Abstract Background Thrombolysis and endovascular thrombectomy are the primary treatment for ischemic stroke. However, due to the limited time window and the occurrence of adverse effects, only a small number of patients can genuinely benefit from recanalization. Intraarterial injection of rtPA (recombinant tissue plasminogen activator) based on arterial thrombectomy could improve the prognosis of patients with acute ischemic stroke, but it could not reduce the incidence of recanalization‐related adverse effects. Recently, selective brain hypothermia has been shown to offer neuroprotection against stroke. To enhance the recanalization rate of ischemic stroke and reduce the adverse effects such as tiny thrombosis, brain edema, and hemorrhage, we described for the first time a combined approach of hypothermia and thrombolysis via intraarterial hypothermic rtPA. Methods and Results We initially established the optimal regimen of hypothermic rtPA in adult rats subjected to middle cerebral artery occlusion. Subsequently, we explored the mechanism of action mediating hypothermic rtPA by probing reduction of brain tissue temperature, attenuation of blood–brain barrier damage, and sequestration of inflammation coupled with untargeted metabolomics. Hypothermic rtPA improved neurological scores and reduced infarct volume, while limiting hemorrhagic transformation in middle cerebral artery occlusion rats. These therapeutic outcomes of hypothermic rtPA were accompanied by reduced brain temperature, glucose metabolism, and blood–brain barrier damage. A unique metabolomic profile emerged in hypothermic rtPA‐treated middle cerebral artery occlusion rats characterized by downregulated markers for energy metabolism and inflammation. Conclusions The innovative use of hypothermic rtPA enhances their combined, as opposed to stand‐alone, neuroprotective effects, while reducing hemorrhagic transformation in ischemic stroke. Keywords: blood–brain barrier, hypothermia, inflammation, ischemic stroke, middle cerebral artery occlusion, tissue plasminogen activator Subject Categories: Blood-Brain Barrier, Ischemic Stroke, Translational Studies __________________________________________________________________ Clinical Perspective. What Is New? * Adverse events limit the usage of recombinant tissue plasminogen activator, which might be suppressed by hypothermic therapy. * Therefore, this study combined recombinant tissue plasminogen activator and infused hypothermic recombinant tissue plasminogen activator into the middle cerebral artery in focal cerebral ischemia to explore its neuroprotective effect against ischemic stroke. What Are the Clinical Implications? * The present study aims to prove the safety and practicability of selective hypothermic recombinant tissue plasminogen activator injection in rats by various methods, leading to broad clinical application prospects. Nonstandard Abbreviations and Acronyms ^18F‐FDG 2‐Fluoro‐2‐deoxy‐D‐glucose (2‐deoxy‐2‐[18F]fluoro‐D‐glucose BBB blood–brain barrier cFn cellular fibronectin EB Evans blue HT hypothermic rtPA MCAO middle cerebral artery occlusion MMP matrix metalloproteinase rCBF regional CBF NT normothermic rtPA rtPA recombinant tissue plasminogen activator uPAR urokinase‐type plasminogen activator receptor As most neuroprotectant strategies failed in clinical trials, recanalization remains the only proven effective treatment for ischemic stroke.[44] ^1 With the extensive application of mechanical thrombectomy, intraarterial injection of recombinant tissue plasminogen activator (rtPA) could improve the vascular recanalization rate and potentially affect the long‐term prognosis of patients as a supplementary treatment.[45] ^2 However, rtPA could not avoid the adverse effects, such as vascular edema and hemorrhagic transformation, caused by vascular recanalization.[46] ^3 Hypothermia was widely used in treating ischemic stroke in neurological intensive care units.[47] ^4 A clinical trial proved that intraarterial selective cooling infusion is safe for patients with stroke treated with mechanical thrombectomy.[48] ^5 Another preclinical study demonstrated that intraarterial selective hypothermic addition could effectively protect against the brain damage induced by ischemia in rhesus monkeys.[49] ^6 Previous research proved that intraarterial selective hypothermic injection could reduce the volume of cerebral infarction in middle cerebral artery occlusion (MCAO) rats[50] ^7 and reduce the damage to the blood–brain barrier (BBB).[51] ^8 The main reason for limiting the usage of rtPA was its potential serious adverse events, such as hemorrhagic transformation and malignant cerebral edema, which might be suppressed by hypothermic therapy.[52] ^9 Therefore, this study combined hypothermic therapy and intraarterial injection of rtPA infused hypothermic rtPA (HT) into the middle cerebral artery in focal cerebral ischemia to examine whether mild hypothermic rtPA could limit the brain injury related to vascular recanalization treatment and enhance the neuroprotection against ischemic stroke. Reducing the adverse events of rtPA application can further broaden the application scenarios of rtPA and benefit as many patients as possible. The present study aims to prove the safety and practicability of selective hypothermic rtPA injection by various methods and explore the possible mechanism of the neuroprotective effect. Methods The data that support the findings of this study are available from the corresponding author upon reasonable request. Thrombolysis In Vitro The thrombus was performed as previously described.