Search

It is currently accepted that cancer-associated fibroblasts (CAF) participate in T-cell exclusion from tumor nests. To unbiasedly test this, we used single-cell RNA sequencing coupled with multiplex imaging on a large cohort of lung tumors. We identified four main CAF populations, two of which are associated with T-cell exclusion: (i) MYH11+αSMA+ CAF, which are present in early-stage tumors and form a single cell layer lining cancer aggregates, and (ii) FAP+αSMA+ CAF, which appear in more advanced tumors and organize in patches within the stroma or in multiple layers around tumor nests. Both populations orchestrate a particular structural tissue organization through dense and aligned fiber deposition compared with T cell–permissive CAF. Yet they produce distinct matrix molecules, including collagen IV (MYH11+αSMA+ CAF) and collagen XI/XII (FAP+αSMA+ CAF). Hereby, we uncovered unique molecular programs of CAF driving T-cell marginalization, whose targeting should increase immunotherapy efficacy in patients bearing T cell–excluded tumors. Significance: The cellular and molecular programs driving T-cell marginalization in solid tumors remain unclear. Here, we describe two CAF populations associated with T-cell exclusion in human lung tumors. We demonstrate the importance of pairing molecular and spatial analysis of the tumor microenvironment, a prerequisite to developing new strategies targeting T cell–excluding CAF. See related commentary by Sherman, p. 2501. This article is highlighted in the In This Issue feature, p. 2483

Supplementary Figure from Spatial Positioning and Matrix Programs of Cancer-Associated Fibroblasts Promote T-cell Exclusion in Human Lung Tumors

Supplementary Table from Spatial Positioning and Matrix Programs of Cancer-Associated Fibroblasts Promote T-cell Exclusion in Human Lung Tumors

It is currently accepted that cancer-associated fibroblasts (CAF) participate in T-cell exclusion from tumor nests. To unbiasedly test this, we used single-cell RNA sequencing coupled with multiplex imaging on a large cohort of lung tumors. We identified four main CAF populations, two of which are associated with T-cell exclusion: (i) MYH11+αSMA+ CAF, which are present in early-stage tumors and form a single cell layer lining cancer aggregates, and (ii) FAP+αSMA+ CAF, which appear in more advanced tumors and organize in patches within the stroma or in multiple layers around tumor nests. Both populations orchestrate a particular structural tissue organization through dense and aligned fiber deposition compared with T cell–permissive CAF. Yet they produce distinct matrix molecules, including collagen IV (MYH11+αSMA+ CAF) and collagen XI/XII (FAP+αSMA+ CAF). Hereby, we uncovered unique molecular programs of CAF driving T-cell marginalization, whose targeting should increase immunotherapy efficacy in patients bearing T cell–excluded tumors.Significance:The cellular and molecular programs driving T-cell marginalization in solid tumors remain unclear. Here, we describe two CAF populations associated with T-cell exclusion in human lung tumors. We demonstrate the importance of pairing molecular and spatial analysis of the tumor microenvironment, a prerequisite to developing new strategies targeting T cell–excluding CAF.See related commentary by Sherman, p. 2501.This article is highlighted in the In This Issue feature, p. 2483

B cells, which are critical for intestinal homeostasis, remain understudied in ulcerative colitis (UC). In this study, we recruited three cohorts of patients with UC (primary cohort, n = 145; validation cohort 1, n = 664; and validation cohort 2, n = 143) to comprehensively define the landscape of B cells during UC-associated intestinal inflammation. Using single-cell RNA sequencing, single-cell IgH gene sequencing and protein-level validation, we mapped the compositional, transcriptional and clonotypic landscape of mucosal and circulating B cells. We found major perturbations within the mucosal B cell compartment, including an expansion of naive B cells and IgG(+) plasma cells with curtailed diversity and maturation. Furthermore, we isolated an auto-reactive plasma cell clone targeting integrin αvβ6 from inflamed UC intestines. We also identified a subset of intestinal CXCL13-expressing TFH-like T peripheral helper cells that were associated with the pathogenic B cell response. Finally, across all three cohorts, we confirmed that changes in intestinal humoral immunity are reflected in circulation by the expansion of gut-homing plasmablasts that correlates with disease activity and predicts disease complications. Our data demonstrate a highly dysregulated B cell response in UC and highlight a potential role of B cells in disease pathogenesis.

Here, we leveraged a large neoadjuvant PD-1 blockade trial in patients with hepatocellular carcinoma (HCC) to search for correlates of response to immune checkpoint blockade (ICB) within T cell-rich tumors. We show that ICB response correlated with the clonal expansion of intratumoral CXCL13+ CH25H+ IL-21+ PD-1+ CD4 T helper cells (CXCL13+ Th) and Granzyme K+ PD-1+ effector-like CD8 T cells, whereas terminally exhausted CD39hi TOXhi PD-1hi CD8 T cells dominated in non-responders. Strikingly, most T cell receptor (TCR) clones that expanded post-treatment were found in pre-treatment biopsies. Notably, PD-1+ TCF-1+ progenitor-like CD8 T cells were present in tumors of responders and non-responders and shared clones mainly with effector-like cells in responders or terminally differentiated cells in non-responders, suggesting that local CD8 T cell differentiation occurs upon ICB. We found that these progenitor CD8 T cells interact with CXCL13+ Th cells within cellular triads around dendritic cells enriched in maturation and regulatory molecules, or “mregDC”. Receptor-ligand analysis revealed unique interactions within these triads that may promote the differentiation of progenitor CD8 T cells into effector-like cells upon ICB. These results suggest that discrete intratumoral niches that include mregDC and CXCL13+ Th cells control the differentiation of tumor-specific progenitor CD8 T cell clones in patients treated with ICB.

Increasing evidence shows that cancer cells can disseminate from early evolved primary lesions much earlier than the classical metastasis models predicted. Here, we reveal at a single-cell resolution that mesenchymal-like (M-like) and pluripotency-like programs coordinate dissemination and a long-lived dormancy program of early disseminated cancer cells (DCCs). The transcription factor ZFP281 induces a permissive state for heterogeneous M-like transcriptional programs, which associate with a dormancy signature and phenotype in vivo. Downregulation of ZFP281 leads to a loss of an invasive, M-like dormancy phenotype and a switch to lung metastatic outgrowth. We also show that FGF2 and TWIST1 induce ZFP281 expression to induce the M-like state, which is linked to CDH1 downregulation and upregulation of CDH11. We found that ZFP281 not only controls the early dissemination of cancer cells but also locks early DCCs in a dormant state by preventing the acquisition of an epithelial-like proliferative program and consequent metastases outgrowth.

Checkpoint blockade therapies have improved cancer treatment, but such immunotherapy regimens fail in a large subset of patients. Conventional type 1 dendritic cells (DC1s) control the response to checkpoint blockade in preclinical models and are associated with better overall survival in patients with cancer, reflecting the specialized ability of these cells to prime the responses of CD8+ T cells1–3. Paradoxically, however, DC1s can be found in tumours that resist checkpoint blockade, suggesting that the functions of these cells may be altered in some lesions. Here, using single-cell RNA sequencing in human and mouse non-small-cell lung cancers, we identify a cluster of dendritic cells (DCs) that we name ‘mature DCs enriched in immunoregulatory molecules’ (mregDCs), owing to their coexpression of immunoregulatory genes (Cd274, Pdcd1lg2 and Cd200) and maturation genes (Cd40, Ccr7 and Il12b). We find that the mregDC program is expressed by canonical DC1s and DC2s upon uptake of tumour antigens. We further find that upregulation of the programmed death ligand 1 protein—a key checkpoint molecule—in mregDCs is induced by the receptor tyrosine kinase AXL, while upregulation of interleukin (IL)-12 depends strictly on interferon-γ and is controlled negatively by IL-4 signalling. Blocking IL-4 enhances IL-12 production by tumour-antigen-bearing mregDC1s, expands the pool of tumour-infiltrating effector T cells and reduces tumour burden. We have therefore uncovered a regulatory module associated with tumour-antigen uptake that reduces DC1 functionality in human and mouse cancers. After taking up tumour-associated antigens, dendritic cells in mouse and human tumours upregulate a regulatory gene program that limits dendritic cell immunostimulatory function, and modulating this program can rescue antitumor immunity in mice.

