Epitranscriptomic Blood Biomarkers for Coronary Artery Disease - A Prospective Cohort Study (IHD-EPITRAN)

Background - Ischemic Heart Disease

With a global prevalence of 126.5 million people and a yearly mortality of 8.9 million, ischemic heart disease (IHD) is the leading cause of death (GBD Collaborators 2017). IHD develops as a result of an ongoing atherogenesis, a process of lipid buildup into the walls of the coronary arteries. Eventually, the accumulation of cholesterol and calcium form deposits, i.e. plaques, that narrow down the vessel lumina. At the onset of myocardial infarction (MI), such a plaque ruptures, which leads to anoxic death of the myocardium relying on the blood supply of the vessel. In the short term, the heart's function is rescued by the formation of a stiff scar. Over time, the pumping function typically deteriorates with ensuing heart failure (HF).

Optimally, disease biomarkers are analyzed from blood and provide both insight into disease progression, severity and aid the evaluation of therapy efficacy. Unfortunately, optimal biomarkers for IHD have not been identified to date. The vast but uncounted number of those living with undiagnosed IHD, benefitting from an early diagnosis, underscore the need for an IHD biomarker.

Background - Epitranscriptomics

Epitranscriptomics, the field of the posttranscriptional modifications of RNA, analogous to DNA epigenetics, has recently gained more scientific remark (Saletore 2012). This expanding field is uncovering a new layer of biological regulation that controls processes ranging from cell proliferation to death.

Although more than 170 RNA modifications have been identified in RNA species (Yang 2018), adenosine methylation at the nitrogen-6 position (m6A) stands out as one the most common and intensively studied (Liu 2020). Following the initial findings of m6A methylated mRNAs in 1970s (Desrosiers 1974), epitranscriptomics has only reignited during the last decade with the development of reliable in vivo methods to assess m6A (Dominissini 2012). These early immunoprecipitation-based methodologies (i.e. meRIP-seq) facilitated the discoveries of the specific m6A punctuation patterns and the specifically denoted writers (methyltransferases; METTL3/14/WTAP1 complex and METTL16), readers (binding proteins; e.g. YTH-protein family), and erasers (demethylases; ALKBH5; Zaccara 2019).

Multiple readers mediating the m6A effects from the YTH domain family have been identified. YTHDF1-3 promote mRNA translation, degradation or both, respectively. YTHDC1-2, eIF2, METTL3 and ribosomes recognize m6A as well and promote translation and cytoplasmic compartmentalization (Zaccara 2019).

Changes in m6A levels have been associated with numerous pathologies, including cancers, cardiovascular and neurological disorders (Frye 2016). Studies using both knockout and overexpression of the m6A writers and erasers have revealed their role in driving immune reactivity, cell proliferation, migration and apoptosis (Delaunay 2019).

Background - Epitranscriptomics \& heart

A recently published set of studies suggest a role for both the m6A and A-to-I editing (adenosine-to-inosine, another common modification diversifying the transcriptome by altering base pairing) during hypertension (Jain 2018), neovascularization (Kwast 2019), myocardial hypertrophy (Dorn 2019, Kmietczyk 2019), ischemia (Mathiyalagan 2019) and HF (Berulava 2020). While a body of evidence indicate epitranscriptomics as a contributor in cardiac health and disease, all of the initial studies have focused on the myocardium.

Rationale, objectives and significance

We hypothesize that the pathophysiology of IHD is reflected in the epitranscriptome of the circulating RNA. For example, ribosomal RNA has a long half-life and its modifications in the coronary circulation can respond to the state of the heart.

The objectives of the IHD-ERPITRAN project are: (1) to establish screening protocols for epitranscriptomic biomarkers, (2) provide epitranscriptomic insight into the IHD pathophysiology, (3) identify a set of IHD biomarker candidates and (4) open avenues for therapeutic development that were previously disregarded due to lack of research and methodological limitations. For example, our research group has reported a discovery of small molecule ligands for RNA methyltransferases (Selberg 2019).

Since epitranscriptomics has been suggested to play a regulatory role in heart and the IHD-EPITRAN is among the first to evaluate epitranscriptomic blood biomarkers in IHD, the project holds keys for a scientific breakthrough with expectable global clinical benefits.

Methods

Study cohorts - 200 patients are recruited into four cohorts, counting 50 patients each. First cohort consists of patients presenting to Heart Unit with an acute MI revascularized with percutaneous coronary intervention (PCI), a cohort that gives valuable insight to the way the acute IHD phenotype is mirrored to the blood epitranscriptomes. Second, the main study cohort of the project will be formed from patients with stable IHD phenotype treated with coronary artery bypass grafting (CABG) and offers critical information in respect to chronic ischemia and coronary atherogenesis, both ongoing processes in the silent, i.e. undiagnosed, IHD. Third, patients destined for an aortic valve replacement (AVR) therapy due to stenosis with no IHD provide the positive control group with a differing cardiac pathology, and, fourth, together with the healthy non-IHD patients verified with the coronary computed tomography (CT), form the control cohorts of the project.

Study samples - The IHD-EPITRAN uncovers blood cellular and cell-free RNA epitranscriptomes from the twice collected TEMPUS(TM) and EDTA blood samples, respectively. In addition, for MI and CABG cohorts, the epitranscriptomes from the right atrial appendage tissue samples are deciphered for a reference point. Finally, a set of supplementary HF and IHD biomarkers: NT-proBNP, hsCRP, sST2 and TMAO (Aimo 2019, Ahmadmehrabi 2017 and Tibaut 2019) are measured.

