What this is
- This research investigates the role of () in stroke patients receiving reperfusion therapy.
- It focuses on how different treatments impact miRNA expression and subsequent functional outcomes.
- The study identifies specific associated with poor recovery, suggesting their potential as biomarkers for treatment efficacy.
Essence
- Reduced levels of certain exosomal in stroke patients receiving combined rt-PA and mechanical thrombectomy (MT) correlate with poorer functional outcomes. Specifically, lower expression of miR-17, miR-20, miR-186, and miR-222 is linked to unfavorable recovery at 90 days post-stroke.
Key takeaways
- Exosomal miRNA levels decreased significantly in patients treated with combined rt-PA/MT therapy by day 10 compared to other treatments. This suggests that the combined therapy may negatively impact miRNA expression, which could be linked to poorer recovery outcomes.
- Patients with a () score of 3–6 at 10 days post-stroke had lower levels of miR-17-5p, miR-20a-5p, miR-186-5p, and miR-222-3p. This indicates that these may serve as biomarkers for predicting functional recovery.
- The study emphasizes the need for further research to validate the role of in stroke recovery and their potential as therapeutic targets.
Caveats
- The study involved a limited sample size of 72 patients, which may affect the generalizability of the findings. Larger cohorts are needed to confirm the results.
- Individual patient factors such as medication, rehabilitation, and mental health could influence functional outcomes, potentially confounding the results.
Definitions
- exosomal microRNAs (miRNAs): Small non-coding RNA molecules found in exosomes that regulate gene expression and may play a role in cellular responses.
- modified Rankin Scale (mRS): A scale used to measure the degree of disability or dependence in daily activities of people who have suffered a stroke.
Simplified
1. Introduction
Stroke is one of the leading causes of death and long-term disabilities in adults worldwide. The Global Burden of Diseases report states that the number of strokes increased worldwide between 1990 and 2019, particularly in the population of patients under the age of 65 [1]. The modern reperfusion treatment (RT) of ischemic stroke allows for lysis and/or the removal of thrombus from the closed lumen of the carotid or cerebral arteries via intravenous thrombolysis with RT-plasminogen activator (rt-PA), and/or mechanical thrombectomy (MT). These methods increase the likelihood of penumbra reperfusion, which occurs in the ischemic region around the ischemic core. RT (i.e., intravenous rt-PA and/or MT) aims to preserve as many hypoxic cells as possible in this zone, including neuronal, endothelial, and glial cells.
Intravenous thrombolysis (rt-PA) has been the standard treatment for stroke since 1995. The administration time, initially 3 h post-disease onset, has increased to 4.5 h [2]. However, in about half of the patients, rt-PA may not adequately restore blood flow, or may increase the risk of bleeding in the brain [3]. Introduction of MT in 2016 was a breakthrough in the treatment of acute ischemic stroke caused by large-vessel occlusion (LVO). Based on the results of the meta-analysis, the use of MT in stroke patients results in the recanalization of over 80% of the arteries that underwent the intervention and a return to full patient independence in approximately 45% of patients within 3 months after stroke. However, according to international stroke registries, mechanical reperfusion has been unsuccessful in up to 30% of treated patients, with potentially detrimental consequences for functional outcomes [2,3,4]. This reveals that a significant proportion of patients have futile recanalization (as their angiogram showed recanalization). To date, evidence of the efficacy and safety of rt-PA preceding MT is scarce. There has been a debate whether MT alone is as effective as the combined treatment of rt-PA with MT regarding clinical outcomes of LVO-stroke patients. Some authors suggest that a combined treatment is associated with higher improvement in the functional outcomes of patients at 90 days post-stroke compared with MT alone, while others did not report any additional clinical benefit of a combined treatment [5,6,7]. We expect the novel thrombolytic agent tenecteplase to enhance the clinical effectiveness of combined reperfusion therapy. There are likely various factors that influence the clinical outcome of MT. The unfavorable prognostic parameters identified in the subpopulation of endovascularly treated stroke patients include older age, diabetes mellitus, severe neurological deficit on the first day of stroke, and a low ASPECTS scale score [8,9,10,11]. Two recent studies (SWIFT DIRECT and DIRECT-SAFE) did not show noninferiority of MT in the comparison to rt-PA with MT in terms of functional outcomes [12,13]. Recent studies and meta-analyses have highlighted MT as an important treatment modality for stroke because of its potential to increase the length of the therapeutic window [14,15]. According to the latest guidelines, the MT can be performed up to 24 h after stroke onset in precisely selected patients, based on radiological and clinical findings [16]. Regardless, up to 9% of patients reocclude within 48 h after complete recanalization. Administration of antiplatelet therapy (aspirin, clopidogrel) is recommended in patients with AIS within 24 to 48 h after the onset. For those treated with reperfusion therapy (MT, rt-PA), antiplatelet treatment is started after a CT of the head performed 24 h after procedure. For patients with double occlusion disease (severe carotid stenosis and MCA occlusion) undergoing carotid angioplasty with stent placement, and potentially additional MT, antiplatelet therapy begins up to 12 h post-procedure. For vulnerable patients, such as those with atrial fibrillation, anticoagulants are used in the acute stroke phase following a brief period of antiplatelet therapy.
