![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
CHEMOTHERAPY, ANTIBIOTICS, AND GENE THERAPY
Bone Marrow Transplant Program, Arizona Cancer Center (M.N., N.H., M.M.O., R.B.I., J.L.M.-R., A.F.L., B.W.F.), Department of Pharmacology & Toxicology (B.W.F.), and ARL Biotechnology Computing Facility (M.L.N., S.J.M., N.C.M.), University of Arizona, Tucson, Arizona
Received June 9, 2004; accepted August 6, 2004.
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
translocation in acute promyelocytic leukemia (APL) is an example where the resultant fusion protein recruits histone deacetylase complexes to target genes resulting in their inappropriate transcriptional repression. All-trans-retinoic acid (ATRA) acts as a ligand that relieves this repression and produces an epigenetic transcriptional reprogramming of the cancer cell. CpG island microarrays were used to analyze the DNA methylation and histone acetylation state of the human APL cell line NB4 before and after differentiation with ATRA as well as normal peripheral blood mononuclear cells (PBMC). Over 70 CpG islands within 1 kb of transcription start of a known gene are aberrantly methylated in NB4 cells compared with PBMC; however, no changes in cytosine methylation were detected following ATRA-induced differentiation. With respect to histone H4 acetylation, over 100 single-copy CpG islands within 1 kb of transcription start of a known human gene became hyperacetylated following ATRA-induced differentiation. One CpG island was aberrantly methylated in NB4 cells, but became hyperacetylated and was induced following ATRA treatment and was associated with the HoxA1 gene, suggesting it may be a target gene of ATRA in APL. In addition to single-copy sequences, a selective increase in acetylation was detected in satellite DNA when compared with other high-copy sequences, such as Alu or rDNA. In summary, ATRA stimulates complex epigenomic changes during leukemic cell differentiation, and monitoring these changes may help to identify new targets of epigenetic dysfunction.
; Alcalay et al., 2001; Chakraborty et al., 2001). One example of these chimeras is the PML/RAR fusion protein seen in acute promyelocytic leukemia (APL), which is a result of a reciprocal translocation t(15;17)(q22; q21) (Larson et al., 1984; Borrow et al., 1990; de The et al., 1990). In the absence of the retinoid ligand, PML/RAR represses transcription of target genes through the recruitment of histone deacetylase complexes to the target gene's regulatory region. In the presence of retinoids, this transcriptional repression is relieved (Grignani et al., 1998; Lin et al., 1998).
In APL, the PML/RAR protein blocks terminal differentiation in the promyelocyte stage and supports continuous proliferation (Wang et al., 1998). All-trans-retinoic acid (ATRA) is a clinically useful retinoid that induces a transcriptional reprogramming of APL cells that result in inhibition of cell proliferation and induction of a mature myeloid phenotype (Flynn et al., 1983). These gross cellular changes are associated with the activation of a host of genes, a notable target gene being retinoic acid receptor 
(RAR ) (Ruthardt et al., 1997; Di Croce et al., 2002). A model cell line of APL, NB4, has been used extensively to study the differentiation process of leukemic cells in vitro (Lanotte et al., 1991). Because ATRA produces a ligand-induced activation of gene expression with concomitant changes in chromatin structure, and subsequent terminal differentiation in NB4 cells, we set out to assess the genome-wide changes in histone acetylation and DNA methylation induced by this retinoid.
Others have successfully used genome-wide scanning approaches to analyze tumor cells and their epigenomic characteristics (Costello et al., 2000; Yan et al., 2000; Zardo et al., 2002; Shi et al., 2003). Based on these successes, we used 6800 element CpG island microarrays to analyze the epigenomic state of NB4 cells, NB4 cells differentiated with ATRA treatment, and normal peripheral blood mononuclear cells (PBMC). CpG island microarray hybridization results show that the untreated NB4 cells have an overall increase in methylation of their CpG islands compared with PBMC. The CpG island arrays identified targets already known to be inappropriately methylated in NB4 cells specifically (e.g., RAR ), as well as identifying potential new CpG island targets of epigenetic dysregulation in cancer. In addition, the CpG island microarray results indicate that ATRA-induced differentiation had no detectable effect on the cytosine methylation state of the CpG islands analyzed.
To assess changes in histone acetylation state, DNA isolated from chromatin immunoprecipitations from the various cells were used to probe the CpG island microarrays. Although ATRA did not induce detectable changes in genome-wide methylation levels, it did induce complex changes in histone acetylation throughout the genome of ATRA-treated NB4 cells. First, approximately 282 single-copy CpG islands displayed increased levels of histone acetylation including the known target, RAR , whereas only 34 clones displayed a decrease in acetylation. Second, increased levels of histone H4 acetylation were seen in the high-copy satellite sequences, suggesting that high-copy elements localized to centromeric regions become preferentially acetylated during leukemic cell differentiation. Third, histone acetylation changes observed occurred independently of changes in 5-methylcytosine.
In summary, CpG island microarray analysis of NB4 cells, NB4 cells terminally differentiated with ATRA, and normal PBMC have 1) revealed new potential targets of aberrant methylation in APL, 2) shown that ATRA-mediated differentiation of NB4 cell targets increased histone H4 acetylation in not only single-copy sequences, but also pericentromeric satellite sequences, and 3) shown that ATRA does not induce detectable changes in genomic methylation.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Nucleic Acid Isolation. Total RNA was isolated from cells using an RNeasy mini or midi kit (QIAGEN, Valencia, CA), and genomic DNA was isolated using the QIAamp DNA mini kit (QIAGEN). RNA and DNA samples were quantitated by UV absorbance measurements at 260 nm.
CpG Island Library, Probe Preparation, and Database. An aliquot of a human CpG island library, originally constructed by Cross et al. (1994), was purchased from the UK HGMP (http://www.hgmp.mrc.ac.uk/). Library aliquots were grown in LB media plus ampicillin and plated on LB agar plates plus ampicillin and 5-bromo-4-chloro-3-indolyl--D-galactopyranoside. Approximately 6800 individual white colonies were arrayed into 96-well plates containing 100 µl of LB plus ampicillin. Colonies were grown in culture overnight at 37°C, followed by the addition of glycerol to a final concentration of 20%, replica plated, and then stored at –80°C for further use.
