A Novel Traditional Chinese Medicine, Ramulus Mori (Sangzhi) Alkaloids Ameliorate Glucose Metabolism Accompanied by the Modulation of Gut Microbiota and Ileal Inammatory Damage in Type 2 Diabetic KKAy Mice

Background: The novel Traditional Chinese Medicine Ramulus Mori (Sangzhi) alkaloid tablets (SZ-A) are approved by The China National Medical Products Administration for the treatment of type 2 diabetes mellitus (T2DM). However, the extensive pharmacological characteristics and the underlying mechanism are unknown. This study investigated the mechanisms by which SZ-A ameliorates glucose metabolism in KKAy mice, an animal model of T2DM. Methods: Diabetic KKAy mice were treated intragastrically with SZ-A once daily for 8 weeks, after which glucose levels, lipid metabolism, gut microbiome, systemic inammatory factors, luminal concentrations of short-chain fatty acids (fecal samples), and ileal proteomic changes were evaluated. The ileum tissues were collected, and the effects of SZ-A on pathological inammatory damage were evaluated by hematoxylin and eosin staining, immunouorescence, and immunohistochemistry. The mRNA and protein expression levels of various inammatory markers, including monocyte chemoattractant protein-1 and phosphorylated nuclear factor kappa B p65, were detected in the ileum tissues. Results: SZ-A improved glucose metabolism with enhanced insulin response and elevated glucagon-like peptide 1 (GLP-1) during the glucose tolerance test in diabetic KKAy mice. Gut microbiota analysis demonstrated that SZ-A administration elevated the abundance of Bacteroidaceae and Verrucomicrobia, reduced the levels of Rikenellaceae and Desulfovibrionaceae; and increased the concentrations of fecal acetic and propionic acids compared to the diabetic model group. Additionally, SZ-A markedly improved ileal inammatory injury and pro-inammatory macrophage inltration and improved intestinal mucosal barrier function in diabetic KKAy mice. SZ-A also attenuated the levels of circulating endotoxin, proinammatory cytokines, and chemokines in the mice sera. Conclusions: SZ-A ameliorated the overall metabolic prole including glucose and lipid metabolism in KKAy mice, which may


