| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cell Growth & Differentiation |
2 Is a Growth Suppressor Epigenetically Silenced In MCF-7 Human Breast Cancer Cells1
Departments of Medicine [E. F. F., A. A., S. W., R. M. L.] and Pathology (I. J. B.), Mount Sinai School of Medicine, New York, New York 10029-6574, and Leukaemia Research Fund Centre, Institute of Cancer Research, London SW3 6JB, United Kingdom [A. Z.]
 |
Abstract |
|---|
 TOP Abstract IntroductionResultsDiscussionMaterials and MethodsReferences |
|---|
2 is also underexpressed in MCF-7 breast cancer cells, at both the message and the protein level, relative to normal or nontumorigenic breast epithelial cells. Bisulfite sequencing of the CpG island in the RAR2 promoter revealed highly penetrant and uniform cytosine methylation in MCF-7 cells. Pretreatment with the DNA methyltransferase inhibitor, azacytidine, followed by treatment with RA and a histone deacetylase inhibitor, trichostatin A, resulted in partial promoter demethylation and RAR2 induction, which strongly suggested that promoter hypermethylation is responsible for RAR2 underexpression. We compared the outcome of ectopic expression in MCF-7 cells of matched levels of RAR2 and RARß2. On the basis of a clonogenic assay, RAR2 displayed ligand-dependent growth-suppressive activity similar to that of RARß2; thus, 10 and 20 nM RA inhibited clonogenic growth by 52 and 80%, respectively, in RAR2-transfected cells compared with 75 and 77%, respectively, in RARß2-transfected cells. We conclude that the silencing of the RAR2 promoter by hypermethylation may play a contributory role in the dysregulation of RA signaling in mammary tumorigenesis. |
Introduction |
|---|
|
TOPAbstractIntroductionResultsDiscussionMaterials and MethodsReferences |
|---|
, ß, 
) and any of the three known RXR genes (, ß, ). Biological processes dependent on the activity of RXR-RAR heterodimers include mammalian development, as demonstrated by extensive gene targeting studies in the mouse (reviewed in Ref. 1
), and cancer, as demonstrated by the role of RAR in the pathogenesis and therapy of acute promyelocytic leukemia (reviewed in Ref. 2
) and as suggested by the chemopreventive activity of retinoids toward solid tumors (reviewed in Ref. 3
). Particularly interesting in connection to the possible role of RARs in regulating epithelial carcinogenesis is the recent observation that RARß regulates branching morphogenesis in the lung (4)
.
Another line of investigation suggestive of a role of RARs in the regulation of epithelial carcinogenesis is the down-regulation of RAR gene expression in cancer. In particular, the expression of RARß2, which is normally induced by RA through the DR5 response element in its promoter (5)
, has been shown to be down-regulated in a wide range of cancer cell types and tissues (6, 7, 8, 9, 10)
. Several recent reports have argued that the down-regulation of RARß2 in cancer is secondary to hypermethylation of the gene (11, 12, 13, 14)
. The finding that cancer cells that display no detectable RARß message usually retain RAR and RAR expression underlies the interest in the loss of RARß expression in cancer. However, the growth-suppressive activity of RARß2 (15, 16, 17)
is not unique, because other RARs, such as RAR1, are also growth suppressive (18
, 19)
. Furthermore, several authors have demonstrated that the selective activation of RAR is growth inhibitory (20, 21, 22)
.
RAR2 expression, like RARß2, is induced by RA through the DR5 element in the RAR2 promoter, which is remarkably conserved with that of RARß2 (23)
. We showed earlier that RAR2 expression is down-regulated in several human breast cancer cell lines relative to normal breast fibroblasts (24)
. However, the possible significance of this observation has been masked by the fact that most breast cancer cells maintain RAR1 expression and by our use of normal mesenchymal cells as reference.
