Limosilactobacillus fermentum MG4295 Improves Hyperglycemia in High-Fat Diet-Induced Mice

Hyperglycemia due to uncontrolled glucose regulation is widely known as cause of diabetes, non-alcoholic fatty liver disease (NAFLD), and other complications. NAFLD refers to a condition in which fat is excessively accumulated, whether inflamed or not, and has caused serious medical problems in recent years. The aim of this study was to explore the antihyperglycemia effects of Limosilactobacillus fermentum MG4295 (L. fermentum MG4295) in high-fat diet (HFD)-induced in vivo. We demonstrated the suitability of L. fermentum MG4295 as a probiotic by observing its stability, survivability, and proliferation under simulated gastrointestinal conditions, and safety, antibiotic susceptibility, hemolysis, and enzyme activity. The potential antihyperglycemic activity of L. fermentum MG4295 was investigated in an HFD and sugar-water-induced mouse model. Administration of this strain for 12 weeks showed an improved trend in glucose tolerance, insulin, alanine amino transferase, total cholesterol, low-density lipoprotein cholesterol, and glucagon-like peptide-1. Histopathological analysis revealed that L. fermentum MG4295 significantly reduced the histopathological scores of hepatic steatosis, inflammation, and hepatocellular hypertrophy in liver tissues and lipid content in adipose tissues. Administration of L. fermentum MG4295 upregulated IRS-1, AKT, and GLUT4 and downregulated G6Pc and PEPCK expression in liver and/or muscle tissues. Our results suggest that L. fermentum MG4295 can improve hyperglycemia. Furthermore, it can be used as a dietary functional supplement to manage blood glucose.


Introduction
Hyperglycemia includes n several mechanisms of insulin resistance, enhanced glucose production, and high glucose loads [1]. These mechanisms have been reported to be closely related to diabetes mellitus (DM) and NAFLD [1,2]. DM is a complex chronic metabolic disease associated with type 1 diabetes (T1DM; juvenile-onset diabetes mellitus and insufficient insulin production) and type 2 diabetes (T2DM; insulin resistance) [3]. DM may cause damage to the blood vessels and various organs, such as kidneys and eyes [4,5]. NAFLD is a sign of obesity-related metabolic syndrome. If T2DM is associated with NAFLD, it will become very complex and make diabetes management more difficult. DM appears to promote the development of NAFLD and elevate the risk of cirrhosis and liver cancer [6]. Although antidiabetic agents have been developed to date, most drugs for the T2DM display side effects, including hyperglycemia, leukopenia, sore throat, hemolytic anemia, indigestion, ketoneuria, or lactic acidosis [7,8]. Eventually, patients are at risk of life-threatening challenges, high medical costs, or lowered quality of life [9]. Therefore, the development of safe treatment materials without side effects and drug resistance is required. T2DM causes susceptibility to NAFLD and advancement through insulin resistance and hyperglycemia [10].
Accumulating evidence from clinical trials and animal studies has revealed that T2DM and NAFLD are closely related to gut microbiota [11][12][13]. Environmental factors, such

L. fermentum MG4295 Cultivation
The probiotic strain L. fermentum MG4295, which was isolated from human, was provided by MEDIOGEN Co., Ltd. (Jecheon, Korea). L. fermentum MG4295 was cultivated in de Man, Rogosa, and Sharpe broth (MRS, Difco, Detroit, MI, USA) at 37 • C for 15 h. The GenBank accession number for the 16S rRNA gene sequence of strain MG4295 is MW404503 as assessed in the National Center of Biotechnology Information (NCBI).

Evaluation of Food Safety of the Strain L. fermentum MG4295
To evaluate food safety of the strain L. fermentum MG4295, the following tests were performed.

