3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) products were obtained from Sigma Chemical Co. (St. Louis, MO). GLUTag and INS-1 cells were acquired from Invitrogen (Carlsbad, CA, USA). RPMI-1640 (Gibco BRL, NY, USA), Dulbecco’s Modified Eagle Medium (DMEM), HEPES (pH 7.3), RPMI-1640 (Gibco BRL, NY, USA), GLP-1 Total ELISA kit, Millipore (Merck, USA), and Rat/mouse insulin was measured by ELISA (Millipore, Billerica, MA, U.S.A). RPMI medium, penicillin, and streptomycin were acquired from HyClone Laboratories (Logan, UT, USA).

Leaf samples were obtained from a cultivar in Amecameca, Mexico State, and a specimen was authenticated by botanist Aurora Chimal. A voucher specimen (6543) is stored in the Department of Botany of the Escuela Nacional de Ciencias Biologicas-IPN for preservation.

The samples were sliced into smaller pieces and then dried under shade at room temperature before being ground finely and subjected to extraction. One thousand grams of powder was extracted at reflux temperature for 4 h. The methanol extract was then filtered, and the process was repeated twice using the same quantity of solvent to extract the celery exhaustively. Methanol was removed from the extract using a rotavapor (BUCHI Rotavapor R-200, Switzerland) under reduced pressure at 45°C. The residues were combined and weighed (CEL) and calculated for percentage yield (19% w/w), then stored in a deep freeze. The methanol extract was reconstituted with distilled water for assays.

Flavonoids were characterized using a Shimadzu HPLC system (Kyoto, Japan), consisting of an LC-20AT pump, an SPD-M20A DAD detector, a CTO-10AS VP column oven, and a SIL-20A autoinjector. A 10 µL aliquot of each sample was eluted through a C18 column of Inertsil ODS-SP (4.6 mm×250 mm, 5 µm) at a column temperature of 28 °C and a flow rate of 0.8 mL/min. The mobile phase consisted of solvent A (0.1% formic acid) and solvent B (acetonitrile). The gradient elution program was as follows: 0–5 min, 5%–15% B; 5–25 min, 15%–30% B; 25–39 min, 30%–38% B; 39–47 min, 38%–55% B; 47–50 min, 55%–70% B; 50–56 min, 70%–75% B; 56–60 min, 75%–100% B; 60–65 min, 100%–5% B; 65–75 min, 5% B. Chromatograms were obtained at 350 nm, and UV diode array detection spectra were scanned from 190–800 nm. Identification of flavonoids from methanol extract was performed by HPLC-ESI-MS using a Thermo Ion trap mass spectrometer (Waltham, MA, USA) equipped with an electrospray ionization (ESI) interface. The ESI-MS was applied in positive modes, and the product ions from protonated molecular ions were further analyzed by MS/MS (MS2). Nitrogen was used as the drying and nebulizing gas. Other parameters consisted of a drying gas temperature of 350 °C, capillary voltage of 3.5 kV, capillary exit voltage of 120.4 V, cap exit offset voltage of 77.2 V, and a range of m/z 100–1000 for full-scan MS analysis.

Cell lines INS-1 and GLUTag cells were cultured in DMEM consisting of 11.1 mmol/L D-glucose supplemented with 100 µg/mL streptomycin, 100 U/mL penicillin, 10% fetal bovine serum, 2 mmol/L L-glutamine, 10 mmol/L HEPES, 1 mmol/L 50 µmol/L mercaptoethanol, and sodium pyruvate, in 5% CO₂ at 37˚C. C2C12 mouse muscle cells (myotubes) were obtained from the American Type Culture Collection (Manassas, VA, USA). To assay insulin-resistant cells, myotubes were preincubated with 5.5 mM glucose in DMEM for 2 h without FBS and supplemented with 1% BSA and then incubated for 18 h in Modified Eagle Medium (DMEM) without fetal bovine serum (FBS), with 5.5 mM glucose, 0.75 mM palmitate, 1% bovine serum albumin (BSA), and 200 mM insulin.

