Liangxue Tongyu Prescription Alleviates Brain Damage in Acute Intracerebral Hemorrhage Rats by Regulating Intestinal Mucosal Barrier Function

Background Liangxue Tongyu prescription (LTP) is a commonly used formula for acute intracerebral hemorrhage (AICH) in clinical practice that has significant ameliorative effects on neurological deficits and gastrointestinal dysfunction, yet the mechanism remains elusive. The aim of this study was to investigate the pathway by which LTP alleviates brain damage in AICH rats. Methods The AICH rat models were established by autologous caudal arterial blood injection. The neurological function scores were evaluated before and after treatment. The water content and the volume of Evans blue staining in the brain were measured to reflect the degree of brain damage. RT-PCR was used to detect the inflammatory factors of the brain. Western blotting was used to detect the expression of the tight junction proteins zonula occludens 1 (ZO-1), occludin (OCLN), and claudin (CLDN) in the brain and colon, followed by mucin 2 (MUC2), secretory immunoglobulin A (SIgA), and G protein-coupled receptor 43 (GPR43) in the colon. Flow cytometry was used to detect the ratios of helper T cells 17 (Th17) and regulatory T cells (Treg) in peripheral blood, and the vagus nerve (VN) discharge signals were collected. Results LTP reduced the brain damage of the AICH rats. Compared with the model group, LTP significantly improved the permeability of the colonic mucosa, promoted the secretion of MUC2, SigA, and GPR43 in the colon, and regulated the immune balance of peripheral T cells. The AICH rats had significantly faster VN discharge rates and lower amplitudes than normal rats, and these abnormalities were corrected in the LTP and probiotics groups. Conclusion LTP can effectively reduce the degree of brain damage in AICH rats, and the mechanism may be that it can play a neuroprotective role by regulating the function of the intestinal mucosal barrier.


