Type 2 diabetes (T2D) is one of the most common and clinically important metabolic disorders¹⁾, is characterized by chronic hyperglycemia resulting from insulin deficiency and insulin resistance²⁾. Obesity is a major risk factor in patients with diabetes, and weight loss in prediabetes has been reported to delay onset or reduce T2D diagnosis³⁾. To prevent T2D progression, clinicians provide behavioral counseling for obesity to prevent T2D progression and prescribe various glucose-lowering agents such as biguanides, sulfonylureas, meglitinides, dipeptidyl peptidase-4 (DPP-4) inhibitors, and glucagon like peptide-1 (GLP-1) receptor agonists. Although these biomedical drugs helps patients with T2D manage their symptoms to a certain extent, they still face multiple long-term health complications with poor glycemic control⁴⁾.

Currently, to manage different types of diabetes with efficacy and safety, biomedical drugs have been used along with other traditional, complementary, and alternative medicines, including herbs, herbal materials, and finished herbal products. Traditional Chinese medicine (TCM) and Korean medicine are becoming increasingly popular in Western countries as alternative intervention for T2D⁵⁾. Traditional medicine interprets diabetes from a different perspective than Western medicine. In traditional medicine, diabetes involves Sogal syndrome with three types of Samso symptoms depending on the progression of diabetes, that is, early, middle, and late stages, using customized prescriptions. The safety of traditional medicines has been widely studied and applied in the clinical treatment of T2D.

The clean-diabetes mellitus 3 (C-DM3) is a herbal formula consisting of Trichosanthis Radix (TrR, Snakegourd Root, the dried root of Trichosanthes Kirilowii), Coptidis Rhizoma (CoR, Chinese Goldthread Rhizome, the dry rhizome of Coptis chinensis Franch), Crataegi Fructus (CrF, Hawthorn Fruit, the dried ripe fruit of Crataegus pinnatifida), and Cinnamomi Cortex (CiC, Cassia Bark, the dried bark of Cinnamomum cassia) which have been clinically used for T2D in traditional medicines via their herbological functions, lowering the blood stasis, turbidity, and lipid accumulation and clearing heat and dampness, and digestion.

This study evaluated the effects of C-DM3 extract as an anti-diabetic drug on the symptoms of T2D in long-term high-fat diet (HFD)-induced obese mice. The mechanism of action of C-DM3 is responsible for its anti-diabetic effects, focusing on the liver and pancreas as representative organs for glucose metabolism.

1. Preparation of C-DM3 extract

Each herb in C-DM3 was purchased from a Korean herbal company (Kwangmyungdang, Ulsan, Korea) and identified by YK Park, an expert in herb discrimination, Dongguk University College of Korean Medicine (Gyeongju, Korea). TrR (60 g), CoR (30 g), CrF (60 g), and CiC (60 g) was chopped and extracted twice in boiling water (2 L) for 3 h, filtered through Whatman No. 1 filter paper (Whatman plc, Maidstone, UK), and concentrated using a vacuum rotary evaporator (Eyela Co. Ltd., Tokyo, Japan) at 60 °C, and then lyophilized in a freeze-dryer (IlShin Lab Co., Yangju, Korea) at -80 °C under 5 mTorr. The powder of C-DM3 extract (yield=32.81%) was stored at -20 °C until use and dissolved in saline before in vivo study.

2. Animals and experiment design

Male C57BL/6 mice (Koatech, Gyeonggi-do, Korea) were housed at an ambient temperature at 20~24 °C and 40±5% humidity, it was maintained on a 12/12 h light/dark cycle in a pathogen-free animal facility. All laboratory animals were handled in accordance with the animal welfare guidelines issued by the Korean National Institute of Health and the Korean Academy of Medical Sciences for the care and use of laboratory animals and were approved by the Institutional Animal Care and Use Committee of Dongguk University (IACUC-2020-10).

The mice were allowed to adjust to the laboratory conditions before the experiment began. Normal group mice were given regular diet (3.1 kcal/g, 18% kcal from fat, 58% kcal from carbohydrates, 24% protein; Huntingdon, UK) and all the remaining mice were given a HFD (5.24 kcal/g, 60% kcal from fat, 20% kcal from carbohydrates, and 20% protein; New Brunswick, NJ, USA) for 8 weeks. Mice were randomly divided into five groups of seven: normal group (Nor), control group (HFD), C-DM3-administrated groups (500 mg/kg, p.o.) and C-DM3-administrated groups (1,000 mg/kg, 3-1,000, p.o.) in diabetic mice, and metformin-administered group (500 mg/kg, p.o., body weight [BW]). After the first 8 weeks, the normal group was administered saline (p.o.) while maintaining a normal diet, and the control group was administered saline (p.o.) while maintaining the HFD. The C-DM3 extract-administered groups were administered C-DM3 extract (p.o.) at a low dose (500 mg/kg) or a high dose (1,000 mg/kg), and the metformin-administered group (Met) was administered metformin (500 mg/kg, p.o.) to maintain the HFD in mice. During drug administration, changes in physiological parameters were measured once weekly. After the experiments, all mice were sacrificed and whole blood, liver, pancreas, and gastrocnemius tissues were harvested for subsequent analysis.

