12p-Amplicon structure analysis in testicular germ cell tumors of adolescents and adults by array CGH

Article metrics


All invasive testicular germ cell tumors of adolescents and adults (TGCTs), that is, seminomas and nonseminomas, show gain of 12p sequences, mostly as isochromosomes. Although several candidate genes have been suggested, the relevant gene(s) have not been identified yet. About 10% of testicular seminomas, however, show a more restricted amplification of the 12p11.2–p12.1 region, in which the various amplicons show an apparent overlap, allowing for the shortest region of amplification overlap approach, aiming at the identification of pathogenetically relevant sequences residing in this region. Here we report on a high-resolution 12p-amplicon architecture analysis using microarray-based comparative genomic hybridization, the results of which were subsequently confirmed by fluorescent in situ hybridization studies. The 12p-specific microarray contained 63 positionally selected BAC clones, which are more or less evenly distributed over the short arm of chromosome 12 (average spacing: less than 500 Kb), including 20 clones within the region of amplification. Out of a series of 17 seminomas, seven seminomas showed amplification of the whole amplicon region, of which three showed a dip in T/R value in the center of the amplified area. A more complex amplification pattern was found in the other 10 seminomas: three showed predominant amplification at the centromeric border; one mainly at the telomeric border; six showed a balanced amplification of both the centromeric and telomeric regions. The only nonseminoma investigated showed a structure in which the centromeric border was only amplified. These data support a mechanistic model in which at least two 12p genes, situated at the border regions of the amplicon, are positional candidates capable of actively supporting tumor progression in TGCTs.


High-level amplification of genomic fragments points towards the existence of one or more relevant, dosage-dependent proto-oncogene genes, even if the amplification is present in only 10% of the tumors. This has, for example, been demonstrated in a number of tumor types, including the breast (Albertson et al., 2000; Monni et al., 2001), prostate (Reiter et al., 2000), esophageal (Nonet et al., 2001; Yasui et al., 2001), ovarian (Guan et al., 2001), cervical and colon (Zhou et al., 1998), and pancreatic cancers (Huntsman et al., 1999). Final identification of the responsible gene(s), however, might be complicated due to the fact that multiple genes are often residing within the region of amplification, which may encompass several megabases in size in individual tumors. Determination of the so-called shortest region of amplification overlap (SRO) is the first step to reduce the region of interest, and consequently the number of positional candidate genes. Subsequently, a detailed investigation of the amplicon architecture itself is informative for pinpointing the responsible gene(s) (Albertson et al., 2000), and for elucidating the mechanism underlying amplicon development.

All invasive human testicular germ cell tumors (TGCTs) of adolescents and adults, both seminomas and nonseminomas, show gain of (part of) the short arm of chromosome 12 (Looijenga and Oosterhuis (2002) for review). Therefore, it is generally accepted that this chromosomal arm harbors genes whose increased expression, caused by increased copy numbers, triggers pathogenesis of this cancer. Furthermore, it has been demonstrated that gain of 12p is not present in the precursor lesion of TGCTs, known as carcinoma in situ (Looijenga et al., 2000; Rosenberg et al., 2000; Summersgill et al., 2001). This supports a model in which overexpression of a gene or genes on 12p is related to the invasive growth of the tumor cells rather than initiation of tumorigenesis. Various attempts have been undertaken to identify the gene of interest, including KRAS2 (Mulder et al., 1989; Moul et al., 1992; Ridanpää et al., 1993; Olie et al., 1995) and CCND2 (Sicinski et al., 1996; Houldsworth et al., 1997; Schmidt et al., 2001; Skotheim et al., 2002). However, formal proof that these genes are involved is missing so far.

Several years ago, we initiated a study on TGCTs, in particular seminomas, with a restricted 12p amplification (Suijkerbuijk et al., 1994; Mostert et al., 1996, 1998). We subsequently showed that the breakpoints of the amplicon cluster significantly (Roelofs et al., 2000; Zafarana et al., 2002), which could indicate that two relevant genes are present within the SRO, one at the centromeric and the other at the telomeric side. Here, we extend these studies by performing a high-resolution copy number analysis on a series of 17 testicular seminomas and one nonseminoma with known high-level 12p amplifications. To this end, BAC array CGH and FISH were used. Data obtained in this way support the notion that at least two different 12p loci may co-operate in the pathogenesis of TGCTs.


