Article Text
PDF
Abstract
Introduction There has been increasing evidence that the gut microbiota is closely related to type 2 diabetes (T2D). Metformin (Met) is often used in combination with saxagliptin (Sax) and repaglinide (Rep) for the treatment of T2D. However, little is known about the effects of these combination agents on gut microbiota in T2D.
Research design and methods A T2D mouse model induced by a high-fat diet (HFD) and streptozotocin (STZ) was employed. The T2D mice were randomly divided into six groups, including sham, Met, Sax, Rep, Met+Sax and Met+Rep, for 4 weeks. Fasting blood glucose level, serum biochemical index, H&E staining of liver, Oil red O staining of liver and microbiota analysis by 16s sequencing were used to access the microbiota in the fecal samples.
Results These antidiabetics effectively prevented the development of HFD/STZ-induced high blood glucose, and the combination treatment had a better effect in inhibiting lipid accumulation. All these dosing regimens restored the decreasing ratio of the phylum Bacteroidetes: Firmicutes, and increasing abundance of phylum Desulfobacterota, expect for Met. At the genus level, the antidiabetics restored the decreasing abundance of Muribaculaceae in T2D mice, but when Met was combined with Rep or Sax, the abundance of Muribaculaceae was decreased. The combined treatment could restore the reduced abundance of Prevotellaceae_UCG-001, while Met monotherapy had no such effect. In addition, the reduced Lachnospiraceae_NK4A136_group was well restored in the combination treatment groups, and the effect was much greater than that in the corresponding monotherapy group. Therefore, these dosing regimens exerted different effects on the composition of gut microbiota, which might be associated with the effect on T2D.
Conclusions Supplementation with specific probiotics may further improve the hypoglycemic effects of antidiabetics and be helpful for the development of new therapeutic drugs for T2D.
- Metformin
- Type 2 Diabetes
- Diabetes Mellitus, Type 2
- Microbiology
Data availability statement
Data are available upon reasonable request.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
Metformin monotherapy may play a part in hypoglycemic effects by altering the composition of gut microbiota.
WHAT THIS STUDY ADDS
We discuss the effects of saxagliptin and repaglinide alone or in combination with metformin on gut microbiota.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
This study provides a reference for the development of new therapeutic drugs. Supplementation with specific probiotics may further improve the hypoglycemic effects of antidiabetics.
Introduction
The WHO has predicted type 2 diabetes (T2D) to be the seventh leading cause of death by 2030, and there is an urgent need for the development of more efficient prevention and treatment strategies.1 A growing body of evidence demonstrates the importance of the gut microbiota in the development of a multitude of complex human diseases, including T2D. In recent years, much attention has been given to the gut microbiota and its potential regulatory effects on T2D. Numerous metagenomic studies have found that the composition profile of the gut microbiota, especially the ratio of Bacteroidetes to Firmicutes, directly affects the development of T2D by influencing intestinal permeability, inflammation, the immune system, and energy metabolism.2 3
Twelve classes of drugs for T2D are classified as biguanides (eg, metformin), dipeptidylpeptidase-4 inhibitors (eg, saxagliptin), meglitinides (eg, repaglinide), sulfonylureas, thiazolidinediones, Sodium-glucosecotransporter-2 inhibitors, Glucagon-like peptide 1 receptor agonists, insulins, α-glucosidase inhibitors, dopaminergic antagonists, bile acid sequestrants, meglitinides and amylinomimetics.4 Metformin, a first-line antidiabetic, alone or in combination with other antidiabetics (such as saxagliptin or repaglinide) can effectively control the blood glucose of T2D, especially obese or overweight diabetic patients.5–7
Saxagliptin is a highly potent, selective, reversible and competitive DPP-4 inhibitor that shows major effects on lowering blood glucose and body weight.8 Saxaglitpin in combination with metformin leads to better improvement in blood glucose control (fasting plasma glucose, postprandial glucose, HbA1c) than either compound separately.9 Repaglinide, a meglitinide hypoglycemic agent, is effective in controlling postprandial hyperglycemia with minimal risk of hypoglycemia.10 Repaglinide in combination with metformin improves glycemic control and weight neutrality without adverse effects on lipid profiles.11
Some studies have reported that metformin may play a part in hypoglycemic effects by altering the composition of gut microbiota to maintain the integrity of intestinal flora, promote the production of short-chain fatty acids (SCFAs), regulate bile acid metabolism and improve glucose homeostasis.12 However, the effects of saxagliptin and repaglinide alone or in combination with metformin on gut microbiota have not been reported. Therefore, for the first time, we compared the effects of the three drugs alone or in combination on the gut microbiota in T2D mice.
