Skip Navigation


Carcinogenesis Advance Access originally published online on May 16, 2006
Carcinogenesis 2006 27(10):2124-2132; doi:10.1093/carcin/bgl075
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow An erratum has been published
Right arrow A corrigendum has been published
Right arrow All Versions of this Article:
27/10/2124    most recent
bgl075v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (1)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Wang, L.G.
Right arrow Articles by Chiao, J.W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wang, L.G.
Right arrow Articles by Chiao, J.W.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2006. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Repression of androgen receptor in prostate cancer cells by phenethyl isothiocyanate

L.G. Wang*, X.M. Liu and J.W. Chiao1

New York University Cancer Institute, New York University School of Medicine New York, NY 10010, USA
1 Department of Medicine, Vosbargh 207, New York Medical College Valhalla, NY 10595, USA

*To whom correspondence should be addressed at: NYU Cancer Institute, NYU School of Medicine, Manhattan VAMC 18th Floor, Room 18003 W, 423 East 23rd Street, New York, NY 10010, USA. Tel: +1 212 263 4274; Fax: +1 212 263 6091; Email: Longgui.Wang{at}med.NYU.edu


   Abstract
Top
 Abstract
Background
 Materials and methods
 Results
 Discussion
 Authors' contributions
 References
 
Background: Prostate cancer usually progresses to androgen refractory after an initial anti-androgen treatment. The androgen receptor (AR) is a pivotal factor for the androgen-mediated growth and maintenance of the prostate. Abnormality of the AR, such as overexpression has been postulated to be related to the hormone independent growth of the cancer. Although we previously demonstrated that the AR expression could be modulated by isothiocyanates, which are natural constituents of cruciferous vegetables, the mechanism, however, remained to be clarified. We have since investigated the mechanism of phenethyl isothiocyanate (PEITC) in AR regulation. Methods: A human androgen dependent prostate cancer cell line LNCaP (AD) and its sub-line LNCaP (AI), i.e. androgen independent but overexpressing AR, were exposed to PEITC. The effects of PEITC on cell growth and AR expression/transcription were analyzed with MTT assay, real-time PCR and western blotting. The AR promoter activity was analyzed with the reporter activity after transfection with pAR-luc. The effects on Sp1, the major transcription factor of the AR, were tested with Sp1-luc activity, western blotting and electrophoretic mobility shift assay. Results: PEITC induced a significant growth inhibition, with equal IC50, in both AD and AI cells. The AR present in both cells was repressed as demonstrated with real-time PCR and western blot. PEITC mediates dual effects at transcriptional and post-translational levels to regulate the AR. At transcriptional level the AR level was reduced via inhibition of the transcription factor Sp1, and at post-translational level by accelerating protein degradation. Conclusion: PEITC represses AR transcription and expression, and mediates growth arrest in androgen dependent and independent prostate cancer cells. With the AR modulation and growth attenuation, PEITC and possibly other isothiocyanates, may prevent and inhibit hormone sensitive and refractory prostate cancer.

Abbreviations: AR, androgen receptor; CHX, cycloheximide; CSFBS, charcoal-stripped, heat-inactivated FBS; FBS, fetal bovine serum; PEITC, phenethyl isothiocyanate


   Background
Top
 Abstract
 Background
Materials and methods
 Results
 Discussion
 Authors' contributions
 References
 
Prostate cancer is the most commonly diagnosed cancer among men in the United States. For patients who are not cured by local treatment and are experiencing metastasis, neither androgen ablation nor chemotherapy abrogates progression of the androgen-independent/hormone refractory disease. Thus, creating a novel strategy for preventing prostate cancer initiation, as well as treating cancer in progression, has become a practical goal to reducing morbidity and mortality.

Abnormalities of the androgen receptor (AR) are decisive factors in the failure of treatment for advanced prostate cancer. The mechanisms postulated for the development of hormone refractory prostate cancer include wild-type AR amplification, excessive recruitment of AR transcriptional co-activators and AR gene mutation (1–3). Although hormone refractory, the growth of the cancer cells need the AR (4,5). One such example is a sub-line developed from androgen-dependent prostate carcinoma LNCaP, which becomes androgen-independent, and strongly overexpress functional AR, and is phosphorylated in an androgen-depleted medium (4). It is therefore reasonable to assume that regulating the overexpressed AR and its function may prove to be an effective approach to prevent and overcome the hormone refractory state (6,7).

Recent epidemiological reports indicated that consumption of cruciferous vegetables is associated with reduced incidence of prostate cancer cases (8–10). The crucifereous vegetables include the Brassica family, and examples are broccoli, cabbages, Brussels sprouts, kale, watercress and cauliflower. The isothiocyanates from these vegetables have been investigated as the responsible dietary factors for chemoprevention. The isothiocyanates occur naturally as thioglucoside conjugates, i.e. glucosinolates, in a wide variety of cruciferous vegetables (11). They are released from glucosinolates by the hydrolytic action of the enzyme myrosinase when the vegetable tissues are crushed or masticated (12). Hydrolysis of the glucosinolate gluconasturtin, found in high amounts in watercress, yields phenethyl isothiocyanate (PEITC). The isothiocyanates are potent cancer chemopreventive agents, as demonstrated with a number of carcinogen-induced cancers in rodents (13,14). The major mode of action is cytoprotection, i.e. by inducing enzymes that prevent cells from the assaults of carcinogens (15,16). More recently, isothiocyanates and their metabolites have been demonstrated to mediate growth regulation in rapidly growing cancer cells, suggesting that the isothiocyanates can inhibit post-initiation progression of carcinogenesis. We have described that the growth regulatory mechanism may include inhibition of cdk activity and/or directly upregulating inhibitor p21WAF1/CIP1, and downregulating cyclins (17–19). The upstream mechanisms to regulate cell cycle progression, as well as the relationship with the AR, remain to be elucidated.

