Study population
The patients who were admitted within 7 days of symptom onset and were consecutively enrolled between May 2020 and September 2021 in the department of Neurology of Xiangya Hospital of Central South University (Hunan, China) were retrospective screened for this study. We initially identified and screened stroke patients with large-artery atherosclerosis classified by the Trial of Org 10172 in Acute Stroke Treatment (TOAST) criteria (29), while other determined or undetermined stroke etiologies were excluded (Fig. 1). The inclusion criteria of this study were as follows: aged >18 years; acute first-ever ischemic stroke; without undergoing intravenous thrombolysis or endovascular thrombectomy, and availability of MRI or CT for evaluating HT. Patients with advanced cancer, other internal of neurological diseases, diabetes (type 1 and 2), or gastrointestinal disease (including inflammatory bowel disease, celiac disease, intestinal surgery, chronic pancreatitis or other malabsorption disorder) in the past 3 months were excluded. Individuals were also excluded if they had received antibiotics or probiotics within the last 3 months. Attending neurologists assessed neurological status for each patient daily. Patients underwent baseline neuroimaging and follow-up imaging with MRI or CT within 7 days post-stroke. Finally, 16 target patients with HT and 17 matching patients with negative HT were recruited. A control group of 15 healthy subjects without inflammatory diseases or neurological disorders was also studied. Blood and fecal samples were collected at admission after the initial diagnosis. The study was approved by the Ethics Committee of Xiangya Hospital of Central South University and all participants provided written informed consent in accordance with the Declaration of Helsinki.
Clinical data collection and detection of HT
The baseline characteristics, vascular risk factors, acute stroke management, and other laboratory findings of the included patients were obtained by reviewing electronic medical records. Clinical stroke severity was assessed at admission using the National Institutes of Health Stroke Scale (NIHSS) (30). Study patients had a brain MRI or CT at hospital admission but had no evidence of hemorrhage. One or more follow-up brain images were performed in all patients according to an institutional stroke imaging protocol. Follow-up brain images were carried out at the time of any clinical worsening or at 5-7 days after admission. HT was detected on those follow-up brain images. CT-detected HT was defined as the part of increase density in an area of low attenuation (31). In MRI images, the combination of diffusion-weighted imaging (DWI), fluid-attenuation inversion recovery (FLAIR), T1-weighting and T2-weighted imaging is used to detect the characterization of HT (32). The presence of HT on brain image was determined by 2 neurologists independently, who were blinded to clinical outcome. When there was a disagreement between the 2 examiners, a decision was reached by consensus after joint review.
Laboratory Tests
12 h fasting blood samples were collected using ethylenediaminetetraacetic acid (EDTA) tubes on admission. Suspensions of plasma were centrifuged at 3000 g for 10 mins and immediately stored at -80℃ until analysis. The serum concentration of lipopolysaccharides (LPS) was measured with a Limulus amebocyte lysate assay (Houshiji Cod Inc., Xiamen, China). The amounts of LBP, sCD14, zonulin and Matrix metalloproteinase 9 (MMP9) in human plasma samples were measured using commercial kits according to the manufacturer’s instructions (Cusabio Science Co, Ltd, Wuhan, China). The LBP and sCD14 levels in plasma of rats were determined using enzyme linked immunosorbent assay (ELISA) kits according to the manufacturer’s protocols (Elabscience, Wuhan, China). All the samples were run in duplicate.
Fecal DNA extraction and sequencing
Fresh stool samples were collected within 48 hours admission and stored at -80℃ until analysis. Total genome DNA from samples was extracted using CTBA method (33). DNA concentration and purity was monitored on 1% agarose gels. According to the concentration, DNA was diluted to 1ng/μL using sterile water. Variable regions V3-V4 of the bacterial 16S rRNA gene were amplified with degenerate polymerase chain reaction (PCR) primers. All PCR reactions were carried out with 15 μL of Phusion® High-Fidelity PCR Master Mix (New England Biolabs); 2 μM of forward and reverse primers, and about 10 ng template DNA. Thermal cycling consisted of initial denaturation at 98℃ for 1 min, followed by 30 cycles of denaturation at 98℃ for 10 s, annealing at 50℃ for 30 s, and elongation at 72℃ for 30 s. Finally 72℃ for 5 min. For the quantification and qualification of PCR products, mix same volume of 1X loading buffer (contained SYB green) with PCR products and operate electrophoresis on 2% agarose gel for detection. PCR products was mixed in equidensity ratios. Then, mixture PCR products was purified with Qiagen Gel Extraction Kit (Qiagen, Germany). Sequencing libraries were generated using TruSeq® DNA PCR-Free Sample Preparation Kit (Illumina, USA) following manufacturer's recommendations and index codes were added. The library quality was assessed on the Qubit@ 2.0 Fluorometer (Thermo Scientific) and Agilent Bioanalyzer 2100 system. At last, the library was sequenced on an Illumina MiSeq platform and 250 bp paired-end reads were generated.
