The landscape of RNA-chromatin interaction reveals small non-coding RNAs as essential mediators of leukemia maintenance.

RNA constitutes a large fraction of chromatin. Spatial distribution and functional relevance of most of RNA-chromatin interactions remain unknown. We established a landscape analysis of RNA-chromatin interactions in human acute myeloid leukemia (AML). In total more than 50 million interactions were captured in an AML cell line. Protein-coding mRNAs and long non-coding RNAs exhibited a substantial number of interactions with chromatin in cis suggesting transcriptional activity. In contrast, small nucleolar RNAs (snoRNAs) and small nuclear RNAs (snRNAs) associated with chromatin predominantly in trans suggesting chromatin specific functions. Of note, snoRNA-chromatin interaction was associated with chromatin modifications and occurred independently of the classical snoRNA-RNP complex. Two C/D box snoRNAs, namely SNORD118 and SNORD3A, displayed high frequency of trans-association with chromatin. The transcription of SNORD118 and SNORD3A was increased upon leukemia transformation and enriched in leukemia stem cells, but decreased during myeloid differentiation. Suppression of SNORD118 and SNORD3A impaired leukemia cell proliferation and colony forming capacity in AML cell lines and primary patient samples. Notably, this effect was leukemia specific with less impact on healthy CD34+ hematopoietic stem and progenitor cells. These findings highlight the functional importance of chromatin-associated RNAs overall and in particular of SNORD118 and SNORD3A in maintaining leukemia propagation.

A Dynamic rRNA Ribomethylome Drives Stemness in Acute Myeloid Leukemia.

The development and regulation of malignant self-renewal remain unresolved issues. Here, we provide biochemical, genetic, and functional evidence that dynamics in ribosomal RNA (rRNA) 2'-O-methylation regulate leukemia stem cell (LSC) activity in vivo. A comprehensive analysis of the rRNA 2'-O-methylation landscape of 94 patients with acute myeloid leukemia (AML) revealed dynamic 2'-O-methylation specifically at exterior sites of ribosomes. The rRNA 2'-O-methylation pattern is closely associated with AML development stage and LSC gene expression signature. Forced expression of the 2'-O-methyltransferase fibrillarin (FBL) induced an AML stem cell phenotype and enabled engraftment of non-LSC leukemia cells in NSG mice. Enhanced 2'-O-methylation redirected the ribosome translation program toward amino acid transporter mRNAs enriched in optimal codons and subsequently increased intracellular amino acid levels. Methylation at the single site 18S-guanosine 1447 was instrumental for LSC activity. Collectively, our work demonstrates that dynamic 2'-O-methylation at specific sites on rRNAs shifts translational preferences and controls AML LSC self-renewal. SIGNIFICANCE: We establish the complete rRNA 2'-O-methylation landscape in human AML. Plasticity of rRNA 2'-O-methylation shifts protein translation toward an LSC phenotype. This dynamic process constitutes a novel concept of how cancers reprogram cell fate and function. This article is highlighted in the In This Issue feature, p. 247.

CD33-directed immunotherapy with third-generation chimeric antigen receptor T cells and gemtuzumab ozogamicin in intact and CD33-edited acute myeloid leukemia and hematopoietic stem and progenitor cells.

Immunotherapies, such as chimeric antigen receptor (CAR) modified T cells and antibody-drug conjugates (ADCs), have revolutionized the treatment of cancer, especially of lymphoid malignancies. The application of targeted immunotherapy to patients with acute myeloid leukemia (AML) has been limited in particular by the lack of a tumor-specific target antigen. Gemtuzumab ozogamicin (GO), an ADC targeting CD33, is the only approved immunotherapeutic agent in AML. In our study, we introduce a CD33-directed third-generation CAR T-cell product (3G.CAR33-T) for the treatment of patients with AML. 3G.CAR33-T cells could be expanded up to the end-of-culture, that is, 17 days after transduction, and displayed significant cytokine secretion and robust cytotoxic activity when incubated with CD33-positive cells including cell lines, drug-resistant cells, primary blasts as well as normal hematopoietic stem and progenitor cells (HSPCs). When compared to second-generation CAR33-T cells, 3G.CAR33-T cells exhibited higher viability, increased proliferation and stronger cytotoxicity. Also, GO exerted strong antileukemia activity against CD33-positive AML cells. Upon genomic deletion of CD33 in HSPCs, 3G.CAR33-T cells and GO preferentially killed wildtype leukemia cells, while sparing CD33-deficient HSPCs. Our data provide evidence for the applicability of CD33-targeted immunotherapies in AML and its potential implementation in CD33 genome-edited stem cell transplantation approaches.

