Expression of bcl-2 and the Progression of Human and ... · of prostatic cancer was determined using immunocytochemical analysis in a series of human and rodent prostatic cancers

Vol. 2, 389-398, February. 1996 Clinical Cancer Research 389

Expression of bcl-2 and the Progression of Human and Rodent Prostatic Cancers 1

Yuzo Furuya, Stanislaw Krajewski, Jonathan I. Epstein, John C. Reed, and John T. I s a a c s 2

Johns Hopkins Oncology Center [Y. F., J. T. I.] and Department of Urology [J. I. E., J. T. I.], Johns Hopkins University School of Medicine, Baltimore, Maryland 21205, and La Jolla Cancer Research Foundation, Oncogene and Tumor Suppressor Gene Program, La Jolla, California [S. K., J. C. R.]

A B S T R A C T

The frequency of bcl-2 protein expression was evalu- ated using immunocytochemical staining during the progression of human and rat prostate cancer from an androgen-sensitive nonmetastatic to an androgen-independent metastatic phenotype. Previous studies (A. S. Shabaik et al., J. Urol. Pathol., 3: 17-27, 1995) demonstrated that 0 of 20 high-grade prostatic intraepithelial neoplasias and only 3 (7%) of 41 pathologically localized stage B human prostatic cancers had detectable bcl-2 staining. In the present study, 5 (17%) of 30 lymph node metastases from pathologically disseminated O 1 disease and 14 (52 %) of 27 bone metastases from pathologically disseminated D 2 disease expressed detectable bcl-2 protein. These data demonstrate that there is a statistically significant (P < 0.05) association between expression of bcl-2 and the progression of human prostatic cancer cells to a metastatic phenotype. Such bcl-2 expression is not absolutely required, however, for either androgen independence or metastatic ability by human prostatic cancer cells. Likewise, within a series of eight distinct Dunning R3327 rat prostatic cancer sublines, which differ widely in their progressional state, there is also a significant association (P < 0.05) between bcl-2 expression and progression (four of six androgen-independent rat sublines expressed bcl-2 protein). Again in this rodent system, bcl-2 expression is not an absolute requirement for either androgen independence or metastatic ability. For example, the androgen- independent highly metastatic Dunning AT-3 subline, while expressing bax protein, does not express bcl-2 protein. If such AT-3 cells are genetically engineered to express bcl-2, these expressing cells are now cross-resistant to a variety of mechanistically diverse noxious insults (e.g., viral infection or exposure to antimetabolites, alkylating agents, or agents which elevate the intracellular free Ca2+). The ability of bcl-2 to inhibit the programmed death of AT-3 cells induced by these agents involves a late step in the death process, since

the early induction of expression of a series of genes associated with apoptosis is not impaired by bcl-2 expression. These data demonstrate that the development of androgen independence and/or metastatic ability can be associated with the expression of bcl-2 protein but that bcl-2-independent mechanisms also exist for such progression.

INTRODUCTION Following androgen ablation, androgen-dependent normal

and malignant prostatic cancer cells stop entering the cell cycle (i.e., cells remain out of the cycle in Go), and instead these G O cells undergo an energy-dependent process of cellular suicide termed programmed cell death or apoptosis (1-5). Programmed death of nonproliferating prostatic cells involves the induced expression of specific genes resulting in a change in the profile of proteins synthesized (2, 5, 7). This epigenetic shift in protein expression leads to a disruption in the regulation of the intracellular pools of the calcium (8, 9). As a result of this disruption, nuclear endonucleases are activated, resulting in the double- strand fragmentation of the cells' genomic DNA, irreversibly committing these cells to undergo cellular fragmentation into apoptotic bodies (8, 9).

Due to its coupled ability to inhibit proliferation and activate programmed death, androgen ablation eliminates the androgen-dependent prostatic cancer cells present within patients with metastatic prostatic cancer. Although 80-90% of patients with metastatic disease initially respond to androgen ablation, essentially all progress to an androgen-independent state due to the continuing growth of androgen-independent prostatic cancer cells heterogeneously present within the patient (10). This is because androgen ablation does not activate the programmed death of the androgen-independent prostatic cancer cells within the patient because of a defect in the initiation step (11, 12). Thus, androgen ablation is not curative, no matter how complete (13). Even with this defect, however, these androgen-independent cells retain the ability to undergo programmed cell death if initiated by nonandrogen ablation methods (11, 12, 14, 15).

For example, the androgen-independent, highly metastatic Dunning AT-3 rat prostatic cancer cells have been treated in vitro with a variety of nonandrogen-ablative agents that inhibit thymidylate synthetase and thus induce "thymineless death" of the cells (e.g., cells treated with 5-FdUrd 3 or trifluorothymidine; Refs. 11 and 14). This thymineless death requires cells to be proliferating, and the cells die during progression through the S-phase (16, 17). Additional studies have demonstrated that programmed death of the androgen-independent AT-3 prostatic

Received 7/17/95; revised 10/13/95; accepted 10/17/95. 1 Supported by NIH Grant CA58236 and by Grant DHP-32C from the American Cancer Society. J. C. R is a Scholar of the Leukemia Society of America. 2 To whom requests for reprints should be addressed, at Johns Hopkins Oncology Center, 422 North Bond Street, Baltimore, MD 21231.

3 The abbreviations used are: 5-FdUrd, 5-fluorodeoxyuridine; 4HC, 4-hydroperoxycyclophosphamide; PIN, prostatic intraepithelial neoplasia; TG, thapsigargin; ECso, 50% effective dose; GRP-78, glucose- regulated M r 78,000 protein; TTG, tissue transglutaminase.

Cancer Research. on September 21, 2020. © 1996 American Association forclincancerres.aacrjournals.org Downloaded from

http://clincancerres.aacrjournals.org/

390 bcl-2 and Prostatic Cancer

cancer cells also can be induced in a manner which, unlike thymineless death, does not require entrance and progression through the cell cycle (i.e., death occurs out of the cycle in Go). Such cell proliferation-independent programmed death is induced in androgen-independent AT-3 prostatic cancer cells if intracellular free calcium is elevated 3-4-fold for >12 h by exposure to either the Ca 2+ ionophore, ionomycin (12), or the sesquiterpene "y-lactone, TG (14, 15, 18).

