ursaPGx: a new R package to annotate pharmacogenetic star alleles using phased whole-genome sequencing data

Long-read sequencing technologies offer new opportunities to generate high-confidence phased whole-genome sequencing data for robust pharmacogenetic annotation. Here, we describe a new user-friendly R package, ursaPGx, designed to accept multi-sample phased whole-genome sequencing data VCF input files and output star allele annotations for pharmacogenes annotated in PharmVar.

Star allele search: a pharmacogenetic annotation database and user-friendly search tool of publicly available 1000 Genomes Project biospecimens

Here we describe a new public pharmacogenetic (PGx) annotation database of a large (n = 3,202) and diverse biospecimen collection of 1000 Genomes Project cell lines and DNAs. The database is searchable with a user friendly, web-based tool ( www.coriell.org/StarAllele/Search ). This resource leverages existing whole genome sequencing data and PharmVar annotations to characterize *alleles for each biospecimen in the collection. This new tool is designed to facilitate in vitro functional characterization of *allele haplotypes and diplotypes as well as support clinical PGx assay development, validation, and implementation.

Tissue-infiltrating alloreactive T cells require Id3 to deflect PD-1-mediated immune suppression during GVHD

ABSTRACT

Persisting alloreactive donor T cells in target tissues are a determinant of graft-versus-host disease (GVHD), but the transcriptional regulators that control the persistence and function of tissue-infiltrating T cells remain elusive. We demonstrate here that Id3, a DNA-binding inhibitor, is critical for sustaining T-cell responses in GVHD target tissues in mice, including the liver and intestine. Id3 loss results in aberrantly expressed PD-1 in polyfunctional T helper 1 (Th1) cells, decreased tissue-infiltrating PD-1+ polyfunctional Th1 cell numbers, impaired maintenance of liver TCF-1+ progenitor-like T cells, and inhibition of GVHD. PD-1 blockade restores the capacity of Id3-ablated donor T cells to mediate GVHD. Single-cell RNA-sequencing analysis revealed that Id3 loss leads to significantly decreased CD28- and PI3K/AKT-signaling activity in tissue-infiltrating polyfunctional Th1 cells, an indicator of active PD-1/PD-L1 effects. Id3 is also required for protecting CD8+ T cells from the PD-1 pathway-mediated suppression during GVHD. Genome-wide RNA-sequencing analysis reveals that Id3 represses transcription factors (e.g., Nfatc2, Fos, Jun, Ets1, and Prdm1) that are critical for PD-1 transcription, exuberant effector differentiation, and interferon responses and dysfunction of activated T cells. Id3 achieves these effects by restraining the chromatin accessibility for these transcription factors. Id3 ablation in donor T cells preserved their graft vs tumor effects in mice undergoing allogeneic hematopoietic stem cell transplantation. Furthermore, CRISPR/Cas9 knockout of ID3 in human CD19-directed chimeric antigen receptor T cells retained their antitumor activity in NOD/SCID/IL2Rg-/- mice early after administration. These findings identify that ID3 is an important target to reduce GVHD, and the gene-editing program of ID3 may have broad implications in T-cell-based immunotherapy.

Non-pathogenic microbiota accelerate age-related CpG Island methylation in colonic mucosa

DNA methylation is an epigenetic process altered in cancer and ageing. Age-related methylation drift can be used to estimate lifespan and can be influenced by extrinsic factors such as diet. Here, we report that non-pathogenic microbiota accelerate age-related methylation drift in the colon when compared with germ-free mice. DNA methylation analyses showed that microbiota and IL10KO were associated with changes in 5% and 4.1% of CpG sites, while mice with both factors had 18% alterations. Microbiota, IL10KO, and their combination altered 0.4%, 0.4%, and 4% of CpG island methylation, respectively. These are comparable to what is seen in colon cancer. Ageing changes were accelerated in the IL10KO mice with microbiota, and the affected genes were more likely to be altered in colon cancer. Thus, the microbiota affect DNA methylation of the colon in patterns reminiscent of what is observed in ageing and colorectal cancer.

Postmortem toxicology findings from the Camden Opioid Research Initiative

The United States continues to be impacted by decades of an opioid misuse epidemic, worsened by the COVID-19 pandemic and by the growing prevalence of highly potent synthetic opioids (HPSO) such as fentanyl. In instances of a toxicity event, first-response administration of reversal medications such as naloxone can be insufficient to fully counteract the effects of HPSO, particularly when there is co-occurring substance use. In an effort to characterize and study this multi-faceted problem, the Camden Opioid Research Initiative (CORI) has been formed. The CORI study has collected and analyzed post-mortem toxicology data from 42 cases of decedents who expired from opioid-related toxicity in the South New Jersey region to characterize substance use profiles. Co-occurring substance use, whether by intent or through possible contamination of the illicit opioid supply, is pervasive among deaths due to opioid toxicity, and evidence of medication-assisted treatment is scarce. Nearly all (98%) of the toxicology cases show the presence of the HPSO, fentanyl, and very few (7%) results detected evidence of medication-assisted treatment for opioid use disorder, such as buprenorphine or methadone, at the time of death. The opioid toxicity reversal drug, naloxone, was detected in 19% of cases, but 100% of cases expressed one or more stimulants, and sedatives including xylazine were detected in 48% of cases. These results showing complex substance use profiles indicate that efforts at mitigating the opioid misuse epidemic must address the complications presented by co-occurring stimulant and other substance use, and reduce barriers to and stigmas of seeking effective medication-assisted treatments.

Decitabine disrupts EBV genomic epiallele DNA methylation patterns around CTCF binding sites to increase chromatin accessibility and lytic transcription in gastric cancer

IMPORTANCE

Epstein-Barr virus (EBV) latency is controlled by epigenetic silencing by DNA methylation [5-methyl cytosine (5mC)], histone modifications, and chromatin looping. However, how they dictate the transcriptional program in EBV-associated gastric cancers remains incompletely understood. EBV-associated gastric cancer displays a 5mC hypermethylated phenotype. A potential treatment for this cancer subtype is the DNA hypomethylating agent, which induces EBV lytic reactivation and targets hypermethylation of the cellular DNA. In this study, we identified a heterogeneous pool of EBV epialleles within two tumor-derived gastric cancer cell lines that are disrupted with a hypomethylating agent. Stochastic DNA methylation patterning at critical regulatory regions may be an underlying mechanism for spontaneous reactivation. Our results highlight the critical role of epigenetic modulation on EBV latency and life cycle, which is maintained through the interaction between 5mC and the host protein CCCTC-binding factor.

TET1 and TDG Suppress Inflammatory Response in Intestinal Tumorigenesis: Implications for Colorectal Tumors With the CpG Island Methylator Phenotype

BACKGROUND & AIMS

Aberrant DNA methylation is frequent in colorectal cancer (CRC), but underlying mechanisms and pathologic consequences are poorly understood.

METHODS

We disrupted active DNA demethylation genes Tet1 and/or Tdg from ApcMin mice and characterized the methylome and transcriptome of colonic adenomas. Data were compared to human colonic adenocarcinomas (COAD) in The Cancer Genome Atlas.

RESULTS

There were increased numbers of small intestinal adenomas in ApcMin mice expressing the TdgN151A allele, whereas Tet1-deficient and Tet1/TdgN151A-double heterozygous ApcMin colonic adenomas were larger with features of erosion and invasion. We detected reduction in global DNA hypomethylation in colonic adenomas from Tet1- and Tdg-mutant ApcMin mice and hypermethylation of CpG islands in Tet1-mutant ApcMin adenomas. Up-regulation of inflammatory, immune, and interferon response genes was present in Tet1- and Tdg-mutant colonic adenomas compared to control ApcMin adenomas. This up-regulation was also seen in murine colonic organoids and human CRC lines infected with lentiviruses expressing TET1 or TDG short hairpin RNA. A 127-gene inflammatory signature separated colonic adenocarcinomas into 4 groups, closely aligned with their microsatellite or chromosomal instability and characterized by different levels of DNA methylation and DNMT1 expression that anticorrelated with TET1 expression. Tumors with the CpG island methylator phenotype (CIMP) had concerted high DNMT1/low TET1 expression. TET1 or TDG knockdown in CRC lines enhanced killing by natural killer cells.

CONCLUSIONS

Our findings reveal a novel epigenetic regulation, linked to the type of genomic instability, by which TET1/TDG-mediated DNA demethylation decreases methylation levels and inflammatory/interferon/immune responses. CIMP in CRC is triggered by an imbalance of methylating activities over demethylating activities. These mice represent a model of CIMP CRC.

Promoter-Independent Synthesis of Chemically Modified RNA by Polymerase θ Variants

Synthetic RNA oligonucleotides composed of canonical and modified ribonucleotides are highly effective for RNA antisense therapeutics and RNA based genome engineering applications utilizing CRISPR-Cas9. Yet, synthesis of synthetic RNA using phosphoramidite chemistry is highly inefficient and expensive relative to DNA oligonucleotides, especially for relatively long RNA oligonucleotides. Thus, new biotechnologies are needed to significantly reduce costs, while increasing synthesis rates and yields of synthetic RNA. Here, we engineer human DNA polymerase theta (Polθ) variants and demonstrate their ability to synthesize long (95-200 nt) RNA oligonucleotides with canonical ribonucleotides and ribonucleotide analogs commonly used for stabilizing RNA for therapeutic and genome engineering applications. In contrast to natural promoter-dependent RNA polymerases, Polθ variants synthesize RNA by initiating from DNA or RNA primers which enables the production of highly pure RNA products. Remarkably, Polθ variants show lower capacity to misincorporate ribonucleotides compared to T7 RNA polymerase. Automation of this enzymatic RNA synthesis technology can potentially increase yields while reducing costs of synthetic RNA oligonucleotide production.

