Altered Prostatic Epithelial Proliferation and Apoptosis ... - BioOne

Download PDF
3 downloads 0 Views 377KB Size Report
Comment
Jul 13, 2005 - involve altered expression of the cell-cycle regulators p27 and p21 (also known as CDKN1B and CDKN1A, respectively). To determine the role ...
BIOLOGY OF REPRODUCTION 73, 951–958 (2005) Published online before print 13 July 2005. DOI 10.1095/biolreprod.105.040980

Altered Prostatic Epithelial Proliferation and Apoptosis, Prostatic Development, and Serum Testosterone in Mice Lacking Cyclin-Dependent Kinase Inhibitors1 Motoko Mukai,3 Qiang Dong,4 Matthew P. Hardy,4 Hiroaki Kiyokawa,5 Richard E. Peterson,6 and Paul S. Cooke2,3 Department of Veterinary Biosciences,3 University of Illinois, Urbana, Illinois 61802 Population Council and Rockefeller University,4 New York, New York 10021 Department of Molecular Genetics,5 College of Medicine, University of Illinois, Chicago, Illinois 60607 School of Pharmacy,6 University of Wisconsin, Madison, Wisconsin 53705 ABSTRACT

INTRODUCTION

Normal prostatic development and some prostatic diseases involve altered expression of the cell-cycle regulators p27 and p21 (also known as CDKN1B and CDKN1A, respectively). To determine the role of these proteins in the prostate, we examined prostatic phenotype and development in mice lacking p27 and/or p21. In p27-knockout (p27KO) mice, epithelial proliferation was increased 2- and 3.8-fold in the ventral and dorsolateral prostate, respectively, versus wild-type (WT) mice, although prostatic weights were not different. Epithelial apoptosis was increased in p27KO mice and may account for the lack of a concurrent increase in weight. Testosterone deficiency observed in this group was not the cause of this increase, because vehicleand testosterone-treated p27KO mice had similar percentages of apoptotic cells. Also observed was a trend toward a decreased functional epithelial cytodifferentiation, indicating a potential role of p27 in this process. Conversely, dorsolateral prostate and seminal vesicle (SV) of p21-knockout (p21KO) mice, and all prostatic lobes and SV of p21/p27 double-knockout mice, weighed significantly less compared to the WT mice, and their epithelial proliferation was normal. Decreased testosterone concentrations may contribute to the decreased prostatic weights. However, other factors may be involved, because testosterone replacement only partially restored prostatic weights. We conclude that loss of p27 increases prostatic epithelial proliferation and alters differentiation but does not result in prostatic hyperplasia because of increased epithelial cell loss. The p21KO mice showed phenotypes distinctly different from those of p27KO mice, suggesting nonredundant roles of p21 and p27 in prostatic development. Loss of p27 or of both p21 and p27 results in serum testosterone deficiency, complicating analysis of the prostatic effects of these cell-cycle regulators.

Prostate cancer and benign prostatic hyperplasia (BPH) are two leading health problems in men worldwide, and dysregulated epithelial cell proliferation is characteristic of each. Elucidating the mechanism by which regulation of prostatic cell proliferation and differentiation normally occurs is critical for understanding these diseases. Cell-cycle regulators, such as p27Kip1 and p21Cip1 (also known as CDKN1B and CDKN1A, respectively), are important in control of cell proliferation and differentiation in many organs, and recent studies have indicated abnormal expression of these proteins in prostatic diseases. A major site for regulation of cell proliferation is the G1 phase of the cell cycle. If a cell passes the G1 restriction point, then it is committed to proceed into the S phase and, subsequently, to divide. This G1/S progression is regulated by cyclin-dependent kinases (CDKs), which are activated through association with cyclins (for review, see [1]). When these kinases are active, they phosphorylate retinoblastoma proteins, which are normally bound to E2F transcription factors. Following phosphorylation of retinoblastoma proteins, E2F proteins are released and activate genes that cause cells to proceed into the S phase [2]. Activity of cyclin/CDK complexes is inhibited by cyclin-dependent kinase inhibitors (CDKIs), comprised of the Ink4 and Cip/ Kip families. The latter Cip/Kip family consists of p27Kip1 [3, 4], p21Cip1 [5–7], and p57Kip2 [8, 9], and it regulates the activity of cyclin E/CDK2 and cyclin D/CDK complexes. Recent studies suggest both p27 and p21 are involved in normal prostatic development and may be regulated by androgens [10, 11]. In normal mouse prostate, p27 protein expression begins during development and is high in adult prostatic epithelium (unpublished results), whereas p21 is only transiently expressed in prostatic epithelium during development [12]. Castration results in decreased androgenic stimulation, apoptosis of prostatic epithelial cells, and prostatic atrophy. Waltregny et al. [10] reported that this postcastration atrophy was associated with significant increases in p27. Furthermore, testosterone (T) treatment of castrated mice decreased prostatic p27 protein concentrations, allowing epithelial proliferation, which continued until p27 and p21 levels rose and inhibited further proliferation. Altered expression of both p27 and p21 occur in prostatic diseases. For example, p27 is decreased in prostatic cancer and BPH [13], and this may be linked to increased epithelial proliferation in these diseases. Treating LNCaP human prostate cancer cells with interleukin-6 [14] or the flavonoid-antioxidant silibinin [15] resulted in G1 arrest and differentiation, which coincided with increased p27 expres-

apoptosis, male reproductive tract, prostate, testosterone Supported by NIH grants AGO24387 (to P.S.C.), HD3005 (to H.K.), HD32588 (to M.P.H), ES01332 (to R.E.P.), and the Thanis A. Field Endowment. M.M. was an Environmental Toxicology Scholar at the University of Illinois and was also supported by an Eli Lilly Predoctoral Fellowship (Eli Lilly Co., Indianapolis, IN). The present investigation was conducted in a facility constructed with support from Research Facilities Improvement Program Grant Number C06 RR16515 from the National Center for Research Resources, National Institutes of Health. 2 Correspondence: Paul S. Cooke, Department of Veterinary Biosciences, University of Illinois at Urbana-Champaign, 2001 S. Lincoln Ave, Urbana, IL 61802. FAX: 217 244 1652; e-mail: [email protected] 1

