Scalable integration of multiomic single-cell data using generative adversarial networks.

MOTIVATION: Single-cell profiling has become a common practice to investigate the complexity of tissues, organs, and organisms. Recent technological advances are expanding our capabilities to profile various molecular layers beyond the transcriptome such as, but not limited to, the genome, the epigenome, and the proteome. Depending on the experimental procedure, these data can be obtained from separate assays or the very same cells. Yet, integration of more than two assays is currently not supported by the majority of the computational frameworks avaiable. RESULTS: We here propose a Multi-Omic data integration framework based on Wasserstein Generative Adversarial Networks suitable for the analysis of paired or unpaired data with a high number of modalities (>2). At the core of our strategy is a single network trained on all modalities together, limiting the computational burden when many molecular layers are evaluated. AVAILABILITY AND IMPLEMENTATION: Source code of our framework is available at https://github.com/vgiansanti/MOWGAN.

Revealing and harnessing CD39 for the treatment of colorectal cancer and liver metastases by engineered T cells.

OBJECTIVE: Colorectal tumours are often densely infiltrated by immune cells that have a role in surveillance and modulation of tumour progression but are burdened by immunosuppressive signals, which might vary from primary to metastatic stages. Here, we deployed a multidimensional approach to unravel the T-cell functional landscape in primary colorectal cancers (CRC) and liver metastases, and genome editing tools to develop CRC-specific engineered T cells. DESIGN: We paired high-dimensional flow cytometry, RNA sequencing and immunohistochemistry to describe the functional phenotype of T cells from healthy and neoplastic tissue of patients with primary and metastatic CRC and we applied lentiviral vectors (LV) and CRISPR/Cas9 genome editing technologies to develop CRC-specific cellular products. RESULTS: We found that T cells are mainly localised at the front edge and that tumor-infiltrating T cells co-express multiple inhibitory receptors, which largely differ from primary to metastatic sites. Our data highlighted CD39 as the major driver of exhaustion in both primary and metastatic colorectal tumours. We thus simultaneously redirected T-cell specificity employing a novel T-cell receptor targeting HER-2 and disrupted the endogenous TCR genes (TCR editing (TCRED)) and the CD39 encoding gene (ENTPD1), thus generating TCREDENTPD1KOHER-2-redirected lymphocytes. We showed that the absence of CD39 confers to HER-2-specific T cells a functional advantage in eliminating HER-2+ patient-derived organoids in vitro and in vivo. CONCLUSION: HER-2-specific CD39 disrupted engineered T cells are promising advanced medicinal products for primary and metastatic CRC.

Nizp1 is a specific NUP98-NSD1 functional interactor that regulates NUP98-NSD1-dependent oncogenic programs.

NSD1, NSD2 and NSD3 proteins constitute a family of histone 3 lysine 36 (H3K36) methyltransferases with similar domain architecture, but diversified activities, in part, dependent on their non-enzymatic domains. These domains, despite their high sequence identity, recruit the hosting proteins to different chromatin regions through the recognition of diverse epigenetic marks and/or associations to distinct interactors. In this sense, the PHDvC5HCH finger tandem domain represents a paradigmatic example of functional divergence within the NSD family. In this work, we prove and give a structural rationale for the uniqueness of the PHDvC5HCH domain of NSD1 in recognizing the C2HR Zinc finger domain of Nizp1 (NSD1 interacting Zn finger protein). Importantly, we show that, in a leukaemogenic context, Nizp1 is pivotal in driving the unscheduled expression of HoxA genes and of genes involved in the type I IFN pathway, triggered by the expression of the fusion protein NUP98-NSD1. These data provide the first insight into the pathophysiological relevance of the Nizp1-NSD1 functional association. Targeting of this interaction might open new therapeutic windows to inhibit the NUP98-NSD1 oncogenic properties.

A Novel RGD-4C-Saporin Conjugate Inhibits Tumor Growth in Mouse Models of Bladder Cancer.

