The central problem the researchers set out to solve was a familiar one in pharmacognosy, the study of medicines derived from natural sources. Mangiferin has been repeatedly associated with antidiabetic effects in the scientific literature, yet the specific protein through which it exerts this activity had never been conclusively defined. Traditional approaches to target identification can take years, requiring extensive biochemical screening of candidate proteins one at a time. The team instead turned to supervised machine learning, training computational models on a curated collection of known TGFBR1 inhibitors obtained from BindingDB, a public database of protein-small molecule binding data. By learning the molecular features that distinguish compounds capable of inhibiting this receptor, the models could then predict whether previously untested phytochemicals would behave similarly.

Several supervised learning algorithms were trained and compared, and the standout performer was a Random Forest model, an ensemble method that builds many decision trees during training and averages their predictions to reduce error and overfitting. The Random Forest classifier achieved an accuracy of 94.68 percent with a Cohen’s kappa coefficient of 0.8981, a statistic that measures agreement between predicted and actual classifications while correcting for chance. Crucially, the model also demonstrated robust performance on external validation, reaching 79.35 percent accuracy on 868 compounds it had never encountered during training. External validation of this kind is a critical safeguard in computational drug discovery, because models that perform well only on their training data often fail when confronted with chemically novel structures in the real world.

With a validated predictive model in hand, the researchers screened a library of 545 phytochemicals derived from Mangifera indica. Five compounds emerged as high-confidence predicted inhibitors of TGFBR1: mangiferin itself, along with fisetin, isochinomin, mangiferolic acid, and astragalin. The identification of mangiferin at the top of this list provided computational support for the hypothesis that this abundant mango xanthone could interact with the TGF-beta signaling pathway, a cascade involved in cell growth, immune regulation, and metabolic processes that has been increasingly implicated in the tissue damage associated with diabetes and its complications.

To place these individual predictions in a broader biological context, the team employed network pharmacology, an approach that maps the interactions among drugs, targets, diseases, and signaling pathways to reveal the most influential nodes in a biological system. Their analysis identified TGF-beta as the most central protein in the interaction network, with a cumulative score of 350, reinforcing the idea that the TGF-beta pathway sits at the heart of mangiferin’s antidiabetic mechanism. Molecular docking studies then examined how mangiferin fits into the binding site of TGFBR1 at atomic resolution. The compound was found to form stable hydrogen bonds with three amino acid residues in the receptor’s binding pocket, serine 287, isoleucine 211, and glutamic acid 208, supplemented by hydrophobic contacts that help anchor the ligand in place.

Docking provides a static snapshot, so the researchers next turned to molecular dynamics simulations to test whether the predicted binding pose would hold up under realistic, thermally fluctuating conditions. Over a simulation lasting 100 nanoseconds, the mangiferin-TGFBR1 complex remained structurally stable, with root-mean-square deviation values between 1.50 and 1.75 angstroms, a range indicating that the protein-ligand complex did not drift apart or undergo disruptive conformational changes. Ligand-residue contact occupancy exceeded 85 percent, meaning that for the vast majority of the simulation the key interactions between mangiferin and the binding site residues were maintained. This level of dynamic stability is an encouraging sign that the predicted interaction reflects genuine physical binding rather than a fleeting computational artifact.

The team also assessed whether mangiferin possesses the chemical properties expected of a viable drug candidate. ADMET profiling, which evaluates absorption, distribution, metabolism, excretion, and toxicity, showed favorable drug-likeness, including compliance with Lipinski’s rule of five, a widely used heuristic for oral bioavailability, and low predicted toxicity. One limitation emerged: mangiferin’s topological polar surface area was calculated at 201.28 square angstroms, a value suggesting moderate permeability limitations that could affect how efficiently the compound crosses biological membranes. Complementary density functional theory calculations yielded a HOMO-LUMO gap of 2.85937 electron volts, an electronic property consistent with antioxidant activity, which is relevant given the role of oxidative stress in diabetic pathology.

