Core Viewpoint - The research published by a team from China Pharmaceutical University reveals a novel brain-liver neurocircuit involving VMH Galnt2 neurons that counterregulate hypoglycemia by increasing hepatic glucose production, providing new insights into the mechanisms of hypoglycemia resistance and potential innovative treatment strategies for metabolic diseases related to glucose regulation [3][7]. Summary by Sections Research Findings - The study identifies a biphasic pattern of glucose dynamics in the blood and hypothalamus during prolonged fasting, highlighting an additional threshold-dependent counterregulatory mechanism [4]. - This mechanism is mediated by a neural pathway from the ventromedial hypothalamus (VMH) to the paraventricular nucleus (PVH), then to the lateral paragigantocellular nucleus (LPGi), and finally to the liver, which detects neuroglycopenic states and activates sympathetic signals to drive hepatic glucose production [4]. Key Highlights - The biphasic model explains the counterregulatory response to hypoglycemia [5]. - VMH glucose-inhibitory neurons play a critical functional role in sensing neuroglycopenia [5]. - The brain-liver neural circuit emphasizes the counterregulatory response to hypoglycemia [5]. - Galnt2 serves as both a genetic marker and a molecular brake for VMH glucose-inhibitory neurons, regulating the glucose sensing threshold and metabolic homeostasis [5]. Implications - The study underscores a brain-liver neural pathway originating from VMH Galnt2 neurons that can sense and counterregulate hypoglycemia, potentially guiding the development of innovative treatment strategies for metabolic diseases characterized by abnormal glucose regulation [7].

撰文丨王聪 编辑丨王多鱼 排版丨水成文 植物进化出了一套具有强化管壁和创新壁结构的导管系统，以确保水分和养分的高效运输。 导管纹孔 （ Vessel pits ） ——精细的三维空腔结构——是决定植物水力系统和生长可塑性的关键因素。然而，其超微 结构及形成机制一直未知。 2025 年 10 月 14 日，中国科学院遗传与发育生物学研究所 张保才 、 周奕华 研究团队 （ 张兰军 、 高 易宏 为共同第一作者 ） 在国际顶尖学术期刊 Cell 上发表了题为： Shaping pit structure in vessel walls sustains xylem hydraulics and grain yield 的研究论文。 该研究利用新兴体电子显微镜技术， 首次揭示了水稻木质部导管纹孔的完整精细三维结构 ，通过克服多项 技术瓶颈，功能鉴定了塑造纹孔三维结构的关键分子模块，发现纹孔三维结构天然突变受 MYB61-PS1 分 子模块调控，并在维持木质部稳健性、促进氮素运输与水稻高产中发挥关键作用。 由此，研究团队发现了一条可提升木质部输水能力及作物产量的分子通路，为可持续农业发展提供了富有 前景的新策略。 论 ...

Core Insights - The article discusses the revolutionary advancements in structural biology through the use of cryo-electron microscopy (cryo-EM) and AI tools like AlphaFold, enabling high-resolution observation and prediction of protein structures [3][6] - A paradigm shift is occurring in structural biology, moving from targeted structure determination to structure-guided discovery of previously uncharacterized biological entities [3][6] - The CryoSeek strategy, proposed by Yan Ning's team, combines cryo-EM with AI-assisted modeling and bioinformatics analysis to discover entirely new biological entities in nature [3][6] Summary by Sections CryoSeek Strategy - In October 2024, Yan Ning's team introduced CryoSeek, utilizing cryo-EM as an observational tool combined with AI for automatic modeling and bioinformatics analysis [3][6] - The strategy aims to identify unknown biological entities from any accessible source, showcasing its potential in discovering novel nanostructures [6] Discoveries in Glycobiology - In October 2025, the team published two papers detailing the discovery of new glycofibril structures using the CryoSeek strategy, highlighting the importance of sugars in biological assembly [4][11] - The research emphasizes that sugars can act as information molecules and structural scaffolds, opening new avenues in glycobiology [4][11] New Algorithms and Techniques - The team developed a new algorithm named Ahaha for determining the absolute handedness of glycofibrils, addressing a significant challenge in sugar science [12][14] - Ahaha has been deployed as an online service, facilitating global access for researchers to measure the absolute handedness of sugar fibers [16] Structural Insights - The research identified five new glycofibrils with varying compositions and structural assemblies, revealing a continuum from protein-dominant to sugar-dominant structures [8][9] - The findings indicate that glycan-mediated interactions are crucial for the assembly of these diverse fibers, challenging the traditional view that proteins are the primary builders of biological structures [11][12]