We have now analyzed RAR1, RAR2, and RARß2 expression in MCF-7 cells relative to immortalized but nontumorigenic MTSV1–7 human breast ductal epithelial cells. Furthermore, we studied the growth effects of ectopic RAR2 as compared with RARß2 and analyzed the methylation status of the endogenous RAR2 promoter. In aggregate, our results suggest that RAR2 down-regulation contributes to the impaired RA responsiveness of cancer cells.
|
Results |
|---|
|
TOPAbstractIntroductionResultsDiscussionMaterials and MethodsReferences |
|---|
2 expression in tumorigenic MCF-7 and nontumorigenic MTSV1–7 breast epithelial cells with the expression of RAR1 and RARß2 in the same cell pair. MTSV1–7 cells expressed both RAR2 and RARß2 under basal conditions, and expression levels increased on treatment with 1 µM RA, as expected (Fig. 1A)
. Similar results were obtained for HMEC cultures generated from reduction mammoplasty tissue and enriched in luminal cells (Fig. 1B)
. In contrast, neither RAR2 nor RARß2 expression could be detected in MCF-7 cells even after treatment with 1 µM RA (Fig. 1A)
. Thus, RAR2 mirrors the expression pattern of RARß2. On the other hand, both MTSV1–7 and MCF-7 cells expressed RAR1, but expression levels were clearly higher in MCF-7 cells (HMECs expressed RAR1 at levels comparable with those of MTSV1–7 cells, data not shown). This is because MTSV1–7 cells, like most of the normal breast epithelium but unlike MCF-7 cells, do not express ER (25)
, which is known to up-regulate RAR1 transcription (26)
. These results suggest that the total RAR mRNA concentration (1 + 2) in MCF-7 cells is comparable with that of MTSV1–7 cells. However, when considering only RAR2 expression, the pattern is analogous to the loss of RARß2 expression in MCF-7 breast cancer cells.
|
and RARß protein expression in MTSV1–7 and MCF-7 cells. Antibodies against the F domain of RAR (recognizing both RAR1 and 2) or RARß (likewise recognizing multiple subtypes) were used. The slightly shorter RAR2 protein migrates sufficiently ahead of RAR1 to allow unambiguous identification of each receptor subtype (27)
. Interestingly, despite the RT-PCR differences described above, RAR1 protein levels were essentially the same in MTSV1–7 and MCF-7 cells (Fig. 1C)
. In sharp contrast, RAR2 was expressed in RA-treated MTSV1–7 but not in RA-treated MCF-7 cells, and its expression level was comparable with that of RAR1. These results suggest that the total RAR protein concentration (1 + 2) in MCF-7 cells is 
50% lower than in nontumorigenic cells. Like RAR2, RARß was detected only in MTSV1–7 cells; its expression was evident under basal conditions but increased after RA treatment (Fig. 1C)
. On the basis of the lack of RARß1 expression in the breast and the good agreement between relative RARß2 mRNA (Fig. 1A)
and RARß protein expression levels (Fig. 1C)
, we suspect that the band detected by the anti-RARß antibody corresponds to RARß2.
In sum, nontumorigenic MTSV1–7 cells express both RAR2 and RARß2 mRNA and RAR2 and RARß protein, whereas MCF-7 tumor cells fail to express these inducible receptors.
Mechanism Underlying RAR2 Silencing in MCF-7 Cells.
The RAR2 promoter and 5'-UTR region are highly CpG rich. This is actually in contrast to the same region of the RARß2 gene, which is CpG poor. Nevertheless, there is good evidence that hypermethylation of the RARß2 promoter and 5'-UTR is the proximal cause of RARß2 underexpression in MCF-7 cells (11)
. We, therefore, analyzed the methylation status of the RAR2 promoter by bisulfite sequencing. The results (Fig. 2A)
demonstrated uniform and highly penetrant methylation of the region studied. This includes three CpG sites near the DR5 response element (highlighted in Fig. 2A
), which are conserved and also methylated in the RARß2 promoter in MCF-7 cells. We have proposed that methylation of these sites may be particularly relevant to promoter silencing (28)
.
|
2 silencing, we pretreated exponentially growing MCF-7 cells with azaC, which is incorporated into DNA and inhibits DNA methyltransferase. Three days later, we treated the cells with a combination of RA and TSA, a histone deacetylase inhibitor; control cultures were pretreated with growth medium alone and then treated as above. As shown in Fig. 2B
, MCF-7 cells, pretreated with azaC and then treated with RA and TSA, expressed readily detectable amounts of RAR2 message, whereas the same treatment had no effect in cells not pretreated with azaC. Bisulfite sequencing of DNA isolated from cultures pretreated with azaC and then treated with RA and TSA revealed a modest and variable degree of RAR2 promoter demethylation (Fig. 2A)
. We suspect that more extensive demethylation occurred in a small fraction of cells and that RAR2 reexpression in these cells accounted for the RT-PCR signal detected in Fig. 2B
. Alternatively, a modest degree of demethylation may be sufficient to initiate chromatin remodeling. These results implicate RAR2 hypermethylation as a proximal cause of RAR2 silencing in MCF-7 cells.
