Differential distribution of PINK1 and Parkin in the primate brain implies distinct roles.

JOURNAL/nrgr/04.03/01300535-202504000-00028/figure1/v/2024-07-06T104127Z/r/image-tiff The vast majority of in vitro studies have demonstrated that PINK1 phosphorylates Parkin to work together in mitophagy to protect against neuronal degeneration. However, it remains largely unclear how PINK1 and Parkin are expressed in mammalian brains. This has been difficult to address because of the intrinsically low levels of PINK1 and undetectable levels of phosphorylated Parkin in small animals. Understanding this issue is critical for elucidating the in vivo roles of PINK1 and Parkin. Recently, we showed that the PINK1 kinase is selectively expressed as a truncated form (PINK1-55) in the primate brain. In the present study, we used multiple antibodies, including our recently developed monoclonal anti-PINK1, to validate the selective expression of PINK1 in the primate brain. We found that PINK1 was stably expressed in the monkey brain at postnatal and adulthood stages, which is consistent with the findings that depleting PINK1 can cause neuronal loss in developing and adult monkey brains. PINK1 was enriched in the membrane-bound fractionations, whereas Parkin was soluble with a distinguishable distribution. Immunofluorescent double staining experiments showed that PINK1 and Parkin did not colocalize under physiological conditions in cultured monkey astrocytes, though they did colocalize on mitochondria when the cells were exposed to mitochondrial stress. These findings suggest that PINK1 and Parkin may have distinct roles beyond their well-known function in mitophagy during mitochondrial damage.

Meningeal lymphatic vessel crosstalk with central nervous system immune cells in aging and neurodegenerative diseases.

Meningeal lymphatic vessels form a relationship between the nervous system and periphery, which is relevant in both health and disease. Meningeal lymphatic vessels not only play a key role in the drainage of brain metabolites but also contribute to antigen delivery and immune cell activation. The advent of novel genomic technologies has enabled rapid progress in the characterization of myeloid and lymphoid cells and their interactions with meningeal lymphatic vessels within the central nervous system. In this review, we provide an overview of the multifaceted roles of meningeal lymphatic vessels within the context of the central nervous system immune network, highlighting recent discoveries on the immunological niche provided by meningeal lymphatic vessels. Furthermore, we delve into the mechanisms of crosstalk between meningeal lymphatic vessels and immune cells in the central nervous system under both homeostatic conditions and neurodegenerative diseases, discussing how these interactions shape the pathological outcomes. Regulation of meningeal lymphatic vessel function and structure can influence lymphatic drainage, cerebrospinal fluid-borne immune modulators, and immune cell populations in aging and neurodegenerative disorders, thereby playing a key role in shaping meningeal and brain parenchyma immunity.

Fluorescence and photoacoustic (FL/PA) dual-modal probe: Responsive to reactive oxygen species (ROS) for atherosclerotic plaque imaging.

Accurate and early detection of atherosclerosis (AS) is imperative for their effective treatment. However, fluorescence probes for efficient diagnosis of AS often encounter insufficient deep tissue penetration, which hinders the reliable assessment of plaque vulnerability. In this work, a reactive oxygen species (ROS) activated near-infrared (NIR) fluorescence and photoacoustic (FL/PA) dual model probe TPA-QO-B is developed by conjugating two chromophores (TPA-QI and O-OH) and ROS-specific group phenylboronic acid ester. The incorporation of ROS-specific group not only induces blue shift in absorbance, but also inhibits the ICT process of TPA-QO-OH, resulting an ignorable initial FL/PA signal. ROS triggers the convertion of TPA-QO-B to TPA-QO-OH, resulting in the concurrent amplification of FL/PA signal. The exceptional selectivity of TPA-QO-B towards ROS makes it effectively distinguish AS mice from the healthy. The NIR emission can achieve a tissue penetration imaging depth of 0.3 cm. Moreover, its PA775 signal possesses the capability to penetrate tissues up to a thickness of 0.8 cm, ensuring deep in vivo imaging of AS model mice in early stage. The ROS-triggered FL/PA dual signal amplification strategy improves the accuracy and addresses the deep tissue penetration problem simultaneously, providing a promising tool for in vivo tracking biomarkers in life science and preclinical applications.

Impact of Irreversible Adsorption on Molecular Ordering and Charge Transport in Poly(3-hexylthiophene) Thin Films on Solid Substrates.