To generate CpG island probes for microarray production, 1 µl of bacterial culture was added to 50 µl of PCR master mix plus PCR primers in a 96-well format (Marsh Bio Products, Rochester, NY). After PCR, an aliquot of products were analyzed by gel electrophoresis, followed by "PCR clean up" of the products using QIAGEN kits (QIAGEN, Chatsworth, CA). After clean up, PCR products were analyzed and quantitated with Agilent Technologies' DNA LabChips (Palo Alto, CA). Typical yields ranged from 1 to 5 µg. After quantitation, the purified PCR products were dried and resuspended in 10 µl of 3x SSC for printing onto slides.
CpG island clones were also DNA sequence validated. CpG island clones were grown overnight at 37°C in 2 ml of LB broth, and plasmid DNA was purified using the Nucleospin robot-96 plasmid kit (Machery-Nagel, Duren, Germany). Plasmid DNA was sequenced on an automated ABI DNA sequencer, and resultant sequences were BLASTed against the complete human genome provided by the University of California Santa Cruz genome browser (November 2002). These CpG island sequences and the associated genome annotation, including but not limited to GC content, observed to expected CpG ratios, and association with a transcribed unit were used to populate a web-accessible MySQL database driven for a customized web accessible portal for further examination and lead identification.
CpG Island Microarray Production. CGI DNAs were printed onto chemically activated glass slides (SuperAmine substrates; TeleChem International, Sunnyvale, CA) under 35 to 40% humidity, using four quill-type pins (TeleChem International, San Jose, CA) mounted onto an OmniGrid robot (GeneMachines, San Carlos, CA). In addition to the CGI clones, Cy3-labeled oligonucleotides were placed in the corners of the array for array quadrant identification and alignment. After printing, the slides were processed for hybridization by briefly placing the slides in a humidity chamber, and then UV cross-linking the DNA to the glass. This was followed by a 2-min wash in 1% sodium dodecyl sulfate, followed by washing twice in double-distilled water. The slides were then spun dry at 1000g for 1 min.
Target Preparation. Genomic DNA was cut by MseI (New England Biolabs, Beverly, MA), and then a catch-linker was ligated to the MseI fragments. These fragments were then cut with a methylation-sensitive restriction enzyme, McrBC (New England Biolabs). "Mock-cut" samples were samples exposed to the same conditions and reagents of the digested samples, except no GTP was added to drive the restriction digest. Twenty nanograms of the mock-cut or uncut samples and 20 ng of McrBC-cut MseI-linked genomic DNA was amplified by PCR. The PCR products were purified with the QIAquick PCR purification kit (QIAGEN). Fluorescent Cy3 or Cy5 dye was incorporated into the PCR product using the BioPrime DNA labeling system (Invitrogen, Carlsbad, CA). Purified DNA after chromatin immunoprecipitation (ChIP) or input (IP) DNA (0.3–0.5 µg) was directly labeled with fluorescent Cy3 or Cy5 dye using a BioPrime DNA-labeling system (Invitrogen), except that the 10x dNTPs mix provided was replaced by our own containing 1.2 mM dATP, dCTP, dGTP and 0.6 mM dTTP, and 3 µl of 1 mM Cy3/Cy5-dUTP was added per reaction.
After the labeling, the cut and mock reactions are mixed together and again cleaned with the QIAquick PCR purification kit (QIA-GEN). After purification, the labeled target was lyophilized to dryness and resuspended in 10 µl of hybridization buffer [2x SSC, 0.1% SDS, 100 ng/µl Cot1 DNA, 100 ng/µl oligo(dA)], denatured by boiling for 10 min, and added to a processed array slide. A coverslip (22 x 22 mm) was applied, and the array was placed in a hybridization chamber (GeneMachines) at 62°C for 4 to 8 h. Following hybridization, slides were washed by placing them into 50-ml conical tubes containing 2x SSC, 0.1% SDS for 5 min, 0.06x SSC, 0.1% SDS for 5 min, and 0.06x SSC for 2 min, all at room temperature. Slides were scanned for Cy3 and Cy5 fluorescence using an Axon GenePix 4000 microarray reader (Axon Instruments, Inc., Foster City, CA).
For chromatin immunoprecipitation experiments (ChIP-Chip), immunoprecipitated DNA, as well as IP DNA, were used for labeling. Reaction conditions for fluorescent labeling, target purification, and microarray hybridization were the same as for methylation analysis of genomic DNA.
Microarray Data Analysis. The data from scanned microarray images were extracted using GenePix software. Median of pixel intensity of each spot and median of its local background were used for further analysis by an Excel data analysis package. To normalize Cy3 and Cy5 signal intensities, we used the "interactive linear regression" approach with minor modifications (Finkelstein et al., 2002).
First, all intensity values were log transformed, then linear regression was performed only on values from the selected clones used for normalization (see below). Residuals were calculated and outliers (those residuals where e > 2 x standard deviation of e) were removed and regression function was recalculated. No further residuals were removed if the difference between the values of r-squared of the new regression line is less than 0.001 of the old. Intercept values were applied as correction factors to the log transformed channel 2 values of all clones. The result is that the function of log channel 1 and log channel 2 of clones closely approximates y = x. All values were then made exponential and ratios were calculated.
To identify clones with differential fluorescence ratios that would accurately reflect relevant differences in DNA methylation or histone acetylation, a simple algorithm was employed. Only spots with original signal intensities greater or equal to 2-fold local background, in at least one channel, were used to calculate ratios and further analyzed. Average fluorescence ratios were calculated from normalized ratios of three independent experiments only if the percentage of coefficient of variation was equal to or less than 30%, allowing us to analyze only the most reproducible clones. If the clone in question passed the first two filters, then the average ratios were compared. A clone was considered differentially methylated (or unmethylated) between two populations when the difference between the average ratios was 30% or greater, as determined empirically from early results. The same approach was taken for the identification of differential histone H4 acetylated clones, except in this case, immunoprecipitated DNA was compared with a reference DNA (input DNA isolated from the same experiment, but not immunoprecipitated).