INS-1 cells were incubated with 400, 700, 1000, 1250, 1500 µg/mL of CEL, and 250 µM H₂O₂ for 15 min, and cell viability was evaluated using the test 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT; Sigma-Aldrich Co., St. Louis, MO, USA). A stimulation of 250 µM H₂O₂ reduced cell viability to ~68%. Thus, oxidative stress was induced by the addition of 200 µM H2O2 for 15 min. In another assay, INS-1 cells were incubated with 100, 125, 150, 200 µg/mL, and 250 µM H₂O₂ for 15 min. Cell integrity was measured by LDH assay, determined as lactate dehydrogenase released in the culture medium. Briefly, after treatment, 0.2 mL culture medium from each group was evaluated for LDH activity using a commercial kit (Cayman Chemical, MI, USA) according to the manufacturer’s instructions. Mitochondrial function was measured by cellular ATP analysis. The culture medium was supplemented with CEL (150, and 200 µg/mL) and maintained for 24 h before cellular ATP analysis. H₂O₂ (250 mM) was incorporated 2 or 4 h before the end of the experiment. Cellular ATP levels were then measured in both cells exposed to CEL and H₂O₂ using a fluorometric assay kit (Abcam, Boston, MA, USA) according to the manufacturer’s instructions.

Glucotoxicity was induced in INS-1 cells incubated for five days with glucose at a concentration of 5.5 mM and 30 mM. INS-1 cells (3 X 10⁵ cells/well) were incubated with 30 mM glucose and 25, 50, or 100 µg/mL of CEL or GLP-1 (10 nM) for 48 h. After removing supernatants, they were replaced with a medium containing a mixture of FBS (2%) and glucose (3 mM) and incubated for 5 h. Immediately afterward, the INS-1 cells were stimulated with 5.5- or 30-mM glucose in Krebs–Ringer buffer (20 mM HEPES, pH 7.4, 5 mM NaHCO3, 1.2 mM KH2PO4, 1.2 mM MgSO4, CaCl₂, KCl, 2.54 mM and 119 mM NaCl) for 65 min at 37 °C. The supernatant was used for the detection of insulin secretion, employing an Elisa Kit according to the manufacturer’s instructions (Lingo Research, Lingo, MO, USA). GLP-1 (10 nM) was used as a standard.

INS-1 cells (3 X 10⁵ cells/well) were seeded in 96-well microplates with 100 µL culture medium for a period of 24 h before the addition of CEL and/or H₂O₂. After 24 h of treatment with CEL (25, 50, and 100 µg/mL) and/or H₂O₂ (250 µM), the cell supernatant was removed and rinsed with phosphate-buffered saline. Insulin secretion (GSIS) was measured using an insulin enzyme-linked immunosorbent assay kit (Lingo Research, Lingo, MO, USA). Untreated cells served as a control.

Myotubes were preincubated in 500 µL of DMEM with 5.5 mM glucose for 4 h and various concentrations (125, 150, 200 µg/mL) of CEL. Then, after 4 h of incubation, 500 µL of the supernatant was collected, and the glucose concentration was measured using the glucose oxidase/peroxidase technique with a commercial kit. To calculate glucose utilization, the remaining glucose in the culture medium after incubation was determined. Metformin (125 µg/mL) was used as a positive control.

GLUTag cells (1 × 10⁵ cells/well), after 24 h of incubation, were washed with KRB buffer (pH 7.2) containing HEPES (20 mM), BSA (0.2%), NaHCO₃ (25.5 mM), CaCl₂ (2.5 mM), MgSO4•7(H2O) (1.16 mM), KH2PO4 (1.2 mM), KCl (4.7 mM), and NaCl (0.114 M). Then, KRB buffer containing a dipeptidyl peptidase-4 inhibitor (0.2 µM), glucose (10 mM), and extract (CEL) at different concentrations were added to the cells, followed by incubation for 30 min. The total GLP-1 secretion was evaluated in the supernatants using a GLP-1 Total ELISA assay kit. Lithocholic acid (LCA; 30 µM) was used as a standard.

