Me-LLaMA: Foundation Large Language Models for Medical Applications.

Recent advancements in large language models (LLMs) such as ChatGPT and LLaMA have hinted at their potential to revolutionize medical applications, yet their application in clinical settings often reveals limitations due to a lack of specialized training on medical-specific data. In response to this challenge, this study introduces Me-LLaMA, a novel medical LLM family that includes foundation models - Me-LLaMA 13/70B, along with their chat-enhanced versions - Me-LLaMA 13/70B-chat, developed through continual pre-training and instruction tuning of LLaMA2 using large medical datasets. Our methodology leverages a comprehensive domain-specific data suite, including a large-scale, continual pre-training dataset with 129B tokens, an instruction tuning dataset with 214k samples, and a new medical evaluation benchmark (MIBE) across six critical medical tasks with 12 datasets. Our extensive evaluation using the MIBE shows that Me-LLaMA models achieve overall better performance than existing open-source medical LLMs in zero-shot, few-shot and supervised learning abilities. With task-specific instruction tuning, Me-LLaMA models outperform ChatGPT on 7 out of 8 datasets and GPT-4 on 5 out of 8 datasets. In addition, we investigated the catastrophic forgetting problem, and our results show that Me-LLaMA models outperform other open-source medical LLMs in mitigating this issue. Me-LLaMA is one of the largest open-source medical foundation LLMs that use both biomedical and clinical data. It exhibits superior performance across both general and medical tasks compared to other open-source medical LLMs, rendering it an attractive choice for medical AI applications. We release our models, datasets, and evaluation scripts at: https://github.com/BIDS-Xu-Lab/Me-LLaMA.

Quantifying the contributions of meteorology, emissions, and transport to ground-level ozone in the Pearl River Delta, China.

Ozone pollution presents a growing air quality threat in urban agglomerations in China. It remains challenge to distinguish the roles of emissions of precursors, chemical production and transportations in shaping the ground-level ozone trends, largely due to complicated interactions among these 3 major processes. This study elucidates the formation factors of ozone pollution and categorizes them into local emissions (anthropogenic and biogenic emissions), transport (precursor transport and direct transport from various regions), and meteorology. Particularly, we attribute meteorology, which affects biogenic emissions and chemical formation as well as transportation, to a perturbation term with fluctuating ranges. The Community Multiscale Air Quality (CMAQ) model was utilized to implement this framework, using the Pearl River Delta region as a case study, to simulate a severe ozone pollution episode in autumn 2019 that affected the entire country. Our findings demonstrate that the average impact of meteorological conditions changed consistently with the variation of ozone pollution levels, indicating that meteorological conditions can exert significant control over the degree of ozone pollution. As the maximum daily 8-hour average (MDA8) ozone concentrations increased from 20 % below to 30 % above the National Ambient Air Quality Standard II, contributions from emissions and precursor transport were enhanced. Concurrently, direct transport within Guangdong province rose from 13.8 % to 22.7 %, underscoring the importance of regional joint prevention and control measures under adverse weather conditions. Regarding biogenic emissions and precursor transport that cannot be directly controlled, we found that their contributions were generally greater in urban areas with high nitrogen oxides (NOx) levels, primarily due to the stronger atmospheric oxidation capacity facilitating ozone formation. Our results indicate that not only local anthropogenic emissions can be controlled in urban areas, but also the impacts of local biogenic emissions and precursor transport can be potentially regulated through reducing atmospheric oxidation capacity.

Nano-microflora Interaction Inducing Pulmonary Inflammation by Pyroptosis.

Antimicrobial nanomaterials frequently induce inflammatory reactions within lung tissues and prompt apoptosis in lung cells, yielding a paradox due to the inherent anti-inflammatory character of apoptosis. This paradox accentuates the elusive nature of the signaling cascade underlying nanoparticle (NP)-induced pulmonary inflammation. In this study, we unveil the pivotal role of nano-microflora interactions, serving as the crucial instigator in the signaling axis of NP-induced lung inflammation. Employing pulmonary microflora-deficient mice, we provide compelling evidence that a representative antimicrobial nanomaterial, silver (Ag) NPs, triggers substantial motility impairment, disrupts quorum sensing, and incites DNA leakage from pulmonary microflora. Subsequently, the liberated DNA molecules recruit caspase-1, precipitating the release of proinflammatory cytokines and activating N-terminal gasdermin D (GSDMD) to initiate pyroptosis in macrophages. This pyroptotic cascade culminates in the emergence of severe pulmonary inflammation. Our exploration establishes a comprehensive mechanistic axis that interlinks the antimicrobial activity of Ag NPs, perturbations in pulmonary microflora, bacterial DNA release, macrophage pyroptosis, and consequent lung inflammation, which helps to gain an in-depth understanding of the toxic effects triggered by environmental NPs.