The bcl-2 gene is located on chromosome 18q21 and encodes a membrane-bound M r 26,000 protein that resides pri- marily in the outer mitochondrial membrane, nuclear envelope, and endoplasmic reticulum (19-21). bcl-2 has been demonstrated to be an oncogene because its induced overexpression can lead to malignant transformation (22-23). bcl-2 is unique among the oncogenes, however, in that its expression does not enhance the rate of cell proliferation but instead decreases the rate of programmed cell death (24, 25). The mechanism for this inhibition of programmed cell death is believed to be via bcl-2' s ability to heterodimerize with a series of "death" proteins. One such death protein is termed bcl-2-associated X protein or bax (26, 27). When bax is allowed to homodimerize within the cell, the rate of programmed cell death is enhanced (26, 27). bcl-2's antiapoptotic effect may thus be via inhibition of such homodi- meration through heterodimeration with bax (26, 27). bcl-2 protein expression decreases the rate of programmed cell death by malignant hematopoietic, lymphopoietic, and neuroblastoma cells induced by a wide variety of chemotherapeutic drugs (28-3 I). Enhanced expression of the bcl-2 protein likewise can inhibit programmed cell death of epithelial-derived cells. For example, Levine et al. (32) demonstrated that overexpression of the bcl-2 protein induced by DNA transfection of an exogenous bcl-2 gene in the Dunning AT-3 rat prostatic cancer cells pre- vents this programmed death initiated by lyric infection with the RNA Sindbis virus.

These combined results suggest that a critical level of bcl-2 protein expression allows prostatic cancer cells to become more resistant to programmed cell death induced by a wide variety of noxious signals. Thus, enhanced expression of bcl-2 protein could be causally involved in the progression of prostatic cancer from an androgen-dependent nonmetastatic to an androgen- independent metastatic state. Indeed, a series of previous studies have demonstrated that bcl-2 expression is associated with progression of human prostatic cancer to an androgen-independent state, and that this effect may be due to bcl-2's inhibition of programmed cell death (33-34). To further test such a possibil- ity, the frequency of expression of bcl-2 during the progression of prostatic cancer was determined using immunocytochemical analysis in a series of human and rodent prostatic cancers. In addition, to evaluate the spectrum of cross-resistance induced by bcl-2 expression, androgen-independent AT-3 rat prostatic cancer bcl-2 and control transfectants generated in our earlier study (32) were tested for their response to a series of mechanistically diverse, nonandrogen-ablative cytotoxic agents.

MATERIALS AND METHODS Immunocytochemical Determination of bcl-2 Expres-

sion. Expression of bcl-2 protein by human and rodent prostatic cancer cells was determined by immunocytochemical stain-

ing of histological sections using procedures and criteria described in detail previously (35). For human tissue, rabbit polyclonal antiserum specific for human bcl-2 was used as described previously (35). For rodent tissue, rabbit polyclonal antiserum specific for murine bcl-2 was used as described previously (36). For detection of expression of bax in rodent prostatic cancer cells, rabbit polyclonal antiserum specific for murine bax was used as described previously (36, 37). For human studies, pelvic lymph nodes containing metastatic prostatic cancer cells from 30 previously untreated patients with pathological D~ metastatic prostatic cancer and bone metastases obtained from a series of 15 untreated patients with pathological D 2 metastatic disease and 12 D 2 patients failing androgen ablation therapy were analyzed. For rodent studies, analysis was performed on prostatic cancer tissue obtained from Copenhagen rats inoculated with one of a series of eight well-characterized sublines of the Dunning R-3327 system of serially transplantable rat prostatic cancers. These sublines were all derived from the original slow-growing, morphologically well-differentiated, androgen- responsive nonmetastatic parental R-3327 tumor via clonal selection and/or genetic instability (37). Details of serial passage in syngeneic Copenhagen male rats and the methods for determination of androgen responsiveness, growth rate, and metastatic ability are as described previously (38).

In Vitro Studies. The androgen-independent, highly metastatic Dunning R-3327 AT-3 rat prostatic cancer cell line has been characterized previously and is maintained in RPMI 1640 (M. A. Bioproducts, Walkerville, MD) containing 10% FCS (HyClone, Logan, UT), 250 nM dexamethasone (Sigma Chem- ical Co., St. Louis, MO), 1 n ~ glutamine, 100 IU/ml potassium penicillin G, and 100 IU/ml streptomycin sulfate (M. A. Bio- products) equilibrated with 5% C02-95% air at 37°C (39). The bcl-2 and neomycin control AT-3 transfectants used were those described previously by Levine et al. (32). To induce proliferation-dependent programmed death, exponentially proliferating cells were exposed to either the thymidylate synthetase inhibitor 5-FdUrd obtained from Sigma Chemical Co. or the alkylating agent 4HC [i.e., an alkylating derivative of cyclophosphamide produced synthetically (40), generously provided by Dr. John Hilton, Johns Hopkins Oncology Center]. To induce the proliferation-independent programmed death, the cells were exposed to either the calcium ionophore, ionomycin (Calbiochem Corp., La Jolla, CA), or plant sesquiterpene ~/-lactone, TG (LC Service Corp., Woburn, MA). The kinetics of morphological changes following treatments were determined using time-lapse videomicroscopy as described previously (12). The kinetics of morphological changes in the cell nucleus following treatments were determined by fixing cells in 70% methanol and then staining the cells with 1 Ixg/ml Hoescht 33342 stain (Calbio- chem) and examining them using epifluorescence illumination. Determination of the number of viable cells at various times of treatment was performed using trypan blue exclusion as described previously (12). Cell survival following treatment was also assessed based on clonogenic assays performed as described previously (12). Cell cycle distribution was determined by bivariant analysis using flow cytometry as described previously (15).

DNA Fragmentation Analysis. Double-strand DNA fragmentation was quantitated according to a modification of



Clinical Cancer Research 391

the pulse field gel electrophoresis method of Stamato and Denko (41) as described in detail previously (14, 15). Qualitative analysis of DNA fragmentation was performed using ethidium bromide staining of 1.5% agarose gels as described previously (14, 15).

RNA Isolation and Northern Blot Analysis. Total RNA was prepared using the acid guanidium thiocyanate-phe- nol-chloroform method as described previously (7). The North- ern blot analysis and probes used are described in detail previously (7, 14).

hcl-2 Western Blot Analysis. Immunoprecipitations and immunoblotting were performed as described by Reed et al. (42) using polyclonal rabbit anti-bcl-2 peptide antibody developed by Reed et al. (42).

