69P Novel monoclonal antibodies can distinguish Cripto-1 from Cripto-3 proteins: Clinical implications and potential new biomarkers

Cripto-1 (CR1) is an oncofetal protein required for implantation and expressed in the adult during wound repair, inflammation, and tumorigenesis. A second gene codes for a very homologous protein, Cripto-3 (CR3). Both proteins contain 188 amino acids and differ only in 6 residues. Heretofore, all available antibodies cannot discriminate between these two proteins, preventing any investigation on their differential contributions to cancer biology. Here, we present the generation, characterization, and clinical implications of highly specific monoclonal antibodies (MoAbs) for these two proteins. Different peptide fragments of either protein were used as antigens to generate mouse MoAbs (233 clones for CR1, 319 clones for CR3) that were selected by their target protein specificity through SPR and ELISA assays. Selected MoAbs were used for immunohistochemistry on tissue arrays including normal tissues as well as cancers of the lung, breast, colon, ovary, and prostate. MoAbs were also used for Western blotting and ELISA characterization of human serum (breast cancer vs healthy controls). MoAbs NCI 5G1-1 and NCI 5G11-2 were highly specific for CR1 and CR3, respectively. No crossreactivity was observed with the other protein. Immunohistochemical analysis of cancer specimens showed differential staining patterns, where some tumor cells expressed both proteins, others expressed only CR3, and still others show vascular endothelial cells stained for CR1 while tumor cells express CR3. In prostate, colon, and breast cancer, CR1 and/or CR3 protein expression correlates with clinical parameters, such as TMN_N, TMN_M, tumor stage, tumor grade, or PR expression. Lastly, both CR1 and CR3 interact with established binding proteins Nodal, GRP78 and Alk4 and competitively interfere with one another for targeted binding. We describe for the first time the development of MoAbs that discriminate human CR1 from CR3, confirming that CR3 is translated into a protein in human cells, thus removing its pseudogene status. We also confirm CR1, and propose CR3, as cancer prognosis and severity markers.