Co-incidental increase in gene copy number of ERBB2 and LRIG1 in breast cancer

The LRIG1 gene (leucine-rich repeats and immunoglobulin like domains-1) at chromosome 3p14 is a proposed tumour suppressor gene whose gene product negatively regulates various receptor tyrosine kinases. This function has been the basis for classifying LRIG1 as a potential tumour suppressor gene (TSG). The ERBB receptor family is important in malignant cellular functions such as proliferation, survival, adhesion, migration and differentiation. In breast cancer, amplification of the ERBB2 proto-oncogene is an important negative prognostic factor. The epidermal growth factor receptor (EGFR/ERBB1), is expressed in colorectal cancer and has been correlated to a worse prognosis. Until recently, immunohistochemical analysis of EGFR expression was used to select patients suitable for treatment with EGFR targeted antibodies. This thesis characterizes LRIG1 in breast and colorectal cancer to gain further knowledge of the gene and its expression. Also, the EGFR expression in metastases and the invasive margin of colorectal cancers was investigated to correlate changes to clinical factors. Breast cancer samples and matched normal tissues were evaluated for LRIG1 and the ERBB receptors at gene, RNA and protein levels. An increase in copy number of the LRIG1 gene was evident. Also, increased LRIG1 copy number was associated with high levels of ERBB2 mRNA. Another set of breast cancer tumours were analysed for LRIG1 by FISH analysis. The results were coherent with the previous results. To further analyze the correlation to ERBB2, tumours with LRIG1 increased copy number were analysed for ERBB2. The data showed that 89% of tumours with increased LRIG1 copy number were either ERBB2 amplified or had an increased copy number of ERBB2. To investigate LRIG1 and the EGFR in colorectal cancer, the gene and protein expression was analysed by several methods in tumours and corresponding normal tissues. There were no significant changes at gene level found, but at the protein level, both over- and under expression were seen. No evident correlation between LRIG1 and EGFR expression was detected. The ERBB receptor family expression in colorectal cancer tumours and corresponding metastases was investigated to explore if the expression was altered in the metastatic lesion. The results showed that the EGFR expression was lost in the corresponding metastases in 33% of the tumours and that the same percentage of tumours gained expression in the metastases. Co-expression of the ERBB family members was also analysed; there was a significant increase of ERBB3/ERBB4 co-expression in late stage tumours. EGFR expression at the invasive margin of colorectal cancers was analysed to clarify whether expression correlated to the patient’s prognosis. Significant correlation to survival and the presence of budding was seen. In conclusion, 34% of the breast cancer tumours studied had an increased copy number of LRIG1 with a significant co-incidental increase in ERBB2 copy number. This raises the question of a functional correlation between LRIG1 and ERBB2, a finding that might be of clinical importance. The studies of EGFR and the ERBB receptors in colorectal cancer reflect the heterogeneity of EGFR expression in tumours. In addition, these findings suggest that survival of the patients correlates to an increasing EGFR expression at the invasive margin.


(page number not for citation purposes)
Using fluorescence in situ hybridization (FISH), we previously showed that the LRIG1 gene had an increased copy number in 11 of 28 (39%) breast cancer tumours [1]. The LRIG1 gene (leucine-rich repeats and immunoglobulin-like domains 1) at chromosome 3p14 is a proposed tumour suppressor gene that negatively regulates various receptor tyrosine kinases, including the breast cancer proto-oncogene product ERBB2 [2,3].
Recently, however, Miller and colleagues [4] showed that 10 of 13 (76%) ERBB2 + tumours had decreased LRIG1 protein levels compared to normal breast tissue. As their data showed down-regulation at the protein level whereas our data showed an increased copy number at the genomic level, we analysed 45 additional breast tumours by FISH as previously described [1]. Thus, out of 73 tumours analysed to date, 25 (34%) did indeed have increased LRIG1 copy number. To further analyse the relationship between LRIG1 and ERBB2 at the genomic level, we evaluated the ERBB2 gene copy numbers in 18 tumours with increased LRIG1 copy number using FISH analysis according to standard procedures. Interestingly, 16 (89%) out of the 18 tumours displayed increased copy number of ERBB2 (Figure 1). This suggests that the majority of breast cancer tumours with increased copy number of ERBB2 simultaneously had increased LRIG1 copy number (our data) and decreased LRIG1 protein levels [4].
We draw the following major conclusions from these results. First, as previously shown, a significant proportion of breast tumours have an increased LRIG1 gene dosage. Second, there is a correlation between increased gene copy numbers of ERBB2 and LRIG1. Third, based on the Miller protein data, most of the tumours with increased LRIG1 gene dosage express reduced levels of the LRIG1 protein. This indicates a negative selection against LRIG1 protein expression, supporting the notion that LRIG1 is a tumour suppressor in breast cancer. Although the mechanism behind the down-regulation of LRIG1 protein in breast cancer is not known, it has been reported that increased gene copy Letter

Figure 1
Increased copy number of LRIG1 and ERBB2 in human breast cancer in the same patient. Interphase nuclei from a breast cancer tumour were analysed by FISH. (a) A specific LRIG1 probe (red) showed increased LRIG1 copy number (five copies) whereas a specific centromere probe (CEP3) (green) showed normal chromosome 3 copy number (two copies). (b) A specific ERBB2 probe (red) showed amplification of the ERBB2 gene whereas a specific centromere probe (CEP17; green) showed three copies of chromosomes 17. numbers in some cases are associated with decreased mRNA expression [5]. In any case, the high frequency (34%) of tumours with increased LRIG1 gene copy number implies a positive selection for tumour cells with this genomic alteration. It remains, however, to be elucidated whether the molecular driver behind the selective advantage associated with this alteration is LRIG1 down-regulation per se. Other possibilities include activation of nearby proto-oncogenes or the generation of novel oncogenic fusion genes.
In summary, the co-incidental increase in copy number of ERBB2 and LRIG1 in breast cancer is a novel finding, pointing at a functional co-operation between these genetic events, where the biological and clinical importance need to be clarified further.