European Heart Journal (2003) 24, 501–503
Editorial
Well kept secrets of the genome
H. Schunkert*, J. Erdmann
Department of Internal Medicine 2, University of Regensburg, Franz-Josef-Strauss-Allee 11, 93053
Regensburg, Germany
See doi:10.1016/S1095-668X(02)00524-9, for the article to which this editorial refers. individual variability of cardiovascular phenotypes.5,6 In fact, a number of editorialists of this journal discussed the difficulty of identifying the genetic roots of complex cardiovascular traits.7–9
Some of the points raised by these editorialists certainly apply to the present paper as well. For example, the present analysis of intra-pair differences found in hearts of twins lacks a power calculation. Most likely, such evaluation would have shown that the small number of individuals studied would produce negative findings even if the polymorphisms were biologically relevant.9 The adequate consideration of confounding factors is another matter of critical interest. Swan and coworkers corrected for age, sex, blood pressure and weight, which was very reasonable. However, other factors that are more difficult to estimate precisely, e.g. the history of hypertension, detailed measures of body composition, vascular compliance, physical exercise, or growth factors may all affect LV mass in an integrated fashion. In addition, it is important to clarify which model was used for the adjustments. Other caveats that apply to negative association studies in general have been discussed in a previous editorial and include the lack of in vitro and in vivo functionality of polymorphisms under study and too broad definitions of the phenotypes.7
When, lo, as they reached the mountain-side,
A wondrous portal opened wide,
As if a cavern was suddenly hollowed,
And the Piper advanced and the children followed.
Gebru¨der Grimm/Robert Browning
The Pied Piper of Hamelin: A Child's Story (Der
Rattenfa¨nger von Hameln)
Geneticists interested in complex cardiovascular diseases follow the tune that of the risk of suffering from hypertension, left ventricular hypertrophy or myocardial infarction is 30–50% inherited. This melody lives again in the study by Swan et al., published in this issue.1 The authors determine in 110 pairs of twins that 53% of the variability of left ventricular mass is explained by genetic factors. This is reassuring news given that previous studies produced similar or only slightly smaller numbers.2–4 But this clear-cut finding asks for further details. Particularly, individuals genes that cause or facilitate cardiac hypertrophy, are of interest.
However, in the same paper bad news lurks.1
Namely, previously published polymorphisms of genes that, hypothetically, may explain some of the genetic background of LVH were not found to be associated with LV mass. These findings again corroborate, some albeit not all, previously published findings in the field.5,6
What can we learn from the dilemma? Firstly, the prediction of candidate genes is more difficult than expected. Indeed, out of the 32 000–38 000 genes of the human genome 21 000–27 000 appear to affect the cardiovascular system.10 At least hypothetically, many of these genes would make a nice case for a disease-related gene to be tested in an association study. Even more bothersome for the hypothesis-driven investigator is the fact that some of these genes come with 50 or more variants.11
Which of these many genes, gene variants or
Thus, the paper is one of many that fails to shed light on the genes that contribute to the inter-
* Corresponding author. Tel.: +49-941-944-7233; fax: +49-
941-944-7235
E-mail address: heribert.schunkert@klinik.uni-regensburg.de
(H. Schunkert).
0195-668X/03/$ - see front matter © 2003 The European Society of Cardiology. Published by Elsevier Science Ltd. All rights reserved. doi:10.1016/S0195-668X(02)00695-4

502 Editorial
Fig. 1 A multitude of physiological systems and components may dilute the effect of a single variant on a complex phenotype such as LV mass. The diagram highlights hypothetically this fact with respect to the aldosterone-344C/T polymorphism. Other phenotypes such as aldosterone plasma or tissue levels may allow a better estimation of the functional relevance of this polymorphism. However, such measurements did not unequivocally suggest a strong effect of this variant.6 Even more unlikely is a measurable effect on LV mass in a small number of subjects. derived haplotypes is the best for association studies? For true optimists we might add that complex cardiovascular phenotypes are the product of an integrated action of several genes and their variants acting along signalling cascades, e.g. physiological systems such as the sympathetic nervous system. Such physiological systems interact, resulting in an interactome in which many genetic and environmental factors cross talk with goal. each other.8 Such dilution of individual gene effects is hypothesized and schematically depicted in
Looking at this scenario a decade after publication of the first major association study on a complex cardiovascular trait,12 we must admit that nature is more puzzling than we liked to believe.
