Ameliorated effects of a lipopeptide surfactin on insulin resistance in vitro and in vivo

Abstract Surfactin, produced by Bacillus amyloliquefaciens fmb50, was used to treat insulin‐resistant (IR) hepatocyte. It was found that surfactin increased glucose consumption in insulin‐resistant HepG2 (IR‐HepG2) cells and ameliorated IR by increasing glucose transporter 4 (GLUT4) protein expression and AMP‐activated protein kinase (AMPK) mRNA expression, promoting GLUT4 translocation and activating phosphatidylinositol 3‐kinase (PI3K)/protein kinase B (Akt) in IR‐HepG2 cells. Meanwhile, surfactin downregulated protein expression of phosphoenolpyruvate carboxy kinase (PEPCK) and glucose‐6‐phosphatase (G6Pase), further inhibiting hepatic gluconeogenesis. In addition, surfactin played important roles in eliminating reactive oxygen species (ROS), improving mitochondrial dysfunction, and inhibiting proinflammatory mediators. We observed that surfactin promoted glucose consumption, meanwhile increased translocation and protein expression of GLUT4 in Caco‐2 cells. These results confirmed the conclusion in hepatic cells. Furthermore, surfactin supplement decreased body weight, food intake, and fasting blood glucose of type 2 diabetes mellitus (T2DM) mice induced by streptozotocin (STZ)/high‐fat diet (HFD). Our data indicated that surfactin ameliorated insulin resistance and lowered blood glucose in intro and in vivo.

Diabetes mellitus (DM) is a major cause of morbidity and mortality in the modern era and also decreases both quality of life and life expectancy (Riaz et al., 2020). Increased blood glucose level is a typical symptom for patients with diabetes. Metabolic glucose absorption in target cells is dependent on the insulin and glucagon levels, which are mainly secreted from the pancreatic islet cells. Type 1 diabetes mellitus (T1DM) is characterized by an insufficient insulin level, while type 2 diabetes mellitus (T2DM) is accompanied by insulin resistance (IR) (Riaz et al., 2020) and dysregulation of glucose utilization. It is well known that IR is often accompanied with ROS overproduction, which leads to oxidative stress in the liver. In normal state, high blood glucose level induces β cells to synthesize insulin. If persisting, this overwhelms the endoplasmic reticulum capacity and leads to the accumulation of misfolded protein (Riaz et al., 2020).
This disturbed endoplasmic reticulum environment induces islets β-cell impairment and insulin resistance or production insufficient, leading to the efficiency of insulin-mediated glucose uptake and utilization reduction and glucose regulation disorders in the liver Jiang et al., 2014). In addition, IRE1α plays a major role in insulin biosynthesis and keeping the oxidative balance in islets β cell (Hassler et al., 2015). Severe high glucose could promote dephosphorylation of IRE1α, resulting in the attenuation of IRE1α activity and insulin resistance or reduced insulin production (Qiu et al., 2010). Therefore, amelioration of hepatic IR is regarded as an important strategy in the prevention and treatment of T2DM.
It was reported that mitochondrial dysfunction can also stimulate excessive ROS generation (Tang et al., 2020). Furthermore, PI3K, Akt, AMPK, and GLUT4 are closely involved in the regulation of glucose uptake and metabolism (Lv et al., 2019;Wang, He, et al., 2020). In particular, AMPK, a phylogenetically conserved intracellular energy sensor, can regulate insulin sensitivity in the liver (Ren et al., 2018), and it is a potential target for ameliorating IR.
GLUT4 translocation from cytoplasm to membrane surface is basal and essential for glucose transport and uptake. Food-derived substances also exhibited the effects of alleviation IR via above signaling pathways. Dendrobium officinale polysaccharide could increase protein expression of PI3K, Akt, and GLUT4 to improve glucose metabolism disorder (Wang et al., 2018). Zhenjiang aromatic vinegar extract exhibited positive effects on elimination ROS through phosphorylating c-Jun N-terminal kinase (JNK) in IR-HepG2 cells (Xia et al., 2021). It is reported that glucose metabolism disorder can be improved by activating the IRS-1/PI3K/Akt and AMPK signaling pathways in hepatocytes (Ren et al., 2018;Wang et al., 2020b). Two gluconeogenic enzymes, including PEPCK and G6Pase, were suppressed by upregulating p-Akt (Ren et al., 2018). It is demonstrated that glucose uptake in insulin-sensitive tissues can be promoted by activating insulin-mediated signaling pathways or directly upregulating GLUT4 protein expression without enhancing insulin signaling (Choi et al., 2011).
In this study, the effects of a lipopeptide surfactin on high insulininduced IR in HepG2 cells were explored. Moreover, the effects of surfactin on insulin signaling pathways, gluconeogenic enzymes, inflammatory factors, and ROS level were investigated in HepG2 cells.
Subsequently, the effects of surfactin on glucose consumption in Caco-2 were further confirmed. Furthermore, the effects of surfactin on T2DM mice induced by STZ/HFD were investigated. | 2457 CHEN Et al.