Introduction
Due to certain factors, such as rapid onset, long duration, and severe sequelae, stroke poses a considerable burden on public health. Te results of the Global Burden of Disease (GBD) 2019 study showed that China has the highest incidence of stroke at over 30%, and intracranial hemorrhage (ICH) accounts for approximately 15% of all strokes [1]. Compared with cerebral infarction (CI), ICH has higher mortality and disability rates [2]. In the acute phase, a hemorrhage mass can directly cause primary brain injury, and the continued existence and expansion of hematoma can lead to secondary brain injury [3]. Currently, conventional treatments that rely on surgery, dehydration to lower intracranial pressure, and blood pressure control have no defnite clinical beneft for the prognosis of patients with ICH [4]. Terefore, exploring efective means to improve stroke prognosis and increase survival rates is necessary.
With the increased understanding of the physiopathological process of ICH, mitigation of secondary brain injury has become a therapeutic focus. Tis is not only related to the infammatory response, neurotoxicity, blood-brain barrier (BBB) disruption, and brain edema but Cheng et al. also confrmed that intestinal mucosal structure and function are disrupted in ICH mice [5,6]. Furthermore, when gut fora homeostasis was improved, neurological defcits were relieved [7]. Based on these fndings, the interaction between the gut and the central nervous system (CNS) has received our attention; moreover, the previously reported theory of the gut-brain axis demonstrates that the two can regulate each other through humoral, cellular immune, and neuronal pathways [8]. Te intestinal mucosal barrier (IMB) has a complex barrier structure to defend against pathogen colonization and endotoxin invasion, with intestinal epithelial cells (IECs) playing the primary protective role [9]. IECs are interconnected by basolateral membrane apical tight junction proteins (TJPs) such as ZO-1, OCLN, CLDN-1, subjacent adherens junctions, and desmosomes, which form a physical barrier that separates the intestinal contents from the underlying tissues [10]. Similarly, the TJPs involved in BBB composition were ZO-1, OCLN, and CLDN-5, which maintained structural permeability as well [11]. Catalyzed by the transcription factors Atoh1 and SPDEF, intestinal epithelial stem cells at the intestinal crypts diferentiate into cup cells that secrete mucins (MUCs) [12]. Short-chain fatty acids (SCFAs) and metabolites of intestinal fora, can combine with GPR43 (Ffar2) located on the surface of IECs, activate the mTOR signaling pathway, and promote Paneth cells to produce antimicrobial peptides (AMPs) [13], and MUCs are seen as chemical barriers [14]. In addition, intestinal endotoxins that cross the MUC layer prompt the conversion of B lymphocytes in the lamina propria into plasma cells that secrete SIgA, initiating humoral immunity [15]. Te intestinal resident CX3CR1+ macrophage population produces interleukin 10 (IL-10), which inhibits the secretion of infammatory factors by colonic T cells and increases the Treg cell ratio, regulating cellular immune homeostasis [16,17]. Studies have confrmed that proteins and amino acids in the duodenum induce the secretion of cholecystokinin (CCK), an enteric neuropeptide that acts on VN sensory fbers under the intestinal epithelium to regulate gastrointestinal motility and transmits satiety signals to the brain [18,19]. Tese previous fndings support the idea that the intestinal mucosa has complex physiological functions and that there is an interaction with the CNS.
Zhongying Zhou, a master of Chinese medicine with more than 70 years of clinical experience, established an LTP formula to treat AICH with obvious efects [20]. Te decoction signifcantly enhances gastrointestinal motility while promoting the recovery of neurological function in AICH [21]. LTP is based on the classic recipe of Xijiao Dihuang decoction from the ancient Chinese medical book "Qian Jin Yao Fang," with some innovations. LTP consists of eight crude drugs such as Rhei Radix et Rhizoma, Bubali Cornu, Rehmanniae Radix, Paeoniae Radix Rubra, Moutan Cortex, Acori Tatarinowii Rhizoma, Pheretima, and Notoginseng Radix et Rhizoma; among them, Bubali Cornu was substituted for Rhinoceri Asiatici Cornu to comply with the Endangered Species Protection Act [22]. According to traditional Chinese medicine (TCM), the main pathological factors in the acute phase of ICH are blood stasis and freheat, categorized as an excess syndrome by syndrome differentiation. Tese generalizations are based on the nature of symptoms and signs. Te formula was for stasis-heat pathology. Rhubarb, the king drug, is mainly used for "purging fu-organs to eliminate heat," treating excess syndrome with purgation, while the remaining drugs exert an efect of "cooling the blood and dissipating the blood stasis." Briefy, LTP not only promotes the dissipation of the stasis-heat pathogen at the lesion but also innovatively improves intestinal dynamics, promotes metabolic circulation, and accelerates the clearance of pathogenic factors in the brain through intestinal function. Interestingly, this concept of organic wholeness coincides with the brain-gut axis theory. In our preliminary studies, we found that LTP had a signifcant reparative efect on intestinal mucosal injury by detecting the levels of lipopolysaccharide (LPS), diamine oxidase (DAO), and D-lactic acid (DLA) in the serum of the AICH rats [23]. Among other things, excess LPS in blood circulation also exacerbates BBB damage [24]. In the case of BBB damage, harmful substances of enteric origin can reach the brain via the blood and aggravate the condition.
Based on the previous results, in the current study, we aimed to explore the following: (1) whether LTPs could improve neurological defcits, brain edema, and BBB disruption in the AICH rats; (2) whether LTPs could enhance IMB functions such as epithelial permeability, MUCs, immune factors, and GPR43 secretion, and maintain T-cell homeostasis and VN activity; and (3) whether there is a more efcacious pathway for LTPs in the treatment of AICH than simply supplementing the intestinal fora by comparing the efects with those of the probiotic group.

