Diabetes has always been a matter of concern to health experts as well as the community due to the increasing number of patients with diabetes and the severe consequences it may cause. Many attempts have been made to discover new treatment options for diabetes, and herbal medicines are currently considered to have great potential. This study was conducted to evaluate the effect of Camellia flava flower extract on the degeneration of the islets of Langerhans and insulin resistance in an alloxan-induced hyperglycemia model in Swiss albino mice. Hyperglycemic conditions were induced by alloxan (55 mg/kg, i.v.). The animals were then treated with glibenclamide (10 mg/kg, p.o.) and flower extract at doses of 1.09 and 2.19 g/kg, p.o. The results showed that the blood glucose, AUC, HbA1c, and HOMA-IR levels of two groups of mice receiving flower extract were considerably lower than those of the hyperglycemic untreated group (p < 0.05). The body weights of these two groups were also lower than the untreated group on the last day of the experiment, though the differences were not significant (p > 0.05). However, this was not observed when assessing insulin levels as well as relative organ weights. In biochemical tests, creatinine and AST and ALT concentrations were evaluated. There was no significant variation in creatinine and AST concentrations between the five experimental groups, whereas mice treated with glibenclamide and flower extract at both doses showed a remarkable decline in ALT concentration (p < 0.05). The hepatic histomicrographs were consistent with ALT results, while the H&E staining of kidneys showed no difference between groups. Histomicrographs of the pancreas revealed that the treatment groups using glibenclamide and flower extract had larger islets of Langerhans than those of the alloxan-treated group. Based on these results, this study demonstrated that Camellia flava flower extract exerted several beneficial effects, including blood sugar level reduction, weight loss promotion, and organ protection, hence making it a new potential herbal medication for the management of diabetes.

alloxan, Camellia flava, hypoglycemic effect, islets of Langerhans, insulin resistance

Diabetes is a chronic disease that occurs when the body can neither synthesize enough insulin nor use insulin effectively (WHO, 2023). The lack of insulin results in sugar not being consumed by the tissues, thus leading to high blood glucose levels. Patients with diabetes often experience many dangerous complications, both chronic and acute, such as macro- and microvascular diseases, diabetic ketoacidosis (DKA), and hyperosmolar hyperglycemic state (HHS). As of 2019, according to the World Health Organization (WHO), there were approximately 422 million people worldwide suffering from diabetes, with the majority coming from low- and middle-income nations (WHO 2023). The incidence of diabetes was 9.3% of the global population in 2019, and this figure is predicted to rise to 10.9% in 2045 (Saeedi et al. 2019). In Vietnam, there were about 5 million people with diabetes in 2021, making up 7.1% of the adult population. The proportion of diabetic patients in Vietnam has nearly doubled over the last 10 years, and it is estimated that one in every 20 Vietnamese adults is living with diabetes (WHO 2016).

Due to this increase in the incidence of diabetes worldwide, there have been many attempts at discovering and developing new medications that help with the treatment of diabetes. According to the American Diabetes Association 2023, there are seven groups of medications approved to be used in curing diabetes, including metformin, iSGLT2 (sodium-glucose transporter 2 inhibitors), GLP-1 RAs (glucagon-like peptide 1 receptor agonists), iDPP-4 (dipeptidyl peptidase 4 inhibitors), thiazolidinediones, sulfonylureas, and insulin. All of the above-mentioned medications have proved to be effective in the management of diabetes, but they also carry some mild to severe side effects such as renal toxicity, heart failure, and obesity (ElSayed et al. 2023). Many herbal medicines or medicinal plants are remarkably considered an alternative therapy to pharmaceutical drugs because of their significant potential in the treatment of diabetes and fewer adverse effects.

Camellia flava (Pitard) Sealy, one golden camellia species belonging to the Theaceae family, is mainly distributed in Asian countries such as China, India, and Vietnam. It has been reported to contain several phytochemical constituents, such as polysaccharides, phenolics, saponins, flavonoids, and so on. These compounds exhibited various pharmacological effects, including antioxidant, antimicrobial, antihyperglycemic, and metabolic balance (Hakoda and Ninh 2001).

