Disruption of Growth Hormone Signaling Retards Early Stages of Prostat
http://www.100md.com
《内分泌学杂志》
Departments of Medicinal Chemistry and Pharmacognosy (Z.W., S.M.S.) and Urology (G.S.P.), Surgical Oncology (S.M.S., K.T.Ch.), Math, Statistics
Computer Science (S.H.)
Medicine (T.G.U.), University of Illinois at Chicago
Department of Veterans Affairs Jesse Brown Medical Center (T.G.U.), Chicago, Illinois 60612
Edison Biotechnology Institute and Department of Biomedical Sciences (K.T.Co., J.J.K.), Ohio University, Athens, Ohio 45701
Laboratory of Cell Regulation and Carcinogenesis (J.E.G.), National Cancer Institute, Bethesda, Maryland 20892
Provident Hospital of Cook County (V.H.R.), Chicago, Illinois 60615
Abstract
Recent epidemiological studies suggest that elevated serum titers of IGF-I, which are, to a large degree, regulated by GH, are associated with an increase in prostate cancer risk. The purpose of the current study was to develop the first animal models to directly test the hypothesis that a normal, functional GH/IGF-I axis is required for prostate cancer progression. The GH receptor (GHR) gene-disrupted mouse (Ghr–/–), which has less than 10% of the plasma IGF-I found in GHR wild-type mice, was crossed with the C3(1)/T antigen (Tag) mouse, which develops prostatic intraepithelial neoplasia driven by the large Tag that progress to invasive prostate carcinoma in a manner similar to the process observed in humans. Progeny of this cross were genotyped and Tag/Ghr+/+ and Tag/Ghr–/– mice were killed at 9 months of age. Seven of eight Tag/Ghr+/+ mice harbored prostatic intraepithelial neoplasia lesions of various grades. In contrast, only one of the eight Tag/Ghr–/– mice exhibited atypia (P < 0.01, Fischer’s exact test). Disruption of the GHR gene altered neither prostate androgen receptor expression nor serum testosterone titers. Expression of the Tag oncogene was similar in the prostates of the two mouse strains. Immunohistochemistry revealed a significant decrease in prostate epithelial cell proliferation and an increase in basal apoptotic indices. These results indicate that disruption of GH signaling significantly inhibits prostate carcinogenesis.
Introduction
PROSTATE CANCER IS the most common and second deadliest form of cancer afflicting American men (1). Androgens are important regulators of prostate proliferation, differentiation, and apoptosis, and androgen antagonism remains the primary treatment for prostate cancer. Although initially effective, most patients’ tumors reemerge as androgen-independent disease. Clearly, other treatment modalities are urgently needed. Recent clinical and epidemiologic studies suggest that GH and IGF-I are important for normal human prostate growth as well as prostate cancer. For example, chronic GH deficiency in adulthood is associated with reduced prostate volume (2). The prevalence of prostate hyperplasia has been reported to be lower in GH-deficient patients than controls (3). Also, acromegalic subjects are known to have enlarged prostates that decrease significantly in size on treatment with somatostatin analogs or surgery to lower GH levels (4).
Many of the biological activities of GH are mediated by IGF-I. IGF-I is unique among growth factors in that it is also an endocrine hormone. GH stimulates IGF-I production in the liver and peripheral tissues. In the blood, IGFs are bound to high-affinity IGF binding proteins that serve as both carrier proteins and modulators of IGF bioactivity. The major IGF binding protein is IGF binding protein-3, which accounts for more than 75% of the bound IGF in the circulation. The endocrine aspect of IGF physiology has facilitated epidemiologic studies on the relationship of circulating levels of IGF-I and their binding proteins to cancer risk.
A number of epidemiologic studies have been conducted to evaluate the role of the GH/IGF axis in prostate carcinogenesis (5). Whereas three prospective studies found a positive association between serum IGF-I and prostate cancer risk (6, 7, 8), one prospective study reported an inverse relationship (9). Case-control studies are also divided, with some studies suggesting that elevated serum IGF-I is associated with increased prostate cancer risk (10, 11, 12, 13, 14, 15, 16), whereas others find little or no association (17, 18, 19, 20, 21). Both metaanalyses of the literature conducted to date agree that there is a positive association between serum IGF-I level and prostate cancer risk (22, 23). With regard to GH, a recent case-control study suggests that elevated basal GH serum titers lowered prostate cancer risk (24). Therefore, whereas epidemiologic studies suggest that the GH/IGF axis may influence carcinogenesis in the prostate, more work is needed to clarify this issue. Animal models in which hormone signaling can be better controlled may be helpful in determining the role of the GH/IGF axis in prostate carcinogenesis.
The GH/IGF system has proven to be important in regulating proliferation of cancer cells in laboratory-based studies. Human prostate cancer cell lines such as LNCaP and PC3 express GH receptors (GHRs) at levels greater than observed in normal tissue (25). Pollak et al. (26) reported that the growth of androgen-independent PC3 cells is slowed in GH-deficient Little mice (Ghrhrlit/lit) relative to control mice. Schally et al. (27) published studies recently that demonstrate an inhibitory effect of GHRH antagonists on the growth of human prostate cancer cells, including androgen-independent lines, in immunodeficient mice. For example, these investigators reported that the GHRH antagonist MZ-4-71 decreased IGF-I levels in not only serum of treated animals but also the tumors (28). These studies suggest that the GH/IGF axis is important for the growth of advanced human prostate cancers.
Whereas previous laboratory-based studies have shown that disruption of the GH/IGF axis can inhibit the proliferation of advanced human prostate cancers propagated either in vitro as cell cultures or in vivo as xenografts in immunodeficient mice, we asked in the present studies whether the GH/IGF axis plays a role in the progression of prostate cancers from initiated cells to preneoplastic lesions. Given the role that the GH/IGF-I axis plays in regulating prostatic cell proliferation and differentiation and prostate gland growth and differentiation, we hypothesized that an intact GH/IGF-I axis is required for prostate cancer cells harboring the T antigen (Tag) oncogene to progress to prostatic intraepithelial neoplasia (PIN). Our approach was to cross the Laron mouse, in which the gene coding for both GHR and GH binding protein has been disrupted or knocked out, with the C3(1)/Tag mouse, which develops prostate cancers due to the Tag oncogene it harbors. We chose the C3(1)/Tag mouse from the many available transgenic models of prostate cancer because it develops prostate cancer relatively slowly, and disease progression from low-grade PIN through high-grade PIN to invasive carcinoma is well characterized (29). This aspect of the model makes it particularly well suited for studies on the prevention of prostate cancer progression.
Materials and Methods
Animals
All studies involving animals were conducted in accordance with mandated standards of humane care as stipulated in the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals (30). Furthermore, the Institutional Animal Care and Use Committee of the University of Illinois at Chicago approved all experimental protocols involving animals before the initiation of any procedures. All the animals were bred at the Biologic Resources Laboratory (University of Illinois at Chicago). They were fed Teklad 8640 diet (Harlan Teklad, Madison, WI) and given water ad libitum and housed in a temperature- and humidity-controlled environment with a regular light/dark illumination cycle (lights on at 0600 h and off at 1900 h).
The C3(1)/Tag transgenic mouse was developed by Green and colleagues (31) in an FVB/N background. The transgene includes the early region of simian virus 40 with the large tumor antigen. Expression is targeted to the prostate by the 5'-flanking region of the rat C3(1) gene. By 8 wk of age, male mice develop foci of PIN that appear identical with human PIN (29).
The GHR and GH binding protein are encoded by a single GHR/binding protein gene in mammalian species (32). Homozygous GHR/binding protein knockout mice (referred to here as Ghr–/–) display postnatal growth retardation, proportionate dwarfism, and absence of the GHR and binding protein. Serum GH levels for Ghr–/– mice are greatly elevated, compared with either Ghr+/+ or Ghr+/– mice (33). Serum IGF-I levels in Ghr–/– mice are decreased by about 90%, compared with Ghr+/+ and Ghr+/– mice (34). No other abnormalities are evident in the homozygous or heterozygous knockout mice; their behavior is indistinguishable from that of their wild-type littermates, and lactation in Ghr+/– mice is adequate to feed their young.
For the current studies, FVB mice heterozygous for the C3(1)/Tag transgene were crossed with Ghr–/– mice of a BALB/c background. Genotyping was conducted by PCR as previously described (34). Offspring of this initial cross were used to generate mice for the current studies that carried the C3(1)/Tag transgene in the presence (Tag/Ghr+/+) or absence (Tag/Ghr–/–) of a wild-type GHR gene.
Histopathology
Male Tag/Ghr+/+ mice and Tag/Ghr–/– mice were killed by CO2 asphyxiation at 38 wk of age. The entire genitourinary bloc (prostate lobes, seminal vesicles ampullary glands, proximal ductus deferens, bladder, and proximal urethra), were excised and fixed in 10% neutral buffered formalin. The lobes of prostates were dissected with the aid of a dissecting microscope and were embedded in paraffin. Two sets of sections, each consisting of at least 15 serial sections, were cut from each block. The sets of sections were separated by at least 100 μm within each block. Four-micrometer sections were placed on SuperFrost/Plus slides (Fisher Scientific Co., Pittsburgh, PA) and stained with hematoxylin and eosin, visualization of dorsal prostate, lateral prostate, and ventral prostate (VP). The slides were read by a board-certified pathologist and a second pathologist experienced in rodent prostate pathology. Both pathologists were blinded to the genotype of the specimens. The dorsolateral and ventral lobes were separately analyzed for absence or presence of hyperplasia, nuclear atypia characterized as PIN, microinvasion, and tumors according to criteria established by the Mouse Models of Human Cancer Consortium Prostate Cancer Committee (35). The observed PIN lesions were further divided into two stages, low grade and high grade, as described by Shibata et al. (29). Finally, the area of PIN lesions within each specimen was measured using image analysis software (MetaVue, Universal Imaging, Downingtown, PA).
