ORIGINAL RESEARCH published: 29 October 2020 doi: 10.3389/fneur.2020.552470 Inflammatory Response of Ischemic Tolerance in Circulating Plasma: Preconditioning-Induced by Transient Ischemic Attack (TIA) Phenomena in Acute Ischemia Patients (AIS) Laura Colàs-Campàs 1 , Joan Farre 1,2 , Gerard Mauri-Capdevila 1,3 , Jessica Molina-Seguín 1 , Núria Aymerich 4 , Ángel Ois 5 , Jaume Roquer 5 , Silvia Tur 6 , María del Carmen García-Carreira 7 , Joan Martí-Fàbregas 8 , Antonio Cruz-Culebras 9 , Edited by: Tomás Segura 10 , Gloria Arque 1† and Francisco Purroy 1,3*† Thanh G. Phan, Monash Health, Australia 1 Clinical Neurosciences Group, Institut de Recerca Biomèdica de Lleida, Lleida, Spain, 2 Medical Laboratory, Hospital Universitari Arnau de Vilanova, Lleida, Spain, 3 Stroke Unit, Department of Neurology, Hospital Universitari Arnau de Vilanova, Reviewed by: Lleida, Spain, 4 Complejo Hospitalario de Navarra, Pamplona, Spain, 5 Hospital del Mar, Barcelona, Spain, 6 Hospital Son Christopher G. Sobey, Espases, Palma de Mallorca, Spain, 7 Corporació Sanitària Parc Taulí, Sabadell, Spain, 8 Hospital de la Santa Creu i Sant La Trobe University, Australia Pau, Barcelona, Spain, 9 Hospital Universitario Ramón y Cajal, Madrid, Spain, 10 Complejo Hospitalario Universitario de Lauren M. Sanders, Albacete, Albacete, Spain St Vincent’s Hospital, Australia *Correspondence: Francisco Purroy Introduction: Ischemic tolerance (IT) refers to a state where cells are resistant to fpurroygarcia@gmail.com the damaging effects caused by periods of ischemia. In a clinical scenario, the IT † These authors have contributed phenomenon would be activated by a recent transient ischemic attack (TIA) before an equally to this work and share senior ischemic stroke (IS). The characterization of inflammatory protein expression patterns will authorship contribute to improved understanding of IT. Specialty section: Methods: A total of 477 IS patients from nine hospitals, recruited between January This article was submitted to Stroke, 2011 and January 2016, were included in the current study and divided in three groups: a section of the journal 438 (91.9%) patients without previous TIA (group 1), 22 (4.6%) patients who suffered TIA Frontiers in Neurology 24 h before IS (group 2), and 17 (3.5%) patients who suffered TIA between 24 h and 7 Received: 17 April 2020 days prior to IS (group 3). An inflammatory biomarker panel (IL-6, NT-proBNP, hsCRP, Accepted: 30 September 2020 Published: 29 October 2020 hs-Troponin, NSE, and S-100b) on plasma and a cytokine antibody array was performed Citation: to achieve the preconditioning signature potentially induced by TIA phenomena. Primary Colàs-Campàs L, Farre J, outcome was modified rankin scale (mRs) score at 90 days. Mauri-Capdevila G, Molina-Seguín J, Aymerich N, Ois Á, Roquer J, Tur S, Results: Recent previous TIA was associated with better clinical outcome at 90 days García-Carreira MdC, (median mRS of group 1: 2.0 [1.0–4.0]; group 2: 2.0 [0.0–3.0]; group 3: 1.0 [0–2.5]; Martí-Fàbregas J, Cruz-Culebras A, Segura T, Arque G and Purroy F p = 0.086) and smaller brain lesion (group 1: 3.7 [0.7–18.3]; group 2: 0.8 [0.3–8.9]; group (2020) Inflammatory Response of 3: 0.6 [0.1–5.5] mL; p = 0.006). All inflammation biomarkers were down regulated in Ischemic Tolerance in Circulating the groups of recent TIA prior to IS compared to those who did not suffer a TIA events. Plasma: Preconditioning-Induced by Transient Ischemic Attack (TIA) Moreover, a cytokine antibody array revealed 30 differentially expressed proteins between Phenomena in Acute Ischemia the three groups. Among them, HRG1-alpha (Fold change 74.4 between group 1 and Patients (AIS). Front. Neurol. 11:552470. 2; 74.2 between group 1 and 3) and MAC-1 (Fold change 0.05 between group 1 and 2; doi: 10.3389/fneur.2020.552470 0.06 between group 1 and 3) expression levels would better stratify patients with TIA 7 Frontiers in Neurology | www.frontiersin.org 1 October 2020 | Volume 11 | Article 552470 Colàs-Campàs et al. Inflammation Response of Prior TIA days before IS. These two proteins showed an earlier inflammation profile that was not detectable by the biomarker panel. Conclusion: Inflammatory pathways were activated by transient ischemic attack, however the period of time between this event and a further ischemic stroke could be determined by a protein signature that would contribute to define the role of ischemic tolerance induced by TIA. Keywords: ischemic stroke, transient ischemic attack (TIA), plasma, ischemic preconditioning (IPC), endogenous neuroprotection, biomarker (BM) INTRODUCTION mainly released into circulation. Although the basic concept of inflammatory response after stroke is quite well-known, there is Ischemic tolerance (IT) refers to a state where cells are resistant still missing information and the concept is constantly evolving. to the damaging effects caused by periods of ischemia. IT was first Biomarkers are defined as cellular, biochemical or molecular identified in the heart (1), but it was subsequently found to also alterations that are objectively measurable in biological samples occur in the brain (2, 3). The IT mechanisms induced by brief such as human tissues, cells, or fluids; they are used as periods of ischemia appear complex; it is an active process that an indicator of a biological or clinical condition, often with triggers gene expression, metabolic signatures, and activation of potential diagnostic or prognostic value. Protein array is an signaling pathways (4, 5). The IT could be induced either by local, emerging technology, it deems to be a versatile and robust global and/or by remote ischemic preconditioning (IPC) (3, 6, 7). tool to detect and/or quantify a large number of proteins IPC consists of brief periods of ischemia that confers protection present in a complex biological sample. Serum contains against later episodes of prolonged tissue ischemia (6). IPC massive amounts of potentially pathophysiological information phenomenon has been evidenced both in vitro and in vivo models of ischemic stroke, so protein array of human plasma would (3, 8, 9). The IPC/IT paradigm allows study of neuroprotection provide a so-called signature which would significantly increase and transient ischemic attacks (TIAs) are experimental strategies diagnostic/prognostic accuracy. to study it. IT phenomenon is activated by TIA and it has a similar According to our understanding, the protein profile of the underlying mechanism as acute ischemic stroke (AIS) (10, 11). preconditioning signature induced by TIA phenomena has not In a clinical scenario, IT phenomenon is activated by a been characterized. The first aim of the current study was recent TIA before an AIS (12–16). An ischemic event is to determine the inflammatory protein expression of an acute characterized by the occlusion of an arterial vessel that supplies ischemic stroke patients cohort, all patients were assessed for blood to an area of the brain, resulting in a corresponding an inflammation panel of six biomarkers [interleukin-6 (IL-6), loss of neurological function. TIA consists in a brief episode of N-terminal pro-B type natriuretic peptide (NT-proBNP), high- neurological dysfunction caused by ischemia, without structural sensitivity C-reactive protein (hsCRP), high sensitive troponin damage, and it has the same underlying mechanism as AIS (hsTroponin), neuron-specific enolase (NSE), S-100b protein]. (17). Thus, patients with previous TIA before an AIS had Moreover, the second aim was to assess a biomarker discovery by better outcome than patients without previous TIA (12–16, 18). a cytokine antibody array on a cohort of acute ischemic stroke Moreover, a beneficial effect of TIAs on final infarct volumes patients. For that, the differential expression profile of 1,000 has been demonstrated (12, 18). Similarly, ischemic presentations inflammation cytokines was performed by an antibody array- prior to acute myocardial infarction (AMI) are associated with based technology on AIS patients with and without previous lower short-term mortality after AMI, suggesting a natural recent TIA phenomena. ischemic preconditioning effect (19). Inflammation plays a dual role in the ischemic stroke’s