A Combined Probiotic-Morus alba Strategy Enhances Glucose Homeostasis in an In Vitro Gut-Pancreas-Liver Axis Model: A Preliminary Mechanistic Screening Study

Abstract

Type 2 Diabetes Mellitus (T2DM) is a chronic metabolic disorder linked to gut microbiota dysbiosis and impaired inter-organ metabolic signalling. This study investigated the combined effects of the probiotic Lactiplantibacillus plantarum TJA7 and Mulberry Leaf extract (Morus alba) on cellular processes relevant to T2DM-related metabolic dysfunction. An advanced in vitro gut–pancreas–liver axis model, using Caco-2, EndoC-βH5, and HepG2 cells, was employed under hyperglycemic and oxidative stress conditions. The combined treatment consistently outperformed the individual components by improving intestinal barrier integrity, as indicated by increased transepithelial electrical resistance (TEER), and by enhancing butyrate translocation across the intestinal layer. Metabolites derived from the combination attenuated pancreatic β-cell dysfunction, reducing reactive oxygen species (ROS) levels and increased insulin secretion (1.7-fold compared with Mulberry Leaf extract alone). At the hepatic level, co-administration modulated key glucose metabolism pathways, including Insulin Receptor Substrate 1 (IRS1), Protein Kinase B (AKT), AMP-Activated Protein Kinase (AMPK), and Peroxisome Proliferator-Activated Receptor Gamma Coactivator 1 Alpha (PGC-1α), suggesting improved cellular glucose handling. Collectively, these findings support a positive dose-specific interaction under the tested conditions and provide a biologically plausible, hypothesis-generating framework for probiotic–phytochemical cooperation along the gut–pancreas–liver axis. Further in vivo and clinical studies are required to establish causality and translational relevance.

1. Introduction

Type 2 diabetes mellitus (T2DM) is a complex multifactorial metabolic disorder, mainly characterised by chronic hyperglycemia due to insulin resistance and insufficient insulin secretion caused by pancreatic β-cell dysfunction [1]. To date, T2DM is one of the most common non-infectious human diseases, and its prevalence is increasing throughout the world; indeed, according to the 11th edition of the International Diabetes Federation (IDF) Diabetes Atlas, in 2024 approximately 589 million adults aged 20–79 years worldwide were living with diabetes, corresponding to an estimated 11.1% of the global adult population, and this number is projected to further increase to 853 million by 2050. This disease is increasingly observed in children as well [2].

Lifestyle factors and genetics influence the development of T2DM. Among the lifestyle factors, there are physical inactivity, sedentary lifestyle, cigarette smoking, large consumption of alcohol and obesity [3]. The majority of patients with T2DM were obese or had higher body fat percentage; indeed, adipose tissue promotes insulin resistance through inflammation, particularly by enhancing free fatty acid release and adipokine deregulation [4]. Several conditions have been associated with an increased risk of T2DM, in particular metabolic syndrome, gestational diabetes, and high systolic blood pressure. In addition, patients taking antihypertensive medications have also shown a higher risk [5].

Over the past 20 years, scientists have identified more than 400 distinct genomic regions influencing T2DM, most of which confer a modest predisposition to the disease [6]. Often, the genes involved are related to β-cell function, leading to a more fragile β-cell that results in early destruction and apoptosis [7].

Dysfunction of β-cells is also due to interactions between environmental factors and molecular pathways implicated in cell biology. In an excessive nutritional state, β-cells are subjected to inflammatory, endoplasmic reticulum, metabolic/oxidative, and amyloid stress, which can lead to loss of islet integrity [4]. More specifically, at the molecular level, one of the leading causes of insulin resistance is oxidative stress; it has been demonstrated that elevated glucose levels in T2DM patients enhance oxidative stress in cells, playing a vital role in the pathogenesis of gastrointestinal complications of T2DM [8].

Data collected from cell culture studies using pancreatic β-cells, aortic smooth muscle cells, and endothelial cells have provided evidence of increased ROS production in diabetes. Exposure of β-cell lines to oxidative stress has been shown to inhibit insulin gene promoter activity and mRNA expression, thereby decreasing insulin gene expression. In addition, oxidative stress is also strongly suspected to be involved in chronic hyperglycemia-induced insulin resistance [9].

Currently, effective treatments are mainly based on the control of glycemia along with the management of comorbidities such as hypertension and hyperlipidemia. As first-line glucose-lowering therapy, clinical guidelines recommend metformin in combination with lifestyle changes. However, metformin therapy is not always able to control glycated haemoglobin (HbA_1c) levels, so guidelines advise selecting drugs based on age, duration of diabetes, presence of comorbidities and risk of adverse events, including hypoglycemia and weight gain [10].

In recent years, the microbiome has been linked to the pathophysiology of many chronic diseases, including T2DM. There is some evidence that the microbiota affects glucose metabolism in both preclinical animal models of T2DM and in healthy animals [11]. Bacteroidetes and Firmicutes are the main components of the human microbiome, which are normally present in a ratio favouring Bacteroidetes over Firmicutes (B/F > 1). However, several studies have shown that this ratio is altered in obese individuals [12]. Moreover, studies show that Gram-negative bacteria can activate the immune system via Toll-like Receptors (TLRs). Indeed, the roles of different TLRs have already been established in various studies, with TLR-5 associated with obesity and metabolic syndrome, and TLR-4 associated with insulin resistance in adipocytes [13].

In the past 10 years, data from different sources have shown an important role for the gut microbiota in the pathogenesis of insulin resistance and T2DM [14]. Indeed, specific imbalances in the composition and function of gut bacteria are known to lead to insulin resistance and thus increase the risk of developing T2DM; so, it is logical that the increasing loss of gut microbial homeostasis correlates with insulin resistance and increased circulating markers of inflammation [15]. In the latter, a moderate degree of gut microbiota dysbiosis was found, with a decrease in some universal butyrate-producing bacteria and an increase in opportunistic pathogens, as well as enrichment of some other functions related to oxidative stress resistance and sulfate reduction [16].

Therefore, a novel strategy for managing T2DM might be to target the gut microbiota with probiotics. In this context, administering probiotics can increase insulin sensitivity by increasing beneficial bacteria and reducing harmful bacteria in the intestine. In addition, probiotics can reduce inflammatory markers and oxidative stress, thereby delaying the progression and development of T2DM [17]. Further, Bifidobacteria and Lactobacillus strains can modulate hyperglycemia, oxidative stress, and inflammation; these bacteria can reduce endotoxemia, stimulate short-chain fatty acid secretion (SCFAs, mainly butyrate, acetate, and propionate), having an important effect on the gastrointestinal hormone secretion, maintaining gastrointestinal homeostasis, controlling the differentiation and proliferation of epithelial cells, and participating in the synthesis of certain vitamins [17]. For example, it was found that Lactobacillus gasseri and Lactobacillus plantarum can help normalise the expression of metabolic genes, inhibit inflammatory responses and suppress stress in obesity and insulin resistance [18]. Indeed, L. plantarum is utilised in both food and pharmaceutical industries, where it is proving a valuable asset due to its ability to be employed without adverse effects. L. plantarum has recently been employed in the medical field for the treatment of various chronic and cardiovascular diseases, including Alzheimer’s, Parkinson’s, diabetes, obesity, hypertension and many others [19]. Among these, diabetes is one of the major global health challenges, and one of the most effective therapeutic strategies for controlling postprandial hyperglycaemia is the inhibition of carbohydrate-digesting enzymes, such as α-glucosidase and α-amylase, which delay intestinal glucose absorption. Acarbose, a widely prescribed antidiabetic pharmaceutical agent, exerts its hypoglycaemic effect by binding α-glucosidase [20].

In a previous study, Kim et al. [21] demonstrated that Lactiplantibacillus plantarum K10 effectively inhibited α-amylase activity by 94.6%, indicating hypoglycemic activity by inhibiting enzymes that metabolise carbohydrates. In addition to their hypoglycemic potential, L. plantarum strains are known to produce bioactive compounds that reduce oxidative stress by preventing the formation of reactive oxygen species [22]. Their antioxidant activity is associated with structural components such as teichoic acids and peptidoglycan, which enable metal ion chelation and reducing activity. Previous studies have demonstrated that L. plantarum strains can effectively inhibit lipid peroxidation and exhibit strong radical-scavenging activity. For instance, L. plantarum AS1 inhibited linoleic acid peroxidation by 50.9%, while L. plantarum C88 showed high hydroxyl radical scavenging activity (44.3%) and DPPH radical scavenging activity (53.1%), highlighting the substantial antioxidant potential of this species [23,24,25,26]. Moreover, the study by García-Cayuela et al. [27] determined the self-aggregation rates of four L. plantarum species at the intestinal cell level. The rate was greater than 50.0%, and all four strains tested showed a high survival rate in the simulated intestinal environment.

