Amplification of 6p22 occurs in about 10–20% of bladder cancers and is associated with enhanced tumour cell proliferation. Candidate target genes for the 6p22 amplicon include E2F3 and the adjacent gene NM_017774. To clarify which gene is representing the main target, we compared the prevalence of the amplification and the functional role of both genes. Amplification of E2F3 and NM_017774 was analysed by fluorescence in situ hybridization on a bladder cancer tissue microarray composed of 2317 cancer samples. Both genes showed amplification in 104 of 893 (11.6%) interpretable tumours and were exclusively found co-amplified. Additional gene expression analysis by real-time polymerase chain reaction in 12 tumour-derived cell lines revealed that amplification of 6p22 was always associated with co-overexpression of E2F3 and NM_017774. Furthermore, RNA interference was used to study the influence of reduced gene expression on cell growth. In tumour cells with and without the 6p22 amplicon, knockdown of E2F3 always lead to unequivocal reduction of proliferation, whereas knockdown of NM_017774 was only capable to slow down cell proliferation in non-amplified cells. Our findings point out that E2F3 but not NM_017774 is driving enhanced proliferation of 6p22 amplified tumour cells. We conclude that E2F3 must be responsible for the growth advantage of 6p22 amplified bladder cancer cells.
Amplification of the chromosomal region 6p22 is one of the most frequent genetic alterations in urinary bladder cancer, affecting up to 20% of high grade, invasively growing tumours (Bruch et al., 2000; Tomovska et al., 2001; Hurst et al., 2004; Oeggerli et al., 2004). In a recent study, we narrowed down the amplicon to a region spanning approximately 1.6 megabases at 6p22.3 enclosing 13 different genes (Tomovska et al., 2001). Only two of these genes, the transcription factor E2F3 and the adjacent gene NM_017774, are consistently overexpressed in 6p22.3 amplified bladder cancers and are therefore considered candidate genes driving the amplification (Bruch et al., 2000; Feber et al., 2004; Hurst et al., 2004).
E2Fs play an important role in the retinoblastoma (Rb) pathway (Hunter and Pines, 1994). The regulatory function of Rb largely depends on the ability to bind and inhibit E2F family members of transcription factors including E2F3 (Hiebert et al., 1992; Qian et al., 1992). We have recently shown that amplification of the E2F3 gene locus is associated with protein overexpression, invasive tumour growth and enhanced cell proliferation (Oeggerli et al., 2004). The function of NM_017774 is currently not known, but it shows some homology to a protein that is associated with cyclin–dependent kinase 5 (CDK5RAP1), which is the reason why it has originally been termed cyclin–dependent kinase 5–associated protein 1–like 1. However, it is important to note that there is no experimental evidence for any functional similarities between these two proteins.
In order to determine, whether E2F3 or NM_017774 is the main amplification target, or if both genes might contribute jointly to the aggressive features of 6p22.3–amplified bladder cancers, we first inspected amplification frequencies in 18 bladder cancer cell lines by fluorescence in situ hybridization (FISH), as described previously (Wagner et al., 1997). We utilized digoxigenated BAC (NM_017774: BAC RP3444C7, RZPD, Berlin, Germany) and PAC (E2F3: PAC dJ177P22, Sanger Centre, Cambridge, UK) probes containing the target genes and a Spectrum Red-labeled chromosome 6 centromeric probe (CEP6) as a reference (Vysis, Downers Grove, IL, USA). Amplifications were found in four of 18 (22%) bladder cancer cell lines (HTB-5, HTB-9, CRL-1472 and HB-CLS-439), showing that E2F3 and NM_017774 were always co-amplified.
A bladder cancer prognosis tissue microarray (TMA), composed of 2317 formalin-fixed paraffin–embedded tissues (Oeggerli et al., 2004), was then used to comprehensively compare the amplification frequencies of both genes. We hypothesized that the main amplification target would be present in all tumours that reveal amplification of the 6p22.3 genomic region. Initial amplification frequencies were 9.8% for NM_017774 and 11.4% for E2F3. A small subset of 34 tumours (3.8%) could be identified, exhibiting amplification of only one gene (see Table 1a and Table 1b). The following case-by-case comparison of these tumours using conventional large tissue sections demonstrated, however, that every tumour with E2F3 amplification had also NM_017774 amplification, and vice versa. The initially observed discrepancies were either due to variable interpretation of borderline findings in low level amplified tumours (15 cases) or counting errors because of low FISH signal intensities, high background, tissue damage or technical artefacts (19 cases). In summary, co-amplification of E2F3 and NM_017774 was found in all 6p22.3 amplified tumours (11.6%). As a consequence, amplification of NM_017774 is identically associated with invasive and high-grade phenotype, and patient prognosis as already published for E2F3 (Oeggerli et al., 2004).