MATERIALS AND METHODS pathophysiology: tissue repair and secondary brain injury effects. Inflammatory response is uniform in ischemic stroke, including Study Participants—Cohorts Description activation of microglia and endothelial cells by oxidative stress A cohort of suspected acute ischemic stroke (AIS) patients and excitotoxicity, infiltration of peripheral leukocytes in affected (n = 756) was consecutively prospective recruited between areas and causing tissue damage (20, 21). Proinflammatory January 2011 and January 2013 from the Hospital Universitari factors leads to upregulation of cell adhesion molecules on Arnau de Vilanova (HUAV) (Lleida, Spain), and between January endothelial cells causing inflow of blood to the ischemic area 2014 and January 2016 from nine Hospitals in Spain (Hospital and further activation of microglia and astrocytes. Cytokines Universitari Arnau de Vilanova-Lleida, Hospital Universitario- are immunomodulating agents and they play a major role in Pamplona, Hospital del Mar-Barcelona, Hospital de son Espases- cell activation, proliferation, and differentiation (22), they are Mallorca, Hospital Parc Taulí-Sabadell, Hospital de Sant Pau- barely detectable in the brain and they provoke and aggravate Barcelona, Hospital Ramón y Cajal-Madrid and Hospital de inflammatory response after stroke (23). Proinflammatory Albacete). The same recruitment protocol was followed at cytokines help to activate endothelial cells locally, but they are each site: TIA/AIS diagnose, neuroimaging confirmation and Frontiers in Neurology | www.frontiersin.org 2 October 2020 | Volume 11 | Article 552470 Colàs-Campàs et al. Inflammation Response of Prior TIA blood extraction. At Hospital admission, AIS patients were the Declaration of Helsinki developed by the World Medical attended by a Neurologist within 24 h from symptoms onset Association (WMA) regarding ethical principles for medical and underwent neuroimaging diagnosis with at least cranial research involving human subjects. computed tomography. Patients with symptoms that persisted < 24 h without clear evidence of acute ischemic lesion in the Inflammatory Biomarkers Panel neuroimaging were excluded from the study (17) (tissue criteria). IL-6, NT-proBNP, hsCRP, hs-Troponin, NSE, and S-100b The other exclusion criteria were a modified Rankin Scale Score levels were assessed in plasma samples (Hoffmann-La Roche, (mRS) > 3, duration of stroke symptoms more than 24 h and/or Basel, Switzerland) by an electrochemical chemiluminescence patients under 18 years old. immunoassay using the COBAS 6000 e601 (Hoffmann-La Roche, All blood samples were obtained within the first 24 h after Basel, Switzerland) at the Medical Laboratory of the HUAV. the onset of symptoms acute ischemic stroke symptoms at Three hundred microliter of plasma was used to perform the the Hospital admission at the emergency room by standard inflammatory biomarkers determination. venipuncture. Plasma, serum and buffy coat were obtained after centrifugation at 3,000 g at 4 ◦ C for 10 min, and aliquoted Inflammation Antibody Array into cryovials for immediate storage at −80 ◦ C (Plataforma To identify differentially expressed circulating proteins a cytokine Biobancos PT17/0015/0027). antibody array was performed on eight groups of plasma pools. Patients were classified etiologically according to TOAST The cytokines were detected using RayBio R Biotin Label-based criteria (Trial of Org 10,172 in Acute Stroke Treatment) (24) as Antibody Array (L-Series, RayBiotech Life, Peachtree Corners, large-artery occlusive disease (LAA), small-vessel disease (SV), GA), a quantitative array platform using multiplexed sandwich cardioembolic (CE), other cause (OC), or undetermined cause ELISA-based technology, which detect 1,000 proteins. (UND). The following patient’s characteristics were collected: In the group of stroke patients without previous TIA, 30 age at admission, gender, vascular risk factors, prior treatments, patients were selected by sex, age, and etiology. Three subgroups revascularization therapies (recombinant tissue plasminogen were established with ten samples per subgroup and samples activator: rt-PA; thrombectomy; rt-PA and thrombectomy; non- pooled within each subgroup. Same procedure was performed revascularization treatment), premorbid modified Rankin scale in the group of stroke with previous TIA 24 h before, where (mRS) and stroke severity by the National Institutes of Health fifteen samples were pooled in three subgroups, and 10 samples Stroke Scale (NIHSS) at admission, 24 h and 7 days. were pooled in the group stroke with previous TIA between According to previous evidence (17, 25), patients were 24 h and 7 days in two subgroups (Figure 1). Briefly, 10 µl of classified in three experimental groups defined by the time each patient plasma sample were pooled in the corresponding interval between recent TIA and ischemic stroke: (i) previous subgroup. For each pooled subgroup, 30 µl were used according TIA within 24 h, (ii) previous TIA between 24 h and 7 days, and to the manufacturer’s protocol. Finally, the signals were visualized (iii) stroke without previous TIA. by chemiluminescence (reading wave length 600PMT) and Previous TIA was defined as a reversible episode of then imaged by GenePix Personal 4100A microarray scanner neurological deficit of ischemic origin that resolved completely (Molecular Devices, San José, California, USA). The obtained within 24 h (26). Families of patients with language impairment pictures were analyzed with GenePix Pro Microarray Analysis or disturbances of consciousness and their general practitioners (Molecular Devices, San José, California, USA) program. were interviewed to identify prior TIA events. All baseline data was centrally monitored and queries were Data Analysis sent to the enrolling physicians of each center. The main Descriptive and frequency statistical analyses of demographic, clinical outcome was the degree of functional impairment etiological, and management variables; and inflammatory at 3 months measured by mRS at 90 days. Patients were biomarkers panel were performed with SPSS for Mac, version followed up by in-person interview in each recruiting Hospital. 20. Categorical variables were shown as frequencies and Moreover, final infarct volume was measured in 366 (76.7%) percentages, and continuous variables as means and standard patients by diffusion-weighted magnetic resonance imaging deviations. Inflammatory biomarkers were not normally (DWI) performed within 7 days (median 3.2 days: SD 1.8). distributed (P-P plot) and values were expressed as median One Neuroradiologist blinded to clinical features established the (interquartile range). presence of DWI abnormalities. Furthermore, OsiriX v4.0 semi- First, we compared demographic, etiological, and automated segmentation tool and imaging software (Bernex, management variables between the three groups attending Switzerland) (27) was used to calculate the total volume of the precedent of previous recent TIA. Statistical significance DWI abnormal signal intensity. Polygon tool was used to for intergroup differences was assessed by Pearson’s chi-square assess the presence of abnormalities, regions of interest (ROIs) test for categorical variables and the ANOVA t-test, Student’s were delineated between segmented slices and automatically t-test, Kruskal-Wallis test or Mann-Whitney test for continuous interpolated, and the infarct volume determined (28). variables. To calculate the sensitivity and specificity for This study protocol was approved by the Ethics Committee of biomarker cut-off values which allowed to discriminate AIS the HUAV and all participating Hospitals (approval code: 928). patients with previous recent TIA from AIS patients without Written informed consent was obtained from the patients or previous TIA, a receiver operator characteristic (ROC) analysis patients’ surrogates. All procedures followed the guidelines of was performed. Variables associated in the univariate analysis Frontiers in Neurology | www.frontiersin.org 3 October 2020 | Volume 11 | Article 552470 Colàs-Campàs et al. Inflammation Response of Prior TIA FIGURE 1 | Flow diagram of recruited patients, excluded patients, and number of final included patients in inflammation markers and antibody arrays. showing p < 0.10 were entered into a sequential logistic- Statistical significance was considered when p ≤ 0.05 and fold regression model to identify variables independently associated change (FC) criteria of FC > 2 or FC < 0.5. with previous recent TIA (odds ratios, 95% confidence interval). mRs at 90 days was the primary clinical endpoint. mRs ≤ RESULTS 3 was considered as good outcome. We compared patients’ characteristics between groups. In addition, simple logistic Clinical Characterization of Experimental regression analysis was performed to identify variables associated Groups/Cohort with the occurrence of good outcome (odds ratios, 95% Of 756 patients, 477 patients met the inclusion criteria (“stroke”), confidence interval). 