Background
The gut microbiome plays important roles in the regulation of glucose and energy homeostasis. It also plays a critical role in obesity, glycemic control, and type 2 diabetes mellitus (T2DM) [1], which is a chronic and multifactorial disease in which diverse physiopathologic mechanisms lead to a persistent state of hyperglycemia. T2DM is fundamentally the result of beta cell and alpha cell dysfunction, and insulin resistance in different tissues of the body [2]. T2DM may also be due to the activation of proin ammatory mechanisms that involve several factors. Gut microbiota-mediated low-grade in ammation is also involved in the onset and progression of T2DM. Studies in mice and humans have shown that there is dysregulation in the gut microenvironment accompanied by immunological and metabolic dysfunctions in individuals who have T2DM [3].
Metabolites derived from the gut microbiota, such as short-chain fatty acids (SCFAs) and lipopolysaccharides (LPS), may act as potent immune modulators [4]. During eubiosis, the production of SCFAs is essential for maintaining the integrity of the intestinal barrier as well as for immunogenic tolerance. In addition, the effects of SCFAs are not limited to immunomodulatory functions [5], as they can also stimulate the secretion of intestinal peptides that participate in the regulation of appetite and insulin secretion such as glucagon-like peptide 1 (GLP-1) [6]. Conversely, in the presence of gut dysbiosis during the progression of T2DM, diet-driven unfavorable microbiota composition can lead to the increased production of pro-in ammatory LPS, which are associated with alterations in gut permeability [7]. Subsequently, these in ammatory states might exacerbate the disruption of the mucus layer barrier and increase the epithelial permeability of the small intestine, resulting in elevated LPS levels in the bloodstream, metabolic endotoxemia [8], increased levels of systemic in ammatory mediators, adiposity, obesity, insulin resistance, and hyperglycemia.
Thus, restoration of gut dysbiosis could potentially treat metabolic disorder. Modulation of the intestinal microbiota by interventions has led to a major impact on both the immunological and metabolic functions of the host. In recent years, there has been increasing interest in investigating the use of prebiotics (non-digestible carbohydrates), probiotics (life bacteria), and anti-diabetic drugs for the modulation of gut dysbiosis [9,10]. Traditional Chinese Medicine (TCM) has been used to manage T2DM. A large number of studies have shown that the effects of TCM may be, at least in part, via modulation of gut microbiota [11][12][13].
The novel TCM Ramulus Mori (Sangzhi) alkaloid (SZ-A) tablets, also known as Sangzhi Zong Shengwujian, is approved by The China National Medical Products Administration (NMPA, formerly known as the China Food and Drug Administration) for the treatment of patients with T2DM (Approve Number Z20200002). The main components of SZ-A powder (materials for SZ-A tablets) include alkaloids, avonoids, polysaccharides, coumarin, quercetin, resveratrol, amino acids, and organic acids.
SZ-A is a group of effective polyhydroxy alkaloids (50% or more by weight) that potently inhibit αglucosidase, including 1-deoxyno-jirimycin (1-DNJ), fagomine (FAG), 1,4-dideoxy-1, 4-iminod-D-arabinitol (DAB), and other soluble polyhydroxy alkaloids or glycosides with a similar structure [14]. In preclinical pharmacological studies, chronic treatment of SZ-A was shown to lower fasting and postprandial blood glucose levels in alloxan-induced diabetic mice and rats. SZ-A has also been shown to reduce the peak of postprandial blood glucose in sucrose/starch loading tests in both healthy and diabetic mice after a single dose through the inhibition of intestinal disaccharidases. Moreover, available evidence from other studies suggests that SZ-A improves dyslipidemia and glucose-stimulated insulin secretion in high-fat diet-induced obese C57 mice after long-term intragastrical administration [15]. These data suggest that the bene cial role of SZ-A may involve multiple mechanisms in addition to α-glucosidase inhibition.
Previous studies have systematically investigated the tissue distribution of the three major active alkaloids, and they found that 1-DNJ, FAG, and DAB are mainly found in the gastrointestinal tract, liver, and kidney, respectively [16]. Therefore, we hypothesized that the anti-diabetic effects of SZ-A may be via regulation of gut microbiota and intestinal metabolites.
In this study, we evaluated the anti-diabetic effects and underlying mechanisms of SZ-A, especially on modulation of gut microbiota, intestinal metabolites, and gut barrier integrity in type 2 diabetic KKAy mice.

Methods
Materials and reagents SZ-A powder (lot number: 201707008, The total polyhydroxy alkaloid content in SZ-A powder is about 63% by weight, which was mainly composed of 39% of DNJ, 10.5% of FAG, and 7% of DAB), was kindly provided by the Department of Research & Development of Beijing Wehand-bio Pharmaceutical Co Ltd.

