Diabetes mellitus is a chronic metabolic disease caused by insufficient insulin secretion, insulin resistance, or a combination of both. Natural products possess diverse structural classes and multiple biological targets and can be used in studies of glucose-metabolism regulation, bioactive compound screening, and the discovery of lead structures for glucose-lowering drug development.

Keywords: diabetes mellitus; natural products; insulin resistance; pancreatic β cells; hepatic gluconeogenesis; glucose uptake; oxidative stress; gut microbiota; drug screening

1 Metabolic Abnormalities in Diabetes and the Research Value of Natural Products

1.1 Major Types and Pathological Basis of Diabetes

Diabetes mellitus mainly includes type 1 diabetes, type 2 diabetes, gestational diabetes, and other specific types. Type 1 diabetes is primarily caused by autoimmune destruction of pancreatic β cells and is characterized by absolute insulin deficiency. Type 2 diabetes is mainly characterized by insulin resistance and progressive loss of compensatory β-cell function. As insulin resistance persists, hepatic glucose output increases, glucose uptake by skeletal muscle and adipose tissue decreases, and pancreatic β cells gradually lose the capacity to secrete sufficient insulin to maintain normal blood glucose, ultimately resulting in persistent hyperglycemia.

1.2 Multiorgan Basis of Glucose-Metabolism Dysregulation

Type 2 diabetes is not caused by dysfunction of a single organ but results from metabolic imbalance involving the liver, skeletal muscle, adipose tissue, pancreatic islets, kidneys, intestine, and other systems. Enhanced hepatic gluconeogenesis and glycogenolysis increase endogenous glucose production. Insufficient GLUT4 translocation in skeletal muscle reduces postprandial glucose clearance. Increased lipolysis in adipose tissue elevates circulating free fatty acids and aggravates insulin resistance. Increased renal tubular glucose reabsorption, abnormal glucagon secretion, and impaired incretin effects also contribute to persistent hyperglycemia.

1.3 Diabetic Complications and Therapeutic Objectives

Long-term hyperglycemia can promote the accumulation of advanced glycation end products, oxidative stress, endothelial dysfunction, and chronic inflammation, thereby contributing to diabetic nephropathy, retinopathy, peripheral neuropathy, cardiovascular and cerebrovascular disease, diabetic foot complications, and impaired wound healing. Diabetes treatment should control both fasting and postprandial blood glucose while also considering hypoglycemia risk, body-weight changes, blood lipids, blood pressure, cardiorenal function, and pancreatic β-cell protection.

1.4 Major Sites of Action of Common Glucose-Lowering Drugs

Figure 1 Mechanisms of metformin regulation of hepatic glucose output and lipid metabolism

2 Major Structural Classes of Antidiabetic Natural Products

2.1 Flavones and Flavonols

Flavonoid natural products possess polyhydroxylated aromatic structures and can influence oxidative stress, inflammatory responses, glucose uptake, and insulin signaling. Compounds such as baicalein, quercetin, luteolin, and fisetin are used to investigate PI3K/AKT, AMPK, NF-κB, and Nrf2-related pathways and have shown effects on insulin resistance, high-glucose-induced injury, or pancreatic β-cell protection in different models. Because flavonoids generally exhibit multitarget activity, their glucose-lowering effects should be further validated through target intervention and metabolic endpoints.

2.2 Polyphenolic Compounds

Polyphenolic compounds such as curcumin, resveratrol, epigallocatechin gallate, chlorogenic acid, and gallic acid can regulate redox homeostasis, inflammatory signaling, mitochondrial metabolism, and energy-sensing pathways. Some polyphenols activate AMPK or Nrf2, suppress NF-κB and inflammatory cytokine expression, and improve glucose uptake and lipid metabolism in experimental models. However, poor water solubility, limited stability, and insufficient systemic exposure may restrict the translation of their in vitro activity.

2.3 Alkaloids

Alkaloids such as berberine can affect AMPK, mitochondrial energy metabolism, hepatic gluconeogenesis, lipid metabolism, and the gut microbiota. Berberine has relatively low oral absorption, and its metabolic effects may involve intestinal local activity, microbiota-mediated metabolism, and enterohepatic circulation. Its biological activity therefore cannot be evaluated solely on the basis of plasma drug concentrations. Studies should distinguish among the parent compound, active metabolites, and microbiota-mediated effects.

