Skip Navigation


Human Molecular Genetics Advance Access originally published online on January 28, 2004
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
13/7/715    most recent
ddh070v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (19)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Rodríguez, S.
Right arrow Articles by Day, I. N.M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Rodríguez, S.
Right arrow Articles by Day, I. N.M.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Human Molecular Genetics, 2004, Vol. 13, No. 7 715-725
DOI: 10.1093/hmg/ddh070

Haplotypic analyses of the IGF2-INS-TH gene cluster in relation to cardiovascular risk traits

Santiago Rodríguez1,*,{dagger}, Tom R. Gaunt1,{dagger}, Sandra D. O'Dell1,{ddagger}, Xiao-he Chen1, Dongfeng Gu1,§, Emma Hawe2, George J. Miller3, Stephen E. Humphries2 and Ian N.M. Day1

1Human Genetics Division, University of Southampton, School of Medicine, Duthie Building (MP 808), Southampton General Hospital, Tremona Road, Southampton SO16 6YD, UK, 2Centre for the Genetics of Cardiovascular Disease, British Heart Foundation Laboratories, The Rayne Building, Royal Free and University College London Medical School, University Street, London WC1E 6JJ, UK and 3Medical Research Council Cardiovascular Research Group, Wolfson Institute of Preventive Medicine, Barts and the London Queen Mary's School of Medicine and Dentistry, Charterhouse Square, London EC1M 6BQ, UK

Received November 20, 2003; Accepted January 23, 2004


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 NOTE ADDED IN PROOF
 REFERENCES
 
The IGF2–INS–TH genomic region has been implicated in various common disorders including the metabolic syndrome, type 2 diabetes and coronary heart disease (CHD). Here we present detailed haplotype analysis of 2743 males 51–62 years old in relation to body weight and composition, blood pressure (BP) and plasma triglycerides (TG). Use of the total data set was complicated by the number of loci typed, missing data, multi-allelic markers and continuous trait phenotypes. Different algorithms and subsets of the data were analysed using the programmes haplotype trend regression, haplo.score, evolutionary-based haplotype analysis package and Phase, in conjunction with SPSS. Ten haplotypes designated in frequency order *1(20.0%) to *10(3.4%) represented 89% of all haplotypes. Haplotype *5 protected against obesity. Haplotype *4 carriers exhibited elevated BP and fat mass, haplotype *6 was associated with raised plasma TG levels. Haplotype *8 also showed similar magnitude effects as *4. These cohort trait analyses and detailed haplotypic analyses enable integration with published case data. Haplotypes *4, *6 and *8 are the only INS VNTR class III-bearing haplotypes, although differing in flanking haplotype, whereas *5 displays unique features in all three genes (with significant commonality with type 1 diabetes-predisposition haplotypes). We propose that long repeat insertion in the insulin gene promoter (‘class III’), reported to result in low insulin production, predisposes to the metabolic syndrome features of elevated BP, fat mass or TG level, therefore appearing more frequently in type 2 diabetic, polycystic ovary syndrome and CHD cases. The functional element(s) of *5 for weight-lowering could reside in any of the three genes.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 NOTE ADDED IN PROOF
 REFERENCES
 
Adult obesity, human essential hypertension and hypertriglyceridemia are interrelated common disorders that have a significant impact on conditions including the metabolic syndrome, type 2 diabetes and coronary heart disease (1–3). Determination of body weight and composition, blood pressure (BP) and plasma triglycerides (TG) is multifactorial, with both genetic and environmental contributions. The genetic factors involved are likely to be multiple and interacting, with most single variants producing only a moderate effect (2,4). A large number of genes, markers and chromosomal regions have been associated or linked with human obesity phenotypes (5), human essential hypertension (2,6) and high plasma TG levels (7).

The IGF2–INS–TH region on human chromosome 11p15 represents a strong candidate polygene for cardiovascular risk. These genes respectively represent insulin-like growth factor II, a major fetal growth factor also expressed in adult life, insulin, the best known regulator of intermediary metabolism, and tyrosine hydroxylase, the rate-limiting enzyme in biogenic amine synthesis. Each has a number of potential pathways by which it may act upon cardiovascular risk traits, both at the cellular and systems levels. Several studies have also shown differential messenger RNA expression or transcription according to genotype. These studies are summarized in Table 1 and some are considered in more detail in the Discussion. Previous genetic epidemiological studies, mainly case studies, mostly of single polymorphic sites within these genes, have suggested that inter-individual variation of this region influences risk of CHD, hypertension, hypertriglyceridemia, glucose tolerance, early and late life diabetes mellitus, polycystic ovary syndrome (PCOS), longevity, growth in early and late life and body mass. This literature is summarized in Table 2. However, although linkage disequilibrium across this gene cluster has been recognized previously, no study has yet attempted to examine markers across the whole cluster in a systematic way in relation to a set of cardiovascular risk traits. In this study, we have undertaken an integrated haplotypic analysis in 2743 Caucasian males (51–62 years) comprising the UK Northwick Park Heart Study II, a UK-wide cohort study of cardiovascular risk traits.


View this table:
[in this window]
[in a new window]
  		Table 1. Summary of functional studies on the effect of variation at the IGF2–INS–TH region on cardiovascular risk traits
 

View this table:
[in this window]
[in a new window]
  		Table 2. Summary of genetic epidemiological studies on the association of polymorphisms at the IGF2–INS–TH region with cardiovascular risk traits
 

   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 NOTE ADDED IN PROOF
 REFERENCES
 
Tagging of the haplotypic information in the IGF2–INS–TH region
Table 3 shows the association of the five commonest IGF2 11-SNP haplotypes in relation to the TH01 and INS VNTR alleles. It can be seen that haplotypes that we have designated A, C, D and E are strongly associated with alleles 6, 9, 7 and 9.3, respectively. Consideration of TH01 and IGF2 ApaI allows a nearly complete tagging of all the information given by the combination of TH01 and the 11 IGF2 SNPs: haplotypes 6–1, 9–2, 7–1 and 9.3–2 tag haplotypes A, C, D and E respectively and haplotype 9.3–1 tags mainly haplotype B and haplotype A to a minor extent. The information given by TH01 and the 11 IGF2 SNPs can thus be reduced to a combination of the microsatellite and one single IGF2 SNP (ApaI). Addition of the INS HphI site permits distinction between classI/classIII alleles at the INS VNTR locus (8). IGF2 ApaI-INS HphI-TH01 haplotypes thus tag the information given by 14 polymorphisms of this region. This increased the number of individuals available with complete data (from 282 to 2061), reduced the uncertainties of indirect haplotype deduction and reduced the number of rare or artefactual haplotypes, the total number then being 21 compared with 203. The latter problem would have been a major issue for haplotype regression analyses (9). Consistent with approaches to reduce risk of false positive findings related to low-frequency haplotypes (10), we considered the 10 haplotypes with a frequency higher than 3%, representing 89% of all haplotypes (Table 4).