Animal experimental design
Animal experiments were performed following the "3R" principles and guidelines for laboratory animals (GB14925-2001 and MOST 2006a) established by the People's Republic of China. 14-week-old male KKAy mice (30 g) were purchased from Beijing Huafukang Bioscience Co., Ltd. (Beijing, China). Animals were maintained at 22 ± 2 °C with a 12 h light-dark cycle with free access to food and water. The 14-week-old male KKAy mice were fed with high-fat diets (45% of energy from fat; D12451; Research Diets, USA). And after 4 weeks of high-fat diets feeding, KKAy mice were randomly divided into three groups (n = 8) according to the levels of blood glucose, triglyceride, total cholesterol, body weight, and percentage of blood glucose increase at 30 min after oral glucose loading: diabetic model group (DM), SZ-A-low dose-treated group (SZ-A 100, 100 mg/kg), SZ-A-high dose-treated group (SZ-A 200, 200 mg/kg). All mice were treated intragastrically with SZ-A solution or an equivalent volume of water once daily for 8 weeks. After 56 days of treatments, all of the mice were fasted overnight and weighted, then were sacri ced via cervical dislocation. Subsequently, the ileum tissues were isolated, xed in paraformaldehyde solution.
Blood glucose, lipid, and glycated hemoglobin measurements After 4 weeks of treatment, fasting blood glucose (FBG) and postprandial blood glucose (PBG) levels were measured using the glucose oxidase method (Biosino Bio-Technology and Science Inc., Beijing, China). After 42 days of treatment, blood triglycerides, total cholesterol levels, and glycated hemoglobin Gut microbiota pro ling and fecal short-chain fatty acids analysis Mice were sacri ced and the luminal contents were collected from the ileum (as the fecal samples) and snap-frozen in liquid nitrogen after 40 days and the end of treatment, followed by storage at −80°C. The gut microbiome in feces was assayed and the abundance and diversity of gut microbiota were analyzed using Illumina MiSeq sequencing (Major Bio-Pharm Technology, Shanghai, China) according to the standard protocol as previously described [17]. The sequence data were processed and analyzed on the free online Majorbio I-Sanger Cloud Platform (www.i-sanger.com).
SCFAs in these fecal samples were detected based on our previous report [18]. Brie y, the SCFAs in each sample were assayed by gas chromatography coupled to a mass spectrometer detector (GC-MS) (Agilent Technologies Inc. CA, USA) and quanti ed using Masshunter quantitative software. Correlation analysis of SCFAs and gut microbiota was performed on the platform of Majorbio I-Sanger Cloud (www.isanger.com). R and p values were obtained using Spearman's rank correlation.
Histopathological evaluation, Immuno uorescence, and immunohistochemistry assay of the ileum About 4 cm of ileum was xed in 4% paraformaldehyde to prepare 5-μm para n slides. The ileum sections were stained with hematoxylin and eosin (H&E) for the analysis of in ammatory changes (n=8). Histopathological assessment of in ammatory and crypt damages was assessed as previously stated by a light microscope [19]. Eight mice in each group and six randomly selected elds were analyzed. The ileum sections were also stained with the rst antibodies against F4/80 (ARG22476) and CD11c (n=5) (ARG59698; Arigo Biolaboratories Corp, Taiwan). For immunohistochemistry analysis, we used Anti-NF-κB p65 (phospho S536) (ab86299, Abcam, Cambridge, United Kingdom) (n=5). Images were captured with a Mirax scanner (3DHISTECH, Hungary), and the area of positive points was calculated with Image Pro (MediaCybernetics, Rockville, MD).

Cytokines and chemokines assay in serum
Blood was collected and serum was prepared as previously stated. The concentration of Endotoxin was determined by ELISA kit, and the concentrations of IL-1β, IL-5, IL-10, IL-13, IL-1a, IL-6, IL-12b, Ccl11, Cxcl1, Ccl4, and Ccl5 in serum were determined by Luminex liquid suspension chip detection, which was performed by Wayen Biotechnologies (Shanghai, China). The mouse 23-plex Multi-Analyte kit (Bio-Plex suspension Array System; Bio-rad, Hercules, CA, USA) was used following the manufacturer's instructions.
The exact protocol was administered according to what had been reported before [20].

Tandem Mass Tagging (TMT) proteomics analysis
The primary experimental procedures for TMT proteomics analysis include protein preparation, trypsin digestion, TMT labeling, HPLC fractionation, LC-MS/MS analysis, and data analysis. The detailed procedure is presented in the Supplementary Methods. The TMT proteomics analysis in our research is supported by Jingjie PTM BioLabs.

Statistical Analysis
The data are presented as the mean ± SEM. Statistical analysis was performed using GraphPad Prism 7.0. Differences in FBG, PBG, insulin, GLP-1, lipid levels, and body weight were assessed using a two-way analysis of variance (ANOVA) with Tukey's test. Data sets involved in two groups or multiple groups were analyzed using unpaired two-tailed Student's t-test or one-way ANOVA depending on the experiments. Differences with P< 0.05 were considered statistically signi cant.