Materials and methods
T2D mouse model
Four-week-old male C57BL/6J mice were purchased from the Nanjing Biomedical Research Institute of Nanjing University (Nanjing, China). The mice were maintained on a 12-hour light and dark cycle at 22°C~24°C and 50%~60% humidity, given water and food ad libitum and housed in plastic cages (6 mice per cage).
After 1 week of acclimation, the mice were randomly divided into two groups: the control group and the T2D group. Mice in the control group were fed a normal diet. To induce diabetes, mice in the T2D group were fed a high-fat diet (HFD, 60 kcal % fat, Dowsentec, China) for 6 weeks and given an intraperitoneal injection of 45 mg/kg streptozotocin (STZ) in 0.01 M citrate buffer, pH 4.2, for three consecutive days. Meanwhile, the control mice were treated with a citrate buffer solution. A T2D mouse model was considered successfully established with fasting blood glucose (FBG) levels over 11.1 mM.
Drug treatment
The T2D mice were randomly divided into six groups according to different treatments: sham group, metformin, saxagliptin, repaglinide, metformin in combination with saxagliptin or repaglinide (6 mice per group). C57BL/6J mice were administered certain solutions via oral gavage: metformin (Yiling Pharmaceutical, China, H20054790, 100 mg/kg/day), saxagliptin (AstraZeneca Pharmaceuticals, UK, J20171033, 2.5 mg/kg/day) and repaglinide (Honsoh Pharma, China, H20000362, 1 mg/kg/day) daily for 4 weeks. The mice in the control group and sham group were administered saline. During the period of drug treatment, mice were fasted overnight but had ad libitum access to water. The experimental operators were blinded after the animals were grouped.
Measurement of FBG
Tail venous blood samples for FBG were collected after fasting overnight for 12 hours with free access to water. FBG was measured every 5 days post drug administration.
Serum biochemical index
The mice were sacrificed at 16 weeks of age. Blood samples were collected, allowed to stand for 30 min and then centrifuged at 3000×g for 15 min to obtain serum. total cholesterol (TC), triglyceride (TG), high-density lipoprotein cholesterol (HDL-c), low-density lipoprotein cholesterol (LDL-c), aspartate aminotransferase (AST), alanine aminotransferase (ALT) and γ-glutamyl transferase (γ-GT) were analyzed using kits (Nanjing Jiancheng Bioengineering Institute, China).
H&E staining
Liver samples were isolated and fixed in 4% neutral-buffered formalin (Servicebio, Wuhan, China) and embedded in paraffin. After cutting into 5 µm-thick sections, the sections were stained with H&E. Histological changes were observed and photographed with a Nikon Eclipse E600 equipped with a Nikon Digital Sight DS-U1 unit (Spach Optics, USA).
Oid red O staining
Liver samples were isolated and fixed in 4% paraformaldehyde. After cutting into 10 µm-thick sections, the sections were stained with Oil Red O solution, destained with 60% isopropanol, and then incubated with hematoxylin for counterstaining.