22 (4.6%) patients had a previous TIA 24 h before the acute Gene Ontology (GO) (29) and pathway analysis (PANTER: ischemic stroke (“stroke with previous TIA ≤ 24 h”) and 17 protein annotation through evolutionary relationship) (30) (3.5%) patients had a previous TIA between 24 h and 7 days were conducted for proteins of interest using web-based before the acute ischemic stroke (“stroke with previous TIA tools (http://geneontology.org and http://www.pantherdb.org). 24 h−7 d”) (Figure 1). Frontiers in Neurology | www.frontiersin.org 4 October 2020 | Volume 11 | Article 552470 Colàs-Campàs et al. Inflammation Response of Prior TIA The average age was 74.3 (±11.6) years and 253 (53.0%) in two subgroups (5 samples each). Each group/subgroup of patients were female. Patients included in the group stroke pooled samples were balanced by sex, age, and etiology (Table 4). with previous TIA ≤ 24 h were significantly younger, they had A total of 27 proteins exhibit significantly different expression more smoking habits, they were more under renin-angiotensin between stroke and stroke with previous TIA ≤ 24 h (Table 5). Of system blockers (Table 1). LAA subtype was significantly more these, 26 proteins showed lower expression levels in patients with frequent among patients with recent previous TIA than in TIA ≤ 24 h prior to ischemic stroke. Only HRG1-alpha showed patients without. In contrast, CE was the most frequent etiologic higher expression levels in this group (fold change = 74.474). subtype in AIS without recent previous TIA. The comparison of stroke patients without previous TIA and Clinical characteristics of patients with previous TIA showed stroke patients with previous TIA between 24 h and 7 days a significant less severe stroke at admission by NIHSS’s score. before ischemic stroke showed five statically significant proteins: Infarct volumes of stroke with previous TIA ≤ 24 h group and AMPHIREGULIN, HRG1-alpha, MAC-1, ONCOSTATIN M, stroke with previous TIA 24 h−7 d group were significantly UBIQUITIN+1 (Table 6). Three of them were down-regulated reduced. Furthermore, outcome at 7 and 90 days measured by (AMPHIREGULIN, MAC-1, and ONCOSTATIN M) and two disability or dependence in daily activities (mRS) was better in up-regulated (HRG1-alpha and UBIQUITIN+1). Interestingly, previous TIA ≤ 7 d patients than non-previous TIA patients MAC-1 and HRG1-alpha showed a different relative expression (Table 1), these patients had better significant outcomes at 7 days level on both comparisons by prior TIA. The presence of a and a tendency at 90 days as well (p = 0.086), measured by previous TIA showed decreased levels of MAC-1 and increased NIHSS and mRs score. Table 2 shows variables associated with levels of HRG1-alpha in both time-windows (≤ 24 h and 24 h−7 mRS at 90 days < =3. In addition to previous recent TIA (Odds d). Finally, PANTHER was used to perform gene ontology (GO) ratio 2.98, 95% CI 1.04–8.58) sex male and small vessel disease analysis on the inflammation antibody array’s obtained results, etiology were related to better outcomes. In contrast, age, basal the significantly enriched molecular function was the transition NIHSS, previous atrial fibrillation and cardioembolic etiology metal ion binding and the major number of proteins were were associated to mRS > 3. associated with cytokine activity and receptor regulator activity. The enriched biological processes were cellular component disassembly and negative regulation of secretion; most of the Inflammatory Biomarkers Panel enriched proteins were related with negative regulation of cellular NSE, IL-6, NT-proBNP, S-100b, hsCRP levels, and hsTroponin processes, locomotion and regulation of cell death (Figure 3). showed significantly lower circulating levels in stroke patients with previous TIA ≤ 24 h (Figure 2). Different cut-off values were obtained for the biomarkers on different endpoints: TIA ≤ DISCUSSION 24 h and TIA between 24 h and 7 days. IL-6 < 5.1 pg/ml has a sensitivity of 68% and a specificity of 27% for TIA ≤ 24 h and a In our study we observed a clinical translation of the sensitivity of 65% and a specificity of 29% for TIA between 24 h preconditioning phenomena induced by an episode of TIA and 7 days. NT-proBNP < 82.25 pg/ml has a sensitivity of 64% before acute ischemic stroke (AIS). Recent previous TIA and a specificity of 13% for TIA ≤ 24 h. NT-proBNP < 203.6 was associated with better clinical outcome and smaller pg/ml has a sensitivity of 65% and a specificity of 28% for TIA brain lesion. Interestingly, we observed significant low levels between 24 h and 7 days. S100b < 41.4 has a sensitivity of 64% of inflammation-related biomarkers and using a cytokine and a specificity of 21% for TIA ≤ 24 h. S100b < 43.2 pg/ml with array technology we describe up to 30 proteins, which a sensitivity of 65% and a specificity of 24% for TIA between 24 h would be potential inflammatory biomarkers with differential and 7 days. HsTroponin < 6.05 pg/ml has a sensitivity of 64% and expression among ischemic stroke patients with previous a specificity of 15% for TIA ≤ 24 h. HsTroponin < 9.35 pg/ml has recent TIA. Only two of them were upregulated: HRG1- a sensitivity of 65% and a specificity of 28% for TIA between 24 h alpha and Ubiquitin+1. Interestingly, MAC-1, and HRG1- and 7 days. alpha were significantly expressed in both groups of TIA Table 3 shows the binary regression model to assess patients (TIA ≤ 24 h and TIA 24 h−7 d) and they might the relation of recent TIA prior to acute ischemic stroke. be potential biomarkers to stratify patients with previous Unfortunately, none of the assessed biomarkers showed a TIA events. statistically significant prediction of previous recent TIA. Prior TIA induces IT and it seems to activate an additional adaptive response based on inflammation biomarkers. The role of inflammation in the ischemic cascade after AIS is well- Inflammation Antibody Array known. Following ischemic stroke, microglia and astrocytes are A cytokine antibody array was performed in three groups activated within hours, leading to the production of cytokines of pooled patients to identify differentially expressed cytokine and chemokines and resulting in infiltration of leukocytes circulating proteins. Group 1 contained 30 ischemic stroke (IS) (31, 32). During the acute phase of stroke, lasting from minutes patients without previous TIA divided in three subgroups (10 to hours, the injured tissue releases reactive oxygen species samples each); group 2 contained 15 patients with TIA 24 h prior (ROS) and proinflammatory factors, such as chemokines and to IS divided in three subgroups (5 samples each), and group 3 cytokines, which induce the expression of adhesion molecules had 10 patients with TIA between 24 h and 7 days prior IS divided on leukocytes and on cerebral endothelial cells that in turn Frontiers in Neurology | www.frontiersin.org 5 October 2020 | Volume 11 | Article 552470 Colàs-Campàs et al. Inflammation Response of Prior TIA TABLE 1 | Clinical characteristics of all included patients by experimental groups. All Stroke (no Stroke with Stroke with Comparison Comparison Comparison previous TIA) previous previous between all between stroke between stroke TIA ≤ 24 h TIA 24 groups p-valuea vs. stroke with vs. stroke with h−7 d previous TIA ≤ TIA 24 h−7 d 24 h p-valuea p-valuea Total n 477 438 (91.8) 22 (4.6) 17 (3.5) Age mean ± SD