With the lack of TATA and CCAAT box of the AR gene promoter, the Sp1 has been described to predominantly and positively influence AR transcription (20–25). Cross talk between AR and growth factor signaling pathways may occur through activation of Sp1, as indicated by our observation that the epidermal growth factor directly stimulates AR transcription (unpublished data). In the present study, we have demonstrated that PEITC significantly downregulates the transcription of the AR via an inhibition of Sp1, and stimulates AR protein degradation, which in turn modulates the cell growth of androgen-dependent and -independent prostate cancer cells. Our observations thus established a new basis for further exploration of PEITC and the isothiocyanates, for preventing and inhibiting the hormone sensitive and hormone refractory prostate cancer.


   Materials and methods
Top
 Abstract
 Background
 Materials and methods
Results
 Discussion
 Authors' contributions
 References
 
Reagents
The PEITC was purchased from LKT Labs (Minneapolis, MN) with 98% purity. The MG-132 was purchased from EMD Biosciences (San Diego, CA). The paclitaxel and other chemicals were purchased from Sigma (St Louis, MO). The Sp1-luc and mtSp1-luc containing three tandem repeats of consensus Sp1 sites and mutant Sp1 sites driving luciferase gene (26) were kindly provided by Dr Yoshihiro Sowa (University of Medicine, Kamigyo-ku, Kyoto 602, Japan). The AR luciferase reporter pLARS-1 was constructed in our laboratory as described previously (27). AR-Sp1 luciferase reporter pARSp1-3 was constructed by inserting three repeats of the double Sp1 sequences from +429 to +442 or its mutant of 5'-untranslated region (UTR) of the AR gene into a basic luciferase reporter pGL3 and the DNA sequence verified. Effectene transfection reagent and luciferase assay kit were acquired from Qiagen (Valencia, CA) and Promega (Madison, WI). Antibodies against the AR, Sp1 and ß-actin, and the second antibody were purchased from Santa Cruz Biotechnology (Santa Cruz, CA), and BD Biosciences (Franklin Lakes, NJ). The ELC Western detection kit and films were purchased from Amersham Biosciences (Buckingamshire, England).

Cell culture
The androden-dependent LNCaP cells (AD) were maintained in RPMI-1640 (GIBCO, Gaithersburg, MD) with 10% heat-inactivated fetal bovine serum (FBS). The androgen-independent sub-line (AI), derived from LNCaP cells, was maintained in RPMI-1640 medium containing 10% charcoal-stripped, heat-inactivated FBS (CSFBS) (Hyclone Laboratories, Logan, UT) and 5 µg/ml insulin as described previously (28).

Cell growth
The MTT assay was used to estimate cell growth and their response to drug treatment as described previously (29). To evaluate androgen-dependent and -independent growth of AD and AI cells, exponentially growing AD, AI or PC-3 cells in 10% FBS RPMI 1640 medium, or in 10% CSFBS RPMI 1640 medium, were seeded into 96-well dishes at a density of 2500 cells/well. After incubation for different time periods, 100 µl of the medium was removed from each of the wells and 50 µl of a 1 mg/ml solution of 3-(4,5-dimethyl-thiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) was added for additional 4 h incubation. Two hundred microliters of 0.04 N HCl–isopropanol was added to each well to dissolve the black formazan precipitates, and absorbance at 540 nm, with a reference wavelength 630 nm, was measured on a 96-well microplate reader (Dynex Technologies, Chantilly, VA).

To examine the effects of PEITC on AD and AI cell growth, AD or AI cells grown exponentially, were aliquoted into 96-well plates at a density of 2500 cells/200 µl/well in RPMI 1640 medium containing 10% FBS for AD, and 10% CSFBS for AI. After 24 h the cells were exposed to serial dilutions of PEITC or paclitaxel for 7 days. Cell growth was then measured by MTT as described.

Cell transfection and luciferase assay
The AD or AI cells grown exponentially were seeded into 60 mm dishes at a density of 1 x 105 cells/ml. After incubation at 37°C, the cells were transfected with 1 µg/dish of luciferase reporter Sp1-luc, mtSp1-luc, pLARS-1 or pARSp-1-3 using effectene (Qiagen, Valencia, CA). Twenty-four hours after transfection, the cells were exposed for an additional 24 h period to various concentrations of PEITC or mithramycin. The cells were washed, lysed and the lysates used for luciferase activity assay with a Promega luciferase assay system. The luciferase activities were expressed as units per milligram of proteins.

Quantitative real-time PCR analysis
The LNCaP AD and AI cells grown exponentially were treated with different concentrations of PEITC for 24 h. The cells were washed, harvested and total RNA extracted with Qiagen kits (Chatsworth, CA). All RNA was treated with DNase (Qiagen). A total 1 µg RNA was reverse-transcribed per reaction using first-strand complementary DNA synthesis with random primers (Invitrogen, Carlsbad, CA). Quantitative real-time PCR of AR was done using AR specific primers provided by Applied Biosystems (Foster City, CA). All values were normalized to the levels of glyceraldehyde-3-phosphate dehydrogenase (GAPDH), and all experiments were done in triplicate.

Chromatin immunoprecipitation (ChIP)
ChIP was performed as described previously (30). Briefly, DNA samples of AD and AI cells, purified from 4 µg Sp1-antibody immunoprecipitates, were amplified by PCR using a high fidelity PCR system and the primers hAR5F: 5'-AGC TGC TAA AGA CTC GGA GG-3' and hAR3R: 5'-GGA GTT ACC TCT CTG CAA AC-3' that covers DNA sequence containing two adjacent Sp1 sites, located at 429–442 of 5'-UTR of AR gene. PCR products were analyzed by electrophoresis on 2% agarose gels and stained with SYBR Gold nucleic acid gel stain (Molecular Probes).

Electrophoretic mobility shift assay (EMSA)
The assay was performed as described previously (31). Nuclear proteins extracted from AD or AI cells were reacted with 32P-labeled Sp1 oligos, and their mutants corresponding to the 5'-UTR of AR gene containing two adjacent Sp1 sites located at 429–442, in binding buffer containing 1 µg of poly(dI–dC) for 30 min at room temperature. The reaction mixtures were then subjected to electrophoresis in 8% native polyacrylamide gel, and the binding complexes visualized by exposure on X-ray films. Specificity of the Sp1 binding complexes were established using specific and non-specific competitions (31).


