Jessica C.M. Pole

A 1 Mb minimal amplicon at 8p11–12 in breast cancer identifies new candidate oncogenes

Amplification of 8p11–12 is a well-known alteration in human breast cancers but the driving oncogene has not been identified. We have developed a high-resolution comparative genomic hybridization array covering 8p11–12 and analysed 33 primary breast tumors, 20 primary ovarian tumors and 27 breast cancer cell lines. Expression analysis of the genes in the region was carried out by using real-time quantitative PCR and/or oligo-microarray profiling. In all, 24% (8/33) of the breast tumors, 5% (1/20) of the ovary tumors and 15% (4/27) of the cell lines showed 8p11–12 amplification. We identified a 1 Mb segment of common amplification that excludes previously proposed candidate genes. Some of the amplified genes did not show overexpression, whereas for others, overexpression was not specifically attributable to amplification. The genes FLJ14299, C8orf2, BRF2 and RAB11FIP, map within the 8p11–12 minimal amplicon, two have a putative function consistent with an oncogenic role, these four genes showed a strong correlation between amplification and overexpression and are therefore the best candidate driver oncogenes at 8p12.

High-resolution analysis of chromosome rearrangements on 8p in breast, colon and pancreatic cancer reveals a complex pattern of loss, gain and translocation.

The short arm of chromosome 8, 8p, is often rearranged in carcinomas, typically showing distal loss by unbalanced translocation. We analysed 8p rearrangements in 48 breast, pancreatic and colon cancer cell lines by fluorescence in situ hybridization (FISH) and array comparative genomic hybridization, with a tiling path of 0.2 Mb resolution over 8p12 and 1 Mb resolution over chromosome 8. Selected breast lines (MDA-MB-134, MDA-MB-175, MDA-MB-361, T-47D and ZR-75-1) were analysed further. Most cell lines showed loss of 8p distal to a break that was between 31 Mb (5′ to NRG1) and the centromere, but the translocations were accompanied by variable amplifications, deletions and inversions proximal to this break. The 8p12 translocation in T-47D was flanked by an inversion of 4 Mb, with a 100 kb deletion at the proximal end. The dicentric t(8;11) in ZR-75-1 carries multiple rearrangements including interstitial deletions, a triplicated translocation junction between NRG1 and a fragment of 11q (unconnected to CCND1), and two separate amplifications, of FGFR1 and CCND1 . We conclude that if there is a tumour suppressor gene on 8p it may be near 31 Mb, for example WRN; but the complexity of 8p rearrangements suggests that they target various genes proximal to 31 Mb including NRG1 and the amplicon centred around ZNF703/FLJ14299.

A recurrent chromosome breakpoint in breast cancer at the NRG1/neuregulin 1/heregulin gene

Most studies of genomic rearrangements in common cancers have focused on regional gains and losses, but some rearrangements may break within specific genes. We previously reported that five breast cancer cell lines have chromosome translocations that break in the NRG1 gene and that could cause abnormal NRG1 expression. NRG1 encodes the Neuregulins 1 (formerly the Heregulins), ligands for members of the ErbB/epidermal growth factor-receptor family, which includes ErbB2/HER2. We have now screened for breaks at NRG1 in paraffin sections of breast tumors. Tissue microarrays were screened by fluorescence in situ hybridization, with hybridization probes proximal and distal to the expected breakpoints. This screen detects breaks but does not distinguish between translocation or deletion breakpoints. The screen was validated with array-comparative genomic hybridization on a custom 8p12 high-density genomic array to detect a lower copy number of the sequences that were lost distal to the breaks. We also precisely mapped the breaks in five tumors with different hybridization probes. Breaks in NRG1 were detected in 6% (19 of 323) of breast cancers and in some lung and ovarian cancers. In an unselected series of 213 cases with follow-up, breast cancers where the break was detected tended to be high-grade (65% grade III compared with 28% of negative cases). They were, like breast tumors in general, mainly ErbB2 low (11 of 13 were low) and estrogen receptor positive (11 of 13 positive).

The NRG1 gene is frequently silenced by methylation in breast cancers and is a strong candidate for the 8p tumour suppressor gene.

Neuregulin-1 (NRG1) is both a candidate oncogene and a candidate tumour suppressor gene. It not only encodes the heregulins and other mitogenic ligands for the ERBB family, but also causes apoptosis in NRG1-expressing cells. We found that most breast cancer cell lines had reduced or undetectable expression of NRG1. This included cell lines that had translocation breaks in the gene. Similarly, expression in cancers was generally comparable to or less than that in various normal breast samples. Many non-expressing cell lines had extensive methylation of the CpG island at the principal transcription start site at exon 2 of NRG1. Expression was reactivated by demethylation. Many tumours also showed methylation, whereas normal mammary epithelial fragments had none. Lower NRG1 expression correlated with higher methylation. Small interfering RNA (siRNA)-mediated depletion of NRG1 increased net proliferation in a normal breast cell line and a breast cancer cell line that expressed NRG1. The short arm of chromosome 8 is frequently lost in epithelial cancers, and NRG1 is the most centromeric gene that is always affected. NRG1 may therefore be the major tumour suppressor gene postulated to be on 8p: it is in the correct location, is antiproliferative and is silenced in many breast cancers.