Background Acute kidney injury (AKI) is a known complication of COVID-19 and is associated with an increased risk of in-hospital mortality. Unbiased proteomics using biological specimens can lead to improved risk stratification and discover pathophysiological mechanisms. Methods Using measurements of ~4000 plasma proteins in two cohorts of patients hospitalized with COVID-19, we discovered and validated markers of COVID-associated AKI (stage 2 or 3) and long-term kidney dysfunction. In the discovery cohort (N= 437), we identified 413 higher plasma abundances of protein targets and 40 lower plasma abundances of protein targets associated with COVID-AKI (adjusted p Results We demonstrate that COVID-AKI is associated with increased markers of tubular injury (NGAL) and myocardial injury. Using estimated glomerular filtration (eGFR) measurements taken after discharge, we also find that 25 of the 62 AKI-associated proteins are significantly associated with decreased post-discharge eGFR (adjusted p desmocollin-2, trefoil factor 3, transmembrane emp24 domain-containing protein 10, and cystatin-Cindicating tubular dysfunction and injury. Conclusions Using clinical and proteomic data, our results suggest that while both acute and long-term COVID-associated kidney dysfunction are associated with markers of tubular dysfunction, AKI is driven by a largely multifactorial process involving hemodynamic instability and myocardial damage.

Predicting COVID-19 severity is difficult, and the biological pathways involved are not fully understood. To approach this problem, we measured 4,701 circulating human protein abundances in two independent cohorts totaling 986 individuals. We then trained prediction models including protein abundances and clinical risk factors to predict adverse COVID-19 outcomes in 417 subjects and tested these models in a separate cohort of 569 individuals. For severe COVID-19, a baseline model including age and sex provided an area under the receiver operator curve (AUC) of 65% in the test cohort. Selecting 92 proteins from the 4,701 unique protein abundances improved the AUC to 88% in the training cohort, which remained relatively stable in the testing cohort at 86%, suggesting good generalizability. Proteins selected from different adverse COVID-19 outcomes were enriched for cytokine and cytokine receptors, but more than half of the enriched pathways were not immune-related. Taken together, these findings suggest that circulating proteins measured at early stages of disease progression are reasonably accurate predictors of adverse COVID-19 outcomes. Further research is needed to understand how to incorporate protein measurement into clinical care.

Background & AimsAnti–GM-CSF autoantibodies (aGMAb) are detected in ileal Crohn’s Disease (CD) patients. Their induction and mode of action impacting homeostasis during, or prior to disease are not well understood. We aimed to investigate the underlying mechanisms leading to the induction of aGMAb, from functional orientation to recognized epitopes, for their impact on intestinal immune homeostasis and use as predictive biomarker for complicated CD.MethodsUsing longitudinally collected sera from active component US personnel, we characterize naturally occurring aGMAb in a subset of CD patients years before disease onset. We employed biochemical, cellular, and transcriptional analysis to uncover a mechanism that governs the impaired immune balance in CD years prior to diagnosis.ResultsNeutralizing aGMAb are specific to posttranslational glycosylations on GM-CSF, detectable years prior to diagnosis, and associated with complicated CD at presentation. Glycosylation and production of GM-CSF change in CD patients, altering myeloid homeostasis and destabilizing group 3 innate lymphoid cells. Perturbations in immune homeostasis precede the inflammation and are detectable in the non-inflamed CD mucosa of patients presenting with anti-GM-CSF autoantibodies.ConclusionsAnti-GM-CSF autoantibodies predict the diagnosis of complicated CD, have unique epitopes, and impair myeloid cell homeostasis across the ILC3-GM-CSF-myeloid cell axis, altering intestinal immune homeostasis long before the diagnosis of disease.

Anti-granulocyte macrophage-colony stimulating factor autoantibodies (aGMAbs) are detected in patients with ileal Crohn's disease (CD). Their induction and mode of action during or before disease are not well understood. We aimed to investigate the underlying mechanisms associated with aGMAb induction, from functional orientation to recognized epitopes, for their impact on intestinal immune homeostasis and use as a predictive biomarker for complicated CD.We characterized using enzyme-linked immunosorbent assay naturally occurring aGMAbs in longitudinal serum samples from patients archived before the diagnosis of CD (n = 220) as well as from 400 healthy individuals (matched controls) as part of the US Defense Medical Surveillance System. We used biochemical, cellular, and transcriptional analysis to uncover a mechanism that governs the impaired immune balance in CD mucosa after diagnosis.Neutralizing aGMAbs were found to be specific for post-translational glycosylation on granulocyte macrophage-colony stimulating factor (GM-CSF), detectable years before diagnosis, and associated with complicated CD at presentation. Glycosylation of GM-CSF was altered in patients with CD, and aGMAb affected myeloid homeostasis and promoted group 1 innate lymphoid cells. Perturbations in immune homeostasis preceded the diagnosis in the serum of patients with CD presenting with aGMAb and were detectable in the noninflamed CD mucosa.Anti-GMAbs predict the diagnosis of complicated CD long before the diagnosis of disease, recognize uniquely glycosylated epitopes, and impair myeloid cell and innate lymphoid cell balance associated with altered intestinal immune homeostasis.

Macrophages have a key role in shaping the tumour microenvironment (TME), tumour immunity and response to immunotherapy, which makes them an important target for cancer treatment1,2. However, modulating macrophages has proved extremely difficult, as we still lack a complete understanding of the molecular and functional diversity of the tumour macrophage compartment. Macrophages arise from two distinct lineages. Tissue-resident macrophages self-renew locally, independent of adult haematopoiesis3–5, whereas short-lived monocyte-derived macrophages arise from adult haematopoietic stem cells, and accumulate mostly in inflamed lesions1. How these macrophage lineages contribute to the TME and cancer progression remains unclear. To explore the diversity of the macrophage compartment in human non-small cell lung carcinoma (NSCLC) lesions, here we performed single-cell RNA sequencing of tumour-associated leukocytes. We identified distinct populations of macrophages that were enriched in human and mouse lung tumours. Using lineage tracing, we discovered that these macrophage populations differ in origin and have a distinct temporal and spatial distribution in the TME. Tissue-resident macrophages accumulate close to tumour cells early during tumour formation to promote epithelial–mesenchymal transition and invasiveness in tumour cells, and they also induce a potent regulatory T cell response that protects tumour cells from adaptive immunity. Depletion of tissue-resident macrophages reduced the numbers and altered the phenotype of regulatory T cells, promoted the accumulation of CD8+ T cells and reduced tumour invasiveness and growth. During tumour growth, tissue-resident macrophages became redistributed at the periphery of the TME, which becomes dominated by monocyte-derived macrophages in both mouse and human NSCLC. This study identifies the contribution of tissue-resident macrophages to early lung cancer and establishes them as a target for the prevention and treatment of early lung cancer lesions. Single-cell RNA sequencing and imaging of macrophages in human non-small cell lung cancer and in a mouse model of lung adenocarcinoma show that tissue-resident macrophages have a key role in early tumour progression.

Human tissues comprise trillions of cells that populate a complex space of molecular phenotypes and functions and that vary in abundance by 4-9 orders of magnitude. Relying solely on unbiased sampling to characterize cellular niches becomes infeasible, as the marginal utility of collecting more cells diminishes quickly. Furthermore, in many clinical samples, the relevant cell types are scarce and efficient processing is critical. We developed an integrated pipeline for index sorting and massively parallel single-cell RNA sequencing (MARS-seq2.0) that builds on our previously published MARS-seq approach. MARS-seq2.0 is based on >1 million cells sequenced with this pipeline and allows identification of unique cell types across different tissues and diseases, as well as unique model systems and organisms. Here, we present a detailed step-by-step procedure for applying the method. In the improved procedure, we combine sub-microliter reaction volumes, optimization of enzymatic mixtures and an enhanced analytical pipeline to substantially lower the cost, improve reproducibility and reduce well-to-well contamination. Data analysis combines multiple layers of quality assessment and error detection and correction, graphically presenting key statistics for library complexity, noise distribution and sequencing saturation. Importantly, our combined FACS and single-cell RNA sequencing (scRNA-seq) workflow enables intuitive approaches for depletion or enrichment of cell populations in a data-driven manner that is essential to efficient sampling of complex tissues. The experimental protocol, from cell sorting to a ready-to-sequence library, takes 2-3 d. Sequencing and processing the data through the analytical pipeline take another 1-2 d.