RNA methods - The RNA isolation and fractionation with in-lab-validated methods is followed by the quantitative analysis of seven base modifications utilizing an UHPLC-triple quadrupole liquid chromatography-tandem mass spectrometry (LC-MS/MS) system (Selberg 2019). Following the quantitative analysis, anti-m6A antibody-based RNA pre-selection rich in m6A marks in tandem with sequencing, i.e. meRIP-seq, is performed. Ultimately, after the UHPLC-LC-MS/MS and meRIP-seq, direct nanopore long-read sequencing enabling identification of the m6A marks in their native sequences is carried out using the novel and recently developed algorithm by Liu et al. (2019). The algorithm bases on the recognition of the characteristic disruptions of electrical current by the m6A flowing through specific protein nanopores.

DNA sequencing - In order to identify the A-to-I RNA editing events, a whole genome next generation sequencing is performed for all study participants to enable a matched DNA-to-RNA base comparison (Park et al., 2012).

Bioinformatics - In order to interpret the massive datasets yielded from the sequencing, bioinformatics protocols are developed in collaboration with the bioinformatics teams from Folkhälsan (Karolinska Institutet, Dr. Shintaro Katayma) and Middle East Technical University (professor Nurcan Tuncbag). The sequences are evaluated against known transcript libraries.

Clinical methods - The echocardiography gives insight into the functional state of the heart and is performed with pre-specified parameters both during the initial hospitalization and 3 months after either PCI, CABG or AVR. These cardiologists' appointments include the severity assessment of angina pectoris and exertional dyspnea with CCS and NYHA gradings, respectively. In addition, morbidity self-assessment with the standardized Short Form 36 Health Survey is included. Moreover, a patient record system search including hospital admissions, deaths and medication changes associable to cardiovascular health are collected 6 months prior and after operation.

Power analysis - The cohort sizes were determined with RNAseqPS web tool designed for evaluating statistical power for RNAseq experiments (Guo et al., 2014). Parameter values used (false discovery rate [FDR] <0.05, total number of genes for testing 20 000, predicted prognostic genes 1500, minimum fold change threshold 2 for differential expression and average read counts of 10 for prognostic genes) were derived from an applicable report concerning epitranscriptomics during HF (Berulava et al., 2020). Prognostic gene dispersion value of 0,215 was applied based on the article focusing on the issue by Yoon \& Nam 2017, which analysed dispersions from ten publicly available RNA-seq datasets, four of which reflected unrelated replicates applicable for the IHD-EPITRAN with the dispersions ranging between 0.15 and 0.28.

Based on the power analysis described above, n=25 per cohort provides sufficient power (P≥0,95). To fluently tackle possible preanalytical errors, perform subsequent validation and follow-up analyses, the cohorts were doubled.

Patient registries

Individual participant identification information data collected during the IHD-EPITRAN will be mainly stored in electrical form on HUS's and Tays's network hard drives that are protected by role-based access rights. Two separate registries are created for the IHD-EPITRAN: (1) A Key Registry, which contains all participant identification information and links between pseudonymization codes that are used in the (2) Research Registry (fully pseudonymized) that contains all other collected study information from the participants. Antti Vento, the director of the IHD-EPITRAN, has the accession rights to the Key Registry. The registries are stored for 10 years to enable possible later follow-up projects.

Processing description documents for both registries, alongside self assessment document for the risks of the study, have been created for the IHD-EPITRAN in Finnish. These documents has been reviewed and approved by the Ethics Board of the Hospital District of Helsinki and Uusimaa (Dnr. HUS/1211/2020).

Research group

Docent Antti Vento is the director of both the IHD-EPITRAN project and the HUS Heart and Lung Center. The interdisciplinary team composed of clinicians, basic researchers and analytical experts enables the IHD-EPITRAN to synergistically perform recruitment, sample collection and preparation followed by the bioinformatic analyses of the epitranscriptomic landscape of circulating RNA in IHD. Docents Mika Laine, Pasi Karjalainen, Helena Rajala and Satu Suihko, all cardiologists from HUS heart station, will carry out the CT patients' recruitment, imaging analyses, treatment of the MI patients and, thirdly, MI, CABG and AVR patients' control appointments. Heart and vascular surgeon Kari Teittinen from the HUS Heart and Lung Center, together with the cardiothoracic surgeon Jahangir Khan PhD and professor Jari Laurikka, both from Tays Heart Hospital, carry out the recruitment of the CABG and AVR patients and their operations. Research nurses Kati Oksaharju and Kati Helleharju, from HUS Heart and Lung Center and Tays Heart Hospital, respectively, coordinate patient recruitment and, in close collaboration with laborant Lahja Eurajoki, the sample collection, transport and prompt storage after freezing. The project will be carried out in the laboratory of docent Esko Kankuri and professor Eero Mervaala (University of Helsinki, CardioReg Group, Department of Pharmacology). B.M. Vilbert Sikorski and MSc Daria Blokhina, will collaboratively perform RNA isolation, fractionation, UHPLC-LC-MS/MS analysis and protocol article formulation. Additionally, Sikorski has collaboratively planned study protocol with the other IHD-EPITRAN co-authors, applied study permissions and continues to coordinate information exchange between collaborators and carries out results article writing with the co-authors.

Ethics

This study will be conducted in compliance with the protocol approved by the HUS Ethics Board (Dnr. HUS/1211/2020). Respective regional and national boards will evaluate the protocol subsequently for the multicenter sample collection. Care will be taken so that each study participant has enough time to familiarize himself/herself with the study protocol. Thereafter, an informed consent discussion will be held and written informed consent obtained from the participant.