As FISH analysis could not identify either E2F3 or NM_017774 as the primary amplification target, we next performed mRNA expression analysis. We expected that the main amplification target gene would show a particularly strong mRNA expression increase. Because of the superior RNA quality in freshly collected tissues as compared to formalin-fixed paraffin–embedded samples, we compared mRNA levels of three amplified and four non-amplified bladder cancer cell lines. Detailed results of our gene expression analysis can be taken from Figure 1. E2F3 was generally expressed at higher levels than NM_017774 (average difference 4.25-fold). However, 6p22.3 amplification had a comparable influence on E2F3 and NM_017774 expression, scaling up individual mRNA levels at least 10-fold. These findings are in line with a previous analysis in cell lines HTB-5, HTB-9, JO'N and CRL-1472 by Hurst et al. (2004) who also found significantly increased expression of NM_017774 and E2F3 following the amplification of 6p22.3.
These results point to the hypothesis that both genes might jointly contribute to the aggressive features of 6p22.3–amplified bladder cancers. Clearly, DNA amplification is a perfect method to co-overexpress neighbouring genes. Evidence of clusters of co-overexpressed genes have already been detected in human, fly and worm (Wang et al., 1995). Examples in humans include the non-I-integrin alpha-chain genes located in clusters on chromosomes 2, 12 and 17. It has been suggested before that keeping functionally related genes near could be advantageous for a cell because it may ease the burden of unpacking of DNA for transcription (Lee and Sonnhammer, 2003). It appears possible that amplification might not always target only one particular gene, but two or more genes that contribute to a common function or pathway. Although only little is known about the possible function of NM_017774, the presence of particular functional domains in the predicted protein structure have linked it to the protein translation machinery (Altschul et al., 1997). It can be expected that such a cooperative effect of E2F3 and NM_017774 would result in a particular strong growth advantage and that any reduction in the quantity of one of these two genes should be sufficient to reverse the effect.
In order to test this hypothesis, we decided to perform gene silencing experiments in 6p22.3–amplified cell lines. RNA interference (RNAi) is an established method to specifically inactivate mRNA of selected target genes (Elbashir et al., 2001; Paddison et al., 2002). Two 6p22.3–amplified (HTB-5, CRL-1472) and two non-amplified cell lines (CRL-7930, PC-3) were tested for their suitability for RNAi treatment. SYBR Green real-time polymerase chain reaction (PCR) (LightCycler, Roche, Basel, Switzerland) was employed to measure the effect of RNAi on target gene expression.
Applying this technique resulted in an always more than 50% decrease of mRNA levels for both potential target genes over a period, starting from 12 h after transfection and lasting until the end of the experiments (after 6 days). Based on these studies, non-amplified cell line CRL-7930 and amplified cell line HTB-5 were selected for subsequent experiments.
In cell line CRL-7930, mRNA levels decreased until day 4 after transfection and did not rise again until the end of the experiment (day 6). Knockdown levels were slightly higher for E2F3 (73% decrease of mRNA level; average from days 4 to 6) as compared to NM_017774 (59% decrease of mRNA level; average from days 4 to 6; P=0.0016).
In the amplified cell line HTB-5, the lowest mRNA expression levels were reached already 24 h after transfection. No difference between E2F3 (53%) and NM_017774 (50%) was detectable (P=0.4186, average knockdown from days 1 to 4). Combined knockdown did not result in a further decrease of individual mRNA levels as compared to separate knockdown, in all tested cell lines. Decreased protein expression of E2F3 in HTB-5, induced by E2F3-specific siRNA, was additionally confirmed by Western blot analysis and results are visualized in Figure 2. The silencing power of NM_017774-specific siRNA could not yet be documented by Western blot, because NM_017774-specific antibodies are currently not available.