   Results
Top
 Abstract
 Background
 Materials and methods
 Results
Discussion
 Authors' contributions
 References
 
Androgen-dependent and independent growth of AD and AI cells
We established an androgen-independent (AI) prostate cancer cell line from androgen-dependent (AD) LNCaP cells, and characterized the differences between the AD and AI lines. The major differences of AI cells from their parental AD cells are androgen-independent growth and resistance to chemotherapeutic agents, due to at least in part the overexpression of AR (5,27,28,32). Figure 1 shows the growth characteristics of AD, AI and PC-3 cells, in the presence or absence of androgen, as measured by MTT. An identical growth rate between AD and AI cells was observed when both cell lines were incubated in the culture medium containing androgens (10% FBS) (Figure 1, left panel). However, the AD cell culture became static when the cells were supplied with an androgen-depleted medium (10% CSFBS), confirming that androgens are required for the growth of the AD cells (Figure 1 right panel). In contrast, AI cells grew well in the charcoal stripped medium just as that with regular FBS medium, demonstrating that the cells became androgen-independent (Figure 1, left and right panels). The growth pattern of AI cells is comparable with the androgen-independent PC-3 cells (Figure 1).


Figure 1
View larger version (19K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1 Growth behaviors of AD and AI cells in the presence (A) or absence (B) of androgen. AD, AI or PC-3 prostate cancer cells at exponential growth phase at density of 2500 cells were seeded per well of 96-well dishes, and cell growth at different time points was determined by MTT as described in Materials and methods.

 
Effects of PEITC on AD and AI cell growth
The effects of PEITC on the cell growth of AD and AI were examined by MTT assay. As shown in Figure 2 panel A, a significant growth inhibition was achieved in both cell lines after exposure to PEITC for 7 days, exhibiting an identical IC50 of ~0.6 µM. The two cell lines were also exposed to paclitaxel, which showed a stronger growth inhibitory activity than PEITC. The AD and AI cells displayed different sensitivity toward paclitaxel (4,5), with IC50 <0.01 nM for AD cells and 0.6 nM for AI cells, representing ~60-fold difference (Figure 2 panel B). The results are in line with previous observations that the AI cells are more resistant to chemotherapeutic agents, such as paclitaxel, retinoic acid and doxorubicin, than it's parental AD cells (4,28).


Figure 2
View larger version (18K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2 The effects of PEITC (A) and paclitaxel (B) on the growth of androgen-dependent (AD) and independent (AI) LNCaP cells. AD and AI cells grown exponentially were seeded into 96-well dishes at density of 2500 cells/well. After 24 h the cells were exposed to serial dilutions of PEITC (A) or Paclitaxel (B) for 7 days. The cell growth was measured by MTT as described in Materials and methods.

 
PEITC represses AR transcription via Sp1 inhibition
The effects of PEITC on the expression of AR were examined by a real-time PCR and western blotting. Figure 3 panel A shows that exposure of AD and AI cells to PEITC for 24 h resulted in a concentration-dependent decrease of the mRNA of AR. The maximal transcriptional repression occurred at 10 µM, with ~60% inhibition in both AD and AI cells. Concurrently, the protein levels of AR were also decreased, to a similar extent after exposure to PEITC (Figure 3, panel B).


Figure 3
View larger version (39K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3 Repression of AR transcription in LNCaP AD and AI cells by PEITC. (A) AD and AI cells at exponential growth phase were exposed to the indicated concentrations of PEITC for 24 h. Total RNA was extracted, and the mRNA of AR was determined by a real-time PCR as described in Materials and methods. All experiments were done in triplicate, and the data represent mean ± SD. (B) Effects of PEITC on protein levels of the AR in AD and AI cells: AD and AI cells at exponential growth phase were exposed to the indicated concentrations of PEITC for 24 h and the levels of AR were measured by western blotting (ß-actin was used as internal loading control) as described in Materials and methods. (C) Effects of PEITC on the activity of AR-luciferase reporter. AD and AI cells were transfected with 1 µg/6-well dish of pLARS-1 in serum-free medium using effectene. Twenty-four hours after the transfection, the cells were exposed to the indicated concentrations of PEITC in a serum containing medium for additional 24 h. The cells were then washed and lysed for determination of luciferase activity using Promega assay kit. Data represent mean ± SD from two independent experiments.

 
To further examine, whether the decrease in mRNA and protein levels of the AR was the result of repression of the AR promoter activity, AD cells were transfected with pLARS-1 (pAR-luc) that covers the human AR promoter from –741 to +907 (27). The exposure of the transfected AD and AI cells to PEITC for 24 h, resulted in a significant decrease of the reporter activities (Figure 3C), with AI cell line appearing to be more sensitive to PEITC than AD (Figure 3C, right panel). The decreased pLARS-1 activity matches the results obtained from real-time PCR (Figure 3A), indicating that repression of AR transcription may be the result of inhibiting the AR promoter.

Since Sp1 predominantly drives AR transcription, we speculated that the repression of AR transcription by PEITC could be mediated by an inhibition of Sp1. To examine this possibility, we tested the effects of PEITC on the activities of Sp1-luc, containing three repeats of Sp1 consensus sites, or its mutant mSp1-luc. Figure 4A shows that exposure of AD cells to a specific Sp1 inhibitor, mithramycin, caused a strong inhibition of the Sp1-luc activity, whereas trichostatin A (TSA) mediated a strong induction, as that described by others (4,26). Similar to mithramycin, PEITC mediated a significant inhibition of Sp1-luc activity in both AD and AI cells, in a concentration-dependent manner (Figure 4B). Mithramycin, TSA or PEITC showed no effect when the mutant mSp1-luc was used (data not shown).


Figure 4
View larger version (24K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4 The effects of PEITC on the activity of Sp1. (A and B) AD and AI cells at exponential growth phase were transfected with 1 µg/6-well dish of pSp1-luc plasmid DNA in a serum-free RPMI 1640 medium for 24 h. The culture was replaced with the medium containing 10% FBS, and the cells treated with indicated concentrations of mithramycin or TSA (A), or PEITC (B) for additional 24 h. The cells were washed, lysed and luciferase activity assayed (unit per mg of the proteins). Data represent mean ± SD from three independent experiments. (C) AD and AI cells at exponential growth phase were exposed to the indicated concentrations of PEITC for 24 h, and the Sp1 levels measured by western blotting (ß-actin was used as internal loading control) as described in Materials and methods.

