Purpose: Histone deacetylase (HDAC) inhibitors have been shown to reverse epigenetic repression of certain genes, including retinoic acid receptor β2 (RARβ2). In this study, we examined whether RARβ2 expression is repressed in human renal cell carcinoma (RCC) and whether the HDAC inhibitor MS-275 may revert its epigenetic repression.
Experimental Design: Six human tumor RCC cell lines were analyzed for RARβ2 gene expression and for methylation and acetylation status at the promoter level. Modulation of RARβ2 expression and correlation with antitumor activity by combination of MS-275 with 13-cis-retinoic acid (CRA) was assessed in a RARβ2-negative RCC cell line.
Results:RARβ2 expression was either strongly present, weakly expressed, or absent in the RCC cell lines analyzed. Methylation-specific PCR indicated that the RARβ2 promoter was partially methylated in three of the cell lines. CRA treatment did not inhibit clonogenic growth in the RARβ2-negative cell line RCC1.18, whereas MS-275 induced a dose-dependent inhibitory effect. A greater inhibitory effect was observed with combination treatment (MS-275 + CRA). Treatment with MS-275 was associated with histone acetylation at the promoter level and synergistic gene reexpression of RARβ2 in combination with CRA. RARβ2 reexpression was associated with synergistic induction of the retinoid-responsive gene HOXA5. In vivo, single-agent CRA treatment showed no significant effect, whereas MS-275 and the combination induced a regression of RCC1.18 tumor xenografts. Discontinuation of treatment produced tumor recurrence in MS-275-treated mice, whereas animals treated with the combination remained tumor free.
Conclusion: The HDAC inhibitor MS-275 seems to revert retinoid resistance due to epigenetic silencing of RARβ2 in a human RCC model and has greater antitumor activity in combination with CRA compared with single agents. Thus, the combination of HDAC inhibitors and retinoids may represent a novel therapeutic approach in patients with RCC.
It is estimated that renal cell carcinoma (RCC) has been diagnosed in >35,000 patients and has caused the death of >12,000 people in the United States during 2004 (1). Metastatic RCC is characterized by a high level of resistance to systemic treatment, including immunotherapy and chemotherapy. Thus, novel therapeutic approaches are needed to control this disease.
Retinol (vitamin A) and its active metabolites and derivatives, such as retinoids [i.e., 13-cis-retinoic acid (CRA) and all-trans-retinoic acid], have been shown to have some chemopreventive and therapeutic activity in cancer (2, 3). However, retinoid resistance represents a major hurdle in cancer treatment, including for RCC patients (3, 4). RA exerts its effects mainly via members of the nuclear receptor superfamily, the retinoic acid receptors (RAR) and the retinoid X receptors, which form heterodimers (5–7). The human RARβ gene is expressed as three different isoforms: β1, β2, and β4 (8). The biologically active RARβ2 isoform is under the regulation of the P2 promoter containing a high-affinity retinoic acid response element (RARE), which is associated with the transcriptional activation of RARβ2 by RA in a variety of cells (7).
Pharmacologic doses of RA induce cell differentiation and cell cycle arrest in some epithelial tumor cell lines but not in others (9). Retinoid resistance has been associated with loss or down-regulation of RARβ2 expression in breast, prostate, colon, lung, and kidney cancers (3). Hoffman et al. have reported that RARβ2 was not expressed by retinoid-resistant RCC cell lines but was present in a retinoid-sensitive RCC cell line and increased following incubation with RA (10). Accumulated evidence has shown RARβ2 to be the principal mediator of the differentiation and antiproliferative effects of retinoids in epithelial tumor cells (11, 12). Exogenous RARβ2 restores RA-induced inhibition in RARβ2-negative cells and RARβ antagonist or antisense mRNA block the effect of all-trans-retinoic acid. A possible cause of RARβ2 gene expression modulation has been associated with the aberrant methylation of CpG islands in the promoter region and histone deacetylation of associated chromatin (13, 14). The methyl CpG binding protein MeCP2 binds to the methylated CpG islands and induces recruitment of chromatin-associated factors, including Sin3a and histone deacetylase (HDAC) corepressor complexes (15, 16).
