Akt/CREB/Cyclin D1 network: a novel target for prostate cancer inhibition in transgenic adenocarcinoma of mouse prostate (TRAMP) model mediated by Nexrutine®, a Phellodendron amurense bark extract

Abstract

Purpose

Development of prostate cancer (PCA) prevention strategies is an important priority to overcome high incidence, morbidity & mortality. Recently, we demonstrated that Nexrutine®, an herbal extract inhibits prostate cancer cell proliferation through modulation of Akt and CREB-mediated signaling pathways. However, it is unknown if Nexrutine® can be developed as a dietary supplement for the prevention of PCA. In this study, we used the transgenic adenocarcinoma of mouse prostate (TRAMP) model to examine the ability of Nexrutine® to protect TRAMP mice from developing prostate cancer.

Experimental Design

8-week-old TRAMP mice were fed pelleted diet containing (300 and 600 mg/kg Nexrutine®) for 20 weeks. Efficacy of Nexrutine® was evaluated by magnetic resonance imaging (MRI) at 18 and 28 weeks of progression and histological analysis of prostate tumor or tissue at the termination of the experiment. Tumor tissue was analyzed for modulation of various signaling molecules.

Results

We show that Nexrutine® significantly suppressed palpable tumors and progression of cancer in the TRAMP model. Expression of total and phosphorylated Akt, cAMP response element binding protein (CREB), and Cyclin D1 was significantly reduced in prostate tissue from Nexrutine® intervention group compared to tumors from control animals. Nexrutine® also inhibited Cyclin D1 transcriptional activity in androgen independent PC-3 cells. Overexpression of kinase dead Akt mutant or phosphorylation-defective CREB inhibited Cyclin D1 transcriptional activity.

Conclusions

The current study demonstrates that Nexrutine®-mediated targeting of Akt/CREB-induced activation of Cyclin D1 prevents the progression of prostate cancer. Expression of CREB and pCREB increased in human prostate tumors compared to normal tissue suggesting their potential use as prognostic markers.

Introduction

Prostate cancer is the second leading cause of cancer related deaths in men and expected to lead to about 27,350 deaths in 2006 (1). African American men have the highest incidence of prostate cancer in the world, whereas Asian men native to their countries who consume a low fat, high fiber diet have the lowest risk (2). Epidemiological studies suggest that a reduced risk of cancer is associated with the consumption of a phytochemical-rich diet that includes fruits and vegetables (3). Evidence suggests that prostate cancer progresses from normal epithelium to proliferative inflammatory atrophy (PIA), to low grade prostatic intraepithelial neoplasia (LGPIN), to high grade PIN (HGPIN) that eventually progresses to the more aggressive-metastatic and clinically evident prostate cancer (4 and references therein). Such preneoplastic lesions have been found in young men in their twenties and are common in men in their fifties (5). However, clinically detectable prostate cancer does not generally manifest itself until the sixties. In addition, the occurrence of precancerous lesions is more prevalent (~1 in 3 men) than the incidence of carcinoma (~1 in 9 men; 6). Therefore, the development of effective strategies for the prevention of early-stage prostate cancer is of utmost importance to ensure quality of life for elderly men. The long latency involved in the development of clinically significant prostate cancer provides plethora of opportunities for intervention, including the use of phytochemicals.

Many components derived from dietary or medicinal plants have been found to possess substantial chemopreventive properties both in rodents and in humans (7). Further phytochemicals such as green tea polyphenols, tomato products and pomegranate juice have shown promising activity in human clinical trials (8–16) as part of the treatment regimen. Consistent with this observation 69% of cancer patients used at least one complementary and alternative medicine (CAM) therapy as part of their cancer treatment (17). Given the fact that cancer arises due to deregulation of multiple signaling pathways, targeting multiple signaling pathways using a combination of agents or complex botanicals offers an added advantage of providing a synergistic or additive effect (18). These data indicate a potential for developing novel non-toxic agents from plants (phytoceuticals) for successful management of prostate cancer.

Nexrutine®, is a commercially available herbal extract from Phellodendron amurense (Phellodendron is “cork tree” in Greek), which is widely used for the treatment of inflammation, gastroenteritis, abdominal pain and diarrhea (19–20). The tree is native to Asia and has been reported to contain isoquinoline alkaloids, phenolic compounds and flavone glycosides (19–20). Recently, we demonstrated for the first time that Nexrutine® inhibits prostate cancer cell proliferation through modulation of Akt and CREB-mediated signaling pathways (21). However, it is unknown if Nexrutine® can be developed as a dietary supplement for the prevention of PCA. In this study, we used the transgenic adenocarcinoma of mouse prostate (TRAMP) model to examine the ability of Nexrutine® to protect TRAMP mice from developing prostate cancer.

