Explore the vast range of titles available.

Leptomeningeal disease (LMD) is a devastating complication of solid tumor malignancies, with dire prognosis and no effective systemic treatment options. Over the past decade, the incidence of LMD has steadily increased due to therapeutics that have extended the survival of cancer patients, highlighting the need for new interventions. To examine the efficacy of immune checkpoint inhibitors (ICI) in patients with LMD, we completed two phase II clinical trials. Here, we investigate the cellular and molecular features underpinning observed patient trajectories in these trials by applying single-cell RNA and cell-free DNA profiling to longitudinal cerebrospinal fluid (CSF) draws from enrolled patients. We recover immune and malignant cell types in the CSF, characterize cell behavior changes following ICI, and identify genomic features associated with relevant clinical phenomena. Overall, our study describes the liquid LMD tumor microenvironment prior to and following ICI treatment and demonstrates clinical utility of cell-free and single-cell genomic measurements for LMD research. Leptomeningeal disease (LMD) is a serious complication of metastatic solid tumors with a poor prognosis. Here, by using single-cell RNA sequencing of cerebrospinal fluid, the authors report genomic and immune correlates of response to immunotherapy in two cohorts of patients with LMD treated with immune checkpoint inhibitors.

Many patients infected with coronaviruses, such as SARS-CoV-2 and NL63 that use ACE2 receptors to infect cells, exhibit gastrointestinal symptoms and viral proteins are found in the human gastrointestinal tract, yet little is known about the inflammatory and pathological effects of coronavirus infection on the human intestine. Here, we used a human intestine-on-a-chip (Intestine Chip) microfluidic culture device lined by patient organoid-derived intestinal epithelium interfaced with human vascular endothelium to study host cellular and inflammatory responses to infection with NL63 coronavirus. These organoid-derived intestinal epithelial cells dramatically increased their ACE2 protein levels when cultured under flow in the presence of peristalsis-like mechanical deformations in the Intestine Chips compared to when cultured statically as organoids or in Transwell inserts. Infection of the intestinal epithelium with NL63 on-chip led to inflammation of the endothelium as demonstrated by loss of barrier function, increased cytokine production, and recruitment of circulating peripheral blood mononuclear cells (PBMCs). Treatment of NL63 infected chips with the approved protease inhibitor drug, nafamostat, inhibited viral entry and resulted in a reduction in both viral load and cytokine secretion, whereas remdesivir, one of the few drugs approved for COVID19 patients, was not found to be effective and it also was toxic to the endothelium. This model of intestinal infection was also used to test the effects of other drugs that have been proposed for potential repurposing against SARS-CoV-2. Taken together, these data suggest that the human Intestine Chip might be useful as a human preclinical model for studying coronavirus related pathology as well as for testing of potential anti-viral or anti-inflammatory therapeutics.

A cell’s phenotype and function are influenced by dynamic interactions with its microenvironment. To examine cellular spatiotemporal activity, we developed SPACECAT—Spatially PhotoActivatable Color Encoded Cell Address Tags—to annotate, track, and isolate cells while preserving viability. In SPACECAT, samples are stained with photocaged fluorescent molecules, and cells are labeled by uncaging those molecules with user-patterned near-UV light. SPACECAT offers single-cell precision and temporal stability across diverse cell and tissue types. Illustratively, we target crypt-like regions in patient-derived intestinal organoids to enrich for stem-like and actively mitotic cells, matching literature expectations. Moreover, we apply SPACECAT to ex vivo tissue sections from four healthy organs and an autochthonous lung tumor model. Lastly, we provide a computational framework to identify spatially-biased transcriptome patterns and enriched phenotypes. This minimally perturbative and broadly applicable method links cellular spatiotemporal and/or behavioral phenotypes with diverse downstream assays, enabling insights into the connections between tissue microenvironments and (dys)function. Spatial transcriptomics aims to pair omic data with tissue structure. Here the authors report Spatially PhotoActivatable Colour Encoded Cell Address Tags (SPACECAT) to track and isolate live cells by location; this enables spatially informed downstream assays like scRNA-seq and flow cytometry.

