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

Acute Ischemic Stroke Thrombus Composition

Ischemic stroke is caused by a thrombus blocking one or multiple arteries in the brain, resulting in irreversible damage in the associated brain tissue. The aim of therapy is to restore the blood flow as fast as possible. Two recanalization strategies are currently available: pharmacological thrombolysis using recombinant tissue plasminogen activator (rt-PA) and mechanical removal of the thrombus. Despite recent advancements, achieving efficient recanalization remains a challenge. The precise causes of therapy failure are not fully understood but thrombus composition is likely a key factor in successful recanalization. This review explores acute ischemic stroke thrombus composition, its recently identified components, and how it affects stroke treatment. It also discusses how new insights could enhance current recanalization strategies for ischemic stroke patients.

Histone content, and thus DNA content, is associated with differential in vitro lysis of acute ischemic stroke clots

BACKGROUND

Fibrin, von Willebrand factor, and extracellular DNA from neutrophil extracellular traps all contribute to acute ischemic stroke thrombus integrity.

OBJECTIVES

In this study, we explored how the proteomic composition of retrieved thromboemboli relates to susceptibility to lysis with distinct thrombolytics.

METHODS

Twenty-six retrieved stroke thromboemboli were portioned into 4 segments, with each subjected to 1 hour of in vitro lysis at 37 °C in 1 of 4 solutions: tissue plasminogen activator (tPA), tPA + von Willebrand factor-cleaving ADAMTS-13, tPA + DNA-cleaving deoxyribonuclease (DNase) I, and all 3 enzymes. Lysis, characterized by the percent change in prelysis and postlysis weight, was compared across the solutions and related to the corresponding abundance of proteins identified on mass spectrometry for each of the thromboemboli used in lysis.

RESULTS

Solutions containing DNase resulted in approximately 3-fold greater thrombolysis than that with the standard-of-care tPA solution (post hoc Tukey, P < .01 for all). DNA content was directly related to lysis in solutions containing DNase (Spearman's ρ > 0.39 and P < .05 for all significant histones) and inversely related to lysis in solutions without DNase (Spearman's ρ < -0.40 and P < .05 for all significant histones). Functional analysis suggests distinct pathways associated with susceptibility to thrombolysis with tPA (platelet-mediated) or DNase (innate immune system-mediated).

CONCLUSION

This study demonstrates synergy of DNase and tPA in thrombolysis of stroke emboli and points to DNase as a potential adjunct to our currently limited selection of thrombolytics in treating acute ischemic stroke.

Inflammation biomarkers in the intracranial blood are associated with outcome in patients with ischemic stroke

BACKGROUND

Performing endovascular treatment (EVT) in patients with acute ischemic stroke (AIS) allows a port of entry for intracranial biological sampling.

OBJECTIVE

To test the hypothesis that specific immune players are molecular contributors to disease, outcome biomarkers, and potential targets for modifying AIS.

METHODS

We examined 75 subjects presenting with large vessel occlusion of the anterior circulation and undergoing EVT. Intracranial blood samples were obtained by microcatheter aspiration, as positioned for stent deployment. Peripheral blood samples were collected from the femoral artery. Plasma samples were quality controlled by electrophoresis and analyzed using a Mesoscale multiplex for targeted inflammatory and vascular factors.

RESULTS

We measured 37 protein biomarkers in our sample cohort. Through multivariate analysis, adjusted for age, intravenous thrombolysis, pretreatment National Institutes of Health Stroke Scale and Alberta Stroke Program Early CT scores, we found that post-clot blood levels of interleukin-6 (IL-6) were significantly correlated (adjusted P value <0.05) with disability assessed by the modified Rankin Scale (mRS) score at 90 days, with medium effect size. Chemokine (C-C) ligand 17 CCL17/TARC levels were inversely correlated with the mRS score. Examination of peripheral blood showed that these correlations did not reach statistical significance after correction. Intracranial biomarker IL-6 level was specifically associated with a lower likelihood of favorable outcome, defined as a mRS score of 0-2.

CONCLUSIONS

Our findings show a signature of blood inflammatory factors at the cerebrovascular occlusion site. The correlations between these acute-stage biomarkers and mRS score outcome support an avenue for add-on and localized immune modulatory strategies in AIS.

Stroke emboli from patients with atrial fibrillation enriched with neutrophil extracellular traps

Background

Recent literature has demonstrated remarkable heterogeneity in the composition of acute ischemic stroke (AIS) emboli, which may impact susceptibility to therapy.

Objectives

In this study, we explored differences in proteomic composition of retrieved embolic material from patients with stroke with and without atrial fibrillation (AF) (AF+ and AF-, respectively).