We investigate the irreversible adsorption of poly(3-hexylthiophene) (P3HT) polymer thin films on silicon dioxide/silicon (SiO2/Si) substrates during thermal annealing at a temperature below the melting temperature (Tm) but far above the glass transition temperature (Tg), i.e., Tg ≪ T = 170 °C < Tm, and its effect on their crystalline ordering and charge transport properties. It was found that short-time annealing enhances the molecular ordering of P3HT films, while prolonged thermal annealing gradually disrupts the crystalline structures and reduces the overall crystallinity of the film. Concurrently, thermal annealing at this temperature facilitates the slow irreversible adsorption of P3HT chains at the polymer-solid interface, resulting in the formation of a 1.7 Rg-thick (∼18 nm thick) adsorbed layer on SiO2/Si substrates that is fully amorphous and contains a large fraction of loosely adsorbed chains. We postulate that such irreversible adsorption is responsible for the reduced crystalline packing of P3HT at the polymer-solid interface at Tg ≪ T < Tm, which further disrupts the molecular ordering of the entire 46 nm thick P3HT film by a long-range perturbation effect. Electrical measurements using an organic field-effect transistor (OFET) device reveal that the enhanced charge carrier mobility of P3HT films correlates with an optimized annealing time at Tg ≪ T < Tm, which achieves a balance between maximizing molecular ordering and minimizing the impact of irreversible chain adsorption. These findings provide new insights into the underlying mechanism of thermal annealing in tailoring the structure and property of conjugated polymer thin films prepared on solid substrates.

Typical Tire Additives in River Water: Leaching, Transformation, and Environmental Risk Assessment.

Tire wear particles (TWPs) released during vehicle driving can enter water bodies, leading to leaching of tire additives (TAs) in aquatic environments. However, the transformation behavior and related ecological impacts of TAs and their transformation products (TPs) remain unclear. In this study, laboratory-based simulation experiments and field investigations were conducted to explore the transformation mechanisms and ecological risks of TAs. After being placed in river water for 24 h, about 7-95% of 12 investigated TAs in TWPs were leached. Forty-eight TPs from eight TAs were tentatively identified along with different transformation pathways via suspect screening by high-resolution mass spectrometry. Semiquantitative results indicated that TPs derived from N-(1,3-dimethylbutyl)-N'-phenyl-p-phenylene-diamine (6PPD) were predominant in leachates, while aryl hydrolysis and quinone pathways were the main transformation pathways. Field investigations on urban surface water samples from 16 sites in Hong Kong revealed the occurrence of 17 TAs and 1 TP, with concentrations ranging from 13.9 to 2230 ng/L (median ± standard deviation: 226 ± 534 ng/L). Sixteen TPs from six TAs were additionally identified via suspect screening. It is estimated that 6PPD-quinone and seven TAs could pose medium to high ecological risk, while N-(1,3-dimethylbutyl)-N'-phenyl-p-quinonediimine, a frequently detected TP, was identified as a persistent-bioaccumulative-toxic substance.

Formation of bovine serum albumin-galangin nanoparticles and their potential to inhibit ROS-induced inflammation: Enthanol desolvation vs. pH-shiftng method.

pH-shifting method, as an eco-friendly approach, is a promising alternative to desolvation method, yet systematic comparison of their property is still lacking. In this study, bovine serum albumin-galangin nanoparticles (BSA-GA NPs) were designed for alleviating ROS-mediated macrophage inflammation by the 2 separate methods. Compared with the desolvation method, BSA exhibited a higher loading capacity for GA under the pH-shifting method, which was attributed to the exposure of the binding site leading to enhanced affinity for GA and a more compact particle structure. Further analyses evidenced that the electron arrangement and crystal structure of GA changed with different methods. The content of random coil of BSA was elevated after pH-shifting method. Besides, the smaller size rendered the pH-shifting treated BSA-GA NPs easier to be taken up by macrophages, while the enhanced specific surface area conferred excellent ROS scavenging and anti-inflammatory performances. This study may provide new insights into the choice of loading methods.

Machine learning-assisted amidase-catalytic enantioselectivity prediction and rational design of variants for improving enantioselectivity.

Biocatalysis is an attractive approach for the synthesis of chiral pharmaceuticals and fine chemicals, but assessing and/or improving the enantioselectivity of biocatalyst towards target substrates is often time and resource intensive. Although machine learning has been used to reveal the underlying relationship between protein sequences and biocatalytic enantioselectivity, the establishment of substrate fitness space is usually disregarded by chemists and is still a challenge. Using 240 datasets collected in our previous works, we adopt chemistry and geometry descriptors and build random forest classification models for predicting the enantioselectivity of amidase towards new substrates. We further propose a heuristic strategy based on these models, by which the rational protein engineering can be efficiently performed to synthesize chiral compounds with higher ee values, and the optimized variant results in a 53-fold higher E-value comparing to the wild-type amidase. This data-driven methodology is expected to broaden the application of machine learning in biocatalysis research.