Circulating tumor cells (CTCs) play a fundamental role in cancer progression. However, in mice, limited blood volume and the rarity of CTCs in the bloodstream preclude longitudinal, in-depth studies of these cells using existing liquid biopsy techniques. Here, we present an optofluidic system that continuously collects fluorescently labeled CTCs from a genetically engineered mouse model (GEMM) for several hours per day over multiple days or weeks. The system is based on a microfluidic cell sorting chip connected serially to an unanesthetized mouse via an implanted arteriovenous shunt. Pneumatically controlled microfluidic valves capture CTCs as they flow through the device, and CTC-depleted blood is returned back to the mouse via the shunt. To demonstrate the utility of our system, we profile CTCs isolated longitudinally from animals over 4 days of treatment with the BET inhibitor JQ1 using single-cell RNA sequencing (scRNA-Seq) and show that our approach eliminates potential biases driven by intermouse heterogeneity that can occur when CTCs are collected across different mice. The CTC isolation and sorting technology presented here provides a research tool to help reveal details of how CTCs evolve over time, allowing studies to credential changes in CTCs as biomarkers of drug response and facilitating future studies to understand the role of CTCs in metastasis.

Viral variant and host vaccination status impact infection with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), yet how these factors shift cellular responses in the human nasal mucosa remains uncharacterized. We performed single-cell RNA sequencing (scRNA-seq) on nasopharyngeal swabs from vaccinated and unvaccinated adults with acute Delta and Omicron SARS-CoV-2 infections and integrated with data from acute infections with ancestral SARS-CoV-2. Patients with Delta and Omicron exhibited greater similarity in nasal cell composition driven by myeloid, T cell and SARS-CoV-2 hi cell subsets, which was distinct from that of ancestral cases. Delta-infected samples had a marked increase in viral RNA, and a subset of PER2 + EGR1 + GDF15 + epithelial cells was enriched in SARS-CoV-2 RNA + cells in all variants. Prior vaccination was associated with increased frequency and activation of nasal macrophages. Expression of interferon-stimulated genes negatively correlated with coronavirus disease 2019 (COVID-19) severity in patients with ancestral and Delta but not Omicron variants. Our study defines nasal cell responses and signatures of disease severity across SARS-CoV-2 variants and vaccination. Ordovas-Montanes and colleagues describe the composition of the nasal cellular ecosystem and signatures of disease severity in vaccinated and unvaccinated adults during infection with the ancestral, Delta and Omicron variants of SARS-CoV-2.

The success of transformer-based foundation models on natural language and images has motivated their use in single-cell biology. Single-cell foundation models have been trained on increasingly larger transcriptomic datasets, scaling from initial studies with 1 million cells to newer atlases with over 100 million cells. This study investigates the role of pre-training dataset size and diversity on the performance of single-cell foundation models on both zero-shot and fine-tuned tasks. Using a large corpus of 22.2 million cells, we pre-train a total of 400 models, which we evaluate by conducting 6,400 experiments. Our results show that current methods tend to plateau in performance with pre-training datasets that are only a fraction of the size of current training corpora.

Clustering is commonly used in single-cell RNA-sequencing (scRNA-seq) pipelines to characterize cellular heterogeneity. However, current methods face two main limitations. First, they require user-specified heuristics which add time and complexity to bioinformatic workflows; second, they rely on post-selective differential expression analyses to identify marker genes driving cluster differences, which has been shown to be subject to inflated false discovery rates. We address these challenges by introducing nonparametric clustering of single-cell populations (NCLUSION): an infinite mixture model that leverages Bayesian sparse priors to identify marker genes while simultaneously performing clustering on single-cell expression data. NCLUSION uses a scalable variational inference algorithm to perform these analyses on datasets with up to millions of cells. Through simulations and analyses of publicly available scRNA-seq studies, we demonstrate that NCLUSION (i) matches the performance of other state-of-the-art clustering techniques with significantly reduced runtime and (ii) provides statistically robust and biologically relevant transcriptomic signatures for each of the clusters it identifies. Overall, NCLUSION represents a reliable hypothesis-generating tool for understanding patterns of expression variation present in single-cell populations.

Foundation models for single-cell transcriptomics have the potential to augment (or replace) purpose-built tools for a variety of common analyses, especially when data are sparse. Recent work with large language models has shown that training data composition greatly shapes performance; however, to date, single-cell foundation models have ignored this aspect, opting instead to train on the largest possible corpus. We systematically investigate the consequences of training dataset composition on the behavior of deep learning models of single-cell transcriptomics, focusing on human hematopoiesis as a tractable model system and including cells from adult and developing tissues, disease states, and perturbation atlases. We find that (1) these models generalize poorly to unseen cell types, (2) adding malignant cells to a healthy cell training corpus does not necessarily improve modeling of unseen malignant cells, and (3) including an embryonic stem cell differentiation atlas during training improves performance on out-of-distribution tasks. Our results emphasize the importance of diverse training data and suggest strategies to optimize future single-cell foundation models.