Abstract Although Moringa oleifera, Vernonia amygdalina, and Centella asiatica are prominent tropical medicinal plants with various biological activities, the potential to allevate pancreatic oxidative stress and safety profile of a novel polyherbal formulation containing their leaf extracts have not been established. Therefore, this study was designed to evaluate the improvement in pancreatic oxidative stress and preclinical safety of this formulation. The pharmacological effects were evaluated in a streptozotocin-induced hyperglycemic mouse model. Preclinical safety was assessed according to OECD guidelines 423 (acute toxicity) and 407 (subchronic toxicity). After 7 days of treatment, the formulation (400 mg/kg) exhibited significant hypoglycemic activity. It also protected the pancreas against streptozotocin-induced oxidative stress by decreasing malondialdehyde (MDA) levels and increasing glutathione (GSH) concentrations in pancreatic tissue. In terms of safety, the formulation showed no acute toxicity, with an LD₅₀ value greater than 5,000 mg/kg. During the 56-day subchronic toxicity study, the formulation (at 400 and 800 mg/kg) caused no mortality or significant changes in behavior and body weight. However, it did alter some biochemical, hematological, and histopathological parameters. Taken together, these findings suggest that the polyherbal formulation of M. oleifera, V. amygdalina, and C. asiatica extracts is a promising candidate for further development as a novel anti-hyperglycemic agent, although careful monitoring of its long-term effects is warranted.

Keywords: Acute oral toxicity, Anti-hyperglycemia, Pancreatic protection, Polyherbal formulation, Subchronic toxicity

Citation:  Tran, T.H., Le, U.V., Nguyen, H.T.T., and Ly, T.H. 2026. Preclinical study on alleviating pancreatic oxidative stress and safety of novel anti-hyperglycemic polyherbal formulation. Natural and Life Sciences Communications. 25(4): e2026075.

Graphical Abstract:

INTRODUCTION

The World Health Organization estimates that approximately 80% of the population in developing countries relies on traditional medicine for alternative treatment and primary health care, largely due to its accessibility, cultural acceptance, and perceived safety (Al Akeel et al., 2018). However, the phytochemicals present in medicinal plants may exhibit toxicity, necessitating rigorous toxicological evaluation to establish safety profile, particularly for chronic administration (Jităreanu et al., 2023).

Moringa oleifera Lam. (family Moringaceae) is a tropical plant whose leaves are extensively utilized as both a nutritional source and a therapeutic agent. Rich in bioactive compounds, including vitamins (A, C), carotenoids, polyphenols, and various flavonoids (quercetin, isoquercetin, rutin, vitexin, kaempferol, apigenin, luteolin, myricetin,…), the leaves are traditionally employed to manage ailments ranging from infections and diabetes to chronic inflammatory diseases (Bhattacharya et al., 2018; Meireles et al., 2020). Regarding its safety, a 70% ethanol extract demonstrated a low acute toxicity profile (LD₅₀ > 2,000 mg/kg) in mice, although it was associated with mild anemia and moderate hepato-renal toxicity (Aliyu et al., 2021). Furthermore, while acute doses up to 5,000 mg/kg of leaf infusion or powder caused only transient behavioral changes, subchronic findings highlight preparation-dependent risks. Specifically, a 28-day administration of leaf infusion showed no adverse effects, whereas the powder form at 500–1,000 mg/kg induced hepatic and renal damage. Notably, no genotoxicity or mutagenicity was observed up to 2,000 mg/kg (de Barros et al., 2022). Collectively, these data suggest that while M. oleifera leaf extracts are generally safe in acute applications, long-term consumption of specific preparations, particularly the raw powder, may pose risks to hematological, hepatic, and renal functions.

Vernonia amygdalina Delile (Asteraceae), commonly known as bitter leaf, is a tropical shrub recognized for its significant nutritional and medicinal properties. Traditionally, leaf infusions have been used as an adjuvant therapy for diabetes and liver disorders, as well as for managing hypertension and respiratory infections such as pneumonia and tonsillitis (Asante and Wiafe, 2023; Degu et al., 2024). These therapeutic effects are attributed to a diverse array of phytochemicals, including polyphenols, saponins, sesquiterpene lactones (e.g., vernoniosides), and flavonoids (e.g., luteolin and its glycosides), which collectively exhibit antioxidant, anti-obesity, anti-diabetic, hepatoprotective, and antihypertensive activities (Ugbogu et al., 2021; Degu et al., 2024). Regarding its safety profile, studies indicate that V. amygdalina leaf extracts demonstrate low toxicity; no acute toxicity was observed at doses up to 3,200 mg/kg (Adedapo et al., 2014), and no adverse effects were reported in a 22-day subacute toxicity evaluation (Rachmaini et al., 2024).

Centella asiatica (L.) Urb., a tropical herbaceous plant in the Apiaceae family commonly known as gotu kola, is widely utilized as both a culinary vegetable and a medicinal herb. C. asiatica is recognized for a broad range of pharmacological effects, including prominent neuroprotective activities, such as cognitive enhancement, anxiolytic effects; as well as wound healing, antidiabetic, and hepatoprotective properties (Gohil et al., 2010; Sun et al., 2020). These activities are primarily attributed to its pentacyclic triterpenoids: the aglycones (asiatic acid, madecassic acid) and their respective glycosides (asiaticoside, madecassoside) (Randriamampionona et al., 2007; Sun et al., 2020). In terms of safety, C. asiatica is widely regarded as non-toxic, a conclusion supported by multiple toxicological assessments. Various extracts have demonstrated a very low acute oral toxicity, with LD₅₀ values exceeding 2,000 mg/kg, 4,000 mg/kg (Chauhan and Singh, 2012), and even reaching 10 g/kg in rat models (Chivapat and Tantisira, 2011). Sub-chronic studies further affirm its safety profile; for instance, administration of a standardized extract at doses up to 1,000 mg/kg for 90 days produced no significant toxicity. Although minor, gender-specific alterations in white blood cell counts and sodium levels were noted at the highest dose (Chivapat and Tantisira, 2011). Furthermore, the plant has shown no evidence of mutagenicity in rats (Deshpande et al., 2015).

Our previous work demonstrated that a polyherbal formulation comprising M. oleifera, V. amygdalina, and C. asiatica extracts exerts anti-hyperglycemic, hepato-renal protective effects, and improves pancreatic histology (Tran et al., 2024). While the safety of these herbs has been individually documented, the safety profile of their specific combination and the underlying mechanisms governing its protection against pancreatic oxidative stress remain to be elucidated. Therefore, the present study aimed to: (1) evaluate the protective effect of this formulation against pancreatic oxidative stress in a streptozotocin-induced hyperglycemic mouse model by assessing malondialdehyde (MDA) and glutathione (GSH) levels; and (2) establish its preclinical safety profile through comprehensive acute and subchronic oral toxicity assessments in mice. These findings are essential to validate the therapeutic potential and safety margins of the formulation for future clinical consideration.

MATERIALS AND METHODS

Materials

Moringa oleifera Lam. leaves were collected in An Giang Province, Vernonia amygdalina Del. leaves in Long An Province, and Centella asiatica (L.) Urb. aerial parts in Ho Chi Minh City, Southern Vietnam. The extracts, specifically M. oleifera leaf extract (MO), V. amygdalina leaf extract (VA), and C. asiatica aerial part extract (CA), were prepared as described in our previous studies (Le, 2022; Tran et al., 2024). Briefly, the dried powdered material of each plant was separately extracted via percolation using 45% (v/v) ethanol. The process involved a 24-h pre-extraction maceration, followed by percolation at a constant flow rate of 2 mL/min, maintaining a solid-to-solvent ratio of 1:20 (w/v). The resulting percolates were concentrated under reduced pressure at 60-70 °C to yield crude extracts, which were subsequently stored at 2-8 °C until further use. The final Polyherbal Formulation of Extracts (PFE) was developed by blending MO, VA, and CA in a 1:1:3 (w/w/w) ratio. This specific ratio was established based on its optimal anti-diabetic effect observed in our preliminary in vitro and in vivo screening (Le, 2022; Tran et al., 2024).

Experimental animals

Male and female Swiss albino mice (6–7 weeks old) were obtained from the Institute of Vaccines and Medical Biologicals (IVAC), Nha Trang, Vietnam. The animals were housed in polypropylene cages (33 cm × 21 cm × 15 cm) under standard laboratory conditions, including a controlled temperature of 25 ± 1°C, relative humidity of 65 ± 5%, and a 12-h light/dark cycle (lights on from 06:00 to 18:00).  The mice had ad libitum access to standard pellet chow (supplied by IVAC) and purified water. All experimental procedures were reviewed and approved by the Institutional Animal Ethics Committee of the National Institute of Medicinal Materials (NIMM), Hanoi, Vietnam (Protocol No. 05/2021-HĐ-ĐTCS-TTS). The study was conducted in strict accordance with national regulations (Decision No. 141/QĐ-K2ĐT, Ministry of Health, Vietnam) and international ethical principles for laboratory animal research, as outlined in the Guide for the Care and Use of Laboratory Animals (8^th Edition, 2011, National Research Council, USA).

