Human Molecular Genetics, 2002, Vol. 11, No. 23 2969-2977
© 2002 Oxford University Press
Angiotensin-1-converting enzyme (ACE) plasma concentration is influenced by multiple ACE-linked quantitative trait nucleotides
1Medical Research Council, Mammalian Genetics Unit, Harwell, Didcot OX11 0RD, UK, 2Department of Biological Anthropology, University of Oxford, Oxford, UK, 3INSERM U 525, Faculté de Médecine Pitié-Salpêtrière, Université Paris VI, Paris, France, 4Centre National de Génotypage, Evry, France, 5Department of Preventive Medicine and Epidemiology, Loyola University Medical Center, Maywood, Illinois, USA, 6Department of Pediatrics/Institute for Child Health, College of Medicine, University of Ibadan, Ibadan, Nigeria and 7Department of Cardiovascular Medicine, University of Oxford, The Wellcome Trust Centre for Human Genetics, Roosevelt Drive, Oxford OX3 7BN, UK
Received July 26, 2002; Accepted September 9, 2002
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Circulating angiotensin-1-converting enzyme (ACE) is a highly heritable trait, and a major component of the genetic variance maps to the region of the ACE gene. The strong effect of the locus, and the interest in ACE as a candidate gene for cardiovascular disorders, has led to extensive investigation of its relationship to the ACE phenotype, providing one of the most complete examples of quantitative trait locus (QTL) analysis in humans. Resequencing of ACE followed by haplotype analysis in families of British and French origin has shown that the genetic variants that are primarily associated with the ACE trait map to an 18 kb interval flanked by two intragenic, ancestral recombination breakpoints. This critical interval contains dozens of ACE-associated variants in Caucasians, but identification of which of these directly influence ACE concentration is ambiguous because of the almost complete linkage disequilibrium in European populations. In a complementary sequencing and genotyping study of individuals from West African families, we show that this population has much greater haplotype diversity across the gene. Through analysis of the contrasting relationships of the trait phenotype with haplotypes that carry different allelic combinations from those observed in Caucasians, we demonstrate that (at least) two major intragenic sites within the critical interval and (at least) one minor promoter site are associated with the ACE quantitative trait through additive effects. These results point to the importance of analysing diverse populations with different gene genealogies in gene-association studies.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
A major objective in the study of common multifactorial diseases is to identify novel genes and biochemical pathways to gain increased insight into biological processes and gene function, and to suggest new opportunities for therapeutic intervention. To this end, positional gene identification strategies, in which genetic mapping studies day a central role, are being applied to many multifactorial diseases and associated phenotypes (complex traits). Mapping of complex traits in human is usually attempted through a combination of linkage with association studies. The latter depends on linkage disequilibrium between polymorphic markers and the (unknown) variants that modify disease susceptibility or directly influence quantitative traits. Once the relationship between a gene and a complex trait has been established, fine-mapping studies are usually undertaken to establish which gene variants are likely to have functional consequences related to disease susceptibility or trait variability, as a prerequisite to the difficult and time-consuming biological investigations that are required to explore the functional consequences of such variation. The power of fine-mapping studies is constrained by the patterns of linkage disequilibrium in the study population, which may make it difficult to distinguish functional variants from neutral polymorphisms occurring on the same haplotypes. Limitations in mapping resolution are a function both of the size of the sample that can reasonably be investigated (which limits the number of rare informative haplotypes that allow statistical discrimination between sites) and, more generally, of the evolutionary history of the population under study. Moreover, the population in which the relationship between the gene and trait was originally established may have led to a successful study precisely because of the presence of strong linkage disequilibrium, which allowed the association to be established with characterization of a limited number of makers in a relatively small sample. Thus, this population may not be ideal for subsequent fine-mapping studies. The obvious experimental strategy then is to choose a variety of populations with different evolutionary histories (and gene genealogies) and hope that the major variants influencing the complex trait predate the divergence of the populations.