Advancing entity recognition in biomedicine via instruction tuning of large language models.

MOTIVATION: Large Language Models (LLMs) have the potential to revolutionize the field of Natural Language Processing, excelling not only in text generation and reasoning tasks but also in their ability for zero/few-shot learning, swiftly adapting to new tasks with minimal fine-tuning. LLMs have also demonstrated great promise in biomedical and healthcare applications. However, when it comes to Named Entity Recognition (NER), particularly within the biomedical domain, LLMs fall short of the effectiveness exhibited by fine-tuned domain-specific models. One key reason is that NER is typically conceptualized as a sequence labeling task, whereas LLMs are optimized for text generation and reasoning tasks. RESULTS: We developed an instruction-based learning paradigm that transforms biomedical NER from a sequence labeling task into a generation task. This paradigm is end-to-end and streamlines the training and evaluation process by automatically repurposing pre-existing biomedical NER datasets. We further developed BioNER-LLaMA using the proposed paradigm with LLaMA-7B as the foundational LLM. We conducted extensive testing on BioNER-LLaMA across three widely recognized biomedical NER datasets, consisting of entities related to diseases, chemicals, and genes. The results revealed that BioNER-LLaMA consistently achieved higher F1-scores ranging from 5% to 30% compared to the few-shot learning capabilities of GPT-4 on datasets with different biomedical entities. We show that a general-domain LLM can match the performance of rigorously fine-tuned PubMedBERT models and PMC-LLaMA, biomedical-specific language model. Our findings underscore the potential of our proposed paradigm in developing general-domain LLMs that can rival SOTA performances in multi-task, multi-domain scenarios in biomedical and health applications. AVAILABILITY AND IMPLEMENTATION: Datasets and other resources are available at https://github.com/BIDS-Xu-Lab/BioNER-LLaMA.

Faithful AI in Medicine: A Systematic Review with Large Language Models and Beyond.

Objective: While artificial intelligence (AI), particularly large language models (LLMs), offers significant potential for medicine, it raises critical concerns due to the possibility of generating factually incorrect information, leading to potential long-term risks and ethical issues. This review aims to provide a comprehensive overview of the faithfulness problem in existing research on AI in healthcare and medicine, with a focus on the analysis of the causes of unfaithful results, evaluation metrics, and mitigation methods. Materials and Methods: Using PRISMA methodology, we sourced 5,061 records from five databases (PubMed, Scopus, IEEE Xplore, ACM Digital Library, Google Scholar) published between January 2018 to March 2023. We removed duplicates and screened records based on exclusion criteria. Results: With 40 leaving articles, we conducted a systematic review of recent developments aimed at optimizing and evaluating factuality across a variety of generative medical AI approaches. These include knowledge-grounded LLMs, text-to-text generation, multimodality-to-text generation, and automatic medical fact-checking tasks. Discussion: Current research investigating the factuality problem in medical AI is in its early stages. There are significant challenges related to data resources, backbone models, mitigation methods, and evaluation metrics. Promising opportunities exist for novel faithful medical AI research involving the adaptation of LLMs and prompt engineering. Conclusion: This comprehensive review highlights the need for further research to address the issues of reliability and factuality in medical AI, serving as both a reference and inspiration for future research into the safe, ethical use of AI in medicine and healthcare.

Genetic and prenatal developmental evaluation of anthraquinone.

