TEL, a putative tumor suppressor, modulates cell growth and cell morphology of ras-transformed cells while repressing the transcription of stromelysin-1

Abstract

TEL is a member of the ETS family of transcription factors that interacts with the mSin3 and SMRT corepressors to regulate transcription. TEL is biallelically disrupted in acute leukemia, and loss of heterozygosity at the TEL locus has been observed in various cancers. Here we show that expression of TEL in Ras-transformed NIH 3T3 cells inhibits cell growth in soft agar and in normal cultures. Unexpectedly, cells expressing both Ras and TEL grew as aggregates. To begin to explain the morphology of Ras-plus TEL-expressing cells, we demonstrated that the endogenous matrix metalloproteinase stromelysin-1 was repressed by TEL. TEL bound sequences in the stromelysin-1 promoter and repressed the promoter in transient-expression assays, suggesting that it is a direct target for TEL-mediated regulation. Mutants of TEL that removed a binding site for the mSin3A corepressor but retained the ETS domain failed to repress stromelysin-1. When BB-94, a matrix metalloproteinase inhibitor, was added to the culture medium of Ras-expressing cells, it caused a cell aggregation phenotype similar to that caused by TEL expression. In addition, TEL inhibited the invasiveness of Ras-transformed cells in vitro and in vivo. Our results suggest that TEL acts as a tumor suppressor, in part, by transcriptional repression of stromelysin-1.

FIG. 1

TEL inhibits growth in soft-agar assays. (A) Immunoblot analysis showing the expression of TEL and TEL mutant proteins in the stable cell lines used in panel B. The right-hand panel shows the level of the endogenous (Control) and expressed TEL using anti-N-terminal TEL. (B) Cellular localization of TEL and TEL mutants. Cells expressing TEL or the indicated TEL mutants were incubated with 0.5% NP-40 in an isotonic buffer for 5 min before being subjected to low-speed centrifugation to collect the nuclei. Equal amounts of cytoplasmic (C) and nuclear (N) fractions were analyzed by immunoblotting with antibodies directed to mSin3A as a nuclear protein control (upper panel), TEL (middle panel), or Ras as a cytoplasmic protein (bottom panel). (C) TEL inhibits soft-agar colony formation of Ras-transformed cells. Cells were grown for 10.5 days in medium containing 0.375% agarose, as overlays on 0.8% agarose beds. A total of 1,500 cells of the indicated cell lines were plated per dish. Values shown are the mean and range of duplicate samples.

FIG. 2

TEL inhibits cell growth. A total of 8 × 10⁵ cells of each cell line were plated in 100-mm tissue culture dishes containing 20 ml of medium. Cells were counted every 24 h for 3 days. Values shown are the mean and standard deviation of quadruplicate samples (A) and the mean and range of duplicate samples (B).

FIG. 3

TEL causes aggregation of Ras-transformed NIH 3T3 cells. Ras-transformed NIH 3T3 cells infected with recombinant retroviruses expressing TEL and the indicated TEL mutants were plated and allowed to grow for 72 h. Cells were visualized by bright-field microscopy (magnification, ×62.5) and recorded as digital images using Adobe Photoshop. The panel labeled Ras+TEL (LP) contains cell photographed at threefold-lower power (LP, lower power).

FIG. 4

TEL inhibits transcription from the stromelysin-1 promoter. (A) Total cellular RNA was isolated from the cell lines described in Fig. 1. Samples (40 μg) were electrophoresed through a 2.5 M formaldehyde–1% agarose gel. (Upper panel) The RNA was transferred to a nylon membrane, UV cross-linked, and probed with a radiolabeled stromelysin-1 cDNA. (Lower panel) RNA stained with ethidium bromide. The third panel (left side) contains the same blot probed for actin expression. (B) NIH 3T3 cells (left panel) or Ras-expressing NIH 3T3 cells (right panel) were transfected with 1 μg of the rat stromelysin-1 luciferase plasmid and 100 ng of the pCMV-TEL expression constructs shown. Luciferase activity was determined 44 h later. Values were normalized for transfection efficiency by using CMV-SEAP activity as an internal control. Values shown are the fold repression (mean and range) of duplicate samples.

FIG. 5

Mapping TEL-responsive elements in the stromelysin-1 promoter. (A) Schematic diagram of the stromelysin-1 promoter and deletion mutants used in panel B. The known ETS factor-binding site is labeled EBS, and a putative TEL-binding site is labeled TBS. (B) Deletion analysis of the stromelysin-1 promoter. The ability of TEL to repress the full-length stromelysin-1 promoter and the deletion mutants shown in panel A is demonstrated. RLU, relative luciferase activity. Assays were performed with NIH 3T3 cells, as in Fig. 4, using 20 ng (left panel) or 100 ng (right panel) of TEL-expressing plasmid. (C) Gel mobility shift assays using GST-TEL and a consensus ETS factor-binding site as the probe or the TEL-binding site (−111 to −88) from the stromelysin-1 promoter as the probe and bacterially produced GST-TEL DNA-binding domain fusion protein (right panel). Oligonucleotides used for competition are designated above each lane (−205, −213 to −183; −93, −93 to −74). The sequences of these oligonucleotides are provided in Materials and Methods. W, wild type; M, mutant. The numbers above the lanes in the left panel are the amounts of the wild-type competitor added to each sample.

FIG. 6

TELΔETS is an inhibitor of TEL-mediated repression. NIH 3T3 cells were transfected with the rat stromelysin-1 luciferase plasmid and the pCMV-TEL or pCMV-TELΔETS expression constructs shown. The numbers below the graph are the amounts of each expression plasmid transfected. Luciferase activity was determined 44 h later. Values were normalized for transfection efficiency using CMV-SEAP as an internal control. Values shown are the mean and standard deviation of triplicate samples. Fold repression represents the promoter activity from cells transfected with expression plasmids compared to those transfected with the empty vector.

FIG. 7

An MMP inhibitor produces a phenotype similar to TEL expression. Ras-transformed NIH 3T3 cells were cultured in the presence or absence of BB-94, a specific MMP inhibitor (67). Cells were grown for 72 h and photographed as described in Materials and Methods.

FIG. 8

TEL inhibits Ras-dependent cell invasion in vitro. NIH 3T3 cells (control) or cells expressing TEL, Ras, Ras plus TEL, or Ras plus the indicated TEL mutants (lower panel) were cultured in Matrigel chambers for 3 days using fibronectin in the lower chamber as a chemoattractant. The filters were fixed by immersion in 1% glutaraldehyde and then stained with 0.2% crystal violet. The upper surface of the membrane was cleaned with a cotton swab to remove all noninvasive cells. The stained, invasive cells were then photographed using a digital camera. The two rows are duplicate samples.