Logo of neoplasiaLink to Publisher's site
Neoplasia. 2006 Jun; 8(6): 465–469.
PMCID: PMC1601467

Three-Color FISH Analysis of TMPRSS2/ERG Fusions in Prostate Cancer Indicates That Genomic Microdeletion of Chromosome 21 Is Associated with Rearrangement1


The recent description of novel recurrent gene fusions in ∼80% of prostate cancer (PCa) cases has generated increased interest in the search for new translocations in other epithelial cancers and emphasizes the importance of understanding the origins and biologic implications of these genomic rearrangements. Analysis of 15 PCa cases by reverse transcription-polymerase chain reaction was used to detect six ERG-related gene fusion transcripts with TMPRSS2. No TMPRSS2/ETV1 chimeric fusion was detected in this series. Three-color fluorescence in situ hybridization confirms that TMPRSS2/ERG fusion may be accompanied by a small hemizygous sequence deletion on chromosome 21 between ERG and TMPRSS2 genes. Analysis of genomic architecture in the region of genomic rearrangement suggests that tracts of microhomology could facilitate TMPRSS2/ERG fusion events.

Keywords: TMPRSS2/ERG rearrangements, microhomology, genome architecture, FISH, genomic microdeletions


As the most common neoplasm and second leading cause of cancer mortality in North American men [1], remarkably little is known about crucial events in prostatic carcinogenesis. Two related discoveries over the last 12 months, however, have the potential to considerably contribute to knowledge in this area. Initially, Petrovicsetal. [2] described an ETS-related gene (ERG1) in the prostate cancer (PCa) transcriptome, suggesting that it was the most commonly overexpressed proto-oncogene in malignant prostatic tissues. Independently, Tomlins et al. [3] described novel gene fusions involving either ERG1 or a related gene ETV1 or ETV4 [4], which is thought to underlie the mechanism of overexpression. These studies used a novel computational analysis of microarray data called Cancer Outlier Profile Analysis (COPA) and was subsequently confirmed by reverse transcription-polymerase chain reaction (RT-PCR) and fluorescence in situ hybridization (FISH). A confirmatory study also detected ERG rearrangements but did not detect ETV alteration [5]. The gene fusions described to date involve the androgen-sensitive TMPRSS2 gene and three aforementioned members of the ETS family of transcription factors (ERG, ETV1, and ETV4). We provide independent confirmation of translocation results in 6 of 15 (40%) PCa specimens and additionally describe two novel variant transcripts in the same multicentric tumor. In addition, break-apart three-color FISH was used to confirm that a deletion between TMPRSS2 and ERG on chromosome 21 was associated with gene fusion events.

Materials and Methods

Fifteen typical PCa tissue samples were obtained from radical prostatectomies. Part of the tissue was embedded in frozen section medium and stored at -80°C until a tumor-rich tissue had been selected for RNA extraction. FISH analysis was performed on adjacent sections. Tissue sections were also stained with hematoxylin and eosin and subjected to standard histopathological evaluation to determine pathological grade, tumor content, and the presence or absence of single/multifocal PCa. Gleason scores ranged from 6 to 9, and one tumor sample (78-01) was considered to have multicentric histology. To determine the prevalence of ETS rearrangement, RT-PCR amplification (GeneAmp RNA PCR Core Kit; Applied Biosystems, Foster City, CA) was carried out as described by Tomlins et al. [3]. Duplicated RT-PCR products from 15 PCa cases were sized by electrophoresis on a 1.5% agarose gel and by DNA 1000 LabChip Kit (Agilent 2100 Bioanalyzer; Agilent Technologies, Inc., Palo Alto, CA). These products were then gel-purified and sequenced directly using an ABI PRISM 377 (Applied Biosystems) sequencer (Figure 1, B and C).

Figure 1
Rearrangement of the TMPRSS2 and ERG genes in PCa. (A) RT-PCR products from six PCa cases were sized using the Agilent 2100 Bioanalyzer. The fragments were analyzed with a ladder marker to determine the size of each variant TMPRSS2/ERG transcript. Depending ...