RAR2 Is a Growth Suppressor.
We wished to compare the effect on cell growth of the ectopic expression of matched levels of RAR2 and RARß2. To this end, we constructed a bicistronic mammalian expression vector encoding RAR2 or RARß2 and EGFP (pRAR2-IRES-2-EGFP and RARß2-IRES-2-EGFP, respectively). Our goal was to use EGFP to infer RAR expression levels rather than rely on the use of antibodies with different titer and affinity. Preliminary experiments demonstrated that transfection of MCF-7 cells with a fixed amount of vector (2 µg) resulted in the same overall expression level of EGFP and, thus, by inference, of RAR2 and -ß2 (Fig. 3A)
. For the experiment proper, we transfected MCF-7 cells with the same fixed dose of each vector and, as a control, 2 µg of an "empty" (no RAR insert) bicistronic vector (pIRES-2-EGFP). Fig. 3B
shows the pattern of EGFP fluorescence of parallel cultures transfected as above and selected in G418 in the absence of RA; comparable fluorescent signals were obtained for each cell group. The experimental cultures were placed in selection medium with or without RA supplementation and colonies allowed to grow for a period of 3 weeks. All of the groups of transfected cells formed multiple colonies in selection medium without RA. Treatment with 10 and 20 nM RA had a minimal effect on the clonogenic potential of control, pIRES-2-EGFP-transfected cells (Fig. 4A)
. In contrast, the same treatment had a marked inhibitory effect on the clonogenic potential of pRAR2-IRES-2-EGFP- and RARß2-IRES-2-EGFP-transfected cells. More importantly, the magnitude of this effect, as can be seen visually (Fig. 4A)
and after colorimetry (Fig. 4B)
, was comparable for RAR2 and -ß2. We conclude that RAR2, like RARß2, is a growth suppressor. This conclusion is based on the interpretation that the results obtained were attributable to the transfected RAR2 and -ß2 receptors per se and not to secondary effects on other nuclear receptors.
|
|
|
Discussion |
|---|
|
TOPAbstractIntroductionResultsDiscussionMaterials and MethodsReferences |
|---|
2, can function as an RA-dependent growth suppressor. Our results suggest that RAR2, besides being RA-inducible, shares three properties with RARß2: it is underexpressed in MCF-7 cells relative to normal and nontumorigenic cells, it is epigenetically silenced, and its product acts as a growth suppressor. We, therefore, suggest that RAR2 should be considered as a factor of equal importance to RARß2 in studies of aberrant RAR expression and function in breast cancer and perhaps other cancers as well.
Although homozygous deletion of RARß2 has been shown to impair RA-induced growth arrest in F9 cells (31)
, and mice harboring an RARß2 antisense transgene have been shown to develop lung tumors (32)
, similar studies with RAR2 have not been reported. On the other hand, several studies have shown that ectopic RAR1 is growth suppressive (18
, 19)
and that selective RAR ligands are growth inhibitory (20, 21, 22)
. Interestingly, it has been proposed that the ability of RAR1 to restore RA growth regulation in ER-nonexpressing breast cancer cells is secondary to the induction of RARß2 (17)
. However, induction of RAR2 likely occurred under the same conditions because they have conserved promoters (23)
. In this context, it is interesting to point out that the recent demonstration that constitutively expressed RAR protein (i.e., RAR1) is rapidly down-regulated after RA activation (33
, 34)
suggests that the induction of RAR2 and RARß2 may be critical for sustaining a RA response. Undoubtedly, the best way to dissect the relative contribution of each isoform would be to extend the targeted gene deletion approach undertaken in F9 cells (31)
to a diploid breast cell line that is sensitive to RA-induced growth arrest.
