Differential CARM1 expression in prostate and colorectal cancers
Coactivator-associated arginine methyltransferase 1 (CARM1) functions as a transcriptional coactivator of androgen receptor (AR)-mediated signaling. Correspondingly, overexpression of CARM1 has been associated with the development of prostate cancer (PCa) and its progression to androgen-independent PCa. In our preliminary study, however, the promoting effects of CARM1, with regard to androgen-stimulated AR target gene expression were minimal. These results suggested that the AR target gene expression associated with CARM1 may result primarily from non-hormone dependent activity. The goal of this study was to confirm the pattern of expression of CARM1 in human tumors and determine the mechanism of action in CARM1 overexpressed tumors.
Tissue microarray was used to determine the pattern of expression of CARM1 in human cancers by immunohistochemistry. CARM1 expression was also evaluated in prostate and colorectal surgical specimens and the clinical records of all cases were reviewed. In addition, a reporter transcription assay using the prostate-specific antigen (PSA) promoter was used to identify the signaling pathways involved in non-hormone-mediated signal activation associated with CARM1.
The tissue microarray showed that CARM1 was particularly overexpressed in the colorectal cancers while CARM1 expression was not prevalent in the prostate and breast cancers. Further studies using surgical specimens demonstrated that CARM1 was highly overexpressed in 75% of colorectal cancers (49 out of 65) but not in the androgen-independent PCa. In addition, CARM1's coactivating effect on the entire PSA promoter was very limited in both androgen-dependent and androgen-independent PCa cells. These results suggest that there are other factors associated with CARM1 expression in PSA regulation. Indeed, CARM1 significantly regulated both p53 and NF-κB target gene transcription.
The results of this study suggest that, in addition to its role in activation of steroid receptors, CARM1 functions as a transcriptional modulator by altering the activity of many transcriptional factors, especially with regard to androgen independent PCa and colorectal cancers.
KeywordsProstate Cancer Androgen Receptor LNCaP Cell Mouse Mammary Tumor Virus P160 Coactivators
Coactivator-associated arginine methyltransferase 1
Nuclear factor-kappa beta
androgen response element
mouse mammary tumor virus
myocyte enhancer factor 2c.
CARM1 is a protein with arginine-specific histone methyltransferase activity ; it initially was described as a transcriptional activator of the p160 family of nuclear receptor-associated proteins (Src-1, GRIP1/TIF2/Src-2, ACTR/AIB1/SRC-3)[2, 3, 4, 5]. The p160 coactivators act as primary coactivators through direct binding to the C-terminal region of nuclear receptors in a ligand-dependent manner. The p160 coactivators are involved in transcriptional activation by bringing secondary activators with them to the promoter. While p300 and CBP bind to the activation domain (AD) 1 of p160 , CARM1 binds to AD2 . Ultimately, CARM1 promotes nuclear receptor activity through the formation of a tertiary complex of CARM1, p300/CBP, and p160 [3, 5]. CARM1 has been shown to be a molecular switch that controls multiple classes of gene-specific transcription factors, including p53, NF-κB, LEF1/TCF4, and E2Fs [7, 8, 9, 10]. Therefore, CARM1 appears to play a pleiotropic role in cell proliferation and survival.
CARM1 has primarily been studied in association with nuclear receptors, including the estrogen receptors (ER) and androgen receptors (AR). Cancers of the breast and prostate have been shown to overexpress CARM1 [7, 11, 12, 13]. Overexpression of CARM1 has also been suggested in androgen-independent prostate cancers (PCa) [12, 13]. The altered expression of CARM1 could provide a growth advantage for the PCa cells by enhancing the AR transactivation function and target gene activation. Here, we show that CARM1 overexpression was surprisingly prevalent in colorectal cancers while only a fraction of tumors from the breast and prostate overexpressed CARM1. Accordingly, we demonstrate that CARM1-mediated AR target gene expression, namely prostate-specific antigen, was minimal. These findings suggest that CARM1's role is mainly linked to the p53 and NF-κB response in non-nuclear receptor mediated cancer development similar to colorectal cancers and androgen-independent progression of PCa.
