Email updates

Keep up to date with the latest news and content from BMC Genomics and BioMed Central.

Open Access Research article

The use of whole genome amplification to study chromosomal changes in prostate cancer: insights into genome-wide signature of preneoplasia associated with cancer progression

Simon Hughes1, Maisa Yoshimoto1, Ben Beheshti2, Richard S Houlston3, Jeremy A Squire146* and Andrew Evans56

Author Affiliations

1 Applied Molecular Oncology, Ontario Cancer Institute, Princess Margaret Hospital, Toronto, Ontario M5G 2M9, Canada

2 Faculty of Medicine, University of Toronto, Toronto, Ontario M5G 2M9, Canada

3 Section of Cancer Genetics, Institute of Cancer Research, Sutton SM2 5NG, UK

4 Department of Medical Biophysics, Faculty of Medicine, University of Toronto, Toronto, Ontario M5G 2M9, Canada

5 Department of Pathology, Princess Margaret Hospital, Toronto, Ontario M5G 2M9, Canada

6 Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, Ontario M5G 1L5, Canada

For all author emails, please log on.

BMC Genomics 2006, 7:65  doi:10.1186/1471-2164-7-65

The electronic version of this article is the complete one and can be found online at: http://www.biomedcentral.com/1471-2164/7/65


Received:27 November 2005
Accepted:30 March 2006
Published:30 March 2006

© 2006 Hughes et al; licensee BioMed Central Ltd.

This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

Background

Prostate cancer (CaP) is a disease with multifactorial etiology that includes both genetic and environmental components. The knowledge of the genetic basis of CaP has increased over the past years, mainly in the pathways that underlie tumourigenesis, progression and drug resistance. The vast majority of cases of CaP are adenocarcinomas that likely develop through a pre-malignant lesion and high-grade prostatic intraepithelial neoplasia (HPIN). Histologically, CaP is a heterogeneous disease consisting of multiple, discrete foci of invasive carcinoma and HPIN that are commonly interspersed with benign glands and stroma. This admixture with benign tissue can complicate genomic analyses in CaP. Specifically, when DNA is bulk-extracted the genetic information obtained represents an average for all of the cells within the sample.

Results

To minimize this problem, we obtained DNA from individual foci of HPIN and CaP by laser capture microdissection (LCM). The small quantities of DNA thus obtained were then amplified by means of multiple-displacement amplification (MDA), for use in genomic DNA array comparative genomic hybridisation (gaCGH). Recurrent chromosome copy number abnormalities (CNAs) were observed in both HPIN and CaP. In HPIN, chromosomal imbalances involving chromosome 8 where common, whilst in CaP additional chromosomal changes involving chromosomes 6, 10, 13 and 16 where also frequently observed.

Conclusion

An overall increase in chromosomal changes was seen in CaP compared to HPIN, suggesting a universal breakdown in chromosomal stability. The accumulation of CNAs, which occurs during this process is non-random and may indicate chromosomal regions important in tumourigenesis. It is therefore likely that the alterations in copy number are part of a programmed cycle of events that promote tumour development, progression and survival. The combination of LCM, MDA and gaCGH is ideally suited for the identification of CNAs from small cell clusters and may assist in the discovery of potential genomic markers for early diagnosis, or identify the location of tumour suppressor genes (TSG) or oncogenes previously unreported in HPIN and CaP.

Background

The worldwide occurrence CaP is increasing and it has now overtaken lung cancer as the most commonly diagnosed malignancy in men in the Western World. In spite of significant progress in its clinical management, comparatively little is known about the aetiology of the disease and predicting outcome is still a challenge for most clinicians. Histopathologically, CaP displays considerable heterogeneity and can contain a substantial admixture of pre-malignant HPIN glands within cancerous foci. HPIN is currently considered to be the most likely precursor to invasive CaP [1]. However, most of these foci are latent and rarely develop into clinically detectable cancer. Unfortunately, a significant number (3%) do progress and can become aggressive and lethal [2]. The major difficulty facing clinicians is the identification of patients presenting with early stage CaP, who are likely to develop life-threatening disease. As a result, intensive research is currently underway to identify the key alterations that may prove to be important for both classification and prognosis prediction.

The advent and development of CGH, which started in the early 1990's [3], has revolutionised cytogenetics and allowed for genome-wide screening of numerous cancer types and the identification of genomic CNA, which may contribute to cancer development and progression. The accumulation of genetic changes that occurs during the stepwise evolution from normal tissue to metastasis, although likely due to increased genetic instability, may indicate the chromosomal locations of TSG or oncogenes that are important in tumourigenesis. When first introduced, CGH used metaphase chromosomes targets to identify CNAs. However, recent advances have substituted these with arrayed DNA sequences that provide higher resolution, 1 Mb versus 10 Mb, and greater ease of analysis.