With Camellia flava being recognized as a potential herbal medicine for diabetes treatment, there is growing interest in investigating the pharmacological effects of this plant. In our preliminary study, we have already evaluated the acute toxicity and hypoglycemic effect of Camellia flava flower extract on 60 mg/kg, i.v., alloxan-induced hyperglycemic mice (Nguyen et al. 2023). The mice were treated with flower extract orally at 4 different doses of 0.22, 0.66, 1.09, and 2.19 g/kg. However, only the mice receiving treatment at doses of 1.09 and 2.19 g/kg showed markedly hypoglycemic signs. Therefore, following this study, we decided to further investigate the effect of Camellia flava flower extract at these 2 doses (1.09 and 2.19 g/kg) on the degeneration of the islets of Langerhans and insulin resistance in a hyperglycemia model induced by alloxan 55 mg/kg, i.v., on Swiss albino mice.

Flavonoids are a widely found group of natural compounds with numerous phytochemical constituents that have been successfully extracted, isolated, and structurally identified. Moreover, this group of compounds is primarily responsible for many biological effects such as antitumor, antioxidant, antimicrobial, anti-inflammatory, antihypertensive, anti-depressant, and, notably, hypoglycemic effects (Rana and Gulliya 2019). Several studies have reported the blood glucose-lowering effects and reduction of diabetic complications of flavonoid compounds such as (i) apigenin, baicalein, and catechin (lowering blood glucose through antioxidant processes); (ii) hesperidin and myricetin (improving neuropathic complications); (iii) glycyrrhizin (improving gestational diabetes); (iv) quercetin (reducing complications on the kidneys); (v) kaempferol and puerarin (beneficial for cardiovascular diseases); and (vi) dihydromyricetin (improving cognitive impairment caused by diabetes) (Bai et al. 2019).

Golden camellia is a precious and expensive group of medicinal herbs belonging to the Camellia genus, and it is known as the “Queen of Tea” in Vietnam. Unlike green tea (Camellia sinensis), which is well studied, there is limited comprehensive research on the chemical composition and pharmacological effects of golden camellia species, particularly Camellia flava, both in Vietnam and worldwide (Chaudhary et al. 2023; Wei et al. 2024). Camellia flava is a native golden camellia species that thrives mainly in Vietnam (Hakoda and Ninh 2001; Beech et al. 2017; POWO). It has a large biomass and is widely used in the treatment of various ailments based on folk knowledge. In previous studies, we conducted preliminary research on Camellia flava, including plant identification, quantification of certain compound groups, and evaluation of its biological effects, particularly its hypoglycemic effects (Truong et al. 2022). Therefore, we have chosen this particular species for further in-depth research on the mechanisms of its hypoglycemic properties.

The Camellia flava species has been used in traditional medicine for its health benefits, and many could be deemed non-toxic according to the GHS classification. Likewise, our previous study recorded no signs of acute toxicity at the dose of 11.75 g/kg body weight of mice (Nguyen et al. 2023). Subsequently, mice were treated with flower extract orally at 4 different doses of 0.22, 0.66, 1.09, and 2.19 g/kg. However, only the mice receiving treatment at doses of 1.09 and 2.19 g/kg showed markedly hypoglycemic signs. Therefore, following this study, we decided to further investigate the effect of Camellia flava flower extract at these 2 doses (1.09 and 2.19 g/kg) on the degeneration of the islets of Langerhans and insulin resistance in a hyperglycemia model induced by alloxan 55 mg/kg, i.v., on Swiss albino mice.

The determination of total flavonoid content using an aluminum chloride reagent was a common quantitative method widely applied in flavonoid and herbal medicine research. This reagent reacted with neighboring hydroxyl and/or carbonyl groups in the structure of flavonoids, forming colored complex products with maximum light absorption at a specific wavelength. The quantification results were typically expressed as equivalent amounts of reference substances in amounts of dry material or extract, in terms of rutin and quercetin as two commonly used standard compounds (Shraim et al. 2021).