To confirm the presence of TMPRSS2/ERG fusions, we used interphase FISH assays on corresponding frozen sections. A break-apart FISH strategy was employed in the analysis of ERG gene rearrangement using bacterial artificial chromosome (BAC) DNA probes published previously [3]. This approach consisted of two DNA probes positioned at opposite sides of the breakpoint region of the ERG gene (ERG 5′ and ERG 3′ loci) and differential labeling using the Ulysis Nucleic Acid Labeling kit (Molecular Probes, Eugene, OR). The OregonGreen-labeled RP11-95I21 BAC probe spans the ERG 5′ and extends inward into exon 10. The red-fluorescein-labeled RP11-476D17 BAC probe spans the ERG 3′ locus and extends inward past exon 4. There is a 35-kb gap between the 3′ and 5′ ERG probes. The TMPRSS2 gene was identified using the Pacific Blue-labeled RP11-35C4 BAC probe, which starts 2.7 Mb from the 5′ end of the ERG gene (Figure 2). The FISH criteria used to evaluate TMPRSS2/ERG rearrangement were as follows: [1] visualization of separate green 5′ ERG and red 3′ ERG signals, and [2] enumeration of each green, red, and blue signal.

Figure 2
Location and names of the BAC probes and gene locations used in the analysis. Gene locations are taken from the May 2004 assembly of the UCSC Genome Browser. Numbers indicate basepair location along the chromosome. Colors correspond to fluorochromes used ...

DAPI-stained tumor nuclei (dark blue) were identified in an adjacent H&E-stained frozen tissue. Normal signal patterns of the probes were confirmed by the colocalization of OregonGreen-labeled 5′ ERG (green signals), AlexaFluor 594-labeled 3′ ERG (red signals), and Pacific Blue-labeled TMPRSS2 (pale blue signals) in normal peripheral lymphocyte metaphase cells and in normal interphase cells (Figure 3A). ERG rearrangement was confirmed by the split of one of the colocalized signals, in addition to a fused signal of the unaffected chromosome 21 (Figure 3B). A minimum of 300 signals per probe was counted to confirm the TMPRSS2/ERG rearrangement in PCa specimens previously analyzed by RT-PCR. A decreased ratio for the 5′ ERG probe mapping to the genomic interval between the TMPRSS2 region and 3′ ERG (Figure 2) was indicative of hemizygous deletion. These experiments were optimized using FISH ratios present in normal adjacent tissues, and deletion cutoff values were defined as a ratio of green 5′ ERG signal to red 3′ ERG (≤ 0.80) [6] when break-apart FISH analysis indicated that a fusion genomic rearrangement was present.

Figure 3
FISH analysis showing rearrangement of TMPRSS2 and ERG genes in PCa. (A) FISH confirms the colocalization of OregonGreen-labeled 5′ ERG (green signals), AlexaFluor 594-labeled 3′ ERG (red signals), and Pacific Blue-labeled TMPRSS2 (light ...

