| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Endocrine Care |
Department of Physiology, The University of Melbourne, Victoria 3010, Australia
Address all correspondence and requests for reprints to: Professor Stephen B. Harrap, Department of Physiology, The University of Melbourne, Victoria 3010, Australia. E-mail: s.harrap{at}physiology.unimelb.edu.au
Abstract
The determination of human adult height is dependent on both environmental and genetic factors. Rare causes of abnormal stature have been identified, including mutations in the gene encoding aromatase (CYP19) and regions on the Y chromosome. However, the possible role of these loci in the genetic control of normal adult height is unknown. We have performed an association study using common biallelic polymorphisms within CYP19 and the Y chromosome to determine whether these loci are associated with variation in height in 413 adult males and 335 females drawn at random from a large population sample. An association between CYP19 and height was found (difference, 2.0 cm; 95% confidence interval, 0.16–3.8; P = 0.003), but this was more evident in men (difference, 2.3 cm; 95% confidence interval, 0.38–4.4; P = 0.05) than women (difference, 0.2 cm; 95% confidence interval, -2.1 to 1.6; P = 0.94). An association was also found with the Y chromosome (P = 0.009; difference of 1.9 cm; 95% confidence interval, 0.5–3.4). Additionally, when men were grouped according to haplotypes of the CYP19 and Y chromosome polymorphisms, a difference of 4.2 cm (95% confidence interval, 0.67–7.3) was detected (P = 0.004). These results suggest that in men, genetic variation in CYP19 and on the Y chromosome are involved in determining normal adult height, and that these loci may interact in an additive fashion.
DETERMINATION OF HUMAN adult height depends on a complex interaction between genetic and environmental factors that influence growth at key times of life. The evidence for environmental influences comes from the increases in population average height over generations (1). Genetic influence is evident in twin and family studies, where the correlation between monozygotic twins is 0.9 and the heritability of height is 55% (2). In physiological terms, height depends on the balance and timing of mechanisms that stimulate or limit linear growth. Growth during puberty accounts for up to 18% of final adult height (3). Closure of the epiphyses of the long bones coincides with cessation of linear growth. Sex hormones are important both for adolescence and in relation to epiphyseal closure (4). The importance of sex chromosomes is evident from the sexual dimorphism of the timing of the adolescent growth spurt and eventual adult heights, such that girls begin their growth spurt early but women are, on average, shorter than men (3). The difference in height between men and women is, to a large extent, explained by the longer legs in men, given that the torso heights are not greatly different (5).
In recent years, the understanding of these biological events has accelerated greatly, and it is now clear that estrogen regulates growth and maturation of bone in both sexes, in relation to the timing of epiphyseal closure and coincident cessation of longitudinal bone growth (4). The relevant constitutional factors are genes that control the formation and action of the sex hormones, such as key enzymes and receptors. Genes on the Y chromosome are also candidates because of the sexual dimorphism of height.
Certain genes have been associated with height determination, based on clinical syndromes that are the result of major mutations (1, 4, 6). Mutations in the aromatase gene (CYP19) that converts androgens to estrogens can result in estrogen deficiency, and estrogen resistance can be caused by mutations of the estrogen receptor genes. In both these conditions, males and females exhibit tall stature, probably as a consequence of continued longitudinal bone growth into adulthood. From the case study literature, it would seem that the resultant excess linear growth might be more obvious in males (4, 7, 8).
Both of the sex chromosomes have been implicated in height determination. For example, subjects with Turner syndrome who lack one X chromosome (45,X) are typically short (1). In men, the Y chromosome is important for height (9). Short stature has been demonstrated in persons with deletions in the sex chromosome pseudoautosomal region that includes a homeobox-containing gene designated SHOX (1). Also, deletions in the nonrecombining region of Yq, which result in reduced adult height, suggest another stature locus, which has been mapped to a region just distal to the Y chromosome centromere (between DYS11 at interval 5C and DYS246 at interval 5D; refer to Ref. 10) (6). The candidate gene in this region has not been characterized.
Given the role (in rare conditions) of the aromatase gene, CYP19, and the Y chromosome, it is possible that less dramatic mutations or chromosomal disruptions might have more subtle (but perhaps more frequent) effects on height in otherwise healthy individuals. To determine the potential role of CYP19 and the Y chromosome in the population variation in height, we performed an association study in 748 Caucasian adults (335 females, 413 males) from a representative sample of the Victorian population. We used a previously described biallelic single-nucleotide polymorphism (SNP) in exon 3 of CYP19 (11) and a restriction fragment-length polymorphism (RFLP) in the nonrecombining centromeric region of the Y chromosome (12). The relationship between alleles and height was determined for each locus separately, and a combined haplotype analysis was undertaken to test for interaction between the 2 loci.
