19 Identification and evaluation of novel protein biomarkers for atrial fibrillation

• Published in: Heart (British Cardiac Society) Vol. 108; no. Suppl 3; pp. A16 - A17
• Main Authors: Tonry, C, Russell-Hallinan, A, Glezeva, N, Collier, P, McDonald, K, Ledwidge, M, Collins, BC, Watson, C
• Format: Journal Article
• Language: English
• Published: London BMJ Publishing Group Ltd and British Cardiovascular Society 06.10.2022; BMJ Publishing Group LTD
• Subjects: Biomarkers; Cardiac arrhythmia; Oral abstract presentations 3; Proteins
• ISSN: 1355-6037; 1468-201X
• DOI: 10.1136/heartjnl-2022-ICS.19

BackgroundThere is a need for improved biomarkers to diagnose atrial fibrillation (AF) earlier and reduce risk of future serious comorbidities. Quantitative protein profiling of atrial appendage tissue from patients with atrial fibrillation (AF n=10) and age/sex matched controls with normal sinus rhythm (control, n=10) was performed using mass spectrometry. Similarly, serum samples, collected longitudinally from patients with and without AF (n=186), were analysed to establish a comprehensive dataset that depicts changes in both the atrial tissue and circulating proteome as result of AF.MethodsSections of formalin fixed paraffin embedded (FFPE) tissue were mechanically homoginsed in Preomics™ LYSE buffer. Protein lysates were digested with trypsin and Lys-C using the Preomics ™ iST kit. Serum samples were enriched for low abundant serum proteins using High Select™ Top Abundant Protein Depletion Resin (Thermo). Unbiased, deep proteomic profiling of individual tissue and serum samples was performed using the diaPASEF workflow on a timsTOF Pro mass spectrometer. Nonparametric statistical tests were applied for subsequent data analysis in R and SPSS (version 27). Pathway analysis was performed using Ingenuity Pathway Analysis (IPA) software.ResultsLabel-free MS analysis led to the identification of over 6,000 proteins in FFPE tissue and over 500 serum proteins. More than 300 proteins were found to be significantly differentially expressed between AF and control samples at tissue level, with stringent cut off criteria applied (observed fold change of ≥ 1.5 or ≤-1.5 and p-value ≤ 0.005). Pathway analysis revealed that significantly up and down-regulated proteins mapped to Epithelial Adherens Junction Signalling and Atherosclerosis Signalling canonical pathways. The most up-regulated protein in AF correlated with tissue BNP levels (r=1.0, p<0.0001) and markers of tissue ischaemia (r=1.0, p<0.0001). The most down-regulated protein was inversely correlated with tissue levels of TGFb, the primary pro-fibrotic cytokine in the heart (r = -0.9, p=0.037). Thirty-one significantly changed tissue proteins were also identified in serum samples and were found to be associated with (i) new-onset AF, (ii) paroxysmal AF and (iii) risk of future stroke and/or heart failure in patients with AF.ConclusionsThe dataset has highlighted significant proteins associated with AF. We have verified that circulating levels of a number of these proteins are significantly associated with AF and, importantly, may be predictive of future cardiovascular comorbidities in patients with AF.