Microbiota analysis by 16S sequencing
Total genomic DNA from fecal samples was extracted using the MagPure Soil DNA LQ Kit (Magen) according to the manufacturer’s protocols. The V3–V4 region of the prokaryotic 16s rRNA was amplified with forward primers 343F: 5’-TACGGRAGGCAGCAG-3’ and reverse primers 798R: 5’-AGGGTATCTAATCCT-3’. Final amplified products were quantified by Qubit3.0 Fluorometer. After amplification, sequencing was performed by the Illumina MiSeq platform (Illumina, San Diego, California). Paired-end reads were merged and trimmed to the same length. The raw reads were quality filtered to exclude chimeric sequences. Then, based on Silva 138, the sequence reads were binned into operational taxonomic units (OTUs) by VSEARCH (V.1.9.6) with a 97% similarity threshold. Alpha diversity was calculated with QIIME to assess the complexity of species diversity (V.1.8.0). Beta diversity was calculated by Principal Co-ordinates Analysis (PCoA) entify significantly different species at the OTU level, a linear discriminant analysis (LDA) effect size (LEfSe) algorithm was conducted using LEfSe V.1.0.0 software.
Statistical analysis
The results are represented as the means±SEs of the means and were analyzed using GraphPad Prism V.8.0. Significant differences between groups were assessed by one-way Analysis of Variance with Bonferroni’s post hoc tests. p<0.05 was considered statistically significant.
Results
Effects of antidiabetics on hyperglycemia in T2D mice
HFD induces insulin resistance, but a robust β-cell response protects against hyperglycemia. Therefore, STZ is frequently used to prevent β-cell compensation and in combination with HFD to establish the T2D mouse model.13 Here, we used HFD/STZ-induced T2D diabetic mice to investigate the effects and mechanisms of the five dosing regimens. A T2D mouse model was considered successfully established at 12 weeks with FBG levels of ≥11.1 mmol/L (figure 1A). Then, the T2D mice were randomly divided into six groups according to different drug treatments for 4 weeks. The FBG decreased significantly in each drug treatment group after a 2-week administration period and then decreased to the level of the control mice at 4 weeks except for the Rep and Sax group (figure 1B). We observed that both food and water consumption in Met+Rep and Met+Sax groups decreased to the level of the control mice after 4 weeks of administration (figure 1C,D). However, there was no significant increase in the weight of the drug-treated mice compared with the sham group (figure 1E).
The antidiabetics ameliorated hyperglycemia in type 2 diabetes (T2D) mice. (A) Experimental design. (B) Fasting blood glucose (FBG) level measurement in control group and T2D mice groups. Food (C) and water (D) consumption in control group and T2D mice groups. (E) Body weight in control group and T2D mice groups. *p<0.05, **p<0.01, ***p<0.001, compared with sham group.
Taken together, our results showed that these antidiabetics effectively prevented the development of HFD/STZ-induced high blood glucose, whether used alone or in combination, but only the combination treatment attenuated excessive intake of water and food in T2D mice.
Effects of antidiabetics on liver injury and lipid accumulation in T2D mice
HFD usually leads to liver injury and lipid accumulation in the liver, which is one of the prominent symptoms of HFD-fed mice. Interestingly, our blood biochemistry analysis showed that there was no significant intragroup variation between the control and sham groups in ALT, AST, TC, TG, LDL-c and γ-GT, but serum levels of γ-GT in the Met, Sax and Met+Sax groups were significantly reduced compared with those in the sham group after 4 weeks of treatment (figure 2A–G). Although the serum levels of TC and LDL-c were not significantly increased in T2D mice, we observed that there was a lower level of TC in the Met+Rep group than in the sham group. However, the sham group showed conspicuously lower serum levels of HDL-c than the control group, and only Met+Rep and Met+Sax treatment reversed the decrease in HDL-c levels (figure 2D).
The antidiabetics reduced liver injury and lipid accumulation in T2D mice. (A–G) ALT, AST, γ-GT, HDL-c, LDL-c, TC and TG were tested in control group and T2D mice groups. (H) H&E staining of liver sections in mice. (I) Oil red O staining of liver sections in mice. (J) Percentage of positive area of liver stained with Oil red O staining. (Magnification: x400).* p<0.05, **p<0.01, ***p<0.001, compared with sham group. T2D, type 2 diabetes; TC, total cholesterol; TG, triglyceride; HDL-c, high-density lipoprotein cholesterol; LDL-c, low-density lipoprotein cholesterol; AST, aspartate aminotransferase; ALT, alanine aminotransferase; γ-GT, γ-glutamyl transferase.
