RNA cytosine methylation and methyltransferases mediate chromatin organization and 5-azacytidine response and resistance in leukaemia

Abstract

The roles of RNA 5-methylcytosine (RNA:m⁵C) and RNA:m⁵C methyltransferases (RCMTs) in lineage-associated chromatin organization and drug response/resistance are unclear. Here we demonstrate that the RCMTs, namely NSUN3 and DNMT2, directly bind hnRNPK, a conserved RNA-binding protein. hnRNPK interacts with the lineage-determining transcription factors (TFs), GATA1 and SPI1/PU.1, and with CDK9/P-TEFb to recruit RNA-polymerase-II at nascent RNA, leading to formation of 5-Azacitidine (5-AZA)-sensitive chromatin structure. In contrast, NSUN1 binds BRD4 and RNA-polymerase-II to form an active chromatin structure that is insensitive to 5-AZA, but hypersensitive to the BRD4 inhibitor JQ1 and to the downregulation of NSUN1 by siRNAs. Both 5-AZA-resistant leukaemia cell lines and clinically 5-AZA-resistant myelodysplastic syndrome and acute myeloid leukaemia specimens have a significant increase in RNA:m⁵C and NSUN1-/BRD4-associated active chromatin. This study reveals novel RNA:m⁵C/RCMT-mediated chromatin structures that modulate 5-AZA response/resistance in leukaemia cells, and hence provides a new insight into treatment of leukaemia.

Fig. 1

Differential expression of RNA:m⁵C, RCMTs and hnRNPK in 5-AZA-sensitive and 5-AZA-resistant leukaemia cells and the binding of hnRNPK to unmethylated and cytosine-methylated RNA. a Western blot analysis of expression of RCMTs, hnRNPK and other proteins in the 5-AZA-sensitive OCI-M2 and SC leukaemia cells and the 5-AZA-resistant M2AR and SCAR leukaemia cells. b Dot blot analysis of 5-methylcytosine (m⁵C) in RNA and DNA from OCI-M2 and M2AR cells. c Dot blot analysis of 5-methylcytosine (m⁵C) in RNA and DNA from SC and SCAR cells. d, e Visualization and measurement of the binding of purified recombinant hnRNPK to the unmethylated and cytosine-methylated fluorescein (FAM)-labelled RNA oligos by an antibody-coupled bead assay

Fig. 2

Distinct functional hnRNPK/RCMT complexes required for leukaemia cell survival. a RNase digestion-coupled immunoprecipitation (IP) and co-immunoprecipitation (co-IP) to examine the interactions between hnRNPK and RCMTs, including DNMT2, NSUN3, NSUN1 and NSUN2, in OCI-M2 cells. b Measurements of the expression of hnRNPK, DNMT2, NSUN3 and other proteins in SC leukaemia cells treated with control siRNAs and the siRNAs targeting HNRNPK, DNMT2 and NSUN3 at day 3. c Measurements of the expression of NSUN1, NSUN2 and other proteins in SC leukaemia cells treated with control siRNAs and the siRNAs targeting NSUN1 and NSUN2 at day 3. d Measurements of the growth of SC leukaemia cells treated with control siRNAs and siRNAs targeting hnRNPK and individual RCMTs by MTT assay at day 3. Data are presented as the mean ± SEM of n = 3 independent samples. *P < 0.05, **P < 0.01 and ***P < 0.001 by Student’s t test

Fig. 3

Lineage-associated drug-responsive chromatin structural changes in leukaemia cells. a STED confocal microscopy analysis of the subcellular localization and co-localization of GATA1 and TET2 in OCI-M2 cells with no drug and 1 μM 5-AZA for 3 h. b STED confocal microscopy analysis of the subcellular localization and co-localization of GATA1 and TET2 in SC cells with no drug and 1 μM 5-AZA for 3 h. c Differential growth inhibition of OCI-M2 and SC leukaemia cells induced by 1 μM 5-AZA for 3 days. d A schematic illustration of the locations of the PCR primers designed in the regulatory region of SPI1/PU.1. e The PCR signals detected by 3C assays in the erythroid leukaemia cell lines, OCI-M2 and K562, and the monocytic leukaemia cell lines, SC and THP1, with no drug and 1 μM 5-AZA for 3 h. f ChIP assays with an antibody against the active form (CTD-S2P/S5P) of RNA-pol-II at the SPI1/PU.1 locus in the OCI-M2, K562, SC and THP1 cells, with no drug and 1 μM 5-AZA for 3 h. g A schematic summary of these data suggesting opposite transformations of the chromatin conformation at SPI1/PU.1 in erythroid vs. monocytic leukaemia cells in response to 5-AZA. Data are presented as the mean ± SEM of n = 3 independent samples. *P < 0.05, **P < 0.01 and ***P < 0.001 by Student’s t test