Modeling respiratory tract diseases for clinical translation employing conditionally reprogrammed cells.

Preclinical models serve as indispensable tools in translational medicine. Specifically, patient-derived models such as patient-derived xenografts (PDX), induced pluripotent stem cells (iPSC), organoids, and recently developed technique of conditional reprogramming (CR) have been employed to reflect the host characteristics of diseases. CR technology involves co-culturing epithelial cells with irradiated Swiss-3T3-J2 mouse fibroblasts (feeder cells) in the presence of a Rho kinase (ROCK) inhibitor, Y-27632. CR technique facilitates the rapid conversion of both normal and malignant cells into a "reprogrammed stem-like" state, marked by robust in vitro proliferation. This is achieved without reliance on exogenous gene expression or viral transfection, while maintaining the genetic profile of the parental cells. So far, CR technology has been used to study biology of diseases, targeted therapies (precision medicine), regenerative medicine, and noninvasive diagnosis and surveillance. Respiratory diseases, ranking as the third leading cause of global mortality, pose a significant burden to healthcare systems worldwide. Given the substantial mortality and morbidity rates of respiratory diseases, efficient and rapid preclinical models are imperative to accurately recapitulate the diverse spectrum of respiratory conditions. In this article, we discuss the applications and future potential of CR technology in modeling various respiratory tract diseases, including lung cancer, respiratory viral infections (such as influenza and Covid-19 and etc.), asthma, cystic fibrosis, respiratory papillomatosis, and upper aerodigestive track tumors. Furthermore, we discuss the potential utility of CR in personalized medicine, regenerative medicine, and clinical translation.

Cross-species comparison of airway epithelium transcriptomics.

Studies of lung transcriptomics across species are essential for understanding the complex biology and disease mechanisms of this vital organ. Single-cell RNA sequencing (scRNA-seq) has emerged as a key tool for understanding cell dynamics across various species. However, comprehensive cross-species comparisons are limited. Therefore, the aims of this study was to investigate the transcriptomic similarities and differences in lung cells across four species-humans, monkeys, mice, and rats-in healthy and asthma conditions using scRNA-seq. The results revealed significant transcriptomic similarities between monkeys and humans and significant cross-species conservation of cell-specific marker genes, transcription factors (TFs), and biological pathways. Additionally, we explored sex differences, identifying distinct sex-specific expression patterns that may influence disease susceptibility. These insights refine our understanding of the mechanism underlying airway cell biology across species and have important implications for studying lung diseases, particularly the mechanisms of mucus clearance in asthma.

MBD2 regulates the progression and chemoresistance of cholangiocarcinoma through interaction with WDR5.

BACKGROUND: Cholangiocarcinoma (CCA) is a highly malignant, rapidly progressing tumor of the bile duct. Owing to its chemoresistance, it always has an extremely poor prognosis. Therefore, detailed elucidation of the mechanisms of chemoresistance and identification of therapeutic targets are still needed. METHODS: We analyzed the expression of MBD2 (Methyl-CpG-binding domain 2) in CCA and normal bile duct tissues using the public database and immunohistochemistry (IHC). The roles of MBD2 in CCA cell proliferation, migration, and chemoresistance ability were validated through CCK-8, plate cloning assay, wound healing assays and xenograft mouse model. In addition, we constructed a primary CCA mouse model to further confirm the effect of MBD2. RNA-seq and co-IP-MS were used to identify the mechanisms by how MBD2 leads to chemoresistance. RESULTS: MBD2 was upregulated in CCA. It promoted the proliferation, migration and chemoresistance of CCA cells. Mechanistically, MBD2 directly interacted with WDR5, bound to the promoter of ABCB1, promoted the trimethylation of H3K4 in this region through KMT2A, and activated the expression of ABCB1. Knocking down WDR5 or KMT2A blocked the transcriptional activation of ABCB1 by MBD2. The molecular inhibitor MM-102 targeted the interaction of WDR5 with KMT2A. MM-102 inhibited the expression of ABCB1 in CCA cells and decreased the chemoresistance of CCA to cisplatin. CONCLUSION: MBD2 promotes the progression and chemoresistance of CCA through interactions with WDR5. MM-102 can effectively block this process and increase the sensitivity of CCA to cisplatin.