Dose calculation

To evaluate the hypoglycemic efficacy and the potential mitigation of pancreatic oxidative stress, PFE doses of 200 and 400 mg/kg were selected based on the effective range identified in our previous research (Tran et al., 2024). Acute oral toxicity testing was conducted in accordance with the OECD Guideline 423, utilizing a stepwise dosing procedure (OECD, 2002). For the 56-day subchronic toxicity study, doses of 400 and 800 mg/kg/day were selected. The lower dose (400 mg/kg) corresponds to the established pharmacologically effective dose, while the higher dose (800 mg/kg) was chosen to assess the safety margin at twofold the effective dose (Tran et al., 2024). For all experiments, the PFE was suspended in distilled water. To ensure stability and homogeneity, the suspension was freshly prepared daily, immediately prior to oral administration by gavage. The administration volume for each mouse was individually calculated and adjusted based on the animal’s most recent body weight recording.

Evaluating anti-hyperglycemic effect and potential mitigation of pancreatic oxidative stress

Streptozotocin (STZ) was employed as a diabetogenic agent to induce hyperglycemia and pancreatic injury (Nindita et al., 2023). Following an overnight fast, mice received a single intraperitoneal (i.p.) injection of streptozotocin (STZ; Sigma-Aldrich, USA) at 170 mg/kg, freshly dissolved in cold 0.1 M sodium citrate buffer (pH 4.5) (Ly et al., 2024). To prevent initial hypoglycemic shock due to massive insulin release, animals were provided with a 5% glucose solution ad libitum for the subsequent 24 hours. Seven days post-STZ injection, mice with fasting plasma glucose (FPG) levels exceeding 200 mg/dL were confirmed as hyperglycemic and randomly assigned to five groups (n = 6 per group). The treatment regimens were as follows: 1) Normal Control: Received citrate buffer (i.p., once) and was orally administered distilled water for 7 days, 2) STZ-Control (Diabetic Control): Received STZ (i.p., once) and was orally administered distilled water for 7 days, 3) PFE 200: Received STZ (i.p., once) and was orally administered PFE at 200 mg/kg for 7 days, 4) PFE 400: Received STZ (i.p., once) and was orally administered PFE at 400 mg/kg for 7 days, 5) Reference: Received STZ (i.p., once) and was orally administered glibenclamide (supplied by Domesco Medical Import Export Joint Stock Corporation, Vietnam; GLI; 5 mg/kg) for 7 days. Treatments were administered daily for 7 consecutive days. On day 7, sixty minutes after the final dose, blood was collected from the tail vein of 12-h fasted mice for FPG measurement. Subsequently, animals were euthanized via an approved method, and pancreatic tissues were rapidly harvested for the quantification of malondialdehyde (MDA) and glutathione (GSH) levels to evaluate the PFE’s capacity to mitigate pancreatic oxidative stress.

Plasma glucose quantification

Mouse blood (10 μL) was put into 10 μL EDTA 2%, then centrifuged at 5,000 rpm/5 min to obtain 5 μL plasma, which was incubated gently with 250 μL test indicator agent (Erba, Germany) at 25°C for 10 mins. The absorbance (A) was measured at 500 nm. The level of glucose was estimated by the expression: Plasma glucose concentration (mg/dL) = (A_s – A_b)/(A₀ – A_b) × 100 × dilution. Where A_s, A_b, and A₀ are absorbance values of test, blank, and standard samples, respectively (Ly et al., 2024).

MDA and GSH quantification

MDA and GSH contents in the pancreatic tissue homogenate were determined to evaluate lipid peroxidative and endogenous antioxidant levels. The pancreatic tissues were homogenized in phosphate buffer (pH = 7.4) at a ratio of 1/2 (w/v) at 4°C for 2 mins. The tissue homogenate was incubated at 37°C for 60 mins. Then, the reaction was stopped by trichloroacetic acid. The reaction was centrifuged at 10,000 rcf at 4°C for 10 mins. For MDA quantification, 50 µL of the supernatant was mixed with 50 µL of 0.37% thiobarbituric acid (Sigma-Aldrich). The reaction was incubated at 90°C for 40 mins and cooled rapidly. The absorbance was measured at λ = 532 nm (Ly et al., 2024). For GSH quantification, 10 µL of the supernatant was mixed with 115 µL of phosphate buffer and 25 µL of DTNB reagent (Sigma-Aldrich). The reaction was kept at room temperature in dark for 5 mins. The absorbance was measured at λ = 412 nm (Ly et al., 2024). The content of MDA (µM/mg protein) and GSH (mM/mg protein) was calculated by the linear regression equation of a malondialdehyde tetrabutylammonium and L-glutathione reduced standard (Sigma-Aldrich), respectively.

Acute toxicity study

The acute oral toxicity of PFE was evaluated using the limit test procedure in accordance with OECD Guideline 423 (OECD, 2002) (Figure A1 in Supplementary file-SF). PFE was freshly prepared in distilled water at a constant dose volume of 0.2 mL/10 g. Following an overnight fast, mice were assigned to two groups: (I) Vehicle control group: Received distilled water (0.2 mL/10 g; n = 3 mice per sex) and (II) PFE-treated group: Received a single oral dose of 5,000 mg/kg (n = 6 mice per sex). This dose level complies with the OECD limit test criteria and represents the maximum feasible dose under national regulatory guidelines for herbal products in Vietnam. Following administration, animals were observed continuously for the first 30 min, periodically during the subsequent 4 h, and once daily thereafter for a total of 14 days. Monitoring focused on clinical manifestations of toxicity, behavioral changes, body weight fluctuations, and mortality. Any animal that succumbed during the study was subjected to immediate gross necropsy. On day 14, all surviving mice were euthanized, and major organs (liver, kidneys, heart, lungs, and spleen) were macroscopically examined for any pathological alterations.

Subchronic oral toxicity study

The subchronic toxicity study was conducted in accordance with OECD Guideline 407 (OECD, 2008), with minor modifications adapted to laboratory conditions (Figure A2, Supplementary File). Swiss albino mice were randomly allocated into three groups (n = 12 per group; 6 males, 6 females): a vehicle control group (distilled water, 10 mL/kg/day) and two treatment groups receiving PFE at 200 and 400 mg/kg/day, respectively. All substances were administered daily via oral gavage for 56 consecutive days.

Throughout the study, animals were monitored daily for clinical signs of toxicity, behavioral alterations, and mortality. Blood samples were collected from the tail vein and stored in EDTA-coated tubes for hematological and clinical biochemistry analysis at three time points: prior to dosing (Day 0), on Day 28, and on Day 56. To comply with ethical guidelines regarding blood volume, mice within each group were sub-divided for specific biochemical panels (e.g., liver or kidney function), while hematological parameters were assessed for all animals. On Day 57, all mice were euthanized for a comprehensive gross necropsy. Key organs, including the heart, liver, and kidneys, were harvested, examined for macroscopic abnormalities, and weighed to determine relative organ weights. Finally, the tissues were preserved in 10% neutral buffered formalin for subsequent histopathological examination.

Clinical observations and body weight monitoring

All mice were monitored daily for morbidity and mortality throughout the 56-day study period. Detailed clinical observations were conducted to assess overall health and neurobehavioral status, including changes in posture, gait, fur texture, skin integrity, and pupillary reflex. Additionally, the condition of mucous membranes, respiratory patterns, and responses to handling were recorded, alongside any occurrences of convulsions, stereotypy, or abnormal movements. Individual body weights were measured at monthly intervals using a calibrated digital balance to monitor growth patterns and allow for dose adjustments.

Hematological and biochemical analysis

Blood samples were collected on Days 0, 28, and 56 for comprehensive laboratory assessments. Hematological parameters were measured using an automated hematology analyzer and included: red blood cell count (RBC, 10⁶/µL), hematocrit (HCT, %), hemoglobin (HGB, g/dL), mean corpuscular volume (MCV, fL), mean corpuscular hemoglobin (MCH, pg), mean corpuscular hemoglobin concentration (MCHC, g/dL), red cell distribution width (RDW, %), white blood cell count (WBC, 10³/µL), and platelet count (PLT, 10³/µL). 

Clinical biochemistry profiles were assayed using an automated chemistry analyzer with commercial diagnostic kits (Khonsung et al., 2020). The evaluated panel comprised: alanine aminotransferase (ALT, U/L), aspartate aminotransferase (AST, U/L), triglycerides (TG, mg/dL), creatinine (mg/dL), urea (mg/dL), glucose (mg/dL), and total protein (TP, g/dL).