Haploinsufficiency of ZNF251 causes DNA-PKcs-dependent resistance to PARP inhibitors in BRCA1-mutated cancer cells

Poly (ADP-ribose) polymerase (PARP) inhibitors represent a promising new class of agents that have demonstrated efficacy in treating various cancers, particularly those that carry BRCA1/2 mutations. The cancer associated BRCA1/2 mutations disrupt DNA double strand break (DSB) repair by homologous recombination (HR). PARP inhibitors (PARPis) have been applied to trigger synthetic lethality in BRCA1/2-mutated cancer cells by promoting the accumulation of toxic DSBs. Unfortunately, resistance to PARPis is common and can occur through multiple mechanisms, including the restoration of HR and/or the stabilization of replication forks. To gain a better understanding of the mechanisms underlying PARPi resistance, we conducted an unbiased CRISPR-pooled genome-wide library screen to identify new genes whose deficiency confers resistance to the PARPi olaparib. Our study revealed that ZNF251, a transcription factor, is a novel gene whose haploinsufficiency confers PARPi resistance in multiple breast and ovarian cancer lines harboring BRCA1 mutations. Mechanistically, we discovered that ZNF251 haploinsufficiency leads to constitutive stimulation of DNA-PKcs-dependent non-homologous end joining (NHEJ) repair of DSBs and DNA-PKcs-mediated fork protection in BRCA1-mutated cancer cells (BRCA1mut + ZNF251KD). Moreover, we demonstrated that DNA-PKcs inhibitors can restore PARPi sensitivity in BRCA1mut + ZNF251KD cells ex vivo and in vivo. Our findings provide important insights into the mechanisms underlying PARPi resistance and highlight the unexpected role of DNA-PKcs in this phenomenon.

Novel treatment strategy of targeting epigenetic dysregulation in pancreatic neuroendocrine tumors

BACKGROUND

Epigenetic dysregulation is an integral step in the progression of pancreatic neuroendocrine tumors. We hypothesized that tumor suppressor repression by DNA methyltransferase 1 in pancreatic neuroendocrine tumors could be targeted with epigenetic treatment.

METHODS

Resected pancreatic neuroendocrine tumors from 32 patients were stained for DNA methyltransferase 1 and scored. Human (BON1) and murine (STC) pancreatic neuroendocrine tumor cells were treated with DNA methyltransferase 1 inhibitor 5-azacytidine and chemotherapeutic agents 5-fluorouracil and temozolomide. Cell proliferation assay and tumor suppressor gene analysis were performed with qRT-PCR and Clarion S microarray. Tumor measurements were compared in a murine treatment model.

RESULTS

High DNA methyltransferase scores were associated with high Ki-67 (6.7% vs 70.6% P < .01), mitotic rate (0.0% vs 31.3%), and grade (20.0% vs 80.4%, P < .01). Treatment with 5-azacytidine and chemotherapy resulted in a reduction of cell proliferation compared to chemotherapy alone in BON1 (77.3% vs 53.1%, P < .001) and STC (73.4% vs 34.2%, P < .001). Treatment with 5-azacytidine and chemotherapy resulted in upregulation of tumor suppressors CDKN1A (7.6 rel. fold, P < .001), BRCA2 (4.3 rel. fold, P < .001), and CDH1 (6.0 rel. fold, P = .026) in BON1 and CDKN1a (14.5 rel. fold, P < .001) and CDH (17.5 rel. fold, P < .001) in STC. In microarray, 5-azacytidine drove global genetic changes in combination treatment. In vivo tumors treated with chemotherapy measured 88.6 ± 19.54 mm³ vs 52.89 ± 10.51 mm³ in those treated with combination therapy (P = .009).

CONCLUSION

Epigenetic dysregulation with DNA methyltransferase 1 is associated with pancreatic neuroendocrine tumors and is a potential targetable strategy. 5-azacytidine and chemotherapy in combination can reduce cell proliferation, upregulate silenced tumor suppressor genes, and decrease in vivo tumors in pancreatic neuroendocrine tumors.

ABL1 kinase as a tumor suppressor in AML1-ETO and NUP98-PMX1 leukemias

Deletion of ABL1 was detected in a cohort of hematologic malignancies carrying AML1-ETO and NUP98 fusion proteins. Abl1-/- murine hematopoietic cells transduced with AML1-ETO and NUP98-PMX1 gained proliferation advantage when compared to Abl1 + /+ counterparts. Conversely, overexpression and pharmacological stimulation of ABL1 kinase resulted in reduced proliferation. To pinpoint mechanisms facilitating the transformation of ABL1-deficient cells, Abl1 was knocked down in 32Dcl3-Abl1ko cells by CRISPR/Cas9 followed by the challenge of growth factor withdrawal. 32Dcl3-Abl1ko cells but not 32Dcl3-Abl1wt cells generated growth factor-independent clones. RNA-seq implicated PI3K signaling as one of the dominant mechanisms contributing to growth factor independence in 32Dcl3-Abl1ko cells. PI3K inhibitor buparlisib exerted selective activity against Lin-cKit+ NUP98-PMX1;Abl1-/- cells when compared to the Abl1 + /+ counterparts. Since the role of ABL1 in DNA damage response (DDR) is well established, we also tested the inhibitors of ATM (ATMi), ATR (ATRi) and DNA-PKcs (DNA-PKi). AML1-ETO;Abl1-/- and NUP98-PMX1;Abl1-/- cells were hypersensitive to DNA-PKi and ATRi, respectively, when compared to Abl1 + /+ counterparts. Moreover, ABL1 kinase inhibitor enhanced the sensitivity to PI3K, DNA-PKcs and ATR inhibitors. In conclusion, we showed that ABL1 kinase plays a tumor suppressor role in hematological malignancies induced by AML1-ETO and NUP98-PMX1 and modulates the response to PI3K and/or DDR inhibitors.

A Phase II Trial of Guadecitabine plus Atezolizumab in Metastatic Urothelial Carcinoma Progressing after Initial Immune Checkpoint Inhibitor Therapy

PURPOSE

Based on preclinical evidence of epigenetic contribution to sensitivity and resistance to immune checkpoint inhibitors (ICI), we hypothesized that guadecitabine (hypomethylating agent) and atezolizumab (anti-PD-L1) together would potentiate a clinical response in patients with metastatic urothelial carcinoma (UC) unresponsive to initial immune checkpoint blockade therapy.

PATIENTS AND METHODS

We designed a single arm Phase II study (NCT03179943) with a safety run-in to identify the recommended phase II dose of the combination therapy of guadecitabine and atezolizumab. Patients with recurrent/advanced urothelial carcinoma who had previously progressed on ICI therapy with PD-1 or PD-L1 targeting agents were eligible. Pre-planned correlative analysis was performed to characterize peripheral immune dynamics and global DNA methylation, transcriptome, and immune infiltration dynamics of patient tumors.

RESULTS

Safety run-in enrolled 6 patients and Phase II enrolled 15 patients before the trial was closed for futility. No dose-limiting toxicity was observed. Four patients, with best response of stable disease, exhibited extended tumor control (8-11 months) and survival (>14 months). Correlative analysis revealed lack of DNA demethylation in tumors after 2 cycles of treatment. Increased peripheral immune activation and immune infiltration in tumors after treatment correlated with progression-free survival and stable disease. Furthermore, high IL-6 and IL-8 levels in the patients' plasma associates with short survival.

CONCLUSIONS

No RECIST responses were observed after combination therapy in this trial. Although we could not detect the anticipated tumor-intrinsic effects of guadecitabine, the addition of hypomethylating agent to ICI therapy induced immune activation in a few patients, which associated with longer patient survival.

DNA methylation entropy as a measure of stem cell replication and aging

BACKGROUND

Epigenetic marks are encoded by DNA methylation and accumulate errors as organisms age. This drift correlates with lifespan, but the biology of how this occurs is still unexplained. We analyze DNA methylation with age in mouse intestinal stem cells and compare them to nonstem cells.

RESULTS

Age-related changes in DNA methylation are identical in stem and nonstem cells, affect most prominently CpG islands and correlate weakly with gene expression. Age-related DNA methylation entropy, measured by the Jensen-Shannon Distribution, affects up to 25% of the detectable CpG sites and is a better measure of aging than individual CpG methylation. We analyze this entropy as a function of age in seven other tissues (heart, kidney, skeletal muscle, lung, liver, spleen, and blood) and it correlates strikingly with tissue-specific stem cell division rates. Thus, DNA methylation drift and increased entropy with age are primarily caused by and are sensors for, stem cell replication in adult tissues.

CONCLUSIONS

These data have implications for the mechanisms of tissue-specific functional declines with aging and for the development of DNA-methylation-based biological clocks.

Transcriptome-guided resolution of tumor microenvironment interactions in pheochromocytoma and paraganglioma subtypes

BACKGROUND

Pheochromocytomas and paragangliomas (PCPG) are rare catecholamine-secreting endocrine tumors deriving from chromaffin cells of the embryonic neural crest. Although distinct molecular PCPG subtypes have been elucidated, certain characteristics of these tumors have yet to be fully examined, namely the tumor microenvironment (TME). To further understand tumor-stromal interactions in PCPG subtypes, the present study deconvoluted bulk tumor gene expression to examine ligand-receptor interactions.

METHODS

RNA-sequencing data primary solid PCPG tumors were derived from The Cancer Genome Atlas (TCGA). Tumor purity was estimated using two robust algorithms. The tumor purity estimates and bulk tumor expression values allowed for non-negative linear regression to predict the average expression of each gene in the stromal and tumor compartments for each PCPG molecular subtype. The predicted expression values were then used in conjunction with a previously curated ligand-receptor database and scoring system to evaluate top ligand-receptor interactions.

RESULTS

Across all PCPG subtypes compared to normal samples, tumor-to-tumor signaling between bone morphogenic proteins 7 (BMP7) and 15 (BMP15) and cognate receptors ACVR2B and BMPR1B was increased. In addition, tumor-to-stroma signaling was enriched for interactions between predicted tumor-originating delta-like ligand 3 (DLL3) and predicted stromal NOTCH receptors. Stroma-to-tumor signaling was enriched for interactions between ephrins A1 and A4 with ephrin receptors EphA5, EphA7, and EphA8. Pseudohypoxia subtype tumors displayed increased predicted stromal expression of genes related to immune-exhausted T-cell response, including those for inhibitory receptors HAVCR2 and CTLA4.