View this table:
[in this window]
[in a new window]
  		Table 3. Interallelic associations found between the five commonest IGF2 haplotypes and TH01 and INS VNTR alleles in NPHSII men. Numbers representing more than 5% of inferred haplotypes are indicated in bold. Also shown is the alphabetical code used for identifying each IGF2 haplotype. The 11 IGF2 SNPs are in the order 6815, 8173, 1156, AluI, 2482, 2722, 266, ApaI, 1926, 2207 and 3750
 

View this table:
[in this window]
[in a new window]
  		Table 4. IGF2–INS–TH haplotypes with a frequency higher than 3% observed in NPHSII by Phase. The commonest 11 IGF2 haplotypes (A–E) associated with each one of the 10 IGF2–INS–TH haplotypes considered are indicated (see Table 3 for this deduction). Also identified are the haplotypes with a PH and VPH effect on type 1 diabetes (13)
 
Comparisons of the results obtained by different testing approaches
Table 5 shows the results obtained by the Haplotype Trend Regression (HTR) program and Table 6 shows the P-values computed by haplo.score and the mean for each phenotype computed from the haplotypes deduced by Phase. A complete agreement was obtained by all the approaches used. Six out of the 10 commonest haplotypes at the IGF2–INS–TH region show significant association with at least one of the phenotypes. In total, 12 significant associations were found and two haplotypes showed trend (0.10<=P<0.05) in both HTR and haplo.score analyses (Tables 5 and 6). These results are also in agreement with the empirical P-values obtained by simulation using haplo.score (unpublished data). The mean values of each phenotype for haplotypes deduced via HTR or Phase are qualitatively similar (Tables 5 and 6). The results obtained from the cladistic-based association analysis performed by the program Evolutionary-based Haplotype Analysis Package (EHAP) were similar to those obtained by HTR and haplo.score, but significant values were in most cases around one order of magnitude different in EHAP (for example, *5 associated with weight with a P=0.0002 in EHAP, and 0.00001 and 0.00003 in HTR and haplo.score, respectively, and the significant value of association between *6 and plasma TG was 0.1007 in EHAP, 0.0147 in HTR and 0.0128 in haplo.score). All significant associations by HTR and haplo.score involving terminal nodes in the cladogram were confirmed by EHAP either at the P<0.05 level of significance or at the 0.05<P<0.10 level, with the exception of the association between *10 and fat mass (P=0.17). However, the associations involving *4 found by HTR and haplo.score were not confirmed by EHAP. *4 is an internal node in the cladogram and was analysed together with *6 (a terminal node) as a one-step clade. The joint consideration of two cladistically related haplotypes but different in allelic architecture at IGF2 ApaI opens the possibility that the effect of *4 in percentage fat, fat mass and diastolic BP found by both HTR and haplo.score could be confounded by the effect of *6 in these traits.


View this table:
[in this window]
[in a new window]
  		Table 5. Association, by the program HTR, of IGF2–INS–TH haplotypes with cardiovascular risk traits. For plasma TG, means are geometric. In bold are the significant values and in italics the cases with trend 0.05<P<=0.10
 

View this table:
[in this window]
[in a new window]
  		Table 6. Association, by the program haplo.score, of IGF2–INS–TH haplotypes with cardiovascular risk traits. The mean values for each trait across haplotypes were computed from the haplotype frequencies obtained by Phase. For plasma TG, means are geometric and SEs are approximate. In bold are the significant values and in italics the cases with trend 0.05<P<=0.10
 
Specific haplotypic associations
We focus our comments on the values presented in Table 6. No significant associations were observed for haplotypes IGF2–INS–TH *1, *2, *7, or *9. The most frequent haplotype (IGF2–INS–TH*1) had 825 alleles inferred but was not so frequent (20% of total) as to ‘set’ the overall mean trait values. For the least frequent haplotype (IGF2–INS–TH*10), 142 alleles (3.4%) were inferred. Carriers of haplotype IGF2–INS–TH*4 showed significantly higher mean percentage fat, fat mass and diastolic BP. The percentage differences from the overall sample means were 7.6, 10.1 and 1.5% respectively (Fig. 1). Haplotype IGF2–INS–TH*8 showed a notable (although non-significant) increased mean percentage fat (6%) and fat mass (8.5%). Haplotype IGF2–INS–TH*6 displayed a significant difference from the overall plasma TG sample geometric mean by 6.7%. The diastolic BP and TG observations were also tested in haplo.score with adjustments for smoking, body mass index (BMI), alcohol consumption, practice centre and age, after which the significances for *4 (diastolic BP) and *6 (plasma TG), respectively, were 0.0675 and 0.0166. Haplotype IGF2–INS–TH*5 displayed low weight (3.0%), low BMI (2.9%) and low lean body mass (LBM) (2.1%). The significant association with low plasma TG (1.70 mmol/l, P=0.0177) for IGF2–INS–TH*5 was ablated by adjustment for BMI; percentage fat and fat mass were also modestly lower but not reaching significance. Haplotype IGF2–INS–TH*3 was significantly associated with higher weight and height, (1.7 and 0.6%, respectively), whereas IGF2–INS–TH*8 displayed reduced mean height by 0.5%.



View larger version (19K):
[in this window]
[in a new window]
  		Figure 1. Significant associations (in bold and underlined) between IGF2–INS–TH haplotypes and cardiovascular risk traits. The bars represent the percentage difference observed between the mean value of a given haplotype and the overall sample mean for each trait. Also shown are haplotypes with non-significant associations (in italics) but with a notable percentage difference. Note that *3, *4, *6 and *8 confer an increase in the mean trait values whereas *5 and *10 confer a decrease.

 
Cladistic relationship among haplotypes
Figure 2 shows a cladogram relating the 10 haplotypes considered in the IGF2–INS–TH region. The commonest haplotypes (*1, *2 and *3) are clustered together with *7 and *9 in a clade characterized by the presence of the commonest alleles at both IGF2 ApaI and INS Hph1. Mutations of either INS Hph1 or IGF2 ApaI lead respectively to two different clades, the first containing *4 and *8 and the second containing *5 and *10. Lastly, *6 is in a separate clade distinguished by having the rare alleles at both SNPs. Greater (and presumably faster accruing) diversity is observed within and between clades for the TH01 microsatellite marker.