Angiotensin-1-converting enzyme (ACE) provides one of the best models available in human for studying the genetics of a complex trait and exploring strategies for fine-mapping studies. Investigations of circulating (plasma or serum) ACE concentration (or enzymatic activity) have shown that that a major portion of interindividual quantitative variation is genetically determined (1), and linked to the ACE gene (2–5). Analysis of family data has shown that the haplotypes present in European populations can be conveniently grouped into several clades (6). The two most common clades (labelled A and B) in Europeans are distinguished by contrasting alleles at many polymorphic sites and are associated with very different ACE activities. We have suggested that the two other most frequent clades in Europeans (C and D) are likely to have evolved through two distinct ancestral recombination events between members of the A and B clades; one recombination event occurring between intron 5 and exon 6, and the other occurring 
18 kb downstream from the first region, and 3' of the last exon (6–8). Our studies of the trait variation associated with these four clades suggest that the majority of the variation attributed to the ACE quantitative trait locus (QTL) is associated with genetic variants that map to this 18 kb interval flanked by the two ancestral recombination breakpoints, and that a minor portion of the variation maps to upstream promoter sequences (8). However, although there are a number of potentially informative single-nucleotide polymorphisms (SNPs) in this region, because they are in strong linkage disequilibrium in European-origin populations it is impractical to map the location of quantitative variants in European-origin samples (8). Recent work has demonstrated that African-origin populations tend to have substantially greater haplotype diversity than do European-origin populations (9,10). Studying such populations is likely to reveal informative haplotypes that will allow informative contrasts between SNPs that are tightly held together in European-origin populations, and we have already demonstrated that the existence of different ACE gene genealogies in African-origin populations can lead to novel insights into the location of functional variants (3). The results reported here represent an extension of this strategy. Here, we report an extensive survey of ACE genetic variation in a West African population, which has guided the selection of polymorphisms for further characterization. Haplotype analysis of ACE markers and the ACE trait in families from this population suggests that several sites (at least one in the promoter and at least two in the distal half of the gene) are jointly associated with circulating ACE concentration. These sites now provide the basis for future functional studies, either by directly investigating the effects of the nucleotides themselves or by evaluating the functional attributes of the chromosomal region on which they lie.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Resequencing of ACE
Approximately 30 kb of genomic DNA spanning the ACE gene was resequenced in 10 Nigerian individuals. The ACE gene variation information supplemented a database of ACE polymorphisms generated through an independent survey of polymorphisms in populations of European and African ancestry (http://genecanvas.idf.inserm.fr). In an exploratory association screening exercise, 41 polymorphisms were genotyped in 24 Nigerian individuals; 12 individuals selected for high ACE trait values (z>2.1) and 12 selected for low ACE trait values (z<-1.5). Additional ACE polymorphisms were identified from public sources (11,12).
ANOVA of polymorphic markers and ACE trait
Thirty-five polymorphic sites were selected for genotyping in 765 members of 234 Nigerian families. These sites were chosen to span the totality of the gene, and to include all polymorphisms that appeared most likely to be associated with the trait in this population, based on their correlation with ACE levels in the resequencing experiments described above. Table 1 and Figure 1 summarize the results of a preliminary survey of the association between individual marker genotypes and residual ACE concentration (after adjustment for age and gender effects). Mixed linear models were used to simultaneously model residual intrafamilial correlations in the ACE trait with the marker–trait associations. The most strongly associated polymorphisms, A23495G, C29809T and A31958G, are located in the distal half of the ACE gene. However, it is noteworthy that three of the five upstream (promoter) markers (T6108C, A6138C and A11377T) are also highly significantly associated with the ACE trait. Computational limitations restrict the number of markers that can be simultaneously examined with our haplotype-based analysis methods. Accordingly, we first applied a stepwise, forward selection algorithm to identify a subset of the markers that are independently associated with the trait. Six markers (A23495G, A31958G, 31839insC, A6138C, 29349delT and G28952C) were selected in this analysis; Table 2 shows the statistical significance for each marker as it incremented the model as well as the final main-effects model. The proportion of the trait variance associated with the six markers collectively was 35%; 20% of the variance was determined by intrafamilial correlations and 45% by individual-specific factors. A second-order factorial model failed to detect any significant marker–marker interactions (P>>0.05), suggesting that the markers have independent effects on the trait.
|
|
|
Measured haplotype analysis
Haplotype frequencies were estimated by maximum likelihood in the founders (assuming no recombination) for the six markers identified in the stepwise ANOVA procedure described above. As shown in Table 3, eight haplotypes with frequency estimates
1.94% accounted for 98% of haplotypes in the families, with the three most frequent haplotypes accounting for 75% overall. Two of the markers, G28952C and 29349delT, only showed variation on relatively rare haplotypes (frequencies 
0.23%) for which there were insufficient data to make reliable statistical inferences (data not shown). Consequently, haplotype frequencies were re-estimated for the four markers (A6138C, A23495G, 31839insC, A31958G) that define a diverse series of common ACE haplotypes that are shown in Table 4.
|
|
The measured haplotype analysis therefore required estimation of the effects associated with nine distinct haplotypes (with population frequencies estimated to range from 0.62% to 31%; Table 4), and a pooled class for the remaining seven rare haplotypes. Table 4 and Figure 2 shows that the estimated mean ACE concentration varies considerably for different ACE haplotypes. A number of substitutions distinguish haplotypes that are associated with very different ACE activities. For instance: haplotypes 2 and 4 differ at a single site (31839insC) and have statistically significant differences in associated mean ACE concentration (P<0.01); haplotypes 9 and 10 differ at A31958G, and this substitution is associated with a large change in mean ACE concentration (P<0.001). Interestingly, haplotypes 6 and 9 differ at three sites (A6138C, A23495G and A31958G) and have extremely different ACE activities (P<<<0.001), suggesting that the effects associated with substitutions at multiple sites are cumulative.
