Author Correction: Repurposing endogenous immune pathways to tailor and control chimeric antigen receptor T cell functionality.

An amendment to this paper has been published and can be accessed via a link at the top of the paper.

Repurposing endogenous immune pathways to tailor and control chimeric antigen receptor T cell functionality.

Endowing chimeric antigen receptor (CAR) T cells with additional potent functionalities holds strong potential for improving their antitumor activity. However, because potency could be deleterious without control, these additional features need to be tightly regulated. Immune pathways offer a wide array of tightly regulated genes that can be repurposed to express potent functionalities in a highly controlled manner. Here, we explore this concept by repurposing TCR, CD25 and PD1, three major players of the T cell activation pathway. We insert the CAR into the TCRα gene (TRACCAR), and IL-12P70 into either IL2Rα or PDCD1 genes. This process results in transient, antigen concentration-dependent IL-12P70 secretion, increases TRACCAR T cell cytotoxicity and extends survival of tumor-bearing mice. This gene network repurposing strategy can be extended to other cellular pathways, thus paving the way for generating smart CAR T cells able to integrate biological inputs and to translate them into therapeutic outputs in a highly regulated manner.

Modulation of chimeric antigen receptor surface expression by a small molecule switch.

BACKGROUND: Engineered therapeutic cells have attracted a great deal of interest due to their potential applications in treating a wide range of diseases, including cancer and autoimmunity. Chimeric antigen receptor (CAR) T-cells are designed to detect and kill tumor cells that present a specific, predefined antigen. The rapid expansion of targeted antigen beyond CD19, has highlighted new challenges, such as autoactivation and T-cell fratricide, that could impact the capacity to manufacture engineered CAR T-cells. Therefore, the development of strategies to control CAR expression at the surface of T-cells and their functions is under intense investigations. RESULTS: Here, we report the development and evaluation of an off-switch directly embedded within a CAR construct (SWIFF-CAR). The incorporation of a self-cleaving degradation moiety controlled by a protease/protease inhibitor pair allowed the ex vivo tight and reversible control of the CAR surface presentation and the subsequent CAR-induced signaling and cytolytic functions of the engineered T-cells using the cell permeable Asunaprevir (ASN) small molecule. CONCLUSIONS: The strategy described in this study could, in principle, be broadly adapted to CAR T-cells development to circumvent some of the possible hurdle of CAR T-cell manufacturing. This system essentially creates a CAR T-cell with an integrated functional rheostat.

An Orthologous Epigenetic Gene Expression Signature Derived from Differentiating Embryonic Stem Cells Identifies Regulators of Cardiogenesis.

Here we used predictive gene expression signatures within a multi-species framework to identify the genes that underlie cardiac cell fate decisions in differentiating embryonic stem cells. We show that the overlapping orthologous mouse and human genes are the most accurate candidate cardiogenic genes as these genes identified the most conserved developmental pathways that characterize the cardiac lineage. An RNAi-based screen of the candidate genes in Drosophila uncovered numerous novel cardiogenic genes. shRNA knockdown combined with transcriptome profiling of the newly-identified transcription factors zinc finger protein 503 and zinc finger E-box binding homeobox 2 and the well-known cardiac regulatory factor NK2 homeobox 5 revealed that zinc finger E-box binding homeobox 2 activates terminal differentiation genes required for cardiomyocyte structure and function whereas zinc finger protein 503 and NK2 homeobox 5 are required for specification of the cardiac lineage. We further demonstrated that an essential role of NK2 homeobox 5 and zinc finger protein 503 in specification of the cardiac lineage is the repression of gene expression programs characteristic of alternative cell fates. Collectively, these results show that orthologous gene expression signatures can be used to identify conserved cardiogenic pathways.

Enhancer modeling uncovers transcriptional signatures of individual cardiac cell states in Drosophila.