 
To analyze whether the inhibition of Sp1 by PEITC was a reflection of reduced Sp1 protein level, western blotting was performed. As revealed in Figure 4C, exposure for 24 h to 5.0 or 10 µM of PEITC caused, respectively, >50 and 70% reduction of Sp1 protein in both cell lines. The data thus suggest that the decrease of Sp1 activity may associate with an inhibition of Sp1 expression.

The interaction of the Sp1 site with Sp1 protein was examined by in vivo ChIP assay shown in Figure 5A. A clear PCR product covering the Sp1 site (429–442) was obtained when Sp1 monoclonal antibody was used to immunoprecipitate the chromatin cross-linked proteins from AD or AI cells.


Figure 5
View larger version (32K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 5 The effects of PEITC on the formation of Sp1 nuclear binding complex. (A) Interaction of Sp1 with Sp1 elements present in the AR gene by ChIP. DNA samples of AD and AI cells, purified from the Sp1-antibody immunoprecipitates, were amplified by PCR using the primers covering DNA sequence containing two adjacent Sp1 sites located at 429–442 of 5'-UTR of the AR gene. (B) EMSA was performed using 5 µg of the nuclear proteins, extracted from the cells exposed to mithramycin or PEITC for 24 h, to react with 32P-labeled Sp1 oligo probe corresponding to the 5'-UTR of the AR gene as described in Materials and methods. (C) LNCaP AD cells at exponential growth were transfected with 1 µg/6 mm dish of pARSp1-3 plasmid DNA, in a serum-free RPMI-1640 medium for 24 h. The culture was replaced with the medium containing 10% FBS and exposed to indicated concentrations of PEITC for additional 24 h. The cells were washed and luciferase activity assayed (expressed as unit per milligram of the proteins).

 
To further examine whether a decreased Sp1 activity was involved in the PEITC-induced AR downregulation, the interaction of the Sp1 with Sp1 element in AR promoter was studied by EMSA using 32P-labeled double strands of ARSp1-3 oligonucleotide, corresponding to the sequence located 429–442 in the 5'-UTR of AR gene, as a probe. As shown in Figure 5B, a significant decrease in the formation of Sp1-nuclear binding complex was observed, when the nuclear extracts from the mithramycin-treated AD cells were used to react with 32[P]-labeled Sp1 oligonucleotide probe. Similar to mithramycin, PEITC induced a moderate decrease of the Sp1-nuclear complex formation, in a concentration-dependent manner, suggesting that the decreased Sp1 binding to the AR promoter could account for, at least in part, the decrease of AR transcription. This interpretation was supported by the data shown in Figure 5C, that PEITC inhibited the AD cell reporter activity of pARSp1-3, a Sp1 specific AR promoter driven luciferase reporter.

PEITC accelerates degradations of AR and Sp1 protein
To further establish the relationship between Sp1 and AR expression, a dynamic effect of PEITC on the protein levels of AR and Sp1 was studied. After exposure of AD cells to 10 µM PEITC, the level of AR protein showed a time-related alteration. A >80% reduction of AR protein was observed after 1 h, with maximal reduction reached by 7 h. Then the AR level was recovered to some extent at 24 h until full restoration by 48 h (Figure 6A). In comparison, the magnitude of Sp1 reduction, mediated by the same PEITC treatment, was less extensive and slower, with the maximum reduction observed at 24 h, and the Sp1 level was fully recovered by 48 h. The data suggest that the PEITC-mediated AR reduction does not result solely from a decrease of Sp1 level.


Figure 6
View larger version (29K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 6 Dynamic effects of PEITC on degradation of Sp1 and AR proteins. (A) Time course effect of PEITC on Sp1 and AR proteins: AD cells were exposed to 10 µM PEITC for different time length and western blot performed to measure the levels of Sp1 and AR proteins. (B) Effect of inhibitors of protein synthesis or proteasome on Sp1 and AR degradation by PEITC: AD cells were exposed to PEITC (10 µM), CHX (25 µM) or MG-132 (2.5 µM) alone, or to PEITC plus CHX or MG-132 for 24 h, and the protein levels of Sp1 and AR determined by western blotting. (C) Time course effect of proteasome inhibitor on AR degradation; AD cells were exposed to MG-132 (2.5 µM) for indicated time period, and the protein levels of AR and p21WAF1 measured by western blotting. (D) AD cells were exposed to PEITC (10 µM) for 3 h, in the presence or absence of MG-132 (2.5 µM), and the level of AR protein determined by western blotting.

 
To examine whether the suppression of the AR and Sp1 expression by PEITC requires protein synthesis, or is a result of proteasome-dependent protein degradation, AD cells were exposed for 24 h to 10 µM PEITC, protein synthesis inhibitor cycloheximide (CHX, 25 µM), or proteasome inhibitor MG-132 (2.5 µM) alone, or to the combination of PEITC plus CHX or MG-132. As described in Figure 6B, the PEITC inhibitory effect for Sp1 was completely abrogated by the presence of PEITC plus either CHX or MG-132. The data suggest that downregulation of Sp1 by PEITC may require protein synthesis for a proteasome activity, thereby enhancing Sp1 protein degradation. The results showed further that the level of Sp1 was unchanged in the presence of either CHX or MG-132 without PEITC, indicating that the steady-state levels of Sp1 protein could not be altered by either a mechanism involving protein synthesis inhibition, or by inhibiting protein-degradation alone.