In the age of emerging personalized medicine, biochemical and molecular biomarkers are being investigated to improve the prognostic value regarding the course of stroke and post-stroke disability. Exosomes are the smallest subtype of extracellular vesicles and are secreted by almost all eukaryotic cells. The vesicles are between 50 and 150 nm in size, are surrounded by a lipid bilayer, and are estimated to contain 0.0001% of the cellular volume [17], in which we may find various soluble substances such as lipids, protein, DNA, and RNA, including miRNA [18]. Exosomes released from cells can transfer miRNAs between cells and tissues, significantly impacting recipient cell physiology and gene expression [19,20]. Examination of exosome content in body fluids is a widely used method for biomarker investigation. Various studies currently focus on discovering miRNAs for use as potential biomarkers for stroke diagnosis and prognosis [21]; therefore, by examining the patient's miRNA profile in the first hours and days after ischemic stroke, one could predict which processes (i.e., neuroprotection or neurodegeneration) will predominate in that patient [22,23,24,25].
Identifying parameters that diminish the clinical benefits of MT involves understanding the specific molecular processes linked to restoring blood flow in hypoxic areas and the mechanisms causing reperfusion injury. This understanding could aid in predicting clinical deficits in patients who have experienced an ischemic stroke, ultimately enhancing the guidance of stroke reperfusion treatment. The search for the most effective tools for stroke treatment has been ongoing for years. Incorporating patients' miRNA profiles in the selection process for MT could assist neurologists in minimizing the risk of failure and enhancing the clinical outcomes of the procedure. This study is the first to report on changes in exosomal miRNA expression during the acute phase of stroke in patients who underwent different endovascular treatments. The clinical condition of patients can vary in the hours and days following a stroke. Therefore, we examined how individual therapies (antithrombotic, rt-PA or MT) or combined therapy (rt-PA/MT) influenced the exosomal miRNA profile 10 days after the stroke occurred. We also compared the levels of different miRNAs assessed on days 1 and 10 to the 10-day and 90-day mRS functional outcome scale (good functional outcome—mRS 0–2, and poor functional outcome—mRS 3–6). Finally, we conducted a functional analysis of the target genes of the miRNAs to determine their involvement in metabolic pathways. In order to identify and understand the biological role of miRNAs in various stroke treatments and patients' functional outcomes, we examined how changes in miRNA expression profile induced by reperfusion might impact the estimated targets (ETs) and their relationship to ischemic stroke treatment.
This study is the first to analyze changes in exosomal miRNA expression during the acute phase of stroke in patients who received different endovascular treatments. Patients' clinical condition can change in the hours and days after a stroke. We investigated how individual therapies (antithrombotic, rt-PA, or MT) or combined therapy (rt-PA/MT) affected the exosomal miRNA profile after 10 days post-stroke. We compared miRNA levels on days 1 and 10, with the 10-day and 90-day mRS functional outcome scale in stroke patients (good functional outcome—mRS 0–2, poor functional outcome—mRS 3–6). Additionally, we analyzed the target genes of the miRNAs to understand their role in metabolic pathways. By examining changes in miRNA expression induced by reperfusion, we aimed to understand their impact on estimated targets (ETs) and their relation to ischemic stroke treatment and patients' functional outcomes.
2. Results
2.1. Study Design
We examined 47 miRNAs in exosomes from serum samples from stroke patients on days 1 and 10 after stroke onset. These 47 miRNAs were selected based on the pilot analysis of exosomal miRNAs from human serum using Human TaqMan Advanced miRNA Array Cards A (Thermo Fisher Scientific, Waltham, USA), which contained 384 human miRNAs. MiRNAs well expressed in exosomes were selected for further analysis. Our objective was to identify specific miRNAs that exhibited varying regulation in patients who underwent different treatments (antithrombotic, rt-PA, MT, or rt-PA/MT combination) within first 10 days from stroke onset and had different functional outcomes (good or poor) as evaluated by the modified Rankin scale [26] on day 10 and 90 post-stroke. In addition, we conducted gene target estimation and enrichment analysis to determine the significant gene targets affected by these miRNAs and the biological pathways they are involved in. Figure 1 illustrates the study's design, while Table 1 summarizes the patients' comprehensive clinical characteristics that were analyzed in the study.
2.2. DEmiRNA Identification
In this study, we analyzed 47 selected miRNAs. These miRNAs included let-7g-5p, miR-15a-5p, miR-16-5p, miR-17-5p, miR-20a-5p, miR-21-5p, miR-23a-3p, miR-26b-5p, miR-30b-5p, miR-92a-3p, miR-93-5p, miR-103a-3p, miR-107, miR-125b-5p, miR-126-3p, miR-130a-3p, miR-142-3p, miR-143-3p, miR-148a-3p, miR-150-5p, miR-152-3p, miR-153-3p, miR-181c-5p, miR-185-5p, miR-186-5p, miR-193b-3p, miR-193a-5p, miR-199a-3p, miR-205-5p, miR-210-3p, miR-221-3p, miR-222-3p, miR-223-3p, miR-224-5p, miR-326, miR-339-5p, miR-342-3p, miR-361-5p, miR-376a-3p, miR-423-5p, miR-424-5p, miR-484, miR-486-5p, miR-505-3p, miR-576-5p, miR-652-3p, and miR-744-5p (). Figure S1
We compared the mean Ct levels of the studied miRNAs between day 10 and day 1 after ischemic stroke onset in each group (Table 2). No DEmiRNAs were identified in the group treated with the aspirin or MT alone. However, in the group treated with rt-PA alone, the level of miR-152-3p was higher on the 10th day compared to the 1st day after stroke. In contrast, when patients underwent combined treatment with both rt-PA/MT, we found 13 DEmiRNAs that were downregulated on the 10th day after stroke onset: miR-15a-5p, miR-16-5p, miR-17-5p, miR-20a-5p, miR-92a-3p, miR-93-5p, miR-152-3p, miR-153-3p, miR-185-5p, miR-186-5p, 210-3p, miR-222-3p, miR-424-5p, and miR-486-5p (Figure 2).