CONCLUSION

The current exploratory study predicted stromal and tumor through compartmental deconvolution and yielded previously unrecognized interactions and putative biomarkers in PCPG.

The nuclear lamina binds the EBV genome during latency and regulates viral gene expression

The Epstein Barr virus (EBV) infects almost 95% of the population worldwide. While typically asymptomatic, EBV latent infection is associated with several malignancies of epithelial and lymphoid origin in immunocompromised individuals. In latently infected cells, the EBV genome persists as a chromatinized episome that expresses a limited set of viral genes in different patterns, referred to as latency types, which coincide with varying stages of infection and various malignancies. We have previously demonstrated that latency types correlate with differences in the composition and structure of the EBV episome. Several cellular factors, including the nuclear lamina, regulate chromatin composition and architecture. While the interaction of the viral genome with the nuclear lamina has been studied in the context of EBV lytic reactivation, the role of the nuclear lamina in controlling EBV latency has not been investigated. Here, we report that the nuclear lamina is an essential epigenetic regulator of the EBV episome. We observed that in B cells, EBV infection affects the composition of the nuclear lamina by inducing the expression of lamin A/C, but only in EBV+ cells expressing the Type III latency program. Using ChIP-Seq, we determined that lamin B1 and lamin A/C bind the EBV genome, and their binding correlates with deposition of the histone repressive mark H3K9me2. By RNA-Seq, we observed that knock-out of lamin A/C in B cells alters EBV gene expression. Our data indicate that the interaction between lamins and the EBV episome contributes to the epigenetic control of viral gene expression during latency, suggesting a restrictive function of the nuclear lamina as part of the host response against viral DNA entry into the nucleus.

Epstein-Barr virus (EBV) is a common herpesvirus that establishes a lifelong latent infection in a small fraction of B cells of the infected individuals. In most cases, EBV infection is asymptomatic; however, especially in the context of immune suppression, EBV latent infection is associated with several malignancies. In EBV+ cancer cells, latent viral gene expression plays an essential role in sustaining the cancer phenotype. We and others have established that epigenetic modifications of the viral genome are critical to regulating EBV gene expression during latency. Understanding how the EBV genome is epigenetically regulated during latent infection may help identify new specific therapeutic targets for treating EBV+ malignancies. The nuclear lamina is involved in regulating the composition and structure of the cellular chromatin. In the present study, we determined that the nuclear lamina binds the EBV genome during latency, influencing viral gene expression. Depleting one component of the nuclear lamina, lamin A/C, increased the expression of latent EBV genes associated with cellular proliferation, indicating that the binding of the nuclear lamina with the viral genome is essential to control viral gene expression in infected cells. Our data show for the first time that the nuclear lamina may be involved in the cellular response against EBV infection by restricting viral gene expression.

The three-dimensional structure of Epstein-Barr virus genome varies by latency type and is regulated by PARP1 enzymatic activity

Epstein-Barr virus (EBV) persists in human B-cells by maintaining its chromatinized episomes within the nucleus. We have previously shown that cellular factor Poly [ADP-ribose] polymerase 1 (PARP1) binds the EBV genome, stabilizes CTCF binding at specific loci, and that PARP1 enzymatic activity correlates with maintaining a transcriptionally active latency program. To better understand PARP1's role in regulating EBV latency, here we functionally characterize the effect of PARP enzymatic inhibition on episomal structure through in situ HiC mapping, generating a complete 3D structure of the EBV genome. We also map intragenomic contact changes after PARP inhibition to global binding of chromatin looping factors CTCF and cohesin across the EBV genome. We find that PARP inhibition leads to fewer total unique intragenomic interactions within the EBV episome, yet new chromatin loops distinct from the untreated episome are also formed. This study also illustrates that PARP inhibition alters gene expression at the regions where chromatin looping is most effected. We observe that PARP1 inhibition does not alter cohesin binding sites but does increase its frequency of binding at those sites. Taken together, these findings demonstrate that PARP has an essential role in regulating global EBV chromatin structure and latent gene expression.

TET2 and DNMT3A Mutations Exert Divergent Effects on DNA Repair and Sensitivity of Leukemia Cells to PARP Inhibitors

Somatic variants in TET2 and DNMT3A are founding mutations in hematological malignancies that affect the epigenetic regulation of DNA methylation. Mutations in both genes often co-occur with activating mutations in genes encoding oncogenic tyrosine kinases such as FLT3ITD, BCR-ABL1, JAK2V617F , and MPLW515L , or with mutations affecting related signaling pathways such as NRASG12D and CALRdel52 . Here, we show that TET2 and DNMT3A mutations exert divergent roles in regulating DNA repair activities in leukemia cells expressing these oncogenes. Malignant TET2-deficient cells displayed downregulation of BRCA1 and LIG4, resulting in reduced activity of BRCA1/2-mediated homologous recombination (HR) and DNA-PK-mediated non-homologous end-joining (D-NHEJ), respectively. TET2-deficient cells relied on PARP1-mediated alternative NHEJ (Alt-NHEJ) for protection from the toxic effects of spontaneous and drug-induced DNA double-strand breaks. Conversely, DNMT3A-deficient cells favored HR/D-NHEJ owing to downregulation of PARP1 and reduction of Alt-NHEJ. Consequently, malignant TET2-deficient cells were sensitive to PARP inhibitor (PARPi) treatment in vitro and in vivo, whereas DNMT3A-deficient cells were resistant. Disruption of TET2 dioxygenase activity or TET2-Wilms' tumor 1 (WT1)-binding ability was responsible for DNA repair defects and sensitivity to PARPi associated with TET2 deficiency. Moreover, mutation or deletion of WT1 mimicked the effect of TET2 mutation on DSB repair activity and sensitivity to PARPi. Collectively, these findings reveal that TET2 and WT1 mutations may serve as biomarkers of synthetic lethality triggered by PARPi, which should be explored therapeutically. SIGNIFICANCE: TET2 and DNMT3A mutations affect distinct DNA repair mechanisms and govern the differential sensitivities of oncogenic tyrosine kinase-positive malignant hematopoietic cells to PARP inhibitors.

Abstract CT121: A Phase II trial of guadecitabine (G) plus atezolizumab (A) in patients with metastatic urothelial carcinoma (UC) progressing after initial checkpoint inhibitor therapy

Introduction: We report the results of a phase II trial testing the hypothesis that adding the hypomethylating agent G to the PDL1 inhibitor A in patients with mUC who developed primary or acquired resistance to checkpoint blockade (CB) will overcome this resistance by (1) eliciting viral mimicry in the tumor tissue to potentiate and reinvigorate anti-tumor immunity, (2) epigenetic reprogramming of T lymphocytes to overcome exhaustion.

Methods: Pts with mUC resistant to CB accrued at 3 centers were treated with G 45mg/m2 daily days 1-5 every 6 wks and A 1200mg every 3 wks. After initial safety lead in with 6 patients, trial was designed to add 37 additional patients. The primary endpoint was ORR. Correlative analyses included analysis of peripheral blood T cells and tumor tissue collected at baseline and once during treatment.

Results: 21 pts were enrolled. 20 pts were evaluable for response. Best response was PD (16), SD (4). 10 patients progressed clinically prompting earlier than scheduled (12 wk) imaging. Four pts exhibited a “hyperprogression” phenotype exhibiting rapid acceleration of tumor growth rate starting with initiation of therapy. At presepcified interim analysis it was determined that the trial would not meet its primary endpoint and it closed early. Median PFS 2.6 mo, median OS 8 mo. The 4 patients with SD maintained that status for median 13 months (range 9-15 mo). Global DNA methylome and transcriptome profiles from pre- and post-treatment tumor samples revealed a lack of transposable element-induced viral mimicry activation, which correlated with minimal DNA demethylation being induced in the tumors. Of note, flow cytometry-based immune profiling of peripheral blood from patients suggests a correlation between increased progression-free survival (PFS) with 1) lower expression of DNAM-1 on mature NK cells and 2) lower expression of CD39 on CD8+ effector T cells at time of inclusion on the trial.

Conclusions: While no responses were seen, both prolonged SD and hyperprogression were seen. Further tissue and peripheral blood based analyses are ongoing to elucidate the biological determinates of this dichotomy.

Citation Format: Elizabeth R. Plimack, Kerry Campbell, Jean-Pierre J. Issa, Noah M. Hahn, David I. Quinn, Hyo Sik Jang, Galen Hostetter, Peter W. Nichols, Woonbok Chung, Jozef Madzo, Hitoshi Ohtani, Hui Shen, Toshinori Hinoue, Stephen B. Baylin, Peter A. Jones. A Phase II trial of guadecitabine (G) plus atezolizumab (A) in patients with metastatic urothelial carcinoma (UC) progressing after initial checkpoint inhibitor therapy [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr CT121.

The developmental origins of sex-biased expression in cardiac development

BACKGROUND

Expression patterns between males and females vary in every adult tissue, even in organs with no conspicuous dimorphisms such as the heart. While studies of male and female differences have traditionally focused on the influence of sex hormones, these do not account for all the differences at the molecular and epigenetic levels. We previously reported that a substantial number of genes were differentially expressed in male and female mouse embryonic stem (ES) cells and revealed dose-dependent enhancer activity in response to Prdm14, a key pluripotency factor expressed more highly in female ES cells. In this work, we investigated the role of Prdm14 in establishing sex-specific gene expression networks. We surveyed the sex-specific landscape in early embryogenesis with special reference to cardiac development. We generated sex-specific co-expression networks from mouse ES cells, examined the presence of sex-specific chromatin domains, and analyzed previously published datasets from different developmental time points to characterize how sex-biased gene expression waxes and wanes to evaluate whether sex-biased networks are detectable throughout heart development.