Mitochondrial DNA is unmethylated (Groot and Kroon, 1979), therefore signals intensities of both channels coming from mitochondrial clones are expected to be equal. Data from arrays analyzing methylation were normalized based on signals of spots containing four mitochondrial clones in four serial dilutions. These dilutions were spotted in each of four subarrays. Their pixel intensities covered the whole signal range of the microarray (see light blue spots Fig. 2B). After normalization, unmethylated clones have a ratio close to 1. Ratios less than 1 reflect various levels of methylation.
|
|
|
To be able to compare ratios from different microarrays, where IP DNA was used in one channel and ChIP DNA in the other, Alu clones were also used for normalization (see light green spots Fig. 7B). Ratios ChIP/IP greater or less than 1 indicate higher or lower acetylation, respectively, relative to Alu sequences. Each microarray experiment was done in triplicate using material from two or three independent biological replicates. Dye swap experiments were always employed.
|
Bisulfite Sequencing. Five micrograms of genomic DNA was modified with sodium bisulfite under conditions previously described (Clark et al., 1994). The various CpG islands were amplified from the bisulfite-modified DNA by two rounds of PCR using nested primers specific to the bisulfite-modified sequence under study. Primers were designed to analyze the CpG sites located within the CpG island; all primer sequences are available upon request. Both rounds of PCR were performed under the same parameters, with 1% of the first round PCR product serving as the template in the second round PCR. PCR amplification was performed in an MJ thermal cycler (PTC200) under the following conditions: 94°C for 4 min followed by 5 cycles of 94°C for 1 min, 56°C for 2 min, 72°C for 3 min, then 35 cycles of 94°C for 30 s, 56°C for 2 min, 72°C for 1.5 min, and ending with a final extension of 72°C for 6 min.
The resultant PCR product was cloned into a TA cloning vector according to the manufacturer's instructions (pGEM-T-Easy cloning kit; Promega, Madison, WI). Ten positive recombinants were isolated using a QIAprep Spin Plasmid Miniprep kit (QIAGEN) according to the manufacturer's instructions and sequenced on an ABI automated DNA sequencer. The methylation status of individual CpG sites was determined by comparison of the sequence obtained with the known target sequence. The number of methylated CpGs at a specific site was divided by the number of clones analyzed (minimum of 10 in all cases) to yield a percentage of methylation for each site.
Chromatin Immunoprecipitations. Chromatin immunoprecipitations using the acetyl-histone H4 antibodies were performed according to the manufacturer's instructions (Upstate Biotechnology, Lake Placid, NY) with slight modifications (Futscher et al., 2002). Cells were rinsed in 1x Hanks' balanced salt solution with 0.1% EDTA and treated with 1% formaldehyde for 10 min at 37°C to form DNA protein cross-links. The cells were rinsed in ice-cold 1x Hanks' balanced salt solution with 0.1% EDTA containing protease inhibitors (1 mM phenylmethylsulfonyl fluoride, 1 µg/ml aprotinin, and 1 µg/ml pepstatin A), scraped, and collected by centrifugation at 4°C. Cells were then resuspended in a PIPES buffer (5 mM PIPES pH 8.0, 85 mM KCl, 0.5% NP40) containing protease inhibitors and incubated for 10 min on ice. Cells were then collected by centrifugation, resuspended in a SDS lysis buffer containing protease inhibitors, and incubated on ice for 10 min. The DNA protein complexes were sonicated to lengths between 200 and 1000 base pairs as determined by gel electrophoresis. Samples were centrifuged at 14,000 rpm at 4°C to spin out cell debris, and then the supernatant was diluted 10-fold with ChIPs dilution buffer containing protease inhibitors. One-tenth of the sample was set aside for input control, and the remaining sample was precleared with protein A-Sepharose (Amersham Biosciences Inc., Piscataway, NJ).
Following preclearing, the samples were split into thirds with two of the three samples treated with acetyl-histone H4, whereas the third sample was left as –Ab control. All samples were rotated overnight at 4°C. The chromatin-antibody complexes were collected using protein A-Sepharose, then sequentially washed with the manufacturer's low salt, high salt, and LiCl buffers, and then washed twice with Tris-EDTA.
The chromatin-antibody complexes were eluted, and the DNA protein cross-links were reversed with 5 M NaCl at 65°C for 4 h for all samples, including the input DNA control. All samples were treated with proteinase K, and the acetyl-histone H4 enriched fractions of genomic DNA were recovered by phenol/chloroform extractions and ethanol precipitations, which were later quantitated using a BioPhotometer (Eppendorf Scientific, Westbury, NY).
Real-Time PCR. PCR amplification was performed using Taq-Man primer/probes specific for the CpG island clones. Primer sequences were designed using the ABI Assays by Design service; the primer probe sequences are available upon request. Real-time PCR was used to analyze ChIP DNA, using the ABI Prism 7000 sequence detector following Applied Biosystem's PCR master mix (Foster City, CA) protocol. Real-time PCR was carried out in triplicate on 5 ng of DNA using parameters recommended by ABI.
For each experiment, the threshold bar was set within the linear range of the PCR amplification. For the majority of the experiments, the data were analyzed with the threshold set at 0.1. The resulting Ct and Rn files were exported to Microsoft Excel for data and graphical analysis. The Ct is the number of PCR cycles necessary to reach fluorescence intensity (an indirect measure of PCR product) within the linear range of PCR amplification. Quantification was determined by applying the comparative Ct method, as described in the ABI 7000 sequence detection user guide and others (Litt et al., 2001). Briefly, -fold enrichment was calculated by subtracting the Ct value of the ChIP DNA from the Ct value of the input DNA fraction and using this value as the power that 2 is raised to (i.e., 2Ct(Input)–Ct(ChIP)).
Slot-Blot Hybridization. To evaluate the histone acetylation state of Alu and satellite repeats, three doubling dilutions (20, 10, and 5 ng) of chromatin-immunoprecipitated and input DNA were blotted on BrightStar-Plus nylon membrane (Ambion, Austin, TX). To prepare DNA for Alu repeat probe, nine clones containing more than 85% of Alu repeats from six different families were selected from the CpG island library (BF.17.E3, BF.19.G4, BF.22.A7, BF.25.B9, BF.32.C2, BF.40.A9, BF.52.H3, BF.53.B12, and BF.53.C11), amplified by PCR, purified with the QIAquick PCR purification kit (QIAGEN), and mixed together in equal amounts. DNA for HSATII probe was prepared likewise. Four clones containing HSATII (BF.10.F1, BF.18.B5, BF.30.D5, and BF.52.D7) were amplified and mixed. Probe labeling and membrane hybridizations were done with the AlkPhos direct labeling and detection kit (Amersham) according to manufacturer's recommendations. One hundred nanograms of each DNA mix were used for labeling. Hybridizations and washes were performed at 55°C, and signals were detected using the chemifluorescent substrate ECF and a storm scanner (Amersham) and quantified with ImageQuant software (Amersham).
