Chlorogenic acid improves anti-lipogenic activity of metformin by positive regulating of AMPK signaling in HepG2 cells

Metformin improves lipid profile, however, combination therapy is developing to increase its effectiveness and reduce the deleterious effects of metformin. Chlorogenic acid (CGA) has exhibited lipid-lowering effects. This study aimed to investigate the combined effect of metformin and CGA on lipid accumulation, as well as to elucidate the engaged mechanism in HepG2 cells. To find the non-lethal doses of metformin and CGA, MTT assay was performed. High Glucose (HG) at 33 mM was used to induce lipogenesis in HepG2 cells. Following treatment with different concentrations of metformin and CGA, total lipid content (Oil Red O-staining), triglyceride level, the genes expression of SREBP-1c and FAS, and phosphorylation of AMPK and ACC were measured. Both Metformin and CGA decreased HG-induced lipid accumulation individually, by decreasing total lipid content and triglyceride level. The lowest effective doses of metformin and CGA were 0.25 mM and 5 μM, respectively, which significantly reduced SREBP-1c and FAS genes expression. The combination of these concentrations reinforced these effects. The phosphorylation of AMPK and ACC were more increased by metformin in combination with CGA than both individually. Our findings suggest that CGA synergistically enhances metformin lipid reducing action via the regulating of involved factors in fatty acid synthesis. Therefore, co-administration of metformin with CGA may have further medical value in treating lipid metabolism disorders.


Introduction
Type 2 diabetes mellitus (T2DM) is one of the most common endocrine disorders that affect more than 400 million individuals worldwide [1]. Obese people living with a sedentary lifestyle are more found at risk of T2DM [2,3] and non-alcoholic fatty liver disease (NAFLD), on the other hand, the prevalence of NAFLD is increased in patients with T2DM [4]. The association between NAFLD and T2DM could be explained by insulin resistance and dyslipidemia [5]. At the molecular level, insulin resistance stimulates lipogenesis by activating the sterol regulatory element-binding protein-1c (SREBP-1c) signaling pathway followed by enhanced expression of the genes responsible for fatty acids synthesis, acetyl-CoA carboxylase (ACC), and fatty acid synthase (FAS) [6,7].
Stimulation of lipogenesis results in the accumulation of lipids, especially triglycerides. Lipid accumulation in nonadipose tissues (liver, heart, pancreas, and muscle) leads to impaired function of these organs [8,9]. Therefore, intervention in the lipogenesis process is considered as a therapeutic goal in insulin resistance patients.
Metformin is the first-line therapy for diabetic cases and is the most popular hypoglycemic agent. The decreasing effects of metformin on hepatic glucose production, insulin resistance, and lipid accumulation have been linked to activation of the AMP-activated protein kinase (AMPK) which is the major cellular regulator of lipid and glucose metabolism [10]. Combination therapy strategy or coadministration of metformin with other anti-diabetic drugs is recommended [11,12] to reduce metformin undesirable complications in long-time consumption and enhance its controlling effect on insulin resistance [13][14][15].
Some well-referenced natural agents were revealed for the treatment of metabolic syndrome, T2DM, and obesity [16][17][18]. In addition, it has been confirmed that some beneficial effects of various phenolic compounds such as plant polyphenols and flavonoids were through activation of AMPK [19], deactivation of SREBP-1c as well as reduction of FAS and ACC expression [20]. Chlorogenic acid (CGA) is a phenolic product naturally found in green coffee extracts, and its useful hypoglycemic and hypolipidemic properties have been determined [21][22][23][24][25]. Likewise, long-term consumption of CGA-rich coffee beverages has been correlated to improved T2DM results [26,27]. However, the mechanism underlying combination therapy of metformin and CGA is not yet clear. The current study investigated the synergistically regulating effect of combined treatment of metformin and CGA on lipogenic genes expression and AMPK phosphorylation in HepG2 cells.