Statistical Analysis. Values are expressed as the mean _+ SE. Statistical analyses of significance for differences were made by a one-way ANOVA with the Newman-Keuls test for multiple comparison with P < 0.05 defined as significant. Statistical analysis of the significance for correlation between the bcl-2 staining and histological Gleason grades or pathological stage was according to X 2, with P < 0.05 defined as significant. Unless otherwise stated, each experiment was performed at least three independent times in triplicate for each data point.

R E S U L T S

bcl-2 Expression during the Progression of Human Prostatic Cancers. High-grade PIN is believed to be a major premalignant lesion within the prostate and that progression of this lesion results in the development of localized prostatic cancer (43). Using the same immunochemical reagents, tech- niques, and criteria as in the present study, Shabaik et al. (35) demonstrated that the frequency of bcl-2 protein expression in high-grade PINs and primary prostatic cancers in hormonally untreated patients with pathologically localized stage B disease is very low [0 of 20 PINs and only 3 of 41 (7%) primary prostatic cancer had positive bcl-2 staining immunocytochemically]. In the present study, 5 (17%) of 30 lymph node metastases from hormonally untreated patients with pathologically disseminated stage D 1 disease had positive bcl-2 staining whereas 14 (52%) of 27 bone metastases from patients with pathologically disseminated D 2 disease were bcl-2 positive (see Fig. 1, a - f for representative examples of immunocytochemical staining). Using X 2 analysis, the frequencies of bcl-2 expression in both lymph node and bone metastases were statistically different (P < 0.05) from both PIN and prostatic cancer in pathologically localized stage B patients. In contrast, there was no statistically significant correlation between the histological Gleason grade and bcl-2 expression.

Within the pathologically disseminated D 2 group, 15 patients were hormonally untreated and 12 patients had failed androgen ablation therapy. There was no difference in the frequency of bcl-2 positivity in the bone metastases for these two subsets of patient [bone metastases from 8 (53%) of 15 untreated and 5 (42%) of 12 patients failing hormonal therapy were positive for bcl-2 staining]. These results demonstrate that bcl-2 expression is not an absolute requirement for becoming either metastatic or androgen independent; however, it is statis-

tically (P < 0.05) associated with progression by human prostatic cancer cells to such phenotypes.

bcl-2 Expression during the Progression of Rodent Prostatic Cancers. A series of transplantable Dunning R-3327 rat prostatic cancer sublines were harvested from tumor- bearing rats and immunochemically stained for bcl-2 protein expression (Table 1). This Dunning system of prostatic cancers was chosen because the sublines differ widely in their progression state based on the criteria of in vivo growth rate, androgen responsiveness, and metastatic ability (Table 1). All eight sublines were derived from the slowing-growing, androgen-responsive nonmetastatic parental R-3327 prostatic cancer via genetic instability coupled with clonal selection (38). The cancer cells in the least advanced (slowest growing, androgen-responsive, nonmetastatic) H subline do not have detectable bcl-2 expression (Table 1). The H tumor is androgen responsive, being heterogeneously composed of ~80-90% androgen-dependent and 10-20% independent prostatic cancer cells (44). If male rats bearing such heterogeneous H tumors are castrated, the tumor stops its net growth for 1 to 2 months, during which time the androgen-dependent cells are eliminated (44). After this responsive period, the tumors eventually relapse due to the continued growth of the preexisting androgen-independent cancer cells (44). Such relapsing tumors are androgen-independent, serially passageable in animals, and termed the HIS subline (38, 44). Although androgen independent, the cancer cells in the HIS subline do not have detectable bcl-2 protein expression (Table 1).

The G subline, like the H, is androgen responsive and does not have detectable bcl-2 expression (Table 1 and Fig. lg). The G subline grows more rapidly than the H (Table 1) and is composed of androgen-sensitive but not androgen-dependent prostatic cancer cells (45). When male rats bearing androgen- responsive G cancers are castrated, the growth rate of the cancer decreases (volume doubling time increases from 6 _+ 1 to 11 _+ 3 days) but does not stop (45). This is due to an increase in the death rate of G cancer cells induced by castration, with no decrease in the rate of cell proliferation (45). Even with such an increase in the death rate, however, the rate of G cell proliferation is still greater than death and thus net tumor growth slows but does not stop. Slower growing G cancers were harvested from castrated hosts and stained for bcl-2 expression. These analyses demonstrated that following castration, the G cancer cells have 2+/3+ staining for the bcl-2 protein (Fig. lh). The ability of the G cells to up-regulate bcl-2 protein expression following castration could explain why the G cells are only androgen sensitive and not androgen dependent.

Combining the data from all of the sublines, detectable expression of bcl-2 protein is associated with the progression to an androgen-independent phenotype in 4 (67%) of 6 of the androgen-independent sublines (Table 1 and Fig. 1i). This association is statistically significant (P < 0.05) by X 2 analysis. In contrast, within these eight sublines, there is no consistent relationship between in vivo tumor growth rates and bcl-2 expression.

Characteristics of AT-3 Parental Cells and neo and bel-2 Transfeetants. The combined human and rodent results demonstrate that progression to either an androgen-independent and/or highly metastatic phenotype is often but not always







Table 1 Relationship between bcl-2 expression and in vivo growth rate, androgen sensitivity, and metastatic ability within the Dunning

series of serially transplantable rat prostatic cancers

Growth rate Tumor (volume doubling Androgen Metastatic bcl-2

sublines time in days) responsiveness ability ~ Expression b

H 20 _+ 4 AR ~ Low - - HIS 20 ± 2 AI Low - - G 6 _+ 1 AR Low - - AT-1 3.0 -+ 0.5 AI Low 1+ AT-2 2.8 + 0.4 AI Low 3+ AT-3 1.9 + 0.3 AI High - - MAT-Lu 3.5 ± 0.6 AI High 1 + MAT-Ly-Lu 1.7 ± 0.2 AI High 2+/3+

%ow, < 10% of rats bearing SQ tumors develop distant metastasis before death, high, >90% of rats bearing SQ tumors develop 20-100 lung metastases before death.

b , no detectable specific staining; 1 +, low but detectable specific staining in majority of cells staining, 2+, moderate specific staining in majority of cells; 3+, high specific staining in majority of ceils.

