Genetic polymorphisms at SIRT1 and FOXO1 are associated with carotid atherosclerosis in the SAPHIR cohort

SIRT1 and FOXO1 are evolutionary conserved regulators of aging, metabolic processes and resistance to oxidative stress. Recently, their role in vascular homeostasis has been suggested. SIRT1 seems to have protective properties against atherosclerosis [1]. It controls vascular tone and endothelial function by deacetylation of endothelial nitric oxide synthase and by stimulating its activity [2],[3]. In endothelial cells and macrophages SIRT1 has anti-inflammatory functions by down-regulating the expression of various pro-inflammatory cytokines, interfering with the NFκB signaling pathway, preventing macrophage foam cell formation and suppressing of endothelial tissue factor (coagulation factor III) expression [1]. In vascular smooth muscle cells SIRT1 enhances the activity of tissue inhibitor of metalloproteinase 3 that might stabilize atherosclerotic plaques [4]. Moreover, SIRT1 is a key component in several stress-responsive pathways involved in apoptotic cell death and cellular senescence [3].

FOXO1 is strongly expressed in atherosclerotic plaques, regulates expression of different cell cycle regulators and is involved in multiple atherogenic pathways in endothelial cells. FOXO1 activity is regulated through processes of phosphorylation by Akt, acetylation at multiple lysine residues and deacetylation by several deacetylases, including the NAD⁺-dependent deacetylase SIRT1 [5]-[9].

FOXO1 is deacetylated in response to oxidative stress and hyperglycemia. Deacetylation of FOXO1 leads to its activation and enhanced expression of FOXO1 target genes [10]. Mice bearing constitutively deacetylated alleles of Foxo1 develop larger artherosclerotic lesions despite improved plasma lipid levels [11]. Ablation of the three genes encoding isoforms of FOXO in endothelial cells prevents atherosclerosis in LDL receptor knockout mice [12]. The atheroprotective effect of FOXO1 deletion was associated with a decrease of insulin-dependent Akt phosphorylation in endothelial cells. Growth factors induce phosphorylation and nuclear exclusion of forkhead proteins in vascular smooth muscle cells that inhibits proliferation of these cells and neointimal hyperplasia [13].

Findings from previous studies showed that SIRT1 and FOXO1 interact with each other in regulating the response to oxidative stress, ageing and longevity. SIRT1 has multiple effects on FOXO-induced gene expression that depresses processes of apoptosis and induces cell-cycle arrest and survival [7],[14]. Until now, associations of genetic polymorphisms at SIRT1 and FOXO1 with body mass index (BMI), visceral obesity, insulin sensitivity, type 2 diabetes mellitus (T2DM) and all-cause mortality risk have been described [15]-[19]. Taking into consideration that atherosclerosis is an aspect of the aging process we hypothesized that genetic polymorphisms at SIRT1 and FOXO1 may also play a role in carotid atherosclerosis.

Our study is the first to report a significant association of genetic polymorphisms at SIRT1 and FOXO1 with a carotid atherosclerosis. More pronounced effects were present for two SNPs (rs10507486 and rs2297627) at FOXO1 and common carotid IMT. Also we found one haplotype in FOXO1 with a moderate effect on common carotid IMT and B-score in comparison to the reference haplotype. Finally, we detected a sex-specific effect for the SNPs at SIRT1 with a protective role of the minor alleles in males and an opposite effect in females.

The pathophysiological functions of SIRT1 and FOXO1 in vascular homeostasis and carotid atherosclerosis have been described within the last years and exert protective roles in case of SIRT1 and controversial roles for FOXO1. Also, the interaction between SIRT1 and FOXO1 in different organisms has been demonstrated. SIRT1 deacetylates FOXO1 and regulates its transcriptional activity, in turn, FOXO1 exerts a positive feedback mechanism regulating SIRT1 expression. FOXO1 can directly bind to SIRT1 promoter region and increase SIRT1 transcription and expression [26]. Phosphorylation of FOXO1 makes it retention in the cytoplasm and inhibition of its transcription activity [27],[28]. In response to oxidative stress, FOXO1 translocates to the nucleus, interacts with SIRT1 resulting in the deacetylation of FOXO1 [10]. This depresses the sensitivity of FOXO1 to insulin-induced phosphorylation, resulting in gain function [29]. At the same time, acetylation of FOXO1 facilitates their phosphorylation and nuclear translocation.

Oxidative stress or hyperglycemia induce FOXO1 deacetylation in vascular endothelial cells, that contributes to the increased risk of atherosclerosis in diabetic patients. Experiments with isolated vascular endothelial cells from mice indicated that deacetylated FOXO1 affects its gene expression response to a variety of pathogenic stimuli and increases interaction of monocytes with endothelium [11].

Previous publications have shown an association of genetic variations at SIRT1 with BMI and visceral obesity [16], acute insulin secretion [30], basal energy expenditure and the respiratory quotient [31]. Carriers of the mutant allele were resistant against lifestyle-induced improvement of fasting plasma glucose, had less increase in insulin sensitivity and an attenuated decline in liver fat [31]. Also it was shown that expression of the SIRT1 gene was negatively correlated with carotid IMT [32]. However, no genetic polymorphisms in the SIRT1 gene were associated with expression level of the SIRT1 transcript [16].

