https://orcid.org/0000-0002-5055-5931
Abstract
The aim of this study was to check the effect of long-term oral glutathione (GSH) supplementation on alteration in gut microbiome of Indian diabetic individuals. Early morning fresh stool sample of diabetic individuals recruited in a randomized clinical trial wherein they were given 500 mg GSH supplementation orally once a day for a period of 6 months was collected and gut microbiome was analysed using high throughput 16S rRNA metagenomic sequencing. Long-term GSH supplementation as reported in our earlier work showed significant increase in body stores of GSH and stabilized decreased glycated haemoglobin (HbA1c). Analysis of gut microbiome revealed that abundance of phylum Proteobacteria significantly decreased (P < 0.05) in individuals with GSH supplementation after 6 months compared to those without it. Beneficial dominant genera such as Megasphaera, Bacteroides, and Megamonas were found to be significantly enriched (P < 0.05), while pathogenic Escherichia/Shigella was found to be depleted (P < 0.05) after supplementation. Data clearly demonstrate that GSH supplementation along with antidiabetic treatment helps restore the gut microbiome by enriching beneficial bacteria of healthy gut and reducing significantly the load of pathogenic bacteria of diabetic gut.
Introduction
Several studies have demonstrated involvement of oxidative stress in developing insulin resistance, a cardinal feature of type 2 diabetes (T2D) and in development of diabetic complications due to persistent hyperglycaemia. T2D is a heterogeneous disease, and the aetiology of insulin resistance is multifactorial. Among the known drivers of insulin resistance are, over nutrition that triggers increased inflammation (Gregor and Hotamisligil 2011), changes in lipid metabolism (Samuel and Shulman 2012), oxidative stress (Tangvarasittichai 2015, Hurrle and Hsu 2017), and imbalance of gut microbiota (Kau et al. 2011, Nicholson et al. 2012), all of which are interconnected to varying degrees. Several studies have confirmed that diabetic individuals display significant increase in markers of oxidative stress and have gut dysbiosis (Fernandes et al. 2019, Gaike et al. 2020, Chen et al. 2021). Earlier studies carried out in our lab have clearly shown that an elevated level of oxidative stress is associated with reduced β-cell function in newly diagnosed T2D individuals and controlling hyperglycaemia with antidiabetic treatment leads to significant improvement in oxidative stress parameters (Acharya et al. 2014). Further, among all the parameters checked, glutathione (GSH) was found to be the best covariate of glucose recovery and responded rapidly with changes in glucose irrespective of antidiabetic treatment used (Kulkarni et al. 2014). Based on these observations, a clinical trial was conducted wherein GSH supplementation was given for 6 months to T2D individuals and their glycaemic parameters were checked. Results confirmed that GSH supplementation significantly increased body stores of GSH, stabilized reduced glycated haemoglobin (HbA1c), and decreased oxidative damage to DNA (Kalamkar et al. 2022). This effect was seen more prominently in elderly subgroup wherein additionally increase in fasting insulin was also seen. Since, gut microbiome is also known to be involved in T2D, we were curious to know whether beneficial effects of long-term GSH supplementation could be in part due to changes in gut microbiome of T2D individuals.
The gut microbiome is the entire collection of symbiotic microorganisms in the human gastrointestinal tract and their collective interacting genomes, which influence human physiology, metabolism, nutrition, and immune function (Shreiner et al. 2015). Our earlier studies have clearly demonstrated perturbations in gut microbiome of T2D individuals (Bhute et al. 2017). We also checked the gut microbiome changes in different states of diabetes, where we found that, it changed significantly in newly diagnosed diabetic patients along with reduced total antioxidant capacity (TAC) and recovered partially in patients on antidiabetic treatment compared to healthy ones (Gaike et al. 2020). Additionally, TAC was found to be positively associated with beneficial taxa such as Prevotella and Akkermansia. In this study, we have analysed the effect of long-term GSH supplementation on the gut microbiome of diabetic individuals.