Immunohistochemistry
Expression of androgen receptor (AR), Tag, p63, cytokeratin 18 (CK18), mouse dorsolateral prostate (mDLP) and proliferating cell nuclear antigen (PCNA) were evaluated by immunohistochemistry following protocols by the respective antibody vendors or literature (36). Tissues were fixed and embedded using methods described above for histopathology. Paraffin sections were heat immobilized (60 C, 1 h), deparaffinized in three changes of xylene, hydrated in a series of graded ethanol, and rinsed in several changes of distilled water. Heat-induced antigen retrieval was performed in either a microwave oven (Tag) or a pressure cooker (p63, CK18, mDLP, AR, and PCNA; decloaking chamber; Biocare Medical, Concord, CA). Endogenous peroxidase was quenched using H2O2 (3%, 10 min). After blocking with the appropriate serum, the prostate sections were treated with mouse anti-Tag (1:50, PAb101; BD PharMingen, San Diego, CA), rabbit anti-p63 (1:500, H137; Santa Cruz Biotechnology, Santa Cruz, CA), sheep anti-CK18 (1:200, PH504; Binding Site, Birmingham, UK), rabbit anti-mDLP (1:5000), rabbit anti-AR (2 μg/ml, PG21), or mouse anti-PCNA (1:80, PC-10; Oncogene Research Products, San Diego, CA) and sequentially with secondary antibodies and Vectastain Elite ABC kit (Vector Laboratories, Burlingame, CA) for rabbit and sheep primary antibody or Vector M.O.M kit (Vector) for mouse primary antibodies. Sections were rinsed in PBS and incubated with 3,3'-diamobenzidine (Sigma Chemical, St. Louis, MO). Slides were counterstained in hematoxylin, dehydrated in graded ethanol, cleared in xylene, and mounted using Permount mounting medium. Tissue specimens from each genotype were processed together to eliminate interassay variability as a confounding factor in analysis. Apoptosis was assessed by the terminal deoxynucleotide-transferase-mediated deoxyuridine 5-triphosphate-digoxigenin nick end labeling (TUNEL) assay using ApopTag apoptosis detection systems (Serologicals Corp., Norcross, GA) according to the manufacturer’s protocol. To compare the cell proliferation and apoptosis levels in the Tag/Ghr+/+ mice and Tag/Ghr–/– mice, the number of PCNA-immunoreactive cells staining positively in the nucleus per 1000 cells were scored in normal-appearing prostate epithelium. Sampling was done by two independent pathologists who randomly selected fields of normal prostate to score. Both investigators were blinded as to the genotype of the specimens.
Serum testosterone
For each group of adult (19–22 wk of age) male Ghr+/+, Ghr–/–, or Tag/Ghr+/+ mice, blood was obtained from the aorta under ketamine/xylazine anesthesia. Serum samples were stored at –20 C until RIA for testosterone (Coat-A-Count total testosterone; Diagnostic Products Corp., Los Angeles, CA).
Statistical analysis
All the data are presented as means ± SEM. The significance of intergroup differences in serum hormone levels, cell proliferation levels, and apoptosis levels were analyzed using one-way ANOVA, two-tailed t test, or two-sided individual t test, respectively, unless otherwise indicated.
Results
Characteristics of experimental animals
As expected, body weight and length were reduced in Tag/Ghr–/–, compared with age-matched Tag/Ghr+/+ mice, at 9 wk of age (n = 10, Fig. 1, A and C). The seminal vesicles, coagulating gland (data not shown), ventral prostate, and dorsolateral prostate were all present and reduced in size but of normal appearance in Tag/Ghr–/– mice, compared with Tag/Ghr+/+ mice at 9 wk of age (Fig. 1B) and in mature animals of 38 wk of age (not shown). The average ventral prostate and dorsolateral prostate weights were significantly lower (P < 0.0001) in Tag/Ghr–/– mice than in Tag/Ghr+/+ mice (Fig. 1C). However, no significant difference in the prostate to body weight ratio was observed between Tag/Ghr–/– mice and Tag/Ghr+/+ mice (Fig. 1D), indicating that the reduction in prostate weight is proportionate to the reduction in body weight, consistent with an effect of reduced GH action.
Prostate carcinogenesis is blocked by disruption of the GHR
Mice were killed at 38 wk of age, and their prostates were dissected as described in Materials and Methods. This time point was chosen because previous studies indicate that male C3(1)/Tag mice develop prostate cancers beginning at 7 months of age and that by 8 months of age, the majority of mice had developed prostate cancers. Killing the mice at 38 wk of age (about 9.5 months) was chosen for the current studies in the hope that all control mice would have developed prostate cancers by this time point. The serial sections of prostate lobes were examined histologically for PIN lesions as described previously (29) (Fig. 2, A–D). Seven of eight Tag/Ghr+/+mice harbored PIN lesions of various grades in the dorsolateral and ventral lobes (Fig. 2, A and C). For the dorsolateral lobes, two had low-grade PIN, four had combined low-grade PIN and high-grade PIN lesions, and one had exclusively high-grade PIN. For the ventral lobes, three had low-grade PIN and three had combined low-grade PIN and high-grade PIN lesions. In contrast, of the eight Tag/Ghr–/– mice, only one harbored low- and high-grade PIN lesions, taking up 5% of the prostate (Fig. 2, B and D). This change in incidence was highly significant (P < 0.01) as determined by Fischer’s exact test (Fig. 2E). The area of PIN was also significantly higher (P < 0.04) in all lobes of Tag/Ghr+/+ than in Tag/Ghr–/– mice (Fig. 2F). However, no tumors were detected in either group of animals.
Expression of Tag
Tag/Ghr+/+ and Tag/Ghr–/– mice harbored few cells in normal-appearing prostate epithelium that stained positive for Tag. However, the number of immunoreactive epithelial cells increased progressively from low-grade PIN (Fig. 3, A and B) to high-grade PIN (Fig. 3, C and D), which is consistent with the findings of Shibata et al. (29) for the original C3(1)/Tag mouse. When Tag expression was compared between PIN lesions of similar severity, no difference in the degree of Tag expression was observed between Tag/Ghr+/+ mice and Tag/Ghr–/– mice (Fig. 3). Therefore, the lack of the GHR in this model does not appear to affect Tag expression in prostate epithelium.
Neither Tag expression nor disruption of GH signaling alters testosterone levels or AR expression
Serum testosterone levels were analyzed in groups of adult (19–22 wk) male Ghr+/+, Ghr–/–, and Tag/Ghr+/+ mice. Testosterone levels were not affected by either Tag (Tag/Ghr+/+ vs. Ghr+/+) or GHR (Ghr+/+ vs. Ghr–/–) expression status as determined by ANOVA (Fig. 4). Furthermore, immunohistochemical analysis of AR demonstrated that there was no difference in AR expression between Tag/Ghr+/+ and Tag/Ghr–/– mice in normal or cancerous prostate epithelial cells (Fig. 5).
Analysis of markers of prostate epithelial cell differentiation
To study the effect of GH signaling on prostate development and differentiation, several biomarkers were evaluated by immunohistochemistry. Markers of prostatic epithelial cell differentiation included p63 for basal cells and CK18 for the luminal cell subpopulation. Functional differentiation was assessed by immunostaining for the mDLP proteins. In Tag/Ghr+/+ prostates, basal cells (p63+) were intermittently localized along the basement membrane in the central and distal regions of the ventral and dorsolateral lobes (Fig. 6A), and this pattern was not affected by the loss of GH signaling in the Tag/Ghr–/– prostates (Fig. 6B). The majority of the prostatic epithelium in both Tag/Ghr+/+ and Tag/Ghr–/– mice stained for CK18, a marker of a differentiated luminal cell (Fig 6, C and D). Furthermore, in both genotypes, mDLP strongly stained in the dorsolateral prostate (Fig. 6, E and F), indicating that functional differentiation of the epithelial cells was not compromised by the loss of GHR.
Cell proliferation and apoptosis
In the C3(1)/Tag mouse, which is one of the parental strains used to generate the current Tag/Ghr model, Shibata et al. (29) reported that the severity of prostate preneoplasia correlated with proliferation and apoptosis of prostate epithelial cells. Due to the low incidence and area of PIN in Tag/Ghr–/– mice, we compared proliferation and apoptosis in normal-appearing prostate epithelial cells, which has potential to develop into PIN, using PCNA and the TUNEL assay, respectively. In normal-appearing prostate epithelium, proliferation was significantly decreased (Fig. 7) and apoptosis significantly increased (Fig. 8) in Tag/Ghr–/– mice, compared with Tag/Ghr+/+.