HepG2 cells culture and treatments
HepG2 cells were purchased from the Pasteur Institute of Iran and cultured in DMEM medium containing normal glucose (NG, 5.5 mM), 10% FBS, and 1% streptomycinpenicillin in a humid atmosphere containing 5% CO2. After reaching 70% confluence; the cells were seeded for subsequent evaluations in the corresponding plates. After an overnight fasting, HepG2 cells were exposed to a high concentration of D-glucose (HG, 33 mM) for 24 h to induce lipogenic condition. D-mannitol at 27.5 mM was used in the NG group as an osmotic control. The time and dose of glucose were selected from a previously published work [28]. HepG2 cells were treated to selected concentrations of metformin and CGA similarly to HG-treated cells. In combination treatment, cells were incubated with metformin and CGA 2 h before HG induction.

Cell viability assay
MTT assay was performed to measure the cytotoxic effects of the drugs. Briefly, HepG2 cells with a 2×10 4 / well density were seeded in 96-well plates. After 24 h of drug treatment, cells were treated with 100 μL of MTT solution at a final concentration of 0.5 μg/ml PBS, for 4 h at 37°C. The formazan crystals formed in living cells (indicating cell viability) were then dissolved in 100 μl DMSO, and the absorbance was read at 570 nm by a microplate reader (BioTek Instruments, Inc. Winooski, VT, USA).

Measurement of total lipid content by Oil Red O staining
Treated HepG2 cells (2 × 10 5 /well in 24-well plates) were washed three times in cold PBS. Then cells were fixed with 4% formalin for 30 min at room temperature. The cells were stained with Oil Red O working solution (300 mg of oil red O powder dissolved in 150 mL of 60% Isopropanol) for 10 min. After washing the cells with distilled water, lipid particles were observed under the Olympus upright microscope, and images were captured using an Olympus camera mounted on the microscope. Quantitative analysis was done by adding 250 μL DMSO to each well. The absorbance of the red stain was read at 510 nm. Total lipid content was calculated by normalizing the results against total protein levels (μg/mg) determined by the bicinchoninic acid assay (BCA) method.

Determining triglyceride level
The pellet of treated cells (~10 7 ) was homogenized into 5% NP-40 (v/v), and slowly incubate at 80-100°C in a water bath for 2-5 min. The heating was repeated twice to solubilize all triglycerides. The supernatant was transferred to a new tube after centrifugation at top speed for 2 min. Intracellular triglyceride levels were quantified using the Biovision triglyceride quantification Colorimetric/Fluorimetric kit (Biovision Inc, U.S) according to the manufacturer's instructions. Finally, normalized triglyceride concentrations were calculated relative to total protein levels (μg/mg) assessed by the BCA method.

Western blot analysis
HepG2 cells were lysed by RIPA buffer (25 mM Tris-HCl, pH 8, 1% Triton X-100, 0.5% sodium deoxycholate, 0.1% SDS, 5 mM EDTA, and 1 mM PMSF, 150 mM NaCl and 10 mM NaVO4). After measuring protein concentration by BCA kit, equal amounts (30 μg) of protein were subjected to SDS-PAGE electrophoresis. After transferring proteins to the PVDF membrane, the blocking was performed by TBST (containing 5% BSA). Blots were then incubated overnight at 4°C with primary antibodies against β-actin, ACC, pACC, AMPK, and pAMPK. After three times washing with TBST, membranes were incubated for 1 h with secondary antibodies at room temperature. The bands were visualized using an enhanced chemiluminescent substrate. Finally, the densitometry analysis of the blots was carried out using NIH image J software.

Statistical analyzes
All statistical analyzes were performed using GraphPad Prism 6 software. The comparison between the groups was done by one-way ANOVA and the comparison between two independent groups was performed by unpaired T-test. The values of p < 0.05 were considered statistically significant. The results were presented as Mean ± SD, derived from at least three independent tests.
The half maximal inhibitory concentration values (IC50) and median effective doses (ED50) were calculated using CompuSyn software (ComboSyn Inc, Paramus, NJ, USA). CompuSyn analysis was also used for the assessment of drug interactions according to the Chou-Talalay method [29]. Combination index (CI) < 1, CI = 1, or CI > 1 is corresponding to the effect of synergism, additivity, or antagonism, respectively.