RESULTS

We performed ChIP-seq on male and female mouse ES cells to determine differences in chromatin status. Our study reveals sex-biased histone modifications, underscoring the potential for the sex chromosome complement to prime the genome differently in early development with consequences for later expression biases. Upon differentiation of ES cells to cardiac precursors, we found sex-biased expression of key transcription and epigenetic factors, some of which persisted from the undifferentiated state. Using network analyses, we also found that Prdm14 plays a prominent role in regulating a subset of dimorphic expression patterns. To determine whether sex-biased expression is present throughout cardiogenesis, we re-analyzed data from two published studies that sampled the transcriptomes of mouse hearts from 8.5 days post-coitum embryos to neonates and adults. We found sex-biased expression at every stage in heart development, and interestingly, identified a subset of genes that exhibit the same bias across multiple cardiogenic stages.

CONCLUSIONS

Overall, our results support the existence of sexually dimorphic gene expression profiles and regulatory networks at every stage of cardiac development, some of which may be established in early embryogenesis and epigenetically perpetuated.

Abstract 4331: Potential role of the splicing factor SF3B1 in epigenetic regulation and activation of p53 signaling

To develop novel and improved cancer therapies, we aim to target the epigenome, which is often deregulated in cancer. We identified the splicing factor SF3B1 as a potential epigenetic regulator in a whole genome siRNA screen for reactivation of aberrantly silenced genes alone or in synergy with the DNMT inhibitor Decitabine (DAC). The screen was done in YB5 colon cancer cells which contain a methylated and silenced CMV promoter driving GFP expression, and confirmed in a similar model in HCT116, another colon cancer cell line. SF3B1 has been studied in splicing, but little is known about its effects on gene regulation outside of this function, prompting us to further investigate this. We validated the screen findings by siRNA (siSF3B1) and inhibition of SF3B1 with Pladienolide B (PB). We then performed RNA-Seq and Reduced Representation Bisulfite Sequencing (RRBS) in cells treated with siSF3B1 and DAC, and we performed ChIP-Seq to assess SF3B1 binding. By flow cytometry, we measured a ~7% and ~5% induction of GFP in YB5 cells after siSF3B1 and DAC, respectively. The combination treatment caused a synergistic increase to ~15% GFP positive cells (p &lt; 0.001). siSF3B1 led to 423 up and 338 downregulated genes, while DAC induced 430 up and 135 downregulated genes. With the combination, there was activation of 1190 genes and downregulation of 904 genes (fold change &gt; 2, FDR &lt; 0.1). There were 695, 119, 1584 genes alternatively spliced following siSF3B1, DAC, and the combination treatment, respectively, but there was no significant overlap with the regulated genes (p &lt; 0.05), suggesting a distinct mechanism for gene expression regulation. There were two major subsets of siSF3B1 upregulated genes. A set of genes had low promoter methylation (0-20% methylation) and were p53 activation targets such as CDKN1A and GADD45A. The other set of genes had high promoter methylation (80-100% methylation) and promoters containing TATA-box motifs. RRBS showed global DNA hypomethylation after DAC treatment as expected, with an average methylation of 76.2% compared to 84.7% methylation in siControl cells. Further demethylation occurred in the combination treated cells, which had an average methylation of 72.7% (p&lt;0.001). ChIP-Seq showed differential binding of SF3B1 at the transcription start sites (TSSs) of genes based on their expression. SF3B1 was depleted at TSSs of expressed genes and enriched at nonexpressed genes, resembling the binding pattern of histone H3. Finally, SF3B1 inhibition with PB alone and with DAC also induced reactivation of gene expression and altered DNA methylation. Together, these findings suggest that SF3B1 may be playing a role in gene regulation outside of its role in splicing. In cancer, SF3B1 transcriptional target genes are potentially tumor suppressor genes and upon knockdown or inhibition of SF3B1, their expression is reactivated leading to antitumor effects.

Citation Format: Sandra Deliard, Yasuyuki Okamoto, Jozef Madzo, Somnath Pandey, Jaroslav Jelinek, Jean-Pierre Issa. Potential role of the splicing factor SF3B1 in epigenetic regulation and activation of p53 signaling [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 4331.

Abstract P3-05-03: Identification of epigenetically silenced breast cancer driver genes

Breast cancer is clinically and molecularly complex disease driven by aberrant genetic and epigenetic alterations. Epigenetic alterations in particular DNA methylation changes are one of the most important events involved in breast cancer initiation and progression. Previous reports identified many aberrant DNA methylation signatures associated with molecular subtypes of breast cancer and over 100 candidate genes with promoter hypermethylation in breast cancer. However, it remains elusive which of these genes with promoter hypermethylation play “driver” role in tumorigenesis. In previous studies, the average gain of DNA methylation across all cancer samples compared to the average DNA methylation in normal samples has been the criterion to select for potential targets. However, known tumor suppressor driver genes regulated by methylation are relatively infrequently altered in target cancers. Therefore, we propose the paradoxical hypothesis that identifying hypermethylated cancer drivers require focusing on infrequent rather than frequent events. Hence, to identify these potential driver genes, we developed an algorithm with two unique properties. First, unlike previous studies we focused on targets that gained DNA methylation relatively infrequently (10-40%) and that lost expression in breast cancer. Second, using this algorithm, we distinguished cancer dependent gain of DNA methylation from age-dependent gain of methylation. To discern age dependent and independent DNA methylation changes, we generated DNA methylation sequencing data on 29 normal purified breast epithelium (age range 33-82 years old). Furthermore, to study the biological effects of the overexpression or downregulation of these genes, we generated DNA methylation sequencing data on 6 breast cancer cell lines. We also used DNA methylation and RNA expression datasets (675 cancer, 100 normal) available through the TCGA. Using our algorithm, we identified 53 genes with age independent promoter hypermethylation and loss of expression in TCGA tumor samples. To begin testing the biological effects of these driver genes, we performed canonical pathway enrichment analyses using Ingenuity Pathway Analysis software. We also investigated the mutational status of these genes and their molecular subtype enrichment. Based on these analyses, we picked 12 genes (C10orf125, RUNX3, YOD1, FXYD5, SMOC1, SLC16A5, RNLS, DKK1, PNPLA3, FZD10, RND2, and PLCB1) for further study. We stably overexpressed these potential driver genes in different breast cancer cell lines. Twelve genes out of the 12 tested, slowed cell proliferation and 9 decreased anchorage independent growth. We further validated these driver genes by knocking them out in normal human mammary epithelial cells using CRISPR/Cas9 tool. The loss of these genes, increased cell proliferation rate in normal human mammary epithelial cells compared to the control cells. In conclusion, based on our preliminary data, using bioinformatics tools as well as functional assays, we identified epigenetically altered breast cancer driver genes. Identifying and deciphering true epigenetic cancer drivers could potentially lead to the development of therapeutic drugs targeting these genes and/or targeting pathway dependence.

Citation Format: Panjarian S, Madzo J, Slater C, Jelinek J, Chen X, Issa J-P. Identification of epigenetically silenced breast cancer driver genes [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P3-05-03.

Demethylator phenotypes in acute myeloid leukemia

Acute myeloid leukemia (AML) often harbors mutations in epigenetic regulators, and also has frequent DNA hypermethylation, including the presence of CpG island methylator phenotypes (CIMPs). Although global hypomethylation is well known in cancer, the question of whether distinct demethylator phenotypes (DMPs) exist remains unanswered. Using Illumina 450k arrays for 194 patients from The Cancer Genome Atlas, we identified two distinct DMPs by hierarchical clustering: DMP.1 and DMP.2. DMP.1 cases harbored mutations in NPM1 (94%), FLT3 (71%) and DNMT3A (61%). Surprisingly, only 40% of patients with DNMT3A mutations were DMP.1, which has implications for mechanisms of transformation by this mutation. In contrast, DMP.2 AML was comprised of patients with t(8;21), inv(16) or t(15;17), suggesting common methylation defects connect these disparate rearrangements. RNA-seq revealed upregulated genes functioning in immune response (DMP.1) and development (DMP.2). We confirmed these findings by integrating independent 450k data sets (236 additional cases), and found prognostic effects by DMP status, independent of age and cytogenetics. The existence of DMPs has implications for AML pathogenesis and may augment existing tools in risk stratification.

Abstract 2952: Targeting CDK9 reactivates epigenetically silenced genes in cancer

In cancer, the epigenome is aberrantly reprogrammed leading to a wide range of heritable changes in gene expression, such as silencing of tumor suppressor genes (TSG). Altered epigenetic marks in cancer involve DNA methylation and histone post-translational modifications, and these come about as a result of aging and acquisition of genetic and epigenetic changes in readers/writers/editors of the epigenome. Given the reversible nature of epigenetic modifications, one goal of epigenetic therapy of cancer is to induce TSG reactivation, leading to cancer cell differentiation and cancer cell death.

To identify novel targets that can reactivate epigenetically silenced genes, we developed a phenotypic-based system, YB5. YB5 is a colon cancer cell line generated by stably transfecting SW48 cells with a vector containing GFP driven by a methylated and silenced CMV promoter. GFP re-expression can be achieved by known epigenetic drugs that lead to demethylation or induce active chromatin marks in the CMV promoter.

We screened a natural compound library for GFP activation in YB5 and identified a novel drug class that shares an aminothiazole core structure, and has epigenetic effects that are equivalent to DNA methyltransferase inhibitor (DNMTi). Target deconvolution identified CDK9 as the target of these drugs, which reactivate gene expression without affecting DNA methylation. It is well established that CDK9, the catalytic subunit of p-TEFb, is a transcriptional activator. CDK9 in complex with its regulatory subunit, Cyclin T1 or T2, is recruited by multiple mechanisms to promote RNAPII promoter-proximal pause release by phosphorylating negative elongation factors (DSIF and NELF). In addition, phosphorylation of the C-terminal domain (CTD) of RNAPII on Serine-2 allows recruitment of RNA processing factors, which work on the nascent RNA as it emerges from RNAPII. Our new data show that long-term CDK9 inhibition can reactivate epigenetically silenced genes with minimal downregulation effects, effects which are the opposite of the canonical role of CDK9-mediated transcriptional elongation. Mechanistically, we showed that CDK9 inhibition dephosphorylates the SWI/SNF protein SMARCA4 and represses HP1α expression, both of which contribute to gene reactivation. Based on gene activation, we developed the highly selective and potent CDK9 inhibitor MC180295 (IC50 =5nM) that has broad anti-cancer activity in-vitro and is effective in in-vivo cancer models. Additionally, CDK9 inhibition sensitizes with the immune checkpoint inhibitor α-PD-1 in vivo, making it an excellent target for epigenetic therapy of cancer. This is the first study that links CDK9 to maintaining gene silencing at epigenetically repressed loci in mammals. Excitingly, this is also the first example of kinase inhibitors as potential drugs that reverse epigenetic silencing.