3. Oral glucose tolerance test

An oral glucose tolerance test (OGTT) was performed in mice after drug administration for 4 weeks. After fasting for 12 h, mice in each group were orally administered a glucose solution (Cat no. G820, Sigma-Aldrich, St Louis, MO, USA) at 2 g/kg BW, and glucose levels were measured in tail vein blood at 0, 15, 30, 60, 90, and 120 min using an Accu-Chek Active Blood Glucose Monitor with test strips (Roche Diabetes Care GmbH, Mannheim, Germany).

4. Measurement of physiological parameters

BW, water and food intakes, and fasting blood glucose (FBG) levels were measured. Calorie intake was calculated from food intake data and the calorie index (kcal/g). Organ weights were measured and the ratio of organ weight to BW was calculated.

5. Measurement of serological parameters

To prepare serum samples, whole blood samples were centrifuged twice at 3,000 rpm at room temperature (RT) for 10 min and used to analyze insulin levels using an ELISA kit (Cat no. 90080, Crystal Chem, Elk Grove Village, IL, USA) according to the manufacturer’s protocols. In addition, glucose, alanine aminotransferase (ALT), triglyceride, and low-density lipoprotein cholesterol (LDL-C) levels were measured using an automated clinical chemistry analyzer (FDC7000I, Fujifilm Co, Tokyo, Japan). The value of homeostatic model assessment for insulin resistance (HOMAIR) was calculated using the following formula: HOMAIR=[fasting serum insulin (μU/mL) × fasting serum glucose (mM)]/22.5

6. Western blot analysis

To analyze the expression of target proteins in organs, the liver, pancreas, and muscle tissues were harvested and homogenized in T-PER tissue protein reagent (Cat no.78510, Thermo Fisher Scientific Inc., MA, USA) containing a phosphatase inhibitor (Cat no. P3200-001, Gen DEPOT Inc., Barker, USA). Lysed samples were centrifuged at 14,000 rpm for 20 min at 4 ℃ and the concentration of total protein was determined with the protein assay reagent (Cat no. 5000006, Bio-Rad, CA, USA). For western blotting, equal amounts of protein (30 μg) were electrophoresed on 8~12% sodium dodecyl-sulfate-polyacrylamide gels and transferred to nitrocellulose membranes (Cat no. 66485, Pall Co., NY, USA). The membranes were blocked with 5% skim milk for 1 h at RT, incubated with primary antibodies against insulin receptor substrate 1 (IRS¹⁾, p-IRS1, phosphoinositide 3-kinase (PI3K), p-PI3K, protein kinase B (AKT), p-AKT, glucose transporter 4 (GLUT⁴⁾, and β-actin as am internal control at 4 ℃ overnight. The membranes were washed thrice with Tris-buffered saline with Tween (TBST), incubated with horseradish peroxidase-conjugated anti-mouse (Cat no. 170-6516, Bio-Rad) or anti-rabbit (Cat no. 170-6515, Bio-Rad) secondary antibodies for 3 h at RT, and washed thrice with TBST and subsequently developed using ECL solution (Cat no. 170-5061, Bio-Rad), the bands were analyzed with the ChemiDoc MP Imaging System (Bio-Rad). The intensity was quantified using the ImageJ software (Image J v2, NIH, Bethesda, MD, USA). The expression of each band was calculated relative to that of β-actin.

7. Histopathological observation

Liver, pancreas, and muscle tissues were fixed with 4% paraformaldehyde (Millipore Co, Bedford, MA, USA) in 0.1 M phosphate-buffered saline and embedded in paraffin. The tissue blocks were cut into 5 mm sections and stained with hematoxylin and eosin dye (Seoulin Biosciences Co., Seoul, Korea), and structural changes were observed using a light microscope (LEICA, Wetzler, Germany). The size of pancreatic islets and myofibers of the gastrocnemius were measured in different sections of three different mice of the same group using ImageJ software.

8. Statistical analysis

Data were expressed as mean±standard error calculated by GraphPad Prism 5.0 (GraphPad Software, La Jolla, CA, USA) for each group. Single-factor analysis of variance was used to determine the level of significance of the treatment effect, followed by postoperative analysis using Tukey’s test. Statistical significance was set at P<0.05.