Whole genome scans of CaP patients and cell lines have identified consistent, chromosomal alterations, which include recurrent loss of chromosomal regions from 5p, 6q, 8p, 10q, 13, 16q and 17p, in addition to gain on 1q, 3q 7p, 7q, 8q 11p, 17q and Xpq [4-6]. Nonetheless, one potential limitation with current CGH methodology is its requirement for microgram quantities of genomic DNA. When studying cell lines this is not necessarily a matter for concern, as it is relatively straightforward to produce a sufficient quantity of clonal cells to obtain many micrograms of DNA. However, when studying patient tissue if DNA is "bulk extracted" from a heterogeneous non-clonal tumour mass the data obtained will represent an average value for all cells within the specimen and any clinically informative genetic changes, restricted to small cell clusters, may be masked. As a result, when studying a heterogeneous tissue, such as prostate, it is essential to obtain pure cell populations via methods such as macrodissection or LCM. The DNA extracted from the dissected material can then be amplified using a whole genome amplification (WGA) technique [7], such as MDA [8-11]. The combination of LCM and WGA can generate sufficient amounts of DNA for extensive genome analyses. Several large studies have demonstrated the ability of MDA to accurately amplify human genomic DNA. When used in SNP genotyping [12,13] an estimated 99.82% of the genome is covered with a reported concordance, to non-amplified DNA, of greater than 99% [14]. In addition, the sequence error rate is only 7.6 × 106 [13] and previous results obtained from gaCGH, using amplified DNA, are almost indistinguishable from those obtained from non-amplified DNA [10,15].

An understanding of the molecular mechanisms behind cancer development will enable identification of molecular and cytogenetic biomarker(s) that may be useful in predicting early transformation into an aggressive phenotype, as well as, providing fundamental insights into the regulatory pathways of genome integrity that may lead to multistep field cancerization and HPIN and CaP. In the present study, we describe the use of LCM, MDA and gaCGH, using a 1 Mb array (Spectral Genomics, USA) to investigate copy number changes occurring in prostate cancer by the analysis of 7 prostatic samples containing HPIN and 8 CaP specimens.

Results

To determine whether whole genomic amplification exhibited bias or a distortion of imbalance assignments, control CGH experiments comparing the results obtained before and after amplification (normal male versus normal male, normal male versus normal female and DNA from a cancer cell line versus normal control DNA) from the same original DNA samples were performed. These are presented in Additional Files 1, 2, 3 and it is evident that no major distortion of imbalance was introduced by amplification of genomic DNA.

Additional File 1. Genomic comparison of gaCGH results obtained from non-amplified male DNA versus non-amplified male DNA hybridisation and amplified male DNA versus amplified male DNA hybridisation. Blue line indicates the chromosome position plotting of amplified DNA data. Black line indicates the chromosome position plotting of non-amplified DNA data.

Format: PNG Size: 9KB Download fileOpen Data

Additional File 2. Genomic comparison of gaCGH results obtained from non-amplified male versus non-amplified female DNA hybridisation and amplified male DNA versus female DNA hybridisation. Blue line indicates the chromosome position plotting of amplified DNA data. Black line indicates the chromosome position plotting of non-amplified DNA data.

Format: PNG Size: 12KB Download fileOpen Data

Additional File 3. Genomic comparison of gaCGH results obtained from non-amplified DNA from the colorectal cell line DLD1 versus non-amplified control DNA male hybridisation and amplified DNA from the colorectal cell line DLD1 versus amplified control DNA hybridisation. Blue line indicates the chromosome position plotting of amplified DNA data. Black line indicates the chromosome position plotting of non-amplified DNA data. Detection of gain and loss is shown by cyan and pink bars, corresponding to amplified and non-amplified DNA, respectively.

Format: PNG Size: 9KB Download fileOpen Data

A summary of the chromosomal CNAs detected by CGH for the 7 HPIN and 8 CaP samples is shown in Table 1 and 2 and displayed graphically in Figure 1 (for HPIN blue = loss, yellow = gain; for CaP red = loss, green = gain). Figure 2 shows the frequency of CNAs detected at the level of each chromosomes arm for HPIN and CaP. The general pattern of loss or gain was very similar in the HPIN and CaP samples, however CaP samples possessed significantly more aberrations (90 CNAs) than HPIN samples (41 CNAs) (P<0.001, chi square test). The average number of CNAs for the 7 HPIN samples was 5.8 (range 3 to 8, median 6). For the 8 CaP samples an average of 11.25 (range 5 to 18, median 11.5) amplifications or deletions were observed. When studied at the 850 cytoband level (minus X and Y chromosome cytobands), HPIN samples had the total of 65 affected bands (average 9, range 4 to 14) and CaP samples 143 affected breakpoints bands (average 18, range 8 to 27). This difference is statistically significant (P<0.001, chi square test). In addition, for both the HPIN and CaP samples there were a significantly greater number of cytobands lost than gained (P<0.01, chi squared test) (Table 1).

thumbnailFigure 1. Chromosomal alterations observed for HPIN and CaP DNA samples. Blue bars indicate loss in HPIN and yellow bars indicate gain in HPIN. Red bars indicate loss in CaP and green bars indicate gain in CaP.

Table 1. Frequency of chromosome copy number abnormalities and affected cytoband breakpoints in CaP and HPIN.