In this context, individual strains of L. plantarum have been further studied for their combined probiotic and metabolic regulatory properties [28]. Among these, L. plantarum B19 has demonstrated strong resistance to acidic conditions and bile salts, high self-aggregation capacity, and a favourable biosafety profile, including the absence of haemolytic activity and sensitivity to commonly used antibiotics. In particular, this strain showed an α-glucosidase inhibition rate of over 60.0%, supporting its potential role in modulating carbohydrate digestion. In vitro studies have also indicated that metabolites derived from L. plantarum B19 may influence glucose metabolism in insulin-resistant HepG2 cells, as improvements in glucose utilisation, glycogen synthesis, and reductions in gluconeogenesis and oxidative stress have been observed. These effects have been associated with the modulation of key components of insulin signalling pathways, such as AMPK and PI3K/AKT, suggesting a possible mechanistic basis for its metabolic activity [28].

The activity of probiotics can be implemented by combining their beneficial properties with other hypoglycemic substances, such as Morus alba leaves (M. alba L.), which have shown beneficial effects on blood glucose, and M. alba leaf extracts (commonly named reducose) can significantly reduce glycemia [29]. Mulberry (Morus alba L.) leaves have been used for thousands of years in traditional Chinese medicine. They are widely used in the clinical treatment of diabetes and its complications [30]. The leaf contains various active components, such as the polysaccharide deoxynojirimycin, which can lower glucose levels and blood pressure [31]. This mechanism is due to the inhibition of sucrose, maltase, α-glucosidase, and α-amylase, which degrade starch, thereby blocking carbohydrate absorption and reducing blood sugar levels to prevent and treat T2DM [32]. M. alba leaves also contain antioxidant compounds such as scopolin, isoquercitrin, astragalin, rose oxide, benzyl glucopyranoside, phenolic acids such as gallic, protocatechuic, p-hydroxybenzoic, vanillic, chlorogenic, syringic, p-coumaric, ferulic and m-coumaric [31].

Given the literature on the microbiome’s role in regulating blood glucose and modulating diabetes-related disorders, microbiome-targeted probiotic treatment is an attractive therapeutic option. Accordingly, the beneficial effects of probiotics and Mulberry Leaf extract on glucose markers and insulin sensitivity in vitro model have been demonstrated in this project. Therefore, Lactiplantibacillus plantarum TJA7 (L. plantarum TJA7) and Mulberry Leaf extract were evaluated to analyse their influence on glucose control, albeit partially, by modulating T2DM key mechanisms and oxidative stress through the establishment of a gut–pancreas–liver axis.

2. Results

2.1. Chemical Composition of the Mulberry Leaf Extract

The chemical composition of the Mulberry (Morus alba L.) Leaf extract was characterised by quantitative HPLC analysis of 1-Deoxynojirimycin (DNJ) and by determination of total phenolic content using the Folin–Ciocalteu assay.

HPLC analysis enabled the identification of DNJ by comparing retention times with those of the authentic reference standard. Quantitative evaluation indicated that DNJ was present at 5.03 ± 0.12% (w/w) relative to the extract’s dry weight, confirming the standardised nature of the preparation.

The total phenolic content of the extract, assessed by the Folin–Ciocalteu method, was 52.4 ± 3.1 mg gallic acid equivalents (GAE)/g of dry extract, corresponding to approximately 5.2% (w/w) phenolic compounds.

The quantitative results obtained from both analytical approaches are summarised in Table 1. Collectively, these data indicate that the Mulberry Leaf extract is characterised by a defined chemical profile, with a substantial contribution from DNJ and phenolic constituents, providing a chemical basis for interpreting its biological effects.

Table 1.

Chemical composition of the Mulberry (Morus alba L.) Leaf extract. Data are expressed as mean ± standard deviation. DNJ content is reported as a percentage (w/w) relative to the dry weight of the extract. Total phenolic content is expressed as milligrams of gallic acid equivalents (GAE) per gram of dry extract.

Parameter	Method	Result
1-DNJ	HPLC	5.03 ± 0.12% (w/w)
Total phenolic content	Folin–Ciocalteu assay	52.4 ± 3.1 mg GAE/g dry extract

2.2. Probiotic Strain and Mulberry Leaf Extract Screening in a Dose–Response Study on Caco-2

A preliminary dose–response study was conducted to identify, among the different selected probiotic strains and the compound Mulberry Leaf extract, the most suitable candidates to be used in subsequent experiments on the modulation of the diabetic condition. As shown in Figure 1, L plantarum TJA7 and Mulberry Leaf extract increased cell viability overall compared to the control (p < 0.05), though with considerable variation in effect size and temporal stability. Conversely, certain concentrations of specific strains did not show a significant change, underscoring nonlinear and occasionally inconsistent responses. Amongst all the treatments analysed, L. plantarum TJA7 was the strain with the most marked and reproducible effect. In particular, the 7.5 mg/mL concentration was found to be the most significant factor in increasing viability after five hours of treatment, with statistical significance exceeding that of all other concentrations tested (p < 0.05; Figure 1A). This robust response, coupled with its broader temporal distribution compared to other strains, suggested greater biological efficacy and potential functional relevance in the experimental model of diabetes. For Mulberry Leaf extract, the 100 mg/mL concentration (Figure 1B) induced the greatest increase in viability at 4 h, approximately 34.0% greater than Mulberry Leaf extract 150 mg/mL (p < 0.05) and 48.5% greater than Mulberry Leaf extract 200 mg/mL (p < 0.05). This reversed dose–response behavior, with a clear functional optimum, was considered favourable for treatment selection, as it indicated a well-defined window of efficacy.

Figure 1.

Dose–response study on Caco-2 cells for assessment of cell viability. In (A), cell viability was measured by MTT on Caco-2 after treatment with L. plantarum TJA7 at different concentrations; in (B), cell viability was measured by MTT on Caco-2 after treatment with Mulberry Leaf extract at various concentrations. Data are expressed as mean ± SD (%) of 5 independent experiments performed in triplicate, normalised to the control, and represent a percentage increase in cell viability after treatment compared to the control (0% line). * p < 0.05 vs. control; ^α p < 0.05 vs. other concentrations.

The overall magnitude of the increase was significantly lower than that observed for L. plantarum TJA7 and Mulberry Leaf extract. In addition, the response manifested as greater transience and was characterised by less stable dynamics between treatment times.

The results of the study demonstrated that only L. plantarum TJA7 at a concentration of 7.5 mg/mL (corresponding to 3 × 10⁹ CFU/mL) and Mulberry Leaf extract 100 mg/mL exerted a broad, reproducible, and concentration-dependent biological effect. These findings suggest that these two substances are the most suitable candidates for further investigation into their potential to modulate the diabetic condition. Consequently, strains exhibiting weaker or variable efficacy were excluded from subsequent phases of experimentation, thereby ensuring a robust biological and interpretative foundation. Based on the dose–response screening, the concentrations of 7.5 mg/mL (corresponding to 3 × 10⁹ CFU/mL) for L. plantarum TJA7 and 100 mg/mL for Mulberry Leaf extract were selected for all subsequent experiments, as these concentrations produced the most consistent and reproducible biological effects without evidence of cytotoxicity.

2.3. Effects of Selected Probiotic Strain and Mulberry Leaf Extract on Intestinal In Vitro Model

To evaluate the effects of probiotic formulations and Mulberry Leaf extract on intestinal barrier integrity and homeostasis, a 3D in vitro intestinal model, validated in accordance with EMA and FDA standards, was used. Firstly, cell viability was assessed over 6 h using a colourimetric assay following treatment with Mulberry Leaf extract (100 mg/mL), L. plantarum TJA7 (7.5 mg/mL), and their combination, L. plantarum TJA7 + Mulberry Leaf extract. As shown in Figure 2A, all treatments increased cell viability compared to the control group (p < 0.05). Notably, the combination of L. plantarum TJA7 + Mulberry Leaf extract induced the most pronounced effect, with an increase of approximately 32.2% above the control at 4 h (p < 0.05) and an increase of about 47.0% and 26.0% compared to Mulberry Leaf extract and L. plantarum TJA7 alone, respectively. These results suggested the absence of cytotoxic effects and a potential protective or stimulatory role of the combination on epithelial metabolism. Consequently, TEER was used to assess the structural integrity of the intestinal monolayer (Figure 2B). As shown in the graph, all treatments led to a progressive increase in TEER values, peaking at 5 h. In particular, the agents in combination achieved the highest resistance with values around 618.2 ± 10 Ω·cm², exceeding those observed with L. plantarum TJA7 alone (about 601 Ω·cm²) and Mulberry Leaf extract (about 584.8 Ω·cm²). According to the literature, TEER values above 500 Ω·cm² are consistent with a well-formed, functional epithelial monolayer [33].