Rutin is a flavonoid glycoside compound with a flavonol aglycone structure that has been reported to exhibit various pharmacological effects, including anti-inflammatory, antioxidant, neuroprotective, hepatoprotective, and hypoglycemic activities. Several mechanisms of rutin’s blood glucose-lowering effects have been studied and proposed, including reducing carbohydrate absorption in the small intestine, inhibiting gluconeogenesis and enhancing glucose uptake in the tissue, stimulating insulin secretion in pancreatic β-cells, and preventing degeneration of the islets of Langerhans (Ghorbani 2017). Additionally, using modern LC-MS/MS analysis techniques, rutin has been detected in other golden camellia species such as Camellia euphlebia and Camellia nitidissima (He et al. 2019; Zhang et al. 2020). Therefore, we chose rutin as the reference compound to express the total flavonoid content. The relatively high flavonoid content in the flower extract of Camellia flava (333.58 ± 2.69 mg RE/g dry extract) indicated the potential of this golden camellia species to exhibit multiple biological effects, particularly its anti-hyperglycemic effect.

Hyperglycemic conditions in mice could be induced using several methods, including a high-fat diet (Surwit et al. 1988; Winzell and Ahrén 2004), streptozotocin (Junod et al. 1967), alloxan (Shaw Dunn and McLetchie 1943), etc. Among these methods, streptozotocin and alloxan are the most preferable diabetogenic agents to be used in laboratories (Szkudelski 2001; Lenzen 2008). Regarding efficacy, both could be utilized for induction of type 1 and type 2 diabetes, and their hyperglycemic effects last for 2–4 weeks or even longer after administration. In addition, alloxan is also more affordable and available than streptozotocin (Ighodaro et al. 2017). As a result, alloxan was chosen for hyperglycemic induction in our study.

Alloxan-induced hyperglycemia models commonly have a mortality rate of 20–90% depending on the dose, route of administration, and speed of injection (Bukhari et al. 2015; Bacevic et al. 2020). In the preliminary study, we induced hyperglycemia in mice by injecting alloxan (60 mg/kg, i.v.) (Nguyen et al. 2023). However, a high mortality rate in mice was recorded at the end of our study (data not shown). Therefore, in this study, we decided to evaluate two more different routes of injection and doses (55 mg/kg, i.v., and 150 mg/kg i.p.), which are commonly used for alloxan (Radenković et al. 2016), to determine the appropriate route that may produce a low mortality rate while maintaining stability in blood glucose levels during the experimental period. When administered intravenously, alloxan is delivered directly into the systemic blood circulation and rapidly achieves high concentrations in the pancreas. This may contribute to the fact that the dosage for intravenous injection of alloxan in mice is normally lower than the intraperitoneal route (Radenković et al. 2016). The 150 mg/kg i.p. group showed a rising tendency in blood glucose level throughout 2 weeks (Fig. 3A), which led to a higher mortality rate compared to the 55 mg/kg i.v. group (Fig. 3C). Therefore, injecting alloxan at 55 mg/kg i.v. produced a high survival rate and percentage of mice that met the criteria, as well as stable blood glucose levels.

As mentioned above, only 2 high doses in 4 doses of Camellia flava flower extract (1.09 and 2.19 g/kg) exhibited hypoglycemic effects after 2 weeks of experimentation (Nguyen et al. 2023). This result aligned with several studies on other medicinal herbs rich in flavonoids, which demonstrated a change in blood glucose level after 2 weeks (Tafesse et al. 2017; Yin et al. 2018). It was also suitable for the alloxan-induced diabetes model, which is often conducted within less than a month (Ma et al. 2015; Tafesse et al. 2017; Yin et al. 2018), as the high rate of mortality is a major drawback of this agent (Ighodaro et al. 2017), especially in the untreated group of animals. This is deemed a limitation in inducing hyperglycemia using chemical compounds. When the long-term effects of hyperglycemia are investigated, genetically engineered mice could be taken into consideration (Brockman et al. 2013; Yao et al. 2014; Wang and Zhu 2016). Notwithstanding, this study was merely pioneering pharmacological research on the hypoglycemic activity of Camellia flava flower extract. More thorough investigations into this plant are necessary to evaluate its long-term therapeutic effects.