NOP10 predicts lung cancer prognosis and its associated small nucleolar RNAs drive proliferation and migration.

Non-small cell lung cancer (NSCLC) is the leading cause of cancer death worldwide underlining the urgent need for new biomarkers and therapeutic targets for this disease. Long noncoding RNAs are critical players in NSCLC but the role of small RNA species is not well understood. In the present study, we investigated the role of H/ACA box small nucleolar RNAs (snoRNAs) and snoRNA-bound ribonucleoproteins (snoRNPs) in the tumorigenesis of NSCLC. H/ACA box snoRNPs including the NOP10 core protein were highly expressed in NSCLC. High levels of either NOP10 mRNA or protein were associated with poor prognosis in NSCLC patients. Loss of NOP10 and subsequent reduction of H/ACA box snoRNAs and rRNA pseudouridylation inhibited lung cancer cell growth, colony formation, migration, and invasion. A focused CRISPR/Cas9 snoRNA knockout screen revealed that genomic deletion of SNORA65, SNORA7A, and SNORA7B reduced proliferation of lung cancer cells. In line, high levels of SNORA65, SNORA7A, and SNORA7B were observed in primary lung cancer specimens with associated changes in rRNA pseudouridylation. Knockdown of either SNORA65 or SNORA7A/B inhibited growth and colony formation of NSCLC cell lines. Our data indicate that specific H/ACA box snoRNAs and snoRNA-associated proteins such as NOP10 have an oncogenic role in NSCLC providing new potential biomarkers and therapeutic targets for the disease.

Site-specific methylation of 18S ribosomal RNA by SNORD42A is required for acute myeloid leukemia cell proliferation.

Noncoding RNAs, including small nucleolar RNAs (snoRNAs), play important roles in leukemogenesis, but the relevant mechanisms remain incompletely understood. We performed snoRNA-focused CRISPR-Cas9 knockout library screenings that targeted the entire snoRNAnome and corresponding host genes. The C/D box containing SNORD42A was identified as an essential modulator for acute myeloid leukemia (AML) cell survival and proliferation in multiple human leukemia cell lines. In line, SNORD42A was consistently expressed at higher levels in primary AML patient samples than in CD34+ progenitors, monocytes, and granulocytes. Functionally, knockout of SNORD42A reduced colony formation capability and inhibited proliferation. The SNORD42A acts as a C/D box snoRNA and directs 2'-O-methylation at uridine 116 of 18S ribosomal RNA (rRNA). Deletion of SNORD42A decreased 18S-U116 2'-O-methylation, which was associated with a specific decrease in the translation of ribosomal proteins. In line, the cell size of SNORD42A deletion carrying leukemia cells was decreased. Taken together, these findings establish that high-level expression of SNORD42A with concomitant U116 18S rRNA 2'-O-methylation is essential for leukemia cell growth and survival.

Enhanced expression of the sphingosine-1-phosphate-receptor-3 causes acute myelogenous leukemia in mice.

Acute myeloid leukemia (AML) carries a 10-100 fold lower mutational burden than other neoplastic entities. Mechanistic explanations for why a low number of mutations suffice to induce leukemogenesis are therefore required. Here we demonstrate that transgenic overexpression of the wild type sphingosine-1-phosphate receptor 3 (S1P3) in murine hematopoietic stem cells is sufficient to induce a transplantable myeloid leukemia. In contrast, S1P3 expression in more mature compartments does not cause malignant transformation. Treatment with the sphingosine phosphate receptor modulator Fingolimod, which prevents receptor signaling, normalized peripheral blood cell counts and reduced spleen sizes in S1P3 expressing mice. Gene expression analyses in AML patients revealed elevated S1P3 expression specifically in two molecular subclasses. Our data suggest a previously unrecognized contribution of wild type S1P3 signaling to leukemogenesis that warrants the exploration of S1P3 antagonists in preclinical AML models.

AML1-ETO requires enhanced C/D box snoRNA/RNP formation to induce self-renewal and leukaemia.