years 74.3 ± 11.6 74.7 ± 11.4 67.82 ± 12.5 71.4 ± 11.3 0.013* 0.006* 0.232 Baseline characteristics Male sex 280 (53.0) 253 (57.8) 14 (63.6) 13 (76.5) 0.273 0.586 0.125 Alcoholism 34 (7.1) 32 (7.3) 2 (9.1) 0 (0.0) 0.502 0.675 0.618 Unknown 5 (1.0) 4 (0.9) 1 (5.9) Smoking habits 93 (19.5) 80 (18.3) 8 (36.36) 5 (29.41) 0.067 0.036* 0.252 Unknown 2 (0.4) 2 (0.5) Hypertension 357 (74.8) 332 (75.8) 14 (63.6) 11 (64.7) 0.260 0.190 0.289 Unknown 1 (0.2) 1 (0.2) Diabetes mellitus 129 (27.0) 120 (27.4) 5 (22.73) 4 (23.5) 0.835 0.622 0.717 Unknown 2 (0.4) 2 (0.5) Hyperlipidemia 209 (43.8) 192 (43.8) 12 (54.6) 5 (29.4) 0.292 0.324 0.239 Previous Atrial 117 (24.5) 111 (25.3) 5 (22.7) 1 (5.9) 0.181 0.774 0.085 fibrillation Unknown 2 (0.4) 2 (0.5) Ischemic heart disease 53 (11.1) 51 (11.6) 2 (9.1) 0 (0.0) 0.309 0.712 0.135 Unknown 1 (0.2) 1 (0.2) Previous treatment Antiplatelet treatment 177 (37.1) 167 (38.1) 4 (18.2) 6 (35.3) 0.163 0.071 0.808 Unknown 1 (0.2) 1 (0.2) Anticoagulation 60 (12.6) 56 (12.8) 3 (13.6) 1 (5.9) 0.692 0.933 0.387 Unknown 1 (0.2) 1 (0.2) Statins 167 (35.0) 154 (35.2) 10 (45.5) 3 (17.7) 0.191 0.329 0.194 Unknown 1 (0.2) 1 (0.2) Renin-angiotensin 153 (32.08) 142 (32.42) 6 (27.3) 5 (29.4) 0.899 0.030 0.788 system blockers Unknown 3 (0.63) 2 (0.46) 1 (4.55) Management 0.887 0.894 0.625 rt-PA administration 95 (19.9) 88 (20.1) 4 (18.2) 3 (17.7) rt-PA + Thrombectomy 33 (6.9) 32 (7.3) 1 (4.5) 0 (0.0) Thrombectomy alone 5 (1.0) 5 (1.1) 0 (0.0) 0 (0.0) Stroke severity NIHSS† 5 (3.0–11.0) 5 (3.0–12.0) 1 (0.0–4.3) 4 (0.5–5.5) < 0.001* < 0.001 0.033 At admission 24 h 3 (1.0–8.0) 4 (2.0–9.0) 1 (0.0–4.5) 2 (0.0–3.0) 0.002* 0.006 0.020 7 days 2 (0.0–5.0) 2 (0.0–5.0) 0.5 (0.0–3.5) 1 (0.0–2.0) 0.024* 0.061 0.039 mRs‡ 0 (0.0–1.0) 0 (0.0–1.0) 0 (0.0–1.0) 0 (0.0–0.8) 0.531 0.264 0.930 At admission 7 days 2 (1.0–4.0) 2 (1.0–4.0) 1.5 (0.0–3.0) 2 (0.0–3.0) 0.030* 0.045 0.070 90 days 2 (1.0–3.0) 2 (1.0–4.0) 2 (0.0–3.0) 1 (0.0–2.5) 0.086 0.263 0.050 mRs 90 days ≤ 3 355 (75.9) 320 (90.1) 19 (86.4) 16 (94.1) 0.091 0.213 0.050 Stroke Etiologies 0.037* 0.085 0.055 (TOAST) LAA 77 (16.1) 65 (14.8) 7 (31.8) 5 (29.4) CE 203 (42.6) 194 (44.3) 6 (27.3) 3 (17.7) (Continued) Frontiers in Neurology | www.frontiersin.org 6 October 2020 | Volume 11 | Article 552470 Colàs-Campàs et al. Inflammation Response of Prior TIA TABLE 1 | Continued All Stroke (no Stroke with Stroke with Comparison Comparison Comparison previous TIA) previous previous between all between stroke between stroke TIA ≤ 24 h TIA 24 groups p-valuea vs. stroke with vs. stroke with h−7 d previous TIA ≤ TIA 24 h−7 d 24 h p-valuea p-valuea Lacunar 72 (15.1) 62 (14.2) 5 (22.7) 5 (29.4) Undeterminate 115 (24.1) 109 (24.9) 3 (13.6) 3 (17.7) Other causes 10 (2.1) 8 (1.8) 1 (4.6) 1 (5.9) DWI vol. Median 2.8 (0.6–17.7) 3.7 (0.8–18.3) 0.8 (0.3–2.9) 0.6 (0.15.5) 0.006* 0.044 0.011 (IQR) mL Plus–minus values are means ± SD. Percentages are shown in parentheses as appropriate. Percentages may not total 100 because of rounding. † Scores on the National Institutes of Health Stroke Scale range from 0 to 42, with higher score indicating impairment after stroke. Values are presented as median and interquartile range. ‡ Scores on modified Rankin Score range from 0 to 5, with higher scores indicating major disability or dependence in the daily activities. Values are presented as median and interquartile range. a p-values correspond to the chi-square test for trend for ordered qualitative variables and the Pearson’s. chi-square test for non-ordered qualitative variables and have been obtained excluding the patients with missing values. *p ≤ 0.05. TIA, Transient ischemic attack; rt-PA, recombinant tissue plasminogen activator; mRS, modified Rankin Score; NIHSS, National Institutes of Health Stroke Scale; LAA, large-artery occlusive disease; CE, cardioembolic. TABLE 2 | Univariate analysis and simple logistic regression analysis of variables associated with outcomes at follow-up (90 days). All mRS > 3 mRS ≤ 3 p-valuea OR (95%CI) p-value Total n 468 175 (37.4) 293 (62.6) – – – Age mean ± SD years 74.3 ± 11.5 80.6 ± 9.3 72.3 ± 11.5 < 0.001 0.92 (0.89–0.94) < 0.001* Baseline characteristics Male sex 276 (59.9) 49 (43.4) 226 (63.7) < 0.001 2.29 (1.49–3.52) < 0.001* Hypertension 351 (75.2) 91 (80.5) 260 (73.4) 0.129 0.67 (0.40–1.13) 0.131 Unknown 1 (0.2) Diabetes mellitus 127 (27.3) 28 (25.0) 99 (28.0) 0.539 1.17 (0.72–1.90) 0.539 Unknown 2 (0.4) Hyperlipidemia 203 (43.4) 44 (38.9) 158 (44.5) 0.298 1.26 (0.82–1.94) 0.298 Previous Atrial fibrillation 114 (24.4) 37 (32.7) 77 (21.8) 0.018 0.57 (0.36–0.91) 0.019* Unknown 1 (0.2) Ischemic heart disease 52 (11.1) 12 (10.6) 40 (11.3) 0.849 1.07 (0.54–2.12) 0.849 Unknown 1 (0.2) Stroke with previous 39 (8.3) 4 (3.5) 35 (9.9) 0.034 2.98 (1.04–8.58) 0.043* TIA ≤ 7 days Management rt-PA administration 93 (19.9) 29 (25.7) 64 (18.0) 0.368 0.89 (0.65–1.22) 0.464 rt-PA + Thrombectomy 32 (6.8) 7 (6.2) 25 (7.0) Thrombectomy alone 5 (1.1) 1 (0.9) 4 (1.1) Stroke severity NIHSS at admission† 5.0 (3.0–11.0) 15.0 (9.0–20.0) 4.0 (2.0–8.0) < 0.001 0.82 (0.79–0.86) < 0.001* Basal mRS‡ 0.0 (0.0–1.0) 0.0 (0.0–1.0) 0.0 (0.0–0.0) < 0.001 0.55 (0.43–0.71) < 0.001* Stroke Etiologies (TOAST) LAA 77 (16.5) 17 (15.0) 60 (16.9) 0.643 1.15 (0.64–2.06) 0.643 CE 200 (42.7) 63 (55.8) 138 (38.9) 0.002 0.51 (0.33–0.77) 0.002* Small vessel 70 (15.0) 4 (3.5) 66 (18.6) < 0.001 6.22 (2.22–17.48) 0.001* Undeterminate 111 (23.7) 27 (23.9) 83 (23.4) 0.911 0.97 (0.59–1.60) 0.911 Other causes 10 (2.1) 2 (1.8) 8 (2.3) 0.757 1.28 (0.27–6.11) 0.757 DWI vol. Median (IQR) mL 2.8 (0.6–17.7) 38.6 (10.4–142.4) 1.6 (0.5–9.3) < 0.001 0.97 (0.96–0.98) < 0.001* Plus–minus values are means ± SD. Percentages are shown in parentheses as appropriate. Percentages may not total 100 because of rounding. † scores on the National Institutes of Health Stroke Scale range from 0 to 42, with higher score indicating impairment after stroke. Values are presented as median and interquartile range. ‡ Scores on modified Rankin Score range from 0 to 5, with higher scores indicating major disability or dependence in the daily activities. Values are presented as median and interquartile range. a p-values correspond to the chi-square test for trend for ordered qualitative variables and the Pearson’s. chi-square test for non-ordered qualitative variables and have been obtained excluding the patients with missing values. *p ≤ 0.05. TIA, Transient ischemic attack; rt-PA, recombinant tissue plasminogen activator; mRS, modified Rankin Score; NIHSS, National Institutes of Health Stroke Scale; LAA, large-artery occlusive disease; CE, cardioembolic. Frontiers in Neurology | www.frontiersin.org 7 October 2020 | Volume 11 | Article 552470 Colàs-Campàs et al. Inflammation Response of Prior TIA FIGURE 2 | Plasma levels of inflammatory markers according to preconditioning induced by TIA. The levels of NSE (A), IL-6 (B), NT-proBNP (C), S-100b (D), hsCRP (E), and hsTroponin (F) were determined by chemiluminescence immunoassay. Differences in levels of inflammatory markers were analyzed using non-parametric test with p < 0.05 being significant. Graphs represent median with interquartile range (gray color). Receiver operating characteristic (ROC) curve for stroke with previous TIA ≤ 24 h and stroke with previous TIA (24 h−7 d) were an insert for each biomarker. Area under the curve (AUC) for NSE is 0.359, p = 0.026 for previous TIA ≤ 24 h; and 0.507, p = 0.921 for previous TIA 24 h−7 d. AUC for IL6 is 0.358, p = 0.024 for previous TIA ≤ 24 h; and 0.415, p = 0.236 for previous TIA 24 h−7 d. AUC for NT proBNP is 0.303, p = 0.002 for previous TIA ≤ 24 h; and 0.408, p = 0.200 for previous TIA 24 h−7 d. AUC for S100B is 0.344, p = 0.014 for previous TIA ≤ 24 h; and 0.342, p = 0.027 for previous TIA 24 h−7 d. AUC for hsCRP is 0.374, p = 0.046 for previous TIA ≤ 24 h; and 0.529, p = 0.689 for previous TIA 24 h−7 d. AUC for hs troponin is 0.298, p = 0.001 for previous TIA ≤ 24 h; and 0.443, p = 0.427 for previous TIA 24 h−7 d. *p-value between 0.025 and 0.05; **p-value between 0.01 and 0.025; ***p < 0.01. promote adherence and trans-endothelial transfer of leukocytes IL-6 is an inflammatory cytokine upregulated by cerebral (33). It has been described how