Apoptotic versus proliferative activities in human benign prostatic hyperplasia

Apoptotic Versus Proliferative Activities in Human Benign Prostatic Hyperplasia NATASHA KYPRIANOU, PHD, HUACHENG TU, MD, AND STEPHEN C, JACOBS, MD Cell growth in the normal prostate is regulated by a delicate balance between cell death and cell proliferation (ie, apoptotic v proliferative activity). Disruption of the molecular mechanisms that regulate these two processes may underline the abnormal growth of the gland leading to benign prostatic hyperplasia (BPH). In this study, the incidence of programmed cell death (apoptosis) and cell proliferation was comparatively analyzed among the various cell subpopuladons in the normal and benign hyperplastic human prostate. The authors also examined the relative expression of two proteins involved in the regulation of prostate apoptosis: (1) transforming growth factor (TGF)-~I, a negative growth factor able to induce prostate apoptosis under physiological conditions; and (2) bd.2, a potent apoptosis suppressor. Analysis of the incidence of "spontaneous" apoptosis in situ, using the end-labeling terminal transferase staining technique for the detection of nucleosomal DNA fragmentation, revealed infrequent apoptotic staining in isolated basal and secretory prostate epithelial cells. The basal level of cell proliferation was determined on the basis of the Ki-67 nuclear antigen staining, a nuclear protein that appears primarily dm-it~ the proliferative phases of the cell cycle. The Ki-67-positive nuclei were equally distributed among the basal and secretory epithelial cells of the hyperplastic prostatic acini. The apoptotic index of the secretory and basal cells of the prostate epithelium was higher in the normal prostate compared with BPH tissue, whereas there was a significant increase in the proliferative index of the respective cell populations in the hyperplastic prostate. Balancing the apoptotic versus the proliferative activ-

ires revealed a substantial net decrease (fourfold) in the total number of cells dying via apoptosis in both the glandular and basal epithelial cell compartments of the hypertrophic prostate (BPH) when compared with the normal gland. TGF-~ staining was exclusively identified in the secretory epithelial cells, lining the prostatic lumen with minimal involvement of the basal cells and total lack of immunoreactivity among the stroma elements. Statistical analysis revealed a significant elevation in TGF-/~ expression in the epithelial ceils of BPH tissue compared with the normal prostate (P < .001). Expression of bcl-2 was topologically restricted to the glandular epithelium of the prostate. In the normal prostate, bd-2 immunoreactivity was predominantly identified in the basal cell layer. An increase in both the intensity of immunoreactivity for bd-2 and the number of positive epithelial cells (basal and secretory) was detected in BPH specimens relative to the normal prostate (P < .02). These results suggest a potential involvement of enhanced expression of this antiapoptosis protein in deregulation of the normal apoptotic cell death mechanisms in the human prostate, thus resulting in a growth imbalance in favor of cell proliferation that might ultimately promote prostatic hyperplasia. HUM PATHOL27:668--675. Copyright © 1996 byW.B. Saunders Company Key words: benign prostatic hyperplasia, apoptosis, cell proliferation, transforming growth factor-/~ bcl-2 expression. Abbreviations: BPH, benign prostatic hyperplasia; TGF-~I, transforming growth factor ~1; TUNEL, transferase deoxyuridine end labeling.

Benign prostatic hyperplasia (BPH) is the most commonly occurring neoplastic disease of m a n k i n d ] The incidence of BPH increases with advancing age, so that 85% of all men older than 50 years of age have symptoms arising from BPH.~ Fifty percent of all American men will eventually require treatment for symptomatic relief of clinical BPH by the time they reach 80 years o f a g e . 2 Although these epidemiological facts underscore the importance of BPH as a major health problem, surprisingly little is known about or agreed on the cellular and molecular processes involved in the development and pathogenesis of the disease. Consequently, no effective nonsurgical therapeutic modalities selectively targeted at BPH are currently available. 2 There is an immediate need, therefore, to gain an enhanced understanding of the molecular basis of this disease.

The h u m a n prostate gland undergoes two distinct phases during the lifetime of the host. The first (ie, the growth phase) begins at birth and continues until the prostate reaches its normal adult size. If the host is castrated at birth, the proliferative growth of the prostate is completely blocked, indicating the chronic requirement for androgens during the growth phase of the prostate. BOnce the maximum adult size of the prostate is reached, the gland usually ceases its net growth, and a second maintenance phase of the prostate begins. During this maintenance phase, there is a continuous cell turnover, with the rate of cell proliferation being balanced by the rate of prostatic cell death, so that neither overgrowth nor involution of the gland usually o c c u r s . 4 A n imbalance between the rate of cell proliferation and cell death based, on either an increase in the proliferative activity or a decrease in the apoptotic (cell death) activity, would result in a ratio higher than 1 and consequently an increase in cell number, ultimately resulting in uncontrolled prostatic growth that characterizes BPH. Kinetically, BPH development is characterized by a slow growth, which progresses over decades with a doubling time of at least 10 years. 3 Previous studies by numerous investigators have been targeted at defining the rate of cell proliferation of the different cell subpopulations of benign h u m a n prostate and its potential significance in BPH development. 5-9 There have been no reported attempts, however, to investigate