Methods

The full proteome of retrieved thromboembolic material from 24 patients with AIS was obtained by mass spectrometry. Known marker proteins were assigned groups representing broad classes of embolus components: red blood cells, platelets, neutrophils, eosinophils, histones, complement, and other clotting-associated proteins (eg, fibrinogen). Relative protein abundances were compared between AF+ and AF- samples. Functional implications of differences were explored with gene set enrichment analysis and Gene Ontology enrichment analysis and visualization tool.

Results

One hundred sixty-six proteins were differentially expressed between AF+ and AF- specimens. Eight out of the 15 neutrophil proteins (P < .05; fold change, >2) and 4 of the 14 histone proteins were significantly enriched in AF+ emboli (P < .05; fold change, >2). Gene set enrichment analysis revealed a significant representation of proteins from published neutrophil extracellular trap (NET) proteomic gene sets. The most significantly represented functional Gene Ontology pathways in patients with AF involved neutrophil activation and degranulation (P < 1 × 10-7).

Conclusion

The present analysis suggests enrichment of NETs in emboli of patients with stroke and AF. NETs are a significant though understudied structural component of thrombi. This work suggests not only unique stroke biology in AF but also potential therapeutic targets for AIS in this population.

CT hyperdense cerebral artery sign reflects distinct proteomic composition in acute ischemic stroke thrombus

BACKGROUND

Hyperdense cerebral artery sign (HCAS) is an imaging biomarker in acute ischemic stroke (AIS) that has been shown to be associated with various clinical outcomes and stroke etiology. While prior studies have correlated HCAS with histopathological composition of cerebral thrombus, it is unknown whether and to what extent HCAS is also associated with distinct clot protein composition.

METHODS

Thromboembolic material from 24 patients with AIS were retrieved via mechanical thrombectomy and evaluated with mass spectrometry in order to characterize their proteomic composition. Presence (+) or absence (-) of HCAS on preintervention non-contrast head CT was then determined and correlated with thrombus protein signature with abundance of individual proteins calculated as a function HCAS status.

RESULTS

24 clots with 1797 distinct proteins in total were identified. 14 patients were HCAS(+) and 10 were HCAS(-). HCAS(+) were most significantly differentially abundant in actin cytoskeletal protein (P=0.002, Z=2.82), bleomycin hydrolase (P=0.007, Z=2.44), arachidonate 12-lipoxygenase (P=0.004, Z=2.60), and lysophospholipase D (P=0.007, Z=2.44), among other proteins; HCAS(-) clots were differentially enriched in soluble N-ethylmaleimide-sensitive factor (NSF) attachment protein (P=0.0009, Z=3.11), tyrosine-protein kinase Fyn (P=0.002, Z=2.84), and several complement proteins (P<0.05, Z>1.71 for all), among numerous other proteins. Additionally, HCAS(-) thrombi were enriched in biological processes involved with plasma lipoprotein and protein-lipid remodeling/assembling, and lipoprotein metabolic processes (P<0.001), as well as cellular components including mitochondria (P<0.001).

CONCLUSIONS

HCAS is reflective of distinct proteomic composition in AIS thrombus. These findings suggest that imaging can be used to identify mechanisms of clot formation or maintenance at the protein level, and might inform future research on thrombus biology and imaging characterization.

Mollet I, Mendonça M, Salavisa M, Meira B, Fernandes M, Serrazina F, Cabral G, Ventura R, Sobral‐Pinho A, Beck HC, Vieira HLA, Viana‐Baptista M, Matthiesen R. Proteomics to Identify New Blood Biomarkers for Diagnosing Patients With Acute Stroke

BACKGROUND

Blood biomarkers are a potential tool for early stroke diagnosis. We aimed to perform a pilot and exploratory study on untargeted blood biomarkers in patients with suspected stroke by using mass spectrometry analysis.

METHODS AND RESULTS

This was a prospective observational study of consecutive patients with suspected stroke admitted within 6 hours of last being seen well. Blood samples were collected at admission. Patients were divided into 3 groups: ischemic stroke (IS), intracerebral hemorrhage (ICH), and stroke mimics. Quantitative analysis from mass spectrometry data was performed using a supervised approach. Biomarker-based prediction models were developed to differentiate IS from ICH and ICH+stroke mimics. Models were built aiming to minimize misidentification of patients with ICH as having IS. We included 90 patients, one-third within each subgroup. The median age was 71 years (interquartile range, 57-81 years), and 49 participants (54.4%) were women. In quantitative analysis, C3 (complement component 3), ICAM-2 (intercellular adhesion molecule 2), PLGLA (plasminogen like A), STXBP5 (syntaxin-binding protein 5), and IGHV3-64 (immunoglobulin heavy variable 3-64) were the 5 most significantly dysregulated proteins for both comparisons. Biomarker-based models showed 88% sensitivity and 89% negative predictive value for differentiating IS from ICH, and 75% sensitivity and 95% negative predictive value for differentiating IS from ICH+stroke mimics. ICAM-2, STXBP5, PLGLA, C3, and IGHV3-64 displayed the highest importance score in our models, being the most informative for identifying patients with stroke.

