A potential strategy for bladder cancer treatment: inhibiting autophagy to enhance antitumor effects of Nectin-4-MMAE.

Research and development on Nectin-4 antibody-drug conjugates (ADC) have been greatly accelerated since the approval of enfortumab vedotin to treat uroepithelial cancer. During the course of this study, we identified that autophagy serves as a cytoprotective mechanism during Nectin-4-MMAE treatment and proposed a strategy to enhance the antitumor effects of Nectin-4-MMAE in bladder cancer. Nectin-4-MMAE rapidly internalized into bladder cancer cells in 30 minutes and released MMAE, inducing the onset of caspase-mediated apoptosis and leading to the inhibition of tumor cell growth. Transcriptomics showed significant alterations in autophagy-associated genes in bladder cancer cells treated with Nectin-4-MMAE, which suggested autophagy was activated by Nectin-4-MMAE. Furthermore, autophagy activation was characterized by ultrastructural analysis of autophagosome accumulation, immunofluorescence of autophagic flux, and immunoblotting autophagy marker proteins SQSTM1 and LC3 I/II. Importantly, inhibiting autophagy by LY294002 and chloroquine significantly enhances the cytotoxicity effects of Nectin-4-MMAE in bladder cancer cells. Additionally, we detected the participation of the AKT/mTOR signaling cascade in the induction of autophagy by Nectin-4-MMAE. The combination of Nectin-4-MMAE and an autophagy inhibitor demonstrated enhanced antitumor effects in the HT1376 xenograft tumor model. After receiving a single dose of Nectin-4-MMAE, the group that received the combination treatment showed a significant decrease in tumor size compared to the group that received only one type of treatment. Notably, one mouse in the combination treatment group achieved complete remission of the tumor. The combination group exhibited a notable rise in apoptosis and necrosis, as indicated by H&E staining and immunohistochemistry (cleaved caspase-3, ki67). These findings demonstrated the cytoprotective role of autophagy during Nectin-4-MMAE treatment and highlighted the potential of combining Nectin-4-MMAE with autophagy inhibitors for bladder cancer treatment.

Co-targeting CD47 and VEGF elicited potent anti-tumor effects in gastric cancer.

BACKGROUND: CD47, serving as an intrinsic immune checkpoint, has demonstrated efficacy as an anti-tumor target in hematologic malignancies. Nevertheless, the clinical relevance of CD47 in gastric cancer and its potential as a therapeutic target remains unclear. METHODS: The expression of CD47 in clinical gastric cancer tissues was assessed using immunohistochemistry and Western blot. Patient-derived cells were obtained from gastric cancer tissues and co-cultured with macrophages derived from human peripheral blood mononuclear cells. Flow cytometry analyses were employed to evaluate the rate of phagocytosis. Humanized patient-derived xenografts (Hu-PDXs) models were established to assess the efficacy of anti-CD47 immunotherapy or the combination of anti-CD47 and anti-VEGF therapy in treating gastric cancer. The infiltrated immune cells in the xenograft were analyzed by immunohistochemistry. RESULTS: In this study, we have substantiated the high expression of CD47 in gastric cancer tissues, establishing a strong association with unfavorable prognosis. Through the utilization of SIRPα-Fc to target CD47, we have effectively enhanced macrophage phagocytosis of PDCs in vitro and impeded the growth of Hu-PDXs. It is noteworthy that anti-CD47 immunotherapy has been observed to sustain tumor angiogenic vasculature, with a positive correlation between the expression of VEGF and CD47 in gastric cancer. Furthermore, the successful implementation of anti-angiogenic treatment has further augmented the anti-tumor efficacy of anti-CD47 therapy. In addition, the potent suppression of tumor growth, prevention of cancer recurrence after surgery, and significant prolongation of overall survival in Hu-PDX models can be achieved through the simultaneous targeting of CD47 and VEGF using the bispecific fusion protein SIRPα-VEGFR1 or by combining the two single-targeted agents. CONCLUSIONS: Our preclinical studies collectively offer substantiation that CD47 holds promise as a prospective target for gastric cancer, while also highlighting the potential of anti-angiogenic therapy to enhance tumor responsiveness to anti-CD47 immunotherapy.

