Effect and Correlation of Rosa roxburghii Tratt Juice Fermented by Lactobacillus paracasei SR10-1 on Oxidative Stress and Gut Microflora Dysbiosis in Streptozotocin (STZ)-Induced Type 2 Diabetes Mellitus Mice

Rosa roxburghii Tratt (RRT) is a kind of excellent fruit, with many healthy functions. RRT fruit dietary interventions have demonstrated a remarkable potential to prevent type 2 diabetes mellitus (T2DM). In the present study, the effects of Lactobacillus paracasei SR10-1 fermented RRT juice (FRRT) on the oxidative stress, short-chain fatty acids (SCFAs), and gut microbiota in T2DM mice induced by high-sugar and high-fat diets and streptozotocin (STZ) were investigated using GC–MS and 16S rRNA gene sequencing. The results showed that medium-dose FRRT intervention resulted in significantly decreased levels of TG, TC, LDL-C, BUN, creatinine, and MDA (p < 0.05) and significantly increased levels of HDL-C, GSH-PX, CAT, and SOD of T2DM mice (p < 0.05). The levels of acetic acid, propionic acid, butyric acid, and isovaleric acid were significantly increased, by 142.28%, 428.59%, 1968.66%, and 81.04% (p < 0.05), respectively. The relative abundance of Firmicutes, Lachnospiraceae, Verrucomicrobiaceae, Akkermansia, and Allobaculum was significantly increased (p < 0.05), and the relative abundance of Proteobacteria, Enterobacteriaceae, Veillonellaceae, Phascolarctobacterium, and Klebsiella was significantly decreased (p < 0.05). Correlation analysis showed that Phascolarctobacterium was significantly negatively correlated with weight (p < 0.05), SOD (p < 0.01), CAT (p < 0.05), and T-AOC (p < 0.05). Akkermansia was significantly negatively correlated with weight (p < 0.05). Conclusively, medium-dose FRRT potentially improved T2DM by reversing dyslipidemia, decreasing oxidative stress, increasing SCFAs, and regulating gut microbiota composition. The medium-dose FRRT may serve as a novel T2DM dietary strategy to prevent T2DM.


Introduction
Type 2 diabetes mellitus (T2DM), a metabolic disease characterized by hyperglycemia and insulin resistance that accounts for more than 90% of all cases of diabetes, is a growing global health problem [1][2][3].The onset of T2DM is associated with insufficient insulin secretion or insulin resistance.Modern pharmacological studies have shown that gut microbiota play an essential role in the onset and development of T2DM, and are increasingly recognized as a critical factor in human health [4][5][6][7].Many probiotic-fermented fruit juices have been used to regulate gut microbiota with remarkable effects [8,9].Gao et al. found that fermented Momordica charantia L. juice had better effects on hyperglycemia, hyperinsulinemia, hyperlipidemia, and oxidative stress than unfermented Momordica charantia L. juice in diabetic rats [10].Probiotic-fermented fruit juices can prevent obesity and modulate the gut microbiota, and the prevention of T2DM by probiotic-fermented fruit juices will bring substantial benefits to patients.Foods with antidiabetic properties have attracted increasing attention [1,11,12].suspension concentration of Lactobacillus paracasei SR10-1 was 9 log CFU/mL.The dose of oral administration was adjusted according to the body weight of the mice, and the dose was given at the same time every day.The intervention period was 4 weeks.FBG and body weight were recorded once a week (Figure 1).injection, feeding was continued for 1 week, and mice were fasted for 12 h.Tail blo collected with a glucometer to determine fasting blood glucose (FBG).FBG ≥ 11.1 m is the success standard for the T2DM model [26].After successful T2DM modeling ent samples and doses of interventions were used according to the requirements group.The concentrations of metformin hydrochloride and vitamin C were 100 and the suspension concentration of Lactobacillus paracasei SR10-1 was 9 log CFU/m dose of oral administration was adjusted according to the body weight of the m the dose was given at the same time every day.The intervention period was 4 wee and body weight were recorded once a week (Figure 1).

Sample Collection
On the day before the end of the experiment, the mice were placed in a sterile cage, and fresh feces samples of mice were collected in sterile Eppendorf tubes.last day, the blood samples were collected from the eyeballs of mice after fasting and kept in an anticoagulant tube containing heparin sodium.The blood sampl centrifuged at 3500 rpm and 4 °C for 15 min, and the upper serum was collected.T were dissected under sterile conditions, and the livers and kidneys were weighed tion of the tissue was rinsed with pre-cooled saline water at 4 °C and placed in a yoxymethylene solution for histopathological examination.The remainder of th was stored at −80 °C.The cecum contents of mice were collected in sterile Epp tubes.All samples were stored at −80 °C until analysis.

Oral Glucose Tolerance Test
In the fourth week, mice were fasted for 12 h and orally administered 2.00 g of glucose.Then, their blood was taken by clipping the tails to measure blood levels at 0 min, 30 min, 60 min, and 120 min.The area under the blood glucos (AUC) was calculated from Equation (1).0.5 0.5 ( / ) 2

A B C D AUC hmmol L    
In Equation (1), A, B, C, and D represent the blood glucose values at 0 min, 60 min, and 120 min, respectively, after oral gavage of glucose solution.