Nucleosomes, the repeating units of the human genome, consist of DNA wrapping around a histone octamer formed by one H3-H4 tetramer and two H2A-H2B dimers. The acetylation status of histones alters chromatin structure, which in turn is involved in gene expression. Two enzyme classes can affect the acetylation of histones—histone acetyltransferases and HDACs (17). HDACs are involved in oncogenic transformation by mediating the transcriptional regulation of genes that are involved in cell cycle progression, proliferation, and apoptosis. Thus, HDACs represent a rational target for therapeutic interventions. Several HDAC inhibitors have been characterized that inhibit tumor growth in vitro and in vivo and are in clinical trials (18). A series of synthetic benzamide derivatives with HDAC-inhibitory activity have been originally generated by Mitsui Pharmaceuticals. One of these, MS-275, has shown induction of chromatin hyperacetylation and antitumor activity by inhibition of HDAC enzyme activity (19). MS-275 has also shown inhibition of tumor cell growth in nude mice that was comparable or superior to conventional cytotoxic agents, such as 5-fluorouracil (19). This orally active synthetic benzamide is currently in phase I clinical trial. Our group and others have reported that treatment with HDAC inhibitors, including MS-275, may reverse epigenetic repression of RARβ2 in epithelial tumors, including prostate and breast (20–22).
In this study, the hypothesis tested was that retinoid resistance in RCC is associated with loss of RARβ2 expression due to an epigenetic mechanism, and treatment with the HDAC inhibitor MS-275 may revert RARβ2 silencing. Results show that MS-275 treatment reinduced RARβ2 expression in RARβ2-negative human RCC cell lines and restores retinoid sensitivity.
Materials and Methods
Cell lines and reagents. Six human RCC cell lines, RCC1.1, RCC1.4, RCC1.11, RCC1.18, RCC1.24, and RCC1.26, were kindly provided by Dr. Elisabeth Jaffee (Johns Hopkins University, Baltimore, MD). These cell lines were established from primary renal cell tumors. Repeated morphologic examination and immunohistochemistry assessment (CD10 and RCC positivity) by a cytopathologist was consistent with RCCs. The cells were cultured in RCC medium [containing RPMI 1640 (Life Technologies, Gaithersburg, MD), 10% tryptose phosphate broth (Sigma, St. Louis, MO), 1% l-glutamine (Life Technologies), 1% nonessential amino acids (Life Technologies), 1% sodium pyruvate (Sigma), and 1% penicillin/streptomycin (Life Technologies)] with 20% fetal bovine serum and kept in an incubator at 37°C in an atmosphere containing 5% CO2. For the in vitro experiments, tumor cells were treated with different concentrations of CRA (Sigma) or MS-275 (kindly provided by Schering AG, Berlin, Germany) or vehicle (DMSO) in RCC medium with 10% fetal bovine serum. For in vivo experiments, CRA and MS-275 were suspended in propylene glycol (Sigma) or 0.5% methocel (Fluka, Buchs SG, Switzerland), respectively, and given by gavage.
RNA isolation and RT-PCR. Total RNA was extracted from tumor cells or tumor tissues by TRIzol (Life Technologies), and the first strand was synthesized with oligo(dT) as primer using 1 μg total RNA according to the manufacturer's instructions. Two sets of primers for detecting RARβ2 were used to ensure reliable data. For the tumor cell line experiments, the product was 256 bp long and covered exons 3 and 4 (sense strand 5′-GACTGTATGGATGTTCGTTCAG-3′ and antisense strand 5′-ATTTGTCCTGGCAGACGAAGCA-3′). Samples were processed in a Perkin-Elmer (Norwalk, CT) 9600 GeneAmp thermocycling system under the following conditions: 2-minute denaturation step at 94°C followed by 35 amplification cycles (30 seconds at 94°C for denaturation, 30 seconds at 60°C for primer annealing, and 45 seconds at 72°C for primer extension) and final extension at 72°C for 10 minutes. For the tumor tissue experiments, the product was 1247 bp long and crossed from exons 3 to 9 (sense strand 5′-GTAGTAGGAAGTGAGCTGTTCA-3′ and antisense strand 5′-GCACTGATGCTACGGAGATCT-3′). The conditions for the long PCR product were 5-minute denaturation step at 94°C followed by 35 amplification cycles (45 seconds at 94°C for denaturation, 1 minute at 56°C for primer annealing, and 2 minutes at 72°C for primer extension) and final extension at 72°C for 10 minutes. Real-time PCR (RT-PCR) with primers encoding for β-actin (638 bp product, sense strand 5′-ATGATGATATCGCCGCGC-3′ and antisense strand 5′-CTCCTTAATGTCACGCACGATTTC-3′) was used as an internal RNA control.