Preparation of LTPs.
LTP consists of eight crude drugs: rhubarb (10 g), bufalo horn (30 g), raw Rehmannia (20 g), red peony root (15 g), tree peony bark (10 g), Acorus calamus (10 g), earthworm (10 g), and Panax notoginseng (10 g) ( Table 1). Te aforementioned herbs were concentrated and decocted for adult humans by Nanjing Hospital of Chinese Medicine Afliated with the Nanjing University of Chinese Medicine, containing 1 g crude medicine per mL. In brief, 115 g crude LTP was soaked in 1000 mL pure water for 2 h, and bufalo horn was frst decocted for 30 min. Ten, other medicinal materials were added to a boil, fltered, repeated twice, and mixed.

Preparation of Probiotic Solution.
Te probiotic formula is a live combined bifdobacterium, lactobacillus, and enterococcus capsules (Lot No.: 22120210323) purchased from Shanghai Shangyao Xinyi Pharmaceutical Factory. Te formula was prepared by dissolving 17.5 mg of initial crude drug per mL of solution in saline at 4°C.

AICH Rat Model Establishment and Treatment.
Tirty male specifc-pathogen-free (SPF) Sprague-Dawley (SD) rats (weighing 200 ± 20 g) were produced by Hangzhou Medical College (License No.: SCXK 2019-0002, Hangzhou, China). Rats were housed at the Experimental Animal Center of the Nanjing University of Chinese Medicine at 20.0 ± 1.0°C and 50.0 ± 10% humidity under a 12 h light-dark cycle. Tis experiment was approved by the Experimental Animal Ethics Committee of Nanjing University of Chinese Medicine (Application No. 202105A029). Apart from the normal rats (blank group) and sham-operated rats (sham group), the AICH rat models were established as we previously reported [25]. In brief, the rats were preoperatively fasted and anesthetized by inhalation of 1% halothane with an Animal Microanesthesia Ventilator (Model: R540, RWD Life Science and Technology Co., Shenzhen, China). Ten, the right caudate nucleus was located by a Rat Brain Stereotactic Instrument (Model: 68005, RWD Life Science and Technology Co., Shenzhen, China) (3 mm to the right from fontanel). After cranial drilling, 50 μL autologous caudate arterial blood was slowly injected with a micropump combined injection needle at a depth of 6 mm. After the blood injection, the needle was left for 10 minutes, and then, the needle was withdrawn by 2 mm. Te needle was left for another 5 minutes, and the needle was completely withdrawn. Te sham group was treated similarly, except that the right caudate nucleus was inserted with an empty needle but no blood was injected. Bone wax was used to seal the pinhole and suture the surgical incision. Te neurological function score of the AICH rats was assigned according to Longa's score on a fve-point scale [26], and the models were successfully built for those with a score of 1-3. Ten, according to the random number table method, the AICH rats were randomly divided into the model group, LTP group, and probiotics group, with 6 rats per group.
In addition to the blank group, the other 24 rats (6 in each group) were treated according to the following methods: the sham and model groups were intragastrically (IG) administered normal saline (10 mL/kg), the LTP group was given an LTP decoction (dose: 10 g/kg, IG), and the probiotics group was given probiotics solution (dose: 0.175 g/kg, IG). Tis dose of LTP was confrmed to be more efective in early dose-gradient pharmacological experiments [27,28]. Te frequency of treatment was once a day for a period of 3 days. In the preliminary study, we observed the positive efects of LTP at 24 h, 72 h, and 120 h after the AICH rat model was established; and we found that the results began to be signifcant at 72 h, the efect remained stable until 120 h that we have observed [29].
2.6. Neurological Function Score. According to Longa's scoring method [26], before and after treatment, the degree of neurological defcits in the AICH rats was assessed and the scores were recorded. Te average score was calculated by two professional researchers.

Brain Water Content.
After the rats were sacrifced by anesthesia overdose, the whole brain was removed immediately by decapitation, and then, the olfactory bulb junction, cerebellum, and low brainstem were harvested. Te surface blood stains were rinsed with saline, and the right Evidence-Based Complementary and Alternative Medicine cerebral hemisphere containing the hematoma was separated. Te water stains were dried with disposable absorbent paper and immediately placed in a dry Petri dish and weighed on a microbalance, i.e., wet weight. Te right cerebral hemisphere was held in tin foil, dried in a constant temperature oscillating incubator at 60°C for 24 h, and then weighed, i.e., dry weight. Brain water content (%) � (wet weight − dry weight)/wet weight × 100%.