The samples, substrate, and DPP-IV enzyme were mixed and incubated for 30 min at 37 °C. The liberated product was measured using a DPP-4 inhibitor screening assay kit (Ray Biotech, GA, USA) as per the manufacturer’s instructions.

Statistical analysis was carried out using the SPSS/PC Version 10 package (Chicago, IL, USA). Data are expressed as mean ± SD (n = 5). The experimental data obtained from three repetitions were examined by one-way analysis of variance, followed by post hoc Duncan’s multiple range tests. P < 0.05 or less were considered statistically significant.

Figure 1

Figure 2

Figure 3

HPLC/MS analysis was carried out to characterize the phenolic profile of the methanolic extract. The MS and UV-Vis spectra data of each chromatographic signal were compared with reference compounds and literature data, supporting the presence of 11 compounds such as Apigenin (1); Luteolin-7-galactoside (2); Aesculetin (3); Pelargonidin-3-O-glucoside (4); Peonidin-3-O-beta-galactopyranoside (5); Farnesol (6); Peonidin-3-(6″-acetylglucoside) Mesylate (7); Apigenin-C-hexoxide (8); Apigenin-O-hexosyl pentoside (9); Apiin (10); Chrysoeriol-O-hexosyl pentoside (11) (Figure 1).

In order to estimate the effect of Apium graveolens on diabetes control, we focused our study on the effects of methanolic extract in vitro models on INS-1 cells and C2C12 cell lines. Although some studies have evaluated the anti-hyperglycemic effect of celery, its mechanisms related to its antidiabetic activity remain poorly understood. Here, we characterize the mechanism of action of the hypoglycemic effect, which is mediated by the GLP-1 pathway to maximize glucose-stimulated insulin secretion.

MTT assays were carried out to evaluate pancreatic cell viability. Exposure of these INS-1 cells to a high H₂O₂ level (250 µM) for 48 h significantly (p < 0.05) inhibited their viability, and CEL treatment prevented this cell death. This indicates that celery protected the INS-1 cells against oxidative stress conditions.

LDH, an enzyme found in all tissues, can be used as an index of INS-1 cells damage. When INS-1 cells were exposed for a period of 24 h to a high H₂O₂ treatment, cytotoxicity was produced in a dose-dependent manner, together with intense membrane damage and a subsequent release of LDH.

Mitochondrial function plays an important role in the progression of type 2 diabetes due to its role in controlling cellular redox potential, oxidative phosphorylation, the release of caspase-activating proteins, ATP production, and electron transport. Interruption of these functions can lead to cell death. The preservation of mitochondrial function was examined in pancreatic cells exposed to H₂O₂. We have observed that CEL treatment significantly increases ATP in a dose-dependent manner, reflecting enhanced mitochondrial function and, therefore, increased cell viability with an important inhibition in the activations of death signals.

This study investigated CEL's insulin secretagogue effect in pancreatic β-cells exposed to high glucose-induced and H₂O₂ toxicity in pancreatic INS-1 cells as models. After administration of 125, 150, and 200 µg/mL concentrations of celery extract, we found an increased insulin secretion in pancreatic β-cells under oxidative (H₂O₂) and glucotoxic conditions as a result of a suppression in oxidative stress, oxidative injury, renovation of intracellular ATP, and the conservation of the morphology, function, and mitochondrial content.

In type II diabetes, its pathogenesis is characterized by β-cell dysfunction and insulin resistance. In this study, we used C2C12 cells treated with palmitate to create insulin resistance and increased reactive oxygen species (ROS) production. Previous studies have shown that metformin's capacity to enhance insulin sensitivity is related to decreased mitochondrial ROS production. Consequently, metformin was used as a standard in this experiment. CEL reduced the insulin resistance induced by palmitate in comparison with the resistance control, suggesting that CEL reverted insulin resistance similarly to metformin (metf) at the same concentration. Pancreatic β-cell mass and function can decrease during the development of type 2 diabetes, thus highlighting the necessity of searching for therapeutic agents that can preserve the mass and function of the pancreatic β-cells.