Results and Discussion

Of the 15 tumors analyzed, 6 (40%) possessed an ERG rearrangement, confirming the FISH findings of a previous study (55%; 16 of 29) [3]. Although none of our samples had an ETV1 rearrangement, this observation is not surprising because the original paper only detected the ETV1 fusion transcript in a smaller proportion of samples (25%; 7 of 29) and because the confirmatory study by Soller et al. [5] did not detect ETV1 alteration in 18 tested samples. Using the Agilent 2100 Bioanalyzer, fragment lengths were precisely determined. Five of six positive ERG/TMPRSS2 fusions had lengths consistent with published findings; however, one tumor sample (PCa 78-01) contained two variant TMPRSS2/ERG transcripts (Figure 1A, lane 6) of 430 and 350 bp. Automated DNA sequencing of gel-purified transcripts from PCa 66-01 and PCa 79-01 (both typical TMPRSS2/ERG fusions) and from PCa 78-01 (upper and lower fragments) confirmed the fusion of TMPRSS2 with the ERG gene. Sequence analysis of both gel-purified fragments from PCa 78-01 revealed two distinct in-frame rearrangements generating novel TMPRSS2/ERG fusion transcripts. The variant TMPRSS2/ERG transcript of 430 bp resulted in the fusion of exons 1 and 2 of the TMPRSS2 gene and of exons 5 and 6 of the ERG gene, and the smaller variant TMPRSS2/ERG transcript of 350 bp resulted in the fusion of exon 1 of the TMPRSS2 gene to exons 5 and 6 of the ERG gene (Figure 1C). Although it is conceivable that these fusion events represent independent genomic alterations occurring within one clonal tumor outgrowth, this interpretation was considered less likely given the multicentric histology of this particular tumor. The detection of these two new variant TMPRSS2/ERG fusions brings the total number of ETS gene fusions in PCa described to date to 11 (Table 1).

Table 1
ETS Fusions Documented in PCa to Date.

Applying the break-apart green (ERG 5′ locus) and red (ERG 3′ locus) FISH strategies allowed for the confirmation of TMPRSS2/ERG fusion in frozen sections from six different patients. Within these six patient samples, deletion between TMPRSS2 and ERG was detected in three samples. In all cases, enumeration with flanking TMPRSS2 (pale blue) and 5′ ERG (green) in tumors showed that the ratio was < 0.80, consistent with deletion affecting the intervening genomic DNA. Additional preliminary data have confirmed that a subset of these tumors demonstrated hemizygous deletion by oligonucleotide array-based comparative genomic hybridization (aCGH) (Human Genome CGH Microarray Kit 44B; Agilent Technologies, Inc., Palo Alto, CA) intervening in TMPRSS2/ERG fusion (data not shown).

Genomic Architecture and Origin of Genetic Translocations

Although there are many transcripts recognized, the human ETV1 gene has up to 14 exons with a DNA binding domain in the last exon, whereas the ERG gene has 11 exons with recognized functional domains occurring across exons 5 and 6 [pointed (PNT) domain interaction] and exon 11 (ETS DNA binding domain). TMPRSS2 has 14 exons with functional domains in the latter half of the protein only. Both ERG and TMPRSS2 lie on chromosome 21 at cytobands 21 q22.3 and 21q22.2, respectively, with approximately 3 Mb between them, and TMPRSS2 localized more telomerically than ERG. As can be seen in Figure 2, the 5′ end of both genes faces the telomere. Both genes have the same transcriptional orientation and are separated by 3 Mb of DNA. Given this genomic organization and the observation that 5′ TMPRSS2 fuses in-frame with 3′ ERG, interstitial deletion of the intervening 3 Mb of DNA must take place. Indeed, our three-color FISH analysis confirms the loss of genomic content from this region of chromosome 21. This finding raises the question of concomitant haploinsufficiency of one or more genes mapping to this deleted interval. Deletion of the TMPRSS2 coding region resulting from fusion rearrangement may lead to haplo-insufficiency of the gene. However, a TMPRSS2 knockout mouse with no apparent phenotype was recently reported [7]. Examination of the 13 genes within this region of chromosome 21 (ETS2, DSCR, BRWD1, HMGN1, C21orf13, SH3BGR, B3GALT5, PCP4, DSCAM, BACE2, FAM3, MX2, and MX1) identified one candidate locus HMGN1. Knock-out models of this gene demonstrated that loss increased N-cadherin expression [8] (which has been noted in high-grade PCa) [9] and altered G2/M checkpoint [10]. Interestingly, COPA analysis using the Oncomine database [11] of PCa (an expression microarray dataset) [12] demonstrated that, when ERG was overexpressed, there was concomitant reduction in the expression of HMGN1.

