Core Viewpoint - The development of the Aeneas model by Google's DeepMind aims to assist historians in analyzing and contextualizing ancient Latin inscriptions, addressing the challenges posed by incomplete texts and the vast amount of data available [3][6]. Group 1: Aeneas Model Development - The Aeneas model is a generative AI tool designed to find relationships between Roman-era Latin inscriptions and other texts, helping to determine the context of these inscriptions and predict missing parts [3][6]. - The model was trained on a merged dataset containing 176,861 inscriptions, with 5% accompanied by images, spanning from the 7th century BC to the 8th century AD [6]. Group 2: Model Functionality and Performance - Aeneas consists of three neural networks, each responsible for different tasks: repairing missing text, predicting text origins, and assessing dating [7]. - Historians found that Aeneas provided useful contextual suggestions 90% of the time, improving confidence in key tasks by 44% [7]. - The model's dating accuracy showed a margin of error of less than 13 years, compared to historians' 31 years, indicating a significant improvement in dating inscriptions [7]. Group 3: Practical Applications - Aeneas demonstrated effectiveness in analyzing altars with Latin inscriptions, identifying connections between inscriptions from the same region without prior geographical knowledge [8]. - The model successfully repaired a damaged inscription from a bronze military certificate issued by a Roman emperor, showcasing its ability to handle large datasets that are challenging for humans [10]. - Aeneas provides more logical answers based on its evidence base, contrasting with other popular AI tools that may generate random guesses [10].

Core Insights - The article emphasizes the importance of ensuring food security and sustainable agricultural development in the face of global population growth, climate change, and decreasing arable land resources [1] Group 1: Technological Innovations in Crop Improvement - A review paper published in Nature discusses the integration of multi-omics, genome editing, protein design, high-throughput phenotyping, and artificial intelligence (AI) in crop genetic improvement [2][3] - Modern omics technologies, such as genomics and metabolomics, provide unprecedented capabilities to analyze crop genetic information, revealing new loci for precise trait improvement [4] - High-throughput phenotyping (HTP) technologies utilize drones and sensors for rapid and accurate assessment of crop traits, effectively linking genotype to phenotype [4] Group 2: Genome Editing and Protein Design - Genome editing technologies, exemplified by CRISPR, enable efficient and precise modifications of crop genomes, significantly shortening breeding cycles and rapidly creating desirable traits [4] - AI-driven protein design technologies are emerging, allowing the creation of novel proteins with specific functions, which can lead to breakthroughs in disease resistance and environmental monitoring [4] Group 3: AI-Assisted Crop Design Framework - The review introduces an "AI-assisted crop design" model that integrates and analyzes multimodal big data from genomics, phenomics, environment, and management practices [19] - Breeders can set specific improvement goals, such as yield enhancement or stress resistance, while AI generates optimized breeding plans through deep learning and knowledge reasoning [19] Group 4: Challenges and Future Directions - The article discusses the challenges and future directions for the application of new technologies, highlighting the need for high-quality, standardized data for training AI models [21] - Regulatory policies for genome-edited crops are evolving towards more scientific and simplified frameworks, creating favorable conditions for the widespread application of new technologies [21]

Core Viewpoint - The article discusses a novel strategy to improve CAR-T cell therapy for osteosarcoma by utilizing cuproptosis, which enhances the efficacy of the treatment in a challenging tumor microenvironment [3][4][11]. Group 1: Osteosarcoma and Current Treatment Challenges - Osteosarcoma is a common malignant tumor with a low survival rate of approximately 15%-17% for patients receiving only surgical treatment, and a stagnated 5-year survival rate of about 60% for those undergoing combined surgery and chemotherapy over the past 30 years [1]. - Traditional chemotherapy often leads to drug resistance and tumor recurrence, highlighting the urgent need for improved treatment strategies [1]. Group 2: Novel Research Findings - A study published by a team from Southern Medical University proposes a new approach using cuproptosis to enhance CAR-T cell therapy in osteosarcoma [2]. - The research indicates that copper death can lower PD-L1 expression in osteosarcoma cells, which is positively correlated with copper death-related gene expression [8][11]. - The study developed a biocompatible nanodrug composed of tetrahedral framework nucleic acids (tFNA), elesclomol-Cu, and anti-PD-L1 antibodies to induce copper death and block the PD-1-PD-L1 signaling axis, thereby reshaping the immunosuppressive tumor microenvironment [3][9]. Group 3: Implications for CAR-T Cell Therapy - The results demonstrate that the use of this nanodrug significantly enhances CAR-T cell infiltration and anti-tumor activity in both in situ and recurrent osteosarcoma models [10]. - This research provides insights into the relationship between copper metabolism and PD-L1 expression, offering a potential universal method to improve adoptive cell therapy for solid tumors [4].