Gross necropsy and histopathological assessment

At the conclusion of the 56-day study, all surviving animals were weighed and then euthanized by CO₂ asphyxiation. A comprehensive gross necropsy was performed, involving a systematic examination of the external surfaces and all major internal organs-including the liver, kidneys, heart, brain, lungs, gastrointestinal tract, spleen, and reproductive organs. The heart, liver, and kidneys were carefully excised, trimmed of adipose tissue, and weighed to determine the absolute organ weights. The relative organ weight (ROW) was subsequently calculated using the following formula: ROW (%) = [Absolute Organ Weight (g) / Body Weight on Day 56 (g)] × 100. These representative organs were preserved in 10% neutral buffered formalin for histopathological evaluation. The fixed tissues were dehydrated through a graded series of alcohols, embedded in paraffin wax, and sectioned at a thickness of 5-µm. The sections were stained with hematoxylin and eosin (H&E) and subsequently examined under a light microscope by a pathologist to identify any structural or pathological alterations.

Statistical analysis

All data were expressed as the mean ± standard deviation (SD). Statistical analyses were performed using GraphPad Prism software (Version 8.0.2, La Jolla, CA, USA). Data normality was assessed using the Anderson-Darling test, and potential outliers were identified and addressed appropriately. For comparisons between two independent groups, an unpaired Student's t-test was employed. For comparisons involving multiple groups, one-way analysis of variance (ANOVA) was used, followed by Tukey’s or Dunnett’s post-hoc test for pairwise and control-group comparisons, respectively. Data from the 56-day subchronic toxicity study, involving multiple treatment groups across various time points, were analyzed using a two-way ANOVA. A p-value of less than 0.05 was considered statistically significant.

RESULTS

Hypoglycemic effect of PFE and its potential to alleviate pancreatic oxidative stress in hyperglycemic mice

After 7 days, the fasting plasma glucose level in the STZ-control group was markedly elevated compared to the normal control group (P < 0.0001) (Supplementary Figure A3). Additionally, the mortality rate of mice following STZ injection was approximately 30% throughout the study, consistent with previous reports (Tran et al., 2024). Initial animal numbers were adjusted to account for this mortality rate, ensuring sufficient statistical power. Daily oral administration of PFE for 7 days significantly reduced glucose level at both 200 mg/kg and 400 mg/kg doses (P < 0.0001). A similar potent hypoglycemic effect was observed in the glibenclamide-treated reference group (5 mg/kg, P < 0.0001 vs. STZ-control) (Figure 1a).

The induction of hyperglycemia was associated with significant pancreatic oxidative stress. Compared to the normal control group, the STZ-control group exhibited a significant increase in pancreatic MDA content (P = 0.0044) and a concomitant depletion of GSH levels (P = 0.0016). Treatment with PFE at 400 mg/kg effectively counteracted these changes, significantly lowering MDA content (P = 0.0254) and restoring GSH levels (P = 0.0121) relative to the STZ-control group. The reference drug, glibenclamide (5 mg/kg), also demonstrated a significant protective effect on these markers (P = 0.0429 for MDA; P = 0.0244 for GSH) (Figures 1b, 1c).

Figure 1. Hypoglycemic effect and the potential of alleviating pancreatic oxidative stress of PFE in STZ-induced oxidative damage mouse model. ^nsP > 0.05, *P < 0.05, **P < 0.01, and ****P < 0.0001 compared to STZ group by Tukey’s test (Mean ± SD, n = 6).

Acute oral toxicity of PFE

In the acute oral toxicity limit test conducted according to OECD Guideline 423, a single dose of PFE at 5,000 mg/kg resulted in no mortality in either male or female mice over the 14-day observation period. Throughout the study, no significant differences in body weight gain were observed between the PFE-treated and vehicle control groups (P > 0.05; Supplementary Table A1). Transient and mild clinical signs were observed shortly after administration, primarily in female mice, which exhibited a slight decrease in somatomotor activity, somnolence, and oral pruritus within the first 30-60 minutes. These signs resolved completely within a few hours, while male mice exhibited no abnormal clinical signs (Supplementary Table A2).

To further confirm this safety profile, a separate study was conducted at the maximum feasible dose (Dmax) of 16.13 g/kg. Consistent with the initial findings, this higher dose also resulted in 0% mortality and no significant signs of toxicity over 14 days. Over the 14-day follow-up, there were no abnormalities in behavior, fur appearance, food consumption, or excretion. Consequently, the median lethal dose (LD₅₀) of the PFE was established to be greater than 16.13 g/kg in mice, classifying it as a substance with a very low acute toxicity profile.

Subchronic oral toxicity of PFE

Behaviors and body weight

No treatment-related abnormalities in fur appearance, behavior, or locomotor activity were recorded in mice administered PFE at doses of 400 and 800 mg/kg. Furthermore, no mortality or significant clinical signs of toxicity were observed throughout the 56-day study period.

PFE treatment did not result in any significant effects on body weight in male mice. In females, all groups exhibited a normal pattern of weight gain over the study period. As displayed in Supplementary Table A3, the body weight of female mice in the 800 mg/kg group increased significantly by day 56 (29.83 g) compared to day 0 (23.00 g) (P = 0.0117), a trend consistent with the control group (29.33 g vs. 24.17 g; P = 0.0032). Importantly, there were no statistically significant differences in final body weights or overall weight gain between the PFE-treated female groups and the control group (P > 0.05).

Hematological parameters

Overall, most hematological parameters in PFE-treated animals remained within normal physiological ranges; however, several statistically significant alterations were observed in male mice (Table 1). On day 28, the white blood cell (WBC) count in the 400 mg/kg PFE group was significantly higher than in both the control (P = 0.0004) and the 800 mg/kg PFE (P = 0.0107) groups. By day 56, the 800 mg/kg PFE group exhibited a significantly lower hematocrit (HCT) value compared to the control (P = 0.0269) and the 400 mg/kg PFE (P = 0.0074) groups. Furthermore, significant differences were noted between the two PFE-treated groups (400 vs. 800 mg/kg) on day 56 regarding RBC (P = 0.0281), HCT (P = 0.0074), and Hb (P = 0.0134) parameters.

Time-course analysis within groups revealed that parameters in the 400 mg/kg PFE group were generally stable, except for a significant decrease in mean corpuscular volume (MCV) at days 28 (P = 0.0134) and 56 (P = 0.004) relative to the day 0 baseline. In contrast, the 800 mg/kg PFE group showed significant time-dependent reductions in several red blood cell indices by day 56. These included RBC (vs. day 0: P = 0.0008, vs. day 28: P < 0.0001), HCT (vs. day 0 and day 28: P < 0.0001), and Hb (vs. day 0 and day 28: P < 0.0001). Similar declining trends were observed for MCH (vs. day 0: P < 0.0001, vs. day 28: P = 0.0001), MCHC (vs. day 28: P < 0.0001), and RDW (vs. day 0: P = 0.0452, vs. day 28: P = 0.0003) by day 56.

Table 1. Effect of PFE (400 and 800 mg/kg) on the hematological parameters of male mice in subchronic toxicity study.

Note: Mean ± SD (n = 6), ^aP < 0.05 vs 0 day, ^bP < 0.05 vs 28^th day, ^cP < 0.05 vs 400 mg/kg at the same time, ^dP < 0.05 vs control at the same time (Tukey’s test).

Table 2 summarizes the hematological parameters for the three female mice groups. In inter-group comparisons at each time point revealed no statistically significant differences for most parameters. Notably, on day 28, the MCHC value of the 800 mg/kg PFE group was significantly different from both the control (P = 0.0326) and the 400 mg/kg PFE groups (P = 0.0211). By the end of the study (day 56), the RDW of the 400 mg/kg PFE group also showed a significant increase compared to the control group (P = 0.0035).

Regarding intra-group analysis (changes over time), the 400 mg/kg PFE group exhibited stable RBC, MCHC, and WBC counts throughout the study. Nevertheless, by day 56, this group showed significant reductions in several indices, including HCT (vs. day 0: P = 0.001; vs. day 28: P = 0.0042), Hb (vs. day 28: P = 0.0301), MCV (vs. day 0: P < 0.0001; vs. day 28: P = 0.0178), MCH (vs. day 0: P = 0.0114), and RDW (vs. day 0: P=0.0198). Similarly, in the 800 mg/kg PFE group, while RBC, HCT, Hb, and WBC parameters remained stable, there were significant decreases in MCV (vs. day 0: P = 0.0001) and MCH (vs. day 0: P = 0.0017; vs. day 28: P = 0.0106). Of note, both PFE-treated groups exhibited a significant elevation in PLT count by day 56 (vs. day 0: P < 0.01).