Figure 2.

Permeability and adhesion study on Caco-2 cells. In (A), Cell viability is measured by using the MTT test; in (B), TEER Value using EVOM3; in (C), butyric acid analysis is carried out using an ELISA kit; in (D), the analysis of aggregation activity; and in (E), the analysis of hydrophobicity activity. The tests were performed using L. plantarum TJA7 at 7.5 mg/mL and Mulberry Leaf extract at 100 mg/mL. Data are mean ± SD of five independent experiments performed in triplicate compared to control values (0% line). * p < 0.05 vs. control; ^α p < 0.05 vs. other concentrations.

In addition, all the probiotics tested modulated intestinal barrier activity, with their metabolites detected in the plasma environment (p < 0.05), as observed by butyric acid analysis at the basolateral level (Figure 2C). The data show that while L. plantarum TJA7 alone induced a statistically significant increase in butyric acid levels. L. plantarum TJA7 + Mulberry Leaf extract metabolite production appears to follow a trend similar to cell viability, showing an increase in SCFA production as butyric acid (about 20.7% vs. control and about 66.0% vs. L. plantarum TJA7, p < 0.05), supporting the hypothesis of synergistic activity with the combined treatment. The effective translocation of butyric acid across the epithelial barrier further supports the hypothesis that these microbial metabolites can reach the systemic circulation and exert beneficial metabolic effects.

Finally, aggregation and hydrophobicity activities were examined after the stimulation period (Figure 2D,E), as they are significant variables that influence bacterial adhesion. These two assays demonstrated that the probiotic strain selected possessed hydrophobic cell surfaces and aggregation capacity, thereby enhancing its adhesion to intestinal cells. In particular, L. plantarum TJA7 + Mulberry Leaf extract exhibited a more pronounced effect in comparison to L. plantarum TJA7 (p < 0.05), suggesting a probiotic adherence to the gut barrier. Especially, L. plantarum TJA7 + Mulberry Leaf increased the hydrophobic cell surface and aggregation capacity by about 37.0% and 31.0%, respectively, compared to L. plantarum TJA7 alone (p < 0.05).

In conclusion, the results of this study suggest that L. plantarum TJA7 + Mulberry Leaf extract may be a promising complementary strategy to support intestinal barrier homeostasis and microbial metabolic activity, thereby increasing cell viability, improving TEER, elevating butyric acid production and translocation, and enhancing adhesion-related properties.

2.4. L. plantarum TJA7 and Mulberry Leaf Extract Effects on Pancreas–Liver Crosstalk in a Hyperglycemic Model

Based on the previous results, to understand if the new combination L. plantarum TJA7 and Mulberry Leaf extract could exert beneficial effects not only on the intestinal barrier but also on pancreatic β cells (EndoC-βH5), further experiments were carried out by performing in vitro pancreas/liver model mimicking insulin resistance, inducing oxidative stress, a common hallmark of T2DM, which negatively affects β-cell function and insulin secretion. Cells were exposed to hyperglycemic and oxidative stress conditions as described in the Method. For this reason, the pancreas/liver model was treated with the basolateral metabolite of the in vitro intestinal model. As shown in Figure 3A, all agents tested induced a greater effect than the control (p < 0.05), confirming that cell viability improved in pancreatic cells during insulin resistance. Mulberry Leaf extract (100 mg/mL) and L. plantarum TJA7 (7.5 mg/mL) both increased cell viability by approximately 7.0% and 9.0%, respectively, compared with the control (p < 0.05). However, the combination induced a greater effect compared to untreated cells and to single agents (about 65.5% than Mulberry Leaf extract and 58.0% than L. plantarum TJA7, p < 0.05), confirming the hypothesis of better effect during diabetes and suggesting a cooperative interaction between the probiotic strain and Mulberry Leaf extract, enhancing β-cell resilience during oxidative stress conditions that mimic diabetic pathophysiology. Since mitochondrial activity is crucial for pancreatic function and an imbalance in oxidative stress may cause insulin resistance, additional experiments were performed to analyse ROS production. As shown in Figure 3B, all substances tested lack oxidant activity, confirming their absence of toxicity (p < 0.05 vs control). More precisely, Mulberry Leaf extract reduced ROS production by approximately 15.0% compared to control (p < 0.05), while L. plantarum TJA7 caused a minimal reduction that was not significant compared to the control. However, the combination showed the most substantial effect, decreasing ROS levels by about 90.0% compared to L. plantarum TJA7 (p < 0.05) and by about 47.0% compared to Mulberry Leaf extract (p < 0.05). This combination amplifies the beneficial effects of the single agents, suggesting an active role for this formulation in maintaining well-being and, consequently, improving mitochondrial metabolism and reducing ROS production, which are the main causes of insulin resistance.

Figure 3.

Safety analysis on pancreatic cells. In (A), cell viability was measured by MTT on EndoC-βH5 cells after treatment with L. plantarum TJA7 at 7.5 mg/mL and Mulberry Leaf extract 100 mg/mL, alone and combined; in (B), ROS production was measured by reduction of Cytochrome C in EndoC-βH5 cells; and in (C), insulin production was measured by a specific ELISA Kit. The tests were performed using L. plantarum TJA7 at 7.5 mg/mL and Mulberry Leaf extract at 100 mg/mL. Data are mean ± SD of five independent experiments performed in triplicate compared to control values (0% line). * p < 0.05 vs. control; ^α p < 0.05 vs. single agents.

Since the leading cause of diabetes is low insulin production, to further investigate whether the combination of probiotic and Mulberry Leaf extract could enhance pancreatic β-cell function, additional experiments were conducted to analyse insulin production after treatment with L. plantarum TJA7 and Mulberry Leaf extract, alone and combined, metabolised by intestinal cells. As shown in Figure 3C, all agents examined increased insulin production compared with untreated control cells (p < 0.05). Furthermore, the combination of L. plantarum TJA7 and Mulberry Leaf extract enhanced this effect by improving insulin production (approximately 72.0% compared to L. plantarum TJA7, p < 0.05; about 31.0% compared to Mulberry Leaf extract, p < 0.05), suggesting an active role for the new formulation in modulating insulin resistance. As reported in Appendix A (Table A1), a post hoc synergy independence analysis showed that the observed combination effects exceeded the expected additive outcomes.

The results of the present study indicate that metabolites derived from the combination of L. plantarum TJA7 + Mulberry Leaf extract exert strong protective effects on pancreatic β-cells in conditions of insulin resistance. The combination treatment demonstrated consistent superiority over individual agents by enhancing β-cell viability, reducing ROS production, and markedly augmenting insulin secretion. The data from this study provide compelling evidence for a cooperative interaction between L. plantarum TJA7 and Mulberry Leaf extract, suggesting that this formulation may help preserve β-cell function and attenuate mechanisms underlying insulin resistance.

Glucose uptake and glycogen accumulation play a central role in systemic glucose homeostasis. To investigate the ability of L. plantarum TJA7 and Mulberry Leaf extract to modulate hepatic glucose handling under hyperglycemic conditions, liver cells were cultured in high-glucose medium and treated with L. plantarum TJA7, Mulberry Leaf extract, and their combination. Subsequently, IRS1 and GLUT2 levels, glucose uptake, and glycogen accumulation were evaluated and normalised to untreated controls. As shown in Figure 4A, the data revealed a significant increase in IRS1 levels in liver cells, particularly following stimulation with the combination of L. plantarum TJA7 and Mulberry Leaf extract compared with the individual agents (about 64.0% compared to L. plantarum TJA7, p < 0.05; about 42.0% compared to Mulberry Leaf extract, p < 0.05). Post hoc evaluation using the Bliss independence model (Appendix A, Table A2) indicated that the combined treatment effects were higher than those predicted by additive responses.

