Predicting cellular responses to complex perturbations in high-throughput screens.

Recent advances in multiplexed single-cell transcriptomics experiments facilitate the high-throughput study of drug and genetic perturbations. However, an exhaustive exploration of the combinatorial perturbation space is experimentally unfeasible. Therefore, computational methods are needed to predict, interpret, and prioritize perturbations. Here, we present the compositional perturbation autoencoder (CPA), which combines the interpretability of linear models with the flexibility of deep-learning approaches for single-cell response modeling. CPA learns to in silico predict transcriptional perturbation response at the single-cell level for unseen dosages, cell types, time points, and species. Using newly generated single-cell drug combination data, we validate that CPA can predict unseen drug combinations while outperforming baseline models. Additionally, the architecture's modularity enables incorporating the chemical representation of the drugs, allowing the prediction of cellular response to completely unseen drugs. Furthermore, CPA is also applicable to genetic combinatorial screens. We demonstrate this by imputing in silico 5,329 missing combinations (97.6% of all possibilities) in a single-cell Perturb-seq experiment with diverse genetic interactions. We envision CPA will facilitate efficient experimental design and hypothesis generation by enabling in silico response prediction at the single-cell level and thus accelerate therapeutic applications using single-cell technologies.

A New Role for Conivaptan in Ulcerative Colitis in Mice: Inhibiting Differentiation of CD4+T Cells into Th1 Cells.

BACKGROUND: Conivaptan, a nonselective antagonist of vasopressin receptors V1a and V2, is the first drug of this class to be used for treating euvolemic and hypervolemic hyponatremia. Recently, increasing evidence supports the involvement of vasopressin in immune responses. AIMS: In this study, we investigated the effect of conivaptan on the modulation of CD4+ T cell homeostasis and the progression of experimental colitis. METHODS: The expression of the V1a receptor on CD4+ T cells was detected by immunofluorescence and western blot. The subset of isolated CD4+ T cells were examined after arginine vasopressin (AVP) incubation. CD4+ T cells were injected into DNBS-induced mice through the tail vein. The severity of colitis was evaluated according to weight, disease activity index (DAI), and morphological injury. Intracellular Ca2+ ([Ca2+]i) signaling in CD4+ T cells was measured using the Fluo-3 AM loading method. T-bet and IFN-γ mRNAs in the colon were detected by real-time polymerase chain reaction (qPCR). RESULTS: We found that CD4+ T cells expressed the V1a receptor. Activation of the V1a receptor significantly promoted the differentiation of CD4+ T cells into T helper 1 (Th1) cells. This process was blocked by conivaptan treatment. However, the activation of the V1a receptor did not evoke an increase in [Ca2+]i in CD4+ T cells. Notably, conivaptan markedly alleviated body weight loss, pathological damage, and expression of T-bet and IFN-γ in the colon of DNBS-treated mice. CONCLUSIONS: For the first time, we report that conivaptan attenuated colitis by inhibiting the differentiation of CD4+ T cells into Th1 cells. Mechanistically, the anti-inflammatory role of conivaptan is independent of [Ca2+]i.

Single-Exosome Profiling Identifies ITGB3+ and ITGAM+ Exosome Subpopulations as Promising Early Diagnostic Biomarkers and Therapeutic Targets for Colorectal Cancer.

Tumor metastasis is a hallmark of colorectal cancer (CRC), in which exosome plays a crucial role with its function in intercellular communication. Plasma exosomes were collected from healthy control (HC) donors, localized primary CRC and liver-metastatic CRC patients. We performed proximity barcoding assay (PBA) for single-exosome analysis, which enabled us to identify the alteration in exosome subpopulations associated with CRC progression. By in vitro and in vivo experiments, the biological impact of these subpopulations on cancer proliferation, migration, invasion, and metastasis was investigated. The potential application of exosomes as diagnostic biomarkers was evaluated in 2 independent validation cohorts by PBA. Twelve distinct exosome subpopulations were determined. We found 2 distinctly abundant subpopulations: one ITGB3-positive and the other ITGAM-positive. The ITGB3-positive cluster is rich in liver-metastatic CRC, compared to both HC group and primary CRC group. On the contrary, ITGAM-positive exosomes show a large-scale increase in plasma of HC group, compared to both primary CRC and metastatic CRC groups. Notably, both discovery cohort and validation cohort verified ITGB3+ exosomes as potential diagnostic biomarker. ITGB3+ exosomes promote proliferation, migration, and invasion capability of CRC. In contrast, ITGAM+ exosomes suppress CRC development. Moreover, we also provide evidence that one of the sources of ITGAM+ exosomes is macrophage. ITGB3+ exosomes and ITGAM+ exosomes are proven 2 potential diagnostic, prognostic, and therapeutic biomarkers for management of CRC.

An integrated cell atlas of the lung in health and disease.

