This review article, "Linking tumour angiogenesis and tumour immunity" by Kabir et al., explores the intricate relationship between tumor angiogenesis, the formation of new blood vessels within tumors, and tumor immunity, the body's immune response to cancer. The authors discuss how the abnormal structure and function of tumor blood vessels create an immunosuppressive environment that hinders effective anti-tumor responses. They highlight the concept of vessel normalization, where anti-angiogenic therapies aim to restore normal vessel function, thereby improving immune cell infiltration and therapeutic delivery. The article also examines the synergistic potential of combining anti-angiogenic treatments with immune checkpoint blockade (ICB) therapies and identifies promising new targets beyond traditional VEGF pathways to enhance anti-angiogenic strategies and overcome treatment resistance.

References:

• Kabir A U, Subramanian M, Kwon Y, et al. Linking tumour angiogenesis and tumour immunity[J]. Nature Reviews Immunology, 2025: 1-1

This article presents a comprehensive spatial transcriptomic atlas of the human cerebral cortex at single-cell resolution. The researchers utilized advanced sequencing technologies, snRNA-seq and subcellular-level spatial transcriptomics, to map 14 distinct cortical regions from multiple donors. This research revealed a detailed cellular and molecular architecture, characterizing the organization, expression patterns, and spatial arrangements of various neural cell types, including glutamatergic and GABAergic neurons. Furthermore, the study investigated cell-cell interactions, regional differences in neuron-glia communication, and deciphered transcriptomic variations within specific cortical layers like layer 4 and layer 6b, offering crucial insights into human brain function and neurological diseases.

References:

• Wei S, Luo M, Wang P, et al. Charting the spatial transcriptome of the human cerebral cortex at single-cell resolution[J]. Nature Communications, 2025, 16(1): 7702.

This article explores the relationship between cholinergic neuronal activity and the growth of diffuse midline glioma (DMG), a severe brain cancer. The research demonstrates that midbrain cholinergic neurons promote the proliferation of both healthy oligodendrocyte precursor cells (OPCs) and DMG cells in a circuit-dependent manner. This effect is primarily mediated by muscarinic M1 and M3 receptors on DMG cells, which the study identifies as potential therapeutic targets. Furthermore, the article reveals a bidirectional interaction, where DMG cells reciprocally increase the activity of midbrain cholinergic neurons. The authors utilize various experimental techniques, including optogenetic stimulation, co-culture models, and genetic manipulation, to understand these complex cellular interactions.

References:

• Drexler R, Drinnenberg A, Gavish A, et al. Cholinergic neuronal activity promotes diffuse midline glioma growth through muscarinic signaling[J]. Cell, 2025.

This research explores the paradoxical relationship between glucose restriction and cancer metastasis, specifically focusing on lung metastasis. The study reveals that while low-carbohydrate diets or impaired glucose metabolism can suppress primary tumor growth, they unexpectedly promote lung metastasis by altering the immune landscape. This occurs through exosomes containing TRAIL protein, which are released by glucose-deprived cancer cells. These exosomes then polarize lung macrophages, leading to the exhaustion of natural killer (NK) cells via the PVR-TIGIT pathway, thereby creating an environment conducive to tumor colonization. The research also highlights plasma exosomal TRAIL as a superior biomarker for predicting early postoperative lung metastasis in hepatocellular carcinoma (HCC) patients, offering potential for improved diagnostic and therapeutic strategies like anti-TIGIT therapy.

References:

• Wu C Y, Huang C X, Lao X M, et al. Glucose restriction shapes pre-metastatic innate immune landscapes in the lung through exosomal TRAIL[J]. Cell, 2025.

This academic article introduces a systems-level framework to quantify the amount of fermentation products harvested from the human gut microbiota. The research integrates bacterial metabolism, host physiology, and metagenomic data to provide consistent estimates of these metabolic fluxes. A key finding is that for Western diets, bacteria contribute a relatively small percentage (2%-5%) to human energy demand, a revision of previous estimates. The study further reveals that diet is the primary determinant of the total fermentation product harvest, with microbiota composition having a greater impact on the types of products rather than the overall quantity. The authors also highlight significant differences in microbiota-supplied energy between humans and mice, emphasizing the need for careful consideration when translating findings from mouse models to human health.