Received: 14 February 2005. First decision: 7 March 2005. Accepted: 11 July 2005. Q 2005 by the Society for the Study of Reproduction, Inc. ISSN: 0006-3363. http://www.biolreprod.org

951

952

MUKAI ET AL.

sion. Abnormal expression of p21 also has been reported to be associated with altered prostatic phenotypes. The transient expression of p21 during prostatic development is not observed in rats that have been estrogenized neonatally, and these animals later develop prostatic hyperplasia [12]. Human prostate cancer xenografts in mice showed transient expression of p21 after androgen withdrawal (castration), and androgen-independent prostate cancer was associated with an increase in p21 expression [16]. The importance of p27 and/or p21 in regulating overall body growth as well as cell proliferation in testis [17, 18], ovary [19], adipose [20], and pituitary [21, 22] has been established through use of knockout mouse models [21– 24]. Prostates of p27-knockout mice were reported previously to have an increase in epithelial proliferation without an increase in prostatic weight [13]. However, the prostatic phenotype of mice lacking p21 has not been examined. Therefore, we studied prostatic phenotype and development in mice lacking p21 and/or p27. Our results show that p27 has a role in inhibiting prostatic epithelial proliferation and apoptosis and, perhaps, in promoting differentiation. Although lack of p21 does not increase prostatic epithelial proliferation, determining the effects of p21 and p27 is complicated by our unexpected findings of low androgen in mice lacking p27 or both p21 and p27. MATERIALS AND METHODS

Animals All mice were generated from our colony at the University of Illinois, Urbana-Champaign. Male wild-type (WT) and p27 single-knockout (p27KO) [21] mice were obtained by crossing Cdkn1b1/2 males and females. The p21KO mice [24] and double-knockout (DBKO) mice [19], which lacked both p21 and p27, were obtained by crossing Cdkn1a2/ 2 Cdkn1b1/2 males and females. Genotypes of pups were determined by polymerase chain reaction (PCR) using genomic DNA obtained from tail lysis. Mice were housed under controlled lighting (photoperiod, 12L:12D; lights-on at 0700 h) and temperature (228C), and they were fed a standard diet ad libitum. All animal experiments were approved by the Institutional Animal Care and Use Committee of the University of Illinois and were conducted in accordance with the NIH Guide for the Care and Use of Laboratory Animals.

Tissue Collection, RIA, and Treatments Between 115 and 120 days of age, male p27KO, p21 single-knockout (p21KO), and DBKO mice, as well as WT littermates, were anesthetized and perfused first with saline and then with 10% neutral-buffered formalin. The urogenital complex was removed, and the ventral prostate (VP), dorsolateral prostate (DLP), anterior prostate (AP), and seminal vesicle (SV) were separated under a dissecting microscope, weighed, and immersionfixed in neutral-buffered formalin for an additional 48 h. Prostatic lobes from some mice were removed without perfusion and used to obtain RNA. Blood was collected at approximately 1400 h from mice ranging from 90 to 120 days of age, and the serum was separated and stored at 2208C for T and LH RIA. Serum T concentrations were measured using a tritium-based RIA as described previously [25]. Serum LH concentrations were assayed by the method of Chandrashekar and Bartke [26]. Values for the interassay variation of the T and LH RIAs were between 4% and 8%; the intra-assay variation was less than 15% for both hormones. The sensitivities of the assays for T and LH were 10 pg/ml and 0.12 ng/ml, respectively. Dorsolateral prostates of mice ranging from 120 to 230 days of age were collected to measure DNA content. The DNA was extracted using the QIAamp DNA Mini Kit (Qiagen, Valencia, CA) following the manufacturer’s instructions, including the RNase treatment to remove residual RNA. The amount of DNA was then measured by NanoDrop (NanoDrop Technologies, Wilmington, DE). For T-replacement experiments, 105- to 110-day-old animals were injected daily i.p. with 10 mg/kg per day of T propionate (TP; J.T. Baker, Phillipsburg, NJ) dissolved in corn oil for 10 days and then killed a day

after the final treatment. This dose and duration of TP was chosen because it produces maximal effects on prostatic growth and fully restores prostatic weights in adult mice that were neonatally castrated [27].

Immunohistochemistry Fixed tissues were dehydrated, embedded in paraffin, and sectioned (thickness, 3 mm) for immunohistochemical analysis. Sections were deparaffinized and hydrated, and endogenous peroxidase activity was quenched by 0.3% H2O2 in methanol. Antigen retrieval was performed by boiling sections in 0.01 M citrate buffer (pH 6.0) for 20 min, then allowing them to cool to room temperature. Cell proliferation was assessed using the mouse anti-human Ki-67 antibody (1:3000; Pharmingen, San Diego, CA), and basal cells were detected by rabbit p63 antibody (1:500; Santa Cruz Biotechnologies, Santa Cruz, CA). The Vectastain Elite ABC kit (Vector Laboratories, Burlingame, CA) was used according to the manufacturer’s instructions. Immunostaining was visualized using 3,39-diaminobenzidine-tetrahydrochloride (DAB; Sigma, St. Louis, MO) as a peroxidase substrate for 7 min, and sections were counterstained with hematoxylin and then mounted for light-microscopic analysis. Proliferation of epithelial cells was determined by counting the percentage of luminal epithelial cells positive for Ki-67 (;500 cells counted/animal) in the distal tip region. In addition to immunohistochemistry, slides were stained with hematoxylin and eosin to assess histological structure.