In order to perform a detailed analysis of the DNA copy number within the short arm of chromosome 12 in TGCTs, a BAC array containing 63 clones was produced, that covers the complete length of 12p from the telomere (BAC RP11-110 K11) to the centromere (BAC RP11-88 P4) (see Materials and methods and Table 1 for details). The short arm of this chromosome is about 40 000 Kb in length, but the SRO (Roelofs et al., 2000; Zafarana et al., 2002) is only about 3000 Kb and is located between 24 500 and 27 000 Kb from the telomere. Therefore, the array contains a more dense distribution of clones from this area. A detailed physical map of the BAC clones located in this region is shown in Figure 1. Detailed information on the position and identity of all the BAC clones used in the array is provided in Table 1.

Table 1 Overview of the RP11-BACs present on the array
Figure 1

Representative example of a 12p array-CGH using DNA of a normal diploid male control. The Y-axis shows the T/R value (test over reference intensity value) and the numbering depicted on the X-axis is in Kb, where 0 Kb is the telomere and 40 000 Kb is the centromere. Each point in the graph represents the T/R value for an individual BAC clone, positioned along the 12p arm. A detailed physical map of the BAC clones within the SRO of amplification is shown below the graph. The BACs indicated with an asterisk were not included in the array, but used for the fluorescent in situ hybridization (FISH) experiments. The drawing is to scale and the BAC clones are numbered according to Table 1. The probes used for FISH are indicated, as well as the position of the annotated genes

To test the reliability of the BAC array generated, we hybridized normal male diploid DNA as well as three seminomas with a previously identified gain of the complete short arm of chromosome 12, without a restricted 12p amplification. A representative example of a normal diploid male control array-CGH experiment is depicted in the upper graph of Figure 1. The graph shows a homogeneous pattern along the complete short arm of chromosome 12 from the telomere (0 Kb) to the centromere (40 000 Kb). A series of 23 TGCTs (22 seminomas and one nonseminoma), all known to contain a restricted 12p amplification, were investigated using the same approach. Five seminomas were noninformative in the array-CGH procedure, that is, no restricted 12p amplification besides gain of all the 12p clones was identified. In fact, these five tumors showed a heterogeneous presence of the amplification within the tumor, as determined by FISH on tissue sections (data not shown). The array-CGH analysis on the other 17 seminomas and the one nonseminoma confirmed the presence of a restricted amplification on the short arm of chromosome 12 at the 12p11.2–p12.1 region (see Figure 2a–e, left panel for representative examples). Since the BACs included in the array could all be physically positioned (see Table 1 and Figure 1), the physical size of the various amplicons could be determined precisely. Data regarding the physical description of these amplicons as well as the SRO are summarized in Table 2. The borders of the amplicon, as previously determined by FISH on tissue sections (Mostert et al., 1998; Roelofs et al., 2000; Zafarana et al., 2002), were found to be fully in line with the data obtained by array CGH. Since 20 out of the 68 BACs map within the SRO, the array data also allowed a detailed analysis of the amplicon structure for each individual tumor. Representative examples are shown in Figure 2a–e, and a summary of the structure of the amplicon per tumor is given in Table 2. Overall, four different patterns of amplification were identified in the seminomas: (1) a simple pattern in which the whole SRO was amplified (seven cases, cases 01–07, Table 2 and Figure 2a); (2) a complex pattern with a predominant amplification of the telomeric border (case 08 in Table 2, and Figure 1b); (3) a complex pattern with predominant amplification of the centromeric border (cases 15, 16, and 17 in Table 2, and Figure 1c); and (4) a complex pattern with a balanced amplification of both the telomeric and centromeric border (six cases, cases 09–14, Table 2 and Figure 1d). The only nonseminoma with a restricted 12p amplification included in this study showed a different pattern from the seminomas, in which only the centromeric border was amplified (case 18 in Table 2 and Figure 1f). These findings were independently confirmed in a selected number of tumors by FISH on tissue sections using BACs present on the array as probes, and the results are shown in the right panels of Figure 2a–f. The BACs used for the FISH analysis were RP11-449 P1, RP11-229 F13, and RP11-615 I16, located, respectively, at the peaks of the telomeric border, at the center, and at the centromeric border of the amplicon. BAC RP11-625 L16 was used for the centromeric peak of the nonseminoma case. The coordinates of the BACs are listed in Table 1. The results (see Figure 2, right panels) show that the number of copies of the probes analysed correlated exactly with the array-CGH data for the same tumor case, also shown in Figure 2. In addition, two BACs (RP11-359 J14 and RP11-681 A17) were used for the FISH analyses. Since these were not represented in the BAC array, but map in the gaps of the contig (indicated by an asterisk in Figure 1 and in Table 1), they were used to further refine the pattern of amplification as described above. In line with the array-CGH results, the FISH data for these clones support the presence of a ‘T/R-dip’ in the regions represented by these probes (data not shown).