Core Insights - The research highlights the role of astrocytic Ca2+ signaling as a "gatekeeper" in preventing synaptic depotentiation during motor learning, significantly altering the understanding of astrocytes' function in synaptic strength maintenance and optimization [4][8]. Group 1: Mechanisms of Motor Learning - Motor learning involves the dynamic adjustment of synaptic strength, where synaptic potentiation is crucial for memory and skill formation, while synaptic depotentiation must be regulated to retain newly acquired skills [3][6]. - The study found that motor training induces synaptic potentiation in layer 5 pyramidal neurons of the mouse motor cortex, accompanied by an increase in astrocytic Ca2+ levels [6][7]. Group 2: Role of Astrocytes - Astrocytic Ca2+ activity regulates activity-dependent synaptic plasticity, and its role in learning-related synaptic changes in vivo remains unclear [6]. - Reducing astrocytic Ca2+ levels leads to synaptic depotentiation during motor training, impairing improvements in motor performance [6][7]. Group 3: Implications for Medical Research - The findings suggest potential therapeutic avenues targeting astrocytic signaling pathways for treating conditions like stroke, neurodegenerative diseases, or motor function disorders by modulating astrocytic Ca2+ or adenosine receptor activity [7][8]. - The study emphasizes the complexity and significance of glial cell interactions in functional brain plasticity, marking an exciting new chapter in neuroscience discoveries [8].

Core Viewpoint - Parkinson's disease (PD) is a progressive neurodegenerative disorder characterized by the selective loss of dopaminergic neurons, leading to dopamine depletion and motor symptoms. Current treatments alleviate symptoms but do not halt neuronal loss [2][5]. Group 1: Research and Development - PD has long been considered a suitable candidate for cell replacement therapy due to its clear pathological features, specifically the loss of certain dopaminergic neuron populations [5]. - A recent study published in the journal Cell reported a Phase 1/2a clinical trial of human embryonic stem cell (hESC)-derived dopamine progenitors (A9-DPC) showing good safety and effective improvement in motor symptoms for PD patients [3][12]. - The development of hPSC, including hESC and induced pluripotent stem cells (iPSC), has provided scalable sources for generating dopamine progenitor cells, enabling large-scale production of high-purity dopamine progenitor cells [5][6]. Group 2: Clinical Trial Findings - The Phase 1/2a trial involved 12 patients with severe PD, where A9-DPC was transplanted into the bilateral shell nucleus under an immunosuppressive regimen. The low-dose group received 3.15 million cells, while the high-dose group received 6.3 million cells [8][9]. - Results indicated excellent safety with no dose-limiting toxicities or transplant-related adverse events. Motor function improved significantly, with the high-dose group showing greater improvement [8][9]. - PET imaging confirmed increased dopamine uptake in the bilateral posterior shell nucleus, particularly in the high-dose group, indicating cell survival post-transplant [9][12]. Group 3: Conclusion and Implications - The study concluded that hESC-derived dopamine progenitor cells (A9-DPC) transplantation into the bilateral shell nucleus demonstrates good safety and effectively improves motor symptoms in PD patients, exhibiting dose-dependent efficacy [12].

撰文丨王聪 编辑丨王多鱼 排版丨水成文 T 细胞恶性肿瘤 是一组异质性疾病，复发率和死亡率都很高，其总体预后通常不佳 （尤其是与 B 细胞恶性肿瘤相比） 。尽管目前诸如化疗之类的策略对某些疾 病亚型有益，但对于复发或难治性疾病患者，其疗效仍然有限。因此，开发针对 T 细胞恶性肿瘤的创新疗法势在必行。 CD5 是恶性 T 细胞的特征性标志物，几乎在所有正常 T 细胞上均有表达。因此，在治疗 T 细胞恶性肿瘤时，不表达 CD5 的 自然杀伤 （NK） 细胞 ，可能比基 于 T 细胞的疗法具有更好的安全性。 2025 年 10 月 8 日，四川大学华西医院 仝爱平 教授、 牛挺 教授等在 Cell 子刊 Cell Reports Medicine 上发表了题为： A nanobody-based tri-specific NK cell engager targeting CD5 triggers antitumor immunity 的研究论文。 该研究开发了一种基于 纳米抗体 的 三特异性 NK 细胞接合器 （ tri-NKCE ） ，通过靶向 CD5， 对 CD5 + 恶性肿瘤细胞表现出更强的抗肿瘤效果， 与 ...

然而，一项最新研究颠覆了上述观点，实际上， 在感染 HSV-1 和 IAV 这两种人类病原体后， 宿主细胞编码的 Z-RNA 是主要的且足以激活 ZBP1 的配体 。 该研究于 2025 年 10 月 13 日在国际顶尖学术期刊 Nature 加速上线，论文题为： Host cell Z-RNAs activate ZBP1 during virus infections （ 宿主细胞的 Z- RNA 在病毒感染期间激活 ZBP1 ） 。 这项研究表明，宿主细胞因 病毒感染 导致的 转录终止中断 （ DoTT） 而生成的细胞 Z-RNA 是真正的 ZBP1 配体，并将 ZBP1 激活的细胞死亡定位为宿主细胞 对抗病毒破坏细胞转录的一种防御机制。 四川大学华西医院 张霆 研究员 、美国 Fox Chase 癌症中心 Siddharth Balachandran 、俄罗斯 国家研究型高等经济大学 Alan Herbert 、 德国 维尔茨堡大学 Lars Dölken 为论文共同通讯作者， Yin Chaoran 、 Guo Hongyan 等人为共同第一作者。 撰文丨王聪 编辑丨王多鱼 排版丨水成文 Z-RNA ...