TUNEL Assay Apoptosis was detected in paraffin-embedded sections of WT and knockout DLP using the Fluorescein-FragEL DNA Fragmentation Detection kit (Oncogene, San Diego, CA) following the manufacturer’s protocol with slight modifications. After the terminal deoxynucleotidyl transferase labeling step, an antifluorescein antibody conjugated with peroxidase (TUNEL POD; Roche, Mannheim, Germany) was applied, and immunoreactions were localized by the procedure described above, with DAB as the substrate and methyl green as a counterstain. Apoptosis was quantified by determining the percentage of TUNEL-positive cells (1000 cells counted/ animal) in the epithelium.

Real Time-PCR Total RNA was isolated using RNeasy Mini Kits (Qiagen). The RNA was diluted in diethyl pyrocarbonate-treated H2O and quantified using Gene Quant (Pharmacia Biotech Ltd., UK), and 1 mg of total RNA was used for the reverse-transcriptase reaction to synthesize cDNA. Total RNA samples (1 mg) were reverse transcribed for 50 min at 428C in a 20-ml reaction with 200 U of SuperScript II reverse transcriptase (Invitrogen) and 0.5 mg of Oligo(dT)12–18 primer following the standard protocol of the manufacturer. The synthesized cDNA was diluted 1:25, and 4 ml of this diluted cDNA was used for subsequent 20-ml PCR reactions. Next, 23 SYBR Green PCR Master Mix (Applied Biosystems, CA) was used with a final concentration of 300 nM for each forward and reverse primer. Primer sequences for cytokeratin 8, Sbp (i.e., MP25) and cyclophilin were as described previously [28]. Quantitative real-time PCR analysis was done using the ABI Prism 7000 Sequence Detector. The Ct (threshold cycle) value was obtained, and the relative amount of amplicon was calculated using methods described in Applied Biosystems User Bulletin 2.

Statistics All values in the figures are expressed as the mean 6 SEM. Values for T and LH were transformed into natural log to obtain normal distribution of the data for parametric analysis. Differences between groups were analyzed by the Student t test or by the general linear model followed by the Dunnett or Tukey test for pairwise comparisons using SYSTAT (SSI, Richmond, CA). All P values were two-sided and considered to be statistically significant at P , 0.05 and marginally significant at 0.05 , P , 0.1.

RESULTS

Body, Prostatic, and SV Weights

Body weights of DBKO mice were increased by 45%, whereas weights of both p27KO and p21KO mice did not show a significant difference compared to WT mice (Fig. 1B). Because of these differences in body weights between

p27 AND p21 IN PROSTATIC DEVELOPMENT

953

genotypes, prostatic weights were expressed relative to body weight (Fig. 1, C–F). Weights of all prostatic lobes and SVs in p27KO mice were not significantly different from those in WT mice. The p21KO mice showed significant decreases of 59% and 27% compared to WT mice in DLP and SV weights (Fig. 1, D and F), respectively, whereas VP and AP weights (Fig. 1, C and E) showed trends toward a decrease (these did not reach significance). Although testis sizes were increased in DBKO mice [18], VP, DLP, AP, and SV weights were decreased by 60%, 68%, 68%, and 75%, respectively, compared to those in WT mice (Fig. 1, A and C–F). Significant decreases in organ weights also were seen when absolute instead of relative weight data were used for statistical analysis (41%, 54%, 55%, and 64% decreases in VP, DLP, AP, and SV, respectively, in DBKO vs. WT). The DNA content was significantly decreased in p21KO mice (67% decrease) and marginally increased in p27KO mice (47% increase, P 5 0.07) compared to that in WT mice (Fig. 1G), and DBKO mice showed a slight trend toward a reduction versus control (34% decrease; P 5 0.4 vs. WT). Epithelial Proliferation and Apoptosis

Epithelial proliferation in VP and DLP was increased by 2- and 3.8-fold, respectively, in p27KO mice compared to WT mice (Fig. 2). An 8.7-fold increase was observed in apoptosis in p27KO luminal epithelial cells compared to those of WT mice (Fig. 3). Also observed was an increased epithelial sloughing into the lumen, and these sloughed cells were positive for TUNEL staining (not shown). Conversely, p21KO mice had a trend toward decreased epithelial proliferation compared to WT mice (2.4% 6 0.3% in p21KO vs. 3.9 6 0.9% in WT VP) in both prostatic lobes, which is not surprising considering the trend of weight decreases in this group, but this difference was not statistically significant. Apoptosis in p21KO prostate was comparable to that in WT prostate, and prostatic epithelial proliferation in DBKO mice was similar to that in WT mice (Fig. 2). A trend was found toward increased apoptosis in DBKO mice, but this did not reach significance (P 5 0.2) (Fig. 3). Morphological structure and location of basal cells in DLP of each genotype were assessed by staining for p63, a basal cell marker (Fig. 2C). These results indicated that the vast majority of cells staining for either Ki-67 or TUNEL were luminal rather than basal cells. Cytodifferentiation Factors in p27KOs