Effects of perceived variety-support on middle school students' learning engagement in physical education: the mediating role of motivation.

Introduction: High engagement in physical education (PE) could effectively develop students' motor competence and promote physical activity, which was significantly important for students' physical and mental health. Researches had shown that motivation was an important factor in explaining students' learning engagement, and variety-support as the fourth independent psychological need was a potential factor influencing students' learning motivation. However, there was a lack of empirical research evidence on the effect of perceived variety-support on middle school students' learning engagement in PE and the influencing mechanisms. This study aimed to investigate the direct effect of perceived variety-support on learning engagement in PE and the mediating effect of motivation in PE on the relationship. Methods: A cross-sectional study was conducted and 587 middle school students from Liaoning province filled the paper-and-pencil questionnaires adopting perceived variety-support in PE scale (PVSPES), utrecht work engagement scale-student (UWES-S), and perceived locus of causality in PE scale, which had been proved to have good reliability and validity (294 boys and 293 girls, Mage=13.47 ± 0.94). Results: The results showed three variables were significantly positively correlated with each other (r = 0.323-0.562 p < 0.01) and perceived variety-support in PE could not only directly promote middle school students' learning engagement in PE but also indirectly through the mediating effect of motivation in PE. Discussion: Therefore, in order to better promote students' participation in PE class, we should pay more attention to satisfy students' varied PE learning needs and stimulate students' autonomous learning motivation.

Targeted regulation of 5-aminolevulinic acid enhances flavonoids, anthocyanins and proanthocyanidins accumulation in Vitis davidii callus.

BACKGROUND: Spine grape (Vitis davidii) is a promising source of high-quality anthocyanins, with vast potential for application in food, pharmaceutical, and cosmetic industries. However, their availability is limited by resource constraints. Plant cell culture has emerged as a valuable approach for anthocyanin production and serves as an ideal model to investigate the regulation of anthocyanin biosynthesis. Elicitors are employed to achieve targeted enhancement of anthocyanin biosynthesis. The present study investigated the impact of 5-aminolevulinic acid (ALA) as an elicitor on the accumulation of anthocyanins and flavonoids during spine grape callus growth. Specifically, we examined the effects of ALA on anthocyanin and its component accumulation in callus, and biosynthetic anthocyanin gene expression. RESULTS: ALA at 25 µg/L increased the biomass of spine grape callus. ALA induction enhanced the levels of flavonoids, anthocyanins and proanthocyanidins in callus, with maximum values reaching 911.11 mg/100 g DW, 604.60 mg/100 g DW, and 5357.00 mg/100 g DW, respectively, after callus culture for 45 days. Notably, those levels were 1.47-, 1.93- and 1.83-fold higher than controls. ALA induction modulated the flavonoid profile, and among 97 differential flavonoid metabolites differing from controls, 77 were upregulated and 20 were downregulated. Six kinds of anthocyanins, namely cyanidin (8), delphinidin (6), peonidin (5), malvidin (4), petunidin (3) and pelargonidin (3), were detected in callus, with peonidin most abundant. Compared with controls, anthocyanin components were increased in ALA-treated callus. The key genes PAL1, PAL2, PAL4, CHI, CHS3, F3'H, F3H, FLS, DFR, UFGT, MYBA1, LDOX, OMT3, GT1 and ACT involved in anthocyanin biosynthesis were upregulated following ALA treatment, resulting in anthocyanin accumulation. CONCLUSION: This study revealed a novel mode of ALA-mediated promotion of plant anthocyanin biosynthesis and accumulation at the cellular level, and a strategy for enhancing anthocyanin content in spine grape callus. The findings advance commercial-scale production of anthocyanins via spine grape callus culture. we also explored the accumulation patterns of flavonoids and anthocyanins under ALA treatment. Augmentation of anthocyanins coincided with elevated expression levels of most genes involved in anthocyanin biosynthesis within spine grape callus following ALA treatment.

US/PA/MR multimodal imaging-guided multifunctional genetically engineered bio-targeted synergistic agent for tumor therapy.