撰文丨王聪 编辑丨王多鱼 排版丨水成文 多层结构的 核仁 是 核糖体 生物合成的主要场所，在这里，构成核糖体小亚基 （SSU） 和大亚基 （LSU） 的 rRNA 前体 （ pre-rRNA ） 依次成熟。然而， pre-rRNA 加工与核仁亚结构之间的空间功能关系，以及这种关系如何适应细胞生理需求的变化，一直未被完全理解。 2025 年 7 月 23 日，中国科学院分子细胞科学卓越创新中心 陈玲玲 研究员团队 （博士生 潘宇航 、博士后 单琳 、博士生 张宇瑶 为共同第一作者） 在国际顶 尖学术期刊 Nature 上加速上线了题为： Pre-rRNA spatial distribution and functional organisation of the nucleolus 的研究论文。 该研究系统解析了构成核糖体大小亚基的 rRNA 前体 （ pre-rRNA ） 在核仁中的动态成熟过程，发现了核糖体小亚基 （ SSU ） pre-rRNA 的加工效率直接调 控核仁内层结构的稳定性，提出了 pre-rRNA 加工的区域化模 型 及其在多层结构核仁的功能与进化中具有重要意义 。 真核细胞的细胞核中的 ...

Core Viewpoint - The recent studies reveal a newly identified biosynthetic pathway for salicylic acid in plants, highlighting its conserved nature across seed plants and providing insights into disease resistance mechanisms in major crops [4][7][16]. Group 1: Research Findings - The research team from Sichuan University identified a three-step biosynthesis pathway of salicylic acid from benzoyl-CoA in plants, which includes the formation of benzyl benzoate, followed by hydroxylation to produce benzyl salicylate, and finally hydrolysis to yield salicylic acid [3][4][6]. - The genes encoding the three key enzymes (BEBT, BBO, and BSH) involved in this pathway are widely present in various plants, including willows, poplars, soybeans, and rice, indicating a broad conservation of this biosynthetic route [6][10]. - The studies from Zhejiang University and Zhejiang Normal University further confirmed the conservation of this biosynthetic pathway in rice and other crops, establishing a comprehensive understanding of salicylic acid synthesis from phenylalanine [10][13][16]. Group 2: Implications for Agriculture - The identification of the salicylic acid biosynthesis pathway provides a molecular basis for understanding the differences in disease resistance mechanisms among different plant groups, particularly major food crops [4][7]. - This research opens new directions and targets for breeding disease-resistant crop varieties, which is crucial for enhancing agricultural productivity and sustainability [4][7][16].

Core Viewpoint - The recent studies from Zhejiang Normal University, Sichuan University, and Zhejiang University have successfully elucidated the biosynthesis pathways of salicylic acid in plants, particularly focusing on the phenylalanine ammonia-lyase (PAL) pathway, which is crucial for plant defense mechanisms [2][10][18]. Group 1: Research Findings - The study published by the team from Zhejiang Normal University and Brookhaven National Laboratory provides a complete analysis of the PAL biosynthesis pathway of salicylic acid from phenylalanine in rice, confirming its conservation in most seed plants [2][9]. - The activation of the PAL pathway in rice significantly enhances salicylic acid levels and the plant's immune capacity [9][10]. - The research identifies three key enzymes involved in the biosynthesis of salicylic acid, filling a long-standing gap in understanding this critical defense hormone [18]. Group 2: Methodology - The PAL biosynthesis pathway involves several steps: 1. OSD1 catalyzes the conversion of trans-cinnamic acid to cinnamoyl-CoA, which is then transformed into benzoyl-CoA through β-oxidation in peroxisomes [6]. 2. Benzoyl-CoA is converted to benzyl benzoate by the action of OSD2 [6]. 3. Finally, benzyl benzoate is hydroxylated to benzyl salicylate by cytochrome P450 enzyme OSD3, which is then hydrolyzed to salicylic acid by OSD4 [5][6][10]. Group 3: Broader Implications - The completion of the PAL pathway provides critical insights into the primary biosynthesis route of salicylic acid across different plant species, offering a precise target for regulating crop immunity [10]. - The findings from these studies contribute to understanding the differences in disease resistance mechanisms among various plant groups, particularly major food crops [13][18].

Core Viewpoint - CAR-T cell therapy has significantly transformed the treatment of B-cell malignancies and is showing initial effectiveness in solid tumors, but there is an urgent need to optimize current treatment protocols due to many patients not responding or developing resistance [2]. Group 1: CAR Therapy Mechanisms - The progression of disease after CAR therapy can be attributed to various factors, categorized into intrinsic tumor factors, tumor microenvironment, or intrinsic resistance mechanisms of lymphocytes [2]. - Factors related to lymphocytes expressing CAR are particularly noteworthy, as they may be altered during the in vitro cell manufacturing process [2]. - Recent clinical data indicate that the expansion and persistence of CAR-T cells in vivo are often associated with better therapeutic outcomes, which also applies to CAR-NK cells [2]. Group 2: Research Findings - A study published on July 22, 2025, in Nature Cancer reveals that the persistence of CAR-engineered lymphocytes (CAR-T and CAR-NK cells) is regulated by a FAS ligand–FAS autoregulatory circuit [3][4]. - The expression of FAS ligand (FAS-L) is primarily found in endogenous T cells, natural killer (NK) cells, and CAR-T cells, with minimal expression in tumor and stromal cells [9]. - The research team demonstrated that CAR-T and CAR-NK cell survival is influenced by FAS-L regulation, as evidenced by the enrichment of CAR-T and CAR-NK cells expressing a dominant negative FAS receptor (ΔFAS) post-transplantation [10]. Group 3: Implications for Treatment - The study suggests that the persistence of CAR-engineered lymphocytes is controlled by the FAS-L/FAS autoregulatory circuit, which could have significant implications for enhancing the efficacy of CAR therapies [11].