9. Network pharmacological analysis

A network was constructed to perform network pharmacological analysis of C-DM3 in diabetes. The chemical composition of C-DM3 was retrieved from the Traditional Chinese Medicine Systems Pharmacology Database (TCMSP, http://lsp.nwu.edu.cn/tcmsp.php) and a bioinformatics analysis tool for the molecular mechanisms of TCM (http://bionet.ncpsb.org/batman-tcm/index.php). In view of the complicated variations in the pharmacodynamic chemical composition of Chinese herbal formulas, various compounds were evaluated to determine their pharmacodynamic and pharmacokinetic characteristics (drug likeness >0.18% and oral bioavailability >30%). Relevant obesity targets were collected from GeneCards (http://www.genecards.org) and online Mendelian inheritance databases (http://www.omim.org). Venn analysis was used to comprehensively analyze C-DM3, including the evaluation of directly related genes and obesity targets. In the network pharmacological analysis, overlap was defined as a group of potential targets. An interaction network was established for the putative targets, active ingredients, and obesity-associated targets of C-DM3 The interactive network was visualized using Cytoscape 3.7.1 software (https://cytoscape.org). Functional enrichment analysis of C-DM3 in DM was also performed. A protein-protein interaction (PPI) network was obtained using STRING 11.0 (https://string-db.org/cgi/input.pl). Gene Ontology (GO) functional enrichment analyses were performed using the database for annotation, visualization, and integrated discovery (https://david.ncifcrf.gov). Pathway enrichment analysis of C-DM3 was performed using the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway database (http://www.genome.jp/kegg).

10. High-performance liquid chromatography (HPLC) analysis

After the samples were thawed slowly at 4 °C, an appropriate amount of C-DM3 extract (dry powder) was added to pre-cooled methanol/acetonitrile/aqueous solution (2:2:1, v/v), followed by vortex mixing, low temperature ultrasonic treatment for 30 min, centrifugation at 14,000 g, cooling at 4 °C for 20 min, and vacuum drying of the supernatant. Next, mass spectrometry analysis was conducted with 100 μL acetonitrile aqueous solution (acetonitrile:water=1:1, v/v), with re-dissolving, vortexing and, centrifuging at 14,000 g, then cooling at 4 °C for 15 min. The supernatant was then prepared for analysis. The analytical conditions used are listed in Table 1.

Table 1

Conditions of chromatographic and mass spectrometric analyses

Item	Condition
High Performance Liquid Chromatography (HPLC)
Instrument:	LC-20AD pump, SIL-20A autosampler, CTO-20A column oven, CBM-20A system controller (Shimadzu, Japan)
Analytical column:	Hector C18 column (5 μm; 4.6 × 250 mm)
Column temperature:	35 °C
Mobile phase:	Solvent A: water with 0.1 formic acid
Solvent B:	acetonitrile with 0.1% formic acid
Gradient elution:	0–10 min, 90% A; 10–35 min, 90–75% A; 35–85 min, 75–65% A; 15–17 min, 28–36% A; 17–40 min, 36–38% A
Flow rate:	0.5 mL/min
Injection volume:	10 μL
Diode array detector (DAD)
Instrument:	SPD M20A system (Shimadzu, Japan)
Wavelength range:	190–400 nm
Mass spectrum (MS)
Instrument:	LCMS-8040 system (Shimadzu, Japan)
Ionization mode:	ESI
Nebulizing gas:	3 L/min
Drying gas:	15 L/min
Interface voltage:	4.5 kV for positive mode; -3.5 kV for negative mode
Desolvation temperature:	250 °C
Heat block temperature:	400 °C
Instrument control and data acquisition software:	LabSolutions, version 5.93

1. Effects of C-DM3 extract on the changes of physiological markers in obese mice

The BW of the mice was measured twice during the experiment: at the s tart of drug t reatment a t 8 weeks and at the end of drug treatment at 12 weeks before sacrifice. Compared to the Nor, BW was significantly higher in the HFD control group (P<0.001). Treatment with C-DM3 500 (P<0.01), C-DM3 1,000 (P<0.01), and Met (P<0.001) significantly decreased the BW compared to the control group (Table 2). BW gain, which was calculated as the BW difference between 8 and 12 weeks, significantly increased in the HFD group (P<0.001). Weight gain significantly reduced in the C-DM3 500 (P<0.001), C-DM3 1,000 (P<0.001), and Met (P<0.001) groups. The total calorie intake was also calculated to convert the total calorie of foods in each mouse per the normal diet (3.1 kcal/g) or the HFD (5.24 kcal/g). Total calorie intake was significantly increased in the HFD group compared to that in the normal group (P<0.001), which was significantly decreased by the administration of the C-DM3 extract at low (P<0.001) and high (P<0.001) doses and Met (P<0.001). Next, we measured the changes in organ weight and calculated the ratio of each organ weight to BW. As shown in Table 2, the ratio index of organ weight to BW showed a significant decrease in the pancreatic (P<0.001), liver (P<0.05), and muscle (P<0.001) tissues in the HFD group compared to the normal group. The weight index of the pancreas significantly increased after administration of the C-DM3 extract at low (P<0.05) and high (P<0.05) doses, but we did not observe significant changes in the weights of the liver and muscle tissues. The Met group showed a significant increase in muscle weight (P<0.01), but there were no significant changes in the liver and pancreatic tissues.