In response to the ingestion of food, the hormone glucagon-like peptide-1 (GLP-1) is secreted from enteroendocrine cells, then binds to its related receptor, and improves glucose-stimulated insulin secretion in pancreatic β-cells, ameliorating the survival of β-cells and suppressing apoptosis. In 2006, a new class of DPP-4 inhibitor was approved for the treatment of diabetic patients. DPP-4 inhibitors improve meal-stimulated insulin secretion by increasing the GLP-1 level and avoiding degradation by the enzyme DPP-4. CEL treatment, in a dose-dependent manner, was observed to significantly (p < 0.05) increase GLP-1 levels and, consequently, insulin secretion was consistent with the raised GLP-1 levels. These findings indicate that celery inhibits DPP-4 and increases GLP-1 half-lives, leading to the secretion of insulin from beta cells, which is the probable hypoglycemic mechanism of celery.

Our findings indicate that celery may protect pancreatic β-cells against glucotoxicity and H₂O₂ toxicity, displaying a significant effect on the recovery of mitochondrial function as a consequence of reducing oxidative stress. This leads to the diminution of apoptosis. In addition, celery increases insulin secretion and enhances insulin resistance in vitro through DPP-4 inhibition and GLP-1 activation, consequently protecting pancreatic β-cell function. This provides a possible mechanism for its antidiabetic effects. A. graveolens is a natural food that can be administered as a supplement to treat and prevent type 2 diabetes.

In conclusion, this research elucidated the promising hypoglycemic efficacy of Apium graveolens (celery) methanol extract through several mechanisms pertinent to diabetes management. Our findings highlighted celery's ability to enhance insulin secretion, augment pancreatic β-cell viability, and mitigate both glucotoxicity and H₂O₂-induced toxicity. Moreover, the inhibition of dipeptidyl peptidase-4 (DPP-4) activity by celery extract, resulting in elevated glucagon-like peptide-1 (GLP-1) levels, underscores its therapeutic potential in the regulation of glucose homeostasis. Additionally, the protective effect of celery on pancreatic β cells and its role in improving insulin sensitivity by reversing insulin resistance provide further evidence of its value in managing type 2 diabetes. These outcomes suggest that celery, rich in phenolic compounds as demonstrated through HPLC/MS analysis, offers an effective natural alternative for diabetes treatment, meriting further investigation in clinical settings. Given its accessibility and the rising interest in plant-based remedies, Apium graveolens emerges as a promising candidate for inclusion in diabetes management protocols, proposing a natural, adjunctive treatment option that complements existing therapies. This study paves the way for further research to explore the full potential of celery in diabetes care and encourages the integration of functional foods into therapeutic strategies for chronic metabolic disorders.

ATP - Adenosine Triphosphate, BSA - Bovine Serum Albumin, CEL - Celery Methanol Extract, DMEM - Dulbecco’s Modified Eagle Medium, DPP-4 - Dipeptidyl Peptidase-4, ELISA - Enzyme-Linked Immunosorbent Assay, FBS - Fetal Bovine Serum, GLP-1 - Glucagon-like Peptide-1, GSIS - Glucose-stimulated Insulin Secretion, H₂O₂ - Hydrogen Peroxide, HEPES - 4-(2-Hydroxyethyl)-1-piperazineethanesulfonic acid, HPLC/MS - High-performance liquid chromatography / mass spectrometry, INS-1 cells - A rat pancreatic β-cell line, KRB buffer - Krebs-Ringer Buffer, LDH - Lactate Dehydrogenase, LCA - Lithocholic Acid, Metf - Metformin, MTT - 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide, ROS - Reactive Oxygen Species, RPMI-1640 - Roswell Park Memorial Institute Medium

None.

RMPG and JMMF carried out all studies, edited figures and revised the manuscript, checked the published data. All authors approved the final manuscript.

None.

Data and materials used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Not applicable.

Not applicable.

The authors declare that they have no competing interests.