Anthraquinone is a recently identified contaminant present in teas globally, and its potential teratogenic and genotoxic impacts have yet to be fully comprehended. Hence, this study's objective was to determine anthraquinone's genotoxicity using various studies such as the Ames test, Mammalian erythrocyte micronucleus test, and in-vitro mammalian chromosome aberration study. Additionally, the study assessed its effects on maternal gestational toxicity and the fetus's teratogenicity through prenatal developmental toxicity research in rats. Results indicated that anthraquinone did not manifest mutagenic effects on Salmonella typhimurium histidine-deficient, did not cause chromosomal aberrations in Chinese hamster ovary cell subclone CHO-K1, and did not exhibit a genotoxic effect on mouse bone marrow erythrocytes. However, in the prenatal developmental toxicity study, administering anthraquinone orally to pregnant rats from day 5 to day 19 of gestation resulted in decreased body weight and food consumption of pregnant rats, along with a higher number of visceral malformations in the fetuses in the highest dose group (217.6 mg/kg BW). Additionally, two pregnant rats died in this group. The study has established the no observed adverse effect level (NOAEL) as 21.76 mg/kg BW, while the lowest observed adverse effect level (LOAEL) was 217.6 mg/kg BW.

Knowledge-enhanced Graph Topic Transformer for Explainable Biomedical Text Summarization.

Given the overwhelming and rapidly increasing volumes of the published biomedical literature, automatic biomedical text summarization has long been a highly important task. Recently, great advances in the performance of biomedical text summarization have been facilitated by pre-trained language models (PLMs) based on fine-tuning. However, existing summarization methods based on PLMs do not capture domain-specific knowledge. This can result in generated summaries with low coherence, including redundant sentences, or excluding important domain knowledge conveyed in the full-text document. Furthermore, the black-box nature of the transformers means that they lack explainability, i.e. it is not clear to users how and why the summary was generated. The domain-specific knowledge and explainability are crucial for the accuracy and transparency of biomedical text summarization methods. In this article, we aim to address these issues by proposing a novel domain knowledge-enhanced graph topic transformer (DORIS) for explainable biomedical text summarization. The model integrates the graph neural topic model and the domain-specific knowledge from the Unified Medical Language System (UMLS) into the transformer-based PLM, to improve the explainability and accuracy. Experimental results on four biomedical literature datasets show that our model outperforms existing state-of-the-art (SOTA) PLM-based summarization methods on biomedical extractive summarization. Furthermore, our use of graph neural topic modeling means that our model possesses the desirable property of being explainable, i.e. it is straightforward for users to understand how and why the model selects particular sentences for inclusion in the summary. The domain-specific knowledge helps our model to learn more coherent topics, to better explain the performance.

A scoping review on multimodal deep learning in biomedical images and texts.

OBJECTIVE: Computer-assisted diagnostic and prognostic systems of the future should be capable of simultaneously processing multimodal data. Multimodal deep learning (MDL), which involves the integration of multiple sources of data, such as images and text, has the potential to revolutionize the analysis and interpretation of biomedical data. However, it only caught researchers' attention recently. To this end, there is a critical need to conduct a systematic review on this topic, identify the limitations of current work, and explore future directions. METHODS: In this scoping review, we aim to provide a comprehensive overview of the current state of the field and identify key concepts, types of studies, and research gaps with a focus on biomedical images and texts joint learning, mainly because these two were the most commonly available data types in MDL research. RESULT: This study reviewed the current uses of multimodal deep learning on five tasks: (1) Report generation, (2) Visual question answering, (3) Cross-modal retrieval, (4) Computer-aided diagnosis, and (5) Semantic segmentation. CONCLUSION: Our results highlight the diverse applications and potential of MDL and suggest directions for future research in the field. We hope our review will facilitate the collaboration of natural language processing (NLP) and medical imaging communities and support the next generation of decision-making and computer-assisted diagnostic system development.

A scoping review on multimodal deep learning in biomedical images and texts.

Computer-assisted diagnostic and prognostic systems of the future should be capable of simultaneously processing multimodal data. Multimodal deep learning (MDL), which involves the integration of multiple sources of data, such as images and text, has the potential to revolutionize the analysis and interpretation of biomedical data. However, it only caught researchers' attention recently. To this end, there is a critical need to conduct a systematic review on this topic, identify the limitations of current work, and explore future directions. In this scoping review, we aim to provide a comprehensive overview of the current state of the field and identify key concepts, types of studies, and research gaps with a focus on biomedical images and texts joint learning, mainly because these two were the most commonly available data types in MDL research. This study reviewed the current uses of multimodal deep learning on five tasks: (1) Report generation, (2) Visual question answering, (3) Cross-modal retrieval, (4) Computer-aided diagnosis, and (5) Semantic segmentation. Our results highlight the diverse applications and potential of MDL and suggest directions for future research in the field. We hope our review will facilitate the collaboration of natural language processing (NLP) and medical imaging communities and support the next generation of decision-making and computer-assisted diagnostic system development.