In summary, the 56-day administration of PFE induced several alterations in the hematological profile of Swiss albino mice.

Table 2. Effect of PFE (400 and 800 mg/kg) on the hematological parameters of female mice in subchronic toxicity study.

Note: Mean ± SD (n = 6), ^aP < 0.05 vs 0 day, ^bP < 0.05 vs 28^th day, ^cP < 0.05 vs 400 mg/kg at the same time, ^dP < 0.05 vs control at the same time (Tukey’s test).

Serum biochemical parameters

Table 3 summarizes the biochemical parameters for male mice. Regarding liver function, no significant differences in AST and ALT levels were observed between the PFE-treated and control groups at any time point. Although a slight upward trend in these enzymes was noted in the treated groups by day 56, it did not reach statistical significance (P > 0.05). Similar to the control group, serum protein content increased in both PFE-treated groups over the 56-day period. For kidney function, creatinine levels remained stable, showing no significant variation either between the groups or over time. However, by day 56, the urea level in the 800 mg/kg PFE group was significantly higher than that of the control group (P = 0.0062).

Analysis of other parameters revealed a significant intra-group increase in triglyceride levels in the 400 mg/kg PFE group by day 56 (P = 0.0035), whereas no significant changes were observed in the control or 800 mg/kg groups. Furthermore, while inter-group glucose concentrations remained comparable, a significant temporal reduction was observed within the 800 mg/kg PFE group by the end of the study (P < 0.05 vs day 28).

In summary, 56-day administration of PFE altered specific biochemical markers in male Swiss albino mice, notably inducing elevations in urea and triglyceride levels alongside a reduction in glucose levels at specific doses.

Table 3. Effect of PFE (400 and 800 mg/kg) on the biochemical parameters of male mice in subchronic toxicity study.

Note: Mean ± SD (n = 6), ^aP < 0.05 vs day 0, ^bP < 0.05 vs 28^th day, ^cP < 0.05 vs 400 mg/kg at the same time,  ^dP < 0.05 vs control at the same time (Tukey’s test).

Table 4 details the biochemical parameters for the female mice. Regarding liver function, the AST level in the 800 mg/kg PFE group was significantly elevated compared to the control group (P = 0.0006) by day 56. In contrast, ALT levels remained statistically comparable across all groups and time points. Total protein content exhibited a significant increase from baseline (day 0) in all groups at both day 28 and day 56.

In terms of kidney function, both PFE-treated groups showed a decreasing trend in creatinine and urea levels relative to their respective baseline values. However, inter-group analysis revealed that the urea concentrations in the 800 mg/kg group were significantly higher than in the 400 mg/kg group at both day 28 (P < 0.0001) and day 56 (P = 0.0017). Furthermore, urea levels in the 800 mg/kg group were significantly higher than the control group throughout the post-baseline observation period (P = 0.0023 at both time points).

Regarding other biochemical parameters, opposing trends were observed in triglyceride levels. In the PFE-treated groups, triglycerides significantly decreased at both 400 mg/kg (P = 0.0158 at day 28) and the 800 mg/kg (P = 0.005 at days 28 and 56), while an increasing trend was noted in the control group. Furthermore, glucose levels exhibited a temporal increase across all three groups by day 56 compared to their respective baselines.

Collectively, these results indicate that 56-day administration of PFE altered several biochemical markers in female Swiss albino mice. Most notably, the treatment induced an elevation in AST levels and significant fluctuations in urea and triglyceride concentrations.

Table 4. Effect of PFE (400 and 800 mg/kg) on the biochemical parameters of female mice in subchronic toxicity study.

Note: Mean ± SD (n = 6), ^aP < 0.05 vs day 0, ^bP < 0.05 vs 28^th day, ^cP < 0.05 vs 400 mg/kg at the same time, ^dP < 0.05 vs control at the same time (Tukey’s test).

Relative organ weight

Relative organ weights, including the heart, liver, and kidneys, showed no significant deviations from the control group following PFE administration at either 400 or 800 mg/kg (Supplementary Table A4).

Histopathology

Macroscopic examination revealed no abnormalities in the morphology or structure of the heart, liver, or kidneys in any group, regardless of sex.

Histologically, cardiac tissues from all groups appeared normal, with no evidence of vascular endothelial injury, myocardial interstitial fibrosis, epicardial inflammation, or congestion (Figure 2). Liver tissues from the control group were also histologically unremarkable, showing no signs of degeneration, necrosis of the central vein, portal or intermediate region, intrahepatic cholestasis, hemorrhage, portal fibrosis, or inflammation (Figure 3). In contrast, 83% of the PFE-treated mice exhibited mild histological alterations, characterized by mild chronic hepatitis accompanied by minimal focal microvesicular steatosis (at 800 mg/kg).

Regarding renal histopathology in male mice, the control group exhibited a baseline pathology of mild chronic interstitial nephritis; however, no signs of acute tubular injury—such as loss of the brush border, nuclear shrinkage, or vacuolization—were observed. Similarly, 50% of the PFE-treated mice presented with mild chronic interstitial nephritis, suggesting these findings may be incidental or strain-related rather than induced by PFE treatment (Figure 4).

Figure 2. Representative histopathological images of heart tissue of control and PFE treated groups at the doses (200× magnification, scale bars of 50 μm).

Figure 3. Representative histopathological images of liver tissue of control and PFE treated groups at the doses (200× magnification, scale bars of 50 μm). INF: Inflammation, SVD: Small vacuolar degeneration.

Figure 4. Representative histopathological images of kidney tissue of control and PFE treated groups at the doses (200× magnification, scale bars of 50 μm). INF: Inflammation.

DISCUSSION

Streptozotocin (STZ) is a well-established diabetogenic agent that induces hyperglycemia while concurrently generating severe oxidative stress in animal models. By promoting the formation of reactive oxygen and nitrogen species (ROS/RNS), STZ triggers mitochondrial dysfunction and lipid peroxidation, ultimately leading to organ damage, particularly in the pancreas (Goyal et al., 2016). In the present study, PFE (400 mg/kg) demonstrated a significant protective effect in the STZ-induced diabetic model. This was evidenced by its dual ability to lower blood glucose and shield the pancreas from oxidative damage, highlighted by a marked reduction in pancreatic MDA levels—a critical biomarker of lipid peroxidation.

Furthermore, the protective mechanism likely involves the glutathione antioxidant system. The balance between reduced glutathione (GSH) and its oxidized glutathione disulfide (GSSG) is crucial for cellular defense; this equilibrium is maintained by glutathione reductase (GR), which catalyzes the recovery of GSH from GSSG. Our results showed a significant increase in pancreatic GSH content following PFE treatment, suggesting that PFE may enhance the antioxidant capacity possibly by upregulating GR activity. This enhancement would facilitate the scavenging of reactive oxygen species and consequently inhibit MDA formation (Khelfi et al., 2023). Given that STZ selectively targets pancreatic beta cells via GLUT2 transporters, the observed increase in GSH levels suggests that PFE specifically fortifies these vulnerable cells against oxidative stress, thereby preserving their functional integrity.

These findings are consistent with our previous work (Tran et al., 2024), where a similar polyherbal formulation exhibited antihyperglycemic effects and preserved pancreatic histological structure. Although pancreatic histology was not performed in the current study, the strong correlation between our biochemical data and previous morphological findings provides a robust basis for these antioxidant claims. Collectively, this evidence supports the conclusion that PFE effectively mitigates STZ-induced oxidative stress. Future molecular studies focusing on specific signaling pathways, such as the Nrf2/ARE pathway, and other enzymes like superoxide dismutase (SOD), catalase (CAT), and glutathione peroxidase (GPx), are warranted to fully elucidate the extract's comprehensive mechanism of action.

The hypoglycemic and antioxidant properties of PFE are likely attributable to its diverse phytochemical profile, with each constituent herb contributing potent bioactive compounds. For instance, Moringa oleifera leaves are abundant in flavonoids such as rutin, quercetin, isoquercetin, kaempferol, apigenin, and myricetin (Bandopadhyay et al., 2018), while Centella asiatica is characterized by triterpenoids like asiatic acid, asiaticoside, and madecassoside (Bandopadhyay et al., 2023). Similarly, Vernonia amygdalina contains pharmacologically active compounds including luteolin and vernodalol (Djeujo et al., 2023). It is plausible that the therapeutic efficacy observed in this study arises from the synergistic interactions among these diverse compounds. However, the mild histological alterations noted during long-term administration suggests a complex interplay of effects that warrants caution. Consequently, further research is essential to isolate the principal active constituents, elucidate their specific molecular pathways, and conduct extended safety assessments. Such studies will be crucial to fully validate the therapeutic potential of this polyherbal formulation.