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
, CD11b, HCK, OS-9, and HoxA1 (Quintrell et al., 1987; Lanotte et al., 1991; Hu et al., 1993; Kimura et al., 1998; Di Croce et al., 2002; Park et al., 2003). A number of genes were also repressed following ATRA treatment, many of which are consistent with a reduction in proliferative capacity and are likely indirect downstream targets of ATRA treatment, such as hTERT, c-myc, and c-myb (Lee et al., 1987; Albanell et al., 1996) (Table 1). Taken together, these cytologic, immunologic, and molecular biologic assessments indicate ATRA is capable of inducing terminal differentiation of NB4 cells and suggests that this pharmacologic approach could be used to identify epigenomic changes that occur during ATRA-induced leukemic cell differentiation.
|
Genome-Wide Methylation Analysis. To assess genome-wide methylation levels, DNA from the respective cells was isolated and analyzed using our CpG island microarrays. The basic approach is shown in Fig. 1. Genomic DNA is isolated from tumor cells, digested with MseI, and the resultant restriction fragments are ligated to linkers that are specific for MseI restriction ends and contain PCR primer sequences. The linker-ligated genomic DNA is then divided into two aliquots. One aliquot is the test sample and is digested with a McrBC, whereas the other aliquot is the reference and is not digested with the methylation-sensitive restriction enzyme. The McrBC recognition sequence is 5'... PumC (N40–3000)Pum C... 3' and therefore digests the DNA if the region is methylated. In addition, the probability of this recognition site within a CpG island clone is near 100%, thereby rendering all the clones on the CpG island microarray potentially informative.
These two aliquots (undigested mock and McrBC-digested) were then amplified by PCR, followed by a direct random-labeling step that incorporates fluorescent Cy3 or Cy5 dUTP. In the test sample, if the CpG island is unmethylated, then McrBC will not digest sites within the linkered genomic fragments, and the DNA fragment will remain intact and production of a Cy5-labeled PCR product will result. If, however, the CpG island is methylated, then McrBC will digest the linkered genomic fragments preventing production of a labeled microarray target. In contrast, the reference sample is PCR-amplified and then direct-labeled with Cy3 dCTP. All the linkered genomic fragments are intact, resulting in the production of a Cy3-labeled PCR product. The fluorescently labeled PCR products obtained from the test and reference samples are then mixed, denatured, applied to a CGI microarray for hybridization, washed to remove nonspecific binding, and laser scanned to obtain quantitative fluorescent emissions. The undigested reference sample serves as a reference to which the McrBC-digested fragments are compared, resulting in fluorescence ratios calculated for each CpG island with the reference sample as the denominator and the test sample as the numerator (test/reference). The resultant ratio reflects the degree of methylation for each CpG island on the CGI microarray in the following manner. A ratio that approaches 0 indicates a methylated CpG island—no production of labeled PCR product following McrBC and digestion although the undigested reference will yield a labeled PCR product. A ratio approaching 1 indicates an unmethylated CpG island—fluorescently labeled PCR product will be obtained in the McrBC digested test sample and the undigested reference. Because mitochondrial DNA is unmethylated DNA (Groot and Kroon, 1979), these elements serve as control for unmethylated DNA, and more importantly they were used to normalize the data, controlling small variations in Cy dye incorporation, input DNA amounts, etc. Positive controls for methylated sequences, thereby ensuring the digestion by McrBC, were the satellite DNA clones (Hassan et al., 2001).
The types and distribution of genomic clones in the DNA sequence-validated genomically annotated CpG island library used for microarray production in this study is in excellent agreement with the library characteristics first reported by Cross et al. (1994). The library consists of approximately 75% single-copy CpG islands with the remaining clones consisting of repetitive elements such as ALU, satellite DNA, rDNA, and mitochondrial elements. To ease analysis, these clones were divided into categories. Rank 1 clones (R1) are clones within 0.5 kb of transcription start (TS); rank 2 clones (R2) are clones within 1 kb of TS; rank 3 clones (R3) are within 2 kb of TS; rank 4 clones (R4) are clones located farther than 2 kb from TS; SAT are clones containing satellite repeats; and ALU are clones that are more than 25% Alu repetitive element, based on clone length. Ranks are based on the November 2002 freeze of the human genome from the University of California Santa Cruz database (http://genome.ucsc.edu/index.html).
Using the CpG island microarray approach, we analyzed the cytosine methylation status in NB4 cells, NB4 cells treated with ATRA, and normal PBMC. An example of a microarray hybridization result is shown in Fig. 2A. Graphical representation of the quantitative hybridization results is shown in Fig. 2B. The top graph shows the results from normal PBMC; the light blue dots show the fluorescent ratios for all the mitochondrial clones and reflects the location of unmethylated DNA. Four CpG island clones are highlighted in different colors; three have similar fluorescent ratios as mitochondria and are therefore considered unmethylated, and one has a fluorescent ratio indicative of the genomic element being methylated in PBMC. The bottom graph shows the results for NB4; the four highlighted clones in this case show differing methylation status compared with normal PBMC.
From these types of results, comparison of the methylation states revealed an overall increase in methylation of CpG islands in NB4 cells when compared with normal PBMC (Fig. 3). Quantitatively, NB4 cells displayed an overall increased methylation in all the single-copy elements—clone ranks R1 to R4. In contrast, significant changes in the methylation of repetitive elements could not be detected. Finally, although there was a net increase of approximately 157 individual single-copy CpG island clones in ranks 1 to 4, there were only two CpG island clones in NB4 cells that showed hypomethylation when compared with normal PBMC.
|
We set out to validate a subset of these methylation results using bisulfite sequencing. Clones from ranks R1 to R4 were chosen based on differential methylation between normal PBMC and the APL cell line, NB4. One CpG island scored as methylated in NB4 and unmethylated in PBMC on the CpG island microarray was the CpG island associated with RAR , which has previously been shown to be aberrantly methylated in NB4 cells. In addition, we also analyzed R1 and R2 clones, such as the CpG island elements associated with HoxA1, and MARK2, as well as an R4 clone most closely associated with the MAD1L1 gene, which was chosen as an example of a CpG-rich area normally methylated in PBMC that becomes demethylated in NB4 tumor cells.