Although toxin may have some advantages compared to chemotherapeutic drugs in cancer therapy, e.g. a potent cytotoxic activity and a reduced risk of resistance, their successful application in the treatments to solid tumors still remains to be fully demonstrated. In this study, we genetically modified the structure of the plant-derived single-chain ribosome inactivating protein saporin (SAP) by fusing its N-terminus to the ACDCRGDCFCG peptide (RGD-4C), an αv-integrin ligand, and explored the anti-tumor activity of the resulting protein (called RGD-SAP) in vitro and in vivo, using a model of muscle invasive bladder cancer. We found that the RGD-4C targeting domain enhances the cytotoxic activity of SAP against various tumor cell lines, in a manner dependent on αv-integrin expression levels. In a subcutaneous syngeneic model of bladder cancer, RGD-SAP significantly reduced tumor growth in a dose-dependent manner. Furthermore, systemic administration of RGD-SAP in combination with mitomycin C, a chemotherapeutic drug currently used to treat patients with bladder cancer, increased the survival of mice bearing orthotopic bladder cancer with no evidence of systemic toxicity. Overall, the results suggest that RGD-SAP represents an efficient drug that could be exploited, either alone or in combination with the state-of-the-art therapies, for the treatment of bladder cancer and, potentially, of other solid tumors.

Chromatin Velocity reveals epigenetic dynamics by single-cell profiling of heterochromatin and euchromatin.

Recent efforts have succeeded in surveying open chromatin at the single-cell level, but high-throughput, single-cell assessment of heterochromatin and its underlying genomic determinants remains challenging. We engineered a hybrid transposase including the chromodomain (CD) of the heterochromatin protein-1α (HP-1α), which is involved in heterochromatin assembly and maintenance through its binding to trimethylation of the lysine 9 on histone 3 (H3K9me3), and developed a single-cell method, single-cell genome and epigenome by transposases sequencing (scGET-seq), that, unlike single-cell assay for transposase-accessible chromatin with sequencing (scATAC-seq), comprehensively probes both open and closed chromatin and concomitantly records the underlying genomic sequences. We tested scGET-seq in cancer-derived organoids and human-derived xenograft (PDX) models and identified genetic events and plasticity-driven mechanisms contributing to cancer drug resistance. Next, building upon the differential enrichment of closed and open chromatin, we devised a method, Chromatin Velocity, that identifies the trajectories of epigenetic modifications at the single-cell level. Chromatin Velocity uncovered paths of epigenetic reorganization during stem cell reprogramming and identified key transcription factors driving these developmental processes. scGET-seq reveals the dynamics of genomic and epigenetic landscapes underlying any cellular processes.

Genomic Instability and Replicative Stress in Multiple Myeloma: The Final Curtain?

Multiple Myeloma (MM) is a genetically complex and heterogeneous hematological cancer that remains incurable despite the introduction of novel therapies in the clinic. Sadly, despite efforts spanning several decades, genomic analysis has failed to identify shared genetic aberrations that could be targeted in this disease. Seeking alternative strategies, various efforts have attempted to target and exploit non-oncogene addictions of MM cells, including, for example, proteasome inhibitors. The surprising finding that MM cells present rampant genomic instability has ignited concerted efforts to understand its origin and exploit it for therapeutic purposes. A credible hypothesis, supported by several lines of evidence, suggests that at the root of this phenotype there is intense replicative stress. Here, we review the current understanding of the role of replicative stress in eliciting genomic instability in MM and how MM cells rely on a single protein, Ataxia Telangiectasia-mutated and Rad3-related protein, ATR, to control and survive the ensuing, potentially fatal DNA damage. From this perspective, replicative stress per se represents not only an opportunity for MM cells to increase their evolutionary pool by increasing their genomic heterogeneity, but also a vulnerability that could be leveraged for therapeutic purposes to selectively target MM tumor cells.

ATR addiction in multiple myeloma: synthetic lethal approaches exploiting established therapies.

Therapeutic strategies designed to tinker with cancer cell DNA damage response have led to the widespread use of PARP inhibitors for BRCA1/2-mutated cancers. In the haematological cancer multiple myeloma, we sought to identify analogous synthetic lethality mechanisms that could be leveraged upon established cancer treatments. The combination of ATR inhibition using the compound VX-970 with a drug eliciting interstrand cross-links, melphalan, was tested in in vitro, ex vivo, and most notably in vivo models. Cell proliferation, induction of apoptosis, tumor growth and animal survival were assessed. The combination of ATM inhibition with a drug triggering double strand breaks, doxorucibin, was also probed. We found that ATR inhibition is strongly synergistic with melphalan, even in resistant cells. The combination was dramatically effective in targeting myeloma primary patient cells and cell lines reducing cell proliferation and inducing apoptosis. The combination therapy significantly reduced tumor burden and prolonged survival in animal models. Conversely, ATM inhibition only marginally impacted on myeloma cell survival, even in combination with doxorucibin at high doses. These results indicate that myeloma cells extensively rely on ATR, but not on ATM, for DNA repair. Our findings posit that adding an ATR inhibitor such as VX-970 to established therapeutic regimens may provide a remarkably broad benefit to myeloma patients.