The therapeutic potential of the PFE can be further elucidated by examining the established molecular mechanisms of its constituent herbs and their primary phytochemicals. Existing literature provides substantial evidence supporting their roles in maintaining glucose homeostasis and promoting cellular protection. For instance, previous studies have demonstrated that the extract and asiatic acid from C. asiatica not only reduce blood glucose levels by improving insulin sensitivity but also mitigate cellular oxidative stress. These effects are mediated through the upregulation of key antioxidant enzymes, including GSH, GST (Glutathione-S-transferase), GPx, CAT, and SOD (Masola et al., 2018; Oyenihi et al., 2020). Furthermore, asiaticoside and madecassoside have been reported to exert potent cytoprotective activity by inhibiting the mitogen-activated protein kinase (MAPK) signaling pathway, thereby reducing inflammation and apoptosis under hyperglycemic conditions (Bandopadhyay et al., 2023).

The leaf extract of M. oleifera further contributes to the observed hypoglycemic effect, likely through its capacity to improve insulin sensitivity. This is achieved by stimulating the insulin-dependent Akt pathway and upregulating the expression of glucose transporter 4 (GLUT4) in skeletal muscle tissue (Attakpa et al., 2017). Notably, quercetin, a prominent flavonoid found in Moringa leaves, has been shown to activate AMPK (adenosine monophosphate-activated protein kinase). Activation of AMPK promotes glucose uptake via GLUT4 and suppresses hepatic gluconeogenesis by downregulating key enzymes, specifically phosphoenolpyruvate carboxykinase (PEPCK) and glucose-6-phosphatase (G6Pase) (Eid et al., 2015). Additionally, polyphenol-rich extracts from M. oleifera exhibit potent antioxidant properties by inhibiting protein oxidation and the formation of advanced glycation end products (AGEs), thereby enhancing cellular protection against chronic hyperglycemic damage (Nunthanawanich et al., 2016).

Similarly, V. amygdalina leaf extract has been shown to downregulate the expression of G6Pase and PEPCK in the hepatic tissue of diabetic mice. The extract also ameliorates insulin resistance by enhancing AMPK phosphorylation (Wu et al., 2018). Other studies have indicated that polyphenol-rich fractions of V. amygdalina effectively lower blood glucose, protect pancreatic pancreatic β-cells from oxidative stress, and upregulate GLUT4 expression in skeletal muscle (Ong et al., 2011).

The global utilization of herbal preparations in healthcare is expanding, with these remedies serving as essential dietary supplements, adjunctive polyherbal therapies, or alternatives to conventional medicine. However, herbal products still pose potential risks of toxicity due to their complex phytochemical profiles. These risks are further compounded in polyherbal formulations, where the vast array of constituents may lead to unpredictable herb-herb interactions, including pharmacological antagonism. Consequently, rigorous safety evaluations of herbal combinations remain a critical prerequisite for their clinical translation (Jităreanu et al., 2023). In alignment with these requirements, the present study evaluated the preclinical safety of a novel formulation comprising M. oleifera, V. amygdalina, and C. asiatica, building upon our prior evidence of its significant anti-hyperglycemic efficacy (Tran et al., 2024).

In the acute oral toxicity test, no mortality or signs of toxicity were observed in any group administered with PFE at doses up to 5,000 mg/kg body weight during the 14-day observation period. Consequently, the median lethal dose (LD₅₀) of the PFE was determined to be greater than 5,000 mg/kg. Transient, mild clinical signs exclusively observed in female mice at the 5,000 mg/kg dose. These signs included reduced somatomotor activity and lethargy within the first four hours post-administration. The animals subsequently recovered thereafter, exhibiting normal behavior for the remainder of the study. No such abnormalities were observed in male mice at the same dose. Furthermore, no statistically significant differences in body weight gain were observed between the PFE-treated and control groups (P > 0.05). All mice exhibited a steady increase in body weight, indicating that the PFE did not adversely affect their normal growth.

The acute oral toxicity study revealed a high safety profile for the PFE. Based on these findings, PFE is classified as Category 5 or 'unclassified' according to the Globally Harmonized System (GHS), indicating a very low potential for acute oral toxicity. This high safety profile is consistent with previous toxicological evaluations of the individual herbal components, where oral LD₅₀ values for M. oleifera, V. amygdalina, and C. asiatica extracts were reported to exceed 2,000 mg/kg (Deshpande et al., 2015; Moodley, 2017; Aliyu et al., 2021), or even 5,000 mg/kg (Zainul et al., 2016; de Barros et al., 2022), 3,200 mg/kg (Adedapo et al., 2014), and 10,000 mg/kg (Chivapat and Tantisira, 2011). To the best of our knowledge, the present study is the first acute toxicity evaluation of this specific polyherbal combination.

Interestingly, transient sedative effects were observed exclusively in female mice at the highest dose. This sex-specific sensitivity aligns with OECD guideline 423, which notes that female animals are often more susceptible to chemical-induced toxic effects. The observed sedation and reduced locomotor activity are likely attributable to the established central nervous system (CNS) depressant properties of the constituent herbs. Specifically, previous studies have documented the sedative effects of M. oleifera (Bakre et al., 2013; Pareek et al., 2023), V. amygdalina (Imoru et al., 2014), and particularly C. asiatica extract which is well-known for its anxiolytic and neuropharmacological activities (Kumar and Gupta, 2002; Gohil et al., 2010).

Furthermore, the absence of significant impacts on body weight supports the conclusion that PFE, at the doses tested, does not interfere with the animals' fundamental physiological or metabolic functions. Collectively, these results demonstrate that this polyherbal formulation possesses a high safety margin for acute oral administration. This lack of acute mortality and significant behavioral alterations provided a solid rationale for proceeding with the subchronic safety assessment over a 56-day period to evaluate long-term therapeutic efficacy.

Subchronic toxicity studies represent a critical component of preclinical safety assessment, typically conducted following preliminary acute toxicity and pharmacological screenings. These studies are essential for identifying potential health risks arising from prolonged repeated exposure to a substance and serve as a regulatory prerequisite for advancing investigational compounds into clinical trials (OECD, 2008). Consequently, this 56-day subchronic oral toxicity study was designed to further characterize the safety profile of PFE in mice. The dose levels (400 and 800 mg/kg) were selected based on our established therapeutic efficacy (Tran et al., 2024). Throughout the study period, daily administration of PFE did not elicit any observable signs of clinical toxicity, behavioral abnormalities, or mortality in either sex. Furthermore, no significant differences in body weight gain were observed compared to the vehicle control group, indicating that chronic PFE consumption did not adversely affect the animals' growth kinetics or general physiological state.

While the PFE did not alter general health indicators such as body weight or behavior, the hematological findings warrant careful interpretation. Analysis of hematological parameters revealed significant alterations primarily in male mice receiving 800 mg/kg PFE for 56 days. Specifically, this group exhibited a statistically significant reduction in red blood cell (RBC) count, hematocrit (HCT), hemoglobin (Hb), mean corpuscular volume (MCV), and mean corpuscular hemoglobin (MCH) relative to the control group (P < 0.05). The concurrent decline in RBC, Hb, and HCT is indicative of induced anemia. Furthermore, the corresponding decrease in MCV and MCH suggests a microcytic, hypochromic morphology, which is typically associated with impaired iron metabolism or defective hemoglobin synthesis. These findings align partially with previous reports on the individual constituents; for instance, M. oleifera hydroethanolic extracts have been noted to induce mild anemia in rats (Aliyu et al., 2021), while V. amygdalina has been shown to alter erythrocytic parameters (Shenkut, 2015). Consequently, the observed effects may result from an additive or synergistic interaction between these phytochemicals.

Additionally, the age of the animals must be considered a potential confounding factor. As the study utilized pre-adult mice, certain physiological shifts—such as increased platelet counts and serum protein levels—are characteristic of normal maturation. Although the anemia was significant compared to age-matched controls, these baseline physiological transitions due to maturation should be acknowledged. Further investigations, including serum iron analysis, ferritin levels, and bone marrow biopsies, are essential to elucidate the precise mechanism underlying these hematological changes.

Serum biochemical markers for hepatotoxicity and nephrotoxicity were evaluated following the 56-day treatment period. The subchronic toxicity assessment suggests that PFE possesses the potential for mild, sex-specific hepatotoxicity. In male mice, AST, ALT, creatinine, and urea levels remained statistically comparable to the control, despite subtle, non-significant upward trends in transaminase activities. Conversely, female mice treated with 800 mg/kg PFE exhibited a statistically significant elevation in AST levels (P < 0.05). While ALT levels in this group were also elevated, the increase did not reach statistical significance.