In Fig. 4, a clonal representation of the bisulfite sequencing result is shown for the above-mentioned samples and their CpG islands. Overall, there was complete concordance between the methylation state as determined by the CpG island microarrays and the methylation state as determined by bisulfite sequencing. Based on these results, it is clear that the CpG island microarrays employed in this study can accurately determine the methylation status of the CpG islands queried.
|
Table 2 is a list of R1 and R2 clones that show aberrant methylation in the NB4 cell line when compared with normal PBMC. This table reveals approximately 60 islands not previously known to be methylated in leukemia and provides a number of new potential genes that may participate in oncogenesis through methylation-mediated transcriptional silencing. Interestingly, it is noted that a number of these CpG islands show low-level sporadic methylation in the samples of PBMC. This is likely due to the fact that many of these clones are found on the edges of the CpG island and reflect the means by which they were first isolated—specifically using the restriction enzyme MseI to release the CpG islands. MseI cuts at 5'-TTAA-3' which will typically cut outside of CpG islands, thereby capturing CpG island edges, which often display varying amounts of DNA methylation.
|
In subsequent experiments, we wanted to determine whether the observed terminal differentiation of the leukemic cells induced by ATRA was associated with demethylation of critical targets, and the impetus was based on a recent report that showed ATRA could induce demethylation of RAR in NB4 cells (Di Croce et al., 2002). To this end, we analyzed the methylation patterns of NB4 before and after ATRA-induced terminal differentiation. The patterns obtained for both samples were identical; ATRA appears to have little or no influence on DNA methylation. Overall results are shown graphically in Fig. 5. Because our CpG island microarray contains a variety of different genomic elements including CpG islands associated with single-copy genes, Alu sequences, satellite sequences, and CG-rich regions not currently associated with a transcribed unit, it seems unlikely that any particular type of genomic element is selectively demethylated by ATRA treatment. In addition, we were unable to detect demethylation of the RAR CpG island by the CpG island microarray analysis or bisulfite sequencing (Fig. 6). Results from these experiments suggest that ATRA does not induce changes in DNA methylation in association with leukemic cell differentiation.
|
|
Genome-Wide Histone H4 Acetylation Analysis. Experiments were performed to assess genome-wide differences in histone acetylation state between NB4 and PBMC and between NB4 cells before and after ATRA treatment. Chromatin immunoprecipitations with antibodies specific for acetylated histones H4 were used to capture genomic regions enriched for histone acetylation. The DNA purified from the chromatin of NB4 cells, ATRA-differentiated NB4 cells, and normal PBMC was randomly labeled and hybridized to the CpG island microarray; nonimmunoprecipitated input DNA was used as the reference. A representative hybridization is shown in Fig. 7. Dramatic differences in genome-wide histone acetylation states were detected between NB4 cells and PBMC. R1 through R4 clones, as well as satellite sequences, showed a generalized hypoacetylation when compared with the terminally differentiated PBMC (Fig. 8). Overall, these results suggest that leukemic cells have a more compacted, topologically constrained genome.
|
ATRA-induced differentiation of NB4 cells produced striking changes in histone H4 acetylation levels. Increased histone acetylation was seen throughout R1 and R4 clones and surprisingly was also seen in a vast majority of satellite clones (Fig. 9) producing an epigenetic state with respect to histone acetylation that is similar to normal terminally differentiated PBMC. Approximately 282 CpG island microarray probes in ranks R1 to R4 showed increased histone acetylation following ATRA treatment, whereas only 34 showed a decrease in acetylation. No changes in histone acetylation could be detected in ALU elements. Table 3 presents a list of rank 1 and 2 clones that showed a >1.5-fold increase in histone H4 acetylation state following ATRA-induced differentiation.
|
|
Histone acetylation changes in specific genomic regions detected by microarray profiling were confirmed through examination of selected clones using real-time PCR analysis of DNA recovered from the same ChIP assay. Primer probe sets to four CpG islands identified by ChIP microarray were designed and used to validate the microarray results (Fig. 10). Three of the CpG islands were associated with increased histone H4 acetylation: RAR , HoxA1, and MARK2. A CpG island associated with MAD1L1 showed decreased histone H4 acetylation. In each case, the real-time PCR results were in complete concordance with the CpG island microarray results.
|
In addition to the single-copy elements that increased their histone acetylation state, satellite sequences associated with pericentromeric heterochromatin also showed an increase in histone acetylation in NB4 cells after differentiation with ATRA. As can be seen in Fig. 9, 
80% of satellite clones show a consistent increase in the acetylation of histone H4. Validation of these results was confirmed by slot-blot analysis of the immunoprecipitated DNA and the input DNA. Varying amounts of DNA immunoprecipitated with anti-acetyl-histone H4 antibody as well as the nonimmunoprecipitated input from -untreated and ATRA-treated NB4 cells were spotted down on nylon membranes and probed sequentially with alkaline phosphatase-labeled probes specific for human satellite sequences and then ALU sequences. These results are shown in Fig. 11 and are in agreement with the CpG island microarray results. Specifically, satellite sequences showed a 2-fold increase in histone H4 acetylation after ATRA treatment confirming the results obtained by CpG island microarray. These results also show that the satellite DNA is still underacetylated before and after treatment relative to input DNA. In contrast, Alu sequences showed higher acetylation level than satellite DNA before ATRA treatment, relative to input; however their acetylation state does not change after ATRA treatment. It appears unlikely that the changes in H4 acetylation simply reflects an overall nonselective increase in genomic acetylation because the widely dispersed Alu elements found in our microarray do not display this increase in histone H4 acetylation. In summary, the histone H4 acetylation results obtained by CpG island microarray indicate that: 1) the NB4 cells are relatively underacetylated compared with normal PBMC; 2) ATRA induces an increase in H4 acetylation selective for single-copy CpG islands and satellite sequences, but not ALU sequences; 3) the increase in acetylation can occur despite the presence of methylation in the region; and 4) histone H4 acetylation can partially overcome the repressive effects of aberrant methylation to reactivate expression of our RAR target genes.