Figure 2

Left panels: representative examples of the array CGH using DNA of (a) a seminoma with amplification of the whole 12p11.2–p12.1 region, case 01; (b) a seminoma with predominant amplification of the telomeric region, case 08; (c) a seminoma with predominant amplification of the centromeric region, case 15; (d) a seminoma with amplification of both the telomeric and centromeric region, case 11; (e) a nonseminoma with only amplification of the centromeric region, case 18; Right panels: pictures of the FISH analyses of the corresponding tumors. The tumors shown in (ad) were studied using the BACs RP11-449 P1, RP11-229 F13, and RP11-615 I16. The nonseminoma shown in (e) was studied using the BACs RP11-449 P1, RP11-229 F13, and RP11-625 L16 (see Figure 1 for physical map and Table 1). Note that, in accordance with the array-CGH data, all tumors have lower copy number of the RP11-229 F13 probe compared to the flanking probes, apart from case 15 (c)

Table 2 Summary of the size and architecture of the amplicon found in primary TGCTs (seminoma=01–17; nonseminoma=18), as identified by array-CGH


Identification of 12p genes involved in the development of invasive TGCTs is one of the major goals in the pathobiology of this tumor type (Bourdon et al., 2002; Rodriguez et al., 2003). Identification of these genes will also have direct implications on the diagnosis/subclassification of seminoma- and nonseminoma-like tumors found at other anatomical localizations, including the anterior mediastinum and midline of the brain, as well as the ovary, because these GCTs also show a consistent gain of 12p (Looijenga and Oosterhuis (2002) for review). About 10% of the testicular seminomas contain a restricted 12p amplification (Suijkerbuijk et al., 1994; Korn et al., 1996; Mostert et al., 1996, 1998; Roelofs et al., 2000; Zafarana et al., 2002). These tumors lack the most frequent cytogenetic anomaly leading to gain of 12p in (T)GCTs, the isochromosome 12p (i(12p)) (Sandberg et al., 1996). This suggests the existence of two independent mechanisms resulting in increased copy numbers of genes from 12p, one via i(12p) formation and the other via another, yet unknown, mechanism, which might be followed by a high level amplification of the 12p11.2–p12.1 region (Looijenga and Oosterhuis, 2002; Zafarana et al., 2002). It is of interest to note that a similar situation seems to exist in pancreatic carcinomas (Heidenblad et al., 2002). Investigation of, specifically, the 12p11.2–p12.1 region is of relevance for all (T)GCTs. This is illustrated by our recent finding of overexpression of genes from 12p11.2–p12.1 in all TGCTs, irrespective of the presence of high-level amplification (Rodriguez et al., 2003). Interestingly, it was also found that nonseminomas, which infrequently contain a restricted amplification of the 12p11.2–p12.1 region, show even higher levels of expression of genes mapped to that region than seminomas with the amplification. This illustrates the value of analysis of nonseminomas with a restricted 12p amplification, as carried out in this study, although it is found rarely. In addition, the data support our model that, overall, nonseminomas can upregulate the expression of relevant 12p genes, independent of the number of gene copies (Zafarana et al., 2002). Therefore, one could reason that seminomas are more prone to develop visible genomic amplification of the 12p11.2–p12.1 region than nonseminomas.