[53] ^10 Whole blood was drawn into a polyethylene‐10 tube and allowed to clot at 37 °C for 2 hours. At the end of this period, the clot was extruded from the catheter into a saline‐filled petri dish and stored at 4 °C for 22 hours. Then the clot was dissected into single 5‐mm and 2‐mm sections. Each clot was placed in 2 mL normothermic rtPA (NT) or HT for 10 minutes. The volume of the clots was measured before and after thrombolysis. Animals Adult male Sprague–Dawley rats weighing 300–320 g (SPF [Beijing] Biotechnology Co., Ltd.) were enrolled in this study. The Institutional Animal Investigation Committee approved all experimental procedures of Xuanwu Hospital, Capital Medical University, and all animal treatments were performed by experienced operators according to the Care and Use of Laboratory Animals guidelines from the National Institutes of Health and the Animal Research: Reporting of In Vivo Experiments guidelines. A total of 64 adult rats were randomly divided into 4 groups: a Sham group, an MCAO group (subjects underwent MCAO with saline injection), an NT group (subjects underwent MCAO followed by NT infusion), and an HT group (subjects underwent MCAO followed by HT injection). All investigators involved in behavioral testing, histological analyses, and other outcomes assays were blinded to the treatment groups. MCAO Model The suture model of transient cerebral artery occlusion in rats was selected to simulate mechanical thrombectomy after ischemic stroke. Rats were anesthetized with 5% isoflurane maintained with 1% to 2% isoflurane, 70% N[2]O, and 30% O[2] with a facemask. The right middle cerebral artery was occluded for 90 minutes and subsequently reperfused. MCAO models were accomplished using the intraluminal suture occlusion method, as previously described.[54] ^11 Rats were excluded from further analysis if they died before the end of the study or if they did not exhibit any signs of ischemic injury based on the Longa score and T2‐weighted sequences of magnetic resonance (MR) imaging. Hypothermia Infusion A 2 mL dose of 4 °C rtPA (0.07 mg/mL) was administered to induce hypothermia as soon as reperfusion. The dosage is converted according to clinical trials (patients underwent intraarterial alteplase in a maximum dose of 22.5 mg in CHOICE [Chemical Optimization of Cerebral Embolectomy in Patients With Acute Stroke Treated With Mechanical Thrombectomy],[55] ^2 and patients underwent intraarterial selective cooling infusion with 350 mL 0.9% saline at 4 °C for 15 minutes in a previous cohort study[56] ^5 ). A 2 mL dose of rtPA represents 10% blood volume of rats. In all 3 groups except the Sham group, a modified polyethylene‐10 catheter (0.2‐mm outer diameter and 0.1‐mm inner diameter) was inserted through the incision in the carotid artery. A 2 mL dose of HT, NT, or saline was infused in 10 minutes. Brain and Body Temperature Monitoring The brain and rectal temperatures were monitored before reperfusion, during (for 10 minutes), and after hypothermia treatment (for another 40 minutes) by needle thermistor probes (Harvard Apparatus, Inc). Regional Cerebral Blood Flow CBF Measurements With Laser Speckle Contrast Imager Regional cerebral blood flow (CBF) of MCAO rats was monitored using a laser speckle contrast imager (PeriCam PSI HR System, Perimed Inc., Jarfalla‐Stockholm, Sweden). We performed the measurements before surgery (baseline), after MCAO, and after injection. After induction of anesthesia, the rat was placed in the prone position under a laser speckle contrast imager. A middle parietal incision was made to expose the skull. The speckle imager was placed 10 cm above the skull at a working distance to detect CBF. Over time, CBF measurements were made in the same areas (MCA territory). CBF changes were expressed as percentages of baseline perfusion. Positron Emission Tomography‐MR Scanning The most widely used metabolic imaging agent in clinical practice is 2‐deoxy‐2‐[18F]fluoro‐D‐glucose [^18F‐FDG], which can detect tumor, cardiac, and glucose metabolism in the brain and which is an analog of glucose labeled with fluorine 18. After intravenous injection into the human body, it can be absorbed by different tissues. The metabolic pathway is also similar to that of natural glucose. The uptake distribution of ^18F‐FDG in other tissues can reflect the glucose metabolism status of different tissues and organs. Head positron emission tomography‐MR imaging was performed on rats at 24 hours and 72 hours after the operation to evaluate the metabolic level of the rat brain. The rats fasted with water for 8 hours before scanning, and the rats were anesthetized and injected with ^18F‐FDG slowly via the tail vein at a dose of 11.1 Mbq/kg. After administration, the rats rested for 40 minutes until the FDG stabilized before scanning. Neurological Deficits Following surgery, the