Single-cell technologies have transformed our understanding of human tissues. Yet, studies typically capture only a limited number of donors and disagree on cell type definitions. Integrating many single-cell datasets can address these limitations of individual studies and capture the variability present in the population. Here we present the integrated Human Lung Cell Atlas (HLCA), combining 49 datasets of the human respiratory system into a single atlas spanning over 2.4 million cells from 486 individuals. The HLCA presents a consensus cell type re-annotation with matching marker genes, including annotations of rare and previously undescribed cell types. Leveraging the number and diversity of individuals in the HLCA, we identify gene modules that are associated with demographic covariates such as age, sex and body mass index, as well as gene modules changing expression along the proximal-to-distal axis of the bronchial tree. Mapping new data to the HLCA enables rapid data annotation and interpretation. Using the HLCA as a reference for the study of disease, we identify shared cell states across multiple lung diseases, including SPP1+ profibrotic monocyte-derived macrophages in COVID-19, pulmonary fibrosis and lung carcinoma. Overall, the HLCA serves as an example for the development and use of large-scale, cross-dataset organ atlases within the Human Cell Atlas.

m6A Regulator-Based Exosomal Gene Methylation Modification Patterns Identify Distinct Microenvironment Characterization and Predict Immunotherapeutic Responses in Colon Cancer.

Recent studies have highlighted the biological significance of exosomes and m6A modifications in immunity. Nonetheless, it remains unclear whether the m6A modification gene in exosomes of body fluid has potential roles in the tumor microenvironment (TME). Herein, we identified three different m6A-related exosomal gene modification patterns based on 59 m6A-related exosomal genes, which instructed distinguishing characteristics of TME in colon cancer (CC). We demonstrated that these patterns could predict the stage of tumor inflammation, subtypes, genetic variation, and patient prognosis. Furthermore, we developed a scoring mode-m6A-related exosomal gene score (MREGS)-by detecting the level of m6A modification in exosomes to classify immune phenotypes. Low MREGS, characterized by prominent survival and immune activation, was linked to a better response to anti-PDL1 immunotherapy. In contrast, the higher MREGS group displayed remarkable stromal activation, high activity of innate immunocytes, and a lower survival rate. Hence, this work provides a novel approach for evaluating TME cell infiltration in colon cancer and guiding more effective immunotherapy strategies.

Machine learning for perturbational single-cell omics.

Cell biology is fundamentally limited in its ability to collect complete data on cellular phenotypes and the wide range of responses to perturbation. Areas such as computer vision and speech recognition have addressed this problem of characterizing unseen or unlabeled conditions with the combined advances of big data, deep learning, and computing resources in the past 5 years. Similarly, recent advances in machine learning approaches enabled by single-cell data start to address prediction tasks in perturbation response modeling. We first define objectives in learning perturbation response in single-cell omics; survey existing approaches, resources, and datasets (https://github.com/theislab/sc-pert); and discuss how a perturbation atlas can enable deep learning models to construct an informative perturbation latent space. We then examine future avenues toward more powerful and explainable modeling using deep neural networks, which enable the integration of disparate information sources and an understanding of heterogeneous, complex, and unseen systems.

The m6A Reader IGF2BP2 Regulates Macrophage Phenotypic Activation and Inflammatory Diseases by Stabilizing TSC1 and PPARγ.

Phenotypic polarization of macrophages is regulated by a milieu of cues in the local tissue microenvironment. Currently, little is known about how the intrinsic regulators modulate proinflammatory (M1) versus prohealing (M2) macrophages activation. Here, it is observed that insulin-like growth factor 2 messenger RNA (mRNA)-binding protein 2 (IGF2BP2)-deleted macrophages exhibit enhanced M1 phenotype and promote dextran sulfate sodium induced colitis development. However, the IGF2BP2-/- macrophages are refractory to interleukin-4 (IL-4) induced activation and alleviate cockroach extract induced pulmonary allergic inflammation. Molecular studies indicate that IGF2BP2 switches M1 macrophages to M2 activation by targeting tuberous sclerosis 1 via an N6-methyladenosine (m6A)-dependent manner. Additionally, it is also shown a signal transducer and activators of transcription 6 (STAT6)-high mobility group AT-hook 2-IGF2BP2-peroxisome proliferator activated receptor-γ axis involves in M2 macrophages differentiation. These findings highlight a key role of IGF2BP2 in regulation of macrophages activation and imply a potential therapeutic target of macrophages in the inflammatory diseases.

Androgen Maintains Intestinal Homeostasis by Inhibiting BMP Signaling via Intestinal Stromal Cells.

Research shows a higher incidence of colorectal cancer in men. However, the molecular mechanisms for this gender disparity remain unknown. We report the roles of androgen in proliferation and differentiation of intestinal stem cells via targeting of the androgen receptor (AR) on intestinal stromal cells by negatively regulating BMP signaling. Orchidectomy (ORX) or the AR antagonist promotes expansion of intestinal epithelium but suppresses intestinal stem cell (ISC) proliferation. Conversely, the AR agonist inhibits ISC differentiation but augments proliferation in ovariectomized mice. Mechanistically, activation of the AR increases expression of BMP antagonists but lowers expression of BMP4 and Wnt antagonists in primary stromal cells, which promotes intestinal organoid growth. Interestingly, the BMP pathway inhibitor LDN-193189 reverses the ORX-induced effects. Our results highlight that stromal cells constitute the intestinal stem cell niche and provide a possible explanation for higher incidence rates of colorectal cancer in men.