From the Division of Urology, Departments of Surgery and Biochemistry, and the Cancer Center, University of Maryland School of Medicine, Baltimore, MD. Accepted for publication December 13, 1995. Supported by a Research Scholarship to N. Kyprianou from the American Foundation for Urologic Disease with funds contributed by Searl and by departmental funds from the Division of Urology (University of Maryland Medical Center). Address correspondence and reprint requests to Natasha Kyprianou, PhD, Division of Urology, 22 South Greene St, University of Maryland Hospital, Baltimore, MD 21201. Copyright © 1996 by W.B. Saunders Company 0046-8177/96/2707-000555.00/0

668

CELL DEATH AND CELL PROLIFERATION IN BPH (Kyprianou et al)

the role of the other equally dynamic parameter in the prostatic growth equilibrium (ie, programmed cell death [apoptosis] in promoting BPH. Apoptosis is the molecular mechanism of physiologically relevant cell death in eukaryotic cells and is primarily responsible for the programmed elimination of cells, which accompanies embryonic development, and regulation of the size of organs in adult life. l° Despite the recognition of the critical homeostatic role played by apoptosis in the regulation of tissue dynamics,~l little is known about the involvement of this molecular process of cellular suicide in the control of h u m a n prostate growth and its potentially important role in the development of BPH. In the normal adult man, prostate growth equilibrium is primarily maintained by androgens. Castration-induced androgen deprivation leads to a decrease in cell proliferation and a dramatic increase in cell death so that by 7 days postcastration more than 70% of the epithelial cells die via apo~ptosis, ultimately resulting in involution of the gland.~ Cells undergoing apoptosis display characteristic morphological features, such as condensed chromatin, nuclear disintegration, cell surface blebbing, and formation of membrane-bound apoptotic bodies. An early biochemical event that characterizes induction of apoptosis as a cellular commitment step is the irreversible fragmentation of genomic DNA into nucleosomal oligomers because of the activation of a Ca ++ Mg++-dependent endonuclease.l~'~~ TGF-fll is a member of a family of nmltifunctional regulators of cell growth, controlling diverse cellular processes, such as proliferation, differentiation, extracellular matrix formation, and morphogenesis) 4 Recent evidence has linked the increase of expression transforming growth factor/3 (TGF-/31), with a concomitant induction of apoptosis in diverse cellular systems 1 "ncluding the prostate. 15 -17 Significantly enough, TGF/3 has been shown to induce the apoptotic death of androgen-dependent prostatic epithelial ceils even in the presence of physiological levels of androgens, 15'I8 evidence that points to a role for TGF-/3 as a physiological regulator of epithelial cell apoptosis in the prostate. Active continuous protection from programmed cell death is a widespread and important p h e n o m e n o n that has been conserved from nematodes to humans. The bcl-2 proto-oncogene is a powerful inhibitor of apoptosis in numerous mammalian systems.~9 This gene was initially identified by its involvement in a nonrandom chromosomal translocation t(14:18)(q32;q21), found in most human follicular lymphornas. 19" Its protein product is expressed in many types of long-lived cells, such as memory lymphocytes, neurons, and regenerating stem cells that line the basement membrane in many continuously replaced epithelia including the skin, colon, and prostate. 2° Through abnormally prolonging the life span of cells otherwise destined to die, bcl-2 provides a powerful biological mechanism for prolonging cell survival, which may be ultimately responsible for neoplastic development. In view of the recent identification of the oncogenic role of bcl-2 in the initiation and progression of h u m a n malignancies, including the prostate, 2123 and its known expression in normal

669

prostatic e p i t h e l i u m y the authors investigated the potential deregulation of bcg2 expression in BPH. In this study, the hypothesis under investigation is that escape from the normal apoptotic process controlling epithelial cell elimination in the human prostate could be responsible for the uncontrolled growth that characterizes BPH. For this purpose, the authors specifically examined (1) whether a decrease in the apoptotic index of the different cell subpopulations in the prostate (ie, basal and secretory epithelial, and stroma) could account for the hyperplastic growth of the gland; (2) whether a potential deregulation of the apoptotic activities among the prostate cell populations in BPH are paralleled by changes in the proliferative activities of the same prostate cell populations; and (3) the expression pattern and topological distribution of two proteins that have been associated with apoptosis regulation (ie, bcl-2, a potent suppressor of apoptosis; and TGF-/3, a physiological signal for triggering prostate apoptosis). The present findings showed a significant decrease in the number of cells undergoing apoptosis in both the epithelial and basal components of the BPH prostate compared with the normal gland, an imbalance in favor of cell proliferation. The study further suggests that this resistance to "spontaneous" apoptosis of the basal and secretory epithelial cells in the hyperplastic prostate might be caused by the ability of these aging cells to express and maintain high levels of bcl-2 (the potent apoptosis suppressor). In addition, this study identified the glandular epithelium as the major localization of TGF-fl expression. In the hyperplastic prostate, the continuous presence of enhanced levels of TGF-fll, which potentially dictates induction of the apoptotic process among the epithelial cell populations, was unable to overcome the strong "antiapoptotic" ability Of bcl-2.