The computational predictions were then subjected to experimental scrutiny in cultured cells. When atorvastatin was combined with mangiferin, the treatment increased apoptosis through the mitochondrial pathway, a form of programmed cell death characterized by a specific molecular sequence. The researchers observed downregulation of the anti-apoptotic protein Bcl-xL, upregulation of the pro-apoptotic protein BAK, release of cytochrome c from mitochondria into the cytoplasm, and activation of caspase-7 and caspase-8, the enzymatic executioners of the apoptotic program. Alongside these effects, the combination enhanced insulin sensitivity, demonstrated by increased glucose uptake in a 2-NBDG assay, a technique that uses a fluorescent glucose analog to track sugar transport into cells, and by elevated expression of the glucose transporter GLUT-4 and the phosphorylated insulin receptor substrate pIRS-1 at tyrosine position 896, both key markers of active insulin signaling.

The most clinically suggestive results came from animal studies. Rats fed a high-fat diet to induce metabolic dysfunction were treated with atorvastatin plus mangiferin for 24 weeks. Compared with untreated high-fat diet controls, the treated animals showed significantly improved glucose tolerance and reduced levels of HbA1c, the long-term glycemic marker reflecting average blood sugar over weeks to months, as well as lower total cholesterol, triglycerides, and random blood sugar measurements. Histological examination revealed preserved pancreatic architecture, indicating structural protection of the insulin-producing tissue, and muscle tissue from treated animals displayed increased GLUT-4 expression, consistent with enhanced peripheral glucose disposal. Together, these results suggest that the combination acts on multiple fronts simultaneously, improving lipid profiles through the statin component while the mango-derived compound contributes glycemic control through its newly characterized target.

The study, led by Raghuvir Keni, Farmiza Begum, Niraja Ranadive, Pawan G. Nayak, and Nandakumar Krishnadas of Manipal College of Pharmaceutical Sciences, together with collaborators at institutions including JSS Academy of Higher Education and Research in Mauritius and Vels Institute of Science, Technology and Advanced Studies in Chennai, represents a template for how artificial intelligence can compress the early stages of drug discovery. Rather than years of trial-and-error biochemical screening, a trained machine learning model evaluated hundreds of natural compounds in silico, flagged the most promising candidates, and directed experimental resources toward the hypotheses most likely to succeed. The authors caution that their findings, while spanning computation, cell culture, and animal models, remain preclinical, and the moderate permeability limitation identified in the ADMET analysis highlights a formulation challenge that would need to be addressed before any human application. Nevertheless, by naming TGFBR1 as a plausible molecular target of mangiferin and demonstrating that a mangiferin-atorvastatin combination improves metabolic parameters in a disease-relevant animal model, the research provides a concrete mechanistic foundation for a therapy concept that pairs a common fruit-derived molecule with one of the world’s most widely used prescription drugs. As diabetes continues to affect hundreds of millions of people worldwide according to the International Diabetes Federation, such repurposing strategies, grounded in computational prediction and validated step by step in the laboratory, offer a potentially faster and cheaper path to new treatment options than de novo drug design.

Subject of Research: Machine learning-driven identification of TGFBR1 as a molecular target of mangiferin from Mangifera indica and evaluation of a mangiferin-atorvastatin combination for antidiabetic therapy

Article Title: Machine learning-driven discovery of TGFBR1 inhibitors from Mangifera indica for Atorvastatin-based antidiabetic therapy

Article References: Keni, R., Begum, F., Ranadive, N., Nayak, P. G., Murugesan, V., Pradeep, K., Deshpande, M., Krishnadas, N., & Theivendren, P. (2026). Machine learning-driven discovery of TGFBR1 inhibitors from Mangifera indica for Atorvastatin-based antidiabetic therapy. 3 Biotech, 16(11), Article 458. https://doi.org/10.1007/s13205-026-05088-0

Image Credits: AI Generated

DOI: 10.1007/s13205-026-05088-0

Keywords: TGFBR1, mangiferin, Mangifera indica, machine learning, atorvastatin, diabetes, network pharmacology, molecular dynamics, drug discovery, phytochemicals, insulin sensitivity, Random Forest