Citation Format: Hanghang Zhang, Somnath Pandey, Meghan Travers, Jittasak Khowsathit, George Morton, Hongxing Sum, Carlos A. Barrero, Carmen Merali, Yasuyuki Okamoto, Takahiro Sato, Judit Garriga, Natarajan V. Bhanu, Johayra Simithy, Bela Patel, Jozef Madzo, Noël Raynal, Benjamin A. Garcia, Marlene A. Jacobson, Salim Merali, Yi Zhang, Wayne Childers, Magid Abou-Gharbia, John Karanicolas, Stephen B. Baylin, Cynthia A. Zahnow, Jaroslav Jelinek, Xavier Graña, Jean-Pierre J. Issa. Targeting CDK9 reactivates epigenetically silenced genes in cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 2952.

TET1-Mediated Hypomethylation Activates Oncogenic Signaling in Triple-Negative Breast Cancer

Both gains and losses of DNA methylation are common in cancer, but the factors controlling this balance of methylation remain unclear. Triple-negative breast cancer (TNBC), a subtype that does not overexpress hormone receptors or HER2/NEU, is one of the most hypomethylated cancers observed. Here, we discovered that the TET1 DNA demethylase is specifically overexpressed in about 40% of patients with TNBC, where it is associated with hypomethylation of up to 10% of queried CpG sites and a worse overall survival. Through bioinformatic analyses in both breast and ovarian cancer cell line panels, we uncovered an intricate network connecting TET1 to hypomethylation and activation of cancer-specific oncogenic pathways, including PI3K, EGFR, and PDGF. TET1 expression correlated with sensitivity to drugs targeting the PI3K-mTOR pathway, and CRISPR-mediated deletion of TET1 in two independent TNBC cell lines resulted in reduced expression of PI3K pathway genes, upregulation of immune response genes, and substantially reduced cellular proliferation, suggesting dependence of oncogenic pathways on TET1 overexpression. Our work establishes TET1 as a potential oncogene that contributes to aberrant hypomethylation in cancer and suggests that TET1 could serve as a druggable target for therapeutic intervention.Significance: This study addresses a critical gap in knowledge of how and why methylation is prognostic in breast cancer and shows how this information can be used to stratify patients with TNBC for targeted therapy. Cancer Res; 78(15); 4126-37. ©2018 AACR.

DREAM: A Simple Method for DNA Methylation Profiling by High-throughput Sequencing

The digital restriction enzyme analysis of methylation (DREAM) is a simple method for DNA methylation analysis at tens of thousands of CpG sites across the genome. The method creates specific signatures at unmethylated and methylated CpG sites by sequential digests of genomic DNA with restriction endonucleases SmaI and XmaI, respectively. Both enzymes have the same CCCGGG recognition site; however, they differ in their sensitivity to CpG methylation and their cutting pattern. SmaI cuts only unmethylated sites leaving blunt 5'-GGG ends. XmaI cuts remaining methylated CC(me)CGG sites leaving 5'-CCGGG ends. Restriction fragments with distinct signatures at their ends are ligated to Illumina sequencing adaptors with sample-specific barcodes. High-throughput sequencing of pooled libraries follows. Sequencing reads are mapped to the restriction sites in the reference genome, and signatures corresponding to methylation status of individual DNA molecules are resolved. Methylation levels at target CpG sites are calculated as the proportion of sequencing reads with the methylated signature to the total number of reads mapping to the particular restriction site. Aligning the reads to the reference genome of any species is straightforward, since the method does not rely on bisulfite conversion of DNA. Sequencing of 25 million reads per human DNA library yields over 50,000 unique CpG sites with high coverage enabling accurate determination of DNA methylation levels. DREAM has a background less than 1 % making it suitable for accurate detection of low methylation levels. In summary, the method is simple, robust, highly reproducible, and cost-effective.

Ezh2 phosphorylation state determines its capacity to maintain CD8+ T memory precursors for antitumor immunity

Memory T cells sustain effector T-cell production while self-renewing in reaction to persistent antigen; yet, excessive expansion reduces memory potential and impairs antitumor immunity. Epigenetic mechanisms are thought to be important for balancing effector and memory differentiation; however, the epigenetic regulator(s) underpinning this process remains unknown. Herein, we show that the histone methyltransferase Ezh2 controls CD8⁺ T memory precursor formation and antitumor activity. Ezh2 activates Id3 while silencing Id2, Prdm1 and Eomes, promoting the expansion of memory precursor cells and their differentiation into functional memory cells. Akt activation phosphorylates Ezh2 and decreases its control of these transcriptional programs, causing enhanced effector differentiation at the expense of T memory precursors. Engineering T cells with an Akt-insensitive Ezh2 mutant markedly improves their memory potential and capability of controlling tumor growth compared to transiently inhibiting Akt. These findings establish Akt-mediated phosphorylation of Ezh2 as a critical target to potentiate antitumor immunotherapeutic strategies.

During an immune response naive CD8⁺ T cells can differentiate into either effector or memory T cells. Here the authors show that Akt-mediated phosphorylation of the epigenetic regulator Ezh2 is critical for the generation of an anti-tumor CD8 T cell response and promotes the expansion of memory-precursors.

A novel isoform of TET1 that lacks a CXXC domain is overexpressed in cancer

TET1 oxidizes methylated cytosine into 5-hydroxymethylcytosine (5hmC), resulting in regulation of DNA methylation and gene expression. Full length TET1 (TET1FL) has a CXXC domain that binds to unmethylated CpG islands (CGIs). This CXXC domain allows TET1 to protect CGIs from aberrant methylation, but it also limits its ability to regulate genes outside of CGIs. Here, we report a novel isoform of TET1 (TET1ALT) that has a unique transcription start site from an alternate promoter in intron 2, yielding a protein with a unique translation start site. Importantly, TET1ALT lacks the CXXC domain but retains the catalytic domain. TET1ALT is repressed in embryonic stem cells (ESCs) but becomes activated in embryonic and adult tissues while TET1FL is expressed in ESCs, but repressed in adult tissues. Overexpression of TET1ALT shows production of 5hmC with distinct (and weaker) effects on DNA methylation or gene expression when compared to TET1FL. TET1ALT is aberrantly activated in multiple cancer types including breast, uterine and glioblastoma, and TET1 activation is associated with a worse overall survival in breast, uterine and ovarian cancers. Our data suggest that the predominantly activated isoform of TET1 in cancer cells does not protect from CGI methylation and likely mediates dynamic site-specific demethylation outside of CGIs.

Abstract LB-100: Potential role of the splicing factor SF3B1 in epigenetic regulation

Driver mutations in splicing factors have been reported in multiple cancer types and SF3B1 is among the most frequently mutated splicing factors. The mechanism(s) of transformation associated with these mutations remain incompletely understood. The role of SF3B1 has been studied in splicing, but little is known about its effects on gene regulation outside of its splicing function in cancer. We identified SF3B1 as a potential epigenetic regulator in a whole genome siRNA screen for reactivation of aberrantly silenced genes alone or in synergy with DNMT inhibitors. This surprising link between splicing factors and epigenetic silencing by DNA methylation and chromatin deregulation could explain in part the effects of SF3B1 in cancer. To confirm this link, we studied effects of SF3B1 on gene expression and DNA methylation, and determined SF3B1 chromatin localization by ChIP-seq. We first used RNA-Seq data from all available Uveal Melanoma (UVM) and breast cancer patients in TCGA. Hierarchical cluster analyses of the UVM data showed distinct clusters with high expression of SF3B1 clustering independently of low SF3B1 expression in both cancer types. We identified genes co-regulated with SF3B1 expression in UVM and breast cancer and observed a significant overlap of 40% of upregulated genes (p &lt; 0.0001) and 41.1% of down regulated genes (p &lt; 0.0001) in patients with low SF3B1 expression in UVM and breast cancer. This similarity in the co-regulated gene sets suggests a possible direct regulation of gene expression by SF3B1. Next, we analyzed DNA methylation patterns in UVM samples stratified by SF3B1 expression and found that in patients with high SF3B1 expression relative to low SF3B1 expression, there was a significant increase of methylation at 755 CpG sites (Δ β-value&gt; 0.1, p&lt;0.001) and decrease of methylation at 42 CpGs (Δ β-value&lt; -0.1, p&lt;0.001) of all (&gt;450,000) CpG sites tested. At promoter CpG sites, 197 CpG sites had significantly increased methylation (Δ β-value&gt; 0.1, p&lt;0.001) while only 10 CpG sites had decreased methylation (Δ β-value&lt; -0.1, p&lt;0.001). Thus, high levels of SF3B1 are associated with increased DNA methylation, and inhibition of SF3B1 is associated with reactivation of gene expression from methylated promoters. Finally, we performed ChIP-Seq analyses for SF3B1 in YB5 colon cancer cells. In addition to SF3B1 enrichment at exons, we observed that SF3B1 was depleted at transcription start sites (TSSs) overall, but stratifying genes by expression reveled SF3B1 depletion at TSSs of expressed genes but not in unexpressed genes. This was similar to the pattern seen for histone H3 ChIP-Seq, suggesting SF3B1 association with nucleosomes. Taken together, our data suggest that SF3B1 plays a role in DNA methylation associated epigenetic silencing. In cancers, these silenced genes are potentially tumor suppressor genes which upon knockdown or inhibition of SF3B1, their expression could be reactivated leading to antitumor effects.