Effect of metformin and CGA on cell viability
HepG2 cells were treated by increasing concentrations of metformin and CGA for 24 h to evaluate their cytotoxic effects. The IC50 of metformin and CGA were calculated as 7.816 mM and 306.196 μM, respectively (Fig. 1A). As shown in Fig. 1B, C, the concentrations lower than 2 mM of metformin and 50 μM of CGA did not affect the viability of HepG2 cells. Therefore, these sub-lethal concentrations were treated for subsequent assessments. High concentration of glucose used to induce lipogenesis did not alter cells viability compared to Glucose at 5.5 Mm. Meanwhile, sub-lethal doses of metformin and CGA did not have a deleterious effect on HepG2 cells' survival in lipogenic conditions (Fig. 1D, E). The combined treatment of the three components showed no decrease in cells' viability (data not shown).

Metformin and CGA reduce lipid accumulation in HepG2 cells
HG significantly induced lipid accumulation in HepG2 cells in comparison with NG (p < 0.01). As indicated by the results of Oil Red-O in Fig. 2A, metformin and CGA dosedependently reduced HG-induced total lipid content. Although it was not significant at 0.1 mM of metformin and 2 μM of CGA compared with the HG group. The ED50 values were calculated as 0.986 mM and 15.04 μM, respectively from Oil red O results (Fig. 2B).
Intracellular triglyceride levels were decreased by metformin and CGA similar to Oil Red-O trend (Fig. 2C). Moreover, the effects of sub-lethal doses of each compound were compared in pairs. The most striking result was the stronger lipid-lowering effect of CGA than metformin at the used concentrations. As shown in Fig. 2A, C, the lowest effective dose of metformin and CGA for lipid reducing was 0.25 mM and 5 μM, respectively, confirmed by the results of ED50.
To characterize the drugs' synergy, the combination of metformin and CGA was treated in HG-exposed cells. Combination Index (CI) analysis demonstrated a clear synergistic lipid-lowering effect between the two drugs combination at doses lower than 2 mM and 50 μM, respectively (Fig. 2D). CI values were 0.691 and 0.586, for the combination of 0.25 mM metformin and 5 µM CGA for total lipid and TG, respectively, which indicate the synergism of the lowest effective doses.

CGA potentiates the inhibitory effect of metformin on lipid accumulation
To investigate the synergistic effect of CGA in combination with metformin in HG-mediated lipogenic conditions, HepG2 cells were treated simultaneously with the lowest effective doses of metformin (0.25 mM) and CGA (5 μM). The Oil Red-O staining results proved the synergistic effect of the combination. In addition to further reducing lipid accumulation compared with metformin (p < 0.01), CGA significantly improved the lipidlowering effect of metformin (p < 0.001) (Fig. 3A, B). Furthermore, the synergistic effect of CGA as an antilipogenic drug was determined when CGA significantly enhanced the triglyceride-lowering activity of metformin (p < 0.001) (Fig. 3C). The simultaneous effects of these two drugs are interesting since they reduced lipid levels to an almost normal range. In fact, HG + metformin + CGAcotreated cells had no significant difference neither in total lipid nor in triglyceride in comparison with NGtreated cells.

