The D. melanogaster (w¹¹¹⁸ strain) was obtained from the Core Facility of Drosophila Resource and Technology, Shanghai Institute of Biochemistry and Cell Biology, CAS. The flies were cultured on normal diet (ND) containing 10.5% (w/v) corn meal, 7.5% (w/v) sucrose, 4% (w/v) yeast, 0.75% (w/v) agar, and 1% (v/v) propionic acid and maintained in an incubator (25 ± 1°C; relative humidity of 60%; 12-h dark/light cycle). The high-sugar diet (HSD) was prepared made up of 10.5% (w/v) corn meal, 30% (w/v) sucrose, 4% (w/v) yeast, 0.75% (w/v) agar, and 1% (v/v) in distilled water.

BLPs were extracted from dried bayberry leaves powder (Cixi, Zhejiang Province, China) and purified by HPD-500 column and Sephadex LH-20 column according to our laboratory methods (Zhang et al., 2017), the structural information of which were shown in Supplementary Materials (Supplementary Figure S1; Supplementary Table S1).

In this study, we exposed HSD-fed flies to BLPs in two ways, including pre-treatment and post-treatment (Figure 1 ). At the pre-treatment stage, 24 h-newly eggs were treated with the following diets for 21 days passing through embryonic and larval stages to adult flies: ND, HSD, HSD supplemented with 1 mg/ml BLPs (0.1% BLPs/HSD), HSD supplemented with 2 mg/ml BLPs (0.2% BLPs/HSD), HSD supplemented with 5 mg/ml BLPs (0.5% BLPs/HSD), respectively. Thereafter, at the post-treatment stage, ND flies and HSD flies were treated with the following diets for another 21 days: ND flies reared on ND (ND+ND), HSD flies reared on ND (HSD+ND), HSD flies reared on ND supplemented with 1 mg/ml BLPs (HSD+0.1% BLPs/ND), HSD flies reared on ND supplemented with 2 mg/ml BLPs (HSD+0.2% BLPs/ND), HSD flies reared on ND supplemented with 5 mg/ml BLPs (HSD+0.5% BLPs/ND), respectively.

Growth behaviors of HSD-fed Drosophila in the periods of egg-adult were recorded using photographs every day, and the first time observed larvae, pre-pupa, pupa, and flies were recorded as well. Meanwhile, the pupas climbed up the walls were collected using a brush and then mounted onto the slide using glycerin for microscope photography (Jun et al., 2016).

At the end of each experiment, flies were transferred into a pre-weighed EP tube under brief anesthesia. The total body weight of 10 flies per group was assessed using the microbalance.

Glucose, triglyceride (TAG) and protein levels were estimated using Glucose Assay Kit (GOPOD format) (Megazyme, Ireland), Triglyceride Assay Kit (GPO-PAP format) (Nanjing Jiancheng Bioengineering Institute, China), and BCA protein Assay Kit (Beyotime Institute of Biotechnology, China) following manufacturer’s protocols. Sample preparation was followed by the previous report with some modifications (Westfall et al., 2018). For homogenate extraction, five flies were liquid nitrogen frozen and homogenized in 300 μl PBS. After centrifuging at 12,000 rpm for 10 min, 2 μl of supernatant was obtained for TAG content assay and 20 μl of supernatant was used for protein content determination. Following, the homogenate was heat-treated for 20 min at 70°C to remove any complexes. After centrifuging, the body glucose content was obtained and analyzed. Quantities of glucose and triglyceride were both standardized against the corresponding protein content to account for variations in fly mass.

The α-amylase and α-glucosidase activity assays were carried out according to the previous report (Oboh et al., 2019). Flies were liquid nitrogen frozen, homogenized in PBS, and centrifuged (12,000 rpm, 10 min) to obtain supernatant. The supernatant was pre-incubated with the substrate (starch paste or pNPG), and the enzymatic reaction began immediately once the enzyme solution was added. After incubation at 37°C for 5 min, the absorbance was measured. The digestive enzyme activity was standardized against the protein content.

Total RNA was extracted from frozen flies with Trizol reagent (GenStar Biosolutions Co., Ltd, China) following the manufacturer’s protocol (DeAngelis and Johnson, 2019). The purity and content of RNA were measured by NanoDrop spectrophotometer (Thermo Fisher Scientific Inc., United States). Reverse transcription was then carried out from isolated RNA using PrimeScript RT reagent Kit with gDNA Eraser Kit (Takara Bio Inc., China). Synthesized cDNA was used as the template for expression profiling, together with specific primers and PowerUp SYBR Green Master Mix (Applied Biosystems, United States) through Quant Studio 3 Real-Time PCR System (Applied Biosystems, United States). The relative transcription level of candidate genes was calculated using the 2^−ΔΔCt equation and rp49 (internal control). Specific primers in this study were listed in Table 1.

