miR-199b, a novel tumor suppressor miRNA in acute myeloid leukemia with prognostic implications

Abstract

Background: Dysregulation of miRNAs that can act as tumor suppressors or oncogenes can result in tumorigenesis. Previously we demonstrated that miR-199b was significantly downregulated in acute myeloid leukemia (AML) and targets podocalyxin and discoidin domain receptor 1. Herein we investigated the functional role of miR-199b in AML and its prognostic implications.

Methods: Major approaches include transduction of hematopoietic stem cells and bone marrow transplantation, analyses of blood lineages, histone deacetylases (HDAC) inhibitors, and molecular and clinical data analyses of AML patients using The Cancer Genome Atlas (TCGA).

Results: We first examined the relative miR-199b expression in steady state hematopoiesis and showed CD33(+) myeloid progenitors had the highest miR-199b expression. Further, silencing of miR-199b in CD34(+) cells resulted in significant increases in CFU-GM colonies. Via TCGA we analyzed the molecular and clinical characteristics of 166 AML cases to investigate a prognostic role for miR-199b. The Kaplan-Meier curves for high and low expression values of miR-199b and the observed distribution of miRNA expression revealed the highly expressed group had significantly better survival outcomes (p < 0.016, log rank test). Additionally, there was significant difference between miR-199b expression across the AML subtypes with particularly low expression found in the FAB-M5 subtype. Furthermore, FAB-M5 subtype showed a poor prognosis with a 1-year survival rate of only 25 %, compared with 51 % survival in the overall sample (p < 0.024). Furthermore, significant inverse correlation of HoxA7 and HoxB6 expression with miR-199b was observed in FAB-M5 AML patients. Molecular mutations were analyzed among miR-199b high and low AML cases. Significant correlations in terms of association and survival outcomes were observed for NPMc and IDH1 mutations. Treatment of THP-1 cells (represents M5-subtype) with HDAC inhibitors AR-42, Panobinostat, or Decitabine showed miR-199b expression was significantly elevated upon AR-42 and Panobinostat treatment. To further understand the hematopathological consequences of decreased miR-199b, we employed a bone-marrow transduce/transplant (BMT) mouse model. Interestingly, in vivo miR-199b silencing per-se in HSCs did not result in profound perturbations.

Conclusions: Loss of miR-199b can lead to myeloproliferation while HDAC inhibitors restore miR-199b expression and promote apoptosis. Low miR-199b in AML patients correlates with worse overall survival and has prognostic significance for FAB-M5 subtype.

Keywords: Acute myeloid leukemia; Bone marrow transplant; FAB-M5; HSC; NPM1; miRNA-199b.

Fig. 1

Intronic miR-199b expression in steady state hematopoiesis and its myelopoietic and hematopoietic potential. a miR-199b’s location within the genome; intronic of the DNM1 gene. b Human primary BM CD34⁺, BM CD33⁺, PB monocytes, PB eosinophils, and PB basophils expression of miR-199b in steady state hematopoiesis. c Lentiviral particles for anti-Has-miR-199b-5p and control vector were used to transduce human BM derived CD34 + cells. Silencing of miR-199b levels were confirmed in HSCs post transduction via RT-qPCR. d Lentiviral particles for anti-hsa-miR-199b-5p and scrambled/control vector were used to transduce human BM derived CD34 + cells. CFU-GM assay was performed and representative photomicrographs of the colonies are shown

Fig. 2

Clinical relevance of miR-199b-5p across AML subtypes. a Kaplan–Meier analysis of 166 AML patients from the TCGA database for high/low miR-199b expression and overall survival, p = 0.0164. b miR-199b cytogenetic risk for same patients set. c High/low miR-199b expression by FAB subtype in the same patients set. d Least squared means plot of 199b expression for different FAB subtypes using ANOVA (p < 0.0001). e miR-199b levels for n = 128 AML patients analyzed for miR-199b/NPM1 with n = 63 miR-199b-high samples, of which n = 2 were positive for NPM1 mutation, and n = 65 miR-199b-low samples of which n = 41 positive for NPM1 mutation (Fisher p value = 1.941e-08, odds ratio 19.6112) (left panel). miR-199b levels for n = 139 patients with n = 59 miR-199b-high samples (of which n = 5 positive for IDH1 mutation) whereas n = 80 were miR-199b-low samples with n = 25 positive for IDH1 mutation (Fisher p value = 0.01175, odds ratio 3.663005) (right panel)

Fig. 3

Expression of HoxA7 and HoxB6 inversely correlates with miR-199b levels in AML with FAB M5 subtype. a Least square (LS) means plots of conserved miR-199b target HOXA7 (ANOVA p < 0.0001) versus different AML FAB categories. ANOVA *p < 0.0001 b Least square (LS) means plot of conserved miR-199b target HOXB6 (ANOVA p < 0.0017) versus different AML FAB categories. ANOVA *p < 0.0017

Fig. 4

HDAC inhibitors restore miR-199b-5p expression in THP-1 cells to induce apoptosis. a miR-199b expression in THP-1 cells upon 24 h treatment with vehicle (DMSO), Decitabine (DB, 5uM), AR-42 (2uM), or Panobinostat (0.7uM) ****p < 0.0001. b Western blot for THP-1 cells treated with DMSO [27], Panobinostat [28] or AR-42 [29] blotted for acetylated (Ac) and total histones H2A, H2B, H3 and H4. c Apoptosis via Annexin V staining for the AR-42 and Panobinostat treated cells compared to DMSO treatment

Fig. 5

miR-199b silencing in vivo via a transduce and transplant approach. a Bone marrow transplant (BMT) approach schematic to delete miR-199b in hematopoietic cells. Mice were treated with 5-flourouracil, LSK (Lin⁻Sca⁺Kit⁺) cells were obtained and silenced with an anti-miR-199b lentivirus. These cells were then transplanted in irradiated Ly5.1 mice. b Left panel: Lentiviral particles for anti-miR-199b-5p and control vector were used to transduce mouse HSCs. Silencing of miR-199b levels were confirmed in HSCs post transduction via RTqPCR. Right panel: Engraftment efficiencies post BMT were measured using staining for Ly5.2 versus Ly5.1 via flow cytometry. A representative flow cytometry data is shown from the analysis. c–g Peripheral blood analysis for control and low-miR-199b transplant mice (n = 5). h Flow cytometry of peripheral blood B (CD19 staining) and T (CD3 staining) cells in control and low-miR-199b transplant mice at 27 weeks