To directly assess the impact of different drug treatments on the liver, we collected the livers of mice for H&E staining and Oil red O staining. The results showed apparent hepatic steatosis and inflammatory cell infiltration in liver sections of T2D mice (figure 2H–J). The antidiabetics could reduce inflammatory cell infiltration and fat vacuolation, especially in the Met+Rep and Met+Sax groups. Therefore, combined with the above results, it is suggested that the combination treatment had a better effect in inhibiting lipid accumulation than these drugs used alone.
Effects of antidiabetics on the α-diversity and β-diversity of the gut microbiota in T2D mice
By comparing the five different antidiabetic treatments, we found that antidiabetics had significant beneficial effects on relieving symptoms in T2D mice, especially combination administration. Since gut microbiota metabolism was reported to play a critical role in blood glucose regulation, we set out to determine whether the glucose regulation effect of antidiabetics in T2D mice was associated with gut microbiota.
Feces samples were collected at 16 weeks of age, when the mice had been treated with hypoglycemic drugs for 4 weeks. Alpha diversity indices showed that the richness (chao1) and diversity (shannon) were significantly decreased in the sham group compared with the control group. Except for metformin, the other four dosing regimens significantly increased the richness and diversity of the microbiota in T2D mice (figure 3A–C). Beta diversity was used to evaluate the differences and similarities in microbiota composition among the samples. The PCoA and Non-metric Multidimensional Scaling (NMDS) exhibited significantly different clusters between the antidiabetic groups and sham group (figure 3D,E).
The antidiabetics increased α-diversity and β-diversity of the gut microbiota in T2D mice. (A) Observed OTUs (B) Chao 1 index (C) Shanon index (D) PCoA analysis (E) NMDS analysis. *p<0.05, **p<0.01, ***p<0.001, compared with sham group. OTUs, operational taxonomic units; T2D, type 2 diabetes. PCoA, principal co-ordinates analysis; NMDS, Non-metric Multidimensional Scaling.
Effects of antidiabetics on the composition and diversity of the gut microbiota in T2D mice
Compared with the control group, the phyla Firmicutes, Desulfobacterota, Actinobacteriota and Deferribacterota were increased, while Bacteroidetes and Verrucomicrobiota were reduced in the sham group (figure 4A and online supplemental figure S1). Antidiabetics could increase the ratio of Bacteroidetes to Firmicutes, and metformin could increase it by 21.8-fold in particular. However, there were no significant differences among the other four groups, but all had significant differences compared with the Met group (p<0.01) (figure 4D). In addition, we found increased abundance of Desulfobacterota was reversed in the Met+Rep group, but not in the Met group.
Supplemental material
The antidiabetics effectively restored phylum Bacteroidetes and genus muribaculaceae populations in the gut microbiota composition of type 2 diabetes (T2D) mice. (A) The taxonomic composition distribution at the phylum level. (B) The relative abundance of phylum Bacteroidetes. (C) The relative abundance of phylum Firmicutes. (D) The relative abundance of phylum Bacteroidetes/Firmicutes. (E) The taxonomic composition distribution at the genus level. (F) The relative abundance of genus muribaculaceae. (G) Cladogram generated from linear discriminant analysis (LDA) coupled with effect size (taxa with LDA score>3.6 and significance of a<0.05 determined by Wilcoxon signed-rank test). *p<0.05, **p<0.01, and ***p<0.001, compared with sham group.