Genome-wide association studies have identified ~170 loci associated with Crohn's disease (CD) and defining which genes drive these association signals is a major challenge. The primary aim of this study was to define which CD locus genes are most likely to be disease related. We developed a gene prioritization regression model (GPRM) by integrating complementary mRNA expression datasets, including bulk RNA-Seq from the terminal ileum of 302 newly diagnosed, untreated CD patients and controls, and in stimulated monocytes. Transcriptome-wide association and co-expression network analyses were performed on the ileal RNA-Seq datasets, identifying 40 genome-wide significant genes. Co-expression network analysis identified a single gene module, which was substantially enriched for CD locus genes and most highly expressed in monocytes. By including expression-based and epigenetic information, we refined likely CD genes to 2.5 prioritized genes per locus from an average of 7.8 total genes. We validated our model structure using cross-validation and our prioritization results by protein-association network analyses, which demonstrated significantly higher CD gene interactions for prioritized compared with non-prioritized genes. Although individual datasets cannot convey all of the information relevant to a disease, combining data from multiple relevant expression-based datasets improves prediction of disease genes and helps to further understanding of disease pathogenesis.

Crohn's disease is a chronic inflammatory intestinal disease that is frequently accompanied by aberrant healing and stricturing complications. Crosstalk between activated myeloid and stromal cells is critical in the pathogenicity of Crohn's disease1,2, and increases in intravasating monocytes are correlated with a lack of response to anti-TNF treatment3. The risk alleles with the highest effect on Crohn's disease are loss-of-function mutations in NOD24,5, which increase the risk of stricturing6. However, the mechanisms that underlie pathogenicity driven by NOD2 mutations and the pathways that might rescue a lack of response to anti-TNF treatment remain largely uncharacterized. Here we use direct ex vivo analyses of patients who carry risk alleles of NOD2 to show that loss of NOD2 leads to dysregulated homeostasis of activated fibroblasts and macrophages. CD14+ peripheral blood mononuclear cells from carriers of NOD2 risk alleles produce cells that express high levels of collagen, and elevation of conserved signatures is observed in nod2-deficient zebrafish models of intestinal injury. The enrichment of STAT3 regulation and gp130 ligands in activated fibroblasts and macrophages suggested that gp130 blockade might rescue the activated program in NOD2-deficient cells. We show that post-treatment induction of the STAT3 pathway is correlated with a lack of response to anti-TNF treatment in patients, and demonstrate in vivo in zebrafish the amelioration of the activated myeloid-stromal niche using the specific gp130 inhibitor bazedoxifene. Our results provide insights into NOD2-driven fibrosis in Crohn's disease, and suggest that gp130 blockade may benefit some patients with Crohn's disease-potentially as a complement to anti-TNF therapy.

Increasing evidence shows that cancer cells can disseminate from early-evolved primary lesions much earlier than the classical metastasis models predicted. It is thought that a state of early disseminated cancer cell (early DCC) dormancy can precede genetic maturation of DCCs and metastasis initiation. Here we reveal at single cell resolution a previously unrecognized role of mesenchymal- and pluripotency-like programs in coordinating early cancer cell spread and a long-lived dormancy program in early DCCs. Using in vitro and in vivo models of invasion and metastasis, single cell RNA sequencing and human sample analysis, we provide unprecedented insight into how early DCC heterogeneity and plasticity control the timing of reactivation. We identify in early lesions and early DCCs the transcription factor ZFP281 as an inducer of mesenchymal- and primed pluripotency-like programs, which is absent in advanced primary tumors and overt metastasis. ZFP281 not only controls the early spread of cancer cells but also locks early DCCs in a prolonged dormancy state by preventing the acquisition of an epithelial-like proliferative program and consequent metastasis outgrowth. Thus, ZFP281-driven dormancy of early DCCs may be a rate-limiting step in metastatic progression functioning as a first barrier that DCCs must overcome to then undergo genetic maturation.

Macrophages (MΦ) form a highly heterogenous compartment within tumor lesions. Using lineage tracing, we recently established that advanced NSCLC lesions in both mice and humans are depleted of tissue-resident macrophages (TRM) and are mainly populated by monocyte-derived MΦ (MoMΦ). Here, we show that TREM2+ MoMΦ dominate murine and human NSCLC lesions. Importantly, using mice reconstituted with mixed TREM2-deficient and -proficient bone marrow cells, we identified a cell-intrinsic ability for TREM2 to drive an efferocytosis-associated molecular program that profoundly reduced the expression of immunogenic molecules in tumor-infiltrating MoMΦ. Trem2 deletion significantly dampened this program, reversed MoMΦ immunosuppression, and significantly enhanced the accumulation of activated NK cells and cDC1 in tumor lesions. Notably, Trem2 deletion reduced lung cancer progression in an NK cell-dependent manner. Taken together, our results underscore the potential for therapeutic inhibition of TREM2 to mount a potent anti-tumor immune response by unleashing cooperative NK/cDC1 cell function.

Given that gastrointestinal (GI) symptoms are a prominent extrapulmonary manifestation of coronavirus disease 2019 (COVID-19), we investigated the impact of GI infection on disease pathogenesis in three large cohorts of patients in the United States and Europe. Unexpectedly, we observed that GI involvement was associated with a significant reduction in disease severity and mortality, with an accompanying reduction in key inflammatory proteins including IL-6, CXCL8, IL-17A and CCL28 in circulation. In a fourth cohort of COVID-19 patients in which GI biopsies were obtained, we identified severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) within small intestinal enterocytes for the first timein vivobut failed to obtain culturable virus. High dimensional analyses of GI tissues confirmed low levels of cellular inflammation in the GI lamina propria and an active downregulation of key inflammatory genes includingIFNG, CXCL8, CXCL2andIL1Bamong others. These data draw attention to organ-level heterogeneity in disease pathogenesis and highlight the role of the GI tract in attenuating SARS-CoV-2-associated inflammation with related mortality benefit.

SUMMARYImmunotherapy is becoming a mainstay in the treatment of NSCLC. While tumor mutational burden (TMB) has been shown to correlate with response to immunotherapy, little is known about the relation of the baseline immune response with the tumor genotype. Here, we profiled 35 early stage NSCLC lesions using multiscale single cell sequencing. Unsupervised clustering identified in a subset of patients a key cellular module consisting of PDCD1+ CXCL13+ activated T cells, IgG+ plasma cells, and SPP1+ macrophages, referred to as the lung cancer activation module (LCAMhi). Transcriptional data from two NSCLC cohorts confirmed a subset of patients with LCAMhi enrichment, which was independent of overall immune cell content. The LCAMhi module strongly correlated with TMB, expression of cancer testis antigens, and with TP53 mutations in smokers and non-smokers. These data establish LCAM as a key mode of immune cell activation associated with high tumor antigen load and driver mutations.

The coronavirus disease 2019 (COVID-19) pandemic, caused by Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) has affected millions of people worldwide, igniting an unprecedented effort from the scientific community to understand the biological underpinning of COVID19 pathophysiology. In this review, we summarize the current state of knowledge of innate and adaptive immune responses elicited by SARS-CoV-2 infection and the immunological pathways that likely contribute to disease severity and death. We also discuss the rationale and clinical outcome of current therapeutic strategies as well as prospective clinical trials to prevent or treat SARS-CoV-2 infection.