| Cell culture and treatments
HepG2 cells were obtained from the cell bank of Chinese Academy of Science and cultured in high-glucose (4.5 g/L) DMEM supplemented with 10% FBS and 1% penicillin-streptomycin solution under 5% CO 2 at 37°C. After activation, all cells used in experiments were within 20 generations. An insulin-induced IR-HepG2 cells model was established using high-glucose medium according to the previous method (Wang et al., 2019a) with slight modification. Briefly, when HepG2 cells reached over 80% confluence, the cells were cultured in a medium with normal DMEM (10% FBS and 1% penicillinstreptomycin mixed solution) as the control group, cells were treated with 15 µg/ml insulin for 48 h as the model group (IR-HepG2 cells), and cells were treated with insulin and different concentrations of surfactin (6.25, 12.5, 25, 50, and100 µg/ml) as the surfactin group.
Caco-2 cells were obtained from the Cell Bank of Chinese Academy of Science (Shanghai, China) and cultured in DMEM supplemented with 20% FBS and 1% penicillin-streptomycin solution under 5% CO 2 at 37°C. Cells were cultured with DMEM (20% FBS and 1% penicillin-streptomycin mixed solution) for 48 h as the control group, and cells were treated with different concentrations of surfactin (6.25, 12.5, 25, 50, and 100 µg/ml) as the surfactin group.

| Cell viability assay
Cell viability was assessed according to a previous literature (Wang & Dong, 2019b). In brief, HepG2 and Caco-2 cells were seeded in 96-well plates at a density of 3 × 10 3 cells/well and incubated for 48 h. Cells were treated with 15 µg/ml of insulin and various concentrations of surfactin (6.25, 12.5, 25, 50, and 100 µg/ml). Then, the culture medium was removed and 100 µl of 1 mg/ml MTT was added into each well and incubated for an additional 4 h at 37°C.
Subsequently, 50 µl of dimethyl sulfoxide (DMSO) was added into each well after removing the MTT solution, and the absorbance at 490 nm was detected by a microplate reader. The absorbance of the control group was regarded as 100%. The cell viability was calculated using the following equation:

| Glucose consumption assay
HepG2 cells were seeded in 96-well plates and treated with insulin and different concentrations of surfactin. After discarding medium, 200 µl FBS-free medium was added and cells were incubated for 12 h.
Caco-2 cells were seeded in 96-well plates and treated without or with different concentrations of surfactin. The glucose level in collected supernatant was determined by a glucose assay kit (Beijing Applygen Gene Technology Co., Ltd). The absorbance values were measured at 550 nm. MTT analysis was used to adjust the glucose consumption.
The glucose consumption was calculated by the following equation: Glucose consumption = initial glucose concentration in the medium (no cells, only medium) -terminal glucose concentration in each group (control, model, and treatment groups). The relative levels of glucose consumption in HepG2 cells were normalized by cell viability as glucose consumption/cell viability.

| Measurement of intracellular ROS generation
HepG2 cells were seeded in six-well plates. Intracellular ROS levels were determined by a fluorescent probe 2′,7′-dichlorofluoresc ein-diacetate (DCFH-DA) (Keygen biotech Institute). Firstly, 10 µM DCFH-DA was added to each well and cells were incubated at 37°C for 30 min in the dark. Then, the cells were washed three times with phosphate-buffered saline (PBS), and fluorescence intensity was measured by flow cytometry within 30 min.