neurological deficits of the rats were assessed using the Zea‐Longa 5‐point scoring, modified neurological severity score scale, and tape removal test. Rats performed these tests at 1, 3, 7, 14, and 21 days after reperfusion, and all tests were overseen by the same individual, who was blinded to the treatment groups. Blood–Brain Barrier Integrity Determination Evans blue (EB) leakage in brain tissues and the loss of TJs (tight junction proteins) in cerebral microvessels were determined to detect the BBB integrity. At 72 hours after reperfusion, EB (2% wt/vol in PBS, 3 mL/kg; Solarbio, Beijing, China) was infused into the tail vein, and the rats were killed by cardiac perfusion of saline 2 hours later. The brain was then removed, sectioned, and photographed to visualize EB extravasation. The tissues were collected in 1.5 mL tubes for measurement of EB content to quantify BBB disruption. Cerebral microvessels were isolated from the brain after 90‐minute ischemia/72‐hour reperfusion. Homogenates (20 μg protein) of microvessels were prepared for Western blotting. The primary antibodies were occludin (1:500, Thermo Fisher Scientific, Waltham, MA); claudin‐5 (1:500, Thermo Fisher Scientific); matrix metalloproteinase‐9 (MMP 9) (1:500, Santa Cruz Biotechnology), MMP 2 (Santa Cruz Biotechnology), and β‐actin (1:2000, Abcam Biotechnology). Protein levels were expressed as the ratio to β‐actin. Multiplex Inflammatory Cytokines Detection On the third day after MCAO, the rats were anesthetized and cerebrospinal fluid (CSF) samples collected from the foramen magnum. Multiplex inflammatory cytokine detection was performed using Bio Legend's LEGEND plex Rat Inflammation Pane. The inflammatory markers included IL (interleukin)‐1α, IL‐1β, IL‐10, IL‐12p70, IL‐17A, IL‐18, chemokine (C‐X‐C motif) ligand 1 (CXCL1), GM‐CSF (granulocyte macrophage colony stimulating factor), IFN‐γ (interferon‐γ), IL‐33, TNF‐α (tumor necrosis factor alpha), IL‐6, and chemokine (C‐C motif) ligand 2. The samples were treated according to the instructions, and the standards and samples were processed according to the instructions. Flow cytometry was used to detect the inflammatory cytokines. Enzyme‐Linked Immunosorbent Assay uPAR (urokinase‐type plasminogen activator receptor) levels in the plasma were measured using ELISA Kit for rat uPAR (No. SEA141Ra) and cFn (cellular fibronectin) (No. SEA214Ra) (Cloud‐Clone Corp, Wuhan, China). The test is carried out according to the steps of the instructions. Metabolomics Study Liquid chromatographic separation for processed plasma was achieved on Waters ACQUITY UPLC BEH C8 (1.7 μm 2.1 mm*100 mm) column using Q Exactive System (Thermo Scientific), whereas mass spectrometry was performed on Thermo Scientific Dionex ΜltiMate 3000 Rapid Separation LC. Data pretreatment procedures, such as nonlinear retention time alignment, peak discrimination, filtering, alignment, matching, and identification, were performed by Progenesis QI 2.3 Software (Waters Corporation, Milford, MA). Orthogonal projections to latent structures discriminant analysis was performed on the comparison group using the “ropls” R package, and the fold changes in abundance between the 2 groups were demonstrated using volcano plots to visualize the regulation of differential metabolites. Metabolites were investigated in the Kyoto Encyclopedia of Genes and Genomes metabolic pathway for enrichment analysis, and the top 20 pathways that 122 showed significant enrichment were selected for bubble mapping based on P values. The analysis of volcano plots and Kyoto Encyclopedia of Genes and Genomes enrichment was performed using Metaboanalyst, a free online platform for data analysis[57] ^12 ([58]https://dev.metaboanalyst.ca/MetaboAnalyst/home.xhtml). The metabolomics data were normalized by log transformation. Statistical Analysis All experimental data were analyzed using GraphPad Prism8 (San Diego, CA) and are expressed as mean±SEM. Experimental data were statistically analyzed by Student's t test, 1‐way ANOVA, and Tukey's test. Analyses involving repeated measurements were statistically analyzed by permutational test. Differences between the groups were considered statistically significant at P<0.05 and adjusted by q value (q<0.05). Results Hypothermic rtPA Reduces Brain Temperature, Glucose Metabolism, and Infarct Volume in MCAO Rats Through in vitro experiments, we found that the efficiency of NTA in the thrombolysis of larger clots (5 mm) was significantly higher than that of HT, whereas when dissolving smaller clots (2 mm), the dissolution rate of NT and HT showed no significant difference (Figure [59]1A). As a supplementary treatment, HT was used to dissolve tiny thrombosis after mechanical thrombectomy. The effect on the thrombolytic efficiency of HT appears mild and tolerable, and the neuroprotective effects of HT were further detected in vivo (n=6). Figure 1. Hypothermic rtPA decreases brain temperature, glucose metabolism, and cerebral infarction volume in MCAO rats. Figure 1 [60]Open in a new tab A, The thrombolytic efficiency of NT in large thrombus (5 mm) was significantly higher than that of HT, but there was no significant difference in small thrombus (2 mm). Results are shown as