Figure 4.

Glucose metabolism analysis. In (A), IRS1 levels measured by ELISA Kit; in (B), GLUT2 levels measured by ELISA Kit; in (C), Glucose uptake measured by ELISA Kit; and in (D), Glycogen accumulation measured by a specific kit. The tests were performed using L. plantarum TJA7 at 7.5 mg/mL and Mulberry Leaf extract at 100 mg/mL, alone or in combination. Data are expressed as mean ± SD (%) of 5 independent experiments performed in triplicate, normalised to control. * p < 0.05 vs. control; ^α p < 0.05 vs. single agents.

An increase in IRS1 in liver cells after treatment with L. plantarum TJA7 and Mulberry Leaf extract suggests an improvement in the insulin pathway and, therefore, a potential reduction in insulin resistance, with benefits on glucose metabolism. To corroborate these findings, GLUT2 levels were examined following treatment with L. plantarum TJA7 and Mulberry Leaf extract (Figure 4B). Indeed, treatment with L. plantarum TJA7 significantly increased GLUT2 levels by approximately 5.5% compared to control (p < 0.05), whereas Mulberry Leaf extract alone induced a greater effect, reaching approximately 13.4% compared to control (p < 0.05). Notably, the combination of L. plantarum TJA7 and Mulberry Leaf extract showed a cooperative effect, elevating GLUT2 levels by approximately 76.0% compared to L. plantarum TJA7 alone (p < 0.05) and by about 42.0% compared to Mulberry Leaf extract alone (p < 0.05). As also observed for IRS1, post hoc evaluation with the Bliss independence model (Table A3 in Appendix A) showed that the combined effects of treatment were higher than those predicted by additive responses.

Glucose uptake followed a similar trend (Figure 4C): the combined treatment further enhanced glucose uptake, increasing glucose uptake approximately 49.5% compared to L. plantarum (p < 0.05) and by about 35.0% compared to Mulberry Leaf extract (p < 0.05).

Moreover, intracellular glycogen levels were significantly higher in all treatment groups than in the control (Figure 4D). L. plantarum TJA7 alone enhanced glycogen accumulation to about 1150 pg/μL, and Mulberry Leaf extract to about 2585 pg/μL (p < 0.05 vs. control). The combination treatment yielded the highest accumulation, reaching about 4380 pg/μL, significantly exceeding the effects observed with either compound alone by about 74.0% (vs. L. plantarum TJA7, p < 0.05) and 41.0% (vs. Mulberry Leaf extract, p < 0.05). Overall, our results show that L. plantarum TJA7 and Mulberry Leaf extract, when combined, significantly increase IRS1 and GLUT2 levels, increasing glucose uptake and improving glycogen storage in liver cells. This suggests a cooperative activity that facilitates glucose entry into hepatocytes and its conversion to glycogen, highlighting the potential of this combined approach to improve hepatic glucose metabolism under hyperglycemic conditions.

Finally, to explore the molecular mechanisms by which L. plantarum TJA7 and Mulberry Leaf extract affect hepatic glucose metabolism, the activity of key intracellular signalling pathways, including PI3K/AKT, AMPK, and PGC-1α, has been assessed in liver spheroids under high-glucose conditions. As shown in Figure 5, treatment with L. plantarum TJA7 (7.5 mg/mL) induced a significant increase in PI3K/AKT activity (about 6.0% vs. control, p < 0.05). Mulberry Leaf extract (100 mg/mL) showed a more pronounced effect, enhancing activity to approximately 28.3% over control (p < 0.05). Still, the combination of L. plantarum TJA7 and Mulberry Leaf extract led to an increase, reaching nearly 56.0% above baseline, corresponding to an increase by about 49.0% compared to Mulberry Leaf extract and an increase by about 90.0% compared to L. plantarum TJA7 alone (p < 0.05 vs. all groups). Similarly, L. plantarum and Mulberry Leaf extract alone increased AMPK activity by about 7.8% and 26.5%, respectively (p < 0.05 vs. control). In contrast, the combined treatment further enhanced levels by approximately 83.0% compared to L. plantarum TJA7 (p < 0.05) and by about 42.0% compared to Mulberry Leaf extract (p < 0.05).

Figure 5.

Analysis of the intracellular pathway. In (A) PI3K/AKT levels, in (B) PGC-1α levels and in (C) AMPK levels, all obtained through a specific ELISA kit. The tests were performed using L. plantarum TJA7 at 7.5 mg/mL and Mulberry Leaf extract at 100 mg/mL, alone or in combination. Data are expressed as mean ± SD (%) of 5 independent experiments conducted in triplicate, normalised to control. * p < 0.05 vs. control; ^α p < 0.05 vs. single agents.

To conclude, PGC-1α levels, a downstream effector of both AMPK and PI3K/AKT and a regulator of glucose homeostasis and mitochondrial biogenesis, also showed significant modulation. While L. plantarum TJA7 and Mulberry Leaf extract alone induced modest but significant increases compared to control (about 8.0% and about 14.2%, respectively, p < 0.05), the combination resulted in a significant activity enhancement of about 69.0% (vs. L. plantarum TJA7, p < 0.05) and by about 45.0% (vs. Mulberry Leaf extract, p < 0.05), indicating a cooperative activation of this metabolic pathway. For PGC-1α and AMPK levels, post hoc assessment using the Bliss independence model (Appendix A, Table A4 and Table A5) indicated that the combined effects of treatment were greater than those predicted by additive responses.

3. Discussion

The gut microbiota plays an important role in regulating metabolic homeostasis in insulin-resistant subjects. Alterations in microbial composition and function characterised by reduced diversity, loss of beneficial taxa, and expansion of pro-inflammatory species contribute to low-grade systemic inflammation, increased intestinal permeability, and metabolic endotoxemia [34]. These changes disrupt SCFA biosynthesis, impair gut–liver–muscle signalling, and exacerbate defects in insulin receptor signalling. With the rising global prevalence of T2DM, there is a pressing need to explore novel complementary therapeutic strategies alongside conventional pharmacological treatments. For this reason, the interest in natural compounds, including probiotics and natural extracts, has emerged due to their potential role in disease management [34]. Recent studies have reported beneficial effects of combined probiotic–plant extract formulations on glucose metabolism, mainly focusing on systemic glycemic outcomes or single-organ responses. However, most of these approaches do not address inter-organ metabolic communication or mechanistic combined effect across the gut–pancreas–liver axis, which represents a distinctive feature of the present study [35,36].

In this context, the present study explores the synergistic effects of L. plantarum TJA7 and Mulberry Leaf extract in modulating insulin resistance in T2DM. The Mulberry Leaf extract utilised in the present study was subjected to chemical characterisation. The HPLC analysis revealed the presence of DNJ at 5.03 ± 0.12% (w/w). Concurrently, the total phenolic compound content, as determined through the Folin–Ciocalteu assay, was determined to be 52.4 ± 3.1 mg GAE/g. These bioactive metabolites, recognised for their antioxidant and hypoglycaemic properties, endow the extract with the capacity to modulate blood sugar and reduce oxidative stress. The chemical standardisation of the extract is pivotal to ensuring reproducibility and consistency in the results, thereby supporting its use in combination with L. plantarum TJA7 to investigate synergistic effects on insulin resistance.