Methods
Study population and sample collection
Subjects recruited in this study were from the project approved by the Institutional Ethical Committee of Jehangir Hospital Development Centre, Pune (JCDC ECN- ECR/352/Inst/NIH/2013); and Institutional Biosafety Committee of SPPU (Bot/27A/15), Pune. Total 37 diabetic individuals were recruited in this study. Out of which, 17 of them were on antidiabetic treatment (D group) alone, whereas 20 received oral GSH supplementation (500 mg/day) for 6 months in addition to antidiabetic treatment (DG group). Participants who were on any antibiotic consumption in last 3 months, or had any major gastrointestinal surgery or presence of any known chronic or clinical disorders were excluded from the study. Early morning stool sample was collected on Day 0 and 180 (at the end of 6 months) in a sterile stool container and stored at −80°C until further processing. Fasting blood sample was also collected on the same days by phlebotomists from Golwilkar Metropolis, Pune, India, for analysis of biochemical parameters. Samples collected on Day 0 are named as alpha and those collected on Day 180 as gamma.
Biochemical analysis
Fasting plasma glucose (FPG), post prandial glucose (PPG), fasting plasma insulin (FPI), and postprandial insulin (PPI), and HbA1c were analysed using automated analyser at Golwilkar Metropolis, Pune, following CLSI (Clinical and Laboratory Standards Institute, Malvern, PA, USA) guidelines. Reduced and total GSH content was estimated using GSH assay kit (Cayman Chemical, Ann Arbour, USA). The homeostasis model assessment of beta-cell function (HOMA-β) is a method employed to evaluate beta-cell function based on the levels of fasting glucose and insulin concentrations. The HOMA-β was calculated using formula: HOMA-β = 20 × fasting insulin (µU/ml)/[fasting plasma glucose (mg/dl)–63] (Yoon et al. 2016). The 8-hydroxy-2-deoxyguanosine (8-OHdG) content in the DNA was assessed using a competitive ELIZA (Modak et al. 2009).
DNA extraction and 16S rRNA gene amplicon sequencing
Total bacterial community DNA was extracted from 74 samples using QIAamp Stool DNA Mini Kit (Qiagen, Germany) as per the manufacturer’s instructions and eluted in 30 ul buffer for further use. DNA was quantitated using NanoDrop (Thermo Fisher Scientific, USA) and its quality was checked by gel electrophoresis. DNA samples were subjected to amplification of 16S rRNA gene using V4 region specific primers (V4 Forward 5′GTGCCAGCMGCCGCGGTAA3′ and V4 Reverse 5′GGACTACHVGGGTWTCTAAT3′). Polymerase chain reaction (PCR) was performed using following conditions: initial denaturation at 95°C for 3 minutes, followed by 25 cycles of denaturation at 95°C for 30 seconds, annealing at 55°C for 30 seconds, polymerization at 72°C for 30 seconds, and final extension at 72°C for 7 minutes. A volume of 2.5 μl DNA (5 ng/μl concentration) was used as a template for each 25 μl PCR reaction. After amplification, products were cleaned using AMPure XP beads (A63882, Beckman Coulter, Inc., USA) and subjected to library preparation using NextraXT Library preparation kit (Illumina, USA) followed by limited cycle PCR to enrich the adapter ligated DNA molecules. Final clean-up was performed using AMPure XP beads to obtain libraries, which were assessed for fragment size distribution using TapeStation (Agilent Technologies, USA) and were quantified using Qubit DNA (Thermo Fisher Scientific, USA) before sequencing. Libraries were clonally amplified on cBOT and sequenced using Illumina MiSeq (Illumina Inc., USA) with 2 × 250 bp paired end chemistry.
Bioinformatics analysis for identification of amplicon sequence variants of taxa
Paired end reads were analysed using DADA2 software package. The raw reads were trimmed to remove adapter sequences from both the ends. The quality filtered sequences were used for further analysis using same DADA2 pipeline v. 1.16, using default parameters, except for filterAndTrim for which parameters truncLen = c(220 200) and, trimLeft = c(19,20) were used. Amplicon sequence variants (ASVs) were cleaned by chimera removal step and ASV table was generated. ASVs were assigned to different taxonomic units using Silva database v138 and species level by add species function in DADA2. Representative sequences from each ASV were combined in phyloseq object. Sequences representing chloroplast, mitochondria, and archaea were removed from ASV table to retain only bacterial ASVs and were normalized for least number of sequences using rarefaction and used for downstream analysis.