Here, we report on our findings using a BAC array-CGH approach on a series of preselected TGCTs containing high-level amplifications of the 12p11.2–p12.1 region. This method allows a detailed investigation of both the size and the architecture of the amplicon. The 12p-specific array used included probes spaced about 0.5 Mb apart, together covering approximately one-third of the short arm of chromosome 12, and representing the complete short arm of chromosome 12, of which a significant number shows a high-level coverage of the hot-spot region of 12p amplification. Within the group of seminomas, overall two amplification patterns were identified: a simple pattern including the whole region and complex patterns in which multiple regions were specifically amplified. The tumors showing involvement of the complete region had the lowest level of amplification (up to a T/R value of three in the array analysis, see Figure 2a and Table 2), whereas the complex patterns were found in the tumors with the highest level of amplification (up to a T/R value of eight in the array analysis) (see Table 2 and Figure 2b–e). The tumors with the complex patterns showed involvement of both the centromeric and telomeric regions. This suggests that the complex patterns of amplification evolved from an initially simpler pattern.

Several genes map within the region of interest on 12p, of which some are interesting candidates. The transcription factor SOX5 is located at the center of the amplicon, a region consistently showing a less-amplified level than the flanking sequences in some of the TGCTs investigated. This result is in line with our previous observation that this gene is not overexpressed in TGCTs with a 12p amplification and is, therefore, not a prime candidate (Roelofs et al., 2000; Rodriguez et al., 2003). The only annotated gene mapped to the telomeric border of the amplicon is EKI1 (Zafarana et al., 2002). This region is predominantly amplified in one of the seminomas, and in combination with the centromeric region in six others. So far, some of the tumors with a restricted 12p amplification showed upregulation of this gene (Zafarana et al., 2002; Rodriguez et al., 2003), which might suggest that EKI1 is the gene of interest in a subgroup of the tumors. The centromeric border of the amplicon contains the previously described DAD-R as candidate (Zafarana et al., 2002). We demonstrated that DAD-R is predominantly overexpressed in all nonseminomas, and in seminomas with a restricted 12p amplification. Recent data from our laboratory indicate that this sequence does not code for any functional protein but, instead, represents an untranslated intronic sequence of BCAT1 which encodes the branched chain aminotransferase 1, a target for cMYC (Benvenisty et al., 1992). Therefore, BCAT1 is the only annotated gene located in the centromeric border region of the amplification. In support to this result, we found that the BCAT1 gene shows the same expression pattern as the putative DAD-R gene (data not shown). BCAT1 is highly expressed in nonseminomas and in seminomas with restricted 12p amplifications, and the expression levels are gene dosage dependent in seminomas without the restricted amplification (Rodriguez et al., 2003; and data not shown). The specific pattern of amplification in the invasive TGCT supports the notion that BCAT1 may be an interesting candidate gene. Other putative genes mapped within the amplified region are currently under investigation.

In conclusion, array CGH-based analysis of TGCTs with restricted 12p amplifications describes in detail the size and architecture of the amplicons present in these tumors. The results obtained provide, for the first time, an in-depth analysis of the superstructure of the 12p amplicons in TGCTs, and thus provide critical information facilitating the identification of a gene(s) causally related to tumor progression in TGCTs.

