H-REV107-1 stimulates growth in non-small cell lung carcinomas via the activation of mitogenic signaling

Abstract

H-REV107-1, a known member of the class II tumor suppressor gene family, is involved in the regulation of differentiation and survival. We analyzed H-REV107-1 in non-small cell lung carcinomas, in normal lung, and in immortalized and tumor-derived cell lines. Sixty-eight percent of lung tumors revealed positive H-REV107-1-specific staining. Furthermore, survival analysis demonstrated a significant association of cytoplasmic H-REV107-1 with decreased patient survival. This suggested that H-REV107-1, known as a tumor suppressor, plays a different role in non-small cell lung carcinomas. Knock-down of H-REV107-1 expression in lung carcinoma cells inhibited anchorage-dependent and anchorage-independent growth whereas overexpression of H-REV107-1 induced tumor cell proliferation. Consistent with results of the survival analysis, cytoplasmic localization of the protein was essential for this growth-inducing function. Analysis of signaling pathways potentially involved in this process demonstrated that overexpression of H-REV107-1 stimulated RAS-GTPase activity, ERK1,2 phosphorylation, and caveolin-1 expression in the cell lines analyzed. These results indicate that H-REV107-1 is deficient in its function as a tumor suppressor in non-small cell lung carcinomas and is required for proliferation and anchorage-independent growth in cells expressing high levels of the protein, thus contributing to tumor progression in a subset of non-small cell lung carcinomas.

FIGURE 1

H-REV107-1 expression in human NSCLCs and in normal tissue. A: The Cancer Profiling Array I, representing cDNA probes from matched tumor and normal tissue, was hybridized with H-REV107-1 probe. cDNA represented normal lung tissue is spotted in the upper panel (N), cDNA from corresponding tumors is spotted in the bottom panel (T). Spots were evaluated by phospho-image analysis, and the ratio of tumor/normal is shown below. *Cases with higher expression of H-REV107-1 in tumors; ^▴tumors with lower H-REV107-1 expression compared with the corresponding normal tissue. B: Left: H-REV107-1 protein was detected in the nuclei of the differentiated mature layer in normal bronchial epithelium (red arrow), but not in the proliferating basal layer (black arrow). Right: Absorption of the H-REV107-1 antiserum against the specific H-REV107-1 peptide before immunohistochemistry was used to control for the specificity of the H-REV107-1 antibody. No H-REV107-1 staining is seen with the preabsorbed antiserum. C: Left: Preferably nuclear localization of the H-REV107-1 protein was found in a subset of NSCLCs. Right: Alternatively, cytoplasmic localization was observed. D: H-REV107-1 protein was detected in bronchiolar epithelium. Left: CC16 protein was used as a marker for Clara cells. Right: H-REV107-1 antigen revealed a homogenous staining in bronchiolar epithelium. E: Univariate survival analysis for 51 patients with a 5-year follow-up documentation. Cumulative survival curves are calculated according to the Kaplan-Meyer method; differences in survival time are assessed with the log rank test. Patients with cytoplasmic H-REV107-1 have a mean survival time of 24 months. In contrast, the patients, which did not harbor H-REV107-1 in the cytoplasm, had a mean survival time of 41 months, P = 0.0165.

FIGURE 2

H-REV107-1 expression in cell lines derived either from normal bronchial epithelium or from lung tumors. A: Northern blot analysis of H-REV107-1 mRNA in immortalized lung epithelial cell lines and lung carcinoma cell lines. HBE and SAE are primary bronchial epithelial cells, whereas 9442, 9609, YP440, and 2078 are immortalized lung epithelial cells. H322, H2030, D51, D54, D117, A549, A427, H2030, H125, COLO699, and H23 are ADC cell lines; SHP77 and D97 are large cell lung carcinoma lines; H266, H2170, and H157 are SCC lines; and H378, H82, H446, CPC-N, H209, DMS114, H526, H187, COLO668, COLO677, DMS79, and N417 are small cell lung carcinoma lines. The ovarian carcinoma cell line SKOV-3 was used as a positive control. The 28S mRNA is shown to control for equal loading of the gels. B: Expression of the H-REV107-1 protein in human lung carcinoma cell lines. Protein extracts were isolated from the cell lines, and immunoblotted first against an anti-H-REV107-1 antibody, and then against an anti-actin antibody used as a loading control. COS-7 cells, transiently transfected with an H-REV107-1 expression plasmid, were used as a positive control for the anti-H-REV107-1 antibody (control). The low levels of the H-REV107-1 protein in some NSCLC cell lines suggest that either this amount is sufficient to exert a function. Alternatively, we cannot exclude that H-REV107-1 has simply lost its function here and is not required for growth. Intensity of signals was semiquantitatively evaluated using ImageJ freeware. The ratio between the relative amounts of H-REV107-1 and actin was formed and is depicted graphically.