Survival Rate under Simulating Gastrointestinal Tract (GIT)
To simulate gastric juice, acidic pH conditions were performed by Mainville et al. methods [28]. Concisely, the harvested cells were centrifuged for 5 min (3750× g at 4 • C) after culture for overnight and discard supernatant. Next, pellets were washed with PBS (pH 7.0) and adjusted to 10 8 CFU/mL. To test for resistance to pepsin and pancreatin, L. fermentum MG4295 was suspended in simulated gastric fluid (pH 2, 3, and 4 with 1 N HCl) and simulated intestinal fluid (pH 7 and 8 with 1 N NaOH). The cell suspensions were cultivated at 37 • C at 3-4 h. The survival rate of strain in each condition was evaluated by observing viable cells. The strain viability was decided by counting colonies on MRS agar plate using colony-forming units per ml (CFU/mL).

Auto-Aggregation
To investigate the capability of L. fermentum MG4295 to adhere to intestinal cells indirectly, an auto-aggregation was conducted according to Kos et al. [29]. The cell culture of the strain was inoculated (2%, v/v) in MRS broth and cultivated at 37 • C for overnight. Then, pellets were obtained by centrifuge (4000× g, 4 • C, and 15 min). After resuspending L. fermentum MG4295 in buffered saline (pH 7.0) to a final concentration of 1.0 at OD 600 , 4-mL aliquots (0.1 mL of L. fermentum MG4295 resuspension with 3.9 mL PBS) were taken and vortexed for 10 s. Auto-aggregation (%) was calculated as the rate of change according to the time (5 h at one-hour intervals) compared to the initial number of viable cells.

Antibiotic Susceptibility
Antibiotic susceptibility of L. fermentum MG4295 was demonstrated by the minimum inhibitory concentration (MIC) strip (Liofilchem, Via Scozia, Abruzzi TE, Italy). The antibiotic susceptibility of the strain to ampicillin, chloramphenicol, clindamycin, erythromycin, gentamicin, kanamycin, streptomycin, and tetracycline was tested according to the European Food Safety Authority (EFSA) guidelines [30]. The strain was cultivated at 37 • C for 18 h in MRS medium. Then, cell pellets were obtained by centrifuge (3750× g, 4 • C, 5 min) and resuspended in PBS (pH 7.4) to a McFarland standard of 0.5. The suspended strain was inoculated on BHI agar. Next, the plates with MIC test strips were incubated at 37 • C, and growth inhibition was confirmed after 48 h. The cut-off values for different antibiotics were measured by the EFSA guidelines [30].

Hemolytic Activity
For the hemolysis test, L. fermentum MG4295 was plated on tryptic soy agar (Difco, Detroit, MI, USA) containing 5% sheep blood (MB cell, Seoul, Korea) and incubated for 18 h at 37 • C in a CO 2 incubator [31]. After incubation, the plates were observed for the product of a greenish zone (α-hemolysis) and no such zone (γ-hemolysis) around the colonies.

Enzyme Production
To evaluate enzyme activity, the L. fermentum MG4295 was estimated by API ZYM system with colonies as described by the manufacturer's manual (bioMérieux, Marcyl'Étoile, France). The enzyme production was confirmed to the strong degree of color from 0 (no activity) to 5 (≥40 nM of product released) at 10 nM intervals.

Sample Preparation and Animals Experimental Design
To prepare the MG4295 sample, the harvested bacterial cell pellet (obtained after centrifugation at 4000× g, 10 min, 4 • C) was combined with lyoprotectant cocktail and then freeze-dried [32].
Seven-week-old male C57BL/6 mice were obtained from DooYeol Biotech (Seoul, Korea). All mice were caged under climate-controlled conditions (55 ± 5% humidity, 23 ± 2 • C, and 12-h day/night cycle), fed normal diet (ND) chow (14% fat, 21% protein, and 64% carbohydrate), and supplied with water ad libitum. The animal experiments were tested by the Food and Nutrition Laboratory (FNLab; No. 20-FNL-504), and every exertion was made to minimize pain of the mice. The mice were distributed into three groups at random (n = 8 per group; ND, high-fat diet (HFD) and HFD plus L. fermentum MG4295). The HFD group was fed chow containing 60% fat and drinking water (55% fructose + 45% sucrose, 42 g/L). In the HFD plus L. fermentum MG4295 group, probiotic powder was prepared in sterilized PBS (pH 7.4) and orally administered (1 × 10 9 CFU/mouse/day) for 12 weeks. At the end of the study, all animals were food deprived for 12 h and anesthetized by CO 2 inhalation and euthanized by cardiac puncture. Liver, muscle, and adipose tissues were rapidly excised and fixed in formalin or stored at−80 • C until analysis.