Results
SZ-A ameliorates glucose metabolism, enhances the insulin response, and elevates GLP-1 in glucose levels during oral glucose tolerance tests in diabetic KKAy mice After a 4-week treatment, the levels of fasting blood glucose (P < 0.01, P < 0.001) and postprandial blood glucose (P < 0.01, P < 0.01) in both SZ-A-treated groups were signi cantly decreased compared to the DM group ( Fig. 1A, B). Hemoglobin A1c (HbA1c) levels in SZ-A-treated groups were lower than those in the DM group after a 6-week treatment (P < 0.05, P < 0.05; Fig. 1C), indicating that SZ-A glycemic control in the KKAy mice during chronic treatment. As shown in Fig. 1D, compared to the DM group, both doses of SZ-A signi cantly reduced blood glucose levels at 15 min after oral glucose loading (P < 0.05, P < 0.01).
We further detected the blood insulin content and active GLP-1 levels as an indication of insulin and GLP-1 secretory function, respectively. As shown in Fig. 1E and 1F, there was no notable increase in blood insulin content and active GLP-1 level in the DM group after oral glucose loading; however, SZ-A-treated groups had increased insulin content and active GLP-1 levels at both baseline and 15 min after glucose stimulation. SZ-A 100 and SZ-A 200 signi cantly enhanced insulin secretion nearly 1.73-fold and 1.88fold from baseline at 15 min after glucose stimulation, respectively (P < 0.01, P < 0.05), compared to the DM group (1.11-fold). In addition, both doses of SZ-A elevated active GLP-1 levels nearly 2.7-fold and 2.6fold at 15 min after glucose stimulation from baseline (P < 0.01, P < 0.05), respectively, compared to the DM group (2.1-fold). Moreover, both doses of SZ-A resulted in decreased blood triglyceride levels after 6 weeks (P < 0.05, P < 0.05), and induced signi cant weight loss compared to the DM group at the end of treatment (P < 0.01, P < 0.001).

