BACKGROUND. The identification of patients with high-risk atherosclerotic plaques prior to the manifestation of clinical events remains challenging. Recent findings question histology- and imaging-based definitions of the “vulnerable plaque,” necessitating an improved approach for predicting onset of symptoms. METHODS. We performed a proteomics comparison of the vascular extracellular matrix and associated molecules in human carotid endarterectomy specimens from 6 symptomatic versus 6 asymptomatic patients to identify a protein signature for high-risk atherosclerotic plaques. Proteomics data were integrated with gene expression profiling of 121 carotid endarterectomies and an analysis of protein secretion by lipid-loaded human vascular smooth muscle cells. Finally, epidemiological validation of candidate biomarkers was performed in two community-based studies. RESULTS. Proteomics and at least one of the other two approaches identified a molecular signature of plaques from symptomatic patients that comprised matrix metalloproteinase 9, chitinase 3-like-1, S100 calcium binding protein A8 (S100A8), S100A9, cathepsin B, fibronectin, and galectin-3-binding protein. Biomarker candidates measured in 685 subjects in the Bruneck study were associated with progression to advanced atherosclerosis and incidence of cardiovascular disease over a 10-year follow-up period. A 4-biomarker signature (matrix metalloproteinase 9, S100A8/S100A9, cathepsin D, and galectin-3-binding protein) improved risk prediction and was successfully replicated in an independent cohort, the SAPHIR study. CONCLUSION. The identified 4-biomarker signature may improve risk prediction and diagnostics for the management of cardiovascular disease. Further, our study highlights the strength of tissue-based proteomics for biomarker discovery. FUNDING. UK: British Heart Foundation (BHF); King’s BHF Center; and the National Institute for Health Research Biomedical Research Center based at Guy’s and St Thomas’ NHS Foundation Trust and King’s College London in partnership with King’s College Hospital. Austria: Federal Ministry for Transport, Innovation and Technology (BMVIT); Federal Ministry of Science, Research and Economy (BMWFW); Wirtschaftsagentur Wien; and Standortagentur Tirol.

There is strong and consistent evidence that oxidative stress is crucially involved in the development of atherosclerotic vascular disease. Overproduction of reactive oxygen species (ROS) in mitochondria is an unifying mechanism that underlies micro- and macrovascular atherosclerotic disease. Given the central role of mitochondria in energy and ROS production, mitochondrial DNA (mtDNA) is an obvious candidate for genetic susceptibility studies on atherosclerotic processes. We therefore examined the association between mtDNA haplogroups and coronary artery disease (CAD) as well as diabetic retinopathy.

A small proportion of lean patients develop non-alcoholic fatty liver disease (NAFLD). We aimed to report the histological picture of lean NAFLD in comparison to overweight and obese NAFLD patients. Biopsy and clinical data from 466 patients diagnosed with NAFLD were stratified to groups according to body mass index (BMI): lean (BMI ≤ 25.0 kg/m², n confirmed to be appropriate = 74), overweight (BMI > 25.0 ≤ 30.0 kg/m², n = 242) and obese (BMI > 30.0 kg/m², n = 150). Lean NAFLD patients had a higher rate of lobular inflammation compared to overweight patients (12/74; 16.2% vs. 19/242; 7.9%; p = 0.011) but were similar to obese patients (25/150; 16.7%). Ballooning was observed in fewer overweight patients (38/242; 15.7%) compared to lean (19/74; 25.7%; p = 0.014) and obese patients (38/150; 25.3%; p = 0.006). Overweight patients had a lower rate of portal and periportal fibrosis (32/242; 13.2%) than lean (19/74; 25.7%; p = 0.019) and obese patients (37/150; 24.7%; p = 0.016). The rate of cirrhosis was higher in lean patients (6/74; 8.1%) compared to overweight (4/242; 1.7%; p = 0.010) and obese patients (3/150; 2.0% p = 0.027). In total, 60/466; 12.9% patients were diagnosed with non-alcoholic steatohepatitis (NASH). The rate of NASH was higher in lean (14/74; 18.9% p = 0.01) and obese (26/150; 17.3%; p = 0.007) compared to overweight patients (20/242; 8.3%)). Among lean patients, fasting glucose, INR and use of thyroid hormone replacement therapy were independent predictors of NASH in a multivariate model. Lean NAFLD patients were characterized by a severe histological picture similar to obese patients but are more progressed compared to overweight patients. Fasting glucose, international normalized ratio (INR) and the use of thyroid hormone replacement may serve as indicators for NASH in lean patients.

