The cis-Regulatory Atlas of the Mouse Immune System.

A complete chart of cis-regulatory elements and their dynamic activity is necessary to understand the transcriptional basis of differentiation and function of an organ system. We generated matched epigenome and transcriptome measurements in 86 primary cell types that span the mouse immune system and its differentiation cascades. This breadth of data enable variance components analysis that suggests that genes fall into two distinct classes, controlled by either enhancer- or promoter-driven logic, and multiple regression that connects genes to the enhancers that regulate them. Relating transcription factor (TF) expression to the genome-wide accessibility of their binding motifs classifies them as predominantly openers or closers of local chromatin accessibility, pinpointing specific cis-regulatory elements where binding of given TFs is likely functionally relevant, validated by chromatin immunoprecipitation sequencing (ChIP-seq). Overall, this cis-regulatory atlas provides a trove of information on transcriptional regulation through immune differentiation and a foundational scaffold to define key regulatory events throughout the immunological genome.

The gut microbiota promotes distal tissue regeneration via RORγ+ regulatory T cell emissaries.

Specific microbial signals induce the differentiation of a distinct pool of RORγ+ regulatory T (Treg) cells crucial for intestinal homeostasis. We discovered highly analogous populations of microbiota-dependent Treg cells that promoted tissue regeneration at extra-gut sites, notably acutely injured skeletal muscle and fatty liver. Inflammatory meditators elicited by tissue damage combined with MHC-class-II-dependent T cell activation to drive the accumulation of gut-derived RORγ+ Treg cells in injured muscle, wherein they regulated the dynamics and tenor of early inflammation and helped balance the proliferation vs. differentiation of local stem cells. Reining in IL-17A-producing T cells was a major mechanism underlying the rheostatic functions of RORγ+ Treg cells in compromised tissues. Our findings highlight the importance of gut-trained Treg cell emissaries in controlling the response to sterile injury of non-mucosal tissues.

Maternal gut bacteria drive intestinal inflammation in offspring with neurodevelopmental disorders by altering the chromatin landscape of CD4+ T cells.

Children with autism spectrum disorders often display dysregulated immune responses and related gastrointestinal symptoms. However, the underlying mechanisms leading to the development of both phenotypes have not been elucidated. Here, we show that mouse offspring exhibiting autism-like phenotypes due to prenatal exposure to maternal inflammation were more susceptible to developing intestinal inflammation following challenges later in life. In contrast to its prenatal role in neurodevelopmental phenotypes, interleukin-17A (IL-17A) generated immune-primed phenotypes in offspring through changes in the maternal gut microbiota that led to postnatal alterations in the chromatin landscape of naive CD4+ T cells. The transfer of stool samples from pregnant mice with enhanced IL-17A responses into germ-free dams produced immune-primed phenotypes in offspring. Our study provides mechanistic insights into why children exposed to heightened inflammation in the womb might have an increased risk of developing inflammatory diseases in addition to neurodevelopmental disorders.

The transcription factor FoxP3 can fold into two dimerization states with divergent implications for regulatory T cell function and immune homeostasis.

FoxP3 is an essential transcription factor (TF) for immunologic homeostasis, but how it utilizes the common forkhead DNA-binding domain (DBD) to perform its unique function remains poorly understood. We here demonstrated that unlike other known forkhead TFs, FoxP3 formed a head-to-head dimer using a unique linker (Runx1-binding region [RBR]) preceding the forkhead domain. Head-to-head dimerization conferred distinct DNA-binding specificity and created a docking site for the cofactor Runx1. RBR was also important for proper folding of the forkhead domain, as truncation of RBR induced domain-swap dimerization of forkhead, which was previously considered the physiological form of FoxP3. Rather, swap-dimerization impaired FoxP3 function, as demonstrated with the disease-causing mutation R337Q, whereas a swap-suppressive mutation largely rescued R337Q-mediated functional impairment. Altogether, our findings suggest that FoxP3 can fold into two distinct dimerization states: head-to-head dimerization representing functional specialization of an ancient DBD and swap dimerization associated with impaired functions.