CAR, androgen responsive as defined by inhibition in net tumor growth rate in tumor-bearing rats following castration; AI, androgen independent as defined by no change in net tumor growth rate in tumor-beating rats following castration.

associated with bcl-2 expression. For example, the AT-3 rat prostatic cancer cells do not have detectable levels of bcl-2

protein detectable by either immunocytochemically or immuno- precipitation analysis (32), and yet these cells are both androgen

independent and highly metastatic. Additional immunocyto-

chemical studies demonstrated that AT-3 cells growing as tu-

mors in animals express detectable levels of the bax protein.

Thus, the response of the AT-3 cells genetically engineered via

DNA transfection to express bcl-2 was tested. Multiple bcl-2/ neo and neo control transfectants were isolated and screened using Northern and Western blotting for their level of expres-

sion of the human bcl-2 gene. Several bcl-2/neo-resistant trans-

fectants expressing high levels of the human bcl-2 mRNA and protein were identified (32). In contrast, the parental untransfected AT-3 cells and all neo-only control transfectants had undetectable levels of either bcl-2 mRNA or protein. In vitro

growth kinetic analysis demonstrated that the doubling times (16.8 + 0.2 h) and clonogenic abilities (41 +__ 6% of cells producing clonal growth) of the high bcl-2-expressing and neo control transfectants were unchanged relative to the untransfected parental AT-3 cells. Likewise, there were no obvious morphological differences between the untransfected parental

and clonally derived AT-3 transfectants.

Table 2 In vitro cell cycle distribution of parental AT-3 cells 24 h after initiation of treatment with various agents

% AT-3 cells in various cell cycle phases 24-h Treatment

(n = 3) GoG 1 S Gz/mitosis

None (control) 58 --- 11 28 _ 6 14 ± 4 I[~M 5-FdUrd 24 ± 7 a 64 ± 9 a 12 ___ 4 5 ~M 4HC 20 ___ 3 ~ 25 ~ 4 a 55 ± 7 a 10 ~M ionomycin 85 ± 5 a 7 ± 2 a 8 ,,, 3 a 500 nM TG 90 ± 7 a 4 ___ 1 a 6 ,,, 2 ~

ap < 0.05 compared to value in control group.

Effect of bcl-2 Protein Expression on Proliferation-dependent and -independent Programmed Death of AT-3 Cells. 5-FdUrd (1 pLM) inhibits thymidytate synthetase, producing a thymineless state that results in an increase in the

percentage of parental AT-3 cells in the S-phase of proliferation (Table 2). During this delayed S-phase progression, a futile

DNA repair process is induced secondary to the misincorpora- tion of uridine into DNA which activates the programmed cell death pathway (17). DNA damage also can be directly induced by pulse exposure to DNA-alkylating agents. For example, the

synthetic derivative of cyclophosphamide, 4HC, can directly

alkylate the DNA without any requirement for enzymatic acti-

vation (40). Thus, when cells are exposed to effective doses of 4HC (i.e., 5 p~M), their DNA is rapidly damaged [the chemical

half-life for 4HC as an alkylating agent at pH 7.4 is -~120 min

(40)]. Acutely exposed cells accumulate in G 2 (Table 2). Such a

G 2 delay is due to activation of DNA repair pathways (30).

Eventually, the cells leave G2, undergo mitosis, and begin a new cycle of proliferation, as demonstrated by the observation that the percentage of AT-3 cells in S remains unchanged during the first 24 h after acute 4HC exposure (Table 2). If sufficient DNA

damage is not repaired during the initial G 2 delay, the cells

undergo programmed death during the next one to two cell cycle

progressions as determined by time lapse videomicroscopy. Thus, the programmed death of AT-3 cells induced by both 5-FdUrd and 4HC is cell proliferation dependent.

In contrast, within 24 h of continuous exposure to either 10 IXM ionomycin or 500 nM TG, AT-3 cells leave the proliferative cell cycle (>85% of cells arrest in Go; Table 2). That these cells are in G O and not G 1 has been previously demonstrated by the fact that there is more than a 95% reduction in the mRNA expression for the G1 cyclins (i.e., cyclins D, C, and E) as well as for thymidine kinase within 24 h of drug treatment (15). Once

Fig. 1 (opposite page) a, human lymph node with infiltrating bcl-2-negative prostate carcinoma metastasis. The tumor cells destroyed the regular histoarchitecture of the lymph node and partially replaced the lymphoid cells in the germinal center presented in the middle of the photomicrograph. X250. b, higher magnification from the same case showing bcl-2-negative tumor cells and 4+ bcl-2 immunoreactivity in normal lymphocytes within the lymph node (internal bcl-2-positive control). X400. c, example of moderately positive prostatic cancer cells in the pelvic lymph node metastasis presenting 2+ and 4+ bcl-2 immunoreactivity in tumor cells and lymphocytes, respectively, x250. d, same as c except at x400. e, strong bcl-2 protein expression in the cytoplasma and in the nucleus of carcinoma cells of bone marrow metastasis within a patient failing androgen ablation therapy. X400. f, focus of squamous cell metaplasia in another case of prostatic cancer within a bone metastasis in a patient treated with estrogens. The tumor cells revealed moderate bcl-2 expression; however, the highest expression was observed in the focus of squamous epithelial cell metaplasia. X250. g, negative bcl-2 protein expression in Dunning R-3327 G rat rodent prostatic cancer subline growing in an intact animal, x400. h, bcl-2 overexpression in the same G subline growing in a castrated host. X400. i, Dunning R-3327 AT-2 rat prostatic cancer growing in an intact host with 3 + bcl-2 immunoreactivity, x400.



O 106

J~

"6 ,- 105

n

E - I

Z

1 0 7

104

5-FrdU lOO

" " '- 75 ' 1DO S u

, * - 0

---® 25

o

5-FrdU

214 4'8 7~2 916 120

Time (hrs)

o 106 m J~ t~ .m

:>

"6 ~, 10S

JQ E

104

i

4 H e

1 0 7

o 2'4 4'8 7'2 9'6 Time (hrs)


o 2'4 4'8 ;2 Time (hrs)

AT3neo

4HC 100

80

z~

40 0 214 4'8 7'2 916

Time (hrs)

AT3bcl2 c lone3 ~ : ) - - AT3bcl2 c lone4

- - z~-- AT3neo(no treatment)

Fig. 2 Kinetics of cytotoxicity (upper panels) and DNA fragmentation (lower panels) in AT-3 neo-only control transfectant cells versus bcl-2-expressing transfectant cells treated with 1 ~M of 5-FdUrd (left side) or 5 p~M 4HC (right side). Fragmented DNA is -< 1 megabase.

in Go, these cells subsequently undergo DNA fragmentation and programmed cell death (15). Both of these treatments thus induce cell proliferation-independent programmed death of AT-3 cells.