MATERIALS AND METHODS Tissue Selection

Fifty-two prostate specimens were obtained from the Pathology Department of the University of Maryland Medical Center. Five normal prostate specimens were obtained from four organ donors (ages 20 to 35), admitted for multiorgan extraction at the University of Maryland Medical System. Prostatic tissue specimens obtained from 47 BPH patients undergoing transurethral resection of the prostate or radical prostatectomy for BPH were processed for routine formalin fixation, paraffin embedding, and histological sectioning. Consecutive 6-#m-thick sections of formalin-fiixed, paraffin-embedded tissue were cut from trimmed blocks and were collected on adhesive-coated glass slides. All tissue specimens were histologically diagnosed by a pathologist before immunostaining, using hematoxylin-eosin staining of serial sections. I m m u n o h i s t o c h e m i c a l Analysis

The slides were deparaffinized with three washes of xylene, rehydrated with gradual changes in ethanol (from higher to lower concentration), and subsequently incubated in hydrogen peroxide to quench endogenous peroxidase. To unmask the antigens, the slides were treated with enzymatic digestion, depending on each specific antibody.

HUMAN PATHOLOGY Volume27, No. 7 (July 1996) Detection of Apoptosis

Statistical Analysis

Spontaneous apoptosis was identified histologically in the paraffin-embedded prostatic tissue sections using the in situ end-labeling procedure for the detection of nucleosomal DNA fragmentation, which uses the terminal transferase reaction (TUNEL).24'25 This was essentially performed using the ApoTag kit (Oncor, Inc, Gaithersburg, MD) according to manufacturer's instructions. Negative controls for the apoptotic staining consisted of consecutive sections of each case in which the TdT enzyme was omitted.

The data obtained from the in situ labeling of fragmented DNA (apoptosis) and the immunostaining analysis (Ki-67, bcl-2, and TGF-fl) were analyzed for statistical significance using the Student's t-test and the Kruskal-WaUis oneway analysis of variance. Values are presented as the mean plus or minus the standard error of the mean.

RESULTS Determination of the Dynamics of Cell Death and Cell Proliferation in the Normal and Hypeffrophic Prostate Cells

Determination of the Proliferative Index and bcl-2 and TGF-~ Expression The proliferative index of the different cell types (ie, basal and secretory epithelial and strorna cells) was determined by immunohistochemical staining of histological sections using the mouse monoclonal antibody MIBI (AMAC, Westbrook, ME). This antibody has been previously shown to specifically recognize Ki-67 antigen in microwave-processed, formalin-fixed, paraffin-embedded sections. 26 The Ki-67 antigen is a nuclear protein that is highly expressed during all the phases of the proliferative cycle (ie, G1, S, G2, and mitosis) but is totally absent when the cells are out of cycle (ie, resting phase, GO). The antihuman bcl-2 mouse rnonoclonal antibody was obtained from DAKO (Glostrup, Denmark). The antiTGF-/3 rabbit polyclonal antibody, against the mature intracellular human TGF-fll (aminoacids 1-30) was a generous gift from Dr M. Sporn (National Cancer Institute, Bethesda, MD). The immunoperoxidase procedure for the detection of immunoreactivity was performed using the antimouse ABC kit (Oncogene Science, Uniondale, NY) for Ki-67 and bcl-2 staining and the anti-rabbit ABC kit (Signet, Dedham, MA) for TGF-/3 staining, respectively, according to manufacturer's instructions. The color reaction produced was developed with diaminobenzidine and hydrogen peroxide. Sections were counterstained with methyl green for the TUNEL end-labeling reaction, bct2 and TGF-/3 staining, and with hematoxylin for the Ki-67 staining. Negative control slides were processed with each experiment and excluded the primary antibody, but retained all the other steps of staining procedure.