Faithful AI in Medicine: A Systematic Review with Large Language Models and Beyond.

Artificial intelligence (AI), especially the most recent large language models (LLMs), holds great promise in healthcare and medicine, with applications spanning from biological scientific discovery and clinical patient care to public health policymaking. However, AI methods have the critical concern for generating factually incorrect or unfaithful information, posing potential long-term risks, ethical issues, and other serious consequences. This review aims to provide a comprehensive overview of the faithfulness problem in existing research on AI in healthcare and medicine, with a focus on the analysis of the causes of unfaithful results, evaluation metrics, and mitigation methods. We systematically reviewed the recent progress in optimizing the factuality across various generative medical AI methods, including knowledge-grounded LLMs, text-to-text generation, multimodality-to-text generation, and automatic medical fact-checking tasks. We further discussed the challenges and opportunities of ensuring the faithfulness of AI-generated information in these applications. We expect that this review will assist researchers and practitioners in understanding the faithfulness problem in AI-generated information in healthcare and medicine, as well as the recent progress and challenges in related research. Our review can also serve as a guide for researchers and practitioners who are interested in applying AI in medicine and healthcare.

Fusion protein and hemagglutinin of canine distemper virus co-induce apoptosis in canine mammary tumor cells.

BACKGROUND: Canine distemper virus (CDV) has been shown to have oncolytic activity against primary canine tumors. Previous studies from this laboratory had confirmed that CDV induces apoptosis in canine mammary tumor (CMT) cells, although the molecular mechanism remains unknown. METHODS: The CDV N, P, M, F, H, L, C, and V genes were identified in CDV-L and cloned separately. Mutants with deletions in the 5' region (pCMV-F L△60, pCMV-FL△107, and pCMV-FL△114) or with site-directed mutagenesis in the 3' region (pCMV-FLA602-610) of the F gene were generated. Late-stage apoptotic cells were detected by Hoechst 33342. Early-stage apoptotic cells were detected by AnnexinV-FITC/PI. Quantitative real-time PCR was performed to detect the mRNA levels of target genes of apoptotic and NF-κB pathway. Western blot analysis was performed to detect the expression or phosphorylation levels of target proteins of apoptotic or NF-κB pathway. Immunofluorescence assay was performed to detect the nuclear translocation of p65 protein. Recombinant viruses (rCDV-FL△60 and rCDV-FLA602-610) were rescued by a BHK-T7-based system. 5-week-old female BALB/c nude mice were used to detect the oncolytic activity of recombinant viruses. RESULTS: In this study, it was first confirmed that none of the structural or non-structural proteins of CDV-L, a vaccine strain, was individually able to induce apoptosis in canine mammary tubular adenocarcinoma cells (CIPp) or intraductal papillary carcinoma cells (CMT-7364). However, when CIPp or CMT-7364 cells were co-transfected with glycoprotein fusion (F) and hemagglutinin (H) proteins of CDV-L, nuclear fragmentation was observed and a high proportion of early apoptotic cells were detected, as well as cleaved caspase-3, caspase-8 and poly (ATP ribose) polymerase (PARP). Cleaved caspase-3 and PARP were down-regulated by apoptosis broad-spectrum inhibitor Z-VAD-FMK and caspase-8 pathway inhibitor Z-IETD-FMK, confirming that the F and H proteins coinduced apoptosis in CMT cells via the caspase-8 and caspase-3 pathways. F and H proteins co-induced phosphorylation of p65 and IκBα and nuclear translocation of p65, confirming activation of the NF-κB pathway, inhibition of which down-regulated cleaved caspase-3 and cleaved PARP. Recombinant F protein with enhanced fusion activity and H protein co-induced more cleaved caspase-3 and PARP than parental F protein, while the corresponding recombinant virus exhibited the same properties both in CIPp cells and in a subcutaneous xenograft mouse model. CONCLUSIONS: F and H proteins of CDV-L co-induce apoptosis in CMT cells, while the NF-κB pathway and fusion activity of F protein paly essential roles in the process.

Using Chicken Embryos to Identify the Key Determinants of Nanoparticles for the Crossing of Air-Blood Barriers.

