30 March 1998
UROT98\REITER.MS--
785 words
Los Angeles--What tissue do prostate cancer cells come from, and how do they turn cancerous? A newly-discovered gene hints at the answer.
Prostate cancer starts in the basal cell epithelium, argues assistant professor Robert E. Reiter, MD, department of urology, UCLA. The new gene, prostate stem cell antigen (PSCA), is expressed strongly in prostate cancer, and resembles other genes involved in apoptosis and proliferation.
"An understanding of normal biology has helped us understand other cancers, like leukemia and hematopoietic cancer," said Dr. Reiter. But prostate cancer isn't understood that well, he said.
"These basal cells that express PSCA have the potential to divide and proliferate," said Dr. Reiter. "We don't know where prostate cancer arises from. The conventional wisdom is that prostate cancer arises from the mature cells that make PSA," based on the observation that basal cells are lost, he said. "However, prostate cancers are known to express a number of genes that are only present in basal cells, such as bcl-2." And now they're known to express another gene.
To find oncogenes, Dr. Reiter compared the RNA expressed by LAPC-4 to the RNA expressed by normal prostate tissue. He found several proteins only in LAPC-4.
PSCA is a 123-amino acid protein whose function is still unknown. But it's part of a family of genes with known functions. PSCA is expressed on the cell surface, mostly in a subset of prostate basal epithelial cells, which are the progenitor cells for the terminally differentiated secretory cells. PSCA is expressed at smaller levels in the placenta, kidney and small intestine. Dr. Reiter reported his work in the Feb. 17 PNAS (95(4);1735-1740).
There was moderate to strong PSCA expression in 111 of 126 (88%) prostate cancer specimens examined by in situ analysis, including high-grade PIN and androgen-dependent and androgen-independent tumors. "In the Northern blot, we found that PSCA is expressed much more strongly in the cancer xenograft than in benign tissue, and we found that it was shed from the cell," said Dr. Reiter. "If that turns be true in vivo in prostate cancer--and based on the in situ's it seems to be--then it may be more specific for cancer than PSA." It may also be a better way to detect metastatic prostate cancer, he added.
But PSCA RNA expression didn't correlate with stage or grade of tumor, so unlike HER-2/neu and p53 it didn't predict outcomes.
Once the UCLA group had the DNA sequence, they could look for other proteins that resembled PSCA. They found two: one human, one mouse.
The human protein was stem cell antigen 2 (SCA-2), with 30% similarity. SCA-2 is expressed in developing T-cells in the thymus, and seems to prevent their apoptosis.
"These stem cell markers have enabled us to sort out the lineages of hematpoietic cells," said Dr. Reiter. "We don't know much about normal prostate development, and we don't have any markers for it. PSCA may help us sort out the cells that are involved in the development of normal and cancerous prostate in the same way."
SCA-2 is a member of a family of proteins that transmit signals from the cell surface to the cell interior, the Thy-1/Ly-6 family of glycosylphosphatidylinositol (GPI)-anchored cell surface antigens.
"This family is unique because the proteins are anchored to the surface itself," said Dr. Reiter. "They're not transmembrane proteins."
Thy-1 is involved in T-cell activation and transmits signals through tyrosine kinases. Ly-6 genes have been implicated in tumorogenesis and cell adhesion between cells of the same type.
"From its restricted expression in basal cells and its homology to SCA-2, we hypothesize that PSCA may play a role in stem progenitor cell functions," such apoptosis and proliferation, said Dr. Reiter.
Another sequence was 70% similar, and seemed to be a mouse version of PSCA. The protein was also found predominantly in the prostate. Dr. Reiter is now breeding mice with that gene knocked out, to see whether it has any effect.
With the sequence, they could also find the PSCA chromosome. It was chromosome 8, band q24.2, at a well-known cancer hot spot. In most prostate cancers, the long, or q, arm of chromosome 8 gains sequences, and the short, or p, arm loses sequences. SCA-2 and another Ly-6 gene are nearby. So is c-myc, which undergoes allelic gain or amplification in prostate cancer metastases.
Dr. Reiter is a consulting scientist to UroGenesys, Inc., Santa Monica, CA, a biotechnology company with commercial rights to PSCA-2.