Baicalin Regulates Autophagy to Intervene with Intestinal Acute Graft-Versus-Host Disease via AMPK/Mtor Pathway and Abnormal Intestinal Flora

Acute graft versus host disease (aGVHD) is the main complication of allogeneic hematopoietic stem cell transplantation and the main cause of death. To verify if baicalin can protect the intestinal mucosal barrier by regulating abnormal autophagy and interfering with acute intestinal graft-versus-host disease, a mouse model of aGVHD was established. CB6F1 mice were transplanted with a mononuclear cell suspension from Balb/c donor mice bone marrow and spleen via vein injection after treatment with 60co X rays. After treatment with different doses of baicalin for 15 days, survival time, survival time,the serum level of TNF-α and IL-10, the autophagy markers from intestinal and bacterial ora were detected.A intestinal barrier dysfunction cell model was also used to verify the effect of baicalin. The results showed that baicalin could signicantly prolong survival time, reduced aGVHD score compared to the model control group, signicantly reduce TNF-α level, and increase IL-10 level. TEM examination showed that baicalin treatment increased the number of autophagic vacuoles and led to recovery of mitochondrial structure in the intestinal mucosal epithelial cells of mice and in Caco-2 cells. Western blot results showed that baicalin treatment enhanced autophagy in vivo by regulating the AMPK/mTOR autophagy pathway. Similar results were observed in vitro in Caco-2 cells. Furthermore, the effect of baicalin was reduced after combined treated with autophagy inhibitors 3-MA.The results of intestinal microbial sequencing also conrmed that there were signicant changes in the type and quantity of bacterial ora before and after the application of baicalin. Baicalin can reduce the severity of intestinal acute graft-versus-host disease by regulating autophagy with inuencing the unbalanced inammatory cytolines and bacterial ora, which may be a new treatment for aGVHD. after transplantation The human colon carcinoma cell line Caco-2 was grown in Dulbecco's modied Eagle's minimum essential medium (DMEM, pH 7.4) (Invitrogen) supplemented with 25 mM glucose, 10% inactivated fetal bovine serum (FBS) (Lonza), 1% penicillin streptomycin (PS) and 1% non-essential amino acid solution (Invitrogen). After the cells were stimulated by TNF-α (100 ng/ml) for 21 h at 37°C, transepithelial electrical resistance (TER) was tested to veried if the intestinal barrier dysfunction cell model was established. For the measurement of TER, 1×10 5 cells were seeded on polycarbonate transwell inserts with a diameter of 12 mm and a pore size of 0.4 µm. TERs were recorded using the Millicell system, and resistances obtained over a blank lter were subtracted from the obtained values. For assessment of Lucifer Yellow, or dextran, permeable transwell inserts were washed with HBSS followed by apical application of the tracer compounds. After 2 h of gentle agitation at 37°C, basal tracer concentration was measured using a uorescence plate reader, and then used to determine relative sodium-to-chloride permeability according to the Goldmann-Hodgkin-Katz equation. After cell model was established, cells were treated with 10, 20,40µg/ml baicalin, respectively. To conrm the autophagy effect 20µg/ml baicalin was chosen to be combined with 3-MA.


Introduction
Acute graft versus host disease(aGVHD) is one of the most fatal complications in the early period after hematopoietic stem cell transplantation. Autophagy is the process of a cell phagocytosing its own cytoplasmic proteins or organelles, and encapsulating them into vesicles which fuse with lysosomes forming autophagic lysosomes. The autophagic lysosomes then degrade their contents which contributes to the metabolic needs of the cell and the renewal of certain organelles. Our work prior to this study suggests that autophagy protects the intestinal mucosa by degrading damaged organelles and other substances, and this slows aGVHD progression. Intervention with the mTOR pathway may protect against acute graft-versus-host disease after bone marrow transplantation [1] . A study found that the intraperitoneal administration of metformin, which activates AMPK signaling, ameliorated the clinical severity of aGHVD and lethality [2] .
Multiple studies have shown that Huangqin decoction can regulate the Th17 / Treg balance and reduce intestinal mucosal barrier damage [3] . This suggests that Huangqin decoction could help treat colitis via immunomodulation. Huangqin decoction contains a variety of ingredients, of which baicalin seems most promising. The baicalin molecular formula is C 21 H 18 O 11 , and it is derived from the dried roots of the dicotyledon plant Scutellaria baicalensis. Baicalin has been demonstrated to signi cantly improve cerebral ischemic injury via enhanced autophagy by up-regulating the Beclin-1 protein [4] . Additionally, baicalin induces autophagy in PC12 cells by reducing p62 expression and increasing LC3-II expression which results in increased PC12 cell viability [5] . Furthermore, baicalin can suppress neural apoptosis to protect the brain by ameliorating the mitochondrial apoptotic pathway [6] . Despite its potential as a treatment, there have been few studies on baicalin's effects on intestinal aGVHD. Therefore, we performed experiments to explore the mechanism of baicalin regulated autophagy in the context of acute intestinal graft-versus-host disease to explore its potential as a novel clinical treatment of aGVHD.

