Search Results (45)

Huperzine A (HupA) is a natural Lycopodium alkaloid known for its potent neuroprotective properties through the inhibition of acetylcholinesterase. Nevertheless, the limited understanding of its biosynthesis restricts its broader application. This study integrates full-length and second-generation transcriptomes with the quantification of HupA and its precursor, huperzine B, across various tissues of Huperzia serrata, the primary source plant. By employing phylogenetic clustering, expression profiling, and correlation analysis between gene expression and metabolite abundance, we identified 71 candidate genes from seven enzyme families potentially involved in the synthesis of the HupA backbone, including lysine/ornithine decarboxylases, copper amine oxidases (CAOs), chalcone synthases, and cytochrome P450 monooxygenases. Additionally, 28 genes from two families were identified for modification reactions, specifically 2-oxoglutarate/Fe(II)-dependent dioxygenases and caffeoyl shikimate esterases. Comparative analysis between young and mature leaves revealed 3801 genes with higher expression in young leaves, with 84 showing a high correlation with HupA content across seven families. Protein–protein interaction network analysis indicated possible interactions with transcription factors from the MYB, NF-YC, GRAS, ERF, BHLH, and SAP families. Functional validation of two candidate CAOs in planta confirmed their catalytic roles in amine/alkaloid metabolism. This study provides a theoretical foundation and a set of candidate genes for elucidating the biosynthetic pathway of HupA and related alkaloids in H. serrata. Full article

Foundation time-series models have recently shown potential for forecasting complex temporal signals, but their behavior in patient-specific continuous glucose monitoring (CGM) forecasting remains insufficiently understood, particularly when only glucose history is available. This study provides a patient-level benchmark of foundation models for 30 min ahead glucose prediction in adults with type 1 diabetes mellitus (T1DM) under a strictly univariate CGM-only setting. Using the HUPA–UCM dataset from 25 individuals, we evaluated TimeGPT, Chronos, and Sundial against representative statistical, machine learning, and deep learning forecasters, including ARIMA, ETS, gradient-boosting models, recurrent networks, and neural forecasting architectures. Models were assessed using a local walk-forward validation strategy over the final 24 h of CGM data for each patient. Foundation models achieved the strongest global performance, with Sundial obtaining the lowest overall MAE ($6.06\mathrm{mg}/\mathrm{dL}$), while TimeGPT and Chronos remained among the most competitive approaches. However, patient-level analyses showed that this advantage was not uniform: ARIMA remained highly competitive in selected individuals, and no single model consistently dominated across the cohort. These findings suggest that foundation time-series models are promising tools for short-horizon CGM forecasting, but their use should be framed within patient-specific model selection rather than as universal replacements for classical forecasting methods. Full article

Background/Objectives: Trichomonas vaginalis is one of the most prevalent non-viral sexually transmitted infections worldwide. Although Gram staining is routinely performed in clinical microbiology laboratories for the evaluation of vaginal samples, it is not considered a diagnostic method for T. vaginalis, which represents a missed diagnostic opportunity in routine practice. This study aimed to evaluate an artificial intelligence (AI)-assisted diagnostic approach for the identification of T. vaginalis directly from routine Gram-stained vaginal smears. Methods: A retrospective dataset of Gram-stained vaginal smear images was analysed using a cascaded AI-based framework combining image processing and classification. The image selection and quality control were performed under the supervision of a specialised clinical microbiologist. All cases were independently confirmed by polymerase chain reaction (PCR), which served as the reference diagnostic standard. Model performance was assessed using standard diagnostic metrics, including accuracy, sensitivity, specificity, area under the receiver operating characteristic curve (AUC), Cohen’s kappa, and Matthews correlation coefficient (MCC). Held-out independent testing was used to assess generalisability beyond the internal validation subset. Results: The proposed AI-assisted approach demonstrated high diagnostic performance for the identification of T. vaginalis, achieving an AUC of 0.973, Cohen’s kappa of 0.87, and an MCC of 0.87. The system showed high diagnostic concordance with PCR results across both internal and external validation datasets, supporting the feasibility and reproducibility of the approach under routine laboratory conditions. Conclusions: This study shows that artificial intelligence may enhance the diagnostic utility of routinely performed Gram-stained vaginal smears by enabling reliable identification of T. vaginalis. The proposed approach could be integrated into standard microbiology workflows as an objective decision-support or triage adjunct, facilitating early identification and supporting clinical decision-making without altering existing laboratory procedures. Full article