Materials and Methods
Subject recruitment and phenotype measurement
Subjects were drawn from the Victorian Family Heart Study (VFHS), which is a population-based study of cardiovascular risk. A total of 2959 healthy Caucasians were recruited between 1991 and 1996. This group comprised 783 families consisting of 2 parents (40–70 yr old) and at least 1 natural offspring (18–30 yr old). Recruitment was limited to Caucasian families. A family history of heart disease was not relevant to recruitment, the aim being to enroll a representative sample of Caucasian families exhibiting a broad cross-section of cardiovascular risk factor levels (2).
These studies were approved by the Ethics Review Committee of the Alfred Hospital, Melbourne, and informed consent was obtained from all participants. Height was measured to the nearest 0.5 cm, with subjects in stocking feet, using a Holtain stadiometer. The research nurses were carefully trained in standardized measurement techniques, and the instrument was calibrated at regular intervals. A blood sample was taken for extraction of DNA (2). For this study, we studied 350 mothers and 350 fathers from the VFHS, who comprised the parental generation of a core set of families. These families were selected because they had 2 or more offspring in the VFHS and, therefore, were suitable for sibling pair linkage analyses. The individuals from these families were representative, in all respects, of the entire VFHS sample of 783 families. In addition, we included an extra 70 males from the parental generation of the VFHS who were not part of the core set for linkage analyses. These men had been genotyped for the Y chromosome as part of a previous study of blood pressure variation (13). They had been selected at random and were representative of their peers in the VFHS. Genotypes were unsuccessful in some subjects and were excluded. This paper reports the results from the remaining 413 males and 335 females with successful genotypes (see below).
Analysis of the CYP19 SNP
Males and females from the parental generation of the VFHS were surveyed for the CYP19 single SNP located in exon 3. The SNP was detected by amplification of a 210-bp fragment of exon 3 using forward primer 5'-CTAAGATGTTGCTTATGCTC-3' and reverse primer 5'-GTCTTCGATTATGAACAGAC-3' (11). Approximately 50 ng DNA from each individual was added to a mix containing 0.5 µM of each primer, 1x PCR buffer (Perkin-Elmer Applied Biosystems, Norwalk, CT), 250 µM deoxynucleotide triphosphate (Perkin-Elmer Applied Biosystems), 1.5 mM MgCl2 (Perkin-Elmer Applied Biosystems), and 1 U Amplitaq Gold DNA polymerase (Perkin-Elmer Applied Biosystems), to give a total reaction vol of 20 µl. Thermal conditions required for the reaction were 95 C for 10 min (for activation of the Amplitaq Gold enzyme), followed by 35 cycles of 95 C for 1 min, 50 C for 1 min, and 72 C for 1 min, followed by a final extension time of 72 C for 10 min.
All PCR products were electrophoresed through agarose gel, and PCR product was extracted from the gel using the Qiaquick Gel Extraction kit (QIAGEN, Hilden, Germany) according to manufacturer’s recommended protocol. Concentration of the purified PCR product was determined using a spectrophotometer. Cycle sequencing of the PCR product was achieved using the BigDye terminator cycle sequencing kit (Perkin-Elmer Applied Biosystems) and sequence determined using an ABI 377 DNA sequencer. Individual genotypes of the CYP19 SNP were determined directly from the sequence. Two alleles of the CYP19 gene were detected, representing the presence of either an A or G nucleotide at the polymorphic site, and are referred to here as A and B alleles, respectively. This resulted in three genotypes: AA, AB, and BB.
Y chromosome RFLP
Alleles of the Y chromosome RFLP, located in the nonrecombining region, were determined in males from the parental generation of the VFHS, as previously described (12, 13). Briefly, DNA from each individual was amplified by PCR using primers flanking an alphoid satellite region close to the centromere, which contains a HindIII RFLP. Two alleles of the Y chromosome RFLP were detected in males, corresponding to the absence or presence of the restriction enzyme cut site, referred to here as a and b, respectively.
Statistical analyses
Phenotypic data are summarized as mean and SD of unadjusted data. Where relevant, the 95% confidence interval (CI) for the difference between means is also given. Differences in height between genotype and haplotype groups were compared using ANOVA, which adjusted within the analyses for the effects of sex where appropriate. Statistical analyses were performed using the SPSS statistical software package (Macintosh version 6.1; SPSS, Inc., Chicago, IL).