View larger version (7K):
[in this window]
[in a new window]
  		Figure 2. Cladogram relating the 10 commonest haplotypes at the IGF2–INS–TH region. The corresponding alleles for IGF2 ApaI, INS HphI and TH01, respectively, are indicated below each haplotype code. Haplotype 1–2–9, with a frequency of 2.3%, is also shown.

 
Retrospective power analyses based on observed haplotype frequencies
A retrospective analysis of the power for the significant associations detected revealed that the probability for detecting the differences observed in this work is higher than 70% for all associations, with the exception of the associations *8 with height (45.6%), *6 with plasma TG (56.3%) and *5 with plasma TG (63.4%).

Also computed (data available on Southampton Genetics Epidemiology Laboratories web site, www.sgel.humgen.soton.ac.uk/) were the percentage of detectable differences between the mean values for a given haplotype and the observed mean values for all individuals, for each trait. The results obtained indicate that there is a 90% power to detect very small differences for weight, systolic BP, diastolic BP, BMI and height (less than 4% for the rarest haplotype, *10, and around or less than 1.5% for the commonest one, *1). The total number of haplotypes observed for these four traits was 4100. Detectable differences for the rarest haplotype were around 10% for fat mass, percentage fat and LBM, and ~20% for plasma TG, whereas considerably smaller differences (2–12%) could be detected with a 90% power for the more frequent haplotypes. The total number of haplotypes observed for fat mass, percentage fat and LBM was 1180, and 4061 for plasma TG. The lowest range of observed percentage of detectable differences for five haplotypes representative of our haplotype frequencies (*1, *2, *3, *5 and *10) was for height (0.5–1%), followed by weight, systolic BP, diastolic BP (1.5–3.5%), LBM (2–10%), percentage fat (6–12%), fat mass (7–13%) and plasma TG (10–22%).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 NOTE ADDED IN PROOF
 REFERENCES
 
This study represents the first detailed description both of the IGF2 haplotypes and of the haplotypic effects of the IGF2–INS–TH region on a spectrum of cardiovascular risk traits. The study was undertaken in a cohort of 2743 middle-aged Caucasian males from a UK multicentre study of cardiovascular risk. Haplotypes were identified which may be contributors to combinations of fat mass, plasma TG level and raised BP, while others may influence anthropometric traits. Comparison of these haplotypes permits an integrating hypothesis and, in conjunction with identification of a marker set for tagging haplotypes, opens the way to more thorough studies of this ‘polygene’ in cardiovascular risk traits and coronary case studies.

Haplotype IGF2–INS–TH*4 is associated with significantly higher fat mass and percentage fat, and with significantly higher diastolic BP: the means for systolic BP and plasma TG are also respectively the highest and second highest for any haplotype, although not significant. Interestingly, similar magnitudes of effects were observed for IGF2–INS–TH*8, although there would not have been sufficient power (145 compared with 419 inferred haplotypes) to observe statistical signal for this magnitude of effect. Haplotypes *4 and *8 both contain the INS VNTR class III allele and the IGF2 ApaI G allele, although differing for the TH01 allele. Only one other haplotype contains INS VNTR class III, namely IGF2–INS–TH*6. This haplotype displays significantly higher geometric mean plasma TG but not the other phenotypic features; however, it differs in containing the IGF2 ApaI A allele and a quite different 11-SNPs IGF2 haplotype (Table 3). In general, previous literature concerned with case studies in coronary disease, type 2 diabetes and related traits such as PCOS has implicated INS VNTR class III in disease causation (Table 7) involving metabolic syndrome features. In addition, class III alleles have been associated with lower insulin secretion and irregular insulin pulsatility in class III/III-positive individuals (Table 2), and with lower INS mRNA transcript levels in pancreatic beta cells (Table 1). However, plasma TG has remained essentially uninvestigated for INS VNTR, IGF2 or TH genotype effects, although this is an obvious hypothesis for INS VNTR. There is a single early case study of type 2 diabetics which showed that long alleles were more prevalent in hypertriglyceridemic than in normotriglyceridemic type 2 diabetics (11), which accords with our findings in a population-based sample. The higher diastolic BP observed for haplotype *4 could be related to the reported insulin-mediated vasodilatation via nitric oxide release (Table 1); class III alleles at *4 would lead to a reduced availability of insulin and subsequent reduction in the expression of endothelial nitric oxide synthase. This would result in a diminished vasodilatation and increased BP. Other factors may be implicated in the effect observed, given that haplotype *6, also including class III alleles, does not show higher BP. In a case study, Sharma et al. (12) implicated TH01 allele 9.3 in hypertension; it is possible that this could represent the effect of the *4 haplotype observed here, although TH01 allele 9.3 is also found in haplotypes *3, *6 and *10. By contrast, the other four TH01 alleles display rather more restricted distributions on IGF2 and IGF2-INS haplotypes. Thus, in general, these observations in this cohort study identify haplotypes consistent with case studies of INS VNTR and TH01 alleles.


View this table:
[in this window]
[in a new window]
  		Table 7. Summary of apparent traits or disease effects associated with particular alleles and haplotypes at the IGF2–INS–TH region
 
Haplotype *8 contains TH01 allele 8 and therefore (13) represents the INS VNTR lineage/class IIIB which is recognized to be the very protective haplotype (VPH) in type 1 diabetes. By contrast, both *4 and *6 haplotypes contain TH01 allele 9.3 and they represent the INS VNTR lineage/class IIIA, which is known to be a pure lineage and is the protective haplotype (PH) in type 1 diabetes (13). However, here we show that the *6 haplotype represents a different IGF2 haplotype (E) which is identical to IGF2 haplotype C and *5 haplotype at its 3' end. It is possible therefore that the INS VNTR ‘class III effects’ observed for haplotypes *4 and *8 are modulated by the different IGF2 haplotype evident for *6, a haplotype which in fact has features in common with the weight-lowering haplotype *5.