Based on the results obtained, the study highlights the ability of L. plantarum TJA7 and Mulberry Leaf extract, individually and in combination, to preserve and reinforce intestinal epithelial barrier function as evidenced by increased TEER and production of SCFAs, specifically butyrate. According to the literature, a functional and intact intestinal barrier is crucial for preventing systemic endotoxemia and chronic low-grade inflammation, which are both hallmarks of T2DM [15]. In our model, TEER values exceeded 500 Ω·cm² in all treatment groups, with the combination treatment achieving the highest resistance (about 610 Ω·cm²). This enhancement of epithelial integrity is in line with previous findings showing that probiotics can modulate TJ proteins, such as occludin, claudin-1, and ZO-1, and attenuate permeability induced by high concentrations of glucose and lipids. In addition, Mulberry Leaf extract may contribute by lowering luminal glycemic load, preventing non-enzymatic glycation of TJ proteins, and limiting NF-κB-driven pro-inflammatory cascades triggered by translocated lipopolysaccharides (LPS) [33]. Although TEER serves as a functional proxy for epithelial integrity, direct assessment of tight junction protein expression (e.g., ZO-1, occludin, claudins) was beyond the scope of the present study and constitutes a limitation. Moreover, recent evidence further emphasises that gut-derived metabolites such as SCFAs exert systemic effects through the coordinated modulation of immune and metabolic pathways, including activation of FFAR2/3 and regulation of inflammatory tone. Current in vitro systems are unable to fully reproduce this microbiota–immune–host crosstalk, which may contribute to the metabolic improvements observed in vivo. From a translational perspective, future studies should therefore employ in vivo models that better capture this complexity, such as diet-induced diabetic rodents receiving probiotic–phytochemical combinations, together with parallel assessment of gut microbiota composition, circulating SCFAs, inflammatory markers, and insulin sensitivity [37,38]. Notably, the combined treatment significantly increased butyrate translocation across the intestinal monolayer compared to L. plantarum TJA7 alone. Butyrate plays a key role in modulating host energy metabolism, activating FFAR2 and FFAR3. It also stimulates GLP-1 and PYY (Peptide YY) secretion and reduces inflammation by inhibiting histone deacetylase activity. This hormonal cascade not only enhances postprandial insulin secretion but also reduces appetite and delays gastric emptying, thereby attenuating postprandial glycemic excursions [39]. Butyrate was selected as the primary SCFA endpoint due to its well-established role in intestinal barrier integrity, β-cell protection, and metabolic signalling, while acetate and propionate were not analysed in the present experimental setting.

Further, the EndoC-βH5 cell line represents a highly relevant and physiologically faithful human β-cell model. This system not only recapitulates the functional characteristics of primary β-cells but also reliably mirrors their responses to pathological stimuli [40]. Oxidative stress in EndoC-βH5 cells has been shown to markedly impair insulin gene transcription, trigger apoptotic pathways, and exacerbate insulin resistance, thereby faithfully modelling the molecular events that underlie β-cell dysfunction in diabetes [9,41]. Consequently, EndoC-βH5 provides an indispensable platform for dissecting the mechanisms of β-cell failure and evaluating novel therapeutic strategies to preserve β-cell integrity and function [42]. In our oxidative stress-induced insulin resistance model, both L. plantarum TJA7 and Mulberry Leaf extract increased β-cell viability by about 7–8% compared to the control. At the same time, the combination improved viability compared to the most effective single agent. This synergistic protection may be attributed to the complementary mechanisms between the L. plantarum TJA7, which modulates immune responses and reduces inflammatory cytokines [18] and Morus alba, which inhibits digestive enzymes and lowers postprandial glycaemia, reducing glucotoxic stress on β-cells, limiting advanced glycation end-product (AGE) formation and subsequent RAGE-mediated oxidative stress [30]. Oxidative stress, quantified as ROS production, was also significantly attenuated, with the combination reducing ROS levels by about 25.0%, surpassing the effect of either agent alone. This aligns with prior studies showing that probiotics and plant polyphenols can exert antioxidant effects by upregulating endogenous antioxidant defenses (e.g., SOD, GPx, catalase) and attenuating ROS-producing pathways such as NADPH oxidase, thereby ameliorating oxidative stress a key driver of mitochondrial dysfunction, β-cell failure and insulin resistance in diabetes [43,44]. In addition to parallel antioxidant actions, probiotic-derived metabolism may enhance the bioavailability of mulberry polyphenols, while mulberry compounds may support probiotic activity, contributing to amplified protective effects in β-cells [45]. In vitro evidence from pancreatic β-cell models supports the relevance of dietary antioxidants in preserving β-cell function and survival under oxidative challenge. Furthermore, advanced in vitro study combining EndoC-βH5 with other cells, such as endothelial and hepatic cells, have been established to better recapitulate islet microenvironments in vitro and could be adapted to evaluate the impact of combined nutraceutical and probiotic interventions on β-cell integrity and oxidative stress responses [42,46].

One of the key findings of this study is enhanced insulin secretion and improved IRS1 regulation. The combination tested increased insulin production by 72.0% compared with L. plantarum TJA7 and 31.0% compared with Mulberry Leaf extract, suggesting a strong synergistic effect. The observed increase in IRS1 levels in the liver supports the hypothesis that activation of mechanisms involved in glucose metabolism results from insulin binding to IRS1. These data are consistent with the literature [47]. This suggests restoration of insulin signalling via the PI3K/AKT pathway, as tyrosine-phosphorylated IRS1 efficiently recruits PI3K, activating downstream AKT and GLUT translocation in peripheral tissues [48]. This effect may be linked to what is observed in early stages of T2DM in humans, when β-cell responsiveness is impaired but not irreversibly lost. Therefore, the involvement of gut-derived metabolites, particularly butyrate, could mediate this effect via activation of free fatty acid receptors (FFAR2/3) and incretin pathways, leading to improved glucose-stimulated insulin secretion [49]. Beyond pathway activation, reciprocal microbial–phytochemical interactions may converge on shared metabolic targets, providing a mechanistic basis for the amplified response observed with the combined treatment.

In addition, the combination of 7.5 mg/mL (corresponding to 3 × 10⁹ CFU/mL) L. plantarum TJA7 and 100 mg/mL Mulberry Leaf extract appears to potentiate hepatic glucose clearance and storage capacity by activating insulin-dependent and independent signalling pathways. At the hepatocyte level, the PI3K/AKT pathway is a central driver of glucose uptake, initiated by insulin receptor activation, which promotes GLUT2 translocation to the plasma membrane. AKT phosphorylation subsequently inhibits glycogen synthase kinase-3β (GSK-3β), relieving its suppression of glycogen synthase (GS) and facilitating glycogen synthesis. Insulin signalling also reduces glycogen phosphorylase activity, limiting glycogen breakdown and reinforcing an anabolic state within hepatic carbohydrate metabolism [50].

The polyphenolic compounds from Mulberry Leaf extract activate AMP-activated protein kinase (AMPK), a metabolic regulator that enhances insulin sensitivity and promotes glucose uptake even under conditions of reduced insulin availability [51]. AMPK activation exerts a dual influence: directly promoting GS activation by altering its phosphorylation status and inhibiting gluconeogenesis by downregulating the transcription of phosphoenolpyruvate carboxykinase (PEPCK) and glucose-6-phosphatase (G6Pase), thereby shifting hepatic metabolic balance toward glucose storage. Moreover, AMPK activation acts on the PI3K/AKT pathway and stimulates PGC-1α, a transcriptional coactivator involved in mitochondrial biogenesis and oxidative metabolism. In the liver, PGC-1α not only enhances oxidative phosphorylation capacity but also improves lipid handling, thereby mitigating hepatic steatosis, which is frequently comorbid with T2DM [52]. Although PGC-1α is involved in multiple metabolic processes, its modulation here is interpreted within the context of improved mitochondrial and metabolic signalling under the experimental conditions employed. Enhanced mitochondrial function increases the energetic efficiency of hepatocytes, supporting sustained glucose uptake and storage without triggering detrimental lipid accumulation. Additionally, L. plantarum TJA7 may indirectly amplify these effects by modulating the gut microbiota to increase SCFA production, particularly propionate and butyrate. These SCFAs activate G-protein-coupled receptors (GPR41/43). They may further stimulate AMPK and AKT signalling, creating a positive feedback loop that consolidates hepatic insulin sensitivity and glycogen storage [53]. Such bidirectional microbial–phytochemical interactions may create a positive feedback loop that consolidates metabolic improvements beyond those achievable by single-agent interventions. The combined role of these pathways places this combination of L. plantarum TJA7 and Mulberry Leaf extract in a favourable strategy compared to single-agent interventions, as it targets both glucose handling and mitochondrial health. This integrated metabolic modulation may be particularly advantageous in T2DM, where defects in insulin signalling, glycogen synthesis, and mitochondrial function coexist and mutually exacerbate glycaemic dysregulation. However, given the promising results obtained from these preliminary analyses, the next step could be more in-depth analyses in animal models or through clinical trials to confirm the safety and efficacy of this new combination. Before that, in vivo research or human clinical trials must validate these findings. Nonetheless, there is still reason to believe that oral administration of this novel formulation comprising L. plantarum TJA7 and Mulberry Leaf extract could represent a complementary mechanistic approach. In this perspective, the present findings should be interpreted as a comprehensive and detailed mechanistic screening of the molecular pathways involved in the gut-pancreas-liver axis. While the results clearly indicate an enhanced combined effect, it is important to acknowledge the inherent limitations of in vitro systems Nevertheless, this study serves as a necessary and rigorous step to provide a mechanistic rationale for these effects, establishing a solid foundation for the potential of this novel formulation.

