Introduction

The 7th edition of the International Diabetes Federation Atlas projects, approximately 515 million people with diabetes worldwide and India alone harbors more than 70 million diabetics. A study conducted to determine the prevalence of diabetes in India (ICMR-INDIAB Study) has shown that the prevalence of prediabetes and diabetes is alarmingly higher in both urban and rural areas and that the number is expected get higher and higher in the coming years [1]. Along with diabetes, overweight and obesity are also emerging as health burden worldwide and in developing countries like India. Due to new economic policy, India and other developing countries are undergoing rapid urbanization. Urbanization has led to rapid changes in lifestyle, decreased physical activity, and easy accessibility to food which contains high fat, sugar, and calorie. This might be one of the important factors for metabolic disorders including type 2 diabetes (T2DM). Epidemiological studies demonstrated that not only genetic factors, but environmental risk factors also influence the prevalence of diabetes and obesity in India [2]. While several omics studies are underway to dissect out the molecular pathogenesis of diabetes and obesity related disorders, gut microbiota has been proposed as an environmental factor, accountable for the development of diabetes and obesity.

Gut microbiota control intestinal permeability, which determines the threshold at which metabolic endotoxemia could induce the metabolic disorders [3]. Energy homeostasis and proinflammatory signals are key factors for the pathogenesis of diabetes, obesity, and associated metabolic disorders. The gut microbiota also plays major role in the regulation of energy homeostasis and inflammatory signatures. While modulating the gut microbiota could be a potential therapeutic target for the treatment of diabetes and obesity, the best way for modulating the gut flora balance in intestine can be achieved by probiotic treatment. Probiotics are live microorganisms which, when administered in adequate amounts confer a health benefit on the host [4]. Species of Lactobacillus and Bifidobacterium are most commonly used microbes as a probiotics. Among these probiotics, lactobacillus species have been claimed to confer health benefits on the management of diabetes and obesity [5, 6]. Recent clinical trial also demonstrated the efficacy of Lactobacillus gasseri treatment in improving the glucose homeostasis and decreasing the abdominal adiposity and body weight [7]. Selective increase in lactic acid bacteria/bifidobacteria supplement has been shown to improve the enteroendocrine L cell proliferation, thereby modulating the gut peptide production (glucagon-like peptide-1, peptide YY and ghrelin) [8, 9]. Type 2 diabetes patients treated with specific lactobacillus strains exhibited appropriate insulin secretion through augmented incretin release implying that oral ingestion of one specific strain may serve as a novel therapeutic approach to improve glucose-dependent insulin release [10].

While the scientific evidence of probiotics has been well studied worldwide, such studies are lacking in India. India being a country with potential microbial diversity, most of the probiotics benefits are claimed as either anectodal or with scarce scientific evidence. Recently, two probiotic strains viz., Lactobacillus plantarum MTCC5690 and L. fermentum MTCC5689 of Indian gut origin were isolated from the fecal samples by the National Dairy Research Institute, Karnal, India and characterized for their identity by 16S rRNA, PCR, RAPD, and whole-genome sequencing [11, 12] and deposited at the International repository under Budapest treaty at MTCC, IMTECH, Chandigarh, India. The efficacy of these two probiotic strains with medicinal properties was also reported previously [13–18]. Therefore, we evaluated the efficacy of these two potential probiotic strains against HFD-induced insulin-resistant diabetic mice model with a focus on disease biology and molecular investigations.

Materials and methods

Animal experiments

Adult (age 8–10 week) male C57BL/6J mice were housed in polypropylene cages in an animal room maintained with a 12-h light/dark cycle at 24 ± 2 °C and 55 ± 5 % humidity. Animals were allowed free access to water and respective diet throughout the study. The animals used in the present study were cared according to the principles and guidance of the Committee for the Purpose of Control and Supervision of Experiments on Animals (CPCSEA). All the protocols and procedures were approved by the Institutional Animal Ethics Committee (IAEC). Normal pellet diet (NPD) and high-fat diet (HFD-57 % kcal energy from fat source) were procured from the National Institute for Nutrition, Hyderabad. Fecal and urine sample were collected on every month with body weight and feed intake recorded periodically.