Tranylcypromine-Based LSD1 Inhibitors: Structure-Activity Relationships, Antiproliferative Effects in Leukemia, and Gene Target Modulation.

LSD1 is a lysine demethylase highly involved in initiation and development of cancer. To design highly effective covalent inhibitors, a strategy is to fill its large catalytic cleft by designing tranylcypromine (TCP) analogs decorated with long, hindered substituents. We prepared three series of TCP analogs, carrying aroyl- and arylacetylamino (1 a-h), Z-amino acylamino (2 a-o), or double-substituted benzamide (3 a-n) residues at the C4 or C3 position of the phenyl ring. Further fragments obtained by chemical manipulation applied on the TCP scaffold (compounds 4 a-i) were also prepared. When tested against LSD1, most of 1 and 3 exhibited IC50 values in the low nanomolar range, with 1 e and 3 a,d,f,g being also the most selective respect to monoamine oxidases. In MV4-11 AML and NB4 APL cells compounds 3 were the most potent, displaying up to sub-micromolar cell growth inhibition against both cell lines (3 a) or against NB4 cells (3 c). The most potent compounds in cellular assays were also able to induce the expression of LSD1 target genes, such as GFI-1b, ITGAM, and KCTD12, as functional read-out for LSD1 inhibition. Mouse and human intrinsic clearance data highlighted the high metabolic stability of compounds 3 a, 3 d and 3 g. Further studies will be performed on the new compounds 3 a and 3 c to assess their anticancer potential in different cancer contexts.

Novel potent inhibitors of the histone demethylase KDM1A (LSD1), orally active in a murine promyelocitic leukemia model.

BACKGROUND: Histone lysine demethylases (KDMs) are well-recognized targets in oncology drug discovery. They function at the post-translation level controlling chromatin conformation and gene transcription. KDM1A is a flavin adenine dinucleotide-dependent amine oxidase, overexpressed in several tumor types, including acute myeloid leukemia, neuroblastoma and non-small-cell lung cancer. Among the many known monoamine oxidase inhibitors screened for KDM1A inhibition, tranylcypromine emerged as a moderately active hit, which irreversibly binds to the flavin adenine dinucleotide cofactor. MATERIAL & METHODS: The KDM1A inhibitors 5a-w were synthesized and tested in vitro and in vivo. The biochemical potency was determined, modulation of target in cells was demonstrated on KDM1A-dependent genes and the anti-clonogenic activity was performed in murine acute promyelocytic Leukemia (APL) blasts. An in vivo efficacy experiment was conducted using an established murine promyelocytic leukemia model. RESULTS: We report a new series of tranylcypromine derivatives substituted on the cyclopropyl moiety, endowed with high potency in both biochemical and cellular assays. CONCLUSION: The most interesting derivative (5a) significantly improved survival rate after oral administration in a murine model of promyelocitic leukemia.

Thieno[3,2-b]pyrrole-5-carboxamides as New Reversible Inhibitors of Histone Lysine Demethylase KDM1A/LSD1. Part 1: High-Throughput Screening and Preliminary Exploration.

Lysine specific demethylase 1 KDM1A (LSD1) regulates histone methylation and it is increasingly recognized as a potential therapeutic target in oncology. We report on a high-throughput screening campaign performed on KDM1A/CoREST, using a time-resolved fluorescence resonance energy transfer (TR-FRET) technology, to identify reversible inhibitors. The screening led to 115 hits for which we determined biochemical IC50, thus identifying four chemical series. After data analysis, we have prioritized the chemical series of N-phenyl-4H-thieno[3, 2-b]pyrrole-5-carboxamide for which we obtained X-ray structures of the most potent hit (compound 19, IC50 = 2.9 µM) in complex with the enzyme. Initial expansion of this chemical class, both modifying core structure and decorating benzamide moiety, was directed toward the definition of the moieties responsible for the interaction with the enzyme. Preliminary optimization led to compound 90, which inhibited the enzyme with a submicromolar IC50 (0.162 µM), capable of inhibiting the target in cells.