Emergence of machine language: towards symbolic intelligence with neural networks.

Inspired by human language, machine language is a novel discrete representation learned from visual data only through playing the speak, guess, and draw game.

Revisiting Confidence Estimation: Towards Reliable Failure Prediction.

Reliable confidence estimation is a challenging yet fundamental requirement in many risk-sensitive applications. However, modern deep neural networks are often overconfident for their incorrect predictions, i.e., misclassified samples from known classes, and out-of-distribution (OOD) samples from unknown classes. In recent years, many confidence calibration and OOD detection methods have been developed. In this paper, we find a general, widely existing but actually-neglected phenomenon that most confidence estimation methods are harmful for detecting misclassification errors. We investigate this problem and reveal that popular calibration and OOD detection methods often lead to worse confidence separation between correctly classified and misclassified examples, making it difficult to decide whether to trust a prediction or not. Finally, we propose to enlarge the confidence gap by finding flat minima, which yields state-of-the-art failure prediction performance under various settings including balanced, long-tailed, and covariate-shift classification scenarios. Our study not only provides a strong baseline for reliable confidence estimation but also acts as a bridge between understanding calibration, OOD detection, and failure prediction.

Novel nitroxoline derivative combating resistant bacterial infections through outer membrane disruption and competitive NDM-1 inhibition.

ABSTRACTNew Delhi metallo-ß-lactamase-1 (NDM-1) has rapidly disseminated worldwide, leading to multidrug resistance and worse clinical prognosis. Designing and developing effective NDM-1 inhibitors is a critical and urgent challenge. In this study, we constructed a library of long-lasting nitroxoline derivatives and identified ASN-1733 as a promising dual-functional antibiotic. ASN-1733 can effectively compete for Ca2+ on the bacterial surface, causing the detachment of lipopolysaccharides (LPS), thereby compromising the outer membrane integrity and permeability and exhibiting broad-spectrum bactericidal activity. Moreover, ASN-1733 demonstrated wider therapeutic applications than nitroxoline in mouse sepsis, thigh and mild abdominal infections. Furthermore, ASN-1733 can effectively inhibit the hydrolytic capability of NDM-1 and exhibits synergistic killing effects in combination with meropenem against NDM-1 positive bacteria. Mechanistic studies using enzymatic experiments and computer simulations revealed that ASN-1733 can bind to key residues on Loop10 of NDM-1, hindering substrate entry into the enzyme's active site and achieving potent inhibitory activity (Ki = 0.22 µM), even in the presence of excessive Zn2+. These findings elucidate the antibacterial mechanism of nitroxoline and its derivatives, expand their potential application in the field of antibacterial agents and provide new insights into the development of novel NDM-1 inhibitors.

CMAUP database update 2024: extended functional and association information of useful plants for biomedical research.

Knowledge of the collective activities of individual plants together with the derived clinical effects and targeted disease associations is useful for plant-based biomedical research. To provide the information in complement to the established databases, we introduced a major update of CMAUP database, previously featured in NAR. This update includes (i) human transcriptomic changes overlapping with 1152 targets of 5765 individual plants, covering 74 diseases from 20 027 patient samples; (ii) clinical information for 185 individual plants in 691 clinical trials; (iii) drug development information for 4694 drug-producing plants with metabolites developed into approved or clinical trial drugs; (iv) plant and human disease associations (428 737 associations by target, 220 935 reversion of transcriptomic changes, 764 and 154121 associations by clinical trials of individual plants and plant ingredients); (v) the location of individual plants in the phylogenetic tree for navigating taxonomic neighbors, (vi) DNA barcodes of 3949 plants, (vii) predicted human oral bioavailability of plant ingredients by the established SwissADME and HobPre algorithm, (viii) 21-107% increase of CMAUP data over the previous version to cover 60 222 chemical ingredients, 7865 plants, 758 targets, 1399 diseases, 238 KEGG human pathways, 3013 gene ontologies and 1203 disease ontologies. CMAUP update version is freely accessible at https://bidd.group/CMAUP/index.html.