Quantitative Analysis of Staining Precise quantitative evaluation of the positive staining among the different components of tissue sections was performed by manual counting. The stained slides were viewed under a 1-cmz reticule, which was divided into 100 smaller squares and attached to the eyepiece of the fight microscope. The cells were counted in randomly selected fields at ×400 magnification. At least 1,000 to 2,000 cells per section were counted, depending on the size of each section using the random sampling technique as the authors have previously described, z3 The index of positive cells was expressed as the percentage of the ratio of positive cells over the total number of cells counted. Cells positive for TGF-/31 and bcl-2 staining were identified by two independent observers, and the intensity of immunostaining was scored as follows: 1+, weak; 2+ moderate; and 3+, intense. For TGF-/3, the staining intensity was assessed relative to the intensely stained polyrnorphonuclear leukocytes (arbitrarily designated as 3+). For bcl-2, the strength of intensity was judged relative to the infiltrating lymphocytes, which also served as internal positive controls and were arbitrarily designated as 3+. For all the immunoreactivity patterns, a value of <0.5% was used as a cutoff to define negative staining.

670

Because the p h e n o m e n o n of nucleosomal DNA fragmentation is a hallmark characteristic of apoptosis, terminal transferase e n d labeling was used specifically to identify cells u n d e r g o i n g apoptosis in histological sections of BPH prostate. This technique is highly specific for all the apoptotic (but n o t necrotic) cell populations. On histological evaluation o f the prostate tissue sections, necrosis (the nonspecific m o d e of cell death) was not c o m m o n l y identified using morphological criteria in any of the tissue specimens analyzed. Similar observations regarding the specificity of the assay have b e e n described previously. 25 Apoptotic DNA fragmentation was histologically detected a m o n g b o t h the luminal epithelial and basal cells of the n o r m a l prostatic glandular epithelium using the T U N E L p r o c e d u r e (Fig 1A). Characteristically, the apoptosis-positive cells were individually scattered within a population of n o r m a l prostate epithelial cells. T h e T U N E L labeling was predominantly confined to the nucleus of the apoptotic cells, but occasionally weak cytoplasmic labeling was observed, possibly reflecting potential leakage of the small nucleosomal fragments ( 2 0 0 - b a s e pair DNA pieces), f r o m the nucleus into the cytoplasm. Adjacent stroma areas were consistently negative for apoptotic e n d labeling. Spontaneous apoptosis was m o r e morphologically conspicuous in sections of hypertrophic prostate (Fig 1B). As docum e n t e d in Fig 2, the apoptotic index of basal and secretory epithelial cells in the n o r m a l prostate was higher than that in BPH specimens, although this difference failed to reach statistical significance (P > .05). Balancing the proliferative and the apoptotic indices a m o n g the secretory and basal epithelial cell populations, however, revealed a ratio higher than 1, indicating a net decrease in the n u m b e r of cells u n d e r g o i n g apoptosis in BPH specimens c o m p a r e d with the normal prostate. Considering the large statistical e r r o r that characterizes the m e a n values for the apoptotic indices of b o t h the glandular epithelial a n d basal cell populations (Fig 2) in the n o r m a l prostatic epithelium, one cannot present the same a r g u m e n t to justify an imbalance in the growth equilibrium in the n o r m a l gland (Fig 3). Positive immunoreactivity for Ki-67 antigen was specifically confined to nuclei of the basal and luminal epithelial cells of hyperplastic prostatic acini, with infreq u e n t stroma staining (Fig 1D). As shown in Fig 3, the proliferative indices of b o t h the epithelial and basal cells in the n o r m a l prostate are remarkably low (0.8%

CELL DEATH AND CELL PROLIFERATION IN BPH (Kyprianou et al)

FIGURE 1. Immunohistochemical detection of apoptosis, Ki-67, bct-2, and TGF-/~ expression in prostate tissue sections. Detection of apoptosis in situ using the terminal transferase end-labeling technique in normal prostate (A) and BPH specimen (B). Apoptotic DNA fragmentation was identified in a few secretory epithelial cells and more infrequently among basal cells. (Original magnification x100.) (C) A serial section stained with the bct-2 antibody and representing an area of high bct-2 immunoreactivity among the basal and the secretory glandular cells of the prostate epithelium, The tall columnar epithelial cells surrounding the prostatic secretion exhibited strong positivity for bcl-2 (3+). (Original magnification x 100.) (D) Representative Ki-67 antigen staining indicating proliferating individual epithelial (secretory and basal) and stromal cells in BPH prostate (Original magnification ×400.) (E) Cytoplasmic TGF-/~ immunoreactivity among the epithelial cells of BPH prostate; note that stroma elements are totally negative. (Original magnification ×400.) All areas are representative of random fields.

