Possible presence of an embryonal carcinoma-associated proteoglycan in the serum of patients with testicular germ cell tumours

Human embryonal carcinoma tumor antigen, Gp200/GCTM-2, is podocalyxin

We previously characterized a peanut agglutinin-binding tumor antigen, gp200, a surface membrane glycoprotein expressed on human embryonal carcinoma, a malignant stem cell of testicular tumors. Gp200 is remarkably similar to another embryonal carcinoma antigen, GCTM-2, a cell differentiation marker that is also detected in blood of testis cancer patients, yet neither molecular identity is known. We now report the identity of gp200 as podocalyxin. Protein sequence results of gp200 peptides match with podocalyxin sequence. Furthermore, two distinct monoclonal antibodies, specific for podocalyxin, react positively with gp200. Therefore, gp200 is a testicular tumor form of podocalyxin, a surface membrane glycoprotein that was originally discovered as a scaffolding extracellular matrix protein of kidney podocyte cells. Podocalyxin is also expressed on subsets of hematopoietic cells where it has a putative function as a cell adhesion protein. This is the first report of podocalyxin expression on malignant cells.

New tumor markers of testis cancer

Potential tumor markers for testis cancer have become numerous with the new molecular techniques available. New protein markers have been evaluated, and histologic factors have shown correlations with stage of disease. Cytogenetic analysis studies have also shown associations with stage progression. Chromosomal markers, oncogenes, and tumor suppressor genes are possible candidates for tumor markers. These new potential tumor markers may become as commonplace as the established markers and may enhance diagnosis, staging, and treatment of testis cancer.

A novel keratan sulphate proteoglycan from a human embryonal carcinoma cell line

We describe here the purification and partial characterization of a 200 kDa keratan sulphate proteoglycan found in the pericellular matrix of human embryonal carcinoma cells. Previously we have shown that this molecule is recognized by a monoclonal antibody (GCTM-2). The antigen was isolated using ion-exchange chromatography and gel filtration, purification being monitored by e.l.i.s.a. using GCTM-2. Metabolic labelling of GCT 27 C-4 embryonal carcinoma cells with sodium [35S]sulphate resulted in the incorporation of [35S]sulphate into the purified molecule. Throughout the purification procedure, the peaks of 35S radioactivity were coincident with the peaks of immunoreactivity, and this label was released both by digestion with keratanase and chondroitinase, confirming the proteoglycan nature of the antigen. The intact molecule ran as a single broad band of 200 kDa, which has been identified by silver staining and immunoblotting following gel electrophoresis. Amino acid analysis of the purified antigen indicated a high content of serine, glycine and aspartic acid/asparagine residues. Visualization by rotary-shadowing electron microscopy suggests that the purified material forms large aggregates, even under denaturing conditions. Deglycosylation of this preparation with trifluoromethanesulphonic acid yielded a major band of 55 kDa and a minor band of 48 kDa. The biochemical nature of the molecule described here, along with tissue distribution studies using GCTM-2, indicates that the antigen is not related to previously described keratan sulphate proteoglycans.

Immunohistochemical and biochemical characterisation of the expression of a human embryonal carcinoma cell proteoglycan antigen in human germ cell tumours and other tissues

In the embryonal carcinoma (EC) cell line GCT 27, monoclonal antibody GCTM-2 recognises an epitope on a 200 kD pericellular matrix keratan sulphate proteoglycan. Immunohistochemical analyses demonstrated staining of tissue sections from 21 out of 22 human non-seminomatous germ cell tumours, and from 22 out of 28 sections of seminomas. In normal human fetal tissues gut epithelium and muscle stained strongly, and certain other epithelia stained moderately. In adult tissues, the distribution of the epitope was similar, but staining intensity was weaker. Neoplastic tissues showed reactivity with embryonal rhabdomyosarcoma and colorectal carcinoma, but no other non-germ cell tumours. Immunofluorescence microscopy showed that GCTM-2 also stained cell lines from human colorectal carcinoma, embryonal rhabdomyosarcoma and choriocarcinoma. In contrast to EC cells the epitope in these other cell types required permeabilization of the cells to be visualised, and the protein bands in immunoblots lacked extensive modification with keratan sulphate and were smaller. Thus, GCTM-2 reacts with an epitope which has a previously unrecognised tissue distribution; its expression as a pericellular matrix proteoglycan is predominantly a characteristic of human EC cells.