Next, we compared the levels of miRNAs (the fold change) between the studied groups (Table 3). In the analysis's result, we found eight DEmiRNAs between studied groups; among them were miR-15a-5p, miR-16-5p, miR-17-5p, miR-142-3p, miR-152-3p, miR-486-5p, miR-505-3p, and miR-744-5p. The levels of miR-15a-5p, miR-17-5p, miR-16-5p, miR-142-3p, miR-486-5p and 505-3p were lower in the group treated with combined treatment (rt-PA/MT), but (i) miR-15a-5p and miR-17-5p compared to aspirin treatment, (ii) miR-15a-5p, miR-16-5p, miR-142-3p, miR-486-5p, and miR-505-3p compared to rt-PA treatment, and (iii) miR-505-3p compared to MT treatment. The level of miR-152-3p was significantly decreased in the group treated with MT compared to aspirin treatment, when miR-744-5p expression was significantly downregulated in groups treated with rt-PA or groups treated with MT when compared to aspirin.
2.3. Identifying DEmiRNA Related to Stroke Patient's Functional Outcome
We used the mRS scale to categorize patients as functionally poor (mRS ≥ 3) or functionally good (mRS < 3) based on their discharge (10-day mRS) and 3-month post-stroke functional status (90-day mRS). Of the study participants, 58.3% (42 patients) had a mRS score of 3 or higher at 10 days, compared to 48.6% (32 patients) at 90 days (Figure 3).
In the presented study, we found that a decline in miR-17-5p, miR-20a-5p, miR-186-5p, and miR-26b-5p levels between days 1 and 10 after a stroke was associated with patients' 10-day mRS scores of 3–6, which is an indicator of a negative stroke outcome (Figure 4A,B,D,F, respectively). This downregulation of miR-17-5p, miR-20a-5p, and miR-186-5p was also observed in patients receiving rt-PA/MT therapy (Figure 2). We compared prevalent miRNA profiles (days 1–10 post-stroke) in patients experiencing poor functional outcomes three months later (90-day mRS 3–6) with good ones (90-day mRS 0–2), for predictive purposes. These patients exhibited downregulation of miR-23a-3p, miR-26b-5p, miR-181c-5p, and miR-222-3p in the first 10 days after stroke (Figure 4C–E,G, respectively). Patients treated with rt-PA/MT also showed decreased miR-222-3p levels (Figure 2).
2.4. Estimating Gene Targets for Differentially Expressed miRNAs and Their Enrichment Analysis
Our study design (Figure 1) involved estimating gene targets and performing enrichment analysis for each group. However, we only analyzed the rt-PA/MT patient group, as it was the only one with more than one differentially expressed miRNA.
2.4.1. Estimating Gene Targets for DEmiRNA in Patients Receiving rt-PA/MT Treatment
In the group of patients receiving the rt-PA/MT treatment, we conducted target estimation analysis for the four DEmiRNAs with the highest fold-change: miR-15a-5p, miR-16-5p, miR-20a-5p, and miR-424-5p, using the MSigDB database [29]. We found that these DEmiRNAs collectively affect 1738 estimated gene targets. To narrow down the targets, we focused on the ones shared by these four DEmiRNAs (Figure S2A). Specifically, we found that miR-15a-5p, miR-16-5p, miR-20a-5p, and miR-424-5p have 64 shared distinct targets, including genes ABHD2, ABL2, ACTR2, AGO4, APP, ARHGAP12, ATG14, ATXN7L3B, BTN3A3, BZW1, CAPZA2, CCND1, CCND2, CCDC88C, CHIC1, CLIP4, CNKSR3, CRIM1, CRK, DCTN5, DNAJC10, EIF2B2, ELK4, FOXK1, FZD9, HSPA4L, HSPA8, ITGA2, KIF23, KLHL15, LAMC1, L2HGDH, MAP2K3, MINK1, MTMR3, N4BP1, NUFIP2, OCRL, PHLPP2, PLRG1, PPP6R3, RAB3IP, RACGAP1, RFK, RPL14, RPRD2, SHOC2, SIK1, SMAD7, SNTB2, SOCS5, SPRED1, SSRP1, TMEM100, TMEM245, TNRC6B, TXNIP, USP48, WNK3, WEE1, VEGFA, YTHDC1, and ZMAT3. Among brain diseases, only 37 ETs have been recorded, with stroke being the subject of 24 of them. In our literature research, we discovered 13 potential neuro-restoring ETs: CAPZA2, KIF23, ACTR2, SHOC2, FZD9, CRIM1, LAMC1, YTHDC1, EIFB2, BZW1, SKI, HSPA8, and ATG14 (Table 4). The 21 remaining ETs, including ARHGAP12, RAB3IP, ABL2, MAP2K3, WNK3, SIK1, AGO4, RPL14, VEGFR2, SPRED1, CRKL, ITGA2, CCDN1, ZMAT3, APP, SMAD7, SOC5, BTN3A3, RFK, TXNIP, and DNAJC10, may contribute to neurodegeneration and lead to unfavorable functional outcomes (Table 4).
2.4.2. Estimating Gene Targets for DEmiRNA in Patients Categorized Based on the 10-Day mRS Score
In our analysis, we estimated the gene targets of four differentially expressed miRNAs: miR-17a-5p, miR-20a-5p, miR-26b-5p, and miR-186-5p. These miRNAs showed significantly decreased expression levels within first 10 days in patients with a 10-day mRS score ≥ 3 compared to those with a 10-day mRS score < 3. Using the MSigDB database [29] for gene target estimation, we identified 2429 gene targets affected by these DEmiRNAs. To further refine our analysis, we focused on the targets shared by all four miRNAs (Figure S2B). Among the miR-17a-5p, miR-20a-5p, miR-26b-5p, and miR-186-5p, we identified 21 common targets: ABHD2, BTG2, BTN3A3, CNOT4, CSNK1A1, DCBLD2, EIF4G2, MCL1, NUFIP1, PANK3, PLS1, PMAIP1, PPP1R15B, PTP4A1, RBM12B, RNASEH1, SLC28A1, SREK1IP1, VPS13C, ZNF426, and ZNF652. Out of these, only six have been reported in brain diseases, with four specifically related to stroke: MCL1, SLC28A1, BTG2, EIF4G2, PMAIP1, and RBM12B (Table 4).