C57BL/6J mice were divided into seven groups with similar body weight. All lactobacillus probiotic cultures viz. LGG, MTCC 5690, and MTCC 5689 were obtained from ICAR-National Dairy Research Institute (NDRI), Karnal, India. The reference Lactobacillus strain, L. rhamnosus GG was used as the positive control for probiotic treatment. Our experimental study design included standard anti-diabetic drugs such as metformin and vildagliptin which were obtained as a gift sample from Orchid Research Laboratories, Chennai. The study groups therefore, comprised of the following: (1) Normal Pellet Diet (NPD), (2) High-Fat Diet (HFD), (3) HFD with LGG treatment (HFD-LGG), (4) HFD with MTCC 5690 treatment (HFD-MTCC 5690), (5) HFD with MTCC 5689 treatment (HFD-MTCC 5689), (6) HFD with metformin treatment (HFD-Met), and (7) HFD with vildagliptin treatment (HFD-Vilda). Each group comprised of six animals. All the dietary regimen and respective probiotic or drug treatment schedules were continued for 6 months. Dose of viable probiotic bacterial strains (LGG, MTCC 5690 and MTCC 5689) was approximately 1.5 × 109 colonies/mouse/day, p.o. Dose of the metformin and viladgliptin was 300 mg/kg and 1 mg/kg; p.o, respectively. PBS (10 mL/kg; p.o) vehicle treatment was given to NPD and HFD groups (which did not receive any probiotic/drug treatment). Detailed study plan and treatment schedule were presented in Table 1.

Table 1 Study design and treatment schedule
Full size table

Biochemical estimation

At the end of the study, all animals were fasted for 12 h. Blood was collected by retro-orbital plexus puncture and approximately 0.4 ml of blood collected in 10 % EDTA tubes. Plasma was separated upon centrifugation @ 3000 RPM for 15 min at 4 °C and stored at −80 °C until the assays performed. Biochemical analyses were carried out on a PerkinElmer auto-biochemical analyzer. Fasting plasma glucose (GOD-POD method), serum cholesterol (CHOD-PAP method), serum triglycerides (GPO-PAP method), and HDL cholesterol (direct method, polyethylene glycol-pretreated enzymes) were measured. Low-density lipoprotein (LDL) cholesterol was calculated using the Friedewald formula [19]. The HOMA-IR index (homeostasis model assessment) was calculated as follows: HOMA index = fasting glucose (mmol/L) × fasting insulin (mU/L)/22.5 [20]. Plasma insulin was estimated by ELISA (Mercodia kit, Sweden). Glycated hemoglobin (HbA1c) was estimated by high-pressure liquid chromatography using the Variant machine (Bio-Rad, Hercules, California, USA). LPS levels were determined using LAL (Limulus Amebocyte Lysate) assay reagent (Lonza, Walkersville, MD, USA) and expressed as EU/ml.

Insulin tolerance test (ITT)

While hyperglycemia is the hallmark of type 2 diabetes, it originates from insulin resistance. Insulin tolerance test (ITT) is the standard protocol that has routinely been used in preclinical studies to report either improvement or deterioration of insulin sensitivity. Therefore, we assessed the dynamic characteristics of blood glucose after short insulin injection by ITT. All animals were fasted for 5 h. Ultra-short action insulin was administered on abdominal cavity (Actarapid Nova Nordisk, Bangalore; Dose: 0.5 IU/kg; ip). Blood was collected by tail cut, and glucose levels measured by glucometer at 0, 15, 30, 60 and 120 min after insulin injection. Area under curve was calculated by Graph pad software. Glucose disposal rate (% KITT) was calculated by standard preclinical method [21].

Oral glucose tolerance test (OGTT)

At the end of 4-month treatment schedule, OGTT was performed in overnight-fasted (12–14 h) animals as per protocols described previously [22]. Glucose load (2 g/kg/10 mL, p.o) was given to all the animals. Blood was collected by tail cut method. Blood glucose was measured by one-touch glucometer (Accu-check glucometer, Life scan, Johnson & Johnson Ltd) at 0, 30, 60, 90, and 120 min after glucose challenge. Area under curve was calculated by Graph pad software.