Materials and Methods

Preparation of Nexrutine® diet

Nexrutine® was provided by Next Pharmaceuticals, Irvine, CA. For in vivo experiments, pelleted diet containing different doses of Nexrutine (300 and 600 mg/kg) was prepared at Dyets, Inc. (Bethlehem, PA). Stability of Nexrutine® in the diet pellets was evaluated every month by thin-layer chromatographic (TLC) analysis. Briefly, 5 g of pelleted diet was extracted with 70% methanol and dried under vacuum. Dried powder was resuspended in water and TLC was performed using dichloromethane and methanol (24:1 w/v) as the solvent systems. The chromatographic profile was observed under UV light. Nexrutine® was used as a positive control. The chromatograms of pure Nexrutine® and the extracted Nexrutine® from the pelleted diet were identical (data not shown).

Transgenic mouse experiments

TRAMP model was developed by prostate specific expression of SV40 large T antigen using the rat probasin promoter (22–23). TRAMP mice develop prostate tumors with 100% frequency, in progressive stages that facilitates preclinical studies in the prevention, intervention and regression setting as demonstrated by various groups, including our own (24–31). TRAMP mice, with a pure C57BL/6 background, were obtained from Jackson Laboratories (The Jackson Laboratories, Bar Harbor, Maine, USA). All mice were maintained in a climate-controlled environment with a 12-h light/dark cycle. Diet and water were supplied ad libitum. For imaging, animals were anesthetized with ketamine (80mg/kg i.m.) and scanned in a Siemens 3T TRIO MRI scanner using a receive-only surface coil that was custom-built for this experiment to provide a full volumetric coverage of the prostate region. A 3-D Fast Low Angle Shot (FLASH) sequence with fat suppression was optimized (TR=27ms, TE=5ms, flip angle=22 degrees) to provide an SNR of 25 at isotropic spatial resolution of 400 microns and total scan time of about 10 min. PSVC volume was measured using MNI Display program¹. Following the final MRI scan, animals were euthanized and all organs were collected for further analysis. At necropsy, animals were examined to determine if there were any gross organ abnormalities. Animal care and handling was conducted in accordance with established humane guidelines and protocols approved by the University of Texas Health Science Center at San Antonio’s Institutional Animal Care and Use Committee.

Tumor histology

Tumors were weighed, harvested and fixed in 10% neutral buffered formalin. Tumors were paraffin embedded, sectioned, placed on poly-lysine slides and stained with H&E to visualize cell nuclei and cytoplasm. Prostate lesions were scored using an established grading system for TRAMP mice (21–22). Non-cancerous lesions were graded as 1, 2 or 3, indicating normal tissue, low PIN and high PIN, respectively. Grades 4, 5 and 6 indicated well-differentiated, moderately differentiated and poorly differentiated cancerous lesions, respectively. Images were recorded using a light microscope.

Preparation of extracts and Western blot analysis

50 mg of prostate tumor or tissue (dorso-lateral) was homogenized in liquid nitrogen and lysed in buffer (50 mM Tris-HCl, 150 mM NaCl, 0.5% NP40, 50 mM NaF, 1 mM NaVO₄, 1 mM phenylmethylsufonyl fluoride, 25 μg/ml leupeptin, 25 μg/ml aprotinin, 25 μg/ml pepstatin and 1 mM DTT; pH 7.4). After passing the lysate through a 25G needle, cell debris was removed by centrifugation at 12,000 rpm for 30 min. Nuclear extracts were prepared according to the method of Dignam and protein content of the extracts was determined by the method of Bradford as described earlier (32–33). Equal amounts of extracts were fractionated on a 10% SDS-polyacrylamide gel. Following electrophoresis, proteins were transferred to a nitrocellulose membrane. The blotted membrane was blocked with 5% non-fat dried milk in Tris-buffered saline containing 0.1% Tween 20, and incubated with indicated antibodies (Santa Cruz Biotechnology, CA; Cell Signaling Technology, Inc. Beverly, MA) followed by incubation with horseradish peroxidase-conjugated anti-rabbit IgG antibody (Sigma) in blocking solution. Bound antibody was detected by enhanced chemiluminescence using Supersignal West Pico Chemiluminescent Substrate, following the manufacturer’s directions (Pierce, Rockford, IL). All the blots were stripped and re-probed with GAPDH to ensure equal loading of protein.

Immunohistochemical analysis

Sections from formalin fixed, paraffin embedded tissue blocks of prostate were cut and stained with pCREB (Cell Signaling Danvers, MA), pAkt (Ser473, rabbit monoclonal, 1:50, Cell Signaling, Danvers, MA); CREB (Cell Signaling Danvers, MA). Proliferation was assessed using the Ki-67 (SP6) antibody (Lab vision, Fremont, CA). The secondary and tertiary antibodies were biotinylated link and streptavidin HRT (Biocare 4 plus Kit, Biocare Medical, Concord, CA or Vector Labs).