|
In order to summarize the impact of the various combinations of the four sites, we devised and fitted an additive, allele-substitution model. In this model, a parameter was specified to measure the mean for haplotype 9 (the selection of this haplotype was arbitrary, but is convenient since it is fairly common and is associated with low ACE concentration). Four additional parameters were then specified to model the effects of substituting alleles at the polymorphic sites relative to their state in haplotype 9. For example, haplotype 1 differs from haplotype 9 at just one site, A6138C. The phenotypic effect fitted to haplotype 1 was specified as the sum of effects for haplotype 9 and the substitution effect for the A6138C site. Haplotype 6 differs from haplotype 9 at three sites, so its mean is the sum of effects measured for haplotype 9 and the three effects of substituting alleles at A6138C, A23495G and A31958G. By maximizing the likelihood of the additive, allele substitution model, we are, in effect, solving a set of nine simultaneous equations numerically. Figure 3 shows the results of this analysis. The goodness of fit of the additive, allele substitution model was satisfactory (P=0.13) when compared with the general four-locus model with nine haplotype-specific means as reported in Figure 2. The effects of independently substituting alleles at the A6138C, 31839insC and A31958G sites were highly significant when tested, with the additive effects of the other sites included under the alternative and null hypotheses (A6138C: P=3x10-4; 31839insC: P=3x10-7; A31958G: P=1x10-13). There was a significant effect (P=0.04) of substituting alleles at A23495G in this haplotype-based analysis. Residual intrafamilial correlations were significant (spouse–spouse r2=0.40±0.07; mother–child r2=0.31±0.06; father–child r2=0.28±0.05; sib–sib r2=0.21±0.06; all correlations P<<<0.05). In the additive, allele-substitution model, 33% of the trait variability was explained by ACE genetic variation, 19% with residual familial correlations and 48% with random, individual-specific factors.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Systematic resequencing of 10 individuals of Nigerian origin has led to the identification of the frequent polymorphisms in a 34 kb segment of DNA spanning the ACE gene. These experiments suggest that the Nigerian sample studied exhibits greater sequence polymorphism than European populations. For instance, of 81 polymorphisms that map downstream from the start of transcription, 38 were polymorphic in both Europeans and Nigerians, 37 were only polymorphic in Nigerians and 6 were only polymorphic in Europeans (http://genecanvas.idf.inserm.fr).
In addition to greater sequence diversity, the Nigerian population also exhibited a greater degree of haplotype diversity. This can be illustrated through a comparison of haplotypes involving a subset of eight ACE markers that were genotyped in the Nigerian families studied here (Table 5) and that have been previously studied in families of French origin (8). Although the allele frequencies estimates are broadly similar for these sites in the two populations, the patterns of allelic association, as revealed by the estimates of haplotype frequencies, were very different. Thus, in Nigerian families, the 15 most frequent haplotypes with frequencies ranging from 14% to 1.5% account for 80% of the observations, whereas in French and other European families, 78% of the observations are accounted for by just three haplotypes, with frequencies of 35%, 31% and 13% (8). Interestingly, the two most common European haplotypes (A and B) are estimated to have frequencies <1% in the Nigerian cohort.
|
Studies of British and French ACE haplotypes have shown that majority of ACE-linked quantitative variation maps to the 18 kb gene segment between G15408A and T33569C, which contains 31 sites of common variation in Europeans (6,8). These variants are in essentially complete linkage disequilibrium in European samples, so that it has not been possible to further limit the potential candidate variants within this interval by analysing European cohorts. However, although the most striking contrast is between the low concentration associated with clade A as compared with the high concentration associated with the other clades, clades B and D are associated with significantly higher mean trait values than clade C in two cohorts of French families (8); a similar trend was observed in a UK family cohort, but failed to attain statistical significance (6). Since clade C differs from clades B and D in the region 5' of T17944C, this suggests that one or more functional variants in the 5' region influence ACE concentration, as hypothesized by Villard and colleagues (12).
