Purpose: The NKX-3.1gene is an androgen regulated prostate specific homeobox gene that is believed to have a vital role in normal prostate development. In mice the homologue NKx3.1 is exclusively expressed in prostate epithelium. In humans NKX3.1 expression is also restricted to the prostate but to our knowledge the cellular location has not been described. Furthermore, since NKX3.1 maps to chromosomal band 8p21, a region with high loss of heterozygosity in prostate cancer, the gene has been proposed to have tumor suppressor function. In this study we demonstrate that in human prostates NKX3.1 is expressed exclusively in secretory epithelial cells and the level of NKX3.1 expression remains invariant in normal tissue and in tissue showing various grades of prostate cancer. In the 19 cases examined the DNA sequences of the NKX3.1 gene were identical and no mutation was detected.
Materials and methods: Frozen tissue from patients who underwent radical prostatectomy was used for this study. For in situ hybridization experiments a 377 bp fragment corresponding to a portion of the 3' untranslated region of the NKX3.1 gene was amplified by polymerase chain reaction and cloned into the pCRII plasmid vector Invitrogen. Antisense or sense [33P] uridine triphosphate labeled RNA probes were generated with SP6 or T7 RNA polymerase and hybridized to the tissue sections. Slides were exposed to photographic emulsion and visualized on autoradiography. Laser capture microdissection was performed to procure pure populations of malignant epithelium. DNA was isolated by digesting samples in proteinase K buffer. Polymerase chain reaction and direct sequencing was performed using standard protocols.
Results: In vitro hybridization showed that NKX3.1 expression was restricted to secretory epithelial cells within benign prostate glands. No expression was detected in stroma or infiltrating lymphocytes. NKX3.1 was expressed in all grades of malignant epithelium in all 25 cases examined. Direct sequencing of the coding region of NKX3.1 revealed the wild-type sequence in all 18 microdissected cancers analyzed.
Conclusions: Based on our studies we propose that NKX3.1 gene expression is restricted to benign and malignant secretory epithelium within the prostate but NKX3.1 does not appear to be a classic tumor suppressor gene responsible for prostate cancer initiation. These findings are consistent with the role of NKX3.1 in the development of normal prostate epithelium and maintenance of normal secretory function. Thus, NKX3.1 may represent a useful molecular marker for benign and malignant prostate epithelium.