The ability of Camellia flava flower extract to lower blood glucose levels in alloxan-induced hyperglycemic mice may be due to the antioxidant activity of the flavonoid content of Camellia flava (Fig. 2B). The antioxidant activity of Camellia flava flower extract was evaluated using the DPPH free radical scavenging model with ascorbic acid as a positive control. Ascorbic acid is mostly known for its antioxidant properties and plays a vital role in maintaining several functions of the human body (Gęgotek and Skrzydlewska 2022). Fig. 2 demonstrates the IC₅₀ values of ascorbic acid and Camellia flava flower extract using 70% ethanol as solvent. Interestingly, the result of the flower extract was merely three times lower than that of ascorbic acid (2.114 μg.mL^-1 compared to 6.793 μg.mL^-1). We also evaluated the antioxidant effect of this plant extracted with 50% and 96% ethanol. The IC₅₀ results were 12.26 μg.mL^-1 and 10.19 μg.mL^-1, respectively (Suppl. material 1: fig. S1). It showed that the flower extract using 70% ethanol brought about the highest antioxidant effect among the three solvents. There have been numerous reports on the antioxidant properties of many species in the Camellia genus. The study of Piao et al. (2011) showed that the ethanol extract of Camellia japonica flower scavenged 49% DPPH free radical at 12.5 μg.mL^-1 (Piao et al. 2011). Therefore, the antioxidant effect of Camellia flava flower extract was much higher, which may contribute to the underlying mechanism by protecting pancreatic β-cells from toxic ROS, hence enhancing insulin sensitivity in tissues.

In addition, Camellia flava has also been proven to possess α-amylase inhibitory activity in vitro, which may support its hypoglycemic effect (Truong et al. 2022). Apart from these suggested mechanisms, the main mechanism of hypoglycemia in Camellia flava flower extract still remains unknown. Another well-known plant in the same Camellia genus (Camellia sinensis) has been reported to have an antihyperglycemic effect with several mechanisms, such as stimulating insulin secretion and serving as a DPP-4 enzyme inhibitor or GLP-1 agonist (Ansari et al. 2022). This may further emphasize the potential of Camellia flava as a new antidiabetic drug to be developed in the future.

Alloxan exerts its effect via two mechanisms of action. It has a similar structure to glucose and is therefore selectively transported into the β-cell via the GLUT2 glucose transporter. Subsequently, it generates reactive oxygen species (Radenković et al.), leading to the necrosis of β-cells. In addition, alloxan also inhibits insulin secretion through the inhibition of glucokinase, which results in a high blood glucose level due to hypoinsulinemia (Lenzen 2008). The mice injected with alloxan and remained untreated showed clear signs of hyperglycemia with elevated glucose concentration (Table 1) and diminished the size of the islets of Langerhans (Fig. 6A) when compared to the control group. Regardless of the mechanism of the testing drugs (Banda et al. 2018), glibenclamide is often used as a positive control for alloxan-induced hyperglycemia based on the mechanisms of action of alloxan. Glibenclamide is widely used as it stimulates the secretion of insulin through the closing of ATP-sensitive K⁺ channels in the β-cell, leading to the depolarization of the cell, which improves the hypoinsulinemia induced by alloxan (Luzi and Pozza 1997).