Fine particulates (FPs) are a major class of airborne pollutants. In mammals, FPs may reach the alveoli through the respiratory system, cross the air-blood barrier, spread into other organs, and induce hazardous effects. Although birds have much higher respiratory risks to FPs than mammals, the biological fate of inhaled FPs in birds has rarely been explored. Herein, we attempted to disclose the key properties that dictate the lung penetration of nanoparticles (NPs) by visualizing a library of 27 fluorescent nanoparticles (FNPs) in chicken embryos. The FNP library was prepared by combinational chemistry to tune their compositions, morphologies, sizes, and surface charges. These NPs were injected into the lungs of chicken embryos for dynamic imaging of their distributions by IVIS Spectrum. FNPs with diameters <16 nm could cross the air-blood barrier in 20 min, spread into the blood, and accumulate in the yolk sac. In contrast, large FNPs (>30 nm) were mainly retained in the lungs and rarely detected in other tissues/organs. In addition to size, surface charge was the secondary determinant for NPs to cross the air-blood barrier. Compared to cationic and anionic particles, neutrally charged FNPs showed the fastest lung penetration. A predictive model was therefore developed to rank the lung penetration capability of FNPs by in silico analysis. The in silico predictions could be well validated in chicks by oropharyngeal exposure to six FNPs. Overall, our study discovered the key properties of NPs that are responsible for their lung penetration and established a predictive model that will greatly facilitate respiratory risk assessments of nanoproducts.

Establishment of reverse genetics for genotype VII Newcastle disease virus and altering the cell tropism by inserting TMPRSS2 into the viral genome.

Newcastle disease (ND) is the most important infectious disease in poultry, which is caused by avian orthoavulavirus type 1 (AOAV-1), previously known as Newcastle disease virus (NDV). In this study, an NDV strain SD19 (GenBank accession number OP797800) was isolated, and phylogenetic analysis suggested the virus belongs to the class II genotype VII. After generating wild-type rescued SD19 (rSD19), the attenuating strain (raSD19) was generated by mutating the F protein cleavage site. To explore the potential role of the transmembrane protease, serine S1 member 2 (TMPRSS2), the TMPRSS2 gene was inserted into the region between the P and M genes of raSD19 to generate raSD19-TMPRSS2. Besides, the coding sequence of the enhanced green fluorescent protein (EGFP) gene was inserted in the same region as a control (rSD19-EGFP and raSD19-EGFP). The Western blot, indirect immunofluorescence assay (IFA), and real-time quantitative PCR were employed to determine the replication activity of these constructs. The results reveal that all the rescued viruses can replicate in chicken embryo fibroblast (DF-1) cells; however, the proliferation of raSD19 and raSD19-EGFP needs additional trypsin. We next evaluated the virulence of these constructs, and our results reveal that the SD19, rSD19, and rSD19-EGFP are velogenic; the raSD19 and raSD19-EGFP are lentogenic; and the raSD19-TMPRSS2 are mesogenic. Moreover, due to the enzymatic hydrolysis of serine protease, the raSD19-TMPRSS2 can support itself to proliferate in the DF-1 cells without adding exogenous trypsin. These results may provide a new method for the NDV cell culture and contribute to ND's vaccine development.

Prognostic value of left ventricular global longitudinal strain on speckle echocardiography for predicting chemotherapy-induced cardiotoxicity in breast cancer patients: A systematic review and meta-analysis.

BACKGROUND: Literature suggests that left ventricular global longitudinal strain (LV-GLS) on speckle echocardiography has the potential to predict cardiotoxicity amongst breast cancer patients receiving chemotherapy such as anthracycline, taxane, cyclophosphamide, and trastuzumab. Our study aimed to collect evidence for the prognostic value of LV-GLS for predicting chemotherapy-induced cardiotoxicity in breast cancer patients. METHODS: A detailed search of the PubMed, Google Scholar, Cochrane Library, and Scopus databases was conducted for published articles up to August 31, 2022. In our meta-analysis, we looked at 13 studies with a total of 1007 breast cancer patients getting chemotherapy that looked at the predictive value of GLS. RESULTS: Absolute GLS change during treatment showed a pooled sensitivity of 84% (95% CI 74% to 91%) and a pooled specificity of 77% (95% CI 68% to 84%).  For a relative change in GLS, we observed a pooled sensitivity of 76% (95% CI 56% to 89%) and a pooled specificity of 83% (95% CI 73% to 90%).  For an absolute change in GLS, we observed a positive likelihood ratio (LR), and the negative LR was 4 and .21. Summary receiver operating characteristics curve with prediction and confidence intervals represents a promising summary area under the curve (sAUC) of .88, 95% CI ranges from .85 to .91 for absolute change in GLS, as well as for relative change (sAUC, .87, 95% CI .84 to .90). CONCLUSION: Our results demonstrated an estimation of LV-GLS after the beginning of required chemotherapy, including anthracyclines and trastuzumab, had a promising prognostic value for predicting the likelihood of cancer therapeutics-related cardiac dysfunction. To confirm our findings, well-designed prospective adequately powered diagnostic randomised trials are necessary.