Materials and methods

Sample handling and characterization

The tumors were collected in the South Western part of the Netherlands. Representative parts of the tumor were snap frozen in liquid nitrogen, or were fixed in 10% formalin for paraffin embedding. Diagnosis was carried out according to the World Health Organization classification (Mostofi and Sesterhenn, 1985), as described previously (Oosterhuis et al., 1989), supported by immunohistochemistry using antibodies directed against germ cell-specific alkaline phosphatase (PLAP), alpha-feto protein (AFP), human chorionic gonadotropin (HCG), and the stem cell factor receptor c-KIT. In addition, immunohistochemistry for OCT3/4 (POU5F1) was applied, as described before (Looijenga et al., 2003). All seminomas were classic seminomas (no anaplastic variant was included). The nonseminoma was diagnosed as a pure embryonal carcinoma. The presence of a restricted 12p amplification was determined by means of FISH on tissue sections using region-specific probes, as described before (Roelofs et al., 2000; Zafarana et al., 2002).

Clones on the BAC array

A total of 68 BAC clones were selected from the June 2002 freeze of the human genome, as represented by the UCSC genome browser (http://genome.ucsc.edu/) and commercially obtained through BACPAC resources (Oakland, USA). These clones were selected to be evenly spread over the entire short arm of chromosome 12, with a higher density in the 12p11.2–12.1 amplification region (an area of 1.7 Mb, see Figure 1 and Table 1). Genomic target DNAs (i.e., BACs) were isolated from pellets originating from 12 ml bacterial cultures using QIAgen R.E.A.L. prep 96 biorobot kits on a QIAgen 9600 biorobot (QIAgen, Valencia, CA, USA), following the instructions of the manufacturer. DNAs from these clones were subsequently used for degenerated oligonucleotide primed (DOP) PCR, for which Taq200 (Stratagene) was found to be superior to all other polymerases tested.

DOP-PCR, spotting and hybridization

DOP-PCR was performed on DNA from all clones essentially as described before (Telenius et al., 1992) with minor modifications (Veltman et al., 2002). The DOP-PCR products were robotically spotted in triplicate onto CMT-GAPS-coated glass slides (Corning, Schiphol-Rijk, The Netherlands) using a Cartesian 5510 Prosys arrayer (Genomic Solutions, Cambridgeshire, UK). DNA preparation, labeling, and hybridization were performed essentially as described before (Veltman et al., 2002).

Image analysis and processing

Slides were scanned and imaged on an Affymetrix 428 scanner (Affymetrix, Santa Clara, CA, USA) using the Affymetrix 428 scanner software package (version 1.0). The acquired microarray images were analysed using Genepix Pro 4.0 (Axon Instruments, Inc., Foster City, CA, USA). DNA spots were automatically segmented, local background was subtracted, and total intensities as well as the fluorescence intensity ratios of the two dyes were calculated for each spot individually. Previously established quality criteria were used for spot inclusion, and data were normalized as described before (Veltman et al., 2002). For all array hybridizations included in this study, 93% of the clones passed these criteria, and were, therefore, included in the final analysis. The average standard deviation for the triplicates was 5%.

Normal clone-to-clone variation in fluorescence intensity ratios was measured in three normal versus normal experiments. The averaged normalized ratio of each clone in these three experiments varied between 0.83 and 1.19 (mean 1.0). Based on the control experiments and on previously published work (Veltman et al., 2002), thresholds for copy number gain and loss were set at 1.2 and 0.8, respectively.

FISH validation experiments

FISH validation experiments were performed on tissue sections of matched tumor samples using routine procedures. Probe labeling, slide preparation, and hybridizations were carried out essentially as described before (Roelofs et al., 2000; Zafarana et al., 2002). A Zeiss epifluorescence microscope equipped with appropriate filters was used for visual examination of the slides. Digital images were captured using a high-performance cooled CCD camera (Photometrics, Trenton, NJ, USA) coupled to a Macintosh Quadra 950 computer. The Image™ FISH software package (Intergen, Purchase, NJ, USA) was used for analysis of the FISH images. Inverted images of DAPI-stained slides were used for chromosome identification. Four tumors, cases 01, 08, 11, and 15, were studied using the BACs RP11-449 P1, RP11-229 F13, and RP11-615 I16. A nonseminoma, case 18, was studied using the BACs RP11-449 P1, RP11-229 F13, and RP11-625 L16. At least 50 nuclei were counted in the cases analysed.