The clarification of the exact genomic architecture will involve the sequencing or the fiber FISH analysis of samples that have been found to express the fusion transcript. Similar rearrangements involving ETS family members in the Ewing family of tumors and hematologic malignancies have been shown to involve classic, complex, or variant translocations. Interstitial deletions have been described in leukemias [13] and congenital syndromes, and are thought to be due to defective homologous recombination [14], which is perhaps related to DNA matrix attachment regions or areas of micro-homology. In this regard, it is noteworthy that, for every case of ERG fusion transcript, there is at least one area of up to 300 bp on the intron following the transcribed TMPRSS2 exon that displays microhomology with up to 90% identity to multiple areas on the intron preceding the relevant ERG exon of the transcript.

Functional Implications

The most pressing need in future work is the need to demonstrate the existence of functional protein from these transcripts. From the analysis of fusion transcripts, it is evident that TMPRSS2 contributes an androgen-responsive regulatory region to ERG, ETV1, and ETV4. Identification of the contribution of other structural regions, such as the PNT domain, which is important for TMPRSS2/ERG and TMPRSS2/ETV1 function, and elucidation of the precise functional properties of each fusion protein are areas of significant future interest. Indeed, the only other study examining ERG expression in PCa, using an antibody to the conserved C-terminus, found expression in 7 of 25 of high-grade cancers [15].

The ETS family of transcription factors encodes nuclear transcription factors with an evolutionarily conserved ETS domain of 85 amino acids that mediate binding to purine-rich DNA residues with more than 400 target genes in the genome that are either positively or negatively regulated by them [16]. One of the most interesting findings of the study by Tomlins et al. [3] is the suggestion that fusion genes only exist in PCa and not in the precursor lesion, high-grade prostatic intraepithelial neoplasia (HPIN). This suggests that a consequence of the expression of these gene products may be the facilitation of the cellular transitions of HPIN precursors to PCa. There is some evidence that ETS-related genes are involved in invasiveness and metastases [16], but these novel transcriptomes are yet to be fully characterized. Evidence to date suggests that ERG may be involved in histone methylation [17], inhibition of apoptosis [18], and transcriptional synergy with Jun/Fos heterodimers [19], and, when overexpressed, display transforming abilities [20]. However, potentially more relevant insights into their dysregulated mechanisms can be gained from examining their role in the Ewing family of tumors and hematologic neoplasms, where similar fusion products are known to occur. For example, both ERG and ETV1 fusion products have been shown to downregulate the TGF-β2 receptor [21]—a potential tumor suppressor in PCa [22].

Clinical Implications

The clinical implications of translocation remain unclear. Intuitively, a translocation would appear to be detrimental; however, Petrovics et al. [2] described ERG overexpression in a subset of 95 PCa patients and noted that high levels were associated with a variety of different positive prognostic variables such as longer prostate-specific antigen (PSA) recurrence-free survival, well and moderately differentiated stages, lower pathological T stage, and negative surgical margins. Subsequently, the paper by Tomlins et al. [3] indicated that, in greater than 90% of cases overexpressing ERG or ETV1, a TMPRSS2 fusion event was detectable. The possibility that the presence of a gene fusion is a positive prognostic factor raises the possibility that there may be multiple pathways toward prostatic carcinogenesis, with varying malignant potentials. In our small cohort, we did not notice any particular clinical outcome with samples carrying the rearrangement, and clinical characteristics were distributed unremarkably; tumor stages ranged from T2a to T3b—five tumors were Gleason 7 and one tumor was Gleason 9. A future area of interest for future investigation is the response to androgen deprivation in PCa. The response is known to be heterogeneous, with the median response before PSA relapse being approximately 24 months [23]. The nature of this genomic rearrangement suggests that it may be fundamental to driving carcinogenesis; thus, further hypotheses may suggest that those with an ETS gene fusion have a prolonged response to androgen deprivation than those without.


aCGHarray-based comparative genomic hybridization
BACbacterial artificial chromosome
PCaprostate cancer
COPACancer Outlier Profile Analysis
FISHfluorescence in situ hybridization
HPINhigh-grade prostatic intraepithelial neoplasia
PCRpolymerase chain reaction
PSAprostate-specific antigen
RT-PCRreverse transcription-polymerase chain reaction


1This work was supported by the National Cancer Institute of Canada, the Canadian Prostate Cancer Research Initiative Program, a predoctoral traineeship award from the Department of Defense Prostate Cancer Research Program from the Congressionally Directed Medical Research Program, and the University of Toronto Graduate Training Program in Molecular Medicine.