671

HUMAN PATHOLOGY

Volume 27, No. 7 (July 1996)

3.5

TABLE

1. Summary of bct-2 and TGF-fl Immunostaining Analysis* bcl-2 Intensity

TGF-/3/Intensity

2.5 t< z o I~

Cell Type

2

Secretory Epithelial Basal Stroma

1.5

o <

1

0 BPH

N

EPITHELIAL

BPH BASAL

N

BPH

Normal Prostate

BPH

0-1 +

2-3 +

1+

3+

1-2+ 0

3+ 0

0 0

0-1+ 0

Abbreviations: BPH, benign prostatic hyperplasia; TGF-fl, transforming growth factor/3. * The intensity of bcl-2 and TGF-/3 immunoreactivity was graded on an arbitrary 0 to 3-point scale: 0, no immunoreactivity; 1 +, weak immunostaining; 2+, moderate; 3+, intense immunoreacfivity.

0.5

N

Normal Prostate

BPH

STROMA

FIGURE 2.

The apoptotic index of the various cell types of human prostate (ie, epithelial, basal, and stroma) in normal (n = 5) and BPH (n : 47) specimens. Values represent the mean percentage of cells that showed positive labeling of DNA fragmentation, as detected in situ with the TUNELtechnique.

and 0.91%, respectively). This low rate of epithelial glandular cell proliferation is sufficient to balance the equally low rate of spontaneous apoptotic cell death of the same cell populations. This finding clearly indicates that the normal prostatic cells are in a steady-state, selfrenewing condition in which neither overgrowth nor accelerated elimination of these cell populations occurs. The proliferative index of the stroma component was remarkably low for both the normal and BPH prostate (0.7 and 1%, respectively), with no statistical difference between the two values, indicating the lack of any significant involvement of the stromal cells in the growth equilibrium during BPH development.

summarized in Table 1. Positive immunoreactivity with bcl-2 was predominantly confined to the basal cells of the normal prostatic glandular epithelium, with minimal involvement of the secretory epithelial cells, while the surrounding stroma lacked staining (Table 1). In the normal prostate, only a small population of basal cells exhibited weak staining for the bcl-2 (+1). The results summarized on Table 2 indicate a significant elevation in the level of bcl-2 expression among both the basal epithelial and secretory cells in BPH specimens compared with their cellular counterparts in the normal prostate (t-test, P < .02). Fourteen out of the 47 prostate specimens analyzed (approximately 30%) showed heterogeneous bcl-2 staining, with approximate 50% of the areas of the section showing diffuse strong positivity (3+) among the glandular epithelial cells (Fig 1C). The epithelial staining pattern revealed specific cytoplasmic immunoreactivity, consistent with previous reports showing bd-2 protein localization to mitochondria, reticulum, and nuclear envelope membranes. In certain BPH sections, most hyperplastic prostatic acini displayed increased bcl-2 staining compared with adjacent normal epithelium (as internal control).

endoP271asmic

D e t e r m i n a t i o n o f bct-2 Expression

The results of the immunohistochemical analysis of the bcl-2 and TGF-fl immunoreactivity patterns are 2 1.8

D e t e r m i n a t i o n o f TGF-fl Expression

1.6

The expression of TGF-fl among the various prostate cell populations was examined within the context of its proposed role as a physiological regulator of apoptosis in the prostate gland. 15']s'2s Cytoplasmic-positive immunoreactivity for TGF-fll was predominantly detected in the secretory cells of the glandular epithelium, with minimal expression in the basal cells and no stroma involvement (Fig 1E). The TGF-fl immunoreactivity pattern in the normal prostate was minimal and focal, localized only to a few luminal epithelial cells and frequently not evident in many glands (Table 1). As shown in Fig 1E, immunoreactivity was predominantly in the cytoplasm of the tall columnar epithelial cells lining the prostatic lumen in BPH sections. The overall TGF-fl staining in BPH tissue was increased in cell number (fourfold) and intensity in comparison with the normal prostatic epithelium (Table 1). This significant increase (P < .001) in TGF-fl expression was exclusively

A 1.4 1.2

,,~ O.B ..J O 0.6 o. 0.4 0.2

N

BPH

EPITHELIAL

N

BPH BASAL

N

BPH

STROMA

FIGURE 3. Comparative analysis of the proliferative index of the epithelial, basal, and stroma components of normal (n = 5) and hypertrophic (n = 47) human prostate. Values represent the mean percentage of cells that exhibited positive immunostaining for the Ki-67 nuclear antigen. 672

CELL DEATH AND CELL PROLIFERATIONIN BPH (Kyprianou et al)

TABLE 2.