The challenge the researchers set out to solve is one of geometry and biology combined. Most laboratory studies of Kawasaki disease rely on flat, two-dimensional sheets of endothelial cells, the thin lining that coats the inside of blood vessels. Such cultures are easy to grow and easy to analyze, but they bear little resemblance to the real thing. A coronary artery is a hollow, curved, multilayered tube in which endothelial cells interface with blood on one side and smooth muscle cells wrap the exterior, providing structural support. Animal models, particularly mice, have been used for decades, but the differences between mouse and human vascular and immune biology often limit how well findings translate to the clinic. The team, led by Xuting Zhang and colleagues at Wenzhou Medical University and collaborating institutions, hypothesized that recreating the true three-dimensional, bilayered architecture of a coronary artery would yield a far more faithful model of the disease.

Their manufacturing method of choice was microfluidic coaxial bioprinting, an elegant technique that exploits the same principle as a pipe within a pipe. The researchers assembled a device from glass capillaries and injection needles, aligning an inner capillary concentrically inside an outer one. Two fluids were pumped through this coaxial nozzle: an outer bioink containing 10 percent gelatin methacryloyl, or GelMA, blended with polyethylene glycol diacrylate, or PEGDA, and an inner sacrificial stream of 10 percent polyvinyl alcohol. When the co-flowing streams emerged from the capillary tip, ultraviolet light at 405 nanometers triggered photopolymerization of the outer hydrogel, locking in a continuous tube. The uncrosslinked PVA core was then simply washed away, leaving behind a hollow hydrogel microfiber with an open lumen running its entire length.

The precision achievable with this setup is remarkable. By adjusting the relative flow rates of the inner and outer fluids, the team could dial in the dimensions of the fibers at will. Holding the inner flow at 15 milliliters per hour and raising the outer flow from 15 to 35 milliliters per hour shifted the ratio of inner to outer diameter from about 65.5 percent down to 58.4 percent, while the outer diameter swelled from roughly 712 to 911 micrometers. Reversing the adjustment, by accelerating the inner stream while holding the outer one constant, enlarged both the overall caliber and the relative lumen. Throughput was equally impressive: a single milliliter of bioink yielded between roughly 4 and 6.5 meters of continuous hollow fiber, depending on the diameter chosen, underscoring the scalability of the approach.

Mechanical tuning came from the PEGDA content. Three formulations containing 2.5, 5, and 7.5 percent PEGDA in 10 percent GelMA produced hydrogels with Young’s moduli of approximately 73, 153, and 221 kilopascals respectively, and rheological testing confirmed the same trend in storage modulus. Swelling behavior plateaued within 24 hours, and the matrices retained more than 95 percent of their mass after a day in culture medium, yet dissolved completely within 24 hours when exposed to the enzyme collagenase, showing that degradation is enzymatically controllable rather than a product of passive hydrolysis. Importantly, the softest formulation, 10 percent GelMA with 2.5 percent PEGDA, proved best for early cell attachment, because PEGDA, while excellent for structural strength, tends to resist cell adhesion. That formulation was selected for all subsequent vascular constructs.

With the scaffold optimized, the researchers populated it with human cells in a deliberate sequence. First, human umbilical vein endothelial cells, or HUVECs, at a density of 10 million cells per milliliter were injected into the lumen and allowed 72 hours to engraft, forming a continuous endothelial monolayer with uniform circumferential alignment. Then human coronary artery smooth muscle cells, at 100 million cells per milliliter, were applied to the outer surface, where they colonized the exterior and formed a distinct smooth muscle layer. Confocal microscopy of the cytoskeleton and immunostaining confirmed the architecture: the endothelial marker CD31 lined the luminal surface, while alpha-smooth muscle actin was confined to the outer layer, faithfully reproducing the bilayered organization of a native coronary artery.

Crucially, the model was more than a structural mimic. When the team perfused fluorescent dextran molecules of 4 and 70 kilodaltons through the lumen, the endothelialized constructs allowed far less leakage into the surrounding hydrogel than acellular controls, demonstrating that the engineered endothelium functioned as a selective barrier to both small solutes and large plasma proteins, just as a real vessel does. As a positive control, exposure to the inflammatory molecule TNF-alpha for six hours triggered a cascade of endothelial activation: elevated lactate dehydrogenase release signaling membrane injury, increased oxidative stress, upregulation of the adhesion molecules ICAM-1, VCAM-1, and E-selectin, a surge of inflammatory cytokines and chemokines including IL-1beta, IL-6, IL-8, CX3CL1, CCL5, and MCP-1, and clear activation of the NF-kappaB signaling pathway, traced by the phosphorylation of I-kappaB-alpha and the p65 subunit.

