Roles of PSF protein and VL30 RNA in reversible gene regulation

Abstract

The mammalian protein PSF contains a DNA-binding domain (DBD) that coordinately represses multiple oncogenic genes in human cell lines, indicating a role for PSF as a human tumor-suppressor protein. PSF also contains two RNA-binding domains (RBD) that form a complex with a noncoding VL30 retroelement RNA, releasing PSF from a gene and reversing repression. Thus, the DBD and RBD in PSF are linked by a mechanism of reversible gene regulation involving a noncoding RNA. This mechanism also could apply to other regulatory proteins that contain both DBD and RBD. The mouse genome has multiple copies of VL30 retroelements that are developmentally regulated, and mouse cells contain VL30 RNAs that have normal and pathological roles in gene regulation. Human chromosome 11 has a VL30 retroelement, and a VL30 EST was identified in human blastocyst cells, indicating that the PSF-VL30 RNA regulatory mechanism also could function in human cells.

Fig. 1.

Effect of PSF expression in MCF7 cells on gene transcription. (A) Transcription of PSF. The MCF7 (low PSF) and MCF7-PSF (high PSF) cells were assayed for PSF transcription by semiquantitative RT-PCR. (B) Microarray analysis. The transcription profiles of the MCF7 and MCF7-PSF cells were analyzed on Affymetrix microarray chips. Each bar in the figure indicates the number of transcripts repressed or induced in MCF7-PSF cells by a factor of 4-8, 8-16, 16-32, 32-64, and >64.

Fig. 2.

Effect of PSF expression in MCF7 cells on cell proliferation and colony formation. (A) Cell proliferation. Cells were grown in small flasks at 37°C as attached monolayers in MEM medium containing 10% FBS. Samples were recovered from three flasks every second day by treatment with trypsin, and the number of viable cells was counted. Each point is the average of three samples that agreed within ± 5%. (B) Colony formation. Single-cell suspensions were plated in soft agar, and the plates were incubated at 37°C for 19 days and photographed.

Fig. 3.

Effect of PSF on transcription of GAGE6 in human tumor lines. Each line was transiently transfected either with a control plasmid (lane 1) or with the same plasmid encoding PSF (lane 2). Semiquantitative RT-PCR assays were used to determine the transcription levels of GAGE6 and also of GAPDH to equalize the amount of mRNA in each pair. Cell lines: Th, A2058, and yusac are melanoma lines; MCF7 is a breast tumor line; H295R is an adrenal tumor line; and colo357 is a pancreatic tumor line.

Fig. 4.

A deletion in the PSF DBD region of the HeLa cell line. (A) Southern blot analysis. The genomic DNA isolated from HeLa (human cervical tumor) and BJ (human normal fibroblast) cell lines were hybridized with a ³²P-labeled PSF DNA probe spanning positions -32 to +154 of the PSF gene. The same amount of genomic DNA was added to each lane. (B) RT-PCR analysis. DBD amd RBD of PSF transcripts were amplified by RT-PCR. (C) PCR analysis. Promoter region of PSF (-422 to -546) was amplified by PCR.

Fig. 5.

Mapping the PSF-binding site in GAGE6. (A) Map of GAGE6 promoter fragments. The fragments were generated in the region -1to -2241 flanking the 5′ end of the GAGE6 coding region. (B and C) Binding of PSF to GAGE6 promoter fragments. ³²P-labeled DNA fragments were mixed with PSF, irradiated with UV, and separated on 7.5% SDS/PAGE. (B) Lanes: 1 (-629 to -1);2(-1206 to -630); 3 (-1601 to -1207); 4 (-1979 to -1602); 5 (-2241 to -1980). (C) Lanes: 1(-2241 to -1980); 2 (-2180 to -1980); 3 (-2153 to -1980); 4 (-2109 to -1980); 5 (-2065 to -1980); 6 (-2021 to -1980). (D) Sequence of the fragment from -2241 to -2181, containing the PSF-binding site.

Fig. 6.

Binding specificities of the DBD and RBD. (A) Binding of PSF protein, DBD, and RBD to GAGE6 promoter DNA. ³²P-labeled GAGE6 DNA (-2241 to -1980), containing the PSF-binding site was mixed with intact PSF protein or a fragment containing the DBD or RBD, and the samples were analyzed by EMSA. (B) Binding of PSF protein, DBD, and RBD to VL30 RNA. ³²P-labeled VL30 RNA was substituted for GAGE6 DNA, and the samples were analyzed as described in (A). (C) Effect of VL30 RNA on binding of PSF protein and DBD to GAGE6 promoter DNA. Unlabeled VL30 RNA was added to samples containing ³²P-labeled GAGE6 DNA and mixed with PSF protein or DBD fragment. The molar ratio of VL30 RNA to GAGE6 promoter DNA was 0 in lanes 1 and 4, 10 in lanes 2 and 5, and 100 in lanes 3 and 6. The samples were analyzed by EMSA.

Fig. 7.

Inhibition of PSF activity in vivo by VL30 RNA. (A) Binding of PSF protein and DBD to GAGE6 promoter DNA in vivo and the effect of VL30 RNA. An anti-PSF antibody was used to immunoprecipitate PSF from human breast tumor lines expressing a low level of PSF (MCF7), a higher level of PSF (MCF7-PSF), or a high level of the DBD fragment (MCF7-DBD). The amount of GAGE6 promoter DNA coprecipitated with PSF was assayed by PCR. VL30 cDNA was transfected into cells in lanes 2, 4, and 6 but not in lanes 1, 3, and 5. (B) Effect of PSF protein, DBD fragment, and VL30 RNA on GAGE6 transcription in vivo. The cell lines were the same as in A. Transcription of GAGE6, PSF, and VL30 was assayed by semiquantitative RT-PCR. VL30 cDNA was transfected into the cells analyzed in lanes 2, 4, and 6 but not in lanes 1, 3, and 5.

Fig. 8.

The effect of PSF on CMV-regulated expression of a luciferase reporter gene. MCF7-PSF cells were transfected with a plasmid carrying a luciferase reporter gene either without a promoter (bar 1), with a wild-type CMV promoter (bar 2), or with a CMV promoter containing a PSF-binding site (Bar 3). The values for luciferase activity are the means ± SE for three experiments.