G protein-coupled receptor 35 attenuates nonalcoholic steatohepatitis by reprogramming cholesterol homeostasis in hepatocytes.

Nonalcoholic fatty liver disease (NAFLD) is the most common chronic liver disease worldwide. Fat accumulation "sensitizes" the liver to insult and leads to nonalcoholic steatohepatitis (NASH). G protein-coupled receptor 35 (GPR35) is involved in metabolic stresses, but its role in NAFLD is unknown. We report that hepatocyte GPR35 mitigates NASH by regulating hepatic cholesterol homeostasis. Specifically, we found that GPR35 overexpression in hepatocytes protected against high-fat/cholesterol/fructose (HFCF) diet-induced steatohepatitis, whereas loss of GPR35 had the opposite effect. Administration of the GPR35 agonist kynurenic acid (Kyna) suppressed HFCF diet-induced steatohepatitis in mice. Kyna/GPR35 induced expression of StAR-related lipid transfer protein 4 (STARD4) through the ERK1/2 signaling pathway, ultimately resulting in hepatic cholesterol esterification and bile acid synthesis (BAS). The overexpression of STARD4 increased the expression of the BAS rate-limiting enzymes cytochrome P450 family 7 subfamily A member 1 (CYP7A1) and CYP8B1, promoting the conversion of cholesterol to bile acid. The protective effect induced by GPR35 overexpression in hepatocytes disappeared in hepatocyte STARD4-knockdown mice. STARD4 overexpression in hepatocytes reversed the aggravation of HFCF diet-induced steatohepatitis caused by the loss of GPR35 expression in hepatocytes in mice. Our findings indicate that the GPR35-STARD4 axis is a promising therapeutic target for NAFLD.

Engineering active heterojunction architecture with oxygenated-Co, Mo bimetallic sulfide heteronanosheet and graphene oxide for peroxymonosulfate activation.

Bimetallic sulfides have distinctive catalytic property in activating peroxymonosulfate (PMS) for water remediation. Polyoxometalates as potential precursors have rarely been reported for the catalytic degradation of refractory organic pollutants. Herein, a composite catalyst of Co-Mo bimetallic sulfides supported onto graphene oxide (O-CoMoS/GO) with a heterojunction architecture was synthesized through a hydrothermal strategy with polyoxometalates ((NH4)4[CoIIMo6O24H6]·6H2O) as the precursor and applied in the PMS activation. This material showed a superior performance for the catalytic degradation of the model organic pollutant, 4-chlorophenol (rapidly removed within 10 min with an apparent reaction rate constant of 0.5458 min-1). O-CoMoS/GO outperformed most of the reported catalysts in terms of activity and had a strong tolerance towards common organic and inorganic compounds in water, and could perform well in different real water systems. Experimental and theoretical results indicated that the introduction of GO could achieve the enrichment of electrons on the metals and reduce the d band center (εd) of Co close to the Fermi level (εF), thereby facilitating the interfacial electron transfer process. The activation mechanism was due to the as-prepared bimetallic sulfides and the formation of heterojunction structure with GO, where Co(II) as the active center could be regenerated by the adjacent Mo element (as co-catalyst) and by gathering electrons from GO through the Co/Mo-O-C coupling. This work provides insights into the design of bimetallic sulfide catalysts in activating PMS for water remediation.

Doped Graphene To Mimic the Bacterial NADH Oxidase for One-Step NAD+ Supplementation in Mammals.