We have previously reported evidence of associations of markers in IGF2, INS and TH to low weight and BMI (14–16). However, for the first time here, it can be seen definitively that one unique haplotype, *5, contains these markers and shows a highly significant association with low weight and low BMI. Additionally, for the first time, we show that this low BMI reflects a significantly low LBM and also low fat mass. Power calculations show that the study had considerably more power to determine LBM than fat mass effects. LBM represents 75% of body mass and fat mass represents 25% of body mass, the latter showing greater variance, whereas there is a 1.33 kg effect for LBM and a 1.03 kg trend for fat mass. This haplotype (*5) involves the IGF2 ApaI A allele which associates with high plasma total IGF-II level and with low IGF2 mRNA levels in leucocytes, and also with INS VNTR class I alleles. We have also previously shown that it is alleles of the INS VNTR lineage IC+ (the smallest class I subclasses), not of lineage ID– (class I alleles 814, 828 and 843) (13) or ID+ (class I alleles 770 and 786) (13) that correlate with weight-lowering effects (17). Indeed, in the present analyses (unpublished data), we did not detect significant differences for either lineage ID– or ID+ for any of the traits analysed. The trend toward low plasma TG for *5 is likely to be dependent on the low BMI, since the trend is ablated by adjustment for BMI. Haplotype *10 is the only other to share the INS VNTR class I and IGF2 ApaI A features and interestingly it shows a significant association with low mean fat mass and a trend to low percentage fat. LBM is lower than for *5, but does not reach significance for this rarer haplotype. These haplotypes may thus have common features that are protective against cardiovascular risk. There is a suggestion that the TH01 allele 9 may associate with higher noradrenaline levels, that INS VNTR class I alleles yield greater INS expression and secretion from the beta cell, that IGF2 ApaI A alleles associate with higher total plasma IGF-II and that low circulating IGF-II concentrations predict weight gain and obesity in humans (Table 1). Detailed comparison of mRNA and protein expression for each gene between haplotype *5 and suitably matched haplotypes may give further insight into the mechanism of action.

Each continuous trait phenotype distribution was examined for each haplotype showing significant association. None was significantly different from normal distribution (log normal for TG). This argues against a subgroup of any haplotype being responsible for the effects found. On the other hand, the cladistic relationship among haplotypes found in this study permits speculation on the evolution of variation at the IGF2–INS–TH region in relation to cardiovascular traits. The commonest haplotypes [which are likely to be the oldest (18,19) although there are exceptions to this generality (64)], marked by the presence of the common alleles at IGF2 ApaI and INS HphI, do not show association with the traits analysed. From this clade, two new clades with effects on cardiovascular traits emerged and these are tagged by alleles either of IGF2 ApaI or INS HphI. Mutation at IGF2 ApaI marks weight-lowering haplotypes, whereas mutation at INS HphI (equivalent to class III alleles at the INS VNTR), marks haplotypes with increased risk for metabolic syndrome traits. Finally, a more recent clade marked by the presence of the rarer alleles at both single-nucleotide polymorphisms (SNPs) has effects on plasma TG variation.

Power to detect effects is greatest for the most frequent haplotypes, and we can conclude important negative findings comparing haplotypes *1, *2 and *3 for the cardiovascular risk traits analysed. These equate to IGF2 haplotypes A, D and A. The haplotype frequency descriptions and trait means and SDS also permitted, for the first time, prospective power calculations for future studies. Although the use of power calculations is a controversial topic because the observed effect sizes will not necessarily be good estimates of the population values, our retrospective power analysis gives some indication of the probabilities that an exact replication study would successfully detect the differences observed in this work for each of the traits analysed if the present study is representative of true effect sizes. The genetic model tested here has been solely haplotype-based, measuring per-allele effects. Power to determine effects of haplotype combinations as diplotypes would need far greater sample sizes. In some cases we did explore combinations and, for example, INS VNTR class III homozygotes showed twice as great a difference from class I homozygotes for geometric mean plasma TG than did heterozygotes.

This descriptive study of haplotypes contains nine traits and 10 haplotypes (Table 6), at worst demanding a Bonferroni correction of 90. Under this model, all significances except the low weight and BMI effect of haplotype *5 would be eliminated. More realistically, many of the traits are interrelated, such as fat with percentage fat, (LBM plus fat mass) with weight, systolic with diastolic BP, weight and height with BMI. Some of the haplotypes are cladistically related, such as *4 and *6 and *8 (INS VNTR class III), and *5 and *10 differing only at TH01. More modest Bonferroni correction (e.g. 20) would leave most signals significant. However, known gene function and literature permit prior hypotheses with justifiable pooling of some haplotypes (e.g. *4, *6 and *8). Stringent trait selection would omit height, percentage fat and systolic BP. In addition, the congruence between the different approaches used to test for significance supports the validity of the associations found. Replication of the results found in this work in an independent sample would provide stronger evidence of the presence of a quantitative trait locus for cardiovascular traits in the IGF2–INS–TH gene cluster. However, the effect measurements in our study are mutually consistent with an increasing body of literature that suggests that this region may be important in obesity, metabolic disease and cardiovascular disease.

Table 7 summarizes existent genotypic associations—some but not all can be integrated but several need further data at the haplotypic level. The weight-lowering haplotype *5 contains the INS VNTR class IC+ alleles most prominently linked and associated with type 1 diabetes, and it also contains the TH01 allele 9 over-represented in normotensive controls in a hypertension case study (12). INS VNTR class III associations with type 2 diabetes, PCOS and high TG levels in diabetics must reflect some or all of haplotypes *4, *6 and *8, of which *4 and *8 appear to associate with raised fat mass and diastolic BP and *6 with raised TG level, but further case studies will be needed to determine whether the IGF2 haplotype of *6 modulates INS VNTR class III effects. However, observations in hypertension and longevity relating to TH01 allele 9.3 are difficult to interpret because this allele is found in association with a variety of IGF2 and INS haplotypes.

The potential significance of haplotype *4 (10.2% of alleles) to cardiovascular risk is not trivial. A 1.29 mmHg higher diastolic BP equates to about a 6–8% increment in stroke risk and a 4–5% increment in CHD risk (20). Comparable estimations can be made for the 10% fat mass increment and for the 10% higher plasma TG of haplotype *6 in relation to the metabolic syndrome. The weight-lowering effect of *5 (1 BMI unit) represents about a 10% increase in the rate of coronary events (21). This systematic description identifies tagging markers and specific haplotypes (particularly *4 for risk and *5 for protection), permitting deeper epidemiological and functional investigation of IGF2–INS–TH as a cardiovascular risk polygene.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
NOTE ADDED IN PROOF
 REFERENCES
 
Study sample
The Northwick Park Heart Study II (NPHSII) (22) comprises an unselected group of 3052 individuals taken from nine general practices distributed throughout the UK with DNA available on 2743 subjects. All individuals were healthy adult males aged 51–62 years, with the following exclusions: a history of unstable angina or myocardial infarction, regular anticoagulant medication, cerebrovascular disease, malignancy or other conditions precluding informed consent. Body mass indicators considered for each individual were BMI, defined as weight (kg)/height (m2), percentage of fat, fat mass and LBM. Plasma TG concentrations, systolic and diastolic BP were determined for each individual as previously described (23).