SZ-A modulates gut microbiota pro ling and SCFA concentration in feces
The effects of high-dose SZ-A (SZ-A 200, 200 mg/kg) on intestinal microbiota composition were examined by Illumina sequencing-based analysis of bacterial 16S ribosomal RNA in fecal samples collected at the end of 8-week treatment. Compared to the DM group, the operational taxonomic unit (OTU) numbers were reduced in the SZ-A 200 group ( Fig. 2A; P < 0.001). The Shannon and Chao indices re ect the diversity and richness of gut microbiota, respectively. As shown in Fig. 2B and 2C, SZ-A diminished the indices of Shannon and Chao (P < 0.05, P < 0.05). Unweighted Unifrac principal coordinate analysis (PCoA) based on OTU levels revealed distinct clustering of microbiota composition in each group (Fig. 2D). Multivariate analysis of variance of PCoA matrix scores revealed that the microbiota community of mice in the SZ-A 200 group differed from that of the DM group (P < 0.001). Additionally, the bacterial community of SZ-A 200-treated mice differed from that of the DM group. Taxonomic pro ling at the family level revealed that SZ-A treatments elevated the abundance of Bacteroidaceae, Erysipelotrichaceae, and Verrucomicrobia and reduced that of Rikenellaceae, Desulfovibrionaceae, and Aerococcaceae compared with DM mice (Fig. 2E). Similar results were also observed at the genus level. SZ-A 200 decreased the abundance of Alistipes, Desulfovibrio, and Aerococcus, and increased the abundance of Bacteroides, Faecalibaculum, and Allobaculum compared with the DM group (Fig. 2F). Collectively, these ndings indicate that SZ-A 200 modulates the composition of gut microbiota.
Subsequently, the relationship between fecal SCFAs and intestinal bacterial at the family level was analyzed (Fig. 3J). The results showed that Enterococcaceae and Corynebacteriaceae, Aerococcaceae, Desulfovibrionaceae, and Rikenellaceae were positively correlated with the decreased SCFAs, including butyric, isobutyric, hexanoic, isohexanoic, pentanoic, and isopentanoic acids (Fig. 3J). Enterococcaceae and Corynebacteriaceae abundance was negatively correlated with propionic acid, Corynebacteriaceae was negatively correlated with acetic acid, and Verrucomicrobiaceae was positively correlated and propionic acid was observed (Fig. 3J). Several transport systems play a role in the cellular uptake of SCFAs in the gut, including monocarboxylate transporter-1 (MCT1) and sodium-coupled monocarboxylate transporter 1 (SMCT1) (SLC5A8). As the transporters responsible for the entry and transcellular transfer of these bacterial products in epithelium are critical determinants of gut function, we detected MCT1 and SLC5A8 protein expression levels in the ileum tissue after SZ-A treatment in KKAy mice. The results showed that MCT1 and SLC5A8 protein levels in ileum from SZ-A-treated mice were signi cantly increased compared with the DM group.
SZ-A alleviates ileal in ammatory injury and pro-in ammatory macrophage in ltration in KKAy mice Considering microbial SCFAS production (especially acetate, propionate, and butyrate) is essential for gut integrity by regulating the mucus production, providing fuel for epithelial cells and effects on mucosal immune function, the histological alteration of ileum tissue were evaluated by hematoxylin and eosin (H&E) staining. As shown in Fig. 4A, in the DM group, a dense in ammatory cellular in ltrate was present in the mucosa and submucosa and crypts showed typical shortening. Focal crypts were lost and the surface epithelium was damaged (Fig. 4A). Microscopic total score and scores for the three features (in ammation, extent of in ammation, and crypt damage) were given for each group (Fig. 4B-E).
In ammation scores were signi cantly reduced by both doses of SZ-A (P < 0.05, P < 0.001). Moreover, microscopic total score, crypt damage score, and in ammation score were signi cantly reduced in the SZ-A 200 group (P < 0.001, P < 0.01, and P < 0.001). Collectively, long-term SZ-A treatment prevented the development of in ammation and restored ileal barrier integrity in diabetic KKAy mice.
The KKAy mice were fed a high-fat diet (HFD) to induce diabetic syndrome. Given the critical role of M1 macrophages in HFD-induced intestinal in ammation, in parallel to those histological changes, macrophage-speci c F4/80 and CD11c expression was measured to verify whether SZ-A treatment was able to modulate macrophage in ltration in the ileum tissue. As shown in Fig. 4F, compared to the DM group, fewer pro-in ammatory CD11c-positive macrophages were observed in both doses of SZ-A-treated groups (P < 0.05, P < 0.001). This indicates a reduced in ammatory state after SZ-A treatment, and fully consistent with this result, we found downregulated mRNA expression of F4/80 (P < 0.05, P < 0.05) and multiple pro-in ammatory factors, including MCP1 (P < 0.05, P < 0.01) and TNF-α (P < 0.05, P < 0.05; Fig. 4G), and also reduced CD11c protein expression in the ileum of SZ-A-treated groups (P < 0.05, P < 0.05; Fig. 4H), compared to the DM group. Given that in ammation damages gut permeability and integrity, we also detected the protein expression levels of zonula occludens-1 (ZO-1), an intestinal tight junction component. SZ-A 200 markedly elevated ZO-1 protein levels (P < 0.05) compared to the DM group.
Nuclear factor kappa B (NF-κB) is critically associated with the progression of in ammation and cell proliferation in the intestinal mucosa. Therefore, the effects of SZ-A on NF-κB activity on the ileal mucosa were investigated. The indices of the phosphorylated (p-NF-κB) p65-positive area were markedly reduced with SZ-A treatment (P < 0.01, P < 0.01; Fig. 4I). These ndings indicate that SZ-A signi cantly alleviated ileal in ammatory injury and attenuated the in ammatory state induced by pro-in ammatory macrophage in ltration of the ileum tissue in diabetic KKAy mice.

SZ-A attenuates endotoxin content, pro-in ammatory cytokine, and chemokine levels in serum of diabetic KKAy mice
The literature suggests that gut dysbiosis not only leads to increased intestinal permeability, but it also results in the translocation of bacteria or bacterial products into circulation, inducing a state of chronic low-grade in ammation, such as LPS in HFD-induced diabetes. Considering the effects of SZ-A on modulating gut microbiota pro ling and alleviating ileal in ammatory injury, serum endotoxin content, and levels of cytokines and chemokines were determined after SZ-A treatment of diabetic KKAy mice.