Statistical analysis
All biochemical parameters, HbA1c, FPG, FPI, PPG, PPI, GSH, GSSG, HOMA-β and 8-OHdG were analysed using nonparametric pairwise comparisons using a Wilcoxon matched-pair signed-rank test to understand the comparison between two study groups. Alpha diversity and differentially abundant genera were analysed by mean difference with Wilcoxon matched-pair signed-rank tests using stat_compare_means function in the R package ggpubr (v0.2.5) (ggpubr.pdf). To evaluate differentially abundant supplementation-related biomarkers (ASVs), we performed core microbiome analysis to obtain ASVs prevalent in 80% subjects.
Alpha diversity analysis based on phylogenetic diversity, Simpson’s evenness (Overall), Simpson’s evenness for dominant taxa and beta-diversity (Generalized UniFrac, weighted and unweighted Unifrac) analyses were performed on rarefied data subsampled at 1742 reads per sample using R packages picante (v1.8.1), microbiome (v2.1.1), and MiSPU (v1.0) (Kembel et al. 2010, Lozupone et al. 2011, Chen et al. 2012, Wu et al. 2016, Lahti 2017).
Association between beta-diversity and study groups was assessed using a nonparametric analysis of similarities (ANOSIM, vegan R package) with 999 permutations.
The ‘envfit’ function from the ‘vegan’ package (available at https://cran.r-project.org/web/packages/vegan/index.html) was employed to assess the effect size and significance of each covariate. This involved comparing the variance in centroids among different groups in relation to the overall variation. To achieve this, ordination was carried out using nonmetric multidimensional scaling (NMDS) based on Bray–Curtis dissimilarity. The significance value was determined through 9999 permutations. All P-values resulting from the ‘envfit’ analysis were adjusted for multiple comparisons using the False Discovery Rate (FDR) adjustment method, specifically the Benjamini–Hochberg procedure. In total, 9 covariates with known associations to gut microbiome variation in D alpha, D gamma, DG alpha, and DG gamma were included in the envfit analysis. Specifically, we tested T2D-associated biochemical factors including FPG, PPG, FPI, PPI, HbA1c, GSSG, GSH, HOMA-β, and 8-OHdG.
Spearman’s correlation was performed to identify associations among biochemical parameters and microbial genera using R package Hmisc (https://cran.r-project.org/web/packages/Hmisc/index.html) and visualization was done using ggplot2 package in R.
Results
The study groups included control diabetic (D group) and diabetic individuals with GSH supplementation (DG group), with mean age of 57 and 54 years, respectively. D group consisted of 6 males and 12 females while DG group consisted of 9 males and 13 females.
Effect of GSH supplementation on blood biochemical parameters
Oral GSH supplementation for 6 months led to significant decrease in FPG (P < 0.05) and HbA1c (P = 0.0006) in DG gamma compared to DG alpha. However, such change was not observed for PPG, FPI, and PPI. All these parameters remained unchanged in D gamma compared to D alpha. Erythrocytic GSH as well as glutathione disulphide (GSSG) were found to be significantly high in DG gamma compared to DG alpha group (P < 0.05). While both GSH and GSSG did not change in D gamma compared to D alpha group. HOMA-β, an indicator of insulin secretion capacity exhibited a significant increase in DG gamma compared to DG alpha. However, no change was observed in the D gamma compared to D alpha groups (Fig. 1). An 8-OHdG, an oxidative DNA damage marker, was found to be significantly decreased in DG gamma compared to DG alpha, while it remained unchanged in D gamma compared to D alpha.
Longitudinal changes in concentration of serum biomarkers in different groups. The statistical comparison was done using paired Wilcoxon’s t-test between alpha and gamma time points of each group. Significance level is *** for P < 0.001, ** for P < 0.01, and * for P < 0.05.