Histopathological examination further corroborated these biochemical findings; liver sections from both sexes treated with PFE displayed evidence of mild chronic hepatitis, characterized by inflammatory cell infiltration in the portal tracts and occasional focal necrosis. The significant rise in AST in high-dose females, coupled with these histological alterations, confirms that PFE can induce a mild inflammatory response in the hepatic parenchyma. AST serves as a sensitive biomarker for hepatocellular injury, and its elevation is a clear indicator of hepatic stress or damage. These results align with previous studies on the individual components. For instance, hydroethanolic extract of M. oleifera was reported to cause moderate liver injury, characterized by increased AST and creatine kinase levels, liver degeneration, necrosis, and non-obstructive sinusoidal dilatation in rats (Aliyu et al., 2021). It is plausible that the mild hepatotoxic effects observed in our study may be partially attributed, at least in part, to the M. oleifera component. The more pronounced response in female mice suggests a higher sensitivity to the formulation's constituents, a phenomenon that warrants further investigation into the underlying metabolic or hormonal mechanisms, particularly the basis for the observed sex-dependent differences.

The impact of 56-day repeated PFE administration on renal function was assessed through biochemical and histological analyses. No statistically significant changes were observed in serum creatinine levels in any PFE-treated groups, for both male and female mice, when compared to their respective controls. However, serum urea levels were altered in a dose- and time-dependent manner. In male mice, a significant elevation in urea concentration was observed in the 800 mg/kg PFE group by day 56 (P < 0.05). In female mice, a transient but significant urea increase was noted in the 800 mg/kg group at day 28, which subsequently normalized to levels comparable to the control by day 56. Histopathological examination corroborated these findings, revealing mild to moderate interstitial nephritis in approximately 50% of the PFE-treated animals, primarily characterized by focal inflammatory cell infiltrates within the renal interstitium. The findings of this study suggest that subchronic administration of the PFE may induce mild nephrotoxic effects, as evidenced by urea fluctuations and histopathological changes, despite stable creatinine levels.

Since creatinine is a robust indicator of the glomerular filtration rate (GFR), its stability suggests that PFE does not cause overt glomerular damage. Conversely, the urea elevation—particularly in high-dose males—points toward potential impairment of renal tubular function or altered protein metabolism. It is important to note that urea is synthesized in the liver from protein metabolism and excreted by the kidneys; thus, its levels can be influenced by both hepatic function and renal tubular handling, not just glomerular filtration. The histopathological observation of interstitial nephritis provides a structural basis for these biochemical changes. Inflammation in the renal interstitium can impair tubular function, potentially affecting urea transport and leading to its elevation in the blood without a corresponding rise in creatinine. The transient nature of the urea increase in female mice may reflect a physiological adaptation to prolonged exposure. These observations align with previous toxicological data on the formulation's components. Notably, M. oleifera leaf hydoethanolic extract was previously reported to cause moderate kidney injury by slightly increasing creatinine, tubular necrosis, interstitial nephritis, and renal interstitial edema (Aliyu et al., 2021). Therefore, the mild renal effects observed in our study could be attributed to constituents within the polyherbal formulation. Further investigations utilizing more sensitive biomarkers, KIM-1 (kidney injury molecule-1) or NGAL (neutrophil gelatinase-associated lipocalin) are warranted to fully elucidate the underlying mechanisms of the PFE.

Beyond the toxicological assessment, this study revealed potential metabolic benefits of PFE, particularly regarding lipid profiles in female mice. While male mice showed no significant alterations in triglyceride or glucose concentrations, female mice receiving the high dose (800 mg/kg) exhibited a statistically significant reduction in serum triglyceride levels. Hypertriglyceridemia is a well-established risk factor for cardiovascular disease; thus, this lipid-lowering effect suggests a therapeutic potential for PFE that extends beyond its previously reported anti-hyperglycemic activity (Tran et al., 2024). This effect is likely attributable to the synergistic actions of its phytochemical constituents, such as those from M. oleifera, which have been documented for their hypolipidemic properties (Meireles et al., 2020). Regarding blood glucose, there were no statistically significant differences in glucose levels between any of the PFE-treated groups and the control group at the final time point. An upward trend in glucose levels was noted across all female groups, including the control, over the 56-day period, suggesting a physiological shift rather than a treatment-induced effect. The lack of hyperglycemia or hypoglycemia compared to controls indicates that PFE does not adversely disrupt glycemic regulation in either sex. The observed sex-specific lipid-lowering effect aligns with other sex-dependent variations noted in this study (e.g., hepatotoxicity) and may be governed by hormonal influences on lipid metabolic pathways. Hormonal differences between sexes are known to play a crucial role in regulating lipid metabolism, which may explain the differential response. Further research is warranted to confirm these lipid-lowering findings and elucidate the mechanisms underlying this female-specific selectivity.

Collectively, the findings from this 56-day subchronic toxicity study raise safety concerns regarding the prolonged use of this polyherbal formulation (PFE), particularly concerning hematological, hepatic, and renal parameters. The observed toxicity may be substantially contributed by C. asiatica, a primary ingredient in the formulation. Although generally regarded as safe at traditional dosages, emerging evidence suggests potential toxicity associated with high-dose or chronic administration of C. asiatica. For instance, subchronic exposure to C. asiatica extract in rats has been shown to alter hematological indices and induce significant hepatorenal damage, as evidenced by both biochemical and histopathological data (Oruganti et al., 2010). Furthermore, recent studies have identified that its key triterpenoids, asiatic acid and madecassic acid, exhibit cardiotoxic potential by inducing apoptosis and oxidative stress (Guo et al., 2024). Asiatic acid and madecassic acid showed cardiotoxicity in zebrafish and induced death in H9C2 embryonic cardiomyocytes by altering the expression of apoptosis-related genes in cardiomyocytes. Furthermore, high concentrations of these two compounds increased systemic inflammation, neutrophil recruitment in the heart, and oxidative stress-induced damage (Guo et al., 2024). While the aerial part of C. asiatica is typically considered safe in acute, low-dose scenarios (Prasesti et al., 2021), 30-day repeated exposure at doses of 500–1,000 mg/kg has been linked to significant elevations in serum biomarkers (AST, ALT, BUN, and creatinine) and apoptosis index, and notably reduced viable cell count in liver and kidney tissues (Oruganti et al., 2010). Histopathological examination also showed significant liver and mild kidney damage These literature reports align closely with the hepatorenal and hematological effects identified in the present study, suggesting that the concentration of C. asiatica within the PFE may be a critical factor in its long-term safety profile.

Given these findings, a rigorous reassessment of the therapeutic dose is warranted. Although the 400 mg/kg dose achieves the desired pharmacological effect, its association with toxicological alterations suggests that a lower dose, such as 200 mg/kg, may offer a more favorable benefit-risk profile. Our previous findings confirms that the 200 mg/kg dose also exerts significant therapeutic efficacy (Tran et al., 2024), potentially representing a safer alternative for chronic administration.

As the first investigation into the subchronic safety of this specific PFE, further investigations are imperative to establish a clear safety margin. Future research should prioritize: (1) a comprehensive subchronic toxicity study at lower, clinically relevant doses (e.g., 200 mg/kg); and (2) reversibility studies incorporating a treatment-free recovery period to determine whether the observed adverse effects are transient. Such evidence will be crucial for guiding the safe clinical translation of this promising polyherbal formulation.

CONCLUSION

In conclusion, the polyherbal formulation (PFE) —comprising Moringa oleifera, Vernonia amygdalina, and Centella asiatica exhibits significant hypoglycemic and pancreatic protective effects by mitigating oxidative stress through the modulation of MDA and GSH levels. This study was conducted to evaluate its preclinical safety through acute and subchronic oral toxicity tests in Swiss albino mice. The acute toxicity assessment revealed a high safety margin, with a median lethal dose (LD₅₀) greater than 5,000 mg/kg. However, the 56-day subchronic assessment reveals the potential target-organ toxicity at higher doses. Repeated administration of PFE at 400 and 800 mg/kg induced significant alterations in hematological, hepatic, and renal biomarkers, corroborated by histopathological evidence of mild chronic damage in hepatic and renal tissues.

These findings underscore that both dosage and duration are critical determinants for the safe clinical application of PFE. While the formulation remains a promising therapeutic candidate, further investigations focusing on lower dose regimens (e.g., 200 mg/kg) and the reversibility of the observed adverse effects are essential to ensure its long-term safety and clinical viability.

ACKNOWLEDGEMENTS

We express our sincere gratitude to MSc. Ngo Thi Minh Huyen, the Research Center of Ginseng and Medicinal Materials Ho Chi Minh City, for timely authentication of herbs under study. We are also grateful to the Research Center of Ginseng and Medicinal Materials Ho Chi Minh City, National Institute of Medicinal Materials for their invaluable support and help in the successful conduction of this study.