Bioactive glass (BAG) S53P4 is a clinically approved bone substitute with antibacterial, osteoconductive and osteostimulatory properties. Its antibacterial effect is associated with ion release, local pH elevation and osmolality, but the precise biochemical and biophysical mode-of-action is unclear. This study investigates the antibacterial mechanism of BAG S53P4 eluates. BAG eluates, collected at 2, 4, 8, and 24 h, eradicated Staphylococcus aureus. Elemental analysis revealed an early increase in concentrations of Si and Na, a later rise in Ca, depletion of P over time and rapid loss of Mg. Membrane disturbances occurred within 5 min, evident by permeability for SYTOX, aligning with time-kill kinetics for S. aureus and Bacillus subtilis. In B. subtilis, 2h-BAG-eluate induced rapid delocalization of marker proteins for cell division and DNA repair, signaling membrane potential collapse and nucleoid condensation. Transcriptomics revealed early transcription remodeling reflecting ionic and energetic imbalance, including disruption of central metabolism, redox homeostasis, and translational stability. Scanning electron microscopy revealed severe cell surface damage and particulate deposits on S. aureus. Transmission electron microscopy showed cell envelop disruptions and cytoplasmic leakage. Energy dispersive X-ray analysis identified Si on bacterial cell surface at 4 h and intracellular accumulation in punctured, empty cells at 24 h. Overall, BAG ionic dissolution products kill bacteria through a stepwise mechanism involving membrane damage, protein delocalization and metabolic impairment, accompanied by Si deposition on bacterial surfaces and loss of Mg. This finally leads to cell wall degradation, cytoplasmic content leakage and further Si deposition on the cells and inside cell ghosts. Full article

Managing blood glucose in type 1 diabetes (T1D) remains a daily clinical challenge, and accurate short-term prediction of glucose levels can meaningfully improve insulin dosing decisions while reducing the risk of dangerous hypoglycaemic episodes. Although numerous machine learning approaches have been proposed for this task, comparing their relative merits is difficult because published studies differ widely in datasets, preprocessing choices, and evaluation criteria. In this work, we address this research gap by benchmarking ten machine learning methods—from a naïve persistence baseline through classical linear regressors, gradient-boosted ensembles, and recurrent neural networks to a novel hybrid that couples LightGBM with stochastic differential equation (SDE)-based glucose–insulin simulation—on two multi-patient datasets comprising 34 T1D subjects, across prediction horizons of 15, 30, 60, and 120 min. Every method is trained and tested under identical preprocessing and temporal splitting conditions to ensure a fair comparison. The proposed Hybrid LightGBM-SDE model consistently outperforms all alternatives, recording RMSE values of 22.42 mg/dL at 15 min, 28.74 mg/dL at 30 min, 33.89 mg/dL at 60 min, and 37.22 mg/dL at 120 min—an improvement of between 13.6% and 27.0% relative to standalone LightGBM. At the clinically important 30 min horizon, 99.7% of predictions lie within the acceptable A and B zones of the Clarke Error Grid. Wilcoxon signed-rank tests confirm that performance differences are statistically significant (p < 10⁻¹⁰), and SHAP-based analysis shows that the SDE-derived simulation features are among the most influential predictors, especially at longer horizons. All source code and evaluation scripts are publicly released to support reproducibility. Due to temporary data access constraints, all experiments reported here use physics-based synthetic datasets generated from the Bergman minimal model, replicating the structural properties of the D1NAMO and HUPA-UCM collections; validation on the original clinical recordings is planned. Among the two synthetic datasets, the D1NAMO-equivalent cohort (nine patients) proves more challenging, with systematically higher per-patient RMSE variance. The clinically acceptable prediction accuracy at the 30 min horizon (99.7% in Clarke zones A + B) suggests potential for integration into insulin dosing decision-support systems. Full article