Data were expressed as the mean value ±standard deviation. Analysis of variance (ANOVA) with Duncan’s difference analysis was applied to all data using Version 20.0 SPSS Statistics (SPSS Inc., Chicago, IL).

In this study, Drosophila offspring were reared on a high-sugar diet (HSD) starting from eggs to simulate T2DM. To explore the appropriate concentration of HSD (20%–40% sucrose), growth behavior including developmental rate and pupation morphology of Drosophila were observed.

Ordinarily, in the fly life cycle, eggs laid by adult flies hatch into larvae and grow into first-instar and second-instar larvae in 5–6 days. At the stage of the third instar, larvae begin to leave the food and “wander” to the wall. After about 11–12 days, third-instar larvae metamorphose into pre-pupa and pupa and then adult flies (Abrat et al., 2018). In the present study, we observed that the growth process of normal diet (ND) fed flies was broadly consistent with expectations, but that of 20% or 30% HSD fed flies had a delay of 2–3 days whether in the larval stage or pupa stage. Meanwhile, a sugar tolerance with growth retardation happened in 40% HSD intake, which was observed to stop growing at the stage of third-instar larvae and even cannot reach the pupal state (Supplementary Figure S2). Furthermore, pupa treated with 30% HSD were observed to show a more “transparent-like” appearance (Figure 2), which tended to be in the pre-pupa stage, indicating that 30% of sucrose induced a more severe growth-deficiency phenotype during the growth and development period of Drosophila.

Besides developmental delay, 30% HSD fed-flies exhibited an evident increase in body glucose and triglyceride level (Supplementary Figure S3), as well as carbohydrate digestive enzymes activities (Supplementary Figure S4). Subsequently, the mRNA levels of genes involved in glucose and lipid metabolism significantly altered in flies with HSD consumption (Supplementary Figure S5). Our results revealed that 30%HSD intake led to the pathophysiological and transcriptional changes in Drosophila, which are consistent with the T2DM phenotype as previously reported (Meshrif et al., 2022) (Eickelberg et al., 2022). In the present study, HSD participated the growth and development stages of Drosophila, therefore affecting the organ development and insulin action, which in turn influenced the glucose homeostasis and insulin sensitivity of flies (Cassim et al., 2018).

To examine the ameliorative effect of BLPs on physiological markers associated with hyperglycemia in Drosophila, the body weight, and the levels of glucose and triglyceride were determined (Figure 3).

As can be observed in Figure 3A, flies exposed to BLPs supplemented with HSD appeared a slight weight loss compared to HSD-fed counterparts in pre-treatment. In post-treatment (Figure 3B), a high dosage of 0.5% BLPs consumption led to a 10.73% weight loss in HSD flies. A similar result was observed in green tea polyphenols (GTP), which were reported to reduce the size and weight of Drosophila at the 2.5–10 mg/ml doses of GTP, was the result of reduced cell size or cell numbers (Lopez et al., 2016). Previous findings illustrated that body weight and visceral index in HFD mice decreased after a 4-week intervention of BLPs (Zhou et al., 2017).

Moreover, we observed that the elevated total body glucose and triglyceride levels in HSD-fed flies were diminished with the administration of BLPs. In detail, those flies treated with BLPs supplemented with HSD had a great reduction in body glucose level by 64.08% (Figure 3C) and in triglyceride level by 47.92% occurring at 0.5% BLPs (Figure 3E). In post-treatment, HSD induction was no longer continued, and the hyperglycemia and accumulation of lipid have improved but remained. HSD+ND group had 1.17-fold higher levels of body glucose and 1.33-fold higher levels of total triglyceride compared to the ND+ND group. After BLPs intervention, flies exhibited hypoglycemic and lipid-lowering effects, with a reduction in body glucose level by 46.78% (Figure 3D) and in triglyceride level by 59.01% occurring at 0.5% BLPs (Figure 3F). The HSD-induced glucose and triglyceride levels were reported to significantly lowered in w¹¹¹⁸ flies exposed to HSD supplemented with Solanum anguivi Lam. fruit (Nakitto et al., 2021) and Avens root extract (Günther et al., 2021), respectively.

Our findings indicated that the administrated BLPs whether in pre-treatment or post-treatment had an alleviative effect on HSD-induced accumulation of glucose and triglyceride, considered an anti-diabetic property (Alfa and Kim, 2016). It was consistent with the previous reports in rodent models that BLPs supplementation significantly reduced the blood glucose and AUC of OGTT (Zhang et al., 2022), as well as serum total cholesterol (TG) and triglyceride (TC) content in high-fat-fed rats (Zhou et al., 2017). To further investigate the hypoglycemic mechanism of BLPs, we focused on the role of BLPs in starch digestion and the glucose metabolism process, contributing to glucose homeostasis in HSD-fed flies.