Organ Coefficient
The organ coefficient is mainly used as an index to assess whether the fee FRRT causes the abnormalities of the liver, spleen, kidney, and heart of mice.An i in the organ coefficient indicates congestion, edema, or hypertrophy of the organ

Sample Collection
On the day before the end of the experiment, the mice were placed in a sterile mouse cage, and fresh feces samples of mice were collected in sterile Eppendorf tubes.On the last day, the blood samples were collected from the eyeballs of mice after fasting for 12 h and kept in an anticoagulant tube containing heparin sodium.The blood samples were centrifuged at 3500 rpm and 4 • C for 15 min, and the upper serum was collected.The mice were dissected under sterile conditions, and the livers and kidneys were weighed.A portion of the tissue was rinsed with pre-cooled saline water at 4 • C and placed in a 4% polyoxymethylene solution for histopathological examination.The remainder of the tissue was stored at −80 • C. The cecum contents of mice were collected in sterile Eppendorf tubes.All samples were stored at −80 • C until analysis.

Oral Glucose Tolerance Test
In the fourth week, mice were fasted for 12 h and orally administered 2.00 g/kg BW of glucose.Then, their blood was taken by clipping the tails to measure blood glucose levels at 0 min, 30 min, 60 min, and 120 min.The area under the blood glucose curve (AUC) was calculated from Equation (1).
In Equation (1), A, B, C, and D represent the blood glucose values at 0 min, 30 min, 60 min, and 120 min, respectively, after oral gavage of glucose solution.

Organ Coefficient
The organ coefficient is mainly used as an index to assess whether the feeding of FRRT causes the abnormalities of the liver, spleen, kidney, and heart of mice.An increase in the organ coefficient indicates congestion, edema, or hypertrophy of the organ, and a decrease in the organ coefficient indicates atrophy of the organ and other degenerative changes.The organ coefficient was calculated based on Equation (2).

Biochemical Analysis of Liver and Kidney
Commercial kits (Nanjing Jiancheng Bioengineering Institute, Nanjing, China) were used to determine the levels of triacylglycerol (TG), total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), and low-density lipoprotein cholesterol (LDL-C) in the liver and the levels of blood urea nitrogen (BUN) and creatinine in the kidney.

Histopathological Analysis
The liver and kidney tissues were fixed with a 4% polyoxymethylene solution, dehydrated, embedded in paraffin wax, sliced, and stained with hematoxylin and eosin (H&E), respectively.Samples were then observed with an optical microscope.

Analysis of Short-Chain Fatty Acids
Saturated sodium chloride solution, 3 mol/L HCl, and ether, with a volume of 0.4, 0.05, and 0.5 mL, respectively, were added to a 50 mg fecal sample for extraction.Mixtures were subsequently shaken for 10 min at 4 • C and centrifuged for 10 min at 12,000 rpm at 4 • C to obtain the supernatant.The CaSO 4 was added to the supernatant to absorb the water, and a 0.22 µm filter was used before the supernatant was injected into the GC-MS system.Analysis was performed using aTRACE1300-TSQ9000 gas chromatography (Thermo Fisher Scientific, Waltham, MA, USA).The extracts were separated using an Agilent HP-FFAP (25 m × 0.32 µm × 0.50 mm) for analysis of SCFAs (acetic acid, propionic acid, butyric acid, and isovaleric acid).Two microliters of the sample were manually injected into a split/splitless inlet (in split mode, 30:1), which was kept at 250 • C. The oven program was set at an initial temperature of 100 • C, and increased to 160 • C at a rate of 5 • C/min for 2 min.The temperatures of the transfer line and electron impact ion source were set at 2800 • C and 250 • C, respectively.The solvent delay was set at 3.5 min.The chromatogram was mapped in SIM mode with the characteristic ions of SCFAs obtained.

Analysis of Gut Microbiota
The DNA was extracted with the TGuide S96 Magnetic Soil DNA Kit (Model: DP812; Tiangen Biotech Co., Ltd., Beijing, China) according to the manufacturer's instructions.The DNA concentration of the samples was measured with the Qubit dsDNA HS Assay Kit and Qubit 4.0 Fluorometer (Invitrogen, Thermo Fisher Scientific, St. Bend, OR, USA).After amplification, PCR products were detected by electrophoresis with 1.8% agarose (Beijing Bomei Fuxin Technology Co., Ltd., Beijing, China).The 27F: AGRGTTTGATYNTG-GCTCAG and 1492R: TASGGHTACCTTGTTASGACTT universal primer set was used to amplify the full-length 16S rRNA gene from the genomic DNA extracted from each sample [27].SMRTbell libraries were prepared from the amplified DNA by SMRTbell Express Template Prep Kit 2.0 according to the manufacturer's instructions (Pacific Biosciences, Menlo Park, CA, USA).Purified SMRTbell libraries from the pooled and barcoded samples were sequenced on a single PacBio Sequel II 8M cell using the Sequel II Sequencing kit 2.0.UCHIME software was used to filter chimeric sequences in the obtained clean tags [28].The resulting sequences with an identity more significant than 97% were classified as operational taxonomic units (OTUs) [29].Bioinformatic analysis was performed on the cloud platform (https://www.bioincloud.tech,accessed on 22 November 2021) of Shenzhen Weikemeng Technology Group Co., Ltd.(Shenzhen, China).

Statistical Analysis
Statistical analyses were performed using IBM SPSS Statistics 26 (International Business Machines Corporation, Armonk, NY, USA), and the statistical results were expressed as mean ± standard deviation (S.D.).Analysis of variance (ANOVA) with Duncan's range was used to compare the differences between groups, and p < 0.05 was considered a significant difference.The sequencing data are analyzed on the cloud platform of Weikemeng Technology Group Co., Ltd.(https://www.bioincloud.tech/,accessed on 22 November 2021).