In vivo Intestinal permeability assay

Gut permeability was assessed during the 5th month (i.e., around day 150 of the trial period). This measurement is based on the intestinal permeability of 4000 Da fluorescent dextran–FITC (DX-4000–FITC) in systemic circulation (DX-4000–FITC; FD4000; Sigma-Aldrich, St. Louis, Missouri, USA) [23]. Briefly, mice that had fasted for 6 h were administrated DX-4000–FITC through oral gavages (500 mg/kg body weight; 125 mg/ml). After 4 h of DX-4000–FITC administration, 100 µl of blood was collected from the tip of the tail vein and plasma was separated upon centrifugation. DX-4000-FITC concentration (which is a surrogate marker of intestinal permeability) was estimated in plasma by using fluorescence reader (Modulus Micro plate reader, Turner Biosystems, Sunneyvale, CA, USA) set at excitation 485 nm and emission 535 nm.

mRNA quantification by real-time PCR

Total RNA from gastrointestinal tract (GIT) or visceral fat tissue/skeletal muscle/liver of the study animals was isolated by using Illustra™ RNA spin Mini isolation Kit (GE Life sciences Ltd, USA). cDNA was synthesized as described previously [24]. Quantitative real-time PCR was performed for specific genes (β actin, occludin, Zonula occludens-1, TNFα, IL6, Adiponectin, FIAF, GRP78, PERK, IRE1-α, XBP1, CHOP, SREBP-1c, GLUT4, MCP-1, GCK, PEPCK, G6Pc, and FOXO1) using SYBR green master mix (Applied Biosystems). PCR amplification was carried out using ABI-7000 (Applied Biosystems) with cycle conditions (initial cycle: 50 °C for 2 min, initial denaturation 95 °C for 15 s, 40 cycles of denaturation 95 °C for 15 s, and annealing/extension of 60 °C for 1 min). Expression levels of genes were determined using \(2^{{ - \Delta \Delta c_{t} }}\) method; data normalized using β-actin (\(2^{{ - \Delta c_{t} }}\)) and a control condition (\(2^{{ - \Delta \Delta c_{t} }}\)). The primer sequences of specific genes probed in this study are listed in Table 2.

Table 2 Primer sequence of specific genes
Full size table

Glucagon-like peptide-1 (GLP-1) estimation

Animals were fasted for overnight. 400 µL of blood was collected in fasting stage to estimate fasting plasma glucose and insulin estimation. 2gm/kg oral glucose challenge was given at 0 min to all the animals. After 30 min, animals were bled and 500 µL of blood was collected into the tube containing 10 % EDTA and 30 µM dipeptidyl peptidase-4 (DPP4) inhibitors (obtained as gift sample from Glenmark Research center, Mumbai, India).This DPP-4 inhibitor prevents the GLP-1 degradation. Plasma was separated upon centrifugation @ 3000 RPM for 15 min at 4 °C and stored at −80 °C until the assays performed. Glucose-induced GLP-1 levels were estimated by using ELISA as per the manufacture instruction (USCN life sciences Inc. USA). Plasma glucose and insulin were also estimated after glucose challenge.

Statistical analysis

Experimental values are expressed as Mean ± SD or Mean ± SEM. Data were analyzed either by using Student’s t test or by using one-way ANOVA followed by Tukey’s multiple post hoc test. Differences were considered significant at p < 0.05. All analyses were done using GraphPad Prism and Windows-based SPSS statistical package (version 10.0, Chicago, IL).

Results

Preclinical characterization and Biochemical markers

Table 3 depicts the biochemical status of the study groups. Significant (p < 0.05) elevation of plasma glucose, total cholesterol, and triglyceride levels were observed in mice fed with high-fat diet (HFD) compared to normal pellet diet (NPD) group. All the glycemic and lipid parameters were significantly reduced in mice with probiotic treatments with the potential effects seen much more with MTCC 5689. The standard drugs metformin and vildagliptin as expected showed glucose and lipid profile lowering actions. While the mice fed with HFD exhibited hyperinsulinemic state as evident from the increased insulin levels, all the treatment arms significantly reduced hyperinsulinemia. The HbA1c test is one of the best methods to determine the previous glycemic episodes of the patients. In this study HFD feeding increases the HbA1c levels, compared to NPD (3.3 ± 0.2 vs 2.7 ± 0.1; p < 0.05). However, metformin and vildagliptin treatment as well as all probiotic treatments (LGG, MTCC 5690 and MTCC 5689) alleviated the HFD-induced elevation of HbA1c levels. Homeostatic model assessment (HOMA-IR) is a method used to quantify insulin resistance, which was significantly increased in HFD group compared with NPD group (4.6 ± 0.2 vs 28 ± 5). All probiotic strains and drug treatments were significantly reduced the elevated HOMA-IR implying amelioration of insulin resistance. Moreover, it was noted that MTCC 5689 pronounced more efficacious effects than other two probiotic strains. Six months of HFD feeding caused significant increase in body weight in mice compared to NPD-fed animals (46.2 ± 3.2 vs 35 ± 3.2; p < 0.05). MTCC 5689 (38.1 ± 1.5; p < 0.05) and metformin (36.2 ± 2.2; p < 0.05) significantly reduced the body weight and restored to normal. Other treatment regimen also showed nonsignificant body weight reductions in the mice. It is important to note that all probiotic/drug treatments reduced bodyweight along with improvements in glucose homeostasis without any significant changes in feed intake. It is interesting to note that, mice treated with the probiotic strains, viz., MTCC5690 and MTCC5689, had improved glycemic and lipid control along with reduction of hyperinsulinemia and HOMA-IR indices that is comparable to anti-diabetic drugs such as metformin and vildagliptin.