mean±SEM, *P<0.05, n=6 (Student's t test). B, The temperature in the cortex, basal ganglia, and rectum decreased after the hypothermic rtPA injection. Results are shown as mean±SEM, *P<0.05, ****P<0.0001, n=6 (permutational test). C, rCBF was monitored in MCAO rats using the 2‐dimensional laser speckle technique before surgery, after MCAO, and immediately after injection. rCBF in the infarct regions was quantified and expressed as the percentage change from baseline. Results are shown as mean±SEM, ****P<0.0001, n=6 (permutational test). D, The 18F‐FDG positron emission tomography‐magnetic resonance imaging scan at 24 hours and 72 hours after reperfusion. The metabolic intensity of [^18F]FDG in the cerebrum was normalized to the metabolic intensity of the cerebellar. E, The infarct volume on T2‐weighted images at 24 hours and 72 hours after reperfusion. Results are shown as mean±SEM, n=6 (permutational test). ^18F‐FDG indicates 2‐Fluoro‐2‐deoxy‐D‐glucose (2‐deoxy‐2‐[18F]fluoro‐D‐glucose; HT, hypothermic rtPA; MCAO, middle cerebral artery occlusion; NT, normothermic rtPA; rCBF, regional cerebral blood flow; and rtPA, recombinant tissue plasminogen activator. The cortex, striatum, and rectum temperatures were maintained within a normothermic range using a feedback‐controlled heating pad. The temperatures of the cortex and striatum decreased dramatically as soon as HT started to inject. Cortical temperatures, starting at 36.5±0.3 °C (NT group) and 36.3±0.4 °C (HT group) at the onset of reperfusion, dropped to 35.9±0.2 and 34±0.3 °C, respectively, after 10 minutes' injection by NT or HT. The cortex did not exceed 35 °C until 20 minutes after the injection finished in the HT group (Figure [61]1B). Striatal temperatures experienced a similar degree of cooling. In the NT group, temperatures fell from 37.5±0.4 °C to 37.1±0.3 °C after 10 minutes. In the HT group, temperatures fell from 37.3±0.4 °C to 35.7±0.6 °C after 10 minutes (Figure [62]1B). Rectal temperatures were less affected by hypothermia. Starting at 37.5±0.3 °C and 37.4±0.6 °C in the NT and HT groups, the rectal temperatures decreased to 37.8±0.2 °C (NT) and 36.7±0.4 °C (HT) after injection, respectively (n=6) (Figure [63]1B). To evaluate the effect of HT on CBF perfusion in rats, we monitored the regional CBF in MCAO rats using the 2‐dimensional laser speckle technique before surgery, after MCAO, and immediately after injection. Compared with the baseline, the regional CBF of the infarcted side decreased significantly. After administration, the regional CBF increased slightly. Compared with the NT treatment group, the cortical perfusion of the HT group was reduced somewhat by HT treatment, but there was no significant difference (n=6) (Figure [64]1C). Positron emission tomography‐MR imaging was performed at 24 hours and 72 hours after ischemia–reperfusion in rats (n=6). T2‐weighted sequences of nuclear magnetic resonance calculated the size of the infarct volume and the metabolic of ^18F‐FDG. In T2‐weighted sequences in 24 hours, the infarct volume of rats in the HT group was significantly smaller than that in the MCAO and NT groups. There was no significant difference in infarct volume between the NT group and the MCAO group (Figure [65]1D). According to the positron emission tomography‐MR imaging, the metabolic intensity of the infarcted brain tissues in the MCAO group and NT group was significantly lower than that of the healthy brain tissues. In contrast, the metabolic intensity of the infarcted brain tissues in the HT group was not significantly different from the healthy tissues. The metabolic intensity of the whole brain tissues was substantially lower in the HT group than in the MCAO group and NT group (Figure [66]1D). The T2‐weighted sequences showed that there was no brain edema in the brain tissue of the HT group, and the infarct volume did not increase in 72 hours after reperfusion. However, edema occurred in the brain tissues of the MCAO group and NT group, and the infarct volume also increased compared with that at 24 hours (Figure [67]1D and [68]1E). At 72 hours after reperfusion, the metabolic intensity of the infarcted brain tissues in the MCAO group was significantly lower than that of the healthy tissues. The metabolic intensity was significantly decreased in the HT group compared with the NT group (Figure [69]1D). Hypothermic rtPA Improves Neurological Scores While Limiting Hemorrhagic Transformation in MCAO Rats The neurological function of rats on the first, third, fifth, seventh, 14th, and 21st days was evaluated by Longa score, stagger step test, tape removal test, and modified neurological severity score (n=4). The Longa score was used to test the neurological damage of MCAO rats briefly. The neurological scores of the MCAO and NT group were higher than those of the HT group. Next, we performed a comprehensive assessment of sensory and motor function in rats using the foot fault test, tape removal test, and modified neurological severity score. These data suggest that the incidence rate of foot faults in the MCAO and NT group remained higher than in the HT group on the 21st day. In the tape removal test, the removal time in the HT group was significantly improved from day 5 to