Evaluation of Weight of Body and Tissue
Weight of body was measured each 12 weeks. Diets were fed to mice simultaneously. The liver and adipose tissues were dissected, washed with PBS, and weighed. The dissected tissues were stored in a deep-freezer until further experiments.

Measurement of Oral Glucose Tolerance Test (OGTT)
The animals were subjected to an OGTT during the week prior to euthanasia. Briefly, all mice were fasted for 12 h and administered glucose (2 g/kg body weight) by oral route; blood glucose levels were measured at 0 and 15, 30, 60, 90, and 120 min after administration of the glucose load. Area under the curve (AUC) of the BGL was analyzed to glucose tolerance.

Histopathological Examination and Diameter of Adipocytes
The effect of L. fermentum MG4295 on steatosis was measured by histopathological changes in the tissues of liver and/or adipose. The tissues were fastened in 10% formalin for at least 24 h, and subsequently, 4-µm liver and adipose tissue sections were obtained and were stained with hematoxylin and eosin (H&E). Histopathological changes in the liver tissues were expressed by the summation of four grades as follows: 0 = normal, 1 = minimal, 2 = moderate, and 3 = severe.

RNA Extraction and Quantitative Real-Time Polymerase Chain Reaction (qRT-PCR)
Total RNA in liver or muscle tissue was extracted by NucleoZOL (MACHEREY-NAGEL, Duren, Germany), and cDNA was synthesized using first strand Reverse Transcription Premix kit (Elpis Biotech Co. Ltd., Daejeon, Korea) according to the manufacturer's instructions. qPCR amplification was executed by CFX ConnectTM Real-Time System (Bio-Rad, Hercules, CA, USA) using iQTM SYBR ® Green Supermix (Bio-Rad) with cycle conditions as follows: pre-denaturation 95 • C for 3 min, denaturation 95 • C for 10 s, annealing 50-65 • C for 30 s, followed by 39 amplification cycles; melt curve was 65 to 95 • C for 0.05 s. The target genes were insulin receptor substrate-1 (IRS-1), protein kinase B (AKT), and glucose transporter-4 (GLUT4) for insulin signaling pathways and glucose-6-phosphatase catalytic subunit (G6Pc) and phosphoenolpyruvate carboxykinase (PEPCK) for gluconeogenesis. The primer sequences used for expression analyses of the mouse genes primers are indicated (Table S1). The target genes expression was normalized to GAPDH and calculated using the 2 − Ct method [24,33,34].

Statistical Analysis
All data are expressed as mean ± standard deviation (SD) of triplicate determinations. The statistical significance of differences was calculated using a one-way analysis of variance (ANOVA) using Duncan's multiple range test, and p < 0.05 was regarded to be statistically significant using the SPSS (Statistical Package for the Social Sciences, ver. 21.0, Chicago, IL, USA). L. fermentum MG4295 was exposed to simulated gastric fluid and intestinal conditions. Simulated gastrointestinal conditions showed cell viability from 4.5 ± 0.03 to 7.6 ± 0.04 log CFU/mL (Table 1). These results indicate that L. fermentum MG4295 can survive in gas- trointestinal conditions of the host body. Among the pH conditions, L. fermentum MG4295 showed a lower growth rate in the pH 2 condition than in other pH conditions. However, there is a need to improve the survivability by performing culture optimization or applying cell-coating technology. The adhesion ability of the strain MG4295 was 58.0 ± 11.4% of the aggregation ability at 5 h.