FOXP3 recognizes microsatellites and bridges DNA through multimerization.

FOXP3 is a transcription factor that is essential for the development of regulatory T cells, a branch of T cells that suppress excessive inflammation and autoimmunity1-5. However, the molecular mechanisms of FOXP3 remain unclear. Here we here show that FOXP3 uses the forkhead domain-a DNA-binding domain that is commonly thought to function as a monomer or dimer-to form a higher-order multimer after binding to TnG repeat microsatellites. The cryo-electron microscopy structure of FOXP3 in a complex with T3G repeats reveals a ladder-like architecture, whereby two double-stranded DNA molecules form the two 'side rails' bridged by five pairs of FOXP3 molecules, with each pair forming a 'rung'. Each FOXP3 subunit occupies TGTTTGT within the repeats in a manner that is indistinguishable from that of FOXP3 bound to the forkhead consensus motif (TGTTTAC). Mutations in the intra-rung interface impair TnG repeat recognition, DNA bridging and the cellular functions of FOXP3, all without affecting binding to the forkhead consensus motif. FOXP3 can tolerate variable inter-rung spacings, explaining its broad specificity for TnG-repeat-like sequences in vivo and in vitro. Both FOXP3 orthologues and paralogues show similar TnG repeat recognition and DNA bridging. These findings therefore reveal a mode of DNA recognition that involves transcription factor homomultimerization and DNA bridging, and further implicates microsatellites in transcriptional regulation and diseases.

Aire regulates chromatin looping by evicting CTCF from domain boundaries and favoring accumulation of cohesin on superenhancers.

Aire controls immunological tolerance by driving promiscuous expression of a large swath of the genome in medullary thymic epithelial cells (mTECs). Its molecular mechanism remains enigmatic. High-resolution chromosome-conformation capture (Hi-C) experiments on ex vivo mTECs revealed Aire to have a widespread impact on higher-order chromatin structure, disfavoring architectural loops while favoring transcriptional loops. In the presence of Aire, cohesin complexes concentrated on superenhancers together with mediator complexes, while the CCCTC-binding factor (CTCF) was relatively depleted from structural domain boundaries. In particular, Aire associated with the cohesin loader, NIPBL, strengthening this factor's affiliation with cohesin's enzymatic subunits. mTEC transcripts up-regulated in the presence of Aire corresponded closely to those down-regulated in the absence of one of the cohesin subunits, SA-2. A mechanistic model incorporating these findings explains many of the unusual features of Aire's impact on mTEC transcription, providing molecular insight into tolerance induction.

Deep learning of immune cell differentiation.

Although we know many sequence-specific transcription factors (TFs), how the DNA sequence of cis-regulatory elements is decoded and orchestrated on the genome scale to determine immune cell differentiation is beyond our grasp. Leveraging a granular atlas of chromatin accessibility across 81 immune cell types, we asked if a convolutional neural network (CNN) could learn to infer cell type-specific chromatin accessibility solely from regulatory DNA sequences. With a tailored architecture and an ensemble approach to CNN parameter interpretation, we show that our trained network ("AI-TAC") does so by rediscovering ab initio the binding motifs for known regulators and some unknown ones. Motifs whose importance is learned virtually as functionally important overlap strikingly well with positions determined by chromatin immunoprecipitation for several TFs. AI-TAC establishes a hierarchy of TFs and their interactions that drives lineage specification and also identifies stage-specific interactions, like Pax5/Ebf1 vs. Pax5/Prdm1, or the role of different NF-κB dimers in different cell types. AI-TAC assigns Spi1/Cebp and Pax5/Ebf1 as the drivers necessary for myeloid and B lineage fates, respectively, but no factors seemed as dominantly required for T cell differentiation, which may represent a fall-back pathway. Mouse-trained AI-TAC can parse human DNA, revealing a strikingly similar ranking of influential TFs and providing additional support that AI-TAC is a generalizable regulatory sequence decoder. Thus, deep learning can reveal the regulatory syntax predictive of the full differentiative complexity of the immune system.