α-Amylase and α-glucosidase are considered the major carbohydrate hydrolyzing enzymes in Drosophila, same as human beings. The starch and other polysaccharides in the diet were hydrolyzed to disaccharides by pancreatic α-amylase, followed by being digested to glucose by α-glucosidase in the small intestine. Therefore, the activity of these enzymes is relevant to the glucose release from carbohydrates in the diet (Sun and Miao, 2020).

The activity of two typical carbohydrate digestive enzymes (α-amylase and α-glucosidase) in flies was shown in Figure 4. Our findings showed that HSD feeding significantly simulated α-glucosidase activity. Similar results were observed in the report of Oyeniran et al. (2020), which indicated that flies fed with a diet supplemented with 30% sucrose obviously increased the digestive enzyme activity. This report was inclined to the view that the changes in enzyme activity reflected a change in enzyme quantity rather than a change in catalytic efficiency in the presence of high sugar. In this study, HSD caused a relative increase in α-glucosidase activity in response to the increase in sucrose (as substrate) amount. Moreover, carbohydrate digestive enzyme activities were closely associated with glucose homeostasis in Drosophila flies (Bezzar-Bendjazia et al., 2017). The higher activities of digestive enzymes after HSD-feeding could be one explanation for the dysglycemia as above mentioned.

With the exposure of BLPs, it seemed that BLPs inhibited α-amylase and α-glucosidase activity to some extent, indicating the hypoglycemic potential. In detail, 0.5% of BLPs inhibited α-amylase activity by 20.47% but were not significant in pre-treatment, while greatly inhibited α-amylase activity by 27.79% in post-treatment. Effective inhibition of α-glucosidase activity has been achieved at the low dosage of 0.1% BLPs, around a 73.95% reduction in pre-treatment and a 51.55% reduction in post-treatment. Comparatively, BLPs had a more dramatic inhibition effect on α-glucosidase activity than α-amylase activity. This result was following our previous in-vitro studies, which suggested that BLPs had the inhibitory activity to α-glucosidase of 517.01 mM acarbose equivalents/g extract (Wang et al., 2019) and the inhibitory activity to α-amylase of 2.92 mM acarbose equivalents/g extract (Wang et al., 2020), respectively. The inhibitory effect of BLPs on the activity of these two enzymes was mainly attributed to affecting the conformational structure and micro-environment of the enzymes through BLPs-enzyme interaction.

Further, previous reports suggested that the digestive enzyme activity is involved in the transcription level of multi-duplicated genes. Drosophila genomes involved Amy (Amy-alleles and Amy-genotypes) and ten duplicated Mal genes, and these genes were regulated and associated with carbohydrate changes in the food substrate (Inomata et al., 2019), (Lai et al., 2014). In this study, the expression levels of the main digestive enzymes genes (Amy and Mal-5b) were investigated (Figure 5). mRNA expressions of Amy and Mal-5b in the BLPs-exposed flies were both downregulated, which led to the changes in enzyme activity. The retardation of BLPs on carbohydrate digestion is one of the main ways to achieve hypoglycemic action, especially through inhibiting the α-glucosidase activity.

Regulation of the insulin signaling pathway was closely relevant to glucose metabolism. To further determine whether BLPs made a difference in glucose metabolism in Drosophila, the transcriptional level of some key genes regulating the insulin signaling pathway was accessed (Figure 6).

As can be observed in Figures 6A–D, mRNA expression of energy metabolism regulators, insulin like-peptide 2 and 3 (dilp2 and dilp3), were elevated in HSD-fed flies compared to ND-fed flies, while decreased after BLPs exposure. At the dosage of 5 mg/ml, BLPs greatly downregulated dilp2 and dilp3 levels to 0.41-fold and 0.91-fold that of the HSD group in pre-treatment, and to 0.64-fold and 0.61-fold that of the HSD+ND group in post-treatment. Moreover, insulin receptor (InR) transcript level also showed an increase in HSD flies, which had a remarkable down-regulation in BLPs-exposed flies (Figures 6E,F). It exhibited a dose-dependent decrease in pre-treatment, with a maximum drop ratio of 83.60%. However, there was no significant difference between the multiple dosages of BLPs, accompanied by a 70.54%–86.04% decrease.