Mature prostatic ducts are composed mainly of three cell types: luminal secretory epithelial cells, basal epithelial cells, and stromal smooth muscle cells. These cell types can be distinguished by expression of specific differentiation makers. Cytokeratin 8 mRNA expression was used as a marker for luminal epithelial cells [28, 29]. Results from real-time PCR showed that p27KO cytokeratin 8 mRNA was increased 40% compared to WT (3.7 6 0.3 in WT vs. 5.2 6 0.3 in p27KO) (Fig. 4A). Also called MP25, SBP, a major prostatic secretory glycoprotein in mouse VP [30], was used as a marker for functional cytodifferentiation of luminal epithelial cells. Although p27KO VPs showed 28% decreases (4.20 6 0.59 in WT vs. 3.04 6 0.86 in p27KO) (Fig. 4B) in Sbp mRNA expression, this did not reach statistical significance. Expression of Sbp mRNA relative to that of cytokeratin 8 mRNA, which allows an approximation of Sbp mRNA per epithelial cell, was decreased in p27KO mice by 46% (1.13

FIG. 1. Effects of loss of p27 and/or p21 on prostatic and seminal vesicle weights in mice. Urogenital complex of 120-day-old WT and Cdkn1a2/2 Cdkn1b2/2 (DBKO) mice (A). For comparison, a testis of each respective genotype is displayed above the urogenital complex. Body weight (B) and relative weights of the ventral prostate (C), dorsolateral prostate (D), anterior prostate (E), and seminal vesicles (F) are shown. The DNA contents of dorsolateral prostate also are shown (G). The DNA content, as well as the DNA concentration, was significantly decreased in p21KOs. Data are presented as the mean 6 SEM (n 5 9 per group). *P , 0.05 vs. WT.

6 0.13 in WT vs. 0.61 6 0.2 in p27KO) (Fig. 4C) compared to that in WT mice, and this difference was marginally significant (P 5 0.08).

Serum T and LH levels

Because an unexpected decrease was observed in prostatic and SV sizes in p21KO and DBKO mice, serum T was measured for all genotypic groups (n 5 10–19 per group). All knockout groups showed trends of lower T levels compared to WT mice (0.73 6 0.16, 0.16 6 0.06, 0.50

954

MUKAI ET AL.

FIG. 3. Increased epithelial cell death in p27KO mice. A) Epithelial apoptosis was significantly increased in p27KO dorsolateral prostate. Results are shown as the mean 6 SEM (n 5 4/group). An asterisk indicates a significant difference (P , 0.05). B) Increased epithelial cell death in the absence of p27 visualized by TUNEL assay. The arrowhead shows a positively stained nucleus. Bar 5 50 mm.

T Replacement

FIG. 2. Epithelial proliferation in ventral and dorsolateral prostate. A) Ki67 labeling index in WT, p27KO, p21KO, and DBKO mice. Results are shown as the mean 6 SEM (n 5 4–5 per group). An asterisk indicates a significant difference (P , 0.05 vs. WT). Epithelial proliferation of p27KO ventral prostate was increased, but this difference was only marginally significant (P 5 0.09 vs. WT). B) Ki-67 immunostaining of mouse dorsolateral prostate in WT, p27KO, p21KO, and DBKO mice. The arrowhead shows an example of positively stained nucleus. Note that epithelial proliferation is increased in the p27KOs compared to all other genotypes. C) The increased epithelial proliferation occurred in the luminal epithelial cells rather than the basal cells, as shown by immunolocalization of p63, a marker of prostatic basal cells, in WT dorsolateral prostate. Basal cell distribution was similar in WT and KO mice (not shown). Bar 5 50 mm.

6 0.12, and 0.20 6 0.07 ng/ml for WT, p27KO, p21KO, and DBKO mice, respectively) (Fig. 5A). Both the p27KO and DBKO groups were significantly decreased, but concentrations in p21KO mice were not significantly decreased compared to those in WT mice. To determine further whether low T levels were caused by functional defects in Leydig cells or other reasons, serum LH levels were measured. The LH values in all knockout groups were not statistically different compared to those in WT mice (Fig. 5B).

To establish whether T supplementation could increase prostatic and SV weights, mice of all genotypes were treated with TP. The dose and duration of TP used in the present study were sufficient to produce maximal effects on growth of accessory sex organs; therefore, it would be expected to restore prostatic and SV weights of p21KO and DBKO mice if the growth deficits in these organs resulted from low T. No significant differences were found in prostatic and SV weights of TP-treated p27KO mice compared to TP-treated WT mice (Fig. 6). The effects of TP on prostatic weight of p21KO mice were minimal. Similar to vehicletreated groups, DLP and SV weights of TP-treated p21KO mice were significantly smaller than those of TP-treated WT mice. Both VP and AP weights of vehicle-treated p21KO mice were not different from those of the vehicletreated WT group. However, the TP-treated p21KO VP and AP weights were statistically less than in the TP-treated WT mice because of a greater increase in prostatic weight after TP treatment in WT. Treatment with TP increased the prostatic weight of DBKO mice (1.9- to 2.9-fold increases vs. vehicle-treated mice). However, prostatic and SV weights still remained significantly smaller compared to TP-treated WT mice. Percentages of TUNEL-positive epithelial cells were 0.7% 6 0.4% and 0.7 6 0.2% in p27KO DLP with and without TP treatment, respectively. Similarly, no significant differences were found in the percentages of apoptotic cells with or without TP treatment in WT, p21KO, and DBKO mice (not shown). DISCUSSION