An in situ hybridization study by Xu et al. (8)
demonstrated that RAR (total) mRNA is undetectable in 16% of invasive human breast cancers as compared with 2% of normal breast tissues. Taking into account that RAR1 mRNA levels in nontumorigenic ER-nonexpressing cells (MTSV1–7) approximate those in ER-nonexpressing breast cancer cells (i.e., they are detectable but lower than the levels expressed by ER-expressing cells; Fig. 1A
; Ref. 21
), we hypothesize that the RAR down-regulation detected by Xu et al. represents down-regulation of RAR2 (the true extent of RAR2 down-regulation may in fact be greater). An in situ hybridization study using RAR1- and RAR2-specific riboprobes is needed to test this hypothesis and reveal the extent, if any, of in situ RAR2 down-regulation in breast cancer.
The evidence implicating aberrant RAR expression in cancer remains largely circumstantial, and this report is no exception. Further progress might be achieved, for instance, through a study of RAR methylation and expression in existing mouse models of breast cancer, which afford the opportunity to sample tissues at linear time points. RAR isoform and subtype-specific knockout mice represent another opportunity for progress.
|
Materials and Methods |
|---|
|
TOPAbstractIntroductionResultsDiscussionMaterials and MethodsReferences |
|---|
Normal HMECs were derived from reduction mammoplasty material that was residual to pathologic analysis and bore no identifiers. After obtaining epithelial organoids as described previously (36) , enrichment in luminal cells was performed by negative immuno-selection (37) and the cells cultured in collagen I-coated dishes in mammary epithelium growth medium (Clonetics, San Diego, CA) supplemented with 5 µg/ml transferrin and 10 µM isoproterenol (cells were used after 0–1 passages).
RT-PCR Analysis.
Aliquots of 1 µg of total RNA were reverse transcribed in 20-µl reaction volumes using the SuperScript Preamplification System (Life Technologies, Inc. Gaithersburg, MD). The PCR mixture consisted of 10x PCR buffer [200 mM Tris-HCl (pH 8.4), 500 mM KCl], 1 mM MgCl2, 200 µM dNTPs, 0.4 µM each primer, 2 µl cDNA, and 5 units of Taq (Promega) in a total volume of 50 µl. Primers for the amplification of RAR1 and -2 were as described previously (38)
, and cycling parameters were: 94°C for 5 min; 30 cycles of 94°C for 1 min, 60°C for 1 min, 72°C for 1 min; and 72°C for 5 min. RAR1 and -2 were amplified, separately or in the same reaction, using sense primers that anneal to the A1 or A2 region, respectively, and an antisense primer that anneals to the common B region, yielding products of 220 bp and 287 bp, respectively. PCR products (25 µl aliquots) were separated in 1.5% agarose gels, blotted, and hybridized to probes labeled with [32P]dCTP by random priming. The RAR1 amplification product was detected using an autologous probe (this was validated by probing with full-length RAR1 cDNA) and the RAR2 product was detected using full-length RAR2 cDNA. When RAR1 and -2 were amplified in the same reaction, the two products were detected by hybridization to full-length RAR2 cDNA and could be unambiguously identified based on their different sizes (separate amplification reactions validated the results obtained after coamplification). RARß2 and GAPDH RT-PCR were performed as described previously (28)
.
Western Blot Analysis.
Cells were lysed on ice for 30 min with radioimmunoprecipitation assay buffer (1x PBS, 1% NP40, 0.5% sodium deoxycholate, and 0.1% SDS) plus proteinase inhibitors (1 mM phenylmethylsulfonyl fluoride, 1 µg/ml aprotinin, 1 µg/ml leupeptin, and 1 µg/ml pepstatin) and phosphatase inhibitors (1 mM Na3VO4 and 1 mM NaF); for more efficient cell lysis, the samples were frozen (-80°C) and thawed once. Protein was quantitated by the method of Bradford (39)
and 50 µg of total protein/lane were resolved by SDS-PAGE (10% polyacrylamide). Rabbit polyclonal antibodies raised against the F region of RAR (number 115) was a generous gift from Pierre Chambon (Institut de Genetique et de Biologie Moleculaire et Cellulaire, Strasbourg, France); rabbit polyclonal antibodies raised against the F region of RARß (C-19) were obtained from Santa Cruz Biotechnology. Polyvinylidene difluoride membranes were blocked with TBST plus 5% milk for 1 h and incubated with first antibody diluted to 1 µg/ml in TBS plus 1% milk for 1 h at room temperature. The membranes were washed five times with TBST and once with TBS and treated with secondary antirabbit antibody diluted to 0.3 µg/ml in TBS plus 1% milk for 45 min at room temperature. The membranes were then washed as above and reaction products visualized using Amersham’s chemiluminiscence detection system.