Quantitative real-time RT-PCR analysis. Quantitative RT-PCR for specific genes was done to confirm the differences in genes identified by RT-PCR. Single-strand cDNA was synthesized from RCC total RNA (1 μg) by reverse transcription using oligo(dT) as the primer. According to the manufacturer's protocol, quantitative RT-PCR was done using an ABI PRISM 7700 sequence detector system (PE-Applied Biosystems, Foster City, CA) with a 2× SYBR Green PCR Master Mix (PE-Applied Biosystems), reverse-transcribed cDNA, and gene specific primers. To quantify the amount of target mRNA in the samples, a standard curve of RARβ2 was prepared for each run using the plasmid containing the target gene as well as a standard curve for glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as internal control. This enabled standardization of the initial mRNA content of cells relative to the amount of GAPDH. The sequences of the specific primers were as follows: RARβ2 primers are the same as in RT-PCR (256 bp), and the sense sequence was 5′-TGAACGGGAAGCTCACTGG-3′ and the antisense sequence was 5′-TCCACCACCCTGTTGCTGTA-3′ for GAPDH. Other primers were used in this experiment, including RARα (23), RARγ (24), and HOXA5 (25). The relative expression of target gene was determined by the difference of the threshold cycle (Ct) between target gene and GAPDH (relative expression = 2ΔCT, ΔCt = CtGAPDH − Cttarget gene; ABI User Bulletins).
DNA extraction and methylation-specific PCR. Genomic DNA was isolated from cell lines and primary tissues following the instruction of the DNeasy Tissue kit (Qiagen, Valencia, CA). DNA (∼1 μg) was modified by bisulfite treatment and subjected to methylation-specific PCR (MSP; ref. 26). The first MSP primers were designed from 80 to 284 bp (upstream 5′-TATGYGAGTTGTTTGAGGATTGGGA-3′ and downstream 5′-AATAATCATTTACCATTTTCCAAACTTA-3′). The next MSP primer sequences that specifically recognized methylated RARβ2 sequence (105-254 bp) were 5′-TGTCGAGAACGCGAGCGATTC-3′ (upstream or sense) and 5′-CGACCAATCCAACCGAAACGA-3′ (downstream) and the unmethylated RARβ2 sequence (100-261 bp) were 5′-TTGGGATGTTGAGAATGTGAGTGATTT-3′ (upstream) and 5′-CCTACTCAACCAATCCAACCAAAACAA-3′ (downstream or antisense).
Sodium bisulfite DNA sequencing. Modified DNA was amplified by PCR with primer 1 (5′-GTATAGAGGAATTTAAAGTGTGGGTTGGG-3′, upstream, nucleotides −415 to −386, Genbank accession no. X56849) and primer 2 (5′-CCTATAATTAATCCAAATAATCATTTACC-3′, downstream, sequence position from +269 to +298). The conditions were as follows: 5 minutes at 95°C and 6 minutes at 80°C followed by 37 cycles (20 seconds at 95°C, 45 seconds at 55°C, and 45 seconds at 72°C) and final extension for 5 minutes at 72°C. PCR products were cloned into the TA vector pCR2.1-TOPO and transformed into bacteria according to the manufacturer's instructions (Invitrogen, Carlsbad, CA). Plasmid DNA from isolated clones containing modified RARβ2 sequence was purified using Wizard Plus Minipres (Promega, Madison, WI) and subjected to automated DNA sequence analysis (ABI automated sequencing).
