Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Genome-wide tracking of unmethylated DNA Alu repeats in normal and cancer cells

Genome-wide tracking of unmethylated DNA Alu repeats in normal and cancer cells 770–784 Nucleic Acids Research, 2008, Vol. 36, No. 3 Published online 15 December 2007 doi:10.1093/nar/gkm1105 Genome-wide tracking of unmethylated DNA Alu repeats in normal and cancer cells 1 2,3 1 1 2,3 Jairo Rodriguez , Laura Vives , Mireia Jorda , Cristina Morales , Mar Mun˜ oz , 1 1,2, Elisenda Vendrell and Miguel A. Peinado * 1 2 ´ ` Institut d’Investigacio Biomedica de Bellvitge (IDIBELL), L’Hospitalet, Institut de Medicina Predictiva i Personalitzada del Ca` ncer (IMPPC), Badalona and Institut Catala` d’Oncologia (ICO), L’Hospitalet, Catalonia, Spain Received September 19, 2007; Revised October 19, 2007; Accepted November 27, 2007 is not just a direct outcome of the number of coding ABSTRACT sequences and that the presence of multiple regulatory Methylation of the cytosine is the most frequent mechanisms accounts for a significant part of biological epigenetic modification of DNA in mammalian cells. complexity (1,2). Among these mechanisms, repetitive In humans, most of the methylated cytosines elements may play a key role in gene regulation and geno- are found in CpG-rich sequences within tandem mic structure. Active transposable elements are involved and interspersed repeats that make up to 45% of the in genome rearrangement and illegitimate recombination and can also influence gene expression by altering splicing human genome, being Alu repeats the most common or by acting as enhancers or promoters (3–7). Advances in family. Demethylation of Alu elements occurs in the understanding of epigenetic mechanisms that regulate aging and cancer processes and has been asso- these repetitive elements may contribute to elucidate their ciated with gene reactivation and genomic instabil- specific participation in biological processes (8). ity. By targeting the unmethylated SmaI site within Silenced regions in mammals and other vertebrates are the Alu sequence as a surrogate marker, we have differentiated, although not exclusively, by the presence quantified and identified unmethylated Alu elements of DNA methylation (9). Methylation of the cytosine is on the genomic scale. Normal colon epithelial cells an epigenetic modification of DNA that plays an impor- contain in average 25 486 10 157 unmethylated tant role in the control of gene expression and chromo- Alu’s per haploid genome, while in tumor cells this some structure in mammalian cells (10–13). Most of the figure is 41 995 17 187 (P = 0.004). There is an 5-methylcytosines are found in CpG-rich sequences within inverse relationship in Alu families with respect to tandem and interspersed repeats (9,12) of which the their age and methylation status: the youngest previous estimates indicate that constitute up to 45% of the human genome (14). Among these repeats, Alu’s, with elements exhibit the highest prevalence of the SmaI more than one million copies per haploid genome, are site (AluY: 42%; AluS: 18%, AluJ: 5%) but the lower considered the most successful family (15). Interestingly, rates of unmethylation (AluY: 1.65%; AluS: 3.1%, Alu’s are not randomly distributed within the human AluJ: 12%). Data are consistent with a stronger genome, as they tend to accumulate in gene-rich regions silencing pressure on the youngest repetitive ele- (14,16,17). Previous works have estimated that Alu ele- ments, which are closer to genes. Further insights ments harbor up to 33% of the total number of CpG sites into the functional implications of atypical unmethy- in the genome (18) and have been reported to be highly lation states in Alu elements will surely contribute methylated in most somatic tissues (18–20). Methylation to decipher genomic organization and gene regula- represents the primary mechanism of transposon suppres- tion in complex organisms. sion and active transposons are demethylated in mamma- lian genomes (12). It has been proposed that regions of the genome containing repetitive elements might be masked by compartmentalization of the chromatin, resulting in INTRODUCTION a reduction of the effective size of the genome (21). Progress in large-scale sequencing projects is critical to Noteworthy, even though a vast number of CpG dinucle- identify and decipher gene organization and regulation in otides are provided by the collection of repetitive sequences many species including human. Nevertheless, cumulated in the human genome, this dinucleotide is greatly under- evidences indicate that the complexity of living organisms represented throughout the genome, but it can be found *To whom correspondence should be addressed. Tel: +34 934978693; Fax: +34 934978697; Email: [email protected] 2007 The Author(s) This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/ by-nc/2.0/uk/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited. Nucleic Acids Research, 2008, Vol. 36, No. 3 771 at close to its expected frequency in small genomic regions (200 bp to a few kb), known as CpG islands (22). These areas are ‘protected’ from methylation and are located in the proximal promoter regions of 75% of human genes (12,13,22). Methylated CpG islands are strongly and hereditably repressed (12). Hence DNA methylation is usually considered as a sign of long-term inactivation (9,10,12). Cancer cells are characterized by the accumulation of both genetic and epigenetic changes. Widespread genomic hypomethylation is an early alteration in carcinogenesis and has been associated with genomic disruption and genetic instability (23–27). Repeats unmasked by demeth- ylation are likely to facilitate rearrangements due to mito- tic recombination and unwanted transcription (28–30). Alternatively, aberrant de novo methylation of CpG islands is a hallmark of human cancers and is associated with epigenetic silencing of multiple tumor suppressor genes (31–37). Therefore, the screening for differentially methylated sequences in tumors appears as a key tool to further understand the molecular mechanisms under- lying malignant transformation of cells. Although, the repertoire of methylation screening methodologies has expanded widely (37–39), and different approaches have been used to make bulk estimates of methylation in repetitive elements (40,41), there is still a lack of screening strategies that specifically allow a feasible identification of DNA methylation alterations in repetitive elements (21). Here we report two variants of a novel methodology to quantify and identify unmethylated Alu sequences. The CpG site within the consensus Alu sequence AACCC GGG is used as a surrogate reporter of methylation. Figure 1. Schematic diagram of the QUMA and AUMA methods. DNA Unmethylated sites are cut with the methylation-sensitive is depicted by a solid line, Alu elements are represented by dashed boxes. restriction endonuclease SmaI (CCCGGG) and an adap- The QUMA and AUMA recognition sites (AACCCGGG) are repre- tor is ligated to the DNA ends. Quantification of sented by dashed/gray boxes. CpGs at SmaI sites are shown as full circles UnMethylated Alus (QUMA) is performed by real-time when methylated and as open circles when unmethylated. The methyla- tion-sensitive restriction endonuclease SmaI can only digest unmethylated amplification of the digested and adaptor-ligated DNA targets, leaving blunt ends to which adaptors can be ligated. (A) QUMA is using an Alu consensus primer that anneals upstream performed by real-time PCR of an inner Alu fragment using a primer of the SmaI site and an adaptor primer extended with the complementary to the Alu consensus sequence upstream of the SmaI site and the primer complementary to the adaptor to which two Alu TT dinucleotide in its 3 end (Figure 1A). The product homologous nucleotides (TT) have been added. (B) In AUMA, sequences generated by this approach is completely inside the flanked by two ligated adaptors are amplified by PCR using a single Alu element and hence it is not possible to make a primer, the same adaptor primer plus the TT nucleotides. When only a few unique identification. As an alternative approach, we have nucleotides are added to the primer, i.e. TT, as illustrated here, other non- also performed restrained amplification of digested and Alu sequences may be amplified. This allows the amplification of a large number of sequences that typically range from 100 to 2000 bp. adaptor-ligated DNA fragments that are flanked by two close SmaI sites. In this case, the same primer homologous to the adaptor with the additional TT nucleotides at the Application of QUMA and AUMA to a series of colo- 3 end to enrich for Alu sequences is used in absence of the rectal carcinomas and their paired normal mucosa has Alu consensus primer (Figure 1B). This second approach offered global estimates of unmethylation of Alu elements is named Amplification of UnMethylated Alu’s (AUMA) in normal and cancer cells and has revealed a large collec- and results in a complex representation of unique DNA tion of unique sequences that undergo highly recurrent sequences flanked by two unmethylated SmaI sites. When hypomethylation and hypermethylation in colorectal resolved by high-resolution electrophoresis, the AUMA tumors. generated sequences appear as a fingerprint characteristic of each sample (Figure 2) and individual scoring and identification of each band can be performed. Because MATERIALS AND METHODS AUMA’s stringency is based on a short sequence Tissues and cell lines (AACCCGGG) that is found preferentially but not exclu- sively in Alu elements, other unmethylated sequences are Fifty colorectal carcinomas and their paired non- also present in AUMA fingerprints. adjacent areas of normal colonic mucosa were included 772 Nucleic Acids Research, 2008, Vol. 36, No. 3 AUMA products performing Monte Carlo simulations. One thousand Monte Carlo simulations were performed using an Excel Add-in (available at www.wabash.edu/ econometrics). In Monte Carlo simulations, it was assumed that 80–100% of SmaI sites at CpG islands are unmethylated and that 50–100% of SmaI sites in other genomic regions different from Alu’s and CpG islands are unmethylated. Quantification of QUMA One microgram of DNA was digested with 20 U of the methylation sensitive restriction endonuclease SmaI (Roche Diagnostics GmbH, Mannheim, Germany) for 16 h at 308C, leaving cleaved fragments with blunt ends (CCC/GGG). Adaptors were prepared incubating the oligonucleotides Blue (CCGAATTCGCAAAGCTC TGA) and the 5 phosphorylated MCF oligonucleotide (TCAGAGCTTTGCGAAT) at 658C for 2 min, and then cooling to room temperature for 30–60 min. One micro- Figure 2. AUMA of normal (N)–tumor (T) pairs of two different patients performed using primer BAu-TT. A highly reproducible band gram of the digested DNA was ligated to 2 nmol of patterning is observed among the four replicates. Representative bands adaptor using T4 DNA ligase (New England Biolabs, showing gains (hypomethylations) and losses (hypermethylations) are Beverly, MA, USA). Subsequent digestion of the ligated marked with up and down arrowheads, respectively. products with the methylation insensitive restriction endonuclease XmaI (New England Biolabs) was per- formed to avoid amplifications from non-digested methy- lated Alu’s. The products were purified using the GFX in this analysis. Samples were collected simultaneously as Kit (Amersham Biosciences, Buckinghamshire, UK) and fresh specimens and snap-frozen within 2 h of removal and eluted in 250ml of sterile water. then stored at 808C. All samples were obtained from the Quantitative real-time PCR was performed using 1 ng Ciutat Sanita` ria i Universita` ria de Bellvitge (Barcelona, (the equivalent of 333 genomes) of DNA in a LightCycler Spain). The study protocol was approved by the Ethics 480 real-time PCR system with Fast Start Master SYBR Committee. Human colon cancer cell lines (HT29, SW480, Green I kit (Roche). Mastermix was prepared to a final HCT116, LoVo, DLD-1, CaCo-2 and LS174T) were concentration of 3.5 mM MgCl and 1mM of each primer. obtained from the American Type Culture Collection The downstream BAu-TT primer (constituted by the 3 (ATCC; Manassas, VA). KM12C and KM12SM cells were end of Blue primer, and the GGGTT sequence including generously provided by A. Fabra. DNA from tumor– the GGG 3 side of the cut SmaI site and the Alu normal pairs was obtained by conventional organic extrac- homologous TT dinucleotide, ATTCGCAAAGCTCTG tion and ethanol precipitation. DNA purity and quality AGGGTT) and the upstream primer was an Alu was checked in a 0.8% agarose gel electrophoresis. RNA consensus sequence (CCGTCTCTACTAAAAATACA) from cell lines was obtained by phenol–chloroform (see Supplementary Data). Magnitudes were expressed extraction and ethanol precipitation, following standard as number of unmethylated Alus per haploid genome procedures. after DNA input normalization. The number of haploid genomes present in the test tube was determined in the Bioinformatic analysis same multiwell plate by quantification of Alu sequences irrespectively of the methylation state. A real-time PCR The distribution of SmaI sites, putative amplification hits, using Alu consensus primers upstream of the CCCGGG PCR homologies, CpG islands and repetitive elements site was performed (see Supplementary Data) and the was assessed using the human genome assembly 36.1 from number of genomes was calculated against a standard NCBI. Data were obtained from the Repbase (http:// curve constructed with a reference genomic DNA mea- www.girinst.org/repbase/index.html) and the Genome Browser Databases (http://hgdownload.cse.ucsc.edu/ sured by UV spectrophotometry. To determine the efficiency of the assay and to perform goldenPath/hg18/database/). Only assembled chromo- some fragments were considered. A Perl routine was absolute quantification, an external Alu product generated used to score all positions containing the target sequences by PCR from a DNA fragment containing an AluSx in all chromosomes (available from the authors upon element was used as standard (Supplementary Methods). request). Data were analyzed using Excel spreadsheets. The number of copies of the external control were spectro- To calculate the proportion of unmethylated Alu photometrically quantified and dilution curves were gene- elements at the genomic level, the number of AUMA rated and treated as samples. Comparison of dilution hits identified in bioinformatic analysis were corrected curves before and after sample processing indicated that according to the distribution of experimentally generated the mean recovery was 73%. DNA samples overdigested Nucleic Acids Research, 2008, Vol. 36, No. 3 773 with the methylation insensitive XmaI endonuclease were analogous to CGH. Briefly, an AUMA product obtained spiked with different amounts of the external standard and from a normal tissue DNA was purified using Jet quick processed. The sensitivity of the QUMA detection was 100 PCR product purification kit (Genomed, Lohne, unmethylated Alu’s per haploid genome (Supplementary Germany) and labeled with SpectrumRed dUTP (Vysis, Figure 1) using 1 ng of genomic DNA per PCR. A linear Downers Grove, IL, USA) using a Nick Translation kit response was observed between 1000 and 100 000 unmethy- (Vysis). Similarly, genomic DNA of the same normal lated Alu’s per haploid genome (Supplementary Data). sample was labeled with SpectrumGreen dUTP (Vysis) and both probes were cohybridized to metaphase chromo- somes. Procedures and image analysis were performed as Amplification of AUMA described (44). Differential normal–tumor representation of AUMA DNA digestion with SmaI enzyme and ligation to the linker at the genomic scale was performed by competitive was performed as described above for QUMA, except for hybridization of AUMA products to BAC arrays. the XmaI digestion that was skipped. The product was AUMA products from two normal–tumor pairs were purified using the GFX Kit (Amersham Biosciences) and purified using Jet quick PCR product purification kit eluted in 250ml of sterile water. Six different chimeric (Genomed, Lo¨ hne, Germany) and 1mg was labeled with primers constituted by the 3 end of the Blue primer dCTP-Cy3 or dCTP-Cy5 (Amersham Biosciences, UK) by sequence (ATTCGCAAAGCTCTGA), the cut SmaI site use of the Bioprime DNA Labeling System (Invitrogen, (GGG) and two, four or seven additional nucleotides Carlsbad, CA, USA). Probes were hybridized to Spectral- homologous to the Alu consensus sequence were used to Chip 2600 BAC arrays (Spectral Genomics, Houston, TX, enrich for Alu sequences (see Supplementary Methods). Three primers were designed to amplify ‘upstream’ of the USA) following the manufacturer’s instructions. Arrays SmaI site (towards ALU promoter): BAu-TT, BAu-TTCA, were scanned with a ScanArray 4000 (GSI Lumonics, BAu-TTCAAGC. Three other primers were designed to Watertown, MA, USA) and processed with GenePix amplify ‘downstream’ (towards ALU poly-A): BAd-AG, software (Axon Instruments, Union City, CA, USA). The BAd-AGGC, BAd-AGGCGGA. Letters after the dash resulting data were processed to filter out low-quality correspond to the 3 sequence of the primer (see Supple- spots based on spot area and similarity of readings mentary Data). Data reported here were obtained by using between the two replicates of each BAC. Data manipula- the BAu-TT primer. tion was performed using Excel spreadsheets. Because In each PCR reaction only one primer was used at AUMA products are not evenly distributed along chro- a time. Products were resolved on denaturing sequencing mosomes, only BACs with intensities above the 10% of gels. Although bands can be visualized by silver staining maximum intensity in at least one of the two channels of the gels, radioactive AUMA’s were performed for were considered for ratio calculations. The pattern of normal–tumor comparisons. A more detailed description chromosomal alterations in these two tumors was deter- of the PCR and the visualization of the bands are given mined by conventional CGH as described (44). as Supplementary Data. Only sharp bands that were reproducible and clearly Isolation and cloning of AUMA tagged bands distinguishable from the background were tagged and DNA excised from gels was directly amplified with the included in the analysis. Faint bands with inconsistent same primer used in AUMA (BAu-TT) (Supplementary display due to small variations in gel electrophoresis Figure 2). The amplified product was cloned into plasmid resolution were not considered. Band reproducibility was vectors using the pGEM-T easy vector System I cloning assessed with the analysis of PCR duplicates of three kit (Promega, Madison, WI, USA). Automated sequenc- independent sample digests from two different samples ing of multiple colonies was performed using the Big Dye and PCR replicates from the same digest from four paired Terminator v3.1 Cycle Sequencing kit (Applied Biosys- tumor–normal samples. AUMA fingerprints were visually tems, Foster City, CA, USA) to ascertain the unique checked for methylation differences between bands in the identity of the isolated band. Sequence homologies were tumor with regard to its paired normal mucosa. Under searched for using the Blat engine (http://genome.ucsc. these premises, a given band was scored according to three edu/). Selected clones corresponding to AUMA isolated possible behaviors: hypomethylation (increased intensity bands were radioactively labeled and used as a probe to in the tumor), hypermethylation (decreased intensity in confirm the identity of the excised band by hybridization the tumor) and no change (no substantial difference in to AUMA fingerprints as previously described (45). intensity between normal and tumor samples) (Figure 2). Only those bands showing clear changes in their intensities in the fingerprint were considered to represent methylation Bisulfite genomic sequencing changes. This is consistent with previous studies done Differential methylation observed in some AUMA using a related technique (42,43). tagged bands was confirmed by direct sequencing of bisulfite treated normal and tumor DNA as previously Competitive hybridization of AUMA products to metaphase described (46). Prior to sequencing, DNA was amplified chromosomes and BAC arrays using a nested or semi-nested PCR approach, as appro- The origin and chromosomal distribution of sequences priate. Three independent PCRs were done and products generated by AUMA was analyzed using procedures were pooled to ensure a representative sequencing. 