Leukaemogenesis requires enhanced self-renewal, which is induced by oncogenes. The underlying molecular mechanisms remain incompletely understood. Here, we identified C/D box snoRNAs and rRNA 2'-O-methylation as critical determinants of leukaemic stem cell activity. Leukaemogenesis by AML1-ETO required expression of the groucho-related amino-terminal enhancer of split (AES). AES functioned by inducing snoRNA/RNP formation via interaction with the RNA helicase DDX21. Similarly, global loss of C/D box snoRNAs with concomitant loss of rRNA 2'-O-methylation resulted in decreased leukaemia self-renewal potential. Genomic deletion of either C/D box snoRNA SNORD14D or SNORD35A suppressed clonogenic potential of leukaemia cells in vitro and delayed leukaemogenesis in vivo. We further showed that AML1-ETO9a, MYC and MLL-AF9 all enhanced snoRNA formation. Expression levels of C/D box snoRNAs in AML patients correlated closely with in vivo frequency of leukaemic stem cells. Collectively, these findings indicate that induction of C/D box snoRNA/RNP function constitutes an important pathway in leukaemogenesis.

MYST2 acetyltransferase expression and Histone H4 Lysine acetylation are suppressed in AML.

Chromatin-modifying enzymes are frequently altered in acute myeloid leukemia (AML). In the current study, we identified MYST2, a core histone acetyltransferase, to be suppressed in blast cells from AML patients compared with nonmalignant hematopoietic progenitor cells. Functionally, loss of MYST2 accelerated leukemic growth and colony formation, while forced expression of MYST2 induced H4K5 acetylation (H4K5Ac) and suppressed hematopoietic progenitor cell growth. Consistently, global H4K5Ac levels were frequently decreased in AML blasts. Low levels of H4K5Ac were most prominent in patients with complex karyotype AML and were associated with inferior overall survival in univariate but not multivariate analysis. ChIP-seq experiments in primary AML patients' blasts revealed widespread H4K5Ac deregulation, most prominent at gene promoters. Taken together, MYST2 is a repressed growth suppressor in AML mediating reduced acetylation of histone 4 at residue 5 and is associated with inferior AML patient survival.

Regulation of the beta1,4-Galactosyltransferase I promoter by E2F1.

Cell surface carbohydrate chains are widely known to contribute to cell migration, recognition and proliferation. beta1,4-Galactosyltransferase I (beta1,4GalT I) transfers galactose to the terminal N-acetylglucosamine of complex-type N-glycan, and contributes to cell proliferation, differentiation and migration. Here, we identified beta1,4GalT I as a novel target gene of cell cycle regulator E2F1. E2F1 proteins interact with the promoter of the beta1,4GalT I gene in vivo, and E2F1 over-expression stimulates the activity of beta1,4GalT I promoter and the mRNA and protein expression of beta1,4GalT I, and augments the level of beta1, 4-galactosyltion. Site-specific mutagenesis revealed that this region which contains two E2F1 binding site (nt -215 to -207 and +1 to +6) is necessary for beta1,4GalT I activation by E2F1. Furthermore, down-regulation of beta1,4GalT I expression attenuates E2F1-induced DNA synthesis and cell cycle progression as well as the expression of cell-cycle regulator Cyclin D1. Thus, beta1,4GalT I is an important E2F1 target gene that is required for cell cycle progression in mammalian cells, which elicits a new mechanism of cell growth and a new mechanism of beta1,4GalT I transcription.

Sox2 is translationally activated by eukaryotic initiation factor 4E in human glioma-initiating cells.

Sox2, a master transcription factor, contributes to the generation of induced pluripotent stem cells and plays significant roles in sustaining the self-renewal of neural stem cells and glioma-initiating cells. Understanding the functional differences of Sox2 between glioma-initiating cells and normal neural stem cells would contribute to therapeutic approach for treatment of brain tumors. Here, we first demonstrated that Sox2 could contribute to the self-renewal and proliferation of glioma-initiating cells. The following experiments showed that Sox2 was activated at translational level in a subset of human glioma-initiating cells compared with the normal neural stem cells. Further investigation revealed there was a positive correlation between Sox2 and eukaryotic initiation factor 4E (eIF4E) in glioma tissues. Down-regulation of eIF4E decreased Sox2 protein level without altering its mRNA level in glioma-initiating cells, indicating that Sox2 was activated by eIF4E at translational level. Furthermore, eIF4E was presumed to regulate the expression of Sox2 by its 5' untranslated region (5' UTR) sequence. Our results suggest that the eIF4E-Sox2 axis is a novel mechanism of unregulated self-renewal of glioma-initiating cells, providing a potential therapeutic target for glioma.