Relative Expression of bcl-2 and TGF-/~I Among the Different Cellular Components of Normal and BPH Prostate*

Epithelial

bcl-2

TGF-/3

Basal

Stroma

Normal

BPH

Normal

BPH

Normal

BPH

3.1 _+ 0.4 0.5 -4- 0.1

15.5 _+ 2.6t 1.81 -4- 0.3t

12.2 -+ 3.9 Negative

37.5 _+ 3.3 t Negative

Negative + Negative

3.8 _+ 0.4t Negative

Abbreviations: BPH, benign prostatic hyperplasia; TGF-fl, transforming growth factor ft. * Values represent mean (plus or minus standard error) percentage of cells (epithelial, basal, or stroma) positive for bcl-2 or TGF-/3 over the total number of cells counted per field. 1"Statistically significant difference compared with the normal ceils (P < .02). ++A value of <0.5% was used as a cut-off to define negative tissue sections for both immunoreactivity patterns.

detected in the secretory epithelial cells of BPH prostates (Table 2). Polymorphonuclear leukocytes infiltrating the stroma matrix and existing within capillaries exhibited a strong immunoreactivity for TGF-/3, consistent with recent observations. 29 These findings, however, on the lack of TGF-fl staining in the surrounding stroma, are in contrast with a previous report showing detection of strong TGF-/31 positivi~ among the stroma elements of the hyperplastic gland." 0 DISCUSSION This study is the first to examine the incidence of apoptotic cell death in situ in human BPH and to establish its significance in regulating the dynamics of prostate growth. The present findings show that in the normal prostate the apoptotic index of both the glandular epithelial and basal cells was low (1.6%), which was paralleled by a low proliferative index among the same cell populations (0.9%). The low proliferative index of the epithelial cells within the normal prostate (0.9%) shown in the present study is consistent with previous reports that documented extremely low levels of cell proliferation in normal prostatic glandular epithelial cells. 5-8 '31 These low apoptotic and proliferative activities result in a steady-state, self-renewing condition in which there is no net growth of the normal gland. Although apoptosis was histologically conspicuous in the prostatic tissue sections examined, one has to consider that this mode of cell death is an active and rapid process, and a low apoptotic index detectable in normal tissue usually translates to a high rate of cell death. For instance, in the normal liver an apoptotic index of 2% results in an approximately 25% loss of cells per day. n The stoma component of both the normal and BPH prostatic sections were consistently negative for apoptotic DNA fragmentation, indicating the extremely low frequency of apoptotic cell death within the stroma element of the prostate. These observations gained support from previous studies, suggesting that the average life span of prostatic stromal cells is approximately 30 years, whereas the life span of epithelial cells is about only 2 years. 9

In the normal prostate, weak bcl-2 staining was detected in the glandular cells of the prostate epithelium, but the immunoreactivity was clearly enhanced among 673

all the basal cells. These findings are in agreement with the early observations defining the basal cell compon e n t of the prostate as the predominant topological localization of bcl-2 expression. 2°'21 The dramatic increase (sixfold) in the cell number and intensity of bcL 2 expression among the prostatic epithelial cells (both secretory luminal and basal cell populations), shown in the present study, suggests that an abnormal increase in the expression of this antideath protein might be potentially involved in the growth deregulation of the prostate epithelium. Recent studies have shown that the prostate secretory epithelial ceils as well as the basal cells of the normal prostate, express high levels of bclX, a bcl-2-related protein, implicated in blocking/regulating apoptosis in different stages of cell differentiation than bcl-2. 32 Because low bcl-2 immunoreactivity was detected in the normal prostatic epithelium, it might be that bcl-X, rather than bcl-2, is the primary regulator of prostate apoptosis in the normal gland. The exclusive localization of TGF-fl expression in the secretory epithelium of benign glands and absence of expression among the prostatic stroma component is in full accord with a recent study showing that intense immunoreactivity against the native TGF-fll was only detected in certain epithelial cells in BPH. 29 It seriously challenges, however, another immunohistochemical that documented TGF-/31 expression among the epithelial as well as the stroma components of the hyperplastic prostate. 3° This discrepancy could be caused by differences in the methodology that result in potential masking of the epitope. Activation of the latent form of TGF/31 can be achieved in vitro by heat, proteolytic enzymes, extreme pH changes, sodium dodecyl sulfate (SDS) denaturation, or a high urea concentration. In the present study, all prostatic tissue sections analyzed for TGF-fll immunoreactivity were treated before the addition of the primary antibody by hyaluronidase to eliminate the possibility of hidden epitopes. In view of the present findings and considering the previously defined role for TGF-fl as a critical regulator of prostate apoptosis, 1"~'18one could speculate on a potential involvement of TGF-/3 in promoting the apoptotic program of cell death in the epithelial ceils of the hyperplastic prostate. The concept gains solid support from in vitro studies implicating TGF-/3 as a negative growth factor with potent cell-killing function specifically targeted at h u m a n benign prostatic epithelial