774 Nucleic Acids Research, 2008, Vol. 36, No. 3 Table 1. Content and distribution of QUMA and AUMA hits in the human genome Sequence Mb Number of SmaI sites AACCCGGG Virtual AUMA Unmethylated Unmethylated b c d e f g elements (CCCGGG) hits AUMA hits hits hits hits (%) Total 3080.4 1 118 195 486 835 168 309 5498 201 14332 2418 8.52  1.4% Alu (S+J+Y) 227.3 1 091 110 198 201 155 226 5109 59 (29.3%) 4104 688 2.64  0.44% AluS 141.2 660 415 122 459 97 951 3382 45 (22.4%) 3028 510 3.09  0.51% AluJ 54.0 283 104 14 017 1235 38 2 (1.0%) 151 25 12.25  1.97% AluY 32.1 147 591 61 725 56 040 1689 12 (6.0%) 925 156 1.65  0.27% CpG islands 16.2 27 085 49 430 1673 63 55 (27.4%) 1501 97 90.5  5.79% Rest 2836.9 – 239 204 11 410 326 87 (43.3%) 8530 1650 75.9  14.63% Genome Mb represented by each type of element. Total number corresponds to the number of megabases analyzed for the presence of hits. Only assembled chromosome fragments were considered. Elements considered in the analysis as obtained from the Repbase and the Genome Browser Databases (see Material and Methods section). Number of occurrences of the sequence AACCCGGG (or CCCGGGTT) within each type of element. Number of AUMA hits present in virtual PCR products of up to 1000 bp. Hits of actual AUMA products. Only bands appearing in normal tissue were considered. Eighty-seven bands contributed two hits each (174 hits) and 27 bands contributed only one due to poor sequence or incomplete homology with the NCBI Build 36.1 of the human genome (hg18 assembly, March 2006). Twenty-three additional bands were detected mainly in tumor tissue and were not considered to perform calculations. Estimated number of unmethylated sites using Monte Carlo simulations (Material and Methods section). In respect to the total number of AACCCGGG (or CCCGGGTT) hits. The sequence of PCR primers is described in RESULTS Supplementary Data. Genomic estimation of the targets and evaluation of the adequacy of the approach by computational analysis The availability of the human genome map has allowed Histone modification analysis by chromatin us to make a detailed estimation of the frequency and immunoprecipitation (ChIP) distribution of the sites targeted by our approaches on 6 the genomic scale. A Perl routine was used to score all Briefly, 6 10 cells were washed twice with PBS and cross- positions containing the target sequences in all chromo- linked on the culture plate for 15 min at room temperature somes and was also applied to perform a virtual AUMA in the presence of 0.5% formaldehyde. Cross-linking (see Material and Methods section). Some of the most reaction was stopped by adding 0.125 M glycine. All subse- important data derived from the bioinformatic analysis quent steps were carried out at 48C. All buffers were pre- are shown in Table 1 and Figure 3. chilled and contained protease inhibitors (Complete Mini, Because of the C to T mutational bias at CpG sites (47), Roche). Cells were washed twice with PBS and then any amplification method relying on the consensus scraped. Collected pellets were dissolved in 1 ml lysis buffer sequence (see Supplementary Data) will only cover a (1% SDS, 5 mM EDTA, 50 mM Tris pH 8) and were fraction of all the Alu’s. Therefore it is important to sonicated in a cold ethanol bath for 10 cycles at 100% estimate the degree of representativity of the methods amplitude using a UP50H sonicator (Hielscher, Teltow, used here if genome-wide estimations are to be made. Alu Germany). Chromatin fragmentation was visualized in 1% repeats constitute 7.4% of the human genome but accu- agarose gel. Obtained fragments were in the 200–500 pb mulate 40.7% of all SmaI sites (Table 1). Nearly 200 000 range. Soluble chromatin was obtained by centrifuging the Alu’s (18% of all Alu’s) contain a SmaI site and 155 000 sonicated samples at 14 000g for 10 min at 48C. The soluble retain the AACCCGGG consensus sequence (Table 1 and fraction was diluted 1/10 in dilution buffer (1% Triton Figure 3) and are therefore potential targets of QUMA X-100, 2 mM EDTA, 20 mM Tris pH 8, 150 mM NaCl) and AUMA. While 38.0% of the youngest AluY elements then aliquoted and stored at 808C until use. contain this sequence, the proportions drop to 14.8 and Immunoprecipitation was carried out at 48C by adding 0.4% in AluS and AluJ families, respectively (Table 1 and 5–10mg of the desired antibody to 1 ml of chromatin. Figure 3). These frequencies are consistent with a higher C Chromatin–antibody complexes were immunoprecipitated to T transition trend at CpG sites in older Alu’s (47). with specific antibodies using a protein A/G 50% slurry The representativity of AUMA was analyzed by a (Upstate, Millipore, Billerica, MA, USA) and subse- virtual bioinformatic assay of the human genome quently washed and eluted according to the manufac- sequence. A total of 168 309 AACCCGGG (or CCCGG turer’s instructions. Antibodies against acetylated H3 GTT) hits were identified throughout the genome, with K9/K14 (Upstate), dimethylated H3 K79 and trimethyl- 92.9% of all hits within Alu elements (Table 1 and ated H3 K9 (Abcam, Cambridge, UK) were used. Enrich- Figure 3). This implies that 14.2% of all Alu elements ment for a given chromatin modification was quantified as contained the AACCCGGG sequence. Another 1.0% of a fold enrichment over the input using quantitative real- the hits were in CpG islands and 6.8% in the rest of the time PCR (Roche). For every PCR, a standard curve was genome (including unique sequences and other repeats) obtained to assess amplification efficiency. All quantifica- (Table 1 and Figure 3). As expected, Alu elements con- tions were performed in duplicate. taining the target sequence mostly belonged to the AluS Nucleic Acids Research, 2008, Vol. 36, No. 3 775 Figure 4. Quantitation of unmethylated Alu’s in 17 paired normal mucosa and colorectal carcinoma by QUMA. The values represent the estimated number of unmethylated Alu’s per haploid genome. Most tumors exhibited a higher level of hypomethylation when compared with the respective normal. Figure 3. Relative distribution the Alu elements and sequence targets considered in bioinformatic and experimental QUMA and AUMA. Mb: number of megabases occupied by each type of element; elements: number of elements considered (‘Rest’ has been set arbitrarily to 50%); while this figure is 24.9% in the cancer cell. Considering SmaI site: CCCGGG sequence; vQUMA hits: AACCCGGG (or GGG that the human genome contains 1.1 million Alu ele- CCCTT) sites in Alu elements; vAUMA hits: AACCCGGG (or GGGC ments, these estimates indicate that unmethylated Alu’s CCTT) sites; vAUMA ends: vAUMA hits considering only putative constitute the 2.3 and 3.8% of all Alu’s in the normal and AUMA products of <1 kb (see Material and Methods section); tumor tissues, respectively. AUMA: elements at each one of the two ends of actual AUMA products. Set-up and optimization of AUMA fingerprinting Because QUMA products are fully contained within the (2/3) and AluY families (1/3), with a minimal representa- Alu sequence, it is not possible to identify and position tion of the older family AluJ (<1%). Virtual AUMA in the genome the unmethylated Alu elements. To achieve determined the presence of 5498 putative products of this it is necessary to amplify the targeted unmethylated <1 Kb (the sequence AACCCGGG in the up strand and Alu element together with an adjacent unique sequence. the sequence CCCGGGTT in the down strand at a This was attained through the use of the second method, distance of <1 kb). Although actual AUMA PCR the Amplification of UnMethylated Alus (AUMA). products may reach 2 kb length (see below), we used the AUMA also targets the unmethylated AACCCGGG 1 kb limit to compare with AUMA-isolated bands, which sequence, as in QUMA, but in this case a single primer were shorter than 1 kb. Most virtual AUMA products is used in the PCR (BAu-TT) (see Figure 1 and Supple- contained an Alu element at one of the ends at least mentary Data). Moreover, the product is resolved on (93%). a high-resolution gel electrophoresis resulting in a band- rich fingerprint. AUMA bands correspond to sequences Quantification of unmethylated Alu in normal flanked by two unmethylated target sequences in opposite and tumor tissues strands and sufficiently close to allow PCR amplification The QUMA approach was applied to quantify unmethy- (Figure 1 and Supplementary Data). Since 92% of the lated Alu’s in a series of 18 colorectal carcinomas and their AACCCGGG occurrences in the human genome are in paired normal colonic mucosa. An external DNA frag- Alu’s (Table 1), the approach is largely biased towards ment containing an AluSx element was used as a standard the amplification of unmethylated Alu’s. The presence (see Materials and Methods section and Supplementary of non-Alu sequences at one of the ends or between two Data) in order to make an absolute quantification of the repetitive elements allows the positioning within the number of unmethylated Alu’s. Replicates and dilution genome map of all products. curves of the samples and standard were performed AUMA products generated using the BAu-TT primer to assess reproducibility, sensitivity and accuracy produced highly reproducible fingerprints consisting of (Supplementary Figure 1). Results were normalized by bands ranging from 100 to 2000 bp when resolved in assessment of the number of haploid genomes per test tube high-resolution sequencing gels (Figure 2 and Supplemen- (see Material and Methods section). The average number tary Figure 2). Some well-identifiable bands (up to 5 per of unmethylated Alu’s per haploid genome was experiment) showed random display in both intra-assay 25 486 10 157 in normal mucosa, and 41 995 17 187 and inter-assay replicates (Supplementary Figure 2) and in tumor samples (P = 0.004, paired t-test) (Figure 4). were not considered for analysis. A subset of 110 bands We computationally identified a total of 168 309 Alu with consistent display among all the experiments were elements containing the AACCCGGG sequence (potential tagged and selected for comparative analysis between targets of QUMA) (Table 1). Therefore we estimate samples (see Materials and Methods section). that 15.1% of Alu repeats with a AACCCGGG site are It should be noted that different fingerprints containing unmethylated in the average normal colonic mucosa cell, alternative representations may be obtained by AUMA 776 Nucleic Acids Research, 2008, Vol. 36, No. 3 just by using primers that either amplify from the SmaI strands. Nevertheless, in 27 sequences the AACCCGGG site towards the Alu promoter (upstream Alu amplifica- site was only present at one of the ends, with the other tion) or towards the Alu poly-A tail (downstream Alu end showing high homology with the primer although amplification). Also the stringency of the Alu selection it was not a perfect match. The presence of these may be increased by using longer primers containing sequences suggests that, in some instances, a single cut additional nucleotides corresponding to the Alu consensus in the sequence may be enough to produce an amplifiable sequence (see Material and Methods section). An illus- fragment. This is not considered an artifact since these trative example of AUMA fingerprints generated with bands still represent an unmethylated AACCCGGG site. different Alu-upstream and Alu-downstream primers is shown in Supplementary Figure 2. All the data reported Genome-wide estimations of unmethylation in Alu’s in this article regarding AUMA were obtained using the and distribution by subfamily BAu-TT primer. Of the 137 identified loci represented in AUMA, 114 were Chromosomal origin of AUMA products isolated from normal tissue DNA and 23 from tumor DNA. Half of the sequences contained an Alu sequence at Competitive hybridization between AUMA products one of the ends and two were flanked by two inverted and genomic DNA on metaphase chromosomes yielded Alu’s. AUMA sequences isolated from tumor tissue and a characteristic hybridization pattern demonstrating not present in normal tissue (this corresponds to a tumor- the unequal distribution of AUMA products along the specific hypomethylation) showed a higher proportion human genome (Figure 5A). Competitive hybridization of of Alu elements (16 out of 23, 70%), and included one AUMA products to BAC arrays showed profiles consis- sequence flanked by two inverted Alu’s. Globally, 78 tent with those obtained on metaphase chromosomes unmethylated Alu elements were identified and positioned (Figure 5B). The highest AUMA signal was detected in the human genome map. in whole chromosomes 16, 17 and 19 in contrast with To study the genomic distribution of unmethylated chromosomes 2, 13, 18 and X which were mainly labeled sequences in normal colon mucosa, we only considered the by genomic DNA. Other chromosomes showed a discrete 114 sequences obtained from normal tissue. This resulted pattern of AUMA product hybridization, in which telo- in a total of 201 unmethylated hits characterized through- meric bands in chromosomes 1, 4, 5, 9, 12 and X, and out the genome. The nature of the sequences represented interstitial bands in chromosomes 1, 3, 7, 11 and 12 are the most prominent examples. in actual AUMA showed striking differences with the distribution expected from the virtual AUMA analysis. Identification of AUMA amplified DNA products The methylation status of the sequence is likely to be the main (if not the only) source of these differences because To determine the identity of bands displayed by AUMA, the virtual AUMA did not consider this state. Therefore 38 tagged bands were isolated and cloned. Multiple clones we can use these differences to estimate the degree of from each band were sequenced, resulting in a total of unmethylation of the Alu repeats. Only 29.4% of the 49 different sequences due to the coincidence of more AUMA ends consisted of Alu’s, as compared with the than one sequence in some bands. Characterized bands expected 92.9% resulting from the bioinformatic analysis. included bands displaying no changes in the normal– The highest downrepresentation corresponded to the tumor comparisons and bands recurrently altered in the youngest AluY family, which was present in 6.0% of the tumor. Table 2 summarizes the main features of a subset AUMA ends, while it was expected to add up to 33.1% in of the bands showing recurrent alterations. A list of all the virtual AUMA. AluS representation in actual and virtual sequences isolated from AUMA fingerprints is provided AUMA was 22.3% and 60%, respectively. Interestingly, as Supplementary Table 1. All sequenced bands contained AluJ representations, in both actual and virtual AUMA, a region of non-repetitive sequence and matched with the were closer (1.0% and 0.7%, respectively) (Figure 3 and BLAST reference sequence, allowing the assignment of Table 1). These results suggest that there is a stronger a unique chromosomal localization. The BLAST reference pressure to methylate younger Alus. Alternatively, the hits sequence corresponding to the 49 sequences isolated from corresponding to CpG islands were overrepresented in the AUMA fingerprint presented the target sequence actual versus virtual AUMA by a factor of nearly 25-fold CCCGGGTT including the SmaI at both ends. Southern (27.4% versus 1.1%), consistently with the unmethylated blot analysis of selected cloned sequences showing coinci- status of most CpG islands (Figure 3 and Table 1). The dental size was performed to confirm its correspondence rest of the hits were located in different types of repetitive with the band displayed in AUMA fingerprints (Supple- elements (MIR, MER, LTR, LINE, etc.) and unique mentary Figure 4). sequences (Supplementary Table 1). The miscellaneous To obtain a more representative collection of AUMA collection of sequences (‘Rest’) was over-represented by bands, 200 clones obtained from normal tissue AUMA about 7-fold (observed hits: 43.3%; expected hits: 5.9%, products were sequenced. The analysis revealed 88 addi- Table 1). The 46 AUMA hits represented by the 23 bands tional sequences. This resulted in a total of 137 different loci represented in AUMA (Supplementary Table 1). specific of tumor tissue showed a higher proportion of Most sequences obtained by random cloning were also Alu’s compared with those obtained from normal tissue flanked by two AACCCGGG sequences in opposite DNA (41% versus 29%, respectively), but similar distribution Nucleic Acids Research, 2008, Vol. 36, No. 3 777 Figure 5. (A) Chromosomal origin of AUMA products. A competitive hybridization of AUMA product obtained from normal tissue DNA (red) and genomic DNA (green) to metaphase chromosomes was performed. AUMA products showed an unequal distribution along chromosomes, displaying highest densities at most telomeric regions and some interstitial bands. Chromosomes 16, 17 and 19 yielded the highest AUMA density. (B) Intensity distribution of AUMA products hybridized to BAC arrays in selected chromosomes. The average intensity (X-axis) of the two normal (blue) and tumor samples analyzed (red) for each BAC is shown. BACs are arranged along the Y-axis according to its position in the chromosome. (C) Differential methylation profiles determined by competitive hybridization of AUMA products from normal and tumor tissue to BAC arrays. Illustrative examples are shown for chromosomes 7 and 8 from the two cases analyzed (81 and 151). X-axis indicates log2 ratio of tumor/normal intensities. Positive values (to the right) indicate hypomethylations, negative values (to the left) indicate hypermethylations. Additional examples are shown in Supplementary Figure 5. 