Evans Blue (EB) Staining.
After the rats were anesthetized, 4% EB solution (dose: 2 mL/kg) was injected into the femoral vein, and the terminal skin and eyes of the rats gradually turned blue during the injection. One hour after the injection, the thoracic cavity of the rats was opened, and normal saline (300 ± 20 mL) was injected through the left ventricle while the right atrium was cut open, and perfusion was stopped when the fuid fowing from the right atrium became clear. Following cardiac perfusion, the right cerebral hemisphere of the rats was stripped and the surface was rinsed well with cold normal saline. Water stains on the surface were dried with absorbent paper and the brain was weighed. Ten, formamide (dose: 10 mL/g) was added and dissolved in a constant temperature shock incubator at 60°C for 24 h. Te EB standard solution was prepared by the gradient titration method with a suitable concentration and added to the NO. 1-6 centrifuge tubes, which were similarly dissolved as the samples to be tested. Te supernatants of the EB standard and untested samples were extracted after centrifugation (2000 rpm × 20 min) and added to a 96-well plate at a volume of 200 μL per well. Te optical density (OD) value was measured at 632 nm using an EnSpire Multimode Plate Reader (PerkinElmer, Waltham, MA, USA). A regression curve was drawn based on the OD value of the standard substance, the concentration of EB in the sample was calculated by the curve equation, and the volume of EB infltration in the rat brain tissue was obtained. Te value of EB infltration (μg·g −1 ) � Concentration of EB in the sample (μg·mL −1 ) × volume of formamide (mL)/wet weight of brain (g).

Reverse Transcription-Polymerase Chain Reaction (RT-PCR).
Total RNA was extracted from the brain hematoma tissues, the purity and concentration were determined, and was then reverse transcribed with PrimeScript RT Master Mix according to the manufacturer's protocol. A real-time PCR was operated following the two-step quantitative RT-PCR method with SYBR Mixture. Te amplifcation of genes was analyzed through a 7500 real-time PCR system (ABI, Waltham, MA, USA). Primer sequences are as follows: All primers were designed and synthesized by Shenggong Bio (Shanghai, China).

Western Blot (WB) Analysis.
Cold RIPA Lysis Bufer (P0013B, Beyotime) with protease inhibitors (P1005, Beyotime) was used to extract the total protein from the tissue around the brain hematoma (Radius: 1 mm) and the distal colon (Below cecum: 7-10 cm). Te protein concentration was quantifed by a BCA kit (P0010, Beyotime), normalized, mixed with Laemmli Sample Bufer (161-0747, Bio-Rad), and heated in a 95°C water bath for 5 minutes. Ten micrograms of protein samples from each group were electrophoresed through a 10% SDS-PAGE gel, transferred onto a polyvinylidene fuoride (PVDF) membrane (Merck Millipore Ltd., Darmstadt, Germany), and blocked with TBST (containing 5% nonfat dry milk) for 2 h at room temperature. After sequential incubation with appropriate primary antibodies and secondary antibodies conjugated to HRP, the positive western blots were detected with an ImageQuant LAS4000 mini biomolecular imager system (GE Healthcare Life Sciences, Fairfeld, CT, USA) by chemiluminescence using an ECL kit (P0018S, Beyotime). Te internal reference for expression levels was β-actin.
2.11. Flow Cytometry (FCM) Analysis. Peripheral blood mononuclear cells (PBMCs) from rats were prepared with lymphocyte separation medium by density gradient centrifugation. Te cells in the PBMC suspension were counted and the concentration was adjusted to 1 × 10 6 /mL. A mixture of ionomycin [fnal concentration (FC): 1 μg/mL], PMA (FC: 100 ng/mL), and brefeldin A (FC: 10 μg/mL) was added to the appropriate volume of cell suspension sample and then placed in a 5% CO 2 incubator at 37°C for 6 h to stimulate