For EGFP expression, total cell lysates were resolved by SDS-PAGE and probed with EGFP antibodies obtained from Clontech (Palo Alto, CA).
Bisulfite Sequencing.
Briefly, 5 µg of XbaI-digested genomic DNA was denatured with 0.5 M NaOH at 75°C for 15 min, deaminated with freshly prepared 4.0 M (pH 5.0) sodium metabisulfite (BDH Laboratory Supplies, Poole, Dorset, England) and incubated at 55°C for 8 h in a total volume of 1.2 ml under mineral oil. Desalting, desulfonation by alkali, neutralization, and ethanol precipitation were as described previously (40)
; the DNA was resuspended in 50 µl of H2O. First-round PCR was carried out in a 50-µl reaction mixture of 10x PCR buffer, 1.2 mM MgCl2, 160 µM dNTPs, 0.4 µM each primer, 10 µl bisulfite-treated DNA, and 5 units of Taq (Promega). Subsequently, two rounds of seminested PCR were carried out using the same conditions as above and 5 µl of the reaction product generated in the preceding round as template. Primers and parameters were: 5'-TTAGGGATTTATTTAAGTTAGGTTTTTT-3' (sense) and 5'-AACTCCCCCAAAATTTAATCAATCCCTAAC-3' (antisense); 94°C for 5 min; 40 cycles of 94°C for 1 min, 48°C for 1 min, 72°C for 1 min; 72°C for 5 min. Primers and parameters for the second round were: 5'-TTTTTGTTTGTGTTTGTGTTAATAGTATT-3' (sense) and the same antisense as above with parameters as above, except annealing temperature was 46°C, and the reaction was terminated after 20 cycles. Third round primers and parameters were: 5'-TATATATATGTTATAGATAATGATAT-3' (sense) and the same antisense as above; parameters were the same as for second round parameters but for 40 cycles. The purified PCR fragment was subcloned into pCR2.1, and individual clones were submitted for sequencing (T7 primer).
Cell Transfection.
Monolayers of MCF-7 cells at 70–80% confluence were transfected (FuGene system, Roche) with 2 µg of pIRES-2-EGFP vector alone or carrying full-length RAR2 or RARß2 cDNA inserts. Two days later, the cells were trypsinized and plated at the desired concentration in selection medium (800 µg/ml Geneticin). Mock-transfected MCF-7 cells were used to follow cell selection and transfection efficiency was monitored by fluorescence microscopy.
RA Treatment.
For testing the effect of 1 µM RA on RAR expression, cells in regular growth medium were treated with RA for 24 h and harvested for analysis. For testing the effect of low RA doses on MCF-7 clonogenicity, FBS was charcoal stripped as follows: 1.25 g of activated charcoal and 0.125 g of dextran T-70 were added to a 500-ml FBS bottle followed by incubation at 55°C for 30 min and centrifugation at 3000 rpm at 4°C for 20 min. The resulting single-stripped serum was collected into a fresh 500-ml flask and the procedure repeated with a second round of charcoal-dextran, but incubated at 37°C for 30 min and, finally, sterile-filtered. MCF-7 breast cancer cells transfected as above were seeded in 6-well (5 x 104 cells/well) or 12-well (2 x 104 cells/well) plate dishes in medium supplemented with 10% charcoal-stripped FBS, 5 µg/ml insulin, 800 µg/ml Geneticin, and antibiotics. After 3 weeks, Geneticin-resistant colonies were fixed in 10% formalin in PBS and stained with 0.05% crystal violet. After photography, the stain was solubilized with 10% acetic acid-10% methanol, and the A595 was measured to quantitate adherent cell mass.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
1 Supported by NIH Grant CA54273 (to RML), a grant from the Leukemia Research Fund of Great Britain (to AZ), The Samuel Waxman Cancer Research Foundation and The Norman and Rosita Winston Foundation. 