  1. Albertson DG, Ylstra B, Segraves R, Collins C, Dairkee SH, Kowbel D, Kuo WL, Gray JW and Pinkel D . (2000). Nat. Genet., 25, 144–146.

  2. Benvenisty N, Leder A, Kuo A and Leder P . (1992). Genes. Dev., 6, 2513–2523.

  3. Bourdon V, Naef F, Rao PH, Reuter V, Mok SC, Bosl GJ, Koul S, Murty VV, Kucherlapati RS and Chaganti RS . (2002). Cancer Res., 62, 6218–6223.

  4. Guan XY, Sham JS, Tang TC, Fang Y, Huo KK and Yang JM . (2001). Cancer Res., 61, 3806–3809.

  5. Heidenblad M, Jonson T, Mahlamaki EH, Gorunova L, Karhu R, Johansson B and Hoglund M . (2002). Genes Chromosom. Cancer, 34, 211–223.

  6. Houldsworth J, Reuter V, Bosl GJ and Chaganti RS . (1997). Cell Growth Differ., 8, 293–299.

  7. Huntsman DG, Chin SF, Muleris M, Batley SJ, Collins VP, Wiedemann LM, Aparicio S and Caldas C . (1999). Oncogene, 18, 7975–7984.

  8. Korn MW, Olde Weghuis DEM, Suijkerbuijk RF, Schmidt U, Otto T, Du Manoir S, Geurts van Kessel A, Seeber S and Becher R . (1996). Genes Chromosom. Cancer, 17, 78–87.

  9. Looijenga LHJ and Oosterhuis JW . (2002). Analyt. Quant. Cytol. Histol., 24, 263–279.

  10. Looijenga LHJ, Rosenberg C, Van Gurp RJHLM, Geelen E, Van Echten-Arends J, De Jong B, Mostert MC and Oosterhuis JW . (2000). J. Pathol., 19, 187–192.

  11. Looijenga LHJ, Stoop H, De Leeuw PJC, De Gouveia Brazao CA, Gillis AJM, Van Roozendaal KEP, Van Zoelen EJJ, Weber RFA, Wolffenbuttel KP, CVan Dekken H, Honecker F, Bokemeyer C, Perlman EJ, Schneider DT, Kononen J, Sauter G and Oosterhuis JW . (2003). Cancer Res., 63, 2244–2250.

  12. Monni O, Barlund M, Mousses S, Kononen J, Sauter G, Heiskanen M, Paavola P, Avela K, Chen Y, Bittner ML and Kallioniemi A . (2001). Proc. Natl. Acad. Sci. USA, 1, 1.

  13. Mostert MC, Van de Pol M, Olde Weghuis D, Suijkerbuijk RF, Geurts van Kessel A, Van Echten-Arends J, Oosterhuis JW and Looijenga LHJ . (1996). Cancer Genet. Cytogenet., 89, 146–152.

  14. Mostert MC, Verkerk AJMH, Van de Pol M, Heighway J, Marynen P, Rosenberg C, Geurts van Kessel A, van Echten J, Oosterhuis JW and Looijenga LHJ . (1998). Oncogene, 16, 2617–2627.

  15. Mostofi FK and Sesterhenn IA . (1985). Prog. Clin. Biol. Res., 203, 1–34.

  16. Moul JW, Theune SM and Chang EH . (1992). Genes Chromosom. Cancer, 5, 109–118.

  17. Mulder MP, Keijzer W, Verkerk A, Boot AJM, Prins MEF, Splinter TAW and Bos JL . (1989). Oncogene, 4, 1345–1351.

  18. Nonet GH, Stampfer MR, Chin K, Gray JW, Collins CC and Yaswen P . (2001). Cancer Res., 61, 1250–1254.

  19. Olie RA, Looijenga LHJ, Boerrigter L, Top B, Rodenhuis S, Mulder MP and Oosterhuis JW . (1995). Genes Chromosom. Cancer, 12, 110–116.

  20. Oosterhuis JW, Castedo SMMJ, De Jong B, Cornelisse CJ, Dam A, Sleijfer DT and Schraffordt Koops H . (1989). Lab. Invest., 60, 14–20.