Body Weight and Fasting Blood Glucose
Weight loss is a typical symptom of T2DM [30][31][32].The body weight of mice in the M, P, V, R, CL, FL, FM, and FH groups was significantly decreased (p < 0.05) compared with the N and FN groups before administration.After the four-week intervention, the weight loss of the V, CL, FL, and FM groups was significantly suppressed (p < 0.05) compared with the M group (Figure 2A).The FBG was significantly decreased, by 56.08%, 52.82%, and 45.18% in the FL, FM, and FH groups (p < 0.05), respectively, and the effect of the FBG decrease was better in the FL, FM, and FH groups than in the P, V, R, and CL groups (Figure 2B).These results indicated that the FL, FM, and FH groups were shown to be more effective than the P, V, R, and CL groups in suppressing weight loss and decreasing FBG levels in T2DM mice.

Statistical Analysis
Statistical analyses were performed using IBM SPSS Statistics 26 (International Business Machines Corporation, Armonk, NY, USA), and the statistical results were expressed as mean ± standard deviation (S.D.).Analysis of variance (ANOVA) with Duncan's range was used to compare the differences between groups, and p < 0.05 was considered a significant difference.The sequencing data are analyzed on the cloud platform of Weikemeng Technology Group Co., Ltd.(https://www.bioincloud.tech/,accessed on 22 November 2021).

Body Weight and Fasting Blood Glucose
Weight loss is a typical symptom of T2DM [30][31][32].The body weight of mice in the M, P, V, R, CL, FL, FM, and FH groups was significantly decreased (p < 0.05) compared with the N and FN groups before administration.After the four-week intervention, the weight loss of the V, CL, FL, and FM groups was significantly suppressed (p < 0.05) compared with the M group (Figure 2A).The FBG was significantly decreased, by 56.08%, 52.82%, and 45.18% in the FL, FM, and FH groups (p < 0.05), respectively, and the effect of the FBG decrease was better in the FL, FM, and FH groups than in the P, V, R, and CL groups (Figure 2B).These results indicated that the FL, FM, and FH groups were shown to be more effective than the P, V, R, and CL groups in suppressing weight loss and decreasing FBG levels in T2DM mice.(B) FBG.All data are presented as the mean ± standard deviation.N: the normal group (normal saline, 10 mL/kg), FN: the normal control group (FRRT, 10 mL/kg), M: the model group (normal saline, 10 mL/kg), P: the positive control group (metformin hydrochloride, 10 mL/kg), V: the vitamin C control group (vitamin C, 10 mL/kg), R: the Lactobacillus paracasei SR10-1 control group (Lactobacillus paracasei SR10-1 suspension, 10 mL/kg), CL: the Rosa roxburghii Tratt fruit juice group (RRTJ, 10 mL/kg), FL: the low-dose FRRT group (FRRT, 5 mL/kg), FM: the medium-dose FRRT group (FRRT, 10 mL/kg), and FH: the high-dose FRRT group (FRRT, 15 mL/kg).

Oral Glucose Tolerance
The glucose levels rose rapidly in all groups within the first 30 min, peaked at 30 min, and then began to decline slowly.After 120 min, glucose in the N and FN groups returned to normal levels; the glucose levels of the P, V, R, CL, FL, FM, and FH groups were significantly decreased (p < 0.05) compared with the M group; and the FL group had a better All data are presented as the mean ± standard deviation.N: the normal group (normal saline, 10 mL/kg), FN: the normal control group (FRRT, 10 mL/kg), M: the model group (normal saline, 10 mL/kg), P: the positive control group (metformin hydrochloride, 10 mL/kg), V: the vitamin C control group (vitamin C, 10 mL/kg), R: the Lactobacillus paracasei SR10-1 control group (Lactobacillus paracasei SR10-1 suspension, 10 mL/kg), CL: the Rosa roxburghii Tratt fruit juice group (RRTJ, 10 mL/kg), FL: the low-dose FRRT group (FRRT, 5 mL/kg), FM: the medium-dose FRRT group (FRRT, 10 mL/kg), and FH: the high-dose FRRT group (FRRT, 15 mL/kg).

Oral Glucose Tolerance
The glucose levels rose rapidly in all groups within the first 30 min, peaked at 30 min, and then began to decline slowly.After 120 min, glucose in the N and FN groups returned to normal levels; the glucose levels of the P, V, R, CL, FL, FM, and FH groups were significantly decreased (p < 0.05) compared with the M group; and the FL group had a better effect than the other groups (Figure 3A).The AUC of the P, V, R, CL, FL, FM, and FH groups was significantly decreased (p < 0.05) compared with the M group, of which the FL group had a most significant decrease, of 26.80% (Figure 3B).effect than the other groups (Figure 3A).The AUC of the P, V, R, CL, FL, FM, and groups was significantly decreased (p < 0.05) compared with the M group, of which FL group had a most significant decrease, of 26.80% (Figure 3B).All data are presented as the mean ± standard deviation.N: the norm group (normal saline, 10 mL/kg), FN: the normal control group (FRRT, 10 mL/kg), M: the mo group (normal saline, 10 mL/kg), P: the positive control group (metformin hydrochloride, mL/kg), V: the vitamin C control group (vitamin C, 10 mL/kg), R: the Lactobacillus paracasei SR1 control group (Lactobacillus paracasei SR10-1 suspension, 10 mL/kg), CL: the Rosa roxburghii Tratt fr juice group (RRTJ, 10 mL/kg), FL: the low-dose FRRT group (FRRT, 5 mL/kg), FM: the medium-d FRRT group (FRRT, 10 mL/kg), and FH: the high-dose FRRT group (FRRT, 15 mL/kg).