CGA attenuates lipid accumulation by regulating lipogenic factors
The relative expression of genes involved in lipogenesis, SREBP-1c, and FAS, were evaluated to confirm the lipidlowering effect of metformin and CGA. As expected, under HG-induced lipogenesis, the expressions of SREBP-1c and FAS were significantly increased (54% and 63%, respectively, Fig. 4A). The elevated SREBP-1c and FAS expression were reduced significantly by both metformin (20% and 19%, respectively) and CGA (25% and 27%, respectively) compared to HG (p < 0.01). These findings may Fig. 1 The effect of metformin (Met) and chlorogenic acid (CGA) on the survival rate of the HepG2 cells. Cells were exposed to intended concentrations of Met, CGA, and HG for 24 h. Their viability was measured by MTT assay. A IC50 of Met and CGA were 7.816 mM and 306.196 μM, respectively. B, C Cells viability was evaluated under Met and CGA individually treatment. D, E Cells viability was evaluated under Combined treatment of HG with Met or CGA. Data were obtained from three independent experiments. Values are expressed in mean ± SD, statistically significant at **p < 0.01 compared to the untreated control indicate that the anti-lipogenic effects of CGA were through the regulation of SREBP-1c and FAS. In addition, it was found that metformin + CGA combination suppressed SREBP-1c and FAS (33% and 36% reduction, respectively) significantly greater than when they were used alone. Interestingly, the concomitant inhibitory effects were to such an extent that no significant difference was seen with the NG as a control group.
As we know, AMPK is the upstream regulator in the lipogenesis pathway. Next, the phosphorylation of AMPK and ACC were examined to provide further confirmatory evidence about the inhibitory effects of metformin + CGA combination on fatty acid synthesis. Following HG treatment, a significant decrease was observed in AMPK and ACC phosphorylation levels (54% and 48%, respectively), compared with NG (Fig. 4B). Metformin and CGA significantly reversed the down-regulation of the phospho-AMPK/total AMPK ratio (31% and 55% increase, respectively) caused by HG treatment. Likewise, we detected some changes in the phosphorylation of ACC by metformin (21% increase) and CGA (42% increase). In the presence of metformin, CGA led to more increase in pAMPK (76%) and pACC (63%) levels in HG treated cells, suggesting an additive effect of CGA on metformin action.