Focused ultrasound ablation surgery (FUAS) is a minimally invasive treatment option that has been utilized in various tumors. However, its clinical advancement has been hindered by issues such as low safety and efficiency, single image guidance mode, and postoperative tumor residue. To address these limitations, this study aimed to develop a novel multi-functional gas-producing engineering bacteria biological targeting cooperative system. Pulse-focused ultrasound (PFUS) could adjust the ratio of thermal effect to non-thermal effect by adjusting the duty cycle, and improve the safety and effectiveness of treatment.The genetic modification of Escherichia coli (E.coli) involved the insertion of an acoustic reporter gene to encode gas vesicles (GVs), resulting in gas-producing E.coli (GVs-E.coli) capable of targeting tumor anoxia. GVs-E.coli colonized and proliferated within the tumor while the GVs facilitated ultrasound imaging and cooperative PFUS. Additionally, multifunctional cationic polyethyleneimine (PEI)-poly (lactic-co-glycolic acid) (PLGA) nanoparticles (PEI-PLGA/EPI/PFH@Fe3O4) containing superparamagnetic iron oxide (SPIO, Fe3O4), perfluorohexane (PFH), and epirubicin (EPI) were developed. These nanoparticles offered synergistic PFUS, supplementary chemotherapy, and multimodal imaging capabilities.GVs-E.coli effectively directed the PEI-PLGA/EPI/PFH@Fe3O4 to accumulate within the tumor target area by means of electrostatic adsorption, resulting in a synergistic therapeutic impact on tumor eradication.In conclusion, GVs-E.coli-mediated multi-functional nanoparticles can synergize with PFUS and chemotherapy to effectively treat tumors, overcoming the limitations of current FUAS therapy and improving safety and efficacy. This approach presents a promising new strategy for tumor therapy.

A case report: Comorbidity of Rhinocerebral mucormycosis and pulmonary aspergillosis with challenging diagnosis.

Background: Mucormycosis is a rare opportunistic invasive fungal disease. Rhinocerebral mucormycosis (RCM) is clinically difficult to diagnose, and patients often die due to delayed diagnosis. Case description: A patient with concurrent pulmonary aspergillosis was diagnosed with RCM caused by Rhizopus through metagenomic Next-Generation Sequencing (mNGS). Despite comprehensive treatment including surgery, amphotericin B, and posaconazole, the patient tragically passed away. The treatment was delayed due to repeated cultures of secretions were negative and pathological examination could not clarify which fungus is infected. Conclusion: The clinical manifestations of RCM are not specific in the early stage, but the infection progresses rapidly. Therefore, early and accurate diagnosis is very important. mNGS is helpful for patients suspected of RCM, especially when conventional microbiology tests were negative.

Machine learning in female urinary incontinence: A scoping review.

Introduction and Hypothesis: The aim was to conduct a scoping review of the literature on the use of machine learning (ML) in female urinary incontinence (UI) over the last decade. Methods: A systematic search was performed among the Medline, Google Scholar, PubMed, and Web of Science databases using the following keywords: [Urinary incontinence] and [(Machine learning) or (Predict) or (Prediction model)]. Eligible studies were considered to have applied ML model to explore different management processes of female UI. Data analyzed included the field of application, type of ML, input variables, and results of model validation. Results: A total of 798 papers were identified while 23 finally met the inclusion criteria. The vast majority of studies applied logistic regression to establish models (91.3%, 21/23). Most frequently ML was applied to predict postpartum UI (39.1%, 9/23), followed by de novo incontinence after pelvic floor surgery (34.8%, 8/23).There are also three papers using ML models to predict treatment outcomes and three papers using ML models to assist in diagnosis. Variables for modeling included demographic characteristics, clinical data, pelvic floor ultrasound, and urodynamic parameters. The area under receiver operating characteristic curve of these models fluctuated from 0.56 to 0.95, and only 11 studies reported sensitivity and specificity, with sensitivity ranging from 20% to 96.2% and specificity from 59.8% to 94.5%. Conclusion: Machine learning modeling demonstrated good predictive and diagnostic abilities in some aspects of female UI, showing its promising prospects in near future. However, the lack of standardization and transparency in the validation and evaluation of the models, and the insufficient external validation greatly diminished the applicability and reproducibility, thus a focus on filling this gap is strongly recommended for future research.

Autophagy activator-loaded bicomponent peptide nanocarriers for phototherapy-triggered immunity enhancement against metastatic breast cancer.