2.4.3. Estimating Gene Targets for DEmiRNA in Patients Categorized Based on the 90-Day mRS Score
We conducted the gene target estimation analysis for miR-23a-3p, miR-26b-5p, miR-181c-5p, and miR-222-3p. These DEmiRNAs showed significantly decreased expression within first 10 days after stroke in patients with 90-day mRS scores ≥ 3 compared to patients with 90-day mRS scores < 3. From the gene target estimation analysis in the MSigDB database [29], we identified 3941 gene targets affected by these DEmiRNAs. We then focused on the targets that were common for these DEmiRNAs (Figure S2C). Among miR-23a-3p, miR-26b-5p, miR-181c-5p, and miR-222-3p, we found one shared target: PTEN (Table 4).
2.5. Analysis of the Enriched Pathways Targeted by DEmiRNAs Using Various Databases, Including PANTHER, Disease Alliance, HALLMARK, WIKI, KEGG, and Elsevier
2.5.1. Analysis of the Enriched Pathways Targeted by DEmiRNAs in the Patients Receiving rt-PA/MT Treatment
Enrichment analysis for rt-PA/MT group was performed for hsa-miR-15a-5p, hsa-miR-16-5p, hsa-miR-20a-5p, and hsa-miR-424-5p, which showed the highest fold change in the group. We used six open access databases: PANTHER, Disease alliance, HALLMARK, WIKI, KEGG, and Elsevier (Figure 5).
PANTHER database analysis showed that the ETs were enriched in various pathways such as ATP synthesis, Hedgehog pathway, p53 pathway by glucose deprivation, hypoxia response to HIF activation, oxidative stress response, PDGF signaling, apoptosis signaling, angiogenesis, and VEGF signaling (Figure 5A). The Disease alliance database revealed the ETs were primarily associated with conditions like diabetes, portal vein thrombosis, neuropathy, multiple system atrophy, cognitive disorder, and vascular dementia (Figure 5B). In the HALLMARK database, the ETs were found to be involved in pathways such as TGFβ signaling, unfolded protein response, mTORC1 signaling, TNFα signaling via NFκB, NOTCH signaling, apoptosis, p53 pathway, hedgehog signaling, and ROS pathway (Figure 5C). According to the WIKI database, the ETs were primarily associated with copper metabolism, mitochondrial β-oxidation, neuroinflammation, PDGFR-β pathway, extracellular vesicle-mediated signaling in recipient cells, hippocampal synaptogenesis and neurogenesis, VEGFA-VEGHR2 signaling, ferroptosis, BDNF, and neurodegeneration with brain iron accumulation (NBIA) (Figure 5D). In the KEGG database, the ETs were enriched in pathways such as cellular senescence, mitophagy, neurotrophin signaling, autophagy, HIF-1 signaling, ferroptosis, Parkinson's disease, VEGF signaling, apoptosis, and long-term depression (Figure 5E). Lastly, the Elsevier database screening revealed that the ETs were enriched in pathways such as neurons necrosis caused by energy deficiency, brain cell necrosis by vascular dementia, blood-brain barrier disruption by epileptiform disorders, Ca2+ cytosolic overload, vascular smooth muscle cell/pericyte differentiation and proliferation, endothelial cell dysfunction in arterial hypertension, vascular endothelial cell activation by NO, and smooth muscle cell dysfunction in arterial hypertension (Figure 5F).
2.5.2. Analysis of the Enriched Pathways Targeted by DEmiRNAs in Patients Categorized Based on the 10-Day mRS Score
Enrichment analysis (Figure 6) was conducted using the estimated targets (ETs) of DEmiRNAs in the first 10 days after stroke onset in patients categorized based on a 10-day mRS score (Figure 4).
PANTHER database analysis showed that ETs were enriched in various pathways, such as circadian clock system, FAS signaling, oxidative stress response, apoptosis signaling, axon guidance, p53 pathway, Ang II-stimulated signaling, Toll receptor signaling, TGF-beta signaling, and p53 pathway feedback loops 2 (Figure 6A). The Disease alliance database revealed ETs in conditions like intracranial aneurysm, carotid artery disease, bipolar disorder, transient cerebral ischemia, brain ischemia, and arteriosclerosis (Figure 6B). In the HALLMARK database, the ETs were involved in pathways like apoptosis, TGF-beta signaling, mTORC1 signaling, TNF-alpha signaling via NF-kB, unfolded protein response, angiogenesis, ROS pathway, protein secretion, p53 pathway, and hypoxia (Figure 6C). According to the WIKI database, the ETs primarily augmented pathways related to unfolded protein response, apoptosis modulation by HSP70, angiogenesis, angiotensin II receptor type 1 pathway, mitochondrial long-chain fatty acid beta-oxidation, omega-9 fatty acid synthesis, effects of nitric oxide, neuroinflammation, and Il-10 anti-inflammatory signaling pathway (Figure 6D). In the KEGG database, we found ETs enriched in pathways, such as p53 signaling, circadian rhythm, protein export, apoptosis, TGF-beta signaling, ferroptosis, HIF-1 signaling pathway, IL-17 signaling, oxidative phosphorylation, and TNF signaling (Figure 6E). Lastly, screening the Elsevier database revealed ETs enriched in pathways related to brain cell necrosis in vascular dementia, vascular smooth muscle cell/pericyte differentiation and proliferation, HIF-1 signaling, exocytosis vesicle trafficking, macroautophagy decline, apoptosis, ROS in triggering vascular inflammation, and vascular endothelial cell activation by blood coagulation factors (Figure 6F).
