CHAPTER 10 212 10 analysis from our group into lipid levels of 2,800 migraine patients and 7,353 controls using a 1H-NMR metabolomics platform did show an indication for alterations in HDL metabolism.9 Another, possibly less eloborate study looking at both the phenotypic and the genotypic associations independently between lipoproteins and migraine in only women with data from the Women’s Genome Health study (over 22,000 subjects) found the association with triglyceride-rich lipoproteins (TRLP) subfractions to be more evident than LDL or HDL.11 The authors from the USA claim that their large sample size, detailed information on migraine subtypes, consideration of important confounders including menopause status and postmenopausal hormone therapy in the phenotypic analysis may be a factor explaining the differences. All in all the studies support the view that it may be possible to use lipoprotein measures in blood to predict migraine status. As shown in Chapter 2 it was indeed possible to produce a predictive metabolite profile that was largely dependent on lipids. Taking the results of the different studies together suggests that a limited number of lipids do play a role in migraine pathophysiology, but that there are many factors to take into account in the study design before meaningful conclusions can be drawn at the individual lipoprotein compound level. Regarding causality of these lipid metabolites and migraine the data are also contradictory, where one study,11 reports no causality between migraine and lipoprotein subfractions, another study found a causally protective role for a longer length of fatty acids against migraine as well as a causal role of a higher level of a lysophosphatidylethanolamine on migraine.7 The second most associated type of metabolite in the above mentioned Australian study7 was amino acids. A genome-wide genetic overlap was seen for higher levels of phenylalanine, isoleucine and gamma-glutamylisoleucine and lower levels of proline betaine and 2-hydroxyisobutyric acid in migraineurs. Amino acids are a subgroup of amines, which were also investigated in Chapter 3. Amino acids have been previously investigated in several biochemical studies in migraine,12, 13 and one particular amino acid, namely glutamate/glutamine has been repeatedly linked to migraine.14-16 In Chapter 2 our study into the metabolomics profile of migraine patients did show that multiple amino acids were part of our prediction model able to predict active migraine status, namely: leucine, isoleucine, proline, methionine, valine and serine. Noteworthy is the overlap found for isoleucine with Chapter 2, but also with phenylalanine in Chapter 3. In Chapter 3 six amine plasma/CFS ratios were found to be different in migraineurs compared to controls. In two of the ratios phenylalanine was present and the overall ratio correlation between and was different between individuals with migraine compared to controls. Whether the metabolomic ratio changes are the consequence of amino acid changes in the brain itself or malfunctioning of the blood brain transporter that co-regulates concentrations inside and outside the brain still needs to be investigated. Taken together, there is ample evidence that lipids and amino acids and their related biomolecules remain interesting targets in migraine research. In addition it is demonstrated that when
RkJQdWJsaXNoZXIy MTk4NDMw