Figure 6C shows that in contrast to Sp1, the combination of PEITC with CHX or MG-132 revealed a synergistic effect to downregulate the AR protein level, as compared with PEITC, CHX or MG-132 alone (Figure 6B). A control experiment performed in parallel showed that PEITC prevented degradation of cyclin inhibitor p27, which required new protein synthesis (Figure 6B). Our observation, that AR level is sensitive to protein synthesis inhibitor CHX, is consistent with a previous report (33). However, contrary to our observation, the proteasome inhibitor MG-132 was described to accumulate AR protein in LNCaP cells after a 2 h exposure (34). Considering that the difference of exposure time could be a factor, a time course study of MG-132 on the level of AR protein was initiated. As described in Figure 6C, the expression of the AR protein was found to be a function of exposure time to MG-132. The AR protein gradually accumulated after exposure to MG-132 for up to 7 h, which was followed by a decrease at 16 h, with the lowest level reached by 24 h. As a control experiment, the level of p21WAF1/CIP1 protein under the same experimental conditions was shown accumulates in direct proportion to the MG-132 exposure time, with the highest level reached at 16–24 h (Figure 6C). This indicates that post-translational regulation of AR and p21WAF1/CIP1 is mediated by different mechanisms. To examine whether MG-132 can prevent a PEITC-mediated decrease of AR protein at its optimal time points, AD cells were treated with 10 µM PEITC, in the presence or absence of MG-132 for 3 h, and the AR protein level determined. As depicted in Figure 6D, the PEITC-mediated AR decrease was blocked by MG-132. Thus, our data indicated that the degradation of the AR protein by PEITC could be reduced or accelerated by the presence of proteasome.


   Discussion
Top
 Abstract
 Background
 Materials and methods
 Results
 Discussion
Authors' contributions
 References
 
We have demonstrated in this article, for the first time to our knowledge, that PEITC may have dual effects to downregulate the AR, at transcriptional level via an inhibition of Sp1 expression, and at post-translational level by mediating protein degradation. These PEITC activities were demonstrated in both androgen dependent and independent prostate cancer cells. Several lines of evidence have been presented in this regard. On the transcriptional level, the exposure to PEITC decreased the AR mRNA level and the AR promoter activity. In parallel, the Sp1 consensus-driven luciferase reporter was inhibited, the protein level of Sp1 reduced and the interaction of Sp1 with the AR promoter decreased. On the post-translational level, the data revealed that the degradation of the AR and Sp1 proteins by PEITC could be blocked by a proteasome inhibitor, indicating that PEITC mediates a proteasome-dependent degradation. The downregulation of AR and Sp1 by PEITC is consistent with our recent observations in rats. Oral feeding of rats with PEITC was shown to inhibit the testosterone-stimulated prostatic cell growth via a decrease of Sp1 expression, and the Sp1 binding complex formation which modulates the AR and cell cycle progression (35). Recently, the effect of other chemopreventive agents, such as green tea extracts and indole-3-carbinol, were also reported to downregulate the AR (36,37).

The AR belongs to the family of genes that do not contain the classical regulatory sequences TATA and CCAAT boxes for its transcription. Instead, there are multiple Sp1 located near the transcriptional initiation site (38–43), that are known to predominantly drive AR transcription (20,21,44). We recently identified two adjacent Sp1 sites located at 429–442 of 5'-UTR of AR gene and named as ARSp1-3. To examine the transcriptional activity of ARSp1-3, we constructed a pARSp1-3 luciferase reporter by inserting three repeats of the two adjacent Sp1 sites, or its mutant into a basic vector pGL3-basic and observed 7- and 11-fold increase of the reporter activities in AD and AI cells, respectively, as compared with its mutant (45). The pARSp1-3 luciferase activity in both AD and AI cells was also found to correlate with the Sp1-nuclear binding complex formation (45). In this study, we further confirmed by in vivo ChIP (Figures 5A) the binding of Sp1 to the two adjacent Sp1 sites. In addition, the formation of ARSp1-3 nuclear binding complex could be blocked moderately by PEITC in a concentration-dependent manner (Figures 5B), which paralleled the inhibition of pARSp1-3 luciferase reporter activity. The data thus demonstrated that inhibition of Sp1 activity by PEITC led to a repression of AR transcription. As a Sp1 inhibitor, the mechanism by which PEITC inhibits Sp1 expression is different from a classical Sp1 inhibitor, mithramycin. Mithramycin interferes the binding of Sp1 protein to its CG rich regulatory sites, via its specific association with the CG rich DNA sequence (46).

We have demonstrated that the dynamic characteristics of Sp1 and the AR are different. While the steady-state Sp1 level was stable and unchanged by inhibitors for protein synthesis or for proteasome alone, the effects of PEITC to reducing the Sp1 level could be blocked by CHX or MG-132 (Figure 6B). This finding indicated that the PEITC-mediated Sp1 degradation involves new protein synthesis and is proteasome dependent. We have described that the reduction of the AR after 24 h exposure to PEITC was synergistically enhanced by CHX or MG-132, supporting the conclusion that the AR is a fast turnover protein. On the other hand, exposure to a protein synthesis or proteasome inhibitor resulted in a reduction of the steady-state AR level. In fact, the activity of the proteasome inhibitor MG-132 on the AR stability was intriguing, i.e. short time exposure to MG-132 resulted in accumulation of the AR protein while longer exposure caused reduction (Figure 6C). This has indicated that mechanism independent of proteasome may also be involved in the regulation of AR stability. A similar observation of a conflicting effect of MG-132 on AR stability was reported previously (47).

The detailed processes of proteasome-dependent AR degradation by PEITC, however, need to be further analyzed. One possible mechanism could involve the chaperone complex, which binds to the AR and is essential for AR stability and maturation. The proteasome is known to associate with the chaperone complex including the heat-shock protein 90 (Hsp90) (48–51). Chen et al. (47) recently reported that while the histone deacetylase inhibitor LAQ824 repressed AR transcription, LAQ824 also stimulated AR degradation via inactivation of the Hsp90 molecular chaperone. The inactivation by LAQ824 was due to an enhanced acetylation of Hsp90, thereby inhibiting its ATP binding activity. The study also showed that LAQ824 reduced the stability of Akt, which stimulates AR phosphorylation and increases AR stability. Since we recently reported that PEITC is an inhibitor of histone deacetylases (52), PEITC could induce similar acetylation of Hsp90 molecular chaperone and inactivate its binding to AR, promoting AR degradation. Since our data described that the AR protein level was diminished earlier than Sp1 by PEITC, the AR could be degraded by proteasome ahead of Sp1. As a transcription factor, Sp1 is mainly located in the nucleus rather than cytoplasma, and it is unlikely that Sp1 plays a direct role in proteasome-dependent AR degradation. One possible interpretation could be that the AR reduction is mediated by Sp1-independent proteasome degradation, and compounded by a downregulated Sp1 effect on AR transcription. In considering PEITC as an inhibitor of histone deacetylases, experiments need to be performed for the effects of PEITC on acetylation of AR and Sp1 proteins, which could affect their stability (53). In addition, the effect of PEITC on AR transcription, in the presence or absence of the protein synthesis inhibitor CHX, at different time points, may further help to clarify the role of Sp1.