Antibiotic Susceptibility and Hemolysis
To demonstrate food safety, L. fermentum MG4295 was investigated for antibiotic resistance using the MIC and hemolytic tests ( Table 2). The strain was within the epidemiological cut-off values suggested by the EFSA (2018). The strain showed neither alpha nor beta hemolysis. However, it showed gamma hemolysis activity as a negative response (data not shown).

Weight of Body and Tissues
To evaluate the change in weight, we measured the body weight of mice and tissue weight of the liver and adipose tissue ( Figure 1A; Table S2). The HFD group exhibited significantly higher body, liver, and adipose tissue weights than the ND group (p < 0.05) ( Figure 1A,C; Table S2). The L. fermentum MG4295-treated group did not show a change in the weight of adipose tissues compared to HFD group (Table S2). Although there were no significant differences, the L. fermentum MG4295-treated group showed reduced body weight (14.3%) and liver weight (13.9%) than HFD group ( Figure 1B,C).

Weight of Body and Tissues
To evaluate the change in weight, we measured the body weight of mice and tissue weight of the liver and adipose tissue ( Figure 1A; Table S2). The HFD group exhibited significantly higher body, liver, and adipose tissue weights than the ND group (p < 0.05) ( Figure 1A,C; Table S2). The L. fermentum MG4295-treated group did not show a change in the weight of adipose tissues compared to HFD group (Table S2). Although there were no significant differences, the L. fermentum MG4295-treated group showed reduced body weight (14.3%) and liver weight (13.9%) than HFD group ( Figure 1B,C).

Effect of L. fermentum MG4295 on OGTT
After 12 weeks of administration, an OGTT was performed to evaluate glucose tolerance ( Figure 2). The HFD-induced group showed greater glucose content represented as AUC after oral glucose loading and glucose tolerance test compared to the ND group (p < 0.05) (Figure 2A). Moreover, administration of L. fermentum MG4295 caused a significant reversal of the disrupted glucose homeostasis induced by the HFD diet ( Figure 2B).

Effect of L. fermentum MG4295 on OGTT
After 12 weeks of administration, an OGTT was performed to evaluate glucose tolerance ( Figure 2). The HFD-induced group showed greater glucose content represented as AUC after oral glucose loading and glucose tolerance test compared to the ND group (p < 0.05) (Figure 2A). Moreover, administration of L. fermentum MG4295 caused a significant reversal of the disrupted glucose homeostasis induced by the HFD diet ( Figure 2B).

Weight of Body and Tissues
To evaluate the change in weight, we measured the body weight of mice and tissue weight of the liver and adipose tissue ( Figure 1A; Table S2). The HFD group exhibited significantly higher body, liver, and adipose tissue weights than the ND group (p < 0.05) ( Figure 1A,C; Table S2). The L. fermentum MG4295-treated group did not show a change in the weight of adipose tissues compared to HFD group (Table S2). Although there were no significant differences, the L. fermentum MG4295-treated group showed reduced body weight (14.3%) and liver weight (13.9%) than HFD group ( Figure 1B,C).

Effect of L. fermentum MG4295 on OGTT
After 12 weeks of administration, an OGTT was performed to evaluate glucose tolerance ( Figure 2). The HFD-induced group showed greater glucose content represented as AUC after oral glucose loading and glucose tolerance test compared to the ND group (p < 0.05) (Figure 2A). Moreover, administration of L. fermentum MG4295 caused a significant reversal of the disrupted glucose homeostasis induced by the HFD diet ( Figure 2B).