ACE concentration is also strongly associated with polymorphisms within the gene in the Nigerian cohort studied here. The degree of diversity and divergence of haplotypes observed in the Nigerians compared with Europeans suggested that analysis of this cohort could provide further information for identifying the variants implicated in ACE concentration. As an example of how results from the Nigerian cohort can contribute to this, consider the Alu insertion polymorphism (I/D) in intron 16 of the gene, which is within the region of essentially complete linkage disequilibrium and identity of variants in clades B, C and D in Europeans, and was the first polymorphism of the gene identified as being associated with the trait. The deletion variant (‘D’ allele) is present on these three clades, associated with high ACE concentration, while the I allele is present on the A clade, associated with low ACE concentration. Based on the genetic information from Europeans, therefore, I/D cannot be excluded as a candidate polymorphism that could directly affect the trait. However, our mixed modelling analysis (Table 2) demonstrates that this polymorphism is not an independent predictor of ACE concentrations in the Nigerian families. Further, when the I/D polymorphism is incorporated into the results of the measured haplotype analysis (Table 4), it becomes apparent that in this Nigerian cohort, haplotypes carrying the I allele can be associated with either high or low values of the trait. Even though haplotype 9 and haplotype 1 are both associated with low levels of ACE, the I allele is carried by only one third of haplotype 1 chromosomes, compared with 86% of haplotype 9 chromosomes. Further, the I allele is carried by 50% of haplotype 2 and haplotype 10 chromosomes, both of which are associated with high levels of ACE. Overall, the results of a supplemental measured haplotype analysis indicate a highly significant difference (P<<<0.05) in the ACE levels associated with the different haplotypes carrying the I allele (data available on request). Thus, the compression of genotypes, or haplotypes from the Nigerian sample eliminates I/D as a candidate QTL with a direct role in ACE concentrations, at least in a manner independent of other variants within the gene.
To identify ACE sites that could be candidates for direct implication in ACE concentration, we examined 35 variants that had been identified from the resequencing experiments in the complete Nigerian family cohort. A preliminary, stepwise regression analysis was performed using a mixed model that incorporated fixed, additive effects of the genotypes for common polymorphisms, and a random familial component of variance. Genotypic effects at six sites (A23495G, A31958G, 31839insC, A6138C, 29349delT and G28952C) were retained in the model. No significant interactions were found between the sites, and four sites (A23495G, A31958G, 31839insC, and A6138C) were confirmed to contribute to the trait variation in a measured haplotype analysis using an allele substitution (i.e. additive) model. No other site was found to contribute significantly (P>0.46) to the measured haplotype analysis once the four sites were incorporated in the model. Allele substitution at these sites accounted for 33% of the trait variability, with 19% being attributed to residual familial correlations.
Our results suggest that multiple sites within the ACE gene influence circulating levels of ACE concentration. The four sites that we have identified are candidates to have a direct effect on the trait. One of these (A6138C) is located in upstream non-coding sequence, a second (A23495G) is a silent substitution and the other two are located within introns. The A31958G polymorphism has also been identified as the site most associated with the trait in analysis of Nigerian and Jamaican volunteers by Zhu and colleagues (13). We note that three of the sites (A23495G, 31839insC and A31958G), are located in the region between G15408A and T33569C, i.e. the region likely to contain variants with major effects on the ACE trait, based on the analysis of European cohorts. However, only A23495G and A31958G are polymorphic in Europeans. Also, in European populations, these latter two sites are in very strong linkage disequilibrium, so their individual effects are not distinguishable. In the Nigerian families, substitution of alleles at these two sites accounted for 20% of the trait variance, compared with 30% of the trait variance attributed to the substitution of the entire segment between T17944C and T33569C in French families (estimated by comparing the trait means from the measured haplotype analysis for clades A, B and D). Moreover, the same alleles, 23495–A and 31958–A, are associated with low ACE concentration in both European and Nigerian populations. Allelic substitution at the A6138C site in the upstream non-coding sequence accounted for 5% of the trait variance. Above, we noted that European cohorts provide some evidence of an effect on the trait of genetic variation in the region 5' of T17944C, which includes A6138C. This region accounted for 7% the variance of the trait in the French family cohorts. However, different alleles (A in Europeans and C in Nigerians), are associated with low ACE concentration. The most likely explanation is that this polymorphism is indirectly associated with the ACE trait and that another 5' site in strong linkage disequilibrium with A6138C directly influences the ACE trait. Collectively, these observations suggest that a number of different variants within the ACE gene region influence levels of circulating ACE, and that the magnitude of the effect of each variant may differ considerably in different populations. This clearly makes the challenge of localizing individual quantitative trait alleles (QTAs) all the more difficult, and requires the use of analytic strategies that are capable of discriminating between, and localizing the effect of, different variants scattered throughout a candidate locus.