DNA markers
Eleven SNPs at the IGF2 region were studied, named 6815, 8173, 1156, AluI, 2482, 2722, 266, ApaI, 1926, 2207 and 3750 (15). These SNPs span 30 kb in the IGF2 gene (Fig. 3). One SNP (HphI) and the INS VNTR locus located at 596 bp 5' to the INS start site were analysed in the INS gene region. The tetranucleotide microsatellite TH01 was studied as a marker of the TH gene region.



View larger version (20K):
[in this window]
[in a new window]
  		Figure 3. Genetic map of the IGF2–INS–TH region spanning around 40 kb of 11p15. Exons of TH, INS and IGF2 are numbered and represented by boxes.

 
Genotyping
Experimental conditions for the genotyping of polymorphisms are as previously described (15,16). In brief, the 11 SNPs at IGF2 and HphI were typed by PCR adopting the Amplification Refractory Mutation System (ARMS) (24) or by digesting each amplification product by the corresponding restriction enzyme. The TH microsatellite TH01 and the INS minisatellite were amplified with primers and PCR conditions previously described (25–27). The amplification products (VNTR, microsatellite and SNP assays) in all cases were resolved by Microplate Array Diagonal Gel Electrophoresis (MADGE) as previously described (28–32).

Statistical analyses
A linkage disequilibrium (LD) map of the region, carried out by means of a recently published approach based on the {rho} measure of association (33), indicated the existence of three small blocks of LD connected by steps, in a scale of 1.2 LD units (unpublished data). The narrow range of LD units found confirmed previous evidence indicating strong association in this region (15,34). Therefore all statistical analyses treated the genotypic data in one haplotype block.

Four statistically different approaches were adopted, HTR (35); haplo.score (36); a cladistic-based association analysis (19) and Phase (37) (combined with SPSS) to test for association of indirectly deduced haplotypes with weight, height, BMI, LBM, percentage fat, fat mass, plasma TG levels, systolic BP and diastolic BP. Plasma TG values were log normalized prior to analysis. Each analysis used the richness of the continuous phenotypic variable, in preference to approaches using binary or other categorizations, in conjunction with between-group haplotype frequency comparisons. These approaches were compared to determine the statistical robustness of their conclusions, given that haplotype uncertainty is a major issue complicating regression analyses in samples of unrelated individuals (9).

HTR estimates the haplotype frequencies by use of the expectation–maximization (EM) algorithm and then relates the inferred haplotype frequencies to the observed phenotype using a regression model (35). The approach gives a significance value (deduced by an F-test) and a mean trait value for each haplotype in relation to each trait.

Haplo.score assigns the probability for each haplotype pair in each individual and then directly models an individual's phenotype as a function of each inferred haplotype pair, weighted by their estimated probability, to account for haplotype ambiguity (9,36). This program has the advantage that adjustment for covariates and computation of simulation P-values for each haplotype can be performed. TG, systolic and diastolic BP analyses were performed unadjusted and adjusted for age, BMI, smoking, practice centre and reported alcohol consumption. The number of simulations for empirical P-values was set as 1000. Haplo.score gives a significance test but not a mean trait value for each haplotype.

The cladistic-based association analysis was performed using the program EHAP (19). This program utilizes the information contained in the evolutionary relationships among haplotypes and in the sample using generalized linear models. A sequential series of test is performed first considering haplotypes which fall into the external nodes of the cladogram (zero-step clades) and then considering one-step clades (produced by moving backward one mutational step from the zero-step clades toward internal nodes), two-step clades and so forth. As in haplo.score, EHAP gives a significance test but not a mean trait value for each haplotype.

The Phase program (37) uses Gibbs sampling, a type of Markov-chain Monte Carlo algorithm (38), which may increase the accuracy in haplotype estimation, potentially reducing the error rates of EM algorithm by >50% (37). The numbers of iterations and burns-in performed was 10 000, each iteration consisting of performing 100 steps through the Markov chain, and the predefined level of confidence was set at >=90%. Mean values for each trait for each haplotype were then calculated, for comparison with values obtained by HTR.

The distributions of the haplotypes significantly associated with each trait were tested for normality in SPSS using the Kolmogorov–Smirnov goodness-of-fit test (39). Retrospective power analyses were also conducted for the detected associations and computed the percentage of detectable differences for each trait for a given power of 90%. Power calculations were performed as previously described (40) using a Normal Power Calculator (http://calculators.stat.ucla.edu/powercalc/) assuming normal distributions for the traits and unequal variances between the mean trait values for each haplotype and those observed for all individuals.

Cladogram
A cladogram was constructed using the principle of parsimony from the haplotypes considered in the analyses. This was guided using cladistic analyses performed in EHAP (19).


   ACKNOWLEDGEMENTS
 
Dr Andy Collins is thanked for advice on use of the {rho} measure of LD blocks. S.R. was an EC Marie Curie Research Fellow. I.N.M.D. was a Lister Institute Professor. The UK Medical Research Council and the British Heart Foundation are thanked for support (RG 2000 015).


   NOTE ADDED IN PROOF
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 NOTE ADDED IN PROOF
REFERENCES
 
A recent paper (65) presents data consistent with directional selection of INS VNTR class I alleles in Europeans, although possibly not of more distal markers in the gene cluster. This could imply that, overall in human history, the IGF2-INS-TH haplotypes which we identify as common (e.g. *1, *2 and *3, Fig. 2), may not be ancestral.


   FOOTNOTES
 
* To whom correspondence should be addressed. Tel: +44 2380794141; Fax: +44 2380794264; Email: santi{at}soton.ac.uk

{dagger} The authors wish it to be known that, in their opinion, the first two authors should be regarded as joint First Authors. Back

{ddagger} Present address: Department of Clinical Developmental Sciences, St George's Hospital Medical School, Cranmer Terrace, London SW17 0RE, UK. Back

§ Present address: Cardiovascular Institute, Fu Wai Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, 167 Beilishi Road, Beijing 100037, People's Republic of China. Back


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 NOTE ADDED IN PROOF
 REFERENCES
 

  1. Kopelman, P.G. (2000) Obesity as a medical problem. Nature, 404, 635–643.[Medline]

  2. Sankaranarayanan, K., Chakraborty, R. and Boerwinkle, E.A. (1999) Ionizing radiation and genetic risks. VI. Chronic multifactorial diseases: a review of epidemiological and genetical aspects of coronary heart disease, essential hypertension and diabetes mellitus. Mutat. Res., 436, 21–57.[CrossRef][ISI][Medline]

  3. Haffner, S.M. (2002) Lipoprotein disorders associated with type 2 diabetes mellitus and insulin resistance. Am. J. Cardiol., 90, 55i–61i.[ISI][Medline]