Animals and experimental protocol
Mice were purchased from Beijing Weitong Lihua Experimental Animal Technology Co., Ltd. and the animal production license number was SCXK (Beijing) 2012-0001. 10 healthy SPF-grade Balb/c H-2d male mice were used as donors (weighing 18-22 g and aged 8-10 weeks); 56 SPF-grade CB6F1 female mice were used as transplant recipients(weighing 18-22 g and aged 8-10 weeks).
Inbred Balb/c H-2d mice were sacri ced by cervical dislocation, and the bone marrow and spleen mononuclear cells were obtained aseptically. Next, CB6F1 mice were irradiated with 60co X rays, and then the transplant was completed by infusing the mononuclear cell suspension (1×10 7 bone marrow cells + 1×10 7 spleen cells), obtained from donor Balb/c mice (above), via tail vein [7] .
Mice were randomly divided into the model control, low dose, medium dose, and high dose groups. Model control group mice were fed with normal saline and treated group mice were fed with 15, 30 or 60 mg/(kg·d) baicalin for 15 days immediately after modeling.
In order to examine the in uecing autophagy effect of baicalin, we also compared treated with 30mg baicalin alone group, to a group treated with 30mg baicalin combined with the autophagy inhibitor 3-MA [8] .All animal procedures were performed with the approval of the Animal Ethics Committee of the A liated Hospital of Shandong University of Traditional Chinese Medicine. All experiments were performed in accordance with relevant guidelines and regulations.
Cell lines culture, tested by transepithelial electrical resistance The human colon carcinoma cell line Caco-2 was grown in Dulbecco's modi ed Eagle's minimum essential medium (DMEM, pH 7.4) (Invitrogen) supplemented with 25 mM glucose, 10% inactivated fetal bovine serum (FBS) (Lonza), 1% penicillin streptomycin (PS) and 1% non-essential amino acid solution (Invitrogen). After the cells were stimulated by TNF-α (100 ng/ml) for 21 h at 37°C, transepithelial electrical resistance (TER) was tested to veri ed if the intestinal barrier dysfunction cell model was established. For the measurement of TER, 1×10 5 cells were seeded on polycarbonate transwell inserts with a diameter of 12 mm and a pore size of 0.4 µm. TERs were recorded using the Millicell system, and resistances obtained over a blank lter were subtracted from the obtained values. For assessment of Lucifer Yellow, or dextran, permeable transwell inserts were washed with HBSS followed by apical application of the tracer compounds. After 2 h of gentle agitation at 37°C, basal tracer concentration was measured using a uorescence plate reader, and then used to determine relative sodium-to-chloride permeability according to the Goldmann-Hodgkin-Katz equation.
After cell model was established, cells were treated with 10, 20,40µg/ml baicalin, respectively. To con rm the autophagy effect 20µg/ml baicalin was chosen to be combined with 3-MA.