Finally, we note that we cannot exclude the possibility that our identification of A6138C, A23495G, 31839insC and A31958G could be due to complex patterns of linkage disequilibrium with other sites, or that additional polymorphisms may also affect the trait. However, it is noteworthy that none of the common variants within the ACE gene would be predicted to have obvious biological effects based on knowledge of their position within the gene. The observation that multiple associated polymorphisms could influence a quantitative trait, even though these effects cannot necessarily be predicted based on sequence information, was first reported in genetic studies of the alcohol dehydrogenase gene in Drosophila, in which three sites (one intronic, one exonic and one in the 3' non-coding sequences) jointly influence the enzyme expression (14). Further, studies of the Delta-linked QTL in Drosophila have shown that two intronic nucleotides are strikingly associated with variation in bristle number (15). Even more relevant is the fact that non-coding SNPs are associated with measurable differences in mRNA levels in 6 out of 13 human loci examined (16). One of the six loci was calpain-10, in which the variation in mRNA was shown to co-segregate with the allelic variant, and which had previously been shown to have intronic SNPs, associated with type 2 diabetes, and with altered mRNA levels (17).
Our data for the ACE gene suggest at least four functional variants influence trait levels, and that these variants are distributed in complex patterns along the gene, depending on the specific evolutionary history of the populations. Consequently, in some populations these variants will segregate independently, while in others they will not. The magnitude of effect may also differ by population as a function of the influence of genetic background and environmental factors. It might be expected that similar situations would pertain frequently in the study of quantitative variation in humans. The recourse to haplotype analysis in large cohorts and multiple populations is likely to be the key strategy in identifying subsets of candidate variants as a prelude to functional explorations.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
234 Nigerian families (765 individuals) that were collected through the multicentre International Collaborative Study on Hypertension in Blacks sampling frame (5,18–20), were selected for high-resolution mapping. Plasma ACE concentrations had been previously measured in a modified immunosorbent assay (5,21). DNA was extracted from buffy coats by standard methods as previously described (5). Genotypes for 13 polymorphisms detected by PCR/RFLP analysis have been previously published (5); genotypes for 22 new polymorphisms were detected by DNA resequencing (Table 1). The PCR reactions for resequencing were performed in a 15 µl mix containing 25 ng of DNA. Primer sequences are available from the authors on request. Sequencing reactions were performed according to the dye terminator method using an ABI PRISM 3700 DNA Analyzer (Applied Biosystems, Foster City, CA). Alignment of experimental results, SNP discovery and genotyping were performed with the Genalys software (22).
Statistical procedures
ACE concentrations were standardized after adjustment for age and sex co-variation. Parameters estimates and tests statistics under mixed linear ANOVA models were calculated using PROC MIXED in SAS 8.2 (SAS Institute, Cary, NC), with genotypes specified as fixed effects and KINDRED (a sequential indicator variable to distinguish families) specified as a random effect. A stepwise, forward selection procedure was implemented manually, with entry threshold P<0.05 based on type III sums-of-squares tests. A forward selection (as opposed to a backward elimination) model selection procedure was adopted to efficiently allow for individuals with incomplete genotypes.
Haplotype frequencies were estimated in the founders (assuming no recombination) by maximum likelihood using the computer program Merlin (23) and an accompanying unpublished program Fugue (Frequencies with graphs). A recursive procedure based on a modified version of the UNKNOWN program [in the LINKAGE package (24)] was used to compile a list of alternative genotypes for individuals, taking into account undetermined genotype status. The list was input to a new haplotyping procedure (HAPLOTRY), which enumerates all sets of non-recombinant haplotypes consistent with the genotype data. A final exhaustive check for Mendelian inconsistencies was made using the Pedcheck program (25), since the current implementation of HAPLOTRY makes simple rule-based checks. The alternative haplotype configurations for each family were linked to the adjusted (for sex and age effects) and standardized ACE phenotypic data. Likelihoods were calculated using a modified version of PAP (26) assuming additive (co-dominant) effects of haplotypes on the trait. Intra-nuclear-family correlations not accounted for by shared haplotype effects were modelled using the PAP papwgfc subroutine (27). The additive, allele substitution model was parameterized in a modified version of the PAP qmlprmv subroutine. For example, consider a two-locus model with alleles A and a and B and b that define four haplotypes AB, Ab, aB, ab. SA
a measures the additive effect of substituting a for an A allele and SBb measures the additive effect of substituting b for a B allele. The haplotype-specific means (µAb, µaB, µab) can then be defined in addition to and relative to µAB as follows:
 |
 |
 |
The mean measured for each genotype is then the sum of the haplotype-specific effects (given that phase has been assigned by the HAPLOTRY procedure), for example for the genotype µAA/BB=µAB+µAB.