Table 3 Biochemical characteristics of the study animals (C57BL/6J mice) fed with High-Fat diet in the absence and presence of probiotics/drug treatments
Full size table

Assessment of insulin tolerance test (ITT)

In order to assess the insulin-resistant versus insulin sensitivity status of the study animals, we measured the dynamic characteristics of blood glucose after insulin injection by short insulin tolerance test (ITT). Figure 1a depicts the glucose levels at time points 0, 15, 30, 60, and 120 min after insulin injection in all the study groups. From this glucose disposal curves, the area under curve were plotted to calculate the percentage of glucose disposal rate (% KITT). Interestingly, the  % K ITT (percentage of glucose disposal rate) value decreased markedly in HFD group compared to animals fed with NPD diet (0.34 ± 0.06 vs 0.68 ± 0.14; p < 0.05) (Fig. 1b) implying that HFD-fed mice were insensitive to exogenous insulin, i.e., insulin resistance was established in peripheral tissues. All the probiotic and drug treatments significantly increased the % K ITT values implying a significant (p < 0.05) improvement in insulin sensitivity.

Fig. 1
figure 1

a Blood glucose values during insulin tolerance test (ITT) under the experimental conditions at different time points (all values are Mean ± SEM). b Percentage KITT (glucose disposal rate) values during insulin tolerance test (ITT) under the experimental conditions (all values are Mean ± SEM; # p < 0.05 compared to NPD; *p < 0.05 compared to HFD)

Full size image

Assessment of glucose intolerance

Oral glucose tolerance test (OGTT) was employed to determine the glycemic responses to exogenously administered glucose in all the groups at 4 months of drug or probiotic treatment. Figure 2a depicts the glucose levels at time points 0, 30, 60, 90, and 120 min during OGTT, and from this, the incremental AUCs (area under the curves) of plasma glucose were calculated for all the study groups and are presented in Fig. 2b. Mice fed with HFD exhibited significantly (p < 0.05) increased AUC of plasma glucose compared to NPD-fed animals (Fig. 2b), demonstrating that HFD-fed mice become glucose-intolerant and diabetic. HFD-fed animals treated with LGG, MTCC 5689, metformin, or vildagliptin showed significantly (p < 0.05) decreased AUC levels plasma glucose implying that mice fed with probiotics/drugs become more glucose-tolerant and resisted the genesis of diabetic state.

Fig. 2
figure 2

a Glucose excursion pattern measured during oral glucose tolerance test (OGTT) under the experimental condition at different time points (all values are Mean ± SEM). b Area under curve (AUC) calculated from blood glucose values measured at different time point (0–120 min) during OGTT under the experimental conditions. All values are Mean ± SEM; # p < 0.05 compared to NPD; *p < 0.05 compared to HFD

Full size image

Circulating GLP-1 level estimation

GLP-1 is one of the key incretin hormones, and it plays a major role in nutrient-induced insulin secretion from the β cells. Apart from this, GLP-1 also reduces gastric emptying thereby inducing satiety. In our study, glucose-induced GLP-1 levels were significantly decreased in animals fed with HFD compared to NPD-fed mice (0.6 ± 0.2 vs 3.2 ± 1 pg/mL, p < 0.05) (Fig. 3). All the probiotic treatments increased and normalized the levels of GLP-1 equal to the conditions seen in control animals. Interestingly HFD mice treated with either metformin or vildagliptin showed several fold increase in the glucose-induced GLP-1 levels.

