[CANCER RESEARCH 52, 4313-4319. August 15, 1992] Hormone-regulated Apoptosis Results from Reentry of Differentiated Prostate Cells onto a Defective Cell Cycle1 Marc Colombel, Carl A. Olsson, Po-Ying Ng, and Ralph Buttyan2 Department of Urology, Columbia University, College of Physicians and Surgeons, New York, New York 10032 ABSTRACT Castration initiates extensive apoptosis of the secretory epithelial cells lining the ducts of the rat ventral prostate, resulting in the striking regression of this male sexual accessory tissue. We had previously described the paradox of finding similar cascades of gene activity (c-fos > c-myc > hsp-70) induced during the early period of ventral prostate regression and during the regrovrth of the ventral prostate gland initi ated by testosterone replenishment. This common pattern of protoon cogene expression during periods of predominant cellular apoptosis or proliferation caused us to examine further the possibility that the two cellular events occur through identical early molecular pathways. In the present study we demonstrate that apoptotic prostate epithelial cells incorporate bromodeoxyuridine into nuclear high-molecular-weight DNA prior to nuclear DNA fragmentation. The DNA synthetic activity- occurs in coordination with a massive induction of proliferai ivo cell nuclear antigen, a proliferation marker, in the nuclei of androgen-de- prived prostatic epithelial cells. Moreover, this activity is also associ ated with the increased expression of mRNA encoding p53, a suppressor gene well known as a cell cycle-blocking agent. Our data indicate that quiescent (G0) prostate epithelial cells undergo apoptosis due to two sequential events initiated by testosterone depletion. The first event is the active reentry of these cells into the cell cycle. The second event is the apoptotic destruction resulting from the inability of the differenti ated cells to successfully complete this cycle. INTRODUCTION One of the more unusual aspects of hormone action is the ability to induce an apoptotic response in cells of hormone- sensitive tissues. This unique form of cell death has increasingly become the subject of attention due to the biological implica tions of a natural cell death process that is genetically regulated (1). Apoptosis, also referred to as a "programmed death," has been recognized most often during embryogenesis of higher organisms, wherein it provides a sculpting force, giving shape to developing masses of cells (2). In adult tissues, the abnormal onset of apoptosis is known to be a determinant in the devel opment of benign disease conditions (3). Moreover, based on several recent studies of human lymphomas (4, 5), we are be coming more cognizant of apoptosis as an influence that must be considered in our theories concerning the development of human tumors and in efforts to generate more effective cancer therapies. Our current understanding of this process, especially as it is evoked by hormone signals, is limited to certain morphological, biochemical, and molecular characterizations. We can identify apoptotic cell death by the presence of apoptotic bodies: cells with condensed and fragmented nuclei, shrunken to the extent that they pull away from neighboring cells and the basement Received 3/9/92; accepted 6/8/92. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accord ance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 Supported by a grant from the National Cancer Institute (NIH CA47848). 2 To whom requests for reprints should be addressed, at Department of Urology, Columbia University. College of Physicians and Surgeons, 630 W. 168 Street, New York, NY 10032. membrane (6, 7). The nuclear DNA of apoptotic cells demon strates a degradation paradigm in which intranucleosomal DNA is preferentially digested, resulting in a visual "ladder" of DNA fragments in multimers of 180-base pair units upon elec- trophoresis (8, 9). However, the most intriguing characteristic of this form of cell death is the apparent need for genetic in volvement. In many cases, apoptosis can be abrogated by RNA synthesis and protein synthesis inhibitors (10, 11), demonstrat ing the requirement for some recently synthesized gene prod- uct(s) for its progression. Based on the latter characteristic, hormone-regulated apop tosis can be described as a genetic suicide process, potentially driven by the expression of a lethal gene product. Therefore reports of specific and seemingly novel gene products, such as sulfated glycoprotein 2 (12) or tissue transglutaminase (13), that are highly induced during apoptotic cell death have gener ated some degree of interest. To date, none of the identified apoptosis-associated genes can be proved to be lethal to the cell that synthesizes them. Moreover, there is no current evidence that a novel nuclease is involved in hormone-induced apoptotic DNA degradation. Rather, we document here a study of andro- gen-regulated apoptosis in the prostate gland, potentially dem onstrating that apoptosis results from the defective progression of a normal cellular process. This study was initiated because of an earlier study in which castration-induced regression of the rat ventral prostate gland was shown to be accompanied by a protooncogene cascade (transient induction of c-fos prior to the induction of c-myc transcripts) (14) that is most notable because it is usually asso ciated with cellular proliferative responses in vitro ( 15) and in vivo (16, 17). At least one of these molecular signals (induced c-myc) has been localized to the glandular epithelial cells (18), the apoptotic target associated with androgen depletion of the ventral prostate gland. Moreover, castration-induced apoptosis of ventral prostate cells can be suppressed by calcium channel- blocking agents (19, 20) in the same manner that these drugs inhibit androgen-dependent tumor cell proliferation (21). This seeming paradox wherein apoptotic prostate cells re spond in a manner similar to that of proliferating prostate cells might be resolved if apoptosis and proliferation were shown to share an early molecular response pathway. Indeed, this hy pothesis would serve to consolidate our current understanding of apoptosis induced by chemotherapeutic agents wherein the cells must be in cycle to access the apoptotic pathway (22, 23). To determine the extent with which the progress of androgen- regulated prostate cell apoptosis might coincide with normal cell cycle activities, we searched for the presence of other pro liferative markers in the rat ventral prostate gland subsequent to castration. The markers that we have studied indicate that pros tate epithelial cells proceed at least through S phase prior to their apoptotic demise. The coordinately increased activity of a well-known cell cycle suppressor, the p53 gene product, sug gests one potential reason that this cell cycle is never com pleted. 4313 Research. on November 5, 2015. © 1992 American Association for Cancer cancerres.aacrjournals.org Downloaded from