Oxymatrine ameliorates white matter injury by modulating gut microbiota after intracerebral hemorrhage in mice

Abstract Introduction White matter injury (WMI) significantly affects neurobehavioral recovery in intracerebral hemorrhage (ICH) patients. Gut dysbiosis plays an important role in the pathogenesis of neurological disorders. Oxymatrine (OMT) has therapeutic effects on inflammation‐mediated diseases. Whether OMT exerts therapeutic effects on WMI after ICH and the role of gut microbiota involved in this process is largely unknown. Methods Neurological deficits, WMI, gut microbial composition, intestinal barrier function, and systemic inflammation were investigated after ICH. Fecal microbiota transplantation (FMT) was performed to elucidate the role of gut microbiota in the pathogenesis of ICH. Results OMT promoted long‐term neurological function recovery and ameliorated WMI in the peri‐hematoma region and distal corticospinal tract (CST) region after ICH. ICH induced significant and persistent gut dysbiosis, which was obviously regulated by OMT. In addition, OMT alleviated intestinal barrier dysfunction and systemic inflammation. Correlation analysis revealed that gut microbiota alteration was significantly correlated with inflammation, intestinal barrier permeability, and neurological deficits after ICH. Moreover, OMT‐induced gut microbiota alteration could confer protection against neurological deficits and intestinal barrier disruption. Conclusions Our study demonstrates that OMT ameliorates ICH‐induced WMI and neurological deficits by modulating gut microbiota.


| INTRODUC TI ON
Intracerebral hemorrhage (ICH) is the second most common subtype of stroke (10%-15%) with high morbidity and mortality, remaining a major cause of death and disability globally. 1 A series of pathological changes, such as the mass effect of hematoma-and hemorrhageinduced secondary injury, including neuronal death, demyelination, axonal degeneration, and glial scarring in the peri-hematoma area, occur after ICH. 2 White matter consists of axons, myelin sheaths, and supporting glial cells. Due to its crucial role in neural signal transmission, both subcortical white matter injury (WMI) and distal white matter fiber tracts disruption produce severe neurological deficits. 3,4 A clinical study reported that WMI was a common pathological event in patients with lobar ICH (present in 77% of patients), 5 which is highly related to the motor impairment. 6 Interestingly, WMI displays potential repairable property, which raises emerging attention from more researchers. However, more attention was paid to the progression and repairment of axonal degeneration and demyelination around primary brain lesions, and little was known about the pathological changes in whiter matter fiber tracts of the spinal cord, especially the corticospinal tract (CST), as well as its role in neurobehavioral recovery. The severity of ipsilateral CST injury was closely associated with the activity of daily living impairment in ICH patients. 7 Our previous work had firstly revealed that striatum hemorrhage induced progressive destruction of the ultrastructure integrity of the CST at the cervical enlargement, which could prolong for at least 5 weeks. 8 Accumulating evidence implicates intestinal bacteria is a crucial factor for the bidirectional communication between the gut and central nervous system. 9 Ischemic stroke patients and transient ischemic attack patients displayed an obvious dysbiotic microbiota, characterized by lower abundance of beneficial commensal taxa and higher abundance of opportunistic pathogens. 10 In an animal model, ICH caused significant gut dysbiosis and gastrointestinal function impairment, and bacteriotherapy by transplanting with healthy microbiota improved neurological function recovery by modulating immune response. 11 The nucleotide-binding oligomerization domain-, leucine-rich repeat and pyrin domain-containing 3 (NLRP3) inflammasome is a molecular complex of the innate immune system and considered as a key contributor to the development and progression of a variety of neuropsychiatric disorders. 12,13 A growing body of research demonstrates that NLRP3 inflammasome plays an important role in the pathophysiology of ICH. [14][15][16] In our previous study, selectively inhibiting NLRP3 inflammasome was sufficient to modulate the gut microbial composition, ameliorate CST injury of the cervical enlargement, and alleviate neurological deficits after ICH. 17 Therefore, gut microbiota may be an important therapeutic target for the progression of ICH. However, the study on the signature of gut microbiota after ICH is still insufficient.
Oxymatrine (OMT) is a main quinolizidine alkaloid extracted from the traditional Chinese herb Sophorae Flavescentis Radix. Emerging studies demonstrated that OMT conferred neuroprotective effects on several neurological disorders, such as ischemic stroke and spinal cord injury. [18][19][20] In preclinical models, OMT protected against cerebral ischemia-reperfusion injury by inhibiting pro-inflammatory signaling pathways. 18 It was reported that OMT could inhibit the progression of acute spinal cord injury by modulating inflammation, oxidative stress, and apoptosis through the TLR4/NF-κB pathway. 19 However, whether OMT exerted therapeutic effects on WMI after ICH, and the role of gut microbiota in these processes are largely unknown.
Therefore, the present study aimed to investigate whether OMT could attenuate hemorrhage-induced WMI, especially secondary CST injury in cervical enlargement, and the role of gut microbiota in the underlying mechanism in an ICH model. We hypothesized that OMT alleviated ICH-induced WMI and neurobehavioral deficits by modulating gut microbiota.