Qualitative Analysis of Components in the Aqueous Extract of LTP.
For the quality evaluation of diferent batches of drugs, the components in the aqueous extract of LTP samples were analyzed by a UPLC-QE-MS method. A total ion chromatogram of LTP is shown in supplementary fgure s1. Te ion fragment peaks of unknown metabolites that were hard to defne in each group of samples were discarded through searching and matching with the laboratory's self-built database, integrated public database, AI database, and metDNA. 157 compounds were included in subsequent analyses. Comparing the superimposed fngerprints of three batches of LTP samples and a quality control (QC) sample, there is no obvious diference. Tus, it was confrmed that the quality of the drugs used in our study is reliable and representative, and the metabolites of LTP could have a higher detection rate in the negative ion mode (Figure 1(a)). As shown in the secondary mass spectrometric heatmap (Figure 1(b)) of total metabolites, the retention time of MS2 fragments in LTP components was mainly concentrated in the range of 0-10 minutes. Te analysis of ion characteristics from the heatmap was consistent with the base peak chromatogram (Figure 1(c)), and 12 active ingredients with relatively high abundance were identifed. Te detailed information on these components was presented in supplementary table s2. Te names of the main compounds are as follows: (1)

LTP Alleviated Brain Damage in AICH Rats.
Longa's scores of rats in both the blank and sham groups were 0. Longa's scores of the AICH rats were not signifcantly diferent between the model, LTP, and probiotics groups in the baseline period before treatment. After 3 d of treatment, the scores were signifcantly lower in the LTP group than in the model group (p < 0.01), and there was no signifcant diference between the probiotics and model groups. Longa's scores were signifcantly lower in the LTP and probiotics groups after treatment than before treatment (p < 0.01, p < 0.05), and there was no signifcant diference before and after treatment in the model group (Figure 2(a)). Te values of brain wet weight and dry weight of rats in each group are shown in Figure 2(b). Te brain water contents in the model, LTP, and probiotics groups were signifcantly higher than those in the blank and sham groups (p < 0.01), and those of the model group increased the most. Compared with the model group, the LTP group was significantly lower (p < 0.01), and the probiotics group was higher than those in the LTP group (p < 0.05) (Figure 2(c)). Compared with the blank and sham groups, the volumes of EB infltration in brain tissue were signifcantly increased in the model and probiotics groups (p < 0.01), and signifcantly decreased in the LTP group when compared with the model group (p < 0.01), with no signifcant diference between the probiotics and model groups. Te volumes of the probiotics group were higher than those of the LTP group (p < 0.05) (Figures 2(d) and 2(e)).   Evidence-Based Complementary and Alternative Medicine between the probiotics and LTP groups in TNF-α relative expression. Te level of IL-1β mRNA expression in the probiotics group was higher than that in the LTP group (p < 0.05) (Figures 3(a) and 3(b)).

LTP Repaired the BBB Permeability in AICH Rats.
Compared with the blank and sham groups, OCLN and CLDN-5 protein expression in the brain tissue was significantly downregulated in the model, LTP, and probiotics groups (p < 0.01), and ZO-1 protein expression was only downregulated in the model and probiotics groups (p < 0.01).
Te aforementioned three TJP levels were signifcantly upregulated in the LTP group compared with the model group (p < 0.01), and there was no signifcant diference between the probiotics and model groups, with the levels in the probiotics group being lower than those in the LTP group (p < 0.01) (Figures 4(a)-4(d)).