Mild autophagy accompanied with immunogenic cell death (ICD) effect destructs immune-associated antigens, weakening the immune response against tumor growth. To address this dilemma, we develop a peptide-based bicomponent nanocarrier with encapsulation of a cellular hyperautophagy activator (STF-62247) for near-infrared (NIR) photo/immunotherapy to eliminate primary and metastatic breast tumors. The electrostatic-driven nanodrug (PPNPs@STF) with active-targeting and efficient endosomal escape can induce specific ICD effect upon NIR laser irradiation, and trigger autophagy to a mild activation state. Notably, the simultaneously released STF-62247 precisely promotes autophagy to an overactivated state, resulting in autophagic death of tumor cells and further boosting ICD-related antigen presentation. More importantly, the combined photo/immunotherapy of PPNPs@STF not only inhibits tumor cell proliferation, but also promotes dendritic cells (DCs)-associated immune response. In 4 T1 tumor-bearing mice, PPNPs@STF effectively inhibits growth of primary and distant tumors, and suppresses lung metastasis with a minimized side effect. This study provides a hyperautophagy activator-assisted strategy that can enhance ICD-based antitumor immune response for the treatment of metastatic breast cancer.

Nanozyme linked multi-array gas driven sensor for real-time quantitative detection of Group A streptococcus.

Group A streptococcus (GAS) is a pathogen typically transmitted through respiratory droplets and skin contact, causing an estimated 700 million mild non-invasive infections worldwide each year. There are approximately 650 000 infections that progress to severe invasive infections, even resulting in death. Therefore, the ability to detect GAS rapidly, accurately and in real time is important. Herein, we developed a nanozyme linked multi-array gas driven sensor (NLMAGS) to point-of-care testing of GAS within 2 h. The NLMAGS demonstrated excellent performance as it combined the advantages of nanozyme techniques, immunoassay techniques, and 3D printing techniques. Platinum- and palladium-rich nanozyme particles (Au@Pt@PdNPs) were synthesized and used to label monocloning antibodies as detection probes. Magnetic beads were labeled with monocloning antibodies as capture probes to establish a double-antibody sandwich immunoassay for the detection of GAS. The sandwich immune complex can catalyze the H2O2 substrate and produce O2. GAS quantification can be achieved by measuring the distance that the O2 pushes the ink drops forward in the sensor. Under optimized conditions, the NLMAGS quantitatively detected 24 spiked samples with a limit of detection (LOD) of 62 CFU mL-1, which was 5 times lower than that of ELISA (334 CFU mL-1). A strong correlation with the conventional ELISA was found (r = 0.99, P < 0.001). In comparison, the traditional lateral flow immunoassay based on Au@Pt@PdNPs-mAb2 (Au@Pt@PdNPs-LFIA) had a LOD of 104 CFU mL-1, which was significantly higher than that of NLMAGS. The NLMAGS demonstrated excellent sensitivity to GAS. The intra- and inter-assay precisions of the sensor were below 15%. Overall, the established NLMAGS has promising potential as a rapid and quantitative method for detecting GAS and can also be used to detect various pathogens.

Transition from sulfur autotrophic to mixotrophic denitrification: Performance with different carbon sources, microbial community and artificial neural network modeling.

To address the limitations inherent in both sulfur autotrophic denitrification (SAD) and heterotrophic denitrification (HD) processes, this study introduces a novel approach. Three carbon sources (glucose, methanol, and sodium acetate) were fed into the SAD system to facilitate the transition towards mixotrophic denitrification. Batch experiments were conducted to explore the effects of influencing factors (pH, HRT) on the denitrification performance of the mixotrophic system. Carbon source dosages were varied at 12.5%, 25%, and 50% of the theoretical amounts required for HD (18, 36, and 72 mg/L, respectively). The results showed distinct optimal dosages for each of the three organic carbon sources. The mixotrophic system, initiated with sodium acetate at 25% of the theoretical value, demonstrated the highest denitrification performance, achieving NO3--N removal efficiency of 99.8% and the NRR of 6.25 mg/(L·h). In contrast, the corresponding systems utilizing glucose (at 25% of the theoretical value) and methanol (at 50% of the theoretical value) achieved lower removal efficiency of 77.0% and 88.4%, respectively. The corresponding NRRs were 4.85 mg/(L·h) and 5.65 mg/(L·h). Following the transition from SAD to a mixotrophic system, the abundance of Thiobacillus decreased from 78.5% to 34.4% at the genus level, and the mixotrophic system cultivated a variety of other denitrifying bacteria (Thauera, Aquimonas, Azoarcus, and Pseudomonas), indicating an enhanced microbial community structure diversity. The established artificial neural network (ANN) model accurately predicted the effluent quality of mixotrophic denitrification, which predicted values closely aligning with experimental results (R2 > 0.9). Furthermore, initial pH exerted greater relative importance for COD removal and sulfur conversion, while the relative importance of HRT was more pronounced for NO3--N removal.