**Background:** Type 2 diabetes mellitus (T2DM) accounts for approximately 90% of diabetes cases worldwide and is characterized by hyperglycemia, insulin resistance, and dyslipidemia. Current anti-diabetic drugs have side effects such as hepatotoxicity and gastrointestinal symptoms, driving the search for natural alternatives. Marine microorganisms, particularly mangrove endophytic fungi, produce exopolysaccharides with novel structures and biological activities, but their anti-diabetic potential has been rarely studied. This study aimed to isolate, characterize, and evaluate the anti-diabetic activity of a novel exopolysaccharide from the mangrove endophytic fungus Penicillium janthinellum N29.
**Methods:** The exopolysaccharide PJ1-1 was isolated from 150 L of liquid culture broth of P. janthinellum N29 via ethanol precipitation, Q Sepharose Fast Flow ion-exchange chromatography, and Sephacryl S-100/HR gel filtration. Structural characterization was performed using high-performance gel permeation chromatography (HPGPC), reversed-phase HPLC, methylation analysis, and NMR spectroscopy (1H, 13C, COSY, TOCSY, HSQC, NOESY, HMBC). In vitro α-glucosidase inhibitory activity was assessed using p-nitrophenyl-α-d-glucopyranoside as substrate with acarbose as reference. For in vivo studies, four-week-old male C57BL/6J mice were fed a high-fat diet for 6 weeks followed by intraperitoneal STZ injection (30 mg/kg on days 1–3). Diabetic mice (FBG ≥ 11.1 mmol/L) were divided into five groups (n=6 per group): model control (MC, saline), positive control (PC, rosiglitazone 200 mg/kg/day), and three PJ1-1 groups (100, 200, or 400 mg/kg/day) via intragastric administration for 35 days. Outcomes measured included body weight, FBG (weekly), oral glucose tolerance test (OGTT), fasting insulin (ELISA), QUICKI, HOMA-IR, HOMA-β, and serum lipid profiles (TC, TG, LDL-C, HDL-C).
**Key Results:** PJ1-1 was a homogeneous galactomannan (molecular weight ~10.24 kDa, total sugar content 97.45%) composed of mannose (52.71%) and galactose (47.29%) in d-configuration. The backbone consisted of →2)-α-d-Manp-(1→, →4)-α-d-Manp-(1→, →3)-β-d-Galf-(1→, and →2)-β-d-Galf-(1→ units with branches at C-3 of →2)-β-d-Galf-(1→. In vitro, PJ1-1 inhibited α-glucosidase activity in a concentration-dependent manner with an inhibitory rate of 70.52% at 5 mg/mL. In vivo, after 5 weeks of treatment, FBG levels in the PJ1-1-H (400 mg/kg), PJ1-1-M (200 mg/kg), and PJ1-1-L (100 mg/kg) groups decreased by 42.16%, 29.10%, and 22.26%, respectively, compared to the MC group (p < 0.01). OGTT-AUC was significantly reduced in the PJ1-1-H group compared to MC (p < 0.05). Fasting insulin content significantly decreased in the PJ1-1-H group (p < 0.05 vs. MC). QUICKI index increased by 12.80%, 7.69%, and 7.66% in the PJ1-1-H, PJ1-1-M, and PJ1-1-L groups, respectively. HOMA-IR index decreased in a dose-dependent manner (p < 0.05 vs. MC). PJ1-1-H significantly reduced TG (p < 0.01), TC (dose-dependent), LDL-C (p < 0.01), and increased HDL-C (p < 0.05) compared to MC.
**Clinical Implications:** PJ1-1 demonstrates multi-faceted anti-diabetic activity in a T2DM mouse model, including blood glucose reduction, improved glucose tolerance, enhanced insulin sensitivity, and correction of dyslipidemia. These effects are comparable to rosiglitazone, a standard anti-diabetic drug. The structural novelty of PJ1-1, particularly its galactofuranose units and α-(1→4) glycosidic linkages, may contribute to its biological activity. While these preclinical results are promising, further investigation into the anti-diabetic mechanism of PJ1-1 is needed, and the findings require validation in human studies before clinical application can be considered.