Terminal differentiation of human liposarcoma cells induced by ligands for peroxisome proliferator-activated receptor gamma and the retinoid X receptor

Abstract

Induction of terminal differentiation represents a promising therapeutic approach to certain human malignancies. The peroxisome proliferator-activated receptor gamma (PPAR gamma) and the retinoid X receptor alpha (RXR alpha) form a heterodimeric complex that functions as a central regulator of adipocyte differentiation. Natural and synthetic ligands for both receptors have been identified. We demonstrate here that PPAR gamma is expressed at high levels in each of the major histologic types of human liposarcoma. Moreover, primary human liposarcoma cells can be induced to undergo terminal differentiation by treatment with the PPAR gamma ligand pioglitazone, suggesting that the differentiation block in these cells can be overcome by maximal activation of the PPAR pathway. We further demonstrate that RXR-specific ligands are also potent adipogenic agents in cells expressing the PPAR gamma/RXR alpha heterodimer, and that simultaneous treatment of liposarcoma cells with both PPAR gamma- and RXR-specific ligands results in an additive stimulation of differentiation. Liposarcoma cell differentiation is characterized by accumulation of intracellular lipid, induction of adipocyte-specific genes, and withdrawal from the cell cycle. These results suggest that PPAR gamma ligands such as thiazolidinediones and RXR-specific retinoids may be useful therapeutic agents for the treatment of liposarcoma.

Figure 1

Expression of PPARγ mRNA in human tissues (A), human liposarcomas (B), and other soft tissue sarcomas (C). Total RNA (15 μg per lane) was isolated from human tumors, electrophoresed through formaldehyde-containing agarose gels, blotted to nylon, and hybridized with ³²P-labeled hPPARγ cDNA. Equivalent amounts of intact RNA was run in each lane as indicated by hybridization to a 36B4 cDNA probe (not shown). MPNS, malignant peripheral nerve sheath tumor; MFH, malignant fibrous histiocytoma.

Figure 2

Pioglitazone induces differentiation of cultured human liposarcoma cells. Primary LS857 (A and B) and LS707 (C and D) cells were isolated and cultured in 60-mm dishes. At confluence, cells were cultured for 7 days in the presence (B and D) or absence (A and C) of 10 μM pioglitazone. After an additional 4 days of culture, cells were fixed and stained with oil red O. (×40.)

Figure 3

PPARγ- and RXR-specific ligands induce expression of markers of terminal adipocyte differentiation in PPARγ-expressing fibroblasts and human liposarcoma cells. NIH PPARγ and LS857 cells were treated for 7 days with no activator, pioglitazone alone, LG268 alone, or pioglitazone and LG268 as indicated (both). Total RNA (10 μg per lane) was blotted to nylon and hybridized with ³²P-labeled human or murine PPARγ, aP2, and adipsin cDNA. Equivalent amounts of intact RNA were run in each lane as indicated by hybridization to a 36B4 cDNA probe (not shown). Two PPARγ transcripts are present in NIH PPARγ cells, the viral transcript (Upper) and the endogenous transcript (Lower).

Figure 4

Induction of differentiation in human liposarcoma cells by thiazolidinediones and RXR-specific retinoids. LS857 cells were treated at confluence for 7 days with no activator, 5 μM thiazolidinedione (BRL 49653 or pioglitazone), 50 nM LG268, or both as indicated. After an additional 4 days of culture cells were fixed and stained with oil red O. Macroscopic view of the 60-mM culture dishes is shown.