Freezing of water and melting of ice: theoretical modeling at the nanoscale.

Freezing of water and melting of ice at the nanoscale play critical roles in science and technology fields, including aviation systems, infrastructures, and other broad spectrum of technologies. To cope with the icing challenge, nanoscale anti-icing surface technology has been developed. The freezing and melting temperatures can be tailored by manipulating the size (the radius of water or ice); however, it lacks systemic research. In this work, the size effect on the melting temperature of ice nanocrystals was first established, which considered the variation of bond energy and equivalent heat energy from the perspective of the force-heat equivalence energy density principle. Based on the heterogeneous nucleation mode and by further considering the size and temperature effects on the interface energy involved solid-liquid energy and liquid-vapor energy as well as the above developed melting temperature model, another model is established to accurately predict the freezing temperature of water nanodroplets. The parameters required by the two models established in this paper have a clear physical meaning and establish the quantitative relationships among freezing temperature, melting temperature, surface stress, interface energy, and other thermodynamic parameters. The agreement between model prediction and experimental simulation data confirms the validity and universality of the established models. The higher prediction accuracy of this work compared to the other theoretical models, due to the more detailed consideration and the reference point, captures the errors introduced by the experiment or simulation. This study contributes to a deeper understanding of the underlying mechanism of freezing of water and melting of ice nanocrystals and provides theoretical guidance for the design of cryopreservation systems and anti-icing systems for aviation.

TROP2-directed nanobody-drug conjugate elicited potent antitumor effect in pancreatic cancer.

BACKGROUND: Pancreatic cancer is a highly aggressive malignancy with limited treatment options and a poor prognosis. Trophoblast cell surface antigen 2 (TROP2), a cell surface antigen overexpressed in the tumors of more than half of pancreatic cancer patients, has been identified as a potential target for antibody-drug conjugates (ADCs). Almost all reported TROP2-targeted ADCs are of the IgG type and have been poorly studied in pancreatic cancer. Here, we aimed to develop a novel nanobody-drug conjugate (NDC) targeting TROP2 for the treatment of pancreatic cancer. RESULTS: In this study, we developed a novel TROP2-targeted NDC, HuNbTROP2-HSA-MMAE, for the treatment of TROP2-positive pancreatic cancer. HuNbTROP2-HSA-MMAE is characterized by the use of nanobodies against TROP2 and human serum albumin (HSA) and has a drug-antibody ratio of 1. HuNbTROP2-HSA-MMAE exhibited specific binding to TROP2 and was internalized into tumor cells with high endocytosis efficiency within 5 h, followed by intracellular translocation to lysosomes and release of MMAE to induce cell apoptosis in TROP2-positive pancreatic cancer cells through the caspase-3/9 pathway. In a xenograft model of pancreatic cancer, doses of 0.2 mg/kg and 1 mg/kg HuNbTROP2-HSA-MMAE demonstrated significant antitumor effects, and a dose of 5 mg/kg even eradicated the tumor. CONCLUSION: HuNbTROP2-HSA-MMAE has desirable affinity, internalization efficiency and antitumor activity. It holds significant promise as a potential therapeutic option for the treatment of TROP2-positive pancreatic cancer.

Genome-Wide Identification and Characterization of the Phytochrome Gene Family in Peanut.