  4. Barsh, G.S., Farooqi, I.S. and O'Rahilly, S. (2000) Genetics of body-weight regulation. Nature, 404, 644–651.[Medline]

  5. Chagnon, Y.C., Rankinen, T., Snyder, E.E., Weisnagel, S.J., Perusse, L. and Bouchard, C. (2003) The human obesity gene map: the 2002 update. Obes. Res., 11, 313–367.[ISI][Medline]

  6. Munroe, P.B. and Caulfield, M.J. (2000) Genetics of hypertension. Curr. Opin. Genet. Dev., 10, 325–329.[CrossRef][ISI][Medline]

  7. Talmud, P.J. (2001) Genetic determinants of plasma triglycerides: impact of rare and common mutations. Curr. Atheroscler. Rep., 3, 191–199.[Medline]

  8. Bennett, S.T. and Todd, J.A. (1996) Human type 1 diabetes and the insulin gene: principles of mapping polygenes. A. Rev. Genet., 30, 343–370.[CrossRef][ISI][Medline]

  9. Zhao, H., Pfeiffer, R. and Gail, M.H. (2003) Haplotype analysis in population genetics and association studies. Pharmacogenomics, 4, 171–178.[CrossRef][ISI][Medline]

  10. Comeron, J.M., Kreitman, M. and De La Vega, F.M. (2003) On the power to detect SNP/phenotype association in candidate quantitative trait loci genomic regions: a simulation study. In Pacific Symposium on Biocomputing, pp 478–489.

  11. Jowett, N.I., Williams, L.G., Hitman, G.A. and Galton, D.J. (1984) Diabetic hypertriglyceridaemia and related 5' flanking polymorphism of the human insulin gene. Br. Med. J. (Clin. Res. Ed), 288, 96–99.

  12. Sharma, P., Hingorani, A., Jia, H., Ashby, M., Hopper, R., Clayton, D. and Brown, M.J. (1998) Positive association of tyrosine hydroxylase microsatellite marker to essential hypertension. Hypertension, 32, 676–682.[Abstract/Free Full Text]

  13. Stead, J.D., Buard, J., Todd, J.A. and Jeffreys, A.J. (2000) Influence of allele lineage on the role of the insulin minisatellite in susceptibility to type 1 diabetes. Hum. Mol. Genet., 9, 2929–2935.[Abstract/Free Full Text]

  14. O'Dell, S.D., Miller, G.J., Cooper, J.A., Hindmarsh, P.C., Pringle, P.J., Ford, H., Humphries, S.E. and Day, I.N. (1997) Apal polymorphism in insulin-like growth factor II (IGF2) gene and weight in middle-aged males. Int. J. Obes. Relat. Metab. Disord., 21, 822–825.[CrossRef][ISI][Medline]

  15. Gaunt, T.R., Cooper, J.A., Miller, G.J., Day, I.N. and O'Dell, S.D. (2001) Positive associations between single nucleotide polymorphisms in the IGF2 gene region and body mass index in adult males. Hum. Mol. Genet., 10, 1491–1501.[Abstract/Free Full Text]

  16. Gu, D., O'Dell, S.D., Chen, X.H., Miller, G.J. and Day, I.N. (2002) Evidence of multiple causal sites affecting weight in the IGF2-INS-TH region of human chromosome 11. Hum. Genet., 110, 173–181.[CrossRef][ISI][Medline]

  17. O'Dell, S.D., Bujac, S.R., Miller, G.J. and Day, I.N. (1999) Associations of IGF2 ApaI RFLP and INS VNTR class I allele size with obesity. Eur. J. Hum. Genet., 7, 821–827.[CrossRef][ISI][Medline]

  18. Zapata, C., Rodríguez, S., Visedo, G. and Sacristán, F. (2001) Spectrum of nonrandom associations between microsatellite loci on human chromosome 11p15. Genetics, 158, 1235–1251.[Abstract/Free Full Text]

  19. Seltman, H., Roeder, K. and Devlin, B. (2003) Evolutionary-based association analysis using haplotype data. Genet. Epidemiol., 25, 48–58.[CrossRef][ISI][Medline]

  20. MacMahon, S., Peto, R., Cutler, J., Collins, R., Sorlie, P., Neaton, J., Abbott, R., Godwin, J., Dyer, A. and Stamler, J. (1990) Blood pressure, stroke, and coronary heart disease. Part 1, Prolonged differences in blood pressure: prospective observational studies corrected for the regression dilution bias. Lancet, 335, 765–774.[CrossRef][ISI][Medline]

  21. Shaper, A.G., Wannamethee, S.G. and Walker, M. (1997) Body weight: implications for the prevention of coronary heart disease, stroke, and diabetes mellitus in a cohort study of middle aged men. Br. Med. J., 314, 1311–1317.[Abstract/Free Full Text]

  22. Miller, G.J., Bauer, K.A., Barzegar, S., Cooper, J.A. and Rosenberg, R.D. (1996) Increased activation of the haemostatic system in men at high risk of fatal coronary heart disease. Thromb. Haemost., 75, 767–771.[ISI][Medline]

  23. Cooper, J.A., Miller, G.J., Bauer, K.A., Morrissey, J.H., Meade, T.W., Howarth, D.J., Barzegar, S., Mitchell, J.P. and Rosenberg, R.D. (2000) Comparison of novel hemostatic factors and conventional risk factors for prediction of coronary heart disease. Circulation, 102, 2816–2822.[Abstract/Free Full Text]

  24. Newton, C.R., Graham, A., Heptinstall, L.E., Powell, S.J., Summers, C., Kalsheker, N., Smith, J.C. and Markham, A.F. (1989) Analysis of any point mutation in DNA. The amplification refractory mutation system (ARMS). Nucl. Acids Res., 17, 2503–2516.[Abstract/Free Full Text]

  25. Weber, J.L. and May, P.E. (1989) Abundant class of human DNA polymorphisms which can be typed using the polymerase chain reaction. Am. J. Hum. Genet., 44, 388–396.[ISI][Medline]

  26. Polymeropoulos, M.H., Xiao, H., Rath, D.S. and Merril, C.R. (1991) Tetranucleotide repeat polymorphism at the human tyrosine hydroxylase gene (TH). Nucl. Acids Res., 19, 3753.[Free Full Text]

  27. Bennett, S.T., Lucassen, A.M., Gough, S.C., Powell, E.E., Undlien, D.E., Pritchard, L.E., Merriman, M.E., Kawaguchi, Y., Dronsfield, M.J., Pociot, F. et al. (1995) Susceptibility to human type 1 diabetes at IDDM2 is determined by tandem repeat variation at the insulin gene minisatellite locus. Nat. Genet., 9, 284–292.[CrossRef][ISI][Medline]