Functional enrichment analysis of differentially abundant proteins in the ileum after SZ-A treatment
Proteomics were used to determine the molecular characteristics of the ileum in the high-dose SZ-Atreated group (SZA) and DM group in KKAy mice. Liquid chromatography tandem mass spectrometry identi ed 208,219 secondary spectra. A total of 42,677 matched effective spectra were obtained. Using a false discovery rate (FDR) < 1% at the peptide and protein levels, 25,816 of the 25,060 peptides were identi ed as speci c, and 5043 of the 4352 proteins were quanti able (Fig. S1A). A total of 34 proteins were differentially expressed (fold change > 1.2, P < 0.05; Fig. 6A) between the DM group and SZA group, of which 24 proteins were upregulated and 10 were downregulated. To determine the characteristics of the differentially expressed proteins, we annotated the subcellular localization, Clusters of Orthologous Group, and Gene Ontology (GO) of the 34 proteins. Annotation of the subcellular localization showed that 44.12% of all identi ed differentially expressed proteins were localized in the cytoplasm, 14.71% in the plasma membrane, 14.71% in the mitochondria, 11.76% in the nucleus, 8.82% in the extracellular space, and 5.88% in the endoplasmic reticulum (Fig. S1B). Most differentially abundant proteins participated in and regulated the cellular and metabolic processes (Fig. S1C). COG functional classi cation revealed that most of these differentially abundant proteins played a role in posttranslational modi cation, protein turnover, and chaperones (Fig. S1D).
Bioinformatics analysis was performed to identify the main biological pathways and functional categories of the differentially abundant proteins (fold change > 1.2; P < 0.05). Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis showed that the most signi cantly altered pathways were involved in primary bile acid biosynthesis, peroxisome proliferator-activated receptor (PPAR) signaling pathway, synthesis and degradation of ketone bodies, and terpenoid backbone biosynthesis (Fig. 6B). We identi ed 34 abundant proteins that were mainly involved in the above-mentioned pathways (Fig. 6C). These proteins included downregulation of cluster of differentiation 36 (CD36), a protein related to the PPAR pathway; CYP27a1, the representative differentially abundant protein related to primary bile acid biosynthesis; and histocompatibility 2 class II antigen E beta, which is critical in the antigen processing and presentation pathway and is also described as major histocompatibility complex class II (MHC II).
The expression level of these key regulators identi ed via proteomics was also con rmed by Western blotting. The results showed the level of CD36 (SZ-A 200 group; P < 0.05) and MHC II (P < 0.05, P < 0.01) were signi cantly reduced in SZ-A-treated groups compared to the DM group (Fig. 6D), which is consistent with proteomics analysis.