Background: Sphingolipids have recently emerged as a biomarker of recurrence and mortality after myocardial infarction (MI). The increased ceramide levels in mammalian heart tissues during acute MI, as demonstrated by several groups, is associated with higher cell death rates in the left ventricle and deteriorated cardiac function. Ceramidase, the only enzyme known to hydrolyze proapoptotic ceramide, generates sphingosine, which is then phosphorylated by sphingosine kinase to produce the prosurvival molecule sphingosine-1-phosphate. We hypothesized that Acid Ceramidase (AC) overexpression would counteract the negative effects of elevated ceramide and promote cell survival, thereby providing cardioprotection after MI. Methods: We performed transcriptomic, sphingolipid, and protein analyses to evaluate sphingolipid metabolism and signaling post-MI. We investigated the effect of altering ceramide metabolism through a loss (chemical inhibitors) or gain (modified mRNA [modRNA]) of AC function post hypoxia or MI. Results: We found that several genes involved in de novo ceramide synthesis were upregulated and that ceramide (C16, C20, C20:1, and C24) levels had significantly increased 24 hours after MI. AC inhibition after hypoxia or MI resulted in reduced AC activity and increased cell death. By contrast, enhancing AC activity via AC modRNA treatment increased cell survival after hypoxia or MI. AC modRNA-treated mice had significantly better heart function, longer survival, and smaller scar size than control mice 28 days post-MI. We attributed the improvement in heart function post-MI after AC modRNA delivery to decreased ceramide levels, lower cell death rates, and changes in the composition of the immune cell population in the left ventricle manifested by lowered abundance of proinflammatory detrimental neutrophils. Conclusions: Our findings suggest that transiently altering sphingolipid metabolism through AC overexpression is sufficient and necessary to induce cardioprotection post-MI, thereby highlighting the therapeutic potential of AC modRNA in ischemic heart disease.

Mutations in genes involved in DNA methylation (DNAme; for example, TET2 and DNMT3A) are frequently observed in hematological malignancies1–3 and clonal hematopoiesis4,5. Applying single-cell sequencing to murine hematopoietic stem and progenitor cells, we observed that these mutations disrupt hematopoietic differentiation, causing opposite shifts in the frequencies of erythroid versus myelomonocytic progenitors following Tet2 or Dnmt3a loss. Notably, these shifts trace back to transcriptional priming skews in uncommitted hematopoietic stem cells. To reconcile genome-wide DNAme changes with specific erythroid versus myelomonocytic skews, we provide evidence in support of differential sensitivity of transcription factors due to biases in CpG enrichment in their binding motif. Single-cell transcriptomes with targeted genotyping showed similar skews in transcriptional priming of DNMT3A-mutated human clonal hematopoiesis bone marrow progenitors. These data show that DNAme shapes the topography of hematopoietic differentiation, and support a model in which genome-wide methylation changes are transduced to differentiation skews through biases in CpG enrichment of the transcription factor binding motif. Single-cell analysis of mouse hematopoietic stem cells shows that mutations in DNA methylation genes change the frequencies of erythroid versus myelomonocytic progenitors, owing to differential CpG enrichment in transcription factor binding motifs.

Immunotherapy is a mainstay of non-small cell lung cancer (NSCLC) management. While tumor mutational burden (TMB) correlates with response to immunotherapy, little is known about the relationship between the baseline immune response and tumor genotype. Using single-cell RNA sequencing, we profiled 361,929 cells from 35 early-stage NSCLC lesions. We identified a cellular module consisting of PDCD1(+)CXCL13(+) activated T cells, IgG(+) plasma cells, and SPP1(+) macrophages, referred to as the lung cancer activation module (LCAM(hi)). We confirmed LCAM(hi) enrichment in multiple NSCLC cohorts, and paired CITE-seq established an antibody panel to identify LCAM(hi) lesions. LCAM presence was found to be independent of overall immune cell content and correlated with TMB, cancer testis antigens, and TP53 mutations. High baseline LCAM scores correlated with enhanced NSCLC response to immunotherapy even in patients with above median TMB, suggesting that immune cell composition, while correlated with TMB, may be a nonredundant biomarker of response to immunotherapy.

BackgroundHepatocellular carcinoma (HCC) has a dismal prognosis, and though checkpoint blocking antibodies have significantly improved patient outcome, many patients remain left out, highlighting the need to identify additional immune target to enhance therapeutic immunity. Macrophages (MF) are an abundant and heterogeneous population in the tumor microenvironment (TME), and are associated with a poor prognosis in multiple tumor types, including HCC, however, their molecular and functional diversity is still poorly understood.MethodsWe analyzed the molecular and spatial organization patterns of immune cells within the TME and adjacent tissue of 26 resected HCC lesions using single-cell RNA sequencing and multiplex immunohistochemistry (IHC).ResultsWe found that Kupffer cells, the self-renewing tissue-resident macrophages in liver tissue, are lacking from the TME, which is dominated by monocyte-derived macrophages. ScRNAseq followed by high-resolution clustering identified distinct MF molecular programs within the monocyte-derived macrophage compartment. One MF subset expressed a shared signature with monocytes including FCN1, S100A8 and T cell activation genes like CXCL9 and IL32. Conversely, one subset of MF expressed FOLR2, SEPP1 and genes of the complement (C1Qs), a program shared with KC in the adjacent tissue, and include another intratumoral subset enriched for the expression of TREM2 and GPNMB. Guided by these results, we are developing an IHC antibody panel that allows to visualize distinct MF localization in the TME. Intratumoral MF interface with, and potentially regulate, the T cell compartment within the TME. We are analyzing HCC tumor lesions in our treatment-naïve cohort and in patients treated with neoadjuvant anti-PD-1 therapy (NCT03916627) to study co-localization and direct interaction of MF and T cells using physically-interacting cell sequencing (PICseq). This analysis enables us to identify MF with direct cell-cell contact with T cells, and our preliminary analysis demonstrates an enrichment in MF with an immunosuppressive phenotype. We are also using spatial transcriptomic to map molecular programs of MF in the TME, that we will corroborated with additional patients.ConclusionsTaken together, our data provide a new understanding of intratumoral MF diversity and highlight the presence of specific immunoregulatory MF programs unique to tumor lesions, with subsets of these MF found to be directly interacting with T cells, potentially modulating anti-tumor responsiveness. Our analysis of resected tumor from anti-PD-1 treated patients, will allow us to correlate MF programs, and direct T cell interaction, with clinical response, and will inform therapeutic trials targeting specific MF populations so as to improve clinical efficacy of cancer immunotherapy.Trial RegistrationNCT03916627Ethics ApprovalSamples of tumor and non-involved liver were obtained from surgical specimens of patients undergoing resection at Mount Sinai Hospital (New York, NY) after obtaining informed consent in accordance with a protocol reviewed and approved by the Institutional Review Board at the Icahn School of Medicine at Mount Sinai (RUTH Human Subjects Electronic Submission System 18–00407 and 20–04150) and in collaboration with the Biorepository and Department of Pathology.

Blockade of the PD-1 pathway is a therapeutic strategy to reinvigorate T cell responses against tumors, and when combined with other biologic therapies in the first line setting this achieves significant clinical response in about 25% of hepatocellular carcinoma (HCC) patients. We hypothesize that phenotypic diversity of tumor infiltrating T cells can explain, at least partially, the disparate clinical responses to immunotherapy. Here, we analyze the molecular diversity of T cells in tumor, adjacent tissue and tumor-draining lymph node (dLN) by single-cell RNA sequencing of tissue from 23 patients with early stage HCC treated by neoadjuvant PD-1 blockade (NCT03916627; final data to be submitted to AACR 2021 as late breaking abstract). We identify distinct subsets of PD-1hi CD8 T cells with varying degrees of exhaustion and effector gene programs. Compared to parallel analysis of untreated HCC tumors, we observed that PD-1 blockade resulted in expansion of PD-1hi CD8 T cells in the tumor, regardless of clinical response. PD-1hi CD8 T cell subsets were highly clonal and enriched in the tumor compared to adjacent tissue, suggesting specificity to tumor antigens. Remarkably, we find an association between a specific transcriptomic phenotype within PD-1hi CD8 T cells and response to PD-1 blockade. Using T cell receptor (TCR) sequencing to study the differentiation patterns between T cell states, we found that expanded clonotypes present among PD-1hi CD8 T cells were also found in CD8 effector cells; these data identify characteristic clonally related T cell populations that are enriched in clinical responders. Furthermore, we find the dLNs harbor clonotypes of PD-1hi CD8 T cells expanded in the tumor. These clonally expanded CD8 T cells in dLN display features of activation and exhaustion, suggesting a continuous role of dLN in anti-tumor responses. These results will be corroborated with 8 additional patient samples in which we will further analyze the molecular characteristics of tumor-specific CD8 T cells in the TME and dLN. Furthermore, our ongoing sequencing of pre- and post-treatment lesions will enable measurement of CD8 T cell clonal expansion and molecular programs during response or resistance to PD-1 blockade. These results should help unravel the contribution of dLN T cells to response to immunotherapy as well as identify T cell molecular programs at baseline that correlate with response. Citation Format: Assaf Magen, Pauline Hamon, Etienne Humblin, Alessandra Schanoski, Joel Kim, Jessica Le Berichel, Ephraim Kenigsberg, Myron Schwartz, Thomas Marron, Alice Kamphorst, Miriam Merad. Heterogeneity of PD-1hi CD8 T cells associates with response to PD-1 blockade in hepatocellular carcinoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 515.