Fig. 4

RNA/hnRNPK-mediated selective interactions between lineage-determining TFs and chromatin modifiers. a Western blotting, IP with GATA1 and SPI1/PU.1 antibodies and co-IP with antibodies against various DNA and histone modifiers in OCI-M2 and SC cells. b Examination of the effects of 5-AZA on the interactions between SPI1/PU.1 and DNMT1 as well as EZH2 by co-IP in OCI-M2 cells treated with 1 μM 5-AZA at various time points. c Examination of the effects of 5-AZA on the interaction between SPI1/PU.1 and TET2 by co-IP in SC cells treated with 1 μM 5-AZA at various time points. d Examination of the RNA-dependence of the interaction between GATA1 and TET2 and the interactions between SPI1/PU.1 and DNMT1 as well as EZH2 by RNase digestion-coupled co-IP in OCI-M2 cells. e Examination of the RNA-dependence of the interaction between SPI1/PU.1 and TET2 by RNase digestion-coupled co-IP in SC cells. f Identification of hnRNPK as a direct binder of both GATA1 and TET2 by IP with GATA1 antibody and co-IPs with antibodies against various RNA-binding proteins, such as hnRNPA1 and AUF1, in OCI-M2 cells. g A schematic illustration of the nature of the interactions between hnRNPK and GATA1 as well as TET2 in OCI-M2 cells

Fig. 5

5-AZA disrupts the interaction between active RNA-pol-II complex and hnRNPK/RCMT complex. a Examination of the effects of various epigenetic drugs on the growth of OCI-M2 and SC leukaemia cells. b Examination of the interactions between hnRNPK and the components of active RNA-pol-II complex and their RNA-dependence by RNase digestion-coupled co-IPs in OCI-M2 cells. c Elucidation of the subcellular localization and co-localization of hnRNPK with active RNA-pol-II CTD-S2P at nascent RNA in OCI-M2 cells by 5-ethynyluridine (EU) clicking chemistry STED confocal microscopy in OCI-M2 and M2AR cells. d Examination of the effects of 5-AZA on the interactions between hnRNPK and the components of active RNA-pol-II complex, including RNA-pol-II CTD-S2P, RNA-pol-II S5P, CDK7 and CDK9/P-TEFb in OCI-M2 cells treated with 1 μM 5-AZA for various times. e Examination of the effects of 5-AZA on the interactions between hnRNPK and DNMT2 as well as NSUN3 by co-IPs in OCI-M2 cells treated with 1 μM 5-AZA for various times. f A schematic presentation of the hnRNPK/RCMT-associated, 5-AZA-sensitive, active chromatin structure at nascent RNA in 5-AZA-sensitive leukaemia cells

Fig. 6

Identification of unique interactions between RNA-pol-II CTD-S2P and NSUN1 as well as BRD4 in 5-AZA-resistant leukaemia cells. a Examination of the protein expression of the components of active RNA-pol-II complex, BRD4 and BRD2 in the 5-AZA-sensitive OCI-M2 and SC cells and the 5-AZA-resistant M2AR and SCAR cells by western blotting. b Examination of the effects of 5-AZA on the interactions between hnRNPK and RNA-pol-II CTDS2P as well as GATA1 in OCI-M2 and M2AR cells by IP and co-IP. c Elucidation of a unique 5-AZA-resistant interaction between RNA-pol-II CTD-S2P and NSUN1 in the 5-AZA-resistant M2AR cells by IP and co-IP. d Elucidation of unique interactions between BRD4 and RNA-pol-II CTD-S2P as well as NSUN1 in the 5-AZA-resistant M2AR cells by IP and co-IP and the effects of JQ1 on these interactions