Discussion
In this study, we evaluated the therapeutic e cacy of SZ-A in glucose and lipid metabolism in vivo in a model of HFD T2DM, KKAy mice. We found that SZ-A ameliorated glucose metabolism and enhanced the insulin response to oral glucose tolerance tests and elevated active GLP-1 levels in diabetic KKAy mice. We suspect that the increased glucose-stimulated insulin secretion might be primarily linked to the improved β-cell function after SZ-A treatment, or related to the elevated blood levels of active GLP-1 in KKAy mice. However, the fact that oral (current study) but not intraperitoneal injection of glucose was associated with improved glucose tolerance in KKAy mice after SZ-A treatment indicated the potential involvement of incretin hormones.
GLP-1, an incretin hormone released in response to the ingestion of nutrients, acts as a hypoglycemic hormone to improve postprandial glucose homeostasis by enhancing meal-induced insulin secretion. GLP-1 activity is mediated by the GLP receptor 1R (GLP-1R). GLP-1Rs are highly expressed in pancreatic β-cells and other tissues including neurons in speci c central brain regions, the kidney, and the gut tract [21]. Findings from the clinical trials have also revealed that the administration of GLP-1R agonists (GLP-1RAs) induces weight loss in addition to glucose improvement [22]. In our study, we also found that both doses of SZ-A induced signi cant weight loss at the end of long-term treatment, whereas the glycemiclowering effects of GLP-1RAs were mainly attributed to endocrine actions at the level of pancreatic islets. Thus, we evaluated changes in α-and β-cell mass or distribution to analyze the functional state of islets after SZ-A treatment. In Figure S2, immuno uorescence staining showed that the glucagon-positive area was signi cantly scattered around the islets in the DM group, suggesting the imbalanced distribution of α-cells. SZ-A decreased the ratio of glucagon-positive area and improved the distribution of α-cells.
GLP-1 is predominantly secreted by neuroendocrine L-cells in the gut. L cells, which produce both GLP-1 and peptide YY, are predominantly expressed in the distal ileum. In humans, prebiotic treatment increases microbiome diversity, which in turn may modulate levels of GLP-1 [6]. Moreover, with ileal transposition surgery, L cells located in the transposed ileum are rapidly stimulated by ingested nutrients to produce GLP-1 [23]. Given the close anatomical proximity, gut microbiota could potentially alter the nutrientsensing capacity of enteroendocrine cells and subsequent gut peptide release. Additionally, the major components of SZ-A (DNJ, FAG, and DAB) were mainly distributed in the gastrointestinal tract. Therefore, we hypothesized that SZ-A caused the altered active GLP-1 levels, possibly by regulating gut microbiota composition. No study has investigated the effect of SZ-A on the distal ileum microbiota, a site that is crucial for intestinal sensing and absorption of nutrients. Therefore, in this study, we evaluated the interaction between SZ-A and ileal microbiota. Then the effects of SZ-A on the gut microbiota were investigated.
We rst characterized the microbiota composition from DM and the high dose-SZ-A treated group (SZ-A 200). Mice were sacri ced and the luminal contents were collected from the ileum (nearly 12 cm distal to the pyloric sphincter). SZ-A increased the abundance of Bacteroidaceae, Erysipelotrichaceae, and Verrucomicrobia. SZ-A also markedly decreased the abundance of Rikenellaceae, Desulfovibrionaceae, and Aerococcaceae at the family level compared with DM mice (Fig. 2E). Similar results were observed at the genus level. SZ-A reduced the abundance of Alistipes, Desulfovibrio, and Aerococcus, and increased the abundance of Bacteroides, Faecalibaculum, and Allobaculum. Bacteroidetes and Verrucomicrobia play crucial roles in producing SCFAs [1]. We also detected measurable fecal SCFA concentrations in the DM and SZ-A-treated groups, which is consistent with previous studies characterizing the ileal microbiota composition of diabetic KKAy mice. We observed elevated propionate and acetate in the SZ-A-treated groups (Fig. 3A, B). Propionate and acetate are two of the main SCFAs that are produced by bacteria as a result of resistant starch fermentation. SZ-A-induced microbiota changes facilitated propionate and acetate production in the ileum, which might explain the elevated basal and glucose-stimulated blood active GLP-1 levels in diabetic KKAy mice after SZ-A treatment.
The microbiota-derived metabolites in the luminal contents also represent a potential link to expression changes of nutrient sensors, ligand-receptor, or transporter. Figure 3 shows that with the exception of propionate and acetate, other SCFA concentrations in ileal luminal contents were decreased after SZ-A treatment. Several transport proteins are involved in the uptake of SCFAs in the gut, including MCT1 and SMCT1 (SLC5A8). These transporters are critical determinants of the entry and transcellular transfer of SCFAs into intestinal epithelium under physiological conditions and in disease states [9]. We detected the protein expression levels of these two transporters. There was signi cant upregulation in the expression of MCT1 and SMCT1 (SLC5A8) in the ileum of SZ-A-treated groups (Fig. 3K), which is consistent with the decreased level of SCFAs following SZ-A treatment. Collectively, we determined a connection between SZ-A action and changes in ileal microbiota to regulate SCFA production, which in turn, affect glucose homeostasis through the regulation of GLP-1 secretion.