SIRT1 and FOXO1 interact with each other in multiple pathways regulating aging, metabolism and resistance to oxidative stress and control different pathways involved in atherosclerotic process. It is not known, if genetic polymorphisms (SNPs) at the SIRT1 and FOXO1 have an influence on carotid atherosclerosis. Intima-media thickness (IMT) was measured on the common and internal carotid arteries. Morphological alterations of the carotid arteries and size of these alterations were included in the B-score grading on a five point scale. Eleven SNPs at SIRT1 and FOXO1 gene loci were genotyped in the SAPHIR cohort (n = 1742). The association of each SNP with common carotid IMT, internal carotid IMT and B-score was analyzed using linear regression models. A significant association was found between common carotid IMT and two SNPs at FOXO1 - rs10507486, rs2297627 (beta = -0.00168, p = 0.0007 and beta = -0.00144, p = 0.0008 respectively) and at least a trend for rs12413112 at SIRT1 (beta = 0.00177, p = 0.0157) using an additive model adjusting for age and sex. Additional adjustment for traditional cardiovascular risk factors and markers (BMI, smoking status, hypertension, total cholesterol, HDL-cholesterol, hsCRP) even improved the strength of this association (p = 0.0037 for SIRT1 and p = 0.0002 for both SNPs at FOXO1). Analysis for internal carotis IMT and B-score did not reveal any significant association. One haplotype in FOXO1 showed a moderate effect on common carotid IMT and B-score in comparison to the reference haplotype of this gene. Several SNPs within SIRT1 showed differential effects for men and women with higher effect sizes for women: rs3740051 on all three investigated phenotypes (interaction p-value < 0.0069); rs2236319 on common and internal carotid IMT (interaction p-value < 0.0083), rs10823108, rs2273773 on common carotid IMT and rs1467568 on B-score (interaction p-value = 0.0007). The latter was significant in women only (betawomen = 0.111, pwomen = 0.00008; betamen = -0.009, pmen = 0.6464). This study demonstrated associations of genetic variations at the SIRT1 and FOXO1 loci with carotid atherosclerosis and highlighted the need for further investigation by functional studies.

Aims: Imbalances of iron metabolism have been linked to the development of atherosclerosis. However, subjects with hereditary haemochromatosis have a lower prevalence of cardiovascular disease. The aim of our study was to understand the underlying mechanisms by combining data from genome-wide association study analyses in humans, CRISPR/Cas9 genome editing, and loss-of-function studies in mice. Methods and results: Our analysis of the Global Lipids Genetics Consortium (GLGC) dataset revealed that single nucleotide polymorphisms (SNPs) in the haemochromatosis gene HFE associate with reduced low-density lipoprotein cholesterol (LDL-C) in human plasma. The LDL-C lowering effect could be phenocopied in dyslipidaemic ApoE−/− mice lacking Hfe, which translated into reduced atherosclerosis burden. Mechanistically, we identified HFE as a negative regulator of LDL receptor expression in hepatocytes. Moreover, we uncovered liver-resident Kupffer cells (KCs) as central players in cholesterol homeostasis as they were found to acquire and transfer LDL-derived cholesterol to hepatocytes in an Abca1-dependent fashion, which is controlled by iron availability. Conclusion: Our results disentangle novel regulatory interactions between iron metabolism, KC biology and cholesterol homeostasis which are promising targets for treating dyslipidaemia but also provide a mechanistic explanation for reduced cardiovascular morbidity in subjects with haemochromatosis.

Background and Aims: Although non-alcoholic fatty liver disease (NAFLD) is linked to obesity, a proportion of lean subjects also have NAFLD with potentially distinct clinical features. We studied the outcome of lean NAFLD subjects. Methods: 299 consecutive patients (215 male, 84 female, 49.5 ± 13.5years) with biopsy-proven NAFLD and a follow-up of 8.4 years (±4.1; range: 0.9-18.0) were stratified by body mass index (BMI) at the time of liver biopsy: lean (BMI ≤25.0 kg/m, n=38), overweight (BMI 25.0-29.9 kg/m2, n=165), obese (BMI ≥30.0 kg/m2, n=93). A control group of 1,013 subjects (547 male, 52.4 ± 5.8) was used for comparison. The time to the event was recorded. Multivariable Cox regression analyses were performed to assess associations with 10-year-mortality. Hazard ratios (HR) and adjusted hazard ratios (aHR) with 95% confidence intervals (CI) were calculated. Results: Age and gender were similar, while components of the metabolic syndrome were less frequent in lean subjects. The proportion of subjects with significant fibrosis and the number of subjects with cirrhosis was increased in lean subjects while the proportion of non-alcoholic steatohepatitis was not different. Mortality in the NAFLD groups was significantly higher than in the control group. Multivariable analysis adjusting for age, gender, and glucose confirmed lower mortality in overweight (aHR 0.21; 95% CI 0.07-0.62, p=0.005) and in obese (aHR 0.22; 95% CI 0.06-0.76, p=0.02) compared to lean subjects. Further adjustment for fibrosis weakened the difference between lean and obese (p=0.12) while the difference to overweight subjects remained intact (p=0.01). Conclusion: Lean subjects with NAFLD have a high risk of liver-related death. Our data support that lean NAFLD subjects deserve particular attention with regard to clinical follow-up.