2.5.3. Analysis of the Enriched Pathways Targeted by DEmiRNAs in Patients Categorized Based on the 90-Day mRS Score
Enrichment analysis (Figure 7) was conducted using the estimated targets (ETs) of DEmiRNAs in the first 10 days after stroke onset in patients categorized based on the 90-day mRS score (Figure 4).
PANTHER database analysis showed that the ETs were enriched in various pathways, including the insulin/IGF pathway, FAS signaling, PI3 kinase pathway, apoptosis signaling, p53 pathway, oxidative stress response, interleukin signaling pathway, TGFβ signaling, PDGF signaling, and inflammation (Figure 7A). The Disease alliance database revealed the ETs were associated with carotid artery disease, muscular atrophy, bipolar disorder, transient cerebral ischemia, hypertension, arteriosclerosis, middle cerebral artery infraction, and Alzheimer's disease (Figure 7B). In the HALLMARK database, the ETs were found to be involved in pathways such as apoptosis, unfolded protein response, TNFα signaling via NFκB, ROS pathway, TGFβ signaling, p53 pathway, mTORC1 signaling, inflammatory response, hypoxia, and angiogenesis (Figure 7C). According to the WIKI database, the ETs were primarily associated with non-classical roles of vitamin D, unfolded protein response, IL-5 signaling, IL-7 signaling, Interleukin-1 induced activation of NFκB, apoptosis, neuroinflammation, TGFβ receptor signaling, omega-9 fatty acid synthesis, and cells and molecules in local acute inflammation (Figure 7D). In the KEGG database, the ETs were enriched in pathways such as p53 signaling, apoptosis, FoxO signaling, IL-17 signaling, lipid signaling, fluid shear stress and atherosclerosis, ferroptosis, HIF-1 signaling, TGFβ signaling, and TNF signaling (Figure 7E). Finally, in the Elsevier database, the ETs were enriched in pathways related to vascular smooth muscle cell/pericyte differentiation and proliferation, antiphospholipid antibodies in endothelial cells, ROS in triggering vascular inflammation, IL1R STAT3 signaling, apoptosis, low-density lipoproteins and chemokines in atherosclerosis, ER stress (unfolded protein response), leptin in insulin synthesis and secretion, and arterial hypertension (Figure 7F).
3. Discussion
The endovascular and pharmacological treatment for ischemic stroke has improved over the years, leading to a higher survival rate and better recovery prognosis for patients. However, there is still a high percentage of patients who do not recover or have a low recovery rate compared to others with similar clinical profiles. Currently, the main method of treating cerebral infarction is rt-PA, but its application is limited because of the short time window of 4.5 h or 9 h for precisely selected acute ischemic stroke (AIS) patients. Endovascular treatment (MT) has increased the recanalization rate of occluded vessels and extended the time window for stroke intervention to 24 h. The combined treatment of rt-PA/MT is considered the most effective, but the biological mechanism behind its effectiveness is not fully understood. The complex network of miRNAs, a type of non-coding RNA molecules, plays a role in neurological changes during and after ischemic stroke. Some miRNAs have been suggested as potential biomarkers for stroke risk assessment and early detection [77,78], but no studies have focused on the relevance of exosomal miRNAs in stroke therapy effectiveness or their role in patient enrollment for endovascular recanalization treatment or recovery rate. Further research is necessary to fully understand MT's effects, particularly by analyzing the exosomal miRNA profile, given exosomes' ability to transfer miRNA between cells and tissues via bodily fluids. Thus, gene expression may be impacted in organs both near and far from the brain.
The aim of this research project was to determine the impact of stroke treatment on specific miRNAs found in circulating blood exosomes. In addition, we assessed the miRNA that varied between patients of different functional outcomes at day 10 and 90 post-stroke. In this research, we integrated preliminary data of 47 exosomal miRNA's expressions and functional outcomes in stroke patients after one of four different treatments: (i) antithrombotic therapy, (ii) rt-PA, (iii) MT and (iv) combined therapy of rt-PA/MT, with detailed bioinformatics–enrichment analyses.
Our study revealed a significant decrease in the expression levels of miR-15a-5p, miR-16-5p, miR-17-5p, miR-20a-5p, miR-92a-3p, miR-93-5p, miR-153-3p, miR-185-5p, miR-210-3p, miR-222-3p, miR-424-5p and miR-486-5p in the EVs of patients treated with rt-PA/MT upon their hospital discharge. The miR-152-3p was the only miRNA that showed a significant decrease in expression levels between the time of rt-PA treatment and the day of discharge. However, when comparing different treatment groups, the analysis showed different results of fold-change levels analysis for each miRNA. Compared to antithrombotic therapy, miR-15a-5p and miR-17-5p were lower in the rt-PA/MT group, miR-152-3p and miR-744-5p in MT group, and miR-744-5p in rt-PA group. Compared to rt-PA, we found decreased level of miR-15a-5p, miR-16-5p, miR-142-3p, miR-486-5p, and miR-505-3p in the rt-PA/MT group. Only the miR-505-3p expression level differed between the MT group and the rt-PA/MT group.
We compared the levels of specific miRNAs between the 1st and 10th day after stroke, and looked at the functional status of the patients at the 10th and 90th day after stroke.