4. Materials and Methods

4.1. Agents Preparation

L. plantarum TJA7 (scientific name Lactiplantibacillus plantarum) was donated by Probionova (Lugano, Switzerland), while Mulberry Leaf extract (Morus alba L.; extracted from Leaf; water and ethanol as extraction solvent, 5.03% of n, DNJ) were donated by Vivatis Pharma (Gallarate, Milan, Italy) and both were prepared immediately before use. For that, a different pack of the product was reconstituted by mixing probiotics with DMEM without red phenol (Merck Life Science, Rome, Italy), supplemented with 0% FBS (Merck Life Science, Rome, Italy), 50 IU/mL penicillin–streptomycin (Merck Life Science, Rome, Italy) and 2 mM L-glutamine solution (Merck Life Science, Rome, Italy). For each test, performed in triplicate, the samples were diluted in culture medium to obtain a range of concentrations of 1 × 10⁹–4 × 10¹⁰ CFU/mL for L. plantarum, and of 100 mg/mL to 200 mg/mL for Mulberry Leaf extract.

4.2. Phytochemical Characterisation of Mulberry Leaf Extract

The Mulberry Leaf extract used in this study (Morus alba L., leaves) was supplied with a Certificate of Analysis provided by the manufacturer. It was standardised to contain at least 5% (w/w) DNJ, as determined by HPLC analysis. The extract was provided as a yellow-brown fine powder with 100% passing through an 80-mesh sieve. Quality control parameters included loss on drying (3.68%), residue on ignition (2.67%), heavy metal content (<10 ppm), and microbiological purity, all of which complied with pharmacopeial and AOAC standards.

In addition, DNJ is a well-recognised bioactive alkaloid characteristic of Morus alba leaves and represents a major functional marker compound for Mulberry Leaf extracts, particularly in studies investigating metabolic, anti-inflammatory, and cytoprotective effects.

4.3. Determination of DNJ by HPLC

The content of DNJ in the Mulberry (Morus alba L.) Leaf extract was determined by high-performance liquid chromatography (HPLC) [54]. DNJ reference standard (≥98% purity) was used for identification and quantification.

The extract powder (50 mg) was dissolved in 10 mL of ultrapure water, sonicated for 15 min at room temperature, and centrifuged at 10,000 rpm for 10 min. The supernatant was filtered through a 0.22 μm membrane filter before analysis.

Chromatographic separation was carried out on a HILIC amide column (250 × 4.6 mm, 5 μm particle size) maintained at 30 °C. The mobile phase consisted of acetonitrile (solvent A) and 10 mM ammonium acetate in water (solvent B), delivered in isocratic mode at a ratio of 75:25 (A:B, v/v). The flow rate was set at 1.0 mL/min, and the injection volume was 20 μL.

DNJ was detected using ultraviolet detection at 195 nm. Identification was based on comparison of retention time with that of the DNJ standard. Quantification was performed using an external calibration curve spanning 10–500 μg/mL. Results were expressed as percentage (w/w) of DNJ relative to the dry weight of the extract.

4.4. Total Phenolic Content by Folin–Ciocalteu Method

The total phenolic content (TPC) of the Mulberry (Morus alba L.) Leaf extract was determined using the Folin–Ciocalteu colourimetric assay [55]. Briefly, the extract was dissolved in ultrapure water at an appropriate concentration. An aliquot of the sample solution (100 μL) was mixed with 500 μL of Folin–Ciocalteu reagent previously diluted 1:10 (v/v) with water. After incubation for 5 min at room temperature, 400 μL of a 7.5% (w/v) sodium carbonate solution was added to the reaction mixture.

The samples were incubated for 30 min at room temperature in the dark, and the absorbance was measured at 765 nm using a UV–Vis spectrophotometer. Gallic acid was used as the reference standard, and a calibration curve was constructed over the concentration range of 10–200 μg/mL.

The total phenolic content was calculated from the calibration curve and expressed as milligrams of gallic acid equivalents per gram of dry extract (mg GAE/g).

4.5. Cell Cultures

Human colorectal carcinoma cells, Caco-2, were purchased from the American Type Culture Collection (ATCC, Manassas, VA, USA) and cultured in Advanced Dulbecco’s Modified Eagle Medium (Adv-DMEM; GIBCO, Thermo Fisher Scientific, Waltham, MA, USA), containing 2 mM l-glutamine (Merck Life Science, Rome, Italy), 1% penicillin-streptomycin (Merck Life Science, Rome, Italy) and 10% foetal bovine serum (FBS; Merck Life Science, Rome, Italy) at 37 °C in a 5% CO₂ incubator [56]. The cells were plated in different manners to perform several experiments including: 1 × 10⁴ cells in 96-well plates to study cell viability by MTT test and ROS production by Cytochrome C; 2 × 10⁴ cells were plated on 6.5 mm Transwell^® (Corning^® Costar^®, Merck Life Science, Rome, Italy) with a 0.4 μm pore polycarbonate membrane insert (Corning^® Costar^®, Merck Life Science, Rome, Italy) in a 24-well plate to perform the analyses of SCFA (Butyric Acid) and intestinal barrier integrity. Cells plated on the Transwell^® insert were maintained in complete medium, which was changed every other day on both the basolateral and apical sides for 21 days before stimulation. Before stimulation, the medium on the apical side was adjusted to pH 6.5, which corresponds to the physiological pH of the small intestinal lumen. In contrast, pH 7.4 on the basolateral side was similar to that of blood [57].

The human pancreatic β-cell line EndoC-βH5 was purchased from Human Cell Design (Paris, France). The cells, supplied as ready-to-use differentiated human β-cells, were cultured according to the manufacturer’s instructions and as previously described in the literature [40]. EndoC-βH5 were maintained in Ulti-β1^® medium (Human Cell Design, Paris, France), a chemically defined culture medium specifically formulated to preserve the functional identity of human β cells. Cells were seeded on plates previously coated with βCOAT^® (Human Cell Design, Paris, France) to promote cell adhesion [40]. Cultures were maintained at 37 °C in a humidified atmosphere with 5% CO₂. After seeding, the culture medium was replaced according to the manufacturer’s recommendations and the cells were allowed to stabilise for approximately 7 days before the experiments were performed, to ensure stable glucose-dependent insulin secretion [58]. For the experiments, cells were seeded as follows: 1 × 10⁴ cells/well in 96-well plates for the assessment of cell viability by MTT assay and for studying the production of reactive oxygen species (ROS); 1.5 × 10⁵ cells were seeded on Transwell^® inserts for the assessment of insulin secretion following sample treatment after intestinal absorption.

Human epithelial hepatocellular carcinoma cells, HepG2, were acquired from the American Type Culture Collection (ATCC, Manassas, VA, USA) and cultured in Adv-DMEM (GIBCO, Thermo Fisher Scientific, Waltham, MA, USA), supplemented with 10% FBS (Merck Life Science, Rome, Italy), L-glutamine 2 mM (Merck Life Science, Milan, Italy) and penicillin-streptomycin 1% (Merck Life Science, Rome, Italy) at 37 °C and 5% CO₂ incubator [59]. After reaching 80–90% confluence, the cells were cultured in different ways based on different experimental protocols: 1 × 10⁴ cells in 96-well plates to study cell viability by MTT test and ROS production; 2.5 × 10⁵ cells were seeded in the basolateral compartment to measure molecular pathways involved in glucose metabolism using an ELISA assay Kit.

4.6. Experimental Protocol

The experiments were divided into three phases to study the effects of probiotics in modulating the diabetic condition. Firstly, a dose–response study was conducted on intestinal cells to select the better concentration to use in the subsequent phase of the study. Indeed, the cells were treated with L. plantarum ranging from 2.5 mg/mL to 25 mg/mL (corresponding to 1 × 10⁹–1 × 10¹⁰ CFU/mL) and Mulberry Leaf extract ranging from 100 mg/mL to 200 mg/mL. Secondly, intestinal cells were used to create an intestinal barrier model in Transwell^® to evaluate the integrity of the intestinal monolayer, measuring transepithelial electrical resistance (TEER), butyric acid production, and probiotic adherence to the intestinal epithelium after treatment with the best concentration of the probiotic and the substance. Finally, pancreatic cells (EndoC-βH5) were cultured in the apical compartment of a Transwell^® system to mimic insulin resistance and to evaluate mitochondrial metabolism, ROS production, and insulin secretion for 24 h. Moreover, the liver cells seeded in the basolateral compartment were exposed to hyperglycemic conditions (30 mM glucose). This was done to facilitate analysis of cell viability, ROS production, glucose uptake, glycogen production, and the mechanisms underlying glucose metabolism. The mechanisms in question included IRS1, AKT, AMPK, and PGC-1α.