Clinical aGVHD classi cation criteria
Clinical score for aGVHD was determined by observing weight loss, posture, activity, hair, skin integrity, and diarrhea on the 15th day after transplantation ( Table 1). The criteria for determining aGVHD was taken from work by Cooke et al [9] . Individual mice from coded cages received a score of 0 to 2 for each criteria (maximum score of 12), as described above.
Small intestinal mucosa histologic and pathological in ammation score analysis Small intestinal tissue was collected from mice. The tissue was xed in 4% paraformaldehyde for hematoxylin and eosin (H&E) staining following standard protocols. HE images were observed under a light microscope. Histological damage of the tissues was calculated using a pathological in ammation score( Table 2) which was determined by examining damage of the small intestine tissue, crypt destruction, and the degree of mucosal ulcer and the extent of cell in ltration (The scores for each criteria were added to get the pathological score of small intestine [10] ). Motic Images Advanced 3.0 software was used to measure intestinal villi height and crypt depth. Detection of TNF-α and IL-10 level The expression of tumor necrosis factor (TNF)-α and interleukin (IL)-10 was measured using quantitative reverse transcription polymerase chain reaction (RT-PCR). Small intestine tissue was collected and quickly frozen in liquid nitrogen for subsequent mRNA extraction. Total RNA was extracted using the guanidine isothiocyanate-phenol-chloroform method. Extracted RNA was reverse-transcribed using the Sparkjade reverse transcription kit and the resulting complementary DNA was analyzed for the expression of the target molecules using the LightCycler 480 real-time PCR system. The primer sequences are provided in Table 3. The resulting gene expression levels of the target molecules were normalized based on GAPDH expression.  Western Blot Analysis Cells were lysed using RIPA lysis buffer and total protein was collected. Next 20µg of each protein sample was loaded on to 10% SDS-PAGE gels, and the protein was concentrated at 80V for 20 min and separated at 120V for ~ 1h. After the electrophoresis was stopped, the samples were transferred to PVDF at 110V at 4℃ based on the wet blotting method protocol. Each immunoblot was blocked in 5% nonfat milk TBST for 2h,then incubated with primary antibodies like LC3, P62, Beclin-1, AMPK, mTOR or β-tubulin at 4℃ overnight. The membrane was washed 3 times, then incubated with diluted secondary antibodies for 2h. After the secondary antibody solution was fully washed away, the PTG ECL chemiluminescence detection kit was used to develop the blot. Then the blot was transferred to an imaging machine for exposure and analysis.

Immunohistochemistry (IHC) analysis
The tissues of small intestine (2-3 cm) were xed with 10% formaldehyde, decolorized, cleared, para n embedded and sectioned. Antigen retrieval was performed for 20 min in a pressure cooker. The sections were cooled, rinsed with PBS, incubated in a 3% H 2 O 2 solution for 10 min and rinsed with PBS again. The tissue sections were incubated in goat serum for 20 min to remove the serum, incubated with primary antibodies against AMPK and LC3 at 37°C for 1 h, rinsed with PBS, incubated with biotin-labeled secondary antibodies at room temperature for 30 min and again rinsed with PBS. Finally, the sections were incubated in DAB substrate solution for 5 min. After been washed thoroughly in tap water, counterstained in hematoxylin, dehydrated in absolute alcohol, cleared in xylene, and subjected to microscopy. The expression of proteins was quanti ed using an image analysis and measuring system (Image-Pro Plus 6.0). The mean area and mean integrated optical density (mean IOD) of the expression of these proteins were calculated.
Immuno uorescence staining of LC3 in vivo and in vitro Tissue para n sections of the small intestine or Caco-2 cells were incubated with primary antibody at 4 ℃ for 24 h. Then, goat anti rabbit red uorescent second antibody Cy3 was applied for 60 min at 37℃.

DNA extraction and sequencing of mouse feces
Mice Feces were harvested with 2 ml perfectly clean eppendorf tube under sterile conditions 2 h before sacri ce and stored immediately in liquid nitrogen until processing for DNA isolation. The Genomic DNA extracted detection by 1% agarose gel electrophoresis after DNA isolation, PCR ampli cation were performed by ABI GeneAmp® 9700 model PCR instrumen. First, the puri ed DNA templates were ampli ed using the following primers of 16S rDNA V3-V4 regions: 5'-ACTCCTACGGGAGGCAGCAG-3' and 5' -GGACTACHVGG GTWTCTAAT − 3'. AxyPrep DNA gel Recovery Kit (AXYGEN company) was used to recover PCR products by gel cutting.The PCR reaction were used TransStart Fastpfu DNA Polymerase, The reaction system consisted of PCR Mix 20 µL Reaction program: 95℃ predenaturation for3 min, 95℃ denaturation for 30s, annealing-temperature of 55℃ denaturation for 30s, 72℃ extension for 45s, a total of 27 cycles, 72℃ nal extension for 10 min. 10℃ until halted by user. According to the preliminary results of electrophoresis, the PCR products were analyzed by quanti uor ™ -St blue uorescence quantitative system (Promega company) to detection and quanti cation, sodium hydroxide denaturation were used to produce single stranded DNA fragments. Finally, miseq pe300 sequencer (Illumina company) were used for sequencin.