Nicotinamide adenine dinucleotide (NAD) is a critical regulator of metabolic networks, and declining levels of its oxidized form, NAD+, are closely associated with numerous diseases. While supplementing cells with precursors needed for NAD+ synthesis has shown poor efficacy in combatting NAD+ decline, an alternative strategy is the development of synthetic materials that catalyze the oxidation of NADH into NAD+, thereby taking over the natural role of the NADH oxidase (NOX) present in bacteria. Herein, we discovered that metal-nitrogen-doped graphene (MNGR) materials can catalyze the oxidation of NADH into NAD+. Among MNGR materials with different transition metals, Fe-, Co-, and Cu-NGR displayed strong catalytic activity combined with >80% conversion of NADH into NAD+, similar specificity to NOX for abstracting hydrogen from the pyridine ring of nicotinamide, and higher selectivity than 51 other nanomaterials. The NOX-like activity of FeNGR functioned well in diverse cell lines. As a proof of concept of the in vivo application, we showed that FeNGR could specifically target the liver and remedy the metabolic flux anomaly in obesity mice with NAD+-deficient cells. Overall, our study provides a distinct insight for exploration of drug candidates by design of synthetic materials to mimic the functions of unique enzymes (e.g., NOX) in bacteria.

Inhibition of transcription of VP2 by mutations in the DNA binding domains of mink enteritis virus NS1 protein.

The NS1 protein of mink enteritis virus (MEV) is a multidomain and multifunctional protein that plays a critical role in viral replication, with predicted nuclease, helicase and transactivation activities. The nuclease and helicase domains of NS1 protein are involved in interaction with viral DNA. Herein, potential amino acids critical for DNA binding in the MEV NS1 were mutated, all of which resulted in a termination of viral production from an infectious MEV clone. Although E121, H129/131, Y212 and K470/472 mutants retained their P38 and 5'UTR transactivation activity, K196/197 and K406 mutations eliminated this. Interestingly, VP2 protein was produced following transfection of F81 cells with pMEV-NS1-196K2G (K196G and K197G) and pMEV-NS1-K406G when pNS1 was co-transfected in trans, indicating that the substitutions did not affect the integrity of the DNA sequence that bound to NS1 protein but inhibited the biological properties of NS1 protein itself. The ability of NS1 protein to interact with SP1 was inhibited by both 196K2G and K406G substitutions, while 196K2G resulted in failure to bind to the DNA-binding sites in the P38 promoter, and the oligomerization of K406G was inhibited. All of these could explain the transcriptional repression.

Transcriptional activation of Mink enteritis virus VP2 by the C-terminal of its NS1 protein.

Mink enteritis virus (MEV) NS1 is a multidomain and multifunctional protein containing origin binding, helicase, and transactivation domains. In particular, parvoviral NS1 proteins are transactivators of the viral capsid protein promoter although the manner by which they exert these transactivation effects remained unclear. In this study, the region of the transactivation domain of the NS1 C-terminal was found located at aa 557 ~ 668 and any deletion within this region reduced the transactivation activity. A dominant negative mutation of the 63 aa deletion in the C-terminal of NS1 protein resulted in loss of ability to activate P38 and VP2-5'UTR in a dual-luciferase reporter assay system, a VP2 protein expression system, and within the whole MEV genome, independent of downstream genes. Additionally, a full-length MEV clone deficient in its NS1 C-terminal failed to rescue the virus, possibly due to the loss of integrity of DNA sequences interacting with NS1 protein, and expression of VP2 was also inhibited even when normal NS1 protein was supplied in trans.

Structure and function of the parvoviral NS1 protein: a review.

Parvoviruses possess a single-stranded DNA genome of about 5 kb, which contains two open reading frames (ORFs), one encoding nonstructural (NS) proteins, the other capsid proteins. The NS1 protein contains an N-terminal origin-binding domain, a helicase domain, and a C-terminal transactive domain, and is essential for effective viral replication and production of infectious virus. We first summarize the developments in the structure of NS1 protein, including the original binding domain and the helicase domain. We discuss the role of different DNA substrates in the oligomerization of these two domains of NS1. During the parvovirus life cycle, the NS1 protein is closely related to the viral gene expression, viral replication, and infection. We provide the current understanding of the impact of parvovirus NS1 protein mutations on its biological properties. Overall, in this review, we focus on the structure and function of the parvoviral NS1 protein.