Table 2

Effects of C-DM3 Extract on Changes of Physiological Markers in Obese Mice

Group		Nor	HFD	C-DM3 500	C-DM3 1,000	Met
BW (g)		27.16±1.094	2.60±1.65a‡	36.92±3.08b†	37.20±3.90b†	32.16±2.02b‡
BW gain (g)		2.94±0.19	5.20±0.22a‡	0.06±0.50b‡	-0.06±0.95b‡	-3.26±0.76b‡
Calories intake (kcal)		315.58±8.66	381.89±10.66a‡	341.75±10.76b‡	344.58±6.21b‡	281.91±15.33b‡
Organ weight/bodyweight ratio (%)	Pancreas	0.85±0.04	0.58±0.04a‡	0.66±0.01b*	0.66±0.04b*	0.58±0.01
	Liver	1.01±0.05	1.40±0.16a*	1.07±0.17	1.11±0.32	0.97±0.08b*
	Muscle	0.55±0.01	0.39±0.01a‡	0.42±0.01	0.41±0.03	0.45±0.03b†

C-DM3: clean-diabetes mellirus, Nor: normal group, HFD: high-fat diet-induced obesity group, C-DM3 500: C-DM3 500 mg/kg group, C-DM3 1,000: C-DM3 1,000 mg/kg group, Met: metformin 500 mg/kg group, BW: body weight.

*P<0.05, ^†P<0.01, and ^‡P<0.001 vs. normal group (a) or HFD control group (b).

In the OGTT, the HFD group showed significantly increased FBG levels at 0, 30, and 60 min (P<0.01 for 0 min, P<0.001 for 30 min, and P<0.05 for 60 min) compared to the normal group (Fig. 1A). Administration of C-DM3 at low (P<0.001 for 30 and 60 min, respectively) and high (P<0.001 for 30 and 60 min, respectively, and P<0.01 for 90 min) significantly decreased glucose levels compared with the HFD group. However, the Met group did not show a significant decrease in HFD mice. The area under the curve (AUC) value was significantly (P<0.001) higher in the HFD group than that in the normal group (Fig. 1B). Administration of C-DM3 at low (P<0.001) and high (P<0.001) doses significantly decreased the AUC value compared to that in the HFD group. However, Met did not affect glucose tolerance in the HFD mice.

Fig. 1. Effects of C-DM3 extract on impaired glucose tolerance in diabetic mice. (A) Levels of fasting glucose were measured in the sera of mice. (B) AUC index was calculated using the equation described in the Materials and Methods section. The data are expressed as the mean±SD. C-DM3: C-DM3 extract 500 mg/kg group, AUC: area under the curve, Nor: normal group, HFD: high-fat diet-induced diabetes control group, Met: metformin 500 mg/kg group, SD: standard deviation. *P<0.05 (n=5 per a group), ^†P<0.01, and ^‡P<0.001 vs. normal (a) or control group (b).

2. Effects of C-DM3 extract on the changes of serological markers in obese mice

To evaluate the effect of C-DM3 extract on serological changes in obesity, various markers were measured in the sera of mice. As shown in Table 3, the levels of glucose (P<0.001) and insulin (P<0.01) were higher in the control group than in the normal group. These increasing levels were significantly decreased in obese mice after the administration of the C-DM3 extract at low (P<0.001 for glucose and P<0.05 for insulin) and high (P<0.001 for glucose and P<0.01 for insulin) doses, and Met (P<0.001 for glucose and P<0.05 for insulin). In HOMA-IR index, the insulin resistance was significantly (P<0.001) increased in the control group compared with the normal group. In the C-DM3 extract-administered groups at low (P<0.001) and high (P<0.001) doses and Met (P<0.001), a significant decrease in the score was observed compared to the control group. The levels of ALT, triglycerides, and LDL-C were significantly elevated in the HFD group (P<0.001 for ALT, P<0.01 for triglycerides, and P<0.01 for LDL-C) compared with those in the normal group. The C-DM3 groups at low (P<0.01 for ALT, P<0.001 for triglyceride, and P<0.001 for LDL-C) and high (P<0.01 for ALT, P<0.001 for triglyceride, and P<0.01 for LDL-C) doses and Met (P<0.001 for ALT, P<0.001 for triglyceride, and P<0.001 for LDL-C) were shown to be significantly decreased in the control group.