Regulation of the cell cycle is important in prostatic development, growth, and differentiation. The CDKIs may have relevance for understanding major prostatic diseases, in that p27 and p21 expression are altered in prostatic can-

p27 AND p21 IN PROSTATIC DEVELOPMENT

955 FIG. 4. Expression of cytokeratin 8 and Sbp mRNA in ventral prostate of adult WT and p27KO mice. Results are shown as the mean of relative mRNA amount against cyclophilin mRNA (A and B) or against cytokeratin 8 mRNA (C) 6 SEM (n 5 4 per group). Marginal significance (P 5 0.07) was found in the Sbp mRNA level normalized to cytokeratin 8 mRNA. *P , 0.05.

cer and BPH. In the present study, we have shown that lack of p27 and/or p21 in mice led to abnormal proliferation and cell death of luminal prostatic epithelial cells as well as abnormal prostatic weights, partly caused by unexpected T deficiency. The loss of p27 increased prostatic epithelial proliferation without producing prostatic organomegaly, confirming previous findings of Cordon-Cardo et al. [13]. The increased prostatic epithelial proliferation is normally expected to occur in an environment where T is increased or at least normal. However, this increased proliferation in p27KO mice is even more notable, because it occurs in animals with significantly decreased T concentrations, as discussed below. Consistent with the wet weight, the prostatic DNA content in p27KO mice was not significantly increased. Our results indicate that epithelial apoptosis in p27KO prostate was increased, which may be responsible for the lack of a clear prostatic organomegaly despite increased epithelial proliferation. In addition, an increase was seen in sloughed epithelial cells in the lumen, which also stained positively with TUNEL. Whether increased epithelial apoptosis resulted from direct effects of p27 deficiency or indirect increases secondary to decreased T was initially unclear. However, the results from the T replacement experiment, in which no difference was observed in apoptosis between vehicle- and TP-treated p27KO groups, indicate that the decreased level of T in p27KO compared to WT mice is unlikely to explain the increased apoptosis. Therefore, the increased apoptosis in p27KO prostate appears to result from effects in the prostatic epithelium itself rather than from secondary effects on trophic hormones. A number of laboratories have reported that p27 deficiency leads to increased apoptosis in some cell types [31–34], but the mechanism of this effect remains unclear (for review, see [35]). Waltregny et al. [10] have shown that prostatic epithelial proliferation after androgen treatment of a castrated rat was associated with decreased p27. Our findings showing increased prostatic epithelial proliferation in p27KO mice are consistent with their results, and our data emphasize the importance of p27 as a major inhibitory regulator of prostatic epithelial proliferation. Loss of p27 expression is a negative prognostic marker for prostate cancer (for review, see [36]), because decreased p27 expression is associated with tumor aggressiveness [37–39]. However, the prostatic phenotype of p27KO mice indicates that lack of p27 is not, in itself, sufficient to cause prostatic overgrowth. Androgen treatment of castrated rodents induces a transient burst of high epithelial proliferation, but this subsequently falls to a baseline level even in the presence of continuous androgen stimulation. The sharp diminution of prostatic epithelial proliferation following the initial mitotic burst resulting from androgen replacement is associated with increased p27 [10]. However, the ability of exogenous T to partially restore growth but not produce excessive overgrowth in p27KO and DBKO mice indicates that the increase in p27 may not be the only factor important for

termination of the transient high epithelial proliferation that occurs initially following androgen replacement and that other factors may be involved as well [40]. The relative mRNA expression of cytokeratin 8, a marker for luminal epithelial cells [29], suggested an increase in luminal epithelial cells in p27KO compared to WT mice. Thus, despite the increased epithelial apoptosis in p27KO mice, the large increases in prostatic epithelial proliferation may produce an organ in which the epithelial:stromal ratio has been increased. Functional cytodifferentiation of epithelial cells, analyzed by mRNA expression for Sbp, an epithelial secretory protein, showed a trend toward a decrease in p27KO mice, although this did not reach statistical significance. The ratio of Sbp to cytokeratin 8 mRNA expression, which represents an approximation of Sbp mRNA production per luminal epithelial cell, showed a stronger trend toward a decrease compared to WT and was marginally significant. From our previous immunohistochemical findings, p27 protein is highly expressed in all prostatic lobes and SVs in normal adult mouse, but we have not found it to be present in prostatic buds of prenatal mice

FIG. 5. Testosterone (A) and LH (B) serum concentrations in WT and knockout adult mice (age, 90–120 days). Statistics were performed on natural log (Ln)-transformed data. Results are shown as the mean 6 SEM on a logarithmic scale for the y-axis (n 5 10–19 per group). * P , 0.05 vs. WT.

956

MUKAI ET AL.

FIG. 6. Prostatic (A–C) and seminal vesicle (D) weights in WT and knockout adult mice after 10 days of vehicle (white bar) or testosterone propionate (10 mg/kg per day i.p.; black bar) treatment. Results are shown as the mean 6 SEM (n 5 3–7 per group). *P , 0.05 vs. WT for the same treatment group, †P , 0.05 vs. vehicle treatment for the same genotypic group.