Feedforward regulation of Myc coordinates lineage-specific with housekeeping gene expression during B cell progenitor cell differentiation.

The differentiation of self-renewing progenitor cells requires not only the regulation of lineage- and developmental stage-specific genes but also the coordinated adaptation of housekeeping functions from a metabolically active, proliferative state toward quiescence. How metabolic and cell-cycle states are coordinated with the regulation of cell type-specific genes is an important question, because dissociation between differentiation, cell cycle, and metabolic states is a hallmark of cancer. Here, we use a model system to systematically identify key transcriptional regulators of Ikaros-dependent B cell-progenitor differentiation. We find that the coordinated regulation of housekeeping functions and tissue-specific gene expression requires a feedforward circuit whereby Ikaros down-regulates the expression of Myc. Our findings show how coordination between differentiation and housekeeping states can be achieved by interconnected regulators. Similar principles likely coordinate differentiation and housekeeping functions during progenitor cell differentiation in other cell lineages.

TCF7L1 suppresses primitive streak gene expression to support human embryonic stem cell pluripotency.

Human embryonic stem cells (hESCs) are exquisitely sensitive to WNT ligands, which rapidly cause differentiation. Therefore, hESC self-renewal requires robust mechanisms to keep the cells in a WNT inactive but responsive state. How they achieve this is largely unknown. We explored the role of transcriptional regulators of WNT signaling, the TCF/LEFs. As in mouse ESCs, TCF7L1 is the predominant family member expressed in hESCs. Genome-wide, it binds a gene cohort involved in primitive streak formation at gastrulation, including NODAL, BMP4 and WNT3 Comparing TCF7L1-bound sites with those bound by the WNT signaling effector ß-catenin indicates that TCF7L1 acts largely on the WNT signaling pathway. TCF7L1 overlaps less with the pluripotency regulators OCT4 and NANOG than in mouse ESCs. Gain- and loss-of-function studies indicate that TCF7L1 suppresses gene cohorts expressed in the primitive streak. Interestingly, we find that BMP4, another driver of hESC differentiation, downregulates TCF7L1, providing a mechanism of BMP and WNT pathway intersection. Together, our studies indicate that TCF7L1 plays a major role in maintaining hESC pluripotency, which has implications for human development during gastrulation.

Identification of NK Cell Subpopulations That Differentiate HIV-Infected Subject Cohorts with Diverse Levels of Virus Control.

HIV infection is controlled immunologically in a small subset of infected individuals without antiretroviral therapy (ART), though the mechanism of control is unclear. CD8+ T cells are a critical component of HIV control in many immunological controllers. NK cells are also believed to have a role in controlling HIV infection, though their role is less well characterized. We used mass cytometry to simultaneously measure the levels of expression of 24 surface markers on peripheral NK cells from HIV-infected subjects with various degrees of HIV natural control; we then used machine learning to identify NK cell subpopulations that differentiate HIV controllers from noncontrollers. Using CITRUS (cluster identification, characterization, and regression), we identified 3 NK cell subpopulations that differentiated subjects with chronic HIV viremia (viremic noncontrollers [VNC]) from individuals with undetectable HIV viremia without ART (elite controllers [EC]). In a parallel approach, we identified 11 NK cell subpopulations that differentiated HIV-infected subject groups using k-means clustering after dimensionality reduction by t-neighbor stochastic neighbor embedding (tSNE) or linear discriminant analysis (LDA). Among these additional 11 subpopulations, the frequencies of 5 correlated with HIV DNA levels; importantly, significance was retained in 2 subpopulations in analyses that included only cohorts without detectable viremia. By comparing the surface marker expression patterns of all identified subpopulations, we revealed that the CD11b+ CD57- CD161+ Siglec-7+ subpopulation of CD56dim CD16+ NK cells are more abundant in EC and HIV-negative controls than in VNC and that the frequency of these cells correlated with HIV DNA levels. We hypothesize that this population may have a role in immunological control of HIV infection.IMPORTANCE HIV infection results in the establishment of a stable reservoir of latently infected cells; ART is usually required to keep viral replication under control and disease progression at bay, though a small subset of HIV-infected subjects can control HIV infection without ART through immunological mechanisms. In this study, we sought to identify subpopulations of NK cells that may be involved in the natural immunological control of HIV infection. We used mass cytometry to measure surface marker expression on peripheral NK cells. Using two distinct semisupervised machine learning approaches, we identified a CD11b+ CD57- CD161+ Siglec-7+ subpopulation of CD56dim CD16+ NK cells that differentiates HIV controllers from noncontrollers. These cells can be sorted out for future functional studies to assess their potential role in the immunological control of HIV infection.