| Determination of mitochondrial membrane potential
HepG2 and Caco-2 cells were seeded in six-well plates. Detection of mitochondrial membrane potential was carried out using fluorescent probe JC-1, according to manufacturer's instructions (Nanjing Sciben Biotech Co., Ltd). Cells were incubated with FBS-free medium containing JC-1 probe for 20 min, and then cells were washed two times using JC-1 staining buffer to discard excess JC-1 probe. Fluorescence images in different groups were captured by Fluorescence Imager (Nikon Eclipse 80i, Nikon Co.). The image J was applied to analyze the fluorescence intensity of fluorescence images. In addition, when the mitochondrial membrane potential is high, JC-1 is driven into mitochondrial matrix to form JC-1 aggregates, and thus, the intensity of red fluorescence is dramatically increased. By contrast, when the mitochondrial membrane potential is low, JC exists as a monomer and it exhibits green fluorescence. The mitochondrial membrane potential was calculated using the following equation: Mitochondrial membrane potential = the fluorescence intensity of JC-1 aggregate/the fluorescence intensity of JC-1 monomer (Du et al., 2010).

| RT-PCR analysis
HepG2 and Caco-2 cells were cultured in six-well plates. After culturing, total RNA in cells was extracted using commercial kit (TransGen Biotech. Co., Ltd) according to manufacturer's instruction. The concentrations of the extracted RNA were measured using Nanodrop 2000 (Thermo Ficher Scientific Inc.). The extracted RNA (200-800 ng/µl) was reverse transcribed into cDNA using a HiScript II Q RT SuperMix kit (Vazyme Biotech. Co., Ltd.). The primer sequences of target genes were listed in Table 1. RT-PCR analysis was performed using HieffTM SYBR Green Master Mix (Yeasen Biotech Co., Ltd.) and a RT-PCR detection system (Applied Biosystem). About Cell viability ( % ) = OD treat OD control × 100 % 100 ng cDNA was used to measure the mRNA levels of target genes.
The detailed program was as follows: 95°C for 120 s; then, 40 cycles of 10 s at 95°C, 30 s at 60°C, and 30 s at 72°C for extension, and the melting curve was applied to check the specificity of the primers.
The data were normalized to mRNA expression of glyceraldehyde-3phosphate dehydrogenase (GAPDH), which was used as an endogenous control. The relative expression level of genes was calculated using the 2 −ΔΔCt method. In addition, RT-PCR analysis was carried out using 6.25-25 µg/ml of surfactin.

| Measurement of parameters involved to glucose metabolism
The production of glycogen, and the protein contents of AMPK, GK, GLUT4, PEPCK, and G6Pase were determined using ELISA kits (Feiya Biotechnology Institute), according to manufacturers' instructions. In addition, the levels of glycogen and GK were shown in Figure S1(a,b).

Extraction of protein in membrane and cytoplasm in HepG2 and
Caco-2 cells was carried out using commercial kit (Nanjing Jiancheng Bioengineering Institute), according to the manufacturer's instruction. Briefly, the total protein in the membrane and cytoplasm was extracted separately using differential centrifugation, according to the manufacturer's instruction. Subsequently, protein GLUT4 in the membrane and cytoplasm was determined using ELISA kit (Feiya Biotechnology Institute).

| Immunofluorescence staining for GLUT4
In HepG2 and Caco-2 cells, GLUT4 was observed by immunofluorescence staining. Briefly, the cells were treated without or with surfactin, and washed with PBS buffer for three times and fixed using 4% (w/v) paraformaldehyde for 8 min. Cells were blocked using 1% goat serum albumin (GSA) at room temperature for 1 h . Subsequently, cells were treated with anti-GLUT4 primary antibody overnight at 4°C. Cells were incubated with goat antirabbit IgG/FITC secondary antibody at room temperature for 1 h, and then treated with DAPI (2 µg/ml) for 8 min in the dark. Finally, the cells were washed with PBS buffer for three times. In addition, a control for the immunofluorescence GLUT4 was performed only through replacing anti-GLUT4 primary antibody by PBS buffer.
Fluorescence imaging was performed using a Fluorescence Imager (Nikon Eclipse 80i, Nikon Co.). The image J was applied to analyze the fluorescence intensity of fluorescence images.