Plasmids and Reagents
pSG5-CARM1 was generously provided by Dr. Michael Stallcup; p61-luc contains the entire PSA promoter as previously described . The pGL-ARE4-Luc contains a synthetic TATA and four tandem copies of androgen response elements (AREs), from the murine mammary tumor virus promoter, in the context of the pGL3 backbone (Promega). A human AR expression vector, pARO, was a gift from Dr. Jan Trapman. The p53-luc containing two copies of p53 response elements was purchased from BD biosciences. The p21-luc was provided by Dr. Bert Vogelstein. The pE2F1-luc was received from Dr. Chihuei Wang (Kaohsiung Medical University, Taiwan). The pTCF4RE-luc contains four copies of the TCF-4 response elements derived from the cyclin D1 promoter. pBV-cmyc-luc was generously provided by Dr. Bert Vogelstein. pRC/RSV-p300 was received from Dr. Richard Goodman (The Cleveland Clinic). CARM1 S229E mutants were made on pSG5-CARM1 backbone using QuickChange Lightning Site-Directed Mutagenesis kit (Stratagene, La Jolla, CA, USA). Polyclonal antibodies to CARM1 were received from Dr. Meei-Huey Jeng (Indiana University) and monoclonal antibodies to β-actin were purchased from Santa Cruz Biotechnology. Synthetic testosterone, R1881, was purchased from NEN Life Science (Boston, MA, USA) and used at a final concentration of 10 nM.
Human prostate cell lines, including P69, LNCaP, C4-2, CWR22RV, PC3 and DU145 were routinely cultured in RPMI media (Invitrogen) supplemented with 5% FBS at 37°C in an atmosphere containing 5% CO2 as described previously . MDA PCa 2b prostate cancer cells were grown in BRFF-HPC1 medium (Athena Environmental Sciences, Inc., Baltimore, MD, USA) with 20% FBS. CV-1 monkey kidney cells, MDA-MB-231 human breast cancer cells, and caco-2 colon cancer cells were maintained in DMEM containing 10% FBS. All cultures were fed with fresh medium every 3-4 days.
Total RNA extraction from selected cultured cells was performed as previously described . RNA from MDA231 cells was used for standardizing the expression levels of CARM1 and GAPDH. Fifty nanograms (ng) of total RNA from each sample were used to detect real-time RT-PCR (QRT-PCR) products with Taqman probes and an ABI PRISM 7700 sequence detection system (PE Applied Biosystems, Foster City, CA, USA). PCR cycling conditions for all of the samples were as follows: 30 min at 48°C for reverse transcription; 10 min at 95°C for AmpliTaq Gold activation; and 40 cycles for the melting (95°C, 15 s) and annealing/extension (60°C, 1 min) steps. CARM1 and GAPDH primers and probes for QRT-PCR were designed using the PRIMER Express program (PE Applied Biosystems). The sequences of the CARM1 primers and probe were as follows: forward, 5'-ttgatgttggctgtggctctgg-3'; reverse, 5'-atgggctccgagatgatgatgtcc-3'; probe, 5'-FAM-caacctgacggaccgcatcgtg-TAMRA-3'. The sequences of the GAPDH primers and probe were as follows: forward, 5'-gaaggtgaaggtcggagtc-3'; reverse, 5'-gaagatggtgatgggatttc-3'; probe, 5'-VIC-caagcttcccgttctcagcc-TAMRA-3'. All QRT-PCR experiments were performed twice in duplicate in one 96-well plate. Using the comparative CT method (PE Applied Biosystems), the resulting Ct values were converted to picogram(pg) quantities according to each standard curve. Then, the quantity of CARM1 was normalized to GAPDH and subtracted from no reverse transcriptase controls. This value was then averaged for each duplicate.
Approximately 1 × 105 cells were plated in a 24-well plate 16 hours before the transfection. To determine the hormone effect, the cells were grown under 5% charcoal dextran-treated (CDT) FBS for three days before the transfection. The transfections were carried out using the Lipofectamine 2000 (Invitrogen) with 0.1 μg of reporter, 0.1 μg of test plasmid, and 2 ng renilla as described by the manufacturer's protocol. Six hours after transfection, the cells were washed and fed with medium containing 5% CDT-FBS. If needed, the cells were treated with either R1881 synthetic androgen or ethanol. After 36 hours, the cells were washed with PBS, lysed with 100 μl of passive lysis buffer, and assayed for luciferase activity as relative light units using the Dual Luciferase assay system (Promega). The transfection experiments were performed in triplicate and the results are reported as the mean ± S.D. The relative luciferase activity (RLU) was measured. When necessary, the RLU from the hormone-treated group was normalized by the non-hormone-treated group and the values represented as fold change.