Before stimulations, the cells were synchronized overnight with DMEM without red phenol (Merck Life Science, Milan, Italy) and FBS (Merck Life Science, Rome, Italy), supplemented with 1% penicillin/streptomycin (Merck Life Science, Rome, Italy), 2 mM L-glutamine (Merck Life Science, Rome, Italy), and 1 mM sodium pyruvate in an incubator at 37 °C, 5% CO₂, 95% humidity.

4.7. Intestinal In Vitro Model

An experimental model was set up using the Transwell^® system technique to construct a complete intestinal barrier model [60,61,62]. The experimental model used was Caco-2, an intestinal cell line widely used to predict intestinal absorption characteristics after oral intake [63] and approved by the EMA and FDA [64,65]. Caco-2 cells, seeded on a Transwell^® insert, were maintained in complete medium for 21/28 days [61,62], the time required for the cells to polarise and differentiate, forming a continuous epithelium with functional tight junctions (TJ) and microvilli at the apical level. To verify proper junction formation before experiments, TEER was monitored using an EVOM3, coupled with STX2 rod electrodes (World Precision Instruments, Sarasota, FL, USA), testing every other day for 21 days until a TEER value ≥ 400 Ω∙cm² was reached, after which stimulation was initiated [66]. Before the start of the experiments, to set up different pH conditions, the medium in the apical and basolateral environments was changed to obtain a pH around 6.5 at the apical level (acidic pH mimicking small intestine lumen) and a pH around 7.4 at the basolateral level (neutral pH mimicking human blood) [63]. The cells were kept for 15 min at 37 °C and 5% CO₂. After that, the TEER values were measured again before starting the experiment to verify stabilisation.

4.8. Pancreas–Liver Axis Model

The Transwell^® system was utilised to establish a pancreas–liver axis in vitro, employing organoids composed of pancreatic EndoC-βH5 cells and HepG2 hepatocytes. Specifically, approximately 1.5 × 10⁵ pancreatic cells were seeded into the apical compartment of the Transwell^® insert and incubated for 7 days at 37 °C with 5% CO₂ [46,67]. Subsequently, approximately 2.5 × 10⁵ liver cells were seeded in the basolateral compartment, arranged in parallel with the pancreatic cells. Following 7 days, the insert containing the pancreatic cells was transferred to the liver cells, which were then placed under hyperglycemic conditions using 30 mM glucose [68].

4.9. Cell Viability Analysis

The A3-(4,5-Dimethylthiazol-2-yl)-2,5-Diphenyltetrazolium Bromide (MTT) assay was performed according to a standard protocol to exclude potential cytotoxic effects [69]. Cell viability was determined using a spectrometer by measuring absorbance at 570 nm with correction at 690 nm (Infinite 200 Pro MPlex, Tecan, Männedorf, Switzerland) and calculated by comparing the results to control cells that had received no stimulus (baseline 0%) from five independent triplicate experiments.

4.10. Butyric Acid Quantification

After stimulation of Caco-2 cells with probiotics, butyric acid production was quantified using an ELISA kit (Cloud-Clone, Wuhan, China) according to the manufacturer’s instructions [70]. The plate was read at 450 nm (Infinite 200 Pro MPlex, Tecan, Männedorf, Switzerland). The OD was interpolated with a standard curve (from 10.000 pg/mL to pg/mL), expressing the data from five independent triplicate experiments as mean (pg/mL) compared to the control.

4.11. Surface Hydrophobicity and Aggregation Activity

Surface hydrophobicity and aggregation activity were assessed following the procedure reported in the literature [71]. PBS served as the control, and for the hydrophobicity test, 0.4 mL of xylene was added to 4 mL of the probiotic suspension, mixed, and left to stand for 15 min to allow phase separation. The absorbance of both the sample and the control was then measured at 600 nm. All analyses were conducted in triplicate, with 10 parallel samples per replicate. Surface hydrophobicity was calculated using the following equation:

$$\mathrm{Surface} \mathrm{hydrophobicity}={\displaystyle \frac{{\mathrm{OD}}_{600}\left(\mathrm{control}\right)-{\mathrm{OD}}_{600}\left(\mathrm{test}\right)}{{\mathrm{OD}}_{600}\left(\mathrm{control}\right)}}\times 100\%$$

where

• OD600 (control): absorbance of the control sample

• OD600 (test): absorbance of the probiotic sample

For the aggregation assay, 0.1 mL of the probiotic suspension was mixed with 2.9 mL of PBS, and absorbance at 600 nm was recorded. Aggregation activity after 2 h was calculated as:

$$\mathrm{Aggregation} \mathrm{activity}={\displaystyle \frac{{\mathrm{OD}}_{600}\left(0\right)-{\mathrm{OD}}_{600}\left(t\right)}{{\mathrm{OD}}_{600}\left(t\right)}}\times 100\%$$

where:

• OD600 (0): absorbance immediately after mixing with buffer

• OD600 (t): absorbance after 2 h of incubation

4.12. Evaluation of ROS Production

The ROS levels following stimulation were examined by measuring the rate of superoxide anion release [63]. Absorbance was determined at 550 nm using a spectrometer (Infinite 200 Pro-MPlex, Tecan, Männedorf, Switzerland). Results are presented as mean ± SD (%) compared to control (line 0, untreated cells) from five independent triplicate tests.

4.13. Insulin ELISA Kit

Insulin concentration was determined using the Human HS-INS (High-sensitivity Insulin) Accquant^® ELISA Kit (FineTest, Wuhan, China) according to the instructions [72]. The plate was read at 450 nm using a plate reader (Infinite 200 Pro MPlex, Tecan, Männedorf, Switzerland). A standard curve is plotted relating the intensity of the colour to the concentration of standards (range 7.813–500 pg/mL). Results are presented as mean ± SD (%) compared to control (line 0, untreated cells) from five independent triplicate tests.

4.14. IRS1 ELISA Kit

Insulin Receptor Substrate 1 (IRS1) production was determined using an ELISA kit (NovusBio, San Diego, CA, USA) to analyse IRS1 in HepG2 lysates, according to the manufacturer’s instructions. Briefly, 100 µL of the sample was added to the corresponding well and incubated for 90 min at 37 °C. Then, 100 µL of Biotinylated Detection Antibody was added and incubated for 1 h at 37 °C. After three washes with Wash Buffer, 100 µL of HRP Conjugate was added, and the plate was incubated for 1 h at 37 °C. The plate was washed 5 times, and 90 µL of Substrate Reagent was added for 15 min at 37 °C. Finally, 50 µL of Stop Solution was added, and the samples were analysed using a spectrometer (Infinite 200 Pro MPlex, Tecan, Männedorf, Switzerland) at 450 nm. The concentration is expressed as ng/mL relative to a standard curve (range from 0.16 to 10 ng/mL), and the results are expressed as percentage (%) relative to the control (line 0, untreated cells) from five independent triplicate tests.

4.15. Evaluation of Glucose Uptake

The Glucose Uptake Colourimetric Assay Kit (Merck Life Science, Rome, Italy) was used to measure glucose uptake [69]. Absorbance was measured at 412 nm every 5 min (Infinite 200 Pro MPlex, Tecan, Männedorf, Switzerland). The results are expressed as a percentage (%) versus control (line 0, untreated cells) from five independent triplicate tests.

4.16. Glycogen Measurement

Glycogen Assay Kit (Merck Life Science, Rome, Italy) was used to detect glycogen levels in HepG2 cells [69]. Absorbance was measured at 570 nm (Infinite 200 Pro MPlex, Tecan, Männedorf, Switzerland). The results are expressed as a percentage (%) versus control (line 0, untreated cells) from five independent triplicate tests.