| Western blot analysis
Total proteins in HepG2 cells were extracted using radioimmunoprecipitation assay (RIPA) buffer added with 1 mM phenylmethylsulfonyl fluoride (PMSF) (Beyotime Institute of Biotechnology). The supernatant of whole-cell lysates was collected by centrifugation at 12,000 g for 15 min at 4°C, and used for western blot. Proteins were separated using 8% and 12% sodium dodecyl sulfate-polyacrylamide gels (SDS-PAGE), which then was transferred to nitrocellulose (NC) membrane (GE Healthcare Life Science). After blocking in Trisbuffered saline and 0.1% Tween 20 (TBST) supplemented with 5% skim dry milk for 2 h, the membranes were overnight incubated with diluent primary rabbit antibodies at 4°C. After washing with TBST solution for three times, the membranes were incubated with HRP-linked secondary antibody anti-rabbit IgG (Beyotime Institute of Biotechnology) for 2 h. Finally, the bands were scanned using ECL plus solution (Affinity Biosciences). Target band density was quantified using Image J software (National Institutes of Health). GAPDH was the internal standard.

| Measurement of body weight, food intake, and fasting blood glucose
Body weight and food intake were monitored every 7 days. The fasting blood glucose (FBG) was determined from 12-h fasting mice for every 2 weeks during experiments.

| Statistical analysis
The data showed as mean ± standard derivation (SD). Variances among the groups were analyzed by one-way analysis of variance (anova) in SPSS statistics followed by Duncan test. p < .05 indicated significant differences.

| Effects of surfactin on cell viability of IR-HepG2 cells
To evaluate cytotoxic effects of surfactin on hepatic cells, cell viability was examined by MTT assay. The results exhibited that 6.25-25 µg/ml of surfactin had no negative effect on cell viability in HepG2 and IR-HepG2 cells as compared with that in the control group (Figure 1a,b), illustrating that surfactin had no cytotoxic effect on HepG2 cells at selected concentrations.

| Effects of surfactin on the glucose consumption in IR-HepG2 cells
To investigate the potential roles of surfactin in IR-HepG2 cells, the glucose consumption was detected in HepG2 cells. As shown in Figure 1(c), the glucose consumption in IR-HepG2 cells were significantly decreased as compared with that in the control group, indicating that insulin resistance could inhibit glucose consumption. However, it was found that surfactin treatment significantly increased glucose consumption in IR-HepG2 cells, especially at high dose (25 µg/ml) (Figure 1c). These results suggested that surfactin was able to enhance glucose consumption and reduce insulin resistance in IR-HepG2 cells. In addition, 6.25 µg/ml of surfactin as surfactin group was used for the following study in HepG2 cells.

| Effects of surfactin on GLUT4 in IR-HepG2 cells
The protein expression of GLUT4 was detected to further explore whether surfactin improved glucose consumption by promoting protein expression of GLUT4 to the membrane.

| Effects of surfactin on gluconeogenic key enzymes in IR-HepG2 cells
It is well known that G6Pase and PEPCK are key rate-limiting enzymes in hepatic gluconeogenesis process (Wang & Dong, 2019b).
To investigate the effect of surfactin on gluconeogenesis in IR- and PEPCK was markedly downregulated after surfactin treatment as compared with those in IR-HepG2 cells (p < .05) (Figure 2a,b).
Furthermore, these results indicated that surfactin can inhibit hepatic gluconeogenesis through suppressing protein expression of G6Pase and PEPCK.

