Characterization and glutamyl preferring carboxypeptidase function of prostate specific membrane antigen: a novel folate hydrolase.

We have cloned the gene encoding the prostate-specific membrane (PSM) antigen, which is recognized by the 7E11C-5 antibody. The antigen is strongly expressed in prostate cancer, and the antibody has been approved for use as an imaging agent for detection of prostatic cancer metastasis. The gene was unique and encoded a type II membrane protein. The only clue to its potential function was found in the cDNA coding sequences from 1250 to 1700, which had a modest but significant homology with transferrin-receptor, demonstrating a 54% homology of nucleic acid sequence. In comparing the mRNA obtained from normal prostate with that obtained from cancerous or lymph node carcinoma of the prostate (LNCaP) cells, normal cells produced a shorter alternative spliced species that encoded a cytosolic form of the protein, and not a membrane protein. It appeared that, as the prostatic cells became cancerous, there was a nearly 100-fold difference in expression of the ratio of the messages encoding the 2 forms, with the cytosolic form (PSM') predominant in normal cells and the membrane form (PSM) predominant in cancer cells. The other tissue in which the membrane antigen form of PSM is highly expressed is the membrane brush border of the small intestine of the proximal, but not distal, small intestine. This is the location of a unique membrane form of a folate hydrolase. This membrane folate hydrolase and its location are necessary in human nutrition because humans require folate, and the folate in foods is poly-gamma-glutamated. Polyglutamated folates cannot be taken into the cells by folate-transporter systems. The ability to take up folate from foods requires the membrane folate hydrolase to sequentially remove the gamma-linked glutamates, freeing folate that can then be transported. PSM antigen has a similar folate hydrolase activity. Others have reported finding an enzyme in the rat brain that functions as an alpha-neurocarboxypeptidase and acts on the abundant brain peptide N-acetylaspartylglutamate to generate glutamate and N-acetylaspartate. The 3'-end of the rat brain enzyme had 84% sequence homology with PSM antigen. Because this enzyme liberates glutamate in the brain, the enzyme is considered to have regulatory activity related to glutamate receptors. Current investigations are underway to determine whether glutamate receptors are present in prostate. Thus, PSM antigen is a unique folate hydrolase-carboxypeptidase that can release glutamate with either gamma-or alpha-linkage. Its enzymatic activity raises a number of questions for consideration. In the normal prostate where the protein is intracellular, is PSM' antigen keeping folate in nonglutamated forms? If so, folate should be able to readily diffuse out of prostate cells, making the prostate gland an organ at risk for localized folate deficiency and carcinogenesis. In prostate tumor cells, with the enzyme outside of the cell, can PSM antigen be used for the activation of cytotoxic prodrugs?