Hepatocellular carcinoma (HCC) has a dismal prognosis, and though checkpoint blocking antibodies have significantly improved patient outcome, many patients remain left out highlighting the need to identify additional immune target to enhance therapeutic immunity. Macrophages (MΦ) are an abundant and heterogeneous population in the tumor microenvironment (TME), and are associated with a poor prognosis in multiple tumor types, including HCC, however, their molecular and functional diversity is still poorly understood. We analyzed the molecular and spatial organization patterns of immune cells within the TME and adjacent tissue of 23 resected HCC lesions using single-cell RNA sequencing and multiplex immunohistochemistry (IHC). Strikingly, we found that Kupffer cells (KC), the self-renewing tissue-resident macrophages in liver tissue, are lacking from the TME, which is dominated by monocyte-derived macrophages. ScRNAseq followed by high-resolution clustering identified distinct MΦ molecular programs within the monocyte-derived macrophage compartment. One MΦ subset expressed a shared signature with monocytes including FCN1 and S100A8 in addition to T cell activation genes like CXCL9 and IL32. Conversely, one subset of MΦ expressed FOLR2, SEPP1 and genes of the complement (C1Qs), a program shared with KC in the adjacent uninvolved tissue, and include another intratumoral subset enriched for the expression of TREM2 and GPNMB. Guided by these results, we are developing an IHC antibody panel that allows to visualize distinct MΦ localization in the TME. Intratumoral MΦ interface with, and potentially regulate, the T cell compartment within the TME. We are analyzing HCC tumor lesions in our treatment-naïve cohort and in patients treated with neoadjuvant anti-PD-1 therapy (NCT03916627) to study direct interaction of MΦ and T cells using an innovative technology of physically-interacting cell sequencing (PICseq). This analysis enables us to identify MΦ with direct cell-cell contact with T cells, and our preliminary analysis demonstrates an enrichment in MΦ with a highly immunosuppressive phenotype. We are also using PICseq to map the molecular program of MΦ that are interacting with T cells and that we will corroborated with additional patients. Taken together, our data provide a new understanding of intratumoral MΦ diversity and highlight the presence of specific immunoregulatory MΦ programs unique to tumor lesions, with subsets of these MΦ found to be directly interacting with T cells, potentially modulating anti-tumor responsiveness. Our analysis of resected tumor from anti-PD-1 treated patients, will allow us to correlate MΦ programs, and direct T cell interaction, with clinical response, and will inform therapeutic trials targeting specific MΦ populations so as to improve clinical efficacy of cancer immunotherapy. Citation Format: Pauline Hamon, Assaf Magen, Merav Cohen, Alexandra Tabachnikova, Christie Chang, Mark Buckup, Leanna Troncoso, Zhen Zhao, Joel Kim, Amir Giladi, Nausicaa Malissen, Fiona Desland, Jessica Le Berichel, Ido Amit, Ephraim Kenigsberg, Myron Schwartz, Thomas Marron, Miriam Merad. Characterization of molecular and spatial diversity of macrophages in hepatocellular carcinoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 64.

Background & Aims Given that gastrointestinal (GI) symptoms are a prominent extrapulmonary manifestation of COVID-19, we investigated intestinal infection with SARS-CoV-2, its effect on pathogenesis, and clinical significance. Methods Human intestinal biopsy tissues were obtained from patients with COVID-19 (n = 19) and uninfected control individuals (n = 10) for microscopic examination, cytometry by time of flight analyses, and RNA sequencing. Additionally, disease severity and mortality were examined in patients with and without GI symptoms in 2 large, independent cohorts of hospitalized patients in the United States (N = 634) and Europe (N = 287) using multivariate logistic regressions. Results COVID-19 case patients and control individuals in the biopsy cohort were comparable for age, sex, rates of hospitalization, and relevant comorbid conditions. SARS-CoV-2 was detected in small intestinal epithelial cells by immunofluorescence staining or electron microscopy in 15 of 17 patients studied. High-dimensional analyses of GI tissues showed low levels of inflammation, including down-regulation of key inflammatory genes including IFNG, CXCL8, CXCL2, and IL1B and reduced frequencies of proinflammatory dendritic cells compared with control individuals. Consistent with these findings, we found a significant reduction in disease severity and mortality in patients presenting with GI symptoms that was independent of sex, age, and comorbid illnesses and despite similar nasopharyngeal SARS-CoV-2 viral loads. Furthermore, there was reduced levels of key inflammatory proteins in circulation in patients with GI symptoms. Conclusions These data highlight the absence of a proinflammatory response in the GI tract despite detection of SARS-CoV-2. In parallel, reduced mortality in patients with COVID-19 presenting with GI symptoms was observed. A potential role of the GI tract in attenuating SARS-CoV-2–associated inflammation needs to be further examined.