HUMAN PATHOLOGY

Volume 27, No. 7 (July 1996)

cells. 28Moreover, although the prostatic basal epithelial cells (in contrast to the secretory cells) are resistant to apoptosis induced by androgen deprivation, 13 their apoptotic death might be differentially activated by TGF-/3. Because in the prostate TGF431 operates through a signaling system that can override the mitogenic effects of androgens (18), disruption of this system by androgen-independent mechanisms, such as bcl2 overexpression, would lead to escape from apoptosis, unconstrained cell proliferation, and development of prostate hypertrophy. Comparative determination of the apoptotic and the proliferative activities within the normal and BPH prostate revealed a substantial reduction (fourfold) in the total number of epithelial cells (both secretory and basal) that are dying via apoptosis in the hypertrophic gland. It is tempting to suggest that this imbalance in favor of cell proliferation would ultimately promote prostatic epithelial hyperplasia. In view of the present observations, a hypothesis could be advanced on the potential deregulation of the apoptotic cell death machinery in the hyperplastic prostate. In normal prostate epithelium, the low rate of cell proliferation is balanced by a comparably low rate of cell death. When the proliferative index among the secretory epithelial cells is gradually driven to higher levels, the cells start to induce the expression of endogenous TGF-/3, the physiological regulator of cell death, to neutralize the enhanced proliferative activity of the prostate. However, resistance emerges from a molecular brake, bc~2, a powerful suppressor of apoptosis, which is now being expressed in excess by both the basal and secretory epithelial cells (in the hyperplastic gland). The normal apoptotic control mechanism can no longer operate, and the resulting population of uncontrollably proliferating epithelial cells, driven as they might to die by TGF-/3, are blocked from entering the apoptotic suicidal pathway by the excessive overproduction of bcl-2. Once the prostate epithelial cells become refractive to TGF-/3's apoptotic effects, this molecular mechanism allows them to be long-lived and protected from selfdestruction, ultimately resulting in the aberrant prostatic growth that resembles the "embryonic" reawakening characterizing BPHJ In support of this hypothesis is recent evidence showing that senescent human fibroblasts are resistant to apoptosis, as a result of their ability to maintain high levels of bc~2 in the prolonged absence of serum in vitro. 3~ Although resistance to apoptosis observed among the aged prostatic epithelial cell populations in BPH may be related to their ability to express remarkably high levels of bcl-2 (in the continuous presence of TGF-fl), one has also to consider that other novel external factors, such as the presence or absence of stimuli in the cell microenvironment, may play a critical role in ushering the prostate cells toward survival or suicidal fate. In conclusion, the results described in this study suggest that disruption of the molecular mechanisms that regulate both processes in the prostatic growth equilibrium (ie, cell proliferation and cell death) may promote development of BPH. Shifting the equilibrium m favor of cell proliferation may activate the molecular

674

cascade of events, such as TGF-/3 overproduction, that would dictate apoptotic death to restore the growth balance. The enhanced presence of bcg2, however, strongly counteracts this trend, and the prostate cells inhabiting the hypertrophic areas of the aging gland are unable to be eliminated by the apoptotic death program. BPH specimens from an expanded patient population are currently being analyzed to correlate the apoptotic index, proliferative index, and bcl-2 expression in the hyperplastic prostate with the age of the patients. Results from these studies will enable us temporally to define the turning point marking the onset of disruption of the growth regulatory mechanisms in the aging prostate gland.

Acknowledgment. The authors are grateful to Dr Wolfgang Mergner for expert advice on the pathological evaluation of the tissue sections and to Dinnise Felder for excellent secretarial assistance in the preparation of the manuscript.

REFERENCES 1. Walsh PC: Benign prostatic hyperplasia: Etiological considerations, in Kimball FA, Buhl AE, Carter DE (eds): Approaches to the Study of Benign Prostatic Hyperplasia. New York, NY, Liss, 1984, pp 10-25 2. McDonnell JD: Medical management of BPH with androgen suppression. The Prostate 3:49-59, 1990 (suppl) 3. Berry SJ, Coffey DS, Walsh PC, et al: The development of human benign prostatic hyperplasia with age. J Urol 132:474-479, 1984 4. Isaacs JT: Control of cell proliferation and cell death in the normal and neoplastic prostate: A stem cell model, in Rodgers CH, Coffey DS, Cunha G, (eds) : Benign Prostatic Hyperplasia, vol 2 (Publication No 8%2881). Bethesda, MD, NIH 1985, pp 85-94 5. Meyers JA, Sufrin G, Maring SA: Proliferation activity of benign human prostate, prostatic adenocarcinoma and seminal vesicle evaluated by thymidine labeling. J Urol 128:1353-1356, 1982 6. Claus S, Wrenger M, Senge T, et al: Immunohistochemical determination of age related proliferation rates in normal and benign hyperplastic human prostates. Urol Res 21:305-308, 1993 7. Gallee MPW, Visser-De Jong E, Ten Kate FJW, et al: Monoclonal antibody Ki-67 defined growth fraction in benign prostatic hyperplasia and prostate cancer. J Urol 142:1342-1346, 1989 8. Sakr WA, Sarkar FH, Sreepaths P, et al: Measurement of cellular proliferation in human prostate by Agnor, PCNA and SPF. The Prostate 22:14%154, 1993 9. Tunn S, Nass R, Ekkernkamp S, et al: Evaluation of average life span of epithelial and stromal cells of human prostate by superoxide dismutase activity. The Prostate 15:263-271, 1989 10. Raft MC: Social controls on cell survival and cell death. Nature 256:396-400, 1992 11. Arrends MJ, Wyllie AH: Apoptosis: Mechanisms and roles in pathology. Int Rev Exp Pathol 32:233-250, 1991 12. Kyprianou N, IsaacsJT: Activation of programmed cell death in the rat ventral prostate after castration. Endocrinology 122:552562, 1988 13. English HF, Kyprianou N, Isaacs JT: Relationship between DNA fragmentation and apoptosis in the programmed cell death in the rat prostate following castration. The Prostate 15:233-251, 1989 14. Massague J: The transforming growth factor-/3 family. Annu Rev Cell Biol 6:597-641, 1990 15. Kyprianou N, IsaacsJT: Expression of transforming growth factor-/3 in rat ventral prostate during castration-induced programmed cell death. Mol Endocrinol 3:1515-1522, 1989 16. Rotello RJ, Lieberman R, Purchio AF, et al: Coordinated regulation of apoptosis and cell proliferation by transforming factor /31 in cultured uterine epithelial cells. Proc Natl Acad Sci U S A 88:3412-3415, 1991 17. Yanagihara K, Tsumuraya M: Transforming growth factor/3