Chromatin immunoprecipitation assay. The histone acetylation status of the RARβ2 promoter was examined using the chromatin immunoprecipitation assay. An antibody specific for acetylated histone H4 was used to immunoprecipitate formaldehyde-cross-linked, sonicated chromatin from cells treated with MS-275 or the combination. Semiquantitative PCR analysis of DNA bound to immunocomplexes was done to detect a 192-bp fragment of the RARβ2 core promoter region (−165 to +27, sense strand primer 5′-CTCTGGCTGTCTGCTTTTGC-3′; antisense strand primer 5′-CAGCTCACTTCCTACTACTTC-3′), which included βRAREs and TATA sequences (27). Hs578t served as a positive control because its RARβ2 promoter is unmethylated and related histone H3 and H4 are acetylated (20).
Western blotting. The protein was obtained according to the kit manual from cell or tissues treated by different drugs. After electrophoresis, the proteins were blotted onto a nitrocellulose membrane. The membrane were blocked with blocking solution containing 5% nonfat milk for overnight and then incubated with primary antibody (anti-acetylated H3 1:2,000). Incubation with the secondary antibodies was done at room temperature for 1 hour. Strict washing (6 of 10-minute washing with PBS + Tween 20) was done after antibody incubation.
Colony formation assay. Exponentially growing tumor cells were seeded (200 cells/well) in six-well plates (Costar, Corning, NY) or 500 cells in 100 × 20 dish (Corning, Corning, NY) and allowed to attach for 48 hours. RCC1.11 cells were treated with CRA (1-10 μmol/L) or MS-275 (0.5 μmol/L) or the combination in complete medium containing DMSO (<0.1%). Cells were rinsed after 72 hours and fresh medium was added. Cultures were observed for 7 to 10 days and then fixed and stained with crystal violet. Colonies containing >30 cells were scored as survivors. Each condition was counted in triplicate (10 fields per well) on an inverted microscope. Results are expressed as mean ± SE number of colonies.
Tumor growth in vivo. Male 4- to 6-week-old severe combined immunodeficient or athymic nude mice (NCI) were kept in a temperature-controlled room on a 12:12-hour light/dark schedule with food and water ad libitum. Animals were injected s.c. in the flank region with 2 × 106 tumor cells (RCC1.18) resuspended in Hank's solution and mixed with Matrigel (1:1, Collaborative Biomedical Products, Bedford, MA) in a final volume of 0.2 mL. As the tumor volume reached a measurable size (50-100 mm3), 20 animals for each tumor were randomly placed in four groups (five animals per group): control, CRA, MS-275, and combination. Animals in the control group were treated with a daily administration (5 days/wk) of vehicle (polyethylene glycol) by gavage. CRA (30 mg/kg/d) and MS-275 (20 mg/kg/mL) were given by gavage. Tumor volume was measured with a caliper twice weekly and calculated according to the following formula: A (length) × B (width) × C (height) × 0.5236 and reported as mean ± SE. The animals were treated for ∼4 weeks and then sacrificed.
Statistical analysis. Differences between means of unpaired samples were evaluated by Student's t test using the Sigmaplot program. P < 0.05 was taken to indicate statistical significance.
Epigenetic repression of RARβ2 in human renal cell carcinoma cell lines. To determine the frequency of retinoid receptor expression and associated retinoid resistance in human RCC, six different cell lines were used that had been recently isolated from patients with RCC. Semiquantitative RT-PCR was done to detect RARβ2 gene expression. The results showed that two of six cell lines (RCC1.1 and RCC1.11) strongly expressed RARβ2. Two other cell lines (RCC1.24 and RCC1.26) faintly expressed RARβ2, whereas two others (RCC1.4 and RCC1.18) were negative for RARβ2 (Fig. 1A). These results were confirmed by real-time RT-PCR (data not shown). To determine whether lack of RARβ2 expression was due to aberrant methylation at the promoter level, the RARβ2 promoter was analyzed by MSP. The results showed that RCC1.1, RCC1.11, and RCC1.26 cell lines did not present a methylated band, the RCC1.24 cell line exhibited an unmethylated band and weakly positive methylated band, the RCC1.18 cell line presented both unmethylated and methylated bands, whereas RCC1.4 cell line showed only a methylated band (Fig. 1B). To better characterize the RARβ2 methylation status detected by MSP, bisulfite-modified DNA sequencing was done in the RCC1.11, RCC1.18, and RCC1.4 cell lines (Fig. 1C). The flank region of RARE and TATA box of the RARβ2 promoter and the first exon in RCC1.4 and RCC1.18 cell lines were methylated (35% and 28% CpG island methylation, respectively). Only one clone of RCC1.18 was found to have one methylated CpG in the RARE area. There were no methylated CpG islands seen in the RCC1.11 cell line. These results were consistent with the MSP results.