To determine whether bcl-2 expression can affect either the proliferation-dependent or -independent programmed death of AT-3 cells, bcl-2-expressing transfectants, neo control transfectants, and parental AT-3 cells were exposed to either (a) 1 p~M 5-FdUrd; (b) 5 p~M 4HC; (c) 10 p~M ionomycin; or (d) 500 nM TG. The cell cycle distribution, degree of DNA fragmentation, number of viable cells, and cellular morphology determined over a 5-day period. These data demonstrated that the neo control transfectants and the untransfected parental AT-3 cells have an identical cytotoxic response to all four treatments in terms of their temporal increase in DNA fragmentation and decrease in viable cell number (data not shown). Qualitative analysis of the pattern of DNA fragmentation at 24 h after drug exposure revealed a nucleosomal repeat size DNA ladder, characteristic of programmed cell death induced by all four agents. Likewise, characteristics of programmed cell death was the temporal pattern of morphological changes in both AT-3 parental and neo control transfectants determined using time lapse videomicroscopy. By 12-24 h of treatment, all cells are smaller and more rounded. After 24 h, the cells undergo periods of plasma membrane hyperactivity characterized by plasma mere-

brane blebbing and eventually nuclear fragmentation occurs with the production of apoptotic bodies detectable by staining cells with the fluorescent DNA-binding dye Hoescht 33342. In addition, similar perturbation in cell cycle distribution was observed for treated parental and neo control transfectant (S-phase elongation for 5-FdUrd, G 2 arrest for 4 HC, and G O arrest for ionomycin and TG).

In the bcl-2-expressing transfectants, the qualitative extent and kinetics of DNA fragmentation induced by all four drug treatments were inhibited (Figs. 2 and 3). However, the quali-

tative analysis of the pattern of DNA fragmentation still reveals

the characteristic nucleosomal ladder even in the bcl-2 transfec-

tants. Likewise, although net cell accumulation stopped after

2-3 days of exposure to all four of the drugs, the extent and kinetics of loss of viable cell number are inhibited in the bcl-2-expressing transfectant (Figs. 2 and 3). This occurred although there was no significant difference in the kinetics or extent of perturbation of cell cycle distribution as compared to either parental or neo controls transfectants following any of the four treatments. Time-lapse videomicroscopic analysis and Hoescht staining demonstrated that while there are bcl-2-expressing cells undergoing programmed death, their proportion is inhibited by >50% at each time point compared to parental AT-3 or neo control transfectants.



107

Fig. 3 Kinetics of cytotoxicity (upper panels) and DNA fragmentation (lower panels) in AT-3 neo-only control transfectant cells versus bcl-2- expressing transfectant cells treated with 10 ~M ionomycin (left side) or 500 nM TG (right side). Fragmented DNA is -< 1 megabase.

100

1 0 7

104 0 214 4J8 ;2 916 120

Time (hrs)

75

Ionomycin

50

25

Ionomycin

0 106 ¢u

.~

,- 105

E 7

106 .O

>

O 105

JO E #

104 0

Thapsigargin

Thapsigargin

21" 418 ;2 9t6 1:20

Time (hrs)

Z<-6

1 3 0

~ o v ~

100

" o 0 ~ 0 75

so

u=. 25

0


0 2'4 4'8 A 9'6 0 2'4 4'8 ;2 9'6 Time (hrs) Time (hrs)

AT3neo ~ AT3bcl2 clone3 - - - o ~ AT3bcl2 clone4

--z~--- AT3neo(no treatment)

To more quantitatively characterize the inhibition of cell proliferation-dependent and -independent programmed death induced by bcl-2 expression, the dose response of neo-only control transfectant cells versus bcl-2-expressing transfectants was compared for the four different cytotoxic drugs. Based on the kinetic data of Figs. 2 and 3, cells were exposed continuously for 24 h to varying doses of 5-FdUrd, 24 h after a pulse exposure to varying doses of 4HC or for 48 h of continuous exposure to varying doses of ionomycin or TG before being tested in clonogenic survival assays. The con- centration of 5-FdUrd needed to commit 50% of the neo-only control cells to undergo programmed death (i.e., ECso) is 0.05 ~M (Fig. 4A). In contrast, the ECso value for 5-FdUrd in the bcl-2 transfectant cells is 2-3 ~M, a value >40-fold higher (Fig. 4A). An enhancement, although not as large, in the resistance of the AT-3 bcl-2 transfectants (i.e., ECso, 8 ~M) compared to neo-only control transfectant (ECso, 3 ~M) cells is seen also when 4HC is used as the proliferation- dependent cytotoxic-inducing agent (Fig. 4B). The ECso value for ionomycin (Fig. 4C) for the parental and neo-only control transfectants was 5 ~M versus 25-40 ~M for the bcl-2 transfectant cells. Thus, bcl-2-expressing AT-3 cells are at least 5-fold more resistant to ionomycin-induced proliferation-independent programmed cell death. An even larger resistance (i.e., >30-fold) was observed for the bcl-2 trans-

fectants when TG was used as the killing agent (Fig. 4D). The ECso value for TG for the parental and neo-only control transfectant cells is 30 nM versus a value of > 1000 nM for the bcl-2 transfectants.