1. Jemal A, Murray T, Ward E, Samuels A, Tiwari RC, Ghafoor A, Feuer EJ, Thun MJ. Cancer statistics, 2005. CA Cancer J Clin. 2005;55:10–30. [PubMed]
2. Petrovics G, Liu A, Shaheduzzaman S, Furasato B, Sun C, Chen Y, Nau M, Ravindranath L, Chen Y, Dobi A, et al. Frequent overexpression of ETS-related gene-1 (ERG1) in prostate cancer transcriptome. Oncogene. 2005;24:3847–3852. [PubMed]
3. Tomlins SA, Rhodes DR, Perner S, Dhanasekaran SM, Mehra R, Sun XW, Varambally S, Cao X, Tchinda J, Kuefer R, et al. Recurrent fusion of TMPRSS2 and ETS transcription factor genes in prostate cancer. Science. 2005;310:644–648. [PubMed]
4. Tomlins SA, Mehra R, Rhodes DR, Smith LR, Roulston D, Helgeson BE, Cao X, Wei JT, Rubin MA, Shah RB, et al. TMPRSS2:ETV4 gene fusions define a third molecular subtype of prostate cancer. Cancer Res. 2006;66:3396–3400. [PubMed]
5. Soller MJ, Isaksson M, Elfving P, Soller W, Lundgren R, Panagopoulos I. Confirmation of the high frequency of the TMPRSS2/ERG fusion gene in prostate cancer. Genes Chromosomes Cancer. 2006;45:717–719. [PubMed]
6. Staff S, Isola JJ, Johannsson O, Borg A, Tanner MM. Frequent somatic loss of BRCA1 in breast tumours from BRCA2 germ-line mutation carriers and vice versa. Br J Cancer. 2001;85:1201–1205. [PMC free article] [PubMed]
7. Kim TS, Heinlein C, Hackman RC, Nelson PS. Phenotypic analysis of mice lacking the Tmprss2-encoded protease. Mol Cell Biol. 2006;26:965–975. [PMC free article] [PubMed]
8. Rubinstein YR, Furusawa T, Lim JH, Postnikov YV, West KL, Birger Y, Lee S, Nguyen P, Trepel JB, Bustin M. Chromosomal protein HMGN1 modulates the expression of N-cadherin. FEBS J. 2005;272:5853–5863. [PMC free article] [PubMed]
9. Jaggi M, Nazemi T, Abrahams NA, Baker JJ, Galich A, Smith LM, Balaji KC. N-cadherin switching occurs in high Gleason grade prostate cancer. Prostate. 2006;66:193–199. [PubMed]
10. Birger Y, Catez F, Furusawa T, Lim JH, Prymakowska-Bosak M, West KL, Postnikov YV, Haines DC, Bustin M. Increased tumorigenicity and sensitivity to ionizing radiation upon loss of chromosomal protein HMGN1. Cancer Res. 2005;65:6711–6718. [PMC free article] [PubMed]
11. Rhodes DR, Yu J, Shanker K, Deshpande N, Varambally R, Ghosh D, Barrette T, Pandey A, Chinnaiyan AM. ONCOMINE: a cancer microarray database and integrated data-mining platform. Neoplasia. 2004;6:1–6. [PMC free article] [PubMed]
12. Lapointe J, Li C, Higgins JP, van de Rijn M, Bair E, Montgomery K, Ferrari M, Egevad L, Rayford W, Bergerheim U, et al. Gene expression profiling identifies clinically relevant subtypes of prostate cancer. Proc Natl Acad Sci USA. 2004;101:811–816. [PMC free article] [PubMed]
13. Kolomietz E, Al-Maghrabi J, Brennan S, Karaskova J, Minkin S, Lipton J, Squire JA. Primary chromosomal rearrangements of leukemia are frequently accompanied by extensive submicroscopic deletions and may lead to altered prognosis. Blood. 2001;97:3581–3588. [PubMed]
14. Kolomietz E, Meyn MS, Pandita A, Squire JA. The role of Alu repeat clusters as mediators of recurrent chromosomal aberrations in tumors. Genes Chromosomes Cancer. 2002;35:97–112. [PubMed]