MS-275 induces RARβ2 reexpression and restores retinoid sensitivity in a RARβ2-negative cell line in vitro. Based on previous evidence of RARβ2 induction by HDAC inhibitors in other tumor cell lines, it was hypothesized that the epigenetically repressed RARβ2 could be restored by treatment with the HDAC inhibitor MS-275. Thus, the RARβ2-negative cell line RCC1.18 was treated with MS-275 (0.5 and 1.0 μmol/L) in the presence or absence of CRA (10 μmol/L) for 48 hours. RT-PCR analysis revealed a dose-dependent reinduction of RARβ2 by MS-275 in the presence of retinoid (Fig. 2A). There was no induction of RARβ2 in CRA-only-treated cells, and similar results were obtained with the RARβ2-negative cell line RCC1.4 (data not shown). To confirm that RARβ2 was repressed by methylation RCC1.18 cell line was treated with the demethylating agent 5-aza-2′-deoxycytidine (1 μmol/L). Ninety-six-hour but not 48-hour exposure with 5-aza-2′-deoxycytidine was able to reinduce RARβ2 expression (Fig. 2A). The next test was to determine whether reexpression of RARβ2 was due to the direct effect of MS-275 on histone acetylation; this was examined by using the chromatin immunoprecipitation assay. Analysis showed a high constitutive H4 deacetylation at the RARβ2 promoter in RCC1.18, and treatment with CRA alone did not increase histone acetylation status (Fig. 2B). However, MS-275 induced a significant increase of histone acetylation associated with the RARβ2 promoter after 24-hour treatment. Then, based on RT-PCR and MSP data, it was investigated whether treatment with MS-275 modulates retinoid response. To determine the antiproliferative effect of CRA and MS-275, these treatments were tested in a clonogenic assay. The RARβ2-positive cell line RCC1.11 and the RARβ2-negative cell line RCC1.18 were exposed to MS-275 (0.5-1.0 μmol/L) and CRA (10 μmol/L) for 72 hours (Fig. 2C). As expected, RCC1.11 growth was inhibited by CRA in a dose-dependent fashion (Fig. 2C). The lowest dose of CRA (0.2 μmol/L) inhibited 46% clones compared with control (P < 0.001). MS-275 treatment induced a 30% inhibition as single agent. However, combination of MS-275 with CRA had a greater inhibitory effect (up to >90% inhibition). In contrast, the RARβ2-negative cell line RCC1.18 was resistant to CRA treatment. MS-275 treatment induced 15.8% and 64.2% inhibition as single agent. However, MS-275 restored the sensitivity of RCC1.18 to CRA with a greater inhibitory effect on combination treatment.
MS-275 induction of RARβ2 reexpression is associated with increase of the retinoic acid–responsive gene HOXA5. To determine the specificity of the effect of combination treatment with MS-275 and CRA on RCC1.18 cell line, the gene expression of other RARs was assessed by real-time RT-PCR and compared with RARβ2. In contrast to RARβ2, RARα and RARγ were expressed in RCC1.18 cell line and were not significantly modulated by 24-hour treatment with MS-275 and/or CRA (Fig. 3A-C). The synergistic induction of RARβ2 gene expression by MS-275 plus CRA was evident after 12-hour treatment. The gene expression of HOXA5, another gene containing βRARE, was also assessed in RCC1.18 cell line. As reported in Fig. 3D, HOXA5 gene expression, which is present at baseline, was synergistically induced by the combination of CRA and MS-275 compared with single agents. The time course by quantitative PCR also showed that RARβ2 gene expression preceded HOXA5 gene expression.