With the information that AR overexpression plays a predominant role in the progression to advanced prostate cancer, novel therapies aiming at AR repression have drawn much attention (6,7,54–56). Since PEITC has been demonstrated to induce growth arrest and apoptosis in prostate cancer cells, our current observation has provided the basis for further exploring PEITC in preventing and inhibiting the androgen-independent progression of prostate cancer. The experimental results also support the speculation that the isothiocyanates present in cruciferous vegetables, by their action to modulate AR and the excessive growth of the prostate, are important dietary factors for prostate cancer chemoprevention.


   Authors' contributions
Top
 Abstract
 Background
 Materials and methods
 Results
 Discussion
 Authors' contributions
References
 
L.G.W. was instrumental in study design, performing experiments and drafting manuscript. X.M.L. performed experiments of cell transfection, culture analyses and EMSA. J.W.C. participated in study design and manuscript preparation.


   Acknowledgments
 
This work was supported by a US Army grant DAMD17-03-1-011.

Conflict of Interest Statement: None declared.


   References
Top
 Abstract
 Background
 Materials and methods
 Results
 Discussion
 Authors' contributions
 References
 

  1. Heinlein C.A. and Chang C. (2004) Androgen receptor in prostate cancer. Endocr. Rev. 25:276–308.[Abstract/Free Full Text]
  2. Santos A.F., Huang H., Tindall D.J. (2004) The androgen receptor: a potential target for therapy of prostate cancer. Steroids 69:79–85.[CrossRef][Web of Science][Medline]
  3. Taplin M.E. and Balk S.P. (2004) Androgen receptor: a key molecule in the progression of prostate cancer to hormone independence. J. Cell. Biochem. 91:483–490.[CrossRef][Web of Science][Medline]
  4. Wang L.G., Ossowski L., Ferrari A.C. (2001) Overexpressed androgen receptor linked to p21WAF1 silencing may be responsible for androgen independence and resistance to apoptosis of a prostate cancer cell line. Cancer Res. 61:7544–7551.[Abstract/Free Full Text]
  5. Liu X.M., Jiang J.D., Ferrari A.C., Budman D.R., Wang L.G. (2003) Unique induction of p21(WAF1/CIP1) expression by vinorelbine in androgen-independent prostate cancer cells. Br. J. Cancer 89:1566–1573.[CrossRef][Web of Science][Medline]
  6. Culig Z., Steiner H., Bartsch G., Hobisch A. (2005) Mechanisms of endocrine therapy-responsive and -unresponsive prostate tumours. Endocr. Relat. Cancer 12:229–244.[Abstract/Free Full Text]
  7. Burnstein K.L. (2005) Regulation of androgen receptor levels: implications for prostate cancer progression and therapy. J. Cell. Biochem. 95:657–669.[CrossRef][Web of Science][Medline]
  8. Kolonel L.N., Hankin J.H., Whittemore A.S., et al. (2000) Vegetables, fruits, legumes and prostate cancer: a multiethnic case–control study. Cancer Epidemiol. Biomarkers Prev. 9:795–804.[Abstract/Free Full Text]
  9. Cohen J.H., Kristal A.R., Stanford J.L. (2000) Fruit and vegetable intakes and prostate cancer risk. J. Natl Cancer Inst. 92:61–68.[Abstract/Free Full Text]
  10. Kristal A.R. and Lampe J.W. (2002) Brassica vegetables and prostate cancer risk: a review of the epidemiological evidence. Nutr. Cancer 42:1–9.[CrossRef][Web of Science][Medline]
  11. Fenwick G.R., Heaney R.K., Mullin W.J. (1983) Glucosinolates and their breakdown products in food and food plants. Crit. Rev. Food Sci. Nutr. 18:123–201.[Medline]
  12. Sones K., Heaney R.K., Fenwick G.R. (1984) An estimate of the mean daily intake of glucosinolatefrom cruciferous vegetables in the UK. J. Sci. Food Agric. 35:712–720.[CrossRef][Web of Science]
  13. Chung F.L., Morse M.A., Eklind K.I., Lewis J. (1992) Quantitation of human uptake of the anticarcinogen phenethyl isothiocyanate after a watercress meal. Cancer Epidemiol. Biomarkers Prev. 1:383–388.[Abstract]
  14. Hecht S.S. (1995) Chemoprevention by isothiocyanates. J. Cell. Biochem. 22:Suppl., 195–209.[CrossRef]
  15. Zhang Y., Kensler T.W., Cho C.G., Posner G.H., Talalay P. (1994) Anticarcinogenic activities of sulforaphane and structurally related synthetic norbornyl isothiocyanates. Proc. Natl Acad. Sci. USA 91:3147–3150.[Abstract/Free Full Text]
  16. Yang C.S., Smith T.J., Hong J.Y. (1994) Cytochrome P-450 enzymes as targets for chemoprevention against chemical carcinogenesis and toxicity: opportunities and limitations. Cancer Res. 54:1982s–1986s.[Free Full Text]
  17. Chiao J.W., Chung F., Krzeminski J., Amin S., Arshad R., Ahmed T., Conaway C.C. (2000) Modulation of growth of human prostate cancer cells by the N-acetylcysteine conjugate of phenethyl isothiocyanate. Int. J. Oncol. 16:1215–1219.[Web of Science][Medline]
  18. Wang L., Liu D., Ahmed T., Chung F.L., Conaway C., Chiao J.W. (2004) Targeting cell cycle machinery as a molecular mechanism of sulforaphane in prostate cancer prevention. Int. J. Oncol. 24:187–192.[Web of Science][Medline]
  19. Chiao J.W., Wu H., Ramaswamy G., Conaway C.C., Chung F.L., Wang L., Liu D. (2004) Ingestion of an isothiocyanate metabolite from cruciferous vegetables inhibits growth of human prostate cancer cell xenografts by apoptosis and cell cycle arrest. Carcinogenesis 25:1403–1408.[Abstract/Free Full Text]