Discussion
Herein, we are the first to provide in vitro evidence of CGA potency which amplifies the lipid-lowering efficacy of metformin. In line with previous documents on hypoglycemic and hypolipidemic effects of metformin as well as CGA on T2DM control [14,21,25], we investigated their effects on lipid accumulation in a model of human hepatic cells. Impaired liver functions in gluconeogenesis and glucose incorporation into lipids (lipogenesis) pathways are responsible for metabolic disorders such as T2DM and obesity, which in turn lead to insulin resistance and NAFLD [30,31] so various drugs are used to control these metabolic conditions. We observed that HG (33 mM) induced hepatic lipid accumulation as prior studies reported [32][33][34]. Treatment with sub-lethal doses of either metformin or CGA was coupled with a significant decrease in lipid content caused by HG. In addition, their lipid reduction was found in a dose-dependent manner. The lowest effective doses were used to maintain maximum cell survival and correspondingly to minimize the possible side effects of drugs. We unexpectedly found that the lipid accumulated in CGA-treated cells was significantly lower than that of metformin ones, indicating that metformin is an efficient but not significantly more efficient option than CGA. Combination of metformin with natural antioxidants has appeared as an emerging trend for better T2DM management. Although numerous studies confirmed the beneficial health effects of CGA or metformin, the question arose of how the combination of CGA + metformin changes cell lipid status. Our important finding was that CGA significantly improved the lipid-lowering effect of metformin so that the combined treatment returned lipid levels to their normal control levels indicating the synergistic effect of CGA. Lipid content is consisting of triglyceride and other Expression of SREBP-1c and FAS (A), and the phosphorylation of AMPK and ACC (B, C) in the lipogenic state. 33 mM glucose significantly increased SREBP-1c and FAS expression and decreased the AMPK and ACC phosphorylation, compared to control ( # p < 0.01). The lowest effective doses of met and CGA were opposed to these effects. *p < 0.05 & **p < 0.01 related to HG complex lipid molecules that free fatty acids are the fundamental blocks for their synthesis. As we found a reduction in total fat and triglyceride by both compounds. Consistent with Ong et al. [35] findings which showed CGA-mediated reduction in hepatic fat and triglyceride, our current result may suggest the hypothesis that CGA inhibited lipogenesis via inhibiting fatty acid synthesis. In previous studies, the hypolipidemic effects of CGA have been attributed to the regulation of enzymes involved in lipid metabolism through the AMPK pathway [25,35]. Activation of AMPK is accompanied by decreasing lipid accumulation in hepatocytes, where SREBP-1c, a downstream substrate of AMPK, inhibited fatty acid synthesis. AMPK phosphorylates and represses the transcriptional activity of SREBP-1c and its lipogenic targets, including ACC and FAS, in hepatocytes exposed to HG, leading to reduced lipogenesis [36].
Currently, we show that the monotherapy of metformin or CGA improved pAMPK/AMPK accompanied by increasing the phosphorylation of ACC which were more pronounced in CGA as compared with metformin. However, CGA + metformin combined effects were greater than that found by their individual potencies, the combination was said to be synergistic. We also observed that the expression of lipid synthesis-related enzymes SREBP-1c and FAS was increased 1.5-2-fold by HG. Following the AMPK activation, expression of SREBP-1c and FAS was significantly suppressed with either metformin or CGA, however, a stronger suppression to nearly normal levels was found by the combined therapy of CGA + metformin. Taken together, our data revealed the synergistic effects of CGA and metformin combination on lipid metabolism, and their signaling mechanism is illustrated as a schematic model in Fig. 5.
Of note, the AMPK activating effects of metformin or CGA alone have been already reported. Given the past knowledge, metformin activated AMPK, leading to subsequently reduced lipogenesis and lipid accumulation [10,37] and it depended on the concentration of metformin, less than 0.25 Mm [38,39]. In the current study, metformin was also used at a dose of 0.25 mM, which in line with the above-mentioned studies confirmed the activated lipid metabolism pathways by metformin. There are conflicting results of the anti-lipogenic effects of CGA. Sudeep H et al. [40] reported the hyperlipidemic effect of CGA that its administration significantly led to the AMPK activation and improved suppression of hepatic lipid accumulation. The network was created with BioRender.com. In individual treatment, the factor that is represented with green shade is upregulated, and orange shades are downregulated. The purple shade in Met + CGA combination indicates activation while the red shades are inhibited. Solid lines denoted interaction in individual treatment; interrupted lines denoted synergistic conditions a subsequent decrease in ACC activity as evident by ACC phosphorylation in high-fat diet-fed rats. However, Mubarak et al. [41] expressed that CGA could not protect the liver against lipid accumulation induced by a high-fat diet since CGA resulted in decreased AMPK and ACC phosphorylation in mice. CGA inefficiency on AMPK activation had been reported by other limited studies [42,43]. The discrepancy in the results may be explained by the type of studies, the used doses of CGA, the duration of treatment, and the different strains of animals.
To date, it was just our research team that reported the combined effects of metformin and CGA. In our laboratory, two simultaneous studies were designed to evaluate the combination of hepatoma cell line HepG2 (our in vitro study) and on high-fat diet mice (in vivo study conducted by another group of our colleagues). In line with our results, their results showed that the combined therapy reduced fat accumulation and improved hepatic steatosis by AMPK activating. In addition, they found that concomitant use of metformin and CGA had far more impact compared with monotherapy and could even lead to normal status [44].
This study encompasses some possible limitations, including the use of an in vitro hyperglycemic condition. In addition, the effect of combined therapy of metformin and CGA observed in liver cancer cell line HepG2 may differ from how would be present in humans.
This study has presented the first report of metformin and CGA combination exerted a synergistic lipid-lowering effect on HepG2 cells mediated through AMPK, ACC, SREBP-1c, and FAS signaling pathways. Thus, these data might lead to a new approach to drug discovery with possible beneficial outcomes in T2DM and NAFLD.