Our study found that a decline in miR-17-5p, miR-20a-5p, miR-186-5p, and miR-26b-5p levels between days 1 and 10 after a stroke was associated with patients' 10-day mRS scores of 3–6, which is an indicator of a negative stroke outcome. This downregulation of miR-17-5p, miR-20a-5p, and miR-186-5p was also observed in patients receiving rt-PA/MT therapy. We compared miRNA profiles (days 1–10 post-stroke) in patients experiencing poor functional outcomes three months later (90-day mRS 3–6) with good ones (90-day mRS 0–2), for predictive purposes. These patients exhibited downregulation of miR-23a-3p, miR-26b-5p, miR-181c-5p, and miR-222-3p in the first 10 days after stroke. Patients treated with rt-PA/MT also showed decreased miR-222-3p levels.
Previous research, primarily focused on diagnostics, had examined a subset of the miRNAs tested here within serum extracellular vesicles from ischemic stroke patients. These miRNAs include miR-15a-5p, miR-16-5p, miRNA-17-5p, miR-23a-3p, miR-93-5p, miR-152-3p, miR-186-5p, miR-424-5p, miR-505-3p, and miR-744-5p. Compared to healthy controls, the levels of miR-15a-5p [79], miR-152-3p [80], and miR-424 [79] decreased in serum EVs of IS patients. Conversely, the levels of miR-16-5p [81,82], miRNA-17-5p [83], miR-23a-3p [82], and miR-93-5p [83,84] in serum EVs were significantly increased. Notably, there was no difference in the levels of miR-186-5p [85], miR-505-3p, and miR-744-5p between IS patients and healthy controls [82].
Because of the scarcity of miRNA research on circulating EVs, we compared our serum EV stroke findings to those from studies using other blood sources in stroke patients.
However, the reported changes in miRNA levels varied significantly across stroke studies, likely because of differences in the blood samples used. We previously examined serum samples from stroke patients and found that miR-9-3p and miR-9-5p levels increased from day one to day ten after stroke in patients treated with reperfusion therapy [86]; nonetheless, these miRNAs were not found in exosomes in our different investigations, which focused on identifying miRNAs with significant expression in circulating exosomes [87]. In our analysis of existing research on miRNAs, we identified a comparable trend: the expression of miRNAs varied based on the blood sample type employed in the study. The levels of miR-15a-5p were found to be increased in serum from IS patients [88]. Conversely, miR-424-5p concentration was lower in the plasma of AIS patients [89], but higher in their serum compared to the control group [90]. In plasma, the level of miR-16-5p showed an increase [88,91]. Similar results were seen in the plasma [92] and PBMC [93] levels of miR-17-5p, as well as in serum levels [88], which were upregulated in IS patients. However, miR-93-5p levels in plasma and blood neutrophils of AIS patients were notably decreased compared to the control group [94]. Serum levels of miR-186-5p were upregulated in IS patients compared to healthy controls [95,96]. Interestingly, plasma miR-186-5p expression was significantly upregulated in IS patients with normal platelet activation on the first day after stroke onset. After seven days, the levels of plasma miR-186-5p significantly decreased in the same patients with normal platelet reactivity [85].
Even though the expression levels of miR-20a-5p, miR-23a-3p, miR-26b-5p, miR-92a-3p, miR-142-3p, miR-153-3p, miR-181c-5p, miR-185-5p, miR-210-3p, miR-222-3p, and miR-486-5p in serum extracellular vesicles (EVs) have not been studied, they were examined in various blood-related samples taken from patients with IS and compared to those from healthy individuals. There were no significant differences in the plasma levels of miR-20a-5p [97] or the serum levels of miR-23a [98] between IS patients and control subjects. However, in patients with acute ischemic stroke, the expression of plasma miR-26b-5p [81] was downregulated. The serum levels of miR-92a-3p were also reported to decrease in IS patients [81,99]. Interestingly, the use of miR-92a-3p showed high accuracy in diagnosing AIS patients [99]. In a different study, the expression of miR-142-3p in serum [100] and miR-181c-5p in plasma [101] was downregulated in stroke patients compared to healthy controls. Conversely, the serum [81,99] and plasma [92] levels of miR-185-5p increased in IS patients. Another study found a positive correlation between serum miR-185 and NIHSS or mRS score, as well as the area of cerebral infarction [90]. Among the studied miRs, miR-210 was the only one that was examined in time manner. It was found to decrease in serum between admission and 3 months after stroke [102,103]. However, in studies where whole-blood samples from IS patients were tested between 48 h and 10 days from stroke onset [104], and blood leukocytes at 7 and 14 days from stroke onset [105], high levels of miR-210 were correlated with good functional outcomes. In IS patients, the levels of miR-222 decreased in plasma [102]. Conversely, miR-486-5p in peripheral NK cells increased compared to healthy controls [106].