2 To whom requests for reprints should be addressed, at Department of Medicine, Mount Sinai School of Medicine, One Gustave L. Levy Place, New York, NY 10029-6574.
3 The abbreviations used are: RA, retinoic acid; RAR, all-trans-RA receptor; RXR, 9-cis-RA receptor; DR5, direct repeat separated by five basepairs; HMEC, human mammary epithelial cell; ER, estrogen receptor ; RT-PCR, reverse transcription-PCR; azaC, 5-azacytidine; TSA, trichostatin A; EGFP, enhanced green fluorescent protein; FBS, fetal bovine serum; TBS, Tris-buffered saline; TBST, TBS with Tween 20; GAPDH, glyceraldehyde-3-phosphate dehydrogenase.
Received for publication 1/ 7/02. Revision received 4/26/02. Accepted for publication 6/17/02.
|
References |
|---|
|
TOPAbstractIntroductionResultsDiscussionMaterials and MethodsReferences |
|---|
acquire sensitivity to growth inhibition by retinoids. J. Biol. Chem., 269: 21440-7, 1994.1 isoform is induced by estradiol and confers retinoic acid sensitivity in human breast cancer cells. Mol. Cell. Endocrinol., 109: 77-86, 1995.[Medline]
with retinoid inhibition of growth of estrogen receptor-positive MCF-7 mammary carcinoma cells. Cancer Res., 55: 4446-4451, 1995. expression correlates with retinoid-induced growth inhibition of human breast cancer cells regardless of estrogen receptor status. Cancer Res., 57: 2642-2650, 1997. is sufficient for full induction of retinoid responses in SK-BR-3 and T47D human breast cancer cells. Cancer Res., 60: 5479-5487, 2000. are generated by alternative splicing and differential induction by retinoic acid. EMBO J., 10: 59-69, 1991.[Medline]
2, ß2, and 2 in human breast cancer cells. FASEB J., 10: 1064-1070, 1996.[Abstract]
evidence for phosphorylation. Exp. Cell Res., 201: 335-346, 1992.[Medline]
(RAR) and oncogenic RAR fusion proteins. Proc. Natl. Acad. Sci. USA, 96: 14807-12, 1999. and through proteasomes in MCF-7 human breast cancer cells. Biochem. Pharmacol., 61: 1347-1355, 2001.[Medline]
/RAR2-PML mRNA expression in acute promyelocytic leukemia cells: a molecular and laboratory-clinical correlative study. Blood, 90: 306-312, 1997.This article has been cited by other articles:
 |
|
 |
|
  J. Srivastava, C. L. Robertson, D. Rajasekaran, R. Gredler, A. Siddiq, L. Emdad, N. D. Mukhopadhyay, S. Ghosh, P. B. Hylemon, G. Gil, et al. AEG-1 Regulates Retinoid X Receptor and Inhibits Retinoid Signaling Cancer Res., August 15, 2014; 74(16): 4364 - 4377. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. P. Mongan and L. J. Gudas Valproic acid, in combination with all-trans retinoic acid and 5-aza-2'-deoxycytidine, restores expression of silenced RAR{beta}2 in breast cancer cells Mol. Cancer Ther., March 1, 2005; 4(3): 477 - 486. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Bourachot, M. Yaniv, and C. Muchardt Growth inhibition by the mammalian SWI-SNF subunit Brm is regulated by acetylation EMBO J., December 15, 2003; 22(24): 6505 - 6515. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Gumireddy, L. N. Sutton, P. C. Phillips, and C. D. Reddy All-trans-Retinoic Acid-induced Apoptosis in Human Medulloblastoma: Activation of Caspase-3/Poly(ADP-ribose) Polymerase 1 Pathway Clin. Cancer Res., September 15, 2003; 9(11): 4052 - 4059. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cell Growth & Differentiation |
, methylated CpGs; 
, unmethylated CpGs). Top panel, results for untreated cells; bottom panel, results for cells pretreated with azaC and treated with RA and TSA (see below); cells not pretreated with azaC and treated with RA and TSA yielded results similar to those for untreated cells (data not shown). B, semiquantitative RT-PCR analysis of RAR