Here we used discriminative training methods to uncover the chromatin, transcription factor (TF) binding and sequence features of enhancers underlying gene expression in individual cardiac cells. We used machine learning with TF motifs and ChIP data for a core set of cardiogenic TFs and histone modifications to classify Drosophila cell-type-specific cardiac enhancer activity. We show that the classifier models can be used to predict cardiac cell subtype cis-regulatory activities. Associating the predicted enhancers with an expression atlas of cardiac genes further uncovered clusters of genes with transcription and function limited to individual cardiac cell subtypes. Further, the cell-specific enhancer models revealed chromatin, TF binding and sequence features that distinguish enhancer activities in distinct subsets of heart cells. Collectively, our results show that computational modeling combined with empirical testing provides a powerful platform to uncover the enhancers, TF motifs and gene expression profiles which characterize individual cardiac cell fates.

Machine learning classification of cell-specific cardiac enhancers uncovers developmental subnetworks regulating progenitor cell division and cell fate specification.

The Drosophila heart is composed of two distinct cell types, the contractile cardial cells (CCs) and the surrounding non-muscle pericardial cells (PCs), development of which is regulated by a network of conserved signaling molecules and transcription factors (TFs). Here, we used machine learning with array-based chromatin immunoprecipitation (ChIP) data and TF sequence motifs to computationally classify cell type-specific cardiac enhancers. Extensive testing of predicted enhancers at single-cell resolution revealed the added value of ChIP data for modeling cell type-specific activities. Furthermore, clustering the top-scoring classifier sequence features identified novel cardiac and cell type-specific regulatory motifs. For example, we found that the Myb motif learned by the classifier is crucial for CC activity, and the Myb TF acts in concert with two forkhead domain TFs and Polo kinase to regulate cardiac progenitor cell divisions. In addition, differential motif enrichment and cis-trans genetic studies revealed that the Notch signaling pathway TF Suppressor of Hairless [Su(H)] discriminates PC from CC enhancer activities. Collectively, these studies elucidate molecular pathways used in the regulatory decisions for proliferation and differentiation of cardiac progenitor cells, implicate Su(H) in regulating cell fate decisions of these progenitors, and document the utility of enhancer modeling in uncovering developmental regulatory subnetworks.

Contribution of distinct homeodomain DNA binding specificities to Drosophila embryonic mesodermal cell-specific gene expression programs.

Homeodomain (HD) proteins are a large family of evolutionarily conserved transcription factors (TFs) having diverse developmental functions, often acting within the same cell types, yet many members of this family paradoxically recognize similar DNA sequences. Thus, with multiple family members having the potential to recognize the same DNA sequences in cis-regulatory elements, it is difficult to ascertain the role of an individual HD or a subclass of HDs in mediating a particular developmental function. To investigate this problem, we focused our studies on the Drosophila embryonic mesoderm where HD TFs are required to establish not only segmental identities (such as the Hox TFs), but also tissue and cell fate specification and differentiation (such as the NK-2 HDs, Six HDs and identity HDs (I-HDs)). Here we utilized the complete spectrum of DNA binding specificities determined by protein binding microarrays (PBMs) for a diverse collection of HDs to modify the nucleotide sequences of numerous mesodermal enhancers to be recognized by either no or a single subclass of HDs, and subsequently assayed the consequences of these changes on enhancer function in transgenic reporter assays. These studies show that individual mesodermal enhancers receive separate transcriptional input from both I-HD and Hox subclasses of HDs. In addition, we demonstrate that enhancers regulating upstream components of the mesodermal regulatory network are targeted by the Six class of HDs. Finally, we establish the necessity of NK-2 HD binding sequences to activate gene expression in multiple mesodermal tissues, supporting a potential role for the NK-2 HD TF Tinman (Tin) as a pioneer factor that cooperates with other factors to regulate cell-specific gene expression programs. Collectively, these results underscore the critical role played by HDs of multiple subclasses in inducing the unique genetic programs of individual mesodermal cells, and in coordinating the gene regulatory networks directing mesoderm development.

Highly parallel assays of tissue-specific enhancers in whole Drosophila embryos.