Proliferation and TUNEL staining in tumors

Proliferation was assessed using the Ki-67 (SP6) antibody (Lab Vision, Fremont, CA). The secondary and tertiary antibodies were a biotinylated link and streptavidin HRP (Biocare 4 plus Kit, Biocare Medical, Concord, CA). Apoptosis was assessed using in situ the terminal transferase dUTP nick end-labeling (TUNEL) assay, with biotin-16-dUTP (Roche Applied Science, Indianapolis, IN) and terminal deoxynucleotidyl transferase (TdT) according to vendor recommendations (Invitrogen, Carlsbad, CA).

CREB DNA binding activity

CREB DNA binding activity was measured in TRAMP nuclear extracts using TransAM^™ CREB as described earlier (21; Active Motif, Carlsbad, CA). The sequence of the wild type CREB was 5′-AGAGATTGCCTGACGTCAGAGAGCTAG-3′ (Mutated nucleotides shown in bold). Extracts were incubated with CREB consensus oligonucleotide that was immobilized in a 96-well plate. A primary antibody specific for an epitope on the bound and active form of CREB was added, followed by subsequent incubation with secondary antibody and developing solution. Following incubation, CREB activity was measured, colorimetrically, at 450 nm. Nuclear extracts prepared from human fibroblast WI-38 cells stimulated with Forskolin (CREB activator) was used as a positive control. For competition experiments, the wells containing immobilized oligo were pre-incubated with a 100-fold molar excess of wild type and mutant oligonucleotide for 30 min before adding the nuclear extract.

Prostate cancer tissue array

Immunohistochemical staining for CREB, pCREB and pAkt was conducted on a human prostate cancer tissue array containing 15 specimens of different Gleason grades with paired normal prostate. The array was prepared and provided by Dr Dean Troyer (Department of Pathology. University of Texas Health Science Center, San Antonio, TX). Each specimen was analyzed for immunoreactivity using a1-4+ scoring system for stain intensity and percentage of positive cells. Grading scale for intensity ranged from undetectable signal (1+) to strong signal (4+). The % of staining was scored by counting the positive stained cells and total number of cells in four random microscopic fields.

Statistical analysis

Dichotomous measures, such as the presence/absence of (1) prostate tumors, (2) cancerous lesions with grade of 4 or higher were evaluated for treatment group differences using Fisher’s exact tests. Owing to the small sample size and similarity of outcome frequencies for the low- and high-dose Nexrutine® treatment groups, the groups were then combined and compared vs. controls using Fisher’s exact tests. Since the reduction in tumor frequency was a primary hypothesis of the study, a power analysis was performed to determine the likelihood of observing a clinically significant response with the proposed sample size of 15 control and 20 experimental mice. In the population of TRAMP mice, if 50% or more mice receiving the control diet develop prostate tumors, while 10% or fewer mice receiving the experimental diet develop prostate tumors, then this difference in tumor rates can be considered clinically significant. This can be detected with the proposed sample size by Fisher’s Exact Test at the 0.05 level with power of 80%. One-way ANOVA was performed to determine if there were any significant treatment group mean differences for prostate weights. Post-hoc tests adjusting for the number of comparisons were performed to identify group differences if the F-test was significant. The weekly mouse weight data were analyzed using mixed-model ANOVA, assuming a first-order autoregressive covariance structure for the repeated effect by week. The mixed-model ANOVA tested for interaction between treatment group and week, as well as the individual main effects and post-hoc comparisons, were performed when the estimated treatment group means for a given week were significantly different (determined by F-test). Because weight changes were substantial at monthly intervals relative to weekly intervals, the mixed-model ANOVA was performed by restricting the data to the baseline, 4, 8, 12 and 16-week measures. For all statistical tests, p-values < 0.05 were considered significant, with the exception that the test for interaction in the ANOVA model was considered significant if p<0.10. The power analysis was performed using PASS 6.0 software. Statistical analyses and graphics were performed using SPSS and Stata software.

Results and Discussion

Dietary intervention with Nexrutine® suppresses palpable tumors in TRAMP mice

Eight-week old TRAMP mice (at which time they display HGPIN) were randomized into two groups of 10 animals each and fed AIN-96A pelleted diet containing either 300 or 600 mg/kg Nexrutine®. A third group of 15 animals was fed control diet without Nexrutine®. These doses were chosen from a dose-escalation xenograft study (data not shown). The ability of Nexrutine® to prevent prostate cancer progression was assessed at 18 and 28 weeks of progression by determining the volume of the prostate seminal vesicle complex (PSVC) using non-invasive magnetic resonance imaging (MRI). At the termination of the study (28 weeks) the weight of the PSVC was determined and histological evaluation of the prostate gland and tumor (if present) was carried out. Changes in body weight and food consumption were measured weekly and the experiment was terminated when the animals were 28 weeks old. At the time of necropsy, all other organs including lungs, liver, and kidney were collected for histopathological evaluation.