Fig. 3
figure 3

Glucose (nutrient)-induced GLP-1 levels in plasma from the study groups. All values are Mean ± SEM; # p < 0.05 compared to NPD; *p < 0.05 compared to HFD

Full size image

Effect of probiotic treatment on Intestinal permeability in vivo

Gut integrity level was examined by permeability of fluorescent-tagged dye (DX-4000–FITC) toward systemic circulation from gastrointestinal tract. If gut integrity was damaged, movement of fluorescent-tagged dye concentration will be more in the systemic circulation indicating the increase in gut permeability. HFD-fed mice exhibited elevated plasma DX-4000–FITC concentration (0.16 ± 0.01 vs 0.27 ± 0.05; p < 0.05) as compared to NPD-fed group (Fig. 4) implying that HFD feeding is associated with loss of gut integrity and increase in gut permeability. LGG, MTCC 5690, and MTCC 5689 probiotic treatment groups showed significant decrease in plasma DX-4000–FITC levels indicating an association of gut integrity maintenance under these conditions. Standard drugs like metformin and vildagliptin treatment also showed marked decrease in gut permeability.

Fig. 4
figure 4

Gut permeability (gut leakage) measured as FITC-labeled fluorescent dye level in plasma under the different treatment conditions. All values are Mean ± SEM; # p < 0.05 compared to NPD; *p < 0.05 compared to HFD

Full size image

Alterations in intestinal tight junction markers

Since HFD-fed mice showed increased gut permeability state, we next determined the extent of transcriptional levels of intestinal tight junction markers in the study groups. Besides exhibiting increased gut permeability, HFD-fed animals also showed significantly (p < 0.05) reduced mRNA expression of epithelial tight junction markers such as occludin and ZO-1 (Fig. 5a, b). All the probiotic treated mice showed a significantly (p < 0.05) increased gene expression profile of the intestinal tight junction markers, occludin and ZO-1. Standard drugs like metformin and vildagliptin also improved the transcriptional levels of intestinal tight junction markers.

Fig. 5
figure 5

Transcriptional levels of tight junction markers in the small intestine from the study groups. a mRNA expression of occludin, b mRNA expression of Zonula Occludens-1-(ZO-1). All values are Mean ± SEM; # p < 0.05 compared to NPD; *p < 0.05 compared to HFD

Full size image

Reduction of endotoxemia under treatment with probiotics/drugs

Compared to mice fed with NPD, HFD-fed mice showed significantly (p < 0.05) elevated serum levels of LPS (Fig. 6). All the treatments significantly (p < 0.05) decreased the LPS levels and the best reduction of endotoxemia was observed with MTCC5689 and LGG. This clearly implies that reduction of endotoxemia and restoration of intestinal barrier integrity by the probiotics treatment are tightly linked to the improvement in intestinal permeability.

Fig. 6
figure 6

Serum levels of LPS from the study groups. All values are Mean ± SEM, # p < 0.05 compared to NPD and *p < 0.05 compared to HFD

Full size image

Effect of probiotic treatment on gene expression profiles in visceral fat

Since HFD consumption is linked to glucose/lipid dyshomeostasis and proinflammation and there exists gut-adipose axis regulatory networks, we next evaluated certain gene expression markers from the visceral adipose tissue from the study groups. There was a several fold increase (p < 0.05) in IL6 and TNFα gene expression in the visceral fat of HFD-fed mice compared to control animals (Fig. 7a, b). However, animals treated with probiotics or drugs significantly (p < 0.05) reduced the HFD-induced increase in proinflammatory gene expression profiles. While FIAF gene expression was significantly (p < 0.05) downregulated in visceral fat of animals fed with HFD diet compared to NPD diet (Fig. 7c), this was significantly (p < 0.05) improved in the presence of probiotics/drugs. We have also examined the adiponectin gene expression level in the visceral adipose tissue from all the study groups. It was found that adiponectin gene expression was significantly (p < 0.05) downregulated in visceral fat of animals fed with HFD compared to NPD (Fig. 7d). Except for LGG, mice treated with other probiotics/drugs showed significantly (p < 0.05) increased transcriptional levels of adiponectin in visceral fat tissue.

Fig. 7
figure 7

Transcriptional levels of certain markers in visceral adipose tissues from study groups. a mRNA expression of IL6, b mRNA expression of TNFα, c mRNA expression of FIAF. d mRNA expression of adiponectin. All values are Mean ± SEM; # p < 0.05 compared to NPD; *p < 0.05 compared to HFD

Full size image

Effect of probiotic treatment on gene expression profiles of ER stress markers in skeletal muscle

Since endoplasmic reticulum (ER) stress pathway is emerging as a drug target for type 2 diabetes, we have checked the gene expression patterns of ER stress markers in skeletal muscle from the mice. Compared to control animals, skeletal muscle from mice fed with HFD showed significantly (p < 0.05) elevated transcriptional levels of ER stress markers viz., GRP78 (Fig. 8a), PERK (Fig. 8b), IRE1α (Fig. 8c), XBP1 (Fig. 8d), and CHOP (Fig. 8e). Interestingly, all probiotic/drug treatments considerably restored the ER stress markers to near normal levels. Additional experiments (data not shown) also revealed an increasing trend of lipogenic marker viz., SREBP-1c and decreasing trend of GLUT-4 mRNA levels in skeletal muscle from mice fed with HFD compared to control animals. Again, probiotic/drug treatments restored the SREBP-1c and GLUT4 transcriptional levels albeit with statistical insignificance.