Table 3

Effects of C-DM3 Extract on the Changes of Serological Markers in Obese Mice

Group	Nor	HFD	C-DM3 500	C-DM3 1,000	Met
Glucose (mmol/L)	9.23±0.47	16.93±2.96a‡	10.10±0.65b‡	9.70±1.42b‡	6.67±1.34b*
Insulin (uU/mL)	7.72±1.51	17.72±6.06a†	10.06±2.17b*	8.78±2.87b†	9.40±3.85b‡
HOMA-IR Index	3.16±0.60	13.21±4.51a‡	4.55±1.14b‡	3.64±0.75b‡	2.74±1.04b‡
ALT (U/L)	30.40±2.50	112.20±36.73a‡	52.00±18.20b†	51.00±20.60b†	38.60±11.29b‡
Triglyceride (mg/dL)	46.80±8.13	67.20±11.02a†	54.80±4.53b‡	46.00±2.90b‡	42.00±2.90b‡
LDL-cholesterol (mg/dL)	21.48±5.64	38.00±4.96a†	13.56±6.22b‡	24.36±3.25b†	19.04±6.61b‡

The levels of glucose, insulin, ALT, triglycerides, and LDL cholesterol were measured in the sera of the mice. The HOMA-IR index score was calculated using the formula described in the Materials and Methods section.

C-DM3: Clean-Diabetes Mellitus 3, Nor: normal group. HFD: high-fat diet-induced obesity group, C-DM3 500: C-DM3 500 mg/kg group, C-DM3 1,000: C-DM3 1,000 mg/kg group, Met: metformin 500 mg/kg group, HOMA-IR: homeostatic model assessment for insulin resistance, ALT: alanine transaminase, LDL: low-density lipoprotein.

*P<0.05, ^†P<0.01, and ^‡P<0.001 vs. normal group (a) or HFD control group (b).

3. Effects of C-DM3 extract on the IRS1/PI3K/AKT and AMP-activated protein kinase (AMPK) signaling pathways in pancreas tissue of obese mice

To investigate the effects of C-DM3 extract on pancreatic function in obese mice, we determined the phosphorylation of IRS-1/AKT/PI3K and AMPK and the expression of GLUT4 in the pancreatic tissue using western blotting. The HFD group showed increased phosphorylation of IRS1 (Fig. 2B) and decreased phosphorylation of AKT (Fig. 2C), PI3K (Fig. 2D), AMPK (P<0.05, Fig. 2E), and GLUT4 (Fig. 2F) compared to the normal group. However, administration of C-DM3 extract at low and high doses and Met increased the phosphorylation of AKT, PI3K, and AMPK, and the expression of GLUT4 compared to the control group.

Fig. 2. Effects of C-DM3 extract on the AMPK/IRS-1/PI3K/AKT signaling pathway in pancreas tissues of diabetic mice. (A) Phosphorylation of AMPK, IRS, and AKT and the expression of PI3K and GLUT4 were determined in pancreas tissues using western blotting. (B–F) Each band is presented as a representative figure, and a histogram has been constructed using the band density value of at least three independent experiments. Actin was used as the internal control. The data are expressed as the mean±SD (n=5 per a group). (G) Pancreas tissues of mice were stained with hematoxylin and eosin and observed using a microscope (×400). (H) Pancreatic islet areas were calculated using ImageJ software. C-DM3: clean-diabetes mellirus 3, AMPK: AMP-activated protein kinase, IRS-1: insulin receptor substrate 1, PI3K: phosphoinositide 3-kinase, AKT: protein kinase B, GLUT4: glucose transporter type 4, Nor: normal, HFD: high fat diet, SD: standard deviation. *P<0.05 and ^†P<0.001 vs. normal (a) or control group (b).

However, there were no significant changes in the pancreatic tissues of the obese mice (Fig. 2B, D, E, and F). Although no significant changes were observed in the pancreatic tissue of obese mice, treatment with the C-DM3 extract at low and high doses and Met decreased the phosphorylation of IRS1 compared to that in the control group (Fig. 2B).

Next, we observed structural changes in the pancreatic tissues of obese mice. In normal pancreatic tissues, regular size and shape of Langerhans islets were observed (Fig. 2G), but the area of Langerhans islets was significantly increased in the control group (P<0.001) compared with the normal group (Fig. 2H). Administration of the C-DM3 extract at low (P<0.001) and high (P<0.001) doses and Met (P<0.001) significantly decreased the size of islets with a normal structure compared to the control group.

4. Effects of C-DM3 extract on the AMPK/IRS-1/PI3K/AKT signaling pathway in liver tissue of obese mice

To evaluate the effects of the C-DM3 extract on liver function in obese mice, the expression of the regulation-related proteins AMPK, IRS1, PI3K, AKT, and GLUT in the liver tissue was determined using western blotting. The control group showed decreased expression levels of AMPK (Fig. 3B), PI3K (Fig. 3D), AKT (Fig. 3E), and GLUT4 (P<0.05, Fig. 3F) and increased expression of phosphorylated IRS1 (Fig. 3C) in the liver tissue of obese mice compared to the normal group. Administration of C-DM3 at low (P<0.05 for GLUT⁴⁾ and high doses and Met increased the phosphorylation of AMPK, PI3K, AKT, and GLUT4 compared to the control group. However, there was no significant change in the liver tissue of the HFD mice (Fig. 3B, D, E, and F). Administration of C-DM3 extract at low and high doses and Met decreased the phosphorylation of IRS1 compared to that in the control group, but no significant changes were observed (Fig. 3C).