Building gene regulatory networks from scATAC-seq and scRNA-seq using Linked Self Organizing Maps.

Rapid advances in single-cell assays have outpaced methods for analysis of those data types. Different single-cell assays show extensive variation in sensitivity and signal to noise levels. In particular, scATAC-seq generates extremely sparse and noisy datasets. Existing methods developed to analyze this data require cells amenable to pseudo-time analysis or require datasets with drastically different cell-types. We describe a novel approach using self-organizing maps (SOM) to link scATAC-seq regions with scRNA-seq genes that overcomes these challenges and can generate draft regulatory networks. Our SOMatic package generates chromatin and gene expression SOMs separately and combines them using a linking function. We applied SOMatic on a mouse pre-B cell differentiation time-course using controlled Ikaros over-expression to recover gene ontology enrichments, identify motifs in genomic regions showing similar single-cell profiles, and generate a gene regulatory network that both recovers known interactions and predicts new Ikaros targets during the differentiation process. The ability of linked SOMs to detect emergent properties from multiple types of highly-dimensional genomic data with very different signal properties opens new avenues for integrative analysis of heterogeneous data.

FoxP3 recognizes microsatellites and bridges DNA through multimerization.

FoxP3 is a transcription factor (TF) essential for development of regulatory T cells (Tregs), a branch of T cells that suppress excessive inflammation and autoimmunity 1-5 . Molecular mechanisms of FoxP3, however, remain elusive. We here show that FoxP3 utilizes the Forkhead domain--a DNA binding domain (DBD) that is commonly thought to function as a monomer or dimer--to form a higher-order multimer upon binding to T n G repeat microsatellites. A cryo-electron microscopy structure of FoxP3 in complex with T 3 G repeats reveals a ladder-like architecture, where two double-stranded DNA molecules form the two "side rails" bridged by five pairs of FoxP3 molecules, with each pair forming a "rung". Each FoxP3 subunit occupies TGTTTGT within the repeats in the manner indistinguishable from that of FoxP3 bound to the Forkhead consensus motif (FKHM; TGTTTAC). Mutations in the "intra-rung" interface impair T n G repeat recognition, DNA bridging and cellular functions of FoxP3, all without affecting FKHM binding. FoxP3 can tolerate variable "inter-rung" spacings, explaining its broad specificity for T n G repeat-like sequences in vivo and in vitro . Both FoxP3 orthologs and paralogs show similar T n G repeat recognition and DNA bridging. These findings thus reveal a new mode of DNA recognition that involves TF homo-multimerization and DNA bridging, and further implicates microsatellites in transcriptional regulation and diseases.

Mutations from patients with IPEX ported to mice reveal different patterns of FoxP3 and Treg dysfunction.

Mutations of the transcription factor FoxP3 in patients with "IPEX" (immune dysregulation, polyendocrinopathy, enteropathy, X-linked syndrome) disrupt regulatory T cells (Treg), causing an array of multiorgan autoimmunity. To understand the functional impact of mutations across FoxP3 domains, without genetic and environmental confounders, six human FOXP3 missense mutations are engineered into mice. Two classes of mutations emerge from combined immunologic and genomic analyses. A mutation in the DNA-binding domain shows the same lymphoproliferation and multiorgan infiltration as complete FoxP3 knockouts but delayed by months. Tregs expressing this mutant FoxP3 are destabilized by normal Tregs in heterozygous females compared with hemizygous males. Mutations in other domains affect chromatin opening differently, involving different cofactors and provoking more specific autoimmune pathology (dermatitis, colitis, diabetes), unmasked by immunological challenges or incrossing NOD autoimmune-susceptibility alleles. This work establishes that IPEX disease heterogeneity results from the actual mutations, combined with genetic and environmental perturbations, explaining then the intra-familial variation in IPEX.