Huperzine A is a preferred treatment option for Alzheimer’s disease. Huperzia serrata (Thunb. ex Murray) Trev. (H. serrata) has garnered significant attention for its ability to produce Huperzine A (HupA). However, natural populations of wild H. serrata (WH) are rapidly declining. Fortunately, our group obtained two types of H. serrata thalli (OT and ST) capable of stably producing Huperzine A, which have the potential to serve as an alternative resource to WH. To evaluate the feasibility of this strategy, we conducted a comprehensive assessment of both WH and H. serrata thallus. The results indicated that compared to WH, ST and OT exhibited stronger anti-inflammatory and antioxidant activities, with lower cytotoxicity. Notably, ST demonstrated a strong radical scavenging activity, reaching 93.23% (DPPH at 0.2 μg/mL) and 99.87% (ABTS at 4 μg/mL), and reduced nitrite production from 10.29 μM to 6.51 μM at 50 µg/mL. GC-MS and widely targeted metabolomics analyses revealed that the higher antioxidant and anti-inflammatory activities for ST and OT were due to higher concentrations of phenolic acids and flavonoids compared to WH. In addition, the HupA content in ST reached 36.56% of that found in WH. KEGG enrichment analysis revealed that the flavonoid, phenylalanine, and phenylpropanoid biosynthesis pathways may be involved in regulating the antioxidant activity. P-coumaroyl quinic acid and caffeoyl quinic acid are the crucial metabolites for antioxidant activity. These findings suggested that the H. serrata thallus could serve as a sustainable alternative to WH. Full article

Background/Objectives: Accurate blood glucose forecasting is critical for closed-loop insulin delivery systems to support effective disease management in people with type 1 diabetes (T1D). While long short-term memory (LSTM) neural networks have shown strong performance in glucose prediction tasks, the relative performance of individualized versus aggregated training remains underexplored. Methods: In this study, we compared LSTM models trained on individual-specific data to those trained on aggregated data from 25 T1D subjects using the HUPA UCM dataset. Results: Despite having access to substantially less training data, individualized models achieved comparable prediction accuracy to aggregated models, with mean root mean squared error across 25 subjects of 22.52 ± 6.38 mg/dL for the individualized models, 20.50 ± 5.66 mg/dL for the aggregated models, and Clarke error grid Zone A accuracy of 84.07 ± 6.66% vs. 85.09 ± 5.34%, respectively. Subject-level analyses revealed only modest differences between the two approaches, with some individuals benefiting more from personalized training. Conclusions: These findings suggest that accurate and clinically reliable glucose prediction is achievable using personalized models trained on limited individual data, with important implications for adaptive, on-device training, and privacy-preserving applications. Full article

Huperzine A (HupA) can alleviate Alzheimer’s disease due to its reversible inhibition of acetylcholinesterase (AChE). The chemical synthesis and plant extraction of HupA is plagued by route complexity and resource scarcity, respectively. Although some endophytic fungi from Huperzia serrata can independently biosynthesize HupA, their yields are trivial. After a comprehensive investigation of HupA-producing H. serrata across China, we focused on the endophytic fungi from Hunan and Hubei provinces, which demonstrated high-level HupA. Morphological characteristics and internal transcriptional sequence (ITS) analysis revealed their diversity. Among the four HupA-producing endophytic fungi, Colletotrichum kahawae is the best-performing and was thus subjected to fermentation optimization. When its fermentation medium was supplemented with H. serrata flavonoids daidzein and apigenin, HupA yields reached 58.38 μg/g (dry cell weight, dcw) and 72.21 μg/g dcw, respectively. In contrast, the addition of L-lysine and H. serrata extracts led to yields of 50.17 μg/g dcw and 255.32 μg/g dcw, respectively. Transcriptomic analysis revealed that H. serrata extracts substantially upregulated the expression of HupA biosynthesis genes in C. kahawae. Overall, H. serrata extracts outperformed L-lysine, daidzein, and apigenin in boosting HupA production, as they encompass all the necessary nutrients for C. kahawae growth. This study not only connotes a nutritional exchange between H. serrata and C. kahawae during long-term coevolution but also offers insights for harnessing plant extracts for the overproduction of desired metabolites in endophytic fungi. Full article