778 Nucleic Acids Research, 2008, Vol. 36, No. 3 Table 2. A selection of characterized AUMA bands Band ID Size % GC Chromosome map Gene CpG island Repetitive Methylation status a c (bp) (Location ) elements in in tumor 0 0 band ends (5 /3 ) Ai1 c3 509 55 17p11.2 (18206453–18206961) SHMT1 Yes Alu Sx/MIR Hypermethylated Aj2 c1 458 49 1q32.2 (206389082–206389539) MGC29875 Yes Alu Sq/None Hypomethylated Ao1 c4 365 50 19q13.32 (53550179–53550543) AK001784 No Alu Sx/MIRb Hypomethylated Ap1 c6 358 56 5q35.2 (175157321–175157674) CPLX2 Yes None/MIR Hypermethylated Aq3 c6 339 58 8p23.3 (2007343–2007682) MYOM2 No LTR/Alu Y Hypomethylated Ar3 c3 329 57 2q14.3 (127875178–127875506) AF370412 Yes None/MIRb Hypomethylated As3 c6 306 63 16p13.3 (3160477–3160782) None Yes None/None Hypermethylated Au4 c1 268 57 16p13.3 (3162099–3162366) None No tRNA/None Hypermethylated Nucleotide position within the contig (strand +). NCBI Build 36.1 of the human genome. The whole sequence or a fragment of the sequence lays not further than 200 bp of a predicted CpG island. As compared to the paired normal tissue. by Alu family (10 AluS, 5 AluY, 1 AluJ, 4 in CpG islands and 26 in other sequences). To calculate the proportion and distribution of unmethylated Alu elements on a genomic scale, we performed Monte Carlo simulations taking into account the observed and expected distribution of hits in each Alu family and CpG islands and the rest of sequences (see Material and Methods section). We estimate that at least 4104 Alu elements are unmethylated or partially unmethylated in normal colonic mucosa. This corresponds to 2.64% of all Alu elements containing the target sequence AACCCGGG (Table 1). Although AluS and AluY represent the majority of these sequences it should be noted that the methylation pressure is inverse to the conservation of the SmaI site. That is, the most conserved and younger AluY family shows the lowest relative rate of unmethylation; and the older and more degenerated AluJ family exhibits the highest unmethylation. Application of AUMA to detect differential DNA Figure 6. Distribution of hypermethylation and hypomethylation rates methylation in colorectal carcinomas in the 110 AUMA tagged bands. Rates were obtained by comparison of the AUMA fingerprints obtained in 50 colorectal tumors as In order to test the usefulness of the method for the compared to their respective matched normal tissue. detection of new altered methylation targets, we applied AUMA to a series of 50 colorectal carcinomas and their paired normal mucosa. Two cases were excluded from the analysis due to recurring experimental failure of the normal tissue. The average tumor showed 19 7 (range normal or tumor tissue DNA. For the rest of 48 normal– 6–37) hypomethylations and 22 10 (range 1–39) hyper- tumor pairs, consistent and fully readable fingerprints methylations. It is of note that hypomethylations could were generated and evaluated for normal–tumor differ- either be seen as an increase in the intensity of a pre-existing ential representation. A given case presented, on average, band in the normal tissue or as the appearance of an non- 107 2.9 informative bands (range 98–110). The variation existent band in the normal tissue. This contrasts with was due to polymorphic display or variable resolution hypermethylation events, which rarely showed the com- power of gel electrophoresis. plete loss of a band in the tumor sample, most likely due to In this study, only those bands showing clear intensity the unavoidable contamination of normal tissue. differences between normal and tumor tissue fingerprints Virtually, all tagged bands (109 out of 110) were found (Figure 2) have been scored as methylation changes since to be altered in at least one tumor when compared to its they are more likely to reflect tumor-wide alterations. normal paired mucosa. AUMA tagged bands presented a Because the fingerprints represent sequences flanked by wide distribution in the hypomethylation/hypermethyla- two unmethylated sites, a decreased intensity in a given tion rates (proportion of tumors showing differential band in the tumor in regard to the paired normal tissue display compared to the paired normal tissue) (Figure 6). is indicative of hypermethylation, while an increased Hypomethylation and hypermethylation showed a strong intensity corresponds to hypomethylation (Figure 2). negative correlation (r =0.55 and P< 0.0001), indicat- All tumors displayed changes in regard to the paired ing that most bands tended to be either hypomethylated Nucleic Acids Research, 2008, Vol. 36, No. 3 779 or hypermethylated. A large proportion of tagged bands tissues as well as tumors 53 and 99 showed heavy methy- (78 bands) were recurrently altered in over 25% of the lation of this region (Figure 7C). In contrast to this and cases included in this series. in agreement with AUMA results, tumors 17, 63 and In order to determine whether normal–tumor differences 74 exhibited hypomethylation at most CpGs. Cell lines were limited to isolated independent loci or changes that showed variable profiles of DNA methylation, with might affect larger chromosomal regions, we compared the CaCo exhibiting unmethylation of the MLT1A element distribution of AUMA products generated from two paired but heavy methylation of the AluYd3 element, which normal and tumor tissues and hybridized to BAC arrays. was also heavily methylated in HCT116 cells but not in the Differential hybridization was observed in many BACs, rest of the cell lines tested. MYOM2 expression levels suggesting that relatively large regions encompassing from analyzed by real-time RT-PCR were not affected by the several hundred Kbs to a few Mbs may undergo concurrent methylation status of this sequence (data not shown). hypomethylation or hypermethylation. Telomeric regions Further 45 normal–tumor pairs were analyzed for methy- of many chromosomes contained most of the differential lation of the AluYd3 element by real-time dissociation display (Figure 5C). The differential methylation profiles analysis (Supplementary Figure 7) and it was found were unaffected by chromosomal dosage as demonstrated hypomethylated in 26 tumors (58%). by its independence of chromosomal losses and gains (as Next, we wondered whether the DNA methylation detected by Comparative Genomic Hybridization (CGH) status of the AluYd3 element was associated with alter- (Supplementary Figure 5). native chromatin states. We performed Chromatin ImmunoPrecipitation (ChIP) analysis of histone 3 (H3) Validation of methylation changes detected by AUMA modifications indicative of active chromatin: acetylation of lysines 9 and 14 (AcH3K9/K14), and dimethylation To confirm that the changes observed in AUMA finger- of lysine 79 (2mH3K79); and silent chromatin: trimethy- prints corresponded to actual changes in the methylation lation of lysine 9 (3mH3K9). These histone marks were status of the sequence, eight different sequences obtained compared between cell lines HCT116 and LoVo (with from AUMA fingerprints were analyzed in normal and 100% and 30% methylation of the AluY element, respec- tumor tissues by direct sequencing of sodium bisulfite- tively). The silencing mark 3mH3K9 was 3.5-fold higher in treated DNAs (Table 2). Moreover, it was demonstrated HCT116 cells compared to the LoVo cell line (Figure 7D). that methylation changes affected not only the CpG in at No differences in active marks were observed and these least one of the two flanking SmaI sites (whose methyla- were significantly lower than the silencing mark 3mH3K9. tion prevents AUMA representation) but also neighboring When HCT116 cells were treated with the demethy- CpGs (Supplementary Figure 6). In two samples, hyper- lating agent 5-aza-2 -deoxycytidine (5AzaC) and the methylations/hypomethylations detected by AUMA could inhibitor of histone deacetylase trichostatin A (TSA), not be confirmed by bisulfite sequencing, suggesting that a moderate decrease in the amount of the 3mH3K9 mark the change could affect only a small fraction of tumor cells was observed (Figure 7E). As a whole, these data suggest and that both methods may exhibit different sensitivities. that DNA methylation changes in this AluYd3 element The presence of minor subpopulations can be detected are accompanied by alternated chromatin states. The using more sensitive techniques, i.e. the Methylation molecular consequences of such epigenetic changes remain Specific PCR or by sequencing of multiples clones. to be identified. Functional implications of changes detected by AUMA Next, we wondered if DNA methylation changes detected DISCUSSION by AUMA may have any functional consequences. We Epigenetic states of Alu elements chose one of the most recurrent hypomethylated AUMA sequences (Aq3) and performed an insightful epigenetic Full genome sequencing has provided precise maps characterization of the region in a series of normal–tumor of repetitive elements, and several studies have investigated pairs and in colon cancer cell lines. their distribution and relationship with genome structure Aq3 band is recurrently hypomethylated in tumors (48–51). More recently, a few studies have explored according to AUMA fingerprints (Figure 7A). It repre- sequence-dependent associations between repetitive ele- sents a sequence situated in the eighth intron of the ments and the epigenetic landscape. There is a character- MYOM2 gene (Table 2) and does not fall inside or close to istic distribution of interspersed elements along methylated any CpG island. The SmaI sites are located in a MLT1A and unmethylated domains, with most elements in the repeat and an AluYd3 element. The methylation status the methylated compartment of the genome (21). Nevertheless, two flanking regions of the AUMA band (465 bp and SINEs, which include Alu elements, are the repetitive 213 bp long spanning 20 and 11 CpGs, respectively) was sequences most commonly found in unmethylated domains analyzed by bisulfite direct sequencing (Figure 7B). (21) and some Alu elements may contain discriminatory Confirmation of AUMA data was performed in three motifs associated with methylation-resistant CpG islands normal–tumor pairs exhibiting differential display of the (52). Somatic cells show unstable epigenetic profiles in Aq3 band in AUMA fingerprints (cases 17, 63 and 74) repetitive elements as demonstrated by global measure- and two cases lacking this band in both normal and tumor ments of either DNA methylation (18,20,40,41) or histone pair (cases 53 and 99) (Figure 7A), as well as five cell lines modifications (53,54). Recent studies have revealed (HCT116, DLD-1, LoVo, HT29 and CaCo2). All normal interindividual variability in DNA methylation profiles 780 Nucleic Acids Research, 2008, Vol. 36, No. 3 Figure 7. (A) Detail of the AUMA fingerprints generated from five normal–tumor sample pairs. The presence of the Aq3 band is indicated by an asterisk under the three Aq3 positive cases. (B) The relative position of the AUMA Aq3 band, MLT1A and Alu Y repetitive elements, as well as MYOM2 ninth exon are shown. Each vertical line in the CpG distribution represents a CpG dinucleotide along the DNA sequence. Two different 0 0 0 fragments were amplified for the bisulfite sequencing analysis (gray boxes). Sequence is oriented 5 –3 in regard to MYOM2 3 end. (C) Methylation status of the CpG nucleotides in the two fragments amplified were ascertained by direct sequencing of bisulfite-treated DNAs of 5 normal–tumor pairs and 5 colon cancer cell lines. (D) ChIP analysis of the AluY element frequently hypomethylated in cancer revealed loss of trimethylation in histone 3 lysine 9 residue (3mH3K9) in LoVo cells (unmethylated at DNA level) as compared to HCT116 (methylated at DNA level). Treatment of HCT116 cells with 5AzaC and TSA produced a moderate decrease in the levels of trimethylation in H3K9. at specific Alu elements (55), and Fraga and colleagues Beyond these few studies, the extension and nature detected epigenetic changes arising during the lifetime of the epigenetic state of interspersed elements is largely of monozygotic twins in Alu elements and other unknown. Global estimates of DNA methylation in sequences (56). repetitive elements have been obtained by Southern blot Nucleic Acids Research, 2008, Vol. 36, No. 3 781 analyses (30) and, more recently, by using approaches in AUMA, with only 9% of the Alu elements of the old J based on bisulfite conversion of the unmethylated cytosine subfamily containing the SmaI site retain the AA dinu- (40,41,57). These studies have confirmed the global cleotide needed for their amplification, while this figure is hypomethylation of most tumors but they do not provide 91 and 80% in the younger AluY and AluS subfamilies, detailed information on the nature and localization of the respectively (Table 1), making clear that younger Alu elements tend to retain the SmaI site nearly as much as unmethylated elements. In silico analysis has revealed that they retain the AA dinucleotide required for their ampli- a number of Alu elements close to CpG islands retain a fication. This bias is not a handicap, since unmethylated high proportion of CpG sites, and this is presumed to be Alu sequences revealed by AUMA are likely to represent a sign of unmethylation (58), but no experimental proof the most relevant events of this kind, because spurious has been provided. In our point of view, the lack of unmethylation of old Alu elements retaining a single or a simple, specific and sensitive methodologies to screen for few CpG sites is expected to have less biological signif- epigenetic changes in repetitive elements on the genomic icance than unmethylation of younger Alu elements that scale has precluded a clearer understanding of the nature are usually closer to active chromatin regions (21) and and implications of these sequences in cell biology. retain more CpGs. The stronger methylation pressure observed in the AluY class is consistent with this Properties of QUMA and AUMA postulate. Here we report a systematic screening of unmethylated AUMA was designed to amplify DNA fragments Alus as a tool to determine the extent of DNA hypo- containing the target sequence (AACCCGGG), which is methylation, to identify specifically unmethylated ele- present in Alu and other repetitive elements. Because a ments and to detect epigenetic alterations in cancer cells. single primer was used for PCR amplification, the target QUMA is a very simple and specific method and provides sequence must appear in both strands of the DNA at accurate relative estimates of the number of unmethylated relatively nearby positions. As expected, Alu elements, elements. QUMA is specially appropriate for compara- with more than one million copies per human genome tive studies, but also provides a raw quantitation of the (15), were the most frequent repeat in AUMA bands number of unmethylated elements per haploid genome, (50% in sequences isolated from non-tumor tissue), but outlining the extent of hypomethylated Alu’s in normal only two sequenced bands contained two inverted Alu and pathologic cells. QUMA analysis indicates that about repeats (Supplementary Table 1). This observation is 1 out of 6 Alu elements containing the AACCCGGG in concordance with previous works reporting on the site are unmethylated, while in tumors, this figure nearly instability of this inverted repeats, which might have doubles in agreement with previous studies (23). Although caused their exclusion from the human genome (61,62). these analyses are likely to generate good estimates at the More restrictive conditions to select for Alu, or any other comparative level (between samples), absolute values repeat of interest, may be achieved by extending the 3 end should be treated with caution because the determination of the primer specific sequence (see Supplementary Data); refers to a single CpG site within the Alu element. however, the number of sequences we obtained was consi- To date there is still a lack of proper methodologies dered appropriate to accomplish the original aim of the allowing genome-wide screenings for recurrent hypo- study which is to screen for differentially methylated methylated regions that may have some impact on repetitive elements in colorectal cancer. tumor biology. Even though QUMA and other method- It is worth noting that AUMA patterns are highly ologies (41,59) allow quantitation of unmethylated reproducible not only in replicates but also among differ- repeats, they do not provide a straightforward approach ent samples, which indicates that the unmethylated status to identify and map the amplified targets. At this point, of these repeats is tightly controlled, probably by the AUMA takes us a step further, allowing the undoubtful epigenetic status of nearby regions. This is strengthened by identification of hypomethylated sequences, in addition the confirmation that unmethylation extends many CpG to hypermethylated targets. Although AUMA is specially sites beyond the SmaI cut site. Moreover, about 50% of suited to determine the nature of the unmethylated the bands tagged in AUMA fingerprints exhibited variable elements, it also allows the calculation of global unmethy- display among normal tissues (data not shown), suggest- lation in Alu elements. Nevertheless, it should be taken ing the usefulness of this technique to investigate epigene- into account that this is an indirect measure, because it tic polymorphisms. relies in the extent of methylation in CpG islands and Alu’s and other repetitive elements tend to be highly other sequences. Moreover, unmethylation of a second methylated in most somatic tissues (8,9,40,63). Here we SmaI site near the Alu is also required to generate the have identified 78 ‘atypical’ Alu elements exhibiting AUMA band and hence to be detected. While AUMA full or partial unmethylation in normal colonic mucosa shares many technical steps with other techniques, namely cells. Different evidences underscore the adequacy of this MCA (60) and AIMS (43), its design is conceptually approach to track changes with possible functional impli- unique since AUMA scans for the atypically unmethy- cations: (i) a significant portion of the characterized bands lated Alu sequences, unlike the other approaches that are are located inside or nearby CpG islands and genes; enriched for typical methylated sequences. (ii) AUMA products show a characteristic distribution Due to sequence degeneration, both QUMA and in R bands, coincidentally with the distribution of Alu AUMA are more effective in screening for unmethylation sequences (15), indicating a bias toward the gene-richest in younger elements. This trend is more clearly seen portion of the genome known as the H3 isochore (64). 