CELL DEATH AND CELL PROLIFERATION IN BPH (Kyprianou et al) induces apoptotic cell death in cultured human gastric carcinoma cells. Cancer Res 52:4042-4045, 1992 18. Martikainen P, Kyprianou N, Isaacs JT: Effect of transforming growth factor fl~ on proliferation and death of rat prostatic cells. Endocrinology 127:2963-2968, 1990 19. Reed JC: BCL-2 and the regulation of programmed cell death. J Cell Biol 1214:1-6, 1994 20. Hockenherry DM, Zutler M, Hickey N, et al: bcl-2 protein is topographically restricted in tissue characterized by apoptotic cell death. Proc Natl Acad Sci U S A 88:6961-6965, 1991 21. McDonnell T, Troncoso P, Brisbay SM, et al: Expression of the protooncogene bcl-2 in the prostate and its association with emergence of androgen-independent prostate cancer. Cancer Res 52:6940-6944, 1992 22. Brronner MB, Culin C, ReedJC, et al: The bcl-2 protooncogene and the gastrointestinal epithelial tumor progression model. A m J Pathol 146:20-26, 1995 23. Tu H, Jacobs SC, Kyprianou N: Significance of apoptosis in prostate cancer progression: Relationship with cell proliferation, TGF-flI and bcl-2 expression. I n t J Cancer 1996 (in press) 24. Gavrieli Y, Sherman Y, Ben-Sasson SA: Identification of programmed cell death in situ via specific labeling of nuclear DNA fragmentation. J Cell Biol 119:493-501, 1992 25. Wijsman JH, Jonker RR, Keiger R, et al: A new method to detect apoptosis in paraffin sections: In situ end labeling of fragmented DNA. J Histochem 41:7-12, 1993

675

26. Cattoretti G, Becket MH, Key G, et al: Monoclonal antibodies against recombinant parts of the Ki-67 antigen (MIB1) and (MIB3) detect proliferation cells in microwave processed formalin fixed paraffin sections. J Pathol 168:357-363, 1992 27. Krajewski S, Tanaka S, Takayama S, et al: Investigations of the subcellnlar distriution of the bcl-2 oncoproteiu: Residence in the nuclear envelope, endoplasmic reticulum and outer mitochoudrial membranes. Cancer Res 53:4701-4714, 1993 28. Sutkowski DM, Fong C-J, Sensibar JA, et al: Interaction of epidermal growth factor and transforming growth factor fl in human prostatic epithelial cells in culture. The Prostate 2l:133-143, 1992 29. Glynne-Jones E, Harper ME, Goddard L, et al: Transforming growth factor beta 1 expression in benign and malignant prostatic tumors. The Prostate 25:210-218, 1994 30. Eklov S, Funa K, Nordgen H, et al: Lack of the latent transforming growth factor fl binding protein in a malignant but not benign prostatic tissue. Cancer Res 53:3193-3197, 1993 31. Berges RR, Vukanovic J, EpsteinJI, et al: Implication of cell kinetic changes during the progression of human prostate cancer. Clinical Cancer Res 1:473-480, 1995 32. Krajewski S, Krajewska M, Shabaik M, et al: Immunohistochemical analysis of in vivo patterns of bcl-X expression. Cancer Res 54:5501-5507, 1994 33. Wang E: Senescent human fibroblasts resist programmed cell death and failure to suppress bcl-2 is involved. Cancer Res 55:2284-2292, 1995