Discussion
PIN lesions are thought to be precursors to prostate cancer in both man and rodents (35, 37). Morphologically, high-grade PIN and prostate cancer share spatial distribution and cytological characteristics. The transition between high-grade PIN and areas of prostatic adenocarcinoma suggest a progression of prostatic neoplasia from a noninvasive into an invasive form, with high-grade PIN representing the noninvasive phase (38).
Shibata and colleagues (29) have shown that progression of PIN to invasive prostate carcinoma in the C3(1)/Tag mouse is similar to that observed in man. Our present study suggests that Tag/Ghr–/– mice, in which GH signaling has been disrupted, are resistant to prostate carcinogenesis. The Tag/Ghr– /– mouse developed PIN at a lower incidence and longer latency than the parental C3(1)/Tag strain. The Tag/Ghr+/+ animals did not develop any prostate cancers by 9 months of age, whereas the C3(1)/Tag mouse has been reported to develop prostate cancers by 7 months (31). These differences are likely caused by genetic factors introduced by mating the C3(1)/Tag mouse, which is an FVB/N background, with the GHR knockout mouse, which is derived from BALB/c. Others have reported that genetic background can affect the penetrance of the C3(1)/Tag construct (39, 40). This was not a confounding issue in the current study because all mice used were derived identically (i.e. by crossing the C3(1)/Tag mouse with the Laron mouse). Nevertheless, the new model presented here demonstrates that the loss of GHR produced a significant reduction in the level of PIN in the ventral as well dorsal-lateral lobes in terms of incidence and PIN area.
Tag expression was evaluated by immunohistochemical analysis. As described by Shibata et al. (29), expression was detected at very low levels in normal epithelial cells of the prostate but increased in low-grade PIN and high-grade PIN in the Tag/Ghr+/+ mice. A similar pattern of Tag expression is seen in the Tag/Ghr–/– mice. Robertson et al. (40) reported that Tag expression was insensitive to prolactin signaling. Here we found that GH signaling was not essential for Tag expression controlled under C3 promoter fragment. Even though fewer lesions are observed in Tag/Ghr–/– mice than Tag/Ghr+/+ mice, the two genotypes had parallel Tag expression levels within each of the various degrees of PIN severity.
As noted above, it is well established that GH is important for prostate growth in full-grown, adult humans (2, 4, 41). Acromegalics have enlarged prostates that shrink to normal size in response to treatments that lower GH serum levels. Furthermore, the prostate shrinks to below normal volume in acromegalics rendered GH deficient due to aggressive therapy (2). Data from the current study suggest that GH is also important for prostate growth in the mouse. Disruption of the GHR gene in Tag/Ghr–/– mice resulted in a 60% decrease in prostate weight relative to their Tag/Ghr+/+ siblings, and the decrease in prostate weight was proportional to the reduction in overall body weight, consistent with a proportional effect of disrupted GH action on prostate and body weight. Importantly, loss of GH signaling did not appear to affect epithelial cell cyto- or functional differentiation as revealed by similar expression levels and pattern of p63, CK18, and mDLP in the Tag/Ghr+/+ and Tag/Ghr–/– mice. Thus, changes in carcinogenesis between the two genotypes are not likely to be a function of altered epithelial cell differentiation.
Androgens play a critical role in prostate growth, development, and carcinogenesis, and the androgen pathway has been the target of first-line prostate therapies for many years. We asked whether disrupting GH signaling resulted in a down-regulation of androgens or the expression of the AR, which could explain the lack of carcinogenesis in the Tag/ Ghr–/– mice. However, as shown in Fig. 4, serum testosterone levels were affected by neither the presence of Tag (Tag/Ghr+/+ vs. Ghr+/+) nor disruption of the GHR (Ghr+/+ vs. Ghr–/–). Furthermore, AR expression was not compromised in the prostate epithelium of Ghr–/– mice relative to controls (Fig. 5). Therefore, we concluded that the protective effect afforded by disrupting GH signaling is independent of either serum testosterone or AR expression. This is of significance clinically because prostate cancers that initially respond to antiandrogen therapies often evolve into androgen-independent disease, which is currently incurable.
Recently Ormandy and colleagues (40) crossed the C3(1)/Tag mouse used in the preset studies with the prolactin receptor knockout mouse (Prlr–/–) and evaluated prostate carcinogenesis at 50 wk of age. Whereas there was no difference in PIN area in the dorsal prostate lobes, PIN area in the ventral prostate were significantly reduced in Prlr–/– mice relative to control mice. Furthermore, whereas one of 11 Prlr+/+ mice and four of 21 Prlr+/– mice harbored prostate tumors, no prostate tumors were observed in any of the Prlr–/– mice (40). These data indicate that disruption of PRL signaling can impede mouse prostate carcinogenesis. However, PRL levels in the Ghr–/– mouse are not reduced (34), suggesting that disruption of PRL signaling is not responsible for protection from PIN development in Tag/Ghr–/– mice.
Carcinogenesis is characterized by dysregulated cell proliferation or apoptosis. The GH/IGF axis plays an important role in regulating prostate epithelial cell proliferation and apoptosis both in vitro and in vivo (42). One of the parental strains of the Tag/Ghr mouse presented in this communication is the Laron mouse, which, in addition to lacking a functional GHR, also has only about 10% of the serum IGF-I present in wild-type mice (33). We therefore hypothesized that the prostate epithelial cells of Tag/Ghr+/+ mice would have a significant proliferation advantage, compared with Tag/Ghr–/– mice, resulting in more rapid progression of carcinogenesis. Our results indicate that proliferation is significantly lower and apoptosis is significantly higher in the prostate epithelium of Tag/Ghr–/– mice, compared with Tag/Ghr+/+ mice (Figs. 7 and 8). Because all the prostate cells of both groups of mice harbor the same oncogene (Tag), the observed difference in prostate cell proliferation and apoptosis is likely to have a significant impact on prostate carcinogenesis.
In summary, we have crossed the C3(1)/Tag mouse with the GHR/binding protein knockout (Laron) mouse, resulting in a model in which prostate cancer progression can be assessed in the presence or absence of GH signaling. The data indicate that progression of Tag-initiated prostate epithelium is significantly inhibited in the absence of GH signaling. This inhibition is not due to insufficient Tag expression or androgen signaling in Tag/Ghr–/– mice relative to Tag/Ghr+/+ mice. Rather, cancer inhibition appears to be associated with decreased proliferation and increased apoptosis of the prostate epithelium of Tag/Ghr–/– mice. These findings may have important translational implications. It is generally accepted that PINs are precursors to lethal prostate cancers, and these lesions occur at a similar incidence in individuals of populations at either high or low risk for the development of prostate cancer. Thus, the difference in mortality rates between high- and low-risk populations seems to be due to differences in the progression of PIN to prostate cancers. The findings presented here suggest that PIN lesions may require GH signaling for progression, suggesting that the GH signaling pathway or the GH/IGF axis may represent important targets for the development of agents to prevent prostate cancer.
Footnotes
This work was supported by National Institutes of Health (NIH) Grant R03 AG020820 and Department of Defense Grant W81XWH-04-1-0201. J.J.K. is supported by the State of Ohio Eminent Scholars’ Program that includes a gift by Milton and Lawrence Goll, DiAthegen LLC, and NIH.
First Published Online September 1, 2005
Abbreviations: AR, Androgen receptor; CK18, cytokeratin 18; GHR, GH receptor; mDLP, mouse dorsolateral prostate; PCNA, proliferating cell nuclear antigen; PIN, prostatic intraepithelial neoplasia; Tag, T antigen; TUNEL, terminal deoxynucleotide-transferase-mediated deoxyuridine 5-triphosphate-digoxigenin nick end labeling; VP, ventral prostate.
Accepted for publication August 22, 2005.