2.4 Terpenoids and Saponins

Terpenoids, triterpenoids, and saponins can regulate insulin signaling, mitochondrial function, adipose-tissue inflammation, and pancreatic β-cell stress. Ginsenosides, Astragaloside IV, ursolic acid, and oleanolic acid are commonly used in studies of metabolic inflammation, glucose uptake, and diabetic complications. Saponins differ markedly in glycosylation, absorption, and intestinal metabolism, and the effects of a total plant extract cannot be directly attributed to a single saponin component.

2.5 Plant Signaling Molecules and Natural Glycosides

Abscisic acid is an important signaling molecule in plants and can affect LANCL2, AMPK, PPARγ, and GLUT4-associated processes in mammalian experimental systems. Phlorizin is a natural dihydrochalcone glycoside that inhibits sodium-glucose cotransporters and provided a lead scaffold for the structural optimization of SGLT inhibitors. The research value of such natural compounds includes both direct metabolic regulation and their use as starting points for improving target selectivity and pharmacokinetic properties.

3 Major Mechanisms by Which Natural Products Regulate Glucose Metabolism

3.1 Improvement of Insulin Signaling and Glucose Uptake

After insulin binds its receptor, IRS, PI3K, and AKT signaling is activated, promoting GLUT4 translocation to the plasma membrane of skeletal muscle and adipose cells. During insulin resistance, activation of inflammatory kinases, accumulation of lipid metabolic intermediates, and abnormal IRS phosphorylation suppress this process. Some natural products can enhance AKT phosphorylation, activate AMPK, or promote GLUT4 membrane translocation, thereby increasing glucose uptake. Experimental validation should assess glucose uptake, membrane-localized GLUT4, and changes in key signaling proteins.

3.2 Suppression of Hepatic Gluconeogenesis and Glucose Output

Enhanced hepatic gluconeogenesis is an important cause of elevated fasting blood glucose. Phosphoenolpyruvate carboxykinase and glucose-6-phosphatase are commonly used evaluation markers. Natural products may regulate the expression of gluconeogenic genes through AMPK, FOXO1, CREB, and mitochondrial energy status and may also alter the metabolism of substrates such as lactate, pyruvate, and glycerol. A reduction in glucose levels in hepatocyte culture medium may also result from cellular injury or abnormal substrate consumption and should therefore be interpreted together with cell viability and gluconeogenic enzyme expression.

3.3 Protection of Pancreatic β Cells and Regulation of Insulin Secretion

High glucose, free fatty acids, reactive oxygen species, endoplasmic reticulum stress, and inflammatory cytokines can damage pancreatic β cells. Some natural products enhance antioxidant defenses, preserve mitochondrial function, and reduce inflammation, thereby improving β-cell survival and glucose-stimulated insulin secretion. β-cell protective effects should be evaluated using cell viability, intracellular insulin content, glucose-stimulated insulin secretion, and apoptosis markers. Insulin leakage caused by plasma membrane injury should not be misinterpreted as enhanced secretion.

3.4 Regulation of Oxidative Stress and Metabolic Inflammation

Persistent hyperglycemia and lipotoxicity increase mitochondrial reactive oxygen species, activate JNK, NF-κB, and inflammasomes, and further disrupt insulin receptor substrate and AKT signaling. Flavonoids, polyphenols, and alkaloids can regulate Nrf2-mediated antioxidant defenses and NF-κB, MAPK, and inflammatory cytokine expression. Antioxidant activity only indicates the potential to scavenge free radicals or modulate redox status. Improvement of glucose metabolism must still be confirmed through insulin signaling, glucose uptake, and metabolic endpoints in animal models.

3.5 Inhibition of Intestinal Carbohydrate Digestion and Absorption

α-Amylase and α-glucosidase participate in the digestion of dietary carbohydrates. Inhibition of these enzymes delays glucose release and reduces postprandial blood glucose. Many flavonoids, polyphenols, and saponins exhibit glycosidase-inhibitory activity in vitro. However, colorimetric and fluorescence assays may be affected by the intrinsic color, fluorescence, aggregation, or nonspecific protein binding of the test compound. Compound blanks, positive inhibitors, and orthogonal assay methods should therefore be included.

