High-resolution analysis of 16q22.1 in breast carcinoma using DNA amplifiable probes (multiplex amplifiable probe hybridization technique) and immunohistochemistry Emad A. Rakha 1 , John A.L. Armour 2 , Sarah E. Pinder 3 , Claire E. Paish 1 and Ian O. Ellis 1 * 1 The Breast Unit, Department of Histopathology, Nottingham City Hospital, University of Nottingham, United Kingdom 2 Institute of Genetics, University of Nottingham, Queen’s Medical Centre, Nottingham, United Kingdom 3 Histopathology Department, Addenbrooke’s NHS Trust, Cambridge, United Kingdom Loss of the chromosomal material at 16q22.1 is one of the most frequent genetic aberrations found in both lobular and low-grade nonlobular invasive carcinoma of the breast, indicating the pres- ence of a tumour suppressor gene (TSG) at this region in these tumours. However, the TSG (s) at the 16q22.1 in the more fre- quent nonlobular carcinomas is still unknown. Multiplex Ampli- fiable Probe Hybridisation (MAPH) is a simple, accurate and a high-resolution technique that provides an alternative approach to DNA copy-number measurement. The aim of our study was to examine the most likely candidate genes at 16q22.1 using MAPH assay combined with protein expression analysis by immunohis- tochemistry. We identified deletion at 16q22.1 that involves some or all of these genes. We also noticed that the smallest region of deletion at 16q22.1 could be delineated to a 3 Mb region centro- meric to the P-cadherin gene. Apart from the correlation between E-cadherin protein expression and its gene copy number, no cor- relation was detected between the expression of E2F-4, CTCF, TRF2 or P-cadherin with their gene’s copy number. In the malig- nant tissues, no significant loss or decrease of protein expression of any gene other than E-cadherin was seen in association with any specific tumour type. No expression of VE-cadherin or Ksp-cad- herin was detected in the normal and/or malignant tissues of the breast in these cases. However, there was a correlation between increased nuclear expression of E2F-4 and tumours with higher histological grade (p  0.04) and positive lymph node disease (p  0.02), suggesting that it may have an oncogenic rather than a tumour suppressor role. The malignant breast tissues also showed abnormal cytoplasmic cellular localisation of CTCF, compared to its expression in the normal parenchymal cells. In conclusion, we have demonstrated that MAPH is a potential technique for assess- ment of genomic imbalances in malignant tissues. Although our results support E-cadherin as the TSG in invasive lobular carci- noma, they argue against the candidacy of E2F-4, CTCF, TRF2, P-cadherin, Ksp-cadherin and VE-cadherin as TSGs in breast cancer. © 2004 Wiley-Liss, Inc. Key words: breast cancer; 16q22.1; candidate genes; MAPH, immu- nohistochemistry Breast cancer is the most common cancer and second leading cause of cancer deaths among women in Western countries 1 and is recognised to be both genetically and clinically heterogeneous. Molecular analysis of breast cancer indicates that chromosome 16 is one of the most frequently altered chromosomes, with loss of heterozygosity (LOH) at 16q reported to occur in about half of low-grade ductal carcinomas and slightly more frequently in lob- ular carcinomas. 2–4 Physical loss of chromosomal material at 16q has also been detected by comparative genomic hybridisation (CGH). 5,6 The most frequent deletion hot spot at 16q in breast cancer has been demonstrated at 16q22.1, indicating the presence of a tumour suppressor gene (TSG) at this locus. 3,7 In the quest for the target TSG, the E-cadherin gene has been shown to be one TSG by identification of mutations in the remaining allele and loss of its protein expression but this is only commonly seen in the lobular type carcinoma of the breast. 8,9 These data indicate that another TSG(s) at 16q22.1, inactivated in the more frequent non- lobular type carcinoma, remains to be identified. However, deletions at the chromosome band 16q22.1 can be larger than 5 megabases and affect many genes. 10 Previous allelo- type of breast cancer did not further minimise the region of LOH to a size that makes functional studies of all genes located in its vicinity feasible. In addition, previous assessment of genomic copy number in cancer is performed either at a relatively low resolution by comparative genomic hybridisation (CGH) (5–10 Mb) 11,12 or at higher resolution by laborious individual locus specific techniques that can analyse only a few loci simultaneously. 13 Array-based CGH is a sensitive and a high-throughput technique but is tech- nically challenging, requires expensive equipment and uses 1 sample at a time. 14,15 Multiplex Amplifiable Probe Hybridisation (MAPH) provides an alternative DNA-based approach for detecting and quantifying genomic copy number alterations in tumours. 16,17 MAPH is a simple, semiquantitative, high-resolution technique that can be used to measure the copy number of many different regions of interest (represented by probes) at a high throughput. 17 Moreover, it is not affected by unexpected sequence polymorphism at the target sites because the probes used are relatively long. 16 MAPH has previously been used to detect DNA copy number alterations in some forms of genetic disorders 18 –21 and recently in patients with chronic myeloid leukaemia. 22 A physical map covering the 16q22.1 region has been construct- ed 10 and the publication of the draft sequence of the human genome has facilitated the study of multiple genes located in its vicinity. The most obvious candidate TSGs in breast cancer at this region are, firstly, genes that are involved in cell-cell adhesion (members of the cadherin family) [E-cadherin (CDH1), 8,23,24 ], P-cadherin (CDH3), 25,26 VE-cadherin (CDH5) 27 and Ks-cadherin (CDH16) 28 ]; second, genes involved in cell cycle control and cell growth [E2F-transcription factor-4 (E2F-4) 29 and CTCF gene 30 ;] and, third, genes related to telomere function and hence cell senescence and cell immortality [Telomeric repeat binding fac- tor-2 (TRF2) 31,32 ]. The goal of our study was to examine the expression pattern of these candidate genes using immunohistochemistry and at the same time measure their DNA copy number with high precision using the MAPH approach. In this way, the correlation between protein expression and gene copy number as well as the different tumour characteristics could be examined in an attempt to identify the target TSG at 16q22.1 in the low-grade nonlobular type inva- sive carcinomas of the breast. Material and methods Breast tumours Forty-nine invasive lobular, low-grade invasive ductal or tubular breast carcinoma samples that had both fresh frozen material and the corresponding formalin fixed paraffin-embedded tissue blocks *Correspondence to: Department of Histopathology, Nottingham City Hospital, Hucknall Road, Nottingham, NG5 1PB, United Kingdom. Fax: +004-0115-962-7768. E-mail: ian.ellis@nottingham.ac.uk Received 15 June 2004; Accepted after revision 3 August 2004 DOI 10.1002/ijc.20738 Published online 17 December 2004 in Wiley InterScience (www. interscience.wiley.com). Int. J. Cancer: 114, 720 –729 (2005) © 2004 Wiley-Liss, Inc. Publication of the International Union Against Cancer