Human tissues comprise trillions of cells that populate a complex space of molecular phenotypes and functions and that vary in abundance by 4–9 orders of magnitude. Relying solely on unbiased sampling to characterize cellular niches becomes infeasible, as the marginal utility of collecting more cells diminishes quickly. Furthermore, in many clinical samples, the relevant cell types are scarce and efficient processing is critical. We developed an integrated pipeline for index sorting and massively parallel single-cell RNA sequencing (MARS-seq2.0) that builds on our previously published MARS-seq approach. MARS-seq2.0 is based on >1 million cells sequenced with this pipeline and allows identification of unique cell types across different tissues and diseases, as well as unique model systems and organisms. Here, we present a detailed step-by-step procedure for applying the method. In the improved procedure, we combine sub-microliter reaction volumes, optimization of enzymatic mixtures and an enhanced analytical pipeline to substantially lower the cost, improve reproducibility and reduce well-to-well contamination. Data analysis combines multiple layers of quality assessment and error detection and correction, graphically presenting key statistics for library complexity, noise distribution and sequencing saturation. Importantly, our combined FACS and single-cell RNA sequencing (scRNA-seq) workflow enables intuitive approaches for depletion or enrichment of cell populations in a data-driven manner that is essential to efficient sampling of complex tissues. The experimental protocol, from cell sorting to a ready-to-sequence library, takes 2–3 d. Sequencing and processing the data through the analytical pipeline take another 1–2 d. Introduction The remarkably rich repertoire of transcriptional states of cells within tissues has been studied for many decades. However, only recently have experimental and computational advances in the field of single-cell genomics opened the way for unbiased dissection of tissues into single cells and the de novo characterization of cell types, subtypes, and transcriptional states1–13. Single-cell RNA sequencing (scRNA-seq) is emerging as a key molecular tool for elucidating biological complexity, promising to contribute to a variety of fields in both basic research and medicine14–17. Methods for single-cell genome-wide expression analysis are continuously being developed, offering increasing coverage, precision and throughput1,10,18–31. Here, we describe a robust method for massively parallel scRNA-seq that combines indexed FACS sorting (recording of surface marker levels for each sorted single cell) and robotics with multiple layers of molecular barcoding. Although drop-based and microwell-based methods increased the throughput of scRNA-seq methods17,29,30, the complexity of the cellular hierarchy remains a major challenge for unbiased sampling of mammalian tissues. Human tissues comprise trillions of cells that populate a complex space of molecular phenotypes and functions such that states within this space may vary in their abundance by 4, 5 and up to 9 orders of magnitude32,33. A classic example of this remarkable property of the human cell state space is the difference in abundance between highly abundant red blood cells (N ~ 1013) and rare hematopoietic stem cells (N ~ 103−4)34,35. A well-designed single-cell technology must consider the orders-of-magnitude variation in abundance of human cell lineages and develop experimental, computational and statistical approaches to overcome it. Because cells, and especially combinations of stem cells, progenitors and various types of mature cells, span many orders of magnitude in abundance, relying solely on unbiased sampling to accurately characterize rare populations becomes infeasible, as the marginal utility of collecting more cells diminishes quickly (Fig. 1a). Combining unbiased transcriptional maps with protein markers enables iterative refinement of cell sampling using FACS-based enrichment and/or depletion strategies for more efficient and deeper sampling of key rare subpopulations (Fig. 1a). Combining index sorting and scRNA-seq has many other potential utilities, including refinement and generation of new sorting panels for both diagnostics and basic research and for mapping fluorescent reporter markers36,37 for lineage tracing and CRISPR pooled screens (CRISP-seq38; Fig. 1b,c). Overview of the experimental protocol MARS-seq2.0 provides a complete experimental and computational framework for massively parallel scRNA-seq. It integrates index sorting into the scRNA-seq pipeline, minimizes doublets (two independent cells that are captured and processed together) and technical cell-to-cell contamination rates (background noise), and supports processing of tens of thousands of cells a day in a simple and costeffective manner. MARS-seq2.0 is based on our MARS-seq protocol, which we presented in February 2014 (ref. 10). So far, we have used MARS-seq2.0 to process > 1 million cells from mice, non-model organisms and human studies10,35,36,38–42, generating between ~100 and ~10,000 unique molecular identifiers (UMIs; molecular tags that serve to reduce quantitative bias) for each, with an average of ~1,700 UMIs and ~700 genes per cell (Supplementary Fig. 1). Using MARS-seq2.0, we identified a unique microglia type associated with restricting development of Alzheimer’s disease40,43, characterized hematopoietic progenitors9,35, mapped the cellular composition of the lung during development42 and in response to influenza virus infection44, and detailed the heterogeneity of epithelial cells in the thymus41. We also used MARS-seq2.0 for whole-organism mapping of early metazoan cell type–specific transcription45,46. Further, we applied MARS-seq2.0 to large clinical studies characterizing the heterogeneity of plasma cells in healthy donors and multiple myeloma patients47, as well as infiltrating immune cells in lung and melanoma patients48,49. MARS-seq2.0 was also combined with other modalities, including single-cell CRISPR-pooled screens to dissect immune circuits in vitro and in mice38, and photoactivatable reporters (NICHE-seq), single-molecule fluorescence in situ hybridization and paired-cell sequencing (gene expression profiling of pairs of joined cells) to spatially reconstruct immune niches36,50,51. In the MARS-seq2.0 framework, single cells are index-sorted by FACS into 384-well plates, lysed and automatically barcoded (with a cellular and a molecular barcode) by reverse transcription (RT) using a poly-T primer in a low-volume (500 nl) reaction. Then the single cells are pooled together, using a liquid-handling robot for subsequent molecular reactions to be performed on the pooled and labeled material. Each group of single cells amplified together is referred to as an ‘amplification batch’. The pooled cDNA is converted into double-stranded DNA (dsDNA) and is linearly amplified by T7 RNA polymerase in vitro transcription (IVT). The amplified RNA (aRNA) is fragmented and a pool barcode is added by RNA–DNA ligation, allowing efficient sequencing of 8,000–10,000 cells in a single run. Illumina adaptors are then added by RT and PCR, which enables quantification of the mRNA (Fig. 2). During library preparation, we routinely evaluate the complexity of the libraries, yielding a quality control (QC) score that is used to monitor MARS-seq2.0 performance before sequencing. On the basis of these scores, amplified libraries suitable for sequencing of 20–40 384-well plates are pooled and sequenced together on a single flow cell. Sequenced reads are automatically processed in the MARS-seq2.0 computational pipeline (http://compgenomics.weizmann.ac.il/tanay/? page_id=672), in which the molecular signal is separated from the background noise. Optimization of MARS-seq Background noise is a major problem in all massively parallel single-cell protocols. Our prior study10 suggested that this cell-to-cell contamination noise could partially be associated with a molecular process that randomly links a gene, a UMI and a cell barcode during library preparation. To enhance the experimental stability and detection rate while controlling for potential sources of systematic noise in MARS-seq2.0, we set out to optimize RNA retention and amplification. Starting from the MARSseq protocol, we optimized all steps of the single-cell processing, including lysis conditions, reaction volume, primer design and the composition of enzymatic reactions. We evaluated the possibility of reducing the RT reaction volume by analyzing 768 single mouse embryonic stem (ES) cells: after sorting, we performed complete evaporation of the 2 μl of lysis buffer containing the RT primer at 95 °C, followed by addition of 0.5 μl of RT mix, using a liquid-handling robot. For these conditions, RT primer quantities were considerably reduced. In low-volume reactions, we did not observe a substantial increase in the number of molecules measured per cell as compared with those of high-volume reactions, but rather saw an improved signal-to-noise ratio as evaluated by the number of molecules in the wells containing cells as compared with our negative-control empty wells (Supplementary Fig. 2a). This is also a consequence of the 200-fold reduction in the RT primer final concentration that prevents well-to-well contamination during the pooling and downstream enzymatic reactions. In addition, such low-volume reactions are cost effective (reducing enzyme consumption) and scalable. Because second-strand-synthesis enzymes are potentially a major sensitivity-limiting factor but can also generate unwanted background noise post pooling, we further evaluated several different second-strand-synthesis mixtures. We observed a small but considerable inverse correlation between the number of molecules obtained in the different conditions and the signal-to-noise ratio, suggesting that although some mixtures may be slightly more efficient in converting the hybrid RNA–cDNA template into a double-stranded DNA, this is accompanied by unwanted hybrid molecular products reaching a noise level of up to 40 % of the true signal in several of the mixtures (Supplementary Fig. 2b). These results demonstrate that it is critical to optimize and control for noise level in scRNA-seq analysis and that this factor can be affected by various steps during single-cell library production. Key procedural advancements in MARS-seq2.0 MARS-seq2.0 introduces important modifications of the MARS-seq method in almost every part of the protocol. These improvements are related to throughput, robustness, noise reduction and costs. Specifically, we have performed the optimizations discussed in the following sections. Experimental improvements Lowering of RT volume: we reduced the volume of the RT reaction by eightfold from 4 μl to 500 nl. Lowering of the RT volume was possible due to evaporation of the sample in the 384-well plate before addition of the RT mix. This is a direct sixfold reduction in the cost of the single-cell library preparation because the RT step is the most costly part of the protocol. Optimization of lysis buffer: we optimized the lysis buffer to make it compatible with the aforementioned volume reduction. Reduction of RT primer concentration: we reduced the concentration of the RT primer from 200 to 1 nM (enabled by the lower volume), which further reduced the contamination and background noise. Optimization of RT primer composition: we modified the cell barcode (7 bp) and the UMI (8 bp) to optimize yield with longer oligos that enable efficient error correction. Primer removal by exonuclease I is performed in each well before pooling in order to maximize the exclusion of any RT primer leftovers that were not used in the RT and could be a source of potential noise. Optimization of second-strand-synthesis enzymes: we tested all major commercially available second-strand-synthesis mixtures, in different dilutions, and identified a specific mixture and concentration that maximize yield and reduce PCR contaminants (major problem in the original MARS-seq and all IVT-based protocols). Optimization of barcoded ligation adaptor: we modified the barcoded ligation adaptor sequence (barcode of 4 bp, 5 Ns) to improve barcoding of amplification batches and reduce the noise level between them. Optimization of the conditions indicated above resulted in a sixfold reduction in the cost of library production (from $0.65 to $0.10 per cell) and reduced the background level (from 10–15 % to 2 %) as evaluated by the number of molecules in the wells containing cells as compared with our negative-control empty wells. Analytical and QC improvements QC scheme to monitor MARS-seq2.0 performance during library preparation: we generated a QC pipeline based on quantitative real-time polymerase chain reaction (qPCR) of housekeeping genes in order to evaluate the complexity of the libraries before sequencing. Generation of a complete analytical pipeline: sequenced reads are de-multiplexed and the molecularsignal is separated from the background noise. QC measurements following sequencing: the analytical pipeline automatically generates a set of > 20 QC measurements (e.g., number of reads, percentage mapping to exons and number of UMIs) that can assist the users in evaluating and optimizing many important features in their single-cell data. MARS-seq2.0 evaluation To evaluate the efficiency of the method and data quality, we applied MARS-seq2.0 to 2,256 mouse ES cells and mouse embryonic fibroblasts (MEFs). We detected 11.3 million mRNA molecules (medians of ~6,000 and ~4,000 molecules per ES and MEF cell, respectively, Supplementary Fig. 3a) while sequencing 80,000–100,000 reads per cell (Supplementary Fig. 3b). As expected, the number of detected transcripts per cell is correlated with cell size (Supplementary Fig. 3c). Because MARS-seq2.0 is designed to allow deep population sampling and characterization of known and novel subpopulations, we quantify our detection rate by estimating the distribution of mean gene expression on independent pools of 100 single cells, revealing a dynamic range spreading over three orders of magnitude (Supplementary Fig. 3d). By comparing a pool of 1,128 ES cells with a pool of 752 MEF cells, we detected 2,350 genes with a significant (false-discovery rate (FDR) < 10−5) fourfold difference in expression (Supplementary Fig. 3e). In many cases, these expression differences are also pronounced at the single-cell level (Supplementary Fig. 3f–i). Advances in microfluidics technologies allow processing of several thousands of cells in parallel. However, current microfluidic approaches display a relatively high rate (1–15 %) of sequencing of two cells (or more) in a single droplet (doublets)30,31,52,53, which may complicate downstream analysis. To evaluate the doublet rate in MARS-seq2.0, we mixed human and mouse ES cells and analyzed the percentage of doublets. Our analysis demonstrates that MARS-seq2.0 has a negligible amount of doublet cells (< 0.2 %; 2 out of 1,041 cells) and provides high confidence in cell identity (Supplementary Fig. 4). The computational framework To implement an effective error-correction algorithm and to ensure the quality of sequenced libraries, we designed a computational framework that closely follows the different experimental steps in MARS-seq2.0. Our framework models RNA-seq data generation as a sequential process of multiple samplings and amplifications steps (see Supplementary Methods, Supplementary Fig. 5), in which single mRNA molecules in a single cell are distributed and sampled in different pools. The computational pipeline quantifies the number of tagged mRNA molecules per gene that went through this complex sampling process, while eliminating several types of experimental artifacts. In addition, it generates a graphical and user-friendly library diagnostics report for each amplification batch, highlighting important technical parameters that can be variable between batches (see Supplementary Figs. 6–8 for representative diagnostic reports). The report provides a detailed analysis of the reads’ fraction spent on sequencing of primer and other sequences used during library preparation, which can highlight biases such as those caused by an excess of primers. Our pipeline automatically filters sequencing and other polymerization errors that occur in earlier experimental steps (i.e., IVT) and generate spurious UMIs or alter cellular barcodes. Different sources of barcode contamination may substantially bias biological interpretation of the data by mixing genes between different single-cell subpopulations in the same amplification batch. We therefore developed an experimental design that allows us to routinely estimate the total cell-to-cell contamination levels by measuring the number of molecules associated with four negative-control wells (empty wells that do not contain cells) in each amplification batch. To visually relate the number of molecules per cell (Supplementary Fig. 3a) to the original single-cell position on the 384-well plate matrix, a plate-view heat-map is generated that shows how the number of recovered genes and spike-in molecules are distributed across the plate position. This can highlight potential sorting, robotic or other problems. Although technical QC is critical, as demonstrated above, we stress that the ultimate way to evaluate the effectiveness of a specific scRNA-seq method is to examine the novelty and richness of the biological insights emerging from the data and that these may depend on multiple factors, including the quality of biological materials. Strengths and limitations of the protocol The key strengths of MARS-seq2.0 are as follows: The combination of indexed FACS sorting and scRNA-seq enables refinement and enrichment of the cells of interest, which is especially critical for analysis of rare subpopulations and processing of rare cells in human clinical samples. The improvements made in MARS-seq2.0 provide a cost-effective library preparation method that allows analysis of any number of single cells without compromising the cost per cell value. This is in contrast to other high-throughput methods, such as droplets30,31, that become cost effective only when processing a very large number of cells at once. The detailed experimental and analytical QC pipelines allow the user to carefully monitor the quality of the libraries and data and to troubleshoot specific problems in the scRNA-seq process. The amount of cell doublets in the MARS-seq2.0 data is extremely low and allows for identification of rare cell types and subpopulations. MARS-seq2.0 provides strand-specific information about the transcripts. However, the following limitations remain: MARS-seq2.0 is a 3ʹ-based scRNA-seq method and will yield information only about the 3ʹ end of the transcript. Therefore, it is not suitable for identifying alternative splicing isoforms or specific sequences at the 5ʹ end of the gene (for these purposes, single-cell full-length RNA-sequencing methodologies, such as SMART-seq24, should be used). The method is selective for polyadenylated RNA. The method requires access to FACS facilities, as well as liquid-handling robotics. Future applications The reduction in cost per sample in MARS-seq2.0, and especially the ability to refine the population of interest by enriching or depleting cell populations in a data-driven manner, allows the researcher to focus on specific populations, including very rare cell types or states in animal models and clinical samples47,49. Furthermore, techniques for simultaneous acquisition of RNA and other molecular signatures of single cells, for example, spatial location36,51, genetics54, lineage55 and signaling37, are critical for deep molecular understanding of physiological processes and diseases. MARS-seq2.0 presents a flexible platform for combination of unbiased transcriptional mapping with a large number of fluorescent markers that enables collection of multiple information tiers on the same single cell, such as time, spatial location, signaling, genetics, epigenetics and many more56.