Thieno[3,2-b]pyrrole-5-carboxamides as New Reversible Inhibitors of Histone Lysine Demethylase KDM1A/LSD1. Part 2: Structure-Based Drug Design and Structure-Activity Relationship.

The balance of methylation levels at histone H3 lysine 4 (H3K4) is regulated by KDM1A (LSD1). KDM1A is overexpressed in several tumor types, thus representing an emerging target for the development of novel cancer therapeutics. We have previously described ( Part 1, DOI 10.1021.acs.jmedchem.6b01018 ) the identification of thieno[3,2-b]pyrrole-5-carboxamides as novel reversible inhibitors of KDM1A, whose preliminary exploration resulted in compound 2 with biochemical IC50 = 160 nM. We now report the structure-guided optimization of this chemical series based on multiple ligand/KDM1A-CoRest cocrystal structures, which led to several extremely potent inhibitors. In particular, compounds 46, 49, and 50 showed single-digit nanomolar IC50 values for in vitro inhibition of KDM1A, with high selectivity in secondary assays. In THP-1 cells, these compounds transcriptionally affected the expression of genes regulated by KDM1A such as CD14, CD11b, and CD86. Moreover, 49 and 50 showed a remarkable anticlonogenic cell growth effect on MLL-AF9 human leukemia cells.

Discovery of a Novel Inhibitor of Histone Lysine-Specific Demethylase 1A (KDM1A/LSD1) as Orally Active Antitumor Agent.

We report the stereoselective synthesis and biological activity of a novel series of tranylcypromine (TCPA) derivatives (14a-k, 15, 16), potent inhibitors of KDM1A. The new compounds strongly inhibit the clonogenic potential of acute leukemia cell lines. In particular three molecules (14d, 14e, and 14g) showing selectivity versus MAO A and remarkably inhibiting colony formation in THP-1 human leukemia cells, were assessed in mouse for their preliminary pharmacokinetic. 14d and 14e were further tested in vivo in a murine acute promyelocytic leukemia model, resulting 14d the most effective. Its two enantiomers were synthesized: the (1S,2R) enantiomer 15 showed higher activity than its (1R,2S) analogue 16, in both biochemical and cellular assays. Compound 15 exhibited in vivo efficacy after oral administration, determining a 62% increased survival in mouse leukemia model with evidence of KDM1A inhibition. The biological profile of compound 15 supports its further investigation as a cancer therapeutic.

Redox-Mediated Suberoylanilide Hydroxamic Acid Sensitivity in Breast Cancer.

AIMS: Vorinostat (suberoylanilide hydroxamic acid; SAHA) is a histone deacetylase inhibitor (HDACi) approved in the clinics for the treatment of T-cell lymphoma and with the potential to be effective also in breast cancer. We investigated the responsiveness to SAHA in human breast primary tumors and cancer cell lines. RESULTS: We observed a differential response to drug treatment in both human breast primary tumors and cancer cell lines. Gene expression analysis of the breast cancer cell lines revealed that genes involved in cell adhesion and redox pathways, especially glutathione metabolism, were differentially expressed in the cell lines resistant to SAHA compared with the sensitive ones, indicating their possible association with drug resistance mechanisms. Notably, such an association was also observed in breast primary tumors. Indeed, addition of buthionine sulfoximine (BSO), a compound capable of depleting cellular glutathione, significantly enhanced the cytotoxicity of SAHA in both breast cancer cell lines and primary breast tumors. INNOVATION: We identify and validate transcriptional differences in genes involved in redox pathways, which include potential predictive markers of sensitivity to SAHA. CONCLUSION: In breast cancer, it could be relevant to evaluate the expression of antioxidant genes that may favor tumor resistance as a factor to consider for potential clinical application and treatment with epigenetic drugs (HDACis).

Pure enantiomers of benzoylamino-tranylcypromine: LSD1 inhibition, gene modulation in human leukemia cells and effects on clonogenic potential of murine promyelocytic blasts.