| Effects of surfactin on key genes and protein involved in glucose metabolism in IR-HepG2 cells
It is reported that GLUT4 translocation from cytoplasm to membrane surface is mainly regulated by the IRS-1/PI3K/Akt signaling pathways , and the phosphorylation of Akt will induce GLUT4 translocation (Izela et al. 2017  The ratio of p-AMPK to AMPK. All data are presented as mean ± SD (n ≥ 3) for each group. Different lowercase alphabet letters over bars indicate statistically significant differences between two groups (p < .05) AMPK) were significantly increased in surfactin treatment group as compared with that in IR-HepG2 cells (Figure 4f,g). In addition, Li et al. reported that gymnemic acid had no effect on protein expression of AMPK but increased p-AMPK/AMPK, further activating phosphorylation of AMPK (p-AMPK), which also ameliorated hyperglycemia in T2DM. It was also reported that IR could be alleviated through upregulating p-Akt and activating PI3K/Akt signaling pathway (Wang, He, et al., 2020). Thus, we assumed that

| Effects of surfactin on ROS generation in IR-HepG2 cells
Previous reports showed that oxidative stress is regarded as a major cause of IR development. High glucose induced excessive accumulation of ROS and eventually led to oxidative stress, which contributes to liver injury (Du et al., 2010). To investigate effects of surfactin on oxidative stress, the level of intracellular ROS was determined.
As shown in Figure 5(a,b), high insulin dramatically increased ROS production by 52.2% (p < .05) as compared with that in control group. However, surfactin treatment decreased ROS production in the model group by 30.0% as compared with that in IR-HepG2 cells (p < .05).

| Effects of surfactin on mitochondrial membrane potential in IR-HepG2 cells
As an important intracellular signal, mitochondrial membrane potential can initiate the antioxidative defense system to attenuate cellular oxidative stress (Chandel, 2015). As shown in Figure 5(c,d), the mitochondrial membrane potential in the model group was decreased by 51.3% as compared with that in the control group (p < .05), while surfactin treatment increased mitochondrial membrane potential by 33.7% as compared with that in the model group (p < .05).

| Effects of surfactin on inflammatory responses in IR-HepG2 cells
The protein and gene expression of pro-inflammatory factors were detected to evaluate effects of surfactin on inflammatory responses in IR-HepG2 cells. As shown in Figure 6(a,b), high insulin significantly upregulated mRNA and protein expression of IL-1β as compared with those in the control group. However, surfactin dramatically downregulated mRNA and protein expression of IL-1β as compared with those in IR-HepG2 cells (Figure 6a,b). In addition, insulin-treated cells possessed a higher mRNA expression of IL-6 and TNFα as compared with those in the control group (p < .05) (Figure 6c,d).
Moreover, surfactin treatment significantly downregulated mRNA expression of IL-6 and TNFα as compared with those of IR-HepG2 cells (Figure 6c,d). These revealed that surfactin could inhibit proinflammatory cytokines release. These results are consistent with a previous report .

| Effects of surfactin on the glucose consumption in Caco-2 cells
To explore the effects of surfactin on Caco-2 cells, glucose consumption was also determined in Caco-2 cells. As shown in Figure 7(a), 6.25-50 µg/ml of surfactin had no cytotoxic effect on Caco-2 cells.
Glucose consumption in Caco-2 cells was obviously increased, when compared with that in the control group, in a dose-wise pattern ( Figure 7b). This revealed that surfactin also promoted glucose consumption in Caco-2 cells. In addition, 12.5 µg/ml of surfactin as surfactin group was used for the following study in Caco-2 cells.

| Effects of surfactin on GLUT4 in Caco-2 cells
The protein expression and translocation of GLUT4 were detected in Caco-2 cells. As shown in Figure 8

| Effects of surfactin on body weight, food intake, and FBG in T2DM mice induced by STZ/HFD
The body weight and food intake of mice were recorded during experiment. As shown in Figure 9(b), body weight of mice possessed an increase trend, and was stable after 8 weeks; while the body weight of the T2DM group significantly decreased after 9 weeks, as compared with that control and surfactin administration group (p < .05).
The body weight in T2DM mice dramatically increased as compared with that in the control group (p < .05). Surfactin significantly decreased body weight of T2DM mice. As shown in Figure 9(a), food intake of mice increased in the control and T2DM group, while maintained stable in surfactin supplement group. In addition, food intake of T2DM mice was notably enhanced as compared with that of the control and surfactin group after 7 weeks. Furthermore, food intake in the surfactin groups dramatically lowered than that in the control group. These indicated that surfactin significantly inhibited food intake and maintained body weight stability.
The FBG of all mice was recorded every 2 weeks. As shown in Figure 9(c), the FBG of the T2DM group was dramatically increased compared with that of the control group (p < .05). After surfactin supplement, the FBG of mice in the MSF group was significantly decreased, when compared with that in the T2DM group (p < .05). The