Background: The callus cultures from the fronds of the lycophyte Phlegmariurus taxifolius produce the huperzine A (HupA) alkaloid, which is used in Alzheimer’s disease treatment. This study aimed to establish the growth kinetics and HupA production by the newly HupS21 cell line grown in 250 mL flasks and in a 2 L airlift bioreactor. Methods: Batch-type kinetics were carried out for 60 days in 250 mL flasks and for 20 days in a 2 L airlift bioreactor. Measurements of dry weight (DW), specific growth rate (μ), doubling time (dt), pH, carbohydrate consumption, and HupA quantification were performed. The acetylcholinesterase (AChE) inhibitory assay of the HupS21 alkaloidal extract was determined. Results: The 250 mL flasks kinetic reached a maximum cell growth of 8.17 g/L DW, with a μ of 0.045 day⁻¹ and a dt of 15.40 days. The maximum HupA production was of 2.03 μg/g DW at day 45. In the 2 L airlift reactor, a maximum growth of 16.70 g/L DW, a μ of 0.062 day⁻¹, a dt of 11.20 days, and HupA production of 2.48 μg/g DW at day 15 were obtained. The alkaloidal extract from the HupS21 cell line at 100 μg/mL showed an AChE inhibitory activity of 85.6 ± 1.27%. Conclusions: The airlift reactor outperformed the flask cultures in maximum cell growth, specific growth rate, doubling time, and HupA production. To our knowledge, this research is the first report on the establishment of suspension cell cultures of P. taxifolius in shaken flasks and in an airlift bioreactor, providing a foundation for scaling up HupA production for pharmaceutical use. Full article

Various surface modifications to increase the lifespan of cobalt–chromium (CoCr) joint prostheses are being studied to reduce the wear rate in bone joint applications. One recently proposed modification involves depositing graphene oxide functionalized with hyaluronic acid (a compound present in joints) on CoCr surfaces, which can act as a solid lubricant. This paper analyzes the biological alterations caused by wear–corrosion phenomena that occur in joints, both from the perspective of the worn surface (in vitro model) and the particles generated during the wear processes (in vivo model). The analysis of the inflammatory response of macrophage was performed on CoCr surfaces modified with graphene oxide and functionalized with hyaluronic acid (CoCr-GO-HA), before and after wear–corrosion processes. The wear particles released during the wear–corrosion tests of the CoCr-GO-HA/CoCr ball pair immersed in 3 g/L hyaluronic acid were intra-articularly injected into the experimental animals. The hematological analysis in vivo was made considering a murine model of intra-articular injection into the left knee in male adult Wistar rats, at increasing concentrations of the collected wear particles dispersed in 0.9% NaCl. Non-significant differences in the inflammatory response to unworn CoCr-GO-HA surfaces and control (polystyrene) were obtained. The wear–corrosion of the CoCr-GO-HA disk increased the inflammatory response at both 72 and 96 h of material exposure compared to the unworn CoCr-GO-HA surfaces, although the differences were not statistically significant. The pro-inflammatory response of the macrophages was reduced on the worn surfaces of the CoCr modified and functionalized with graphene oxide (GO) and hyaluronic acid (HA), compared to the worn surfaces of the unmodified CoCr. The hematological analysis and tissue reactions after intra-articular injection did not reveal pathological damage, with average hematological values recorded, although slight reductions in creatinine and protein within non-pathological ranges were found. Some traces of biomaterial particles in the knee at the highest concentration of injected particles were only found but without inflammatory signs. The results show the potential benefits of using graphene in intra-articular prostheses, which could improve the quality of life for numerous patients. Full article

Air particle abrasion (APA) using bioactive glass (BG) effectively decontaminates titanium (Ti) surface biofilms and the retained glass particles on the abraded surfaces impart potent antibacterial properties against various clinically significant pathogens. The objective of this study was to investigate the effect of BG APA and simulated body fluid (SBF) immersion of sandblasted and acid-etched (SA) Ti surfaces on osteoblast cell viability. Another goal was to study the antibacterial effect against Streptococcus mutans. Square-shaped 10 mm diameter Ti substrates (n = 136) were SA by grit blasting with aluminum oxide particles, then acid-etching in an HCl-H₂SO₄ mixture. The SA substrates (n = 68) were used as non-coated controls (NC-SA). The test group (n = 68) was further subjected to APA using experimental zinc-containing BG (Zn4) and then mineralized in SBF for 14 d (Zn4-CaP). Surface roughness, contact angle, and surface free energy (SFE) were calculated on test and control surfaces. In addition, the topography and chemistry of substrate surfaces were also characterized. Osteoblastic cell viability and focal adhesion were also evaluated and compared to glass slides as an additional control. The antibacterial effect of Zn4-CaP was also assessed against S. mutans. After immersion in SBF, a mineralized zinc-containing Ca-P coating was formed on the SA substrates. The Zn4-CaP coating resulted in a significantly lower Ra surface roughness value (2.565 μm; p < 0.001), higher wettability (13.35°; p < 0.001), and higher total SFE (71.13; p < 0.001) compared to 3.695 μm, 77.19° and 40.43 for the NC-SA, respectively. APA using Zn4 can produce a zinc-containing calcium phosphate coating that demonstrates osteoblast cell viability and focal adhesion comparable to that on NC-SA or glass slides. Nevertheless, the coating had no antibacterial effect against S. mutans. Full article