Alpha2,3-Sialylation regulates the stability of stem cell marker CD133.

CD133 is widely used as a marker for the isolation and characterization of normal and cancer stem cells. The dynamic alternation of CD133 glycosylation contributes to the isolation of normal and cancer stem cells, and is supposed to be associated with cell differentiation. Although CD133 has been identified as a N-glycosylated protein, the specific glycosylation status of CD133 remain unclear. Here, we found that CD133 could be sialylated in neural stem cells and glioma-initiating cells, and the sialyl residues attach to CD133 N-glycan terminal via alpha2,3-linkage. Furthermore, desialylation of CD133 by neuraminidase specifically accelerates its degradation in lysosomes-dependent pathway. Taken together, our results characterized CD133 as an alpha2,3-sialylated glycoprotein and revealed that the sialylation modification contributes to the stability of CD133 protein, providing clues to understanding the function of CD133 molecular and to understanding the utility of glycosylated CD133 epitopes in defining neural stem cells and tumour-initiating cells.

Identification and characterization of the promoter of human ATF5 gene.

ATF5, a member of ATF/CREB family of b-ZIP transcription factors, is highly expressed in a wide variety of neoplasms and regulates cell differentiation, cell survival and apoptosis. However, the mechanism of human ATF5 transcriptional regulation has not been clarified. Here, we identified the transcription start site of the ATF5 gene, cloned its 5'-flanking region and identified the region -105 to +3 relative to the transcription start site as that having promoter activity. This region contained potential binding sites for several transcription factors, including EBF1, Sp1 and E2F1. EBF1 transcription factor binds to the ATF5 promoter and regulates the ATF5 transcription in an EBF-binding site independent manner. Thus, our studies not only provided molecular basis of ATF5 transcriptional regulation, but also identified ATF5 as a target gene of EBF1 transcription factor.

Beta1,4-galactosyltransferase V regulates self-renewal of glioma-initiating cell.

Glioma results from unregulated expansion of a self-renewing glioma-initiating cell population. The regulatory pathways which are essential for sustaining the self-renewal of glioma-initiating cells remain largely unknown. Cell surface N-linked oligosaccharides play functional roles in determining cell fate and are associated with glioma malignancy. Previously, we have reported that beta1,4-galactosyltransferase V (beta1,4GalT V) effectively galactosylates the GlcNAcbeta1-->6Man arm of the highly branched N-glycans and positively regulates glioma cell growth. Here, we show that decreasing the expression of beta1,4GalT V by RNA interference in glioma cells attenuated the formation of polylactosamine and inhibited the ability of tumor formation in vivo. Down-regulation of beta1,4GalT V depleted CD133-positive cells in glioma xenograft, and inhibited the self-renewal capacity and the tumorigenic potential of glioma-initiating cells. These data reveal a critical role of beta1,4GalT V in the self-renewal and tumorigenicity of glioma-initiating cells, and indicate that manipulating beta1,4GalT V expression may have therapeutic potential for the treatment of malignant glioma.

Identification of beta-1,4-galactosyltransferase I as a target gene of HBx-induced cell cycle progression of hepatoma cell.

BACKGROUND/AIMS: The hepatitis B virus-encoded HBx protein contributes to hepatocarcinogenesis with largely unknown mechanisms. It is widely known that N-linked oligosaccharides on glycoproteins are structurally altered during malignant transformation and these alterations are often associated with malignant transformation of cells. beta-1,4-galactosyltransferase I (GalT I) contributes to the biosynthesis of Galbeta-->4GlcNAc structure in the outer chain moieties of N-glycans. METHODS: The difference of GalT I expression between normal liver and hepatoma tissues were investigated; the effect of HBx on GalT I expression was investigated; the role of GalT I in hepatoma cell growth and HBx-induced hepatoma cell growth were investigated. RESULTS: GalT I was highly expressed in hepatocellular carcinoma and transcriptionally up-regulated by HBx, and functioned as a positive growth regulator in hepatoma cells. Furthermore, decreasing the expression of GalT I in hepatoma cells reduced the ability of tumor formation in vivo and inhibited HBx-induced cell cycle progression. CONCLUSIONS: HBx-induced GalT I expression might contribute to HBx-mediated HCC development and progression.