3.6 Regulation of the Gut Microbiota and Its Metabolites

Natural products can alter gut microbial composition, short-chain fatty acid production, bile acid conversion, and intestinal barrier function, thereby influencing systemic inflammation and insulin sensitivity. Some natural products with low oral absorption may act predominantly within the gastrointestinal tract. A causal relationship between microbiota changes and glucose-lowering effects cannot be established on the basis of correlation alone and may be investigated using antibiotic treatment, fecal microbiota transplantation, germ-free animals, or supplementation with defined microbial strains.

4 Representative Natural Products and Naturally Derived Glucose-Lowering Drugs

4.1 Galegine Scaffolds and Metformin

Galegine is a guanidine-containing natural product derived from plants of the genus Galega, and its chemical scaffold contributed to the development of biguanide glucose-lowering drugs. Metformin mainly lowers hepatic glucose output, alters cellular energy status, and improves insulin sensitivity. The relationship between natural lead structures and metformin does not represent simple direct extraction but rather a drug-development pathway involving chemical optimization, safety evaluation, and pharmacokinetic modification.

4.2 Phlorizin Scaffolds and SGLT2 Inhibitors

Phlorizin inhibits both SGLT1 and SGLT2, but its oral stability, selectivity, and absorption characteristics limit direct application. Dapagliflozin, empagliflozin, and other drugs developed on the basis of its glycoside scaffold and mechanism show greater SGLT2 selectivity and promote urinary glucose excretion by inhibiting glucose reabsorption in the renal proximal tubule. This process represents a typical pathway from natural-product activity discovery to structural modification and drug optimization.

4.3 Baicalein

Baicalein is a flavonoid natural product used to investigate oxidative stress, inflammation, and pancreatic β-cell injury under high-glucose or lipotoxic conditions. Its effects on glucose tolerance or pancreatic islet function may involve Nrf2, NF-κB, PI3K/AKT, and mitochondrial processes. However, the dominant target may differ according to the model and treatment conditions.

4.4 Curcumin

Curcumin is a natural polyphenol that regulates NF-κB, Nrf2, AMPK, and oxidative stress-related signaling. In experimental models, it is used to investigate metabolic inflammation, hepatic lipid accumulation, insulin sensitivity, and diabetic complications. Curcumin has low water solubility and is rapidly metabolized in vivo, so results obtained with high in vitro concentrations should be interpreted in the context of formulation type and systemic exposure.

4.5 Berberine

Berberine regulates hepatic gluconeogenesis, lipid metabolism, mitochondrial function, AMPK, and the gut microbiota and is widely used in studies of type 2 diabetes and metabolic syndrome. Its effects generally arise from multiple metabolic processes. Studies should avoid attributing all effects to a single target and should not infer its complete mechanism solely from changes in AMPK phosphorylation.

4.6 Abscisic Acid

Abscisic acid can be used to study LANCL2-associated signaling, AMPK activation, GLUT4 translocation, and glucose uptake. Different stereoisomers and formulations may have different activities, so the stereochemical configuration, purity, and treatment conditions should be clearly defined. Increased glucose utilization induced by abscisic acid should not be directly interpreted as increased insulin secretion from pancreatic β cells and should be distinguished using insulin measurements and receptor-intervention experiments.

5 Experimental Evaluation of the Antidiabetic Activity of Natural Products

5.1 In Vitro Cell Models

Hepatocytes can be used to evaluate glucose production, glycogen synthesis, and lipid accumulation. Skeletal muscle cells and adipocytes can be used to measure insulin signaling, GLUT4 translocation, and glucose uptake. Pancreatic β cells or isolated pancreatic islets can be used to evaluate insulin synthesis, glucose-stimulated insulin secretion, and cytoprotection. Experiments should include normal controls, model controls, positive-drug controls, and vehicle controls. Concentrations that cause clear cytotoxicity should not be used to explain glucose-lowering mechanisms.

5.2 Insulin-Resistance Models

High-concentration insulin, dexamethasone, palmitic acid, high glucose, or inflammatory cytokines can be used to induce different forms of insulin resistance. High-insulin models mainly simulate receptor-signaling desensitization, palmitic acid models emphasize lipotoxicity and inflammation, and dexamethasone alters glucose and lipid metabolism through the glucocorticoid receptor. Because the pathological basis differs among models, activity in a single model does not demonstrate applicability to all forms of insulin resistance.