Results
For all 748 subjects, the mean age was 53.4 yr (SD, 5.5), the mean height was 168.8 cm (SD, 8.9), mean weight was 76.1 kg (SD, 14.1), and mean body mass index was 26.6 kg/m2 (SD, 4.1). There was no correlation between age and height, other than accounted for by sex differences. Females (n = 335) were younger (51.7 yr; SD, 5.0) than men (n = 413, 54.7 yr; SD, 5.5; P < 0.0001). Men (174.1 cm, SD, 7.0) were significantly (P < 0.0001) taller than women (162.3 cm; SD, 6.2), by 11.8 cm.
Association of CYP19 with height
Results were obtained from 334 females and 394 males. In the
entire sample, we found a significant difference in height between the
3 genotype groups of the CYP19 SNP (P =
0.003). Analyses for each sex revealed that the association between the
CYP19 genotypes and height was obvious in men
(P = 0.05) but not women (P = 0.94)
(Table 1
).
|
The results of the Y chromosome RFLP were obtained successfully from 409 of the 413 males. A significant difference of 1.9 cm (95% CI, 0.5–3.4) in male height was detected between the 2 genotypic groups (allele a, 174.7 cm; SD, 6.6 vs. allele b, 172.8 cm; SD, 7.7) defined by the Y chromosome RFLP (P = 0.009).
Association of CYP19 and Y chromosome polymorphism haplotypes with male height
Both the CYP19 SNP and the Y chromosome RFLP alleles were available in a total of 390 males, in whom 6 combined haplotype groups (AAa, AAb, ABa, ABb, BBa, BBb) were analyzed. The lowest average height was observed in the 33 men with the AAb combination, and the tallest men were those carrying the BBa combination. The difference in average height between these 2 extremes was 4.2 cm (95% CI, 0.67–7.3; P = 0.004).
For individuals with the a allele of the Y chromosome, height increased
in proportion to the number of B alleles at the CYP19 locus,
with a difference of 3.5 cm (95% CI, 1.3–5.6) between AA and BB
groups (P = 0.008). The height for those with the AB
genotype was intermediate, suggesting that the effects of the B allele
were additive. In contrast, in individuals with the b allele of the Y
chromosome, the overall difference in height between the AA and BB
groups was 0.5 cm (95% CI, -4.4 to 3.6; P = 0.97). In
subjects carrying the b allele, the same heights observed for AA and AB
would be consistent with a dominant effect of the A allele at the
CYP19 locus. Furthermore, the greater height associated with
the Y chromosome a genotype was evident for each of the
CYP19 genotypes. This effect was most obvious in the BB
group, where the effect of the Y chromosome genotype was 3.7 cm (95%
CI, 1.0–6.5; P = 0.007), and least obvious for the AA
group, where the difference was only 0.6 cm (95% CI, -2.4 to 3.8;
P = 0.69) (Table 2).
|
This is the first study to reveal specific genetic loci that are associated with significant differences in adult height in the general population. The effects of these genes seem to be largely limited to men. Moreover, the genetic markers are reasonably common, and the effects are relatively large. The difference in height associated with combined simple polymorphisms in the aromatase gene and the Y chromosome amounts to 4.2 cm in men. The genotypes associated with greater height for the CYP19 gene and the Y chromosome were present in 23.4% and 69.0%, respectively, of men in our population sample. The haplotype associated with the greatest height is found in 15.4% of men, and the haplotype associated with the shortest stature is found in 17.9% of men. Although major mutations or disruptions in both the CYP19 gene and the Y chromosome have been associated previously with large effects on height (1, 4, 6, 7, 8), our findings would suggest that other common variations in these two genetic regions exert less dramatic (but significant and additive) effects on adult height.
In the case of CYP19, the SNP does not have any known functional consequences per se. Instead, it is presumably in linkage disequilibrium with genetic variants that influence height. By analogy with major mutations in CYP19 that severely reduce the activity of aromatase and result in estrogen deficiency, more common variants may have quantitative effects on gene transcription, posttranslational processing, or the amino acid sequence. These actions may alter the aromatase concentration or activity and change estrogen concentrations in a systemic or possibly developmental stage- or tissue-specific manner and affect linear bone growth and the closure of epiphyses (4, 7, 8). Variation in aromatase activity may also be relevant to the interaction between estrogens and regulators of long-bone growth, including GH and somatomedin (14). No data are available from this or other studies regarding aromatase activity or hormonal profiles in relation to the CYP19 SNP associated with adult height. To further examine this hypothesis, it will be necessary to perform physiological studies complemented by molecular mutation detection analyses to localize and characterize the DNA variants that influence aromatase and height.