Effect of long-term GSH supplementation on gut microbial diversity and structure
High-quality sequences were retained with average sequence reads of 110 492 ± 12 901.23 per sample. A total of 2086 ASVs were obtained and after removing singleton ASVs and rarefying the data 1742 ASVs were observed. Taxonomic assignment was done using 97% similarity cut off with Silva reference database v138 followed by species assignment.
To find out distribution of various taxa in gut microbiome across the study groups, i.e. alpha diversity, phylogenetic diversity (Faith’s phylogenetic diversity) was checked and Simpson evenness was calculated. Phylogenetic diversity was found to be decreasing post GSH supplementation in DG group though was not significant (Fig. 2A). Such difference was not observed for Simpson evenness for D or DG group (Fig. 2B). However, when Simpson diversity was analysed based on dominant taxa, it was found to be increased significantly in DG gamma compared to DG alpha (P = .021), whereas such change was not observed in D gamma compared to D alpha (Fig. 2C). Beta diversity was used to analyse changes in microbial diversity within the groups. Principal coordinates analysis was performed on the basis of the Unifrac distances using three measures, i.e. unweighted, generalized, and weighted UniFrac (Fig. 2D). Based on generalized UniFrac, though variation in bacterial community was observed, it was not significant in DG gamma compared to DG alpha. Such change was not observed in D gamma compared to D alpha group (Fig. 2D).
Alpha diversity analysis in four study groups. (A) Phylogenetic diversity, (B) Simpson evenness (overall), and (C) Simpson (dominance). The statistical comparison was done using paired Wilcoxon’s t-test between alpha and gamma time points of each group. Significance level is * for P < 0.05. (D) Comparison of beta-diversity between groups based on principal coordinates of UnWeighted Unifrac (Anosim, R = 0.02269, P = 0.123); Generalized UniFrac (Anosim, R = 0.02908, P = 0.082); and Weighted Unifrac distance (Anosim, R = 0.0259, P = 0.083), with centroids and bars representing standard error across the two axes.
Covariates influencing gut microbiome diversity were determined by assessing the relationship between continuous or categorical phenotypes and the community structure at the ASV level using the ‘envfit’ function within the R-package ‘vegan’. In the context of overall variations in the microbiome, it was found that among several covariates, GSH exhibited a significant impact on microbiome composition (Supplementary Fig. S1; Supplementary Table 1). This observation was corroborated through principal coordinate analysis visualization using Bray–NMDS metrics. It became apparent that each data point within the DG gamma group was remarkably affected by GSH levels compared to other groups (Supplementary Fig. S2). The D alpha, D gamma, and DG alpha were influenced by HbA1c and 8-OHdG.
Gut microbiota composition in different study groups
Initially phylum level distribution in D and DG groups was analysed (Fig. 3A). Phylum Firmicutes was found to be abundant in all the samples (47% in D alpha, 38% in D gamma, 40% in DG alpha, and 43% in DG gamma) followed by phylum Bacteroidetes with 23 and 37% in D alpha and gamma, respectively, and 19% and 32% in DG alpha and gamma, respectively. Similarly, phylum Proteobacteria was found to be the third most abundant (20% in D alpha, 18% in D gamma, 25% in DG alpha, and 15% in DG gamma) in all the groups. Abundance of another major phylum Actinobacteria was found to be 8% in D alpha, 6% in D gamma, 10% in DG alpha, and 7% in DG gamma. Whereas abundance of a beneficial phylum Verrucomicrobiota was found to be low (2% in D alpha, 1% in D gamma, 6% in DG alpha, and 3% in DG gamma) in all the groups. Among these phyla, significant decrease in abundance was seen only in case of Proteobacteria in DG gamma (P < 0.05) (Fig. 3B).
(A) Comparison of differentially abundant significant phyla among all study groups based on mean relative abundance. (B) Differential abundance of dominant phylum Proteobacteria. The statistical comparison was done using paired Wilcoxon’s t-test between alpha and gamma with FDR correction by the Bonferroni method. P < .05.