FBG of LSF and HSF group was not significantly different compared
to that in T2DM group.

| DISCUSS ION
T2DM is a complex and chronic metabolic disease characterized by IR and hyperglycemia, and is also regarded as one of the main threats to human health . In the basal state, Akt is phosphorylated after activating insulin-mediated signaling pathway, and then plays roles in regulating glucose metabolism via multiple pathways. Relative mitochondrial membrane potential after calculation of the ratio of JC-1 aggregates to JC-1 monomers. All data are presented as mean ± SD (n ≥ 3) for each group. Different lowercase alphabet letters over bars indicate statistically significant differences between two groups (p < .05) the stimulation of insulin in hepatic cells. In addition, AMPK dysfunction caused by IR might also lead to inhibition of GLUT4 translocation.
It is reported that the promotion of glucose uptake in skeletal muscle and liver involves two separate pathways: the insulin-dependent mechanism such as PI3K/Akt activation and contraction-mediated AMPK stimulation (Kuo et al., 2016). In this study, we observed that surfactin treatment increased hepatic glucose consumption. The following investigation indicated that surfactin upregulated mRNA expression of AMPK, increased p-Akt/Akt and p-AMPK/AMPK in the protein level, activated PI3K/Akt and AMPK pathway, and increased GLUT4 translocation, meanwhile suppressed protein expression of PEPCK and G6Pase. We speculated that surfactin promoted glucose consumption possibly owing to increasing GLUT4 translocation by activating PI3K/Akt and AMPK signaling pathways. Previous reports also indicated that phosphorylated AMPK and Akt could prevent hepatic gluconeogenesis and lower blood glucose levels, independent of the activation of IRS-1/PI3K/Akt insulin signaling Wang et al., 2020b). This is consistent with our investigation. In addition, we observed that surfactin treatment did not affect hepatic glycogen synthesis ( Figure S1a,b). Taken together, we concluded surfactin ameliorated IR by increasing glucose consumption, reducing gluconeogenesis, and activating the PI3K/Akt and AMPK signaling pathways ( Figure 8). However, this hypothesis needs to be further verified.
Glucose transporters also play a major role in the effective absorption of glucose via the insulin-mediated pathway in peripheral tissues, such as skeletal muscle and adipose tissue (Kuo et al., 2016).
It is reported that increased glucose absorption of intestinal cells and decreased glucose into blood in the intestine-pancreatic axis also could alleviate IR (Yang et al., 2021). We also investigated the effects of surfactin on the glucose consumption in Caco-2 cells. In fact, the peptides in human body might be broken into some small peptides by pepsin, trypsin, and chymotrypsin to weaken their bioactivities during the oral administration and gastrointestinal digestion (Wang et al., 2020a). In view of this, in vitro digestion of surfactin was performed, including oral administration and gastrointestinal digestion.
The results indicated that surfactin could not be broken down and maintains initial bioactivity after digesting ( Figures S2 and S3). In Mitochondria is essential for ATP production, and contributes to GLUT4 translocation and glucose uptake in cells (Osorio-Fuentealba et al., 2013). However, oxidative stress caused by excessive ROS and free radicals leads to cell membrane damage and mitochondria dysfunction (Hervera et al., 2018), blocking insulin signaling through MAPK pathway. In this study, surfactin inhibited hepatic ROS overproduction, and increased mitochondrial membrane potential in HepG2, indicating that surfactin inhibited oxidative stress and restored mitochondria dysfunction. It is speculated that surfactin suppressed ROS overproduction and restored mitochondrial dysfunction probably through activating MAPK pathway, contributing to ameliorate IR. This speculation still needs further investigation to confirm. In addition, it is reported that protein hydrolysate of oysters could donate electrons to free radicals and convert them into more stable and less chemically reactive products . Peptides with high proportion of hydrophobic amino acids also exhibit high radical eliminating activities owing to their hydrophobic interaction with phospholipid bilayer of membrane (Wang et al., 2020a). Based on this, we hypothesized that surfactin eliminated ROS probably due to its high hydrophobic amino acid Leu.