Statistical analysis
The experimental results were analyzed using SPSS 19.0 and GraphPad Prism 8.0. The results were expressed as mean ± standard deviation and analyzed by ANOVA test. P < 0.05 indicated that the difference was statistically signi cant.

Effect of baicalin treatment on the recovery of intestinal aGVHD in mice
At rst, the intestinal aGVHD model was veri ed by small intestinal mucosa histological analysis( Fig. 1. A) and aGVHD score( Fig. 1. D). The IL-10 and TNF-α expression level( Fig. 1.B) was also assessed to con rmed the intestinal aGVHD model established. Heavy in ltration of in ammatory cells was detected in the intestinal mucosa of the model group, which is consistent with the pathological characteristics of intestinal aGVHD (Fig. 1.A). After being treated with medium or high dose baicalin which structure showed in Fig. 1.C, the small intestinal mucosal epithelium structure recovered in multiple ways including decreased in ammatory cell in ltration, more regular arrangement of glandular cells, an increased number of goblet cells, less shedding of cellular debris, and less necrosis of mucosal epithelium. The pathological in ammation score of medium and high dose groups were signi cantly lower compared with to the model and low dose groups ( Fig. 1.A). After using baicalin, the TNF-α expression was signi cantly reduced and IL-10 expression was signi cantly increased in the medium and high dose groups compared with the model group, (P < 0.01, Fig. 1.B).
The model group mice all died by d15 and the low dose group all died by d27, whereas, 50% of the medium dose and high dose group mice were alive beyond day 28 ( Fig. 1.E). These data demonstrate that medium and high dose baicalin treatment effectively reduce morbidity and mortality by improving intestinal mucosal injury in aGVHD model mice.

Baicalin effects on intestinal aGVHD via in uencing autophagy in mice
In order to nd the mechanism of baicalin treatment, the mTOR signal mediated autophagy were tested. Medium dose of baicalin was chosen to be tested. Speci cally, the AMPK autophagy pathway was activated, as indicated by increased levels of LC3 / , Beclin1, and p-ampk with decreased level of P62 and mTOR (P < 0.01), (Fig. 2.A). After autophagy was inhibited with 3-MA, the baicalin induced increase of autophagy related proteins was reversed to the same level as the model group. Additionally, there were more biphasic membrane autophagosomes with normal mitochondrial structure in baicalin group, compared to control and the 3-MA treatment group (Fig. 2.B). There was no signi cant difference between the baicalin + 3-MA group and the model group (Fig. 2.A and B). This indicates that autophagy mediates baicalin's effect on intestinal aGVHD.
As Fig. 3.A showed, the inhibiting autophagy with 3-MA reversed the bene cial effect that baicalin treatment had on aGVHD morbidity. The baicalin + 3-MA group had a far more damaged small intestinal mucosal epithelium structure, a higher level of in ammatory cell in ltration, an abnormal arrangement glandular cells, less goblet cells, and more mucosal epithelium shedding and necrosis when compared to the group given baicalin alone (P < 0.01) (Fig. 3.A). The level of LC3 was assessed by IHC( Fig. 3.C) and immuno uorescence staining (Fig. 3.D) in small intestine, results showed that the LC3 level was improved in tissues of baicalin group compared with the baicalin + 3-MA group and model group. Based on the proteins level of WB ( Fig. 2.A) and IHC (Fig. 3.B), AMPK/mTOR induced autophagy should be the effective target of baicalin on intestinal aGVHD.

Baicalin can improve intestinal mucosal barrier by enhancing autophagy in vitro
After Caco-2 cells induced by TNF-α, the effect of baicalin was tested by using transepithelial electrical resistance (TER), western blot and immuno uorescence staining. We found a signi cant increase of level of LC3 / ratio (Fig. 4.A) and the number of LC3 positive dots (Fig. 4.D) in baicalin group with higher autophagic vesicle numbers( Fig. 4.C). The results were similar to what was found in vivo, which indicated that baicalin can promote autophagy by enhancing AMPK pathway activity and down-regulating mTOR activity in Caco2 cells. It was also found that the mitochondrial structure is rebuilt in the baicalin treated group (Fig. 4.C). If the autophagy inhibitor was added in baicalin treated group, the therapeutic effect of baicalin was decreased with lower autophagy level. Cause the intestinal barrier dysfunction can be alleviated by baicalin in TER test (Fig. 4.B), the mechanism of baicalin is by stimulating on autophagy was veri ed again in vitro experiments.