Fig. 3. Effects of C-DM3 extract on the AMPK/IRS-1/PI3K/AKT signaling pathway in liver tissues of diabetic mice. (A) Phosphorylation of AMPK, IRS, and AKT and the expression of PI3K and GLUT4 were determined in pancreas tissues using western blotting. (B–F) Each band is presented as a representative figure, and a histogram has been constructed using the band density value of at least three independent experiments. Actin was used as the internal control. The data are expressed as the mean±SD (n=5 per a group). (G) Liver tissues of the mice were stained with hematoxylin and eosin and observed using a microscope (×400). Arrows show lipid droplets in liver tissues. C-DM3: clean-diabetes mellirus 3, AMPK: AMP-activated protein kinase, IRS-1: insulin receptor substrate 1, PI3K: phosphoinositide 3-kinase, AKT: protein kinase B, GLUT4: glucose transporter Type 4, Nor: normal, HFD: high fat diet, SD: standard deviation. *P<0.05 vs. normal (a) or control group (b).

Next, we observed structural changes in mouse liver tissues. In normal liver tissues, a regular architecture of the hepatic lobules and substantial number of hepatocytes with clear nuclei were observed; however, numerous lipid droplets were observed in the fatty liver tissues of HFD mice (Fig. 3G). Administration of C-DM3 extract at low and high doses and Met decreased the number of lipid droplets in HFD mice.

5. Network pharmacological analysis of C-DM3 on diabetes

The corresponding gene targets of the active compounds of the C-DM3 extract with four herbs in diabetes-related molecular targets were identified using Traditional Chinese Medicine Systems Pharmacology (TCMSP), online mendelian inheritance in Man, and GeneCards database analysis. Also, the “Herb/compounds- argets-disease” interactive network was established between C-DM3 extract and diabetes (Fig. 4A). Subsequently, to further explore the pharmacodynamic mechanism of C-DM3 extract, PPI networks (Fig. 4B), frequency analysis of protein targets (Fig. 4C), GO functional enrichment (Appendix Fig. 1A), and KEGG pathway enrichment analysis (Appendix Fig. 1B) were established and conducted. Common high-frequency targets, genes, and KEGG pathways between C-DM3 extract and diseases were obtained, which provided a theoretical basis for the study of the active ingredient of C-DM3 extract and its clinical application. The centers of the PPI network for C-DM3 and diabetes were actin beta (ACTB) and AKT1. In addition, the frequency of interleukin 6 (IL-6), insulin (INS), jun proto-oncogene (JUN), IL-1β, vascular endothelial growth factor A (VEGFA), and CASP3 was higher. GO analysis revealed three high- frequency molecular functions: receptor-ligand activity, cytokine receptor binding, and cytokine activity. The KEGG results revealed three high-frequency signaling pathways: lipid and atherosclerosis, the mitogenactivated protein kinase signaling pathway, and hepatitis B.

Fig. 4. Network pharmacology-related network and pathway enrichment analysis. Network construction, pathway analysis, and functional enrichment analysis of the effects of C-DM3 extract on obesity and diabetes. (A) Potential active herb/compound-target-disease network. Different color symbols: disease (red), C-DM3 extract (green), targets (purple), and compounds (blue). (B) PPI network. Node information: query proteins and first shell of interactors (colored nodes), second shell of interactors (white nodes), proteins of unknown 3D structure (empty nodes), some 3D structures are known or predicted (filled nodes), curated databases (), experimentally determined (), gene neighborhood (), gene fusions (), gene co-occurrence (), text mining (), co-expression (), and protein homology (). (C) Frequency analysis of protein targets. C-DM3: clean-diabetes mellitus 3, PPI: protein-protein interaction.

6. HPLC analysis of C-DM3 extract

To confirm the quality of the chemical profile of the CDM3 extract, HPLC-DAD-MS analysis was performed. From the qualitative analysis shown in Fig. 5, we can clearly identify the main chemical components of Trichosanthis Radix, Coptidis Rhizoma, Crataegi Fructus, and Cinnamomi Cortex in C-DM3, including hyperoside, berberine, epiberberine, columbamine, coptisine, coumarin, jatrorrhizine, and citric acid. Table 4 presents the detailed chemical profile of C-DM3. The compound content was then determined by quantitative analysis. The equations of standard curve of hyperoside, berberine, epiberberine, columbamin, coptisine, coumarin, jatrorrhizine, and citric acid were y=12.211x+36.211 (R²=0.9998), y=12.979x+0.1911 (R²=0.9994), y=12.206+14.813 (R²=0.9996), y=8.0401x+1.6566 (R²=0.9995), y=15.336x-1.7632 (R²=0.9997), respectively. The contents of hyperoside, berberine, epiberberine, columbamin, coptisine, coumarin, jatrorrhizine, and citric acid were 2.57, 9.26, 37.06, 20.90, and 64.53 mg·g^-1 respectively.