Effect of L. fermentum MG4295 on Serum and Incretin Hormone Concentrations
At the end of the experiment, the HFD-induced group exhibited increased serum levels of glucose, insulin, ALT, AST, TC, LDL, and GLP-1 (Figure 3). Although the L. fermentum MG4295-treated group did not display a significant difference with the HFD-induced group, administration of L. fermentum MG4295 led to reduced insulin, ALT, TC, and LDL levels and increased GLP-1 levels compared to HFD-induced group.

Effect of L. fermentum MG4295 on Serum and Incretin Hormone Concentrations
At the end of the experiment, the HFD-induced group exhibited increased serum levels of glucose, insulin, ALT, AST, TC, LDL, and GLP-1 (Figure 3). Although the L. fermentum MG4295-treated group did not display a significant difference with the HFD-induced group, administration of L. fermentum MG4295 led to reduced insulin, ALT, TC, and LDL levels and increased GLP-1 levels compared to HFD-induced group.

Histopathological Examination of The Liver and Diameters of Adipocytes in Epididymal Adipose Tissues
Histological changes in liver and adipose tissues were assessed to confirm improvement due to the administration of L. fermentum MG4295 (Figure 4A,B). In the HFD-induced group, the histopathological total score (5.8 ± 0.5), including the level of steatosis, lobular inflammation, and hepatocellular hypertrophy, was markedly increased compared to that in the ND (0.4 ± 0.6) group. In contrast, in the L. fermentum MG4295-treated group, the score was significantly reduced (3.0 ± 1.2) compared to that in HFD group (Figure 4C). Regarding the diameters of adipocytes in the adipose tissue, hypertrophy (increases of mean diameters) of epididymal adipose cells was more markedly observed in HFD-induced group (92.2 ± 4.5 μm) than in ND group (43.5 ± 2.8 μm) ( Figure 4D). However, it was decreased in the L. fermentum MG4295 group (76.1 ± 5.6 μm) compared to the HFD-induced group. In this study, L. fermentum MG4295 showed an improvement in the histopathological score in liver tissues and lipid accumulation in adipocytes.

Histopathological Examination of The Liver and Diameters of Adipocytes in Epididymal Adipose Tissues
Histological changes in liver and adipose tissues were assessed to confirm improvement due to the administration of L. fermentum MG4295 (Figure 4A,B). In the HFD-induced group, the histopathological total score (5.8 ± 0.5), including the level of steatosis, lobular inflammation, and hepatocellular hypertrophy, was markedly increased compared to that in the ND (0.4 ± 0.6) group. In contrast, in the L. fermentum MG4295-treated group, the score was significantly reduced (3.0 ± 1.2) compared to that in HFD group ( Figure 4C). Regarding the diameters of adipocytes in the adipose tissue, hypertrophy (increases of mean diameters) of epididymal adipose cells was more markedly observed in HFD-induced group (92.2 ± 4.5 µm) than in ND group (43.5 ± 2.8 µm) ( Figure 4D). However, it was decreased in the L. fermentum MG4295 group (76.1 ± 5.6 µm) compared to the HFD-induced group. In this study, L. fermentum MG4295 showed an improvement in the histopathological score in liver tissues and lipid accumulation in adipocytes.

Effect of L. fermentum MG4295 on Regulation of Insulin and Gluconeogenesis Signaling Pathway
The effect of L. fermentum MG4295 on insulin and gluconeogenesis signaling pathwayrelated gene expression in the liver and muscle via qPCR was assessed ( Figure 5). Compared to the ND group, the HFD-induced group exhibited significantly downregulated the mRNA expression of IRS-1, AKT, and GLUT4 (0.21-, 0.30-, and 0.26-fold of control, respectively, p < 0.05) and upregulated the mRNA expression of G6Pc and PEPCK (2.52-and 1.42-fold of control, respectively). Conversely, administration of L. fermentum MG4295 significantly upregulated IRS-1 and GLUT4 mRNA expression (0.76-and 0.91-fold of control, respectively), and G6Pc and PEPCK mRNA expression was significantly decreased (2.06-and 0.59-fold of control, respectively). For HFD-induced group and L. fermentum MG4295-treated group in AKT (0.52-fold of control), mRNA expression showed no significant difference. In AKT mRNA expression, the L. fermentum MG4295 group increased that of levels.