Effect of GSH supplementation on core microbial genera
We further analysed the core microbiome of D and DG groups at lower taxonomic level, i.e. at genus level with 50% prevalence and at least 1% abundance threshold among subjects. Genera Prevotella_9, Escherichia/Shigella, Bacteroides, Megasphaera, Lactobacillus, Bifidobacterium, and Dialister were found to be core taxa for both D and DG group.
In both D and DG groups, 240 genera were found after aggregating all the ASVs belonging to same genus. Among them, abundance of genus Bacteroides belonging to phylum Bacteroidetes and genera Megasphaera and Megamonas belonging to phylum Firmicutes was found to be increased significantly in DG gamma, which was not observed in D gamma group. Abundance of genus Escherichia/Shigella from phylum Proteobacteria decreased significantly (P < 0.001) in DG gamma and remained unchanged in D gamma (Fig. 4).
Comparison of differentially abundant dominant genera among all study groups based on relative abundance. The statistical comparison was done using paired Wilcoxon’s t-test between alpha and gamma with FDR correction by the Bonferroni method. P < 0.05.
Differential abundance of core ASVs prevalent in 80% individuals
The effect of GSH supplementation was analysed at individual level to better understand microbiome changes over a period of 6 months. Differential abundance of ASVs prevalent in 80% subjects was performed to check the differences in microbiome at lower taxonomic level (at ASV) between D and DG group (Fig. 5). In this analysis, we found 15 highly abundant core taxa in 80% subjects. Among these, ASV2 belonging to beneficial genus Megasphaera was significantly enriched (P < 0.05) and abundance of ASV1_Escherichia/Shigella was significantly decreased (P < 0.05) in 16 individuals on GSH supplementation in DG gamma compared to DG alpha, whereas these taxa remained unaltered in 4 individuals. Such a change was not observed in D gamma compared to D alpha (Fig. 5).
Comparison of differentially abundant dominant ASVs prevalent in 80% individuals among all study groups based on relative abundance. The statistical comparison was done using paired Wilcoxon’ t-test between alpha and gamma with FDR correction by the Bonferroni method. P < 0.05.
Gut microbiome and its association with serum biomarkers
Association of microbiome with serum biomarkers was analysed based on the highly abundant genera found in all the study groups. Among them, genus Megaphaera was negatively correlated with HbA1c (P < 0.05), Akkermansia was positively associated with GSH (P < 0.05), and Escherichia/Shigella was negatively associated with GSSG (P < 0.05). Another genus Subdoligranulum was positively associated with HbA1c and PP glucose and negatively with GSH (Fig. 6).
Spearman correlation analysis based on top abundant genera and serum biomarkers. Spearman correlation values are shown in the vertical heatmap panel to the right. P-values of < 0.05 are indicated by the plus symbols.
Discussion
This is a case-control study wherein oral supplementation of GSH was given to T2D individuals along with the antidiabetic treatment for 6 months and its effect on gut microbiome was checked. We have already reported that GSH supplementation in T2D individuals significantly increased body stores of GSH and stabilized HbA1c in population overall. This effect was seen more prominently in population above age of 55 years (Kalamkar et al. 2022). As expected, with increase in body stores of GSH significant decrease in 8-OHDG, marker for oxidative DNA damage, was observed among diabetic individuals on GSH supplementation (Kalamkar et al. 2022). Interestingly, we find that decrease in 8-OHdG was correlated with significant decrease in major phyla Proteobacteria (P < 0.05) after GSH supplementation. This, in turn, appears to contribute to the reduction of oxidative stress and enhancement of gut microbiota by promoting an increase in beneficial microorganisms while decreasing the presence of pathogenic taxa. Numerous studies have investigated the intricate relationships between gut bacteria, blood glucose regulation, and insulin control (Kovatcheva-Datchary et al. 2015, Bagarolli et al. 2017, Zhao et al. 2018). However, the precise influence of GSH on the gut microbiome and its correlation with HOMA-β remains underexplored in existing literature. Our study clearly demonstrates that after long-term GSH supplementation in T2D individuals increase in beta-cell function (HOMA-β) observed was correlated with the improved gut microbiome with increase in beneficial taxa such as ASV2 Megasphaera.