Transcriptional enhancers are a primary mechanism by which tissue-specific gene expression is achieved. Despite the importance of these regulatory elements in development, responses to environmental stresses and disease, testing enhancer activity in animals remains tedious, with a minority of enhancers having been characterized. Here we describe 'enhancer-FACS-seq' (eFS) for highly parallel identification of active, tissue-specific enhancers in Drosophila melanogaster embryos. Analysis of enhancers identified by eFS as being active in mesodermal tissues revealed enriched DNA binding site motifs of known and putative, previously uncharacterized mesodermal transcription factors. Naive Bayes classifiers using transcription factor binding site motifs accurately predicted mesodermal enhancer activity. Application of eFS to other cell types and organisms should accelerate the cataloging of enhancers and understanding how transcriptional regulation is encoded in them.

Integrative analysis of the zinc finger transcription factor Lame duck in the Drosophila myogenic gene regulatory network.

Contemporary high-throughput technologies permit the rapid identification of transcription factor (TF) target genes on a genome-wide scale, yet the functional significance of TFs requires knowledge of target gene expression patterns, cooperating TFs, and cis-regulatory element (CRE) structures. Here we investigated the myogenic regulatory network downstream of the Drosophila zinc finger TF Lame duck (Lmd) by combining both previously published and newly performed genomic data sets, including ChIP sequencing (ChIP-seq), genome-wide mRNA profiling, cell-specific expression patterns of putative transcriptional targets, analysis of histone mark signatures, studies of TF cooccupancy by additional mesodermal regulators, TF binding site determination using protein binding microarrays (PBMs), and machine learning of candidate CRE motif compositions. Our findings suggest that Lmd orchestrates an extensive myogenic regulatory network, a conclusion supported by the identification of Lmd-dependent genes, histone signatures of Lmd-bound genomic regions, and the relationship of these features to cell-specific gene expression patterns. The heterogeneous cooccupancy of Lmd-bound regions with additional mesodermal regulators revealed that different transcriptional inputs are used to mediate similar myogenic gene expression patterns. Machine learning further demonstrated diverse combinatorial motif patterns within tissue-specific Lmd-bound regions. PBM analysis established the complete spectrum of Lmd DNA binding specificities, and site-directed mutagenesis of Lmd and additional newly discovered motifs in known enhancers demonstrated the critical role of these TF binding sites in supporting full enhancer activity. Collectively, these findings provide insights into the transcriptional codes regulating muscle gene expression and offer a generalizable approach for similar studies in other systems.

Two forkhead transcription factors regulate the division of cardiac progenitor cells by a Polo-dependent pathway.

The development of a complex organ requires the specification of appropriate numbers of each of its constituent cell types, as well as their proper differentiation and correct positioning relative to each other. During Drosophila cardiogenesis, all three of these processes are controlled by jumeau (jumu) and Checkpoint suppressor homologue (CHES-1-like), two genes encoding forkhead transcription factors that we discovered utilizing an integrated genetic, genomic, and computational strategy for identifying genes expressed in the developing Drosophila heart. Both jumu and CHES-1-like are required during asymmetric cell division for the derivation of two distinct cardiac cell types from their mutual precursor and in symmetric cell divisions that produce yet a third type of heart cell. jumu and CHES-1-like control the division of cardiac progenitors by regulating the activity of Polo, a kinase involved in multiple steps of mitosis. This pathway demonstrates how transcription factors integrate diverse developmental processes during organogenesis.

Differential regulation of mesodermal gene expression by Drosophila cell type-specific Forkhead transcription factors.