The diversity of intestinal ora in mice
The sequencing coverage of intestinal micro ora in mice were about 99.614%, compared with the normal control group, 392 bacteria were same in baicalin group and model group, 176 bacteria were different between baicalin group and model group (Fig. 5.A). There were 18 bacteria in the structure of intestinal micro ora at the phylum level, including 10 dominant phylum: Actinobacteria, Bacteria_unclassi ed, Bacteroidetes, Cyanobacteria, Deferribacteres, Firmicutes, Marinimicrobia_SAR406_clade, Patescibacteria, Proteobacteria, Tenericutes (Fig. 5B). The coverage of intestinal microbial sequencing in aGVHD model group were about 99.663%, which were higher than that in normal control group, and the expression of Bacteria_unclassi ed, Firmicutes, Proteobacteria were higher than that in normal control group.
Actinobacteria, Bacteroidetes, Chloro exi, Cyanobacteria, Deferribacteres, Marinimicrobia_SAR406_clade, Patescibacteria, Tenericutes, Verrucomicrobia were lower than that in normal control group. Compared with the model group, the expression of Bacteroidetes, Proteobacteria, verrucomicrobia were increased and the expression of Firmicutes were decreased after baicalin treatment. The microbial community structure of the mouse intestinal ora is mainly composed of 227 genera at the genus level, of which 67 were dominant bacterial genera. Compared with the model group, the expressions of Akkermansia, Bacteroides, Clostridium_sensu_stricto_1, Eisenbergiella, Escherichia-Shigella, Lachnospiraceae_ NK4A136_ group, Prevotellaceae_ NK3B31_ group were increased after baicalin treatment, and the expressions of Alistipes, Lachnospiraceae_ FCS020_group, Lactobacillus, Muribaculaceae_ norank, Prevotellaceae_ unclassi ed were decreased. And the relative abundance of bacteroidetes rmicutes, proteobacteria, verrucomicrobia, eferribacterres and the others was displayed in Fig. 5.C and D. The distribution abundance of intestinal ora changed signi cantly between baicalin group and model group (Fig. 5E).
The rarefaction curve tends to be at, indicating that the selection of sequencing data is reasonable( Fig. 6. A, B). The range of the curve on the horizontal axis is large, the species abundance is high, and the curve on the vertical axis is gentle, indicating that the species are evenly distributed( Fig. 6.C). We use principal component analysis to extract the two coordinate axes that re ect the differences between samples to the greatest extent, so as to re ect the differences of multidimensional data on the two-dimensional coordinate diagram, and then calculate that the distance between baicalin group and control group is closer than that between model group and control group. This means that the composition of intestinal ora in baicalin group is more similar to that in normal group (Fig. 6.D).