Table 2. Chromosomal changes in microdissected HPIN and CaP DNA samples.

thumbnailFigure 2. Summary of chromosomal losses and gains in a) HPIN (n = 7) and b) CaP (n = 8). Number of samples of each type with gain or loss of the chromosome arm is shown by black and grey bars, respectively.

In the 7 HPIN samples recurrent chromosomal changes, detected in greater than 25% of cases, were losses found on 8p (7 out of 7 cases, 100%), 6q, 13q, 16q (2 out of 7 cases, 28.5%, each), whereas gains were found on 8q and 16p (3 out of 7 cases, 43%, each), 1p, 7p and 20q (2 out of 7 cases, 28.5%, each). For the 8 CaP samples recurrent chromosomal changes, detected in greater than 25% of tumours, were losses on 8p (8 out of 8 cases, 100%), 10q (6 out of 8 cases, 75%), 13q (5 out of 8 cases, 62.5%), 6q (4 out of 8 cases, 50%), 1p, 16q (3 our of 8 cases, 37.5%), 5q, 12p, 18q (2 out of 8 cases, 25%, each) and gains on 7q (6 out of 8 cases, 75%), 7p, 8q (5 out of 8 cases, 62.5%, each), 4p, 16p, 19p (2 out of 8 cases, 25%).

Small consensus regions of consistent CNA were observed for both HPIN and CaP samples (Table 3), in >25% of samples. For example, 8p region was consistently lost in HPIN (100%), but within the cytoband 8p11.23-p23.2 that was commonly lost in >50% of the HPIN samples. Similarly, 8p was consistently lost in CaP (100%), but within the cytoband 8p11.22-p23.2 that was commonly lost in >60% of the CaP samples. A list of the genes present within these consensus or commonly lost regions is displayed in Table 3.

Table 3. Consensus regions of copy number gain. Names in bold indicate those genes that have been implicated in prostate cancer. The * symbol indicates the same candidate genes from HPIN.

Discussion

Chromosome copy number abnormalities (CNAs) are common in most cancers, with specific regions of amplification or deletion being associated with specific tumour types, stages or outcomes [16-18]. The introduction of metaphase CGH [3] to study these CNAs has eliminated the requirement to obtain metaphase spreads from the tumour samples, which was often challenging due to technical difficulties in culturing certain tissues. Though metaphase CGH allows for more samples to be examined, it provides a relatively limited resolution and still requires substantial cytogenetic experience to analyse the results. The recent development of array CGH has opened up the field of CGH research and permitted more laboratories to study an ever-increasing number of tumour types and stages. The application of BAC microarrays to analyse human samples is straightforward and their high-throughput nature makes them the method of choice for rapid detection of genetic alterations.

When analysing heterogeneous tumour samples, the gaCGH results obtained from "bulk extracted" DNA are likely to be inaccurate. However, with the use of LCM and MDA we have been to able obtain highly purified HPIN and CaP DNA and thus identify the particular chromosomal changes associated with the two disease stages.

In this study, we have used a 2,400-element BAC microarray with a resolution of ~1 Mb to study CNAs in a set of 15 patient samples comprised of 7 HPIN cases and 8 CaP cases. For the 7 HPIN cases, 41 genomic alterations (20 gains, 21 losses) were identified, in contrast to the 90 genomic alterations (38 gains, 52 losses) seen for the 8 CaP cases. As with other cancers, CaP development and progression is likely to be the outcome of a series of stepwise genetic changes. The accumulation of CNAs, which occurs during this process, although likely due to increased genetic instability, is non-random and may indicate chromosomal regions important in tumourigenesis. It is suggested that failure in the fidelity of homologous recombination within the repetitive sequences, that comprise the kinetochore complex, could lead to recurrent loss of 8p and gain of 8q by rearrangement of chromosome 8-specific alphoid centromeric sequences. Thus, the high fidelity process of homologous recombination can be the major DNA repair pathway, which is indispensable for the maintenance of genetic stability.

Examination of our array results indicates that aberrations involving parts or all of 1p, 6q, 7p, 7q, 8p, 8q, 10q, 13q 16p and 16q are most common, which is concordant with previous metaphase [19,20] and array CGH [4,21] results. It is therefore likely that the alterations in copy number are part of a programmed cycle of events that promote tumour development and progression as well having an impact on disease-specific survival.

A comparison of the CNAs present in the HPIN and CaP samples identified a significant increase in copy number for 7q (P<0.01, chi square test) and a significantly increased frequency of loss for 10q (P<0.01, chi square test) and 13q (P<0.0001, chi square test). In genotype/phenotype correlations, gain of chromosome 7q [22] and loss of 13q [23] have been associated with advancing tumour stage and aggressiveness, which is in agreement with the results presented here. However, gain of 8q [24] and loss of 16q [25] have also been linked to tumour progression, but our data do not show any significant difference for these CNAs in our HPIN and CaP samples. This would suggest that 8q and 16q CNA's are likely to be early events in tumourigenesis. In addition, they may also identify HPIN and CaP samples that are likely to progress.