At the genus level, Muribaculaceae, belonged to the phylum Bacteroidetes, was dominant in the gut microbiota of the control mice, and antidiabetics could restore T2D-induced downregulation of Muribaculaceae, but the abundance in the combined treatment groups was lower than that in the monotherapy groups (figure 4E,F). A remarkable reduction in the abundance of another genus of Bacteroidetes, namely, Prevotellaceae_UCG-001, was observed in T2D mice. After the therapy, Prevotellaceae_UCG-001 was significantly increased, except for Met group. As the largest proportion of the phylum Firmicutes, the reduced Lachnospiraceae_NK4A136_group could be restored after the combined treatment, while metformin treatment did not have this effect (online supplemental figure S2).
To detect specific bacteria that covaried with different drug treatments, an LEfSe analysis with LDA coupled with effect size was performed to identify the taxonomic levels in the six T2D mouse groups that were significantly different (figure 4G). There was a higher abundance of the phyla Firmicutes and Actinobacteria in the sham mice than in the other mice. Met treatment significantly increased f_Desulfovibrionaceae, belonging to the order Desulfovibrionales. It was also observed that Rep induced two genera of the order Oscillospirales. Sax mainly increased the order of Bacteroidales. However, g_Burkholderia-Caballeronia-Paraburkholderia, classified in the order Burkholderiales, was associated with Met+Rep, while g_Escherichia-Shigella, classified in the order Enterobacterales, was associated with Met+Sax.
Effects of antidiabetics on gut microbiota function in T2D mice
To further understand the functional implications underlying the microbial differences among different mice, metagenomic sequencing was used to explore different Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways in the seven groups. Global and overview maps showed that carbohydrate metabolism, amino acid metabolism, energy metabolism, metabolism of cofactors and vitamins and membrane transport accounted for the top six highest abundances (figure 5A).
The prediction of KEGG function based on PICRUSt analysis. (A) Bar graph of KEGG showed the mean abundance of pathways in each group. (B) Bar graph of KEGG function terms showed different enrichments among each group. KEGG, Kyoto Encyclopedia of Genes and Genomes; PICRUSt, Phylogenetic Investigation of Communities by Reconstruction of Unobserved States.
Moreover, among the top 30 enriched KEGG function terms, KEGG orthologs (KOs) associated with K01990, K06147, K02004, K02003, K03406, K07024 and K06180 were upregulated in T2D mice and restored in the majority of the antidiabetic groups. KOs associated with K03088, K05349 and K01190 were downregulated in T2D mice and restored in the majority of the antidiabetic groups except the metformin group (figure 5B).
Discussion
The pathophysiological factors contributing to hyperglycemia are complicated, including dysbiosis in gut microbiota, inflammation, immune dysregulation and islet amyloid polypeptide (amylin) deposition in the pancreas,14 and these are also potential therapeutic targets for T2D. In our study, we chose the HFD/STZ-induced diabetic model, which is clinically relevant and extensively used to model T2D. The diabetogenic property of STZ is characterized by selective destruction of β cells, insulin deficiency, hyperglycemia, polydipsia and polyuria that resemble human diabetes.15
In our study, alpha diversity analysis showed that antidiabetic treatment could restore the decreased richness and diversity of microbiota in T2D mice, but there was no significant change after metformin treatment. This is consistent with the results previously reported in which the alpha diversity was not changed in Zucker diabetic fatty rats after metformin treatment.16 Interestingly, it has been reported that metformin could significantly reduce both the richness and diversity of the gut microbiota in HFD rats.17 In addition to animal models, a systematic review also revealed interactions between metformin and gut microbiome profiles in patients with T2D, in which the vast majority of studies showed that the α diversity was not affected by metformin treatment.18 Little has been reported about the effects of saxagliptin and repaglinide on the microbiota in diabetes. Wang et al19 reported that saxagliptin could restore the decreased richness and diversity of microbiota in STZ/ApoE−/− mice, which is in keeping with our analysis.