A common theme in developmental biology is the repeated use of the same gene in diverse spatial and temporal domains, a process that generally involves transcriptional regulation mediated by multiple separate enhancers, each with its own arrangement of transcription factor (TF)-binding sites and associated activities. Here, by contrast, we show that the expression of the Drosophila Nidogen (Ndg) gene at different embryonic stages and in four mesodermal cell types is governed by the binding of multiple cell-specific Forkhead (Fkh) TFs - including Biniou (Bin), Checkpoint suppressor homologue (CHES-1-like) and Jumeau (Jumu) - to three functionally distinguishable Fkh-binding sites in the same enhancer. Whereas Bin activates the Ndg enhancer in the late visceral musculature, CHES-1-like cooperates with Jumu to repress this enhancer in the heart. CHES-1-like also represses the Ndg enhancer in a subset of somatic myoblasts prior to their fusion to form multinucleated myotubes. Moreover, different combinations of Fkh sites, corresponding to two different sequence specificities, mediate the particular functions of each TF. A genome-wide scan for the occurrence of both classes of Fkh domain recognition sites in association with binding sites for known cardiac TFs showed an enrichment of combinations containing the two Fkh motifs in putative enhancers found within the noncoding regions of genes having heart expression. Collectively, our results establish that different cell-specific members of a TF family regulate the activity of a single enhancer in distinct spatiotemporal domains, and demonstrate how individual binding motifs for a TF class can differentially influence gene expression.

A machine learning approach for identifying novel cell type-specific transcriptional regulators of myogenesis.

Transcriptional enhancers integrate the contributions of multiple classes of transcription factors (TFs) to orchestrate the myriad spatio-temporal gene expression programs that occur during development. A molecular understanding of enhancers with similar activities requires the identification of both their unique and their shared sequence features. To address this problem, we combined phylogenetic profiling with a DNA-based enhancer sequence classifier that analyzes the TF binding sites (TFBSs) governing the transcription of a co-expressed gene set. We first assembled a small number of enhancers that are active in Drosophila melanogaster muscle founder cells (FCs) and other mesodermal cell types. Using phylogenetic profiling, we increased the number of enhancers by incorporating orthologous but divergent sequences from other Drosophila species. Functional assays revealed that the diverged enhancer orthologs were active in largely similar patterns as their D. melanogaster counterparts, although there was extensive evolutionary shuffling of known TFBSs. We then built and trained a classifier using this enhancer set and identified additional related enhancers based on the presence or absence of known and putative TFBSs. Predicted FC enhancers were over-represented in proximity to known FC genes; and many of the TFBSs learned by the classifier were found to be critical for enhancer activity, including POU homeodomain, Myb, Ets, Forkhead, and T-box motifs. Empirical testing also revealed that the T-box TF encoded by org-1 is a previously uncharacterized regulator of muscle cell identity. Finally, we found extensive diversity in the composition of TFBSs within known FC enhancers, suggesting that motif combinatorics plays an essential role in the cellular specificity exhibited by such enhancers. In summary, machine learning combined with evolutionary sequence analysis is useful for recognizing novel TFBSs and for facilitating the identification of cognate TFs that coordinate cell type-specific developmental gene expression patterns.

Molecular mechanism underlying the regulatory specificity of a Drosophila homeodomain protein that specifies myoblast identity.

A subfamily of Drosophila homeodomain (HD) transcription factors (TFs) controls the identities of individual muscle founder cells (FCs). However, the molecular mechanisms by which these TFs generate unique FC genetic programs remain unknown. To investigate this problem, we first applied genome-wide mRNA expression profiling to identify genes that are activated or repressed by the muscle HD TFs Slouch (Slou) and Muscle segment homeobox (Msh). Next, we used protein-binding microarrays to define the sequences that are bound by Slou, Msh and other HD TFs that have mesodermal expression. These studies revealed that a large class of HDs, including Slou and Msh, predominantly recognize TAAT core sequences but that each HD also binds to unique sites that deviate from this canonical motif. To understand better the regulatory specificity of an individual FC identity HD, we evaluated the functions of atypical binding sites that are preferentially bound by Slou relative to other HDs within muscle enhancers that are either activated or repressed by this TF. These studies showed that Slou regulates the activities of particular myoblast enhancers through Slou-preferred sequences, whereas swapping these sequences for sites that are capable of binding to multiple HD family members does not support the normal regulatory functions of Slou. Moreover, atypical Slou-binding sites are overrepresented in putative enhancers associated with additional Slou-responsive FC genes. Collectively, these studies provide new insights into the roles of individual HD TFs in determining cellular identity, and suggest that the diversity of HD binding preferences can confer regulatory specificity.