Apart from the commonly reported CNAs, additional alterations that have been less frequently reported in earlier CGH studies have also been identified. These include gains on 12q (HPIN) and loss of 4q and 10q (HPIN and CaP). Whether the identification of 12q and 4q regions will provide additional insight into CaP progression is not yet clear. Further analysis using a platform such as tissue microarrays, which permits the screening of different disease stages from large patient cohorts, will help better identify their frequency and also the potential use of genomic imbalance in diagnosis of CaP. The value of genomic analysis in CaP was recently demonstrated by the discovery of a high frequency of chromosomal translocations leading to rearrangement of the fusion oncoproteins ERG or ETV1 with TMPRSS2 [26].

A common problem when analysing archival tissue is the availability of a suitable quality and quantity of RNA to study expression changes. Though RNA amplification techniques [27] can generate a sufficient quantity, obtaining RNA of the required quality is often challenging. Previous reports have demonstrated a relationship between alterations in chromosomal copy number and alterations in gene expression [28,29]. The ability to distinguish these regions will point to genes, which may either directly (tumour suppressor gene loss or oncogene gain) or indirectly contribute to tumour development and progression. As a result, CGH can be used as a surrogate for gene identification. Candidate genes, which have previously been implicated in CaP have been highlighted in bold in Table 3. The roles of MYC [30-32], PSCA [33-35], and MDM2 [36-38] have all been well reported and alterations in gene dosage correlate well with their change in expression. However, other candidate genes have been less well studied. For example, EBAG9, whose increased expression in CaP is a negative prognostic indicator, has a potential role in progression by enabling cancer cells to evade the immune response [39]. NBS1, which has been identified as a founder mutation causing an increased susceptibility to prostate cancer [40], is involved in processing/repair of DNA double strand breaks and in cell cycle checkpoints, thus its deregulation will likely contribute to chromosomal instability. FKHR, which is a member of the FOXO forkhead transcription factor family, is thought to play a regulatory role in several cellular functions including cell proliferation and survival [41]. Loss of FKHR expression, as observed in CaP cell lines, is likely to abrogate this control leading to tumour cell growth. Though these genes have previously been implicated in CaP there are additional genes, including both oncogenes and tumour suppressor genes, which reside within all of the affected regions that may play an important role in the aetiology of the disease. Further analysis of these other candidates may identify their potential as molecular targets for diagnosis and treatment.

Several of the genes present within altered regions of the genome are associated with genetic pathways, indicating that these pathways are likely to be important in prostate tumourigenesis.

Conclusion

The combination of techniques used in this study has allowed for the identification of consistent regions of copy number change, ranging from specific cytobands to whole chromosomes, starting with as little as 5–10 ng of DNA. Through our use of LCM, we have also identified several lower frequency CNAs which may not have been detected when using bulk extracted DNA due to the heterogeneous nature of prostate tissue. In addition, several interesting candidate genes have been observed within the altered chromosomal regions. Although some have already been extensively studied others, which have not, may prove to be of clinical importance. Genetic screening strategies that combine FISH and immunohistochemistry for the detection of various combinations of chromosomal gains and/or losses and altered gene expression are likely to be of great use in diagnosis and prognosis prediction.

Methods

Tissue accrual

Fresh prostate tissue was obtained from radical prostatectomies performed at The University Health Network (UHN), Toronto, with the informed patient consent and approval of the institutional research ethics board. The tissue was embedded in OCT frozen section medium (Stephens Scientific, Riverdale NJ, USA) and stored at -80°C until required. For all samples, the presence of HPIN and CaP was verified by histologic assessment of frozen sections by a urological pathologist (A.E.).

Cohort selection

Patient samples for gaCGH analysis were selected based on two criteria; first, the presence of clearly identifiable regions of HPIN and/or CaP and second, the availability of good quality high molecular weight DNA. DNA quality was assessed by DNA extraction from a single tissue section and visualisation by agarose gel electrophoresis. A cohort consisting of 15 cases (7 HPIN and 8 CaP derived from different patients) were selected based on these criteria.

Laser capture microdissection

The selected fresh-frozen prostate tissue samples were cut onto microscope slides (8 μm thickness) and stained using the HistoGene LCM frozen section staining kit (Arcturus, USA). Foci of HPIN and CaP were identified from stained prostate tissue sections and a minimum of 1000 cells were removed from these regions by LCM using the Cell Robotics LaserScissors system (Cell Robotics Inc., USA).

DNA extraction

DNA was extracted using the QIAamp DNA Micro Kit (Qiagen, USA) and the DNA concentration was determined using the PicoGreen dsDNA Quantitation kit (Molecular Probes Inc., USA). Both procedures were performed following manufacturers instructions.