  21. Reiter RE, Sato I, Thomas G, Qian J, Gu Z, Watabe T, Loda M and Jenkins RB . (2000). Genes Chromosom. Cancer, 27, 95–103.

  22. Ridanpää M, Lothe RA, nfelt A, Fosså SD, BÆrresen AL and Husgafvel-Pursiainen K . (1993). Environ. Health Perspect., 101, 185–187.

  23. Rodriguez S, Jafer O, Goker H, Summersgill BM, Zafarana G, Gillis AJM, Van Gurp RJHLM, Oosterhuis JW, Lu Y-J, Huddart R, Cooper CS, Clark J, Looijenga LHJ and Shipley J . (2003). Oncogene, 22, 1880–1891.

  24. Roelofs H, Mostert MC, Pompe K, Zafarana G, Van Oorschot M, Van Gurp RJHLM, Gillis AJM, Stoop H, Rodenhuis S, Oosterhuis JW, Bokemeyer C and Looijenga LJ . (2000). Am. J. Pathol., 157, 1155–1166.

  25. Rosenberg C, Van Gurp RJHLM, Geelen E, Oosterhuis JW and Looijenga LHJ . (2000). Oncogene, 19, 5858–5862.

  26. Sandberg AA, Meloni AM and Suijkerbuijk RF . (1996). J. Urol., 155, 1531–1556.

  27. Schmidt BA, Rose A, Steinhoff C, Strohmeyer T, Hartmann M and Ackermann R . (2001). Cancer Res., 61, 4214–4221.

  28. Sicinski P, Donaher JL, Geng Y, Parker SB, Gardner H, Park MY, Robker RL, Richards JS, McGinnis LK, Biggers JD, Eppig JJ, Bronson RT, Elledge SJ and Weinberg RA . (1996). Nature, 384, 470–474.

  29. Skotheim RI, Monni O, Mousses S, Fossa SD, Kallioniemi OP, Lothe RA and Kallioniemi A . (2002). Cancer Res., 62, 2359–2364.

  30. Suijkerbuijk RF, Sinke RJ, Olde Weghuis DEM, Roque L, Forus A, Stellink F, Siepman A, Van de Kaa C, Soares J and Geurts van Kessel A . (1994). Cancer Genet. Cytogenet., 78, 145–152.

  31. Summersgill B, Osin P, Lu YJ, Huddart R and Shipley J . (2001). Br. J. Cancer, 85, 213–220.

  32. Telenius H, Carter NP, Bebb CE, Nordenskjold M, Ponder BA and Tunnacliffe A . (1992). Genomics, 13, 718–725.

  33. Veltman JA, Schoenmakers EF, Eussen BH, Janssen I, Merkx G, van Cleef B, van Ravenswaaij CM, Brunner HG, Smeets D and Geurts van Kessel A . (2002). Am. J. Hum. Genet., 70, 1269–1276.

  34. Yasui K, Imoto I, Fukuda Y, Pimkhaokham A, Yang ZQ, Naruto T, Shimada Y, Nakamura Y and Inazawa J . (2001). Genes Chromosom. Cancer, 32, 112–118.

  35. Zafarana G, Gillis AJM, Van Gurp RJHLM, Olsson PG, Elstrodt F, Stoop H, Millan JL, Oosterhuis JW and Looijenga LHJ . (2002). Cancer Res., 62, 1822–1831.

  36. Zhou H, Kuang J, Zhong L, Kuo W-L, Gray JW, Sahin A, Brinkley BR and Sen S . (1998). Nat. Genet., 20, 189–193.

Download references


This work was financially supported by the Dutch Cancer Society (DDHK 98-1685), Interuniversity Poles of attraction program of Belgium, and the EC COST-B19 action ‘Molecular cytogenetics of solid tumors’ (BG).

Author information

Correspondence to Leendert H J Looijenga.

Rights and permissions

Reprints and Permissions

About this article


  • testicular germ cell tumors
  • 12-amplicon
  • array CGH
  • architecture
  • candidate genes

Further reading