At the time of termination, 53% of the animals in the control group (n=15) had palpable tumors. On the other hand, none of the animals in the low-dose group (n=10) and only 10% of the animals in the high-dose group (n=10) had palpable tumors. Analysis of these data showed that dietary intervention with Nexrutine® significantly suppressed the occurrence of palpable tumors (p= 0.0052, control vs. either dose of Nexrutine®). When we combined the data from both experimental dose groups and compared it with the control group, the data reached even greater significance (p=0.0019; figure 1A). These data show that Nexrutine® intervention decreased palpable tumors, indicating its potential for either reducing tumor incidence and/or increasing latency of tumor development.

Figure 1. Effect of Nexrutine® on tumor formation in TRAMP mice.

Eight-week old TRAMP mice were fed control or Nexrutine® (Nx)-containing diet for a total of 20 weeks as described in methods. Animals were observed for tumors by abdominal palpation (A) and body weight changes (B) throughout the study. Mean body weight in grams ± S.E as a function of age (in weeks) is shown. 0 indicates week 8.

Mean body weight changes in response to Nexrutine® intervention with respect to age of animals in weeks are shown in figure 1B. The mixed-model ANOVA indicated a significant interaction between treatment groups and time (F=1.8, p=0.086), indicating that treatment group mean differences varied by week. No significant differences in the body weight of these animals between control and experimental groups were seen through 8 weeks of dietary intervention (F=1.4, p=0.259 at 4 weeks and F=1.1, p=0.346 at the end of 8 weeks of intervention). Furthermore, Nexrutine® treatment produced significant group mean differences (F=5.4, p=0.008 at 12 weeks and F=4.9, p=0.012 at 16 weeks). Animals in the high-dose group had a significant increase in body weight as compared to the low-dose group, using Bonferroni-adjusted t-tests (p=0.008 at 12 weeks and p=0.018 at 16 weeks) with similar results compared to the control group (p=0.062 at 12 weeks and p=0.043 at 16 weeks). Animals in the low-dose group showed no significant difference from controls animals in any week. The observed increase in body weight in the high-dose group could not be attributed to differences in food consumption (data not shown). Since the animals in the experimental group did not show significant loss in their body weights, these data indicate non-toxic nature of Nexrutine®.

Dietary intervention with Nexrutine® retards progression of prostate tumors in TRAMP mice

Volumetric analysis of PSVC at the time of MRI scanning showed that the average PSVC volume varied between 1,025 to 7,000 mm³ in the control animals (n=6) compared to 700 to 1,600 mm³ in the treated animals (n=6). MRI analysis data show an approximately 4-fold increase in the prostate volume in animals on control diet from 18 to 28 weeks. In contrast, animals on the Nexrutine® diet, irrespective of the dose, showed approximately 2-fold increase in the prostate volume during the same period (figures 2 A&B). PSVC volume measured by MNI display program is shown in figure 2C. Following the final imaging, animals were euthanized, all the organs were collected, and wet weight of the PSVC apparatus was determined as described in methods (figure 2D). As shown in Table I, mean (with 95% confidence intervals) PSVC weight from animals on control diet (n=15) was approximately 5.51g (2.57, 8.46). In contrast, the mean weight of PSVC was 1.36g (0.78, 1.95) and 1.11g (0.69, 1.53) from animals on low- and high-dose Nexrutine® (n=10 for each dose) respectively. Owing to the heterogeneous variances, a robust ANOVA was performed which indicated significant group mean differences (F=9.3, p=0.0023). Post-hoc comparisons using Games-Howell tests showed that the observed mean differences were significant (p=0.024 and 0.017 between control vs. low- and control vs. high-dose Nexrutine®, respectively) but not between low and high dose of Nexrutine®. PSVC weight decreased despite 10% of the animals showing palpable tumors which could be due to heterogeneity of prostate cancer or these may be non-responders. We believe that the observed palpable tumors (10%) in the high dose Nexrutine® group but not in low dose could be due to heterogeneity of prostate cancer. One would expect a multi-fold increase in PSVC weight if a palpable tumor were present compared to mice with no palpable tumors; in fact, the one positive high dose mouse had a PSVC weight of 1.3 gm which was well within the range of other mice without palpable tumors (0.24 to 1.6 gm). Thus the PSVC weight for the one positive high dose mouse was not exceptionally high. These data demonstrate that animals fed Nexrutine® displayed a significant reduction in the weight of the prostate gland compared to animals on normal diet (figure 2E and Table I).

Figure 2. MRI images of 18 and 28-week old TRAMP mice.