Fig. 5. Chemical profile in C-DM3 extract using HPLC analysis. Main chemical compositions of Trichosanthis Radix, Coptidis Rhizoma, Crataegi Fructus, and Cinnamomi Cortex, such as hyperoside, berberine, epiberberine, columbamin, coptisine, coumarin, jatrorrhizine, and citric acid were identified using LC-DAD-ESI-MS/MS, as described in the Materials and Methods section. C-DM3: clean-diabetes mellitus 3, HPLC: high performance liquid chromatography.

Table 4

Identification of Chemicals in C-DM3 Extract Using LC-DAD-ESI-MS/MS

Peak no.	tR (min)	λmax (nm)	Precursor ion (m/z)	Product ions (m/z)	Identification
1	6.339	212	191.00 [M-H]-	111.00; 173.00; 175.00; 163.00; 147.00	Citric acid*
2	33.808	327, 242, 217	353.10 [M-H]-	191.00; 179.00; 135.00; 161.00; 173.00	Chlorogenic acid*
3	38.097	301, 268, 221	342 [M]+	297.10; 282.00; 265.00; 237.05; 222.00; 191.00	Phellodendrine acid*
4	47.983	255, 201	463.10 [M+H]+	300.00; 179.00; 151.00	Hyperoside acid*
5	56.025	344, 264, 226	336.10 [M-H]-	320.00; 292.00; 280.00; 262.00	Epiberberine
6	56.731	356, 268, 242, 224	338.00 [M]+	322.00; 308.00; 294.00; 280.00	Columbamine
7	57.728	356, 265, 240, 225	320.00 [M-H]-	292.00; 277.00; 262.00	Coptisine
8	65.747	308, 275	147.00 [M-H]-	103.00; 91.00	Coumarin
96	6.940	344, 266, 226	352.10 [M]+	336.10; 337.05; 308.05; 322.00; 320.05	Palmatine
10	68.12934	6, 264, 229	338.10 [M]+	322.00; 307.00; 294.00; 280.00	Jatrorrhizine
11	68.327	346, 264, 22933	6.05 [M]+	320.10; 321.00; 306.00; 303.95; 292.10	Berberine*

C-DM3: clean-diabetes mellitus 3, PPIs: psychophysiological interactions, LC-DAD-ESI-MS/MS: liquid chromatography coupled to diode array detection and electrospray ionisation tandem mass spectrometry.

*These compounds were compared with corresponding standards.

World Health Organization warns of the seriousness of diabetes that the number of people with diabetes and its prevalence have been steadily increasing over the past few years. Humans require energy from food, but excessive eating cause diabetes, obesity, insulin resistance, and high blood sugar levels. However, it is used to treat diabetes, which is a major public health problem worldwide. In Western medicine, a variety of antidiabetic drugs have been developed to stabilize and control blood glucose levels, including insulin and oral hypoglycemic drugs such as biguanides, sulfonylureas, alpha-glucosidase inhibitors, thiazolidinediones, DPP-4 inhibitors, and GLP-1 receptor agonists. However, these antidiabetic drugs help patients with diabetes maintain their condition under control and reduce the risk of diabetes complications, despite their potential side effects.

Traditional medicines in Korea and China explain diseases based on a different concept than Western medicine and apply different treatment methods. Diabetes is included in So-Gal syndrome with polydipsia, polyphagia, and polyuria according to three steps of disease progression: Samso (upper, middle, and lower), which is closely related to different Zang organs, lungs, heart, pancreas, stomach, liver, and kidneys. Therefore, the treatment of Korean and Chinese medicines is applied differently according to the disease steps with different symptoms. Recently, because of the therapeutic potential of traditional medicines, drug research and development using herbal medicines have been actively performed; unfortunately, herbal medicine-derived drugs for diabetes treatment have not been produced despite of various advantages their safety and efficacy. Therefore, to develop new antidiabetic drugs from traditional medicines, we prepared a C-DM3 extract consisting of four different herbs and evaluated its effects on obesity and diabetes in HFD mice.