The role of tandem duplicator phenotype in tumour evolution in high-grade serous ovarian cancer

High‐grade serous ovarian carcinoma (HGSOC) is characterized by genomic instability, ubiquitous TP53 loss, and frequent development of platinum resistance. Loss of homologous recombination (HR) is a mutator phenotype present in 50% of HGSOCs and confers hypersensitivity to platinum treatment. We asked which other mutator phenotypes are present in HGSOC and how they drive the emergence of platinum resistance. We performed whole‐genome paired‐end sequencing on a model of two HGSOC cases, each consisting of a pair of cell lines established before and after clinical resistance emerged, to describe their structural variants (SVs) and to infer their ancestral genomes as the SVs present within each pair. The first case (PEO1/PEO4), with HR deficiency, acquired translocations and small deletions through its early evolution, but a revertant BRCA2 mutation restoring HR function in the resistant lineage re‐stabilized its genome and reduced platinum sensitivity. The second case (PEO14/PEO23) had 216 tandem duplications and did not show evidence of HR or mismatch repair deficiency. By comparing the cell lines to the tissues from which they originated, we showed that the tandem duplicator mutator phenotype arose early in progression in vivo and persisted throughout evolution in vivo and in vitro, which may have enabled continual evolution. From the analysis of SNP array data from 454 HGSOC cases in The Cancer Genome Atlas series, we estimate that 12.8% of cases show patterns of aberrations similar to the tandem duplicator, and this phenotype is mutually exclusive with BRCA1/2 carrier mutations. Copyright

High-resolution array CGH clarifies events occurring on 8p in carcinogenesis

BackgroundRearrangement of the short arm of chromosome 8 (8p) is very common in epithelial cancers such as breast cancer. Usually there is an unbalanced translocation breakpoint in 8p12 (29.7 Mb – 38.5 Mb) with loss of distal 8p, sometimes with proximal amplification of 8p11-12. Rearrangements in 8p11-12 have been investigated using high-resolution array CGH, but the first 30 Mb of 8p are less well characterised, although this region contains several proposed tumour suppressor genes.MethodsWe analysed the whole of 8p by array CGH at tiling-path BAC resolution in 32 breast and six pancreatic cancer cell lines. Regions of recurrent rearrangement distal to 8p12 were further characterised, using regional fosmid arrays. FISH, and quantitative RT-PCR on over 60 breast tumours validated the existence of similar events in primary material.ResultsWe confirmed that 8p is usually lost up to at least 30 Mb, but a few lines showed focal loss or copy number steps within this region. Three regions showed rearrangements common to at least two cases: two regions of recurrent loss and one region of amplification. Loss within 8p23.3 (0 Mb – 2.2 Mb) was found in six cell lines. Of the genes always affected, ARHGEF10 showed a point mutation of the remaining normal copies in the DU4475 cell line. Deletions within 12.7 Mb – 19.1 Mb in 8p22, in two cases, affected TUSC3. A novel amplicon was found within 8p21.3 (19.1 Mb – 23.4 Mb) in two lines and one of 98 tumours.ConclusionThe pattern of rearrangements seen on 8p may be a consequence of the high density of potential targets on this chromosome arm, and ARHGEF10 may be a new candidate tumour suppressor gene.

High-resolution analysis of genomic alteration on chromosome arm 8p in urothelial carcinoma.

Loss of chromosome arm 8p, sometimes in combination with amplification of proximal 8p, is found in urothelial carcinoma (UC) and other epithelial cancers and is associated with more advanced tumor stage. We carried out array comparative genomic hybridization on 174 UC and 33 UC cell lines to examine breakpoints and copy number. This was followed by a detailed analysis of the cell lines using fluorescence in situ hybridization (FISH) and, in some cases, M‐FISH, to refine breakpoints and determine translocation partners, heterozygosity analysis, and analysis of expression of selected genes. We showed an overall pattern of 8p loss with reduced heterozygosity and reduced gene expression. Amplification was seen in some samples and shown in the cell line JMSU1 to correlate with overexpression of ZNF703, ERLIN2, PROSC, GPR124, and BRF2. Apart from the centromere, no single breakpoint was overrepresented, and we postulate that frequent complex changes without consistent breakpoints reflect the need for alterations of combinations of genes. The region around 2 Mb, which was homozygously deleted in one cell line and includes the gene ARHGEF10 and the micro‐RNA hsa‐mir‐596, is one candidate tumor suppressor gene region.