Multi-tumor tissue microarray slides were obtained from the Cooperative Human Tissue Network under the Tissue Array Research Program (TARP) of the National Cancer Institute, The National Institutes of Health, Bethesda, MD, USA. Human colorectal tumors were obtained from the Indiana University Tissue Bank. There were 65 tumors with matching normal tissues adjacent to the tumors. All tumors were formalin-fixed and paraffin-embedded. After deparaffinizing the tissues, slides were microwaved for 10 min, three times in 10 mM citrate buffer (pH 6.0) to retrieve the antigens. Then, endogenous peroxidase was removed by the treatment of the tissues with 0.3% H2O2 followed by avidine-biotin blocking. To inhibit nonspecific binding of antibodies, the slides were treated with 3% normal goat serum before the overnight incubation of the slides with anti-CARM1 antibodies. Then, the signals were amplified by the horseradish peroxidase-DAB detection method.
Western Blot Assay
The cells were grown to 80% confluence in P60 culture dishes containing 5% FBS-T media. The cells were then lysed in protein extraction buffer (1× TBS, 1% NP-40, 0.5% sodium deoxycholate), 0.1% SDS and protease inhibitors. Twenty μg of total cell lysates were loaded onto a 10% Bis-Tris gel and separated using the electroporation system (Biorad). After the proteins were transferred to a PVDF membrane, primary antibodies were applied, followed by incubation with horse peroxidase-conjugated secondary antibodies. The blots were developed by the ECL detection system (Pierce).
Chromatin immunoprecipitation was performed as described  using antibodies against p53, NF-κB p65, or CARM1. The immunoprecipitated DNA was amplified by using specific primers as follows: p21 (ctcacatcctccttcttcag, cacacacagaatctgactccc), CCND1 (tcagggatggcttttggg, caacttcaacaaaactcccc).
CARM1 was highly expressed in colorectal cancers, but not in breast and prostate cancers
Score of CARM1 expression in human tumors
Score of CARM1 expression*
CARM1 was not overexpressed in androgen-independent prostate cancer
CARM1 was particularly overexpressed in colorectal tumors
CARM1's coactivating function for androgen activity was minimal in androgen-independent prostate cancer cells
CARM1 exerts its effect by regulation of p53 and NF-κB responses
Aberrant expression of CARM1 has been suggested to be important in tumors of the breast and colon [7, 11, 12, 13]. Our study, however, demonstrated that CARM1 was mainly overexpressed in colorectal tumors; with a frequency as high as 68% (Table 1), compared to other tumors from the brain, skin, ovary, lymphocytes, lung, breast, and prostate. CARM1 overexpression was noted in some hormone-dominated tumors such as breast cancer (27%) and ovarian cancer (17%), and in only 6% of prostate cancers. Considering the interest in the role of CARM1 in hormone receptor modulation these frequencies were lower than expected. These findings suggest that CARM1 may act by other transcription events for which there is limited information.
Follow-up studies with surgical specimens further demonstrated that CARM1 was commonly expressed in over 75% of colorectal tumors (49 out of 65 tumors expressed high levels of CARM1). Previously, CARM1 was shown to be overexpressed in both androgen-sensitive tumors and androgen-independent tumors with increased expression in castration-resistant prostate cancers . In a more recent study, CARM1 was not found to be generally overexpressed in androgen sensitive tumors, but has been suggested to play a role in androgen independent progression of prostate cancer . In this study, however, the CARM1 expression was low in most prostate cancers, including both androgen-sensitive and androgen-resistant tumors. The explanation for this difference between studies is not clear. However, one can assume that acquired hormone-resistant tumor samples are quite different depending on the treatment protocol from various institutions. This is especially true considering that acquisition of real hormone dependent tumors is difficult since most patients at this stage fail to respond to hormone ablation therapy. In cultured PCa cells, there was low expression of CARM1 in most androgen-dependent PCa cells while high expression of CARM1 was observed only in androgen-independent PC3 and DU145 cells (Fig. 2A-B). While these cells are described as androgen-independent PCa cells, they are generally considered as non-prostate-like cancer cells mainly due the deficiency of androgen receptors and thereby lack of PSA expression. In cultured cancer cells, mRNA and protein level of CARM1 were not always correlated, especially in 231 breast cancer cells and CWR22RV PCa cells. This discrepancy is not clearly understandable. Ohkura et al have demonstrated that there are at least one other isoform of CARM1 existed in endogenous cell level by alternative splicing . This isoform may have contributed to increased CARM1 protein expression compared to the level of RNA. It is also noteworthy that, in addition to its transcriptional coactivator role, CARM1 also regulates target gene expression by modulating protein stability, including p/CIP , AP-1 , and NF-κB . Taken together, our conclusions imply that CARM1 is not overexpressed in hormone refractory prostate cancers well correlate with this phenomenon.