Effect of L. fermentum MG4295 on Regulation of Insulin and Gluconeogenesis Signaling Pathway
The effect of L. fermentum MG4295 on insulin and gluconeogenesis signaling pathway-related gene expression in the liver and muscle via qPCR was assessed ( Figure 5). Compared to the ND group, the HFD-induced group exhibited significantly downregulated the mRNA expression of IRS-1, AKT, and GLUT4 (0.21-, 0.30-, and 0.26-fold of control, respectively, p < 0.05) and upregulated the mRNA expression of G6Pc and PEPCK (2.52-and 1.42-fold of control, respectively). Conversely, administration of L. fermentum MG4295 significantly upregulated IRS-1 and GLUT4 mRNA expression (0.76-and 0.91fold of control, respectively), and G6Pc and PEPCK mRNA expression was significantly decreased (2.06-and 0.59-fold of control, respectively). For HFD-induced group and L. fermentum MG4295-treated group in AKT (0.52-fold of control), mRNA expression showed no significant difference. In AKT mRNA expression, the L. fermentum MG4295 group increased that of levels.

Discussion
In recent years, probiotics have attracted the attention of scientists as functional foods to be used as dietary supplements and for preclinical and/or clinical trials [35]. Although probiotics are defined as GRAS or belong to species qualified presumption of safety status, novel candidates have to establish this qualification to be used as probiotics. In this regard, prerequisites, such as stability, safety, and health benefits, have been investigated. In stability study, L. fermentum MG4295 survived in simulated gastrointestinal conditions; the percentage viability each gastric and intestinal samples ranged from 58.9 to 98.9% and 97.7 to 99.4%, respectively. A low pH of gastric juice is one of the important health benefits