PMC42, a breast progenitor cancer cell line, has normal-like mRNA and microRNA transcriptomes

IntroductionThe use of cultured cell lines as model systems for normal tissue is limited by the molecular alterations accompanying the immortalisation process, including changes in the mRNA and microRNA (miRNA) repertoire. Therefore, identification of cell lines with normal-like expression profiles is of paramount importance in studies of normal gene regulation.MethodsThe mRNA and miRNA expression profiles of several breast cell lines of cancerous or normal origin were measured using printed slide arrays, Luminex bead arrays, and real-time reverse transcription-polymerase chain reaction.ResultsWe demonstrate that the mRNA expression profiles of two breast cell lines are similar to that of normal breast tissue: HB4a, immortalised normal breast epithelium, and PMC42, a breast cancer cell line that retains progenitor pluripotency allowing in-culture differentiation to both secretory and myoepithelial fates. In contrast, only PMC42 exhibits a normal-like miRNA expression profile. We identified a group of miRNAs that are highly expressed in normal breast tissue and PMC42 but are lost in all other cancerous and normal-origin breast cell lines and observed a similar loss in immortalised lymphoblastoid cell lines compared with healthy uncultured B cells. Moreover, like tumour suppressor genes, these miRNAs are lost in a variety of tumours. We show that the mechanism leading to the loss of these miRNAs in breast cancer cell lines has genomic, transcriptional, and post-transcriptional components.ConclusionWe propose that, despite its neoplastic origin, PMC42 is an excellent molecular model for normal breast epithelium, providing a unique tool to study breast differentiation and the function of key miRNAs that are typically lost in cancer.

Co-amplification of 8p12 and 11q13 in breast cancers is not the result of a single genomic event.

Epithelial cancers frequently have multiple amplifications, and particular amplicons tend to occur together. These co‐amplifications have been suggested to result from amplification of pre‐existing junctions between two chromosomes, that is, translocation junctions. We investigated this hypothesis for two amplifications frequent in breast cancer, at 8p12 and 11q13, which had been reported to be associated in Southern blot studies. We confirmed that both genomic amplification and expression of genes was correlated between the frequently‐amplified regions of 8p and 11q, in array CGH and microarray expression data, supporting the importance of co‐amplification. We examined by FISH the physical structure of co‐amplifications that we had identified by array CGH, in five breast cancer cell lines (HCC1500, MDA‐MB‐134, MDA‐MB‐175, SUM44, and ZR‐75‐1), four breast tumors, and a pancreatic cancer cell line (SUIT2). We found a variety of arrangements: amplification of translocation junctions; entirely independent amplification of the two regions on separate chromosomes; and separate amplification of 8p and 11q sequences in distinct sites on the same rearranged chromosome. In this last arrangement, interphase nuclei often showed intermingling of FISH signals from 8p12 and 11q13, giving a false impression that the sequences were interdigitated. We conclude that co‐amplification of the main 8p and 11q amplicons in breast tumors is not usually the result of a preceding translocation event but most likely reflects selection of clones that have amplified both loci. This article contains supplementary material available at http://www.interscience.wiley.com/jpages/1045‐2257/suppmat.

Genomic analysis of the 8p11-12 amplicon in familial breast cancer.

Amplification of 8p11‐12 has been recurrently reported in sporadic breast cancer. These studies define a complex molecular structure with a set of minimal amplified regions, and different putative oncogenes that show a strong correlation between amplification and over‐expression such as ZNF703/FLJ14299, SPFH2/C8orf2, BRF2 and RAB11FIP. However, none of these studies were carried out on familial breast malignancies. We have studied the incidence, molecular features and clinical value of this amplification in familial breast tumors associated with BRCA1, BRCA2 and non‐BRCA1/2 gene mutations. We detected 9 out of 80 familial tumors with this amplicon by chromosomal comparative genomic hybridization. Next, we used a high‐resolution comparative genomic hybridization array covering the 8p11‐12 region to characterize this chromosomal region. This approach allowed us to define 2 cores of common amplification that largely overlap with those reported in sporadic tumors. Our findings confirm the molecular complexity of this chromosomal region and indicate that this genomic event is a common alteration in breast cancer, present not only in sporadic but also in familial tumors. Finally, we found correlation between the 8p11‐12 amplification and proliferation (Ki‐67) and cyclin E expression, which further proves in familial tumors the poor prognosis association previously reported in sporadic breast cancer.