Inability of bcl-2 Protein to Inhibit the Early Changes Occurring during Programmed Death of AT-3 Cells. Pre- viously we have demonstrated that within the first 24 h of treatment of AT-3 cells with agents which induce either proliferation-dependent programmed cell death (i.e., 5-FdUrd) or proliferation-independent programmed cell death (i.e., ionomycin or TG), there is a major change in the mRNA expression for a series of distinct genes (14). For 5-FdUrd treatment, GRP-78 TTG and clusterin mRNAs are induced (14). For ionomycin and TG treatment, GRP-78, TTG, c-myc, calmodulin, and oL-prothymosin mRNAs are induced, but not clusterin (14). Therefore, the ability of bcl-2 expression to modify these patterns of gene induction following exposure to these agents, as well as 4HC, was tested. These results demonstrated that within 24 h of 4HC exposure, GRP-78, TTG, c-myc, calmodulin, ~-prothymosin, and clusterin are induced in AT-3 cells, and that the pattern and kinetics of mRNA induction are unchanged in the bcl-2 expressing AT-3 transfectant cells following exposure to 4HC. Like- wise, there were no differences in the pattern and kinetics of mRNA induction during the first 24 h of exposure to either 5-FdUrd, ionomycin, or TG.




A 100-

-- 75-

i 50-

a. 25-

C

B 100-

75

50

2 5

0 0 13---- 0 0 1 1 10 100 0 5 10 20 40

5-FrdU (pM) 4HC (pM)

D 100-

.~ 75-

- ' !

u) 5o-

L,. Q

a. 25

1 I I /

0 1 10 25 50

75

5o

Q

~ 25

0 I I I I I

0 10 100 500 1000

Ionomyein (pM) Thapsigargin (nM)

AT3neo - - - o - - AT3bcl2 clone3 ~ AT3bcl2 clone4

Fig. 4 Dose-response relationship between drug exposure and loss of clonogenic ability in AT-3 net-only control versus bcl-2-expressing transfectant cells. A, 5-FdUrd (5-FrdU); B, 4HC; C, ionomycin; D, TG.

D I S C U S S I O N Within the epithelial compartment of the normal prostate,

bcl-2 is expressed by the basal epithelial cells, neuroendocrine cells, and in the intraacinar lymphocytes but not by the glandular epithelial cells (33-35). It is these bcl-2-negative prostatic glandular cells which are the major androgen-dependent cell type present within the gland, and these are the cells of origin for the majority of human prostatic adenocarcinomas (46). Previous studies have demonstrated that the malignant progression of these normal prostatic glandular cells to either high-grade PIN or pathologically localized stage B prostatic cancer is rarely (0 and <7%, respectively) associated with the expression of bcl-2 protein (35). Based on immunocytochemical staining, it is clear that the progression of pathologically localized stage B prostate cancer to pathologically disseminated stage D] (to pelvic lymph nodes) and D 2 (to bone) disease is associated with a signifi- cantly (P < 0.05) increased frequency of bcl-2 protein expression. Likewise, within the well-characterized Dunning R-3327 system of rat prostatic cancers, there is a significant (P < 0.05) association between bcl-2 protein expression and progression to an androgen-independent metastatic phenotype. Although there is a consistent association between enhanced bcl-2 protein expression and progression of both human and rodent prostatic cancer, such bcl-2 expression is not an absolute requirement for progression to an androgen-independent metastatic phenotype in

either species. (Only 17% of lymph node metastases and 52% of bone metastases in humans with prostatic cancer express detectable bcl-2 protein.) In addition, only 42% of androgen-independent bone metastases in humans have detectable bcl-2 expression.

There are several possible explanations for this data. First, there could be multiple pathways for progression to an androgen-independent metastatic phenotype by prostatic cancer cells, only one of which is affected by bcl-2 expression. Alternatively, bcl-2 expression may not have a direct ability to induce the specific development of an androgen-independent metastatic phenotype but instead have an indirect ability to increase survival (clonal expansion) of random genetic variants developing stoicastically due to the genetic instability of prostatic cancer cells (38, 44). Since some of these newly developing clones could have the genetic changes required for the androgen- independent metastatic phenotype, the increased survival of such novel clones stoicastically via bcl-2 expression would indirectly increase the progression to a more malignant phenotype.

Consistent with this indirect mechanism are the data presented in this article demonstrating that when bcl-2 nonexpress- ing androgen-independent highly metastatic AT-3 rat prostatic cancer cells are genetically engineered to overexpress bcl-2, these cells now acquire an increased resistance to a variety of




noxious insults. The mechanism for how bcl-2 expression in-

duces such a multidrug cross-resistant state is unknown. Data

presented in this article demonstrate that bcl-2 expression does

not change the kinetics or extent of early (within the first 24 h)

mRNA induced by the various agents. Nor does such bcl-2

expression inhibit the early (i.e., within 30 min) elevation in

intracellular free calcium induced by ionomycin or TG. bcl-2

protein expression does inhibit, however, both the kinetics and

extent of DNA and cellular fragmentation into apoptotic bodies

induced after 24 h of exposure to all of the agents tested. These

results demonstrate that bcl-2 protein effects a late step in the

biochemical cascade of programmed cell death commonly in-

duced by all of the viral and chemical agents tested. Since the

initial effects of these different agents (Sindbis virus, 5-FdUrd,

4HC, ionomycin, and TG) are very different, this suggests that

although the biochemical pathway for initiating programmed

cell death can be variable, the biochemical pathway for com-

pleting this process eventually converges on some common

stereotypic step(s), at least one of which is inhibited by bcl-2

expression (47).

The present study demonstrates that bcl-2 protein expres-

sion is associated with the progression of both human and rodent

prostatic cancer cells. One mechanism for such bcl-2 expression

during the progression of prostatic cancer could be via defects in

the p53 tumor suppressor pathway. Miyashita et al. (36) have

presented evidence that p53 expression decreases the expression

of bcl-2 and increases the expression of bax. Such an explana-

tion is consistent with the demonstration by Navone et al. (48)

that p53 mutations while infrequent in the early stage of human

prostatic cancer (< 10%) are associated with the progression of

prostatic cancer from a localized to a bone metastatic androgen-

independent phenotype. In this regard, it may be more than

simple coincidence that ~50% of such prostatic cancer bone

metastases have both p53 mutation (48) and bcl-2 protein ex-

pression (present study). Studies to directly test whether p53

mutations and bcl-2 protein expression colocalized within the same human prostatic cancer cells within bone metastases are presently being performed by dual immunocytochemical stain-

ing. Regardless of these results, it is clear that the clinical

significance of detectable expression of bcl-2 by prostate

cancer cells is that it is a predictor of aggressive clinical

behavior. Since bcl-2 expression does not correlate with the Gleason score, immunocytochemica l staining for bcl-2 ex-

pression may well be an independent and thus useful adjuvant

to histological grading to predict the biological behavior of

prostatic cancers.

