid:15297300
  53. 53. Andrulionyte L, Laukkanen O, Chiasson JL, Laakso M (2006) Single nucleotide polymorphisms of the HNF4alpha gene are associated with the conversion to type 2 diabetes mellitus: the STOP-NIDDM trial. J Mol Med 84: 701–708. pmid:16838170
  54. 54. Chavali S, Mahajan A, Tabassum R, Dwivedi OP, Chauhan G, et al. (2011) Association of variants in genes involved in pancreatic beta-cell development and function with type 2 diabetes in North Indians. J Hum Genet 56: 695–700. pmid:21814221
  55. 55. Muller YL, Infante AM, Hanson RL, Love-Gregory L, Knowler W, et al. (2005) Variants in hepatocyte nuclear factor 4alpha are modestly associated with type 2 diabetes in Pima Indians. Diabetes 54: 3035–3039. pmid:16186411
  56. 56. Sookoian S, Gemma C, Pirola CJ (2010) Influence of hepatocyte nuclear factor 4alpha (HNF4alpha) gene variants on the risk of type 2 diabetes: a meta-analysis in 49,577 individuals. Mol Genet Metab 99: 80–89. pmid:19748811
  57. 57. Weissglas-Volkov D, Huertas-Vazquez A, Suviolahti E, Lee J, Plaisier C, et al. (2006) Common hepatic nuclear factor-4alpha variants are associated with high serum lipid levels and the metabolic syndrome. Diabetes 55: 1970–1977. pmid:16804065
  58. 58. Bonnycastle LL, Willer CJ, Conneely KN, Jackson AU, Burrill CP, et al. (2006) Common variants in maturity-onset diabetes of the young genes contribute to risk of type 2 diabetes in Finns. Diabetes 55: 2534–2540. pmid:16936201
  59. 59. Sladek R, Rocheleau G, Rung J, Dina C, Shen L, et al. (2007) A genome-wide association study identifies novel risk loci for type 2 diabetes. Nature 445: 881–885. pmid:17293876
  60. 60. Ladias JA, Hadzopoulou-Cladaras M, Kardassis D, Cardot P, Cheng J, et al. (1992) Transcriptional regulation of human apolipoprotein genes ApoB, ApoCIII, and ApoAII by members of the steroid hormone receptor superfamily HNF-4, ARP-1, EAR-2, and EAR-3. J Biol Chem 267: 15849–15860. pmid:1639815
  61. 61. Metzger S, Halaas JL, Breslow JL, Sladek FM (1993) Orphan receptor HNF-4 and bZip protein C/EBP alpha bind to overlapping regions of the apolipoprotein B gene promoter and synergistically activate transcription. J Biol Chem 268: 16831–16838. pmid:8344962
  62. 62. Pramfalk C, Karlsson E, Groop L, Rudel LL, Angelin B, et al. (2009) Control of ACAT2 liver expression by HNF4{alpha}: lesson from MODY1 patients. Arterioscler Thromb Vasc Biol 29: 1235–1241. pmid:19478207
  63. 63. Shih DQ, Dansky HM, Fleisher M, Assmann G, Fajans SS, et al. (2000) Genotype/phenotype relationships in HNF-4alpha/MODY1: haploinsufficiency is associated with reduced apolipoprotein (AII), apolipoprotein (CIII), lipoprotein(a), and triglyceride levels. Diabetes 49: 832–837. pmid:10905494
  64. 64. Barroso I, Luan J, Wheeler E, Whittaker P, Wasson J, et al. (2008) Population-specific risk of type 2 diabetes conferred by HNF4A P2 promoter variants: a lesson for replication studies. Diabetes 57: 3161–3165. pmid:18728231
  65. 65. Lehman DM, Richardson DK, Jenkinson CP, Hunt KJ, Dyer TD, et al. (2007) P2 promoter variants of the hepatocyte nuclear factor 4alpha gene are associated with type 2 diabetes in Mexican Americans. Diabetes 56: 513–517. pmid:17259399
  66. 66. Tanahashi T, Osabe D, Nomura K, Shinohara S, Kato H, et al. (2006) Association study on chromosome 20q11.21–13.13 locus and its contribution to type 2 diabetes susceptibility in Japanese. Hum Genet 120: 527–542. pmid:16955255
  67. 67. Wanic K, Malecki MT, Wolkow PP, Klupa T, Skupien J, et al. (2006) Polymorphisms in the gene encoding hepatocyte nuclear factor-4alpha and susceptibility to type 2 diabetes in a Polish population. Diabetes Metab 32: 86–88. pmid:16523192
  68. 68. Ek J, Hansen SP, Lajer M, Nicot C, Boesgaard TW, et al. (2006) A novel-192c/g mutation in the proximal P2 promoter of the hepatocyte nuclear factor-4 alpha gene (HNF4A) associates with late-onset diabetes. Diabetes 55: 1869–1873. pmid:16731855
  69. 69. Hara K, Horikoshi M, Kitazato H, Ito C, Noda M, et al. (2006) Hepatocyte nuclear factor-4alpha P2 promoter haplotypes are associated with type 2 diabetes in the Japanese population. Diabetes 55: 1260–1264. pmid:16644680
  70. 70. Saif-Ali R, Harun R, Al-Jassabi S, Wan Ngah WZ (2011) Hepatocyte nuclear factor 4 alpha P2 promoter variants associate with insulin resistance. Acta Biochim Pol 58: 179–186. pmid:21633728
  71. 71. Dhe-Paganon S, Duda K, Iwamoto M, Chi YI, Shoelson SE (2002) Crystal structure of the HNF4 alpha ligand binding domain in complex with endogenous fatty acid ligand. J Biol Chem 277: 37973–37976. pmid:12193589
  72. 72. Yuan X, Ta TC, Lin M, Evans JR, Dong Y, et al. (2009) Identification of an endogenous ligand bound to a native orphan nuclear receptor. PLoS One 4: e5609. pmid:19440305
  73. 73. Adamson AW, Suchankova G, Rufo C, Nakamura MT, Teran-Garcia M, et al. (2006) Hepatocyte nuclear factor-4alpha contributes to carbohydrate-induced transcriptional activation of hepatic fatty acid synthase. Biochem J 399: 285–295. pmid:16800817
  74. 74. Krebs NF, Himes JH, Jacobson D, Nicklas TA, Guilday P, et al. (2007) Assessment of child and adolescent overweight and obesity. Pediatrics 120 Suppl 4: S193–S228. pmid:18055652
  75. 75. Flegal KM, Ogden CL (2011) Childhood obesity: are we all speaking the same language?. Adv Nutr 2: 159S–166S. pmid:22332047
  76. 76. de OM, Onyango A, Borghi E, Siyam A, Blossner M, et al. (2012) Worldwide implementation of the WHO Child Growth Standards. Public Health Nutr 15: 1603–1610. pmid:22717390
  77. 77. Marcovecchio ML, Chiarelli F (2013) Metabolic syndrome in youth: chimera or useful concept?. Curr Diab Rep 13: 56–62. pmid:23054749
  78. 78. D’Adamo E, Santoro N, Caprio S (2013) Metabolic syndrome in pediatrics: old concepts revised, new concepts discussed. Curr Probl Pediatr Adolesc Health Care 43: 114–123. pmid:23582593