For the moment it may be more useful to establish reliable and reproducible associations with well defined intermediate phenotypes in well defined individuals, e.g. blood pressure or aldosterone plasma levels, instead of spanning bridges from single gene polymorphisms to complex phenotypes such as LVH or myocardial infarction. Too many of these bridges have collapsed already.
Secondly, we are still lacking optimal instruments for such analyses. The growing linkage disequilibrium map will be of help. Moreover, combination of adequately powered linkage and association studies may provide further hope to trace relevant genes.13 Further technical and biometrical innovations as well as genomic discoveries are awaited to ultimately achieve this For the moment, investigators in the field have to withstand the temptation of follow blind ending
Fig. 1. ‘short cuts’ to the genetic roots of LVH and other complex cardiovascular phenotypes. If we wish to learn the truth about individual genes and their variants only very large and meticulously phenotyped and genotyped samples will provide the power to trace truly positive findings under the given circumstances. Alternatively, an imminent cause–effect relationship between the genetic variant and phenotype is mandatory. The ideal standards have been proposed.14 We have to pay the price, just like the Mayor of Hamelin who should have met the requests of the Pied Piper in the famous fairy tail. Editorial 503 in the funding of genetic studies? Eur Heart J 2002;
23:1071–4.
References
8. Swynghedauw B. Susceptibility-conferring polymorphic genotypes in cardiovascular multifactorial syndromes. Eur
Heart J 2002;23:271–3.
1. Swan L, Birnie DH, Padmanabhan S et al. The genetic determination of left ventricular mass in healthy adults. Eur
Heart J 2003;24:577–82.
2. Mayosi BM, Keavney B, Kardos A et al. Electrocardiographic measures of left ventricular hypertrophy show greater heritability than echocardiographic left ventricular mass: a family study. Eur Heart J 2002;13:1963–71.
3. Schunkert H, Bro¨ckel U, Hengstenberg C et al. Familial predisposition of left ventricular hypertrophy. J Am Coll
Cardiol 1999;33:1685–91.
4. Post WS, Larson MG, Myers RH et al. Heritability of left ventricular mass: the Framingham Heart Study. Hypertension 1997;30:1025–8.
5. Sedla´cek K, Fischer M, Erdmann J et al. Relation of the GNB3 polymorphism with left ventricular structure and function.
Hypertension 2002;40:162–7.
9. Schunkert H, Fischer M. Old and simple tools may do better—sometimes. Eur Heart J 2002;23:1900–2.
10. Dempsey AA, Dzau VJ, Liew CC. Cardiovascular genomics: estimating the total number of genes expressed in the human cardiovascular system. J Mol Cell Cardiol 2001;
33:1879–86.
11. Rieder MJ, Taylor SL, Clark AG et al. Sequence variation in the human angiotensin converting enzyme. Nat Genet 1999;
22:59–62.
12. Cambien F, Poirier O, Lecerf L et al. Deletion polymorphism in the gene for angiotensin-converting enzyme is a potent risk factor for myocardial infarction. Nature 1992;
359:641–4.
13. Van Eerdewegh P, Little RD, Dupuis J et al. Association of the ADAM33 gene with asthma and bronchial hyperresponsiveness. Nature 2002;25(418):426–30.
14. Editorial. Freely associating. Nat Genet 1999;22:1–2.
6. Schunkert H, Hengstenberg C, Holmer SR et al. Lack of association between a polymorphism of the aldosterone synthase gene and left ventricular structure. Circulation
1999;99:2255–60.
7. Jones A, Montgomery H. The Gly389Arg beta-1 adrenoceptor polymorphism and cardiovascular disease: time for a rethink