782 Nucleic Acids Research, 2008, Vol. 36, No. 3 DNA methylation along Alu families Aq3 band is flanked by two repeats, a LTR and an AluY, which map within an intron between exons 8 and Alu families showed striking differences in their methyla- 9 of the MYOM2 gene at 8p23.3. Both elements are tion level. Most of the Alu elements characterized here are heavily methylated in normal tissue and partially to fully from the younger families AluS and AluY (74% and 22%, unmethylated in tumor tissue. Interestingly, we have respectively). Nevertheless, this observation is mainly due found moderately high levels of the heterochromatin to the depletion of CpG sites in older Alu elements. associated mark 3mH3K9 in the fully methylated AluYd3 Hence, only 1 out 230 AluJ elements maintains the element in the HCT116 cell line, while the levels were AUMA target site (AACCCGGG), while younger ele- significantly lower (3.5-fold) in the partially demethylated ments show higher rates of maintenance in accordance LoVo cell line. Furthermore, none of the classical active with their age (AluS: 1 out of 7; AluY: 2 out of 5) marks AcH3K9/K14 and 2mH3K79, have been found (Table 1). Interestingly, the rate of unmethylation is higher enriched in the LoVo cell line. These data are in concor- in older elements (AluJ: 12.2%; AluS: 3.1%; AluY: 1.6%) dance with preliminary data showing that the hypomethy- (Table 1). As noted by Rollins et al. (21), the boundaries lation does not affect the expression of MYOM2 (data not of unmethylated domains tend to be occupied by methy- shown) but could rather affect chromatin structure in the lated Alu transposons of the younger AluS and AluY region. In agreement with these observations, this genomic families. Other studies have noted that CpG island- region undergoes frequent losses (68–70) and is rearranged associated Alu’s retain a higher proportion of CpG sites, in many different types of tumors (71,72), which hints at suggesting that these elements are unmethylated in the a role for DNA hypomethylation in genomic instability germ line (58). These unmethylated elements that can be (24–27,73). The specific functional consequences of this easily revealed by AUMA are likely to play a regulatory hypomethylation deserve further investigations. role in a significant number of genes (65). AUMA of Another application of AUMA is the detection of tumors was enriched in Alu sequences as compared with genomic regions that have been silenced in cancer. Inter- normal tissue (41% versus 29%, respectively), suggesting spersed elements are concentrated in gene-rich regions and a favored hypomethylation of Alu elements within the due to the intended selection of unmethylated repetitive global genomic hypomethylation associated with tumor- elements in AUMA, it appears reasonable to postulate igenesis (23). that normally unmethylated sequences are likely to pin- point active genomic regions. In this context, AUMA Application of AUMA to detect epigenetic changes provides a large collection of genomic regions undergoing in cancer cells hypermethylation, which are readily seen as bands recur- Cancer-related hypomethylation is well documented (23) rently loss in the fingerprints. DNA methylation asso- and different studies have demonstrated the demethylation ciated epigenetic silencing is probably one of the most of Alu’s and other repetitive elements in different types prevalent mechanisms of tumor suppression inactivation of neoplasias (66). Our data are consistent with previous in cancer (33,37,74). Therefore, AUMA can also be used estimates and go one step forward in the characterization to screen for differential methylation not only in Alu of unmethylated repeats. The AUMA approach was elements but also in unique sequences and repetitive conceived as a straightforward DNA methylation screen- elements other than Alu. ing strategy targeting specific interspersed repeats and In summary, QUMA and AUMA methodologies are suitable to be applied to large series of samples or a simple and novel approach to explore and gain insights experimental conditions as reported here. into the functional significance of interspersed genomic The application of AUMA to a series of colorectal elements and neighboring sequences. Due to its distinctive carcinomas and their paired matched normal tissue has features (bias for unmethylated elements in gene-rich revealed a high rate of alterations. This indicates the regions and detection of both hypomethylation and hyper- plasticity of epigenetic control of the elements screened methylation) we think that these techniques constitute by AUMA in colorectal carcinogenesis. Although some a new and unique tool that should complement global bands show bidirectional changes (hypomethylations and determinations and high-resolution genome-wide scanning hypermethylations), which have been also reported in strategies. Beyond unmethylated repetitive elements, other sequences (42,67), most bands display an alternative AUMA can be also used to detect recurrent epigenetic trend either toward hypermethylations or hypomethyla- changes associated with tumorigenesis including gene tions. Some of these changes are highly recurrent (in more epigenetic inactivation. than 50% of tumors), suggesting that they may represent relevant alterations related to mechanisms frequently disturbed in colon cancer. Because the default status of SUPPLEMENTARY DATA repetitive elements is methylation, hypomethylations Supplementary data are available at NAR Online. are readily detected as the emergence of a new band in the tumor AUMA fingerprint. Since all amplified bands include a unique sequence, it has been possible to identify ACKNOWLEDGEMENTS all of the isolated bands and pinpoint them in the genomic map. We thank Gemma Aiza for technical support and Jessica As an example, we have investigated the Aq3 sequence, Halow for critical review of the manuscript. J.R. was one of the most recurrent hypomethylations in this study. a fellow of the Generalitat de Catalunya; E.V. was a fellow Nucleic Acids Research, 2008, Vol. 36, No. 3 783 23. Ehrlich,M. (2002) DNA methylation in cancer: too much, but also of the Fondo de Investigacio´ n Sanitaria (FIS). This work too little. Oncogene, 21, 5400–5413. was supported by a grant from the Ministry of Education 24. Eden,A., Gaudet,F., Waghmare,A. and Jaenisch,R. (2003) and Science (SAF2006/351) and the Consolider-Ingenio Chromosomal instability and tumors promoted by DNA hypo- 2010 Program (CSD2006-49). Funding to pay the Open methylation. Science, 300, 455. 25. Chen,R.Z., Pettersson,U., Beard,C., Jackson-Grusby,L. and Access publication charges for this article was provided Jaenisch,R. (1998) DNA hypomethylation leads to elevated by the Institute of Predictive and Personalized Medicine of mutation rates. Nature, 395, 89–93. Cancer (IMPPC). 26. Suzuki,K., Suzuki,I., Leodolter,A., Alonso,S., Horiuchi,S., Yamashita,K. and Perucho,M. (2006) Global DNA demethylation Conflict of interest statement. None declared. in gastrointestinal cancer is age dependent and precedes genomic damage. Cancer Cell., 9, 199–207. 27. Rodriguez,J., Frigola,J., Vendrell,E., Risques,R.A., Fraga,M.F., REFERENCES Morales,C., Moreno,V., Esteller,M., Capella,G. et al. (2006) Chromosomal instability correlates with genome-wide DNA 1. Mattick,J.S. (2003) Challenging the dogma: the hidden layer of demethylation in human primary colorectal cancers. Cancer Res., non-protein-coding RNAs in complex organisms. Bioessays, 25, 66, 8462–9468. 930–939. 28. Liu,W.M., Maraia,R.J., Rubin,C.M. and Schmid,C.W. (1994) 2. Zuckerkandl,E. (2002) Why so many noncoding nucleotides? The Alu transcripts: cytoplasmic localisation and regulation by DNA eukaryote genome as an epigenetic machine. Genetica, 115, 105–129. methylation. Nucleic Acids Res., 22, 1087–1095. 3. Kreahling,J. and Graveley,B.R. (2004) The origins and implications 29. Roman-Gomez,J., Jimenez-Velasco,A., Agirre,X., Cervantes,F., of Aluternative splicing. Trends Genet., 20, 1–4. Sanchez,J., Garate,L., Barrios,M., Castillejo,J.A., Navarro,G. et al. 4. Jasinska,A. and Krzyzosiak,W.J. (2004) Repetitive sequences that (2005) Promoter hypomethylation of the LINE-1 retrotransposable shape the human transcriptome. FEBS Lett., 567, 136–141. elements activates sense/antisense transcription and marks the 5. Jordan,I.K., Rogozin,I.B., Glazko,G.V. and Koonin,E.V. (2003) progression of chronic myeloid leukemia. Oncogene, 24, 7213–7223. Origin of a substantial fraction of human regulatory sequences from 30. Ehrlich,M. (2006) Cancer-linked DNA hypomethylation and its transposable elements. Trends Genet., 19, 68–72. relationship to hypermethylation. Curr. Top. Microbiol. Immunol., 6. Hasler,J. and Strub,K. (2006) Alu elements as regulators of gene 310, 251–274. expression. Nucleic Acids Res., 34, 5491–5497. 31. Plass,C. (2002) Cancer epigenomics. Hum. Mol. Genet., 11, 7. Slotkin,R.K. and Martienssen,R. (2007) Transposable elements and 2479–2488. the epigenetic regulation of the genome. Nat. Rev. Genet., 8, 32. Jones,P.A. and Baylin,S.B. (2002) The fundamental role of 272–285. epigenetic events in cancer. Nat. Rev. Genet., 3, 415–428. 8. Schmid,C.W. (1998) Does SINE evolution preclude Alu function? 33. Herman,J.G. and Baylin,S.B. (2003) Gene silencing in cancer in Nucleic Acids Res., 26, 4541–4550. association with promoter hypermethylation. N. Engl. J. Med., 349, 9. Yoder,J.A., Walsh,C.P. and Bestor,T.H. (1997) Cytosine methyla- 2042–2054. tion and the ecology of intragenomic parasites. Trends Genet., 13, 34. Lund,A.H. and van Lohuizen,M. (2004) Epigenetics and cancer. 335–340. Genes Dev., 18, 2315–2335. 10. Jaenisch,R. and Bird,A. (2003) Epigenetic regulation of gene 35. Feinberg,A.P. and Tycko,B. (2004) The history of cancer epige- expression: how the genome integrates intrinsic and environmental netics. Nat. Rev. Cancer, 4, 143–153. signals. Nat. Genet., 33(Suppl.), 245–254. 36. Laird,P.W. (2005) Cancer epigenetics. Hum. Mol. Genet., 14, 11. Weissmann,F. and Lyko,F. (2003) Cooperative interactions between R65–R76. epigenetic modifications and their function in the regulation of 37. Esteller,M. (2007) Cancer epigenomics: DNA methylomes and chromosome architecture. Bioessays, 25, 792–797. histone-modification maps. Nat. Rev. Genet., 6,6. 12. Goll,M.G. and Bestor,T.H. (2005) Eukaryotic cytosine methyl- 38. Laird,P.W. (2003) The power and the promise of DNA methylation transferases. Annu. Rev. Biochem., 74, 481–514. markers. Nat. Rev. Cancer, 3, 253–266. 13. Fazzari,M.J. and Greally,J.M. (2004) Epigenomics: beyond CpG 39. Cottrell,S.E. (2004) Molecular diagnostic applications of DNA islands. Nat. Rev. Genet., 5, 446–455. methylation technology. Clin. Biochem., 37, 595–604. 14. Lander,E.S., Linton,L.M., Birren,B., Nusbaum,C., Zody,M.C., 40. Yang,A.S., Estecio,M.R., Doshi,K., Kondo,Y., Tajara,E.H. and Baldwin,J., Devon,K., Dewar,K., Doyle,M. et al. (2001) Initial Issa,J.P. (2004) A simple method for estimating global DNA sequencing and analysis of the human genome. Nature, 409, methylation using bisulfite PCR of repetitive DNA elements. 860–921. Nucleic Acids Res., 32, e38. 15. Batzer,M.A. and Deininger,P.L. (2002) Alu repeats and human 41. Weisenberger,D.J., Campan,M., Long,T.I., Kim,M., Woods,C., genomic diversity. Nat. Rev. Genet., 3, 370–379. Fiala,E., Ehrlich,M. and Laird,P.W. (2005) Analysis of repetitive 16. Korenberg,J.R. and Rykowski,M.C. (1988) Human genome element DNA methylation by MethyLight. Nucleic Acids Res., 33, organization: Alu, lines, and the molecular structure of metaphase 6823–6836. chromosome bands. Cell, 53, 391–400. 42. Frigola,J., Sole,X., Paz,M.F., Moreno,V., Esteller,M., Capella,G. 17. Chen,C., Gentles,A.J., Jurka,J. and Karlin,S. (2002) Genes, and Peinado,M.A. (2005) Differential DNA hypermethylation and pseudogenes, and Alu sequence organization across hypomethylation signatures in colorectal cancer. Hum. Mol. Genet., human chromosomes 21 and 22. Proc. Natl Acad. Sci. USA, 99, 14, 319–326. 2930–2935. 43. Frigola,J., Ribas,M., Risques,R.A. and Peinado,M.A. (2002) 18. Schmid,C.W. (1991) Human Alu subfamilies and their Methylome profiling of cancer cells by amplification of inter- methylation revealed by blot hybridization. Nucleic Acids Res., 19, methylated sites (AIMS). Nucleic Acids Res., 30, e28. 5613–5617. 44. Vendrell,E., Ribas,M., Valls,J., Sole,X., Grau,M., Moreno,V., 19. Gama-Sosa,M.A., Wang,R.Y., Kuo,K.C., Gehrke,C.W. and Capella,G. and Peinado,M.A. (2007) Genomic and transcriptomic Ehrlich,M. (1983) The 5-methylcytosine content of highly prognostic factors in R0 Dukes B and C colorectal cancer patients. repeated sequences in human DNA. Nucleic Acids Res., 11, 3087–3095. Int. J. Oncol., 30, 1099–1107. 20. Kochanek,S., Renz,D. and Doerfler,W. (1993) DNA methylation in 45. Perucho,M., Welsh,J., Peinado,M.A., Ionov,Y. and McClelland,M. the Alu sequences of diploid and haploid primary human cells. (1995) Fingerprinting of DNA and RNA by arbitrarily primed EMBO J., 12, 1141–1151. polymerase chain reaction: applications in cancer research. Methods 21. Rollins,R.A., Haghighi,F., Edwards,J.R., Das,R., Zhang,M.Q., Enzymol., 254, 275–290. Ju,J. and Bestor,T.H. (2006) Large-scale structure of genomic 46. Stirzaker,C., Song,J.Z., Davidson,B. and Clark,S.J. (2004) methylation patterns. Genome Res., 16, 157–163. Transcriptional gene silencing promotes DNA hypermethylation 22. Bird,A. (2002) DNA methylation patterns and epigenetic memory. through a sequential change in chromatin modifications in cancer Genes Dev., 16, 6–21. cells. Cancer Res., 64, 3871–3877. 784 Nucleic Acids Research, 2008, Vol. 36, No. 3 47. Xing,J., Hedges,D.J., Han,K., Wang,H., Cordaux,R. and 61. Lobachev,K.S., Stenger,J.E., Kozyreva,O.G., Jurka,J., Gordenin,D.A. and Resnick,M.A. (2000) Inverted Alu repeats Batzer,M.A. (2004) Alu element mutation spectra: Molecular clocks unstable in yeast are excluded from the human genome. EMBO J., and the effect of DNA methylation. J. Mol. Biol, 344, 675–682. 19, 3822–3830. 48. Medstrand,P., van de Lagemaat,L.N. and Mager,D.L. (2002) 62. Stenger,J.E., Lobachev,K.S., Gordenin,D., Darden,T.A., Jurka,J. Retroelement distributions in the human genome: variations and Resnick,M.A. (2001) Biased distribution of inverted and direct associated with age and proximity to genes. Genome Res., 12, Alus in the human genome: implications for insertion, exclusion, 1483–1495. and genome stability. Genome Res., 11, 12–27. 49. Grover,D., Majumder,P.P., Rao,C.B., Brahmachari,S.K. and 63. Hellmann-Blumberg,U., Hintz,M.F., Gatewood,J.M. and Mukerji,M. (2003) Nonrandom distribution of alu elements in genes Schmid,C.W. (1993) Developmental differences in methylation of of various functional categories: insight from analysis of human human Alu repeats. Mol. Cell. Biol., 13, 4523–4530. chromosomes 21 and 22. Mol. Biol. Evol., 20, 1420–1424. 64. Saccone,S., Caccio,S., Kusuda,J., Andreozzi,L. and Bernardi,G. 50. Grover,D., Mukerji,M., Bhatnagar,P., Kannan,K. and (1996) Identification of the gene-richest bands in human chromo- Brahmachari,S.K. (2004) Alu repeat analysis in the complete human somes. Gene, 174, 85–94. genome: trends and variations with respect to genomic composition. 65. Oei,S.L., Babich,V.S., Kazakov,V.I., Usmanova,N.M., Bioinformatics, 20, 813–817. Kropotov,A.V. and Tomilin,N.V. (2004) Clusters of regulatory 51. Price,A.L., Eskin,E. and Pevzner,P.A. (2004) Whole-genome signals for RNA polymerase II transcription associated with Alu analysis of Alu repeat elements reveals complex evolutionary family repeats and CpG islands in human promoters. Genomics, 83, history. Genome Res., 14, 2245–2252. 873–882. 52. Feltus,F.A., Lee,E.K., Costello,J.F., Plass,C. and Vertino,P.M. 66. Wilson,A.S., Power,B.E. and Molloy,P.L. (2007) DNA hypo- (2006) DNA motifs associated with aberrant CpG island methyla- methylation and human diseases. Biochim. Biophys. Acta, 1775, tion. Genomics, 15, 15. 138–162. 53. Fraga,M.F., Ballestar,E., Villar-Garea,A., Boix-Chornet,M., 67. Nishiyama,R., Qi,L., Tsumagari,K., Weissbecker,K., Dubeau,L., Espada,J., Schotta,G., Bonaldi,T., Haydon,C., Ropero,S. et al. Champagne,M., Sikka,S., Nagai,H. and Ehrlich,M. (2005) A DNA (2005) Loss of acetylation at Lys16 and trimethylation at Lys20 of Repeat, NBL2, Is Hypermethylated in Some Cancers but histone H4 is a common hallmark of human cancer. Nat. Genet., 13, Hypomethylated in Others. Cancer Biol. Ther., 4,4. 68. Emi,M., Fujiwara,Y., Nakajima,T., Tsuchiya,E., Tsuda,H., 54. Kondo,Y. and Issa,J.P. (2003) Enrichment for histone H3 lysine 9 Hirohashi,S., Maeda,Y., Tsuruta,K., Miyaki,M. et al. (1992) methylation at Alu repeats in human cells. J. Biol. Chem., 278, Frequent loss of heterozygosity for loci on chromosome 8p in 27658–27662. hepatocellular carcinoma, colorectal cancer, and lung cancer. 55. Sandovici,I., Kassovska-Bratinova,S., Loredo-Osti,J.C., Leppert,M., Cancer Res., 52, 5368–5372. Suarez,A., Stewart,R., Bautista,F.D., Schiraldi,M. and Sapienza,C. 69. Sunwoo,J.B., Sun,P.C., Gupta,V.K., Schmidt,A.P., El-Mofty,S. and (2005) Interindividual variability and parent of origin DNA Scholnick,S.B. (1999) Localization of a putative tumor suppressor methylation differences at specific human Alu elements. Hum. Mol. gene in the sub-telomeric region of chromosome 8p. Oncogene, 18, Genet., 14, 2135–2143. 2651–2655. 56. Fraga,M.F., Ballestar,E., Paz,M.F., Ropero,S., Setien,F., 70. Muscheck,M., Sukosd,F., Pesti,T. and Kovacs,G. (2000) High Ballestar,M.L., Heine-Suner,D., Cigudosa,J.C., Urioste,M. et al. density deletion mapping of bladder cancer localizes the putative (2005) Epigenetic differences arise during the lifetime of mono- tumor suppressor gene between loci D8S504 and D8S264 at zygotic twins. Proc. Natl Acad. Sci. USA, 102, 10604–10609. chromosome 8p23.3. Lab Invest., 80, 1089–1093. 57. Xiong,Z. and Laird,P.W. (1997) COBRA: a sensitive and quanti- 71. Wong,N., Wong,K.F., Chan,J.K. and Johnson,P.J. (2000) tative DNA methylation assay. Nucleic Acids Res., 25, 2532–2534. Chromosomal translocations are common in natural killer-cell 58. Brohede,J. and Rand,K.N. (2006) Evolutionary evidence suggests lymphoma/leukemia as shown by spectral karyotyping. Hum. that CpG island-associated Alus are frequently unmethylated in Pathol., 31, 771–774. human germline. Hum. Genet., 119, 457–458. 72. Jin,Y., Jin,C., Wennerberg,J., Hoglund,M. and Mertens,F. (2001) 59. Choi,I.S., Estecio,M.R., Nagano,Y., Kim do,H., White,J.A., Cytogenetic and fluorescence in situ hybridization characterization Yao,J.C., Issa,J.P. and Rashid,A. (2007) Hypomethylation of of chromosome 8 rearrangements in head and neck squamous cell LINE-1 and Alu in well-differentiated neuroendocrine tumors carcinomas. Cancer Genet. Cytogenet., 130, 111–117. (pancreatic endocrine tumors and carcinoid tumors). Mod. Pathol., 73. Karpf,A.R. and Matsui,S. (2005) Genetic disruption of cytosine 20, 802–810. DNA methyltransferase enzymes induces chromosomal instability in 60. Toyota,M., Ho,C., Ahuja,N., Jair,K.W., Li,Q., Ohe-Toyota,M., human cancer cells. Cancer Res., 65, 8635–8639. Baylin,S.B. and Issa,J.P. (1999) Identification of differentially 74. Egger,G., Liang,G., Aparicio,A. and Jones,P.A. (2004) Epigenetics methylated sequences in colorectal cancer by methylated CpG in human disease and prospects for epigenetic therapy. Nature, 429, island amplification. Cancer Res., 59, 2307–2312. 457–463. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Nucleic Acids Research Oxford University Press

Genome-wide tracking of unmethylated DNA Alu repeats in normal and cancer cells

Download PDF
15 pages

Loading next page...