Toward a systems-level understanding of developmental regulatory networks.

Developmental regulatory networks constitute all the interconnections among molecular components that guide embryonic development. Developmental transcriptional regulatory networks (TRNs) are circuits of transcription factors and cis-acting DNA elements that control expression of downstream regulatory and effector genes. Developmental networks comprise functional subnetworks that are deployed sequentially in requisite spatiotemporal patterns. Here, we discuss integrative genomics approaches for elucidating TRNs, with an emphasis on those involved in Drosophila mesoderm development and mammalian embryonic stem cell maintenance and differentiation. As examples of regulatory subnetworks, we consider the transcriptional and signaling regulation of genes that interact to control cell morphology and migration. Finally, we describe integrative experimental and computational strategies for defining the entirety of molecular interactions underlying developmental regulatory networks.

Expression-guided in silico evaluation of candidate cis regulatory codes for Drosophila muscle founder cells.

While combinatorial models of transcriptional regulation can be inferred for metazoan systems from a priori biological knowledge, validation requires extensive and time-consuming experimental work. Thus, there is a need for computational methods that can evaluate hypothesized cis regulatory codes before the difficult task of experimental verification is undertaken. We have developed a novel computational framework (termed "CodeFinder") that integrates transcription factor binding site and gene expression information to evaluate whether a hypothesized transcriptional regulatory model (TRM; i.e., a set of co-regulating transcription factors) is likely to target a given set of co-expressed genes. Our basic approach is to simultaneously predict cis regulatory modules (CRMs) associated with a given gene set and quantify the enrichment for combinatorial subsets of transcription factor binding site motifs comprising the hypothesized TRM within these predicted CRMs. As a model system, we have examined a TRM experimentally demonstrated to drive the expression of two genes in a sub-population of cells in the developing Drosophila mesoderm, the somatic muscle founder cells. This TRM was previously hypothesized to be a general mode of regulation for genes expressed in this cell population. In contrast, the present analyses suggest that a modified form of this cis regulatory code applies to only a subset of founder cell genes, those whose gene expression responds to specific genetic perturbations in a similar manner to the gene on which the original model was based. We have confirmed this hypothesis by experimentally discovering six (out of 12 tested) new CRMs driving expression in the embryonic mesoderm, four of which drive expression in founder cells.

An integrated strategy for analyzing the unique developmental programs of different myoblast subtypes.

An important but largely unmet challenge in understanding the mechanisms that govern the formation of specific organs is to decipher the complex and dynamic genetic programs exhibited by the diversity of cell types within the tissue of interest. Here, we use an integrated genetic, genomic, and computational strategy to comprehensively determine the molecular identities of distinct myoblast subpopulations within the Drosophila embryonic mesoderm at the time that cell fates are initially specified. A compendium of gene expression profiles was generated for primary mesodermal cells purified by flow cytometry from appropriately staged wild-type embryos and from 12 genotypes in which myogenesis was selectively and predictably perturbed. A statistical meta-analysis of these pooled datasets--based on expected trends in gene expression and on the relative contribution of each genotype to the detection of known muscle genes--provisionally assigned hundreds of differentially expressed genes to particular myoblast subtypes. Whole embryo in situ hybridizations were then used to validate the majority of these predictions, thereby enabling true-positive detection rates to be estimated for the microarray data. This combined analysis reveals that myoblasts exhibit much greater gene expression heterogeneity and overall complexity than was previously appreciated. Moreover, it implicates the involvement of large numbers of uncharacterized, differentially expressed genes in myogenic specification and subsequent morphogenesis. These findings also underscore a requirement for considerable regulatory specificity for generating diverse myoblast identities. Finally, to illustrate how the developmental functions of newly identified myoblast genes can be efficiently surveyed, a rapid RNA interference assay that can be scored in living embryos was developed and applied to selected genes. This integrated strategy for examining embryonic gene expression and function provides a substantially expanded framework for further studies of this model developmental system.