In general, an imbalance in pro-and anti-inflammatory factors aggravates metabolic inflammation and induces IR (Tang et al., 2021). Intracellular ROS accumulation might activate NF-κB pathway  and release of TNFα and IL-6, blocking IRS-1/PI3K signaling and aggravating IR . In our study, it is observed that mRNA expression of IL-6, TNFα, and IL-1β, and protein expression of IL-1β were suppressed after treating with surfactin in IR-HepG2 cells. These results are in agreement with a previous study (Tang et al., 2020). These revealed that surfactin inhibited proinflammatory factors production and improved insulin signaling pathway.
At present, food-derived bioactive ingredients are increasingly used to ameliorate IR owning to their nontoxic, non-side effects, and low cost, such as vinegar extracts (Xia et al., 2021), nut-derived peptides (Wang et al., 2020b), and the postbiotics (Zouari et al., 2016).
Common metformin and rosiglitazone could control hyperglycemia due to indirect activation of AMPK by disturbing mitochondrial respiration (Turner et al., 2008). It was reported that C-phycocyanin, a kind of blue protein isolated from Spirulina platensis, can ameliorate hyperglycemia through inhibiting hepatic gluconeogenesis and increasing glycogen synthesis due to activating Akt and AMPK in insulin resistance hepatocytes (Ren et al., 2018). A postbiotic, the extracellular polysaccharide extracted from Lactobacillus plantarum RJF4 can inhibit the activity of α-glucosidase, further lowering blood glucose and playing important roles in antidiabetes (Dilna et al., 2015). A previous report showed that surfactin produced from B. amyloliquefaciens WH1 is regarded as a novel drug for alleviating T1DM (Gao et al., 2014). Surfactin also decreased serum total F I G U R E 1 0 Underlying and hypothesis of molecular mechanism ameliorating IR of surfactin in IR-HepG2 cells. Red arrows denote changes in response to high insulin, and green arrows denote changes in high insulin-induced cells receiving surfactin intervention cholesterol (TC) and low-density lipoprotein cholesterol (LDL-C), and increased serum high-density lipoprotein cholesterol (HDL-C) in mice. These reports revealed that surfactin possessed antidiabetic effects. However, surfactin exhibited beneficial effects on regulation of glucose consumption, the protein expression, and translocation of GLUT4 and gluconeogenesis, in addition to elimination of ROS and inhibition of proinflammatory factors in this study. On the other hand, surfactin also could improve oral delivery of insulin via increasing its penetration through the cell membrane (Zhang et al., 2016). This indicated that surfactin as an insulin carrier can maintain insulin stability and facilitate insulin delivery to target sites.
However, this displays a different action compared to this study. In our study, surfactin as a postbiotic promoted hepatic glucose consumption through regulating PI3K/Akt/GLUT4 and AMPK signaling pathways, further regulating glucose metabolism and ameliorating IR. In addition, it was observed that surfactin supplement also significantly lowered fasting blood glucose of T2DM mice induced by STZ/HFD. However, the detail mechanism still needs to be investigated in animal model.

| CON CLUS ION
In conclusion, surfactin could ameliorate IR probably owing to promotion of GLUT4 translocation and suppression of protein expres- also lowered body weight, food intake, and FBG of T2DM mice induced by STZ/HFD. Therefore, surfactin can attenuate high insulininduced IR and decrease high blood glucose. It is encouraging for further exploration into amelioration of IR and regulation effects on glucose metabolism in animals and humans.

ACK N OWLED G M ENTS
C.X.Y. contributed to study design, preparation, and drafting of the manuscript. C.X.Y. and Z.H.Y. contributed to data collection and analysis. M.F.Q., Z.L.B., P.X.Y., L.Y.J., and L.Z.X revised the manuscript.

CO N FLI C T O F I NTE R E S T
The authors declare no competing financial interest.