day 21 (Figure [70]2A). Similarly, rats in the HT group improved considerably in modified neurological severity score from day 5 to day 21 (Figure [71]2B). Figure 2. Hypothermic rtPA decreases neurological function scores and reduces hemorrhage transformation in MCAO rats. Figure 2 [72]Open in a new tab A, Longa score, foot fault test, and tape removing test after reperfusion. Results are shown as mean±SEM, **P<0.01, n=4 (permutational test). B, mNSS of MCAO rats. Results are shown as mean±SEM, *P<0.05, **P<0.01, n=4 (permutational test). C, Prussian blue stain in brain tissues of MCAO rats. Prussian blue stained the ions in brain tissues. Prussian blue^+ cells were found in the brain tissue of the NT group, which represents the hemosiderin in the brain tissues. In the sham, MCAO, and HT groups, seldom Prussian blue^+ cells were observed in the brain tissues. Results are shown as mean±SEM, **P<0.01, ^# P<0.05, n=6 (permutational test). HT indicates hypothermic rtPA; MCAO, middle cerebral artery occlusion; mNSS, Modified neurological severity score; NT, normothermic rtPA; rCBF, regional cerebral blood flow; and rtPA, recombinant tissue plasminogen activator. Prussian blue staining was used to stain the trivalent iron ions in the paraffin sections of the rat brain and costained with the nuclei to indirectly mark the hemoglobin in the brain parenchyma. The results of Prussian blue staining showed that circular blue particles with positive Prussian blue staining could be observed in the brain tissue of the NT group, that is, hemosiderin containing trivalent iron in the brain parenchyma. There were no significant Prussian blue‐positive staining granules in the brain tissue of the sham, MCAO, and HT groups (n=6) (Figure [73]2C). By Prussian blue staining, we can intuitively prove that NT injection via the middle cerebral artery can lead to a hemorrhagic transformation of ischemic brain tissue, whereas HT can reduce the occurrence of hemorrhagic transformation. Hypothermic rtPA Suppresses Blood–Brain Barrier Damage In EB testing, the EB leakage into the ischemic hemisphere of normothermic rats was significantly higher than MCAO rats at 72 hours after reperfusion. The HT treatment significantly reduced EB leakage (n=6) (Figure [74]3A). Figure 3. Hypothermic rtPA ameliorates blood–brain barrier damage. Figure 3 [75]Open in a new tab A, Evan's blue leakage in ischemic hemisphere of rats after 72‐hour reperfusion. Results are shown as mean±SEM, ****P<0.001 vs MCAO group, ^#### P<0.0001 vs NT group, n=6 (1‐way ANOVA). B, Western blot detection of MMP 9 and MMP 2 in brain microvessels of rats after 72‐hour reperfusion. Results are shown as mean±SEM, *P<0.05, **P<0.01 vs Sham group, ^## P<0.001 vs MCAO group, n=6 (1‐way ANOVA). C, Western blot detection of occludin and claudin‐5 in the brain microvessels of rats after 72‐hour reperfusion. Results are shown as mean±SEM, *P<0.05, **P<0.01 vs Sham group, ^## P<0.001 vs MCAO group, n=6 (1‐way ANOVA). MCAO indicates middle cerebral artery occlusion; MMP 9, matrix metalloproteinase 9; NT, normothermic rtPA; and rtPA, recombinant tissue plasminogen activator. The loss of TJs induced by rtPA leads to the damage of the BBB, which is one of the leading causes of hemorrhagic transformation. We then investigated whether HT injection could reduce rtPA‐induced BBB damage, thereby reducing hemorrhagic transformation and further reducing brain damage. The rat brain microvessels were extracted for Western blot detection. The semiquantitative analysis showed that levels of MMP 2 and MMP 9 in the cerebral microvessels of the MCAO group and NT group were significantly higher than those of the Sham group at 72 hours, and the expression levels of MMP 2 and MMP 9 in the NT group were considerably higher than those of the HT group. It was significantly lower than in the MCAO group (Figure [76]3B). Compared with the Sham group, the expressions of vascular endothelial TJs occludin and claudin‐5 in the MCAO group and NT were decreased, and the vascular endothelial TJ expression in the HT group was higher than that in the MCAO group. However, the difference was not statistically significant (n=6) (Figure [77]3C). Full unedited gel for the Western blot is shown in Figure [78]S1. Hypothermic rtPA Dampens Neuroinflammation in MCAO Rats Adverse effects of rtPA may include the activation of the fibrinolytic system through enzymatic reaction and lead to the secretion of inflammatory factors. At the same time, it can also play a role in the inflammatory response as a cytokine. The fibrinolytic activation‐related proteins uPAR and cFn are not only essential components of the fibrinolytic system but also important cytokines affecting immune cell migration. Inflammatory factors in rat plasma were detected by multifactor detection, and plasma uPAR and cFn were detected by ELISA. The secretion of inflammatory factors in peripheral blood did not show a specific difference (n=4) (Figure [79]4A through [80]4C). However, there was a significant difference in the uPAR and cFn expression levels. Compared with the Sham group, the expression of uPAR in plasma and CSF of rats in the MCAO group and NT group decreased at 72 hours after reperfusion. The uPAR in plasma