LPS is a major component of the cell wall of gram-negative bacteria and is considered an endotoxin when present in the blood. Increased LPS levels can induce a large number of proin ammatory responses and in ammatory cytokine release [7]. Of note, among the changes in the gut microbiota of KKAy mice after SZ-A treatment, the decreased abundance of LPS-containing Desulfovibrionaceae was observed. Desulfovbrionaceae is an endotoxin producer and has been linked to gut permeability [24]. There is increasing evidence of the role of gut microbiota in various in ammatory diseases, especially those affecting gastrointestinal tract in ammation. Additionally, butyrate, propionate and acetate are SCFAs that are produced by bacteria as a result of resistant starch fermentation, and have anti-in ammatory properties [4]. Due to these effects of SZ-A on the composition of the gut microbiota, it is speculated that SZ-A treatment might repair the in ammatory damage of intestine by modulating the abundance of LPS and SCFA-producing gut microbiota.
Subsequently, H&E staining was performed of the ileum of KKAy mice after SZ-A treatment. Three different sections were studied for each animal. Histological in ammation was scored by two blinded investigators using a modi ed scoring system [19]. Considering the degree of in ammation, the transmural vertical extent of in ammation, and the crypt damage score, related to the percentage of involvement of mucosal surface in each slide. As shown in Fig. 4A to 4E, this nding supported the occurrence of damage of the ileal mucosal barrier in diabetic KKAy mice (DM group) and repaired ileal in ammatory damage after treatment with both doses of SZ-A. LPS production is also responsible for inducing monocyte-and macrophage-mediated in ammation in the intestine [25]. Gut macrophages, which reside in the connective tissue underlying the gut epithelium, the lamina propria, are considered key players for the maintenance of intestinal homeostasis and in ammation [26]. We observed that in ileal tissues, the percentage of macrophages (CD11c + /F4/80 + ) and the cell mass of CD11c-positive monocytes were signi cantly decreased with both doses of SZ-A compared with the control (Fig. 4F). The degree of expression of F4/80, MCP-1, and TNFα mRNA in both SZ-A-treated groups was signi cantly decreased compared to the DM group. Similarly, the protein levels of CD11c and MCP1 in the ileum were also reduced by SZ-A treatment (Fig. 4G, H). The NF-κB signaling pathway was also detected in the ileum.
The phosphorylation of p65 and its nuclear translocation were signi cantly decreased by SZ-A treatment. These results suggest that SZ-A treatment ameliorate ileal in ammatory damage by reducing monocyte recruitment, by decreasing in ammatory signals.
Over time, major in ammatory signals (e.g., NF-κB-dependent) become activated in diabetic KKAy mice, thereby stimulating pro-in ammatory cytokine secretion in the small intestine. This in ammatory state might subsequently exacerbate the disruption of the mucus layer barrier and increase epithelial permeability of the small intestine. We observed increased protein levels of the tight junction protein ZO-1 in the SZ-A-treated group, which can re ect the intestinal mucosal barrier function. Our results also con rmed the results that the intestinal mucosal barrier function was improved after treatment of KKAy mice with SZ-A. The persistent in ammatory state not only increases intestinal permeability but also the destruction of tight junction proteins attached to epithelial cells, increasing portal vein and systemic plasma LPS concentrations, and eventually promoting the development of systemic in ammation [24]. Therefore, we subsequently determined the concentrations of endotoxin and in ammatory cytokines in the serum of KKAy mice. The serum endotoxin levels signi cantly decreased after both doses of SZ-A treatment, compared to the DM group. The serum levels of in ammatory factors (e.g., IL-1β, IL-6, CCL4, and CCL5) indicated that low-grade in ammation in diabetic KKAy mice decreased after SZ-A treatment, which corresponded to the endotoxin level. Thus, ileal in ammatory damage in the diabetic KKAy mice was positively associated with low-grade in ammation, which was effectively alleviated by SZ-A treatment.
Given that the production of pro-in ammatory cytokines is related to the activation of LPS-induced macrophage activation and polarization, which were signi cantly suppressed in the ileum by SZ-A treatment in diabetic KKAy mice. The molecular characteristics of the ileal tissues in the high-dose SZA and DM groups in diabetic KKAy mice were investigated through proteomics. Proteomics analysis revealed that monocyte differentiation into intestinal macrophages involves phenotypic changes with MHCII expression was downregulated by SZAH treatment, which may be responsible for the improvement of ileal in ammatory damage after SZ-A treatment. Beyond that, the representative differentially downregulated abundant proteins related to PPAR pathway and fat digestion and absorption, ileal CD36 expression level was signi cantly decreased in SZ-A-treated groups (Fig. S3), which might be related to improved blood triglyceride levels.

Conclusions
Here, we reported that long-term treatment of SZ-A (8 weeks) is su cient to alter the microbiota composition in the ileum of diabetic KKAy mice. Our data, together with enriched literature, provide novel mechanistic insights into the role of SZ-A in mediating gut microbial community in ileal in ammatory damage and glucose metabolism.

Consent for publication
Not applicable.

Declaration of Competing Interest
The authors declare no con ict of interest.