The decisive test came when the constructs were bathed in serum drawn from children during the acute phase of Kawasaki disease, before intravenous immunoglobulin therapy, mixed at a 10 percent volume ratio into the culture medium, with serum from age-matched healthy children as the control. Within six hours, the patient serum drove the engineered arteries into a state that mirrored the disease: heightened endothelial injury and oxidative stress, intensified expression of all three adhesion molecules, a broad transcriptional inflammatory response, and robust NF-kappaB activation. The serum also triggered pyroptosis, a fiery form of programmed cell death, evidenced by upregulation of the inflammasome sensor NLRP3, the executioner protein gasdermin D, and caspase-1. This finding aligns with growing evidence that pyroptotic endothelial death contributes to the coronary artery lesions that make Kawasaki disease dangerous, and it suggests that NF-kappaB signaling and the NLRP3 inflammasome operate in a reinforcing loop rather than as separate, sequential events.

The platform then proved its worth as a drug-testing instrument. Pretreating the constructs with atorvastatin, the widely used statin, at a non-toxic concentration of 10 micromolar significantly blunted the inflammatory response to Kawasaki serum, suppressing adhesion molecule and cytokine expression and dampening NF-kappaB phosphorylation, consistent with the statin’s known anti-inflammatory, lipid-independent pleiotropic effects and with early clinical trials of atorvastatin in Kawasaki patients with coronary aneurysms. The selective NLRP3 inhibitor MCC950, also tested at 10 micromolar, likewise reduced ICAM-1 expression and inflammatory mediator transcripts, and it specifically shut down the pyroptotic machinery, lowering NLRP3, gasdermin D, caspase-1, and IL-1beta. Because intravenous immunoglobulin resistance remains a major clinical problem, the ability to distinguish these two mechanistically distinct protective strategies within a single human tissue model is exactly the kind of preclinical capability the field has lacked.

The authors are candid about the model’s current limits. The constructs were cultured under static conditions, without the fluid flow and shear stress that shape endothelial behavior in living arteries, and future versions will incorporate dynamic perfusion. The endothelial layer consists of HUVECs rather than coronary-specific endothelial cells, a pragmatic choice driven by the robust proliferation needed to form an intact barrier in a bioprinted scaffold. Even so, the achievement stands out. In a single platform, the team has combined tunable biofabrication, native-like multilayered architecture, functional barrier properties, patient-derived disease induction, and mechanistically resolved drug evaluation. For a disease that strikes young children, has no known cause, and can leave permanent cardiac damage, this bioprinted coronary artery equivalent offers a powerful new window into pathogenesis and a versatile testing ground for the therapies of tomorrow.

Subject of Research: Biomimetic 3D bioprinted coronary artery models for Kawasaki disease modeling and drug screening

Article Title: Engineering biomimetic coronary artery equivalents for Kawasaki disease modeling and drug screening

Article References: Zhang, X., Shao, C., Wang, Y., Liu, A., Rong, X., Jia, C., Yu, W., Zhang, H., Wu, R., Zhang, J., Zhang, L., Ye, F., & Chu, M. (2026). Engineering biomimetic coronary artery equivalents for Kawasaki disease modeling and drug screening. Materials Today Bio, 41, Article 103701. https://doi.org/10.1016/j.mtbio.2026.103701

Image Credits: AI Generated

DOI: 10.1016/j.mtbio.2026.103701

Keywords: Kawasaki disease, coronary artery, bioprinting, microfluidics, GelMA, endothelial dysfunction, NF-kB, pyroptosis, NLRP3, atorvastatin, MCC950, drug screening