preconditioning, among ischemia (34), but it has also been described previously how other actions, induces the capacity of the brain for self- IL-6 levels among AIS patients with previous recent TIA are protection by suppressing inflammation and dampening post- lower than AIS patients without prior TIA (12). A recent meta- ischemic microglial activation (6). In addition, the inflammation analysis of 24 studies confirmed an unlikely translation of IL-6 effect plays a role in remote ischemic conditioning mediating into clinical practice for the prognosis of AIS (35); however, drugs neuroprotection against stroke (7). that decreases plasma levels of IL-6 like anakinra (a recombinant Frontiers in Neurology | www.frontiersin.org 8 October 2020 | Volume 11 | Article 552470 Colàs-Campàs et al. Inflammation Response of Prior TIA TABLE 3 | Binary regression model for pre-conditioning effect of TIA on acute ischemic patients. Stroke with previous TIA ≤ 24 h Stroke with previous TIA 24 h−7 d Variables OR (95% CI) p-valuea OR (95% CI) p-valuea Age 0.98 (0.94–1.02) 0.358 0.98 (0.94–1.03) 0.458 LAA 2.27 (0.87–5.96) 0.096 2.36 (0.79–7.05) 0.125 IL-6 < 5.1 pg/mL 0.65 (0.22–1.89) 0.429 1.22 (0.38–3.93) 0.744 NT-ProBNP < 82.25 pg/mL 1.90 (0.57–6.38) 0.300 – – NT-ProBNP < 203.6 pg/mL – – 0.86 (0.25–2.99) 0.808 S100b < 41.4 pg/mL 1.87 (0.59–5.91) 0.288 – – S100b < 43.15 pg/mL – – 1.40 (0.47–4.15) 0.548 hsTroponin < 6.05 pg/mL 1.87 (0.59–5.91) 0.288 – – hsTroponin < 9.35 pg/mL – – 1.03 (0.29–3.65) 0.960 a p ≤ 0.05. OR, Odds ratio; TIA, Transient ischemic attack; LAA, large-artery atherosclerosis; IL-6, interleukin-6; NT-proBNP, N-terminal pro-B type natriuretic peptide; hsTroponin, high sensitive troponin. IL-1 receptor antagonist) reduces infarct volume in experimental constitutively expressed at very low levels. Cytokines play a models (36). Our results showed a similar expression pattern, major role upregulating the expression of cell adhesion molecules specifically a reduction of IL-6 levels in the group of ischemic (CAM) (47, 48). IL-6, interleukin 1β (IL-1β) and tumor necrosis stroke with previous TIA up to 24 h. factor-alpha (TNF-α) are the major proinflammatory cytokines S100b levels were decreased in both groups with prior TIA that provoke and aggravate an inflammatory response after stroke before ischemic stroke. S100b could be found in both CSF (23, 49). Matrix metalloproteinases (MMPs) which are a large and blood serum of different brain diseases like brain tumors, family of proteolytic enzymes that degrade all components of neuroinflammatory and neurodegenerative disorders, psychiatric extracellular matrix were also identified (50). Although, MMPs disorders, cerebral infections, subarachnoid hemorrhage, acute act as proinflammatory factor, they are also important for brain injury, and cerebral infarction (37, 38). It is detected in normal physiological function such as neuronal regeneration, cell glial cells of the central nervous system such as oligodendrocytes, proliferation, angiogenesis, and apoptosis (51). In addition, in ependymocytes, astrocytes, neuronal progenitor cells, and others our study previous TIA was related to cytokines downregulation (12). In the brain ischemia cellular microenvironment, S100b is implicated in molecular functions and biological processes released as a molecule that triggers an inflammatory response previously described in the preconditioning phenomena like heat (39). For that, a reduction on S100b levels would be related with shock proteins (6, 52), receptor of protein tyrosine kinase activity a reduction on the inflammatory reaction. (53, 54), and the response to estradiol (55). Lower levels of hsTroponin and NT-ProBNP were observed HRG1-alpha is an alternative spliced variant of in patients with previous recent TIA. Elevated levels of both NEUROGULIN 1 (NRG1) gene, which is a glycoprotein biomarkers are associated with unfavorable course of stroke and that interacts with the ERBB2 receptor tyrosine kinase to etiological cardioembolic type (40–43). Recently, the American increase its phosphorylation on tyrosine residues. There is a lack Heart Association recommended to obtain a baseline troponin of literature on HRG1-alpha splice variant. However, NRG1 is level in all patients with AIS, because stroke-associated cardiac related with synaptic plasticity and neuroinflammation (56), dysfunction was detected by elevated troponin levels and nerve repair (57), peripheral nervous system development (58). associated with unfavorable short- and long-term outcomes after Neuroprotective and anti-inflammatory effects of NRG-1 are AIS (44). associated with the differential regulation of NF-kB signaling Our results showed that previous TIA was more frequent pathways in microglia (59, 60). MAC-1 gene encodes the integrin among large artery atherosclerosis (LAA) etiology, as described alpha M chain. Integrins are heterodimeric integral membrane previously (12, 14, 45). The overrepresentation of LAA etiology proteins composed of an alpha chain and a beta chain. Activation type and the smaller infarction size could interfere in our of the integrin MAC-1 is known to be critical for mediating obtained results. In the multivariate analysis, no biomarker neutrophil adhesion and migration. The expression of Mac-1 in emerged as an independent predictor of previous recent microglia is heterogeneous both under normal conditions and TIA, although it is well-known the role of inflammation in after stroke (61). the pathogenesis of atherosclerosis (46). This limitation was overcome by adjusting the groups of the inflammation antibody LIMITATIONS array by etiology and other variables, age and sex. Cytokines are immunomodulating agents and they play a The main limitation of the current study is the available TIA major role in cell activation, proliferation, and differentiation patients in our cohort [TIA patients = 8.1%], which have (22). Cytokines are generally small pleiotropic polypeptides lead us to use a pooling method of sample processing in the (8–26 kDa) barely detectable in the brain with their receptors inflammation antibody arrays, instead of the individual sample Frontiers in Neurology | www.frontiersin.org 9 October 2020 | Volume 11 | Article 552470 Frontiers in Neurology | www.frontiersin.org Colàs-Campàs et al. TABLE 4 | Clinical characteristics of included patients in the inflammation array experiment. Stroke (no previous TIA) Stroke and previous TIA < 24 h Stroke and previous TIA 24 h−7 d Intergroup (N = 30) (N = 15) (N = 10) differences Group 1 Group 2 Group 3 p-valuea Group 1 Group 2 Group 3 p-valuea Group 1 Group 2 p-valuea p-valuea Total n 10 10 10 5 5 5 5 5 Age mean ± SD years 70.0 ± 8.8 70.4 ± 7.5 67.5 ± 11.3 0.756 66.8 ± 16.4 66.6 ± 6.5 67.6 ± 9.4 0.990 70.0 ± 12.2 70.0 ± 16.3 1 0.704 Baseline characteristics Sex, male (%) 8 (80.0) 9 (90.0) 8 (80.0) 0.787 4 (80.0) 4 (80.0) 4 (80.0) 1 4 (80.0) 4 (80.0) 1 0.946 Alcoholism 1 (10.0) 3 (30.0) 2 (20.0) 0.535 0 (0.0) 1 (20.0) 0 (0.0) 0.343 0 (0.0) 0 (0.0) – 0.124 Smoking habit 2 (20.0) 2 (20.0) 5 (50.0) 0.276 1 (20.0) 2 (40.0) 2 (40.0) 0.741 2 (40.0) 1 (20.0) 0.114 0.921 Hypertension 9 (90.0) 8 (80.0) 8 (80.0) 0.787 3 (60.0) 2 (40.0) 4 (80.0) 0.435 3 (60.0) 4 (80.0) 0.490 0.198 Diabetes Mellitus 3 (30.0) 4 (40.0) 6 (60.0) 0.387 1 (20.0) 1 (20.0) 1 (20.0) 1 0 (0.0) 2 (40.0) 0.114 0.241 Hyperlipidemia 8 (80.0) 5 (50.0) 6 (60.0) 0.366 2 (40.0) 3 (60.0) 4 (80.0) 0.435 3 (60.0) 2 (40.0) 0.537 0.315 Previous Atrial fibrillation 2 (20.0) 2 (20.0) 1 (10.0) 0.787 1 (20.0) 0 (0.0) 1 (20.0) 0.562 1 (20.0) 0 (0.0) 0.292 0.649 Ischemic heart disease 1 (10.0) 3 (30.0) 4 (40.0) 0.303 1 (20.0) 0 (0.0) 1 (20.0) 0.562 0 (0.0) 0 (0.0) – 0.071 Previous treatment Antiplatelet 2 (20.0) 4 (40.0) 3 (30.0) 0.621 3 (60.0) 0 (0.0) 1 (20.00) 0.092 0 (0.0) 2 (40.0) 0.114 0.707 Anticoagulation 1 (10.0) 0 (0.0) 1 (10.0) 0.585 0 (0.0) 0 (0.0) 0 (0.00) – 1 (20.00) 0 (0.00) 0.292 0.591 Statins 6 (60.0) 4 (40.0) 4 (40.0) 0.585 2 (40.0) 3 (60.0) 3 (60.00) 0.765 1 (20.00) 2 (40.00) 0.490 0.295 Renin-angiotensin system 3 (30.00) 2 (20.00) 3 (30.00) 0.843 2 (40.00) 0 (0.00) 3 (60.00) 0.170 1 (20.00) 1 (20.00) 1 0.634 blockers 10 Management rt-PA 0 (0.0) 2 (20.0) 0 (0.0) 0.307 0 (0.0) 1 (20.0) 1 (20.0) 0.552 0 (0.0) 0 (0.0) 1 0.500 rt-PA + Thrombectomy 2 (20.0) 2 (20.0) 1 (10.0) 0 (0.0) 1 (20.0) 0 (0.0) 0 (0.0) 0 (0.0) Thrombectomy 0 (0.00) 0 (0.0) 1 (10.