References
Jemal A, Murray T, Ward E, Samuels A, Tiwari RC, Ghafoor A, Feuer EJ, Thun MJ 2005 Cancer statistics, 2005. CA Cancer J Clin 55:10–30
Colao A, Di Somma C, Spiezia S, Filippella M, Pivonello R, Lombardi G 2003 Effect of growth hormone (GH) and/or testosterone replacement on the prostate in GH-deficient adult patients. J Clin Endocrinol Metab 88:88–94
Colao A, Spiezia S, Di Somma C, Marzullo P, Cerbone G, Pivonello R, Faggiano A, Lombardi G 2000 Effect of GH and/or testosterone deficiency on the prostate: an ultrasonographic and endocrine study in GH-deficient adult patients. Eur J Endocrinol 143:61–69
Colao A, Marzullo P, Spiezia S, Giaccio A, Ferone D, Cerbone G, Di Sarno A, Lombardi G 2000 Effect of two years of growth hormone and insulin-like growth factor-I suppression on prostate diseases in acromegalic patients. J Clin Endocrinol Metab 85:3754–3761
Roberts C 2004 IGF-I and prostate cancer. Novartis Found Symp 262:193–199
Chan JM, Stampfer MJ, Giovannucci E, Gann PH, Ma J, Wilkinson P, Hennekens CH, Pollak M 1998 Plasma insulin-like growth factor-I and prostate cancer risk: a prospective study. Science 279:563–566
Harman SM, Metter EJ, Blackman MR, Landis PK, Carter HB 2000 Serum levels of insulin-like growth factor I (IGF-I), IGF-II, IGF-binding protein-3, and prostate-specific antigen as predictors of clinical prostate cancer. J Clin Endocrinol Metab 85:4258–4265
Stattin P, Bylund A, Rinaldi S, Biessy C, Dechaud H, Stenman UH, Egevad L, Riboli E, Hallmans G, Kaaks R 2000 Plasma insulin-like growth factor-I, insulin-like growth factor-binding proteins, and prostate cancer risk: a prospective study. J Natl Cancer Inst 92:1910–1917
Woodson K, Tangrea JA, Pollak M, Copeland TD, Taylor PR, Virtamo J, Albanes D 2003 Serum insulin-like growth factor I: tumor marker or etiologic factor A prospective study of prostate cancer among Finnish men. Cancer Res 63:3991–3994
Oliver SE, Gunnell D, Donovan J, Peters TJ, Persad R, Gillatt D, Pearce A, Neal DE, Hamdy FC, Holly J 2004 Screen-detected prostate cancer and the insulin-like growth factor axis: results of a population-based case-control study. Int J Cancer 108:887–892
Mantzoros CS, Tzonou A, Signorello LB, Stampfer M, Trichopoulos D, Adami HO 1997 Insulin-like growth factor 1 in relation to prostate cancer and benign prostatic hyperplasia. Br J Cancer 76:1115–1118
Wolk A, Mantzoros CS, Andersson SO, Bergstrom R, Signorello LB, Lagiou P, Adami HO, Trichopoulos D 1998 Insulin-like growth factor 1 and prostate cancer risk: a population-based, case-control study. J Natl Cancer Inst 90:911–915
Wolk A, Andersson SO, Mantzoros CS, Trichopoulos D, Adami HO 2000 Can measurements of IGF-1 and IGFBP-3 improve the sensitivity of prostate-cancer screening Lancet 356:1902–1903
Khosravi J, Diamandi A, Mistry J, Scorilas A 2001 Insulin-like growth factor I (IGF-I) and IGF-binding protein-3 in benign prostatic hyperplasia and prostate cancer. J Clin Endocrinol Metab 86:694–699
Chan JM, Stampfer MJ, Ma J, Gann P, Gaziano JM, Pollak M, Giovannucci E 2002 Insulin-like growth factor-I (IGF-I) and IGF binding protein-3 as predictors of advanced-stage prostate cancer. J Natl Cancer Inst 94:1099–1106
Li L, Yu H, Schumacher F, Casey G, Witte JS 2003 Relation of serum insulin-like growth factor-I (IGF-I) and IGF binding protein-3 to risk of prostate cancer (United States). Cancer Causes Control 14:721–726
Cutting CW, Hunt C, Nisbet JA, Bland JM, Dalgleish AG, Kirby RS 1999 Serum insulin-like growth factor-1 is not a useful marker of prostate cancer. BJU Int 83:996–999
Janssen JA, Wildhagen MF, Ito K, Blijenberg BG, Van Schaik RH, Roobol MJ, Pols HA, Lamberts SW, Schroder FH 2004 Circulating free insulin-like growth factor (IGF)-I, total IGF-I, and IGF binding protein-3 levels do not predict the future risk to develop prostate cancer: results of a case-control study involving 201 patients within a population-based screening with a 4-year interval. J Clin Endocrinol Metab 89:4391–4396
Chen C, Lewis SK, Voigt L, Fitzpatrick A, Plymate SR, Weiss NS 2005 Prostate carcinoma incidence in relation to prediagnostic circulating levels of insulin-like growth factor I, insulin-like growth factor binding protein 3, and insulin. Cancer 103:76–84
Kurek R, Tunn UW, Eckart O, Aumuller G, Wong J, Renneberg H 2000 The significance of serum levels of insulin-like growth factor-1 in patients with prostate cancer. BJU Int 85:125–129
Finne P, Auvinen A, Koistinen H, Zhang WM, Maattanen L, Rannikko S, Tammela T, Seppala M, Hakama M, Stenman UH 2000 Insulin-like growth factor I is not a useful marker of prostate cancer in men with elevated levels of prostate-specific antigen. J Clin Endocrinol Metab 85:2744–2747
Renehan AG, Zwahlen M, Minder C, O’Dwyer ST, Shalet SM, Egger M 2004 Insulin-like growth factor (IGF)-I, IGF binding protein-3, and cancer risk: systematic review and meta-regression analysis. Lancet 363:1346–1353
Shi R, Berkel HJ, Yu H 2001 Insulin-like growth factor-I and prostate cancer: a meta-analysis. Br J Cancer 85:991–996
Fuhrman B, Barba M, Schunemann HJ, Hurd T, Quattrin T, Cartagena R, Carruba G, Muti P 2005 Basal growth hormone concentrations in blood and the risk for prostate cancer: a case-control study. Prostate 64:109–115
Reiter E, Kecha O, Hennuy B, Lardinois S, Klug M, Bruyninx M, Closset J, Hennen G 1995 Growth hormone directly affects the function of the different lobes of the rat prostate. Endocrinology 136:3338–3345
Pollak M, Beamer W, Zhang JC 1998 Insulin-like growth factors and prostate cancer. Cancer Metastasis Rev 17:383–390
Schally AV, Comaru-Schally AM, Plonowski A, Nagy A, Halmos G, Rekasi Z 2000 Peptide analogs in the therapy of prostate cancer. Prostate 45:158–166
Jungwirth A, Schally AV, Pinski J, Halmos G, Groot K, Armatis P, Vadillo-Buenfil M 1997 Inhibition of in vivo proliferation of androgen-independent prostate cancers by an antagonist of growth hormone-releasing hormone. Br J Cancer 75:1585–1592
Shibata MA, Ward JM, Devor DE, Liu ML, Green JE 1996 Progression of prostatic intraepithelial neoplasia to invasive carcinoma in C3(1)/SV40 large T antigen transgenic mice: histopathological and molecular biological alterations. Cancer Res 56:4894–4903
Institute of Laboratory Animal Resources (United States) 1996 Guide for the care and use of laboratory animals. 7th ed. Washington, DC: National Academy Press
Maroulakou IG, Anver M, Garrett L, Green JE 1994 Prostate and mammary adenocarcinoma in transgenic mice carrying a rat C3(1) Simian virus 40 large tumor antigen fusion gene. Proc Natl Acad Sci USA 91:11236–11240
Leung DW, Spencer SA, Cachianes G, Hammonds RG, Collins C, Henzel WJ, Barnard R, Waters MJ, Wood WI 1987 Growth hormone receptor and serum binding protein: purification, cloning and expression. Nature 330:537–543
Zhou Y, Xu BC, Maheshwari HG, He L, Reed M, Lozykowski M, Okada S, Cataldo L, Coschigamo K, Wagner TE, Baumann G, Kopchick JJ 1997 A mammalian model for Laron syndrome produced by targeted disruption of the mouse growth hormone receptor/binding protein gene (the Laron mouse). Proc Natl Acad Sci USA 94:13215–13220
Chandrashekar V, Bartke A, Coschigano KT, Kopchick JJ 1999 Pituitary and testicular function in growth hormone receptor gene knockout mice. Endocrinology 140:1082–1088
Shappell SB, Thomas GV, Roberts RL, Herbert R, Ittmann MM, Rubin MA, Humphrey PA, Sundberg JP, Rozengurt N, Barrios R, Ward JM, Cardiff RD 2004 Prostate pathology of genetically engineered mice: definitions and classification. The consensus report from the Bar Harbor meeting of the Mouse Models of Human Cancer Consortium Prostate Pathology Committee. Cancer Res 64:2270–2305
Donjacour AA, Cunha GR 1993 Assessment of prostatic protein secretion in tissue recombinants made of urogenital sinus mesenchyme and urothelium from normal or androgen-insensitive mice. Endocrinology 132:2342–2350
Bostwick DG, Brawer MK 1987 Prostatic intra-epithelial neoplasia and early invasion in prostate cancer. Cancer 59:788–794
Bostwick DG 2000 Prostatic intraepithelial neoplasia. Curr Urol Rep 1:65–70
Green JE, Shibata MA, Yoshidome K, Liu ML, Jorcyk C, Anver MR, Wigginton J, Wiltrout R, Shibata E, Kaczmarczyk S, Wang W, Liu ZY, Calvo A, Couldrey C 2000 The C3(1)/SV40 T-antigen transgenic mouse model of mammary cancer: ductal epithelial cell targeting with multistage progression to carcinoma. Oncogene 19:1020–1027
Robertson FG, Harris J, Naylor MJ, Oakes SR, Kindblom J, Dillner K, Wennbo H, Tornell J, Kelly PA, Green J, Ormandy CJ 2003 Prostate development and carcinogenesis in prolactin receptor knockout mice. Endocrinology 144:3196–3205
Colao A, Marzullo P, Spiezia S, Ferone D, Giaccio A, Cerbone G, Pivonello R, Di Somma C, Lombardi G 1999 Effect of growth hormone (GH) and insulin-like growth factor I on prostate diseases: an ultrasonographic and endocrine study in acromegaly, GH deficiency, and healthy subjects. J Clin Endocrinol Metab 84:1986–1991
Moschos SJ, Mantzoros CS 2002 The role of the IGF system in cancer: from basic to clinical studies and clinical applications. Oncology 63:317–332(Zhuohua Wang, Gail S. Prins, Karen T. Co)
Computer Science (S.H.)