Effects of FRRT on Liver and Kidney
FRRT can repair and improve the liver and kidney function of T2DM mice.After fo weeks of FRRT intervention, the liver and kidney coefficients of the FL and FM grou were significantly decreased (p < 0.05), and there were no significant differences in liv and kidney coefficients between the P, V, R, and CL groups (p > 0.05) compared with M group (Figure 4A,B).The abnormality of liver and kidney biochemical indexes is clos related to T2DM [33,34].The TG levels of the P, V, R, CL, FL, FM, and FH groups w significantly decreased, by 23.48%, 36.00%,49.28%, 55.66%, 38.09%, 63.55%, and 31.6 (Figure 4C, p < 0.05), respectively; the TC levels of the V, R, CL, FL, and FM groups w significantly decreased, by 20.30%, 39.36%, 41.52%, 27.85%, and 41.15% (Figure 4D, 0.05), respectively; the HDL-C levels of the P, V, R, CL, FL, FM, and FH groups were s nificantly increased (Figure 4E, p < 0.05); and the LDL-C levels of the P, V, CL, and F groups were significantly decreased (Figure 4F, p < 0.05) compared with the M group.T BUN and creatinine levels are used as indicators to judge renal function.The BUN lev of the P, V, R, CL, FL, FM, and FH groups were significantly decreased, by 5.75%, 4.33 16.09%, 16.46%, 14.70%, 10.98%, and 9.04% (p < 0.05), respectively, and the creatinine l els of the P, CL, FL, FM, and FH groups were significantly decreased, by 21.02%, 18.45 23.28%, 35.30%, and 21.78% (p < 0.05), respectively, compared with the M group (Figu 4G,H).These results indicated that the FM group (medium-dose FRRT) showed bet liver and kidney biochemical indexes than other groups.The FM group was significan better than the P group in TG, TC, BUN, and creatinine levels (p < 0.05); the FM group w slightly lower than the P group in HDL-C levels (p < 0.05); and there was no signific difference between the FM and the P groups in LDL-C levels (p > 0.05).

Effects of FRRT on Liver and Kidney
FRRT can repair and improve the liver and kidney function of T2DM mice.After four weeks of FRRT intervention, the liver and kidney coefficients of the FL and FM groups were significantly decreased (p < 0.05), and there were no significant differences in liver and kidney coefficients between the P, V, R, and CL groups (p > 0.05) compared with the M group (Figure 4A,B).The abnormality of liver and kidney biochemical indexes is closely related to T2DM [33,34].The TG levels of the P, V, R, CL, FL, FM, and FH groups were significantly decreased, by 23.48%, 36.00%,49.28%, 55.66%, 38.09%, 63.55%, and 31.62%(Figure 4C, p < 0.05), respectively; the TC levels of the V, R, CL, FL, and FM groups were significantly decreased, by 20.30%, 39.36%, 41.52%, 27.85%, and 41.15% (Figure 4D, p < 0.05), respectively; the HDL-C levels of the P, V, R, CL, FL, FM, and FH groups were significantly increased (Figure 4E, p < 0.05); and the LDL-C levels of the P, V, CL, and FM groups were significantly decreased (Figure 4F, p < 0.05) compared with the M group.The BUN and creatinine levels are used as indicators to judge renal function.The BUN levels of the P, V, R, CL, FL, FM, and FH groups were significantly decreased, by 5.75%, 4.33%, 16.09%, 16.46%, 14.70%, 10.98%, and 9.04% (p < 0.05), respectively, and the creatinine levels of the P, CL, FL, FM, and FH groups were significantly decreased, by 21.02%, 18.45%, 23.28%, 35.30%, and 21.78% (p < 0.05), respectively, compared with the M group (Figure 4G,H).These results indicated that the FM group (medium-dose FRRT) showed better liver and kidney biochemical indexes than other groups.The FM group was significantly better than the P group in TG, TC, BUN, and creatinine levels (p < 0.05); the FM group was slightly lower than the P group in HDL-C levels (p < 0.05); and there was no significant difference between the FM and the P groups in LDL-C levels (p > 0.05).

Histological Traits of Liver and Kidney
The structure of hepatocytes in the N group was well-defined, regular, wellranged, and evenly distributed.However, in the M group the hepatocytes were vac lated and necrotic, and many fat particles infiltrated the liver.After four weeks of in vention, the infiltration of fat granules and the degree of vacuolation and necrosis hepatocytes in the P, V, and FM groups were alleviated (Figure 5A).Mesangium enlar ment, tubulointerstitial fibrosis, glomerulosclerosis, and vacuolar degeneration of glom ular epithelial cells were observed in the M group.After four weeks of intervention, glomerular interstitial fibrosis, epithelial vacuoles, and glomerular atrophy were proved in the R, FL, and FM groups (Figure 5B).

Histological Traits of Liver and Kidney
The structure of hepatocytes in the N group was well-defined, regular, well-arranged, and evenly distributed.However, in the M group the hepatocytes were vacuolated and necrotic, and many fat particles infiltrated the liver.After four weeks of intervention, the infiltration of fat granules and the degree of vacuolation and necrosis of hepatocytes in the P, V, and FM groups were alleviated (Figure 5A).Mesangium enlargement, tubulointerstitial fibrosis, glomerulosclerosis, and vacuolar degeneration of glomerular epithelial cells were observed in the M group.After four weeks of intervention, the glomerular interstitial fibrosis, epithelial vacuoles, and glomerular atrophy were improved in the R, FL, and FM groups (Figure 5B).