(unpublished results). This suggests that p27 is mainly expressed in differentiated, mature prostatic epithelial cells but not in immature prostate. It has been reported that onset of p27 expression also is associated with terminal differentiation in Sertoli cells [17] and adipocytes [41]. The trend toward decreased Sbp mRNA production in p27KO prostate indicates that prostatic epithelial cells in the p27KO mice may not achieve full functional differentiation. To our knowledge, no previous studies have described the prostate in p21KO mice. Along with p27, p21 is an important determinant of Sertoli cell [18] and adipocyte number [20]. In addition, p21KO mice develop spontaneous tumors in various tissues at approximately 16 mo of age [42]. Thus, increased cell proliferation and/or hyperplasia also were possible in p21KO prostates. Unexpectedly, however, p21KO mice had reduced DLP and SV relative weights compared to those in WT mice, although the VP and AP relative weights were not smaller. The DNA content also was decreased in DLP, which suggests that the overall weight reduction results from decreased cell numbers. In addition, DNA concentration was reduced in p21KO mice, indicating that changes in cell size or amount of secretions also may result from loss of p21. Epithelial proliferation showed a trend toward a decrease compared to that in WT mice, suggesting that p21 is not as critical as p27 for inhibiting prostatic epithelial proliferation. Testosterone levels in p21KO mice were comparable to those in WT mice, and further TP replacement did not increase prostatic and SV weights. In adult rodent prostate, p21 protein is not expressed, or is only slightly expressed, but it is transiently expressed during development. Lack of p21 expression during prostatic development may alter prostatic differentiation and, thus, affect ultimate growth and hormonal responsiveness. Although preliminary immunohistochemical analysis revealed no obvious increases of p27 in p21KO prostate (not shown), p27 or other cell-cycle inhibitory factors still present may compensate for p21 and inhibit further epithelial proliferation on TP treatment. Although redundant roles for these CDKIs are known, single-knockout mice showed distinctly different phenotypes, suggesting unique roles for

p21 and p27 in prostatic cell proliferation and differentiation. Additive or synergistic effects of p21 and p27 loss are observed on body weight [20], as confirmed by our data, and in other cell types [18–20]. Based on the known expression of both p27 and p21 in prostate, we predicted additive or synergistic prostatic effects of p21 deficiency in an animal already lacking p27. Therefore, the decreases in prostatic weight in DBKO mice were unexpected. The smaller accessory organs in DBKO mice may be caused, at least in part, by lack of T, which normally would increase the LH negative-feedback mechanism. However, serum LH was not different in knockout compared to WT mice. The absence of increased LH suggests that a functional alteration exists in the pituitary or feedback loop. The p27KO mice develop intermediate-lobe pituitary tumors [21, 22], which also have been observed in DBKO mice (unpublished results), consistent with a functional pituitary deficit. The decreased T in the knockout mice indicates that a functional Leydig cell deficit may exist as well. The T-treatment experiment confirmed that decreased prostatic and SV weights result, at least in part, from decreased T levels in DBKO mice, because T injection increased the weights of these organs to a level two- to threefold greater than that of the vehicle-treated DBKO mice. Our preliminary results also indicated an increase in epithelial proliferation. Nonetheless, prostatic and SV weights did not recover fully and remained smaller compared to those in WT mice even after extended T treatment. Several possibilities could explain this. Both p21 and p27 may have important, positive roles in cell proliferation and normal prostatic development, such that absence of p21 and p27 results in decreased prostatic weight despite normal androgen levels. The CDKIs also can be activators of the cell cycle, in that primary mouse embryonic fibroblasts lacking both p21 and p27 fail to correctly assemble cyclin D/CDK [43]. Because of this and other findings, roles for both p21 and p27 in cell-cycle regulation have been considered to be far more complicated than simple cell-cycle arrest at the G1/S restriction point [35, 43, 44]. Although low androgen levels are involved in decreased prostatic and

p27 AND p21 IN PROSTATIC DEVELOPMENT

SV weights of DBKO mice, failure of cyclin D/CDK complex assembly may still occur in these epithelial cells and contribute to growth deficits of the DBKO prostate. The possibility also exists that low T levels during the critical period of prostatic development may alter the normal response to androgens even later, in adults, possibly by downregulating the androgen receptor. We have not measured T or LH levels in neonatal mice. However, anogenital distances of all knockout groups were not significantly different from that of WT mice at Postnatal Day 10 (unpublished results), which indicates that neonatal T may not be decreased. Bryja et al. [45] recently demonstrated that T production by Leydig cells of 10- to 11-wk-old p27KO mice was greater than that of WT controls, suggesting the loss of p27 might inhibit Leydig cell differentiation and stimulate Leydig cell function. Our findings with prostatic epithelium are similar, in that they also emphasize the role of p27 in differentiation. However, the explanation for the apparent discrepancy of our results and those of Bryja et al. regarding T production is unclear. Their use of younger animals compared to our adult mice or different sources of knockout mice could be a factor. In summary, lack of p27 leads not only to increased prostatic epithelial proliferation but also to increased epithelial cell loss, which precludes increases in prostatic weight and DNA content despite epithelial hyperplasia. Lack of p21 did not lead to significant changes in epithelial proliferation and resulted in slightly smaller DLP and SV weights compared to those in WT controls. This suggests that the two CDKIs have different roles in prostatic epithelial proliferation. Deletion of both p27 and p21 resulted in decreased prostatic growth. The marked decreases of T levels in knockout mice were unexpected, and they indicate that p27 and/or p21 also may have important roles in Leydig cell development and function. Decreased prostatic growth in DBKO mice was caused, at least in part, by T deficiency; however, other factors also may be involved. ACKNOWLEDGMENTS We thank Dr. David Schaeffer for assistance with statistical analysis, Chantal M. Sottas for assistance with RIAs, and Charity Beals for assistance with immunohistochemistry.