Fig. 8
figure 8

Transcriptional levels of ER stress markers in skeletal muscle tissue from the study groups. a mRNA expression of GRP78, b mRNA expression of PERK, c mRNA expression of IRE1-α, d mRNA expression of XBP1, e mRNA expression of CHOP. All values are Mean ± SEM; # p < 0.05 compared to NPD; *p < 0.05 compared to HFD

Full size image

Effect of probiotic treatment on gene expression profiles of proinflammation, lipogenesis, and gluconeogenesis in liver

Compared to control animals, liver from the HFD-fed diabetic mice showed significantly (p < 0.05) increased expression of proinflammatory genes such as TNFα (Fig. 9a), IL-6 (Fig. 9b) and MCP-1 (Fig. 9c), as well as increased mRNA expression of GCK (Fig. 9d) and SREBP1c (Fig. 9e). More importantly, liver from the HFD-fed diabetic mice showed significantly (p < 0.05) increased expression of key gluconeogenic genes such as phosphoenol pyruvate carboxykinase (PEPCK) (Fig. 9f) and glucose 6 phosphatase catalytic subunit (G6Pc) (Fig. 9g), along with increased expression of FOXO1 (Fig. 9h), the predominant transcription factor connected with hepatic gluconeogenesis. Interestingly, transcription of all these genes were significantly (p < 0.05) reduced and normalized under the treatment of probiotics as well as anti-diabetic agents.

Fig. 9
figure 9

Transcriptional levels of proinflammation, lipogenesis and gluconeogenesis in liver from the study groups. mRNA expression of TNFα (a), IL-6 (b), MCP-1 (c), GCK (d), SREBP1 (e), PEPCK (f), G6Pc (g), and FOXO1 (h). All values are Mean ± SEM; # p < 0.05 compared to NPD; *p < 0.05 compared to HFD

Full size image

Discussion

Although there is hope and hype in the usage of probiotics as nutritional supplements and therapeutic modality for many disease states including type 2 diabetes, there is lack of studies that looked into the scientific basis of probiotics with a focus on disease biology and mode of action benefits. Therefore, our study is significant in reporting the following findings: (a) HFD-fed diabetic mice were characterized to exhibit decreased levels of GLP-1, increased gut permeability and circulatory LPS levels and decreased gene expression patterns of intestinal tight junction markers (viz., occludin and ZO-1). (b) Diabetic mice showed increased proinflammatory gene markers (TNFα and IL6) in visceral fat accompanied by decreased mRNA expression of FIAF and adiponectin and increased transcriptional levels of ER stress makers in the skeletal muscle as well as increased gene expression profiles of gluconeogensis in liver. (c) Indigenous probiotic interventions (most prominently the native MTCC 5689 strain) resisted insulin resistance and development of diabetes in mice under HFD feeding and beneficially modulated all the biochemical and molecular alterations both at the systemic and tissue levels. (d) Interestingly all the metabolic benefits seen with the probiotics strains are more or less comparable to that of standard anti-diabetic drugs such as metformin and vildagliptin.

Intestinal flora has been recently proposed as an environmental factor involved in the control of lipids, glucose, body weight, and energy homeostasis. Food and Agriculture Organization/World Health Organization (FAO/WHO) suggested that use of probiotics/prebiotics (functional food) can result in metabolic benefits through modulation of gut microbiota specifically lactobacilli and bifidobacteria [4]. In our study, indigenous probiotic strains (MTCC 5690 and MTCC 5689) improved glucose tolerance and resisted insulin resistance and type 2 diabetes in mice fed with high-fat diet. Preclinical biochemical investigations revealed that these probiotics were able to reduce glucose and lipid levels along with reduction in body weight gain. Our observations and findings correlated well with previous studies [3, 8, 25, 26].