Likelihoods were maximized with simultaneous estimation of haplotype frequencies, mean haplotype effects, and residual familial and random effects, and standard errors were calculated numerically with the GEMINI function (28) included in the PAP package. The goodness of fit of hierarchical models were evaluated using likelihood ratio tests, and significance levels were calculated by comparison with 
2-distributions with degrees of freedom equal to the difference in numbers of parameters fitted in each model.
|
ACKNOWLEDGEMENTS |
|---|
This work was supported by an internal grant from the University of Oxford, National Heart, Lung, and Blood Institute grants HL45508 and HL 47910, National Institute of General Medical Sciences Grant GM28356, and National Centre for Research Resources Grant RR03655, and by the French Ministry of Research and the Medical Research Council. We are grateful to Charles Rotimi, Olefemi Ogunbiyi, Simon Anderson and Becky Stack for assistance in family collection and DNA preparation.
|
FOOTNOTES |
|---|
* To whom correspondence should be addressed: Tel: +44 1865287601; Fax: +44 1865287501; Email: martin.farrall{at}well.ox.ac.uk
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
1 Cambien, F., Alhenc-Gelas, F., Herbeth, B., Andre, J., Rakotovao, R., Gonzales, M., Allegrini, J. and Bloch, C. (1988) Familial resemblance of plasma angiotensin-converting enzyme level: the Nancy study. Am. J. Hum. Genet., 43, 774–780.[Web of Science][Medline]
2 McKenzie, C.A., Julier, C., Forrester, T., McFarlane-Anderson, N., Keavney, B., Lathrop, G.M., Ratcliffe, P.J. et al. (1995) Segregation and linkage analysis of serum angiotensin I-converting enzyme levels: evidence for two quantitative-trait loci. Am. J. Hum. Genet., 57, 1426–1435.[Web of Science][Medline]
3
McKenzie, C.A., Abecasis, G.R., Keavney, B., Forrester, T., Ratcliffe, P.J., Julier, C., Connell, J.M.C., Bennett, F., McFarlane-Anderson, N., Lathrop, G.M. and Cardon, L.R. (2001) Trans-ethnic fine mapping of a quantitative trait locus for circulating angiotensin I-converting enzyme (ACE). Hum. Mol. Genet., 10, 1077–1084.
4 Tiret, L., Rigat, B., Visvikis, S., Breda, C., Corvol, P., Cambien, F. and Soubrier, F. (1992) Evidence, from combined segregation and linkage analysis, that a variant of the angiotensin I-converting enzyme (ACE) gene controls plasma ACE levels. Am. J. Hum. Genet., 51, 197–205.[Web of Science][Medline]
5 Zhu, X., Bouzekri, N., Southam, L., Cooper, R.S., Adeyemo, A., McKenzie, C.A., Luke, A., Chen, G., Elston, R.C. and Ward, R. (2001) Linkage and association analysis of angiotensin I-converting enzyme (ACE)-gene polymorphisms with ACE concentration and blood pressure. Am. J. Hum. Genet., 68, 1139–1148.[Web of Science][Medline]
6
Keavney, B., McKenzie, C.A., Connell, J.M.C., Julier, C., Ratcliffe, P.J., Sobel, E., Lathrop, M. and Farrall, M. (1998) Measured haplotype analysis of the angiotensin-1 converting enzyme (ACE) gene. Hum. Mol. Genet., 7, 1745–1751.
7 Farrall, M., Keavney, B., McKenzie, C., Delépine, M., Matsuda, F. and Lathrop, G.M. (1999) Fine-mapping of an ancestral recombination breakpoint in DCP1. Nat. Genet., 23, 270–271.[Web of Science][Medline]
8 Soubrier, F., Martin, S., Alonso, A., Visvikis, S., Tiret, L., Matsuda, F., Lathrop, G.M. and Farrall, M. (2002) High-resolution genetic mapping of the ACE-linked QTL influencing circulating ACE activity. Eur. J. Hum. Genet., 10, 553–561.
9 Reich, D.E., Cargill, M., Bolk, S., Richter, D., Ireland, J., Sabeti, P.C., Lavery, T., Kouyoumjian, R., Farhadian, S.F., Ward, R. and Lander, E.S. (2001) Linkage disequilibrium in the human genome. Nature, 410, 199–204.
10
Gabriel, S.B., Schnaffner, S.F., Nguyen, H., Moore, J.M., Roy, J., Blumenstiel, B., Higgins, J., DeFelice, M., Lochner, A., Faggart, M. et al. (2002) The structure of haplotypes blocks in the human genome. Science, 296, 2225–2229.