MS-275 restores retinoid sensitivity in a RARβ2-negative renal cell carcinoma cell line in vivo. To determine the effect of MS-275 and CRA on RCC growth in vivo, RCC1.18 cells were injected s.c. in severe combined immunodeficient mice. Once the tumors were established, animals received either control vehicle, CRA (30 mg/kg/d), MS-275 (20 mg/kg/d), or the combination of both. The tumor growth curves showed that MS-275 and the combination of MS-275 and CRA had a significant inhibitory effect and induced RCC1.18 tumor regression (Fig. 4A). Following 3 weeks of treatment that was without overt toxicity, most xenografts in the MS-275-treated group and the combination-treated group were not detectable. CRA had no significant effect on RCC1.18 growth. After the discontinuation of treatment, all five animals in the MS-275 group eventually developed tumor recurrence, whereas those animals in the combination group remained tumor free for >2 months. A separate experiment done in nude mice had a similar response but with residual visible tumors (no CRA-induced tumor growth inhibition and 66% and 86% growth inhibition in the MS-275 group and in the combination group, respectively). Thus, mRNA was extracted from tumor tissue and analyzed for RARβ2 expression by RT-PCR (Fig. 4B). RARβ2 was not detected in the control and CRA-treated groups. However, some RARβ2 expression was present in the MS-275-treated group and was significantly induced in the combination-treated animals. The status of H3 acetylation in RCC1.18 tumor was determined by Western blot. MS-275 treatment significantly increased acetylated H3 compared with control and CRA group (Fig. 4C). DNA extraction from the tumor samples and MSP analysis revealed persistent methylation at the promoter region in tumors treated with MS-275 (Fig. 4D).
In this study, it is reported that tumor cell lines established from patients with RCC have different levels of RARβ2 gene expression. Loss of RARβ2 in RCC cell lines was associated with retinoid resistance and methylated CpG islands in the RARβ2 promoter and exon region and H4 hypoacetylation at the promoter level. The HDAC inhibitor MS-275 was shown to increase H4 acetylation of the chromatin associated with RARβ2 promoter and to induce reexpression of RARβ2 in RARβ2-negative RCC cell lines in the presence of CRA. In this preclinical model, MS-275 treatment alone had a significant in vivo activity, confirming the results in other tumor models (19). However, restoration of RARβ2 expression in a RCC cell line was associated with a greater inhibitory effect of the combination of MS-275 with CRA on tumor growth both in vitro and in vivo. Our data suggest an association between RARβ2 reexpression and antitumor activity but cannot rule out the possibility that other genes reactivated by CRA and MS-275 may contribute to the anticancer effects observed. Future studies with microarray analysis of MS-275- and CRA-treated RCC cells may provide useful information and implement our understanding of the biological mechanisms responsible for the antitumor activity observed with this drug combination.
Combination of a demethylating agent with a HDAC inhibitor is a rational “epigenetic” therapeutic strategy and is currently being tested in clinical trials. Aberrant DNA methylation is an important mechanism in gene regulation and epithelial tumorigenesis (28). Associated chromatin remodeling also plays a critical role in gene modulation and links between tumorigenesis and altered HDAC activity have been identified (18, 29). Hypermethylation has been reported in pediatric Wilms' tumors and adult RCC and has been associated with inactivation of several genes, such as VHL, RASSF1A, P16, CASP8, MGMT, NORE1A, and P14 (30, 31). Aberrant methylation at the RARβ2 promoter and consequent gene silencing has been reported in breast (13, 14), lung (32), prostate (33), esophagus (34), pancreas (35), colon (36), and stomach (37) tumors. There has been a recent report showing RARβ2 promoter methylation in RCC (38). Our data suggest that the methylation of CpG islands in the promoter and first exon lead to chromatin deacetylation and block the access of transcription factors to the start site of the RARβ2 gene in RCC. Our findings confirm that methylation is a critical step in tumor suppressor genes silencing and “lock in” function (39). However, it is speculated that chromatin remodeling and HDAC inhibition alone may overcome some degree of methylation-induced repression of certain ligand-inducible genes, such as RARβ2, by inducing sufficient acetylation to make the promoter susceptible to RA action without affecting the methylation status. Regardless of the class of HDAC inhibitor used, RARβ2 reexpression is observed, although there is persistence of DNA methylation (13). Interestingly, our results show that the tumor growth inhibition in severe combined immunodeficient mice is transient with the HDAC inhibitor alone but not in combination with CRA. These data suggest that inhibition of HDACs must be followed by active hyperacetylation induced by CRA via the histone acetyltransferase machinery to achieve the optimal biological results.