Citation Format: Sandra Deliard, Yasuyuki Okamoto, Jozef Madzo, Jaroslav Jelinek, Jean-Pierre Issa. Potential role of the splicing factor SF3B1 in epigenetic regulation [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr LB-100. doi:10.1158/1538-7445.AM2017-LB-100

Abstract 3359: TET1 mediated hypomethylation activates oncogenic signaling pathways in triple negative breast cancer

Triple negative breast cancer (TNBC) is a subtype of breast cancer that occurs in 15-20% of patients, and is defined by tumors that do not overexpress the estrogen, progesterone and HER2 receptors. This aggressive subtype has a significantly worse overall survival compared to non-TNBC and importantly, these patients lack options for targeted therapy. To identify novel subtype-specific therapeutic targets, we must first understand the biological underpinnings of the disease. DNA methylation is a hallmark of cancer, as it can regulate gene expression of both tumor suppressor genes and oncogenes. We have found that TN tumors have widespread genome-wide hypomethylation when compared to other breast cancer subtypes and normal breast controls. TET1 is a DNA demethylase that converts 5-methyl cytosine into 5-hydroxymethyl cytosine, which can be further oxidized into un-methylated cytosine. To determine if TET1 could possibly play a role in TNBC hypomethylation, we analyzed genome-wide DNA methylation, DNA mutation and RNA-seq datasets from the TCGA breast cancer cohort. We identified a subset of TN patients that upregulate TET1 and display DNA hypomethylation. To identify the hypomethylated TET1 target genes, we computed Spearman correlations between TET1 expression and methylation % for 450,000 sites (450K array) across 67 TNBC patients. Filtering for sites with r&lt;-0.3 revealed 42,559 sites negatively correlated with TET1 expression. Cluster analysis of the sites that lose methylation compared to normal breast (12,807 CpG sites) revealed two distinct clusters. Cluster 1 (42% of TNBC cases) were TET1 high and hypomethylated, while cluster 2 (58%) looked more like normal breast controls. Gene set enrichment analysis of the hypomethylated genes revealed Hippo Signaling, Pathways in Cancer and PI3K-Akt Signaling as significantly enriched, with p&lt;0.001. In addition, only 4% of patients in cluster 1 have mutations/genomic alterations in the PI3K pathway, compared to 29% of cluster 2 patients, p=0.01. Most strikingly, analysis of phosphorylated EIF4EBP1 RPPA proteomic data revealed TET1 high patients have increased PI3K pathway activity, even though they lack mutations in the pathway, p=0.02. We hypothesize that the PI3K hyper-activation, in part, can be explained by TET1 upregulation and target gene demethylation. In TET1 knock out MDA-MB-231 cells, we observed a reduction in phosphorylated 4E-BP1, suggesting loss of PI3K activity is concomitant with loss of TET1 as well as decreased cellular proliferation (p&lt;0.01). Furthermore, breast cancer cell lines with high TET1 (but not low TET1) are sensitive to drugs targeting the PI3K/ERK pathway (XMD8-85, AZ628, TGX221). In addition to explaining and predicting PI3K inhibitor sensitivity in breast cancer, our studies may establish TET1 as a TNBC specific oncogene that could serve as a novel druggable target for therapeutic intervention.

Citation Format: Charly Ryan Good, Andrew Kelly, Jozef Madzo, Jaroslav Jelinek, Jean-Pierre Issa. TET1 mediated hypomethylation activates oncogenic signaling pathways in triple negative breast cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 3359. doi:10.1158/1538-7445.AM2017-3359

Abstract 5382: A DNMT3A-independent hypomethylator phenotype is a unifying epigenetic signature of AML with good risk cytogenetics

Background: Acute myeloid leukemia (AML) causes the most leukemia-related deaths in the United States, and has frequent mutations in epigenetic regulators, including DNMT3A, IDH, and TET2. Such aberrations have been proposed to transform the epigenetic state in cancer, often involving DNA hypomethylation, however, the genomic specificity, causes, and clinical consequences of such methylation changes in AML remain unclear.

Methods: We queried genome-wide CpG methylation using The Cancer Genome Atlas (TCGA) AML samples (n=194) run on Illumina 450k arrays. We used RNA-seq data to study gene expression changes associated with hypomethylator phenotypes (HP). Analysis was done using R.

Results: Genome-wide analysis of CpG sites that are highly methylated in normal blood, and variably methylated in AML (β-value standard deviation in AML &gt; 0.2; average β-value in normal blood &gt; 0.8) revealed two distinct HPs by hierarchical clustering: Good-risk (GR) HP which included favorable cytogenetics, and DNMT-HP, which was enriched for DNMT3A mutations. We refined DNA methylation signatures of each HP cluster by differential methylation analysis and re-classified patients accordingly. Strikingly, all patients with t(8;21), inv(16), or t(15;17) belonged to the GR-HP+ group, suggesting that a common epigenetic thread connects these otherwise disparate genetic aberrations. From a clinical perspective GR-HP+ patients were younger than GR-HP- patients, and had significantly longer overall survival (median OS, years: GR-HP+ = Not reached; GR-HP- = 1.00; P &lt; 0.001). In contrast, DNMT-HP+ cases were statistically equivalent to DNMT-HP- except for an enrichment for higher WBC counts, including no difference in survival (median OS, years: DNMT-HP+ = 0.92; DNMT-HP- = 1.34; P = 0.27). From an epigenetic perspective the two HP clusters harbored distinct DNA methylation changes; although both favored hypomethylation within non-CpG islands relative to CpG islands, the enrichment was more pronounced for DNMT-HP (Odds ratio: hypomethylated CpG islands/hypomethylated non-CpG islands, GR-HP = 0.64; DNMT-HP = 0.18). Genetic analysis revealed that GR-HP+ leukemia had wild-type IDH, DNMT3A, and NPM1 genes. In contrast, DNMT-HP+ AML had significantly more FLT3, NPM1, and DNMT3A mutations compared to DNMT-HP- patients. RNA-seq revealed significant up-regulation of genes in both HP phenotypes (216, and 150 genes for GR-HP and DNMT-HP, respectively at FDR &lt; 0.01 and FC &gt; 2). Pathway analysis of these genes revealed enrichments for ion channels and the complement pathway in DNMT-HP, and for nervous system and developmental genes in GR-HP.

Conclusions: Our data suggest that two HPs exist in AML with unique epigenetic and transcriptomic signatures. The striking association between GR-HP and different favorable cytogenetic changes suggests that a common set of epigenetic features may contribute to improved survival in these patients.

Citation Format: Andrew D. Kelly, Jozef Madzo, Priyanka Madireddi, Patricia Kropf, Charly R. Good, Jaroslav Jelinek, Jean-Pierre J. Issa. A DNMT3A-independent hypomethylator phenotype is a unifying epigenetic signature of AML with good risk cytogenetics [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 5382. doi:10.1158/1538-7445.AM2017-5382

Transcriptional Selectivity of Epigenetic Therapy in Cancer

A central challenge in the development of epigenetic cancer therapy is the ability to direct selectivity in modulating gene expression for disease-selective efficacy. To address this issue, we characterized by RNA-seq, DNA methylation, and ChIP-seq analyses the epigenetic response of a set of colon, breast, and leukemia cancer cell lines to small-molecule inhibitors against DNA methyltransferases (DAC), histone deacetylases (Depsi), histone demethylases (KDM1A inhibitor S2101), and histone methylases (EHMT2 inhibitor UNC0638 and EZH2 inhibitor GSK343). We also characterized the effects of DAC as combined with the other compounds. Averaged over the cancer cell models used, we found that DAC affected 8.6% of the transcriptome and that 95.4% of the genes affected were upregulated. DAC preferentially regulated genes that were silenced in cancer and that were methylated at their promoters. In contrast, Depsi affected the expression of 30.4% of the transcriptome but showed little selectivity for gene upregulation or silenced genes. S2101, UNC0638, and GSK343 affected only 2% of the transcriptome, with UNC0638 and GSK343 preferentially targeting genes marked with H3K9me2 or H3K27me3, respectively. When combined with histone methylase inhibitors, the extent of gene upregulation by DAC was extended while still maintaining selectivity for DNA-methylated genes and silenced genes. However, the genes upregulated by combination treatment exhibited limited overlap, indicating the possibility of targeting distinct sets of genes based on different epigenetic therapy combinations. Overall, our results demonstrated that DNA methyltransferase inhibitors preferentially target cancer-relevant genes and can be combined with inhibitors targeting histone methylation for synergistic effects while still maintaining selectivity. Cancer Res; 77(2); 470-81. ©2016 AACR.

Abstract 2657: Target specificity of epigenetic therapy in cancer

A major question facing the use of epigenetic therapies in cancer is specificity in modulating gene expression. In addition, combined targeting of DNA and histone methylation remains largely unexplored despite the promising synergistic effects observed from combining DNA methyltransferase inhibitors with HDAC inhibitors. To address these questions, we performed RNA-seq, DNA methylation analysis and ChIP-seq (H3K4me2, H3K9me2, and H3K27me3) to study the effects of inhibitors of DNA methyltransferases (DAC), histone deacetylases (Depsi), histone demethylases (KDM1A inhibitor S2101), and histone methylases (EHMT2 inhibitor UNC0638 and EZH2 inhibitor GSK343) in three different cancer models (colon cancer, breast cancer, and leukemia). In colon cancer cells (YB5), DAC affected 3% of the transcriptome and 93% of the effect was gene upregulation. DAC had a greater effect on genes expressed in normal tissues and silenced in cancer (443 genes) compared to genes that do not change in cancer (194 genes). 90% of DAC targets genes showed no promoter DNA methylation in normal colon but gained methylation in cancer. Depsi changed the expression of 35% of the transcriptome and showed little specificity for gene upregulation or silenced genes. S2101, UNC0638, and GSK343 had limited effects on their own (&lt;1.5% of the transcriptome), but UNC0638 and GSK343 preferentially targeted genes with H3K9me2 or H3K27me3, respectively. DAC combined with histone methylation inhibitors led to synergistic gene upregulation while still maintaining specificity for DNA methylated and silenced genes. These synergistic genes had limited overlap, indicating the possibility to target distinct sets of genes based on different epigenetic therapy combinations. IPA analysis demonstrated that these genes are enriched in cancer pathways, and consistent with this analysis, the combination therapies were able to decrease cancer cell proliferation more effectively than monotherapy. Broadly similar results were seen with genome wide studies in both breast cancer (MCF7) and leukemia (HL-60) cells. These results demonstrate that DNA methyltransferase inhibitors preferentially target cancer relevant genes, and can be combined with inhibitors targeting histone methylation for synergistic effects while still maintaining specificity.