In recent years, many studies have shown that the Bacteroidetes/Firmicutes ratio in T2D is lower than that in healthy individuals,20 21 and the abundant of Firmicutes decreased after metformin treatment.18 We observed that the relative abundance of the phylum Bacteroidetes and the ratio of Bacteroidetes to Firmicutes were increased after antidiabetic treatment, especially in the Met group. However, there are a few opposite results. For instance, Chen et al22 reported that the relative abundance of the phylum Bacteroidetes was decreased in HFD/STZ mice after Met treatment, and the phylum Firmicutes was increased. Given the inconsistency of the results, Sophie et al suggested that analysis only from the phylum level could be too simplistic and misleading, as it could hide important changes at the genus or species level.23 In addition, we found that the phylum Desulfobacterota was increased in the T2D mice, metformin monotherapy could not decrease this phylum, but when metformin was combined with repaglinide, they could decrease this phylum significantly. It was previously shown that the abundance of Desulfobacterota has positive correlation with lipid concentration, which is consistent with our results.24 25
The genera Muribaculaceae and Prevotellaceae_UCG-001 belong to the phylum Bacteroidetes. As a predominant genus in the phylum Bacteroidetes, Muribaculaceae, previously named S24-7, was found to be lower in western style diet-fed and ob/ob mice than in control mice. In addition, these obese mice were characterized by increased penetration of the inner colonic mucus layer and a reduced mucus growth rate,26 27 and this genus has been reported to be associated with intact function of the inner colonic mucus layer.28 In our results, the antidiabetics restored the decreasing abundance of Muribaculaceae in T2D mice, which might be related to its ability to promote recovery of intact function. However, when Met was combined with Rep or Sax, the abundance of Muribaculaceae was decreased, although it was significantly increased compared with that of the sham group. Prevotellaceae_UCG-001 was found to promote indirectly fiber degradation to produce short SCFAs and showed a positive trend with other SCFAs-producing bacteria.29 30 SCFAs could improve glucose homeostasis and insulin function by effectively regulating glucose absorption and utilization in liver, skeletal muscle and adipose tissue.31 In our study, the combined treatment could restore the reduced abundance of Prevotellaceae_UCG-001, while metformin monotherapy had no such effect. As a predominant genus in the phylum Firmicutes, Lachnospiraceae_NK4A136_group has been reported to maintain the integrity of gut barrier in mice and to be negatively associated with intestinal permeability by producing SCFAs (especially acetic and butyric acid).32 33 Moreover, Lachnospiraceae_NK4A136_group was regarded to have the ability to reduce inflammation.34 After the antidiabetics treatment, the reduced Lachnospiraceae_NK4A136_group was well restored in the combination treatment groups, and the effect was much greater than that in the corresponding monotherapy group.
Our analyses above show that the combination group is unique in the gut flora, which is closely related to lipid accumulation and SCFA production. This association may be the advantage and characteristic of the combination treatment of T2D. However, our results differ from some reports. The main difference is likely to be the choice of model.35 Moreover, the gut microbiota is not uniform throughout the digestive tract, so the selection of samples from different parts of the digestive tract will also produce different results.36 In addition, different sequencing approaches used to analyze gut microbiota can also generate differences.23
Conclusions
We found that antidiabetic treatment could afford protection from HFD/STZ-induced hyperglycemia, liver injury and lipid accumulation. The combination treatment attenuated excessive intake of water and food in T2D mice and exerted a better effect in inhibiting lipid accumulation. This may be associated with the difference in the effect on gut microbiota. Our study found that metformin, saxagliptin and repaglinide used alone or in combination differentially affected the gut microbiota. This indicates that supplementation with specific probiotics may further improve the hypoglycemic effects of antidiabetics and be helpful for the development of new therapeutic drugs.
Data availability statement
Data are available upon reasonable request.
Ethics statements
Patient consent for publication
Ethics approval
This study was approved by the ethical and humane committee of Anhui University (protocol code 2021-013).
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Contributors TX and YT are the guarantors of this work. TX and ZF contributed to the study conception and design. YT, MY and SJ contributed to experiment operation. The first draft of the manuscript was written by YT and revised by TX.
Funding This study was supported by Anhui Province key Laboratory Performance Project (Grant Number 2019b12030026).
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer-reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.