| Ethical approval
The experimental protocols were approved by the Institutional Ethics Committee of Zhujiang Hospital of Southern Medical University (LAEC-2020-227) and conducted according to the guidelines of the National Institute of Health Laboratory Animal Care and Use Guidelines.

| Drug preparation and administration
Animals were randomly assigned into three groups, namely Sham, ICH + Vehicle, and ICH + OMT group ( Figure 1A). The mice of the ICH + OMT group daily received OMT (Cat#: A111286, Aladdin) by oral gavage, which was dissolved in deionized water to reach the concentration of 40 mg/ml, at 120 mg/kg body weight according to previous research, 18 until sacrificed. Both the Sham group and the Conclusions: Our study demonstrates that OMT ameliorates ICH-induced WMI and neurological deficits by modulating gut microbiota.

K E Y W O R D S
fecal microbiota transplantation, gut microbiota, intracerebral hemorrhage, oxymatrine, white matter injury ICH + Vehicle group were given equivalent volumes of deionized water in the same manner.

| Statistical analysis
Statistical analysis was performed using the SPSS software version 20.0 (SPSS). Normality was evaluated using Shapiro-Wilk's test. For normal distribution variables, the student t-test was used to compare the means of two groups. Comparisons among multiple groups were analyzed using one-way analysis of variance (ANOVA) with a post hoc analysis with the Bonferroni method.
Continuous neurological function scores were analyzed by twoway ANOVA adjusted by Tukey's post hoc test. Regarding nonnormal distribution variables, the Kruskal-Wallis H-test and Mann-Whitney U-test were conducted for statistical analysis between different groups. Correlation analysis between bacterial species and inflammatory markers was evaluated by Spearman's F I G U R E 1 OMT alleviated the neurological deficits after ICH. (A) The experimental protocol for OMT treatment after ICH. (B, C) hematoma volume (B, n = 6-7/group) and brain water content (%) (C, n = 4-5/group) at 3 days after ICH. (D, E) mNSS score (D, n = 9-10/ group) and corner turn test (E, n = 9-10/group) at Days 1, 3, 7, and 14 after ICH. Sham group versus ICH + Vehicle group, *p<0.05, **p<0.01, ***p<0.001, ICH + Vehicle group versus ICH + OMT group, # p<0.05, ## p<0.01, ### p<0.001. rank correlation. GraphPad Prism Software version 6.0 (GraphPad Prism) was applied to visualize analysis results. The p-Value <0.05 was considered statistically significant. Other materials and methods are described in the Appendix S1.