  20. Takane K.K. and McPhaul M.J. (1996) Functional analysis of the human androgen receptor promoter. Mol. Cell. Endocrinol. 119:83–93.[CrossRef][Web of Science][Medline]
  21. Mizokami A. and Chang C. (1994) Induction of translation by the 5'-untranslated region of human androgen receptor mRNA. J. Biol. Chem. 269:25655–25659.[Abstract/Free Full Text]
  22. Supakar P.C., Song C.S., Jung M.H., Slomczynska M.A., Kim J.M., Vellanoweth R.L., Chatterjee B., Roy A.K. (1993) A novel regulatory element associated with age-dependent expression of the rat androgen receptor gene. J. Biol. Chem. 268:26400–26408.[Abstract/Free Full Text]
  23. Supakar P.C., Jung M.H., Song C.S., Chatterjee B., Roy A.K. (1995) Nuclear factor kappa B functions as a negative regulator for the rat androgen receptor gene and NF-kappa B activity increases during the age-dependent desensitization of the liver. J. Biol. Chem. 270:837–842.[Abstract/Free Full Text]
  24. Grossmann M.E. and Tindall D.J. (1995) The androgen receptor is transcriptionally suppressed by proteins that bind single-stranded DNA. J. Biol. Chem. 270:10968–10975.[Abstract/Free Full Text]
  25. Grossmann M.E., Lindzey J., Blok L., Perry J.E., Kumar M.V., Tindall D.J. (1994) The mouse androgen receptor gene contains a second functional promoter which is regulated by dihydrotestosterone. Biochemistry 33:14594–14600.[CrossRef][Medline]
  26. Sowa Y., Orita T., Minamikawa S., Nakano K., Mizuno T., Nomura H., Sakai T. (1997) Histone deacetylase inhibitor activates the WAF1/Cip1 gene promoter through the Sp1 sites. Biochem. Biophys. Res. Commun. 241:142–150.[CrossRef][Web of Science][Medline]
  27. Wang L.G., Ossowski L., Ferrari A.C. (2004) Androgen receptor level controlled by a suppressor complex lost in an androgen-independent prostate cancer cell line. Oncogene 23:5175–5184.[CrossRef][Web of Science][Medline]
  28. Gao M., Ossowski L., Ferrari A.C. (1999) Activation of Rb and decline in androgen receptor protein precede retinoic acid-induced apoptosis in LNCaP cells and their androgen-independent derivative. J. Cell. Physiol. 179:336–346.[CrossRef][Web of Science][Medline]
  29. Wang L.G., Liu X.M., Kreis W., Budman D.R. (1997) Down-regulation of prostate-specific antigen expression by finasteride through inhibition of complex formation between androgen receptor and steroid receptor-binding consensus in the promoter of the PSA gene in LNCaP cells. Cancer Res. 57:714–719.[Abstract/Free Full Text]
  30. Kinoshita Y. and Johnson E.M. (2004) Site-specific loading of an MCM protein complex in a DNA replication initiation zone upstream of the c-MYC gene in the HeLa cell cycle. J. Biol. Chem. 279:35879–35889.[Abstract/Free Full Text]
  31. Wang L.G., Liu X.M., Li Z.R., Denstman S., Bloch A. (1994) Differential binding of nuclear c-ets-1 protein to an intron I fragment of the c-myb gene in growth versus differentiation. Cell Growth Differ. 5:1243–1251.[Abstract]
  32. Wang L.G., Ossowski L., Ferrari A.C. (2001) Overexpressed Androgen Receptor linked to p21(WAF1) silencing may be responsible for androgen independence and resistance to apoptosis of a prostate cancer cell line. Cancer Res. 61:7544–7551.[Abstract/Free Full Text]
  33. Cinar B., De Benedetti A., Freeman M.R. (2005) Post-transcriptional regulation of the androgen receptor by mammalian target of rapamycin. Cancer Res. 65:2547–2553.[Abstract/Free Full Text]
  34. Sheflin L., Keegan B., Zhang W., Spaulding S.W. (2000) Inhibiting proteasomes in human HepG2 and LNCaP cells increases endogenous androgen receptor levels. Biochem. Biophys. Res. Commun. 276:144–150.[CrossRef][Web of Science][Medline]
  35. Anastasia A., Beklemisheva J.F., Yunshin Y., Wang L.G., Chiao J.W. (2006) Modulating testosterone stimulated prostate growth by phenethyl isothiocyanate via Sp1 and androgen receptor down-regulation. Prostate in press.
  36. Yuan H., Gong A., Young C.Y. (2005) Involvement of transcription factor Sp1 in quercetin-mediated inhibitory effect on the androgen receptor in human prostate cancer cells. Carcinogenesis 26:793–801.[Abstract/Free Full Text]
  37. Hsu J.C., Zhang J., Dev A., Wing A., Bjeldanes L.F., Firestone G.L. (2005) Indole-3-carbinol inhibition of androgen receptor expression and downregulation of androgen responsiveness in human prostate cancer cells. Carcinogenesis 26:1896–1904.[Abstract/Free Full Text]
  38. Tilley W.D., Wilson C.M., Marcelli M., McPhaul M.J. (1990) Androgen receptor gene expression in human prostate carcinoma cell lines. Cancer Res. 50:5382–5386.[Abstract/Free Full Text]
  39. Tilley W.D., Marcelli M., McPhaul M.J. (1990) Expression of the human androgen receptor gene utilizes a common promoter in diverse human tissues and cell lines. J. Biol. Chem. 265:13776–13781.[Abstract/Free Full Text]
  40. Husmann D.A., Wilson C.M., McPhaul M.J., Tilley W.D., Wilson J.D. (1990) Antipeptide antibodies to two distinct regions of the androgen receptor localize the receptor protein to the nuclei of target cells in the rat and human prostate. Endocrinology 126:2359–2368.[Abstract/Free Full Text]