 
/lp/oxford-university-press/genome-wide-tracking-of-unmethylated-dna-alu-repeats-in-normal-and-t3RUm2j105

References (81)

Publisher
Oxford University Press
Copyright
© 2007 The Author(s)
ISSN
0305-1048
eISSN
1362-4962
DOI
10.1093/nar/gkm1105
pmid
18084025
Publisher site
See Article on Publisher Site

Abstract

770–784 Nucleic Acids Research, 2008, Vol. 36, No. 3 Published online 15 December 2007 doi:10.1093/nar/gkm1105 Genome-wide tracking of unmethylated DNA Alu repeats in normal and cancer cells 1 2,3 1 1 2,3 Jairo Rodriguez , Laura Vives , Mireia Jorda , Cristina Morales , Mar Mun˜ oz , 1 1,2, Elisenda Vendrell and Miguel A. Peinado * 1 2 ´ ` Institut d’Investigacio Biomedica de Bellvitge (IDIBELL), L’Hospitalet, Institut de Medicina Predictiva i Personalitzada del Ca` ncer (IMPPC), Badalona and Institut Catala` d’Oncologia (ICO), L’Hospitalet, Catalonia, Spain Received September 19, 2007; Revised October 19, 2007; Accepted November 27, 2007 is not just a direct outcome of the number of coding ABSTRACT sequences and that the presence of multiple regulatory Methylation of the cytosine is the most frequent mechanisms accounts for a significant part of biological epigenetic modification of DNA in mammalian cells. complexity (1,2). Among these mechanisms, repetitive In humans, most of the methylated cytosines elements may play a key role in gene regulation and geno- are found in CpG-rich sequences within tandem mic structure. Active transposable elements are involved and interspersed repeats that make up to 45% of the in genome rearrangement and illegitimate recombination and can also influence gene expression by altering splicing human genome, being Alu repeats the most common or by acting as enhancers or promoters (3–7). Advances in family. Demethylation of Alu elements occurs in the understanding of epigenetic mechanisms that regulate aging and cancer processes and has been asso- these repetitive elements may contribute to elucidate their ciated with gene reactivation and genomic instabil- specific participation in biological processes (8). ity. By targeting the unmethylated SmaI site within Silenced regions in mammals and other vertebrates are the Alu sequence as a surrogate marker, we have differentiated, although not exclusively, by the presence quantified and identified unmethylated Alu elements of DNA methylation (9). Methylation of the cytosine is on the genomic scale. Normal colon epithelial cells an epigenetic modification of DNA that plays an impor- contain in average 25 486 10 157 unmethylated tant role in the control of gene expression and chromo- Alu’s per haploid genome, while in tumor cells this some structure in mammalian cells (10–13). Most of the figure is 41 995 17 187 (P = 0.004). There is an 5-methylcytosines are found in CpG-rich sequences within inverse relationship in Alu families with respect to tandem and interspersed repeats (9,12) of which the their age and methylation status: the youngest previous estimates indicate that constitute up to 45% of the human genome (14). Among these repeats, Alu’s, with elements exhibit the highest prevalence of the SmaI more than one million copies per haploid genome, are site (AluY: 42%; AluS: 18%, AluJ: 5%) but the lower considered the most successful family (15). Interestingly, rates of unmethylation (AluY: 1.65%; AluS: 3.1%, Alu’s are not randomly distributed within the human AluJ: 12%). Data are consistent with a stronger genome, as they tend to accumulate in gene-rich regions silencing pressure on the youngest repetitive ele- (14,16,17). Previous works have estimated that Alu ele- ments, which are closer to genes. Further insights ments harbor up to 33% of the total number of CpG sites into the functional implications of atypical unmethy- in the genome (18) and have been reported to be highly lation states in Alu elements will surely contribute methylated in most somatic tissues (18–20). Methylation to decipher genomic organization and gene regula- represents the primary mechanism of transposon suppres- tion in complex organisms. sion and active transposons are demethylated in mamma- lian genomes (12). It has been proposed that regions of the genome containing repetitive elements might be masked by compartmentalization of the chromatin, resulting in INTRODUCTION a reduction of the effective size of the genome (21). Progress in large-scale sequencing projects is critical to Noteworthy, even though a vast number of CpG dinucle- identify and decipher gene organization and regulation in otides are provided by the collection of repetitive sequences many species including human. Nevertheless, cumulated in the human genome, this dinucleotide is greatly under- evidences indicate that the complexity of living organisms represented throughout the genome, but it can be found *To whom correspondence should be addressed. Tel: +34 934978693; Fax: +34 934978697; Email: [email protected] 2007 The Author(s) This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/ by-nc/2.0/uk/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited. Nucleic Acids Research, 2008, Vol. 36, No. 3 771 at close to its expected frequency in small genomic regions (200 bp to a few kb), known as CpG islands (22). These areas are ‘protected’ from methylation and are located in the proximal promoter regions of 75% of human genes (12,13,22). Methylated CpG islands are strongly and hereditably repressed (12). Hence DNA methylation is usually considered as a sign of long-term inactivation (9,10,12). Cancer cells are characterized by the accumulation of both genetic and epigenetic changes. Widespread genomic hypomethylation is an early alteration in carcinogenesis and has been associated with genomic disruption and genetic instability (23–27). Repeats unmasked by demeth- ylation are likely to facilitate rearrangements due to mito- tic recombination and unwanted transcription (28–30). Alternatively, aberrant de novo methylation of CpG islands is a hallmark of human cancers and is associated with epigenetic silencing of multiple tumor suppressor genes (31–37). Therefore, the screening for differentially methylated sequences in tumors appears as a key tool to further understand the molecular mechanisms under- lying malignant transformation of cells. Although, the repertoire of methylation screening methodologies has expanded widely (37–39), and different approaches have been used to make bulk estimates of methylation in repetitive elements (40,41), there is still a lack of screening strategies that specifically allow a feasible identification of DNA methylation alterations in repetitive elements (21). Here we report two variants of a novel methodology to quantify and identify unmethylated Alu sequences. The CpG site within the consensus Alu sequence AACCC GGG is used as a surrogate reporter of methylation. Figure 1. Schematic diagram of the QUMA and AUMA methods. DNA Unmethylated sites are cut with the methylation-sensitive is depicted by a solid line, Alu elements are represented by dashed boxes. restriction endonuclease SmaI (CCCGGG) and an adap- The QUMA and AUMA recognition sites (AACCCGGG) are repre- tor is ligated to the DNA ends. Quantification of sented by dashed/gray boxes. CpGs at SmaI sites are shown as full circles UnMethylated Alus (QUMA) is performed by real-time when methylated and as open circles when unmethylated. The methyla- tion-sensitive restriction endonuclease SmaI can only digest unmethylated amplification of the digested and adaptor-ligated DNA targets, leaving blunt ends to which adaptors can be ligated. (A) QUMA is using an Alu consensus primer that anneals upstream performed by real-time PCR of an inner Alu fragment using a primer of the SmaI site and an adaptor primer extended with the complementary to the Alu consensus sequence upstream of the SmaI site and the primer complementary to the adaptor to which two Alu TT dinucleotide in its 3 end (Figure 1A). The product homologous nucleotides (TT) have been added. (B) In AUMA, sequences generated by this approach is completely inside the flanked by two ligated adaptors are amplified by PCR using a single Alu element and hence it is not possible to make a primer, the same adaptor primer plus the TT nucleotides. When only a few unique identification. As an alternative approach, we have nucleotides are added to the primer, i.e. TT, as illustrated here, other non- also performed restrained amplification of digested and Alu sequences may be amplified. This allows the amplification of a large number of sequences that typically range from 100 to 2000 bp. adaptor-ligated DNA fragments that are flanked by two close SmaI sites. In this case, the same primer homologous to the adaptor with the additional TT nucleotides at the Application of QUMA and AUMA to a series of colo- 3 end to enrich for Alu sequences is used in absence of the rectal carcinomas and their paired normal mucosa has Alu consensus primer (Figure 1B). This second approach offered global estimates of unmethylation of Alu elements is named Amplification of UnMethylated Alu’s (AUMA) in normal and cancer cells and has revealed a large collec- and results in a complex representation of unique DNA tion of unique sequences that undergo highly recurrent sequences flanked by two unmethylated SmaI sites. When hypomethylation and hypermethylation in colorectal resolved by high-resolution electrophoresis, the AUMA tumors. generated sequences appear as a fingerprint characteristic of each sample (Figure 2) and individual scoring and identification of each band can be performed. Because MATERIALS AND METHODS AUMA’s stringency is based on a short sequence Tissues and cell lines (AACCCGGG) that is found preferentially but not exclu- sively in Alu elements, other unmethylated sequences are Fifty colorectal carcinomas and their paired non- also present in AUMA fingerprints. adjacent areas of normal colonic mucosa were included 772 Nucleic Acids Research, 2008, Vol. 36, No. 3 AUMA products performing Monte Carlo simulations. One thousand Monte Carlo simulations were performed using an Excel Add-in (available at www.wabash.edu/ econometrics). In Monte Carlo simulations, it was assumed that 80–100% of SmaI sites at CpG islands are unmethylated and that 50–100% of SmaI sites in other genomic regions different from Alu’s and CpG islands are unmethylated. Quantification of QUMA One microgram of DNA was digested with 20 U of the methylation sensitive restriction endonuclease SmaI (Roche Diagnostics GmbH, Mannheim, Germany) for 16 h at 308C, leaving cleaved fragments with blunt ends (CCC/GGG). Adaptors were prepared incubating the oligonucleotides Blue (CCGAATTCGCAAAGCTC TGA) and the 5 phosphorylated MCF oligonucleotide (TCAGAGCTTTGCGAAT) at 658C for 2 min, and then cooling to room temperature for 30–60 min. One micro- Figure 2. AUMA of normal (N)–tumor (T) pairs of two different patients performed using primer BAu-TT. A highly reproducible band gram of the digested DNA was ligated to 2 nmol of patterning is observed among the four replicates. Representative bands adaptor using T4 DNA ligase (New England Biolabs, showing gains (hypomethylations) and losses (hypermethylations) are Beverly, MA, USA). Subsequent digestion of the ligated marked with up and down arrowheads, respectively. products with the methylation insensitive restriction endonuclease XmaI (New England Biolabs) was per- formed to avoid amplifications from non-digested methy- lated Alu’s. The products were purified using the GFX in this analysis. Samples were collected simultaneously as Kit (Amersham Biosciences, Buckinghamshire, UK) and fresh specimens and snap-frozen within 2 h of removal and eluted in 250ml of sterile water. then stored at 808C. All samples were obtained from the Quantitative real-time PCR was performed using 1 ng Ciutat Sanita` ria i Universita` ria de Bellvitge (Barcelona, (the equivalent of 333 genomes) of DNA in a LightCycler Spain). The study protocol was approved by the Ethics 480 real-time PCR system with Fast Start Master SYBR Committee. Human colon cancer cell lines (HT29, SW480, Green I kit (Roche). Mastermix was prepared to a final HCT116, LoVo, DLD-1, CaCo-2 and LS174T) were concentration of 3.5 mM MgCl and 1mM of each primer. obtained from the American Type Culture Collection The downstream BAu-TT primer (constituted by the 3 (ATCC; Manassas, VA). KM12C and KM12SM cells were end of Blue primer, and the GGGTT sequence including generously provided by A. Fabra. DNA from tumor– the GGG 3 side of the cut SmaI site and the Alu normal pairs was obtained by conventional organic extrac- homologous TT dinucleotide, ATTCGCAAAGCTCTG tion and ethanol precipitation. DNA purity and quality AGGGTT) and the upstream primer was an Alu was checked in a 0.8% agarose gel electrophoresis. RNA consensus sequence (CCGTCTCTACTAAAAATACA) from cell lines was obtained by phenol–chloroform (see Supplementary Data). Magnitudes were expressed extraction and ethanol precipitation, following standard as number of unmethylated Alus per haploid genome procedures. after DNA input normalization. The number of haploid genomes present in the test tube was determined in the Bioinformatic analysis same multiwell plate by quantification of Alu sequences irrespectively of the methylation state. A real-time PCR The distribution of SmaI sites, putative amplification hits, using Alu consensus primers upstream of the CCCGGG PCR homologies, CpG islands and repetitive elements site was performed (see Supplementary Data) and the was assessed using the human genome assembly 36.1 from number of genomes was calculated against a standard NCBI. Data were obtained from the Repbase (http:// curve constructed with a reference genomic DNA mea- www.girinst.org/repbase/index.html) and the Genome Browser Databases (http://hgdownload.cse.ucsc.edu/ sured by UV spectrophotometry. To determine the efficiency of the assay and to perform goldenPath/hg18/database/). Only assembled chromo- some fragments were considered. A Perl routine was absolute quantification, an external Alu product generated used to score all positions containing the target sequences by PCR from a DNA fragment containing an AluSx in all chromosomes (available from the authors upon element was used as standard (Supplementary Methods). request). Data were analyzed using Excel spreadsheets. The number of copies of the external control were spectro- To calculate the proportion of unmethylated Alu photometrically quantified and dilution curves were gene- elements at the genomic level, the number of AUMA rated and treated as samples. Comparison of dilution hits identified in bioinformatic analysis were corrected curves before and after sample processing indicated that according to the distribution of experimentally generated the mean recovery was 73%. DNA samples overdigested Nucleic Acids Research, 2008, Vol. 36, No. 3 773 with the methylation insensitive XmaI endonuclease were analogous to CGH. Briefly, an AUMA product obtained spiked with different amounts of the external standard and from a normal tissue DNA was purified using Jet quick processed. The sensitivity of the QUMA detection was 100 PCR product purification kit (Genomed, Lohne, unmethylated Alu’s per haploid genome (Supplementary Germany) and labeled with SpectrumRed dUTP (Vysis, Figure 1) using 1 ng of genomic DNA per PCR. A linear Downers Grove, IL, USA) using a Nick Translation kit response was observed between 1000 and 100 000 unmethy- (Vysis). Similarly, genomic DNA of the same normal lated Alu’s per haploid genome (Supplementary Data). sample was labeled with SpectrumGreen dUTP (Vysis) and both probes were cohybridized to metaphase chromo- somes. Procedures and image analysis were performed as Amplification of AUMA described (44). Differential normal–tumor representation of AUMA DNA digestion with SmaI enzyme and ligation to the linker at the genomic scale was performed by competitive was performed as described above for QUMA, except for hybridization of AUMA products to BAC arrays. the XmaI digestion that was skipped. The product was AUMA products from two normal–tumor pairs were purified using the GFX Kit (Amersham Biosciences) and purified using Jet quick PCR product purification kit eluted in 250ml of sterile water. Six different chimeric (Genomed, Lo¨ hne, Germany) and 1mg was labeled with primers constituted by the 3 end of the Blue primer dCTP-Cy3 or dCTP-Cy5 (Amersham Biosciences, UK) by sequence (ATTCGCAAAGCTCTGA), the cut SmaI site use of the Bioprime DNA Labeling System (Invitrogen, (GGG) and two, four or seven additional nucleotides Carlsbad, CA, USA). Probes were hybridized to Spectral- homologous to the Alu consensus sequence were used to Chip 2600 BAC arrays (Spectral Genomics, Houston, TX, enrich for Alu sequences (see Supplementary Methods). Three primers were designed to amplify ‘upstream’ of the USA) following the manufacturer’s instructions. Arrays SmaI site (towards ALU promoter): BAu-TT, BAu-TTCA, were scanned with a ScanArray 4000 (GSI Lumonics, BAu-TTCAAGC. Three other primers were designed to Watertown, MA, USA) and processed with GenePix amplify ‘downstream’ (towards ALU poly-A): BAd-AG, software (Axon Instruments, Union City, CA, USA). The BAd-AGGC, BAd-AGGCGGA. Letters after the dash resulting data were processed to filter out low-quality correspond to the 3 sequence of the primer (see Supple- spots based on spot area and similarity of readings mentary Data). Data reported here were obtained by using between the two replicates of each BAC. Data manipula- the BAu-TT primer. tion was performed using Excel spreadsheets. Because In each PCR reaction only one primer was used at AUMA products are not evenly distributed along chro- a time. Products were resolved on denaturing sequencing mosomes, only BACs with intensities above the 10% of gels. Although bands can be visualized by silver staining maximum intensity in at least one of the two channels of the gels, radioactive AUMA’s were performed for were considered for ratio calculations. The pattern of normal–tumor comparisons. A more detailed description chromosomal alterations in these two tumors was deter- of the PCR and the visualization of the bands are given mined by conventional CGH as described (44). as Supplementary Data. Only sharp bands that were reproducible and clearly Isolation and cloning of AUMA tagged bands distinguishable from the background were tagged and DNA excised from gels was directly amplified with the included in the analysis. Faint bands with inconsistent same primer used in AUMA (BAu-TT) (Supplementary display due to small variations in gel electrophoresis Figure 2). The amplified product was cloned into plasmid resolution were not considered. Band reproducibility was vectors using the pGEM-T easy vector System I cloning assessed with the analysis of PCR duplicates of three kit (Promega, Madison, WI, USA). Automated sequenc- independent sample digests from two different samples ing of multiple colonies was performed using the Big Dye and PCR replicates from the same digest from four paired Terminator v3.1 Cycle Sequencing kit (Applied Biosys- tumor–normal samples. AUMA fingerprints were visually tems, Foster City, CA, USA) to ascertain the unique checked for methylation differences between bands in the identity of the isolated band. Sequence homologies were tumor with regard to its paired normal mucosa. Under searched for using the Blat engine (http://genome.ucsc. these premises, a given band was scored according to three edu/). Selected clones corresponding to AUMA isolated possible behaviors: hypomethylation (increased intensity bands were radioactively labeled and used as a probe to in the tumor), hypermethylation (decreased intensity in confirm the identity of the excised band by hybridization the tumor) and no change (no substantial difference in to AUMA fingerprints as previously described (45). intensity between normal and tumor samples) (Figure 2). Only those bands showing clear changes in their intensities in the fingerprint were considered to represent methylation Bisulfite genomic sequencing changes. This is consistent with previous studies done Differential methylation observed in some AUMA using a related technique (42,43). tagged bands was confirmed by direct sequencing of bisulfite treated normal and tumor DNA as previously Competitive hybridization of AUMA products to metaphase described (46). Prior to sequencing, DNA was amplified chromosomes and BAC arrays using a nested or semi-nested PCR approach, as appro- The origin and chromosomal distribution of sequences priate. Three independent PCRs were done and products generated by AUMA was analyzed using procedures were pooled to ensure a representative sequencing. 