To investigate the potential role of phytochrome (PHY) in peanut growth and its response to environmental fluctuations, eight candidate AhPHY genes were identified via genome-wide analysis of cultivated peanut. These AhPHY polypeptides were determined to possess acidic and hydrophilic physiochemical properties and exhibit subcellular localization patterns consistent with residence in the nucleus and cytoplasm. Phylogenetic analysis revealed that the AhPHY gene family members were classified into three subgroups homologous to the PHYA/B/E progenitors of Arabidopsis. AhPHY genes within the same clade largely displayed analogous gene structure, conserved motifs, and phosphorylation sites. AhPHY exhibited symmetrical distribution across peanut chromosomes, with 7 intraspecific syntenic gene pairs in peanut, as well as 4 and 20 interspecific PHY syntenic gene pairs in Arabidopsis and soybean, respectively. A total of 42 cis-elements were predicted in AhPHY promoters, including elements implicated in phytohormone regulation, stress induction, physiology, and photoresponse, suggesting putative fundamental roles across diverse biological processes. Moreover, spatiotemporal transcript profiling of AhPHY genes in various peanut tissues revealed distinct expression patterns for each member, alluding to putative functional specialization. This study contributes novel insights into the classification, structure, molecular evolution, and expression profiles of the peanut phytochrome gene family, and also provides phototransduction gene resources for further mechanistic characterization.

Innate Immunity in Cancer Biology and Therapy.

Immunotherapies including adaptive immune checkpoint inhibitors (ICIs) and chimeric antigen receptor (CAR) T cells, have developed the treatment of cancer in clinic, and most of them focus on activating T cell immunity. Although these strategies have obtained unprecedented clinical responses, only limited subsets of cancer patients could receive long-term benefits, highlighting the demand for identifying novel targets for the new era of tumor immunotherapy. Innate immunity has been demonstrated to play a determinative role in the tumor microenvironment (TME) and influence the clinical outcomes of tumor patients. A thorough comprehension of the innate immune cells that infiltrate tumors would allow for the development of new therapeutics. In this review, we outline the role and mechanism of innate immunity in TME. Moreover, we discuss innate immunity-based cancer immunotherapy in basic and clinical studies. Finally, we summarize the challenges in sufficiently motivating innate immune responses and the corresponding strategies and measures to improve anti-tumor efficacy. This review could aid the comprehension of innate immunity and inspire the creation of brand-new immunotherapies for the treatment of cancer.

Computational Study of Stiffness-Tuning Strategies in Anguilliform Fish.

Biological evidence demonstrates that fish can tune their body stiffness to improve thrust and efficiency during swimming locomotion. However, the stiffness-tuning strategies that maximize swimming speed or efficiency are still unclear. In the present study, a musculo-skeletal model of anguilliform fish is developed to study the properties of variable stiffness, in which the planar serial-parallel mechanism is used to model the body structure. The calcium ion model is adopted to simulate muscular activities and generate muscle force. Further, the relations among the forward speed, the swimming efficiency, and Young's modulus of the fish body are investigated. The results show that for certain body stiffness, the swimming speed and efficiency are increased with the tail-beat frequency until reaching the maximum value and then decreased. The peak speed and efficiency are also increased with the amplitude of muscle actuation. Anguilliform fish tend to vary their body stiffness to improve the swimming speed and efficiency at a high tail-beat frequency or small amplitude of muscle actuation. Furthermore, the midline motions of anguilliform fish are analyzed by the complex orthogonal decomposition (COD) method, and the discussions of fish motions associated with the variable body stiffness and the tail-beat frequency are also presented. Overall, the optimal swimming performance of anguilliform fish benefits from the matching relationships among the muscle actuation, the body stiffness, and the tail-beat frequency.

Imitating the oracle: Towards calibrated model for class incremental learning.

