Multiplexed transcriptomic profiling of the fate of human CAR T cells in vivo via genetic barcoding with shielded small nucleotides.

The design of chimeric antigen receptor (CAR) T cells would benefit from knowledge of the fate of the cells in vivo. This requires the permanent labelling of CAR T cell products and their pooling in the same microenvironment. Here, we report a cell-barcoding method for the multiplexed longitudinal profiling of cells in vivo using single-cell RNA sequencing (scRNA-seq). The method, which we named shielded-small-nucleotide-based scRNA-seq (SSN-seq), is compatible with both 3' and 5' single-cell profiling, and enables the recording of cell identity, from cell infusion to isolation, by leveraging the ubiquitous Pol III U6 promoters to robustly express small-RNA barcodes modified with direct-capture sequences. By using SSN-seq to track the dynamics of the states of CAR T cells in a tumour-rechallenge mouse model of leukaemia, we found that a combination of cytokines and small-molecule inhibitors that are used in the ex vivo manufacturing of CAR T cells promotes the in vivo expansion of persistent populations of CD4+ memory T cells. By facilitating the probing of cell-state dynamics in vivo, SSN-seq may aid the development of adoptive cell therapies.

Elevated NSD3 histone methylation activity drives squamous cell lung cancer.

Amplification of chromosomal region 8p11-12 is a common genetic alteration that has been implicated in the aetiology of lung squamous cell carcinoma (LUSC)1-3. The FGFR1 gene is the main candidate driver of tumorigenesis within this region4. However, clinical trials evaluating FGFR1 inhibition as a targeted therapy have been unsuccessful5. Here we identify the histone H3 lysine 36 (H3K36) methyltransferase NSD3, the gene for which is located in the 8p11-12 amplicon, as a key regulator of LUSC tumorigenesis. In contrast to other 8p11-12 candidate LUSC drivers, increased expression of NSD3 correlated strongly with its gene amplification. Ablation of NSD3, but not of FGFR1, attenuated tumour growth and extended survival in a mouse model of LUSC. We identify an LUSC-associated variant NSD3(T1232A) that shows increased catalytic activity for dimethylation of H3K36 (H3K36me2) in vitro and in vivo. Structural dynamic analyses revealed that the T1232A substitution elicited localized mobility changes throughout the catalytic domain of NSD3 to relieve auto-inhibition and to increase accessibility of the H3 substrate. Expression of NSD3(T1232A) in vivo accelerated tumorigenesis and decreased overall survival in mouse models of LUSC. Pathological generation of H3K36me2 by NSD3(T1232A) reprograms the chromatin landscape to promote oncogenic gene expression signatures. Furthermore, NSD3, in a manner dependent on its catalytic activity, promoted transformation in human tracheobronchial cells and growth of xenografted human LUSC cell lines with amplification of 8p11-12. Depletion of NSD3 in patient-derived xenografts from primary LUSCs containing NSD3 amplification or the NSD3(T1232A)-encoding variant attenuated neoplastic growth in mice. Finally, NSD3-regulated LUSC-derived xenografts were hypersensitive to bromodomain inhibition. Thus, our work identifies NSD3 as a principal 8p11-12 amplicon-associated oncogenic driver in LUSC, and suggests that NSD3-dependency renders LUSC therapeutically vulnerable to bromodomain inhibition.

SETD5-Coordinated Chromatin Reprogramming Regulates Adaptive Resistance to Targeted Pancreatic Cancer Therapy.

Molecular mechanisms underlying adaptive targeted therapy resistance in pancreatic ductal adenocarcinoma (PDAC) are poorly understood. Here, we identify SETD5 as a major driver of PDAC resistance to MEK1/2 inhibition (MEKi). SETD5 is induced by MEKi resistance and its deletion restores refractory PDAC vulnerability to MEKi therapy in mouse models and patient-derived xenografts. SETD5 lacks histone methyltransferase activity but scaffolds a co-repressor complex, including HDAC3 and G9a. Gene silencing by the SETD5 complex regulates known drug resistance pathways to reprogram cellular responses to MEKi. Pharmacological co-targeting of MEK1/2, HDAC3, and G9a sustains PDAC tumor growth inhibition in vivo. Our work uncovers SETD5 as a key mediator of acquired MEKi therapy resistance in PDAC and suggests a context for advancing MEKi use in the clinic.

High-throughput quantitative histology in systemic sclerosis skin disease using computer vision.