  28. Day, I.N. and Humphries, S.E. (1994) Electrophoresis for genotyping: microtiter array diagonal gel electrophoresis on horizontal polyacrylamide gels, hydrolink, or agarose. Anal. Biochem., 222, 389–395.[CrossRef][ISI][Medline]

  29. O'Dell, S.D., Gaunt, T.R. and Day, I.N. (2000) SNP genotyping by combination of 192-well MADGE, ARMS and computerized gel image analysis. Biotechniques, 29, 500–506.[ISI][Medline]

  30. O'Dell, S.D., Chen, X. and Day, I.N. (2000) Higher resolution microplate array diagonal gel electrophoresis: application to a multiallelic minisatellite. Hum. Mutat., 15, 565–576.[CrossRef][ISI][Medline]

  31. Chen, X.H., O'Dell, S.D. and Day, I.N. (2002) Microplate array diagonal gel electrophoresis for cohort studies of microsatellite loci. Biotechniques, 32, 1080–1082, 1084, 1086.[ISI][Medline]

  32. Gaunt, T.R., Hinks, L.J., Rassoulian, H. and Day, I.N. (2003) Manual 768 or 384 well microplate gel ‘dry’ electrophoresis for PCR checking and SNP genotyping. Nucl. Acids Res., 31, E48.

  33. Maniatis, N., Collins, A., Xu, C.F., McCarthy, L.C., Hewett, D.R., Tapper, W., Ennis, S., Ke, X. and Morton, N.E. (2002) The first linkage disequilibrium (LD) maps: delineation of hot and cold blocks by diplotype analysis. Proc. Natl Acad. Sci. USA, 99, 2228–2233.[Abstract/Free Full Text]

  34. Julier, C., Hyer, R.N., Davies, J., Merlin, F., Soularue, P., Briant, L., Cathelineau, G., Deschamps, I., Rotter, J.I., Froguel, P. et al. (1991) Insulin-IGF2 region on chromosome 11p encodes a gene implicated in HLA-DR4-dependent diabetes susceptibility. Nature, 354, 155–159.[CrossRef][Medline]

  35. Zaykin, D.V., Westfall, P.H., Young, S.S., Karnoub, M.A., Wagner, M.J. and Ehm, M.G. (2002) Testing association of statistically inferred haplotypes with discrete and continuous traits in samples of unrelated individuals. Hum. Hered., 53, 79–91.[ISI][Medline]

  36. Schaid, D.J., Rowland, C.M., Tines, D.E., Jacobson, R.M. and Poland, G.A. (2002) Score tests for association between traits and haplotypes when linkage phase is ambiguous. Am. J. Hum. Genet., 70, 425–434.[CrossRef][ISI][Medline]

  37. Stephens, M., Smith, N.J. and Donnelly, P. (2001) A new statistical method for haplotype reconstruction from population data. Am. J. Hum. Genet., 68, 978–989.[CrossRef][ISI][Medline]

  38. Gilks, W.R., Richardson, S., and Spiegelhalter, D.J. (1996) Markov Chain Monte Carlo in Practice. Chapman & Hall, London.

  39. Massey, F.J., Jr (1951) The Kolmogorov-Smirnov test of goodness of fit. J. Am. Stat. Assoc., 46, 68–78.[CrossRef][ISI]

  40. Thompson, S.K. (2002) Sampling. J Wiley, New York.

  41. Jones, B.K., Levorse, J. and Tilghman, S.M. (2001) Deletion of a nuclease-sensitive region between the Igf2 and H19 genes leads to Igf2 misregulation and increased adiposity. Hum. Mol. Genet., 10, 807–814.[Abstract/Free Full Text]

  42. Da Costa, T.H., Williamson, D.H., Ward, A., Bates, P., Fisher, R., Richardson, L., Hill, D.J., Robinson, I.C. and Graham, C.F. (1994) High plasma insulin-like growth factor-II and low lipid content in transgenic mice: measurements of lipid metabolism. J. Endocrinol., 143, 433–439.[Abstract]

  43. Zaina, S., Pettersson, L., Thomsen, A.B., Chai, C.M., Qi, Z., Thyberg, J. and Nilsson, J. (2003) Shortened life span, bradycardia, and hypotension in mice with targeted expression of an Igf2 transgene in smooth muscle cells. Endocrinology, 144, 2695–2703.[Abstract/Free Full Text]

  44. Scherrer, U. and Sartori, C. (1997) Insulin as a vascular and sympathoexcitatory hormone: implications for blood pressure regulation, insulin sensitivity, and cardiovascular morbidity. Circulation, 96, 4104–4113.[Abstract/Free Full Text]

  45. Kuboki, K., Jiang, Z.Y., Takahara, N., Ha, S.W., Igarashi, M., Yamauchi, T., Feener, E.P., Herbert, T.P., Rhodes, C.J. and King, G.L. (2000) Regulation of endothelial constitutive nitric oxide synthase gene expression in endothelial cells and in vivo: a specific vascular action of insulin. Circulation, 101, 676–681.[Abstract/Free Full Text]

  46. Albanese, V., Biguet, N.F., Kiefer, H., Bayard, E., Mallet, J. and Meloni, R. (2001) Quantitative effects on gene silencing by allelic variation at a tetranucleotide microsatellite. Hum. Mol. Genet., 10, 1785–1792.[Abstract/Free Full Text]

  47. Vafiadis, P., Bennett, S.T., Todd, J.A., Nadeau, J., Grabs, R., Goodyer, C.G., Wickramasinghe, S., Colle, E. and Polychronakos, C. (1997) Insulin expression in human thymus is modulated by INS VNTR alleles at the IDDM2 locus. Nat. Genet., 15, 289–292.[CrossRef][ISI][Medline]

  48. Pugliese, A., Zeller, M., Fernandez, A., Jr, Zalcberg, L.J., Bartlett, R.J., Ricordi, C., Pietropaolo, M., Eisenbarth, G.S., Bennett, S.T. and Patel, D.D. (1997) The insulin gene is transcribed in the human thymus and transcription levels correlated with allelic variation at the INS VNTR-IDDM2 susceptibility locus for type 1 diabetes. Nat. Genet., 15, 293–297.[CrossRef][ISI][Medline]