| OMT alleviated the neurological deficits after ICH
OMT by oral administration for 3 days reduced hematoma volume and brain water content (BWC, in ipsilateral hemilateral) compared with that of vehicle-treated mice after ICH ( Figure 1B,C). Then, we investigated inflammation by qPCR and found that NLRP3 inflammasome complex (Nlrp3, Asc, and Caspase-1) and pro-inflammatory cytokines (Il-1β, Il-6, Tnfα, and Nos2) mRNA levels were significantly elevated in peri-hematoma region on Days 3 and 14, which were down-regulated by OMT treatment (except Il-6 at 3 days and Tnfα at 14 days) ( Figure S1A,B).
The mNSS test showed that neurological function impairment reached the peak on Day 1 and thereafter gradually declined from Day 3 after ICH. Instead, OMT partially alleviated ICH-induced neurological deficits ( Figure 1D). Similarly, OMT administration significantly reduced the proportion of right turn after ICH ( Figure 1E).
Taken together, these results suggested that OMT could suppress hematoma expansion and brain edema development, inhibit neuroinflammation, alleviate neurological deficits.

| OMT alleviated white matter injury and glial scar formation after ICH
The integrity of white matter was assessed by immunostaining with MBP (a marker of myelin sheath) and NF200 (a marker of axons) ( Figure S2A). Our results demonstrated that both MBP and NF200 were markedly destroyed by ICH, resulting in sharp decrease in the mean fluorescence intensity (MFI) of MBP and NF200. The MFI of them was significantly increased in the OMT-treated group Figure S2A). These findings suggested that OMT attenuated the WMI around the hematoma region.
Subsequently, we detected the expression of MBP and NF200 in the CST region of cervical enlargement to further clarify secondary WMI. The results showed that the expression of NF200 was dramatically decreased after ICH, consistent with that in the perihematoma region, which indicated axonal Wallerian degeneration may extended to the CST region of cervical enlargement. The MFI of NF200 was significantly increased after OMT administration ( Figure 2B). Nevertheless, the MFI of MBP in cervical enlargement was not significantly different in ICH mice with or without OMT treatment ( Figure 2B). This may attribute to the heterogeneous pathogenic processes of the axonal degeneration and demyelination.
Our results indicated that OMT attenuated both proximal and distal WMI at 14 days after ICH.
The formation of glial scar is regarded as another injury indicator after ICH. The chondroitin sulfate proteoglycan (CSPG) is an extracellular matrix compound, derived from reactive astrocytes, which is the major inhibitor of axonal growth. A mass of activated astrocytes, detected by the colocalization of GFAP and CSPG, aggregated at the peri-hematoma region after ICH ( Figure 2C, Figure S2B,C).
The same phenomenon was observed in the CST region ( Figure 2D,

| OMT modulated the gut dysbiosis induced by ICH
To investigate the composition of gut microbiota after ICH and the effect of OMT on this alteration, 16 S rRNA gene sequencing was performed. Microbial profiling depicted that in the acute phase after ICH, the α-diversity, calculated by the Shannon index, manifested a significant decline in both the vehicle-and OMT-treated groups compared with the sham group ( Figure 3A, Figure S3A).
Additionally, the β-diversity, reflecting the differences between microbial communities, estimated by PCoA analysis based on the weighted UniFrac distance, displayed that samples from the ICH + Vehicle group were clearly separated from those from the Sham group on Days 1 and 3 after ICH. And this phenomenon was only observed at 3 days after ICH between the ICH + Vehicle group and the ICH + OMT group ( Figure 3B, Figure S3B). Microbial population analysis revealed that ICH remarkably altered the gut microbial composition, and OMT treatment modulated the composition of gut microbiota at the phyla and the genus levels. LEfSe analysis showed that at 1 day after ICH, the relative abundance of Rikenellaceae and S24_7 was significantly higher but those of In the chronic phase after ICH, there was still a significant decline in the species richness in the ICH + Vehicle group as shown by the Shannon index ( Figure 4A). PCoA analysis revealed that samples from different groups were clearly separated from each other, suggesting the composition of gut microbiota was discriminated between different groups ( Figure 4B). Using LEfSe analysis, we determined that Bacteroidetes, S24-7, Paraprevotellaceae, Prevotella, and Akkermansia become the predominated bacteria in the ICH + Vehicle group, but the relative abundance of above-mentioned species (ex-

| OMT ameliorated intestinal barrier disruption and gut-derived endotoxemia after ICH
To determine intestinal barrier function, we detected the mRNA  The serum concentrations of LPS and several inflammatory factors including TNFα, IL-6, and IL-1β were obviously increased after ICH, which were reversed by OMT ( Figure 5E,F). In summary, OMT attenuated the gut-derived endotoxemia after ICH.

