It is well known that short parents have short children and vice versa, and that variation in normal stature has a strong genetic component.  However, despite many decades of interest in the genetics of stature, the relevant genes remain elusive.  In fact, the genetics of most common traits and diseases in humans is not well understood.  The principal explanation is that the geneticist’s primary tool for mapping genes is of only limited power for finding genes that have modest effects, such as those that contribute to common diseases and variable traits such as stature.  Recent advances in genomics, however, have made it feasible to apply genome-wide linkage analysis to such entities.  Indeed, the group led by Eric Lander has used this approach to identify genetic linkage for adult height.¹  

In total, 2,327 individuals from 483 families were studied.  Fifty-eight families resided in the Botnia region of Finland, 183 families were from other areas of Finland, 179 families were from southern Sweden and 63 families were from the Saguenay-Lac-St-Jean region of Quebec.  They were originally ascertained to investigate other genetic traits.  Males were older than 23.5 years and females older than 21.1 years to exclude individuals still growing.  The original genotyping results that were based on average spacing of microsatellite markers from 6.5 cM to 12.5 cM depending on the study population, were reanalyzed using the variance-components method implemented in the GENE-HUNTER 2 protocol.  The method uses nonparametric multipoint approaches to generate LOD scores for chromosomal locations that reflect the likelihood that genotype data being observed is due to linkage relative to the absence of linkage.

Evidence for linkage was detected in four instances.  A LOD score of 3.85 was obtained for linkage at chromosome 6q24-25 in Botnia.  A score of 3.40 was calculated for a marker located at 7q31.3-36 in Sweden.  A LOD score of 3.35 was determined for markers at 12p11.2-q14 in Finland and a score of 3.56 was found in Finland for 13q32-33.  The authors note that a companion study also detected linkage at chromosome 7 site.²

The authors are optimistic that they have identified chromosomal regions where genes that influence stature reside, especially on chromosome 7.  However, they caution that definitive interpretation is difficult in the absence of confirmation of linkage in additional populations.  They observe their results were inconsistent across the four study groups, but note that this is typical in linkage studies of common diseases.  They discuss possible reasons for the inconsistency including variation in sampling, existence of genetic variation in different populations and statistical fluctuations and false positives due to unknown causes.

Editor’s Comment: An additional comment is pertinent to this topic.  Many genes known to influence stature have been identified by searching for disease genes.  Examples include genes that harbor mutations that cause chondrodysplasias and many other syndromes associated with short stature.  They range from homeobox-containing genes such as SHOX to cartilage matrix protein genes, i.e., COL2A1 to transcription factor and receptor genes such as SOX9 and FGFR3, respectively.  Similarly, mutations of Fibrillin 1 lead to tall stature in the Marfan syndrome.  It seems likely that there are genes that influence stature that are not associated with disease.  The approach used here should identify genes in both categories.  It will be interesting to see what genes fall into the latter category.

These papers are the first reported genome-wide studies of genetic linkage and stature.  They probably represent the tip of the iceberg in terms of what will come as genetic markers become more dense, more populations are studied and analytical approaches become more sophisticated.  As noted by Hirschhorn et al, identifying the genetic basis of variation in height raises important ethical issues as the potential for genetic engineering evolves. However, as they point out, a greater understanding of this subject could be beneficial in the contexts of establishing diagnoses and predicting adult stature of “short” children.

William Horton, MD

Print version (pdf) |Printer-Friendly