R E F E R E N C E S

1. Kyprianou, N., and Isaacs, J. T. Activation of programmed cell death in the rat ventral prostate following castration. Endocrinology, 122: 552-562, 1988.

2. Kyprianou, N., English, H., and Isaacs, J. T. Programmed cell death during regression of the PC-82 human prostate cancer following androgen ablation. Cancer Res., 50: 3748-3752, 1990.

3. Berges, R. S., Furuya, Y., Jacks, T., English, H., and Isaacs, J. T. Cell proliferation, DNA repair, and p53 function are not required for programmed death of prostatic glandular cells induced by androgen ablation. Proc. Natl. Acad. Sci. USA, 90: 8910-8914, 1993.

4. Furuya, Y., and Isaacs, J. T. Proliferation independent cell death as a therapeutic target for prostatic cancer. Cancer Bull., 46: 173-180, 1994.

5. Furuya, Y., Lin, X. S., Walsh, J. C., Nelson, W. G., and Isaacs, J. T. Androgen ablation-induced programmed death of prostatic glandular cells does not involve recruitment into a defective cell cycle or p53 induction. Endocrinology, 136:1898 - 1906, 1995.

6. Kyprianou, N., and Isaacs, J. T. Expression of transforming growth factor [3 in the rat ventral prostate during castration-induced programmed cell death. Mol. Endocrinol., 3: 1515-1522, 1989.

7. Furuya, Y., and Isaacs, J. T. Differential gene regulation during programmed death (apoptosis) versus proliferation of prostatic glandular cells induced by androgen manipulation. Endocrinology, I33: 2660- 2666, 1993.

8. Kyprianou, N., English, H., and Isaacs, J. T. Activation of Ca 2+- Mg2+-dependent endonuclease as an early event in castration induced prostatic cell death. Prostate, 13:103-117, 1988.

9. Martikainen, P., and Isaacs, J. T. The role of calcium in the programmed death of rat prostatic glandular cells. Prostate, 17: 175-188, 1990.

10. Carter, H. B., and Isaacs, J. T. Experimental and theoretical basis for hormone treatment of prostatic cancer. Semin. Urol., VI: 262-268, 1988.

11. Kyprianou, N., and Isaacs, J. T. "Thymineless" death in androgen- independent prostatic cancer Cells. Biochem. Biophys. Res. Commun., 165: 73-81, 1989.

12. Martikainen, P., Kyprianou, N., Tucker, R. W., and Isaacs, J. T. Programmed death of non-proliferating androgen independent prostatic cancer cells. Cancer Res., 51: 4693-4700, 1991.

13. Crawford, E. D., Eisenberger, M. A., McLeod, D. C., Spaulding, J., Benson, R., Dorr, F. A., Blumenstein, B. A., Davis, M. A., and Good- man, P. J. A control randomized trial of leuprolide with and without flutamide in prostatic cancer. N. Engl. J. Med., 321: 419-424, 1989.

14. Furuya, Y., and Isaacs, J. T. Proliferation dependent versus independent programmed cell death of prostatic cancer cells involving distinct gene regulation. Prostate, 25: 301-309, 1994.

15. Furuya, Y., Lundmo, P., Short, A. D., Gill, D. L., and Isaacs, J. T. The role of calcium, pH, and cell proliferation in the programmed (apoptotic) death of androgen-independent prostatic cancer cells induced by thapsigargin. Cancer Res., 54: 6167-6175, 1994.

16. Goulian, M., Bleile, B. M., Dickey, L. M., Grafstrom, R. H., Ingraham, H. A., Neynaber, S. A., Peterson, M. S., and Tseng, B. Y. Mechanism of thymineless death. Adv. Exp. Med. Biol., 195: 89-95, 1986.

17. Canman, C. E., Lawrence, T. S., Shewach, D. S., Tang, H-Y., and Maybaum, J. Resistance to fluorodeoxyuridine-induced DNA damage and cytotoxicity correlates with an elevation of deoxyuridine triphos- phatase activity and failure to accumulate deoxyuridine triphosphate. Cancer Res., 53: 5219-5224, 1993. 18. Isaacs, J. T., and Lundmo, P. Chemotherapeutic induction of programmed cell death in non-;Proliferating prostate cancer cells. Proc. Am. Assoc. Cancer Res., 33: 588-589, 1992.

19. Tsujimota, Y., Cossman, J., Jaffe, E., and Croce, C. Involvement of the bcl-2 gene in human follicular lymphoma. Science (Washington DC), 228: 1440-1443, 1985.

20. Chen-Levy, S., and Cleary, M. L. Membrane topology of the Bcl-2 proto-oncogenic protein demonstrated in vitro. J. Biol. Chem., 265: 4929-4933, 1990.

21. Krajewski, S., Tanaka, S., Takayama, S., Schibler, M. J., Fenton, W., and Reed, J. C. Investigations of the subcellular distribution of the bcl-2 oncoprotein: residence in the nuclear envelope, endoplasmic reticulum, and outer mitochondrial membranes. Cancer Res., 53: 4701- 4714, 1993. 22. Reed, J. C., Cuddy, M., Slabiak, T., Croce, C. M., and Nowell, P. C. Oncogenic potential of bcl-2 demonstrated by gene transfer. Nature (Lond.), 336: 259-261, 1988.




23. McDonnell, T. J. Deane, N., Platt, F. H., Nunes, G., Jaeger, U., McKearn, J. P., and Korsmeyer, S. J. Bcl-2-immunoglobulin transgenic mice demonstrate extended B cell survival and follicular lymphoprolif- eration. Cell, 57: 79-88, 1989. 24. Vaux, D., Cory, S., and Adams, J. Bcl-2 promotes haemopoietic cell survival and cooperates with c-myc to immortalize pre-B cells. Nature (Lond.), 335: 440-442, 1988.

25. Hockenbery, D., Nufiez, G., Milliman, C., Schreiber, R. D., and Korsmeyer, S. J. Bcl-2 is an inner mitochondrial membrane protein that blocks programmed cell death. Nature (Lond.), 348: 334-336, 1990.

26. Oltvai, Z. N., Milliman, C. L., and Korsmeyer, S. J. Bcl-2 het- erodimerizes in vivo with a conserved homolog, Bax, that accelerates programed cell death. Cell, 74: 609-619, 1993.