Fig. 7

Distinct NSUN1/BRD4-associated active chromatin structures in 5-AZA-restsant leukaemia cells. a Elucidation of the subcellular localization of nascent RNA, NSUN1 and BRD4 by 5-ethynyluridine (EU)-clicking chemistry-coupled STED confocal microscopy in OCI-M2 and M2AR cells with 0.3 mM EU for 5 h. b Elucidation of the subcellular localization of nascent RNA, NSUN1 and RNA-pol-II CTD-S2P by EU-clicking chemistry-coupled STED confocal microscopy in OCI-M2 and M2AR leukaemia cells treated with 0.3 mM EU for 5 h. c Demonstration of markedly increased co-localization of NSUN1 and BRD4 by proximity ligation and rolling circle amplification (PL-RCA) in OCI-M2 and M2AR cells. d Demonstration of markedly increased co-localization of NSUN1 and RNA-pol-II CTD-S2P by proximity ligation and rolling circle amplification (PL-RCA) in OCI-M2 and M2AR cells

Fig. 8

Differential growth inhibition with CDK7/9 inhibitors, BRD4 inhibitor JQ1 and RCMT siRNAs in 5-AZA-sensitive vs. 5-AZA-resistant leukaemia cells. a Measurements of the cell growth rates in OCI-M2 and M2AR cells treated with the CDK7 inhibitor, BS-181, and the CDK9 inhibitor, PHA-767491, at various concentrations for three days. b Measurements of the growth rates of OCI-M2 and M2AR cells treated with JQ1 at various concentrations for 3 days and the synergic growth inhibition in the M2AR cells treated with a low dose (0.01 μM) of JQ1 plus 5-AZA at various concentrations. c Measurements of the growth inhibition by the siRNAs targeting various RCMTs in SC and SCAR for 3 days and the synergic growth inhibition in the SCAR cells treated with 2 nM NSUN1 siRNAs plus 5-AZA at various concentrations. d A schematic presentation of the proposed working model of distinct 5-AZA responsive/resistant chromatin structures in 5-AZA-sensitive and 5-AZA-resistant leukaemia cells

Fig. 9

Mass spectrometric analysis of mRNA:m⁵C in clinical 5-AZA-sensitive and 5-AZA-resistant AML/MDS specimens. a Representative mass spectra of the mRMA:m⁵C and adenosine peaks using the Ficoll-Paque™ PLUS-purified mononuclear bone marrow cells from 5-AZA-sensitive and 5-resistant AML and MDS specimens. b Quantitative measurement of the m⁵C/A ratios in 5-AZA-sensitive and 5-AZA-resistant clinical AML/MDS bone marrow cells. c Comparison of the manual blast count-normalized mRNA:m⁵C levels in 5-AZA-sensitive vs. 5-AZA-resistant clinical AML/MDS bone marrow cells

Fig. 10

Increased expression of hnRNPK, NSUN1 and BRD4 as well as their co-localization with active RNA-pol-II in 5-AZA-resistant AML/MDS specimens. a Visualization and quantitative measurement of co-localization of NSUN1 and RNA-pol-II CTD-S2P in 5-AZA-sensitive and 5-AZA-resistant clinical AML/MDS bone marrow specimens by PL-RCA. b Visualization and quantitative measurement of co-localization of NSUN1 and BRD4 in 5-AZA-sensitive and 5-AZA-resistant clinical AML/MDS bone marrow specimens by PL-RCA. c Morphological and immunohistochemical studies of the expression of CD34, hnRNPK, NSUN1 and BRD4 in the bone marrow cells from normal controls, cytogenetically normal MDS/refractory cytopenia with multilineage dysplasia (RCMD), and AML, n = 10 for each group, with an Olympus BX43 microscope at ×400 magnification. d STED confocal microscopic analysis of the expression and subcellular localization of hnRNPK and RNA-pol-II CTD-S2P in normal and AML bone marrows