FIGURE 3

H-REV107-1 intracellular localization in A549 and D51 cells and effect of H-REV107-1 suppression on cell proliferation. A: Immunocytochemistry was done using the H-REV107-1-specific antibody. A part of the image was digitally magnified (a small icon in the left corner of the images). In A549 cells, H-REV107-1 is distributed through the cytoplasm. In D51 cells, H-REV107-1 expression seems to be confluency-dependent. Both cytoplasmic and nuclear localization of the protein can be observed. COS-7 cells were used as a negative control for the H-REV107-1-antibody. B: Growth properties of A549 and D51 cells after suppression of H-REV107-1 expression using RNAi. siRNA effect was controlled via RT-PCR and semiquantitatively evaluated using ImageJ freeware. The ratio between H-REV107-1 and GAPDH expression is presented graphically (right). Both anchorage-dependent (left) and anchorage-independent growth (middle) were controlled. Note that >50% reduction of H-REV107-1 expression resulted in a complete abrogation of cellular growth.

FIGURE 4

Role of the cytoplasmic localization for H-REV107-1 function. Two expression constructs, HREVFL and HREVdN, were transfected into A549 and D51 cells. A: Immunofluorescence was used to analyze the distribution of the ectopically expressed H-REV107-1 wild-type protein (HREVFL) and its deletion mutant (HREVdN) in A549 (left) and D51 (right) cells. B: Growth response of A549 and D51 cells on H-REV107-1 overexpression was analyzed using the MTT test. Cells were seeded in 96-wells and incubated for 96 hours. H-REV107-1 protein induced both anchorage-dependent and anchorage-independent growth compared with the N-terminal deletion mutant (HREVdN) and pcDNA3.1 in both cell lines.

FIGURE 5

Effect of the H-REV107-1 overexpression on RAS, MAPK, and caveolin-1 expression and on RAS activity. A: Western blot analysis was used to examine the levels of total and phosphorylated EGFR after transfection of A549 and D51 cells with full-length HREV107-1 construct (HREVFL), its N terminus truncated form HREVdN and pcDNA3 plasmid. H-REV107-1 has as slight inhibitory effect on the phosphorylation of EGFR in A549 cells and no effect in D51 cells. As a loading control, actin immunoblotting was done. B: A549 and D51 cells were fractionated to define the level of active RAS proteins. The antibody used recognizes all three RAS proteins: HRAS, KRAS, and NRAS. Two bands of slightly different electrophoretic mobility seen in the membrane fraction might correspond to a mutated and a wild-type form of one of the RAS proteins, because mutations in the positions V12 and C61 lead to a change of the protein electrophoretic mobility. α-Tubulin was used to control purity of the membrane fractions. C: Western blot analysis was used to examine the levels of RAS and caveolin-1 protein expression, and ERK1,2, Akt/PKB, and FAK phosphorylation after H-REV107-1 expression. A moderate increase of the protein levels of RAS and p-ERK1,2 is seen in both cell lines; caveolin-1 and phosphorylated Akt/PKB are up-regulated after H-REV107-1 overexpression only in D51 cells. The relative levels of H-REV107-1 wild-type and mutated proteins were controlled using the H-REV107-1-specific antibody. As a loading control, actin immunoblotting was performed. D: RAS-GTPase activity assay (Upstate) demonstrated the amount of the activated RAS proteins after H-REV107-1 overexpression in A549 and D51 cells. A GST antibody was used as a control for immunoprecipitation. Intensity of signals was semiquantitatively evaluated using ImageJ freeware and the ratio RAS/GST was calculated. Note that H-REV107-1 induces RAS activity in A549 cells, but only a moderate effect is seen in D51 cells.