BACKGROUND: Skin fibrosis is the clinical hallmark of systemic sclerosis (SSc), where collagen deposition and remodeling of the dermis occur over time. The most widely used outcome measure in SSc clinical trials is the modified Rodnan skin score (mRSS), which is a semi-quantitative assessment of skin stiffness at seventeen body sites. However, the mRSS is confounded by obesity, edema, and high inter-rater variability. In order to develop a new histopathological outcome measure for SSc, we applied a computer vision technology called a deep neural network (DNN) to stained sections of SSc skin. We tested the hypotheses that DNN analysis could reliably assess mRSS and discriminate SSc from normal skin. METHODS: We analyzed biopsies from two independent (primary and secondary) cohorts. One investigator performed mRSS assessments and forearm biopsies, and trichrome-stained biopsy sections were photomicrographed. We used the AlexNet DNN to generate a numerical signature of 4096 quantitative image features (QIFs) for 100 randomly selected dermal image patches/biopsy. In the primary cohort, we used principal components analysis (PCA) to summarize the QIFs into a Biopsy Score for comparison with mRSS. In the secondary cohort, using QIF signatures as the input, we fit a logistic regression model to discriminate between SSc vs. control biopsy, and a linear regression model to estimate mRSS, yielding Diagnostic Scores and Fibrosis Scores, respectively. We determined the correlation between Fibrosis Scores and the published Scleroderma Skin Severity Score (4S) and between Fibrosis Scores and longitudinal changes in mRSS on a per patient basis. RESULTS: In the primary cohort (n = 6, 26 SSc biopsies), Biopsy Scores significantly correlated with mRSS (R = 0.55, p = 0.01). In the secondary cohort (n = 60 SSc and 16 controls, 164 biopsies; divided into 70% training and 30% test sets), the Diagnostic Score was significantly associated with SSc-status (misclassification rate = 1.9% [training], 6.6% [test]), and the Fibrosis Score significantly correlated with mRSS (R = 0.70 [training], 0.55 [test]). The DNN-derived Fibrosis Score significantly correlated with 4S (R = 0.69, p = 3 × 10- 17). CONCLUSIONS: DNN analysis of SSc biopsies is an unbiased, quantitative, and reproducible outcome that is associated with validated SSc outcomes.

METTL13 Methylation of eEF1A Increases Translational Output to Promote Tumorigenesis.

Increased protein synthesis plays an etiologic role in diverse cancers. Here, we demonstrate that METTL13 (methyltransferase-like 13) dimethylation of eEF1A (eukaryotic elongation factor 1A) lysine 55 (eEF1AK55me2) is utilized by Ras-driven cancers to increase translational output and promote tumorigenesis in vivo. METTL13-catalyzed eEF1A methylation increases eEF1A's intrinsic GTPase activity in vitro and protein production in cells. METTL13 and eEF1AK55me2 levels are upregulated in cancer and negatively correlate with pancreatic and lung cancer patient survival. METTL13 deletion and eEF1AK55me2 loss dramatically reduce Ras-driven neoplastic growth in mouse models and in patient-derived xenografts (PDXs) from primary pancreatic and lung tumors. Finally, METTL13 depletion renders PDX tumors hypersensitive to drugs that target growth-signaling pathways. Together, our work uncovers a mechanism by which lethal cancers become dependent on the METTL13-eEF1AK55me2 axis to meet their elevated protein synthesis requirement and suggests that METTL13 inhibition may constitute a targetable vulnerability of tumors driven by aberrant Ras signaling.

Leveraging heterogeneity across multiple datasets increases cell-mixture deconvolution accuracy and reduces biological and technical biases.

In silico quantification of cell proportions from mixed-cell transcriptomics data (deconvolution) requires a reference expression matrix, called basis matrix. We hypothesize that matrices created using only healthy samples from a single microarray platform would introduce biological and technical biases in deconvolution. We show presence of such biases in two existing matrices, IRIS and LM22, irrespective of deconvolution method. Here, we present immunoStates, a basis matrix built using 6160 samples with different disease states across 42 microarray platforms. We find that immunoStates significantly reduces biological and technical biases. Importantly, we find that different methods have virtually no or minimal effect once the basis matrix is chosen. We further show that cellular proportion estimates using immunoStates are consistently more correlated with measured proportions than IRIS and LM22, across all methods. Our results demonstrate the need and importance of incorporating biological and technical heterogeneity in a basis matrix for achieving consistently high accuracy.

Interpretation of biological experiments changes with evolution of the Gene Ontology and its annotations.