Composition Analysis of the Gut Microbiota
FRRT affected the diversity and abundance of the gut microbiota in T2DM mice.T gut microbiota of each group of mice were sequenced, tags with more than 97% similar were clustered, and each OTU obtained was considered as the same species [29].OT varied among intervention groups, with the FN group having the highest OTUs, follow by the FM group (Figure 8A).The alpha diversity of species within the gut microbiom a common outcome of interest that is examined in microbiome research.It contains t sub-constructs: richness and evenness of the species composition in the sample.T Chao1 index is commonly used to indicate microbial species richness, and the Shann and Simpson indexes represent microbial species diversity [39].The Chao1 and Shann indexes were significantly higher in the FN group than in the N group (Figure 8B,C, 0.05), indicating that the abundance and diversity of the gut microbiota were significan increased (p < 0.05) in healthy mice after FRRT intervention.The Simpson indexes in P, V, R, CL, FL, and FM groups were significantly decreased (p < 0.05) compared to the group, indicating that the diversity of the gut microbiota was significantly increased (F ure 8D, p < 0.05) in the P, V, R, CL, FL, and FM groups.Beta diversity compares the co position of microbial communities between different samples, indicating the differen between microbial communities, which is measured by comparing the distance valu between samples; the closer the distance between the samples is, the more similar the s cies composition is.The sample distance matrix is analyzed using dimension-reduc sorting through principal component analysis (PCA) and non-metric multidimensio scaling (NMDS).Then, the sample relationship is intuitively judged on the three-dim sional plane [40,41].These results indicated that the P, R, FL, and FH groups are relativ close, indicating that metformin hydrochloride, Lactobacillus paracasei SR10-1 bacter and FRRT have similar effects on the gut microbiota of T2DM mice.After the intervent of FRRT in the FN group, the gut microbiota were significantly different from the ot (B) butyric acid and isovaleric acid.Different superscript characters indicate differences significantly among groups (p < 0.05).All data are presented as the mean ± standard deviation.N: the normal group (normal saline, 10 mL/kg), FN: the normal control group (FRRT, 10 mL/kg), M: the model group (normal saline, 10 mL/kg), P: the positive control group (metformin hydrochloride, 10 mL/kg), V: the vitamin C control group (vitamin C, 10 mL/kg), R: the Lactobacillus paracasei SR10-1 control group (Lactobacillus paracasei SR10-1 suspension, 10 mL/kg), CL: the Rosa roxburghii Tratt fruit juice group (RRTJ, 10 mL/kg), FL: the low-dose FRRT group (FRRT, 5 mL/kg), FM: the medium-dose FRRT group (FRRT, 10 mL/kg), and FH: the high-dose FRRT group (FRRT, 15 mL/kg).