The untreated alloxan-injected group showed a considerable increase in HbA1c levels in comparison with the control group, indicating its hyperglycemic effect on mice. No significant difference in HbA1c levels between the glibenclamide and flower extract (1.09 and 2.19 g/kg) groups was observed when compared to the untreated group (Fig. 4A). This may be due to the short period of time during which the experiment was conducted, which was not enough for this index to decrease. The HbA1c level in the mutation mouse model was usually determined at a time point of over 40 days (Membrez et al. 2010; Kim et al. 2014). By contrast, the alloxan-induced hyperglycemic model should measure the HbA1c level at an earlier time point (14 days) due to its high mortality rate. Interestingly, our result showed that HbA1c levels start to increase significantly at 14 days, indicating further investigation may commence at an earlier timepoint for preventing hyperglycemia efficiency.

The insulin levels did not significantly differ between the five groups and still remained high in the untreated group (Fig. 4B). This can be attributed to the fact that even though the destruction of β-cells is irreversible due to the limit of β-cell regeneration from stem cells, they can still be induced to secrete insulin when the blood glucose level increases (Bernard et al. 1999). In addition, the increase in glucose concentration can stimulate insulin secretion in β-cells (Takemoto et al. 2016). This might also contribute to the constancy of insulin levels in the untreated group despite several degenerative changes in pancreases in terms of the size of the islets of Langerhans. Mice in two treatment groups using flower extract had rather larger islets of Langerhans than those of the untreated group (Fig. 6). On the other hand, in our study, the experiment only lasted for more than 2 weeks due to the limitation of the alloxan model (Tafesse et al. 2017; Yin et al. 2018), during which β-cells in the islets of Langerhans still maintained the ability to produce insulin. As a result, the changes in insulin level may not be recorded in the alloxan group with the limitation of time. Further experiments need to be conducted for a clearer phenomenon.

With the stability of insulin levels being recorded, we continued to evaluate the extent of insulin resistance between five groups of mice using HOMA-IR. HOMA-IR is a commonly used index to assess β-cell function and insulin sensitivity from basal fasting glucose and insulin concentrations (Matthews et al. 1985; Antunes et al. 2016). Unlike clinical trials on humans, pharmacological tests on animals’ HOMA-IR were conducted at an early point in time after confirming the status of diabetes/obesity (Strage et al. 2021; Mugni et al. 2023). There was a considerable increase in HOMA-IR in the untreated hyperglycemic group in comparison with the control group (Table 2), indicating a reduction in the insulin response of this group at day 14. This contributed to the rise in blood glucose concentration of the untreated group (Table 1) despite the unchanged insulin level in a short period of time, as there was a decrease in insulin sensitivity leading to tissues not being able to utilize insulin for glucose intake. By contrast, the treatment groups using glibenclamide and flower extract at both doses (1.09 and 2.19 g/kg) showed a remarkable decrease in HOMA-IR when compared to the untreated group and a corresponding improvement in insulin resistance (Table 2). Therefore, Camellia flava flower extract exhibited an insulin-sensitivity-enhancing effect. However, it is important to emphasize that the molecular mechanism behind this is still unclear and needs more elucidation.

The body weight of the glibenclamide-treated group appeared to be higher than the other groups (Fig. 5), as glibenclamide belongs to the sulfonylureas, which are associated with weight gain (Cheng and Kashyap 2011). Camellia flava flower extract seemed to be more effective than glibenclamide in reducing body weight to some extent, though the differences were not great. The α-amylase inhibitory activity of the flower extract (Truong et al. 2022) might be one of the reasons for this observation, as it plays a role in impeding the process of converting polysaccharides into glucose and thus reducing glucose intake. This result was consistent with the study of Nguyen et al. (2023), which revealed that the flower extract had no effect on raising body weight in hyperglycemic mice (Nguyen et al. 2023). This showed that Camellia flava flower extract may intend to decrease body weight, yet the differences were not statistically significant, thus opening the way for new research to investigate its potential for promoting weight loss.