27. Yin, X-M., Oltvai, Z. N., and Korsmeyer, S. J. BH1 and BH2 domains of bcl-2 are required for inhibition of apoptosis and het- erodimerization with Bax. Nature (Lond.), 369: 321-323, 1994.

28. Miyashita, T., and Reed, J. C. Bcl-2 gene transfer increase relative resistance of $49.1 and WEHI7.2 lymphoid cells to cell death and DNA fragmentation induced by glucocorticoids and multiple chemotherapeutic drugs. Cancer Res., 52:5407-5411, 1992.

29. Miyashita, T., and Reed, J. C. Bcl-2 oncoprotein blocks chemother- apy-induced apoptosis in a human leukemia cell line. Blood, 81: 151- 157, 1993.

30. Walton, M. I., Whysong, D., O'Connor, P. M., Hockenbery, D., Korsmeyer, S. J., and Kohn, K. W. Constitutive expression of human bcl-2 modulates nitrogen mustard and camptothecin induced apoptosis. Cancer Res., 53: 1853-1861, 1993.

31. Hanada, M., Krajewski, S., Tanaka, S., Cazals-Hatem, D., Spengler, B. A., Ross, R. A., Biedler, J. L., and Reed, J. C. Regulation of bcl-2 oncoprotein levels with differentiation of human neuroblastoma cells. Cancer Res., 53: 4978-4986, 1993.

32. Levine, B., Huang, Q., Isaacs, J. T., Reed, J. C., Griffin, D. E., and Hardwick, J. M. Conversion of lytic to persistent alphavirus infection by the bcl-2 cellular oncogene. Nature (Lond.), 361: 739-742, 1993.

33. McDonnell, T. J., Troncoso, P., Brisbay, S. M., Logothetis, C., Chung, L. W. K., Hsieh, J-T., Tu, S-M., and Campbell, M. L. Expres- sion of the protooncogene bcl-2 in the prostate and its association with emergence of androgen-independent prostate cancer. Cancer Res., 52: 6940-6944, 1992. 34. Colombel, M., Symmans, F., Gil, S., O'Toole, K. M., Chopin, D., Benson, M., Olsson, C. A., Korsmeyer, S., and Buttyan, R. Detection of the apoptosis-suppressing oncoprotein bcl-2 in hormone-refractory human prostate cancers. Am. J. Pathol., 143: 390-400, 1993. 35. Shabaik, A. S., Krajewski, S., Burgan, A., Krajewski, M., and Reed, J. C. bcl-2 protooncogene expression in normal, hyperplastic, and neo- plastic prostate tissue. J. Urol. Pathol., 3: 17-27, 1995.

36. Miyashita, T., Krajewski, S., Krajewska, M., Wang, H. G., Lin, H. K., Liebermann, D. A., Hoffman, B., and Reed, J. C. Tumor suppressor p53 is a regulator of bcl-2 and bax gene expression in vitro and in vivo. Oncogene, 9: 1799-1805, 1994.

37. Krajewski, S., Krajewska, M., Shabaik, A., Miyashita, T., Wang, H. G., and Reed, J. C. Immunohistochemical determination of in vivo

distribution of BAX, a dominant inhibitor of bcl-2. Am. J. Pathol., 145: 1323-1336, 1994.

38. Isaacs, J. T., and Hukku, B. Nonrandom involvement of chromosome 4 in the progression of rat prostatic cancer. Prostate, 13: 165-188, 1988.

39. Isaacs, J. T., Isaacs, W. B., Feitz, W. F. J., and Scheres, J. Estab- lishment and characterization of seven Dunning rat prostatic cancer cell lines and their use in developing methods for predicting metastatic abilities of prostatic cancers. Prostate, 9: 261-281, 1986.

40. Hilton, J. Deoxyribonucleic acid crosslinking by 4-hydroperoxycyclophosphamide in cyclophosphamide-sensitive and -resistant L1210 cells. Biochem. Pharmacol., 33: 1867-1872, 1984.

41. Stamato, T. D.. and Denko, N. Asymmetric field inversion gel electrophoresis: a new method for detecting DNA double-strand breaks in mammalian cells. Radiat. Res., 121: 196-205, 1990.

42. Reed, J. C., Meister, Ly, Tanaka, S., Cuddy, M., Yum, S, Geyer, C., and Pleasure, D. Differential expression of bcl-2 protooncogene in neuroblastoma and other human tumor cell lines of neural origin. Cancer Res., 51: 6529-6538, 1991.

43. Bostwick, D. G. Prostatic intraepithelial neoplasia (PIN): current concepts. J. Cell. Biochem., (Suppl. 16H): 10-19, 1992.

44. Isaacs, J. T., Wake, N., Coffey, D. S., and Sandberg, A. A. Genetic instability coupled to clonal selection as a mechanism for tumor progression in the Dunning R-3327 rat prostatic adenocarcinoma system. Cancer Res., 42: 2353-2361, 1982.

45. Humphries, J. E., and Isaacs, J. T. Unusual androgen sensitivity of the androgen-independent Dunning R-3327 G rat prostatic adenocarcinoma: androgen effect on tumor cell loss. Cancer Res., 42: 3148-3156, 1982.

46. Isaacs, J. T. Role of Androgens in Prostatic Cancer. Vitamins and Hormones, Vol. 49, pp. 433-502. New York: Academic Press, 1994. 47. Reed, J. C. Bcl-2 and the regulation of programmed cell death. J. Cell. Biol., 124: 1-6, 1994.

48. Navone, N. M., Troncoso, P., Pisters, L. L., Goodrow, T. L., Palmer, J. L., Nichols, W. W., von Eschenbach, A. C., and Conti, C. J. p53 protein accumulation and gene mutation in the progression of human prostate carcinoma. J. Natl. Cancer Inst., 85: 1657-1669, 1993.



1996;2:389-398. Clin Cancer Res Y Furuya, S Krajewski, J I Epstein, et al. prostatic cancers.Expression of bcl-2 and the progression of human and rodent

Updated version

http://clincancerres.aacrjournals.org/content/2/2/389

Access the most recent version of this article at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://clincancerres.aacrjournals.org/content/2/2/389To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



Documents

Expression of bcl-2 and the Progression of Human and ... · of prostatic cancer was determined using immunocytochemical analysis in a series of human and rodent prostatic cancers