Composition Analysis of the Gut Microbiota
FRRT affected the diversity and abundance of the gut microbiota in T2DM mice.The gut microbiota of each group of mice were sequenced, tags with more than 97% similarity were clustered, and each OTU obtained was considered as the same species [29].OTUs varied among intervention groups, with the FN group having the highest OTUs, followed by the FM group (Figure 8A).The alpha diversity of species within the gut microbiome is a common outcome of interest that is examined in microbiome research.It contains two sub-constructs: richness and evenness of the species composition in the sample.The Chao1 index is commonly used to indicate microbial species richness, and the Shannon and Simpson indexes represent microbial species diversity [39].The Chao1 and Shannon indexes were significantly higher in the FN group than in the N group (Figure 8B,C, p < 0.05), indicating that the abundance and diversity of the gut microbiota were significantly increased (p < 0.05) in healthy mice after FRRT intervention.The Simpson indexes in the P, V, R, CL, FL, and FM groups were significantly decreased (p < 0.05) compared to the M group, indicating that the diversity of the gut microbiota was significantly increased (Figure 8D, p < 0.05) in the P, V, R, CL, FL, and FM groups.Beta diversity compares the composition of microbial communities between different samples, indicating the differences between microbial communities, which is measured by comparing the distance values between samples; the closer the distance between the samples is, the more similar the species composition is.The sample distance matrix is analyzed using dimension-reducing sorting through principal component analysis (PCA) and non-metric multidimensional scaling (NMDS).Then, the sample relationship is intuitively judged on the three-dimensional plane [40,41].These results indicated that the P, R, FL, and FH groups are relatively close, indicating that metformin hydrochloride, Lactobacillus paracasei SR10-1 bacterial, and FRRT have similar effects on the gut microbiota of T2DM mice.After the intervention of FRRT in the FN group, the gut microbiota were significantly different from the other groups (Figure 8E,F, p < 0.05).The phylum-level gut microbiota were mainly composed of Firmicutes, Proteobacteria, Bacteroidetes, Verrucomicrobia, Cyanobacteria, Deferribacterota, Actinobacteria, Tenericutes, and TM7.The relative abundance of Firmicutes in the P (84.46%),V (94.65%),CL (89.37%), and FM (83.33%) groups was significantly increased (p < 0.05) compared with the M group (74.93%).The relative abundance of Proteobacteria in the M group (13.40%) was significantly increased (p < 0.05) compared with the N group (3.38%).After FRRT intervention, the relative abundance of Proteobacteria returned to 3.26% in the FL group and 5.02% in the FM group (Figure 9A), indicating that the V, CL, FL, and FM groups reduced the ratio of Firmicutes to Proteobacteria (Figure 9B).
For the family-level gut microbiota, the relative abundance of Lachnospiraceae in the R, FL, FM, and FH groups was significantly increased (p < 0.05) compared with the M group.Enterobacteriaceae is thought to be associated with overweight and obesity [42].The relative abundance of Enterobacteriaceae in the M group (11.18%) was significantly increased (p < 0.05) compared with the N group (0.02%).After the intervention, the relative abundance of Enterobacteriaceae in the V (1.12%), CL (0.43%), FL (1.18%), and FM (0.00%) groups was significantly decreased (p < 0.05).Veillonellaceae is a common Gramnegative anaerobe that can cause endogenous intestinal infection as a conditional pathogen.The relative abundance of Veillonellaceae in the gut microbiota of diabetic children was significantly higher than that of healthy children [43,44].In the present study, Veillonellaceae only appeared in the M group, and the relative abundance of 8.28% indicated that high-fat and high-sugar feed-induced diabetic mice increased the number of pathogenic bacteria.However, Veillonellaceae did not appear in the FL, FM, and FH groups after the FRRT intervention (Figure 10A).The phylum-level gut microbiota were mainly composed of Firmicutes, Proteobacteria, Bacteroidetes, Verrucomicrobia, Cyanobacteria, Deferribacterota, Actinobacteria, Tenericutes, and TM7.The relative abundance of Firmicutes in the P (84.46%),V (94.65%),CL (89.37%), and FM (83.33%) groups was significantly increased (p < 0.05) compared with the M group (74.93%).The relative abundance of Proteobacteria in the M group (13.40%) was significantly increased (p < 0.05) compared with the N group (3.38%).After FRRT intervention, the relative abundance of Proteobacteria returned to 3.26% in the FL group and 5.02% in the FM group (Figure 9A), indicating that the V, CL, FL, and FM groups reduced the ratio of Firmicutes to Proteobacteria (Figure 9B).
For the family-level gut microbiota, the relative abundance of Lachnospiraceae in the R, FL, FM, and FH groups was significantly increased (p < 0.05) compared with the M group.Enterobacteriaceae is thought to be associated with overweight and obesity [42].The relative abundance of Enterobacteriaceae in the M group (11.18%) was significantly increased (p < 0.05) compared with the N group (0.02%).After the intervention, the relative abundance of Enterobacteriaceae in the V (1.12%), CL (0.43%), FL (1.18%), and FM (0.00%) groups was significantly decreased (p < 0.05).Veillonellaceae is a common Gram-negative anaerobe that can cause endogenous intestinal infection as a conditional pathogen.The relative abundance of Veillonellaceae in the gut microbiota of diabetic children was significantly higher than that of healthy children [43,44].In the present study, Veillonellaceae only appeared in the M group, and the relative abundance of 8.28% indicated that high-fat and high-sugar feed-induced diabetic mice increased the number of pathogenic bacteria.However, Veillonellaceae did not appear in the FL, FM, and FH groups after the FRRT intervention (Figure 10A).For the genus-level gut microbiota, Akkermansia is a novel probiotic that degrad mucin in the human gut and is inversely associated with obesity, diabetes, inflammatio and metabolic disorders.Decreased abundance of Akkermansia leads to the impairment insulin secretion and glucose homeostasis in T2DM [45].The relative abundance of Akk mansia in the R (13.59%), FL (17.95%),FM (11.40%), and FH (16.01%) groups was sign cantly increased (p < 0.05) compared with the M group (8.80%).Klebsiella, a condition pathogen that parasitizes the respiratory or intestinal tract of animals, is one of the pa ogens causing pneumonia in humans.Various strains in the genus Klebsiella have evolv to become a major clinical and public health threat worldwide [46].The relative abu dance of Klebsiella in the M group (10.20%) was significantly increased (p < 0.05) compar with the N group (0.00%).After the FRRT intervention, the relative abundance of Klebsie reduced to 1.18% in the FL group and 0.00% in the FM group.Allobaculum was positive correlated with the expression of angiopoietin-like protein 4 [47], and the relative abu dance of Allobaculum was 0.02% in the N group, whereas that of the FN group was sign icantly increased after FRRT intervention, with a relative abundance of 5.62% (p < 0.0 Phascolarctobacterium only appeared in the M group (8.28%), and Phascolarctobacterium w an essential biomarker in obese patients with T2DM and might be a potential therapeu target [48] (Figure 10B).For the genus-level gut microbiota, Akkermansia is a novel probiotic that degrades mucin in the human gut and is inversely associated with obesity, diabetes, inflammation, and metabolic disorders.Decreased abundance of Akkermansia leads to the impairment of insulin secretion and glucose homeostasis in T2DM [45].The relative abundance of Akkermansia in the R (13.59%), FL (17.95%),FM (11.40%), and FH (16.01%) groups was significantly increased (p < 0.05) compared with the M group (8.80%).Klebsiella, a conditioned pathogen that parasitizes the respiratory or intestinal tract of animals, is one of the pathogens causing pneumonia in humans.Various strains in the genus Klebsiella have evolved to become a major clinical and public health threat worldwide [46].The relative abundance of Klebsiella in the M group (10.20%) was significantly increased (p < 0.05) compared with the N group (0.00%).After the FRRT intervention, the relative abundance of Klebsiella reduced to 1.18% in the FL group and 0.00% in the FM group.Allobaculum was positively correlated with the expression of angiopoietin-like protein 4 [47], and the relative abundance of Allobaculum was 0.02% in the N group, whereas that of the FN group was significantly increased after FRRT intervention, with a relative abundance of 5.62% (p < 0.05).Phascolarctobacterium only appeared in the M group (8.28%), and Phascolarctobacterium was an essential biomarker in obese patients with T2DM and might be a potential therapeutic target [48] (Figure 10B). .Species composition of gut microbiota in T2DM mice: (A) family-level abundance of g microbiota, and (B) genus-level abundance of gut microbiota.Each family or genus is represent by a unique color.N: the normal group (normal saline, 10 mL/kg), FN: the normal control gro (FRRT, 10 mL/kg), M: the model group (normal saline, 10 mL/kg), P: the positive control gro (metformin hydrochloride, 10 mL/kg), V: the vitamin C control group (vitamin C, 10 mL/kg), R: Lactobacillus paracasei SR10-1 control group (Lactobacillus paracasei SR10-1 suspension, 10 mL/kg), C the Rosa roxburghii Tratt fruit juice group (RRTJ, 10 mL/kg), FL: the low-dose FRRT group (FRRT mL/kg), FM: the medium-dose FRRT group (FRRT, 10 mL/kg), and FH: the high-dose FRRT gro (FRRT, 15 mL/kg).