Citation Format: Takahiro Sato, Matteo Cesaroni, Shoghag Panjarian, Anthony Tran, Jozef Madzo, Yasuyuki Okamoto, Hanghang Zhang, Xiaowei Chen, Jaroslav Jelinek, Jean-Pierre J. Issa. Target specificity of epigenetic therapy in cancer. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 2657.

Abstract 2659: A novel isoform of TET1 that lacks a CXXC domain is overexpressed in cancer

TET1 is a DNA demethylase that regulates DNA methylation, hydroxymethylation and gene expression patterns. It has a catalytic domain that oxidizes methylated cytosine into 5-hydroxymethylcytosine and can then be further oxidized or converted to un-methylated cytosine. TET1 has a CXXC domain that allows it to recognize and bind to long stretches of CpG dinucleotides, known as CpG islands (CGIs). Our previous work found TET1 to protect CGIs from aberrant methylation and that the CXXC domain limits its ability to regulate genes outside of CGIs. However, many of the methylation changes in cancer are outside CGIs. To determine if TET1 could possibly play a role in this hypomethylation, we searched for alternatively spliced forms of the protein. We identified a novel isoform of TET1 (TET1ALT) that lacks the CXXC domain but retains the demethylase domain. TET1ALT has a unique transcription start site from a strong alternate promoter upstream of exon 3, yielding a protein with a unique translation start site. We confirmed promoter activity using Luciferase constructs, and confirmed TET1ALT translation and function through overexpression experiments. Gene expression analyses by qPCR found that TET1ALT is silenced in mESCs but becomes activated in select embryonic and adult tissues while full length TET1 is favored in mESCs, but is repressed in adult tissues. Thus, there is an isoform switch during differentiation and TET1ALT is the major isoform expressed in adult cells. TET1ALT is over expressed in several cancer cell lines, including breast cancer. In conclusion, we have identified a novel isoform of TET1 that has the potential to regulate DNA methylation outside of CGIs and is the predominant isoform expressed in adult cells. We are investigating the hypothesis that TET1ALT may contribute to the methylation changes observed in cancer outside of CGIs, such as in gene bodies and in enhancers.

Citation Format: Charly R. Good, Jozef Madzo, Shinji Maegawa, Nora Engel, Jaroslav Jelinek, Jean-Pierre Issa. A novel isoform of TET1 that lacks a CXXC domain is overexpressed in cancer. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 2659.

Abstract 1071: Age-dependent DNA methylation in normal breast epithelium and breast cancer

Breast cancer risk factors include age, genetic alterations, weight, diet and others. More recently accumulating evidence suggests that epigenetic alterations are frequent and play important role in breast cancer development and progression. Epigenetic alterations include changes in DNA methylation, histone modifications and miRNA expression in the absence of changes in DNA sequence. The epigenetic changes in cancer include global loss of DNA methylation that might lead to genomic instability and gain of methylation at many promoters of tumor suppressor genes that leads to gene silencing. Interestingly, work from our own laboratory and others have established that in many different tissues, DNA methylation changes due to age have similar patterns as to changes that occur in cancer. Therefore, the goal of this study is to identify and characterize age-dependent DNA methylation changes in normal breast epithelium and their contribution to breast cancer development. To differentiate between age-dependent and independent DNA methylation changes, we utilized DNA extracted from 27 primary human mammary epithelial cells (age range 33-82 years old) derived from adjacent or contralateral normal mammary tissue of breast cancer patients. We measured DNA methylation using a highly sensitive methodology developed in our laboratory called Digital Restriction Enzyme Analysis of Methylation (DREAM). We also studied DNA methylation in six different breast cancer cell lines by DREAM. We also used publically available 450K TCGA data for 685 breast tumor and 98 normal samples. Using bioinformatic approach, we have defined age dependent (Spearman R &gt; |0.38|, p &lt;0.05) and independent (Spearman R &lt; |0.37|, p&gt;0.05) sites that exhibit DNA methylation changes in normal breast epithelium. Interestingly, we found that the age-related genes are more seeded (exhibit low level of methylation) in their promoter CpG islands and are enriched for polycomb group target genes compared to the non-age related genes (p value = 0.032). Furthermore, these age-related genes (for example HOXD9, TDRD10, MYOD1, DPYS, GDA, GIPC2, FAM162B, LAMA1, PKDREJ, etc) have increased DNA methylation in tumors compared to the normal samples in TCGA with a methylation difference of up to 50%. These changes are significantly different compared to not-age related genes or random genes (p&lt;0.0001). The age-related genes that gained DNA methylation in tumors in TCGA dataset, were also hypermethylated in breast cancer cell lines compared to immortalized human mammary epithelial cells. Thirty one percent of the common genes between TCGA data set and breast cancer cell lines, showed decreased mRNA expression in RNA-seq data available through the TCGA data portal. Therefore, in this study we have characterized age-dependent epigenetic changes in normal breast epithelium as potential breast cancer risk associated alterations.

Citation Format: Shoghag B. Panjarian, Carolyn Slater, Jozef Madzo, Jaroslav Jelinek, Xiaowei Chen, Jean-Pierre Issa. Age-dependent DNA methylation in normal breast epithelium and breast cancer. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 1071. doi:10.1158/1538-7445.AM2015-1071

TET2 Mutations Affect Non-CpG Island DNA Methylation at Enhancers and Transcription Factor-Binding Sites in Chronic Myelomonocytic Leukemia

TET2 enzymatically converts 5-methylcytosine to 5-hydroxymethylcytosine as well as other covalently modified cytosines and its mutations are common in myeloid leukemia. However, the exact mechanism and the extent to which TET2 mutations affect DNA methylation remain in question. Here, we report on DNA methylomes in TET2 wild-type (TET2-WT) and mutant (TET2-MT) cases of chronic myelomonocytic leukemia (CMML). We analyzed 85,134 CpG sites [28,114 sites in CpG islands (CGI) and 57,020 in non-CpG islands (NCGI)]. TET2 mutations do not explain genome-wide differences in DNA methylation in CMML, and we found few and inconsistent differences at CGIs between TET2-WT and TET2-MT cases. In contrast, we identified 409 (0.71%) TET2-specific differentially methylated CpGs (tet2-DMCs) in NCGIs, 86% of which were hypermethylated in TET2-MT cases, suggesting a strikingly different biology of the effects of TET2 mutations at CGIs and NCGIs. DNA methylation of tet2-DMCs at promoters and nonpromoters repressed gene expression. Tet2-DMCs showed significant enrichment at hematopoietic-specific enhancers marked by H3K4me1 and at binding sites for the transcription factor p300. Tet2-DMCs showed significantly lower 5-hydroxymethylcytosine in TET2-MT cases. We conclude that leukemia-associated TET2 mutations affect DNA methylation at NCGI regions containing hematopoietic-specific enhancers and transcription factor-binding sites.

A phase I and pharmacodynamic study of the histone deacetylase inhibitor belinostat plus azacitidine in advanced myeloid neoplasia

Background We hypothesized that targeting two mechanisms of epigenetic silencing would be additive or synergistic with regard to expression of specific target genes. The primary objective of the study was to establish the maximum tolerated dose (MTD) of belinostat in combination with a fixed dose of azacitidine (AZA). Methods In Part A of the study, patients received a fixed dose of AZA, with escalating doses of belinostat given on the same days 1-5, in a 28 day cycle. Part B was designed to evaluate the relative contribution of belinostat to the combination based on analysis of pharmacodynamic markers, and incorporated a design in which patients were randomized during cycle 1 to AZA alone, or the combination, at the maximally tolerated dose of belinostat. Results 56 patients with myeloid neoplasia were enrolled. Dose escalation was feasible in part A, up to 1000 mg/m(2) dose level of belinostat. In Part B, 18 patients were assessable for quantitative analysis of specific target genes. At day 5 of therapy, MDR1 was significantly up-regulated in the belinostat/AZA arm compared with AZA alone arm (p = 0.0023). There were 18 responses among the 56 patients. Conclusions The combination of belinostat and AZA is feasible and associated with clinical activity. The recommended phase II dose is 1000 mg/m(2) of belinostat plus 75 mg/m(2) of AZA on days 1-5, every 28 days. Upregulation in MDR1 was observed in the combination arm at day 5 compared with the AZA alone arm, suggesting a relative biologic contribution of belinostat to the combination.

TET1-mediated hydroxymethylation facilitates hypoxic gene induction in neuroblastoma

The ten-eleven-translocation 5-methylcytosine dioxygenase (TET) family of enzymes catalyzes the conversion of 5-methylcytosine (5-mC) to 5-hydroxyme-thylcytosine (5-hmC), a modified cytosine base that facilitates gene expression. Cells respond to hypoxia by inducing a transcriptional program regulated in part by oxygen-dependent dioxygenases that require Fe(II) and α-ketoglutarate. Given that the TET enzymes also require these cofactors, we hypothesized that the TETs regulate the hypoxia-induced transcriptional program. Here, we demonstrate that hypoxia increases global 5-hmC levels, with accumulation of 5-hmC density at canonical hypoxia response genes. A subset of 5-hmC gains colocalize with hypoxia response elements facilitating DNA demethylation and HIF binding. Hypoxia results in transcriptional activation of TET1, and full induction of hypoxia-responsive genes and global 5-hmC increases require TET1. Finally, we show that 5-hmC increases and TET1 upregulation in hypoxia are HIF-1 dependent. These findings establish TET1-mediated 5-hmC changes as an important epigenetic component of the hypoxic response.