Our earlier studies have also indicated that there is a gut dysbiosis in T2D individuals and antidiabetic treatment improves microbial diversity to some extent with increasing abundance of rare taxa compared to healthy (Gaike et al. 2020) and reduces load of pathogenic bacteria, similar to that reported in different populations (Forslund et al. 2015, Bhute et al. 2017, Wu et al. 2017, Sun et al. 2018). Findings from current study suggest that though GSH supplementation does not significantly change the overall microbiome diversity compared to those without supplementation (Fig. 2A, B, and D), it significantly increases abundance of dominant taxa as evident from high evenness in Simpson dominance (Fig. 2B). This indicates that GSH supplementation does help alter dominant taxa and enrich beneficial bacteria such as Megasphaera and Bacteroides, and reduce pathogenic bacteria such as Escherichia.
Envfit analysis revealed that, among various parameters tested, GSH was an important covariate, which may have a role in influencing the microbiome diversity. In this study, where DG gamma group was clustered separately apart from other groups, indicating microbiome composition was indeed influenced by GSH in DG gamma on 6 month’s GSH supplementation. This change in microbiome composition may contribute to the maintenance of gut homeostasis.
Previous reports have demonstrated a significant decrease in Bacteroidetes concomitant with increase in Firmicutes in T2D individuals compared to nondiabetic healthy individuals (Larsen et al. 2010, Bhute et al. 2017, Gaike et al. 2020). When gut microbiome structure of T2D individuals on GSH supplementation was analysed at phylum level, we did not observe significant change in these dominant gut bacterial phyla compared to those without supplementation. Proteobacteria is shown to be one of the significant contributing phyla to the pathogenesis and dysbiosis of the gut in metabolic disorders (Rizzatti et al. 2017, Bodkhe et al. 2019). Significant decrease in abundance of phylum Proteobacteria was observed in individuals with GSH supplementation compared to those without it. This indicated that GSH supplementation does help reduce the load of pathogenic bacteria in diabetic gut.
Next, we investigated the effect of GSH supplementation on core and dominant taxa at the genus level. Interestingly, Bacteroides, Megasphaera, and Megamonas were found to be significantly increased in T2D individuals with GSH supplementation. Earlier study by Chaudhari et al. (2020) has demonstrated that decreased abundance of one of the dominant genus Bacteroides is associated with gut dysbiosis. Many other studies have also reported negative association of abundance of Bacteroides with T2D (Munukka et al. 2012, Zhang et al. 2013, Candela et al. 2016, Yamaguchi et al. 2016, Lippert et al. 2017). Administration of B. acidifaciens (Yang et al. 2017) and B. uniformis (Gauffin Cano et al. 2012) has shown to improve glucose intolerance and insulin resistance in diabetic mice, suggesting beneficial role of Bacteroides in glucose metabolism. Furthermore, B. acidifaciens was also found to be involved in fatty acid oxidation in the adipose tissue via TGR5-PPAR-α pathway (Yang et al. 2017) and help prevent obesity by improving insulin sensitivity in mice. Thus, increase in abundance of this genus found in T2D individuals on GSH supplementation may have helped improve glucose homeostasis in them.
Genus Megasphaera (Shetty et al. 2013) and few species of Megamonas (Chevrot et al. 2008, Sakon et al. 2008) are involved in producing beneficial metabolites such as short-chain fatty acids (SCFAs), which are known to alter glucose homeostasis in the host (Zhao et al. 2018). SCFAs such as butyrate, propionate, and acetate reduced expression of PPAR-γ, which resulted in increased fatty acid oxidation, and helped alleviate obesity induced T2D (den Besten et al. 2015). Dietary supplementation of butyrate was shown to improve insulin sensitivity through increased energy expenditure (Gao et al. 2009) in diabetic mice. Genus Megasphaera is known to harbour many beneficial species as they possess unique sets of genes encoding Carbohydrate-Active enZymes (CAZymes) important in dietary carbohydrate metabolism. The presence of CAZyme-encoding genes is also associated with production of SCFAs (Soverini et al. 2017). Operational taxonomic units (OTUs) related to Megamonas genus were found to be enriched in nondiabetic compared to T2D individuals in the Chinese population (Zhang et al. 2013). Thus, significant increase observed in Megasphaera, and Megamonas in individuals on GSH supplementation may have helped improve glucose homeostasis by producing SCFAs.