Strand displacement amplification

DNA was amplified using the GenomiPhi Amplification Kit (Amersham Biosciences, USA) according to manufacturer's instructions. Briefly, amplification was carried out in two individual steps. The step 1 reaction mixture contained 5–10 ng (1000 to 2000 cell equivalents) of DNA (from diluted reference or test DNA) in 1 μl of sterile water and 9 μl of Sample Buffer. This mixture was heated at 95°C for 3 minutes and then chilled on ice. The step 2 reaction (amplification) mixture contained 9 μl of Reaction Buffer, 1 μl of Enzyme Mix and the 10 μl from Step 1. The amplification reaction was incubated at 30°C for 16–18 hours. The enzyme was inactivated by heating at 65°C for 10 minutes, followed by cooling to 4°C. This method of WGA consistently produced in excess of 5 μg of DNA from a starting concentration of 5–10 ng.

Assessment of DNA quality and strand displacement amplification

Five microlitres of each amplification reaction was electrophoresed through a 1% agarose gel and stained with ethidium bromide in order to assess product yield and product length. All amplification products were purified by phenol-chloroform extraction and DNA concentration and purity were determined by measuring absorbance at A260 and A280.

Spectral genomic BAC arrays

The genomic DNA arrays used in these experiments were obtained from Spectral Genomics Inc. and consist of approximately 2400 unique BAC and PAC clones, which provide an average genomic resolution of 1 Mb. CGH experiments were performed using the amplified patient DNA as the "test" and amplified female placental DNA as the "normal". In addition, control experiments were performed, corresponding to DNA before and after amplification (normal non-neoplastic prostate epithelial DNA versus normal non-neoplastic prostate epithelial DNA, normal non-neoplastic prostate epithelial DNA versus normal female placental DNA and DNA from a cancer cell line versus normal control DNA).

Reference DNA and test DNA were first digested overnight at 37°C using 10 units of RsaI (Invitrogen) in a 10 μl reaction. The digested DNA was labeled using the protocol optimized by Spectral Genomics, with separate labeling reactions for Cy3 and Cy5 being set up for both reference and test DNA. Briefly, labeling reaction were set up containing 2 μg of DNA, 20 μl of 2.5× random primer/reaction buffer mix (Invitrogen) and sterile water up to a final volume of 45 μl. The reaction mix was boiled for 5 minutes prior to cooling on ice and the addition of 2.5 μl of Spectral labeling buffer (Spectral Genomics, Houston, U.S.A), 1.5 μl of either Cy3-dCTP (1mM, Applied Biosystems, Foster City, U.S.A) or Cy5-dCTP (1 mM, Applied Biosystems) and 1 μl of Klenow Fragment (BioPrime labeling kit, Invitrogen). The reaction was mixed gently and then incubated for 2 hours at 37°C. Following incubation the reaction was stopped by the addition of 5 μl 0.5 M EDTA (pH8) and heating at 72°C for 10 minutes.

The Cy3 labeled test DNA was combined with the Cy5 labeled normal reference DNA and vice versa. Each combined probe was mixed with 45 μl of Spectral Hybridisation Buffer (Spectral Genomics), 11.3 μl of 5 M NaCl and 110 μl of room temperature isopropanol. The samples were incubated in the dark at room temperature for 10–15 minutes and centrifuged at 16,000 x g for 10 minutes and the supernatant discarded. The pellets were then washed with 500 μl of 70% ethanol. The supernatant was carefully removed and the pellets air-dried at room temperature in the dark. For hybridisation, the pellets were first resuspended in 10 μl of sterile water prior to being mixed with 30 μl of Spectral Hybridisation Buffer II (Spectral Genomics) by pipetting. The reconstituted probes were then incubated at 72°C for 10 minutes, placed on ice for 5 minutes and then incubate for 30 minutes at 37°C. The probes were hybridised to BAC arrays, covered with a 22 × 60 mm coverslip and incubated for 12–16 hours at 37°C in a humidified chamber.

The wash buffers, with the exception of Wash I, were pre-warmed to 50°C. The slides were gently dipped into and out of Wash I (2× SSC, 0.5% SDS) until the coverslip detached from the slide. The slides were then washed once in Wash II (2 × SSC, 50% deionized Formamide, pH 7.5) for 20 minutes, followed by successive washes in Wash III (2× SSC, 0.1% NP-40, pH 7.5) for 20 minutes and Wash IV (0.2× SSC, pH 7.5) for 10 minutes. All washes were performed at 50°C, with the exception of Wash I. The slides were briefly submerged in distilled deionized water for 5–10 seconds and centrifuged for 5 minutes at 750 rpm to dry.

Data collection and analysis

The slides were scanned using an Axon GenePix 4000 A confocal scanner, each fluorescence signal was collected separately and quantified with the GenePix Pro 3.0 software (Axon Instruments, U.S.A). The data was normalised and analysed using Normalise Suite v2.4 [42], all regions of loss or gain were determined as those that were 2 standard deviations above the mean baseline for each separate sample.

Authors' contributions

SH and MY designed and carried out the study, in addition to preparing the final manuscript for publication. BB helped with the analysis of the data. AE helped in the selection of cases and worked closely with JS in the study design and coordination.

Additional File 4. H&E sections show an example of LCM. Dissection of HPIN. Top picture represents the tissue before dissection, middle picture is after dissection, and bottom picture is the cap tissue.