An increased frequency of autoantibody-inducing CD4+ T cells in pre-diseased lupus-prone mice.

Pathogenic autoantibody production in murine models of lupus is dependent on autoreactive CD4+ helper T cells. However, the mechanisms which permit the selection and maintenance of this autoantibody-inducing CD4+ T-cell repertoire are currently unknown. We hypothesized that the peripheral CD4+ T-cell repertoire of lupus-prone mice was enriched with autoantibody-inducing specificities. To test this, we utilized the splenic focus assay to determine if pre-diseased lupus-prone (NZB x NZW)F(1) mice have an elevated frequency of autoreactive CD4+ T lymphocytes capable of supporting autoantibody production. The splenic focus limiting dilution assay permits anti-nuclear antibodies to be generated from contact-dependent T-B interactions in vitro. We show that young, pre-diseased lupus-prone mice have an elevated frequency of autoantibody-inducing CD4+ T cells. Interestingly, these autoantibody-inducing CD4+ T-cell responses are also present in the thymus. Therefore, an elevated frequency of autoantibody-inducing CD4+ T cells predisposes lupus-prone mice to the development of autoantibodies.

Activation of diverse repertoires of autoreactive T cells enhances the loss of anti-dsDNA B cell tolerance.

CD4+ helper T cells play a critical role in the production of the antinuclear autoantibodies that characterize systemic lupus erythematosus in mice and humans. A key issue is whether this help is derived from a diverse repertoire of autoreactive CD4+ T cells or from a select number of T cells of limited specificity. We used the chronic graft-versus-host disease model to define the diversity of the CD4+ T cell repertoire required to induce the autoantibody response. By transferring clonally restricted versus clonally diverse populations of MHC class II-reactive CD4+ T cells, we show that the loss of B cell tolerance to nuclear antigens has two distinct components with different CD4+ cell requirements. Activation of limited repertoires of CD4+ T cells was sufficient for the expansion of anergized anti-double-stranded DNA B cells and production of IgM autoantibodies. Unexpectedly, we found that CD4+ T cell diversity was necessary for CD4+ T cell trafficking into the follicle and for the generation of isotype-switched IgG autoantibodies. Importantly, combining two limited repertoires of T cells provides sufficient CD4+ T cell diversity to drive antinuclear Ab production. These data demonstrate that a diverse CD4+ T cell repertoire is required to generate a sustained effector B cell response capable of mediating systemic autoimmunity.

Antigen presentation by keratinocytes directs autoimmune skin disease.

The antigen-presenting cells that initiate and maintain MHC class II-associated organ-specific autoimmune diseases are poorly defined. We now describe a new T cell antigen receptor (TCR) transgenic (Tg) model of inflammatory skin disease in which keratinocytes activate and are the primary target of autoreactive CD4(+) T cells. We previously generated keratin 14 (K14)-A(beta)b mice expressing MHC class II only on thymic cortical epithelium. CD4(+) T cells from K14-A(beta)b mice fail to undergo negative selection and thus have significant autoreactivity. The TCR genes from an autoreactive K14-A(beta)b CD4 hybridoma were cloned to produce a TCR Tg mouse, 2-2-3. 2-2-3 TCR Tg cells are negatively selected in WT C57BL6 mice but not in 2-2-3K14-A(beta)b mice. Interestingly, a significant number of mice that express both the K14-A(beta)b transgene and the autoreactive 2-2-3 TCR spontaneously develop inflammatory skin disease with mononuclear infiltrates, induction of MHC class II expression on keratinocytes, and T helper 1 cytokines. Disease can be induced by skin inflammation but not solely by activation of T cells. Thus, cutaneous immunopathology can be directed through antigen presentation by tissue-resident keratinocytes to autoreactive TCR Tg CD4(+) cells.