REFERENCES 1. Sherr CJ. Cancer cell cycles. Science 1996; 274:1672–1677. 2. Nevins JR. The Rb/E2F pathway and cancer. Hum Mol Genet 2001; 10:699–703. 3. Polyak K, Lee MH, Erdjument-Bromage H, Koff A, Roberts JM, Tempst P, Massague J. Cloning of p27Kip1, a cyclin-dependent kinase inhibitor and a potential mediator of extracellular antimitogenic signals. Cell 1994; 78:59–66. 4. Toyoshima H, Hunter T. p27, A novel inhibitor of G1 cyclin-Cdk protein kinase activity, is related to p21. Cell 1994; 78:67–74. 5. Xiong Y, Hannon GJ, Zhang H, Casso D, Kobayashi R, Beach D. p21 Is a universal inhibitor of cyclin kinases. Nature 1993; 366:701–704. 6. El-Deiry WS, Tokino T, Velculescu VE, Levy DB, Parsons R, Trent JM, Lin D, Mercer WE, Kinzler KW, Vogelstein B. WAF1, a potential mediator of p53 tumor suppression. Cell 1993; 75:817–825. 7. Harper JW, Adami GR, Wei N, Keyomarsi K, Elledge SJ. The p21 Cdk-interacting protein Cip1 is a potent inhibitor of G1 cyclin-dependent kinases. Cell 1993; 75:805–816. 8. Lee MH, Reynisdottir I, Massague J. Cloning of p57KIP2, a cyclindependent kinase inhibitor with unique domain structure and tissue distribution. Genes Dev 1995; 9:639–649. 9. Matsuoka S, Edwards MC, Bai C, Parker S, Zhang P, Baldini A, Harper JW, Elledge SJ. p57KIP2, A structurally distinct member of the p21CIP1 Cdk inhibitor family, is a candidate tumor suppressor gene. Genes Dev 1995; 9:650–662.

957

10. Waltregny D, Leav I, Signoretti S, Soung P, Lin D, Merk F, Adams JY, Bhattacharya N, Cirenei N, Loda M. Androgen-driven prostate epithelial cell proliferation and differentiation in vivo involve the regulation of p27. Mol Endocrinol 2001; 15:765–782. 11. Lu S, Liu M, Epner DE, Tsai SY, Tsai MJ. Androgen regulation of the cyclin-dependent kinase inhibitor p21 gene through an androgen response element in the proximal promoter. Mol Endocrinol 1999; 13: 376–384. 12. Chang WY, Birch L, Woodham C, Gold LI, Prins GS. Neonatal estrogen exposure alters the transforming growth factor-b signaling system in the developing rat prostate and blocks the transient p21(cip1/ waf1) expression associated with epithelial differentiation. Endocrinology 1999; 140:2801–2813. 13. Cordon-Cardo C, Koff A, Drobnjak M, Capodieci P, Osman I, Millard SS, Gaudin PB, Fazzari M, Zhang ZF, Massague J, Scher HI. Distinct altered patterns of p27KIP1 gene expression in benign prostatic hyperplasia and prostatic carcinoma. J Natl Cancer Inst 1998; 90:1284– 1291. 14. Mori S, Murakami-Mori K, Bonavida B. Interleukin-6 induces G1 arrest through induction of p27(Kip1), a cyclin-dependent kinase inhibitor, and neuron-like morphology in LNCaP prostate tumor cells. Biochem Biophys Res Commun 1999; 257:609–614. 15. Zi X, Agarwal R. Silibinin decreases prostate-specific antigen with cell growth inhibition via G1 arrest, leading to differentiation of prostate carcinoma cells: implications for prostate cancer intervention. Proc Natl Acad Sci U S A 1999; 96:7490–7495. 16. Fizazi K, Martinez LA, Sikes CR, Johnston DA, Stephens LC, McDonnell TJ, Logothetis CJ, Trapman J, Pisters LL, Ordonez NG, Troncoso P, Navone NM. The association of p21(WAF-1/CIP1) with progression to androgen-independent prostate cancer. Clin Cancer Res 2002; 8:775–781. 17. Beumer TL, Kiyokawa H, Roepers-Gajadien HL, van den Bos LA, Lock TM, Gademan IS, Rutgers DH, Koff A, de Rooij DG. Regulatory role of p27kip1 in the mouse and human testis. Endocrinology 1999; 140:1834–1840. 18. Holsberger DR, Buchold GM, Leal MC, Kiesewetter SE, O’Brien DA, Hess RA, Franca LR, Kiyokawa H, Cooke PS. Cell-cycle inhibitors p27Kip1 and p21Cip1 regulate murine Sertoli cell proliferation. Biol Reprod 2005; 72:1429–1436. 19. Jirawatnotai S, Moons DS, Stocco CO, Franks R, Hales DB, Gibori G, Kiyokawa H. The cyclin-dependent kinase inhibitors p27Kip1 and p21Cip1 cooperate to restrict proliferative life span in differentiating ovarian cells. J Biol Chem 2003; 278:17021–17027. 20. Naaz A, Holsberger DR, Iwamoto GA, Nelson A, Kiyokawa H, Cooke PS. Loss of cyclin-dependent kinase inhibitors produces adipocyte hyperplasia and obesity. FASEB J 2004; 18:1925–1927. 21. Kiyokawa H, Kineman RD, Manova-Todorova KO, Soares VC, Hoffman ES, Ono M, Khanam D, Hayday AC, Frohman LA, Koff A. Enhanced growth of mice lacking the cyclin-dependent kinase inhibitor function of p27(Kip1). Cell 1996; 85:721–732. 22. Nakayama K, Ishida N, Shirane M, Inomata A, Inoue T, Shishido N, Horii I, Loh DY. Mice lacking p27(Kip1) display increased body size, multiple organ hyperplasia, retinal dysplasia, and pituitary tumors. Cell 1996; 85:707–720. 23. Fero ML, Randel E, Gurley KE, Roberts JM, Kemp CJ. The murine gene p27Kip1 is haplo-insufficient for tumor suppression. Nature 1998; 396:177–180. 24. Brugarolas J, Chandrasekaran C, Gordon JI, Beach D, Jacks T, Hannon GJ. Radiation-induced cell cycle arrest compromised by p21 deficiency. Nature 1995; 377:552–557. 25. Cochran RC, Ewing LL, Niswender GD. Serum levels of follicle stimulating hormone, luteinizing hormone, prolactin, testosterone, 5 alphadihydrotestosterone, 5a-androstane-3a, 17b-diol, 5a-androstane-3b, 17b-diol, and 17b-estradiol from male beagles with spontaneous or induced benign prostatic hyperplasia. Invest Urol 1981; 19:142–147. 26. Chandrashekar V, Bartke A. Influence of endogenous prolactin on the luteinizing hormone stimulation of testicular steroidogenesis and the role of prolactin in adult male rats. Steroids 1988; 51:559–576. 27. Donjacour AA, Cunha GR. The effect of androgen deprivation on branching morphogenesis in the mouse prostate. Dev Biol 1988; 128: 1–14. 28. Lin TM, Ko K, Moore RW, Simanainen U, Oberley TD, Peterson RE. Effects of aryl hydrocarbon receptor null mutation and in utero and lactational 2,3,7,8-tetrachlorodibenzo-p-dioxin exposure on prostate and seminal vesicle development in C57BL/6 mice. Toxicol Sci 2002; 68:479–487.