Nasopharyngeal carcinoma is a distinctive malignancy arising from the epithelium lining the nasopharynx, the space behind the nose and above the throat. Its geographic distribution is strikingly uneven: while rare in most of Europe and North America, it is endemic in Guangdong province, Hong Kong, and much of Southeast Asia, a pattern linked to Epstein-Barr virus infection, genetic susceptibility, and dietary factors such as salt-cured fish. Decades of progress in diagnostic imaging, intensity-modulated radiotherapy, and systemic chemotherapy have pushed five-year overall survival above 70 percent, transforming the clinical challenge. Yet survivors who beat their primary tumor remain at elevated risk of dying from recurrent disease, secondary cancers, cardiovascular complications of radiotherapy, and pulmonary illness, and until now there has been no established intervention specifically aimed at improving long-term survivorship in this growing population.

The new study, known as the Hong Kong NPC Survivor Study, screened 18,135 patients diagnosed with nasopharyngeal carcinoma between 1997 and 2015 and focused on 7,872 people who survived at least five years after diagnosis, the point at which most treatment-related mortality has subsided. Within this cohort, 2,439 patients had filled prescriptions totaling at least 90 days of statin therapy, while 5,433 had no record of statin use. The researchers drew their data from the Clinical Data Analysis Reporting System, a territory-wide electronic database maintained by Hong Kong’s Hospital Authority that captures more than 90 percent of the region’s cancer services across a population of 7.3 million, including demographics, diagnoses, prescriptions, laboratory results, and coded causes of death.

The headline result was dramatic. Statin users showed markedly better overall survival than non-users, with a hazard ratio of 0.46 in multivariable models, corresponding to roughly a 54 percent reduction in the risk of death from any cause. At ten years after diagnosis, 89.1 percent of statin users were alive compared with 71.8 percent of non-users; at fifteen years the figures were 77.2 percent versus 55.1 percent, and at twenty years 59.2 percent versus 42.4 percent. Because statin users in the raw cohort were older and carried more cardiovascular comorbidities such as diabetes, hypertension, and dyslipidemia, the team applied inverse probability of treatment weighting, a statistical technique that reweights the comparison groups to mimic randomization. After weighting, the apparent benefit grew even stronger, with hazard ratios of 0.39 under both the average treatment effect and average treatment effect on the treated frameworks.

Observational studies of this kind are notoriously vulnerable to immortal time bias, the artifact that arises when patients must survive long enough to receive a treatment before being counted as treated. The investigators confronted this problem head-on by modeling statin exposure as a time-varying covariate, so that each patient contributed unexposed person-time until they actually began taking the drug. Under this more conservative analysis the hazard ratio settled at 0.71, still a highly significant 29 percent reduction in mortality, and the authors candidly note that the unadjusted estimates likely overstate the true effect. They also performed a target trial emulation, a design that mimics the structure of a randomized trial by matching 1,332 statin initiators to non-initiators on age, sex, smoking, alcohol use, and baseline eligibility, and again found a significant protective effect with a hazard ratio of 0.54. Sensitivity analyses excluding patients with early recurrence or metastasis, and a landmark analysis resetting time zero at five years post-diagnosis, produced consistent results.

Perhaps the most intriguing findings concern why statin users died less often. Among the 2,773 deaths in the cohort, statin users had significantly lower risks of death from secondary primary malignancies, a hazard ratio of 0.51 after weighting, from cardiovascular disease, from pulmonary disease, and, in multivariable models, from recurrence or progression of the original cancer. This pattern suggests the drugs may be doing more than protecting the heart. Statins block HMG-CoA reductase, the rate-limiting enzyme of the mevalonate pathway, which supplies cancer cells with essential lipid intermediates for membrane synthesis and protein prenylation. Beyond this, laboratory work has documented anti-inflammatory, antioxidant, anti-angiogenic, and immune-modulatory properties, and statins have been shown to enhance chemotherapy efficacy and reduce radiation-induced vascular injury in nasopharyngeal carcinoma specifically. The reduced mortality from secondary cancers and recurrence hints that suppressing the mevalonate pathway may inhibit both new carcinogenesis and the outgrowth of residual tumor cells.

