Brain-derived neurotrophic factor Val66Met polymorphism in depression and thrombosis: SIRT1 as a possible mediator

This editorial refers to ‘BDNFVal66met polymorphism: a potential bridge between depression and thrombosis’†, by P. Amadio et al., on page 1426. Neurotrophins are a class of growth factors promoting neuronal proliferation, differentiation, and survival. Brain-derived neurotrophic factor (BDNF) is the most abundant neurotrophin and affects the function of both the central and peripheral nervous systems. It is highly expressed in the central nervous system, but it is also present in other tissues such as the myocardium, platelets, kidneys, and spleen, as well as in soluble form in plasma. Interestingly, reduced BDNF activity, resulting from a single nucleotide polymorphism (SNP) in the coding region of the BDNF gene (Val66Met; rs6265), has been associated with neuropsychiatric disorders in humans such as depression.1 Intriguingly, the phenotype of these disorders could be confirmed in transgenic mice carrying the BDNF Val66Met mutation.2 BDNF also assumes important functions in the cardiovascular system. Bdnf-deficient mice display an impaired development of endothelial cells, culminating in endothelial cell apoptosis, impaired cardiac function, and early post-natal death.3 Likewise, studies in genetically engineered mice showed that deletion of Bdnf in neurons or of its receptor, tropomyosin-related kinase B (TrkB), in the heart deteriorated left ventricular function following myocardial infarction, a phenotype that could be rescued by intraperitoneal injection of recombinant BDNF.4 Another study proposed that BDNF may also affect cardiac contraction independent of the central nervous system: cardiomyocyte-specific deletion of Bdnf or of the truncated form of the BDNF receptor, i.e. TrkB.T1, was sufficient to abrogate BDNF-dependent inotropic effects and led to cardiomyopathy in mice.5 These experimental studies suggest that BDNF exerts its cardiac functions through local myocardial receptors and imply protective cardiovascular effects of BDNF. In humans, an increased content of plasma BDNF was detected in the coronary circulation of patients with unstable angina compared with patients with stable coronary artery disease (CAD) or controls without CAD.6 While BDNF expression in non-atherosclerotic arteries was low, it was increased in the intima and adventitia of atherosclerotic coronary arteries, and particularly elevated in smooth muscle cells and macrophages.6 Others proposed that activated platelets express and secrete higher BDNF levels than any vascular cell type,7 suggesting that platelet-derived BDNF contributes to the above-mentioned observations.6 It remains to be determined whether increased plasma BDNF levels in human CAD are part of a compensatory feedback loop. Collectively, these findings suggest that BDNF is secreted in multiple tissues and exerts systemic, autocrine, and paracrine effects in the cardiovascular system. The role of the Val66Met mutation, associated with reduced BDNF activity in cardiovascular disease (CVD). remains controversial. The divergent findings may be explained by the different CVD phenotypes, different risk factor profiles, or different genetic backgrounds of the respective study populations. Importantly, a recent large-scale analysis in the Framingham Heart Study and subsequent Mendelian randomization in the CARDIoGRAM consortium showed that higher serum BDNF levels correlated with a decreased risk of CVD and mortality, while the BDNF Val66Met allele was associated with CAD prevalence.8 Interestingly, two other independent studies demonstrated an association between the BDNF Val66Met polymorphism and the prevalence of depression among CAD patients.9,10 The study in this issue of the journal extends our knowledge about the effects of the BDNF Val66Met polymorphism in experimental thrombosis using transgenic mice carrying Val66Met alleles (BdnfMet/Met).11 The authors applied a carotid artery thrombosis model using either topical administration of FeCl3 or systemic injection of collagen and epinephrine: the time to occlusion was reduced and mortality of BdnfMet/Met mice was enhanced compared with controls. In vitro and in vivo experiments suggest that these findings are a consequence of an increased platelet activation and consecutive enhanced susceptibility to thrombosis observed in BdnfMet/Met mice. Subsequent proteomic analyses of aortic secretomes identified differential contents of gelsolin and alpha1-antitrypsin, which are implicated in fibrinolytic and proinflammatory processes, respectively. Furthermore, the authors noted increased vascular expression and activity of tissue factor as well as reduced expression levels of vascular sirtuin 1 (SIRT1) in BdnfMet/Met mice. Since the expression of tissue factor is modulated by the lysine deacetylase SIRT1,12,13 Amadio et al.11 treated BdnfMet/Met mice with a SIRT1 activator, and were able to rescue the thrombosis phenotype of BdnfMet/Met mice. These findings corroborate previous experimental studies where SIRT1 provided antithrombotic effects and atheroprotection in mice.14,15 Finally, the authors translated their experimental findings to human CAD: they reported an association of Val66Met homozygosity with the prevalence of acute myocardial infarction in a human cohort of 979 CAD patients, a finding in line with previous reports.8–10 (A) Reduced brain-derived neurotrophic factor (BDNF) activity in BDNFMet/Met mice is associated with decreased sirtuin 1 (SIRT1) expression and consecutively increased both tissue factor (TF) expression and activity, promoting platelet activation and arterial thrombosis. Moreover, compared with control mice, aortae of BDNFMet/Met mice secreted less gelsolin and more alpha1-antitrypsin (α1-AT), thereby further contributing to arterial thrombus formation. (B) In humans, the BDNFVal66Met single nucleotide polymorphism, also associated with a reduced BDNF activity, has been associated with both depression and coronary artery disease risk and mortality. However, mechanistic insight linking either phenotype in humans remains to be determined. In summary, this hypothesis-generating study uncovers an intriguing association between the BDNF Val66Met polymorphism in depression and thrombosis. A causal role for SIRT1 may exist but remains mechanistically unproven. Major challenges remain to be addressed for understanding the mechanisms linking human depression and CAD given the polygenic nature of both diseases. We acknowledge support to S.S. by an Ambizione grant from the Swiss National Science Foundation (SNSF) and Novartis Consumer Health Foundation, to C.M.M. by the SNSF, Swiss Heart Foundation, the Hartmann-Müller Foundation, and to S.W. and C.M.M. by Matching Funds UZH, University Research Priority Program Integrative Human Physiology at the University of Zurich and the Zurich Heart House. Conflict of interest: none declared.