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
CpG island microarrays revealed a net increase of 157 more single-copy CpG islands methylated in NB4 tumor cells when compared with the normal PBMC counterpart. Although there was a clear net increase in methylated CpG islands in NB4 cells, there were also incidences where CpG islands became unmethylated, and bisulfite sequencing was used to validate examples of both cases and confirm the CpG island microarray results. Of the aberrantly methylated CpG islands identified, 71 of them were within 1 kb of a transcription start site of a currently known human gene. Importantly, the CpG island microarray was able to correctly ascertain that the RAR CpG island was aberrantly methylated in agreement with previous reports (Di Croce et al., 2002). These results provide greater confidence that a portion of the remaining CpG islands identified as aberrantly methylated likely represent bona fide new targets in APL, whereas, almost assuredly, others are the result of stochastic methylation events that occurred during tumor cell evolution, as well as methylation changes that may be associated with long-term cell culture.
One of the potential targets of aberrant methylation identified in NB4 cells by CpG island microarray was the CpG island associated with HoxA1, which in turn was confirmed by bisulfite sequencing. HoxA1 is a member of the homeobox family of genes whose products function as transcription factors and whose dysregulation has been linked to leukemogenesis (Buske and Humphries, 2000; Owens and Hawley, 2002). Although little is known about the potential role of HoxA1 in APL, it has been recently reported that HoxA1 expression is increased following ATRA-induced differentiation of NB4 cells (Park et al., 2003) with which our own expression results agree. In addition, histone H4 associated with HoxA1 becomes hyperacetylated following treatment with ATRA.
Although CpG island methylation differences were readily discernible between NB4 and PBMC, no differences in methylation could be detected between NB4 cells and NB4 cells differentiated with ATRA either at the level of CpG island microarray or at the level of bisulfite sequencing of a known target, RAR . It is possible that subtle differences in CpG island methylation induced by ATRA were missed or that the demethylation resulting from ATRA treatment is variable. In either case, these results do suggest that ATRA drives transcriptional reprogramming primarily at the level of histone modification and that these modifications can, in part, overcome cytosine methylation-mediated repression.
Using CpG island microarrays to assess the histone modification state of NB4 cells, it revealed that, overall, the genome was relatively underacetylated at histone H4 compared with normal PBMC with both single-copy sequences and satellite sequences being principal targets. Again, the CpG island microarrays were able to correctly ascertain that RAR went from unacetylated to heavily acetylated at histone H4 following treatment with ATRA. The increase in acetylation of the RAR CpG island in ATRA-differentiated NB4 cells is consistent with previous reports that demonstrated ATRA relieved transcriptional repression by releasing the PML/RAR chimera and its associated histone deacetylase complexes (Grignani et al., 1998; Lin et al., 1998; Wang et al., 1998).
In addition, the CpG island microarrays also revealed significant increases in the H4 acetylation of HoxA1 following ATRA treatment; both sets of these data were confirmed by real-time PCR using TaqMan chemistry. Taken together, the DNA methylation results, the histone acetylation data, and the increase in HoxA1 expression seen after ATRA treatment suggest that HoxA1 may be a target of aberrant epigenetic modification in APL and that ATRA may be able to partially overcome the repressive effects of aberrant cytosine methylation. It is unknown if HoxA1 represents a direct or indirect target of ATRA. Evidence in the literature suggests that retinoids induce HoxA1 expression in other systems, with some data indicating that HoxA1 is a direct target of retinoids (Chariot et al., 1995; Kolm and Sive, 1995; Minucci et al., 1996). In addition, RAR response elements can be readily identified in the HOXA1 promoter and future studies will determine directly if the PML/RAR fusion protein binds and represses HoxA1 in APL cells.
It is likely, however, that a number of the elements identified are indirect downstream targets of the relief of transcriptional repression mediated by ATRA on PML/RAR . This is supported by the acetyl H4 ChIP-Chip results that showed an increase in acetylation of CpG island targets were already acetylated prior to ATRA treatment. In addition, another likely example is represented by the significant increases in the H4 acetylation state of a large majority of pericentromeric satellite sequences of NB4 cells after ATRA-induced differentiation. Although these sequences do have many imperfect RAR response elements, preliminary chromatin immunoprecipitation studies indicate that the PML/RAR fusion protein does not bind this region. Interestingly, the acetylation state of histone H4 levels of the ATRA-differentiated NB4 resemble the acetylation state observed in normal PBMC, a population of cells that are largely or completely terminally differentiated, and others have also seen an increase in histone acetylation, albeit a transient event, during differentiation of HL-60 down the granulocytic and monocytic pathways (O'Neill and Turner, 1995). It is tempting to speculate that the small but significant increase in histone acetylation seen in the satellite sequences of NB4 during differentiation allow for the binding of critical proteins that would act to prevent any further inappropriate cell divisions of a terminally differentiated cell.
In summary, large differences in the cytosine methylation and histone acetylation state distinguish the human APL cell line from normal PBMC. Results presented herein demonstrate that NB4 cells have increased methylation at a number of CpG islands with more than 70 new potential targets identified occurring within 1 kb of transcription start of known genes. One intriguing new target may be HoxA1, which shows, in addition to aberrant methylation, increased histone H4 acetylation and increased gene expression following ATRA treatment, suggesting that increased histone acetylation can partially overcome the repressive effects of DNA methylation. In addition, based on the epigenomic changes observed during ATRA-driven differentiation, it appears that the primary mechanism of action of ATRA is mediated through increased histone acetylation of the genome rather than demethylation of the genome. In conclusion, we demonstrate that ATRA stimulates complex epigenomic changes during leukemic cell differentiation, and monitoring these changes offers the opportunity to identify new genetic targets subject to epigenetic dysfunction.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
ABBREVIATIONS: RAR, retinoic acid receptor; APL, acute promyelocytic leukemia; ATRA, all-trans-retinoic acid; PBMC, peripheral blood mononuclear cell(s); PCR, polymerase chain reaction; SSC, standard saline citrate; Cy3, cyanine 3; Cy5, cyanine 5; ChIP, chromatin immunoprecipitation; IP, input; McrBC, methylation-sensitive restriction enzyme; PIPES, 1,4-piperazinediethanesulfonic acid; TS, transcription start; kb, kilobase(s); kbp, kilobase pair(s).