It appears that the beneficial effect of MTCC 5690 and MTCC 5689 in our study could be mediated by mechanism(s) that could improve the gut integrity, decrease systemic levels of LPS (and thereby proinflammation), increase GLP-1, decrease ER stress, and subsequently improve peripheral insulin sensitivity. Supplementation of certain probiotic strains have been shown to beneficially modulate glucose and lipid homeostasis [27]. Previous studies demonstrated that, LPS triggers immune system and proinflammatory cytokines. Prolonged HFD feeding favors the gut environment toward gram-negative bacteria abundance and elevated LPS levels [28, 29]. Normally LPS (endotoxin) leakage was controlled by gut integrity, and gut integrity was maintained by tight junction proteins (TJPs such as occludin and zonula occludens-1). TJPs prevent the translocation of LPS into the systemic circulation. Studies by Cani et al. [8, 30] reported that HFD feeding decreased the TJP expression and resulted in intestinal permeability (leaky gut) as well as elevated LPS in systemic circulation. Four weeks of subcutaneous infusion of LPS resulted in fasting hyperglycemia, hyperinsulinemia, and body weight gain in rodent model, and these metabolic abnormalities resembled those in HFD-fed mice [3, 30]. Since endotoxemia (increased circulatory levels of lipopolysaccharide) has been identified as a triggering factor of insulin resistance in mice, the suppression of endotoxemia by probiotic supplementation is considered as a protective mechanism [31]. In one of our pioneering clinical studies, we have already shown association of increased circulatory LPS levels in patients with type 2 diabetes [32]. Significant reduction of circulatory LPS levels in our present study accompanied by maintenance of gut permeability along with increased expression of intestinal tight junction proteins (occludin and ZO-1) under the probiotic treatment implies that the anti-inflammatory benefits and improvement in peripheral insulin sensitivity might have originated from reduction of endotoxemia and associated signaling cascade. An oral administration of Lactobacillus casei Shirota was able to enhance the expression of plasma lipopolysaccharide-binding protein (LBP) and consequently reduced endotoxemia in murine models of obesity and T2DM [33]. In another study, the consumption of the probiotic strain, Bifidobacteriumanimalis subsp. Lactis 420, suppressed the bacterial translocation process from intestine to tissues, which might lead to metabolic bacteremia in the early onset of T2DM [34]. Together, these finding emphasize that gut microbiota plays a major role in regulating gut integrity, reducing the circulatory levels of LPS and curtailing proinflammation at several sites. While our study demonstrated the mechanistic action of indigenous probiotic strains, further metagenomics work is needed to document the beneficial gut microbiota alterations in such preclinical and clinical studies.

One of the important findings in our study is that all the probiotic/drug treatments increased the nutrient-induced circulatory levels of GLP-1. While vildagliptin regulation of GLP-1 is a direct mechanism [35] and metformin has been shown to increase GLP-1 levels [36], similar effects by the probiotics in our study imply that regulation of GLP-1 by probiotics could be one of the mechanisms of glucose homeostasis regulation. While L cell is responsible for the secretion of GLP-1 [37] and related gut hormones, it has been earlier claimed that probiotics or prebiotics might increase the L cells abundance in the gastrointestinal tract [8]. In a recent study, combining probiotics and/or prebiotics with anti-diabetic drugs (such as metformin and sitagliptin) has been shown to improve glycemic control and insulin sensitivity in mice via beneficial incretin modulation [38]. Based on the literature, we speculate that L cell activation could be linked to beneficial alterations in short-chain fatty acids (SCFAs). Probiotic treatments have been shown to alter a diverse range of pathways outcomes, including alterations SCFAs. While SCFAs (acetate, propionate, and butyrate) are the fermented products of fibers by intestinal microbiota, experimental evidence exists in that GLP-1 level was improved by SCFA-mediated L cell activation [39]. SCFAs upregulate the intestinal proglucagon (a precursor for GLP-1) mRNA expression and peptide-Y (PYY) and act as a ligand for many G-protein-coupled receptors (GPR43). SCFA has been shown to stimulate GPR43 and thereby improves the proliferation and activation of enteroendocrine L cells [40]. While investigating the effect of L. fermentum on representative microbial populations and overall metabolic activity of the human intestinal microbiota using a three-stage continuous culture system, Pereira et al. [41] have demonstrated increased levels of butyrate. Recent metagenomics studies on type 2 diabetes revealed that patients with type 2 diabetes exhibited altered gut microbiota diversity with loss of butyrate-producing bacteria [42, 43]. Connecting all the above, it appears that future studies are warranted to delineate whether the indigenous probiotic strains used in our study could beneficially alter gut microbiota as well as the levels of SCFAs including butyrate.