The effects of E2F3, NM_017774 and combined gene knockdown on cell proliferation are compared against the effect of nonsense siRNA control and are shown in Figure 3: In the non-amplified cell line (CRL-7930), knockdown of E2F3 as well as NM_017774 resulted in a pronounced decrease of the cell proliferation rate (average over 6 days: E2F3: −43.1%, P=0.004; NM_017774: −48.7%, P=0.006; maximum E2F3: −48.2% at day 4; maximum NM_017774: −55.4% at day 4). Simultaneous knockdown exerted even a stronger proliferation decrease amounting to −57.9% (P=0.008; maximum −69.6% at day 4).
In the amplified cell line (HTB-5), knockdown of E2F3 resulted in a comparable decrease of the cell proliferation, like it had been observed in non-amplified cells (average: −36.7%, P=0.018; maximum −44.7% at day 4). In contrast, knockdown of NM_017774 had no negative influence on cell proliferation in amplified cells (average: +5%, P=0.309; maximum –10.9% at day 4). The combined knockdown (average: −27.7%, P=0.011; maximum −36.8% at day 4) reached values analogous to E2F3 alone.
In summary, knockdown of E2F3 strongly inhibited cell proliferation in 6p22.3-amplified cells (−36.7%), whereas no such effect was observed for NM_017774 (+5%). This argues against a cooperative effect of E2F3 and NM_017774 on cell proliferation. Importantly, even after successful knockdown of E2F3 and NM_017774, the residual amount of mRNA left over in the amplified cell line exceeded the standard mRNA levels of non-amplified and non-siRNA–treated cells by a factor 5–10 (see Figure 4). Nevertheless, knockdown of E2F3 severely inhibited regular cell growth in 6p22.3-amplified cells. This emphasizes E2F3 as the relevant target gene of 6p22.3 amplification. Together with our recent observation that E2F3 expression is linked to rapid proliferation (Oeggerli et al., 2004), these data support an important role of E2F3 as a limiting factor for urothelial cell proliferation. It seems that 6p22.3 amplification conveys massive E2F3 overexpression in order to overcome a molecular bottleneck that prevents accelerated cell proliferation.
Simultaneous knockdown levels of NM_017774 did not lead to reduced growth rates of 6p22.3-amplified cells. In non-amplified cells, however, reduced levels of NM_017774 induced a comparable proliferation reduction (−48.7%) as seen for E2F3 (−43.1%; see Figure 3). The growth reduction was even stronger if both E2F3 and NM_017774 were jointly silenced (−57.9%). The biological function of NM_017774 is unknown yet, but this finding adds additional evidence to the hypothesis (see above) that NM_017774 might be involved in regular cell growth. Conclusively, we found no evidence for NM_017774 to have a possible supportive effect on enhanced cellular proliferation when co-amplified and co-overexpressed alongside E2F3. Our results either suggest that NM_017774 is only accidentally co-amplified because of its spatial neighbourhood to E2F3 (like other genes in the area) and does not have a functional role in 6p22.3 amplification, or that co-amplification of NM_017774 could be involved in another, not yet detected aspect of the disease, that is not linked to enhanced cellular proliferation.
Amplification of 6p22.3 has been reported exclusively in bladder cancer so far (Bruch et al., 2000; Feber et al., 2004; Hurst et al., 2004). The finding of a single case of breast cancer with 6p22.3 amplification in a multitumour TMA experiment, and a second breast cancer that was recently found to be E2F3 amplified in a DNA chip-based copy number analysis in our lab (C Ruiz, unpublished personal observation) (see Table 2) indicates that E2F3 amplification is not limited only to bladder cancer.
Conclusively, our data in conjunction with published evidence strongly indicate that amplification of E2F3 is a hallmark of one genetic pathway in invasive bladder cancer that is followed by approximately one-third of these tumours. These results prompt for a review of existing drug compound databases for potential E2F3 inhibitors.
Altschul SF, Madden TL, Schaffer AA, Zhang J, Zhang Z, Miller W et al. (1997). Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res 25: 3389–3402.
Bruch J, Schulz WA, Haussler J, Melzner I, Bruderlein S, Moller P et al. (2000). Delineation of the 6p22 amplification unit in urinary bladder carcinoma cell lines. Cancer Res 60: 4526–4530.