AUTHOR CONTRIBUTIONS

Tran Huyen Tran: Conceptualization (Equal), Methodology (Equal), Investigation (Lead), Formal Analysis (Lead), Validation (Equal), Visualization (Lead), Writing – Original Draft (Lead), Writing – Review & Editing (Equal); Ut Van Le: Investigation (Equal), Methodology (Supporting), Visualization (Equal), Formal Analysis (Equal), Validation (Lead), Writing – Original Draft (Supporting), Writing – Review & Editing (Equal); Huong Thu Thi Nguyen: Conceptualization (Equal), Methodology (Equal), Visualization (Supporting), Formal Analysis (Supporting), Writing – Original Draft (Equal), Writing – Review & Editing (Lead), Supervision (Supporting); Trieu Hai Ly: Conceptualization (Lead), Methodology (Lead), Formal Analysis (Equal), Validation (Lead), Visualization (Equal), Resource (Lead); Writing – Original Draft (Lead), Writing – Review & Editing (Lead), Investigation (Equal), Supervision (Lead), Project Administration (Lead).

CONFLICT OF INTEREST

The authors declare that they have no conflicts of interest.

REFERENCES

Adedapo, A.A., Aremu, O.J., and Oyagbemi, A.A. 2014. Anti-oxidant, anti-inflammatory and antinociceptive properties of the acetone leaf extract of Vernonia amygdalina in some laboratory animals. Advanced Pharmaceutical Bulletin. 4(2): 591-598.  https://doi.org/10.5681/apb.2014.087

Al Akeel, M.M., Al Ghamdi, W.M., Al Habib, S., Koshm, M., and Al Otaibi, F. 2018. Herbal medicines: Saudi population knowledge, attitude, and practice at a glance. Journal of Family Medicine and Primary Care. 7(5): 865-875. https://doi.org/10.4103/jfmpc.jfmpc_315_17

Aliyu, A., Shaari, M.R., Ahmad Sayuti, N.S., Reduan, F.H., Sithambaram, S., Mohamed Mustapha, N., Shaari, K., and Hamzah, H.B. 2021. Moringa oleifera hydorethanolic leaf extract induced acute and sub-acute hepato-nephrotoxicity in female ICR-mice. Science Progress. 104(4): 368504211004272. https://doi.org/10.1177/00368504211004272

Asante, D.B. and Wiafe, G.A. 2023. Therapeutic benefit of Vernonia amygdalina in the treatment of diabetes and its associated complications in preclinical studies. Journal of Diabetes Research. 2023: 3159352. https://doi.org/10.1155/2023/3159352

Attakpa, E.S., Sangaré, M.M., Béhanzin, G.J., Ategbo, J.M., Seri, B., and Khan, N.A. 2017. Moringa olifeira Lam. Stimulates activation of the insulin-dependent Akt pathway. Antidiabetic effect in a diet-induced obesity (DIO) mouse model. Folia Biolica (Praha). 63(2): 42-51. https://doi.org/10.14712/fb2017063020042

Bakre, A.G., Aderibigbe, A.O., and Ademowo, O.G. 2013. Studies on neuropharmacological profile of ethanol extract of Moringa oleifera leaves in mice. Journal of Ethnopharmacology. 149(3): 783-789. https://doi.org/10.1016/j.jep.2013.08.006

Bandopadhyay, S., Mandal, S., Ghorai, M., Jha, N.K., Kumar, M., Radha Ghosh, A., Prockow, J., Perez de la Lastra, J.M., and Dey, A. 2023. Therapeutic properties and pharmacological activities of asiaticoside and madecassoside: A review. Journal of Cellular and Molecular Medicine. 27(5): 593-608. https://doi.org/10.1111/jcmm.17635

Bhattacharya, A., Tiwari, P., Sahu, P.K., and Kumar, S. 2018. A review of the phytochemical and pharmacological characteristics of Moringa oleifera. Journal of Pharmacy and Bioallied Sciences. 10(4): 181-191. https://doi.org/10.4103/jpbs.Jpbs_126_18

Chauhan, P.K. and Singh, V. 2012. Acute and subacute toxicity study of the acetone leaf extract of Centella asiatica in experimental animal models. Asian Pacific Journal of Tropical Biomedicine. 2(2): 511-513. https://doi.org/10.1016/S2221-1691(12)60263-9

Chivapat, S. and Tantisira, M. 2011. Acute and sub-chronic toxicity of a standardized extract of Centella asiatica ECa 233. Thai Journal of Pharmaceutical Sciences. 35: 55-64. https://doi.org/10.56808/3027-7922.2152

de Barros, M.C., Silva, A.G.B., Souza, T.G.dS., Chagas, C.A., Machado, J.C.B., Ferreira, M.R.A., Soares, L.A.L., Xavier, V.L., de Araújo, L.C.C., Borba, E.F.dO., et al. 2022. Evaluation of acute toxicity, 28-day repeated dose toxicity, and genotoxicity of Moringa oleifera leaves infusion and powder. Journal of Ethnopharmacology. 296: 115504. https://doi.org/10.1016/j.jep.2022.115504

Degu, S., Meresa, A., Animaw, Z., Jegnie, M., Asfaw, A., and Tegegn, G. 2024. Vernonia amygdalina: A comprehensive review of the nutritional makeup, traditional medicinal use, and pharmacology of isolated phytochemicals and compounds.  Frontiers in Natural Products. 3: 1347855. https://doi.org/10.3389/fntpr.2024.1347855

Deshpande, P.O., Mohan, V., and Thakurdesai, P. 2015. Preclinical safety assessment of standardized extract of Centella asiatica (L.) urban leaves. Toxicology International. 22(1): 10-20. https://doi.org/10.4103/0971-6580.172251

Djeujo, F.M., Stablum, V., Pangrazzi, E., Ragazzi, E., and Froldi, G. 2023. Luteolin and vernodalol as bioactive compounds of leaf and root Vernonia amygdalina extracts: Effects on α-glucosidase, glycation, ROS, cell viability, and in silico ADMET parameters. Pharmaceutics. 15(5): 1541. https://doi.org/10.3390/pharmaceutics15051541

Eid, H.M., Nachar, A., Thong, F., Sweeney, G., and Haddad, P.S. 2015. The molecular basis of the antidiabetic action of quercetin in cultured skeletal muscle cells and hepatocytes. Pharmacognosy Magazine. 11(41): 74-81. https://doi.org/10.4103/0973-1296.149708

Guo, Q., Li, Q., Liang, W., Zhang, Y., Jiang, C., Zhang, Y., Tan, J., and Zhao, H. 2024. Asiatic acid and madecassic acid cause cardiotoxicity via inflammation and production of excessive reactive oxygen species in zebrafish. Journal of Applied Toxicology. 44(7): 1028-1039. https://doi.org/10.1002/jat.4602

Gohil, K.J., Patel, J.A., and Gajjar, A.K. 2010. Pharmacological review on Centella asiatica: A potential herbal cure-all. Indian Journal of Pharmaceutical Sciences. 72(5): 546-556. https://doi.org/10.4103/0250-474x.78519

Goyal, S.N., Reddy, N.M., Patil, K.R., Nakhate, K.T., Ojha, S., Patil, C.R., and Agrawal, Y.O. 2016. Challenges and issues with streptozotocin-induced diabetes - a clinically relevant animal model to understand the diabetes pathogenesis and evaluate therapeutics. Chemico-Biological Interactions. 244: 49-63. https://doi.org/10.1016/j.cbi.2015.11.032

Imoru, J., Oyemitan, I., and Ilesanmi, O. 2014. Anxiolytic, sedative and hypothermic effects of aqueous leaf extract of Vernonia amygdalina Del. (Asteraceae) in albino mice. British Journal of Pharmaceutical Research. 4: 2210-2225. https://doi.org/10.9734/BJPR/2014/12529

Jităreanu, A., Trifan, A., Vieriu, M., Caba, I.C., Mârțu, I., and Agoroaei, L. 2023. Current trends in toxicity assessment of herbal medicines: A narrative review. Processes. 11(1): 83. https://doi.org/10.3390/pr11010083

Khelfi, S., Zerizer, S., Foughalia, A., Tebibel, S., Bensouici, C., and Kabouche, Z. 2023. The antioxidant activity and the anti-inflammatory effect of Citrus sinensis L. fruit on intestinal inflammation induced by hyperhomocysteinemia in mice. Natural Life Sciences Communications. 22(1): e2023009. https://doi.org/10.12982/NLSC.2023.009

Khonsung, P., Chiranthanut, N., Panthong, A., Chansakaow, S., Intahphuak, S., and Sireeratawong, S. 2020. Acute and ninety-day oral toxicity of the water extract from the fresh leaves of Pseuderanthemum palatiferum in rats. Chiang Mai University Journal of Natural Sciences. 19(4): 983-996. https://doi.org/10.12982/CMUJNS.2020.0061

Kumar, M.H.V. and Gupta, Y.K. 2002. Effect of different extracts of Centella asiatica on cognition and markers of oxidative stress in rats. Journal of Ethnopharmacology. 79(2): 253-260. https://doi.org/10.1016/s0378-8741(01)00394-4

Le, T.K.O., Nguyen, L.C., Ly, H.T., and Le, V.M. 2022. In vitro α-glucosidase inhibitory and antioxidant activities of combined extracts of Moringa oleifera, Vernonia amygdalina, and Centella asiatica. Hong Bang International University Journal of Science. 19: 68-78. 