There are also findings about the role of genetic variations in the FOXO1 and the susceptibility for T2DM. Müssig et al. showed the nominally significant association of rs2297627 with β-cell dysfunction (p = 0.0387) and impaired glucose tolerance (p = 0.0221). Carries of the minor allele had reduced insulin secretion and elevated glucose levels during an oral glucose tolerance test [17]. Kuningas et al. revealed the role of different haplotypes in the human FOXO1a gene in glucose metabolism, all-cause and T2DM mortality risk [33].

Association of genetic polymorphisms at the SIRT1 and FOXO1 loci with carotid atherosclerosis have not been reported before. However, significant associations between genetic variants in other SIRT gene - the SIRT6 and atherosclerotic plaque have been observed [34]. The last meta-analysis of genome-wide association studies by the CHARGE consortium identified some variants associated with common carotid IMT and plaques. However, no SNP achieved significant thresholds on a genome-wide scale for internal carotid IMT [35].

The genetic effect of SIRT1 and FOXO1 on common carotid IMT shown in the current study might be explained by their direct involvement in the pathophysiological mechanisms of atherosclerosis. FOXO1 integrates diverse cellular signals through its post-translational modifications, such as phosphorylation, acetylation, deacetylation resulting in changes in DNA binding properties, transcriptional activity, protein stability and subcellular localization. At the same time, SIRT1 is one of multiple regulators of FOXO1 activity in endothelial and smooth muscle cells.

It has to be noted that the definition of the carotid atherosclerotic phenotype consists of site-dependent IMT measurement and plaque-score. Although IMT and plaques are highly correlated with each other [36],[37], we found in our study an association of genetic polymorphisms at the SIRT1 and FOXO1 only with common carotid IMT. A haplotype specific effect was also found for mean B-score, though, which could not be detected at the single SNP level. The site-specific differences in atherosclerosis phenotype might be based on the important role of blood flow and genetics in manifestation and progression of carotid atherosclerosis [38]-[40]. Elevated carotid IMT may be the result of an increased intima layer as a consequence of atherosclerotic process, but also as an increased media layer due to vascular hypertrophy in hypertension or increasing of both layers. Nevertheless, IMT in carotid arteries is an established marker of the early onset of atherosclerosis and has been included in recent guidelines as a potentially useful marker for cardiovascular disease prediction in clinical and epidemiological studies [41],[42].

It is evident, that there is a sex-specific aspect of cardiovascular disease. In the Tromso Study, carotid atherosclerosis was a stronger predictor of myocardial infarction in women than in men [43]. Also this study showed that only in women triglyceride levels were associated with an increase in IMT, while physical activity and smoking were predictors of IMT in men [44]. Previously we published that the metabolic syndrome is a stronger risk factor for early carotid atherosclerosis in women than in men. In females, blood glucose and triglyceride levels showed the strongest association with IMT, whereas HDL-cholesterol had a stronger effect in men [45]. The available literature describes sex-specific effects only for SIRT1 in relation to increased risk for T2DM in Pima Indians [30] and SIRT6 for atherosclerotic plaque [34]. We cannot explain the mechanism underlying the sex-related association shown in our study. The SNPs that showed sex-specific effects were not significant in the overall analysis. Different effect of genetic polymorphisms in the SIRT1 on carotid atherosclerosis in female and male populations might be partial explained by the different role of sex hormones in regulation of gene expression or important role of sex-specific risk factors for atherosclerosis.

The strength of this investigation is that all ultrasound investigations were done by a single experienced ultrasound operator who was blinded to all clinical and laboratory measurements. It allowed us to reduce inter-observer variability and obtain data independent from the clinical phenotype. Moreover, the gene targeting approach used in this study gave not only the possibility to find association of genetic variations with carotid IMT and B-score, but also brought a new information about role of genetics of SIRT1 and FOXO1 in the atherosclerotic process. There are also some limitations of this investigation. First, replication of the findings in an independent cohort is missing. Due to power issues it was also not possible to perform an internal validation approach. Secondly, some of the included SNPs were in linkage disequilibrium although SNPs have been selected using SNPtagger to exclude such a situation. In the analysis, the LD structure was accounted for by calculating the number of independent SNPs. Therefore, further replication studies and also functional studies are warranted to elucidate the molecular mechanism of this association and the role of these genetic variations at SIRT1 and FOXO1 in carotid atherosclerosis.

Atherosclerosis is a clinically latent process for many years before its manifestation as coronary, cerebrovascular and/or peripheral arteries diseases. The identification of patients with high cardiovascular risk in subclinical phase of atherosclerotic process is critical for primary prevention of cardiovascular diseases. Genetics in combination with clinical parameters and carotid phenotypes may help us to understand pathogenesis of atherosclerotic diseases which might improve the prediction of cardiovascular diseases. However, the functional consequences of SIRT1 and FOXO1 genetics in regulation of carotid atherosclerosis remain incompletely understood and should be investigated in further functional studies.

LK designed and supervised the study, carried out the genotyping and drafted the manuscript. CL performed the statistical analysis and drafted the manuscript. IK carried out the genotyping and managed the SAPHIR study. BK performed the statistical analysis and helped to draft the manuscript. TK carried out the genotyping and helped to draft the manuscript. BI performed ultrasonography and managed the SAPHIR study. FK designed and supervised the study, drafted the manuscript. BP designed and supervised the study, managed the SAPHIR study and drafted the manuscript. All authors read and approved the final manuscript.