774 Nucleic Acids Research, 2008, Vol. 36, No. 3 Table 1. Content and distribution of QUMA and AUMA hits in the human genome Sequence Mb Number of SmaI sites AACCCGGG Virtual AUMA Unmethylated Unmethylated b c d e f g elements (CCCGGG) hits AUMA hits hits hits hits (%) Total 3080.4 1 118 195 486 835 168 309 5498 201 14332 2418 8.52  1.4% Alu (S+J+Y) 227.3 1 091 110 198 201 155 226 5109 59 (29.3%) 4104 688 2.64  0.44% AluS 141.2 660 415 122 459 97 951 3382 45 (22.4%) 3028 510 3.09  0.51% AluJ 54.0 283 104 14 017 1235 38 2 (1.0%) 151 25 12.25  1.97% AluY 32.1 147 591 61 725 56 040 1689 12 (6.0%) 925 156 1.65  0.27% CpG islands 16.2 27 085 49 430 1673 63 55 (27.4%) 1501 97 90.5  5.79% Rest 2836.9 – 239 204 11 410 326 87 (43.3%) 8530 1650 75.9  14.63% Genome Mb represented by each type of element. Total number corresponds to the number of megabases analyzed for the presence of hits. Only assembled chromosome fragments were considered. Elements considered in the analysis as obtained from the Repbase and the Genome Browser Databases (see Material and Methods section). Number of occurrences of the sequence AACCCGGG (or CCCGGGTT) within each type of element. Number of AUMA hits present in virtual PCR products of up to 1000 bp. Hits of actual AUMA products. Only bands appearing in normal tissue were considered. Eighty-seven bands contributed two hits each (174 hits) and 27 bands contributed only one due to poor sequence or incomplete homology with the NCBI Build 36.1 of the human genome (hg18 assembly, March 2006). Twenty-three additional bands were detected mainly in tumor tissue and were not considered to perform calculations. Estimated number of unmethylated sites using Monte Carlo simulations (Material and Methods section). In respect to the total number of AACCCGGG (or CCCGGGTT) hits. The sequence of PCR primers is described in RESULTS Supplementary Data. Genomic estimation of the targets and evaluation of the adequacy of the approach by computational analysis The availability of the human genome map has allowed Histone modification analysis by chromatin us to make a detailed estimation of the frequency and immunoprecipitation (ChIP) distribution of the sites targeted by our approaches on 6 the genomic scale. A Perl routine was used to score all Briefly, 6 10 cells were washed twice with PBS and cross- positions containing the target sequences in all chromo- linked on the culture plate for 15 min at room temperature somes and was also applied to perform a virtual AUMA in the presence of 0.5% formaldehyde. Cross-linking (see Material and Methods section). Some of the most reaction was stopped by adding 0.125 M glycine. All subse- important data derived from the bioinformatic analysis quent steps were carried out at 48C. All buffers were pre- are shown in Table 1 and Figure 3. chilled and contained protease inhibitors (Complete Mini, Because of the C to T mutational bias at CpG sites (47), Roche). Cells were washed twice with PBS and then any amplification method relying on the consensus scraped. Collected pellets were dissolved in 1 ml lysis buffer sequence (see Supplementary Data) will only cover a (1% SDS, 5 mM EDTA, 50 mM Tris pH 8) and were fraction of all the Alu’s. Therefore it is important to sonicated in a cold ethanol bath for 10 cycles at 100% estimate the degree of representativity of the methods amplitude using a UP50H sonicator (Hielscher, Teltow, used here if genome-wide estimations are to be made. Alu Germany). Chromatin fragmentation was visualized in 1% repeats constitute 7.4% of the human genome but accu- agarose gel. Obtained fragments were in the 200–500 pb mulate 40.7% of all SmaI sites (Table 1). Nearly 200 000 range. Soluble chromatin was obtained by centrifuging the Alu’s (18% of all Alu’s) contain a SmaI site and 155 000 sonicated samples at 14 000g for 10 min at 48C. The soluble retain the AACCCGGG consensus sequence (Table 1 and fraction was diluted 1/10 in dilution buffer (1% Triton Figure 3) and are therefore potential targets of QUMA X-100, 2 mM EDTA, 20 mM Tris pH 8, 150 mM NaCl) and AUMA. While 38.0% of the youngest AluY elements then aliquoted and stored at 808C until use. contain this sequence, the proportions drop to 14.8 and Immunoprecipitation was carried out at 48C by adding 0.4% in AluS and AluJ families, respectively (Table 1 and 5–10mg of the desired antibody to 1 ml of chromatin. Figure 3). These frequencies are consistent with a higher C Chromatin–antibody complexes were immunoprecipitated to T transition trend at CpG sites in older Alu’s (47). with specific antibodies using a protein A/G 50% slurry The representativity of AUMA was analyzed by a (Upstate, Millipore, Billerica, MA, USA) and subse- virtual bioinformatic assay of the human genome quently washed and eluted according to the manufac- sequence. A total of 168 309 AACCCGGG (or CCCGG turer’s instructions. Antibodies against acetylated H3 GTT) hits were identified throughout the genome, with K9/K14 (Upstate), dimethylated H3 K79 and trimethyl- 92.9% of all hits within Alu elements (Table 1 and ated H3 K9 (Abcam, Cambridge, UK) were used. Enrich- Figure 3). This implies that 14.2% of all Alu elements ment for a given chromatin modification was quantified as contained the AACCCGGG sequence. Another 1.0% of a fold enrichment over the input using quantitative real- the hits were in CpG islands and 6.8% in the rest of the time PCR (Roche). For every PCR, a standard curve was genome (including unique sequences and other repeats) obtained to assess amplification efficiency. All quantifica- (Table 1 and Figure 3). As expected, Alu elements con- tions were performed in duplicate. taining the target sequence mostly belonged to the AluS Nucleic Acids Research, 2008, Vol. 36, No. 3 775 Figure 4. Quantitation of unmethylated Alu’s in 17 paired normal mucosa and colorectal carcinoma by QUMA. The values represent the estimated number of unmethylated Alu’s per haploid genome. Most tumors exhibited a higher level of hypomethylation when compared with the respective normal. Figure 3. Relative distribution the Alu elements and sequence targets considered in bioinformatic and experimental QUMA and AUMA. Mb: number of megabases occupied by each type of element; elements: number of elements considered (‘Rest’ has been set arbitrarily to 50%); while this figure is 24.9% in the cancer cell. Considering SmaI site: CCCGGG sequence; vQUMA hits: AACCCGGG (or GGG that the human genome contains 1.1 million Alu ele- CCCTT) sites in Alu elements; vAUMA hits: AACCCGGG (or GGGC ments, these estimates indicate that unmethylated Alu’s CCTT) sites; vAUMA ends: vAUMA hits considering only putative constitute the 2.3 and 3.8% of all Alu’s in the normal and AUMA products of <1 kb (see Material and Methods section); tumor tissues, respectively. AUMA: elements at each one of the two ends of actual AUMA products. Set-up and optimization of AUMA fingerprinting Because QUMA products are fully contained within the (2/3) and AluY families (1/3), with a minimal representa- Alu sequence, it is not possible to identify and position tion of the older family AluJ (<1%). Virtual AUMA in the genome the unmethylated Alu elements. To achieve determined the presence of 5498 putative products of this it is necessary to amplify the targeted unmethylated <1 Kb (the sequence AACCCGGG in the up strand and Alu element together with an adjacent unique sequence. the sequence CCCGGGTT in the down strand at a This was attained through the use of the second method, distance of <1 kb). Although actual AUMA PCR the Amplification of UnMethylated Alus (AUMA). products may reach 2 kb length (see below), we used the AUMA also targets the unmethylated AACCCGGG 1 kb limit to compare with AUMA-isolated bands, which sequence, as in QUMA, but in this case a single primer were shorter than 1 kb. Most virtual AUMA products is used in the PCR (BAu-TT) (see Figure 1 and Supple- contained an Alu element at one of the ends at least mentary Data). Moreover, the product is resolved on (93%). a high-resolution gel electrophoresis resulting in a band- rich fingerprint. AUMA bands correspond to sequences Quantification of unmethylated Alu in normal flanked by two unmethylated target sequences in opposite and tumor tissues strands and sufficiently close to allow PCR amplification The QUMA approach was applied to quantify unmethy- (Figure 1 and Supplementary Data). Since 92% of the lated Alu’s in a series of 18 colorectal carcinomas and their AACCCGGG occurrences in the human genome are in paired normal colonic mucosa. An external DNA frag- Alu’s (Table 1), the approach is largely biased towards ment containing an AluSx element was used as a standard the amplification of unmethylated Alu’s. The presence (see Materials and Methods section and Supplementary of non-Alu sequences at one of the ends or between two Data) in order to make an absolute quantification of the repetitive elements allows the positioning within the number of unmethylated Alu’s. Replicates and dilution genome map of all products. curves of the samples and standard were performed AUMA products generated using the BAu-TT primer to assess reproducibility, sensitivity and accuracy produced highly reproducible fingerprints consisting of (Supplementary Figure 1). Results were normalized by bands ranging from 100 to 2000 bp when resolved in assessment of the number of haploid genomes per test tube high-resolution sequencing gels (Figure 2 and Supplemen- (see Material and Methods section). The average number tary Figure 2). Some well-identifiable bands (up to 5 per of unmethylated Alu’s per haploid genome was experiment) showed random display in both intra-assay 25 486 10 157 in normal mucosa, and 41 995 17 187 and inter-assay replicates (Supplementary Figure 2) and in tumor samples (P = 0.004, paired t-test) (Figure 4). were not considered for analysis. A subset of 110 bands We computationally identified a total of 168 309 Alu with consistent display among all the experiments were elements containing the AACCCGGG sequence (potential tagged and selected for comparative analysis between targets of QUMA) (Table 1). Therefore we estimate samples (see Materials and Methods section). that 15.1% of Alu repeats with a AACCCGGG site are It should be noted that different fingerprints containing unmethylated in the average normal colonic mucosa cell, alternative representations may be obtained by AUMA 776 Nucleic Acids Research, 2008, Vol. 36, No. 3 just by using primers that either amplify from the SmaI strands. Nevertheless, in 27 sequences the AACCCGGG site towards the Alu promoter (upstream Alu amplifica- site was only present at one of the ends, with the other tion) or towards the Alu poly-A tail (downstream Alu end showing high homology with the primer although amplification). Also the stringency of the Alu selection it was not a perfect match. The presence of these may be increased by using longer primers containing sequences suggests that, in some instances, a single cut additional nucleotides corresponding to the Alu consensus in the sequence may be enough to produce an amplifiable sequence (see Material and Methods section). An illus- fragment. This is not considered an artifact since these trative example of AUMA fingerprints generated with bands still represent an unmethylated AACCCGGG site. different Alu-upstream and Alu-downstream primers is shown in Supplementary Figure 2. All the data reported Genome-wide estimations of unmethylation in Alu’s in this article regarding AUMA were obtained using the and distribution by subfamily BAu-TT primer. Of the 137 identified loci represented in AUMA, 114 were Chromosomal origin of AUMA products isolated from normal tissue DNA and 23 from tumor DNA. Half of the sequences contained an Alu sequence at Competitive hybridization between AUMA products one of the ends and two were flanked by two inverted and genomic DNA on metaphase chromosomes yielded Alu’s. AUMA sequences isolated from tumor tissue and a characteristic hybridization pattern demonstrating not present in normal tissue (this corresponds to a tumor- the unequal distribution of AUMA products along the specific hypomethylation) showed a higher proportion human genome (Figure 5A). Competitive hybridization of of Alu elements (16 out of 23, 70%), and included one AUMA products to BAC arrays showed profiles consis- sequence flanked by two inverted Alu’s. Globally, 78 tent with those obtained on metaphase chromosomes unmethylated Alu elements were identified and positioned (Figure 5B). The highest AUMA signal was detected in the human genome map. in whole chromosomes 16, 17 and 19 in contrast with To study the genomic distribution of unmethylated chromosomes 2, 13, 18 and X which were mainly labeled sequences in normal colon mucosa, we only considered the by genomic DNA. Other chromosomes showed a discrete 114 sequences obtained from normal tissue. This resulted pattern of AUMA product hybridization, in which telo- in a total of 201 unmethylated hits characterized through- meric bands in chromosomes 1, 4, 5, 9, 12 and X, and out the genome. The nature of the sequences represented interstitial bands in chromosomes 1, 3, 7, 11 and 12 are the most prominent examples. in actual AUMA showed striking differences with the distribution expected from the virtual AUMA analysis. Identification of AUMA amplified DNA products The methylation status of the sequence is likely to be the main (if not the only) source of these differences because To determine the identity of bands displayed by AUMA, the virtual AUMA did not consider this state. Therefore 38 tagged bands were isolated and cloned. Multiple clones we can use these differences to estimate the degree of from each band were sequenced, resulting in a total of unmethylation of the Alu repeats. Only 29.4% of the 49 different sequences due to the coincidence of more AUMA ends consisted of Alu’s, as compared with the than one sequence in some bands. Characterized bands expected 92.9% resulting from the bioinformatic analysis. included bands displaying no changes in the normal– The highest downrepresentation corresponded to the tumor comparisons and bands recurrently altered in the youngest AluY family, which was present in 6.0% of the tumor. Table 2 summarizes the main features of a subset AUMA ends, while it was expected to add up to 33.1% in of the bands showing recurrent alterations. A list of all the virtual AUMA. AluS representation in actual and virtual sequences isolated from AUMA fingerprints is provided AUMA was 22.3% and 60%, respectively. Interestingly, as Supplementary Table 1. All sequenced bands contained AluJ representations, in both actual and virtual AUMA, a region of non-repetitive sequence and matched with the were closer (1.0% and 0.7%, respectively) (Figure 3 and BLAST reference sequence, allowing the assignment of Table 1). These results suggest that there is a stronger a unique chromosomal localization. The BLAST reference pressure to methylate younger Alus. Alternatively, the hits sequence corresponding to the 49 sequences isolated from corresponding to CpG islands were overrepresented in the AUMA fingerprint presented the target sequence actual versus virtual AUMA by a factor of nearly 25-fold CCCGGGTT including the SmaI at both ends. Southern (27.4% versus 1.1%), consistently with the unmethylated blot analysis of selected cloned sequences showing coinci- status of most CpG islands (Figure 3 and Table 1). The dental size was performed to confirm its correspondence rest of the hits were located in different types of repetitive with the band displayed in AUMA fingerprints (Supple- elements (MIR, MER, LTR, LINE, etc.) and unique mentary Figure 4). sequences (Supplementary Table 1). The miscellaneous To obtain a more representative collection of AUMA collection of sequences (‘Rest’) was over-represented by bands, 200 clones obtained from normal tissue AUMA about 7-fold (observed hits: 43.3%; expected hits: 5.9%, products were sequenced. The analysis revealed 88 addi- Table 1). The 46 AUMA hits represented by the 23 bands tional sequences. This resulted in a total of 137 different loci represented in AUMA (Supplementary Table 1). specific of tumor tissue showed a higher proportion of Most sequences obtained by random cloning were also Alu’s compared with those obtained from normal tissue flanked by two AACCCGGG sequences in opposite DNA (41% versus 29%, respectively), but similar distribution Nucleic Acids Research, 2008, Vol. 36, No. 3 777 Figure 5. (A) Chromosomal origin of AUMA products. A competitive hybridization of AUMA product obtained from normal tissue DNA (red) and genomic DNA (green) to metaphase chromosomes was performed. AUMA products showed an unequal distribution along chromosomes, displaying highest densities at most telomeric regions and some interstitial bands. Chromosomes 16, 17 and 19 yielded the highest AUMA density. (B) Intensity distribution of AUMA products hybridized to BAC arrays in selected chromosomes. The average intensity (X-axis) of the two normal (blue) and tumor samples analyzed (red) for each BAC is shown. BACs are arranged along the Y-axis according to its position in the chromosome. (C) Differential methylation profiles determined by competitive hybridization of AUMA products from normal and tumor tissue to BAC arrays. Illustrative examples are shown for chromosomes 7 and 8 from the two cases analyzed (81 and 151). X-axis indicates log2 ratio of tumor/normal intensities. Positive values (to the right) indicate hypomethylations, negative values (to the left) indicate hypermethylations. Additional examples are shown in Supplementary Figure 5. 