Discussion
In recent years, probiotics have attracted the attention of scientists as functional foods to be used as dietary supplements and for preclinical and/or clinical trials [35]. Although probiotics are defined as GRAS or belong to species qualified presumption of safety status, novel candidates have to establish this qualification to be used as probiotics. In this regard, prerequisites, such as stability, safety, and health benefits, have been investigated. In stability study, L. fermentum MG4295 survived in simulated gastrointestinal conditions; the percentage viability each gastric and intestinal samples ranged from 58.9 to 98.9% and 97.7 to 99.4%, respectively. A low pH of gastric juice is one of the important health benefits to inhibit the overgrowth of pathogens that induce gut dysbiosis [36]. However, as an acidic pH destroys many microorganisms that are ingested, resistance to human gastric transit is a major selection criterion for microorganisms present in probiotics [37]. Auto-aggregation is one of the important qualities of probiotics for colonizing epithelial cells [38,39]. Our strain showed relatively high auto-aggregation values (58%) compared to the strains used in other studies (<30%) [40,41]. In this respect, L. fermentum MG4295 is likely to survive in the gastrointestinal condition, as it showed survivability and adhesive ability in the stomach and intestinal conditions. The safety of probiotics is important for the choice as probiotic strains and consumer demand [42,43]. Thus, the safety and functionality of probiotics should be evaluated using global standards. In our study, L. fermentum MG4295 showed no hemolysis or β-glucuronidase activity. Additionally, if probiotics have antibiotic resistance, those probiotics could transfer antibiotic resistance to pathogenic bacteria that can cause serious problems [43,44]. Lower cut-off MIC values for the antibiotics test of L. fermentum MG4295 fulfilled global standards. In conclusion, these results demonstrate that L. fermentum MG4295 could be used as a probiotics.
Supplementation with L. fermentum MG4295 showed a tendency to cause loss of body and liver weight in the HFD group, suggesting that L. fermentum MG4295 could prevent obesity and ameliorate hyperglycemia. Previous studies have reported that probiotic dairy products containing L. acidophilus and L. casei could defer the beginning of glucose intolerance and hyperglycemia in a high fructose-induced rat model [45]. Moreover, the consumption of probiotic product, including L. acidophilus La5 and Bi. lactis Bb12, restored the fasting blood glucose in T2DM patients [46]. Furthermore, treatment with L. plantarum NA136 significantly reduced weight gain and adipose tissue, lipids, AST, and ALT levels in HFD/F-treated mice [47]. In our study, administration of L. fermentum MG4295 for 12 weeks improved diabetic, NAFLD parameters, and oral glucose tolerance in HFD-induced groups.
GLP-1 is a kind of incretin secreted by intestinal enteroendocrine cells [48]. GLP-1 has many physiological effects, such as stimulation of glucose-dependent insulin secretion and increment of B-cell numbers [49,50]. Many studies have described that the probiotics elevate the release of GLP-1 and decrease inflammation and IR caused by HFD [33,51,52]. This hormone affects the GIT by inhibiting the secretion of gastric juice and gastrointestinal peristalsis, delaying the emptying of gastric contents, and stimulating the hypothalamus in the central nervous system to increase the sense of satiety and appetite [53,54]. Thus, weight, fasting blood glucose level, and other indicators related to T2DM can be decreased [52]. Our results are consistent with previous studies that revealed a hyperglycemic effect in HFD-induced groups.
In T2DM, insulin resistance can be defined as a metabolic state with a relatively low insulin action (insulin sensitivity) compared to the normal state of physiological insulin concentrations [4,8]. IR appear in target organs as well as liver, muscles, and adipose fat tissues [7]. IR causes compensatory hyperinsulinemia along with metabolic abnormalities, such as glucose hypersensitivity, hyperlipidemia, hepatic steatosis, and NAFLD [55]. The insulin signaling pathway plays a key role in controlling IR and glucose metabolism [56]. To elucidate the putative mechanisms of L. fermentum MG4295 on insulin signaling pathways and gluconeogenesis in HFD-induced groups, target genes were investigated. The PI3K was initiated by tyrosine phosphorylation of the IRS-1 in reaction to insulin stimulation [57]. Activation of PI3K catalyzes the phosphorylation of Akt that activate downstream proteins containing glycogen synthesis [9]. Eventually, these stages are activated, and GLUT4 from the plasma membrane of myocytes is translocated to enhance glucose transport [34]. In our study, factors related to insulin signaling pathway, including IRS-1, AKT, and GLUT4, were elevated by the administration of L. fermentum MG4295 to HFD-induced mice. Additionally, G6Pc and PEPCK play major roles in gluconeogenesis stimulated by glucagon [58]. In a previous study, L. paracasei CCFM0416, L. fermentum MTCC 5689, and L. plantarum MTCC 5690 suppressed the expression of G6Pc and PEPCK in liver tissues [24,33]. Similarly, our results showed a more significantly reduced expression of G6Pc and PEPCK involved in gluconeogenesis in the MG4295 group than in HFD group. Taken together, this study suggests that Limosilactobacillus fermentum MG4295 could be usable as a functional food and nutraceutical to prevent diabetes by modulating the insulin and gluconeogenesis signaling pathways.

Conclusions
In this study, L. fermentum MG4295 was verified to be a probiotic by investigating its properties and safety. L. fermentum MG4295 improved serum profiles, including glucose, insulin, ALT, TC, LDL, HDL, and GLP-1 levels of HFD-induced mice. It also significantly reduced weight gain, hepatic histopathological score, and diameters of adipocytes of epididymal adipose tissues. Furthermore, L. fermentum MG4295 effectively ameliorated IRS-1, AKT, GLUT4, G6Pc, and PEPCK expressions. This result means that L. fermentum MG4295 could be regulate insulin and gluconeogenesis pathways. Consequently, L. fermentum MG4295 can be used as a probiotic strain to improve hyperglycemia and NAFLD.