It is widely agreed that stroke is a complex condition with a polygenic nature. The findings from the enrichment analyses and DEmiRNAs target estimation correspond to stroke pathology. Within a 10-day timeframe following a stroke, the expression of all delineated DEmiRNAs diminished, which could subsequently induce upregulation of their targets associated with stroke pathology (see Table 4), encompassing both neurorestorative and neurodegenerative potentials. Stroke events induce significant cellular stress in the brain, resulting from an insufficient oxygen and glucose supply. Under stress, there is a marked increase in mitochondrial β-oxidation in response to glucose deficiency; simultaneously, oxidative stress response and cellular senescence pathways are activated, resulting in irreversible cellular arrest. Acute and delayed cell death inevitably occurs through a regulated apoptosis, ferroptosis because of ion accumulation, or uncontrolled necrosis. Autophagy, with its diverse mechanisms, is critical in post-stroke pathobiology. This may have pro-survival potential through the activation of chaperone proteins. However, when dysregulated, it may cause aberrant protein synthesis and accumulation, potentially causing delayed dementia in cognitive disorders such as Alzheimer's and Parkinson's disease, as well as mental disorders like depression and bipolar disorder. Conversely, differentially expressed genes can activate pathways that regulate cellular development through neuro- and synapto-genesis, re-networking of stroke-affected brain regions, and by augmenting synaptic plasticity. Concurrently with resolving post-stroke tissue response, angiogenesis facilitates vascular smooth muscle cell/pericyte differentiation and proliferation, which is vital for efficient blood perfusion to stroke-affected tissues, thereby protecting neural cells from demise; however, its immaturity may concurrently induce intracerebral brain hemorrhages. Analogous to autophagy, neuroinflammation is essential for an appropriate stroke response and the removal of stroke-related debris; however, its progression into chronic inflammation can cause severe secondary brain damage.
The ultimate functional outcome following a stroke is determined by the critical equilibrium between neurodegenerative and neuroprotective elements, influencing whether patients achieve full recovery or experience delayed neuropathy, multiple system atrophy, cognitive and psychological disorders, or secondary transient cerebral ischemia.
Limitations
Our study has a few limitations. First, the were a limited number of patients (total n = 72) and small group sizes of each tested treatment. The individual variations in patients, like the patient's medication dosage, economic situation, rehabilitation, and the patient's mental condition between day 10 and day 90, may significantly affect the functional outcome evaluation on day 90. As a result, it may mask the biological effect of tested miRNA on the patient's functional outcome in the presented preliminary study. Therefore, further studies should be performed on a bigger cohort of patients, together with sub-grouping patients according to their TICI (thrombolysis in cerebral infraction) scores, and a healthy control group to establish the baseline expression level of miRNAs. Second, the target estimation research was based on a recently published data package, but did not include the most recent studies. The estimated targets for analyzed DEmiRNAs were enriched by the analysis of the OMIC databases and need further in situ revalidation.
Although exosomes constitute a tiny fraction (0.0001%) of a cell's volume, and represent only a small part of their parent cells, even minute alterations could significantly affect the intricate pathophysiology of ischemic stroke. The local abundance of miRNAs and their targets determine their ability to optimize cellular responses to abrupt or chronic changes, while miRNA expression can depend on many individual factors [,]. Therefore, further large-scale population studies are needed to confirm the clinical significance of identified DEmiRNAs. 107 108
4. Materials and Methods
This study included patients hospitalized at the Department of Neurology Upper Silesian Medical Center of the Silesian Medical University in Katowice between 2020 and 2022 because of stroke. The patients suffered from ischemic stroke related to large vessel occlusion, and were treated with aspirin (19 patients included in this study), or rt-PA (15 samples included in this study), or MT (11 patients included in this study), or both rt-PA /MT (27 patients included in this study) in ultra-acute stroke phase.
4.1. Inclusion Criteria
We included patients who met the main criteria: (1) age 50–85 years (2), first-ever symptomatic ischemic stroke diagnosed according to WHO definition and head CT and/or MRI result, (3) pre-stroke status of 0–2 mRankin, (4) no history of intracranial bleeding including the hemorrhagic transformation of the ischemic lesion, and (5) lack of other severe and/or disabling neurological disorders. All patients signed their informed consent. The main exclusion criteria were (1) pregnancy, (2) alcohol abuse/chronic use of psychostimulant, (3) brain tumor, (4) chronic infection/active neoplastic disease, (5) renal/hepatic failure, and (6) surgery in the last three months. The 72 patients who finally qualified into the study were divided into four groups: A total of 19 patients received antithrombotic therapy with aspirin; 15 patients received intravenous thrombolysis with alteplase (rt-PA); 11 patients underwent mechanical thrombectomy (MT); and 27 patients were treated with intravenous thrombolysis (alteplase) and mechanical thrombectomy (rt-PA/MT) (see flow-chart, Figure 1). The characteristics of the 72 qualified patients are presented in the Table 1.
4.2. Serum Sampling
Blood samples (5 mL) were collected twice from each patient by venipuncture into serum separator tubes (BD, Franklin Lakes, NJ, USA) on days 1 and 10 after the stroke onset. After incubation at room temperature for 30–45 min. to allow clotting, the samples were then centrifuged at 1940× g for 10 min at room temperature. The supernatant was collected and pipetted into aliquots (500 µL). Samples were stored at −80 °C until further analysis.
4.3. Exosome Isolation from Serum
Exosomes were isolated as described previously [87,108]. Briefly, exosomes were isolated using the Total Exosome Isolation kit (from serum) (Thermo Fisher Scientific, Waltham, MA, USA) according to manufacturer's instruction. Serum samples were subjected to centrifugation at 2000× g for 30 min to pellet cellular debris. Then, supernatant containing the clarified serum (500 µL) was transferred to a new tube without disturbing the pellet, and placed on ice until ready to perform the isolation. Next, for precipitation, the Total Exosome Isolation kit was added to the cell and debris-free serum (1:2 with exosome isolation reagent and serum, respectively). Serum and the exosome isolation reagents were mixed by brief vortexing and incubated at 4 °C for 30 min before being centrifugated at 4 °C at 10,000× g for 1 h. The pellet containing pre-enriched exosomes was resuspended in 200 µL Exosome Resuspension Buffer (Thermo Fisher Scientific, Waltham, MA, USA). Exosomes isolated using this method expressed the exosome markers CD9 and CD63, and ranged in size between 30 to 200 nm, as shown elsewhere [87,109]. MiRNAs in exosomes are protected by a lipid bilayer and therefore resistant to degradation by circulating ribonucleases [110]. Because of the stable expression of the endogenous control miRNA-320a, stability of the miRNA in our exosomal preparations and between treatments can also be assumed.