In our selection of RCC cell lines, we did not encounter a cell line with fully methylated RARβ2. Thus, our data do not rule out the possibility that increases in RARβ2 expression by the HDAC inhibitor may be attributable to enhancement of transcription from the unmethylated alleles. However, in a previous study, we showed that combination treatment with the HDAC inhibitor trichostatin A and all-trans-retinoic acid was able to restore RARβ2 gene expression in MDA-MB-231 breast carcinoma cell line where both alleles are methylated (20). It remains to be elucidated how the degree of promoter methylation may affect the capability of the HDAC inhibitors of overcoming the epigenetic repression and reinducing RARβ2 expression in tumor cells.
To determine the specificity of the effect of MS-275 and CRA on RARβ2 expression, we used real-time RT-PCR and assessed other RARs expression and their modulation by this combination. The result showed that, in contrast to RARβ2, RARα and RARγ are expressed in RCC1.18 cell line and are not significantly modulated by MS-275 and/or CRA (Fig. 3). Several potential RA target genes bear RAREs, including Hox genes (40). To determine whether a functional RARβ2 protein was induced by the combination treatment, we tested the gene expression of one of the Hox genes, HOXA5. According to TRANFAC 4.0 transcription element search system, we found three RARβ binding sequences in the promoter region, ∼2,000 bp upstream from HOXA5 start site (TGACCT, AGGTCA, and GAGGTCAGGG). HOXA5 gene expression, which is present at baseline in RCC1.18 cells, was synergistically induced by the combination of CRA and MS-275. The time course by quantitative PCR also showed that RARβ2 gene expression preceded HOXA5 gene expression, suggesting that a functional RARβ2 protein is expressed.
A proposed model suggests that an inactive RARβ2 promoter may undergo increased HDAC accumulation and associated chromatin acetylation during epithelial cell tumor development (20). The inactive promoter may become silenced due to additional epigenetic mechanisms, such as methylation by HDAC-induced MeCP2 recruitment. RARβ2 promoter inactivity may result from different mechanisms, including low intracellular levels of retinol and its metabolites. Key enzymes in retinoid metabolism and transport, such as lecitin/retinol acyl transferase and cellular retinol or RA-binding proteins, have been also reported to be reduced in epithelial tumors, including RCC (41, 42). Epigenetic mechanisms may be involved in the reduced expression of these enzymes (43). The methylation status of these enzymes in our RCC cell lines and whether HDAC inhibitor treatment may modulate their expression is currently being investigated.
Clinical trials involving CRA as single agents have shown no significant clinical activity in RCC patients. The combination of CRA with IFN-α has been reported to induce significant antitumor responses in preclinical models but to have only modest clinical activity. Interestingly, Berg et al. have shown that in a small group of renal cancer patients who underwent tumor biopsy before and after treatment, up-regulation of RARβ expression—and not baseline expression—correlated with response to CRA and IFN-α (44). These clinical data suggest that RARβ2 induction, rather than constitutive expression, may predict which tumor will respond to retinoid-based therapy. Reexpression of epigenetically silenced RARβ2 with consequent restoration of RARβ2 signaling pathways by concomitant exposure to HDAC inhibitor and pharmacologic doses of RA may be a predictor of response in patients with epithelial tumors, particularly renal cell cancer.
In summary, this report shows that retinoid sensitivity can be restored in retinoid-resistant RCCs via targeted therapy with RARβ2 agonists and chromatin remodeling drugs that produce epigenetic changes at RARβ2. A methylated RARβ2 promoter and an inducible RARβ2 may represent a rational predictor for tumor response in patients undergoing “differentiation” therapy with the combination of a HDAC inhibitor and a retinoid. Based also on these preclinical results, a Cancer Therapy Evaluation Program-National Cancer Institute–sponsored phase I clinical study of MS-275 in combination with CRA in metastatic progressive cancer is currently accruing patients at our institution.
Grant support: Maryland Cigarette Restitution Fund Research Grant and Flight Attendant Medical Research Institute (R. Pili).
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
We thank Dr. Nicoletta Sacchi for critically reviewing the article, Drs. Michael Carducci and John Isaacs for helpful discussion, and Drs. Xinyan Wu, Sushant Kachhap, and Hexin Chen for technical support.