LTP Repaired Epithelial Permeability in the Distal Colon of AICH Rats.
Compared with the blank and sham groups, the ZO-1, OCLN, and CLDN-1 protein expression levels in the colon tissue were signifcantly downregulated in the model  Figure 2: Efect of LTP on brain damage in the AICH rats: (a) Longa's score before and after treatment, (b) brain weight, (c) brain water content, (d) volume of EB infltration in brain tissue, and (e) appearance of EB staining in rat brains. * * p < 0.01and # p < 0.05 vs. before treatment. a p < 0.01 vs. the blank and sham groups. b p < 0.01 vs. the model group. c p < 0.05 vs. the LTP group (n � 6). group (p < 0.05), and those of the LTP and probiotics groups were signifcantly upregulated when compared with the model group (p < 0.01), CLDN-1 levels were signifcantly lower and ZO-1 expression was signifcantly higher in the probiotics group compared to the LTP group (p < 0.05), and there was no signifcant diference in OCLN expression between the two groups ( Figures 5(a)-5(d)).  Evidence-Based Complementary and Alternative Medicine between the probiotics and model groups, and the expression in the probiotics group was lower than that in the LTP group (p < 0.05) (Figures 6(a) and 6(d)). Compared with the blank and sham groups, SIgA expression was signifcantly downregulated in the model group (p < 0.05), and that of the LTP group was signifcantly upregulated when compared with the model group (p < 0.05). Tere was no signifcant diference between the probiotics and model groups (Figures 6(b) and 6(d)). Compared with the blank and sham groups, GPR43 expression was signifcantly downregulated in the model and probiotics groups (p < 0.05), and that in the LTP group was signifcantly upregulated when compared with the model group (p < 0.05). Tere was no signifcant diference between the probiotics and model groups (Figures 6(c) and 6(d)).

LTP Maintained the Balance of T17/Treg Cell in the
Peripheral Blood of AICH Rats. Compared with the blank and sham groups, the ratios of the T17 cells in peripheral blood of the model, LTP, and probiotics groups were signifcantly higher (p < 0.01) and signifcantly lower in the LTP and probiotics groups when compared with the model group (p < 0.01), and those of the probiotics group were higher than those in the LTP group (p < 0.05) (Figures 7(a) and 7(c) A). Compared with the blank and sham groups, the Treg cell ratios were signifcantly lower in the model and probiotics groups (p < 0.01) and signifcantly higher in the LTP group when compared with the model group (p < 0.01), and those of the probiotics group were lower than those in the LTP group (p < 0.05) (Figures 7(b) and 7(c) B). Compared with the blank and sham groups, the values of T17/Treg in the model, LTP, and probiotics groups were signifcantly higher (p < 0.01), and signifcantly lower in the LTP and probiotics groups.
When compared with the model group (p < 0.01), those of the probiotics group were higher than in the LTP group (p < 0.05) (Figure 7(c) C).

LTP Regulated the Discharge of the VN in the Duodenum of AICH Rats.
Compared with the blank and sham groups, the VN discharge rates in the model group were signifcantly faster and the amplitudes were signifcantly lower (p < 0.01).
When compared with the model group, the discharge rates were signifcantly slower and the amplitudes were signifcantly higher in the LTP and probiotic groups (p < 0.01), with no signifcant diference between the LTP and probiotic groups (Figures 8(a)-8(c)).