MRI was used to detect the presence and absence of prostate tumors in response to dietary intervention with Nexrutine®. For imaging, animals were anesthetized with ketamine (80mg/kg i.m.) and scanned in a Siemens 3T TRIO MRI scanner using a receive-only surface coil that was custom-built for this experiment to provide a full volumetric coverage of the prostate region. A 3-D Fast Low Angle Shot (FLASH) sequence with fat suppression was optimized (TR=27ms, TE=5ms, flip angle=22 degrees) to provide an SNR of 25 at isotropic spatial resolution of 400 microns and total scan time of about 10 min. PSVC volume was measured using MNI Display program as described in methods. A. A representative MRI image from a 18-week old TRAMP animal on normal and Nexrutine® diet for 10-wks is shown. B. Same animals in A were imaged at 28 weeks of progression (at the time of termination of the study). Representative images show a large prostate tumor in control mice that infiltrated into seminal vesicles; while the animal on a Nexrutine®-fed diet showed very small prostate with no evidence of tumor. PSVC was determined as described in methods using MNI display program and a representative image is shown in C.

D. Gross appearance of PSVC complex from 28-week old TRAMP animal on control or Nexrutine® diet respectively. PT is prostate tumor, P is prostate, SV is seminal vesicles and B is bladder. Weight of PSVC (mean weight in grams ±s.d) of TRAMP mice on control and Nexrutine® (Nx) diet at 28 weeks is shown in E.

Table I. Effect of Nexrutine on wet weight of prostate seminal vesicle complex (PSVC) in 28 week old TRAMP mice.

Eight-week old TRAMP mice were fed control or Nexrutine® (Nx)-containing diet for a total of 20 weeks as described in methods. Experiment was terminated when animals were 28 week old. Weight of the PSVC was determined as described in methods. Mean weight in grams of the PSVC weight and the statistical analysis of the data was performed as described in methods.

	N	Mean	Std Dev	Std. Error	95% confidence interval for Mean		Minimum	Maximum
					Lower Bound	Upper Bound
Control	15	5.513	5.319	1.373	2.568	8.459	0.24	16.80
Nx Low dose	10	1.361	0.819	0.259	0.775	1.947	0.30	3.40
Nx High dose	10	1.113	0.586	0.185	0.694	1.532	0.35	2.20
Robust F-test	F = 9.3	p = 0.0023
Games-Howell tests		GH p-value
Control vs Nx Low dose		0.0243
Control vs Nx High dose		0.0168
Nx Low vs High dose		0.7209

Histological evaluation of the prostate tumor/tissue from TRAMP mice fed control and experimental diet is shown in figure 3A. Prostate from control animals showed well-differentiated adenocarcinoma characterized by variable nuclear shape with little or no gland formation. This is consistent with published results using this model (22–23). In contrast, prostates from animals fed low dose Nexrutine® exhibited features consistent with HGPIN such as increased variability in nuclear shape with apoptotic and mitotic features (fig. 3A), whereas prostates from mice fed the high-dose of Nexrutine® exhibited pathological features consistent with the appearance of normal prostatic epithelium. Cumulative analysis of the data from all the animals indicates that 6 of 15 (40%) control animals and 2 of 18 (10%) animals in the experimental group (low- and high-dose combined together) were graded as 4–6. 18 of 20 animals in the experimental group were graded 1–3. Statistical analysis of these data indicates that the observed differences are significant (p=0.046; figure 3B). These results demonstrate that intervention with Nexrutine® reduces incidence of adenocarcinoma and retards progression of prostate tumors in TRAMP mice.

Figure 3. A. Histopathological analysis of prostate or tumor from untreated and treated TRAMP mice.

Sections of prostate tissues excised from TRAMP mice were stained with H&E. Histopathological evaluation of prostate was performed on H&E stained slides as described in methods. The prostatic lesions were graded as described earlier on a scale of 1 to 6. Briefly non-cancerous lesions were graded as 1, 2 or 3 indicating normal tissue, low PIN and high PIN respectively. Grades 4, 5 and 6 indicated well differentiated, moderately differentiated and poorly differentiated cancerous lesions respectively. Power analysis was performed using PASS 6.0 software. Statistical analyses and graphics were performed using SPSS and Stata software as described in methods. For all statistical tests, p-values less than 0.05 were considered significant and quantitative analysis of the data is shown in B.

C. Proliferation of prostate tumor tissues: Paraffin-embedded tissues sections were stained with Ki-67 (SP6) antibody (Lab vision, Fremont, CA) to assess proliferation. The secondary and tertiary antibodies were a biotinylated link and streptavidin HRP (Biocare 4 Plus Kit, Biocare Medical, Concord, CA). Negative controls were included by omitting the primary antibody or TdT (for TUNEL).

D. TUNEL staining in prostate tumor tissues: Apoptosis was assessed in prostate tumor or tissues using the in situ terminal transferase dUTP nick end-labeling (TUNEL) assay, with biotin-16-dUTP (Roche Applied Science, Indianapolis, IN) and terminal deoxynucleotidyl transferase (TdT) (Invitrogen, Carlsbad, CA).