Ly, H.T., Pham, K.D., Le, P.H., Do, T.H.T., Nguyen, T.T.H., and Le, V.M. 2024. Pharmacological properties of Ensete glaucum seed extract: Novel insights for antidiabetic effects via modulation of oxidative stress, inflammation, apoptosis and MAPK signaling pathways. Journal of Ethnopharmacology. 320: 117427. https://doi.org/10.1016/j.jep.2023.117427

Masola, B., Oguntibeju, O.O., Oyenihi, A.B. 2018. Centella asiatica ameliorates diabetes-induced stress in rat tissues via influences on antioxidants and inflammatory cytokines. Biomedicine & Pharmacotherapy. 101: 447-457. https://doi.org/10.1016/j.biopha.2018.02.115

Meireles, D., Gomes, J., Lopes, L., Hinzmann, M., and Machado, J. 2020. A review of properties, nutritional and pharmaceutical applications of Moringa oleifera: Integrative approach on conventional and traditional Asian medicine. Advances in Traditional Medicine. 20(4): 495-515. https://doi.org/10.1007/s13596-020-00468-0

Moodley, I. 2017. Acute toxicity of Moringa oleifera leaf powder in rats.  Journal of Medicinal Plants Studies. 5: 180-185.

Nindita, Y., Utomo, A.W., Maharani, N., Mahati, E., Kristiandi, I.F., Kesumayadi, I., Kurniawati, E.D., Sobirin, M.A., and Wijayahadi, N. 2023. Protective effect of Curcuma domestica and Curcuma xanthorrhiza extracts toward kidney, liver, and pancreatic organ dysfunction in streptozotocin-induced diabetes mellitus mice. Natural Life Sciences Communications. 22(2): e2023029. https://doi.org/10.12982/NLSC.2023.029

Nunthanawanich, P., Sompong, W., Sirikwanpong, S., Mäkynen, K., Adisakwattana, S., Dahlan, W., and Ngamukote, S. 2016. Moringa oleifera aqueous leaf extract inhibits reducing monosaccharide-induced protein glycation and oxidation of bovine serum albumin. Springer Plus. 5(1): 1098. https://doi.org/10.1186/s40064-016-2759-3

OECD. 2002. Test No. 423: Acute Oral Toxicity - Acute Toxic Class Method.

OECD. 2008. Test No. 407: Repeated Dose 28-day Oral Toxicity Study in Rodents.

Ong, K.W., Hsu, A., Song, L., Huang, D., and Tan, B.K. 2011. Polyphenols-rich Vernonia amygdalina shows anti-diabetic effects in streptozotocin-induced diabetic rats. Journal of Ethnopharmacology. 133(2): 598-607. https://doi.org/10.1016/j.jep.2010.10.046 

Oruganti, M., Roy, B.K., Singh, K.K., Prasad, R., and Kumar, S. 2010. Safety assemment of Centella asiatica in albino rats. Pharmacognosy Journal. 2(16): 5-13. https://doi.org/10.1016/S0975-3575(10)80043-4

Oyenihi, A.B., Opperman, M., Alabi, T.D., Mpahleni, B., and Masola, B. 2020. Centella asiatica alleviates diabetes-induced changes in fatty acid profile and oxidative damage in rat testis. Andrologia. 52(10): e13751. https://doi.org/10.1111/and.13751

Pareek, A., Pant, M., Gupta, M.M., Kashania, P., Ratan, Y., Jain, V., Pareek, A., and Chuturgoon, A.A. 2023. Moringa oleifera: An updated comprehensive review of its pharmacological activities, ethnomedicinal, phytopharmaceutical formulation, clinical, phytochemical, and toxicological aspects. International Journal of Molecular Sciences. 24(3): 2098. https://doi.org/10.3390/ijms24032098

Prasesti, G.K. and Kurniati, N.F. 2022. Toxicity studies of Centella asiatica for drug development: Mini review. Biointerface Research in Applied Chemistry. 12(6): 8081-8093. https://doi.org/10.33263/BRIAC126.80818093

Rachmaini, F., Abdillah, R., Anshari, H., and Juwita, D.A. 2024. Sub-acute toxicity study of bitter leaf (Vernonia amygdalina Del.) aqueous fraction on haematological parameters. Tropical Journal of Natural Product Research.  8(3): 6519-6524. https://doi.org/10.26538/tjnpr/v8i3.8

Randriamampionona, D., Diallo, B., Rakotoniriana, F., Rabemanantsoa, C., Cheuk, K., Corbisier, A.M., Mahillon, J., Ratsimamanga, S., and El Jaziri, M. 2007. Comparative analysis of active constituents in Centella asiatica samples from Madagascar: Application for ex situ conservation and clonal propagation. Fitoterapia. 78(7-8): 482-489. https://doi.org/10.1016/j.fitote.2007.03.016

Shenkut, M. 2015. Toxic effect of Vernonia amygdalina Delile on blood parameters and histopathology of liver and kidney in rats. Global Journal of Medicinal Plants Research. 1: 1-8.

Sun, B., Wu, L., Wu, Y., Zhang, C., Qin, L., Hayashi, M., Kudo, M., Gao, M., and Liu, T. 2020. Therapeutic potential of Centella asiatica and its triterpenes: A review. Frontiers in Pharmacology. 11: 568032. https://doi.org/10.3389/fphar.2020.568032

Tran, T.T.L., Ly, H.T., Le, T.K.O., and Le, V.M. 2024. Anti-hyperglycemic effect of herbal formula of Moringa oleifera, Vernonia amygdalina and Centella asiatica extracts in streptozotocin-induced hyperglycemic mice. Pharmacological Research - Modern Chinese Medicine. 11: 100428. https://doi.org/10.1016/j.prmcm.2024.100428

Ugbogu, E.A., Emmanuel, O., Dike, E.D., Agi, G.O., Ugbogu, O.C., Ibe, C., and Iweala, E.J. 2021. The phytochemistry, ethnobotanical, and pharmacological potentials of the medicinal plant-Vernonia amygdalina L. (bitter leaf). Clinical Complementary Medicine and Pharmacology. 1(1): 100006. https://doi.org/10.1016/j.ccmp.2021.100006

Wu, X.M., Ren, T., Liu, J.F., Liu, Y.J., Yang, L.C., and Jin, X. 2018. Vernonia amygdalina Delile extract inhibits the hepatic gluconeogenesis through the activation of adenosine-5'monophosph kinase. Biomedicine & Pharmacotherapy. 103: 1384-1391. https://doi.org/10.1016/j.biopha.2018.04.135

Zainul, N., Zakaria, Y., Nik Hassan, N.F., and Muhammad, H. 2016. Phytochemicals and acute oral toxicity studies of the aqueous extract of Vernonia amygdalina from state of Malaysia. Journal of Medicinal Plants Studies. 4(3): 1-5.       

OPEN access freely available online

Natural and Life Sciences Communications

Chiang Mai University, Thailand. https://cmuj.cmu.ac.th

Supplementary

Preclinical study on alleviating pancreatic oxidative stress and safety of novel anti-hyperglycemic polyherbal formulation.

Figure A1. Experimental design of acute oral toxicity study according to OECD No. 423.

Figure A2. Experimental design of sub-acute oral toxicity study according to OECD 407.

Figure A3. STZ-induced hyperglycemic mice. ^nsP > 0.05 and ****P < 0.0001 by Tukey’s test (Mean ± SD, n = 6).

Table A1. Effect of 5,000 mg/kg PFE on body weight of mice in acute toxicity study.

Note: PFE: Polyherbal Formula Extract, Mean ± SD (n = 6), ^nsP > 0.05 compared to the control at the same time by t-test.

Table A2. Behavioral patterns of female mice in 5,000 mg/kg PFE treated group compared to control group.

Note: CT: Control, PFE: Polyherbal Formula Extract, N = Normal, ↓ = Decreased, F = Found, N.F = Not found.

Table A3. Effect of PFE (400 and 800 mg/kg) on body weight of mice in subchronic toxicity study.

Note: Mean ± SD (n = 6), ^aP < 0.05 vs day 0, ^bP < 0.05 vs 28^th day (Tukey’s test).

Table A4. Effect of PFE (400 and 800 mg/kg) on the relative weight of mouse organs in subchronic toxicity study.

Note: Mean ± SD (n = 6).