Identifying circulating biomarkers in a cohort of hypercholesterolaemic patients with statin intolerance from Gauteng, South Africa

Abstract

Cardiovascular disease is one of the leading causes of death worldwide. Dyslipidemia, defined by high circulating levels of atherogenic apoB-containing lipoproteins, contributes significantly to the increasing burden of cardiovascular morbidity and mortality.

At public health care institutions, in South Africa, simvastatin and atorvastatin are the standard therapy for hypercholesterolemia. In combination with dietary and lifestyle changes statins are the gold standard therapy to effectively reduce LDL levels to lower the risk of cardiovascular disease, e.g. myocardial infarction or stroke. However, a proportion of patients do experience statin associated adverse effects, especially muscle related symptoms.

Studies suggest that statins affect hundreds of miRNAs, as one miRNA has the potential of binding to hundreds of mRNAs, the consequences may be enormous. However, whether these epigenetic changes lead to statin related adverse effects in South African statin users is still unknown.

The aim of this pilot study was therefore to identify potential circulating biomarkers of statin intolerance in a cohort of hypercholesterolaemic patients from Gauteng, South Africa. Following institutional ethical clearance, healthy participants (n=100), hypercholesterolemic participants not taking statins (n=100) and statin-treated hypercholesterolemic participants (n=100) were recruited. Informed Consent was obtained from all participants prior to study enrollment.

The median age (range) for participants included in the study was 49 (19 - 78), 47 (22 – 79) and 49 (19 – 78), for the control group, statin users and statin naïve group, respectively. Sampling and enrolment were based on South Africa’s population ratios, with an average of ~80% black South Africans and ~20% white South African included in the study in each group. Of note is the higher prevalence of comorbidities in the statin naïve compared to the statin users. However, this is likely attributed to statin intolerance, delayed treatment initiation, socio-economic disparities, healthcare access and treatment adherence.

All of the statin-treated hypercholesterolaemic participants were requested to complete a questionnaire adapted from the American College of Cardiology (ACC) Statin Intolerance Application. The ACC’s Statin Intolerance Application was developed to assist clinicians to manage cholesterol, as well as managing and treating patients who present statin related adverse events. Of the 100 statin-treated hypercholesterolemic patients enrolled, 15 presented with a high risk of statin intolerance (15%), 49 presented with a moderate risk of statin intolerance (49%) and 36 presented with a low risk of statin intolerance (36%).

In silico screening of miR-33a, miR133a and miR-499a using TargetScan was used to assess the conserved mRNA targets of miR-33a (hsa-miR-33a), miR-133a (hsa-miR-133a) and miR-499a (hsa-miR-499a). The targets generated from TargetScan were filtered and sorted based on a predefined criteria for each miRNA. For miR-33a a total of 545 transcripts with conserved sites were identified using the TargetScan. The table generated from TargetScan was sorted according to the aggregate PCT score, and targets were filtered based on their association to cholesterol homeostasis and atherosclerosis. The list of target genes with conserved binding sites for miR-33a/b identified, includes genes involved in regulating cholesterol homeostasis, specifically ABCA1 and ABCG1, fatty acid metabolism, cyclin dependent kinases and mitogen-activate protein kinases. The in silico screening performed to identify possible target genes for miR-133a using TargetScan generated 703 possible targets, The list of target genes included target genes of the SRF transcription factor, such as AOX1, CCNDBP1, ISCA2, LGALS8, MRPL44 and POLH. From the conserved targets identified for miR-499a using TargetScan of note is FOXO4, PDCD4 and SOX6, which all are genes associated in pathways related to regulation of smooth and skeletal muscle cell differentiation.