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) Stroke severity and outcome NIHSS Basal† 4.5 (3.0–9.8) 4 (0.0–11.8) 4 (2.8–6.5) 0.644 0 (0.0–6.0) 1 (0.5–11.5) 3 (0.0–6.5) 0.688 0 (0.0–2.5) 3 (0.0–9.0) 0.421 0.086 NIHSS 24 h 3.5 (1.8–7.3) 2 (0.0–4.0) 4.5 (2.5–8.0) 0.268 0 (0.0–2.5) 0.5 (0.0–6.3) 2 (0.0–4.5) 0.767 0 (0.0–1.5) 1 (0.0–11.0) 0.421 0.016* NIHSS 7 days 2.5 (1.00–7.25) 1.5 (0.0–3.0) 3 (1.0–5.0) 0.212 0 (0.0–2.5) 1 (0.0–4.5) 0 (0.0–3.5) 0.865 0 (0.00–1.5) 0 (0.00–9.5) 1 0.059 mRs basal‡ 0 (0.0–1.3) 0 (0.0–0.5) 0 (0.0–0.3) 0.819 0 (0.0–0.0) 0 (0.0–0.0) 0 (0.0–0.0) 0.472 0 (0.0–1.5) 0.5 (0.0–1.8) 0.343 0.790 mRS 7 days 3 (1.0–4.0) 1 (1.0–3.0) 2 (1.0–4.0) 0.258 0 (0.0–2.5) 1 (0.0–2.50) 0 (0.0–3.0) 0.939 0 (0.0–2.0) 0 (0.0–4.0) 0.690 0.018 mRS 90 days 3 (0.8–4.0) 1 (0.0–1.5) 1 (0.75–3.25) 0.259 0 (0.0–2.0) 0 (0.0–2.5) 2 (0.0–3.0) 0.578 1 (0.0–2.0) 0 (0.00–2.5) 0.841 0.160 Stroke Etiologies (TOAST) LAA 4 (40.00) 5 (50.00) 3 (30.00) 0.959 2 (40.00) 2 (40.00) 2 (40.00) 1 2 (40.00) 1 (20.00) 0.881 0.999 October 2020 | Volume 11 | Article 552470 CE 2 (20.00) 2 (20.00) 2 (20.00) 1 (20.00) 1 (20.00) 1 (20.00) 1 (20.00) 1 (20.00) Lacunar 2 (20.00) 2 (20.00) 2 (20.00) 1 (20.00) 1 (20.00) 1 (20.00) 1 (20.00) 2 (40.00) Inflammation Response of Prior TIA Indeterminate 2 (20.00) 1 (10.00) 3 (30.00) 1 (20.00) 1 (20.00) 1 (20.00) 1 (20.00) 1 (20.00) DWI vol. Median (IQR) mL 3.3 (0.3–41.3) 0.6 (0.3–57.5) 1.7 (0.5–4.4) 0.963 1.3 (0.3–2.7) 0.8 (0.2–6.5) 0.5 (0.1–7.3) 0.564 0.8 (0.1–14.4) 0.1 (0.1–68.7) 0.548 0.083 Plus–minus values are means ± SD. Percentages are shown in parentheses as appropriate. a p-values correspond to the chi-square test for trend for ordered qualitative variables and the Pearson’s. chi-square test for non-ordered qualitative variables and have been obtained excluding the patients with missing values. *p ≤ 0.05. TIA, Transient ischemic attack; rt-PA, recombinant tissue plasminogen activator; mRS, modified Rankin Score; NIHSS, National Institutes of Health Stroke Scale; LAA, large-artery occlusive disease; CE, cardioembolic; DWI, diffusion weighted imaging. † Scores on the National Institutes of Health Stroke Scale range from 0 to 42, with higher score indicating impairment after stroke. Values are presented as median and interquartile range. ‡ Scores on modified Rankin Score range from 0 to 5, with higher scores indicating major disability or dependence in the daily activities. Values are presented as median and interquartile range. Colàs-Campàs et al. Inflammation Response of Prior TIA TABLE 5 | Differentially expressed proteins in acute ischemic stroke patients (no previous TIA) and stroke patients with previous TIA within 24 h. Protein (relative expression) Stroke Stroke with previous p-valuea Fold-changeb Regulation (TIA ≤ (no previous TIA) TIA ≤ 24 h 24 h vs. stroke) ACTIVIN B 144.054 23.629 0.042 0.164 Down ANGIOPOIETIN-4 178.218 12.983 0.013 0.073 Down ALK-6 7519.000 384.996 0.041 0.051 Down CCR9 10713.000 856.717 0.032 0.080 Down CRIM 1 134.054 12.783 0.007 0.095 Down FGF R4 6885.000 1314.000 0.032 0.191 Down WFIKKNRP 211.300 38.228 0.032 0.181 Down HRG1-alpha 93.684 6977.000 0.034 74.474 Up IFN-alpha 108.243 10.848 0.001 0.100 Down IGFBP-7 157.790 3.920 0.008 0.025 Down IL-1 F10 124.106 17.655 0.012 0.142 Down IL-17 116.802 19.712 0.015 0.169 Down IL-3 7520.000 69.126 0.008 0.009 Down IL-6 R 235.073 77.546 0.019 0.330 Down LEPTIN (OB) 10853.000 291.024 0.018 0.027 Down LIPOCALIN-2 10108.000 478.070 0.041 0.047 Down MAC-1 6955.000 378.802 0.004 0.054 Down MMP-12 136.681 21.665 0.003 0.159 Down MMP-15 169.020 10.369 0.014 0.061 Down MMP-19 149.643 14.739 0.022 0.098 Down GPNMB 2670.000 91.088 0.022 0.034 Down CCL25 124.901 4.834 0.004 0.039 Down TIMP-1 255.223 6.519 0.005 0.026 Down TNFRSF1B 87.092 7.174 0.028 0.082 Down TNFSF10 82.890 3.259 0.026 0.039 Down TNFRSF10D 86.505 9.687 0.007 0.112 Down VEGF R2 (KDR) 121.142 4.217 0.013 0.035 Down a Mann-Whitney U-test and Bonferroni adjustment. Statistical significance was considered when p-value < 0.05. b Fold-change analysis. Statistical significance was considered when FC > 2 or FC < 0.5. ALK-6, Activin-like Kinase; BMPR-IB, bone morphogenetic protein receptor IB; CCR9, C-C motif chemokine receptor 9; CRIM 1, cysteine rich transmembrane BMP regulator 1; FGF R4, fibroblast growth factor receptor 4; GASP-1, G protein-coupled receptor associated sorting protein 1; GPNMB, Glycoprotein non-metastatic melanoma B; HRG-alpha, histidine- rich glycoprotein alpha; IGFBP, insulin like growth factor binding protein; IL, inteleukin; IFN-alpha, interferon-alpha; MAC-1, macrophage-1 antigen; MMP, metalloproteinase; TECK, thymus-expressed chemokine; TIMP-1, tissue inhibitor of matrix metalloproteinase-1; TNFRSF1B, TNF receptor superfamily member 1B; TNFSF10, TNF superfamily member 10; TRAIL, TNF-related apoptosis-inducing ligand; TIA, Transient ischemic attack; TNF, tumor necrosis factor; VEGF, vascular endothelial growth factor; WFIKKNRP, WAP, Follistatin/kazal, Immunoglobulin, Kunitz, and Netrin domain-containing 2. TABLE 6 | Differentially expressed proteins on acute ischemic stroke patients (no previous TIA) and stroke patients with previous TIA between 24 h and 7 days. Protein Stroke Stroke with previous p-valuea Fold-changeb Regulation (TIA 24 (relative expression) (no previous TIA) TIA 24 h−7 d h−7 d vs. stroke) AMPHIREGULIN 149.731 25.127 0.002 0.168 Down HRG1 ALPHA 93.684 6955.000 0.049 74.239 Up MAC-1 6955.000 427.177 0.025 0.061 Down ONCOSTATIN M 4116.000 897.950 0.029 0.218 Down UBIQUITIN+1 182.943 3592.000 0.001 19.635 Up a Mann-Whitney U-test and Bonferroni adjustment. Statistical significance was considered when p-value < 0.05. b Fold-change analysis. Statistical significance was considered when FC > 2 or FC < 0.5. TIA, Transient ischemic attack; MAC-1, macrophage-1 antigen; HRG-alpha, heuregulin 1-alpha. Frontiers in Neurology | www.frontiersin.org 11 October 2020 | Volume 11 | Article 552470 Colàs-Campàs et al. Inflammation Response of Prior TIA FIGURE 3 | Results of GO Enrichment analysis on the set of 30 differentially expressed proteins. Terms are sorted according to P-value with the most significantly enriched terms at the top: GO terms for molecular functions (GO:FM, left, blue); GO terms for biological processes (GO:PB, right, green). The length of each bar correlates with P-value while the width indicates the number of proteins in the set associated with the term. Those functions or process that only one protein was involved were discarded. processing. As in other human diseases, the pathophysiological by other techniques. Finally, the etiology and the infarction size pathway of AIS is complex and heterogeneous. In order to of TIA groups should be taken into account and its effects on the confirm our findings, it would be beneficial to complete the final results dissected. study by single protein expression determination of potential candidate biomarkers in individual plasma samples in a larger DATA AVAILABILITY STATEMENT cohort. Information of previous TIA among aphasic or severe AIS is difficult to obtain, although close familiars and caregivers The raw data supporting the conclusions of this article will be were asked about it, some information might be lacking. made available by the authors, without undue reservation. CONCLUSIONS ETHICS STATEMENT Previous recent TIA before AIS is associated with better The studies involving human participants were reviewed and clinical and imaging outcomes than AIS without previous TIA. approved by Comité ètica de l’Hospital Universitari Arnau de This fact could be explained by the induction of ischemic Vilanova de Lleida. The patients/participants provided their tolerance phenomena that is related to an inflammatory written informed consent to participate in this study. plasma cytokine signature. The determination of previous described biomarkers like IL-6 and S100b confirmed a specific AUTHOR CONTRIBUTIONS inflammation signature of ischemic tolerance. Interestingly, the antibody array technology emerged as a plausible strategy to FP