Gene Ontology (GO) enrichment analysis is ubiquitously used for interpreting high throughput molecular data and generating hypotheses about underlying biological phenomena of experiments. However, the two building blocks of this analysis - the ontology and the annotations - evolve rapidly. We used gene signatures derived from 104 disease analyses to systematically evaluate how enrichment analysis results were affected by evolution of the GO over a decade. We found low consistency between enrichment analyses results obtained with early and more recent GO versions. Furthermore, there continues to be a strong annotation bias in the GO annotations where 58% of the annotations are for 16% of the human genes. Our analysis suggests that GO evolution may have affected the interpretation and possibly reproducibility of experiments over time. Hence, researchers must exercise caution when interpreting GO enrichment analyses and should reexamine previous analyses with the most recent GO version.

Antigen Identification for Orphan T Cell Receptors Expressed on Tumor-Infiltrating Lymphocytes.

The immune system can mount T cell responses against tumors; however, the antigen specificities of tumor-infiltrating lymphocytes (TILs) are not well understood. We used yeast-display libraries of peptide-human leukocyte antigen (pHLA) to screen for antigens of "orphan" T cell receptors (TCRs) expressed on TILs from human colorectal adenocarcinoma. Four TIL-derived TCRs exhibited strong selection for peptides presented in a highly diverse pHLA-A∗02:01 library. Three of the TIL TCRs were specific for non-mutated self-antigens, two of which were present in separate patient tumors, and shared specificity for a non-mutated self-antigen derived from U2AF2. These results show that the exposed recognition surface of MHC-bound peptides accessible to the TCR contains sufficient structural information to enable the reconstruction of sequences of peptide targets for pathogenic TCRs of unknown specificity. This finding underscores the surprising specificity of TCRs for their cognate antigens and enables the facile indentification of tumor antigens through unbiased screening.

Gene Expression Analysis to Assess the Relevance of Rodent Models to Human Lung Injury.

The relevance of animal models to human diseases is an area of intense scientific debate. The degree to which mouse models of lung injury recapitulate human lung injury has never been assessed. Integrating data from both human and animal expression studies allows for increased statistical power and identification of conserved differential gene expression across organisms and conditions. We sought comprehensive integration of gene expression data in experimental acute lung injury (ALI) in rodents compared with humans. We performed two separate gene expression multicohort analyses to determine differential gene expression in experimental animal and human lung injury. We used correlational and pathway analyses combined with external in vitro gene expression data to identify both potential drivers of underlying inflammation and therapeutic drug candidates. We identified 21 animal lung tissue datasets and three human lung injury bronchoalveolar lavage datasets. We show that the metasignatures of animal and human experimental ALI are significantly correlated despite these widely varying experimental conditions. The gene expression changes among mice and rats across diverse injury models (ozone, ventilator-induced lung injury, LPS) are significantly correlated with human models of lung injury (Pearson r = 0.33-0.45, P < 1E-16). Neutrophil signatures are enriched in both animal and human lung injury. Predicted therapeutic targets, peptide ligand signatures, and pathway analyses are also all highly overlapping. Gene expression changes are similar in animal and human experimental ALI, and provide several physiologic and therapeutic insights to the disease.

EMPOWERING MULTI-COHORT GENE EXPRESSION ANALYSIS TO INCREASE REPRODUCIBILITY.

A major contributor to the scientific reproducibility crisis has been that the results from homogeneous, single-center studies do not generalize to heterogeneous, real world populations. Multi-cohort gene expression analysis has helped to increase reproducibility by aggregating data from diverse populations into a single analysis. To make the multi-cohort analysis process more feasible, we have assembled an analysis pipeline which implements rigorously studied meta-analysis best practices. We have compiled and made publicly available the results of our own multi-cohort gene expression analysis of 103 diseases, spanning 615 studies and 36,915 samples, through a novel and interactive web application. As a result, we have made both the process of and the results from multi-cohort gene expression analysis more approachable for non-technical users.

Integrated, multicohort analysis of systemic sclerosis identifies robust transcriptional signature of disease severity.