4.17. GLUT2 ELISA Kit

Human Glucose Transporter 2 (GLUT2) production was determined using an ELISA kit (MyBiosource, San Diego, CA, USA) to analyse GLUT2 in HepG2 cell lysates, according to the manufacturer’s instructions. Briefly, 100 µL of each sample was added to the corresponding well and incubated for 2 h at 37 °C. Then the liquids were aspirated, and 100 µL of Detection Reagent A was added before incubating the plate for 1 h at 37 °C. After that, the wells were washed 3 times with Wash Buffer, and 100 µL of Detection Reagent B was added, followed by incubation of the plate for 1 h at 37 °C. After washing the plate 5 times, 100 µL of Substrate Solution was added, and the plate was incubated for 25 min at 37 °C. Finally, 50 µL of Stop Solution was added to each well, and the samples were analysed using a spectrometer (Infinite 200 Pro MPlex, Tecan, Männedorf, Switzerland) at 450 nm. The concentration is expressed as ng/mL relative to a standard curve (range from 1.56 ng/mL to 100 ng/mL), and the results are expressed as percentage (%) relative to the control (line 0, untreated cells) from five independent triplicate tests.

4.18. AKT ELISA Kit

Phospho-AKT production was determined using an ELISA kit (MyBiosource, San Diego, CA, USA) to analyse AKT in HepG2 cell lysates, according to the manufacturer’s instructions. Briefly, 40 µL of each sample was added to the wells, along with 10 µL of anti-phospho-AKT antibody, and the plate was incubated for 1 h at 37 °C. Then the plate was washed 5 times with Wash Buffer before adding 50 µL of Substrate Solution A and 50 µL of Substrate Solution B to each well for 10 min. Finally, 50 µL of Stop solution was added to each well, and the samples were analysed by a spectrometer (Infinite 200 Pro MPlex, Tecan, Männedorf, Switzerland) at 450 nm. The concentration is expressed as ng/mL compared to a standard curve (range from 20 to 7000 ng/L), and the results are expressed as percentage (%) versus control (line 0, untreated cells) from five independent triplicate tests.

4.19. PGC-1α ELISA Kit

Human peroxisome proliferator-activated receptor gamma coactivator-1 (PGC-1α) production was determined using an ELISA kit (MyBiosource, San Diego, CA, USA) to analyse PGC-1α in HepG2 cell lysates, according to the manufacturer’s instructions. Briefly, 40 µL of each sample was added to the wells with 10 µL of anti-PGC-1α antibody and the plate was incubated for 1 h at 37 °C. Then the plate was washed 5 times with Wash Buffer before adding 50 µL of Substrate Solution A and 50 µL of Substrate Solution B to each well for 10 min. Finally, 50 µL of Stop solution was added to each well and the samples were analysed by a spectrometer (Infinite 200 Pro MPlex, Tecan, Männedorf, Switzerland) at 450 nm. The concentration is expressed as ng/mL compared to a standard curve (range from 0.05 ng/mL to 30 ng/mL) and the results are expressed as percentage (%) versus control (line 0, untreated cells) from five independent triplicate tests.

4.20. AMPK ELISA Kit

Human AMPK alpha-1,2 (Phospho) [pT172] ELISA Kit (Thermo Fisher, Milan, Italy) was used to determine AMPK in HepG2 cell lysates, according to the manufacturer’s instructions [73]. The samples were analysed by a spectrometer (Infinite 200 Pro MPlex, Tecan, Männedorf, Switzerland) at 450 nm. The concentration is expressed as ng/mL relative to a standard curve, and the results are expressed as a percentage (%) relative to the control (line 0, untreated cells) from five independent triplicate tests.

4.21. Statistical Analysis

Data reported were obtained from at least five independent experiments performed in triplicate for each experimental protocol and analysed using Prism GraphPad 5 software 10.2.3 (GraphPad Software, La Jolla, CA, USA). Data were normalised relative to the control by averaging three replicates per sample and dividing each measurement by the control group mean. Results are presented as percentage-normalised values, with the control assigned a baseline of 0% to facilitate comparison across treatment conditions. Synergistic interactions were assessed post hoc using the Bliss independence model, comparing observed combination effects with predicted additive responses.

Results are reported as means ± SD using One-way ANOVA followed by Bonferroni post hoc test for statistical analysis. p-values < 0.05 were considered statistically significant. For certain parameters, synergism was examined post hoc using the Bliss model, in which the predicted additive effect from individual treatments was compared with the observed combined response.

5. Conclusions

The present study suggests a promising mechanistic framework of Lactiplantibacillus plantarum TJA7 (7.5 mg, corresponding to 3 × 10⁹ CFU/mL) and Morus alba leaf extract (100 mg/mL) eliciting coordinated biological effects across an in vitro gut–pancreas–liver axis. This combination modulated key cellular processes associated with glucose homeostasis, including intestinal barrier integrity, short-chain fatty acid production, pancreatic β-cell function, and hepatic glucose handling. The convergence of these effects supports the existence of a mechanistic interaction between the probiotic strain and the plant extract at the cellular and molecular level. Importantly, the experimental model employed does not directly recapitulate the complexity of diabetes in vivo; therefore, the present findings should be interpreted as mechanistic evidence derived from a controlled in vitro system. Further investigations in appropriate animal models and well-designed clinical studies will be necessary to validate the physiological relevance of these observations and determine whether this combined approach may contribute to future strategies supporting metabolic health.

Abbreviations

The following abbreviations are used in this manuscript:

Adv-DMEM	Advanced Dulbecco’s Modified Eagle Medium
AKT	Protein Kinase B
AMPK	AMP-Activated Protein Kinase
ATCC	American Type Culture Collection
CFU	Colony-Forming Units
DMEM	Dulbecco’s Modified Eagle Medium
DNJ	1-Deoxynojirimycin
ELISA	Enzyme-Linked Immunosorbent Assay
EMA	European Medicines Agency
EVOM	Epithelial Volt/Ohm Meter
FBS	Fetal Bovine Serum
FDA	Food and Drug Administration
GLUT2	Glucose Transporter Type 2
IRS1	Insulin Receptor Substrate 1
MTT	3-(4,5-Dimethylthiazol-2-yl)-2,5-Diphenyltetrazolium Bromide
OD	Optical Density
PGC-1α	Peroxisome Proliferator-Activated Receptor Gamma Coactivator 1 Alpha
PI3K	Phosphoinositide 3-Kinase
ROS	Reactive Oxygen Species
SCFAs	Short-Chain Fatty Acids
SOD	Superoxide Dismutase
TEER	Transepithelial Electrical Resistance
TJ	Tight Junction
TLR	Toll-Like Receptor

Appendix A

Table A1.

Bliss independence analysis on insulin production.

Treatment	Expected Data	Observed Data	Interaction
L. plantarum TJA7 + Mulberry Leaf extract	18.45%	19.80%	Synergistic

Table A2.

Bliss independence analysis on IRS1 levels.

Treatment	Expected Data	Observed Data	Interaction
L. plantarum TJA7 + Mulberry Leaf extract	43.30%	53.57%	Synergistic

Table A3.

Bliss independence analysis on GLUT2 levels.

Treatment	Expected Data	Observed Data	Interaction
L. plantarum TJA7 + Mulberry Leaf extract	18.18%	23.20%	Synergistic

Table A4.

Bliss independence analysis on PGC-1α levels.

Treatment	Expected Data	Observed Data	Interaction
L. plantarum TJA7 + Mulberry Leaf extract	21.14%	25.92%	Synergistic

Table A5.

Bliss independence analysis on AMPK levels.

Treatment	Expected Data	Observed Data	Interaction
L. plantarum TJA7 + Mulberry Leaf extract	32.24%	46.96%	Synergistic

Author Contributions

Conceptualization, F.P., R.G. and F.U.; methodology, F.P., R.G., S.M. and M.M.; software, F.P., R.G. and S.M.; validation, F.P., R.G. and S.M.; formal analysis, F.P. and R.G.; investigation, F.P., R.G., S.M. and M.M.; resources, F.U.; data curation, F.P., R.G. and F.U.; writing—original draft preparation, R.G., S.M., and F.U.; visualization, F.U.; supervision, F.U.; project administration, F.U.; funding acquisition, F.U. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The original contributions presented in the study are included in the article, further inquiries can be directed to the corresponding author.

Conflicts of Interest

Authors Francesca Parini and Rebecca Galla were employees of Noivita S.r.l. at the time the study was conducted. These authors had no role in the study design, data collection, analysis and interpretation of results, manuscript writing, or decision to submit it for publication. The other authors declare that the research was conducted without any commercial or financial relationships that could be construed as a potential conflict of interest.

Funding Statement

The research activities did not receive any specific funding. The publication costs of this manuscript were covered by the University of Eastern Piedmont. No external funds were used, and no grant number is associated with this work.