Among the constituents of C-DM3, TrR acts on the lungs and stomach, with cold, sweet, bitter, and sour tastes, to relieve inflammation and fever. This herb has been mainly studied for its anti-tumor effects on colorectal cancer⁶⁾, breast cancer⁷⁾, and non-small cell lung cancer⁸⁾ and for its inhibition of nephrotoxicity in vitro and in vivo⁹⁾. CoR is also an herb that eliminates inflammation by acting on the heart, pancreas, stomach, liver, gallbladder, and large intestine lung with a cold and bitter taste, and has been reported to have various anti-inflammatory effects on keratinocytes¹⁰⁾, acute and chronic inflammation in vivo¹¹⁾, and anti-obesity and anti-diabetic effects in vivo models^12,13). This herb is known for its bioactive compound berberine^14,15). CrF is used as a digestive aid during indigestion after eating meat because of its warm, sour, and sweet taste in the pancreas, stomach, and liver. It has been reported hypolipidemic, antimyocardial, anti-ischemia, antithrombotic, anti-atherosclerotic, anti-inflammatory, antineoplastic neuroprotective activity, etc.¹⁶⁾ The CiC is an herb to warm up inside by acting on the kidney, pancreas, and bladder. This herb is characterized by a fever with brilliant and sweet tastes, and is well known for its various pharmacological effects, including antitumor, anti-inflammatory, analgesic, anti-diabetic, anti-obesity, antibacterial, antiviral, cardiovascular protective, cytoprotective, neuroprotective, immune-regulatory, anti-tyrosinase activity and so on¹⁷⁾. Based on the pharmacological properties of each herb, the effects of the C-DM3 extract on obesity and T2D can be expected. Our HPLC analysis identified several compounds, including hyperoside, berberine, epiberberine, columbamin, coptisine, coumarin, jatrorrhizine, and citric acid in C-DM3 extract. Thus, it is well known that these compounds have a good effect on improving diabetes, such as antihypoglycemic effects of hyperoside¹⁸⁾, protective effects of berberin on pancreatic β-cells with increase of insulin sensitivity¹⁹⁾, antihyperlipidermic effects of epiberberine²⁰⁾, action of columbamin as lipase and cholinesterase inhibitors²¹⁾, action of coptisine as an enhancer of peripheral glucose consumption²²⁾, multiple action of coumarin glycoside on diabetic nephropathy²³⁾, and hypolipidemic properties of jatrorrhizine²⁴⁾. From these reports, it can be assumed that the antidiabetic efficacy of the C-DM3 extract is due to its ingredients.

In our study, the C-DM3 extract improved diabetic symptoms, such as body weight gain, skeletal muscle loss, impaired glucose tolerance, insulin resistance, increases in triglyceride, LDL-C, and ALT levels with the accumulation of lipids in the liver, and pancreatic islet hyperplasia in HFDinduced obese mice. In addition, the antidiabetic effect of C-DM3 extract was associated with the regulation of the IRS1/PI3K/AKT and AMPK signaling pathways in the pancreas (Fig. 2), liver (Fig. 3), and muscle tissues (Appendix Fig. 2). In our network pharmacological prediction analysis, it was provided that the effective active ingredients of C-DM3 extract may be AKT1, IL-6, INS, JUN, VEGFA, IL-6, etc. to play a role in the treatment of diabetes. KEGG signaling pathway enrichment analysis revealed that the active ingredients of the C-DM3 extract play a therapeutic role in diabetes through multiple pathways, in which AKT1 is a common target of multiple pathways, indicating that this target is the central target. In lipid- and atherosclerosis-related pathways, upregulation of PI3K/AKT expression promotes endothelial cell apoptosis, increases endothelial cell permeability, and plays an important role in the pathogenesis of diabetes. In addition, AKT1 is an important factor of the PI3K/AKT pathway, which plays an important role in regulating glucose homeostasis, lipid metabolism, and cell survival²⁵⁾. The AMPK signal pathway is also up-regulated, which can promote energy metabolism and interacts with other molecular pathways such as peroxisome proliferatoractivated receptor gamma coactivator 1-alpha, PI3K/AKT, reduced nicotinamide adenine dinucleotide phosphate oxidase 4, and nuclear factor kappa B 26). The most important feature of herbal medicine in treating diseases is its multi-component and multitarget synergy; therefore, the C-DM3 extract may play a synergistic role through inflammatory and energy-metabolismrelated pathways. AKT1/ PI3K signaling is an important central target, which provides a reference for further exploration of the pharmacological role of C-DM3 extract.

In conclusion, C-DM3 extract is expected to be effective in improving obesity and preventing obesity-induced diabetes (Appendix Fig. 3). In addition, we expect that C-DM3 extract can be used to treat obesity and prevent diabetes in future alternative and complementary applications and clinical trials.

The Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), Ministry of Health and Welfare, Korea (No. HF20C0121), Shanxi Key Laboratory of Tradition Herbal Medicines Processing (No. 20210901), and Innovation Team of Shanxi University of Chinese Medicine (No. 2022TD1014).

No potential conflict of interest relevant to this article was reported.