SummaryLymphoid cells that produce IL-17 cytokines protect barrier tissues from pathogenic microbes, but are also prominent effectors of inflammation and autoimmune disease. T-helper (TH17) cells, defined by RORγt-dependent production of IL-17A and IL-17F, exert homeostatic functions in the gut upon microbiota-directed differentiation from naïve CD4+ T cells. In the non-pathogenic setting, their cytokine production is regulated by serum amyloid A proteins (SAA1 and SAA2) secreted by adjacent intestinal epithelial cells. However, TH17 cell behaviors vary markedly according to their environment. Here we show that SAAs additionally direct a pathogenic pro-inflammatory TH17 cell differentiation program, acting directly on T cells in collaboration with STAT3-activating cytokines. Using loss- and gain-of-function mouse models, we show that SAA1, SAA2, and SAA3 have distinct systemic and local functions in promoting TH17-mediated inflammatory diseases. These studies suggest that T cell signaling pathways modulated by the SAAs may be attractive targets for anti-inflammatory therapies.

Checkpoint blockade therapies have improved cancer treatment, but such immunotherapy regimens fail in a large subset of patients. Conventional type 1 dendritic cells (DC1s) control the response to checkpoint blockade in preclinical models and are associated with better overall survival in patients with cancer, reflecting the specialized ability of these cells to prime the responses of CD8

Lymphoid cells that produce IL-17 cytokines protect barrier tissues from pathogenic microbes, but are also prominent effectors of inflammation and autoimmune disease. T-helper (TH17) cells, defined by RORγt-dependent production of IL-17A and IL-17F, exert homeostatic functions in the gut upon microbiota-directed differentiation from naive CD4+ T cells. In the non-pathogenic setting, their cytokine production is regulated by serum amyloid A proteins (SAA1 and SAA2) secreted by adjacent intestinal epithelial cells. However, TH17 cell behaviors vary markedly according to their environment. Here we show that SAAs additionally direct a pathogenic pro-inflammatory TH17 cell differentiation program, acting directly on T cells in collaboration with STAT3-activating cytokines. Using loss- and gain-of-function mouse models, we show that SAA1, SAA2, and SAA3 have distinct systemic and local functions in promoting TH17-mediated inflammatory diseases. These studies suggest that T cell signaling pathways modulated by the SAAs may be attractive targets for anti-inflammatory therapies.