Nexrutine® reduces proliferation potential of prostate tumors

Previously, we showed that Nexrutine® inhibited proliferation of prostate cancer cells by inducing apoptosis (21). Here we investigated if intervention with Nexrutine® would prevent tumor development in vivo through inhibition of tumor cell proliferation and induction of apoptosis using Ki67, as a proliferation marker and TUNEL staining to determine apoptosis. Immunohistochemical analysis indicated that prostates from TRAMP mice fed control diet are highly proliferative, as indicated by Ki67-positive immunostaining. The number of Ki67-positive cells decreased in Nexrutine® dose-dependent manner (figure 3 C). These in vivo data are consistent with our published work in cells showing Nexrutine® inhibited proliferation in prostate cancer cells (21). Prostate tumors from TRAMP mice fed normal diet also showed more apoptotic cells compared with Nexrutine®-diet fed animals as indicated by TUNEL staining (figure 3 D). There were negligible cells undergoing apoptosis in tissues of animals from Nexrutine® treatment (that were graded as normal prostate lesions). In response to Nexrutine® treatment in vivo the prostate tissue is normal unlike the cancer cells in our earlier in vitro study (21). We did not observe apoptosis in the tissue of animals treated with Nexrutine® since these were normal or near normal prostate tissue.

Mechanism of prostate tumor prevention in vivo

We have shown that Nexrutine® inhibits proliferation and induces apoptosis in prostate cancer cells through modulation of the Akt/CREB signaling pathway (21). We validated these in vitro observations in vivo by measuring alterations in the expression of pAkt, CREB and pCREB in triplicate samples from each group of animals (control; low and high-dose Nexrutine®) using immunohistochemistry. Immunoreactivity was scored based on the percentage of stained cells and graded semi-quantitatively as zero (0% stained cells), 1+ (< 10% stained cells), 2+ (10–20% stained cells), 3+ (20–50% stained cells) and 4+ (> 50% cells stained). Consistent with published literature, pAkt staining was typically localized in the cytoplasm and cell membrane in prostate tumors, while CREB staining was predominantly localized in the nuclei of proliferative epithelial cells (figure 4A). We also observed increased pAkt and CREB staining during prostate carcinogenesis in TRAMP tissues determined by the intensity of staining (graded 4+). As shown in figure 4A, > 50% cells from prostate tumors from control animals showed expression of CREB, pCREB and pAkt. However, prostate tissue from Nexrutine® treated animals showed reduced expression of phosphorylated Akt, CREB and pCREB (graded 2+ to 3+). In contrast very few cells (<10%) showed staining for CREB and pCREB in the prostate from wild type normal prostate (figure 4A labeled as N). We confirmed these studies with western blot analysis using whole-cell extracts prepared from prostate tumors or normal tissue (n=3 from each group). As shown in figure 4B, prostate tumors from control group (labeled as T1, T2 and T3 for tumor) expressed high levels of pAkt and pCREB (normalized to GAPDH, figure 4C). However, prostate tissue from the treated group (L1, L2, L3; low), H1 and H2 (high) dose showed significant reduction in the expression of pAkt (p< 0.01). Although expression of pCREB was decreased with high dose Nexrutine treatment, the decrease was less pronounced at low dose compared to pAkt. The observed increase in pAkt staining in prostate tumors from TRAMP mice is consistent with published results (34). To the best of our knowledge, no prior studies have examined the expression of CREB during prostate carcinogenesis although increased expression of pAkt has been demonstrated in high Gleason-grade prostate cancer specimens (35–38). CREB is a nuclear factor that mediates stimulus induced gene expression through cyclic-AMP response element sequence present in the transcriptional regulatory regions of genes (CRE; 36–39). More than 100 genes involved in various cellular processes including cell growth, survival, apoptosis and differentiation have been shown to be regulated by CREB (39)². Ours is the first report showing the potential involvement of CREB in prostate carcinogenesis. CREB has been shown to (i) be over-amplified and highly expressed in blast cells from patients with acute myeloid or lymphoid leukemia (40–41), (ii) cooperate with oncoprotein to induce cellular transformation (42–44) and (iii) be involved in survival of melanoma cells, endocrine and lung tumors (45–47). In addition, transgenic mice overexpressing dominant negative CREB induces apoptosis in T cells following growth factor stimulation. All these studies implicate CREB as an important factor for cell survival. Our data showing increased expression of CREB in prostate tumors from TRAMP mice is consistent with these published results.

Figure 4. Immunohistochemical analysis of representative tumors or prostate tissue from 28-week old wild type non-transgenic (N), control TRAMP (C) or treated, low (L) and high (H) dose TRAMP mice.