Medicine (T.G.U.), University of Illinois at Chicago
Department of Veterans Affairs Jesse Brown Medical Center (T.G.U.), Chicago, Illinois 60612
Edison Biotechnology Institute and Department of Biomedical Sciences (K.T.Co., J.J.K.), Ohio University, Athens, Ohio 45701
Laboratory of Cell Regulation and Carcinogenesis (J.E.G.), National Cancer Institute, Bethesda, Maryland 20892
Provident Hospital of Cook County (V.H.R.), Chicago, Illinois 60615
Abstract
Recent epidemiological studies suggest that elevated serum titers of IGF-I, which are, to a large degree, regulated by GH, are associated with an increase in prostate cancer risk. The purpose of the current study was to develop the first animal models to directly test the hypothesis that a normal, functional GH/IGF-I axis is required for prostate cancer progression. The GH receptor (GHR) gene-disrupted mouse (Ghr–/–), which has less than 10% of the plasma IGF-I found in GHR wild-type mice, was crossed with the C3(1)/T antigen (Tag) mouse, which develops prostatic intraepithelial neoplasia driven by the large Tag that progress to invasive prostate carcinoma in a manner similar to the process observed in humans. Progeny of this cross were genotyped and Tag/Ghr+/+ and Tag/Ghr–/– mice were killed at 9 months of age. Seven of eight Tag/Ghr+/+ mice harbored prostatic intraepithelial neoplasia lesions of various grades. In contrast, only one of the eight Tag/Ghr–/– mice exhibited atypia (P < 0.01, Fischer’s exact test). Disruption of the GHR gene altered neither prostate androgen receptor expression nor serum testosterone titers. Expression of the Tag oncogene was similar in the prostates of the two mouse strains. Immunohistochemistry revealed a significant decrease in prostate epithelial cell proliferation and an increase in basal apoptotic indices. These results indicate that disruption of GH signaling significantly inhibits prostate carcinogenesis.
Introduction
PROSTATE CANCER IS the most common and second deadliest form of cancer afflicting American men (1). Androgens are important regulators of prostate proliferation, differentiation, and apoptosis, and androgen antagonism remains the primary treatment for prostate cancer. Although initially effective, most patients’ tumors reemerge as androgen-independent disease. Clearly, other treatment modalities are urgently needed. Recent clinical and epidemiologic studies suggest that GH and IGF-I are important for normal human prostate growth as well as prostate cancer. For example, chronic GH deficiency in adulthood is associated with reduced prostate volume (2). The prevalence of prostate hyperplasia has been reported to be lower in GH-deficient patients than controls (3). Also, acromegalic subjects are known to have enlarged prostates that decrease significantly in size on treatment with somatostatin analogs or surgery to lower GH levels (4).
Many of the biological activities of GH are mediated by IGF-I. IGF-I is unique among growth factors in that it is also an endocrine hormone. GH stimulates IGF-I production in the liver and peripheral tissues. In the blood, IGFs are bound to high-affinity IGF binding proteins that serve as both carrier proteins and modulators of IGF bioactivity. The major IGF binding protein is IGF binding protein-3, which accounts for more than 75% of the bound IGF in the circulation. The endocrine aspect of IGF physiology has facilitated epidemiologic studies on the relationship of circulating levels of IGF-I and their binding proteins to cancer risk.
A number of epidemiologic studies have been conducted to evaluate the role of the GH/IGF axis in prostate carcinogenesis (5). Whereas three prospective studies found a positive association between serum IGF-I and prostate cancer risk (6, 7, 8), one prospective study reported an inverse relationship (9). Case-control studies are also divided, with some studies suggesting that elevated serum IGF-I is associated with increased prostate cancer risk (10, 11, 12, 13, 14, 15, 16), whereas others find little or no association (17, 18, 19, 20, 21). Both metaanalyses of the literature conducted to date agree that there is a positive association between serum IGF-I level and prostate cancer risk (22, 23). With regard to GH, a recent case-control study suggests that elevated basal GH serum titers lowered prostate cancer risk (24). Therefore, whereas epidemiologic studies suggest that the GH/IGF axis may influence carcinogenesis in the prostate, more work is needed to clarify this issue. Animal models in which hormone signaling can be better controlled may be helpful in determining the role of the GH/IGF axis in prostate carcinogenesis.
The GH/IGF system has proven to be important in regulating proliferation of cancer cells in laboratory-based studies. Human prostate cancer cell lines such as LNCaP and PC3 express GH receptors (GHRs) at levels greater than observed in normal tissue (25). Pollak et al. (26) reported that the growth of androgen-independent PC3 cells is slowed in GH-deficient Little mice (Ghrhrlit/lit) relative to control mice. Schally et al. (27) published studies recently that demonstrate an inhibitory effect of GHRH antagonists on the growth of human prostate cancer cells, including androgen-independent lines, in immunodeficient mice. For example, these investigators reported that the GHRH antagonist MZ-4-71 decreased IGF-I levels in not only serum of treated animals but also the tumors (28). These studies suggest that the GH/IGF axis is important for the growth of advanced human prostate cancers.
Whereas previous laboratory-based studies have shown that disruption of the GH/IGF axis can inhibit the proliferation of advanced human prostate cancers propagated either in vitro as cell cultures or in vivo as xenografts in immunodeficient mice, we asked in the present studies whether the GH/IGF axis plays a role in the progression of prostate cancers from initiated cells to preneoplastic lesions. Given the role that the GH/IGF-I axis plays in regulating prostatic cell proliferation and differentiation and prostate gland growth and differentiation, we hypothesized that an intact GH/IGF-I axis is required for prostate cancer cells harboring the T antigen (Tag) oncogene to progress to prostatic intraepithelial neoplasia (PIN). Our approach was to cross the Laron mouse, in which the gene coding for both GHR and GH binding protein has been disrupted or knocked out, with the C3(1)/Tag mouse, which develops prostate cancers due to the Tag oncogene it harbors. We chose the C3(1)/Tag mouse from the many available transgenic models of prostate cancer because it develops prostate cancer relatively slowly, and disease progression from low-grade PIN through high-grade PIN to invasive carcinoma is well characterized (29). This aspect of the model makes it particularly well suited for studies on the prevention of prostate cancer progression.
Materials and Methods
Animals
All studies involving animals were conducted in accordance with mandated standards of humane care as stipulated in the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals (30). Furthermore, the Institutional Animal Care and Use Committee of the University of Illinois at Chicago approved all experimental protocols involving animals before the initiation of any procedures. All the animals were bred at the Biologic Resources Laboratory (University of Illinois at Chicago). They were fed Teklad 8640 diet (Harlan Teklad, Madison, WI) and given water ad libitum and housed in a temperature- and humidity-controlled environment with a regular light/dark illumination cycle (lights on at 0600 h and off at 1900 h).
The C3(1)/Tag transgenic mouse was developed by Green and colleagues (31) in an FVB/N background. The transgene includes the early region of simian virus 40 with the large tumor antigen. Expression is targeted to the prostate by the 5'-flanking region of the rat C3(1) gene. By 8 wk of age, male mice develop foci of PIN that appear identical with human PIN (29).
The GHR and GH binding protein are encoded by a single GHR/binding protein gene in mammalian species (32). Homozygous GHR/binding protein knockout mice (referred to here as Ghr–/–) display postnatal growth retardation, proportionate dwarfism, and absence of the GHR and binding protein. Serum GH levels for Ghr–/– mice are greatly elevated, compared with either Ghr+/+ or Ghr+/– mice (33). Serum IGF-I levels in Ghr–/– mice are decreased by about 90%, compared with Ghr+/+ and Ghr+/– mice (34). No other abnormalities are evident in the homozygous or heterozygous knockout mice; their behavior is indistinguishable from that of their wild-type littermates, and lactation in Ghr+/– mice is adequate to feed their young.
For the current studies, FVB mice heterozygous for the C3(1)/Tag transgene were crossed with Ghr–/– mice of a BALB/c background. Genotyping was conducted by PCR as previously described (34). Offspring of this initial cross were used to generate mice for the current studies that carried the C3(1)/Tag transgene in the presence (Tag/Ghr+/+) or absence (Tag/Ghr–/–) of a wild-type GHR gene.
Histopathology
Male Tag/Ghr+/+ mice and Tag/Ghr–/– mice were killed by CO2 asphyxiation at 38 wk of age. The entire genitourinary bloc (prostate lobes, seminal vesicles ampullary glands, proximal ductus deferens, bladder, and proximal urethra), were excised and fixed in 10% neutral buffered formalin. The lobes of prostates were dissected with the aid of a dissecting microscope and were embedded in paraffin. Two sets of sections, each consisting of at least 15 serial sections, were cut from each block. The sets of sections were separated by at least 100 μm within each block. Four-micrometer sections were placed on SuperFrost/Plus slides (Fisher Scientific Co., Pittsburgh, PA) and stained with hematoxylin and eosin, visualization of dorsal prostate, lateral prostate, and ventral prostate (VP). The slides were read by a board-certified pathologist and a second pathologist experienced in rodent prostate pathology. Both pathologists were blinded to the genotype of the specimens. The dorsolateral and ventral lobes were separately analyzed for absence or presence of hyperplasia, nuclear atypia characterized as PIN, microinvasion, and tumors according to criteria established by the Mouse Models of Human Cancer Consortium Prostate Cancer Committee (35). The observed PIN lesions were further divided into two stages, low grade and high grade, as described by Shibata et al. (29). Finally, the area of PIN lesions within each specimen was measured using image analysis software (MetaVue, Universal Imaging, Downingtown, PA).