Address correspondence to: Dr. Bernard W. Futscher, Arizona Cancer Center, 1515 N. Campbell Ave., Tucson AZ 85724-5024. E-mail: bfutscher{at}azcc.arizona.edu
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Albanell J, Han W, Mellado B, Gunawardane R, Scher HI, Dmitrovsky E, and Moore MA (1996) Telomerase activity is repressed during differentiation of maturation-sensitive but not resistant human tumor cell lines. Cancer Res 56: 1503–1508.
Alcalay M, Orleth A, Sebastiani C, Meani N, Chiaradonna F, Casciari C, Sciurpi MT, Gelmetti V, Riganelli D, Minucci S, et al. (2001) Common themes in the pathogenesis of acute myeloid leukemia. Oncogene 20: 5680–5694.[CrossRef][Medline]
Borrow J, Goddard AD, Sheer D, and Solomon E (1990) Molecular analysis of acute promyelocytic leukemia breakpoint cluster region on chromosome 17. Science (Wash DC) 249: 1577–1580.
Buske C and Humphries RK (2000) Homeobox genes in leukemogenesis. Int J Hematol 71: 301–308.[Medline]
Chakraborty S, Senyuk V, and Nucifora G (2001) Genetic lesions and perturbation of chromatin architecture: a road to cell transformation. J Cell Biochem 82: 310–325.[CrossRef][Medline]
Chariot A, Moreau L, Senterre G, Sobel ME, and Castronovo V (1995) Retinoic acid induces three newly cloned HOXA1 transcripts in MCF7 breast cancer cells. Biochem Biophys Res Commun 215: 713–720.[CrossRef][Medline]
Clark SJ, Harrison J, Paul CL, and Frommer M (1994) High sensitivity mapping of methylated cytosines. Nucleic Acids Res 22: 2990–2997.
Costello JF, Fruhwald MC, Smiraglia DJ, Rush LJ, Robertson GP, Gao X, Wright FA, Feramisco JD, Peltomaki P, Lang JC, et al. (2000) Aberrant CpG-island methylation has non-random and tumour-type-specific patterns. Nat Genet 24: 132–138.[CrossRef][Medline]
Cross SH, Charlton JA, Nan X, and Bird AP (1994) Purification of CpG islands using a methylated DNA binding column. Nat Genet 6: 236–244.[CrossRef][Medline]
de The H, Chomienne C, Lanotte M, Degos L, and Dejean A (1990) The t(15;17) translocation of acute promyelocytic leukaemia fuses the retinoic acid receptor alpha gene to a novel transcribed locus. Nature (Lond) 347: 558–561.[CrossRef][Medline]
Di Croce L, Raker VA, Corsaro M, Fazi F, Fanelli M, Faretta M, Fuks F, Lo Coco F, Kouzarides T, Nervi C, et al. (2002) Methyltransferase recruitment and DNA hypermethylation of target promoters by an oncogenic transcription factor. Science (Wash DC) 295: 1079–1082.
Finkelstein D, Ewing R, Gollub J, Sterky F, Cherry JM, and Somerville S (2002) Microarray data quality analysis: lessons from the AFGC project. Arabidopsis Functional Genomics Consortium. Plant Mol Biol 48: 119–131.[CrossRef][Medline]
Flynn PJ, Miller WJ, Weisdorf DJ, Arthur DC, Brunning R, and Branda RF (1983) Retinoic acid treatment of acute promyelocytic leukemia: in vitro and in vivo observations. Blood 62: 1211–1217.
Futscher BW, Oshiro MM, Wozniak RJ, Holtan N, Hanigan CL, Duan H, and Domann FE (2002) Role for DNA methylation in the control of cell type specific maspin expression. Nat Genet 31: 175–179.[CrossRef][Medline]
Grignani F, De Matteis S, Nervi C, Tomassoni L, Gelmetti V, Cioce M, Fanelli M, Ruthardt M, Ferrara FF, Zamir I, et al. (1998) Fusion proteins of the retinoic acid receptor-alpha recruit histone deacetylase in promyelocytic leukaemia. Nature (Lond) 391: 815–818.[CrossRef][Medline]
Groot GS and Kroon AM (1979) Mitochondrial DNA from various organisms does not contain internally methylated cytosine in -CCGG-sequences. Biochim Biophys Acta 564: 355–357.[Medline]
Hassan KM, Norwood T, Gimelli G, Gartler SM, and Hansen RS (2001) Satellite 2 methylation patterns in normal and ICF syndrome cells and association of hypomethylation with advanced replication. Hum Genet 109: 452–462.[CrossRef][Medline]
Hu ZB, Ma W, Uphoff CC, Lanotte M, and Drexler HG (1993) Modulation of gene expression in the acute promyelocytic leukemia cell line NB4. Leukemia 7: 1817–1823.[Medline]
Kimura Y, Nakazawa M, and Yamada M (1998) Cloning and characterization of three isoforms of OS-9 cDNA and expression of the OS-9 gene in various human tumor cell lines. J Biochem (Tokyo) 123: 876–882.
Kolm PJ and Sive HL (1995) Regulation of the Xenopus labial homeodomain genes, HoxA1 and HoxD1: activation by retinoids and peptide growth factors. Dev Biol 167: 34–49.[CrossRef][Medline]
Lanotte M, Martin-Thouvenin V, Najman S, Balerini P, Valensi F, and Berger R (1991) NB4, a maturation inducible cell line with t(15;17) marker isolated from a human acute promyelocytic leukemia (M3). Blood 77: 1080–1086.
Larson RA, Kondo K, Vardiman JW, Butler AE, Golomb HM, and Rowley JD (1984) Evidence for a 15;17 translocation in every patient with acute promyelocytic leukemia. Am J Med 76: 827–841.[CrossRef][Medline]
Lee J, Mehta K, Blick MB, Gutterman JU, and Lopez-Berestein G (1987) Expression of c-fos, c-myb and c-myc in human monocytes: correlation with monocytic differentiation. Blood 69: 1542–1545.
Lin RJ, Nagy L, Inoue S, Shao W, Miller WH Jr, and Evans RM (1998) Role of the histone deacetylase complex in acute promyelocytic leukaemia. Nature (Lond) 391: 811–814.[CrossRef][Medline]
Litt MD, Simpson M, Recillas-Targa F, Prioleau MN, and Felsenfeld G (2001) Transitions in histone acetylation reveal boundaries of three separately regulated neighboring loci. EMBO (Eur Mol Biol Organ) J 20: 2224–2235.[CrossRef][Medline]
Minucci S, Saint-Jeannet JP, Toyama R, Scita G, DeLuca LM, Tiara M, Levin AA, Ozato K, and Dawid IB (1996) Retinoid X receptor-selective ligands produce malformations in Xenopus embryos. Proc Natl Acad Sci USA 93: 1803–1807.