A Paraffin-embedded tissues sections were stained with CREB, pCREB and pAkt antibodies. The antibodies used were at a dilution of 1:100 in PBS and incubated overnight at 4°C. Immune complexes were revealed using a universal secondary antibody (100 μl for 30 minutes) followed by chromogen. Negative controls were included by omitting the primary antibody. As a control, non-transgenic prostate is also shown (N).

B. Modulation in the levels of pAkt, pCREB and SV40 T antigen in TRAMP tissues following Nexrutine® treatment: Whole-cell extracts were prepared separately from TRAMP prostate tumors (three individual animals on control diet) or prostate tissue (dorso-lateral from three individual animals on low-dose of Nexrutine® diet or two different animals on high-dose Nexrutine® diet) was used in immunoblot analysis with the indicated antibodies. Equal loading of protein was confirmed with GAPDH antibody. Bound antibody was detected by enhanced chemiluminescence using Supersignal West Pico Chemiluminescent Substrate, following the manufacturer’s directions (Pierce, Rockford, IL). The blots were imaged using Syngene G Box Fredrick MD and quantified using Gene tools software. Graphical representation of the quantified data is shown in C.

Since TRAMP mice were generated with SV40 large T antigen coupled with probasin promoter (22–23) we examined whether Nexrutine® mediated effects are a consequence of down regulation of the PB-Tag transgene using western blot analysis. As shown in figure 4B, PB-Tag & GAPDH expression was detected both from control and treated tumor/tissue obtained from age-matched TRAMP mice. GAPDH levels did not change between control and treated prostatic tissues. These results demonstrate that Nexrutine®-induced biological effects are due to direct suppression of carcinogenesis and not due to the down regulation of PB-Tag. Graphical representation of quantification data is shown in figure 4C.

CREB and pCREB expression in human prostate tumors

We analyzed the expression of CREB and pCREB in human prostate cancer specimens and normal prostate epithelium using immunohistochemistry. Human prostate cancer tissue arrays containing 15 specimens of different grades and normal tissues were stained for CREB and pCREB. Each specimen was qualitatively analyzed for immunoreactivity (figure 5). We found CREB and pCREB staining was expressed most predominantly in the epithelial nuclei. Of 15 samples analyzed, 9 samples showed positive staining with an intensity of 4+ and other 6 showed 3+. Noteworthy was the heterogeneity observed in the prostate cancer specimens where some of the cancer areas stained stronger (4+) than others. Quantification of the data indicates that more than 50% of the positively stained cells showed a staining intensity of 4+ and other 50% showed staining intensity of 2+ to 3+. Noteworthy is the observation that tumor cells consistently showed high intensity staining for CREB and pCREB compared to normal epithelium. This is further confirmed in our subsequent studies where we have examined the staining of CREB and phospho-CREB in the human tissue array containing more samples. These studies show that >50% of the high Gleason (> 8/10) grade tumors showed 4+ staining whereas 75% of low Gleason grade (< 7/10) tumors showed 2+ staining (Kumar et al, manuscript in preparation). Although the sample size in the current study was small, these observations are very promising and suggest that expression of CREB may be associated with increased tumor grade and therefore greater aggressiveness. Although prostate specific antigen (PSA) is used as a marker to detect PCA, it has questionable reliability (48–49). Due to lack of a molecular marker that can be used to detect a precancerous state or identify which pre-malignant lesions will develop into invasive PCA, CREB should be pursued further to determine if it can potentially be used as a biomarker. Such studies to relate this to grade, biochemical recurrence and other clinical features are currently in progress in the laboratory.

Figure 5. CREB and pCREB staining in human prostate.

Human prostate tissue array containing different grade tumors with paired normal prostate was stained with CREB (top panel) and pCREB (bottom panel). CREB and pCREB were used at a dilution of 1:100 in PBS and incubated overnight at 4°C. Immune complexes were revealed using a universal secondary antibody (100 μl for 30 minutes) followed by chromogen as described in methods. Top panel shows little or no CREB staining in the cancerous tissue; and bottom panel shows pCREB expression in normal prostate. Negative controls were included by omitting the primary antibody (not shown).