Sequencing data analysis
Sequences were clustered into operational taxonomic units (OTUs) using the Upare software (34) (Uparse v7.0.1001, http://drive5.com/uparse/) . Sequences with ≥97% similarity were assigned to the same OTUs. Representative sequence for each OTU was screened for further annotation. For each representative sequence, the Silva Database (http://www.arb-silva.de/) was used based on Mothur algorithm to annotate taxonomic information (35). In order to study phylogenetic relationship of different OTUs, and the difference of the dominant species in different samples(groups), multiple sequence alignment were conducted using the MUSCLE software (Version 3.8.31,http://www.drive5.com/muscle/) (36). α diversity was assessed by two different parameters: the Shannon index (http://www.mothur.org/wiki/Shannon) and the Simpson index (http://www.mothur.org/wiki/Simpson). All these indices in our samples were calculated with QIIME (Version 1.7.0) and displayed with R software (Version 2.15.3). Beta diversity analysis was used to evaluate differences of samples in species complexity. Beta diversity on both weighted and unweighted unifrac were calculated by QIIME software (Version 1.9.1). Linear discriminant analysis (LDA) EffectSize Tools were used to compare the differences of microbial community compositions among groups (37). Finally, functional prediction was conducted by phylogenetic Investigation of Communities by Reconstruction of Unobserved States (PICRUSt) to identify enrichment of Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways (38).
Animal care and experimental procedures
Male Sprague-Dawley (SD) rats were purchased from Hunan SJA Laboratory Animal Co. Ltd (Changsha, China) and kept under a 12 h light/12 h dark cycle in standard specific-pathogen-free (SPF) conditions. After acclimatization for 1 week, rats aged 7 weeks were randomly divided into experimental groups of six to eight mice, as follows: HT – infarct with hemorrhagic transformation; non-HT – infarct without hemorrhagic transformation; ABX – antibiotics-treated; non-ABX – control given by sterilized drinking water; F-HT – antibiotics-treated with fecal microbiota transplant from HT rats; F-non-HT – antibiotics-treated with fecal microbiota transplant from non-HT rats. All animal experimentation protocols were approved by the Institutional Animal Care and Use Committee of Central South University. All experiments were carried out according to animal experimentation animal welfare act guidelines.
The antibiotics treatment was performed and controlled as previously report (39, 40). To deplete gut flora, rats were given broad-spectrum antibiotics (vancomycin 0.5g/L, ampicillin 1g/L, metronidazole 1g/L, neomycin sulfate 1g/L; all from Meilun Biotechnology, Dalian, China) in drinking water for 4 weeks. The non-ABX group was received regular drinking water. ABX-containing water was replaced at every 3-4 d, together with the cage bedding. Rats in both groups were subjected to HT procedure after antibiotics treatment or not and then received regular or ABX-containing drinking water for the remainder of the experiment.
Fecal microbiota transplantation was performed as described elsewhere (41). After depletion of endogenous microbiota by antibiotics for 2 weeks, the rats were given by oral gavage once day with freshly extracted fecal microbiota from HT or non-HT rats. To obtain inoculants, 1 g of fresh feces from donor rats were pooled and suspended with 5 ml sterile PBS. The mixture was vortexed and centrifuged for 3 min at 1000 Í g, and the isolated supernatant was immediately given to recipient rats by oral gavage. Fresh feces were prepared within 15 minutes before oral gavage on the day of fecal transplantation. After administration fecal extract for 2 consecutive weeks, these recipients were experienced MCAO procedure. The rats were humanely euthanized on the 5th day after MCAO procedure to collect blood sample, cecal contents and colon for next analysis. The LPS and LBP in the plasma was detected with ELISA as described above.
MCAO procedure and HT model
HT induction was performed as previously described (42). Rats received 50% dextrose (6ml/kg) intraperitoneally, 15 min before middle cerebral artery occlusion (MCAO), to induce acute hyperglycemia. Blood glucose was measured from the tail vein using a glucometer (Freestyle, Alameda, CA). During surgery and postoperative period, rectal temperature was maintained at 37.0℃ by using a feedback-controlled heating pad. Focal cerebral ischemia was performed as previously described (43). In brief, 90 min MCAO was performed under anesthesia followed by up to 5 days reperfusion. In order to expose the common carotid artery, external carotid artery (ECA) and internal carotid artery (ICA), a midline cervical incision was made. The ECA was separated and ligated. A rounded-tip 4-0 siliconized filament was inserted into the internal carotid artery and advanced until the origin of the middle cerebral artery (MCA), approximately 18-22 mm from the insertion point, to produce a sudden drop of cerebral blood flow (CBF) to below 25% of baseline measured using a laser doppler monitor. Following MCAO, rats were placed in temperature-controlled recovery cages for 2 h to prevent post-surgery hypothermia. After 90 min, the occlusion suture was removed to allow reperfusion. Sham-operated rats underwent the same protocol of HT rats, without occlusion of the MCA. The survival of rats in each group was monitored daily.