Systemic sclerosis (SSc) is a rare autoimmune disease with the highest case-fatality rate of all connective tissue diseases. Current efforts to determine patient response to a given treatment using the modified Rodnan skin score (mRSS) are complicated by interclinician variability, confounding, and the time required between sequential mRSS measurements to observe meaningful change. There is an unmet critical need for an objective metric of SSc disease severity. Here, we performed an integrated, multicohort analysis of SSc transcriptome data across 7 datasets from 6 centers composed of 515 samples. Using 158 skin samples from SSc patients and healthy controls recruited at 2 centers as a discovery cohort, we identified a 415-gene expression signature specific for SSc, and validated its ability to distinguish SSc patients from healthy controls in an additional 357 skin samples from 5 independent cohorts. Next, we defined the SSc skin severity score (4S). In every SSc cohort of skin biopsy samples analyzed in our study, 4S correlated significantly with mRSS, allowing objective quantification of SSc disease severity. Using transcriptome data from the largest longitudinal trial of SSc patients to date, we showed that 4S allowed us to objectively monitor individual SSc patients over time, as (a) the change in 4S of a patient is significantly correlated with change in the mRSS, and (b) the change in 4S at 12 months of treatment could predict the change in mRSS at 24 months. Our results suggest that 4S could be used to distinguish treatment responders from nonresponders prior to mRSS change. Our results demonstrate the potential clinical utility of a novel robust molecular signature and a computational approach to SSc disease severity quantification.

CD47-blocking immunotherapies stimulate macrophage-mediated destruction of small-cell lung cancer.

Small-cell lung cancer (SCLC) is a highly aggressive subtype of lung cancer with limited treatment options. CD47 is a cell-surface molecule that promotes immune evasion by engaging signal-regulatory protein alpha (SIRPα), which serves as an inhibitory receptor on macrophages. Here, we found that CD47 is highly expressed on the surface of human SCLC cells; therefore, we investigated CD47-blocking immunotherapies as a potential approach for SCLC treatment. Disruption of the interaction of CD47 with SIRPα using anti-CD47 antibodies induced macrophage-mediated phagocytosis of human SCLC patient cells in culture. In a murine model, administration of CD47-blocking antibodies or targeted inactivation of the Cd47 gene markedly inhibited SCLC tumor growth. Furthermore, using comprehensive antibody arrays, we identified several possible therapeutic targets on the surface of SCLC cells. Antibodies to these targets, including CD56/neural cell adhesion molecule (NCAM), promoted phagocytosis in human SCLC cell lines that was enhanced when combined with CD47-blocking therapies. In light of recent clinical trials for CD47-blocking therapies in cancer treatment, these findings identify disruption of the CD47/SIRPα axis as a potential immunotherapeutic strategy for SCLC. This approach could enable personalized immunotherapeutic regimens in patients with SCLC and other cancers.

Coordination of stress signals by the lysine methyltransferase SMYD2 promotes pancreatic cancer.

Pancreatic ductal adenocarcinoma (PDAC) is a lethal form of cancer with few therapeutic options. We found that levels of the lysine methyltransferase SMYD2 (SET and MYND domain 2) are elevated in PDAC and that genetic and pharmacological inhibition of SMYD2 restricts PDAC growth. We further identified the stress response kinase MAPKAPK3 (MK3) as a new physiologic substrate of SMYD2 in PDAC cells. Inhibition of MAPKAPK3 impedes PDAC growth, identifying a potential new kinase target in PDAC. Finally, we show that inhibition of SMYD2 cooperates with standard chemotherapy to treat PDAC cells and tumors. These findings uncover a pivotal role for SMYD2 in promoting pancreatic cancer.

Combined inhibition of BET family proteins and histone deacetylases as a potential epigenetics-based therapy for pancreatic ductal adenocarcinoma.

Pancreatic ductal adenocarcinoma (PDAC) is one of the most lethal human cancers and shows resistance to any therapeutic strategy used. Here we tested small-molecule inhibitors targeting chromatin regulators as possible therapeutic agents in PDAC. We show that JQ1, an inhibitor of the bromodomain and extraterminal (BET) family of proteins, suppresses PDAC development in mice by inhibiting both MYC activity and inflammatory signals. The histone deacetylase (HDAC) inhibitor SAHA synergizes with JQ1 to augment cell death and more potently suppress advanced PDAC. Finally, using a CRISPR-Cas9-based method for gene editing directly in the mouse adult pancreas, we show that de-repression of p57 (also known as KIP2 or CDKN1C) upon combined BET and HDAC inhibition is required for the induction of combination therapy-induced cell death in PDAC. SAHA is approved for human use, and molecules similar to JQ1 are being tested in clinical trials. Thus, these studies identify a promising epigenetic-based therapeutic strategy that may be rapidly implemented in fatal human tumors.