Nexrutine® inhibits Cyclin D1 transcriptional activation

The above results suggest a role for Akt and CREB in prostate carcinogenesis; however, the mechanism through which they promote prostate carcinogenesis is not clear. Activation of CREB by phosphorylation promotes transcription of genes involved in cell proliferation, cell cycle regulation and inflammation (36–39). In addition, kinases including Akt have been shown to phosphorylate CREB at Ser 133 (39)². Thus, CREB expression in conjunction with other molecular events could affect the progression of prostate cancer. Cyclin D1 is a critical cell cycle regulatory gene involved in G₁/S progression. We analyzed TRAMP prostate tumor/tissue samples for levels of Cyclin D1 using western blot analysis. As shown in figure 6A and B, prostate tumors from control diet showed a greater expression of Cyclin D1 compared to prostate tissue of treatment group of animals (p<0.001). The promoter region of Cyclin D1 contains potential binding site for transcription factor CREB in addition to AP1, Sp1 and NFκB (50–51). It is possible that any of these or some other unidentified factors will be involved in the transcriptional regulation of Cyclin D1 in response to Nexrutine®. In order to demonstrate this we examined the effect of Nexrutine® on Cyclin D1 transactivation using a full-length 1.7 kb promoter (−1745/+134) of Cyclin D1 gene linked to luciferase reporter in PC-3 cells (53). Treatment of cells with Nexrutine® resulted in 50% reduction in the promoter activity as compared to solvent treated control (figure 6 C). These results indicate that Nexrutine® can down regulate Cyclin D1 promoter activity. Based on our data showing reduced levels of pAkt, pCREB, Cyclin D1 and Akt kinase activity in Nexrutine® treated cells, we reasoned that Akt mediated activation of CREB regulates Cyclin D1 transcriptional activity. We therefore examined the effect of blocking Akt activation using kinase dead Akt mutant expression vector in transient transfection experiments. As shown in figure 6C, inactivation of Akt blocked Cyclin D1 promoter activity indicating the importance of Akt signaling in Cyclin D1 activation. Further co-transfection of phosphorylation defective CREB along with Cyclin D1 promoter also inhibited Cyclin D1 promoter activity. These data collectively indicate an important role for both Akt and CREB in the transcriptional regulation of Cyclin D1.

Figure 6. Nexrutine® treatment reduces Cyclin D1 levels, Cyclin D1 promoter activity and CREB DNA binding activity.

Cyclin D1 levels were determined in TRAMP tissues as described in figure 4B. Quantification of western blotting data is shown in figure 6B. Cyclin D1 promoter activity was determined in PC-3 cells (figure 6C). Briefly, either expression plasmids encoding the kinase dead mutant Akt or dominant negative CREB were transfected with Cyclin D1 reporter plasmid into PC-3 cells. Renilla luciferase activity was measured as described in method and the data shown is an average±s.d of normalized luciferase activity from three independent transfections. CREB DNA binding activity was measured in TRAMP nuclear extracts by using TransAM^™ CREB (Active Motif, figure 6D).

We measured DNA binding activity of CREB using nuclear extracts prepared from TRAMP prostate tumors. As shown in figure 6D, CREB DNA binding activity was reduced in tissues from experimental group of animals, compared to that from the control group. This is consistent with our published studies showing a decrease in CREB DNA activity in PC-3 cells following Nexrutine® treatment (21). Further our results showing regulation of Cyclin D1 by CREB is also consistent with the published data in lymphocytes (51). These data suggest that Akt/CREB-mediated activation of Cyclin D1 plays an important role in the protective effects of Nexrutine® in this prostate cancer model. In addition, it is possible that Nexrutine®-induced apoptosis in prostate cancer cells proceeds through modulation of CREB as aberrant expression of CREB has been shown to regulate apoptosis (53–54). Although our results suggest a potential role for Akt in Cyclin D1 regulation, participation of other kinases such as protein kinase A (PKA), mitogen activated protein kinases (MAPKs), Ca⁺²/Calmodulin-dependent protein kinases (CaMKs) that are known to activate CREB cannot be ruled out. Since a number of growth factors and hormones can activate Akt, Akt/CREB mediated activation of Cyclin D1 may play a role in the development of androgen independent prostate cancer. Further interactions between AR and CREB through CBP have been demonstrated as a potential mechanism of androgen independent prostate cancer progression (55). Recently it was demonstrated that CREB regulates expression of bone matrix proteins such as osteocalcin and sialoprotein in prostate cancer cells (56). Constitutive activation of Src-MEK-1-2-ERK-1/2-CREB signaling pathway has been shown to be associated with the androgen independent phenotype (57). Thus, inhibition of such critical signaling network to prevent prostate cancer progression with a cost-effective non-toxic herbal supplement is a highly significant finding. An added advantage of Nexrutine® is that it has a good safety record in humans and has been shown to be biologically active in human test subjects. The Living Longer Clinic in Cincinnati, Ohio, conducted a 288-subject open-label, single center study to test Nexrutine® as a potential analgesic. These subjects were given Nexrutine® (1–2 capsules thrice daily. Two hundred fifty five of the subjects (88%) reported beneficial effects of Nexrutine®, including reduction in pain and/or inflammation, while the remainder reported no improvement³. To date, the exact nature of the biologically active component(s) in this herbal extract has not been elucidated. Additional studies are currently being carried out in our laboratory to identify and characterize the biologically active principle of Nexrutine® and the precise molecular pathways involved in its anti-cancerous activity. The results of these and future studies with Nexrutine® may yield an important addition to the weak therapeutic armamentarium that exists for the treatment of prostate cancer and perhaps other cancers.