Hydroxymethylation at gene regulatory regions directs stem/early progenitor cell commitment during erythropoiesis

Hematopoietic stem cell differentiation involves the silencing of self-renewal genes and induction of a specific transcriptional program. Identification of multiple covalent cytosine modifications raises the question of how these derivatized bases influence stem cell commitment. Using a replicative primary human hematopoietic stem/progenitor cell differentiation system, we demonstrate dynamic changes of 5-hydroxymethylcytosine (5-hmC) during stem cell commitment and differentiation to the erythroid lineage. Genomic loci that maintain or gain 5-hmC density throughout erythroid differentiation contain binding sites for erythroid transcription factors and several factors not previously recognized as erythroid-specific factors. The functional importance of 5-hmC was demonstrated by impaired erythroid differentiation, with augmentation of myeloid potential, and disrupted 5-hmC patterning in leukemia patient-derived CD34+ stem/early progenitor cells with TET methylcytosine dioxygenase 2 (TET2) mutations. Thus, chemical conjugation and affinity purification of 5-hmC-enriched sequences followed by sequencing serve as resources for deciphering functional implications for gene expression during stem cell commitment and differentiation along a particular lineage.

Perturbations of 5-hydroxymethylcytosine patterning in hematologic malignancies

The recent identification of covalent cytosine modifications derived from the metabolism of 5-methylcytosine (5-mC) and catalyzed by the TET proteins has facilitated molecular insight into a new subclass of acute myeloid leukemias (AMLs). TET2-mutant AMLs have the predicted hypermethylation phenotype expected given the inability of the mutant TET2 protein to convert 5-mC to 5-hydroxymethylcytosine (5-hmC). In addition, IDH1/2 mutations confer a gain-of-function, allowing the enzymes to process α-ketoglutarate to 2-hydroxyglutarate, which inhibits the TET proteins and ultimately induces the same hypermethylation phenotype. New techniques are being developed rapidly that have the unprecedented capacity to distinguish among the various covalent cytosine modifications now known to exist. Soon, these methods will be harnessed to yield a new level of insight into AMLs with altered distribution of 5-hmC, information that may allow new diagnostic and therapeutic approaches for patients with this subtype of AML.

Tet2 loss leads to increased hematopoietic stem cell self-renewal and myeloid transformation

Somatic loss-of-function mutations in the ten-eleven translocation 2 (TET2) gene occur in a significant proportion of patients with myeloid malignancies. Although there are extensive genetic data implicating TET2 mutations in myeloid transformation, the consequences of Tet2 loss in hematopoietic development have not been delineated. We report here an animal model of conditional Tet2 loss in the hematopoietic compartment that leads to increased stem cell self-renewal in vivo as assessed by competitive transplant assays. Tet2 loss leads to a progressive enlargement of the hematopoietic stem cell compartment and eventual myeloproliferation in vivo, including splenomegaly, monocytosis, and extramedullary hematopoiesis. In addition, Tet2(+/-) mice also displayed increased stem cell self-renewal and extramedullary hematopoiesis, suggesting that Tet2 haploinsufficiency contributes to hematopoietic transformation in vivo.

The identification of (ETV6)/RUNX1-regulated genes in lymphopoiesis using histone deacetylase inhibitors in ETV6/RUNX1-positive lymphoid leukemic cells

PURPOSE

Chimeric transcription factor ETV6/RUNX1 (TEL/AML1) is believed to cause pathologic block in lymphoid cell development via interaction with corepressor complex and histone deacetylase. We wanted to show the regulatory effect of ETV6/RUNX1 and its reversibility by histone deacetylase inhibitors (HDACi), as well as to identify potential ETV6/RUNX1-regulated genes.

EXPERIMENTAL DESIGN

We used luciferase assay to show the interaction of ETV6/RUNX1 protein, ETV6/RUNX1-regulated gene, and HDACi. To identify ETV6/RUNX1-regulated genes, we used expression profiling and HDACi in lymphoid cells. Next, using the flow cytometry and quantitative reverse transcription-PCR, we measured differentiation changes in gene and protein expression after HDACi treatment.

RESULTS

Luciferase assay showed repression of granzyme B expression by ETV6/RUNX1 protein and the reversibility of this effect by HDACi. Proving this regulatory role of ETV6/RUNX1, we identified, using complex statistical analysis, 25 genes that are potentially regulated by ETV6/RUNX1 protein. In four selected genes with known role in the cell cycle regulation (JunD, ACK1, PDGFRB, and TCF4), we confirmed expression changes after HDACi by quantitative analysis. After HDACi treatment, ETV6/RUNX1-positive cells showed immunophenotype changes resembling differentiation process compared with other leukemic cells (BCR/ABL, ETV6/PDGFRB positive). Moreover, ETV6/RUNX1-positive leukemic cells accumulated in G(1)-G(0) phase after HDACi whereas other B-lineage leukemic cell lines showed rather unspecific changes including induction of apoptosis and decreased proliferation.

CONCLUSIONS

Presented data support the hypothesis that HDACi affect ETV6/RUNX1-positive cells via direct interaction with ETV6/RUNX1 protein and that treatment with HDACi may release aberrant transcription activity caused by ETV6/RUNX1 chimeric transcription factor.

Prenatal origin of childhood AML occurs less frequently than in childhood ALL

Background

While there is enough convincing evidence in childhood acute lymphoblastic leukemia (ALL), the data on the pre-natal origin in childhood acute myeloid leukemia (AML) are less comprehensive. Our study aimed to screen Guthrie cards (neonatal blood spots) of non-infant childhood AML and ALL patients for the presence of their respective leukemic markers.

Methods

We analysed Guthrie cards of 12 ALL patients aged 2–6 years using immunoglobulin (Ig) and T-cell receptor (TCR) gene rearrangements (n = 15) and/or intronic breakpoints of TEL/AML1 fusion gene (n = 3). In AML patients (n = 13, age 1–14 years) PML/RARalpha (n = 4), CBFbeta/MYH11 (n = 3), AML1/ETO (n = 2), MLL/AF6 (n = 1), MLL/AF9 (n = 1) and MLL/AF10 (n = 1) fusion genes and/or internal tandem duplication of FLT3 gene (FLT3/ITD) (n = 2) were used as clonotypic markers. Assay sensitivity determined using serial dilutions of patient DNA into the DNA of a healthy donor allowed us to detect the pre-leukemic clone in Guthrie card providing 1–3 positive cells were present in the neonatal blood spot.

Results

In 3 patients with ALL (25%) we reproducibly detected their leukemic markers (Ig/TCR n = 2; TEL/AML1 n = 1) in the Guthrie card. We did not find patient-specific molecular markers in any patient with AML.

Conclusion

In the largest cohort examined so far we used identical approach for the backtracking of non-infant childhood ALL and AML. Our data suggest that either the prenatal origin of AML is less frequent or the load of pre-leukemic cells is significantly lower at birth in AML compared to ALL cases.

TEL/AML1 and immunoreceptor gene rearrangements—which comes first?

TEL/AML1 fusion gene is present in 20-25% of childhood acute lymphoblastic leukaemias. In order to unravel at which stage of B-cell precursor development the fusion is originated, we analysed frequency and pattern of immunoreceptor (immunoglobulin and T-cell receptor) gene rearrangements in 47 TEL/AML1-positive and 43 TEL/AML1-negative cases of the same CD10+ immunophenotype. Moreover, we compared corresponding immunoreceptor gene rearrangements in 11 cases of TEL/AML1-positive leukaemia at diagnosis and relapse. More mature immunogenotype of TEL/AML1-positive cases and changes in 37% of rearrangements between diagnosis and relapse suggest that in most cases the TEL/AML1 fusion is formed during immunoreceptor gene rearrangement process.

Myeloid antigens in childhood lymphoblastic leukemia: clinical data point to regulation of CD66c distinct from other myeloid antigens

Background

Aberrant expression of myeloid antigens (MyAgs) on acute lymphoblastic leukemia (ALL) cells is a well-documented phenomenon, although its regulating mechanisms are unclear. MyAgs in ALL are interpreted e.g. as hallmarks of early differentiation stage and/or lineage indecisiveness. Granulocytic marker CD66c – Carcinoembryonic antigen-related cell adhesion molecule 6 (CEACAM6) is aberrantly expressed on ALL with strong correlation to genotype (negative in TEL/AML1 and MLL/AF4, positive in BCR/ABL and hyperdiploid cases).

Methods

In a cohort of 365 consecutively diagnosed Czech B-precursor ALL patients, we analyze distribution of MyAg+ cases and mutual relationship among CD13, CD15, CD33, CD65 and CD66c. The most frequent MyAg (CD66c) is studied further regarding its stability from diagnosis to relapse, prognostic significance and regulation of surface expression. For the latter, flow cytometry, Western blot and quantitative RT-PCR on sorted cells is used.

Results

We show CD66c is expressed in 43% patients, which is more frequent than other MyAgs studied. In addition, CD66c expression negatively correlates with CD13 (p < 0.0001), CD33 (p = 0.002) and/or CD65 (p = 0.029). Our data show that different myeloid antigens often differ in biological importance, which may be obscured by combining them into "MyAg positive ALL". We show that unlike other MyAgs, CD66c expression is not shifted from the onset of ALL to relapse (n = 39, time to relapse 0.3–5.3 years). Although opposite has previously been suggested, we show that CEACAM6 transcription is invariably followed by surface expression (by quantitative RT-PCR on sorted cells) and that malignant cells containing CD66c in cytoplasm without surface expression are not found by flow cytometry nor by Western blot in vivo. We report no prognostic significance of CD66c, globally or separately in genotype subsets of B-precursor ALL, nor an association with known risk factors (n = 254).

Conclusion

In contrast to general notion we show that different MyAgs in lymphoblastic leukemia represent different biological circumstances. We chose the most frequent and tightly genotype-associated MyAg CD66c to show its stabile expression in patients from diagnosis to relapse, which differs from what is known on the other MyAgs. Surface expression of CD66c is regulated at the gene transcription level, in contrast to previous reports.