Similarities and Differences in Lipidomics Profiles among Healthy Monozygotic Twin Pairs

Abstract

Differences in genetic background and/or environmental exposure among individuals are expected to give rise to differences in measurable characteristics, or phenotypes. Consequently, genetic resemblance and similarities in environment should manifest as similarities in phenotypes. The metabolome reflects many of the system properties, and is therefore an important part of the phenotype. Nevertheless, it has not yet been examined to what extent individuals sharing part of their genome and/or environment indeed have similar metabolomes. Here we present the results of hierarchical clustering of blood plasma lipid profile data obtained by liquid chromatography-mass spectrometry from 23 healthy, 18-year-old twin pairs, of which 21 pairs were monozygotic, and 8 of their siblings. For 13 monozygotic twin pairs, within-pair similarities in relative concentrations of the detected lipids were indeed larger than the similarities with any other study participant. We demonstrate such high coclustering to be unexpected on basis of chance. The similarities between dizygotic twins and between nontwin siblings, as well as between nonfamilial participants, were less pronounced. In a number of twin pairs, within-pair dissimilarity of lipid profiles positively correlated with increased blood plasma concentrations of C-reactive protein in one twin. In conclusion, this study demonstrates that in healthy individuals, the individual genetic background contributes to the blood plasma lipid profile. Furthermore, lipid profiling may prove useful in monitoring health status, for example, in the context of personalized medicine.

Hierarchical clustering analysis of blood plasma lipidomics profiles from mono- and dizygotic twin families

“…Furthermore, because HCA is an exploratory data analysis technique, in contrast to SEM it allows for the discovery of novel biological effects causing heterogeneity among study participants. As an example of the latter, in an earlier publication 14 we demonstrated that in HCA of blood plasma lipidomics data obtained in a sample consisting mainly of MZ twin pairs, male and female study participants were separated at the highest level in the clustering dendrogram. This suggested that variation in lipidomics profiles is relatively small among individuals of the same gender.…”

“…17 Characterization of participants, collection of fasting blood and urine samples, and sample preparation were performed as described previously. 14,16,18 Participants completed a number of surveys; for the current study, we used answers to questions regarding current use of any medication, recent subjective health, current and earlier smoking habits and whether participants currently lived together. Female participants reported the day of their menstrual cycle at the time of blood sampling.…”

“…14,16 In brief, lipidomics profiling was performed using an LC-MS method targeted at the analysis of lipids that become positively charged during ionization in the mass spectrometer. These measurements were carried out in two consecutive 'blocks' , denoted as B1 and B2, respectively.…”

“…This number of separating nodes is representative of the relative proximity of individuals according to their lipid profiles, and hence acknowledges that the same Euclidean distance may have a different interpretation in sparsely and densely populated areas of the total multivariate space put up by the lipid profiles of all study participants. 14 For each possible number of nodes separating MZ or DZ cotwins or samesex nontwin siblings in the dendrogram, we compared the observed number of cotwin or sibling pairs separated by that number of nodes with the number of observations that was expected on the basis of chance. Chance distributions were created by permutation of the object labels over the leaves of the clustering dendrogram.…”

“…Lipidomics, that is, the analysis of lipids with metabolomics techniques, is an important branch of metabolomics because lipids are involved in a plethora of (patho) physiological processes. 15 For the current study, we combined the data that provided the basis for our previous article 14 with additional data mainly from DZ twin pairs and from biological nontwin siblings. The initial data and the additional data were acquired with the same LC-MS method but in different measurement 'blocks' , that is, in different sets of measurements consisting of one or more measurement batches.…”