Huperzine A (HupA) is an important drug for treating Alzheimer’s disease (AD) and is primarily extracted from the Huperzia serrata (Lycopodiaceae). Failures in the chemical synthesis of Hup and in vitro culture have put H. serrata in danger of extinction, and there is a need for an extensive investigation of Hup from alternative perspectives. The aim of this study is to identify endophytic fungi that produce high Hup or simultaneously produce many types of Hup and have high genetic stability derived from other Lycopodiaceae species as a source of materials for natural Hup production. In this work, Hup-producing endophytic fungi were isolated from three species: Lycopodium clavatum, Phlegmariurus squarrosus, and P. phlegmaria. Of these, L. clavatum and P. squarrosus were confirmed as novel sources of Hup-producing fungi. Based on morphological characteristics and nuclear ribosomal DNA ITS sequences, four endophytic fungi Colletotrichum siamense THG1-17, Epicoccum sorghinum THG01-18, Phoma sp. TKH3-2, and Phyllosticta sp. THG2-27 were firstly isolated from these Lycopodiaceae plants, which were capable of simultaneously producing both HupA and HupB, as evidenced by high-performance liquid chromatography analysis. The four strains showed stability in Hup yield over 50 generations of culture with an in vitro storage period of 3 months. These isolated fungi will provide a new source of materials for further research to develop drugs containing HupA as well as HupB for AD treatment in the future. Full article

This study aimed to evaluate the surface characteristics of a nanocrystalline hydroxyapatite coating made through atomic layer deposition (ALD-HA) on titanium surfaces and to investigate its effect on blood coagulation and platelet adhesion. Grade 2 square titanium discs (0.7 cm, 1 mm thick) were used (n = 108). Half of the substrates (n = 54) were coated with ALD-HA, and the other half were used as the non-coated control. Surface free energy (SFE), contact angle (CA), surface roughness (Ra), and chemical composition were evaluated. Blood thrombogenic properties were assessed on ALD-HA and non-coated surfaces using the kinetic clotting time method. The platelets’ adhesion and morphology were also evaluated. The ALD-HA-coated surfaces demonstrated significantly higher polar SFE (p < 0.001) and lower CA (p < 0.001) values compared to the non-coated surfaces. In addition, the surface roughness was significantly lower for the ALD-HA (p < 0.001) than for the non-coated surfaces. Platelets adhered to both surfaces; however, there was variability in platelet morphologies in different areas with higher platelet density on the ALD-HA surfaces. There was no significant difference in the overall absorbance values of the hemolyzed hemoglobin for both substrates, and the total clotting time was achieved at 60 min. It can be concluded that the ALD-HA coating of titanium can enhance surface wettability, increase surface free energy, and support blood coagulation and platelet adhesion. Full article

In the era of liquid biopsies, the reliable and cost-effective detection and screening of cancer biomarkers has become of fundamental importance, thus paving the way for the advancement of research in the field of point-of-care testing and the development of new methodologies and technologies. Indeed, the latter ones can help designing advanced diagnostic tools that can offer portability, ease of use with affordable production and operating costs. In this respect, impedance-based biosensing platforms might represent an attractive alternative. In this work, we describe a proof-of-concept study aimed at designing portable impedimetric biosensors for the monitoring of human urokinase-type plasminogen activator (h-uPA) cancer biomarker by employing small synthetic receptors. Aberrant levels of h-uPA were correlated with different types of cancers. Herein, we report the use of two bicyclic peptides (P₂ and P₃) which have been engineered to bind h-uPA with high affinity and exquisite specificity. The synthetic receptors were immobilized via biotin-streptavidin chemistry on the surface of commercial screen-printed electrodes. The impedimetric changes in the electrode/solution interface upon incubation of spiked h-uPA samples in the presence of a redox probe were followed via electrochemical impedance spectroscopy. The P₃-based impedimetric assay showed the best outcomes in terms of dynamic range and linearity (0.01–1 μg mL⁻¹) and sensitivity (LOD = 9 ng mL⁻¹). To fully assess the performances of P₃ over P₂, and to compare the label-free architecture vs. labelled architecture, a voltammetric assay was also developed. Full article