ABSTRACT

The use of carcinoembryonic antigen (CEA) as a recognized tumor marker has affected the prognosis and treatment of colorectal cancer patients for the past thirty years. The cloning of CEA and related family members in the 1980s (Thompson et al., 1991) and assignment of available monoclonal antibodies to distinct CEA-related proteins (Hammarstrom et al., 1989) clearly indicated that multiple CEA antigens are expressed in different normal and diseased tissues. As many research efforts were steered towards the elucidation of functions associated with the human CEA family members, the development of adequate animal models also became a priority. It was with this perspective that we and others attempted to identify the mouse and rat CEA homologs. However, the homolog of the human CEA gene has not yet been found in rodents and may simply be absent from their genomes. Although the first partial cDNA sequence of a mouse CEA-related clone suggested that CEA was present in mouse colon (Beauchemin et al., 1989), the sequence of its full length cDNA provided us with a deduced protein sequence corresponding to that of the biliary glycoprotein (BGP) (Svenberg, 1976; McCuaig et al., 1992, 1993). This glycoprotein is unique within the CEA family in that it contains a cytoplasmic domain which is generated by conserved alternative splicing mechanisms (Barnett et al., 1989; Najjar et al., 1993; Nédellec et al., 1995) and encodes either a short 10 amino acid tail or a longer 71 (human, rat) or 73 (mouse) amino acid domain. In the rat, a cell adhesion molecule orginally called C-CAM 105 (Ocklind and Öbrink 1982), and renamed CCAM, was subsequently shown to be the rat homolog of the human biliary glycoprotein (Margolis et al., 1990; Öbrink, 1991). In this chapter, we will discuss recent findings which identify a novel function for biliary glycoprotein, that of a tumor suppressor.