| The correlation between significantly differential microbiome taxa and injury-associated indexes after ICH
In order to investigate the correlation between the relative abundance of microbial species and injury-associated indexes, Spearman's rank correlation was conducted. At 3 days after ICH, members from the Bacteroidetes phylum, including Rikenellaceae and  Figure 6B). To sum up, the relative abundances of some microbial species were significantly associated with ICH-induced pathological changes.

| Gut microbiota partially mediated the therapeutic effects of OMT on ICH
We further conducted FMT to detect the role of the gut microbiota in the pathogenic processes of ICH ( Figure 7A). At 3 days after ICH, recolonization with ICH + OMT-fecal microbiota significantly reduced mNSS scores and right turn bias, compared with ICH-fecal microbiota ( Figure 7B). For intestinal barrier permeability detection, mice recolonized with ICH-fecal microbiota exhibited higher serum fluorescein concentrations than those recolonized with ICH + OMTfecal microbiota, suggesting lower permeability of intestinal barrier after ICH + OMT-fecal microbiota recolonization ( Figure 7C). In addition, the mRNA levels of tight junction proteins were significantly increased in the FMT-ICH + OMT group compared with the FMT-ICH + Vehicle group ( Figure 7D). Furthermore, 16 S rRNA sequencing was performed to reveal the gut microbial composition in recipient mice who received FMT for 3 days. The α-diversity in the FMT-ICH + Vehicle group was significantly lower than those in the FMT-ICH + OMT group ( Figure 7E).

F I G U R E 6
Correlation between significantly differential microbiome taxa and injury-associated indexes after ICH. (A) The heatmap indicated the correlation between microbial species and injury-associated indexes on Days 3 (left) and 14 (right) after ICH. Red and green cells indicated positive and negative correlations, respectively. *p<0.05, **p<0.01, ***p<0.001.
As the PCoA analysis showed, the distribution of samples from the FMT-ICH + Vehicle group was clearly separated from those from the FMT-ICH + OMT group, and the difference in PC1 axis reached significant ( Figure 7F). Population analysis demonstrated that the FMT-ICH + Vehicle group displayed a higher abundance of Proteobacteria and a lower abundance of Bacteroidetes than the FMT-ICH + OMT group at the phylum level ( Figure S4G). The abundance of the predominant seven genus was different between the two groups at the genus level ( Figure S4H).  Figure 7G, Figure S4I). While the relative abundance of Bacteroidetes, Bacteroidaceae, Bacteroides, Parabacteroides, Proteus, and Myroides in the FMT-ICH + Vehicle group were significantly lower than those in the FMT-ICH + OMT group ( Figure 7G, Figure S4I). Above all, the therapeutic effects of OMT on ICH-induced neurological deficits and intestinal barrier dysfunction were partially mediated by gut microbiota. However, the mechanism underlying requires a more thoroughly designed study to investigate.