Correspondingly, transcript levels of insulin-signaling regulators-protein kinase B (dAKT) and protein synthesis regulators-target of rapamycin (dTOR) were significantly upregulated, with a slight increase in Forkhead Box-O (dFOXO) in response to HSD feeding in pre-treatment, whereas dose-dependent downregulated in BLPs-exposed flies. And with a cessation of HSD induction in post-treatment, elevated mRNA expressions of dAKT, dTOR, and dFOXO in HSD flies have been impaired, which were further decreased when flies were reared on a BLPs-supplemented diet (Figure 6G–L). What’s more, phosphoenolpyruvate carboxykinase (PEPCK) is regulated by dFOXO activation. Elevated PEPCK expression in HFD flies significantly downregulated in the presence of BLPs (Figure 6M–N), reflecting inhibition of gluconeogenesis. Mitogen-activated protein kinase (MAPK), a key regulator of energy metabolism, was also reduced in mRNA expression by BLPs.

Insulin resistance is regarded as an over-production of insulin due to the cells’ reduced sensitivity to insulin (Westfall et al., 2018), which is related to the imbalance in insulin signaling pathway regulation. Generally, the insulin signaling pathway in Drosophila involves the following links. Insulin-like peptides (dilps) are secreted by the insulin-producing cells (IPCs), which subsequently bind to the single insulin receptor (InR) to activate the insulin/insulin-like growth factor signaling pathway (IIS) (Tatar et al., 2001). As a result, InR is activated, and thus leads to the stimulation of dAKT along with the activity regulation of the downstream dTOR pathway, and the subsequent sequestration of transcription factor dFOXO activity (Loreto et al., 2021). Considering that the mRNA expression of dilp2, dilp3, InR, and downstream dAKT-dTOR were all upregulated in HSD flies, it conformed to the insulin resistance-like (Nayak and Mishra, 2021). Our results illustrated that HSD feeding increased the signs of glucose metabolic stress in statistics and successfully induced insulin resistance based on mRNA expression whether in pre-treatment or post-treatment. And decreased dilp, InR, dAKT, and dTOR mRNA expression in BLPs-exposed flies were the consequence of reduced insulin signaling, suggesting the ameliorative effect of BLPs on HSD-induced insulin-resistance. Numerous studies have demonstrated the hypoglycemic action of some proanthocyanidins, such as grape seed procyanidins (Montagut et al., 2010), apple procyanidins (Ogura et al., 2016), proanthocyanidins from I. lacteals (Tie et al., 2020). They were reported to alleviate insulin resistance in type 2 diabetes mellitus via modulating second messenger signaling pathways, including PI3K-AKT, MAPK, and JNK signaling. What’s more, BLPs downregulated dFOXO-PEPCK expression, resulting in gluconeogenesis inhibition, which subsequently contributed to a reduction of glucose output (Gu et al., 2018).

Dysbiosis associated with the insulin signaling pathway would lead to insulin resistance accompanied by obesity in general (Reynés et al., 2017), so the transcriptional level of some key genes related to obesity was also investigated (Figure 7). E78 as the downstream transcriptional target of peroxisome proliferator-activated receptor-γ (PPARγ), mRNA expression of which was down-regulated with BLPs-exposure (Figure 7A–B). In parallel, as the key lipogenesis regulator, mRNA expression of sterol-regulatory element binding protein (SREBP) was slightly upregulated in HSD-fed flies compared to ND-fed flies in pre-treatment. It was aggravated in post-treatment but improved with BLPs treatment (Figure 7C–D). In terms of fatty acid synthase (FAS) and lipid storage droplet 2 (LSD), their transcript level was elevated with HSD feeding, whereas impaired in flies exposed to BLPs.

PPARγ is known as an important regulator of cholesterol, lipid and glucose metabolism (Han et al., 2019). With the activation of PPARγ, the mRNA expression of the downstream gene (E78) and adipogenic and lipogenic gene (such as SREBP) would be in turn stimulated, therefore leading to lipid storage, fatty acid oxidation, triglyceride synthesis as well as insulin sensitivity (Evans et al., 2013). SREBP is also the master transcriptional regulator of lipogenic enzymes (e.g. FAS), committed steps of cholesterol or fatty acid synthesis (Bertolio et al., 2019). And LSD plays a key role in lipid storage control (Kühnlein, 2012). Our results showed that HSD treatment upregulated the mRNA expression of SREBP, FAS, and LSD, which was the markers of aggravated obesity, including lipid storage, triglyceride synthesis and insulin resistance. As expected, BLPs markedly attenuated these dyslipidemia symptoms induced by HSD consumption. BLPs reducing oleic acid-induced lipid accumulation were previously observed in HepG2 cells, by modulating the expression of proteins related to TAG biosynthesis and sterols (Zhang et al., 2017).