15. Gavrilov D, Kenzior O, Evans M, Calaluce R, Folk WR. Expression of urokinase plasminogen activator and receptor in conjunction with the ets family and AP-1 complex transcription factors in high grade prostate cancers. Eur J Cancer. 2001;37:1033–1040. [PubMed]
16. Seth A, Watson DK. ETS transcription factors and their emerging roles in human cancer. Eur J Cancer. 2005;41:2462–2478. [PubMed]
17. Yang L, Mei Q, Zielinska-Kwiatkowska A, Matsui Y, Blackburn ML, Benedetti D, Krumm AA, Taborsky GJ, Jr, Chansky HA. An ERG (ets-related gene)-associated histone methyltransferase interacts with histone deacetylases 1/2 and transcription co-repressors mSin3A/B. Biochem J. 2003;369:651–657. [PMC free article] [PubMed]
18. Yi H, Fujimura Y, Ouchida M, Prasad DD, Rao VN, Reddy ES. Inhibition of apoptosis by normal and aberrant Fli-1 and erg proteins involved in human solid tumors and leukemias. Oncogene. 1997;14:1259–1268. [PubMed]
19. Camuzeaux B, Spriet C, Heliot L, Coll J, Duterque-Coquillaud M. Imaging Erg and Jun transcription factor interaction in living cells using fluorescence resonance energy transfer analyses. Biochem Biophys Res Commun. 2005;332:1107–1114. [PubMed]
20. Hart AH, Corrick CM, Tymms MJ, Hertzog PJ, Kola I. Human ERG is a proto-oncogene with mitogenic and transforming activity. Oncogene. 1995;10:1423–1430. [PubMed]
21. Im YH, Kim HT, Lee C, Poulin D, Welford S, Sorensen PH, Denny CT, Kim SJ. EWS-FLI1, EWS-ERG, and EWS-ETV1 oncoproteins of Ewing tumor family all suppress transcription of transforming growth factor beta type II receptor gene. Cancer Res. 2000;60:1536–1540. [PubMed]
22. Zhu B, Kyprianou N. Transforming growth factor beta and prostate cancer. Cancer Treat Res. 2005;126:157–173. [PubMed]
23. Roudier MP, True LD, Higano CS, Vesselle H, Ellis W, Lange P, Vessella RL. Phenotypic heterogeneity of end-stage prostate carcinoma metastatic to bone. Hum Pathol. 2003;34:646–653. [PubMed]

Articles from Neoplasia (New York, N.Y.) are provided here courtesy of Neoplasia Press
PubReader format: click here to try


Save items

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • Compound
    PubChem chemical compound records that cite the current articles. These references are taken from those provided on submitted PubChem chemical substance records. Multiple substance records may contribute to the PubChem compound record.
  • Gene
    Gene records that cite the current articles. Citations in Gene are added manually by NCBI or imported from outside public resources.
  • GEO Profiles
    GEO Profiles
    Gene Expression Omnibus (GEO) Profiles of molecular abundance data. The current articles are references on the Gene record associated with the GEO profile.
  • HomoloGene
    HomoloGene clusters of homologous genes and sequences that cite the current articles. These are references on the Gene and sequence records in the HomoloGene entry.
  • MedGen
    Related information in MedGen
  • PubMed
    PubMed citations for these articles
  • Substance
    PubChem chemical substance records that cite the current articles. These references are taken from those provided on submitted PubChem chemical substance records.

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...