| DISCUSS ION
WMI induced by ICH plays an important role in the neurological function outcome. 21 In this study, it was shown that ICH can not only cause extensive WMI in the primary lesion but also induce persistent secondary white matter fiber tracts injury at the cervical enlargement, followed by gut dysbiosis, intestinal barrier dysfunction, and systemic inflammation. Oral administration with OMT significantly alleviated ICH-induced WMI in the hematoma region and distal spinal segment, modulated gut microbial composition, and ameliorated intestinal barrier dysfunction. FMT experiment confirmed that gut microbiota was involved in this pathophysiological process after ICH.
Baseline hematoma volume and subsequently hematoma expansion are established prognostic factors of both mortality and functional outcome after ICH. 22,23 Besides, secondary brain injury caused by mass effect and release of blood components, such as edema and inflammation, also plays an important role in the neurological function recovery after ICH. 24 Current study revealed that OMT administration (120 mg/kg) conferred neuroprotective effects on ICH by reducing hematoma volume and inhibiting the formation of edema. Accumulating studies demonstrated that OMT displayed remarkably antiinflammation and immunomodulation effects. NLRP3 inflammasome is a critical receptor and sensor of the innate immune system that aggravated inflammatory response and brain edema progression after ICH, and selectively inhibiting NLRP3 inflammasome activation attenuated ICH-induced brain injury and neurological function deficits. 14,25,26 Our study showed that OMT significantly downregulated NLRP3 inflammasome complex and other pro-inflammatory cytokines mRNA levels at 3 and 14 days after ICH, indicating OMT could effectively inhibit neuroinflammation in the brain. However, the underlying mechanism of the inhibition effect of OMT on NLRP3 inflammasome and neuroinflammation still needs to be verified in further study.
The severity of WMI is closely related to neuromotor dysfunction in ICH patients. 7 In present study, OMT significantly alleviated WMI in the peri-hematoma region at 14 days after ICH. The less extensive WMI in the primary lesion may be account for better neurological function recovery after ICH. Besides, ICH-induced secondary remote white matter fiber tracts (such as CST) injury also plays a pivotal role in the functional outcome. In stroke patients, the integrity of the CST was a determining factor for the proportional recovery of motor impairment. 27 Quantitative tractography revealed that there were more reconstructed CST fiber pathways in ICH patients with favorable outcomes. 28 Based on the results above, treatment with OMT significantly alleviated ICH-induced CST axon injury and demyelination at cervical enlargement, which is perhaps due to the protection of OMT against WMI in the primary lesion, or its inhibition of astrocytes/microglia activation and inflammatory response.
Prolonged activation of microglia induced by myelin debris has long been implicated in aggravating inflammatory action by secreting pro-inflammatory cytokines. 29 Gut microbiota is involved in the development and progression of ischemic stroke, through the microbiota-gut-brain axis. 32 Recolonized with stroke-induced dysbiotic microbiome from special pathogen-free mice, germ-free mice developed a larger infarct F I G U R E 7 Gut microbiota mediated the therapeutic effects of OMT on ICH. (A) The experimental protocol for FMT after ICH. Donor mice, fecal samples were collected from ICH mice with or without OMT treatment on Day 3 after ICH. Recipient mice, C57BL/6J mice were subjected to antibiotics mix by gavage for 1 week, followed by ICH model conduction, and then daily received FMT for 3 days. (B) mNSS score (A, n = 10/group) and corner turn test (n = 9-10/group) on Days 1 and 3 after ICH. (C, D) Serum FD4 concentrations (C, n = 6/group) and the mRNA levels of tight junction proteins (D, ZO-1, Occludin, and Claudin-4, n = 8/group) in colon tissues on Day 3 after ICH. (E, F) Comparison of ɑ-diversity (E, Shannon index), β-diversity (F, PCoA analysis based on weighted UniFrac distance) at 3 days after ICH. (G) The distribution bar plot based on the LEfSe analysis (LDA score (log 10)>3), and the richness of significantly differential taxa between the two groups at 3 days after ICH. n = 9-10/group. Sham group versus ICH + Vehicle group, *p<0.05, **p<0.01, ***p<0.001, ICH + Vehicle group versus ICH + OMT group, # p<0.05, ## p<0.01, ### p<0.001.
volume after distal middle cerebral artery occlusion. 33 However, the role of gut microbiota in the ICH outcome is still poorly understood.
Interestingly, when administrated orally, OMT will be converted into a more absorbable metabolite matrine (MT) by gut bacteria in the gastrointestinal tract. 34

| CON CLUS IONS
In conclusion, our study is the first time to reveal that OMT treatment effectively reduced WMI both at the striatum and distal CST

CO N FLI C T O F I NTE R E S T
The authors declare no conflict of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
All raw data used in this manuscript are available from the corresponding author upon reasonable request.

CO N S E NT FO R PU B LI C ATI O N
Not applicable.