Discussion
RRT is rich in bioactive substances, such as vitamin C, polysaccharides, polyphenols, and SOD, often used in the research and development of functional foods [13].Polysaccharides [17] and polyphenols [22] in RRT have been proven to have hypoglycemic effects.The previous research of our team found that RRT fruit vinegar had an ameliorating effect on dyslipidemia and other phenomena in obese mice [25].However, the relationship between FRRT and T2DM was unclear.In the present study, FRRT was used to intervene in STZ-induced T2DM mice.Body weight, blood glucose, the liver coefficient, the kidney coefficient, oxidative stress, SCAFs, and gut microbiota were measured in both T2DM and healthy mice.
Weight loss is an early symptom of T2DM, mainly due to the relative insulin deficiency caused by pancreatic β-cell dysfunction and insulin resistance in target organs [1].The body does not fully utilize glucose for energy production, leading to increased fat and protein breakdown, over-consumption, negative nitrogen balance, and even weight loss [31].In the present study, the weight loss of mice in the FL and FM groups was significantly suppressed (Figure 2A, p < 0.05).FBG in the FL and FM groups was significantly decreased (Figure 2B, p < 0.05).Glucose tolerance refers to the body's ability to regulate

Discussion
RRT is rich in bioactive substances, such as vitamin C, polysaccharides, polyphenols, and SOD, often used in the research and development of functional foods [13].Polysaccharides [17] and polyphenols [22] in RRT have been proven to have hypoglycemic effects.The previous research of our team found that RRT fruit vinegar had an ameliorating effect on dyslipidemia and other phenomena in obese mice [25].However, the relationship between FRRT and T2DM was unclear.In the present study, FRRT was used to intervene in STZ-induced T2DM mice.Body weight, blood glucose, the liver coefficient, the kidney coefficient, oxidative stress, SCAFs, and gut microbiota were measured in both T2DM and healthy mice.
Weight loss is an early symptom of T2DM, mainly due to the relative insulin deficiency caused by pancreatic β-cell dysfunction and insulin resistance in target organs [1].The body does not fully utilize glucose for energy production, leading to increased fat and protein breakdown, over-consumption, negative nitrogen balance, and even weight loss [31].In the present study, the weight loss of mice in the FL and FM groups was significantly suppressed (Figure 2A, p < 0.05).FBG in the FL and FM groups was significantly decreased (Figure 2B, p < 0.05).Glucose tolerance refers to the body's ability to regulate blood sugar levels.In oral glucose tolerance tests, the FL and FM groups improved the glucose tolerance of T2DM mice (Figure 3A,B).The FL (low-dose FRRT) and FM (medium-dose FRRT) groups showed significantly better results than the P group (metformin hydrochloride) in inhibiting weight loss and controlling FBG and glucose tolerance.
Patients with diabetes are often accompanied by dyslipidemia, leading to increased TG, TC, LDL-C, BUN, and creatinine levels and decreased HDL-C levels [48,51].Dyslipidemia in T2DM mice was ameliorated after metformin hydrochloride (P group), RRTJ (CL group), low-dose FRRT (FL group), and medium-dose FRRT (FM group) interventions, and the effect of medium-dose FRRT was significantly superior to that of metformin hydrochloride, RRTJ, and low-dose FRRT.Fermented juice had significantly better effects on the regulation of blood lipids, oxidative stress, and antidiabetes than unfermented juice [32,52].
Oxidative stress is a pathogenesis of T2DM.The levels of GSH-PX, CAT, SOD, and T-AOC were significantly decreased (p < 0.05), and the levels of MDA were significantly increased (p < 0.05) in T2DM mice (Figure 6).The SOD and ascorbic acid in RRT can balance oxidative stress in humans.Oxidative stress is caused by a physiological imbalance between levels of antioxidants and oxidants (free radicals and reactive species) [20].The GSH-PX and SOD levels of T2DM mice in the CL group were significantly increased (Figure 6A,C, p < 0.05), and the SOD levels of T2DM mice in the R group were significantly increased (Figure 6C, p < 0.05).After RRTJ was fermented by Lactobacillus paracasei SR10-1 (FRRT), the levels of GSH-PX, CAT, SOD, and T-AOC had significantly increased (p < 0.05), indicating the synergistic effect of Lactobacillus paracasei SR10-1, and RRTJ significantly improved oxidative stress in T2DM mice (Figure 6).
Observational findings achieved during the past two decades suggest that gut microbiota are essential for human health [4].Humans benefit from the production of SCFAs by the gut microbiota through carbohydrate fermentation, and it stands to reason that insufficient production of SCFAs is associated with T2DM [37].Su et al. reported that dietary fiber from RRT residue could increase the production of SCFAs by colon fermentation in vitro [14].The SCFAs of T2DM mice in the FM group were more significantly increased (p < 0.05) than in the P, R, and CL groups (Figure 7), and it also suggests that Lactobacillus paracasei SR10-1 fermentation enhanced the functionality of RRTJ, and the organic combination of this strain and RRTJ enhanced the efficiency of FRRT.The healthy and T2DM mice had significantly increased (p < 0.05) SCFAs production after mediumdose FRRT intervention (Figure 7), an intriguing result that suggests that, for a healthy population, FRRT can benefit gut microbiota as a functional food.The SCFAs levels of vitamin C intervention were significantly increased, especially propionic acid and butyric acid [53].Propionic acid can enhance glucose-stimulated insulin release and maintain β-cell mass by inhibiting apoptosis [40].Butyric acid can increase insulin sensitivity, inhibit liver gluconeogenesis, and improve host sugar metabolism and insulin resistance [41].And in the present study, the propionic acid and butyric acid levels in T2DM mice of vitamin C intervention (vitamin C: 1000 mg/100 g) were more significantly increased (p < 0.05) than of medium-dose FRRT intervention (vitamin C: 375.65 ± 7.39 mg/100 g).