Class-incremental learning (CIL) aims to recognize classes that emerged in different phases. The joint-training (JT), which trains the model jointly with all classes, is often considered as the upper bound of CIL. In this paper, we thoroughly analyze the difference between CIL and JT in feature space and weight space. Motivated by the comparative analysis, we propose two types of calibration: feature calibration and weight calibration to imitate the oracle (ItO), i.e., JT. Specifically, on the one hand, feature calibration introduces deviation compensation to maintain the class decision boundary of old classes in feature space. On the other hand, weight calibration leverages forgetting-aware weight perturbation to increase transferability and reduce forgetting in parameter space. With those two calibration strategies, the model is forced to imitate the properties of joint-training at each incremental learning stage, thus yielding better CIL performance. Our ItO is a plug-and-play method and can be implemented into existing methods easily. Extensive experiments on several benchmark datasets demonstrate that ItO can significantly and consistently improve the performance of existing state-of-the-art methods. Our code is publicly available at https://github.com/Impression2805/ItO4CIL.

Characterization and functional analysis of AhGPAT9 gene involved in lipid synthesis in peanut (Arachis hypogaea L.).

GPAT enzymes (glycerol-3-phosphate 1-O-acyltransferase, EC 2.3.1.15) catalyze the initial and rate-limiting step of plant glycerolipid biosynthesis for membrane homeostasis and lipid accumulation, yet little research has been done on peanuts. By reverse genetics and bioinformatics analyses, we have characterized an AhGPAT9 isozyme, of which the homologous product is isolated from cultivated peanut. QRT-PCR assay revealed a spatio-temporal expression pattern that the transcripts of AhGPAT9 accumulating in various peanut tissues are highly expressed during seed development, followed by leaves. Green fluorescent protein tagging of AhGPAT9 confirmed its subcellular accumulation in the endoplasmic reticulum. Compared with the wild type control, overexpressed AhGPAT9 delayed the bolting stage of transgenic Arabidopsis, reduced the number of siliques, and increased the seed weight as well as seed area, suggesting the possibility of participating in plant growth and development. Meanwhile, the mean seed oil content from five overexpression lines increased by about 18.73%. The two lines with the largest increases in seed oil content showed a decrease in palmitic acid (C16:0) and eicosenic acid (C20:1) by 17.35% and 8.33%, respectively, and an increase in linolenic acid (C18:3) and eicosatrienoic acid (C20:3) by 14.91% and 15.94%, respectively. In addition, overexpressed AhGPAT9 had no significant effect on leaf lipid content of transgenic plants. Taken together, these results suggest that AhGPAT9 is critical for the biosynthesis of storage lipids, which contributes to the goal of modifying peanut seeds for improved oil content and fatty acid composition.

Targeting CD47 enhanced the antitumor immunity of PD-L1 blockade in B-cell lymphoma.

Background: Only a subset of B-cell lymphoma (BCL) patients can benefit from immune checkpoint inhibitors targeting PD-1/PD-L1. Materials & methods: In the A20 model, SIRPα-Fc and anti-PD-L1 were employed to target CD47 and PD-L1 simultaneously. Flow cytometry, immunofluorescence and quantitative polymerase chain reaction were used to unravel the potential mechanisms. Results: Simultaneously targeting CD47 and PD-L1 activated CD8+ T cells with an increased release of effector molecules. Furthermore, infiltration of F4/80+iNOS+ M1 macrophages was enhanced by the dual therapy. Conclusion: Anti-CD47 therapy could sensitize BCL tumors to anti-PD-L1 therapy in a CD8+ T-cell- and M1-macrophage-dependent manner by promoting cytotoxic lymphocyte infiltration, which may provide a potential strategy for BCL treatment by simultaneously targeting CD47 and PD-L1.

Immune checkpoint inhibitors targeting PD-1/PD-L1 have become effective agents for cancer treatment. However, only a minority of patients benefit from this treatment in the clinic because of the limited response rate. Targeting CD47/SIRPα restores macrophage function and improves the response of antitumor immunity. Here, combination immunotherapy targeting CD47/SIRPα and PD-1/PD-L1 was investigated to increase the response rate and antitumor effect of PD-L1 monotherapy in B-cell lymphoma (BCL). This study broadens the application of the combination therapy and provided a promising strategy for B-cell lymphoma treatment by simultaneous targeting of PD-1/PD-L1 and CD47/SIRPα axis.