11 Rieder, M.J., Taylor, S.L., Clark, A.G. and Nickerson, D.A. (1999) Sequence variation in the human angiotensin converting enzyme. Nat. Genet., 22, 59–62.[Web of Science][Medline]
12 Villard, E., Tiret, L., Visvikis, S., Rakotovao, R., Cambien, F. and Soubrier, F. (1996) Identification of new polymorphisms of the angiotensin I-converting enzyme (ACE) gene, and study of their relationship to plasma ACE levels by two-QTL segregation-linkage analysis. Am. J. Hum. Genet., 58, 1268–1278.[Web of Science][Medline]
13 Zhu, X., McKenzie, C.A., Forrester, T., Nickerson, D.A., Broeckel, U., Schunkert, H., Doering, A., Jacob, H.J., Cooper, R.S. and Rieder, M.J. (2000) Localization of a small genomic region associated with elevated ACE. Am. J. Hum. Genet., 67, 1144–1153.[Web of Science][Medline]
14 Stam, L.F. and Laurie, C.C. (1996) Molecular dissection of a major gene effect on a quantitative trait: the level of alcohol dehydrogenase expression in Drosophila melanogaster. Genetics, 144, 1559–1564.[Abstract]
15
Long, A.D., Lyman, R.F., Langley, C.H. and Mackay, T.F.C. (1998) Two sites in the Delta gene region contribute to naturally occurring variation in bristle number in Drosophila melanogaster. Genetics, 149, 999–1017.
16
Yan, H., Yuan, W., Velculescu, V.E., Vogelstein, B. and Kinzler, K.W. (2002) Allelic variation in human gene expression. Science, 297, 1143.
17 Horikawa, Y., Oda, N., Cox, N.J., Li, X., Orho-Melander, M., Hara, M., Hinokio, Y., Lindner, T.H., Mashima, H., Schwarz, P.E. et al. (2000) Genetic variation in the gene encoding calpain-10 is associated with type 2 diabetes mellitus. Nat. Genet., 26, 163–175.[Web of Science][Medline]
18
Cooper, R., Rotimi, C., Ataman, S., McGee, D., Osotimehin, B., Kadiri, S., Muna, W., Kingue, S., Fraser, H. and Forrester, T. et al. (1997) Hypertension prevalence in seven populations of African origin. Am. J. Public. Health., 87, 160–168.
19 Guo, X., Rotimi, C., Cooper, R., Luke, A., Elston, R., Ward, R. and Ogunbiyi, O. (1999) Evidence of a major gene effect for angiotensinogen among Nigerians. Ann. Hum. Genet., 63, 293–300.[Web of Science][Medline]
20
Rotimi, C.N., Cooper, R.S., Cao, G., Ogunbiyi, O., Ladipo, M., Owoaje, E. and Ward, R. (1999) Maximum-likelihood generalized heritability estimate for blood pressure in Nigerian families. Hypertension, 33, 874–878.
21 Danilov, S., Savoie, F., Lenoir, B., Jeunemaitre, X., Azizi, M., Tarnow, L. and Alhenc-Gelas, F. (1996) Development of enzyme-linked immunoassays for human angiotensin I converting enzyme suitable for large-scale studies. J. Hypertens., 14, 719–727.[Web of Science][Medline]
22 Takahashi, M., Matsuda, F., Margetic, N. and Lathrop, M. (2002) Automated identification of single nucleotide polymorphisms from sequencing data. In Proceedings of the IEEE Computer Society Bioinformatics Conference, 2002, pp. 87--96.
23 Abecasis, G.R., Cherny, S.S., Cookson, W.O. and Cardon, L.R. (2002) Merlin-rapid analysis of dense genetic maps using sparse gene flow trees. Nat. Genet., 30, 97–101.[Web of Science][Medline]
24 Lathrop, G.M., Lalouel, J.M., Julier, C. and Ott, J. (1985) Multilocus linkage analysis in humans: detection of linkage and estimation of recombination. Am. J. Hum. Genet., 37, 482–498.[Web of Science][Medline]
25 O'Connell, J.R. and Weeks, D.E. (1998) PedCheck: a program for identifying genotype incompatibilities in linkage analysis. Am. J. Hum. Genet., 63, 259–266.[Web of Science][Medline]
26 Hasstedt, S.J. (1994) Pedigree Analysis Package revision 4.0. Department of Human Genetics, University of Utah, Salt Lake City.