CARM1 functions as a coactivator for many nuclear receptors, such as the estrogen receptor (ER), AR, the glucocorticoid receptor, and the thyroid receptor [1, 3, 10]. For example, CARM1 has been implicated in the positive regulation of ERα-mediated gene activation in response to estrogen signaling . Most nuclear receptors, including ARs, are bound directly by p160 coactivators whose N-terminal region interacts with CARM1 to transmit the activating signal to transcription machinery [10, 30]. In fact, CARM1 exerts an effect on methyltransferase activity to mediate this process . CARM1 functions as a coactivator of the AR to activate target promoters containing androgen response elements, such as MMTV, PSA, and probasin [10, 13]. Although in this study we observed the same phenomenon while we used a simple ARE construct in LNCaP cells, a strong coactivating function of CARM1 was not detected in most androgen-independent cells, including C42 and CWR22RV. Moreover, using the entire PSA promoter spanning 5.9 kb in the PSA promoter and enhancer region, CARM1 did not significantly increase promoter activity, especially in C42 and CWR22RV cells. Majumder et al demonstrated that CARM1 exerted strong coactivating effects on androgen-stimulated AR activity; however, they used a PSA promoter region containing only -5,322 to -3,873, which may explain the different findings in the studies . Nevertheless, our results suggest that regulation of PSA expression in PCa cells by CARM1 was mainly mediated by other transcription factors transactivating the PSA promoter region.
CARM1 also coactivates other transcription factors such as the myocyte enhancer factor 2C (MEF2C), p53, NF-κB, and E2Fs [7, 8, 9, 10], suggesting that CARM1's effect may not primarily be associated with the AR function in PCa cells . The results of this study demonstrated that forced expression of CARM1 alone slightly inhibited the activity of the p53-response, in contrast to the previous report by An et al . CARM1 has also been reported to exert coactivator effects on the GADD45 promoter, another p53 target gene, in p53-deficient cells . The reason for this difference is not clear. However, CARM1 may provide coactivator activity in a tissue-specific manner. In addition, we showed that CARM1 can activate the NF-κB response by stimulation of TNF-α. Both NF-κB and p53 pathways are key mediators of genes involved in the control of the cellular proliferation and apoptosis [31, 32]. Antiapoptotic genes that are directly activated by NF-κB include c-IAP1, c-IAP2, and IXAP, TNF receptor-associated factors, the Bcl-2 homologue A1/Bfl-1, and IEX-IL . NF-κB directly induces expression of A1/Bf1-1 by binding to specific sites in its promoter . NF-κB also acts in the control of the cell cycle. NF-κB activates the expression of cyclin D1, a positive regulator of G1-to-S-phase progression, by direct binding to multiple sites in its promoter . Thus, the apoptotic regulation by either NF-κB or p53 involves the regulation of multiple genes involved in different aspects of growth control. Therefore, the results of this study provide evidence that CARM1 may exert enzyme effects to non-hormone receptor, transcriptional regulatory molecules, such as p53 and NF-κB to modulate the target gene expression and subsequently the proliferation of colon cancer cells. At the same time, malignant progression of prostate cancer might be achieved by androgen-independent activity of CARM1.
In the field of steroid receptor-mediated transcriptional regulation, CARM1 functions as a coactivator that confers steroid receptor-mediated transactivation. The results of this study demonstrated that expression of CARM1 was especially high in colorectal cancers, over 75%. In contrast to previous reports, CARM1 was not overexpressed in the androgen sensitive and androgen-resistant prostate cancers. In addition, CARM1 modulated the transcription activity of p53 and NF-κB. These results suggest that CARM1 may exert its methylase activity on non-hormone receptor-related transcriptional machinery to modulate the biology of cancer cells, including androgen-independent prostate cancers and colorectal cancers.