NPASS database update 2023: quantitative natural product activity and species source database for biomedical research.

Quantitative activity and species source data of natural products (NPs) are important for drug discovery, medicinal plant research, and microbial investigations. Activity values of NPs against specific targets are useful for discovering targeted therapeutic agents and investigating the mechanism of medicinal plants. Composition/concentration values of NPs in individual species facilitate the assessments and investigations of the therapeutic quality of herbs and phenotypes of microbes. Here, we describe an update of the NPASS natural product activity and species source database previously featured in NAR. This update includes: (i) new data of ∼95 000 records of the composition/concentration values of ∼1 490 NPs/NP clusters in ∼390 species, (ii) extended data of activity values of ∼43 200 NPs against ∼7 700 targets (∼40% and ∼32% increase, respectively), (iii) extended data of ∼31 600 species sources of ∼94 400 NPs (∼26% and ∼32% increase, respectively), (iv) new species types of ∼440 co-cultured microbes and ∼420 engineered microbes, (v) new data of ∼66 600 NPs without experimental activity values but with estimated activity profiles from the established chemical similarity tool Chemical Checker, (vi) new data of the computed drug-likeness properties and the absorption, distribution, metabolism, excretion and toxicity (ADMET) properties for all NPs. NPASS update version is freely accessible at http://bidd.group/NPASS.

Learning by Seeing More Classes.

Traditional pattern recognition models usually assume a fixed and identical number of classes during both training and inference stages. In this paper, we study an interesting but ignored question: can increasing the number of classes during training improve the generalization and reliability performance? For a k-class problem, instead of training with only these k classes, we propose to learn with k+m classes, where the additional m classes can be either real classes from other datasets or synthesized from known classes. Specifically, we propose two strategies for constructing new classes from known classes. By making the model see more classes during training, we can obtain several advantages. First, the added m classes serve as a regularization which is helpful to improve the generalization accuracy on the original k classes. Second, this will alleviate the overconfident phenomenon and produce more reliable confidence estimation for different tasks like misclassification detection, confidence calibration, and out-of-distribution detection. Lastly, the additional classes can also improve the learned feature representation, which is beneficial for new classes generalization in few-shot learning and class-incremental learning. Compared with the widely proved concept of data augmentation (dataAug), our method is driven from another dimension of augmentation based on additional classes (classAug). Comprehensive experiments demonstrated the superiority of our classAug under various open-environment metrics on benchmark datasets.

Leveraging Balanced Semantic Embedding for Generative Zero-Shot Learning.

Generative (generalized) zero-shot learning [(G)ZSL] models aim to synthesize unseen class features by using only seen class feature and attribute pairs as training data. However, the generated fake unseen features tend to be dominated by the seen class features and thus classified as seen classes, which can lead to inferior performances under zero-shot learning (ZSL), and unbalanced results under generalized ZSL (GZSL). To address this challenge, we tailor a novel balanced semantic embedding generative network (BSeGN), which incorporates balanced semantic embedding learning into generative learning scenarios in the pursuit of unbiased GZSL. Specifically, we first design a feature-to-semantic embedding module (FEM) to distinguish real seen and fake unseen features collaboratively with the generator in an online manner. We introduce the bidirectional contrastive and balance losses for the FEM learning, which can guarantee a balanced prediction for the interdomain features. In turn, the updated FEM can boost the learning of the generator. Next, we propose a multilevel feature integration module (mFIM) from the cycle-consistency branch of BSeGN, which can mitigate the domain bias through feature enhancement. To the best of our knowledge, this is the first work to explore embedding and generative learning jointly within the field of ZSL. Extensive evaluations on four benchmarks demonstrate the superiority of BSeGN over its state-of-the-art counterparts.