Core Viewpoint - The "AI Augmented Biology" conference aims to explore the integration of artificial intelligence with biological sciences, facilitating groundbreaking discoveries and advancements in the field [4]. Group 1: Conference Details - The conference is organized by Nanjing University, the Chinese Biophysics Society, and several prestigious journals including Nature and Nature Biotechnology [2]. - It will take place from October 22 to 24, 2025, in Nanjing, with both online and offline participation [2]. - The agenda includes sessions on the interface of AI and biomedicine, AI in the multi-omics era, and decoding the language of molecules [5]. Group 2: Topics of Discussion - Key topics will include multimodal data mining, protein engineering, molecular and cellular engineering, and the application of large language models to understand complex biological systems and diseases [4]. - The conference will feature important milestones and breakthroughs in AI technology as applied to biological research [4]. Group 3: Editorial Participation - Editors from the Nature series journals will be present, providing opportunities for attendees to engage with them [5]. - Notable editors include Cheng Qian from Nature Communications and Angela Eggleston from Nature [5][6]. Group 4: Accommodation Information - Recommended hotels include Nanjing Zijin Mountain Villa with room rates starting at 450 RMB for a double room [14]. - Alternative accommodation is available at Jiangsu Conference Center, approximately 20 minutes from the venue, with rates starting at 380 RMB [14].

Core Insights - The article emphasizes the growing interest in stem cell research as a revolutionary hope for human health, transitioning from basic research to clinical applications and regenerative medicine [1][3]. Conference Overview - The "Stem Cell Biology: From Embryonic Development to Aging and Therapy" conference will be held from November 19-21, 2025, at Westlake University, Hangzhou, China [14][26]. - The conference aims to explore various aspects of stem cell biology, including embryogenesis, aging, cell reprogramming, tissue repair, and regenerative medicine [12][16]. Key Themes and Topics - The conference will cover core topics such as organ development, cellular clonality and lineage, tissue repair, and stem cell therapy [16][24]. - It will provide a platform for academic exchange and collaboration among global scientists in the field of stem cell research [12][16]. Notable Speakers - Keynote speakers include prominent researchers from renowned institutions, such as Linda Partridge from University College London and Magdalena Zernicka-Goetz from California Institute of Technology [28][31]. - The organizing committee consists of distinguished members from Westlake University and editors from top life science journals [9][14]. Registration and Participation - Registration fees vary based on the timing and participant category, with early registration discounts available [48]. - Participants can submit academic posters for evaluation, with selected posters displayed during the conference [48].

Core Insights - The article discusses the emergence of AI scientists, particularly the DeepScientist system developed by a team at Westlake University, which has demonstrated the ability to achieve significant scientific advancements in a fraction of the time it takes human scientists [3][4][29] - DeepScientist represents a paradigm shift in scientific discovery, moving from being a mere assistant to a true collaborative partner in research [4][29] Group 1: DeepScientist Overview - DeepScientist is capable of autonomous scientific discovery, completing in two weeks what human scientists achieve in three years [3][15] - The system operates by modeling the entire scientific discovery process as a goal-driven Bayesian optimization problem, focusing on maximizing performance metrics [8] - Its core innovation lies in a three-stage exploration cycle: strategy and hypothesis generation, implementation and validation, and analysis and reporting [9] Group 2: Breakthroughs and Performance - DeepScientist has surpassed state-of-the-art (SOTA) methods in three advanced scientific tasks, achieving significant performance improvements: - A2P method for agent failure attribution improved by 183.7% - LLM inference acceleration increased by 1.9% - AI text detection methods established new SOTA with a 7.9% improvement in AUROC [13][16] - The system's generated research papers achieved a 60% acceptance rate, comparable to human-generated papers, with evaluations indicating high quality in conceptualization [19][20] Group 3: Insights from the Exploration Process - The exploration process of DeepScientist involved generating over 5000 unique ideas, with only about 1100 selected for experimental validation, and ultimately leading to 21 significant scientific advancements [23] - The analysis revealed that intelligent filtering is crucial, as random sampling without a selection strategy resulted in a near-zero success rate [23] Group 4: Scaling and Resource Allocation - A promising scaling trend was observed, where increasing computational resources led to a higher rate of scientific discoveries, indicating a near-linear relationship between allocated resources and valuable scientific findings [25][26] - This suggests that scientific breakthroughs can be systematically produced by scaling computational resources, rather than relying solely on individual genius [26] Group 5: Future Implications - The results from DeepScientist indicate a fundamental shift in AI research, suggesting that the pace of scientific discovery may no longer be solely dictated by human cognition [29] - This advancement positions AI as a capable partner in pushing the boundaries of scientific knowledge, potentially transforming the landscape of research and discovery [29]