and CSF of the HT group increased compared with that in the NT group. However, the difference was not statistically significant (n=6) (Figure [81]4D). In plasma, compared with the Sham group, the expression of cFn in the MCAO group showed a downward trend, and the presentation of cFn in NT and HT groups increased slightly, with no statistical significance (Figure [82]4D). In CSF, the changing trend of cFn expression was consistent with that of uPAR. The level of cFn in the CSF of the NT group was significantly lower than that of the Sham group (Figure [83]4D). HT treatment may affect the activation of immune cells by upregulating the expression of fibrinolytic‐related proteins caused by reperfusion treatment. The changing trend of plasma cFn expression in the NT group is different from that in the central nervous system, which may be related to the coagulation function and hemostasis of cFn after rtPA treatment, but it still needs to be further verified. Figure 4. Hypothermic rtPA significantly attenuates the expression of proinflammatory factors in the plasma of MCAO rats. Figure 4 [84]Open in a new tab A, Multiplex inflammatory cytokine detection of CSF, CCL2, and GM‐CSF in the plasma of rats after 24‐hour reperfusion. Results are shown as mean±SEM, n=4 (1‐way ANOVA). B, Multiplex inflammatory cytokine detection of inflammatory factors that induced macrophage activation in the plasma of rats after 24‐hour reperfusion. Results are shown as mean±SEM, n=4 (1‐way ANOVA). C, Multiplex inflammatory cytokine detection of inflammatory factors inducing other immune cells in the plasma of rats after 24‐hour reperfusion. Results are shown as mean±SEM, n=4 (one‐way ANOVA). D, Expression of fibrinolytic related proteins uPAR and cFn in plasma and CSF of rats at 24 hours and 72 hours after reperfusion were detected by ELISA kit. Results are shown as mean±SEM, *P<0.05 vs Sham group, n=6 (1‐way ANOVA). CCL2 indicates chemokine (C‐X‐C motif) ligand 1 (CXCL1), chemokine (C‐C motif) ligand 2; cFn, cellular fibronectin; CSF, cerebrospinal fluid; GM‐CSF, granulocyte macrophage colony stimulating factor; HT, hypothermic rtPA; IFN‐γ, interferon‐γ; IL, interleukin; MCAO, middle cerebral artery occlusion; NT, normothermic rtPA; rtPA, recombinant tissue plasminogen activator; TNF‐α, tumor necrosis factor alpha; and uPAR, urokinase‐type plasminogen activator receptor. We detected the secretion of inflammatory factors in rats' CSF. The expression of chemokine (C‐X‐C motif) ligand 1 and chemokine (C‐C motif) ligand 2 in the CSF of the HT group was significantly lower than that of the MCAO group (n=4) (Figure [85]5A), and the expression of IL‐18 was decreased as well (n=4) (Figure [86]5B). The expression of other inflammatory factors did not show a specific difference (n=4) (Figure [87]5C). This indicated that HT could reduce the chemotaxis of inflammatory cells in peripheral blood, especially myeloid cells. Figure 5. Hypothermic rtPA inhibits the expression of inflammation‐associated chemokines and filters infiltrating neutrophils. Figure 5 [88]Open in a new tab A, Multiplex inflammatory cytokine detection of CCL2, and GM‐CSF in the CSF of rats after 72‐hour reperfusion. Results are shown as mean±SEM, *P<0.05, n=4 (1‐way ANOVA). B, Multiplex inflammatory cytokine detection of inflammatory factors that induced macrophage activation in the CSF of rats after 72‐hour reperfusion. Results are shown as mean±SEM, *P<0.05, n=4 (1‐way ANOVA). C, Multiplex inflammatory cytokine detection of inflammatory factors inducing other immune cells in the CSF of rats after 72 hours reperfusion. Results are shown as mean±SEM, n=4 (1‐way ANOVA). D, Neutrophil infiltration in brain tissue at 72 hours after reperfusion were detected by double immunofluorescence labeling of NCF4 and CD16/CD206 in paraffin sections of brain tissue of MCAO rats after 72 hours reperfusion. The blue channel is nuclear DAPI staining (n=6). CCL2 indicates chemokine (C‐X‐C motif) ligand 1 (CXCL1), chemokine (C‐C motif) ligand 2; cFn, cellular fibronectin; CSF, cerebrospinal fluid; GM‐CSF, granulocyte macrophage colony stimulating factor; HT, hypothermic rtPA; MCAO, middle cerebral artery occlusion; NT, normothermic rtPA; and rtPA, recombinant tissue plasminogen activator. To further verify the effect of low‐temperature rtPA on inflammatory cells, we detected the neutrophil infiltration in brain tissue and found that the neutrophil infiltration in brain tissue of rats in the HT group was reduced, and the classification was transformed from inflammatory phenotype N1 (CD16^+) to anti‐informatory phenotype N2 (CD206^+) (n=6) (Figure [89]5D). Hypothermic rtPA Creates a Unique Metabolomic Profile in MCAO Rats Gas chromatography–mass spectrometry and ultra‐performance liquid chromatography‐mass spectrometry detected 12 137 peaks from all 24 plasma samples by nontargeted metabolomic mass spectrometry analysis (n=6). The orthogonal projections to latent structures discriminant analysis model was used to characterize the metabolic disturbances (Figure [90]6A). The horizontal axis in the figure represents the predicted principal component, that is, the difference among groups. The vertical axis represents the orthogonal principal components, which