The pure enantiomers of the N-(2-, 3-, and 4-(2-aminocyclopropyl)phenyl)benzamides hydrochlorides 11a-j were prepared and tested against LSD1 and MAO enzymes. The evaluation of the regioisomers 11a-j highlighted a net increase of the anti-LSD1 potency by shifting the benzamide moiety from ortho to meta and mainly to para position of tranylcypromine phenyl ring, independently from their trans or cis stereochemistry. In particular, the para-substituted 11a,b (trans) and 11g,h (cis) compounds displayed LSD1 and MAO-A inhibition at low nanomolar levels, while were less potent against MAO-B. The meta analogs 11c,d (trans) and 11i,j (cis) were in general less potent, but more efficient against MAO-A than against LSD1. In cellular assays, all the para and meta enantiomers were able to inhibit LSD1 by inducing Gfi-1b and ITGAM gene expression, with 11b,c and 11g-i giving the highest effects. Moreover, 11b and 11g,h strongly inhibited the clonogenic potential of murine promyelocytic blasts.

Pure Diastereomers of a Tranylcypromine-Based LSD1 Inhibitor: Enzyme Selectivity and In-Cell Studies.

The pure four diastereomers (11a-d) of trans-benzyl (1-((4-(2-aminocyclopropyl)phenyl)amino)-1-oxo-3-phenylpropan-2-yl)carbamate hydrochloride 11, previously described by us as LSD1 inhibitor, were obtained by enantiospecific synthesis/chiral HPLC separation method. Tested in LSD1 and MAO assays, 11b (S,1S,2R) and 11d (R,1S,2R) were the most potent isomers against LSD1 and were less active against MAO-A and practically inactive against MAO-B. In cells, all the four diastereomers induced Gfi-1b and ITGAM gene expression in NB4 cells, accordingly with their LSD1 inhibition, and 11b and 11d inhibited the colony forming potential in murine promyelocytic blasts.

Synthesis, biological activity and mechanistic insights of 1-substituted cyclopropylamine derivatives: a novel class of irreversible inhibitors of histone demethylase KDM1A.

Histone demethylase KDM1A (also known as LSD1) has become an attractive therapeutic target for the treatment of cancer as well as other disorders such as viral infections. We report on the synthesis of compounds derived from the expansion of tranylcypromine as a chemical scaffold for the design of novel demethylase inhibitors. These compounds, which are substituted on the cyclopropyl core moiety, were evaluated for their ability to inhibit KDM1A in vitro as well as to function in cells by modulating the expression of Gfi-1b, a well recognized KDM1A target gene. The molecules were all found to covalently inhibit KDM1A and to become increasingly selective against human monoamine oxidases MAO A and MAO B through the introduction of bulkier substituents on the cyclopropylamine ring. Structural and biochemical analysis of selected trans isomers showed that the two stereoisomers are endowed with similar inhibitory activities against KDM1A, but form different covalent adducts with the FAD co-enzyme.

A dual role for Hdac1: oncosuppressor in tumorigenesis, oncogene in tumor maintenance.

Aberrant recruitment of histone deacetylases (HDACs) by the oncogenic fusion protein PML-RAR is involved in the pathogenesis of acute promyelocytic leukemia (APL). PML-RAR, however, is not sufficient to induce disease in mice but requires additional oncogenic lesions during the preleukemic phase. Here, we show that knock-down of Hdac1 and Hdac2 dramatically accelerates leukemogenesis in transgenic preleukemic mice. These events are not restricted to APL because lymphomagenesis driven by deletion of p53 or, to a lesser extent, by c-myc overexpression, was also accelerated by Hdac1 knock-down. In the preleukemic phase of APL, Hdac1 counteracts the activity of PML-RAR in (1) blocking differentiation; (2) impairing genomic stability; and (3) increasing self-renewal in hematopoietic progenitors, as all of these events are affected by the reduction in Hdac1 levels. This led to an expansion of a subpopulation of PML-RAR-expressing cells that is the major source of leukemic stem cells in the full leukemic stage. Remarkably, short-term treatment of preleukemic mice with an HDAC inhibitor accelerated leukemogenesis. In contrast, knock-down of Hdac1 in APL mice led to enhanced survival duration of the leukemic animals. Thus, Hdac1 has a dual role in tumorigenesis: oncosuppressive in the early stages, and oncogenic in established tumor cells.

