Malignant mammary tumors represent the most common group of neoplasm in female dogs with frequent incidence of tumor-related death due to recurrence or metastases. It has been proven that similarities between human breast cancer and canine mammary cancer exist with respect to pathogenesis, tumor marker profile, genetic alterations and metastatic behaviour. Therefore, naturally occurring mammary tumors in pet dogs may serve as relevant model to improve the understanding of cancer biology and to support drug development. In humans, carcinoembryonic antigen (CEA, CEACAM5) is widely used as tumor marker because of its overexpression in many types of cancer with simultaneously relatively restricted expression in healthy cells. This feature makes CEA the objective of several immunotherapeutical approaches. It is assumed that soluble CEA interacts with the recently discovered carcinoembryonic antigen receptor (CEAR) and may therefore mediate prometastatic properties. In line with the 'Comparative oncology strategy' the aim of the present thesis was to investigate the function of CEA in canine cancer cells, its possible role as target for immunotherapy and the identification of a putative CEAR in the canine species.
Using immunohistochemistry 23 out of 30 tissue sections of canine mammary tumors showed positive reactivity to monoclonal antibody Col-1, directed against human CEACAM5 and previously applied for cancer diagnosis in dogs. In immunoblot analyses, Col-1 did not only detect CEACAM5 in human colon cancer cells HT29, but also a canine antigen in CF33 and CF41 mammary carcinoma cell lines and in primary mammary tumor samples. According to phylogeny canine CEACAM1 family-related molecules are most closely related to human CEACAM5. However, Col-1 did not react with canine CEACAM1, -23, -24, -25, -28 or -30, all members of the canine CEACAM1 family. By flow cytometry, the Col-1 target molecule was localized intracellularly in canine CF33 and CF41 cells, whereas human CEACAM5 showed surface expression in HT29. Furthermore, Col-1 incubation had neither an effect on canine nor human cancer cell proliferation.
Col-1 treatment decreased AKT-phosphorylation in canine CF33 cells, possibly suggesting an anti-apoptotic function, whereas Col-1 increased AKT-phosphorylation in human HT29 cells. Within the phylogenetic tree we demonstrate further a 99% amino acid similarity of human and canine CEA receptor. Indeed, CEAR could be detected in four canine cancer cell lines by immunoblot and in 10 out of 10 mammary cancer specimen from dog by immunohistochemistry.
In conclusion, Col-1 specifically detects an intracellular target antigen in canine mammary carcinoma cell lines and tissue sections of canine patients. Targeting this antigen by Col-1 alone does not affect cancer cell proliferation, whereas Col-1 treatment induced opposing results between human and dog regarding AKT-phosphorylation. Still, it might be a valuable and specific target for immunotherapeutical strategies in the dog. In contrast to the high heterogeneity of CEACAM molecules between species, to our surprise we found CEAR being highly conserved and also expressed in canine cancer cells. The question whether the specific canine Col-1 target molecule may act as ligand to canine CEAR remains unanswered and must be addressed in future studies.
The evidence collected in this thesis demonstrates the complex functional evolution of the different CEACAM molecules between human and canine species. In contrast, CEAR seems to be highly conserved between many species and may therefore be a comprehensive target for novel anti-cancer therapeutics.