Core Viewpoint - The study reveals that lactylation of YTHDC1 at K82 enhances its phase separation, stabilizing oncogenic mRNA and promoting the progression of renal cell carcinoma (RCC) in a hypoxic environment [2][6][9]. Group 1: Research Findings - The research systematically mapped the lactylation profile of proteins under hypoxic conditions in RCC, focusing on the functional mechanism of YTHDC1 K82 lactylation [2][6]. - Elevated levels of global lysine lactylation (Kla) were found in human RCC tissues and cells, which promotes malignant development of RCC [6][7]. - YTHDC1 K82 lactylation, mediated by p300 under hypoxic conditions, promotes the malignancy of RCC both in vitro and in vivo [6][7]. Group 2: Mechanism of Action - YTHDC1 K82 lactylation enhances the phase separation of YTHDC1, leading to the expansion of nuclear condensates that protect oncogenic transcripts BCL2 and E2F2 from degradation by the PAXT-EXO complex [6][7][9]. - The study highlights that the increased lysine lactylation regulates the stability of YTHDC1 target genes, thereby facilitating the progression of RCC [9]. Group 3: Study Highlights - Quantitative lactylation proteomics analysis revealed high levels of lactylation modification proteins under hypoxic conditions [7]. - The study identifies a novel regulatory pathway involving YTHDC1 lactylation that opens new therapeutic targets in the intersection of tumor metabolism and RNA regulation [2][6].

Core Viewpoint - Cancer cachexia significantly alters the metabolism of patients, leading to involuntary weight loss and increased mortality, with no FDA-approved treatments available to fully reverse the condition [2][3][6]. Group 1: Cancer Cachexia Overview - 50%-80% of cancer patients experience cancer cachexia, resulting in functional decline, decreased quality of life, increased chemotherapy toxicity, and higher mortality rates [3]. - Cancer cachexia accounts for at least 20% of cancer-related deaths, highlighting the urgency for effective treatments [3]. Group 2: Research Findings - A study published in Cell identified the liver as a previously overlooked driver of cancer cachexia, revealing that the disruption of the biological clock gene REV-ERBα in the liver promotes the release of hepatokines that enhance catabolism [4][5]. - Reactivating REV-ERBα expression or inhibiting specific hepatokines (LBP, ITIH3, IGFBP1) significantly improved weight loss in mouse models of cancer cachexia [5][11]. Group 3: Mechanisms of Action - The study demonstrated that the liver undergoes metabolic reprogramming in cancer cachexia, with the biological clock gene REV-ERBα becoming inactive, leading to increased release of catabolic hepatokines [10][11]. - In cancer cachexia patients, levels of LBP, ITIH3, and IGFBP1 were significantly elevated compared to weight-stable cancer patients, indicating their role in promoting tissue wasting [12][15]. Group 4: Implications for Treatment - The findings suggest that the liver actively contributes to the progression of cancer cachexia rather than being a passive responder, providing new biomarkers and therapeutic targets for better diagnosis and treatment interventions [16].

Core Viewpoint - Alzheimer's disease (AD) is a prevalent neurodegenerative disorder with no effective treatment to reverse its progression, necessitating urgent action for better therapeutic options [2][3][6]. Group 1: Current State of Alzheimer's Disease - Over 50 million people globally suffer from Alzheimer's or related dementias, with projections indicating this number will double by 2050 [3]. - Annual global spending on Alzheimer's exceeds $1 trillion, making it one of the most costly health issues [3]. - The drug development landscape for Alzheimer's faces significant challenges, with a failure rate of 98% over recent decades [6]. Group 2: Research Findings - A recent study published in Cell identified two FDA-approved cancer drugs, letrozole and irinotecan, that, when used in combination, significantly improved memory and reduced Alzheimer's-related pathology in mouse models [4][17][20]. - The research utilized publicly available data to analyze gene expression changes in neurons and glial cells associated with Alzheimer's, leading to the identification of potential drug candidates [9][11][20]. - The study highlighted the importance of glial cell dysfunction alongside neuronal vulnerability in the disease's heterogeneous pathology [7][8]. Group 3: Methodology and Results - The research team compared gene expression features from Alzheimer's patients with a drug response database, narrowing down 1,300 drugs to 86, then to 10, and finally to 5 promising candidates [10][13]. - The combination therapy of letrozole and irinotecan was shown to reverse multiple Alzheimer's symptoms in mouse models, eliminating toxic protein plaques and restoring memory [17][20]. - The study emphasizes a cell-type directed, multi-target drug discovery strategy based on human data and real-world evidence, paving the way for precision medicine in Alzheimer's treatment [18][20].