Main

The adenomatous polyposis coli (APC) gene, mapped to chromosome 5q21 (Kinzler et al, 1991), plays a prominent role in the development of colorectal cancer, both in the autosomal dominant inherited familial APC syndrome (Bodmer et al, 1987; Groden et al, 1991; Joslyn et al, 1991; Kinzler et al, 1991; Nishisho et al, 1991) and in sporadic colorectal cancer (Fearon and Vogelstein, 1990; Miyoshi et al, 1992; Powell et al, 1992). An impaired function of APC, most often attributable to mutations within the coding sequence of the gene, leads to a lack of degradation and nuclear accumulation of β-catenin, which acts as a transcriptional activator, causing loss of cell growth control (Sparks et al, 1998). Moreover, APC functions in pathways counteracting metastasis by mediating intercellular adhesion and stabilising the cytoskeleton (Fearnhead et al, 2001).

Similar to findings in colorectal cancers, it has been suggested that disruption of the APC/β-catenin pathway may be involved in breast cancer. Loss of APC expression and upregulation of β-catenin have been described in human breast cancer and breast cancer cells (Ho et al, 1999; Jonsson et al, 2000; Schlosshauer et al, 2000). Somatic APC mutations are reported in only a minority of breast cancers (Furuuchi et al, 2000), despite high rates of allelic loss at chromosome locus 5q21 (Thompson et al, 1993; Medeiros et al, 1994). Nevertheless, epigenetic inactivation of APC due to DNA methylation is frequently present in both breast cancer cell lines and breast cancer tissue. In most cultured breast cancer cells, there is a complete concordance between APC promoter methylation and silencing of its transcript (Virmani et al, 2001). Cellular APC expression can be restored after demethylation with 5-aza-2′-deoxycytidine treatment. APC promoter methylation also occurs in a significant number of primary breast tumours (ranging from 28 to 53% of cases, depending on the applied technology) (** et al, 2001; Virmani et al, 2001; Liu et al, 2007). The frequency of APC methylation in primary breast tumours increases with tumour stage and size (Virmani et al, 2001; Roa et al, 2004; Yan et al, 2006). Interestingly, global profiling of DNA methylation revealed more methylated genes in normal adjacent samples than in normal donor control samples. There are several possible explanations for this observation. The intermediate levels of methylation might reflect the infiltration of neoplastic cells in histologically ‘normal’ surrounding breast tissue. Alternatively, the hypermethylation of genes in samples from normal tissue adjacent to a breast tumour could be explained by field cancerisation. This concept was originally proposed by Slaughter et al (1953) to explain the development of multiple primary tumours and locally recurrent cancer (Slaughter et al, 1953). Previous studies have confirmed that genetic abnormalities exist in histologically normal breast tissues immediately adjacent to invasive cancers (Deng et al, 1996). Now, it has also been suggested that the primary tumour might serve as an epicentre from which methylation density progressively diffuses outwards to surrounding tissues (Yan et al, 2006).

We found no significant difference in APC mRNA and protein levels among tissues with low or high APC methylation status. Possible explanations for these results are: (i) samples with APC promoter methylation may have only one allele affected, allowing expression from the unaltered allele (Esteller et al, 2000); (ii) gene silencing is not a single event, but instead a series of events that begin with a marked drop in transcription and ends with its complete cessation (Turker, 2002); (iii) APC methylation patterns among the tumour cells that constitute a given sample might be heterogeneous and (iv) APC expression might be inactivated by gene mutations or allelic losses and not by methylation of CpG sites in the promoter region.

To conclude, we have shown that aberrant methylation of the APC gene promoter characterises the IBC phenotype. It should be emphasised that more investigation is required to determine to what extent the epigenetic inactivation of APC affects the biological behaviour of these tumours. Our study was limited not only by the small sample size, but also by the use of a candidate gene approach that was based on the observation of low gene expression. The analysis of more IBC cases and the consideration of additional genes, which is facilitated by several recently described techniques, such as methylation-sensitive arbitrarily primed PCR, restriction landmark genomic sequencing and CpG-island microarrays (see reference (Ushijima, 2005) and references therein), could present a great opportunity for enriching our knowledge of IBC biology.