Immunohistochemistry
Expression of androgen receptor (AR), Tag, p63, cytokeratin 18 (CK18), mouse dorsolateral prostate (mDLP) and proliferating cell nuclear antigen (PCNA) were evaluated by immunohistochemistry following protocols by the respective antibody vendors or literature (36). Tissues were fixed and embedded using methods described above for histopathology. Paraffin sections were heat immobilized (60 C, 1 h), deparaffinized in three changes of xylene, hydrated in a series of graded ethanol, and rinsed in several changes of distilled water. Heat-induced antigen retrieval was performed in either a microwave oven (Tag) or a pressure cooker (p63, CK18, mDLP, AR, and PCNA; decloaking chamber; Biocare Medical, Concord, CA). Endogenous peroxidase was quenched using H2O2 (3%, 10 min). After blocking with the appropriate serum, the prostate sections were treated with mouse anti-Tag (1:50, PAb101; BD PharMingen, San Diego, CA), rabbit anti-p63 (1:500, H137; Santa Cruz Biotechnology, Santa Cruz, CA), sheep anti-CK18 (1:200, PH504; Binding Site, Birmingham, UK), rabbit anti-mDLP (1:5000), rabbit anti-AR (2 μg/ml, PG21), or mouse anti-PCNA (1:80, PC-10; Oncogene Research Products, San Diego, CA) and sequentially with secondary antibodies and Vectastain Elite ABC kit (Vector Laboratories, Burlingame, CA) for rabbit and sheep primary antibody or Vector M.O.M kit (Vector) for mouse primary antibodies. Sections were rinsed in PBS and incubated with 3,3'-diamobenzidine (Sigma Chemical, St. Louis, MO). Slides were counterstained in hematoxylin, dehydrated in graded ethanol, cleared in xylene, and mounted using Permount mounting medium. Tissue specimens from each genotype were processed together to eliminate interassay variability as a confounding factor in analysis. Apoptosis was assessed by the terminal deoxynucleotide-transferase-mediated deoxyuridine 5-triphosphate-digoxigenin nick end labeling (TUNEL) assay using ApopTag apoptosis detection systems (Serologicals Corp., Norcross, GA) according to the manufacturer’s protocol. To compare the cell proliferation and apoptosis levels in the Tag/Ghr+/+ mice and Tag/Ghr–/– mice, the number of PCNA-immunoreactive cells staining positively in the nucleus per 1000 cells were scored in normal-appearing prostate epithelium. Sampling was done by two independent pathologists who randomly selected fields of normal prostate to score. Both investigators were blinded as to the genotype of the specimens.
Serum testosterone
For each group of adult (19–22 wk of age) male Ghr+/+, Ghr–/–, or Tag/Ghr+/+ mice, blood was obtained from the aorta under ketamine/xylazine anesthesia. Serum samples were stored at –20 C until RIA for testosterone (Coat-A-Count total testosterone; Diagnostic Products Corp., Los Angeles, CA).
Statistical analysis
All the data are presented as means ± SEM. The significance of intergroup differences in serum hormone levels, cell proliferation levels, and apoptosis levels were analyzed using one-way ANOVA, two-tailed t test, or two-sided individual t test, respectively, unless otherwise indicated.
Results
Characteristics of experimental animals
As expected, body weight and length were reduced in Tag/Ghr–/–, compared with age-matched Tag/Ghr+/+ mice, at 9 wk of age (n = 10, Fig. 1, A and C). The seminal vesicles, coagulating gland (data not shown), ventral prostate, and dorsolateral prostate were all present and reduced in size but of normal appearance in Tag/Ghr–/– mice, compared with Tag/Ghr+/+ mice at 9 wk of age (Fig. 1B) and in mature animals of 38 wk of age (not shown). The average ventral prostate and dorsolateral prostate weights were significantly lower (P < 0.0001) in Tag/Ghr–/– mice than in Tag/Ghr+/+ mice (Fig. 1C). However, no significant difference in the prostate to body weight ratio was observed between Tag/Ghr–/– mice and Tag/Ghr+/+ mice (Fig. 1D), indicating that the reduction in prostate weight is proportionate to the reduction in body weight, consistent with an effect of reduced GH action.
Prostate carcinogenesis is blocked by disruption of the GHR
Mice were killed at 38 wk of age, and their prostates were dissected as described in Materials and Methods. This time point was chosen because previous studies indicate that male C3(1)/Tag mice develop prostate cancers beginning at 7 months of age and that by 8 months of age, the majority of mice had developed prostate cancers. Killing the mice at 38 wk of age (about 9.5 months) was chosen for the current studies in the hope that all control mice would have developed prostate cancers by this time point. The serial sections of prostate lobes were examined histologically for PIN lesions as described previously (29) (Fig. 2, A–D). Seven of eight Tag/Ghr+/+mice harbored PIN lesions of various grades in the dorsolateral and ventral lobes (Fig. 2, A and C). For the dorsolateral lobes, two had low-grade PIN, four had combined low-grade PIN and high-grade PIN lesions, and one had exclusively high-grade PIN. For the ventral lobes, three had low-grade PIN and three had combined low-grade PIN and high-grade PIN lesions. In contrast, of the eight Tag/Ghr–/– mice, only one harbored low- and high-grade PIN lesions, taking up 5% of the prostate (Fig. 2, B and D). This change in incidence was highly significant (P < 0.01) as determined by Fischer’s exact test (Fig. 2E). The area of PIN was also significantly higher (P < 0.04) in all lobes of Tag/Ghr+/+ than in Tag/Ghr–/– mice (Fig. 2F). However, no tumors were detected in either group of animals.
Expression of Tag
Tag/Ghr+/+ and Tag/Ghr–/– mice harbored few cells in normal-appearing prostate epithelium that stained positive for Tag. However, the number of immunoreactive epithelial cells increased progressively from low-grade PIN (Fig. 3, A and B) to high-grade PIN (Fig. 3, C and D), which is consistent with the findings of Shibata et al. (29) for the original C3(1)/Tag mouse. When Tag expression was compared between PIN lesions of similar severity, no difference in the degree of Tag expression was observed between Tag/Ghr+/+ mice and Tag/Ghr–/– mice (Fig. 3). Therefore, the lack of the GHR in this model does not appear to affect Tag expression in prostate epithelium.
Neither Tag expression nor disruption of GH signaling alters testosterone levels or AR expression
Serum testosterone levels were analyzed in groups of adult (19–22 wk) male Ghr+/+, Ghr–/–, and Tag/Ghr+/+ mice. Testosterone levels were not affected by either Tag (Tag/Ghr+/+ vs. Ghr+/+) or GHR (Ghr+/+ vs. Ghr–/–) expression status as determined by ANOVA (Fig. 4). Furthermore, immunohistochemical analysis of AR demonstrated that there was no difference in AR expression between Tag/Ghr+/+ and Tag/Ghr–/– mice in normal or cancerous prostate epithelial cells (Fig. 5).
Analysis of markers of prostate epithelial cell differentiation
To study the effect of GH signaling on prostate development and differentiation, several biomarkers were evaluated by immunohistochemistry. Markers of prostatic epithelial cell differentiation included p63 for basal cells and CK18 for the luminal cell subpopulation. Functional differentiation was assessed by immunostaining for the mDLP proteins. In Tag/Ghr+/+ prostates, basal cells (p63+) were intermittently localized along the basement membrane in the central and distal regions of the ventral and dorsolateral lobes (Fig. 6A), and this pattern was not affected by the loss of GH signaling in the Tag/Ghr–/– prostates (Fig. 6B). The majority of the prostatic epithelium in both Tag/Ghr+/+ and Tag/Ghr–/– mice stained for CK18, a marker of a differentiated luminal cell (Fig 6, C and D). Furthermore, in both genotypes, mDLP strongly stained in the dorsolateral prostate (Fig. 6, E and F), indicating that functional differentiation of the epithelial cells was not compromised by the loss of GHR.
Cell proliferation and apoptosis
In the C3(1)/Tag mouse, which is one of the parental strains used to generate the current Tag/Ghr model, Shibata et al. (29) reported that the severity of prostate preneoplasia correlated with proliferation and apoptosis of prostate epithelial cells. Due to the low incidence and area of PIN in Tag/Ghr–/– mice, we compared proliferation and apoptosis in normal-appearing prostate epithelial cells, which has potential to develop into PIN, using PCNA and the TUNEL assay, respectively. In normal-appearing prostate epithelium, proliferation was significantly decreased (Fig. 7) and apoptosis significantly increased (Fig. 8) in Tag/Ghr–/– mice, compared with Tag/Ghr+/+.