Proteins secreted from adipose tissue are increasingly recognized to play an important role in the regulation of glucose and lipid metabolism. Suppression of fasting-induced adipocyte factor (FIAF) gene expression in adipose tissue from the HFD-fed mice and upregulation of FIAF by probiotics/drugs in our study is an important observation. Using transgenic mice that mildly over express FIAF in peripheral tissues, Mandard et al. [44] showed that FIAF is an extremely powerful regulator of lipid metabolism and adiposity. In our study, systemic triglyceride levels were significantly decreased under the treatment of probiotics/drugs in association with upregulation of FIAF in visceral fat tissue. In this context, it is important to note that FIAF has been shown to antagonize the activity of lipoprotein lipase (LPL), thereby preventing the storage of triglycerides as fat [45]. Moreover, Bäckhed et al. [46] have demonstrated a mice model in which the gut microbiota suppresses FIAF expression in response host sensitivity to over nutrition, thereby increasing LPL activity and ultimately fat deposition in adipocytes. In another study [47], mice supplemented with Lactobacillus paracasei ssp. paracasei F19 showed reduction of fat storage and higher levels of circulating FIAF even under a high-fat diet. The same alterations in FIAF expression, but to a lesser extent, were observed in co-culture of colonic cell lines with Bifidobacteriumanimalis subsp. lactis Bb12 [47]. In our study, visceral fat from mice fed with HFD showed reduced mRNA expression of adiponectin, and this was restored to normal levels by the potential probiotics. Kim et al. [48] have also reported probiotics-mediated improvement in insulin sensitivity and reduction of adiposity in high-fat diet-fed mice through enhancement of adiponectin production. Therefore, future studies are warranted to delineate the mechanism(s) by which gut microbiota alterations and/or probiotic interventions could beneficially modulate FIAF and adiponectin signals and thereby regulate lipid and glucose homeostasis.

It is now well known that ER stress plays an important role in the progression of insulin resistance in a tissue-specific manner [49]. In one of our recent clinical studies, we have also demonstrated association of increased ER stress markers and proinflammation in patients with type 2 diabetes [50]. Therefore, increased ER stress markers in the skeletal muscle of mice fed with HFD and the amelioration of ER stress by probiotics in our study is an important observation. Previously, Park et al. [51] have reported that Lactobacillus rhamnosus GG improved glucose tolerance through alleviating ER stress in skeletal muscle and suppressed macrophage activation in db/db mice. Further studies are warranted to clarify whether probiotics could function as ‘chaperones’ and offer insulin-sensitizing effects. Recent studies imply a dynamic crosstalk between intestinal microbiota and multiple organs dysfunction in the genesis of obesity and diabetes [52]. Consistent with this, our study also reports increased gene expression of proinflammation, lipogenesis, and gluconeogenesis in liver from the HFD-fed diabetic mice and all of these transcriptional alterations were significantly reduced under the probiotics treatment. A study by Yoo et al. [53] also revealed that certain strain-specific probiotics could beneficially act against metabolic disorders. While our study exposed the several of the tissue-specific beneficial alterations of indigenous probiotic strains, it is suggested that advancements in such studies would unravel novel opportunities for next-generation probiotics targeting type 2 diabetes and other associated metabolic disorders.

To conclude, HFD-fed mice in our study were demonstrated to be insulin resistant, glucose intolerant, dyslipedimic, and obese, and are characterized to exhibit decreased levels of GLP-1, increased circulatory LPS levels and gut permeability, decrease in the gene expression patterns of intestinal tight junction markers (occludin and ZO-1), increased proinflammatory gene markers (TNFα and IL6) in visceral fat accompanied by decreased mRNA expression of FIAF and adiponectin and increased transcriptional levels of ER stress makers in the skeletal muscle as well as increased gene expression profiles of gluconeogensis in liver. Probiotic interventions (most prominently the indigenous MTCC 5689 strain) resisted insulin resistance and development of diabetes in mice under HFD feeding and beneficially modulated all the biochemical and molecular alterations in a mechanistic way both at systemic and tissue levels. Further metagenomics and metabolomics studies would unravel and extend the therapeutic benefits of probiotics. Since several recent studies on probiotics also reported improved glucose and glycemic factors in healthy volunteers and patients with diabetes [54–58], it appears that the probiotic strain MTCC 5689 on further investigations could be a promising candidate to enter into human clinical trials.