4.4. Exosomal miRNA qPCR/RT
Total RNA was isolated using the Total Exosome RNA and Protein Isolation Kit (Thermo Fisher Scientific, Waltham, MA, USA) according to manufacturer's instruction. 200 µL of exosome solution was mixed in 200 µL of 2× Denaturing Solution. Subsequently, 400 µL of Acid-Phenol:Chloroform was used for phase separation. Then, 312.5 µL of 100% ethanol was added on the aqueous phase (250 µL) from acid-phenol:chloroform extraction, and then mixed thoroughly. Finally, RNA was eluted in 50 µL of Elution Solution after being washed three times in RNA Wash Solution. Total RNA (2 µL) from the serum was reverse transcribed using the TaqMan® Advanced miRNA Assays (Thermo Fisher Scientific, Waltham, MA, USA). TaqManTM Advanced miRNA Custom Array Cards for 48 selected miRNAs (Figure S1) were used with the TaqMan® Fast Advanced Master Mix in the QuantStudioTM 7 Flex Real-Time PCR System (Thermo Fisher Scientific, Waltham, MA, USA). The 48 miRNAs were selected based on the pilot analysis of 384 exosomal miRNAs using Human TaqMan Advanced miRNA Array Cards A (Thermo Fisher Scientific, Waltham, MA, USA). MiRNAs with robust expression levels in serum exosomes (Ct < 35) were selected [90].
4.5. MiRNAs Quantification and Differential Expression Analysis
MiRNA-320a was selected as an endogenous control based on stable expression between samples and the literature [90,111].
Data from Polymerase Chain Reaction was normalized to the endogenous miRNA-320a, and mean expression levels of normalized data were later used for all statistical analyses [112].
Statistical analysis was divided into three parts: analysis of DEmiRNAs between the 1st and 10th day post-stroke within each treatment group, analysis of DEmiRNAs fold-changes between treatment groups, and analysis of DEmiRNAs fold-changes in patients categorized based on the 10-day and 90-day mRS score, where good functional outcome was mRS < 3 and poor functional outcome was mRS ≥ 3, according to the modified Rankin Scale (mRS) [26].
We used Student's t-tests (or Welch's t-tests for non-normally distributed data) and Mann–Whitney U tests (for data that did not meet the assumptions of normality) to analyze differentially expressed miRNAs. Four-group comparisons were analyzed using one-way ANOVA with Tukey's post hoc test if data were normally distributed; otherwise, Kruskal–Wallis tests with Dunn's post hoc tests were employed. Normality of the data was assessed using the Shapiro–Wilk test, with the Q-Q plot used for a visual check to see if data transformation was required. Variance homogeneity was assessed with Levene's test. A p-value below 0.05 was considered statistically significant. The significance of confounding variables, including diabetes mellitus, dyslipidemia, atrial fibrillation, and sex, was assessed using Spearman's correlation and Multiple Linear Regression (Tables S19 and S20). No statistically significant findings were observed. This showed that the DEmiRNA findings were independent of confounding variables.
4.6. Estimation of DEmiRNAs Gene Targets and Enrichment Analysis
In part 2, we identified DEmiRNAs using ANOVA for parametric data with equal variances, Welch's ANOVA for parametric data with unequal variances, and the Kruskal–Wallis test for non-parametric data lapping targets for DEmiRNAs in each analyzed group. We linked previously experimentally validated DEmiRNA targets and linked them to terms in the PANTHER (Protein ANalysis THrough Evolutionary Relationships) Classification System [113,114], DISEASE [115], MSig HALLMARK [29], WiKi Pathways [116], Kyoto Encyclopedia of Genes and Genomes (KEGG) [117,118] and Disease Pathways Elsevier databases [119]. To determine the p-value for enrichment analysis, the target occurrence number was compared to the occurrence expected by random chance. By setting the threshold at p-value < 0.05 and Enrichment FDR < 0.05, we could identify significant pathways of DEmiRNAs targets. A fold-change greater than or equal to 2 was significant according to the Fold-change-Specific Enrichment Analysis (FSEA) [86,120].
5. Conclusions
For the first time, our study showed that administering rt-PA/MT altered the exosomal miRNA profile in stroke patients, which might negatively impact their functional outcome. The levels of all the DEmiRNAs in this group were lower on the day of discharge compared to the day of admission, as well as compared to the fold-changes observed in the other study groups. Furthermore, the study revealed an association between reduced miRNA expression and negative stroke outcomes; patients with unfavorable functional outcomes at discharge presented with lower levels of miR-17, miR-20, miR-186, and miR-26b-5p. Similarly, patients with decreased levels of miR-222 at day 10 had a poor score on the mRS scale measured 90 days after suffering a stroke. The functional analysis revealed gene targets and enriched pathways associated with various processes, such as cytoskeleton remodeling, apoptosis, necrosis, ferroptosis, autophagy, inflammation, and dementia.
We would like to emphasize that it is unclear why only 50% of positive reperfusion therapy outcomes, as confirmed by post-thrombectomy angiography, result in a favorable patient status. Our aim has been to identify supplementary clinical and non-clinical parameters to refine patient selection for reperfusion therapy and potentially serve as targets for intervention during the initial phase following a stroke. This report details the potential correlation between the expression of specific mRNAs and the outcomes of post-stroke patients, with functional analysis indicating that certain processes may contribute to cell death and neurodegeneration. Further investigation is required to fully validate the relationship with larger patient cohorts.