Discussion
Increasing research on neurological disorders has focused attention on the brain-gut axis. AICH is often associated with intestinal system complications such as disturbance of intestinal fora, impaired fecal excretion, and stress ulcers. Hang et al. found that [30] severe histopathological changes occurred in the intestinal mucosa at 3 h after the onset of traumatic brain injury (TBI), with further apoptosis of IECs and disruption of intercellular tight junction structures, intestinal mucosal permeability also increased. In addition, Yu et al. used [31] healthy fecal microbiome transplantation (FMT) to treat ICH mice, which can restore the homeostasis of the intestinal environment, protect IMB function, and promote the transfer of immune lymphocytes from the intestine to the brain, thus reducing neuroinfammatory damage and improving the prognosis of ICH. However, the specifc pathophysiological link between intestinal microecology and ICH is yet to be revealed. Furthermore, the intestinal mucosa, an important carrier of intestinal microecology, has been shown to form a natural protective barrier through IECs, TJPs, the MUC layer, immune cells, and the enteric nervous system (ENS), reducing the invasion of pathogenic bacteria and infammatory factors [32,33]. With the lack of efective drugs against AICH, the use of herbal medicine to improve IMB function may be a potential therapeutic option that deserves in-depth investigation.
According to the favor and efcacy of the herbs used, the compatibility strategy of LTP was aimed at the two pathological factors of blood stasis and fre-heat highlighted in the acute stage of ICH. Blood stasis refers to the stagnation of blood from ruptured cerebral vessels to form a spaceoccupying efect in the brain, and the resulting local hematoma pressure and thrombin toxicity could cause brain edema. Fire-heat manifested as an infammatory response in AICH. Our previous study demonstrated that LTPs reduce the activation of toll-like receptor 4 (TLR4) and inhibit infammatory responses mediated by the nuclear factor kappa-B (NFκB) signaling pathway in AICH rats [34]. Kwon also demonstrated in a rat model of refux esophagitis that rhubarb components could achieve anti-infammatory effects by inhibiting the NFκB pathway [35]. Xu et al. found the same antioxidant efect of rhubarb in a rat model of TBI [36]. Our current study data showed that LTPs reduced the degree of neurological defcits, attenuated brain edema, repaired BBB permeability, and attenuated the neuroinfammation in AICH rats, moreover, the repairing efect of LTPs on the BBB yielded consistent results in both EB staining (Figures 2(d) and 2(e)) and WB analysis (Figures 4(a)-4(d)). Gallic acid and epicatechin are abundant in the LTP compositions. Studies have confrmed that [37,38] gallic acid signifcantly improves neurological damage in TBI rats, and Cheng's study also showed that epicatechin has a restorative efect on neurological defcits in TBI mice [39]. Te antioxidant efect of LTPs was consistent with the expected treatment method of "cooling the blood and dissipating the blood stasis." Combined with another therapeutic principle "purging fu-organs to eliminate heat" of LTP, the fu-organ mentioned here is the intestine, which is responsible for the excretion of feces. Tis efect of the monarch drug rhubarb was demonstrated in a recent study, and rhubarb extract could increase the secretion of MUCs by cupular cells and regulate intestinal fora in the colon to promote stool evacuation [40]. To explore the potential mechanisms of this core therapeutic strategy, we focused our subsequent studies on the intestinal system. Te intestine is the largest immune organ in the body, and the colon has a high abundance of fora and exuberant secretory activity; in particular, high levels of MUCs are maintained in the distal colon, which has a protective efect on the intestinal mucosa [41,42]. Te IMB consists of the IEC junction complex and its secretions, intestine-associated immune cells, and the intrinsic intestinal fora. Te MUCs secreted by cup cells in the intestinal epithelium adhere to the outermost layer of the intestinal mucosa, which is rich in AMPs and SIgA, preventing pathogenic bacteria from reaching the intestinal epithelium [43]. Moreover, IgAsecreting plasma cells in the intestine can be transferred to the brain and exert a protective efect on the central nervous system [44]. TJPs in the intestinal epithelium act as the basis for the intestinal mechanical barrier and prevent endotoxin and infammatory factors from entering the bloodstream for infection, while the BBB also relies on TJPs to maintain permeability and avoid harmful substances from invading the brain. Te intestinal fora and its metabolites can induce Foxp3 expression via GPR43, increase Treg cell diferentiation, and suppress excessive infammatory responses mediated by T17 cells, as has been demonstrated by numerous authors [45][46][47]. GPR43 can also exert immunomodulatory efects by modulating Group 3 innate lymphoid cells (ILC3s) [48]. Te results of our experiment confrmed that LTP signifcantly increased the secretion of MUC2, SIgA, and GPR43 in the colon of the AICH rats, while the probiotics had no signifcant reparative efect. In addition, by analyzing the expression of ZO-1, OCLN, and CLDN-1, the intestinal mucosal TJPs in each group of rats, we found that both LTP and probiotics had signifcant reparative efects on intestinal permeability in AICH rats. At the beginning of the manuscript, we introduced that our recent report also proved that both LTP and probiotics can signifcantly reduce the content of LPS in the serum of AICH rats [23]. Stimulated by LPS, monocytes and macrophages can rapidly release proinfammatory factors, such as TNF-α and IL-1 β. In the current study, we also found that LTP and probiotics decreased the mRNA level of TNF-α and IL-1β in the brain tissue of the AICH rats, and the inhibitory efect of LTP on IL-1 β is better. Accordingly, the protective efect of LTP on IMB might indirectly prevent intestinal proinfammatory toxins from invading the brain in AICH rats. Although this view of explaining the brain-gut axis theory may only be a possibility, it is consistent with the latest report from Pellegrini et al. [49]. We also performed FCM analysis of lymphocytes in the peripheral blood of each group of rats, showing that both LTPs and probiotics had a regulatory efect on T17/Treg cell imbalance, but LTPs were superior to probiotics. Wang et al. found that [50] disorders of intestinal fora can cause histological damage to the intestinal mucosa with disruption of tight junctions and increased bacterial translocation in a mouse model. Chen et al. demonstrated that [51] ginsenoside Rb1 could regulate Lactobacillus helveticus abundance and c-aminobutyric acid (GABA) receptor expression to exert neuroprotective efects in a cerebral ischemia-reperfusion injury rat model. Tese fndings suggest that intestinal fora have a modulatory efect on both intestinal mucosal permeability and T-cell immunity, which was confrmed in the probiotic group of rats in our study. However, the LTP component could also afect more classes of intestinal strains, produce more abundant MUCs, SIgA, and GPR43, better modulate the immune function of the intestinal mucosa, and suppress the excess infammatory response, thus exerting a better therapeutic efect than the probiotics. VN, which passes through the brain and intestine, undergoes signal communication between gut microbes and the brain [52]. Te analysis of VN discharge signals in the AICH rat model was valuable for exploring the neuroimmune regulatory pathways of the brain-gut axis. We collected the waveforms of VN discharge signals from the duodenal segment of rats in each group and found that the excitability of the VN in the AICH model rats was suppressed, as evidenced by a signifcant decrease in amplitude and a compensatory increase in discharge rate. LTP and probiotics could reverse this discharge suppression state of VN to diferent degrees. Comprehensive analysis showed that the LTPs and probiotics had comparable efects in repairing the epithelial structure and VN conduction function in the intestinal mucosa of the AICH rats. Zhou et al.'s study confrmed that [53] VN stimulation attenuates the disruption of intestinal mucosal tight junctions in endotoxemic mice via a cholinergic anti-infammatory pathway. Accordingly, the activity of VN fbers attached to the intestinal wall may be interdependent with intestinal epithelial integrity; however, the exact association needs to be elucidated by additional neurophysiological studies.