The DNA demethylating agent decitabine activates the TRAIL pathway and induces apoptosis in acute myeloid leukemia.

Although epigenetic drugs have been approved for use in selected malignancies, there is significant need for a better understanding of their mechanism of action. Here, we study the action of a clinically approved DNA-methyltransferase inhibitor - decitabine (DAC) - in acute myeloid leukemia (AML) cells. At low doses, DAC treatment induced apoptosis of NB4 Acute Promyelocytic Leukemia (APL) cells, which was associated with the activation of the extrinsic apoptotic pathway. Expression studies of the members of the Death Receptor family demonstrated that DAC induces the expression of TNF-related apoptosis-inducing ligand (TRAIL). Upregulation of TRAIL, upon DAC treatment, was associated with specific epigenetic modifications induced by DAC in the proximity of the TRAIL promoter, as demonstrated by DNA demethylation, increased DNaseI sensitivity and histone acetylation of a non-CpG island, CpG-rich region located 2kb upstream to the transcription start site. Luciferase assay experiments showed that this region behave as a DNA methylation sensitive transcriptional regulatory element. The CpG regulatory element was also found methylated in samples derived from APL patients. These findings have been confirmed in the non-APL, AML Kasumi cell line, suggesting that this regulatory mechanism may be extended to other AMLs. Our study suggests that DNA methylation is a regulatory mechanism relevant for silencing of the TRAIL apoptotic pathway in leukemic cells, and further elucidates the mechanism by which epigenetic drugs mediate their anti-leukemic effects.

HDACs link the DNA damage response, processing of double-strand breaks and autophagy.

Protein acetylation is mediated by histone acetyltransferases (HATs) and deacetylases (HDACs), which influence chromatin dynamics, protein turnover and the DNA damage response. ATM and ATR mediate DNA damage checkpoints by sensing double-strand breaks and single-strand-DNA-RFA nucleofilaments, respectively. However, it is unclear how acetylation modulates the DNA damage response. Here we show that HDAC inhibition/ablation specifically counteracts yeast Mec1 (orthologue of human ATR) activation, double-strand-break processing and single-strand-DNA-RFA nucleofilament formation. Moreover, the recombination protein Sae2 (human CtIP) is acetylated and degraded after HDAC inhibition. Two HDACs, Hda1 and Rpd3, and one HAT, Gcn5, have key roles in these processes. We also find that HDAC inhibition triggers Sae2 degradation by promoting autophagy that affects the DNA damage sensitivity of hda1 and rpd3 mutants. Rapamycin, which stimulates autophagy by inhibiting Tor, also causes Sae2 degradation. We propose that Rpd3, Hda1 and Gcn5 control chromosome stability by coordinating the ATR checkpoint and double-strand-break processing with autophagy.

Interplay between oncogene-induced DNA damage response and heterochromatin in senescence and cancer.

Two major mechanisms have been causally implicated in the establishment of cellular senescence: the activation of the DNA damage response (DDR) pathway and the formation of senescence-associated heterochromatic foci (SAHF). Here we show that in human fibroblasts resistant to premature p16(INK4a) induction, SAHF are preferentially formed following oncogene activation but are not detected during replicative cellular senescence or on exposure to a variety of senescence-inducing stimuli. Oncogene-induced SAHF formation depends on DNA replication and ATR (ataxia telangiectasia and Rad3-related). Inactivation of ATM (ataxia telangiectasia mutated) or p53 allows the proliferation of oncogene-expressing cells that retain increased heterochromatin induction. In human cancers, levels of heterochromatin markers are higher than in normal tissues, and are independent of the proliferative index or stage of the tumours. Pharmacological and genetic perturbation of heterochromatin in oncogene-expressing cells increase DDR signalling and lead to apoptosis. In vivo, a histone deacetylase inhibitor (HDACi) causes heterochromatin relaxation, increased DDR, apoptosis and tumour regression. These results indicate that heterochromatin induced by oncogenic stress restrains DDR and suggest that the use of chromatin-modifying drugs in cancer therapies may benefit from the study of chromatin and DDR status of tumours.