FoxP3 associates with enhancer-promoter loops to regulate Treg-specific gene expression.

Gene expression programs are specified by higher-order chromatin structure and enhancer-promoter loops (EPLs). T regulatory cell (Treg) identity is dominantly specified by the transcription factor (TF) FoxP3, whose mechanism of action is unclear. We applied chromatin conformation capture with immunoprecipitation (HiChIP) in Treg and closely related conventional CD4+ T cells (Tconv). EPLs identified by H3K27Ac HiChIP showed a range of connection intensity, with some superconnected genes. TF-specific HiChIP showed that FoxP3 interacts with EPLs at a large number of genes, including some not differentially expressed in Treg versus Tconv, but enriched at the core Treg signature loci that it up-regulates. FoxP3 association correlated with heightened H3K27Ac looping, as ascertained by analysis of FoxP3-deficient Treg-like cells. There was marked asymmetry in the loci where FoxP3 associated at the enhancer- or the promoter-side of EPLs, with enrichment for different transcriptional cofactors. FoxP3 EPL intensity distinguished gene clusters identified by single-cell ATAC-seq as covarying between individual Tregs, supporting a direct transactivation model for FoxP3 in determining Treg identity.

Interferon-α-producing plasmacytoid dendritic cells drive the loss of adipose tissue regulatory T cells during obesity.

The visceral adipose tissue (VAT) of lean mice hosts a unique population of regulatory T cells (Tregs) that have a distinct transcriptome and T cell receptor (TCR) repertoire and regulate local and systemic inflammation and metabolism. Perplexingly, this population disappears in obese mice, limiting the promise of Treg-based therapies for metabolic disorders. We exploited the power of a VAT-Treg TCR-transgenic mouse model to follow the dynamics of, and phenotypic changes in, the VAT-Treg population throughout the development of diet-induced obesity. Our results show that VAT-Tregs are lost under obesogenic conditions due to downregulation of their defining transcription factor, PPARγ, coupled with their strikingly enhanced responses to pro-inflammatory cytokines. In particular, the VAT from obese mice (and reportedly humans) was strongly enriched in plasmacytoid dendritic cells that actively express interferon-alpha. These cells were directly toxic to PPARγ+ VAT-Tregs. Blocking this pathway in obese mice by multiple approaches substantially restored the VAT-Treg population and enhanced insulin sensitivity.

Slug regulates the Dll4-Notch-VEGFR2 axis to control endothelial cell activation and angiogenesis.

Slug (SNAI2), a member of the well-conserved Snail family of transcription factors, has multiple developmental roles, including in epithelial-to-mesenchymal transition (EMT). Here, we show that Slug is critical for the pathological angiogenesis needed to sustain tumor growth, and transiently necessary for normal developmental angiogenesis. We find that Slug upregulation in angiogenic endothelial cells (EC) regulates an EMT-like suite of target genes, and suppresses Dll4-Notch signaling thereby promoting VEGFR2 expression. Both EC-specific Slug re-expression and reduced Notch signaling, either by γ-secretase inhibition or loss of Dll4, rescue retinal angiogenesis in SlugKO mice. Conversely, inhibition of VEGF signaling prevents excessive angiogenic sprouting of Slug overexpressing EC. Finally, endothelial Slug (but not Snail) is activated by the pro-angiogenic factor SDF1α via its canonical receptor CXCR4 and the MAP kinase ERK5. Altogether, our data support a critical role for Slug in determining the angiogenic response during development and disease.

STATegra, a comprehensive multi-omics dataset of B-cell differentiation in mouse.

