A systems biology approach for the study of smoking and myocardial infarction

Metabolomics has been proven to be a powerful tool to study complex
phenotypes. It can present a snapshot of the current status of
metabolism and provide a functional readout of the gene products.
Complementing with other ‘omics’ techniques in systems biology
studies, the integration of metabolomics with other ‘omics’, e.g.
trancriptomics and epigenomics, will help to illustrate complex
biological processes which are related to disease and environmental
exposure. This thesis presents three studies focusing on a
lifestyle related environmental exposure—smoking and a disease
which is related to the exposure—myocardial infarction (MI). The
general aim of these studies is to establish links between smoking,
intermediated biomarkers of disturbed metabolic pathways and MI.
Establishment of how they are linked might enlarge our knowledge
about the metabolic basis of these links. The first study presented
in this thesis aims to understand the effects of smoking and
smoking cessation on human serum metabolite profile. Whilst smoking
increases the risks of many diseases, including MI, the benefits of
cessation is remarkable as it has shown to reduce the risk of MI in
a very short time frame. The results presented in this thesis
showed significant differences in metabolite profiles between
current smokers, former smokers and never smokers. Amongst the 21
metabolites, which were found to be different between current
smokers and never smokers, 19 were found reversible in former
smokers. The results were furthermore confirmed in the prospective
study of KORA S4->F4. Network analysis was applied to integrate
smoking related genes and metabolites, which consistently showed
the reversibility of the smoking effects on gene expression and
metabolite profile. The reversibility of smoking related changes in
serum metabolites also coincide with the reduced risk of MI, which
gives rise to the possibility of using these metabolites as
potential biomarkers to characterize smoking related diseases.
Inspired by the first study, two other studies were initiated with
different aims. The second study in this thesis aims to use
multi-level ‘omics’ data to illustrate how smoking influences the
metabolite profiles by alteration in DNA methylation and gene
expression. Candidate biomarkers of smoking were first discovered
separately in epigenomic, transcriptomic and metabolomic levels.
Mediation analyzes were applied to assess the potential
interactions between smoking, DNA methylation, gene expression and
metabolites. In general, seven CpG sites showed significant
mediation effects for the expression of the LRRN3 gene. Amongst
these seven, two were also significantly associated with the
concentrations of LPC (18:2) and PC ae C34:3. In the third study,
three metabolites (arginine, LPC (17:0) and LPC (18:2)), which may
serve as novel biomarkers for incident MI, were identified based on
a targeted metabolomics approach in two prospective cohort studies.
These metabolites significantly associated with MI in Cox
regression models after adjustment for other MI risk factors, such
as smoking and C-reactive protein (CRP). Inclusion of these
metabolites in the established MI prediction models provided
significant added predictive value. Additionally, the observation
that these metabolites were associated with CRP indicates potential
inflammatory process they are commonly involved in. Among the three
metabolites listed above, arginine and LPC (18:2) are also
associated with smoking as shown in the first study of this thesis,
which implies the underlying metabolic relationships between
smoking and MI. In summary, this doctoral thesis reveals
metabolites associated with smoking and MI. Using a systems biology
approach, the effects of smoking on DNA methylation and gene
expression, which mediates the corresponding variations on
metabolite concentrations, were analyzed by integrating multi-level
‘omics’ data. The metabolites associated with both smoking and MI
may contribute to a deeper insight into the molecular basis between
the link of MI and its risk factor—smoking.