778 Nucleic Acids Research, 2008, Vol. 36, No. 3 Table 2. A selection of characterized AUMA bands Band ID Size % GC Chromosome map Gene CpG island Repetitive Methylation status a c (bp) (Location ) elements in in tumor 0 0 band ends (5 /3 ) Ai1 c3 509 55 17p11.2 (18206453–18206961) SHMT1 Yes Alu Sx/MIR Hypermethylated Aj2 c1 458 49 1q32.2 (206389082–206389539) MGC29875 Yes Alu Sq/None Hypomethylated Ao1 c4 365 50 19q13.32 (53550179–53550543) AK001784 No Alu Sx/MIRb Hypomethylated Ap1 c6 358 56 5q35.2 (175157321–175157674) CPLX2 Yes None/MIR Hypermethylated Aq3 c6 339 58 8p23.3 (2007343–2007682) MYOM2 No LTR/Alu Y Hypomethylated Ar3 c3 329 57 2q14.3 (127875178–127875506) AF370412 Yes None/MIRb Hypomethylated As3 c6 306 63 16p13.3 (3160477–3160782) None Yes None/None Hypermethylated Au4 c1 268 57 16p13.3 (3162099–3162366) None No tRNA/None Hypermethylated Nucleotide position within the contig (strand +). NCBI Build 36.1 of the human genome. The whole sequence or a fragment of the sequence lays not further than 200 bp of a predicted CpG island. As compared to the paired normal tissue. by Alu family (10 AluS, 5 AluY, 1 AluJ, 4 in CpG islands and 26 in other sequences). To calculate the proportion and distribution of unmethylated Alu elements on a genomic scale, we performed Monte Carlo simulations taking into account the observed and expected distribution of hits in each Alu family and CpG islands and the rest of sequences (see Material and Methods section). We estimate that at least 4104 Alu elements are unmethylated or partially unmethylated in normal colonic mucosa. This corresponds to 2.64% of all Alu elements containing the target sequence AACCCGGG (Table 1). Although AluS and AluY represent the majority of these sequences it should be noted that the methylation pressure is inverse to the conservation of the SmaI site. That is, the most conserved and younger AluY family shows the lowest relative rate of unmethylation; and the older and more degenerated AluJ family exhibits the highest unmethylation. Application of AUMA to detect differential DNA Figure 6. Distribution of hypermethylation and hypomethylation rates methylation in colorectal carcinomas in the 110 AUMA tagged bands. Rates were obtained by comparison of the AUMA fingerprints obtained in 50 colorectal tumors as In order to test the usefulness of the method for the compared to their respective matched normal tissue. detection of new altered methylation targets, we applied AUMA to a series of 50 colorectal carcinomas and their paired normal mucosa. Two cases were excluded from the analysis due to recurring experimental failure of the normal tissue. The average tumor showed 19 7 (range normal or tumor tissue DNA. For the rest of 48 normal– 6–37) hypomethylations and 22 10 (range 1–39) hyper- tumor pairs, consistent and fully readable fingerprints methylations. It is of note that hypomethylations could were generated and evaluated for normal–tumor differ- either be seen as an increase in the intensity of a pre-existing ential representation. A given case presented, on average, band in the normal tissue or as the appearance of an non- 107 2.9 informative bands (range 98–110). The variation existent band in the normal tissue. This contrasts with was due to polymorphic display or variable resolution hypermethylation events, which rarely showed the com- power of gel electrophoresis. plete loss of a band in the tumor sample, most likely due to In this study, only those bands showing clear intensity the unavoidable contamination of normal tissue. differences between normal and tumor tissue fingerprints Virtually, all tagged bands (109 out of 110) were found (Figure 2) have been scored as methylation changes since to be altered in at least one tumor when compared to its they are more likely to reflect tumor-wide alterations. normal paired mucosa. AUMA tagged bands presented a Because the fingerprints represent sequences flanked by wide distribution in the hypomethylation/hypermethyla- two unmethylated sites, a decreased intensity in a given tion rates (proportion of tumors showing differential band in the tumor in regard to the paired normal tissue display compared to the paired normal tissue) (Figure 6). is indicative of hypermethylation, while an increased Hypomethylation and hypermethylation showed a strong intensity corresponds to hypomethylation (Figure 2). negative correlation (r =0.55 and P< 0.0001), indicat- All tumors displayed changes in regard to the paired ing that most bands tended to be either hypomethylated Nucleic Acids Research, 2008, Vol. 36, No. 3 779 or hypermethylated. A large proportion of tagged bands tissues as well as tumors 53 and 99 showed heavy methy- (78 bands) were recurrently altered in over 25% of the lation of this region (Figure 7C). In contrast to this and cases included in this series. in agreement with AUMA results, tumors 17, 63 and In order to determine whether normal–tumor differences 74 exhibited hypomethylation at most CpGs. Cell lines were limited to isolated independent loci or changes that showed variable profiles of DNA methylation, with might affect larger chromosomal regions, we compared the CaCo exhibiting unmethylation of the MLT1A element distribution of AUMA products generated from two paired but heavy methylation of the AluYd3 element, which normal and tumor tissues and hybridized to BAC arrays. was also heavily methylated in HCT116 cells but not in the Differential hybridization was observed in many BACs, rest of the cell lines tested. MYOM2 expression levels suggesting that relatively large regions encompassing from analyzed by real-time RT-PCR were not affected by the several hundred Kbs to a few Mbs may undergo concurrent methylation status of this sequence (data not shown). hypomethylation or hypermethylation. Telomeric regions Further 45 normal–tumor pairs were analyzed for methy- of many chromosomes contained most of the differential lation of the AluYd3 element by real-time dissociation display (Figure 5C). The differential methylation profiles analysis (Supplementary Figure 7) and it was found were unaffected by chromosomal dosage as demonstrated hypomethylated in 26 tumors (58%). by its independence of chromosomal losses and gains (as Next, we wondered whether the DNA methylation detected by Comparative Genomic Hybridization (CGH) status of the AluYd3 element was associated with alter- (Supplementary Figure 5). native chromatin states. We performed Chromatin ImmunoPrecipitation (ChIP) analysis of histone 3 (H3) Validation of methylation changes detected by AUMA modifications indicative of active chromatin: acetylation of lysines 9 and 14 (AcH3K9/K14), and dimethylation To confirm that the changes observed in AUMA finger- of lysine 79 (2mH3K79); and silent chromatin: trimethy- prints corresponded to actual changes in the methylation lation of lysine 9 (3mH3K9). These histone marks were status of the sequence, eight different sequences obtained compared between cell lines HCT116 and LoVo (with from AUMA fingerprints were analyzed in normal and 100% and 30% methylation of the AluY element, respec- tumor tissues by direct sequencing of sodium bisulfite- tively). The silencing mark 3mH3K9 was 3.5-fold higher in treated DNAs (Table 2). Moreover, it was demonstrated HCT116 cells compared to the LoVo cell line (Figure 7D). that methylation changes affected not only the CpG in at No differences in active marks were observed and these least one of the two flanking SmaI sites (whose methyla- were significantly lower than the silencing mark 3mH3K9. tion prevents AUMA representation) but also neighboring When HCT116 cells were treated with the demethy- CpGs (Supplementary Figure 6). In two samples, hyper- lating agent 5-aza-2 -deoxycytidine (5AzaC) and the methylations/hypomethylations detected by AUMA could inhibitor of histone deacetylase trichostatin A (TSA), not be confirmed by bisulfite sequencing, suggesting that a moderate decrease in the amount of the 3mH3K9 mark the change could affect only a small fraction of tumor cells was observed (Figure 7E). As a whole, these data suggest and that both methods may exhibit different sensitivities. that DNA methylation changes in this AluYd3 element The presence of minor subpopulations can be detected are accompanied by alternated chromatin states. The using more sensitive techniques, i.e. the Methylation molecular consequences of such epigenetic changes remain Specific PCR or by sequencing of multiples clones. to be identified. Functional implications of changes detected by AUMA Next, we wondered if DNA methylation changes detected DISCUSSION by AUMA may have any functional consequences. We Epigenetic states of Alu elements chose one of the most recurrent hypomethylated AUMA sequences (Aq3) and performed an insightful epigenetic Full genome sequencing has provided precise maps characterization of the region in a series of normal–tumor of repetitive elements, and several studies have investigated pairs and in colon cancer cell lines. their distribution and relationship with genome structure Aq3 band is recurrently hypomethylated in tumors (48–51). More recently, a few studies have explored according to AUMA fingerprints (Figure 7A). It repre- sequence-dependent associations between repetitive ele- sents a sequence situated in the eighth intron of the ments and the epigenetic landscape. There is a character- MYOM2 gene (Table 2) and does not fall inside or close to istic distribution of interspersed elements along methylated any CpG island. The SmaI sites are located in a MLT1A and unmethylated domains, with most elements in the repeat and an AluYd3 element. The methylation status the methylated compartment of the genome (21). Nevertheless, two flanking regions of the AUMA band (465 bp and SINEs, which include Alu elements, are the repetitive 213 bp long spanning 20 and 11 CpGs, respectively) was sequences most commonly found in unmethylated domains analyzed by bisulfite direct sequencing (Figure 7B). (21) and some Alu elements may contain discriminatory Confirmation of AUMA data was performed in three motifs associated with methylation-resistant CpG islands normal–tumor pairs exhibiting differential display of the (52). Somatic cells show unstable epigenetic profiles in Aq3 band in AUMA fingerprints (cases 17, 63 and 74) repetitive elements as demonstrated by global measure- and two cases lacking this band in both normal and tumor ments of either DNA methylation (18,20,40,41) or histone pair (cases 53 and 99) (Figure 7A), as well as five cell lines modifications (53,54). Recent studies have revealed (HCT116, DLD-1, LoVo, HT29 and CaCo2). All normal interindividual variability in DNA methylation profiles 780 Nucleic Acids Research, 2008, Vol. 36, No. 3 Figure 7. (A) Detail of the AUMA fingerprints generated from five normal–tumor sample pairs. The presence of the Aq3 band is indicated by an asterisk under the three Aq3 positive cases. (B) The relative position of the AUMA Aq3 band, MLT1A and Alu Y repetitive elements, as well as MYOM2 ninth exon are shown. Each vertical line in the CpG distribution represents a CpG dinucleotide along the DNA sequence. Two different 0 0 0 fragments were amplified for the bisulfite sequencing analysis (gray boxes). Sequence is oriented 5 –3 in regard to MYOM2 3 end. (C) Methylation status of the CpG nucleotides in the two fragments amplified were ascertained by direct sequencing of bisulfite-treated DNAs of 5 normal–tumor pairs and 5 colon cancer cell lines. (D) ChIP analysis of the AluY element frequently hypomethylated in cancer revealed loss of trimethylation in histone 3 lysine 9 residue (3mH3K9) in LoVo cells (unmethylated at DNA level) as compared to HCT116 (methylated at DNA level). Treatment of HCT116 cells with 5AzaC and TSA produced a moderate decrease in the levels of trimethylation in H3K9. at specific Alu elements (55), and Fraga and colleagues Beyond these few studies, the extension and nature detected epigenetic changes arising during the lifetime of the epigenetic state of interspersed elements is largely of monozygotic twins in Alu elements and other unknown. Global estimates of DNA methylation in sequences (56). repetitive elements have been obtained by Southern blot Nucleic Acids Research, 2008, Vol. 36, No. 3 781 analyses (30) and, more recently, by using approaches in AUMA, with only 9% of the Alu elements of the old J based on bisulfite conversion of the unmethylated cytosine subfamily containing the SmaI site retain the AA dinu- (40,41,57). These studies have confirmed the global cleotide needed for their amplification, while this figure is hypomethylation of most tumors but they do not provide 91 and 80% in the younger AluY and AluS subfamilies, detailed information on the nature and localization of the respectively (Table 1), making clear that younger Alu elements tend to retain the SmaI site nearly as much as unmethylated elements. In silico analysis has revealed that they retain the AA dinucleotide required for their ampli- a number of Alu elements close to CpG islands retain a fication. This bias is not a handicap, since unmethylated high proportion of CpG sites, and this is presumed to be Alu sequences revealed by AUMA are likely to represent a sign of unmethylation (58), but no experimental proof the most relevant events of this kind, because spurious has been provided. In our point of view, the lack of unmethylation of old Alu elements retaining a single or a simple, specific and sensitive methodologies to screen for few CpG sites is expected to have less biological signif- epigenetic changes in repetitive elements on the genomic icance than unmethylation of younger Alu elements that scale has precluded a clearer understanding of the nature are usually closer to active chromatin regions (21) and and implications of these sequences in cell biology. retain more CpGs. The stronger methylation pressure observed in the AluY class is consistent with this Properties of QUMA and AUMA postulate. Here we report a systematic screening of unmethylated AUMA was designed to amplify DNA fragments Alus as a tool to determine the extent of DNA hypo- containing the target sequence (AACCCGGG), which is methylation, to identify specifically unmethylated ele- present in Alu and other repetitive elements. Because a ments and to detect epigenetic alterations in cancer cells. single primer was used for PCR amplification, the target QUMA is a very simple and specific method and provides sequence must appear in both strands of the DNA at accurate relative estimates of the number of unmethylated relatively nearby positions. As expected, Alu elements, elements. QUMA is specially appropriate for compara- with more than one million copies per human genome tive studies, but also provides a raw quantitation of the (15), were the most frequent repeat in AUMA bands number of unmethylated elements per haploid genome, (50% in sequences isolated from non-tumor tissue), but outlining the extent of hypomethylated Alu’s in normal only two sequenced bands contained two inverted Alu and pathologic cells. QUMA analysis indicates that about repeats (Supplementary Table 1). This observation is 1 out of 6 Alu elements containing the AACCCGGG in concordance with previous works reporting on the site are unmethylated, while in tumors, this figure nearly instability of this inverted repeats, which might have doubles in agreement with previous studies (23). Although caused their exclusion from the human genome (61,62). these analyses are likely to generate good estimates at the More restrictive conditions to select for Alu, or any other comparative level (between samples), absolute values repeat of interest, may be achieved by extending the 3 end should be treated with caution because the determination of the primer specific sequence (see Supplementary Data); refers to a single CpG site within the Alu element. however, the number of sequences we obtained was consi- To date there is still a lack of proper methodologies dered appropriate to accomplish the original aim of the allowing genome-wide screenings for recurrent hypo- study which is to screen for differentially methylated methylated regions that may have some impact on repetitive elements in colorectal cancer. tumor biology. Even though QUMA and other method- It is worth noting that AUMA patterns are highly ologies (41,59) allow quantitation of unmethylated reproducible not only in replicates but also among differ- repeats, they do not provide a straightforward approach ent samples, which indicates that the unmethylated status to identify and map the amplified targets. At this point, of these repeats is tightly controlled, probably by the AUMA takes us a step further, allowing the undoubtful epigenetic status of nearby regions. This is strengthened by identification of hypomethylated sequences, in addition the confirmation that unmethylation extends many CpG to hypermethylated targets. Although AUMA is specially sites beyond the SmaI cut site. Moreover, about 50% of suited to determine the nature of the unmethylated the bands tagged in AUMA fingerprints exhibited variable elements, it also allows the calculation of global unmethy- display among normal tissues (data not shown), suggest- lation in Alu elements. Nevertheless, it should be taken ing the usefulness of this technique to investigate epigene- into account that this is an indirect measure, because it tic polymorphisms. relies in the extent of methylation in CpG islands and Alu’s and other repetitive elements tend to be highly other sequences. Moreover, unmethylation of a second methylated in most somatic tissues (8,9,40,63). Here we SmaI site near the Alu is also required to generate the have identified 78 ‘atypical’ Alu elements exhibiting AUMA band and hence to be detected. While AUMA full or partial unmethylation in normal colonic mucosa shares many technical steps with other techniques, namely cells. Different evidences underscore the adequacy of this MCA (60) and AIMS (43), its design is conceptually approach to track changes with possible functional impli- unique since AUMA scans for the atypically unmethy- cations: (i) a significant portion of the characterized bands lated Alu sequences, unlike the other approaches that are are located inside or nearby CpG islands and genes; enriched for typical methylated sequences. (ii) AUMA products show a characteristic distribution Due to sequence degeneration, both QUMA and in R bands, coincidentally with the distribution of Alu AUMA are more effective in screening for unmethylation sequences (15), indicating a bias toward the gene-richest in younger elements. This trend is more clearly seen portion of the genome known as the H3 isochore (64). 