Genetic associations with lipoprotein subfractions provide information on their biological nature

Ann Kristin Petersen, Klaus Stark, Muntaser D. Musameh, Christopher P. Nelson, Werner Römisch-Margl, Werner Kremer, Johannes Raffler, Susanne Krug, Thomas Skurk, Manuela J. Rist, Hannelore Daniel, Hans Hauner, Jerzy Adamski, Maciej Tomaszewski, Angela Döring, Annette Peters, H. Erich Wichmann, Bernhard M. Kaess, Hans Robert Kalbitzer, Fritz HuberVolker Pfahlert, Nilesh J. Samani, Florian Kronenberg, Hans Dieplinger, Thomas Illig, Christian Hengstenberg, Karsten Suhre, Christian Gieger, Gabi Kastenmüller

Research output: Contribution to journalArticle

24 Citations (Scopus)


Adverse levels of lipoproteins are highly heritable and constitute risk factors for cardiovascular outcomes. Hitherto, genome-wide association studies revealed 95 lipid-associated loci. However, due to the small effect sizes of these associations large sample numbers (>100 000 samples) were needed. Here we show that analyzing more refined lipid phenotypes, namely lipoprotein subfractions, can increase the number of significantly associated loci compared with bulk high-density lipoprotein and low-density lipoprotein analysis in a study with identical sample numbers. Moreover, lipoprotein subfractions provide novel insight into the human lipid metabolism. We measured 15 lipoprotein subfractions (L1-L15) in 1791 samples using 1H-NMR (nuclear magnetic resonance) spectroscopy. Using cluster analyses, we quantified inter-relationships among lipoprotein subfractions. Additionally, we analyzed associations with subfractions at known lipid loci. We identified five distinct groups of subfractions: one (L1) was only marginally captured by serum lipids and therefore extends our knowledge of lipoprotein biochemistry. During a lipid-tolerance test, L1 lost its special position. In the association analysis, we found that eight loci (LIPC, CETP, PLTP, FADS1-2-3, SORT1, GCKR, APOB, APOA1) were associated with the subfractions, whereas only four loci (CETP, SORT1, GCKR, APOA1) were associated with serum lipids. For LIPC, we observed a 10-fold increase in the variance explained by our regression models. In conclusion, NMR-based fine mapping of lipoprotein subfractions provides novel information on their biological nature and strengthens the associations with genetic loci. Future clinical studies are now needed to investigate their biomedical relevance.

Original languageEnglish
Article numberddr580
Pages (from-to)1433-1443
Number of pages11
JournalHuman Molecular Genetics
Issue number6
Publication statusPublished - Mar 2012
Externally publishedYes


ASJC Scopus subject areas

  • Molecular Biology
  • Genetics
  • Genetics(clinical)

Cite this

Petersen, A. K., Stark, K., Musameh, M. D., Nelson, C. P., Römisch-Margl, W., Kremer, W., Raffler, J., Krug, S., Skurk, T., Rist, M. J., Daniel, H., Hauner, H., Adamski, J., Tomaszewski, M., Döring, A., Peters, A., Wichmann, H. E., Kaess, B. M., Kalbitzer, H. R., ... Kastenmüller, G. (2012). Genetic associations with lipoprotein subfractions provide information on their biological nature. Human Molecular Genetics, 21(6), 1433-1443. [ddr580].