Multi-omics approaches use a diversity of high-throughput technologies to profile the different molecular layers of living cells. Ideally, the integration of this information should result in comprehensive systems models of cellular physiology and regulation. However, most multi-omics projects still include a limited number of molecular assays and there have been very few multi-omic studies that evaluate dynamic processes such as cellular growth, development and adaptation. Hence, we lack formal analysis methods and comprehensive multi-omics datasets that can be leveraged to develop true multi-layered models for dynamic cellular systems. Here we present the STATegra multi-omics dataset that combines measurements from up to 10 different omics technologies applied to the same biological system, namely the well-studied mouse pre-B-cell differentiation. STATegra includes high-throughput measurements of chromatin structure, gene expression, proteomics and metabolomics, and it is complemented with single-cell data. To our knowledge, the STATegra collection is the most diverse multi-omics dataset describing a dynamic biological system.

Dynamic Gene Regulatory Networks of Human Myeloid Differentiation.

The reconstruction of gene regulatory networks underlying cell differentiation from high-throughput gene expression and chromatin data remains a challenge. Here, we derive dynamic gene regulatory networks for human myeloid differentiation using a 5-day time series of RNA-seq and ATAC-seq data. We profile HL-60 promyelocytes differentiating into macrophages, neutrophils, monocytes, and monocyte-derived macrophages. We find a rapid response in the expression of key transcription factors and lineage markers that only regulate a subset of their targets at a given time, which is followed by chromatin accessibility changes that occur later along with further gene expression changes. We observe differences between promyelocyte- and monocyte-derived macrophages at both the transcriptional and chromatin landscape level, despite using the same differentiation stimulus, which suggest that the path taken by cells in the differentiation landscape defines their end cell state. More generally, our approach of combining neighboring time points and replicates to achieve greater sequencing depth can efficiently infer footprint-based regulatory networks from long series data.

Prostaglandin E2 Leads to the Acquisition of DNMT3A-Dependent Tolerogenic Functions in Human Myeloid-Derived Suppressor Cells.

Myeloid-derived suppressor cells (MDSCs) and dendritic cells (DCs) arise from common progenitors. Tumor-derived factors redirect differentiation from immune-promoting DCs to tolerogenic MDSCs, an immunological hallmark of cancer. Indeed, in vitro differentiation of DCs from human primary monocytes results in the generation of MDSCs under tumor-associated conditions (PGE2 or tumor cell-conditioned media). Comparison of MDSC and DC DNA methylomes now reveals extensive demethylation with specific gains of DNA methylation and repression of immunogenic-associated genes occurring in MDSCs specifically, concomitant with increased DNA methyltransferase 3A (DNMT3A) levels. DNMT3A downregulation erases MDSC-specific hypermethylation, and it abolishes their immunosuppressive capacity. Primary MDSCs isolated from ovarian cancer patients display a similar hypermethylation signature in connection with PGE2-dependent DNMT3A overexpression. Our study links PGE2- and DNMT3A-dependent hypermethylation with immunosuppressive MDSC functions, providing a promising target for therapeutic intervention.

iPSC-Derived Human Microglia-like Cells to Study Neurological Diseases.

Microglia play critical roles in brain development, homeostasis, and neurological disorders. Here, we report that human microglial-like cells (iMGLs) can be differentiated from iPSCs to study their function in neurological diseases, like Alzheimer's disease (AD). We find that iMGLs develop in vitro similarly to microglia in vivo, and whole-transcriptome analysis demonstrates that they are highly similar to cultured adult and fetal human microglia. Functional assessment of iMGLs reveals that they secrete cytokines in response to inflammatory stimuli, migrate and undergo calcium transients, and robustly phagocytose CNS substrates. iMGLs were used to examine the effects of Aß fibrils and brain-derived tau oligomers on AD-related gene expression and to interrogate mechanisms involved in synaptic pruning. Furthermore, iMGLs transplanted into transgenic mice and human brain organoids resemble microglia in vivo. Together, these findings demonstrate that iMGLs can be used to study microglial function, providing important new insight into human neurological disease.