Format: TIFF Size: 703KB Download fileOpen Data

Acknowledgements

We would like to thank the researchers at Princess Margaret Hospital tissue procurement facility for their help and technical assistance in securing good quality samples for this study. We also like to thank Roger Lasken for his advice. This work was supported by the National Cancer Institute of Canada with funds from the Canadian Cancer Society and by awards to S.H. and M.Y. from the United States Army Medical Research and Materiel Command [DAMD17-03-1-0154].

References

  1. van der Kwast TH, Labrie F, Tetu B: Persistence of high-grade prostatic intra-epithelial neoplasia under combined androgen blockade therapy.

    Hum Pathol 1999, 30:1503-1507. PubMed Abstract | Publisher Full Text OpenURL

  2. Al-Maghrabi J, Vorobyova L, Toi A, Chapman W, Zielenska M, Squire JA: Identification of numerical chromosomal changes detected by interphase fluorescence in situ hybridization in high-grade prostate intraepithelial neoplasia as a predictor of carcinoma.

    Arch Pathol Lab Med 2002, 126:165-169. PubMed Abstract | Publisher Full Text OpenURL

  3. Kallioniemi A, Kallioniemi OP, Sudar D, Rutovitz D, Gray JW, Waldman F, Pinkel D: Comparative genomic hybridization for molecular cytogenetic analysis of solid tumors.

    Science 1992, 258:818-821. PubMed Abstract OpenURL

  4. van Dekken H, Paris PL, Albertson DG, Alers JC, Andaya A, Kowbel D, van der Kwast TH, Pinkel D, Schroder FH, Vissers KJ, Wildhagen MF, Collins C: Evaluation of genetic patterns in different tumor areas of intermediate-grade prostatic adenocarcinomas by high-resolution genomic array analysis.

    Genes Chromosomes Cancer 2004, 39:249-256. PubMed Abstract | Publisher Full Text OpenURL

  5. Wolf M, Mousses S, Hautaniemi S, Karhu R, Huusko P, Allinen M, Elkahloun A, Monni O, Chen Y, Kallioniemi A, Kallioniemi OP: High-resolution analysis of gene copy number alterations in human prostate cancer using CGH on cDNA microarrays: impact of copy number on gene expression.

    Neoplasia 2004, 6:240-247. PubMed Abstract | Publisher Full Text OpenURL

  6. Clark J, Edwards S, Feber A, Flohr P, John M, Giddings I, Crossland S, Stratton MR, Wooster R, Campbell C, Cooper CS: Genome-wide screening for complete genetic loss in prostate cancer by comparative hybridization onto cDNA microarrays.

    Oncogene 2003, 22:1247-1252. PubMed Abstract | Publisher Full Text OpenURL

  7. Hughes S, Arneson N, Done S, Squire J: The use of whole genome amplification in the study of human disease.

    Prog Biophys Mol Biol 2005, 88:173-189. PubMed Abstract | Publisher Full Text OpenURL

  8. Dean FB, Hosono S, Fang L, Wu X, Faruqi AF, Bray-Ward P, Sun Z, Zong Q, Du Y, Du J, Driscoll M, Song W, Kingsmore SF, Egholm M, Lasken RS: Comprehensive human genome amplification using multiple displacement amplification.

    Proc Natl Acad Sci U S A 2002, 99:5261-5266. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  9. Dean FB, Nelson JR, Giesler TL, Lasken RS: Rapid amplification of plasmid and phage DNA using Phi 29 DNA polymerase and multiply-primed rolling circle amplification.

    Genome Res 2001, 11:1095-1099. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  10. Lage JM, Leamon JH, Pejovic T, Hamann S, Lacey M, Dillon D, Segraves R, Vossbrinck B, Gonzalez A, Pinkel D, Albertson DG, Costa J, Lizardi PM: Whole genome analysis of genetic alterations in small DNA samples using hyperbranched strand displacement amplification and array-CGH.

    Genome Res 2003, 13:294-307. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  11. Lasken R: Multiple displacement amplification of genomic DNA. In Whole Genome Amplification: Methods Express. Edited by Hughes SLR. Oxford, Scion Publishing Ltd.; 2005:99-118. OpenURL

  12. Barker DL, Hansen MS, Faruqi AF, Giannola D, Irsula OR, Lasken RS, Latterich M, Makarov V, Oliphant A, Pinter JH, Shen R, Sleptsova I, Ziehler W, Lai E: Two methods of whole-genome amplification enable accurate genotyping across a 2320-SNP linkage panel.

    Genome Res 2004, 14:901-907. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  13. Paez JG, Lin M, Beroukhim R, Lee JC, Zhao X, Richter DJ, Gabriel S, Herman P, Sasaki H, Altshuler D, Li C, Meyerson M, Sellers WR: Genome coverage and sequence fidelity of phi29 polymerase-based multiple strand displacement whole genome amplification.