conceived the study and procured funding. LC-C, JF, and identify the effect to the inflammatory response related to brain FP designed experiments. GM-C, JM-S, NA, ÁO, JR, ST, MG-C, ischemia and new biomarkers or targets for neuroprotection. JM-F, AC-C, and TS patients’ recruitment and clinical data However, our results need to be confirmed in larger cohorts and acquisition. LC-C, JF, GA, and FP sample processing and data Frontiers in Neurology | www.frontiersin.org 12 October 2020 | Volume 11 | Article 552470 Colàs-Campàs et al. Inflammation Response of Prior TIA analysis. GA and FP wrote the paper. All authors commented on supported by IRBLleida Biobank (B.0000682) and Plataforma and approved submission of this manuscript. Biobancos PT17/0015/0027. FUNDING ACKNOWLEDGMENTS This study was supported by the Government of Catalonia- We thank all recruited patients, as well as the members of Agència de Gestió d’Ajuts Universitaris i de Recerca (FP: 2017 Clinical Neuroscience group at IRBLleida, and personnel SGR 1628), Instituto de Salud Carlos III and co-funded by of Neurology Department at Hospital Universitari Arnau European Union (ERDF/ESF, Investing in your future and A way de Vilanova de Lleida for scientific discussions and to build Europe) (FP: PI17-01725 and PI14/01574). This was also instrumental help. REFERENCES 16. Arboix A, Cabeza N, Garcia-Eroles L, Massons J, Oliveres M, Targa C, et al. Relevance of transient ischemic attack to early neurological 1. Murry CE, Jennings RB, Reimer KA. Preconditioning with ischemia: a recovery after nonlacunar ischemic stroke. Cerebrovasc Dis. (2004) delay of lethal cell injury in ischemic myocardium. Circulation. (1986) 18:304–11. doi: 10.1159/000080356 74:1124–36. doi: 10.1161/01.CIR.74.5.1124 17. Easton JD, Saver JL, Albers GW, Alberts MJ, Chaturvedi S, Feldmann E, et al. 2. Kitagawa K, Matsumoto M, Kuwabara K, Tagaya M, Ohtsuki T, Hata R, et al. Definition and evaluation of transient ischemic attack: a scientific statement ’Ischemic tolerance’ phenomenon detected in various brain regions. Brain Res. for healthcare professionals from the American heart association/American (1991) 561:203–11. doi: 10.1016/0006-8993(91)91596-S stroke association stroke council; council on cardiovascular surgery and 3. Narayanan SV, Dave KR, Perez-Pinzon MA. Ischemic anesthesia; council on cardiovascular radiology and intervention; council preconditioning and clinical scenarios. Curr Opin Neurol. (2013) on cardiovascular nursing; and the interdisciplinary council on peripheral 26:1–7. doi: 10.1097/WCO.0b013e32835bf200 vascular disease. The American academy of neurology affirms the value 4. Dirnagl U, Simon RP, Hallenbeck JM. Ischemic tolerance of this statement as an educational tool for neurologists. Stroke. (2009) and endogenous neuroprotection. Trends Neurosci. (2003) 40:2276–93. doi: 10.1161/STROKEAHA.108.192218 26:248–54. doi: 10.1016/S0166-2236(03)00071-7 18. Wegener S, Gottschalk B, Jovanovic V, Knab R, Fiebach JB, Schellinger PD, 5. Schaller B, Graf R. Cerebral ischemic preconditioning. An experimental et al. Transient ischemic attacks before ischemic stroke: preconditioning phenomenon or a clinical important entity of stroke prevention? J Neurol. the human brain? A multicenter magnetic resonance imaging study. Stroke. (2002) 249:1503–11. doi: 10.1007/s00415-002-0933-8 (2004) 35:616–21. doi: 10.1161/01.STR.0000115767.17923.6A 6. Iadecola C, Anrather J. Stroke research at a crossroad: asking the brain for 19. Herrett E, Bhaskaran K, Timmis A, Denaxas S, Hemingway H, directions. Nat Neurosci. (2011) 14:1363–8. doi: 10.1038/nn.2953 Smeeth L. Association between clinical presentations before myocardial 7. Hess DC, Blauenfeldt RA, Andersen G, Hougaard KD, Hoda MN, Ding Y, infarction and coronary mortality: a prospective population- et al. Remote ischaemic conditioning-a new paradigm of self-protection in the based study using linked electronic records. Eur Heart J. (2014) brain. Nat Rev Neurol. (2015) 11:698–710. doi: 10.1038/nrneurol.2015.223 35:2363–71. doi: 10.1093/eurheartj/ehu286 8. Stetler RA, Leak RK, Gan Y, Li P, Zhang F, Hu X, et al. 20. Yan T, Chopp M, Chen J. Experimental animal models and inflammatory Preconditioning provides neuroprotection in models of CNS cellular changes in cerebral ischemic and hemorrhagic stroke. Neurosci Bull. disease: paradigms and clinical significance. Prog Neurobiol. (2014) (2015) 31:717–34. doi: 10.1007/s12264-015-1567-z 114:58–83. doi: 10.1016/j.pneurobio.2013.11.005 21. Jayaraj RL, Azimullah S, Beiram R, Jalal FY, Rosenberg GA. 9. Ramos-Araque ME, Rodriguez C, Vecino R, Cortijo Garcia E, de Lera Neuroinflammation: friend and foe for ischemic stroke. J Neuroinflammation. Alfonso M, Sanchez Barba M, et al. The neuronal ischemic tolerance (2019) 16:142. doi: 10.1186/s12974-019-1516-2 is conditioned by the Tp53 Arg72Prop. Transl Stroke Res. (2018) 22. Stolp HB. Neuropoietic cytokines in normal brain development 10:204–15. doi: 10.1007/s12975-018-0631-1 and neurodevelopmental disorders. Mol Cell Neurosci. (2013) 10. Purroy F, Montaner J, Molina CA, Delgado P, Ribo M, Alvarez- 53:63–8. doi: 10.1016/j.mcn.2012.08.009 Sabin J. Patterns and predictors of early risk of recurrence after 23. Ramiro L, Simats A, Garcia-Berrocoso T, Montaner J. Inflammatory transient ischemic attack with respect to etiologic subtypes. Stroke. (2007) molecules might become both biomarkers and therapeutic 38:3225–9. doi: 10.1161/STROKEAHA.107.488833 targets for stroke management. Ther Adv Neurol Disord. (2018) 11. Purroy F, Jimenez Caballero PE, Gorospe A, Torres MJ, Alvarez-Sabin 11:1756286418789340. doi: 10.1177/1756286418789340 J, Santamarina E, et al. Stroke project of the Spanish cerebrovascular 24. Adams HP, Bendixen BH, Kappelle LJ, Biller J, Love BB, Gordon DL, et al. diseases study, recurrent transient ischaemic attack and early risk of stroke: Classification of subtype of acute ischemic stroke. Definitions for use in a data from the PROMAPA study. J Neurol Neurosurg Psychiatr. (2013) multicenter clinical trial. TOAST. Trial of org 10172 in acute stroke treatment. 84:596–603. doi: 10.1136/jnnp-2012-304005 Stroke. (1993) 24:35–41. doi: 10.1161/01.STR.24.1.35 12. Castillo J, Moro MA, Blanco M, Leira R, Serena J, Lizasoain I, et al. The release 25. Schaller B. Ischemic preconditioning as induction of ischemic tolerance after of tumor necrosis factor-alpha is associated with ischemic tolerance in human transient ischemic attacks in human brain: its clinical relevance. Neurosci Lett. stroke. Ann Neurol. (2003) 54:811–9. doi: 10.1002/ana.10765 (2005) 377:206–11. doi: 10.1016/j.neulet.2004.12.004 13. Weih M, Kallenberg K, Bergk A, Dirnagl U, Harms L, Wernecke KD, et al. 26. Special report from the National institute of neurological disorders Attenuated stroke severity after prodromal TIA: a role for ischemic tolerance stroke. Classification of cerebrovascular diseases III. Stroke. (1990) in the brain? Stroke. (1999) 30:1851–4. doi: 10.1161/01.STR.30.9.1851 21:637–76. doi: 10.1161/01.STR.21.4.637 14. Weber R, Diener HC, Weimar C, German Stroke Study Collaboration. Why 27. Rosset A, Spadola L, Ratib O. OsiriX: an open-source software for do acute ischemic stroke patients with a preceding transient ischemic attack navigating in multidimensional DICOM images. J Digit Imaging. (2004) present with less severe strokes? Insights from the German Stroke Study. Eur 17:205–16. doi: 10.1007/s10278-004-1014-6 Neurol. (2011) 66:265–70. doi: 10.1159/000331593 28. Purroy F, Cambray S, Mauri-Capdevila G, Jove M, Sanahuja J, 15. Aboa-Eboule C, Bejot Y, Osseby GV, Rouaud O, Binquet C, Marie C, et al. Farre J, et al. Metabolomics predicts neuroimaging characteristics Influence of prior transient ischaemic attack on stroke prognosis. J Neurol of transient ischemic attack patients. EBioMedicine. (2016) Neurosurg Psychiatry. (2011) 82:993–1000. doi: 10.1136/jnnp.2010.209171 14:131–8. doi: 10.1016/j.ebiom.2016.11.010 Frontiers in Neurology | www.frontiersin.org 13 October 2020 | Volume 11 | Article 552470 Colàs-Campàs et al. Inflammation Response of Prior TIA 29. Ashburner M, Ball CA, Blake JA, Botstein D, Butler H, Cherry JM, et al. Gene 48. Reyes R, Cardenes B, Machado-Pineda Y, Cabanas C. Tetraspanin