MicroRNA-204 critically regulates carcinogenesis in malignant peripheral nerve sheath tumors

Abstract

Malignant peripheral nerve sheath tumors (MPNSTs) are highly aggressive soft tissue sarcomas accounting for 3%-10% of all soft tissue sarcomas. Neurofibromatosis type 1 (NF1) is the most important known risk factor. MPNSTs are often diagnosed at an advanced stage when distant metastases have developed. Although surgical resection remains the main treatment for MPNSTs, complete surgical resection is rarely possible. The prognosis for patients with MPNSTs is poor. There is an urgent need for improved therapies. To this end, we investigated whether microRNA (miR), specifically miR-204, might be implicated in MPNSTs because it is located at a cancer-associated genomic region exhibiting high frequency of loss of heterozygosity in tumors. We show that miR-204 expression is downregulated in NF1 and non-NF1 MPNST tumor tissues and in tumor cell lines. Restoring miR-204 expression in MPNST cell lines STS26T (non-NF1), ST88-14 (NF1), and T265p21 (NF1) significantly reduces cellular proliferation, migration, and invasion in vitro. Restoring miR-204 expression in STS26T decreases tumor growth and malignant progression in vivo. We also report that miR-204 inhibits Ras signaling and expression of high mobility group gene A2. These findings support the hypothesis that miR-204 plays critical roles in MPNST development and tumor progression. miR-204 may represent a novel biomarker for diagnosis and a candidate target with which to develop effective therapies for MPNSTs.

Fig. 1.

Downregulation of miR-204 expression in MPNSTs. Total RNA was isolated from human MPNST tumor tissues or cell lines. (A) Total RNA was labeled and hybridized on miRNA microarrays. Signals represent median signal values of 3 times. miR-204 along with a few other miRNAs showed low expression levels in NF1 and non-NF1 MPNST tumor tissues (1_1, 1_2, 1_3, 1_4: neurofibromas; 2_2, 2_3: NF1 MPNSTs; 2_4: non-NF1 MPNSTs). (B) The expression of miR-204 was further studied by qRT-PCR. miR-204 expression was significantly lower in NF1 and non-NF1 MPNST tumor tissues, in NF1 MPNST cell lines (ST8814 and T265p21), and the non-NF1 MPNST cell line (STS26T) compared with neurofibromas (**P < .01; ***P < .001; n = 3–12).

Fig. 2.

Restoring miR-204 expression in NF1 and non-NF1 MPNST cell lines. miR-204 gene was stably transfected into human NF1-associated MPNST cell lines ST8814 and T265p21 and the non-NF1-associated MPNST cell line STS26T with a lentiviral system. Fluorescence microscopy and qRT-PCR assay were used to select out the colonies with higher expression of miR-204. Assay by qRT-PCR showed that miR-204 expression was significantly higher in ST8814, T265p21, and STS26T cells with exogenous miR-204 gene than in the controls (***compared with controls, P < .001, n = 3).

Fig. 3.

Restoring miR-204 expression inhibited proliferation in NF1 and non-NF1 MPNST cell lines. miR-204 gene was stably transfected into MPNST cells with a lentiviral system. (A) MTT assay showed that restoring miR-204 in MPNST cells inhibited cell proliferation at most time points during days 6–7 (n = 5 at each time point). (B) Overexpressing miR-204 significantly reduced colony formation in NF1 MPNST cell lines T265p21 and ST8814 cells, and in the non-NF1 MPNST cell line STS26T (n = 3) compared with controls, *P < .05; **P < .01; ***P < .001).

Fig. 4.

Restoring miR-204 expression inhibited migration and invasion in NF1 and non-NF1 MPNST cell lines. The miR-204 precursor gene was stably transfected into MPNST cells with a lentiviral system. Cellular migration and invasion were studied with transwell inserts or invasion chambers. (A) Overexpressing miR-204 significantly reduced cellular migration in NF1 MPNST cell lines T265p21 and ST8814 cells and in the non-NF1 MPNST cell line STS26T. (B) Overexpressing miR-204 significantly reduced cellular invasion in NF1 MPNST cell lines T265p21 and ST8814 cells, and in the non-NF1 MPNST cell line STS26T. (Compared with controls, *P < .05; **P < .01; ***P < .001. n = 3).

Fig. 5.

Restoring miR-204 expression reduced MPNST tumor formation in vivo. STS26T cells containing the exogenous miR-204 or the control construct were injected subcutaneously into 8-week old NSG mice. The mice were euthanized 14 days after injection. The tumor volume was calculated by using formula: 1/2 ab² (a: long axis, b: short axis). Compared with controls, restoring miR-204 reduced tumor size (A) and volume (B) after xenograft (miR-204 group versus control group: ***P < .001, n = 10). (C–F) Compared with tumors in the control group, the tumor in the miR-204 group showed loose connection with surrounding tissue (arrow), collagen around tumor tissue in the miR-204 group was thick, integrated, and continuous (arrow head) (C, D: H&E stain; E, F: Masson's trichrome stain).

Fig. 6.

miR-204 inhibited RAS signaling and HMGA2 expression. (A) NF1 and non-NF1 MPNST cells with the exogenous miR-204 or the control construct were serum-starved for 24 h and then stimulated with serum-free medium (vehicle, VEH) or medium supplemented with 10% FBS for 10 min. Phosphorylated ERK and total ERK were studied by Western blotting. Compared with the VEH controls, treatment with 10% FBS increased phosphorylated ERK in all 3 MPNST cell lines. Cells with the exogenous miR-204 showed less phosphorylated ERK than cells with the control construct. (B) MPNST cells with the exogenous miR-204 or the control construct were serum-starved for 24 h. The expression of HMGA2 was studied by Western blotting. Compared with the controls, HMGA2 expression was lower in MPNST cells with the exogenous miR-204. (C) The 3′ UTR of the HMGA2 gene was analyzed by TargetScan, and a target sequence for miR-204 was identified. The strategy to make point mutations within the target sequence was listed as HMGA2–3′ UTRm. (D) The miR-204 precursor gene or the control was stably transfected into 293T cells. The transfected cells were further cotransfected with pMIR reporter vector with HMGA2 3′ UTR or HMGA2 3′UTRm. The Dual Luciferase Assay kit was used to perform the HMGA2 3′ UTR reporter assay. Overexpression of miR-204 reduced the levels of luciferase activity by about 40% in the cells containing the HMGA2 3′ UTR (HMGA2 3′ UTR) but not in the cells with tmutated seed sequence (HMGA2 3′ UTRm). These results indicated that HMGA2 3′ UTR was directly targeted by miR-204 (***P < .001 vs control; n = 3).