Discussion
Intestinal aGVHD is one of the most common graft-versus-host reactions after transplantation. The intestinal symptoms of patients with acute graft-versus-host disease after transplantation are refractory diarrhea, serious colic, hematochezia and intestinal obstruction that will affect the implantation and endanger life. These symptoms are the same as those recorded in use of huangqin decoction, and that is why we tried to nd an element in huangqin decoction to treat this disease. There are some studies that have shown that the occurrence of intestinal aGVHD is closely related to gastrointestinal ora [12] , in ammatory cell in ltration and other factors. One key cause recognized today is the pretreatment regimen which leads to cytokine storm-mediated in ammation and subsequent low immunity [13] . In our study, the levels of TNF-α and IL-10 changed signi cantly after baicalin was used. Mucositis and infection [14] can trigger the aGVHD via destruction of the intestinal mucosal barrier. So addressing how to protect the intestinal mucosal barrier is key in treating aGVHD.
Autophagy-dependent Treg cells are essential for immune regulation and reconstruction after transplantation [15] . Autophagy-dependent control of Treg aGVHD activity was further supported by a study suggesting that increased autophagy increased the ability of thymus-derived Tregs to interfere with CD8 + T cell mediated aGVHD [16,17] . Our experiment showed an increase in autophagy level changed after the application of baicalin suggesting that baicalin could help stimulate Treg control of aGVHD. Interestingly, if the C3aR/C5aR complex is inhibited, inducing mitochondrial autophagy in dendritic cells will likely reduce the symptoms of graft-versus-host disease on the basis of ensuring graft anti-leukemia effects [18] . Our study showed that baicalin induced autophagy reduced the clinical aGVHD score of mice was decreased mirroring these results. In the initial treatment of standard risk aGVHD, mTOR inhibitor rapamycin had similar complete/partial remission rates as prednisone on day 28. Rapamycin can change the expression of autophagy related proteins, such as inhibit mTOR and promote autophagy [19][20] , this is similar to the effect of baicalin.
Baicalin is the main ingredient of Huangqin decoction which is often used to treat diarrhea symptoms. Baicalin itself can induce autophagy through PI3K/Akt/mTOR pathway which results in an antiin ammatory effect [21] .
In our experiments, abnormal levels of IL-10 and TNF-α, in ammatory factors often dysregulated in aGVHD, were corrected by treatment with baicalin. This effect was most notable in the medium dose group. Additionally, after treatment with baicalin the viability of mice improved and the clinical aGVHD and pathological scores of small intestine decreased. This showed that baicalin had some ability to reduce the pathology of intestinal aGVHD. Next we postulated that mechanism by which baicalin regulated aGVHD was by regulating the unbalanced autophagy found in intestinal aGVHD, and we veri ed this by examining the AMPK/mTOR pathway after baicalin treatment. Baicalin can increase the expression of autophagy related proteins such as LC3-II/I and Beclin 1 so as to increase the production of autophagy vesicles and LC3 level to promote the autophagy of cellular debris and other substances, improve the function of mitochondria, and reduce the apoptosis and abscission of intestinal epithelial cells. There are some studies that suggest that activated autophagy may attenuate intestinal mucosal barrier dysfunction by preventing and reducing oxidative stress [22][23] . Our experimental results show that low levels of autophagy make it di cult to improve the permeability of Caco-2 cells, and with the increased level of TNF in the small intestinal mucosa of acute graft-versus-host bearing mice, the barrier of intestinal mucosa will be destroyed.
To con rm this mechanism, 3-MA, an autophagy inhibitor, was used. The data showed that, after inhibition of autophagy with 3-MA, the effects of baicalin were reversed in vivo. Some studies have found that 3-MA can inhibit Caco-2 cell activity by inhibiting autophagy [24] . For this reason we tested Caco-2 cell treated by 3-MA with or without baicalin. Medium and high dose of baicalin were found to improve intestinal barrier dysfunction by increasing the relative TER level of Caco-2 cells. Then 20µg/ml baicalin was combined with 3-MA for the further study. We found that the effect of baicalin was weakened after the application of 3-MA, providing further evidence that baicalin reduces aGVHD pathology by regulating autophagy.
Many studies have con rmed that intestinal ora is closely related to intestinal diseases [25,26] .For example, the lachnospiraceae decreased with the Firmicutes increased [27] in ulcerative colitis model [28] , Bacteroidetes [29] , verrucomicrobia [30] and lachnospiraceae [31] would increase after treatment. These above bacteria have a certain inhibitory effect on in ammation, and play an important role in colitis and other intestinal diseases, which is consistent with our research results.Our data indicated that baicalin may reduce lachnospiraceae by reducing the diversity of ora, reducing the relative abundance of Firmicutes, increasing the relative abundance of Bacteroidetes, Proteobacteria and verrucomicrobia_ FCS020. The relative abundance of lachnospiraceae was increased. The relative abundance of group bacteria can resist acute graft-versus-host disease.
This study provides strong evidence that baicalin can protect the intestinal mucosal barrier by promoting autophagy, reducing the death of intestinal mucosal epithelial cells, improve the abnormal intestinal ora and interfering with the progression of intestinal aGVHD. Our next experiments will focus on the effect of baicalin on Treg and Th17 cells in our aGVHD model, so as to further reveal baicalin's mechanism of protecting intestinal mucosal barrier by immunomodulation.

Conclusion
Based on the results above, the following conclusions can be drawn: baicalin can interfere with intestinal aGVHD by regulating autophagy-related protein level via the AMPK/mTOR pathway, baicalin increases autophagic vesicles and repairs mitochondrial morphology, and that baicalin intervention is weakened by inhibiting autophagy using 3-MA. Furthermore, baicalin can protect the intestinal mucosal barrier and interfere with the progression of acute graft-versus-host disease.

Declarations
Ethics approval and consent to participate All experiments were conducted in compliance with the ARRIVE guidelines. We con rm that animal care and experimental procedures were carried out in accordance with the guidelines of the Animal Ethics Committee of the A liated Hospital of Shandong University of Traditional Chinese Medicine. The reference number is AWE-2019-037.

Consent for publication
Not applicable.

Availability of data and material
The datasets supporting the conclusions of this article are included within the article. The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.