Discussion
PIN lesions are thought to be precursors to prostate cancer in both man and rodents (35, 37). Morphologically, high-grade PIN and prostate cancer share spatial distribution and cytological characteristics. The transition between high-grade PIN and areas of prostatic adenocarcinoma suggest a progression of prostatic neoplasia from a noninvasive into an invasive form, with high-grade PIN representing the noninvasive phase (38).
Shibata and colleagues (29) have shown that progression of PIN to invasive prostate carcinoma in the C3(1)/Tag mouse is similar to that observed in man. Our present study suggests that Tag/Ghr–/– mice, in which GH signaling has been disrupted, are resistant to prostate carcinogenesis. The Tag/Ghr– /– mouse developed PIN at a lower incidence and longer latency than the parental C3(1)/Tag strain. The Tag/Ghr+/+ animals did not develop any prostate cancers by 9 months of age, whereas the C3(1)/Tag mouse has been reported to develop prostate cancers by 7 months (31). These differences are likely caused by genetic factors introduced by mating the C3(1)/Tag mouse, which is an FVB/N background, with the GHR knockout mouse, which is derived from BALB/c. Others have reported that genetic background can affect the penetrance of the C3(1)/Tag construct (39, 40). This was not a confounding issue in the current study because all mice used were derived identically (i.e. by crossing the C3(1)/Tag mouse with the Laron mouse). Nevertheless, the new model presented here demonstrates that the loss of GHR produced a significant reduction in the level of PIN in the ventral as well dorsal-lateral lobes in terms of incidence and PIN area.
Tag expression was evaluated by immunohistochemical analysis. As described by Shibata et al. (29), expression was detected at very low levels in normal epithelial cells of the prostate but increased in low-grade PIN and high-grade PIN in the Tag/Ghr+/+ mice. A similar pattern of Tag expression is seen in the Tag/Ghr–/– mice. Robertson et al. (40) reported that Tag expression was insensitive to prolactin signaling. Here we found that GH signaling was not essential for Tag expression controlled under C3 promoter fragment. Even though fewer lesions are observed in Tag/Ghr–/– mice than Tag/Ghr+/+ mice, the two genotypes had parallel Tag expression levels within each of the various degrees of PIN severity.
As noted above, it is well established that GH is important for prostate growth in full-grown, adult humans (2, 4, 41). Acromegalics have enlarged prostates that shrink to normal size in response to treatments that lower GH serum levels. Furthermore, the prostate shrinks to below normal volume in acromegalics rendered GH deficient due to aggressive therapy (2). Data from the current study suggest that GH is also important for prostate growth in the mouse. Disruption of the GHR gene in Tag/Ghr–/– mice resulted in a 60% decrease in prostate weight relative to their Tag/Ghr+/+ siblings, and the decrease in prostate weight was proportional to the reduction in overall body weight, consistent with a proportional effect of disrupted GH action on prostate and body weight. Importantly, loss of GH signaling did not appear to affect epithelial cell cyto- or functional differentiation as revealed by similar expression levels and pattern of p63, CK18, and mDLP in the Tag/Ghr+/+ and Tag/Ghr–/– mice. Thus, changes in carcinogenesis between the two genotypes are not likely to be a function of altered epithelial cell differentiation.
Androgens play a critical role in prostate growth, development, and carcinogenesis, and the androgen pathway has been the target of first-line prostate therapies for many years. We asked whether disrupting GH signaling resulted in a down-regulation of androgens or the expression of the AR, which could explain the lack of carcinogenesis in the Tag/ Ghr–/– mice. However, as shown in Fig. 4, serum testosterone levels were affected by neither the presence of Tag (Tag/Ghr+/+ vs. Ghr+/+) nor disruption of the GHR (Ghr+/+ vs. Ghr–/–). Furthermore, AR expression was not compromised in the prostate epithelium of Ghr–/– mice relative to controls (Fig. 5). Therefore, we concluded that the protective effect afforded by disrupting GH signaling is independent of either serum testosterone or AR expression. This is of significance clinically because prostate cancers that initially respond to antiandrogen therapies often evolve into androgen-independent disease, which is currently incurable.
Recently Ormandy and colleagues (40) crossed the C3(1)/Tag mouse used in the preset studies with the prolactin receptor knockout mouse (Prlr–/–) and evaluated prostate carcinogenesis at 50 wk of age. Whereas there was no difference in PIN area in the dorsal prostate lobes, PIN area in the ventral prostate were significantly reduced in Prlr–/– mice relative to control mice. Furthermore, whereas one of 11 Prlr+/+ mice and four of 21 Prlr+/– mice harbored prostate tumors, no prostate tumors were observed in any of the Prlr–/– mice (40). These data indicate that disruption of PRL signaling can impede mouse prostate carcinogenesis. However, PRL levels in the Ghr–/– mouse are not reduced (34), suggesting that disruption of PRL signaling is not responsible for protection from PIN development in Tag/Ghr–/– mice.
Carcinogenesis is characterized by dysregulated cell proliferation or apoptosis. The GH/IGF axis plays an important role in regulating prostate epithelial cell proliferation and apoptosis both in vitro and in vivo (42). One of the parental strains of the Tag/Ghr mouse presented in this communication is the Laron mouse, which, in addition to lacking a functional GHR, also has only about 10% of the serum IGF-I present in wild-type mice (33). We therefore hypothesized that the prostate epithelial cells of Tag/Ghr+/+ mice would have a significant proliferation advantage, compared with Tag/Ghr–/– mice, resulting in more rapid progression of carcinogenesis. Our results indicate that proliferation is significantly lower and apoptosis is significantly higher in the prostate epithelium of Tag/Ghr–/– mice, compared with Tag/Ghr+/+ mice (Figs. 7 and 8). Because all the prostate cells of both groups of mice harbor the same oncogene (Tag), the observed difference in prostate cell proliferation and apoptosis is likely to have a significant impact on prostate carcinogenesis.
In summary, we have crossed the C3(1)/Tag mouse with the GHR/binding protein knockout (Laron) mouse, resulting in a model in which prostate cancer progression can be assessed in the presence or absence of GH signaling. The data indicate that progression of Tag-initiated prostate epithelium is significantly inhibited in the absence of GH signaling. This inhibition is not due to insufficient Tag expression or androgen signaling in Tag/Ghr–/– mice relative to Tag/Ghr+/+ mice. Rather, cancer inhibition appears to be associated with decreased proliferation and increased apoptosis of the prostate epithelium of Tag/Ghr–/– mice. These findings may have important translational implications. It is generally accepted that PINs are precursors to lethal prostate cancers, and these lesions occur at a similar incidence in individuals of populations at either high or low risk for the development of prostate cancer. Thus, the difference in mortality rates between high- and low-risk populations seems to be due to differences in the progression of PIN to prostate cancers. The findings presented here suggest that PIN lesions may require GH signaling for progression, suggesting that the GH signaling pathway or the GH/IGF axis may represent important targets for the development of agents to prevent prostate cancer.
Footnotes
This work was supported by National Institutes of Health (NIH) Grant R03 AG020820 and Department of Defense Grant W81XWH-04-1-0201. J.J.K. is supported by the State of Ohio Eminent Scholars’ Program that includes a gift by Milton and Lawrence Goll, DiAthegen LLC, and NIH.
First Published Online September 1, 2005
Abbreviations: AR, Androgen receptor; CK18, cytokeratin 18; GHR, GH receptor; mDLP, mouse dorsolateral prostate; PCNA, proliferating cell nuclear antigen; PIN, prostatic intraepithelial neoplasia; Tag, T antigen; TUNEL, terminal deoxynucleotide-transferase-mediated deoxyuridine 5-triphosphate-digoxigenin nick end labeling; VP, ventral prostate.
Accepted for publication August 22, 2005.