Identification of E1AF as a target gene of E2F1-induced apoptosis in response to DNA damage.

Transcription factor E1AF plays critical roles in neuronal development and tumour metastasis and is regulated by a number of signalling cascades, including the mitogen-activated protein kinase pathways. Accumulated evidence indicted that E1AF might contribute to cell survival in response to environment factors. Here, we provided evidence the cell cycle and apoptosis regulator E2F1 induces E1AF expression at the transcriptional level. DNA damage by etoposide causes E2F1-dependent induction of E1AF expression at transcriptional level. Furthermore, disruption of E1AF expression by E1AF RNAi decreased E2F1-induced apoptosis in response to etoposide. Thus, we conclude that activation of E1AF provides a means for E2F1 to induce cell apoptosis in response to DNA damage.

E1AF promotes mithramycin A-induced Huh-7 cell apoptosis depending on its DNA-binding domain.

Transcription factor E1AF is widely known to play critical roles in tumor metastasis via directly binding to the promoters of genes involved in tumor migration and invasion. Here, we reported for the first time the pro-apoptotic role of E1AF in tumor cells. The expression of E1AF at protein level was obviously increased during Huh-7 and Hep3B cells apoptosis induced by the anticancer agent mithramycin A. E1AF overexpression markedly enhanced mithramycin A-induced Huh-7 cell apoptosis and the expression of pro-apoptotic protein Bax depending on its DNA-binding domain. And, reduction of E1AF inhibited mithramycin A-induced Huh-7 cell apoptosis. Furthermore, reducing the expression of Bax significantly inhibited E1AF-increased Huh-7 cell apoptosis induced by mithramycin A. Taken together, E1AF increases mithramycin A-induced Huh-7 cells apoptosis and Bax expression depending on its DNA-binding domain, indicating that E1AF might contribute to the therapeutic efficiency of mithramycin A for hepatoma.

Down-regulation of beta1,4GalT V at protein level contributes to arsenic trioxide-induced glioma cell apoptosis.

Arsenic trioxide (As2O3) has considerable efficacy in treating solid tumors with induction of apoptosis with largely unknown mechanisms. Posttranslational processing of proteins by glycosylation could have multiple regulating roles in the process of apoptosis. Here, we found that the expression of beta1,6-linked GlcNAc-bearing N-glycans on cell surface protein was gradually decreased after induction of apoptosis by As2O3-treatment. And, As2O3 significantly decreased the protein expression level of beta1,4GalT V, which effectively galactosylates the beta1,6-GlcNAc branch of N-glycans and functions as a positive regulator in glioma development. Furthermore, interfering with the expression of beta1,4GalT V in human glioma cell markedly promoted As2O3-induced cell apoptosis and beta1,4GalT V overexpression significantly reduced As2O3-induced glioma cell apoptosis. Taken together, our results suggested that down-regulation of beta1,4GalT V expression plays an important role in As2O3-induced apoptosis, providing a new mechanism of As2O3-induced cell apoptosis and indicating that inhibitors of beta1,4GalT V may enhance the therapeutic efficiency of As2O3 for malignant glioma.

Identification of beta1,4GalT II as a target gene of p53-mediated HeLa cell apoptosis.

beta1,4-galactosyltransferase II (beta1,4GalT II) is one of the enzymes transferring galactose to the terminal N-acetylglucosamine of complex-type N-glycans. Previously, we have reported that beta1,4GalT II overexpression increased cisplatin-induced HeLa cell apoptosis. However, the mechanisms of its expression regulation have been rarely investigated. Here, we cloned the 1.8-kb 5'-flanking region of the beta1,4GalT II gene and analysed its promoter activity. The transcriptional activity and mRNA expression level of beta1,4GalT II were dramatically induced by p53 transcription factor in HeLa cells. In response to DNA damage agent adriamycin, the mRNA expression and promoter activity of beta1,4GalT II were significantly up-regulated and the binding of p53 to beta1,4GalT II promoter was obviously increased. Furthermore, decreasing the expression of beta1,4GalT II using RNA interference inhibited p53-mediated HeLa cell apoptosis induced by adriamycin. Collectively, these results suggested that beta1,4GalT II might serve as a target gene of p53 transcription factor during adriamycin-induced HeLa cell apoptosis, which elucidated a new mechanism of p53-mediated cell apoptosis.