Conclusions
Ultimately, we demonstrated that LTPs ameliorate neurological defcits, brain edema, BBB permeability, and neuroinfammation, thereby, slowing the progression of brain hemorrhage in the acute phase. In addition, LTPs could repair intestinal mucosal epithelial structures, promote the secretion of MUC2, SIgA, and GPR43, and regulates the activity of VN in the duodenal segment and the T17/Treg ratio in peripheral blood, thus enhancing IMB function. Meanwhile, we also found that probiotics had a signifcant restorative efect on intestinal mucosal epithelial permeability and VN activity. Te view that a stable intestinal mucosal function could promote the repair of brain damage is relatively clear through analyzing the connection between the anti-infammatory efects of LTP and probiotics on the brain-gut axis of the AICH rats. To sum up, in order to explore the main therapeutic approaches that LTP is superior to probiotics in AICH rats, we will focus on the signaling pathway neuroimmune mediated between the gut and the brain in the future, thereby, providing novel ideas for the clinical treatment of AICH. However, the pharmacological efects of LTP components require more in-depth in vitro experimental studies.

Data Availability
Te data used to support the fndings of this study are available from the corresponding author upon reasonable request.

Ethical Approval
All animal experiments were approved by the Experimental Animal Ethics Committee of Nanjing University of Chinese Medicine (application no. 202105A029).