  49. Malmstrom, R., Packard, C.J., Watson, T.D., Rannikko, S., Caslake, M., Bedford, D., Stewart, P., Yki-Jarvinen, H., Shepherd, J. and Taskinen, M.R. (1997) Metabolic basis of hypotriglyceridemic effects of insulin in normal men. Arterioscler. Thromb. Vasc. Biol., 17, 1454–1464.[Abstract/Free Full Text]

  50. Malmstrom, R., Packard, C.J., Caslake, M., Bedford, D., Stewart, P., Yki-Jarvinen, H., Shepherd, J. and Taskinen, M.R. (1998) Effects of insulin and acipimox on VLDL1 and VLDL2 apolipoprotein B production in normal subjects. Diabetes, 47, 779–787.[Abstract]

  51. Arner, P., Bolinder, J., Engfeldt, P. and Ostman, J. (1981) The antilipolytic effect of insulin in human adipose tissue in obesity, diabetes mellitus, hyperinsulinemia, and starvation. Metabolism, 30, 753–760.[CrossRef][ISI][Medline]

  52. Eckel, R.H., Prasad, J.E., Kern, P.A. and Marshall, S. (1984) Insulin regulation of lipoprotein lipase in cultured isolated rat adipocytes. Endocrinology, 114, 1665–1671.[Abstract]

  53. Nezer, C., Moreau, L., Brouwers, B., Coppieters, W., Detilleux, J., Hanset, R., Karim, L., Kvasz, A., Leroy, P. and Georges, M. (1999) An imprinted QTL with major effect on muscle mass and fat deposition maps to the IGF2 locus in pigs. Nat. Genet., 21, 155–156.[CrossRef][ISI][Medline]

  54. Jeon, J.T., Carlborg, O., Tornsten, A., Giuffra, E., Amarger, V., Chardon, P., Andersson-Eklund, L., Andersson, K., Hansson, I., Lundstrom, K. et al. (1999) A paternally expressed QTL affecting skeletal and cardiac muscle mass in pigs maps to the IGF2 locus. Nat. Genet., 21, 157–158.[CrossRef][ISI][Medline]

  55. Rice, T., Chagnon, Y.C., Perusse, L., Borecki, I.B., Ukkola, O., Rankinen, T., Gagnon, J., Leon, A.S., Skinner, J.S., Wilmore, J.H. et al. (2002) A genomewide linkage scan for abdominal subcutaneous and visceral fat in black and white families: the HERITAGE Family Study. Diabetes, 51, 848–855.[Abstract/Free Full Text]

  56. Sandhu, M.S., Gibson, J.M., Heald, A.H., Dunger, D.B. and Wareham, N.J. (2003) Low circulating IGF-II concentrations predict weight gain and obesity in humans. Diabetes, 52, 1403–1408.[Abstract/Free Full Text]

  57. Puers, C., Hammond, H.A., Jin, L., Caskey, C.T. and Schumm, J.W. (1993) Identification of repeat sequence heterogeneity at the polymorphic short tandem repeat locus HUMTH01[AATG]n and reassignment of alleles in population analysis by using a locus-specific allelic ladder. Am. J. Hum. Genet., 53, 953–958.[ISI][Medline]

  58. Jindra, A., Jachymova, M., Horky, K., Peleska, J., Umnerova, V., Bultas, J. and Savlikova, J. (2000) Association analysis of two tyrosine hydroxylase gene polymorphisms in normotensive offspring from hypertensive families. Blood Press., 9, 250–254.[CrossRef][ISI][Medline]

  59. Ukkola, O., Sun, G. and Bouchard, C. (2001) Insulin-like growth factor 2 (IGF2) and IGF-binding protein 1 (IGFBP1) gene variants are associated with overfeeding-induced metabolic changes. Diabetologia, 44, 2231–2236.[CrossRef][ISI][Medline]

  60. Cocozza, S., Riccardi, G., Monticelli, A., Capaldo, B., Genovese, S., Krogh, V., Celentano, E., Farinaro, E., Varrone, S. and Avvedimento, V.E. (1988) Polymorphism at the 5' end flanking region of the insulin gene is associated with reduced insulin secretion in healthy individuals. Eur. J. Clin. Invest, 18, 582–586.[ISI][Medline]

  61. Ahmed, S., Bennett, S.T., Huxtable, S.J., Todd, J.A., Matthews, D.R. and Gough, S.C. (1999) INS VNTR allelic variation and dynamic insulin secretion in healthy adult non-diabetic Caucasian subjects. Diabet. Med., 16, 910–917.[CrossRef][ISI][Medline]

  62. Waterworth, D.M., Bennett, S.T., Gharani, N., McCarthy, M.I., Hague, S., Batty, S., Conway, G.S., White, D., Todd, J.A., Franks, S. et al. (1997) Linkage and association of insulin gene VNTR regulatory polymorphism with polycystic ovary syndrome. Lancet, 349, 986–990.[CrossRef][ISI][Medline]

  63. Eaves, I.A., Bennett, S.T., Forster, P., Ferber, K.M., Ehrmann, D., Wilson, A.J., Bhattacharyya, S., Ziegler, A.G., Brinkmann, B. and Todd, J.A. (1999) Transmission ratio distortion at the INS-IGF2 VNTR. Nat. Genet., 22, 324–325.[CrossRef][ISI][Medline]

  64. Fullerton, S.M., Clark, A.G., Weiss, K.M., Nickerson, D.A., Taylor, S.L., Stengard, J.H., Salomaa, V., Vartiainen, E., Perola, M., Boerwinkle, E. et al. (2000) Apolipoprotein E Variation at the sequence haplotype level: implication for the origin and maintenance of a major human polymorphism. Am. J. Hum. Genet., 67, 881–900.[CrossRef][ISI][Medline]

  65. Stead, J.D.H., Hurles, M.E. and Jeffreys, A.J. (2003) Global haplotype diversity in the human insulin gene region. Genome Res., 13, 2101–2111.[Abstract/Free Full Text]


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 J. Clin. Endocrinol. Metab.Home page
B. Heude, K. K. Ong, R. Luben, N. J. Wareham, and M. S. Sandhu
Study of Association between Common Variation in the Insulin-Like Growth Factor 2 Gene and Indices of Obesity and Body Size in Middle-Aged Men and Women
J. Clin. Endocrinol. Metab., July 1, 2007; 92(7): 2734 - 2738.
[Abstract] [Full Text] [PDF]

Home page
 J. Appl. Physiol.Home page
J. M. Devaney, E. P. Hoffman, H. Gordish-Dressman, A. Kearns, E. Zambraski, and P. M. Clarkson
IGF-II gene region polymorphisms related to exertional muscle damage
J Appl Physiol, May 1, 2007; 102(5): 1815 - 1823.
[Abstract] [Full Text] [PDF]

Home page