miR-1202 acts as anti-oncomiR in myeloid leukaemia by down-modulating GATA-1S expression

Abstract

Transient abnormal myelopoiesis (TAM) is a Down syndrome-related pre-leukaemic condition characterized by somatic mutations in the haematopoietic transcription factor GATA-1 that result in exclusive production of its shorter isoform (GATA-1_S). Given the common hallmark of altered miRNA expression profiles in haematological malignancies and the pro-leukaemic role of GATA-1_S, we aimed to search for miRNAs potentially able to modulate the expression of GATA-1 isoforms. Starting from an in silico prediction of miRNA binding sites in the GATA-1 transcript, miR-1202 came into our sight as potential regulator of GATA-1 expression. Expression studies in K562 cells revealed that miR-1202 directly targets GATA-1, negatively regulates its expression, impairs GATA-1_S production, reduces cell proliferation, and increases apoptosis sensitivity. Furthermore, data from TAM and myeloid leukaemia patients provided substantial support to our study by showing that miR-1202 down-modulation is accompanied by increased GATA-1 levels, with more marked effects on GATA-1_S. These findings indicate that miR-1202 acts as an anti-oncomiR in myeloid cells and may impact leukaemogenesis at least in part by down-modulating GATA-1_S levels.

Keywords: Down syndrome; GATA-1; alternative splicing; miR-1202; myeloid leukaemia.

Figure 1.

Schematic of post-transcriptional mechanisms generating GATA-1 isoforms.

Figure 2.

Identification of a putative target site of miR-1202 in the second exon of the GATA-1 transcript. (a) Sanger sequencing electropherogram of GATA-1 exon 2 showing a novel 22-nucleotide duplication in patient TAM-P1 (upper line, bold black and red text); the arrow indicates the starting point of the c.153_174 duplication sequence. (b) Northern blot analysis of small RNAs from fetal liver and cord blood samples. The upper panel is a representative image of hybridization signals obtained with an LNA probe corresponding to the 22-mer sequence (exon 2/22mer). Signals compatible with mature miRNA sizes were appreciable in the fetal liver sample. A probe against U6 RNA was used as loading control (lower panel). (c) in silico prediction of hsa-miR-1202 target sites showing pairing nucleotides (red text) between the miR-1202 seed region and its potential complementary site within the 22-mer sequence in the second exon of GATA-1. (d) Quantitative real-time PCR analysis in fetal liver samples showing reduced miR-1202 levels in a T21 sample as compared to a D21 control. (e) Quantitative real-time PCR analysis in cord blood samples showing reduced miR-1202 levels in mosaic or constitutional T21 samples as compared to D21 controls, with a more dramatic reduction in constitutional T21 samples. RNU6B was used as endogenous control. Data are presented as mean ± s.d. of three independent experiments. Statistics: *p-value < 0.05, **p-value < 0.001 relative to control groups (calculated by Student's t-tests and one-way ANOVA, followed by Dunnett's multiple comparisons test, where appropriate).

Figure 3.

miR-1202 expression levels inversely correlate with GATA-1 mRNA levels in the myeloid leukaemia K562 cell line. (a,b) Evaluation of miR-1202 up- and downregulation in K562 cells transfected with (a) pre-miR-1202 or (b) anti-miR-1202, respectively. (c,d) endogenous GATA-1 mRNA levels in K562 cells transfected with (c) pre- and (d) anti-miR-1202, respectively. Data are presented as mean ± s.d. of three independent experiments. Statistics: *p-value < 0.05, **p-value < 0.001 versus untreated negative control groups (calculated by one-way ANOVA, followed by Dunnett's multiple comparisons test).

Figure 4.

miR-1202 differently affects the expression levels of GATA-1 isoforms in K562 cells. (a) Western blot analysis of GATA-1 isoforms in protein extracts from K562 cells transfected with pre-miR-1202. GATA-1_FL is shown above with a protein band at approximately 48 kDa, with GATA-1_S represented by a band at 37 kDa. miR-1202 upregulation correlates with reduced endogenous GATA-1 levels, mostly due to the GATA-1_S fraction. (b,c) Densitometric analysis of western blot results showing (b) total GATA-1 levels, and (c) GATA-1 isoforms levels following miR-1202 up-regulation. (d) Western blot analysis of GATA-1 isoforms in protein extracts from K562 cells transfected with anti-miR-1202. miR-1202 downregulation was accompanied by increased levels of both GATA-1 isoforms with a significant prevalence of the GATA-1_S fraction. (e,f) Densitometric analysis of western blot results showing (e) total GATA-1 levels and (f) GATA-1 isoform levels. Data are presented as mean ± s.d. of three independent experiments. Statistics: *p-value < 0.05, **p-value < 0.001 versus untreated negative control groups and ^#p-value < 0.05, ^##p-value < 0.0001 versus each respective scramble control group (calculated by one-way ANOVA, followed by Dunnett's multiple comparisons test).

Figure 5.