References
Jemal A, Murray T, Ward E, Samuels A, Tiwari RC, Ghafoor A, Feuer EJ, Thun MJ 2005 Cancer statistics, 2005. CA Cancer J Clin 55:10–30
Colao A, Di Somma C, Spiezia S, Filippella M, Pivonello R, Lombardi G 2003 Effect of growth hormone (GH) and/or testosterone replacement on the prostate in GH-deficient adult patients. J Clin Endocrinol Metab 88:88–94
Colao A, Spiezia S, Di Somma C, Marzullo P, Cerbone G, Pivonello R, Faggiano A, Lombardi G 2000 Effect of GH and/or testosterone deficiency on the prostate: an ultrasonographic and endocrine study in GH-deficient adult patients. Eur J Endocrinol 143:61–69
Colao A, Marzullo P, Spiezia S, Giaccio A, Ferone D, Cerbone G, Di Sarno A, Lombardi G 2000 Effect of two years of growth hormone and insulin-like growth factor-I suppression on prostate diseases in acromegalic patients. J Clin Endocrinol Metab 85:3754–3761
Roberts C 2004 IGF-I and prostate cancer. Novartis Found Symp 262:193–199
Chan JM, Stampfer MJ, Giovannucci E, Gann PH, Ma J, Wilkinson P, Hennekens CH, Pollak M 1998 Plasma insulin-like growth factor-I and prostate cancer risk: a prospective study. Science 279:563–566
Harman SM, Metter EJ, Blackman MR, Landis PK, Carter HB 2000 Serum levels of insulin-like growth factor I (IGF-I), IGF-II, IGF-binding protein-3, and prostate-specific antigen as predictors of clinical prostate cancer. J Clin Endocrinol Metab 85:4258–4265
Stattin P, Bylund A, Rinaldi S, Biessy C, Dechaud H, Stenman UH, Egevad L, Riboli E, Hallmans G, Kaaks R 2000 Plasma insulin-like growth factor-I, insulin-like growth factor-binding proteins, and prostate cancer risk: a prospective study. J Natl Cancer Inst 92:1910–1917
Woodson K, Tangrea JA, Pollak M, Copeland TD, Taylor PR, Virtamo J, Albanes D 2003 Serum insulin-like growth factor I: tumor marker or etiologic factor A prospective study of prostate cancer among Finnish men. Cancer Res 63:3991–3994
Oliver SE, Gunnell D, Donovan J, Peters TJ, Persad R, Gillatt D, Pearce A, Neal DE, Hamdy FC, Holly J 2004 Screen-detected prostate cancer and the insulin-like growth factor axis: results of a population-based case-control study. Int J Cancer 108:887–892
Mantzoros CS, Tzonou A, Signorello LB, Stampfer M, Trichopoulos D, Adami HO 1997 Insulin-like growth factor 1 in relation to prostate cancer and benign prostatic hyperplasia. Br J Cancer 76:1115–1118
Wolk A, Mantzoros CS, Andersson SO, Bergstrom R, Signorello LB, Lagiou P, Adami HO, Trichopoulos D 1998 Insulin-like growth factor 1 and prostate cancer risk: a population-based, case-control study. J Natl Cancer Inst 90:911–915
Wolk A, Andersson SO, Mantzoros CS, Trichopoulos D, Adami HO 2000 Can measurements of IGF-1 and IGFBP-3 improve the sensitivity of prostate-cancer screening Lancet 356:1902–1903
Khosravi J, Diamandi A, Mistry J, Scorilas A 2001 Insulin-like growth factor I (IGF-I) and IGF-binding protein-3 in benign prostatic hyperplasia and prostate cancer. J Clin Endocrinol Metab 86:694–699
Chan JM, Stampfer MJ, Ma J, Gann P, Gaziano JM, Pollak M, Giovannucci E 2002 Insulin-like growth factor-I (IGF-I) and IGF binding protein-3 as predictors of advanced-stage prostate cancer. J Natl Cancer Inst 94:1099–1106
Li L, Yu H, Schumacher F, Casey G, Witte JS 2003 Relation of serum insulin-like growth factor-I (IGF-I) and IGF binding protein-3 to risk of prostate cancer (United States). Cancer Causes Control 14:721–726
Cutting CW, Hunt C, Nisbet JA, Bland JM, Dalgleish AG, Kirby RS 1999 Serum insulin-like growth factor-1 is not a useful marker of prostate cancer. BJU Int 83:996–999
Janssen JA, Wildhagen MF, Ito K, Blijenberg BG, Van Schaik RH, Roobol MJ, Pols HA, Lamberts SW, Schroder FH 2004 Circulating free insulin-like growth factor (IGF)-I, total IGF-I, and IGF binding protein-3 levels do not predict the future risk to develop prostate cancer: results of a case-control study involving 201 patients within a population-based screening with a 4-year interval. J Clin Endocrinol Metab 89:4391–4396
Chen C, Lewis SK, Voigt L, Fitzpatrick A, Plymate SR, Weiss NS 2005 Prostate carcinoma incidence in relation to prediagnostic circulating levels of insulin-like growth factor I, insulin-like growth factor binding protein 3, and insulin. Cancer 103:76–84
Kurek R, Tunn UW, Eckart O, Aumuller G, Wong J, Renneberg H 2000 The significance of serum levels of insulin-like growth factor-1 in patients with prostate cancer. BJU Int 85:125–129
Finne P, Auvinen A, Koistinen H, Zhang WM, Maattanen L, Rannikko S, Tammela T, Seppala M, Hakama M, Stenman UH 2000 Insulin-like growth factor I is not a useful marker of prostate cancer in men with elevated levels of prostate-specific antigen. J Clin Endocrinol Metab 85:2744–2747
Renehan AG, Zwahlen M, Minder C, O’Dwyer ST, Shalet SM, Egger M 2004 Insulin-like growth factor (IGF)-I, IGF binding protein-3, and cancer risk: systematic review and meta-regression analysis. Lancet 363:1346–1353
Shi R, Berkel HJ, Yu H 2001 Insulin-like growth factor-I and prostate cancer: a meta-analysis. Br J Cancer 85:991–996
Fuhrman B, Barba M, Schunemann HJ, Hurd T, Quattrin T, Cartagena R, Carruba G, Muti P 2005 Basal growth hormone concentrations in blood and the risk for prostate cancer: a case-control study. Prostate 64:109–115
Reiter E, Kecha O, Hennuy B, Lardinois S, Klug M, Bruyninx M, Closset J, Hennen G 1995 Growth hormone directly affects the function of the different lobes of the rat prostate. Endocrinology 136:3338–3345
Pollak M, Beamer W, Zhang JC 1998 Insulin-like growth factors and prostate cancer. Cancer Metastasis Rev 17:383–390
Schally AV, Comaru-Schally AM, Plonowski A, Nagy A, Halmos G, Rekasi Z 2000 Peptide analogs in the therapy of prostate cancer. Prostate 45:158–166
Jungwirth A, Schally AV, Pinski J, Halmos G, Groot K, Armatis P, Vadillo-Buenfil M 1997 Inhibition of in vivo proliferation of androgen-independent prostate cancers by an antagonist of growth hormone-releasing hormone. Br J Cancer 75:1585–1592
Shibata MA, Ward JM, Devor DE, Liu ML, Green JE 1996 Progression of prostatic intraepithelial neoplasia to invasive carcinoma in C3(1)/SV40 large T antigen transgenic mice: histopathological and molecular biological alterations. Cancer Res 56:4894–4903
Institute of Laboratory Animal Resources (United States) 1996 Guide for the care and use of laboratory animals. 7th ed. Washington, DC: National Academy Press
Maroulakou IG, Anver M, Garrett L, Green JE 1994 Prostate and mammary adenocarcinoma in transgenic mice carrying a rat C3(1) Simian virus 40 large tumor antigen fusion gene. Proc Natl Acad Sci USA 91:11236–11240
Leung DW, Spencer SA, Cachianes G, Hammonds RG, Collins C, Henzel WJ, Barnard R, Waters MJ, Wood WI 1987 Growth hormone receptor and serum binding protein: purification, cloning and expression. Nature 330:537–543
Zhou Y, Xu BC, Maheshwari HG, He L, Reed M, Lozykowski M, Okada S, Cataldo L, Coschigamo K, Wagner TE, Baumann G, Kopchick JJ 1997 A mammalian model for Laron syndrome produced by targeted disruption of the mouse growth hormone receptor/binding protein gene (the Laron mouse). Proc Natl Acad Sci USA 94:13215–13220
Chandrashekar V, Bartke A, Coschigano KT, Kopchick JJ 1999 Pituitary and testicular function in growth hormone receptor gene knockout mice. Endocrinology 140:1082–1088
Shappell SB, Thomas GV, Roberts RL, Herbert R, Ittmann MM, Rubin MA, Humphrey PA, Sundberg JP, Rozengurt N, Barrios R, Ward JM, Cardiff RD 2004 Prostate pathology of genetically engineered mice: definitions and classification. The consensus report from the Bar Harbor meeting of the Mouse Models of Human Cancer Consortium Prostate Pathology Committee. Cancer Res 64:2270–2305
Donjacour AA, Cunha GR 1993 Assessment of prostatic protein secretion in tissue recombinants made of urogenital sinus mesenchyme and urothelium from normal or androgen-insensitive mice. Endocrinology 132:2342–2350
Bostwick DG, Brawer MK 1987 Prostatic intra-epithelial neoplasia and early invasion in prostate cancer. Cancer 59:788–794
Bostwick DG 2000 Prostatic intraepithelial neoplasia. Curr Urol Rep 1:65–70
Green JE, Shibata MA, Yoshidome K, Liu ML, Jorcyk C, Anver MR, Wigginton J, Wiltrout R, Shibata E, Kaczmarczyk S, Wang W, Liu ZY, Calvo A, Couldrey C 2000 The C3(1)/SV40 T-antigen transgenic mouse model of mammary cancer: ductal epithelial cell targeting with multistage progression to carcinoma. Oncogene 19:1020–1027
Robertson FG, Harris J, Naylor MJ, Oakes SR, Kindblom J, Dillner K, Wennbo H, Tornell J, Kelly PA, Green J, Ormandy CJ 2003 Prostate development and carcinogenesis in prolactin receptor knockout mice. Endocrinology 144:3196–3205
Colao A, Marzullo P, Spiezia S, Ferone D, Giaccio A, Cerbone G, Pivonello R, Di Somma C, Lombardi G 1999 Effect of growth hormone (GH) and insulin-like growth factor I on prostate diseases: an ultrasonographic and endocrine study in acromegaly, GH deficiency, and healthy subjects. J Clin Endocrinol Metab 84:1986–1991
Moschos SJ, Mantzoros CS 2002 The role of the IGF system in cancer: from basic to clinical studies and clinical applications. Oncology 63:317–332(Zhuohua Wang, Gail S. Prins, Karen T. Co)