Quantitative polymerase chain reaction (qPCR) was used to determine the levels of miR-33a, miR-133a and miR-499a, enzyme-linked immunosorbent assay (ELISA) was used to quantify serum creatine kinase (CK) and aspartate aminotransferase (AST) levels in plasma. Quantitative analysis showed no significant difference in the ApoA-1 levels between participants in the control, statin-users, or statin naïve groups, (p= 0,4117). A significant difference between the CK levels between the different patient groups, p= <0,0001, was however noted with the average CK level for the statin-users being significantly higher compared to the other 2 groups. The miRNA levels of miR-33a and miR-133a were detectable. A significant difference (p= 0,0011) in the relative fold change of miR-33 was noted between statin-users and statin naïve patients compared to the control group. A significant difference in the relative fold change of miR-133 expression was noted between the control group and statin-users (p= 0,0006), and the statin-users and the statin naïve patients (p= 0,0239). MicroRNA-499a, however, was not detectable in any of the samples, likely due to low abundance and biological variation.

A linear regression and Spearman R correlation analysis was conducted on matched samples between miR-33a and ApoA-1 levels. The Spearman’s rank correlation coefficient was used to assess a possible correlation between the miR-33a expression and ApoA-1 levels. An inverse correlation between miR-33a and ApoA-1, was observed, with the r = -0,3266 (95% confidence interval -0,5885 to -0,002740) and p= 0,0424. Despite the lack of a significant difference in ApoA1 levels between the control, statin users, and statin-naïve groups, the inverse correlation between miR-33a expression and ApoA1 levels observed in the study supports the notion that miR-33a has a regulatory influence on cholesterol metabolism.

Following the quantification of AST and the expression of miR-133a, a linear regression and Spearman R correlation analysis was conducted on matched samples between miR-133a and AST levels to determine miR-133a’s effect on AST levels. No correlation was observed between the miR-133a expression and AST levels for either the hypercholesterolaemic participants on statin therapy or the combined hypercholesterolaemic group, with r = -0,1088 (95% confidence interval -0,6336 to 0,4845) and p= 0,72 and r= 0,07492 (95% confidence interval -0,3326 to 0,4588) and p= 0,7161, respectively.

As there is evidence which supports the notion that miR-levels can influence muscle integrity, an alternative biomarker was also used to assess the possible association between miR-133a and muscle damage. Creatine kinase is the gold-standard biomarker used to assess cardiac- and skeletal muscle damage. The Spearman’s rank correlation coefficient was used to assess a possible correlation between the miR-133a levels and CK levels in statin-users and subsequently in a combined hypercholesterolaemic group (i.e. all the hypercholesterolaemic patient enrolled in the study, statin-treated and statin naïve patients). A positive correlation was observed between the CK levels and miR-133a levels in the combined hypercholesterolaemic group, with r= 0,4567 (95% confidence interval 0,1601 to 0,6777) and p= 0,0031. Although, no correlation was observed between the CK levels and miR-133a levels for the statin-users, a positive correlation was observed between the CK levels and miR-133a levels in the combined hypercholesterolaemic group. It was determined that there was a significant difference in the CK levels between the different patient groups, with the average CK level for the statin-users being significantly higher compared to the other two groups.

The findings of this study contribute to a growing body of evidence which supports the notion that of the three miR’s assessed, miR-33a may serve as a drug target for managing hypercholesterolemia and cardiometabolic diseases, such as atherosclerosis. miR-33a inhibits genes involved in cholesterol efflux, fatty acid oxidation and insulin signaling thus aiding in cholesterol regulation and lipid metabolism. Statin intolerance, particularly myopathy, is associated with disruptions in lipid metabolism and oxidative stress. Statins act as an activator of SREBP2, which can lead to increased miR-33a levels. As miR-33a negatively regulates ApoA-1, increased levels of miR-33a can reduce ApoA-1 levels. This can lead to reduced activity of ApoA-1 which hinders the removal of cholesterol from tissues. Decreased ApoA-1 levels can also exacerbate systemic inflammation, potentially worsening muscle symptoms in statin intolerant patients.

Understanding the role of miRNAs and its potential connection to statin intolerance could provide valuable insights into the adverse effects experienced by statin users. This study helps gain a better understanding of statin intolerance and these findings will allow a more personalized approach to statin therapy, e.g. miR-33a inhibitors which could lead to restored ApoA-1 and HDL levels, which in turn could improve cholesterol efflux and mitigate muscle-related symptoms.