O'Neill LP and Turner BM (1995) Histone H4 acetylation distinguishes coding regions of the human genome from heterochromatin in a differentiation-dependent but transcription-independent manner. EMBO (Eur Mol Biol Organ) J 14: 3946–3957.[Medline]
Owens BM and Hawley RG (2002) HOX and non-HOX homeobox genes in leukemic hematopoiesis. Stem Cells 20: 364–379.
Park DJ, Vuong PT, de Vos S, Douer D, and Koeffler HP (2003) Comparative analysis of genes regulated by PML/RAR alpha and PLZF/RAR alpha in response to retinoic acid using oligonucleotide arrays. Blood 102: 3727–3736.
Quintrell N, Lebo R, Varmus H, Bishop JM, Pettenati MJ, Le Beau MM, Diaz MO, and Rowley JD (1987) Identification of a human gene (HCK) that encodes a protein-tyrosine kinase and is expressed in hemopoietic cells. Mol Cell Biol 7: 2267–2275.
Ruthardt M, Testa U, Nervi C, Ferrucci PF, Grignani F, Puccetti E, Peschle C, and Pelicci PG (1997) Opposite effects of the acute promyelocytic leukemia PML-retinoic acid receptor alpha (RAR alpha) and PLZF-RAR alpha fusion proteins on retinoic acid signalling. Mol Cell Biol 17: 4859–4869.[Abstract]
Saha V, Young BD, and Freemont PS (1998) Translocations, fusion genes and acute leukemia. J Cell Biochem Suppl 30–31: 264–276.
Shi H, Wei SH, Leu YW, Rahmatpanah F, Liu JC, Yan PS, Nephew KP, and Huang TH (2003) Triple analysis of the cancer epigenome: an integrated microarray system for assessing gene expression, DNA methylation and histone acetylation. Cancer Res 63: 2164–2171.
Wang ZG, Delva L, Gaboli M, Rivi R, Giorgio M, Cordon-Cardo C, Grosveld F, and Pandolfi PP (1998) Role of PML in cell growth and the retinoic acid pathway. Science (Wash DC) 279: 1547–1551.
Yan PS, Perry MR, Laux DE, Asare AL, Caldwell CW, and Huang TH (2000) CpG island arrays: an application toward deciphering epigenetic signatures of breast cancer. Clin Cancer Res 6: 1432–1438.
Zardo G, Tiirikainen MI, Hong C, Misra A, Feuerstein BG, Volik S, Collins CC, Lamborn KR, Bollen A, Pinkel D, et al. (2002) Integrated genomic and epigenomic analyses pinpoint biallelic gene inactivation in tumors. Nat Genet 32: 453–458.[CrossRef][Medline]
This article has been cited by other articles:
|
|
 |
|
  P. Novak, T. Jensen, M. M. Oshiro, G. S. Watts, C. J. Kim, and B. W. Futscher Agglomerative Epigenetic Aberrations Are a Common Event in Human Breast Cancer Cancer Res., October 15, 2008; 68(20): 8616 - 8625. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. S. Yan and T. H.-M. Huang Tools to Uncover Altered DNA Methylation in Cancer Genomes Am. Assoc. Cancer Res. Educ. Book, April 14, 2007; 2007(1): 194 - 196. [Full Text] [PDF]
|
|
|
|
 |
|
 S. Pfister, C. Schlaeger, F. Mendrzyk, A. Wittmann, A. Benner, A. Kulozik, W. Scheurlen, B. Radlwimmer, and P. Lichter Array-based profiling of reference-independent methylation status (aPRIMES) identifies frequent promoter methylation and consecutive downregulation of ZIC2 in pediatric medulloblastoma Nucleic Acids Res., April 1, 2007; 35(7): e51 - e51. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. S. Yan, C. Venkataramu, A. Ibrahim, J. C. Liu, R. Z. Shen, N. M. Diaz, B. Centeno, F. Weber, Y.-W. Leu, C. L. Shapiro, et al. Mapping Geographic Zones of Cancer Risk with Epigenetic Biomarkers in Normal Breast Tissue. Clin. Cancer Res., November 15, 2006; 12(22): 6626 - 6636. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Novak, T. Jensen, M. M. Oshiro, R. J. Wozniak, M. Nouzova, G. S. Watts, W. T. Klimecki, C. Kim, and B. W. Futscher Epigenetic Inactivation of the HOXA Gene Cluster in Breast Cancer Cancer Res., November 15, 2006; 66(22): 10664 - 10670. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. E. K. Ibrahim, N. P. Thorne, K. Baird, N. L. Barbosa-Morais, S. Tavare, V. P. Collins, A. H. Wyllie, M. J. Arends, and J. D. Brenton MMASS: an optimized array-based method for assessing CpG island methylation Nucleic Acids Res., November 6, 2006; 34(20): e136 - e136. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.-g. Wang, L. W. Barsky, E. Davicioni, K. I. Weinberg, T. J. Triche, X.-k. Zhang, and L. Wu Retinoic acid induces leukemia cell G1 arrest and transition into differentiation by inhibiting cyclin-dependent kinase-activating kinase binding and phosphorylation of PML/RAR{alpha} FASEB J, October 1, 2006; 20(12): 2142 - 2144. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Balch, P. Yan, T. Craft, S. Young, D. G. Skalnik, T. H-M. Huang, and K. P. Nephew Antimitogenic and chemosensitizing effects of the methylation inhibitor zebularine in ovarian cancer Mol. Cancer Ther., October 1, 2005; 4(10): 1505 - 1514. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. W. Laird Cancer epigenetics Hum. Mol. Genet., April 15, 2005; 14(suppl_1): R65 - R76. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
 |
 |
|
 |
 |
 |
0.67) is graphed for seven different clone categories. Clone categories (ranks): R1, clones starting within 0.5 kbp around TS; R2, clones starting within 1 kbp around TS; R3, clones starting within 2 kbp around TS; R4, clones located within a known gene but more than 2 kbp from TS; All, all clones of R1–R4; SAT, clones containing satellite repeats; Alu, clones consisting of greater than 25% of an Alu sequence.