27 Hasstedt, S.J. (1993) Variance components/major locus likelihood approximation for quantitative, polychotomous and multivariate data. Genet. Epidemiol., 10, 145–158.[Web of Science][Medline]
28 Lalouel, J.M. (1979) GEMINI—a computer program for optimization of a nonlinear function. Department of Medical Biophysics and Computing, University of Utah, Technical Report No. 14.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M.-G. Hong, C. Reynolds, M. Gatz, B. Johansson, J. C. Palmer, H. F. Gu, K. Blennow, P. G. Kehoe, U. de Faire, N. L. Pedersen, et al. Evidence that the gene encoding insulin degrading enzyme influences human lifespan Hum. Mol. Genet., August 1, 2008; 17(15): 2370 - 2378. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Biller, G. Zissel, B. Ruprecht, M. Nauck, A. Busse Grawitz, and J. Muller-Quernheim Genotype-corrected reference values for serum angiotensin-converting enzyme Eur. Respir. J., December 1, 2006; 28(6): 1085 - 1091. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Frere, D.-A. Tregouet, P.-E. Morange, N. Saut, D. Kouassi, I. Juhan-Vague, L. Tiret, and M.-C. Alessi Fine mapping of quantitative trait nucleotides underlying thrombin-activatable fibrinolysis inhibitor antigen levels by a transethnic study Blood, September 1, 2006; 108(5): 1562 - 1568. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F.A. Sayed-Tabatabaei, B.A. Oostra, A. Isaacs, C.M. van Duijn, and J.C.M. Witteman ACE Polymorphisms Circ. Res., May 12, 2006; 98(9): 1123 - 1133. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Catarsi, R. Ravazzolo, F. Emma, D. Fruci, L. Finos, A. Frau, G. Morreale, A. Carrea, and G. M. Ghiggeri Angiotensin-converting enzyme (ACE) haplotypes and cyclosporine A (CsA) response: a model of the complex relationship between ACE quantitative trait locus and pathological phenotypes Hum. Mol. Genet., August 15, 2005; 14(16): 2357 - 2367. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. J. Lehmann, M. Cortina-Borja, D. R. Warden, A. D. Smith, K. Sleegers, J. A. Prince, C. M. van Duijn, and P. G. Kehoe Large Meta-Analysis Establishes the ACE Insertion-Deletion Polymorphism as a Marker of Alzheimer's Disease Am. J. Epidemiol., August 15, 2005; 162(4): 305 - 317. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Lei, I. N. M. Day, and I. Vorechovsky Exonization of AluYa5 in the human ACE gene requires mutations in both 3' and 5' splice sites and is facilitated by a conserved splicing enhancer Nucleic Acids Res., July 14, 2005; 33(12): 3897 - 3906. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Zhang, F. Sun, and H. Zhao HAPLORE: a program for haplotype reconstruction in general pedigrees without recombination Bioinformatics, January 1, 2005; 21(1): 90 - 103. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Katzov, A. M. Bennet, P. Kehoe, B. Wiman, M. Gatz, K. Blennow, B. Lenhard, N. L. Pedersen, U. de Faire, and J. A. Prince A cladistic model of ACE sequence variation with implications for myocardial infarction, Alzheimer disease and obesity Hum. Mol. Genet., November 1, 2004; 13(21): 2647 - 2657. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. F. Gu, S. Efendic, S. Nordman, C.-G. Ostenson, K. Brismar, A. J. Brookes, and J. A. Prince Quantitative Trait Loci Near the Insulin-Degrading Enzyme (IDE) Gene Contribute to Variation in Plasma Insulin Levels Diabetes, August 1, 2004; 53(8): 2137 - 2142. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Farrall Quantitative genetic variation: a post-modern view Hum. Mol. Genet., April 1, 2004; 13(90001): R1 - 7. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. M. Kammerer, N. Gouin, P. B. Samollow, J. F. VandeBerg, J. E. Hixson, S. A. Cole, J. W. MacCluer, and L. D. Atwood Two Quantitative Trait Loci Affect ACE Activities in Mexican-Americans Hypertension, February 1, 2004; 43(2): 466 - 470. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. A. Haiman, S. O. Henderson, P. Bretsky, L. N. Kolonel, and B. E. Henderson Genetic Variation in Angiotensin I-Converting Enzyme (ACE) and Breast Cancer Risk: The Multiethnic Cohort Cancer Res., October 15, 2003; 63(20): 6984 - 6987. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. G Kehoe Review: The renin-angiotensin-aldosterone system and Alzheimer's disease? Journal of Renin-Angiotensin-Aldosterone System, June 1, 2003; 4(2): 80 - 93. [Abstract] [PDF]
|
|
|
|
|
|
X. Zhu, Y.-P. C. Chang, D. Yan, A. Weder, R. Cooper, A. Luke, D. Kan, and A. Chakravarti Associations Between Hypertension and Genes in the Renin-Angiotensin System Hypertension, May 1, 2003; 41(5): 1027 - 1034. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. G. Kehoe, H. Katzov, L. Feuk, A. M. Bennet, B. Johansson, B. Wiman, U. d. Faire, N. J. Cairns, G. K. Wilcock, A. J. Brookes, et al. Haplotypes extending across ACE are associated with Alzheimer's disease Hum. Mol. Genet., April 15, 2003; 12(8): 859 - 867. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. S. Cooper and B. M. Psaty Genomics and Medicine: Distraction, Incremental Progress, or the Dawn of a New Age? Ann Intern Med, April 1, 2003; 138(7): 576 - 580. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||