CD9: a key ontology: tool for the unification of biology. The gene ontology consortium. regulator of cell adhesion in the immune system. Front Immunol. (2018) Nat Genet. (2000) 25:25–9. doi: 10.1038/75556 9:863. doi: 10.3389/fimmu.2018.00863 30. Mi H, Muruganujan A, Thomas PD. PANTHER in 2013: modeling 49. Yang C, Hawkins KE, Dore S, Candelario-Jalil E. Neuroinflammatory the evolution of gene function, and other gene attributes, in mechanisms of blood-brain barrier damage in ischemic stroke. Am J Physiol the context of phylogenetic trees. Nucleic Acids Res. (2013) Cell Physiol. (2019) 316:C135–53. doi: 10.1152/ajpcell.00136.2018 41:D377–86. doi: 10.1093/nar/gks1118 50. Vandooren J, Van Damme J, Opdenakker G. On the structure and functions 31. Jin R, Yang G, Li G. Inflammatory mechanisms in ischemic stroke: role of of gelatinase B/matrix metalloproteinase-9 in neuroinflammation. Prog Brain inflammatory cells. J Leukoc Biol. (2010) 87:779–89. doi: 10.1189/jlb.1109766 Res. (2014) 214:193–206. doi: 10.1016/B978-0-444-63486-3.00009-8 32. Iadecola C, Anrather J. The immunology of stroke: from mechanisms to 51. Morancho A, Rosell A, Garcia-Bonilla L, Montaner J. Metalloproteinase and translation. Nat Med. (2011) 17:796–808. doi: 10.1038/nm.2399 stroke infarct size: role for anti-inflammatory treatment? Ann N Y Acad Sci. 33. Yilmaz G, Granger DN. Cell adhesion molecules and ischemic stroke. Neurol (2010) 1207:123–33. doi: 10.1111/j.1749-6632.2010.05734.x Res. (2008) 30:783–93. doi: 10.1179/174313208X341085 52. Kirino T. Ischemic tolerance. J Cereb Blood Flow Metab. (2002) 34. Bustamante A, Sobrino T, Giralt D, Garcia-Berrocoso T, Llombart V, Ugarriza 22:1283–96. doi: 10.1097/01.WCB.0000040942.89393.88 I, et al. Prognostic value of blood interleukin-6 in the prediction of functional 53. Tanno M, Tsuchida A, Nozawa Y, Matsumoto T, Hasegawa T, Miura T, et al. outcome after stroke: a systematic review and meta-analysis. J Neuroimmunol. Roles of tyrosine kinase and protein kinase C in infarct size limitation by (2014) 274:215–24. doi: 10.1016/j.jneuroim.2014.07.015 repetitive ischemic preconditioning in the rat. J Cardiovasc Pharmacol. (2000) 35. Bustamante A, Dominguez C, Rodriguez-Sureda V, Vilches A, Penalba A, 35:345–52. doi: 10.1097/00005344-200003000-00001 Giralt D, et al. Prognostic value of plasma chitotriosidase activity in acute 54. Gagalo I, Rusiecka I, Kocic I. Tyrosine kinase inhibitor as a new stroke patients. Int J Stroke. (2014) 9:910–6. doi: 10.1111/ijs.12160 therapy for ischemic stroke and other neurologic diseases: is there 36. McDonough A, Weinstein JR. Neuroimmune response any hope for a better outcome? Curr Neuropharmacol. (2015) in ischemic preconditioning. Neurotherapeutics. (2016) 13:836–44. doi: 10.2174/1570159X13666150518235504 13:748–61. doi: 10.1007/s13311-016-0465-z 55. Durham JL, Jordan KA, Devos MJ, Williams EK, Sandstrom NJ. Estradiol 37. Ye H, Wang L, Yang XK, Fan LP, Wang YG, Guo L. Serum S100B levels may protects against hippocampal damage and impairments in fear conditioning be associated with cerebral infarction: a meta-analysis. J Neurol Sci. (2015) resulting from transient global ischemia in mice. Brain Res. (2012) 348:81–8. doi: 10.1016/j.jns.2014.11.010 1443:64–74. doi: 10.1016/j.brainres.2012.01.014 38. Zurek J, Fedora M. The usefulness of S100B, NSE, GFAP, NF-H, secretagogin 56. Li B, Woo RS, Mei L, Malinow R. The neuregulin-1 receptor erbB4 and Hsp70 as a predictive biomarker of outcome in children with controls glutamatergic synapse maturation and plasticity. Neuron. (2007) traumatic brain injury. Acta Neurochir. (2012) 154:93–103; discussion 54:583–97. doi: 10.1016/j.neuron.2007.03.028 103. doi: 10.1007/s00701-011-1175-2 57. Gambarotta G, Fregnan F, Gnavi S, Perroteau I. Chapter nine - neuregulin 39. Chen GY, Nunez G. Sterile inflammation: sensing and reacting to damage. Nat 1 role in schwann cell regulation and potential applications to promote Rev Immunol. (2010) 10:826–37. doi: 10.1038/nri2873 peripheral nerve regeneration. In: Geuna S, Perroteau I, Tos P, Battiston B, 40. Lasek-Bal A, Kowalewska-Twardela T, Gasior Z, Warsz-Wianecka A, Haberka editors. International Review of Neurobiology. Amsterdam: Academic Press M, Puz P, et al. The significance of troponin elevation for the clinical (2013). p.223–56. doi: 10.1016/B978-0-12-410499-0.00009-5 course and outcome of first-ever ischaemic stroke. Cerebrovasc Dis. (2014) 58. Wu L, Walas SJ, Leung W, Lo EH, Lok J. Neuregulin-1 and neurovascular 38:212–8. doi: 10.1159/000365839 protection. In: Kobeissy FH, editor. Brain Neurotrauma: Molecular, 41. Merkler AE, Gialdini G, Murthy SB, Salehi Omran S, Moya A, Lerario MP, Neuropsychological, Rehabilitation Aspects. Boca Raton, FL: CRC Press; Taylor et al. Association between troponin levels and embolic stroke of undetermined & Francis (2015). doi: 10.1201/b18126-46 source. J Am Heart Assoc. (2017) 6:e005905. doi: 10.1161/JAHA.117.005905 59. Simmons LJ, Surles-Zeigler MC, Li Y, Ford GD, Newman GD, Ford 42. Purroy F, Suarez-Luis I, Mauri-Capdevila G, Cambray S, Farre J, Sanahuja J, BD. Regulation of inflammatory responses by neuregulin-1 in brain et al. N-terminal pro-brain natriuretic peptide level determined at different ischemia and microglial cells in vitro involves the NF-kappa B pathway. times identifies transient ischaemic attack patients with atrial fibrillation. Eur J Neuroinflammation. (2016) 13:237. doi: 10.1186/s12974-016-0703-7 J Neurol. (2014) 21:679–83. doi: 10.1111/ene.12222 60. Amantea D, Certo M, Petrelli F, Bagetta G. Neuroprotective properties of a 43. Llombart V, Antolin-Fontes A, Bustamante A, Giralt D, Rost NS, macrolide antibiotic in a mouse model of middle cerebral artery occlusion: Furie K, et al. B-type natriuretic peptides help in cardioembolic characterization of the immunomodulatory effects and validation of the stroke diagnosis: pooled data meta-analysis. Stroke. (2015) efficacy of intravenous administration. Assay Drug Dev Technol. (2016) 46:1187–95. doi: 10.1161/STROKEAHA.114.008311 14:298–307. doi: 10.1089/adt.2016.728 44. Powers WJ, Rabinstein AA, Ackerson T, Adeoye OM, Bambakidis NC, 61. Heldmann U, Mine Y, Kokaia Z, Ekdahl CT, Lindvall O. Selective Becker K, et al. Guidelines for the early management of patients with depletion of Mac-1-expressing microglia in rat subventricular zone does acute ischemic stroke: 2019 update to the 2018 guidelines for the not alter neurogenic response early after stroke. Exp Neurol. (2011) early management of acute ischemic stroke: a guideline for healthcare 229:391–8. doi: 10.1016/j.expneurol.2011.03.005 professionals from the American heart association/American stroke association. Stroke. (2019) 50:e344–418. doi: 10.1161/STR.0000000000 Conflict of Interest: The authors declare that the research was conducted in the 000211 absence of any commercial or financial relationships that could be construed as a 45. Hoshino T, Mizuno S, Shimizu S, Uchiyama S. Clinical features and potential conflict of interest. functional outcome of stroke after transient ischemic attack. J Stroke Cerebrovasc Dis. (2013) 22:260–6. doi: 10.1016/j.jstrokecerebrovasdis.2011. Copyright © 2020 Colàs-Campàs, Farre, Mauri-Capdevila, Molina-Seguín, 08.010 Aymerich, Ois, Roquer, Tur, García-Carreira, Martí-Fàbregas, Cruz-Culebras, 46. Kelly PJ, Murphy S, Coveney S, Purroy F, Lemmens R, Tsivgoulis G, et al. Segura, Arque and Purroy. This is an open-access article distributed under the terms Anti-inflammatory approaches to ischaemic stroke prevention. J Neurol of the Creative Commons Attribution License (CC BY). The use, distribution or Neurosurg Psychiatry. (2017) 89:211–8. doi: 10.1136/jnnp-2016-314817 reproduction in other forums is permitted, provided the original author(s) and the 47. Lakhan SE, Kirchgessner A, Hofer M. Inflammatory mechanisms copyright owner(s) are credited and that the original publication in this journal in ischemic stroke: therapeutic approaches. J Transl Med. (2009) is cited, in accordance with accepted academic practice. No use, distribution or 7:97. doi: 10.1186/1479-5876-7-97 reproduction is permitted which does not comply with these terms. Frontiers in Neurology | www.frontiersin.org 14 October 2020 | Volume 11 | Article 552470