indicates the difference within the group can be seen in the direction of the vertical axis. Figure 6. Hypothermic rtPA produces a metabolomic signature in MCAO rats. Figure 6 [91]Open in a new tab A, The OPLS‐DA scores. B, Volcano plots showed all metabolite alterations in the plasma of rats. C, Metabolic pathway enrichment analysis of differential metabolites in the plasma of rats. All matched pathways are displayed as circles; the color and size of each circle are based on the P value and the pathway impact value, respectively. Redder and bigger circus indicates lower P values and greater count numbers, respectively (n=6). HT indicates hypothermic rtPA; IFN‐γ, interferon‐γ; IL, interleukin; MCAO, middle cerebral artery occlusion; NT, normothermic rtPA; OPLS‐DA, orthogonal projections to latent structures discriminant analysis; rtPA, recombinant tissue plasminogen activator; and TNF‐α, tumor necrosis factor alpha. After peak alignment and missing values were removed, 697 ions showed significant changes (P<0.05) between the MCAO group and the Sham group, and 140 ions had changed considerably between the HT group and the NT group. The ions with variable importance in the projection values >1.0 were considered the potential differential metabolites. We compared the Sham, MCAO, NT, and HT groups, identifying and characterizing specific metabolites and metabolic pathways. Different metabolites were screened according to the differential expression of metabolite (P<0.05, FC > 2) (Figure [92]6B). A total of 882 metabolites were downregulated by ischemic stroke, whereas 1571 metabolites were upregulated. Moreover, 23 metabolites were downregulated by NT treatment, 61 of which are upregulated. Compared with the NT group, HT treatment downregulated 112 metabolites and 272 were upregulated. The primary differential metabolic pathways between groups are shown in Figure [93]6C, which was performed by pathway analysis of Metaboanalyst. The primary differential metabolites are mostly related to energy metabolism, lipid metabolism (cell membrane structure), and inflammatory reaction. Metabolic pathways such as pyrimidine metabolism, tryptophan metabolism, pentose and glucuronic acid interconversion pathway, and phosphatidylinositol signaling pathway changed significantly; Disturbance of pentose and glucuronic acid interconversion pathway, P450 related metabolism pathway, and phosphatidylinositol signaling pathway was the most significant in all paired comparisons ([94]https://www.genome.jp/kegg/), which indicates that HT might reduce the metabolic rate of cells, thus reducing the energy metabolism disorder in ischemia, and HT might reduce the damage of BBB by affecting lipid metabolism and maintaining the stability of cell membrane structure. Discussion Rescuing the ischemic penumbra represents a major theme of neuroprotection research in cerebral ischemic injury. With the development of vascular recanalization, the recanalization rate of cerebral ischemia has significantly improved. However, many patients still present with a poor prognosis, partly due to the expansion of the infarct core area before vascular recanalization and the reduction of ischemic penumbra.[95] ^13 , [96]^14 On the other hand, postoperative no‐reflow may impair microvascular tissue, and the subsequent reperfusion injury may also expand the ischemic penumbra.[97] ^15 , [98]^16 In addition, the risk of hemorrhagic transformation caused by delayed recanalization likely exacerbates the poor prognosis of vascular recanalization treatment.[99] ^17 To this end, hypothermia can effectively protect the ischemic penumbra from recanalization‐associated adverse events.[100] ^18 Indeed, clinical hypothermia trials in patients subjected to recanalization have shown promising results. Intraarterial mild hypothermia significantly reduced infarct volume compared with that in the control group, with no apparent increase in adverse reactions.[101] ^19 Moreover, the hypothermia group's modified Rankin scale scores displayed a positive trend of improvement at 3 months.[102] ^18 In parallel, intravascular hypothermia improved the neurological scores of stroke patients who received intravenous thrombolysis.[103] ^20 Selective hypothermia therapy, as an adjunct to thrombolysis, appears effective when the target temperature reduction encompasses immediately the ischemic penumbra after the stroke onset. Microcatheter delivery of cold saline before intravascular revascularization may facilitate this rapid brain‐region specific temperature reduction.[104] ^5 Our present study established HT as an innovative tool to selectively deliver hypothermia concurrent with rtPA treatment into the MCA territory in an effort to reduce thrombolytic adverse events while enhancing their combined, instead of stand‐alone, neuroprotective effects. Targeted injection of HT into the MCA region improved the neuroprotective effect on brain tissues. We demonstrated that HT could effectively reduce brain temperature and cerebral metabolic rate of MCAO rats without significantly affecting cerebral blood flow through multiple physiological indices. We further showed that HT significantly improved neurological function, reduced BBB leakage, decreased hemorrhagic transformation incidence, and suppressed deleterious inflammatory responses in MCAO rats. The continuous hypoperfusion of oxygen in the ischemic penumbra and its consumption of oxygen entails a “mismatch” in cellular energy demands that can lead to cell death. The perturbed cells in this region correspond to the penumbra that exists around the core area of cerebral ischemia injury.