This study was supported by a grant from the National R&D Program for Cancer Control, Ministry for Health, Welfare and Family Affairs, Republic of Korea (#0720120), the Korea Science & Engineering Foundation through the Medical Research Center for Gene Regulation (R13-2002-013-04002-0) at Chonnam National University, and a research grant (2007-CURIMS-DR009) from Chonnam National University Research Institute of Medical Sciences.
- 7.El Messaoudi S, Fabbrizio E, Rodriguez C, Chuchana P, Fauquier L, Cheng D, Theillet C, Vandel L, Bedford MT, Sardet C: Coactivator-associated arginine methyltransferase 1 (CARM1) is a positive regulator of the Cyclin E1 gene. Proc Natl Acad Sci USA. 2006, 103 (36): 13351-13356. 10.1073/pnas.0605692103.CrossRefPubMedPubMedCentralGoogle Scholar
- 10.Koh SS, Li H, Lee YH, Widelitz RB, Chuong CM, Stallcup MR: Synergistic coactivator function by coactivator-associated arginine methyltransferase (CARM) 1 and beta-catenin with two different classes of DNA-binding transcriptional activators. J Biol Chem. 2002, 277 (29): 26031-26035. 10.1074/jbc.M110865200.CrossRefPubMedPubMedCentralGoogle Scholar
- 12.Hong H, Kao C, Jeng MH, Eble JN, Koch MO, Gardner TA, Zhang S, Li L, Pan CX, Hu Z, et al: Aberrant expression of CARM1, a transcriptional coactivator of androgen receptor, in the development of prostate carcinoma and androgen-independent status. Cancer. 2004, 101 (1): 83-89. 10.1002/cncr.20327.CrossRefPubMedGoogle Scholar
- 16.Wu TT, Sikes RA, Cui Q, Thalmann GN, Kao C, Murphy CF, Yang H, Zhau HE, Balian G, Chung LW: Establishing human prostate cancer cell xenografts in bone: induction of osteoblastic reaction by prostate-specific antigen-producing tumors in athymic and SCID/bg mice using LNCaP and lineage-derived metastatic sublines. Int J Cancer. 1998, 77 (6): 887-894. 10.1002/(SICI)1097-0215(19980911)77:6<887::AID-IJC15>3.0.CO;2-Z.CrossRefPubMedGoogle Scholar
- 19.Foster CS, Falconer A, Dodson AR, Norman AR, Dennis N, Fletcher A, Southgate C, Dowe A, Dearnaley D, Jhavar S, et al: Transcription factor E2F3 overexpressed in prostate cancer independently predicts clinical outcome. Oncogene. 2004, 23 (35): 5871-5879. 10.1038/sj.onc.1207800.CrossRefPubMedGoogle Scholar
- 23.Teyssier C, Ou CY, Khetchoumian K, Losson R, Stallcup MR: Transcriptional intermediary factor 1alpha mediates physical interaction and functional synergy between the coactivator-associated arginine methyltransferase 1 and glucocorticoid receptor-interacting protein 1 nuclear receptor coactivators. Mol Endocrinol. 2006, 20 (6): 1276-1286. 10.1210/me.2005-0393.CrossRefPubMedGoogle Scholar
- 24.Ma H, Baumann CT, Li H, Strahl BD, Rice R, Jelinek MA, Aswad DW, Allis CD, Hager GL, Stallcup MR: Hormone-dependent, CARM1-directed, arginine-specific methylation of histone H3 on a steroid-regulated promoter. Curr Biol. 2001, 11 (24): 1981-1985. 10.1016/S0960-9822(01)00600-5.CrossRefPubMedGoogle Scholar
- 30.Stallcup MR, Kim JH, Teyssier C, Lee YH, Ma H, Chen D: The roles of protein-protein interactions and protein methylation in transcriptional activation by nuclear receptors and their coactivators. J Steroid Biochem Mol Biol. 2003, 85 (2-5): 139-145. 10.1016/S0960-0760(03)00222-X.CrossRefPubMedGoogle Scholar
- The pre-publication history for this paper can be accessed here:http://www.biomedcentral.com/1471-2407/10/197/prepub
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.