Elbashir SM, Harborth J, Lendeckel W, Yalcin A, Weber K, Tuschl T . (2001). Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured mammalian cells. Nature 411: 494–498.
Feber A, Clark J, Goodwin G, Dodson AR, Smith PH, Fletcher A et al. (2004). Amplification and overexpression of E2F3 in human bladder cancer. Oncogene 23: 1627–1630.
Hiebert SW, Chellappan SP, Horowitz JM, Nevins JR . (1992). The interaction of RB with E2F coincides with an inhibition of the transcriptional activity of E2F. Genes Dev 6: 177–185.
Hunter T, Pines J . (1994). Cyclins and cancer. II: Cyclin D and CDK inhibitors come of age. Cell 79: 573–582.
Hurst CD, Fiegler H, Carr P, Williams S, Carter NP, Knowles MA . (2004). High-resolution analysis of genomic copy number alterations in bladder cancer by microarray-based comparative genomic hybridization. Oncogene 23: 2250–2263.
Lee JM, Sonnhammer EL . (2003). Genomic gene clustering analysis of pathways in eukaryotes. Genome Res 13: 875–882.
Leone G, DeGregori J, Yan Z, Jakoi L, Ishida S, Williams RS et al. (1998). E2F3 activity is regulated during the cell cycle and is required for the induction of S phase. Genes Dev 12: 2120–2130.
Lindl TaB J . (1989). Zell- und Gewebekultur. Gustav Fischer Verlag: Stuttgart, New York.
Livak KJ, Schmittgen TD . (2001). Analysis of relative gene expression data using real-time quantitative PCR and the 2(−Delta Delta C(T)) method. Methods 25: 402–408.
Oeggerli M, Tomovska S, Schraml P, Calvano-Forte D, Schafroth S, Simon R et al. (2004). E2F3 amplification and overexpression is associated with invasive tumor growth and rapid tumor cell proliferation in urinary bladder cancer. Oncogene 23: 5616–5623.
Paddison PJ, Caudy AA, Hannon GJ . (2002). Stable suppression of gene expression by RNAi in mammalian cells. Proc Natl Acad Sci USA 99: 1443–1448.
Qian Y, Luckey C, Horton L, Esser M, Templeton DJ . (1992). Biological function of the retinoblastoma protein requires distinct domains for hyperphosphorylation and transcription factor binding. Mol Cell Biol 12: 5363–5372.
Tomovska SRJ, Suess K, Wagner U, Rozenblum E, Gasser TC, Moch H et al. (2001). Molecular cytogenetic alterations associated with rapid tumor cell proliferation in advanced urinary bladder cancer. Int J Oncol 18: 1239–1244.
Wagner S, Beil W, Westermann J, Logan RP, Bock CT, Trautwein C et al. (1997). Regulation of gastric epithelial cell growth by Helicobacter pylori: offence for a major role of apoptosis. Gastroenterology 113: 1836–1847.
Wang W, Wu W, Desai T, Ward DC, Kaufman SJ . (1995). Localization of the alpha 7 integrin gene (ITGA7) on human chromosome 12q13: clustering of integrin and Hox genes implies parallel evolution of these gene families. Genomics 26: 568–570.
We thank Rosmarie Chaffard, Bettina Zwyssig, Veronika Bättig, Barbara Stalder, Kirsten Struckmann and Alexander Rufle for their excellent technical support. This work was supported by the Swiss National Science Foundation (3100A0–100807/1).
About this article
Cite this article
Oeggerli, M., Schraml, P., Ruiz, C. et al. E2F3 is the main target gene of the 6p22 amplicon with high specificity for human bladder cancer. Oncogene 25, 6538–6543 (2006). https://doi.org/10.1038/sj.onc.1209946
- bladder cancer
- amplification target gene
A novel LncRNA transcript, RBAT1, accelerates tumorigenesis through interacting with HNRNPL and cis-activating E2F3
Molecular Cancer (2020)
Nature Reviews Cancer (2009)
Tiling resolution array CGH and high density expression profiling of urothelial carcinomas delineate genomic amplicons and candidate target genes specific for advanced tumors
BMC Medical Genomics (2008)
Laboratory Investigation (2008)