958

MUKAI ET AL.

29. Fuchs E. Keratins as biochemical markers of epithelial differentiation. Trends Genet 1988; 4:277–281. 30. Mills JS, Needham M, Thompson TC, Parker MG. Androgen-regulated expression of secretory protein synthesis in mouse ventral prostate. Mol Cell Endocrinol 1987; 53:111–118. 31. Hiromura K, Pippin JW, Fero ML, Roberts JM, Shankland SJ. Modulation of apoptosis by the cyclin-dependent kinase inhibitor p27(Kip1). J Clin Invest 1999; 103:597–604. 32. Eymin B, Haugg M, Droin N, Sordet O, Dimanche-Boitrel MT, Solary E. p27Kip1 induces drug resistance by preventing apoptosis upstream of cytochrome c release and procaspase-3 activation in leukemic cells. Oncogene 1999; 18:1411–1418. 33. Eymin B, Sordet O, Droin N, Munsch B, Haugg M, Van de Craen M, Vandenabeele P, Solary E. Caspase-induced proteolysis of the cyclin-dependent kinase inhibitor p27Kip1 mediates its antiapoptotic activity. Oncogene 1999; 18:4839–4847. 34. Ophascharoensuk V, Fero ML, Hughes J, Roberts JM, Shankland SJ. The cyclin-dependent kinase inhibitor p27Kip1 safeguards against inflammatory injury. Nat Med 1998; 4:575–580. 35. Coqueret O. New roles for p21 and p27 cell-cycle inhibitors: a function for each cell compartment? Trends Cell Biol 2003; 13:65–70. 36. Macri E, Loda M. Role of p27 in prostate carcinogenesis. Cancer Metastasis Rev 1998; 17:337–344. 37. Cote RJ, Shi Y, Groshen S, Feng AC, Cordon-Cardo C, Skinner D, Lieskovosky G. Association of p27Kip1 levels with recurrence and survival in patients with stage C prostate carcinoma. J Natl Cancer Inst 1998; 90:916–920. 38. Tsihlias J, Kapusta LR, DeBoer G, Morava-Protzner I, Zbieranowski

39.

40. 41.

42. 43.

44. 45.

I, Bhattacharya N, Catzavelos GC, Klotz LH, Slingerland JM. Loss of cyclin-dependent kinase inhibitor p27Kip1 is a novel prognostic factor in localized human prostate adenocarcinoma. Cancer Res 1998; 58:542–548. Yang RM, Naitoh J, Murphy M, Wang HJ, Phillipson J, deKernion JB, Loda M, Reiter RE. Low p27 expression predicts poor diseasefree survival in patients with prostate cancer. J Urol 1998; 159:941– 945. Di Cristofano A, De Acetis M, Koff A, Cordon-Cardo C, Pandolfi PP. Pten and p27KIP1 cooperate in prostate cancer tumor suppression in the mouse. Nat Genet 2001; 27:222–224. Morrison RF, Farmer SR. Role of PPARg in regulating a cascade expression of cyclin-dependent kinase inhibitors, p18(INK4c) and p21(Waf1/Cip1), during adipogenesis. J Biol Chem 1999; 274:17088– 17097. Martin-Caballero J, Flores JM, Garcia-Palencia P, Serrano M. Tumor susceptibility of p21(Waf1/Cip1)-deficient mice. Cancer Res 2001; 61: 6234–6238. Cheng M, Olivier P, Diehl JA, Fero M, Roussel MF, Roberts JM, Sherr CJ. The p21(Cip1) and p27(Kip1) CDK ‘inhibitors’ are essential activators of cyclin D-dependent kinases in murine fibroblasts. EMBO J 1999; 18:1571–1583. Sherr CJ, Roberts JM. CDK inhibitors: positive and negative regulators of G1-phase progression. Genes Dev 1999; 13:1501–1512. Bryja V, Pachernik J, Faldikova L, Krejci P, Pogue R, Nevriva I, Dvorak P, Hampl A. The role of p27(Kip1) in maintaining the levels of D-type cyclins in vivo. Biochim Biophys Acta 2004; 1691:105– 116.