CONCLUSIONS

In this proof-of-concept and exploratory study, our biomarker-based prediction models, including ICAM-2, STXBP5, PLGLA, C3, and IGHV3-64, showed 75% to 88% sensitivity for identifying patients with IS, while aiming to minimize misclassification of ICH. Although our methodology provided an internal validation, these results still need validation in other cohorts and with different measurement techniques.

Gene expression profiles of ischemic stroke clots retrieved by mechanical thrombectomy are associated with disease etiology

BACKGROUND

Determining stroke etiology is crucial for secondary prevention, but intensive workups fail to classify ~30% of strokes that are cryptogenic.

OBJECTIVE

To examine the hypothesis that the transcriptomic profiles of clots retrieved during mechanical thrombectomy are unique to strokes of different subtypes.

METHODS

We isolated RNA from the clots of 73 patients undergoing mechanical thrombectomy. Samples of sufficient quality were subjected to 100-cycle, paired-end RNAseq, and transcriptomes with less than 10 million unique reads were excluded from analysis. Significant differentially expressed genes (DEGs) between subtypes (defined by the Trial of Org 10 172 in Acute Stroke Treatment) were identified by expression analysis in edgeR. Gene ontology enrichment analysis was used to study the biologic differences between stroke etiologies.

RESULTS

In all, 38 clot transcriptomes were analyzed; 6 from large artery atherosclerosis (LAA), 21 from cardioembolism (CE), 5 from strokes of other determined origin, and 6 from cryptogenic strokes. Among all comparisons, there were 816 unique DEGs, 174 of which were shared by at least two comparisons, and 20 of which were shared by all three. Gene ontology analysis showed that CE clots reflected high levels of inflammation, LAA clots had greater oxidoreduction and T-cell processes, and clots of other determined origin were enriched for aberrant platelet and hemoglobin-related processes. Principal component analysis indicated separation between these subtypes and showed cryptogenic samples clustered among several different groups.

CONCLUSIONS

Expression profiles of stroke clots were identified between stroke etiologies and reflected different biologic responses. Cryptogenic thrombi may be related to multiple etiologies.

Advancing Stroke Research on Cerebral Thrombi with Omic Technologies

Cerebrovascular diseases represent a leading cause of disability, morbidity, and death worldwide. In the last decade, the advances in endovascular procedures have not only improved acute ischemic stroke care but also conceded a thorough analysis of patients' thrombi. Although early anatomopathological and immunohistochemical analyses have provided valuable insights into thrombus composition and its correlation with radiological features, response to reperfusion therapies, and stroke etiology, these results have been inconclusive so far. Recent studies applied single- or multi-omic approaches-such as proteomics, metabolomics, transcriptomics, or a combination of these-to investigate clot composition and stroke mechanisms, showing high predictive power. Particularly, one pilot studies showed that combined deep phenotyping of stroke thrombi may be superior to classic clinical predictors in defining stroke mechanisms. Small sample sizes, varying methodologies, and lack of adjustments for potential confounders still represent roadblocks to generalizing these findings. However, these techniques hold the potential to better investigate stroke-related thrombogenesis and select secondary prevention strategies, and to prompt the discovery of novel biomarkers and therapeutic targets. In this review, we summarize the most recent findings, overview current strengths and limitations, and present future perspectives in the field.

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

Background and Aims

Besides the crucial role in the treatment of acute ischemic stroke (AIS), mechanical thrombectomy represents a unique opportunity for researchers to study the retrieved clots, with the possibility of unveiling biological patterns linked to stroke pathophysiology and etiology. We aimed to develop a shotgun proteomic approach to study and compare the proteome of formalin-fixed paraffin-embedded (FFPE) cardioembolic and large artery atherosclerotic (LAA) clots.

Methods

We used 16 cardioembolic and 15 LAA FFPE thrombi from 31 AIS patients. The thrombus proteome was analyzed by label-free quantitative liquid chromatography-tandem mass spectrometry (LC-MS/MS). MaxQuant v1.5.2.8 and Perseus v.1.6.15.0 were used for bioinformatics analysis. Protein classes were identified using the PANTHER database and the STRING database was used to predict protein interactions.