Several studies have reported an increase in the abundance of Escherichia genera in metabolic diseases, particularly in T2D (Qin et al. 2012, Karlsson et al. 2013, Bhute et al. 2017, Gaike et al. 2020). After GSH supplementation, Escherichia/Shigella decreased significantly in T2D individuals. It is known that Escherichia/Shigella are involved in low grade inflammation and insulin resistance due to lipopolysaccharide (LPS) production causing endotoxemia (Cani et al. 2007, Rizzatti et al. 2017). Thus, reduced abundance of these microbes is expected to help restore a healthy gut microbiome on long-term GSH supplementation.
We then examined each individual’s microbiome at the ASV level to better understand the effect of supplementation. Among 15 ASVs found prevalent in 80% of subjects, ASV2: Megasphaera and ASV1: Escherichia/Shigella were found to be significantly changed indicating that bacteria involved in dysbiosis decreased while those with beneficial effects on human gut increased on supplementation. Our data suggest that, in most of the individuals with GSH supplementation, microbial diversity improves with dominant and beneficial taxa compared to T2D individuals. We did not observe such change in few individuals that might be due to their low GSH concentration even after GSH supplementation.
Thus, improvement in glucose homeostasis observed in T2D individuals on GSH supplementation in part seems to be due to improving gut microbiome by enriching population of beneficial bacteria and decreasing that of pathogenic ones. Further metagenomic or metatranscriptomics studies must be done to validate and understand the species or strain level functional response to the supplementation.
Conclusion
Overall, the GSH supplementation in addition to antidiabetic treatment helped in altering dominant taxa in the T2D individuals, which subsequently leads to enrichment in beneficial bacteria and reduction in pathogenic bacteria in diabetic gut. GSH supplementation can be used as an adjuvant therapy for restoration of healthy gut.
Author contributions
P.G., Y.S.S., S.S.G. and A.H.G.: conceived and designed the experiments; S.D.K. and V.G.: sample collection; U.D. and S.K.I. helped in patient recruitment; A.H.G.: DNA extraction and next generation sequencing (NGS) experiment; A.H.G. and S.D.K.: biochemical parameter analysis; A.H.G.: bioinformatics and statistical analysis; A.H.G. and S.S.G.: interpretation; A.H.G., and S.S.G.: manuscript preparation; and A.H.G., S.D.K., and S.S.G.: manuscript and data analysis improvement. All authors read and approved the final manuscript.
Ethics approval
The study was approved by the Institutional Ethical Committee of Jehangir Hospital Development Centre, Pune (JCDC ECN- ECR/352/Inst/NIH/2013); and Institutional Biosafety Committee of SPPU (Bot/27A/15), Pune.
Acknowledgements
Author A.H.G. acknowledges the Indian Council of Medical Research (ICMR), Government of India, for senior research fellowship (ICMR-SRF). Author S.D.K. is a recipient of a Senior Research Fellowship from CSIR, India.We acknowledge Dr. Sudarshan Shetty for his help in R scripts.
Conflict of interest
None declared.
Funding
This work was financially supported by the UGC-XIIth Plan-Innovation Grant, SPPU, Pune. We also acknowledge partial financial support from the UGC-CAS and the DST-PURSE programme of the Department of Zoology, Savitribai Phule Pune University, Pune, India. This study was supported by Savitribai Phule Pune University, India, National Centre for Cell Science, Pune and IISER, Pune.
Data Availability
Raw sequencing data are available on NCBI Sequence Read Archive (SRA) submission under the Bioproject ID PRJNA898687.
