Machine Learning Analysis of the Cerebrovascular Thrombi Proteome in Human Ischemic Stroke: An Exploratory Study

Abstract

Objective: Mechanical retrieval of thrombotic material from acute ischemic stroke patients provides a unique entry point for translational research investigations. Here, we resolved the proteomes of cardioembolic and atherothrombotic cerebrovascular human thrombi and applied an artificial intelligence routine to examine protein signatures between the two selected groups. Methods: We specifically used n = 32 cardioembolic and n = 28 atherothrombotic diagnosed thrombi from patients suffering from acute stroke and treated by mechanical thrombectomy. Thrombi proteins were successfully separated by gel-electrophoresis. For each thrombi, peptide samples were analyzed by nano-flow liquid chromatography coupled to tandem mass spectrometry (nano-LC-MS/MS) to obtain specific proteomes. Relative protein quantification was performed using a label-free LFQ algorithm and all dataset were analyzed using a support-vector-machine (SVM) learning method. Data are available via ProteomeXchange with identifier PXD020398. Clinical data were also analyzed using SVM, alone or in combination with the proteomes. Results: A total of 2,455 proteins were identified by nano-LC-MS/MS in the samples analyzed, with 438 proteins constantly detected in all samples. SVM analysis of LFQ proteomic data delivered combinations of three proteins achieving a maximum of 88.3% for correct classification of the cardioembolic and atherothrombotic samples in our cohort. The coagulation factor XIII appeared in all of the SVM protein trios, associating with cardioembolic thrombi. A combined SVM analysis of the LFQ proteome and clinical data did not deliver a better discriminatory score as compared to the proteome only. Conclusion: Our results advance the portrayal of the human cerebrovascular thrombi proteome. The exploratory SVM analysis outlined sets of proteins for a proof-of-principle characterization of our cohort cardioembolic and atherothrombotic samples. The integrated analysis proposed herein could be further developed Dargazanli et al. Proteomic Composition of Intracranial Thrombi and retested on a larger patients population to better understand stroke origin and the associated cerebrovascular pathophysiology.

Potential Biomarkers of Acute Ischemic Stroke Etiology Revealed by Mass Spectrometry-Based Proteomic Characterization of Formalin-Fixed Paraffin-Embedded Blood Clots

“…We successfully analyzed the proteome of FFPE samples in the current study, identifying 1,581 protein groups as part of the proteome of AIS thrombi using LC-MS/MS. Our results are in line with previous studies performed on FF samples, such as the study of Dargazanli et al who identified 2,455 proteins in 50 samples ( 31 ) and the earlier study by Munoz et al, which used four samples and identified a proteome in the order of 1,600 proteins ( 32 ).…”

“…Both of these earlier studies found a set of proteins of interest helpful to advance AIS thrombus proteome characterization ( 31 , 32 ). We acknowledge that our use of FFPE samples allowed us to identify less proteins than previous studies which used FF samples.…”

“…14-3-3 protein in platelets interacts with several phosphoserine-dependent binding sites in the major platelet adhesion receptor, the glycoprotein Ib-IX complex, regulating its interaction with vWF and initiating thrombus formation ( 33 ). Dargazanli et al compared the proteome of AIS clots of cardioembolic and atheroembolic etiology, and identified seven key proteins that could be indicative of cardioembolic etiology, in particular, coagulation factor XIII and proteins involved in cellular cytoskeletal assembly such as myosin light chain kinase and F-actin capping protein ( 31 ). Our results are in line with these findings, since coagulation factor XIII A chain is one of the most abundant proteins we found in CE clots.…”

Proteomic Analysis of Cardioembolic and Large Artery Atherosclerotic Clots Using Reverse Phase Protein Array Technology Reveals Key Cellular Interactions Within Clot Microenvironments

“…A previous proteomic study found that two inflammation proteins (integrin alpha-M and mitochondrial superoxide dismutase) are associated with high blood low-density lipoprotein (LDL) [10]. In another study, these two proteins failed to differentiate between cardioembolic vs. atherosclerotic etiology [11]. Also, another study analyzed four thrombi, with 1,600 proteins identified [12].…”

Clot Composition Analysis as a Diagnostic Tool to Gain Insight into Ischemic Stroke Etiology: A Systematic Review

“…120 Finally, higher IL-1β expression was measured in LAA thrombi, 112 while coagulation factor XIII was associated with CE. 115 Some studies aimed to compare the features of CRY thrombi with those of known TOAST etiology. Most of them found similar proportions of RBCs, 39,57,64,96,122-124,127 FBR, PLTs (or both together), 57,76,96,[122][123][124]127 and WBCs between CRY and CE thrombi.…”