References:

• Arnoldini M, Sharma R, Moresi C, et al. Quantifying the varying harvest of fermentation products from the human gut microbiota[J]. Cell, 2025.

This article details the identification of gut microbial bile acid metabolic enzymes through an AI-assisted pipeline called BEAUT. The research, conducted by Ding et al., uncovers novel enzymes like MABH, responsible for hydrolyzing specific bile acids, and ADS, which creates an unreported skeleton bile acid, 3-acetoDCA. The paper explores the structural basis and catalytic mechanisms of these enzymes and further investigates 3-acetoDCA's widespread presence in the human gut and its impact on the gut microbiota, particularly its ability to promote the growth of Lactobacillus species. Ultimately, this work offers a top-down approach to discovering microbial enzymes and expands our understanding of the complex relationship between gut microbes and host health.

References:

• Ding Y, Luo X, Guo J, et al. Identification of gut microbial bile acid metabolic enzymes via an AI-assisted pipeline[J]. Cell, 2025.

This article explores the epigenomic and transcriptomic changes in Alzheimer's disease (AD) progression and cognitive resilience, utilizing single-cell multi-region analysis across six brain areas. The authors reveal widespread epigenomic relaxation and information loss in AD, particularly in vulnerable neurons and exhausted glial cells. Conversely, preserved epigenomic stability is shown to correlate with better cognitive function and resilience. The study identifies key genes and regulatory networks associated with these epigenomic dynamics, offering potential insights for therapeutic strategies targeting AD.

References:

• Liu Z, Zhang S, James B T, et al. Single-cell multiregion epigenomic rewiring in Alzheimer’s disease progression and cognitive resilience[J]. Cell, 2025.

This article explores how bone metastases can hinder the effectiveness of immune checkpoint blockade (ICB) therapies in tumors located elsewhere in the body. The research highlights that osteoclasts, cells involved in bone breakdown, are conditioned by bone tumors to produce osteopontin (OPN), a protein that circulates and suppresses anti-tumor T cells in primary tumors. The study demonstrates that targeting these osteoclasts or OPN can successfully restore sensitivity to ICB therapy, suggesting new therapeutic strategies for patients with bone metastases. Clinical data supports these findings, showing that denosumab, a RANKL-blocking antibody, improves ICB outcomes in these patients, unlike bisphosphonates. The article concludes that OPN levels could serve as a biomarker for predicting ICB treatment prognosis in individuals with bone metastasis.

References:

• Cheng J N, Jin Z, Su C, et al. Bone metastases diminish extraosseous response to checkpoint blockade immunotherapy through osteopontin-producing osteoclasts[J]. Cancer Cell, 2025, 43(6): 1093-1107. e9.

This research integrates 249,156 fibroblasts from 73 studies across 10 tissues to construct a comprehensive single-cell atlas of human fibroblasts. The study classifies 18 distinct fibroblast subtypes, including a previously uncharacterized TSPAN8+ chromatin remodeling fibroblast population associated with poor prognosis and specific interactions with endothelial and T cells. It explores the functional heterogeneity, tissue distribution, differentiation trajectories, aging characteristics, and clinical significance of these subtypes. The findings provide a more refined understanding of fibroblast diversity within the tumor microenvironment, including their roles in cancer progression and potential as biomarkers for targeted therapies.

References:

• Liu K, Cui Y, Han H, et al. Fibroblast atlas: Shared and specific cell types across tissues[J]. Science Advances, 2025, 11(14): eado0173.

This paper introduces CellTreeQM, a novel deep learning framework designed to reconstruct cell lineage trees from phenotypic features, such as gene expression profiles. The authors explain how traditional lineage tracing methods are often limited by feasibility or ethical constraints, leading to a need for alternative approaches. CellTreeQM addresses this by formulating lineage reconstruction as a tree-metric learning problem, which optimizes an embedding space for tree-graph inference. The method is benchmarked using synthetic data and lineage-resolved single-cell RNA sequencing datasets from C. elegans, demonstrating its ability to recover lineage structures with minimal supervision and limited data. The paper thoroughly discusses the challenges of phenotype-based lineage reconstruction, including the uncharacterized stochastic processes in gene expression data and the scarcity of ground-truth annotations.

References:

• Kuang D, Qiu G, Kim J. Reconstructing Cell Lineage Trees from Phenotypic Features with Metric Learning[J]. arXiv preprint arXiv:2503.13925, 2025.