Creatinine is commonly used as a biomarker to help estimate kidney function. Increased creatinine concentration in plasma suggests a higher risk of many renal diseases (Kashani et al. 2020). Mice consuming Camellia flava flower extract at both doses (1.09 and 2.19 g/kg) showed no significant variation in plasma creatinine level in comparison with the hyperglycemic untreated group (Table 3). Furthermore, there were no abnormalities in terms of size and structure in the kidneys of all treated groups (Fig. 8). Therefore, Camellia flava flower extract was considered safe for kidneys to be used at these two doses for a long time.

Besides creatinine, plasma AST and ALT levels were also evaluated. AST and ALT are two biochemical indices often used in liver function tests to detect hepatic injuries as well as necrosis of hepatocytes. ALT can be found with a high concentration in the cytosol of the hepatocytes, whereas AST is distributed in many organs such as the heart, liver, kidney, or skeletal muscle. As a result, ALT is considered to be more sensitive and specific to the liver than AST. There was a statistically significant difference between the ALT levels between the 5 experimental groups, but this was not observed when evaluating AST (Table 3). The rising tendency of the ALT level of the disease group aligned with its histomicrograph showing liver damage (Fig. 7), which might stem from the necrosis of hepatocytes, which led to the enzyme being released abundantly into the bloodstream. On the contrary, the decline in ALT concentration (Table 3) as well as the protective effect on liver (Fig. 7) of two treatment groups using flower extract was possibly due to the antioxidant activity of Camellia flava flower extract, which prevented the hepatocytes from being damaged by ROS and regulated the release of ALT. In addition, mice treated with glibenclamide showed a substantial decrease in ALT level (Table 3) and liver damage shown in its histomicrograph (Fig. 7) compared to an untreated one (Table 3), as glibenclamide not only stimulates β-cells to secrete more insulin but also protects them from the destruction of alloxan (Ao et al. 2008). This further suggests the hepatoprotective effect of this drug when faced with liver diseases.

Importantly, alloxan reduces the size of the islets of Langerhans, which leads to the necrosis and degeneration of pancreatic cells (Muthuraman et al. 2009; Ismail et al. 2020). It is clear that Camellia flava flower extract helped enhance this damage by maintaining the size of the islets of Langerhans as well as diminishing the vacuolation and deterioration of cells (Fig. 6). As a result, it showed the effect of recovering insulin secretion and raising insulin sensitivity.

Our research team successfully created a hyperglycemia model in mice by injecting alloxan intravenously at a dose of 55 mg/kg, with the percentage of mice having blood glucose levels above 200 mg/dL being 60% after 2 weeks. This study also determined that Camellia flava flower extract produces many positive outcomes when administered to alloxan-induced hyperglycemic mice, including decreased blood glucose and HbA1c concentrations and improvements in insulin resistance. Additionally, Camellia flava flower extract was proved to be safe for kidneys for long-term use and exhibited a hepatoprotective effect in terms of reducing ALT level and minimizing liver damage, as shown in histomicrographs. Last but not least, it exerted ameliorative activity on the degeneration of the islets of Langerhans. Overall, this medicinal plant has great potential for developing into a new medication for the treatment of diabetes.

The authors have declared that no competing interests exist.

H.L.T.N: First author: Methodology, Supervision, Writing – original draft, Validation; T.T.P.H: Conceptualization, Data curation, Methodology; N.P.N.N: Writing – original draft, Investigation; N.H.L.P: Data curation, Visualization, Formal analysis; D.T.H.N: Methodology, Writing – original draft; V.D.T, M.N.T: Data curation, Investigation, Methodology; T.M.N.H: Writing – original draft, Data curation; T.N.L: Methodology, Animal care; N.P.N: Data curation, Formal analysis, Methodology; T.B.N: H&E staining, Methodology; H.N.M: Corresponding author: Writing – review & editing, Resources, Formal analysis, Supervision.

We greatly thank our colleagues from Department of Pharmacology, University of Medicine and Pharmacy at Ho Chi Minh City, Vietnam for their support in this research.