SummaryClinical benefits to cytokine blockade in ileal Crohn’s disease (iCD) have been limited to a subset of patients. Whether cellular and molecular heterogeneity contributes to variability in treatment responses has been unclear. Using single cell technologies combining scRNAseq, CyTOF and multiplex tissue imaging, we mapped the cellular landscape of inflamed ileum lesions, adjacent non-inflamed ileum and matched circulating blood cells of iCD patients. In inflamed tissues, we identified a pathogenic module characterized by an inflammatory mononuclear phagocyte (Inf.MNP)-associated cellular response organized around inflammatory macrophages and mature dendritic cells in a subset of iCD patients. We confirmed the Inf.MNP-associated cellular response in 4 independent iCD cohorts (n=441) and showed that presence of this pathogenic module at diagnosis correlated with primary resistance to anti-TNF therapy. Single cell mapping of iCD tissues identifies a complex cellular signature of anti-TNF resistance thereby revealing novel biomarkers of treatment response and tailored therapeutic opportunities.

Genome sequence compositions and epigenetic organizations are correlated extensively across multiple length scales. Replication dynamics, in particular, is highly correlated with GC content. We combine genome-wide time of replication (ToR) data, topological domains maps and detailed functional epigenetic annotations to study the correlations between replication timing and GC content at multiple scales. We find that the decrease in genomic GC content at large scale late replicating regions can be explained by mutation bias favoring A/T nucleotide, without selection or biased gene conversion. Quantification of the free dNTP pool during the cell cycle is consistent with a mechanism involving replication-coupled mutation spectrum that favors AT nucleotides at late S-phase. We suggest that mammalian GC content composition is shaped by independent forces, globally modulating mutation bias and locally selecting on functional element. Deconvoluting these forces and analyzing them on their native scales is important for proper characterization of complex genomic correlations.

Summary Clinical benefits of cytokine blockade in ileal Crohn’s disease (iCD) are limited to a subset of patients. Here, we applied single-cell technologies to iCD lesions to address whether cellular heterogeneity contributes to treatment resistance. We found that a subset of patients expressed a unique cellular module in inflamed tissues that consisted of IgG plasma cells, inflammatory mononuclear phagocytes, activated T cells, and stromal cells, which we named the GIMATS module. Analysis of ligand-receptor interaction pairs identified a distinct network connectivity that likely drives the GIMATS module. Strikingly, the GIMATS module was also present in a subset of patients in four independent iCD cohorts (n = 441), and its presence at diagnosis correlated with failure to achieve durable corticosteroid-free remission upon anti-TNF therapy. These results emphasize the limitations of current diagnostic assays and the potential for single-cell mapping tools to identify novel biomarkers of treatment response and tailored therapeutic opportunities.

SummaryMammalian CpG islands are key epigenomic elements that were first characterized experimentally as genomic fractions with low levels of DNA methylation. Currently, CpG islands are defined based on their genomic sequences alone. Here, we develop evolutionary models to show that several distinct evolutionary processes generate and maintain CpG islands. One central evolutionary regime resulting in enriched CpG content is driven by low levels of DNA methylation and consequentially low rates of CpG deamination. Another major force forming CpG islands is biased gene conversion that stabilizes constitutively methylated CpG islands by balancing rapid deamination with CpG fixation. Importantly, evolutionary analysis and population genetics data suggest that selection for high CpG content is not a significant factor contributing to conservation of CpGs in differentially methylated regions. The heterogeneous, but not selective, origins of CpG islands have direct implications for the understanding of DNA methylation patterns in healthy and diseased cells.

Summary Within the bone marrow, stem cells differentiate and give rise to diverse blood cell types and functions. Currently, hematopoietic progenitors are defined using surface markers combined with functional assays that are not directly linked with in vivo differentiation potential or gene regulatory mechanisms. Here, we comprehensively map myeloid progenitor subpopulations by transcriptional sorting of single cells from the bone marrow. We describe multiple progenitor subgroups, showing unexpected transcriptional priming toward seven differentiation fates but no progenitors with a mixed state. Transcriptional differentiation is correlated with combinations of known and previously undefined transcription factors, suggesting that the process is tightly regulated. Histone maps and knockout assays are consistent with early transcriptional priming, while traditional transplantation experiments suggest that in vivo priming may still allow for plasticity given strong perturbations. These data establish a reference model and general framework for studying hematopoiesis at single-cell resolution.

In multicellular organisms, biological function emerges when heterogeneous cell types form complex organs. Nevertheless, dissection of tissues into mixtures of cellular subpopulations is currently challenging. We introduce an automated massively parallel single-cell RNA sequencing (RNA-seq) approach for analyzing in vivo transcriptional states in thousands of single cells. Combined with unsupervised classification algorithms, this facilitates ab initio cell-type characterization of splenic tissues. Modeling single-cell transcriptional states in dendritic cells and additional hematopoietic cell types uncovers rich cell-type heterogeneity and gene-modules activity in steady state and after pathogen activation. Cellular diversity is thereby approached through inference of variable and dynamic pathway activity rather than a fixed preprogrammed cell-type hierarchy. These data demonstrate single-cell RNA-seq as an effective tool for comprehensive cellular decomposition of complex tissues.

Modern functional genomics uncovered numerous functional elements in metazoan genomes. Nevertheless, only a small fraction of the typical non-exonic genome contains elements that code for function directly. On the other hand, a much larger fraction of the genome is associated with significant evolutionary constraints, suggesting that much of the non-exonic genome is weakly functional. Here we show that in flies, local (30–70 bp) conserved sequence elements that are associated with multiple regulatory functions serve as focal points to a pattern of punctuated regional increase in G/C nucleotide frequencies. We show that this pattern, which covers a region tenfold larger than the conserved elements themselves, is an evolutionary consequence of a shift in the balance between gain and loss of G/C nucleotides and that it is correlated with nucleosome occupancy across multiple classes of epigenetic state. Evidence for compensatory evolution and analysis of SNP allele frequencies show that the evolutionary regime underlying this balance shift is likely to be non-neutral. These data suggest that current gaps in our understanding of genome function and evolutionary dynamics are explicable by a model of sparse sequence elements directly encoding for function, embedded into structural sequences that help to define the local and global epigenomic context of such functional elements., Author Summary A key challenge in functional genomics is to predict evolutionary dynamics from functional annotation of the genome and vice versa. Modern epigenomic studies helped assign function to numerous new sequence elements, but left most of the genome essentially uncharacterized. Evolutionary genomics, on the other hand, consistently suggests that a much larger fraction of the un-annotated genome evolves under selective pressure. We hypothesize that this function-selection gap can be attributed to sequences that facilitate the physical organization of functional elements, such as transcription factor binding sites, within chromosomes. We exemplify this by studying in detail the sequences embedding small conserved elements (CEs) in Drosophila. We show that, while CEs have typically high AT content, high GC content levels around them are maintained by a non-neutral evolutionary balance between gain and loss of GC nucleotides. This non-uniform pattern is highly correlated with nucleosome organization around CEs, potentially imposing an evolutionary constraint on as much as one quarter of the genome. We suggest this can at least partly explain the above function-selection gap. Weak evolutionary constraints on “structural” sequences (at scales ranging from one nucleosome to recently described multi-megabase topological domains) may affect genome evolution just like structural motifs shape protein evolution.

SummaryChromosomes are the physical realization of genetic information and thus form the basis for its readout and propagation. Here we present a high-resolution chromosomal contact map derived from a modified genome-wide chromosome conformation capture approach applied to Drosophila embryonic nuclei. The data show that the entire genome is linearly partitioned into well-demarcated physical domains that overlap extensively with active and repressive epigenetic marks. Chromosomal contacts are hierarchically organized between domains. Global modeling of contact density and clustering of domains show that inactive domains are condensed and confined to their chromosomal territories, whereas active domains reach out of the territory to form remote intra- and interchromosomal contacts. Moreover, we systematically identify specific long-range intrachromosomal contacts between Polycomb-repressed domains. Together, these observations allow for quantitative prediction of the Drosophila chromosomal contact map, laying the foundation for detailed studies of chromosome structure and function in a genetically tractable system.