    Nucleic Acids Res 2004, 32:e71. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  14. Tzvetkov MV, Becker C, Kulle B, Nurnberg P, Brockmoller J, Wojnowski L: Genome-wide single-nucleotide polymorphism arrays demonstrate high fidelity of multiple displacement-based whole-genome amplification.

    Electrophoresis 2005, 26:710-715. PubMed Abstract | Publisher Full Text OpenURL

  15. Hughes S, Lim G, Beheshti B, Bayani J, Marrano P, Huang A, Squire JA: Use of whole genome amplification and comparative genomic hybridisation to detect chromosomal copy number alterations in cell line material and tumour tissue.

    Cytogenet Genome Res 2004, 105:18-24. PubMed Abstract | Publisher Full Text OpenURL

  16. Hughes S, Damato BE, Giddings I, Hiscott PS, Humphreys J, Houlston RS: Microarray comparative genomic hybridisation analysis of intraocular uveal melanomas identifies distinctive imbalances associated with loss of chromosome 3.

    Br J Cancer 2005, 93:1191-1196. PubMed Abstract | Publisher Full Text OpenURL

  17. Cleton-Jansen AM, Buerger H, Haar N, Philippo K, van de Vijver MJ, Boecker W, Smit VT, Cornelisse CJ: Different mechanisms of chromosome 16 loss of heterozygosity in well- versus poorly differentiated ductal breast cancer.

    Genes Chromosomes Cancer 2004, 41:109-116. PubMed Abstract | Publisher Full Text OpenURL

  18. Kimura Y, Noguchi T, Kawahara K, Kashima K, Daa T, Yokoyama S: Genetic alterations in 102 primary gastric cancers by comparative genomic hybridization: gain of 20q and loss of 18q are associated with tumor progression.

    Mod Pathol 2004, 17:1328-1337. PubMed Abstract | Publisher Full Text OpenURL

  19. Joos S, Bergerheim US, Pan Y, Matsuyama H, Bentz M, du Manoir S, Lichter P: Mapping of chromosomal gains and losses in prostate cancer by comparative genomic hybridization.

    Genes Chromosomes Cancer 1995, 14:267-276. PubMed Abstract OpenURL

  20. Visakorpi T, Kallioniemi AH, Syvanen AC, Hyytinen ER, Karhu R, Tammela T, Isola JJ, Kallioniemi OP: Genetic changes in primary and recurrent prostate cancer by comparative genomic hybridization.

    Cancer Res 1995, 55:342-347. PubMed Abstract OpenURL

  21. Yano S, Matsuyama H, Matsuda K, Matsumoto H, Yoshihiro S, Naito K: Accuracy of an array comparative genomic hybridization (CGH) technique in detecting DNA copy number aberrations: comparison with conventional CGH and loss of heterozygosity analysis in prostate cancer.

    Cancer Genet Cytogenet 2004, 150:122-127. PubMed Abstract | Publisher Full Text OpenURL

  22. Paiss T, Worner S, Kurtz F, Haeussler J, Hautmann RE, Gschwend JE, Herkommer K, Vogel W: Linkage of aggressive prostate cancer to chromosome 7q31-33 in German prostate cancer families.

    Eur J Hum Genet 2003, 11:17-22. PubMed Abstract | Publisher Full Text OpenURL

  23. Dong JT, Boyd JC, Frierson HFJ: Loss of heterozygosity at 13q14 and 13q21 in high grade, high stage prostate cancer.

    Prostate 2001, 49:166-171. PubMed Abstract | Publisher Full Text OpenURL

  24. Alers JC, Rochat J, Krijtenburg PJ, Hop WC, Kranse R, Rosenberg C, Tanke HJ, Schroder FH, van Dekken H: Identification of genetic markers for prostatic cancer progression.

    Lab Invest 2000, 80:931-942. PubMed Abstract OpenURL

  25. Matsuyama H, Pan Y, Yoshihiro S, Kudren D, Naito K, Bergerheim US, Ekman P: Clinical significance of chromosome 8p, 10q, and 16q deletions in prostate cancer.

    Prostate 2003, 54:103-111. PubMed Abstract | Publisher Full Text OpenURL

  26. Tomlins SA, Rhodes DR, Perner S, Dhanasekaran SM, Mehra R, Sun XW, Varambally S, Cao X, Tchinda J, Kuefer R, Lee C, Montie JE, Shah RB, Pienta KJ, Rubin MA, Chinnaiyan AM: Recurrent fusion of TMPRSS2 and ETS transcription factor genes in prostate cancer.

    Science 2005, 310:644-648. PubMed Abstract | Publisher Full Text OpenURL

  27. Zhao H, Hastie T, Whitfield ML, Borresen-Dale AL, Jeffrey SS: Optimization and evaluation of T7 based RNA linear amplification protocols for cDNA microarray analysis.

    BMC Genomics 2002, 3:31. PubMed Abstract | BioMed Central Full Text | PubMed Central Full Text OpenURL

  28. Collins C, Rommens JM, Kowbel D, Godfrey T, Tanner M, Hwang SI, Polikoff D, Nonet G, Cochran J, Myambo K, Jay KE, Froula J, Cloutier T, Kuo WL, Yaswen P, Dairkee S, Giovanola J, Hutchinson GB, Isola J, Kallioniemi OP, Palazzolo M, Martin C, Ericsson C, Pinkel D, Albertson D, Li WB, Gray JW: Positional cloning of ZNF217 and NABC1: genes amplified at 20q13.2 and overexpressed in breast carcinoma.