Gut microbiota are closely related to human health [54][55][56].It has been confirmed that T2DM is closely associated with the gut microbiota.Das et al. first reported significant differences in the diversity and abundance of gut microbiota at the phylum and genus levels between diabetic retinopathy patients and healthy individuals [57].High-fat and high-sugar diet-induced T2DM leads to strong alterations in the composition of the gut microbiota [58].In the present study, the relative abundance of Enterobacteriaceae that was considered to be associated with overweight and obesity [42] in the M group (11.18%) was significantly increased (p < 0.05) compared with the N group (0.02%).Veillonellaceae is a common Gram-negative anaerobe that can cause endogenous intestinal infections as a conditional pathogen, which occurs only in the M group (8.28%) (Figure 10A).Phascolarctobacterium was significantly negatively correlated with weight (p < 0.05), SOD (p < 0.01), CAT (p < 0.05), and T-AOC (Figure 11C, p < 0.05), which was consistent with the study by Wang et al., who found that Phascolarctobacterium was an essential biomarker in obese patients with T2DM [48].Klebsiella is present in the gut microbiota of T2DM mice, suggesting that people with diabetes are more susceptible to Klebsiella peneumoniae [59].The butyric acid content in the cecum was significantly increased due to the stimulation of Lachnospiraceae [60].Akkermansia is a novel probiotic that degrades mucin in the human gut and is inversely associated with obesity, diabetes, inflammation, and metabolic disorders.Akkermansia is reduced, impairing insulin secretion and glucose homeostasis in T2DM [45].In the correlation analysis, Akkermansia was significantly negatively correlated (p < 0.05) with weight (Figure 11C).RRT treatment can reverse the gut microbiota disturbance from a high-fat diet [61].Metformin hydrochloride intervention increased the relative abundance of Firmicutes; vitamin C intervention increased the relative abundance of Firmicutes and decreased the relative abundance of Enterobacteriaceae; Lactobacillus paracasei SR10-1 intervention increased the relative abundance of Lachnospiraceae and Akkermansia; RRTJ intervention increased the relative abundance of Firmicutes and decreased the relative abundance of Enterobacteriaceae; low-dose FRRT intervention increased the relative abundance of Lachnospiraceae and Akkermansia and decreased the relative abundance of Proteobacteria, Enterobacteriaceae, Veillonellaceae, Klebsiella, and Phascolarctobacterium; medium-dose FRRT intervention increased the relative abundance of Firmicutes, Lachnospiraceae, Akkermansia, and Allobaculum and decreased the relative abundance of Proteobacteria, Enterobacteriaceae, Veillonellaceae, Klebsiella, and Phascolarctobacterium; and high-dose FRRT intervention increased the relative abundance of Lachnospiraceae and Akkermansia.
The medium-dose FRRT intervention (FM group) showed better antidiabetic effects than metformin hydrochloride (P group), vitamin C (V group), Lactobacillus paracasei SR10-1 (R group), RRTJ (CL group), low-dose FRRT (FL group), and high-dose FRRT (FH group) interventions.Therefore, the imbalance in the gut microbiota of T2DM mice was effectively alleviated after FRRT intervention, and medium-dose FRRT intervention was more effective than other interventions in gut microbiota regulation.This finding also provides a reference for FRRT to explore the relationship between the gut microbiome and T2DM and other metabolic syndromes.However, it is not clear whether FRRT improves T2DM by one or more components.

Conclusions
In summary, this study reveals the regulatory effects of metformin hydrochloride, vitamin C, Lactobacillus paracasei SR10-1, RRTJ, low-dose FRRT, medium-dose FRRT, and high-dose FRRT on oxidative stress, dyslipidemia, and gut microbiota in high-sugar and high-fat diets and STZ-induced T2DM.The synergistic effect of Lactobacillus paracasei SR10-1 and RRTJ fermented FRRT with a better antidiabetic effect.The medium-dose FRRT (FM group) reverses dyslipidemia, relieves oxidative stress, increases SCFAs levels, and regulates the composition of gut microbiota better than the other intervention groups.The results show that the improvement of body weight and oxidative stress in T2DM mice after supplementation by medium-dose FRRT are highly correlated with the Odoribacter, Coprobacillus, Coprococcus, Ruminococcus, Dorea, Lactobacillus, Helicobacter, Bifidobacterium, Parabacteroides, Phascolarctobacterium, Sutterella, Bacteroides, Psychrobacter, Escherichia, Klebsiella, Weissella, and Akkermansia.This study demonstrated that FRRT has excellent potential for dietary prevention of T2DM and related chronic diseases.The medium-dose of FRRT supplementation has beneficial health effects on healthy and T2DM mice.

Figure 1 .
Figure 1.Overview of the T2DM model procedure.

Figure 1 .
Figure 1.Overview of the T2DM model procedure.