  41. Tilley W.D., Marcelli M., McPhaul M.J. (1990) Recent studies of the androgen receptor: new insights into old questions. Mol. Cell. Endocrinol. 68:C7–C10.[CrossRef][Web of Science][Medline]
  42. Faber P.W., King A., van Rooij H.C., Brinkmann A.O., de Both N.J., Trapman J. (1991) The mouse androgen receptor. Functional analysis of the protein and characterization of the gene. Biochem. J. 278:269–278.[Medline]
  43. Faber P.W., van Rooij H.C., van der Korput H.A., Baarends W.M., Brinkmann A.O., Grootegoed J.A., Trapman J. (1991) Characterization of the human androgen receptor transcription unit. J. Biol. Chem. 266:10743–10749.[Abstract/Free Full Text]
  44. Baarends W.M., Themmen A.P., Blok L.J., Mackenbach P., Brinkmann A.O., Meijer D., Faber P.W., Trapman J., Grootegoed J.A. (1990) The rat androgen receptor gene promoter. Mol. Cell. Endocrinol. 74:75–84.[CrossRef][Web of Science][Medline]
  45. Wang L.G. and A.C.F. (2005) Mithramycin targets androgen receptor transcription—a potential application for the treatment of hormone refractory prostate cancer. Multidisciplinary Prostate Cancer Symposium Proceedings 138:.
  46. Majee S., Dasgupta D., Chakrabarti A. (1999) Interaction of the DNA-binding antitumor antibiotics, chromomycin and mithramycin with erythroid spectrin. Eur. J. Biochem. 260:619–626.[Web of Science][Medline]
  47. Chen L., Meng S., Wang H., Bali P., Bai W., Li B., Atadja P., Bhalla K.N., Wu J. (2005) Chemical ablation of androgen receptor in prostate cancer cells by the histone deacetylase inhibitor LAQ824. Mol. Cancer Ther. 4:1311–1319.[Abstract/Free Full Text]
  48. Jiang J., Ballinger C.A., Wu Y., Dai Q., Cyr D.M., Hohfeld J., Patterson C. (2001) CHIP is a U-box-dependent E3 ubiquitin ligase: identification of Hsc70 as a target for ubiquitylation. J. Biol. Chem. 276:42938–42944.[Abstract/Free Full Text]
  49. Murata S., Minami Y., Minami M., Chiba T., Tanaka K. (2001) CHIP is a chaperone-dependent E3 ligase that ubiquitylates unfolded protein. EMBO Rep. 2:1133–1138.[CrossRef][Web of Science][Medline]
  50. Pratt W.B. and Toft D.O. (2003) Regulation of signaling protein function and trafficking by the hsp90/hsp70-based chaperone machinery. Exp. Biol. Med. (Maywood) 228:111–133.[Abstract/Free Full Text]
  51. Solit D.B., Scher H.I., Rosen N. (2003) Hsp90 as a therapeutic target in prostate cancer. Semin. Oncol. 30:709–716.[CrossRef][Web of Science][Medline]
  52. Wang L.G., Feng J., Liu M.X., Beklemisheva A., Chiao J.W. (2005) Phenethyl isothiocyanates-mediated cross talk between chromatin remodeling and DNA demethylation—restoring GSTP1 expression in prostate cancer. Am. Assoc. Cancer Res. 46:1226.
  53. Gaughan L., Logan I.R., Neal D.E., Robson C.N. (2005) Regulation of androgen receptor and histone deacetylase 1 by Mdm2-mediated ubiquitylation. Nucleic Acids Res. 33:13–26.[Abstract/Free Full Text]
  54. Eder I.E., Culig Z., Ramoner R., Thurnher M., Putz T., Nessler-Menardi C., Tiefenthaler M., Bartsch G., Klocker H. (2000) Inhibition of LncaP prostate cancer cells by means of androgen receptor antisense oligonucleotides. Cancer Gene Ther. 7:997–1007.[CrossRef][Web of Science][Medline]
  55. Zegarra-Moro O.L., Schmidt L.J., Huang H., Tindall D.J. (2002) Disruption of androgen receptor function inhibits proliferation of androgen refractory prostate cancer cells. Cancer Res. 62:1008–1013.[Abstract/Free Full Text]
  56. Haag P., Bektic J., Bartsch G., Klocker H., Eder I.E. (2005) Androgen receptor down regulation by small interference RNA induces cell growth inhibition in androgen sensitive as well as in androgen independent prostate cancer cells. J. Steroid Biochem. Mol. Biol. 96:251–258.[CrossRef][Web of Science][Medline]
Received October 19, 2005; revised April 28, 2006; accepted May 5, 2006.


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 CarcinogenesisHome page
P.-H. Huang, D. Wang, H.-C. Chuang, S. Wei, S. K. Kulp, and C.-S. Chen
{alpha}-Tocopheryl succinate and derivatives mediate the transcriptional repression of androgen receptor in prostate cancer cells by targeting the PP2A-JNK-Sp1-signaling axis
Carcinogenesis, July 1, 2009; 30(7): 1125 - 1131.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Res.Home page
A. Bommareddy, E.-R. Hahm, D. Xiao, A. A. Powolny, A. L. Fisher, Y. Jiang, and S. V. Singh
Atg5 Regulates Phenethyl Isothiocyanate-Induced Autophagic and Apoptotic Cell Death in Human Prostate Cancer Cells
Cancer Res., April 15, 2009; 69(8): 3704 - 3712.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow An erratum has been published
Right arrow A corrigendum has been published
Right arrow All Versions of this Article:
27/10/2124    most recent
bgl075v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (1)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Wang, L.G.
Right arrow Articles by Chiao, J.W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wang, L.G.
Right arrow Articles by Chiao, J.W.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?