Our data also identify an important role of the Y chromosome in height determination. Again, the causative DNA variants are presumably in linkage disequilibrium with our Y chromosome polymorphism. However, the genetic implications are different from the CYP19 case. Specifically, the localization and identification of etiologic variants on the Y chromosome is difficult because the polymorphic marker is in linkage disequilibrium with the entire nonrecombining region (the majority of the chromosome) and small adjacent stretches of the two pseudoautosomal regions on either end of the chromosome. However, clues to the location of the variants come from Y chromosome regions known to cause short stature when deleted. In this respect, two loci have been determined on the Y chromosome: one residing in the pseudoautosomal region on Yp (1), and one in the nonrecombining region of Yq (6). The Yq region is perhaps more likely because our RFLP is also in the nonrecombining region. The locus in Yp has an X chromosome homologue and, therefore, is less likely to explain the observed sexually dimorphic effect. The putative height locus in the Yq nonrecombining region has been designated as the GCY locus (growth control Y), but the gene has yet to be identified (6).
An intriguing aspect of this study is the observation that the genetic associations between CYP19 and height are found only in men. The explanation is not immediately obvious. The physiology of pubertal growth in males differs from females in that it begins and finishes later and has a higher peak velocity. Men are, on average, taller than women, mostly as a result of longer legs rather than longer torsos (5). These features raise the possibility that the aromatase interacts with the other mechanisms controlling growth of the long bones, in particular. Such an interaction may depend on the prevailing hormonal milieu in men and women during growth. For example, on the predominant androgenic background of males, putative CYP19-dependent variation in estrogen levels may be of greater physiological impact than in females with a predominantly estrogenic milieu. Nevertheless, our Y chromosome findings indicate that not all men are created equal, and the effects of DNA variation in or around CYP19 are almost exactly additive to those of the variation in the Y chromosome.
In conclusion, we have identified two genetic loci that seem to be involved in the normal variation of adult height in males. Together, population variation in the aromatase gene and a stature locus on the Y chromosome seem capable of determining male Caucasian height differences of approximately 4 cm. Further study is required to identify the exact genetic variants at these loci and the physiological mechanisms that are responsible for these effects.
Acknowledgments
We thank Dr. John Hopper (Australian National Health and Medical Research Council Twin Registry), Dr. Graham Giles (Collaborative Cohort Study, Health 2000), the general practitioners and research nurses for their contributions to subject recruitment, Dr. Zilla Wong for management of blood samples, and Angela Lamantia for her technical assistance and DNA extraction.
Footnotes
This work was supported in part by the Victorian Health Promotion Foundation and the National Heart Foundation of Australia.
CI, Confidence interval; RFLP, restriction fragment length polymorphism; SNP, single nucleotide polymorphism; VFHS, Victorian Family Heart Study
Received December 17, 2000.
Accepted May 9, 2001.
References
This article has been cited by other articles:
 |
|
 |
|
  J. D. Veldhuis, J. N. Roemmich, E. J. Richmond, and C. Y. Bowers Somatotropic and Gonadotropic Axes Linkages in Infancy, Childhood, and the Puberty-Adult Transition Endocr. Rev., April 1, 2006; 27(2): 101 - 140. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. L. Tosi, B. D. Boyan, and A. L. Boskey Does Sex Matter in Musculoskeletal Health? The Influence of Sex and Gender on Musculoskeletal Health J. Bone Joint Surg. Am., July 1, 2005; 87(7): 1631 - 1647. [Full Text] [PDF]
|
|
|
|
 |
|
 J. D. Veldhuis, A. Bae, R. S. Swerdloff, A. Iranmanesh, and C. Wang Experimentally Induced Androgen Depletion Accentuates Ethnicity-Related Contrasts in Luteinizing Hormone Secretion in Asian and Caucasian Men J. Clin. Endocrinol. Metab., March 1, 2005; 90(3): 1632 - 1638. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. D. Veldhuis, J. N. Roemmich, E. J. Richmond, A. D. Rogol, J. C. Lovejoy, M. Sheffield-Moore, N. Mauras, and C. Y. Bowers Endocrine Control of Body Composition in Infancy, Childhood, and Puberty Endocr. Rev., February 1, 2005; 26(1): 114 - 146. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