The benefit was strongly duration-dependent. Compared with short-term use of three months to two years, the risk of death fell by 30 percent with two to four years of statin therapy, by 45 percent with four to six years, and by 69 percent with more than six years of use. Notably, whether patients started statins during concurrent chemoradiotherapy or only after completing it made no significant difference to survival, implying that sustained exposure matters more than timing. This duration-response relationship fits a cytostatic rather than cytotoxic model of action: unlike chemotherapy, which rapidly shrinks tumors, statins may hold dormant disease in check over years, an effect that only accumulates with prolonged use. Parallel findings in prostate and breast cancer cohorts, where multi-year statin use was likewise associated with the largest mortality reductions, support this interpretation.

The type of statin also appeared to matter. All statin classes were associated with lower mortality, but the strongest signals came from atorvastatin and rosuvastatin, second- and third-generation agents with longer half-lives and higher affinity for HMG-CoA reductase, with hazard ratios of 0.35 and 0.30 respectively. Compared with first-generation drugs such as simvastatin and pravastatin, second-generation agents cut mortality risk by about a third and third-generation agents by more than half. Interestingly, the researchers found no clear advantage of lipophilic over hydrophilic statins, contrary to some prior laboratory evidence, suggesting that lipid-lowering potency and favorable pharmacokinetics may be more important than membrane penetration alone. In a telling negative control, statin users actually showed a higher risk of death from mental and neural disorders, a biologically unrelated endpoint; if the entire survival benefit were an artifact of healthy-user bias, statins would have appeared protective across the board.

The authors are careful to frame these results as hypothesis-generating rather than practice-changing. The retrospective design cannot exclude residual confounding by unmeasured factors such as lipid levels, body mass index, diet, and physical activity, and excluding patients who died within five years may have selectively enriched the cohort for indolent disease. Previous randomized trials of statin add-on therapy in patients with advanced cancer and poor prognosis showed no survival benefit, and a meta-analysis of 60 observational studies across cancer types found only a 22 percent reduction in cancer-specific mortality, far smaller than the signal reported here. Yet the Hong Kong cohort is by far the largest and longest-followed study in nasopharyngeal carcinoma, and its rigorous bias-control methods lend unusual credibility to the association.

The practical implications are considerable. Immunotherapy has recently demonstrated a 37 percent reduction in the risk of death in recurrent or metastatic nasopharyngeal carcinoma, but its cost places it far beyond most patients in the endemic regions of Southeast Asia and Southern China. Statins, by contrast, are generic, affordable, well tolerated, and carry a decades-long safety record. If prospective randomized trials confirm that sustained statin therapy improves long-term survival in cancer survivors, the drugs could offer an accessible adjunctive strategy for the thousands of patients who beat nasopharyngeal carcinoma each year but remain at lifelong risk, particularly in low- and middle-income countries where the burden of this virus-driven cancer is heaviest.

Subject of Research: The effect of statin use on long-term mortality among survivors of nasopharyngeal carcinoma

Article Title: Effect of statins on the mortality of long-term survivors of nasopharyngeal carcinoma: a population-based study in endemic area

Article References: Xu, R.-Y., Au, P. C.-M., Chan, K. S.-K., Cheung, C.-L., Fong, J. K.-S., Sing, C.-W., Chiu, M. K.-L., Chow, J. C.-H., Cheung, K. K.-M., Wong, E. C.-Y., Lau, T. T.-S., Chan, A. S.-Y., Wong, K. C.-W., Tin, W. W.-Y., Lee, V. H.-F., Lee, A. W.-M., Ng, W.-T., Wong, I. C.-K., & Chiang, C.-L. (2026). Effect of statins on the mortality of long-term survivors of nasopharyngeal carcinoma: a population-based study in endemic area. The Lancet Regional Health – Western Pacific, 75, Article 101993. https://doi.org/10.1016/j.lanwpc.2026.101993

Image Credits: AI Generated

DOI: 10.1016/j.lanwpc.2026.101993

Keywords: statins, nasopharyngeal carcinoma, cancer survivorship, Hong Kong cohort study, overall survival, mevalonate pathway, secondary primary malignancy, cardiovascular mortality, target trial emulation, immortal time bias, atorvastatin, rosuvastatin