[105] ^21 Without intervention (ie, blood flow reconstitution), the penumbra enters a precarious state and may transform into the core area at any time due to lack of energy supply.[106] ^21 Thrombolysis and recanalization through rtPA and mechanical thrombectomy may preserve the ischemic penumbra, but adverse events, particularly hemorrhagic transformation, may accompany such treatments. In ischemic brain injury, a deficient cell metabolism triggers failure of various energy pumps, threatening the ion concentration gradient and leading to cell swelling, activation of autophagic enzymes, and ultimately apoptosis.[107] ^22 , [108]^23 Abrogating this aberrant cell metabolic dysfunction could ameliorate ischemic stroke. Hypothermia stands as a robust strategy to reduce the consumption of energy‐generating adenosine triphosphate under pathological conditions, such as ischemia, providing ample energy supply for key ion pump activities and thereby retarding cell death. In this study, using positron emission tomography‐MR to monitor glucose uptake and metabolism in rat brain tissue, we detected that hypothermia could effectively reduce the metabolism of the stroke brain, taper the neuronal activity and energy demand in the ischemic penumbra, and maintain the physiological balance between oxygen transport and oxygen demand in the ischemic penumbra. In an ideal scenario of recanalization, as the occluded artery is recanalized, the infarcted tissue is reperfused. However, the absence of blood flow after recanalization may persist likely due to the impaired neurovascular unit and onset of microthrombosis. Mild hypothermia treatment may protect the integrity of the neurovascular unit, whereas intraarterial injection of rtPA into the ischemic artery may avoid the formation of microthrombosis. However, because rtPA is an enzyme with a biological activity that is temperature dependent, the thrombolytic efficiency of rtPA treatment under mild hypothermia has been questioned. In vitro studies demonstrated that the catalytic activity of rtPA decreased slightly with decreasing temperatures between 25 and 27 °C.[109] ^24 Clinical trials revealed that point estimates correspond to a benefit of alteplase treatment within the range of 35.5 °C to 37.5 °C but showed a negative trend >37.5 °C. Alteplase did not influence temperature profiles at 72 hours after stroke.[110] ^25 Therefore, the effect on the thrombolytic efficiency of HT appears mild and tolerable. Here, the MCAO model was accomplished using the intraluminal suture occlusion method, so the dissolution effect of rtPA on microemboli could not be explored in the present study. Nonetheless, we examined the neuroprotective effects of HT on reducing rtPA's side effects, that is, hemorrhagic transformation. HT reduced the expression of inflammatory chemokines in the CSF of MCAO rats. In addition, HT upregulated the expression of fibrinolysis‐related proteins, specifically uPAR and cFn, but did not lead to excessive activation of the fibrinolytic system. This modulation of inflammatory chemokines via nonenzymatic reaction of uPAR and cFn accompanied HT‐treated MCAO rats to display a reduced infiltration of neutrophils into the ischemic brain tissue. Through nontargeted metabolomics detection, we found that NT leads to abnormal activation of metabolic pathways such as phosphatidylinositol signaling pathway and pentose and glucuronic acid interconversion pathway, which are associated with abnormal activation of calcium ion channels, harmful inflammatory responses, and pathophysiological processes such as abnormal energy metabolism. Interestingly, HT inhibited these cell death pathways, suggesting the possible neuroprotective mechanisms mediating the therapeutic outcomes of HT in ischemic stroke. Conclusions Altogether, our present observations advance the novel use of HT in the setting of ischemic stroke. Our data showed that selective cooling packaged as a concurrent treatment with rtPA is safe and effective in enhancing neuroprotection while avoiding the adverse hemorrhagic transformation in MCAO rats. Sources of Funding This work was funded by the grants from National Natural Science Foundation of China (82 171 301, 81 971 222) and Health Improvement and Research (2020‐2‐1032). Disclosures None. Supporting information Data S1 [111]Click here for additional data file.^ (193.7KB, pdf) This article was sent to Neel S. Singhal, MD, PhD Associate Editor, for review by expert referees, editorial decision, and final disposition. Supplemental Material is available at [112]https://www.ahajournals.org/doi/suppl/10.1161/JAHA.123.029817 For Sources of Funding and Disclosures, see page 13. Contributor Information Yumin Luo, Email: yumin111@ccmu.edu.cn. Cesario Borlongan, Email: cborlong@usf.edu. Jie Lu, Email: lujie@xwhosp.org. References