Results

We identified 1,581 protein groups as part of the AIS thrombus proteome. Fourteen significantly differentially abundant proteins across the two etiologies were identified. Four proteins involved in the ubiquitin-proteasome pathway, blood coagulation or plasminogen activating cascade were identified as significantly abundant in LAA clots. Ten proteins involved in the ubiquitin proteasome-pathway, cytoskeletal remodeling of platelets, platelet adhesion or blood coagulation were identified as significantly abundant in cardioembolic clots.

Conclusion

Our results outlined a set of 14 proteins for a proof-of-principle characterization of cardioembolic and LAA FFPE clots, advancing the proteome profile of AIS human thrombi and understanding the pathophysiology of ischemic stroke.

Neurovascular unit dysfunction as a mechanism of seizures and epilepsy during aging

The term neurovascular unit (NVU) describes the structural and functional liaison between specialized brain endothelium, glial and mural cells, and neurons. Within the NVU, the blood‐brain barrier (BBB) is the microvascular structure regulating neuronal physiology and immune cross‐talk, and its properties adapt to brain aging. Here, we analyze a research framework where NVU dysfunction, caused by acute insults or disease progression in the aging brain, represents a converging mechanism underlying late‐onset seizures or epilepsy and neurological or neurodegenerative sequelae. Furthermore, seizure activity may accelerate brain aging by sustaining regional NVU dysfunction, and a cerebrovascular pathology may link seizures to comorbidities. Next, we focus on NVU diagnostic approaches that could be tailored to seizure conditions in the elderly. We also examine the impending disease‐modifying strategies based on the restoration of the NVU and, more in general, the homeostatic control of anti‐ and pro‐inflammatory players. We conclude with an outlook on current pre‐clinical knowledge gaps and clinical challenges pertinent to seizure onset and conditions in an aging population.

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

Mechanical thrombectomy renders the occluding clot available for analysis. Insights into thrombus composition could help establish the stroke cause. We aimed to investigate the value of clot composition analysis as a complementary diagnostic tool in determining the etiology of large vessel occlusion (LVO) ischemic strokes (International Prospective Register of Systematic Reviews [PROSPERO] registration # CRD42020199436). Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, we ran searches on Medline (using the PubMed interface) and Web of Science for studies reporting analyses of thrombi retrieved from LVO stroke patients subjected to mechanical thrombectomy (January 1, 2006 to September 21, 2020). The PubMed search was updated weekly up to February 22, 2021. Reference lists of included studies and relevant reviews were hand-searched. From 1,714 identified studies, 134 eligible studies (97 cohort studies, 31 case reports, and six case series) were included in the qualitative synthesis. Physical, histopathological, biological, and microbiological analyses provided information about the gross appearance, mechanical properties, structure, and composition of the thrombi. There were non-unanimous associations of thrombus size, structure, and composition (mainly proportions of fibrin and blood formed elements) with the Trial of Org 10172 in Acute Stroke Treatment (TOAST) etiology and underlying pathologies, and similarities between cryptogenic thrombi and those of known TOAST etiology. Individual thrombus analysis contributed to the diagnosis, mainly in atypical cases. Although cohort studies report an abundance of quantitative rates of main thrombus components, a definite clot signature for accurate diagnosis of stroke etiology is still lacking. Nevertheless, the qualitative examination of the embolus remains an invaluable tool for diagnosing individual cases, particularly regarding atypical stroke causes.

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

Thrombus characteristics are dependent on clot composition, but identification of the etiology based on histological analysis has proved inconclusive. Identification of proteomic signatures may help to differentiate between clots of different etiologies such as cardioembolic, large artery atherosclerotic, and other known etiologies, information that could enhance an individualized medicine approach to secondary stroke prevention. In this study, total protein extracts from cardioembolic (n=25) and large artery atherosclerotic (n=23) thrombus specimens were arrayed in quadruplicate on nitrocellulose slides and immunostained for 31 proteins using a Dako Autostainer (Agilent Technologies, Inc., Santa Clara, USA). We quantified 31 proteins involved in platelet and/or endothelial function, inflammation, oxidative stress, and metabolism. Pathway analysis showed more heterogeneity and protein network interactions in the cardioembolic clots but no specific correlations with clot etiology. Reverse-phase protein arrays are a powerful tool for assessing cellular interactions within the clot microenvironment and may enhance understanding of clot formation and origination. This tool could be further explored to help in identifying stroke etiology in large vessel occlusion patients with embolic stroke of an undetermined source.