    Proc Natl Acad Sci U S A 1998, 95:8703-8708. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  29. Shayesteh L, Lu Y, Kuo WL, Baldocchi R, Godfrey T, Collins C, Pinkel D, Powell B, Mills GB, Gray JW: PIK3CA is implicated as an oncogene in ovarian cancer.

    Nat Genet 1999, 21:99-102. PubMed Abstract | Publisher Full Text OpenURL

  30. Savinainen KJ, Linja MJ, Saramaki OR, Tammela TL, Chang GT, Brinkmann AO, Visakorpi T: Expression and copy number analysis of TRPS1, EIF3S3 and MYC genes in breast and prostate cancer.

    Br J Cancer 2004, 90:1041-1046. PubMed Abstract | Publisher Full Text OpenURL

  31. Cassinelli G, Supino R, Zuco V, Lanzi C, Scovassi AI, Semple SC, Zunino F: Role of c-myc protein in hormone refractory prostate carcinoma: cellular response to paclitaxel.

    Biochem Pharmacol 2004, 68:923-931. PubMed Abstract | Publisher Full Text OpenURL

  32. Bernard D, Pourtier-Manzanedo A, Gil J, Beach DH: Myc confers androgen-independent prostate cancer cell growth.

    J Clin Invest 2003, 112:1724-1731. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  33. Zhigang Z, Wenlv S: Prostate stem cell antigen (PSCA) expression in human prostate cancer tissues and its potential role in prostate carcinogenesis and progression of prostate cancer.

    World J Surg Oncol 2004, 2:13. PubMed Abstract | BioMed Central Full Text | PubMed Central Full Text OpenURL

  34. Fuessel S, Sickert D, Meye A, Klenk U, Schmidt U, Schmitz M, Rost AK, Weigle B, Kiessling A, Wirth MP: Multiple tumor marker analyses (PSA, hK2, PSCA, trp-p8) in primary prostate cancers using quantitative RT-PCR.

    Int J Oncol 2003, 23:221-228. PubMed Abstract OpenURL

  35. Reiter RE, Gu Z, Watabe T, Thomas G, Szigeti K, Davis E, Wahl M, Nisitani S, Yamashiro J, Le Beau MM, Loda M, Witte ON: Prostate stem cell antigen: a cell surface marker overexpressed in prostate cancer.

    Proc Natl Acad Sci U S A 1998, 95:1735-1740. PubMed Abstract | Publisher Full Text | PubMed Central Full Text OpenURL

  36. Chang CJ, Freeman DJ, Wu H: PTEN regulates Mdm2 expression through the P1 promoter.

    J Biol Chem 2004, 279:29841-29848. PubMed Abstract | Publisher Full Text OpenURL

  37. Mu Z, Hachem P, Agrawal S, Pollack A: Antisense MDM2 oligonucleotides restore the apoptotic response of prostate cancer cells to androgen deprivation.

    Prostate 2004, 60:187-196. PubMed Abstract | Publisher Full Text OpenURL

  38. Leite KR, Franco MF, Srougi M, Nesrallah LJ, Nesrallah A, Bevilacqua RG, Darini E, Carvalho CM, Meirelles MI, Santana I, Camara-Lopes LH: Abnormal expression of MDM2 in prostate carcinoma.

    Mod Pathol 2001, 14:428-436. PubMed Abstract | Publisher Full Text OpenURL

  39. Takahashi S, Urano T, Tsuchiya F, Fujimura T, Kitamura T, Ouchi Y, Muramatsu M, Inoue S: EBAG9/RCAS1 expression and its prognostic significance in prostatic cancer.

    Int J Cancer 2003, 106:310-315. PubMed Abstract | Publisher Full Text OpenURL

  40. Cybulski C, Gorski B, Debniak T, Gliniewicz B, Mierzejewski M, Masojc B, Jakubowska A, Matyjasik J, Zlowocka E, Sikorski A, Narod SA, Lubinski J: NBS1 is a prostate cancer susceptibility gene.

    Cancer Res 2004, 64:1215-1219. PubMed Abstract | Publisher Full Text OpenURL

  41. Huang H, Muddiman DC, Tindall DJ: Androgens negatively regulate forkhead transcription factor FKHR (FOXO1) through a proteolytic mechanism in prostate cancer cells.

    J Biol Chem 2004, 279:13866-13877. PubMed Abstract | Publisher Full Text OpenURL

  42. Beheshti B, Braude I, Marrano P, Thorner P, Zielenska M, Squire JA: Chromosomal localization of DNA amplifications in neuroblastoma tumors using cDNA microarray comparative genomic hybridization.

    Neoplasia 2003, 5:53-62. PubMed Abstract | Publisher Full Text OpenURL