5.3 Animal Models

Streptozotocin, alloxan, high-fat diets, and high-fat diets combined with low-dose streptozotocin are commonly used to establish experimental diabetes. Streptozotocin and alloxan primarily damage pancreatic β cells, high-fat diets mainly induce obesity and insulin resistance, and combination models reproduce both insulin resistance and declining β-cell function. Animal models should be selected according to the target mechanism, with clear reporting of species, dose, modeling period, and disease severity.

5.4 Metabolic and Tissue Evaluation Indicators

Antidiabetic activity can be evaluated through fasting blood glucose, random blood glucose, oral glucose tolerance tests, insulin tolerance tests, glycated hemoglobin, blood lipids, insulin levels, and insulin-resistance indices. Tissue-level measurements may include hepatic glycogen, gluconeogenic enzymes, GLUT4 translocation, pancreatic islet morphology, renal injury, and vascular function. Measurement of fasting blood glucose alone cannot distinguish among increased insulin secretion, improved insulin sensitivity, reduced intestinal absorption, and increased urinary glucose excretion.

5.5 Validation of Bioactive Components and Targets

Natural-product studies should preferentially use structurally defined compounds with controlled purity. Plant extracts and compound formulations should be standardized using chromatography, mass spectrometry, and chemical fingerprinting, and major active components should be identified through activity-guided fractionation. Molecular docking can only propose potential binding modes. Direct targets should be validated using enzymatic assays, binding experiments, gene knockdown, gene knockout, or target mutation.

6 Quality Control and Application Boundaries in Natural-Product Research

6.1 Standardization of Composition

Plant species, geographical origin, harvest time, storage conditions, and extraction procedures can all alter natural-product composition. Without standardization, results from different studies are difficult to compare. The source of raw materials, extraction solvent, concentration of major marker compounds, impurity profile, and stability should therefore be specified. For structurally defined compounds, purity, solvent, and storage conditions should also be recorded.

6.2 Bioavailability and Metabolism

Many flavonoids and polyphenols have low water solubility and may rapidly undergo glucuronidation, sulfation, or other metabolic transformations after oral administration. High concentrations used in vitro may not be achievable in vivo. Physiological relevance should therefore be assessed together with plasma and tissue exposure, active metabolites, and administration route. Improving solubility or altering delivery may change both activity and toxicity.

6.3 Safety and Drug Interactions

Natural origin does not imply low toxicity or absence of toxicity. Natural products may affect cytochrome P450 enzymes, drug transporters, and the absorption or metabolism of other glucose-lowering drugs. Combined use with insulin or sulfonylureas may also increase hypoglycemia risk. Studies should evaluate cytotoxicity, liver and kidney function, hematological indices, cardiac safety, and the effects of long-term administration.

6.4 Boundaries of Translational Research

Cellular and animal models can be used to identify activity and explain mechanisms but cannot replace clinical efficacy validation. Candidate natural products require compositional standardization, dose exploration, pharmacokinetic analysis, safety evaluation, and properly designed clinical studies before their value in diabetes prevention, adjunctive treatment, or management of complications can be established.

7 Antidiabetic Natural Products, Model Compounds, and Research Reagents

Research on the antidiabetic activity of natural products should simultaneously address chemical composition, metabolic effects, direct molecular targets, and in vivo exposure. Their glucose-lowering activity and drug-development value can be accurately evaluated only when the experimental model is appropriate, the chemical composition is clearly defined, and the proposed mechanism has been validated.

References

[1] Xu L, et al. Natural products for the treatment of type 2 diabetes mellitus: Pharmacology and mechanisms. Pharmacology Research. 2018;130:451-465.

[2] Beccuti G, et al. Timing of food intake: Sounding the alarm about metabolic impairments? A systematic review. Pharmacology Research. 2017;125(Pt B):132-141.

[3] Foretz M, et al. Metformin: From mechanisms of action to therapies. Cell Metabolism. 2014;20(6):953-966.

For more related articles, please see below:

[1] Establishment and Evaluation Methods of Diabetic Animal Models

[2] Animal model of diabetic nephropathy