Identification of a miR-1202 binding element in the second exon of GATA-1. (a) Biotin-labelled miR-1202 (Bi-miR-1202) pull-down RNA samples showing a substantial enrichment (approx. threefold) in GATA-1 transcript by quantitative real-time PCR analysis. Enrichment analysis of RAB1A and ACTB transcripts were used as positive and negative controls, respectively. (b) Luciferase reporter gene assay in K562 cells co-transfected with pre-miR-1202 and GATA-1 exon 2 psiCHECK2 construct (GATA-1 exon 2) or GATA-1 3′-UTR psiCHECK2 construct. (c) Luciferase reporter gene assay in K562 cells co-transfected with wild-type or mutated GATA-1 exon 2 psiCHECK2 constructs. A reduced luciferase activity was found in cells transfected with the wild-type construct and the constructs bearing point mutations outside the putative miR-1202 binding site (mutant constructs 1–2). This effect was abolished by point mutations within the putative miR-1202 target site or its complete deletion (mutant constructs 3 and 4). Wild-type and mutant sequence details are shown in electronic supplementary material, table S1. Data are presented as mean ± s.d. of three independent experiments. Statistics: *p-value < 0.05, **p-value < 0.001 versus each respective scramble control group (calculated by one-way ANOVA, followed by Dunnett's multiple comparisons test).

Figure 6.

miR-1202 has anti-proliferative effects and increases apoptosis sensitivity in K562 cells. (a,b) Cell viability assessed by MTT assay on K562 cells 24, 48 and 72 h after transfection with (a) pre-miR-1202 or (b) anti-miR-1202. (c,d) Evaluation of apoptosis (Annexin V-FITC+/PI + cells) in K562 cells (c) upregulated or (d) down-modulated for miR-1202 and treated with 20 and 30 µM cisplatin. Data are presented as mean ± s.d. of three independent experiments. Statistics: (a) and (b) *p-value < 0.05, **p-value < 0.001 compared to the scramble control group (calculated as fold-change relative to control cells, arbitrarily set at 100%). (c) and (d) *p-value < 0.05, **p-value < 0.001 versus untreated K562 cells and ^#p-value < 0.05, ^##p-value < 0.0001 versus scramble control cells (calculated by one-way ANOVA, followed by Dunnett's multiple comparisons test).

Figure 7.

miR-1202 upregulation increases the apoptosis rate in K562 cells. (a) Increased expression of the pro-apoptotic protein Bax and decreased expression of the anti-apoptotic protein Bcl-x_L in K562 cells over-expressing miR-1202. (b–d) Densitometric analysis of western blot bands showing statistically significant (b) increased Bax levels (c) decreased Bcl-x_L levels and (d) increased Bax/Bcl-x_L expression ratio following miR-1202 up-modulation. (e) miR-1202 downregulation did not significantly affect the expression of Bax, but caused increased expression of Bcl-x_L. (f–h) Densitometric analysis of western blot results showing (f) no significant variations in Bax levels, (g) significantly increased Bcl-x_L levels and (h) reduced Bax/Bcl-x_L expression ratio following miR-1202 down-modulation. For each analysis, band intensities from three independent experiments were quantified and normalized to GAPDH used as loading control. Data are presented as mean ± s.d. of three independent experiments. Statistics: *p-value < 0.05, **p-value < 0.001 versus untreated negative control groups and ^#p-value < 0.05, ^##p-value < 0.0001 versus scramble control groups (calculated by one-way ANOVA, followed by Dunnett's multiple comparisons test).

Figure 8.

miR-1202 levels inversely correlate with GATA-1 expression levels in TAM and AML-DS patients. (a) miR-1202 mRNA levels detected in peripheral blood samples collected at different phases of disease in patients TAM-P2 and TAM-P3. (b) GATA-1 mRNA levels detected in patients TAM-P2 and TAM-P3 showing increased GATA-1 levels in both patients at diagnosis followed by GATA-1 downregulation at remission in patient TAM-P2, whereas further GATA-1 upregulation was detected in patient TAM-P3 with concomitant clinical worsening and early death. (c) miR-1202 mRNA levels in bone marrow samples from patient AML-P1 at different phases of disease. (d) An opposite trend was found in GATA-1 mRNA levels during MDS progression to AML showing over-expressed mRNA levels in the MDS phase that were further increased during AML progression and, conversely, dramatically reduced at complete remission. (e) Western blot analysis of GATA-1 isoforms in protein extracts from bone marrow samples of patient AML-P1 at different phases of disease. (f) Densitometric analysis of western blot results showing progressive increase of total GATA-1 levels during MDS to AML progression and its reduction to levels comparable to T21 control at remission. (g) Comparison of GATA-1 isoforms levels showing that both increased and reduced total GATA-1 levels were mostly due to the GATA-1_S fraction. (h) Evaluation of the GATA-1_FL/GATA-1_S ratio confirming the altered ratio in favour of the GATA-1_S fraction during the acute phases of the disease and, conversely, an almost normalized ratio at remission. Band intensities were quantified and normalized to α-actin used as loading control. Statistics: *p-value < 0.05, **p-value < 0.001 versus each respective T21 negative control groups and ^#p-value < 0.05, ^##p-value < 0.0001 versus remission stage (calculated by one-way ANOVA, followed by Dunnett's multiple comparisons test).

Figure 9.

Schematic of putative molecular mechanisms involved in miRNA-mediated modulation of GATA-1 levels. (a) Regulation of alternative splicing of the GATA-1 transcript by the PRMT1/RBM15 axis. miR-1202 downregulation could promote PRMT1 activity, thus favouring alternative splicing of the GATA-1 transcript and production of the shorter isoform GATA-1_S. (b) miR-1202 targets the coding region of the GATA-1 transcript to down-modulate GATA-1 levels through a post-transcriptional gene silencing mechanism (PTGS).