Assessment of infarct volume and HT
Infarct volume was calculated 5 d after MCAO, as described previously (23). Brains were isolated, frozen on dry ice and serially sliced into 5 coronal sections (2-mm thick). The brain slices were stained with 1% 2,3,5-triphenyltetrazolium chloride (TTC, Sigma) for 15 min. The cerebral infarct area was calculated using Image J software 6.0 (Fig. 6D; red areas indicate no infarction, and white areas indicate infarction). In order to correct formula, the infarct area was calculated with the formula: the percentage of hemisphere lesion volume after edema correction = [total infarct volume - (ipsilateral hemisphere volume – contralateral hemisphere volume)]/contralateral hemisphere volume Í100%.
The stained brain sections are collected for hemoglobin content, quantified by spectrophotometric assay (43). After homogenization and centrifugation of brain tissue, the supernatant aliquots were collected. A 20-μL aliquot of supernatant was added with 80 μL of Drabkin reagent (Sigma) followed by optical density (OD) measurement at 540 nm. A blinded investigator scored macroscopic HT in brain slices using a four-point rubric (44), as follows: 0 – no hemorrhage; 1 – dispersed individual petechiae; 2 – confluent petechiae; 3 – small diffuse hemorrhage or hematoma; 4 – large diffuse hemorrhage or hematoma (Fig. 7F). The total score for each rat was reported.
Behavioral testing
Neurological function was assessed using Garcia score (45). The test was conducted 1, 3 and 5 d of reperfusion. The Garcia score is a composite of six parts: spontaneous activity, symmetry of movements (four limbs), symmetry of forelimbs (outstretching while held by tail); vibrissa touch, body proprioception, and the capacity of climb. Neurological function was graded on a scale of 0 to 18 points (normal score, 18; maximal deficit score, 3). The lower the score, the more severe the injury.
Sensorimotor deficits were assessed 2 and 4 d after MCAO surgery by the contact and removal adhesive tape test (46). Briefly, the rats were placed in a clear plexiglass box and allowed to explore the new environment for 2 or 3 min. A 10 mm diameter black color adhesive label was placed on the inside surface of each forelimb. Then we returned the rats to the plexiglass box and recorded the time for the rats to remove the first label and all other labels, respectively. Post-MCAO trials were conducted two times for 180 s each time to diminish stress effects related to handing. Results were expressed as mean of the two trials of either the contralateral contact or removal time. The two neurobehavioral tests were performed by investigators blinded to the experiments.
Western bolt analysis
Samples for western blot analysis were prepared as previously reported (47). The colon tissues were collected and lysed in SDS lysis buffer with the protease inhibitor phenylmethane sulfonyl fluoride (Beyotime Biotechnology, Beijng, China). BCA protein assay kit was used to measure the protein concentrations according to the manufacture’s instructions. 5 × loading buffer (20% mercaptoethanol, 16% glycerol, 0.05% bromophenol blue, and 2% sodium dodecyl sulfate) was added to each sample before boiling for 5 min. The samples were stored at -80 °C until analysis. Equivalent amounts of protein (20 μg) were separated on sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). Then the membranes were incubated with primary antibodies at 4 °C overnight (anti-ZO1 tight junction antibody, 1:1000, Proteintech, Wuhan, China, 21773-1-AP; anti-TLR4 antibody, 1:1000, Proteintech, Wuhan, China, 19811-1-AP; anti-NF-κB p65 antibody, 1:1000, Proteintech, Wuhan, China, 10745-1-AP; anti-β-actin antibody, 1:10000, Cell Signaling Technology, MA, USA, BH10D10). The membranes were then washed in TBST and incubated with corresponding secondary antibody (goat anti-rabbit or mouse IgG, Jackson, 1:5000 dilution in blocking buffer) for 1 h at room temperature. The blots were quantified using the ChemiDoc MP System (Bio-Rad, Hercules, CA, USA). The band intensities of target protein were analyzed using Quantity One 4.4.0 software (Bio-Rad, Hercules, CA, USA). Outliers were excluded from the statistical analysis.
Statistical analysis
Data were analyzed using SPSS software 22.0 (IBM Corp., Armonk, NY, USA) and GraphPad Prism 9 (San Diego, CA, USA). A P value < 0.05 was considered statistically significant. All of the data were tested for normality and variance homogeneity. Statistical significance between three groups was determined by a one-way analysis of variance (ANOVA) followed by post-hoc Tukey’s multiple comparison test. Statistical comparisons between two groups were analyzed with Student’s t-test. Nonparametric analyses were performed with Mann-Whitney U test (two groups) or the Kruskal-Wallis H test (three groups). The baseline characteristics of included patients were presented as number (%) and continuous variables with normal distributions were presented as the mean ± SD, while other variables that were not normally distributed were shown as the median (IQR). Patient characteristics were compared using the χ 2 test for trend for categorical data, one way ANOVA for normally distributed continuous data followed by the LSD test (equal variances assumed) or Tamhance’s T2 test (equal variances not assumed), and the Kruskal–Wallis test for non-normally distributed continuous data. Spearman’s correlation coefficient was used to analyzed the association between gut microbiota and plasma biomarkers.