Neuronal differentiation of synovial sarcoma and its therapeutic application

Abstract

Synovial sarcoma is a rare sarcoma of unknown histologic origin. We previously reported the gene expression profile of synovial sarcoma was closely related to that of malignant peripheral nerve sheath tumors, and the fibroblast growth factor (FGF) signal was one of the main growth signals in synovial sarcoma. Here we further demonstrate the neural origin of synovial sarcoma using primary tumors and cell lines. The expression of neural tissue-related genes was confirmed in synovial sarcoma tumor tissues, but the expression of some genes was absent in synovial sarcoma cell lines. Treatment of synovial sarcoma cell lines with BMP4 or FGF2 enhanced or restored the expression of neural tissue-related genes and induced a neuron-like morphology with positive Tuj-1 expression. Treatment with all-trans-retinoic acid also induced the expression of neural tissue-related genes in association with growth inhibition, which was not observed in other cell lines except a malignant peripheral nerve sheath tumor cell line. A growth-inhibitory effect of all-trans-retinoic acid was also observed for xenografted tumors in athymic mice. The simultaneous treatment with FGF signal inhibitors enhanced the growth-inhibitory effect of all-trans-retinoic acid, suggesting the combination of growth signaling inhibition and differentiation induction could be a potential molecular target for treating synovial sarcoma.

Fig. 1

The mRNA expression of neuron- and glia-related genes was analyzed by reverse transcription-PCR in tissue samples of 29 sarcomas, including 18 synovial sarcomas (SSs). Among the SS tumors, seven were SYT-SSX1-positive biphasic, six were SYT-SSX1-positive monophasic, and five were SYT-SSX2-positive monophasic. PLS = pleomorphic liposarcoma; LMS = leiomyosarcoma; MFH = malignant fibrous histiocytoma; MPNST = malignant peripheral nerve sheath tumor.

Fig. 2

The mRNA expression of neuron- and glia-related genes was analyzed by reverse transcription-PCR in 12 tumor cell lines, including five synovial sarcoma (SS) cell lines (YaFuSS, HS-SY-II, SYO-1, Fuji, 1273/99). The SS cell lines expressed a number of neuron-related genes such as CRABP1, which was not detected in other cell lines. The SS cell lines partially retained the original expression profile of the primary SS tumors shown in Figure 1.

Fig. 3A–D

The neuronal differentiation of synovial sarcoma (SS) cells is shown. (A) Immunostaining for Tuj-1 (red) after treatment with vehicle, ATRA, rhFGF2, or rhBMP4 is shown. Nuclei were counterstained with 4′,6-diamidino-2-phenylindole (blue). Original magnification was x100 in all figures. The mRNA expression of the MASH1 gene in SS cells is shown after treatment with (B) ATRA, (C) rhFGF2, or (D) rhBMP4.

Fig. 4

A graph illustrates the growth-inhibitory effect of ATRA in vitro. Cells (1.0 × 10⁴) were seeded on 60-mm dishes and cultured in standard medium containing 10% FBS with or without ATRA (10 mmol/L). The cells were counted on Day 17, and the fraction of surviving cells (number of treated cells divided by number of untreated cells) was calculated. Treatment with ATRA failed to inhibit the growth of fibrosarcoma (HT1080) or colon carcinoma (COLO208) cells but suppressed the growth of all five synovial sarcoma cell lines as well as the MPNST (NMS-2). p = 0.0009.

Fig. 5

A graph illustrates the growth-inhibitory effect of ATRA in vivo. SYO-1 (5 × 10⁶) cells suspended in 100 mL phosphate-buffered saline were subcutaneously injected into the hind flank of male BALB/c nu/nu athymic mice. Mice were randomly assigned to ATRA-treated (●), vehicle-treated (■), and untreated (▴) groups (five mice per group). ATRA treatment (20 mg/kg) was started on Day 2, and tumor volume was measured on Days 3, 7, 10, 14, and 17. In vivo treatment with ATRA achieved growth-inhibitor effect at Day 17 (*p = 0.026 versus vehicle; **p = 0.034 versus no treatment). The inhibitory effect, however, was not complete, and the tumor started to regrow with the same kinetics as pretreated cells.

Fig. 6

The graph illustrates the growth-inhibitory effect of ATRA combined with an FGFR inhibitor in vivo. Cells (1.0 × 10⁴) were seed on 60-mm dishes and cultured in standard medium containing 10% FBS with the indicated concentrations of PD166866 and ATRA (gray, 0 mmol/L; black, 1 mmol/L). The cells were counted on Day 17, and the fraction of surviving cells (number of treated cells divided by number of untreated cells) was calculated. In the case of YaFuSS, the simultaneous administration of ATRA (1 mmol/L) and PD166866 (100 nmol/L) had much greater-inhibitory effects than the single administration of each drug. Similar additive effects were observed in HS-SY-II, SYO-1, and Fuji cells. *p < 0.05; **p < 0.01.