782 Nucleic Acids Research, 2008, Vol. 36, No. 3 DNA methylation along Alu families Aq3 band is flanked by two repeats, a LTR and an AluY, which map within an intron between exons 8 and Alu families showed striking differences in their methyla- 9 of the MYOM2 gene at 8p23.3. Both elements are tion level. Most of the Alu elements characterized here are heavily methylated in normal tissue and partially to fully from the younger families AluS and AluY (74% and 22%, unmethylated in tumor tissue. Interestingly, we have respectively). Nevertheless, this observation is mainly due found moderately high levels of the heterochromatin to the depletion of CpG sites in older Alu elements. associated mark 3mH3K9 in the fully methylated AluYd3 Hence, only 1 out 230 AluJ elements maintains the element in the HCT116 cell line, while the levels were AUMA target site (AACCCGGG), while younger ele- significantly lower (3.5-fold) in the partially demethylated ments show higher rates of maintenance in accordance LoVo cell line. Furthermore, none of the classical active with their age (AluS: 1 out of 7; AluY: 2 out of 5) marks AcH3K9/K14 and 2mH3K79, have been found (Table 1). Interestingly, the rate of unmethylation is higher enriched in the LoVo cell line. These data are in concor- in older elements (AluJ: 12.2%; AluS: 3.1%; AluY: 1.6%) dance with preliminary data showing that the hypomethy- (Table 1). As noted by Rollins et al. (21), the boundaries lation does not affect the expression of MYOM2 (data not of unmethylated domains tend to be occupied by methy- shown) but could rather affect chromatin structure in the lated Alu transposons of the younger AluS and AluY region. In agreement with these observations, this genomic families. Other studies have noted that CpG island- region undergoes frequent losses (68–70) and is rearranged associated Alu’s retain a higher proportion of CpG sites, in many different types of tumors (71,72), which hints at suggesting that these elements are unmethylated in the a role for DNA hypomethylation in genomic instability germ line (58). These unmethylated elements that can be (24–27,73). The specific functional consequences of this easily revealed by AUMA are likely to play a regulatory hypomethylation deserve further investigations. role in a significant number of genes (65). AUMA of Another application of AUMA is the detection of tumors was enriched in Alu sequences as compared with genomic regions that have been silenced in cancer. Inter- normal tissue (41% versus 29%, respectively), suggesting spersed elements are concentrated in gene-rich regions and a favored hypomethylation of Alu elements within the due to the intended selection of unmethylated repetitive global genomic hypomethylation associated with tumor- elements in AUMA, it appears reasonable to postulate igenesis (23). that normally unmethylated sequences are likely to pin- point active genomic regions. In this context, AUMA Application of AUMA to detect epigenetic changes provides a large collection of genomic regions undergoing in cancer cells hypermethylation, which are readily seen as bands recur- Cancer-related hypomethylation is well documented (23) rently loss in the fingerprints. DNA methylation asso- and different studies have demonstrated the demethylation ciated epigenetic silencing is probably one of the most of Alu’s and other repetitive elements in different types prevalent mechanisms of tumor suppression inactivation of neoplasias (66). Our data are consistent with previous in cancer (33,37,74). Therefore, AUMA can also be used estimates and go one step forward in the characterization to screen for differential methylation not only in Alu of unmethylated repeats. The AUMA approach was elements but also in unique sequences and repetitive conceived as a straightforward DNA methylation screen- elements other than Alu. ing strategy targeting specific interspersed repeats and In summary, QUMA and AUMA methodologies are suitable to be applied to large series of samples or a simple and novel approach to explore and gain insights experimental conditions as reported here. into the functional significance of interspersed genomic The application of AUMA to a series of colorectal elements and neighboring sequences. Due to its distinctive carcinomas and their paired matched normal tissue has features (bias for unmethylated elements in gene-rich revealed a high rate of alterations. This indicates the regions and detection of both hypomethylation and hyper- plasticity of epigenetic control of the elements screened methylation) we think that these techniques constitute by AUMA in colorectal carcinogenesis. Although some a new and unique tool that should complement global bands show bidirectional changes (hypomethylations and determinations and high-resolution genome-wide scanning hypermethylations), which have been also reported in strategies. Beyond unmethylated repetitive elements, other sequences (42,67), most bands display an alternative AUMA can be also used to detect recurrent epigenetic trend either toward hypermethylations or hypomethyla- changes associated with tumorigenesis including gene tions. Some of these changes are highly recurrent (in more epigenetic inactivation. than 50% of tumors), suggesting that they may represent relevant alterations related to mechanisms frequently disturbed in colon cancer. Because the default status of SUPPLEMENTARY DATA repetitive elements is methylation, hypomethylations Supplementary data are available at NAR Online. are readily detected as the emergence of a new band in the tumor AUMA fingerprint. Since all amplified bands include a unique sequence, it has been possible to identify ACKNOWLEDGEMENTS all of the isolated bands and pinpoint them in the genomic map. We thank Gemma Aiza for technical support and Jessica As an example, we have investigated the Aq3 sequence, Halow for critical review of the manuscript. J.R. was one of the most recurrent hypomethylations in this study. a fellow of the Generalitat de Catalunya; E.V. was a fellow Nucleic Acids Research, 2008, Vol. 36, No. 3 783 23. Ehrlich,M. (2002) DNA methylation in cancer: too much, but also of the Fondo de Investigacio´ n Sanitaria (FIS). This work too little. Oncogene, 21, 5400–5413. was supported by a grant from the Ministry of Education 24. Eden,A., Gaudet,F., Waghmare,A. and Jaenisch,R. (2003) and Science (SAF2006/351) and the Consolider-Ingenio Chromosomal instability and tumors promoted by DNA hypo- 2010 Program (CSD2006-49). Funding to pay the Open methylation. Science, 300, 455. 25. Chen,R.Z., Pettersson,U., Beard,C., Jackson-Grusby,L. and Access publication charges for this article was provided Jaenisch,R. (1998) DNA hypomethylation leads to elevated by the Institute of Predictive and Personalized Medicine of mutation rates. Nature, 395, 89–93. Cancer (IMPPC). 26. Suzuki,K., Suzuki,I., Leodolter,A., Alonso,S., Horiuchi,S., Yamashita,K. and Perucho,M. (2006) Global DNA demethylation Conflict of interest statement. None declared. in gastrointestinal cancer is age dependent and precedes genomic damage. Cancer Cell., 9, 199–207. 27. Rodriguez,J., Frigola,J., Vendrell,E., Risques,R.A., Fraga,M.F., REFERENCES Morales,C., Moreno,V., Esteller,M., Capella,G. et al. (2006) Chromosomal instability correlates with genome-wide DNA 1. Mattick,J.S. (2003) Challenging the dogma: the hidden layer of demethylation in human primary colorectal cancers. Cancer Res., non-protein-coding RNAs in complex organisms. Bioessays, 25, 66, 8462–9468. 930–939. 28. Liu,W.M., Maraia,R.J., Rubin,C.M. and Schmid,C.W. (1994) 2. Zuckerkandl,E. (2002) Why so many noncoding nucleotides? The Alu transcripts: cytoplasmic localisation and regulation by DNA eukaryote genome as an epigenetic machine. Genetica, 115, 105–129. methylation. Nucleic Acids Res., 22, 1087–1095. 3. Kreahling,J. and Graveley,B.R. (2004) The origins and implications 29. Roman-Gomez,J., Jimenez-Velasco,A., Agirre,X., Cervantes,F., of Aluternative splicing. Trends Genet., 20, 1–4. Sanchez,J., Garate,L., Barrios,M., Castillejo,J.A., Navarro,G. et al. 4. Jasinska,A. and Krzyzosiak,W.J. (2004) Repetitive sequences that (2005) Promoter hypomethylation of the LINE-1 retrotransposable shape the human transcriptome. FEBS Lett., 567, 136–141. elements activates sense/antisense transcription and marks the 5. Jordan,I.K., Rogozin,I.B., Glazko,G.V. and Koonin,E.V. (2003) progression of chronic myeloid leukemia. Oncogene, 24, 7213–7223. Origin of a substantial fraction of human regulatory sequences from 30. Ehrlich,M. (2006) Cancer-linked DNA hypomethylation and its transposable elements. Trends Genet., 19, 68–72. relationship to hypermethylation. Curr. Top. Microbiol. Immunol., 6. Hasler,J. and Strub,K. (2006) Alu elements as regulators of gene 310, 251–274. expression. Nucleic Acids Res., 34, 5491–5497. 31. Plass,C. (2002) Cancer epigenomics. Hum. Mol. Genet., 11, 7. Slotkin,R.K. and Martienssen,R. (2007) Transposable elements and 2479–2488. the epigenetic regulation of the genome. Nat. Rev. Genet., 8, 32. Jones,P.A. and Baylin,S.B. (2002) The fundamental role of 272–285. epigenetic events in cancer. Nat. Rev. Genet., 3, 415–428. 8. Schmid,C.W. (1998) Does SINE evolution preclude Alu function? 33. Herman,J.G. and Baylin,S.B. (2003) Gene silencing in cancer in Nucleic Acids Res., 26, 4541–4550. association with promoter hypermethylation. N. Engl. J. Med., 349, 9. Yoder,J.A., Walsh,C.P. and Bestor,T.H. (1997) Cytosine methyla- 2042–2054. tion and the ecology of intragenomic parasites. Trends Genet., 13, 34. Lund,A.H. and van Lohuizen,M. (2004) Epigenetics and cancer. 335–340. Genes Dev., 18, 2315–2335. 10. Jaenisch,R. and Bird,A. (2003) Epigenetic regulation of gene 35. Feinberg,A.P. and Tycko,B. (2004) The history of cancer epige- expression: how the genome integrates intrinsic and environmental netics. Nat. Rev. Cancer, 4, 143–153. signals. Nat. Genet., 33(Suppl.), 245–254. 36. Laird,P.W. (2005) Cancer epigenetics. Hum. Mol. Genet., 14, 11. Weissmann,F. and Lyko,F. (2003) Cooperative interactions between R65–R76. epigenetic modifications and their function in the regulation of 37. Esteller,M. (2007) Cancer epigenomics: DNA methylomes and chromosome architecture. Bioessays, 25, 792–797. histone-modification maps. Nat. Rev. Genet., 6,6. 12. Goll,M.G. and Bestor,T.H. (2005) Eukaryotic cytosine methyl- 38. Laird,P.W. (2003) The power and the promise of DNA methylation transferases. Annu. Rev. Biochem., 74, 481–514. markers. Nat. Rev. Cancer, 3, 253–266. 13. Fazzari,M.J. and Greally,J.M. (2004) Epigenomics: beyond CpG 39. Cottrell,S.E. (2004) Molecular diagnostic applications of DNA islands. Nat. Rev. Genet., 5, 446–455. methylation technology. Clin. Biochem., 37, 595–604. 14. Lander,E.S., Linton,L.M., Birren,B., Nusbaum,C., Zody,M.C., 40. Yang,A.S., Estecio,M.R., Doshi,K., Kondo,Y., Tajara,E.H. and Baldwin,J., Devon,K., Dewar,K., Doyle,M. et al. (2001) Initial Issa,J.P. (2004) A simple method for estimating global DNA sequencing and analysis of the human genome. Nature, 409, methylation using bisulfite PCR of repetitive DNA elements. 860–921. Nucleic Acids Res., 32, e38. 15. Batzer,M.A. and Deininger,P.L. (2002) Alu repeats and human 41. Weisenberger,D.J., Campan,M., Long,T.I., Kim,M., Woods,C., genomic diversity. Nat. Rev. Genet., 3, 370–379. Fiala,E., Ehrlich,M. and Laird,P.W. (2005) Analysis of repetitive 16. Korenberg,J.R. and Rykowski,M.C. (1988) Human genome element DNA methylation by MethyLight. Nucleic Acids Res., 33, organization: Alu, lines, and the molecular structure of metaphase 6823–6836. chromosome bands. Cell, 53, 391–400. 42. Frigola,J., Sole,X., Paz,M.F., Moreno,V., Esteller,M., Capella,G. 17. Chen,C., Gentles,A.J., Jurka,J. and Karlin,S. (2002) Genes, and Peinado,M.A. (2005) Differential DNA hypermethylation and pseudogenes, and Alu sequence organization across hypomethylation signatures in colorectal cancer. Hum. Mol. Genet., human chromosomes 21 and 22. Proc. Natl Acad. Sci. USA, 99, 14, 319–326. 2930–2935. 43. Frigola,J., Ribas,M., Risques,R.A. and Peinado,M.A. (2002) 18. Schmid,C.W. (1991) Human Alu subfamilies and their Methylome profiling of cancer cells by amplification of inter- methylation revealed by blot hybridization. Nucleic Acids Res., 19, methylated sites (AIMS). Nucleic Acids Res., 30, e28. 5613–5617. 44. Vendrell,E., Ribas,M., Valls,J., Sole,X., Grau,M., Moreno,V., 19. Gama-Sosa,M.A., Wang,R.Y., Kuo,K.C., Gehrke,C.W. and Capella,G. and Peinado,M.A. (2007) Genomic and transcriptomic Ehrlich,M. (1983) The 5-methylcytosine content of highly prognostic factors in R0 Dukes B and C colorectal cancer patients. repeated sequences in human DNA. Nucleic Acids Res., 11, 3087–3095. Int. J. Oncol., 30, 1099–1107. 20. Kochanek,S., Renz,D. and Doerfler,W. (1993) DNA methylation in 45. Perucho,M., Welsh,J., Peinado,M.A., Ionov,Y. and McClelland,M. the Alu sequences of diploid and haploid primary human cells. (1995) Fingerprinting of DNA and RNA by arbitrarily primed EMBO J., 12, 1141–1151. polymerase chain reaction: applications in cancer research. Methods 21. Rollins,R.A., Haghighi,F., Edwards,J.R., Das,R., Zhang,M.Q., Enzymol., 254, 275–290. Ju,J. and Bestor,T.H. (2006) Large-scale structure of genomic 46. Stirzaker,C., Song,J.Z., Davidson,B. and Clark,S.J. (2004) methylation patterns. Genome Res., 16, 157–163. Transcriptional gene silencing promotes DNA hypermethylation 22. Bird,A. (2002) DNA methylation patterns and epigenetic memory. through a sequential change in chromatin modifications in cancer Genes Dev., 16, 6–21. cells. Cancer Res., 64, 3871–3877. 784 Nucleic Acids Research, 2008, Vol. 36, No. 3 47. Xing,J., Hedges,D.J., Han,K., Wang,H., Cordaux,R. and 61. Lobachev,K.S., Stenger,J.E., Kozyreva,O.G., Jurka,J., Gordenin,D.A. and Resnick,M.A. (2000) Inverted Alu repeats Batzer,M.A. (2004) Alu element mutation spectra: Molecular clocks unstable in yeast are excluded from the human genome. EMBO J., and the effect of DNA methylation. J. Mol. Biol, 344, 675–682. 19, 3822–3830. 48. Medstrand,P., van de Lagemaat,L.N. and Mager,D.L. (2002) 62. Stenger,J.E., Lobachev,K.S., Gordenin,D., Darden,T.A., Jurka,J. Retroelement distributions in the human genome: variations and Resnick,M.A. (2001) Biased distribution of inverted and direct associated with age and proximity to genes. Genome Res., 12, Alus in the human genome: implications for insertion, exclusion, 1483–1495. and genome stability. Genome Res., 11, 12–27. 49. Grover,D., Majumder,P.P., Rao,C.B., Brahmachari,S.K. and 63. Hellmann-Blumberg,U., Hintz,M.F., Gatewood,J.M. and Mukerji,M. (2003) Nonrandom distribution of alu elements in genes Schmid,C.W. (1993) Developmental differences in methylation of of various functional categories: insight from analysis of human human Alu repeats. Mol. Cell. Biol., 13, 4523–4530. chromosomes 21 and 22. Mol. Biol. Evol., 20, 1420–1424. 64. Saccone,S., Caccio,S., Kusuda,J., Andreozzi,L. and Bernardi,G. 50. Grover,D., Mukerji,M., Bhatnagar,P., Kannan,K. and (1996) Identification of the gene-richest bands in human chromo- Brahmachari,S.K. (2004) Alu repeat analysis in the complete human somes. Gene, 174, 85–94. genome: trends and variations with respect to genomic composition. 65. Oei,S.L., Babich,V.S., Kazakov,V.I., Usmanova,N.M., Bioinformatics, 20, 813–817. Kropotov,A.V. and Tomilin,N.V. (2004) Clusters of regulatory 51. Price,A.L., Eskin,E. and Pevzner,P.A. (2004) Whole-genome signals for RNA polymerase II transcription associated with Alu analysis of Alu repeat elements reveals complex evolutionary family repeats and CpG islands in human promoters. Genomics, 83, history. Genome Res., 14, 2245–2252. 873–882. 52. Feltus,F.A., Lee,E.K., Costello,J.F., Plass,C. and Vertino,P.M. 66. Wilson,A.S., Power,B.E. and Molloy,P.L. (2007) DNA hypo- (2006) DNA motifs associated with aberrant CpG island methyla- methylation and human diseases. Biochim. Biophys. Acta, 1775, tion. Genomics, 15, 15. 138–162. 53. Fraga,M.F., Ballestar,E., Villar-Garea,A., Boix-Chornet,M., 67. Nishiyama,R., Qi,L., Tsumagari,K., Weissbecker,K., Dubeau,L., Espada,J., Schotta,G., Bonaldi,T., Haydon,C., Ropero,S. et al. Champagne,M., Sikka,S., Nagai,H. and Ehrlich,M. (2005) A DNA (2005) Loss of acetylation at Lys16 and trimethylation at Lys20 of Repeat, NBL2, Is Hypermethylated in Some Cancers but histone H4 is a common hallmark of human cancer. Nat. Genet., 13, Hypomethylated in Others. Cancer Biol. Ther., 4,4. 68. Emi,M., Fujiwara,Y., Nakajima,T., Tsuchiya,E., Tsuda,H., 54. Kondo,Y. and Issa,J.P. (2003) Enrichment for histone H3 lysine 9 Hirohashi,S., Maeda,Y., Tsuruta,K., Miyaki,M. et al. (1992) methylation at Alu repeats in human cells. J. Biol. Chem., 278, Frequent loss of heterozygosity for loci on chromosome 8p in 27658–27662. hepatocellular carcinoma, colorectal cancer, and lung cancer. 55. Sandovici,I., Kassovska-Bratinova,S., Loredo-Osti,J.C., Leppert,M., Cancer Res., 52, 5368–5372. Suarez,A., Stewart,R., Bautista,F.D., Schiraldi,M. and Sapienza,C. 69. Sunwoo,J.B., Sun,P.C., Gupta,V.K., Schmidt,A.P., El-Mofty,S. and (2005) Interindividual variability and parent of origin DNA Scholnick,S.B. (1999) Localization of a putative tumor suppressor methylation differences at specific human Alu elements. Hum. Mol. gene in the sub-telomeric region of chromosome 8p. Oncogene, 18, Genet., 14, 2135–2143. 2651–2655. 56. Fraga,M.F., Ballestar,E., Paz,M.F., Ropero,S., Setien,F., 70. Muscheck,M., Sukosd,F., Pesti,T. and Kovacs,G. (2000) High Ballestar,M.L., Heine-Suner,D., Cigudosa,J.C., Urioste,M. et al. density deletion mapping of bladder cancer localizes the putative (2005) Epigenetic differences arise during the lifetime of mono- tumor suppressor gene between loci D8S504 and D8S264 at zygotic twins. Proc. Natl Acad. Sci. USA, 102, 10604–10609. chromosome 8p23.3. Lab Invest., 80, 1089–1093. 57. Xiong,Z. and Laird,P.W. (1997) COBRA: a sensitive and quanti- 71. Wong,N., Wong,K.F., Chan,J.K. and Johnson,P.J. (2000) tative DNA methylation assay. Nucleic Acids Res., 25, 2532–2534. Chromosomal translocations are common in natural killer-cell 58. Brohede,J. and Rand,K.N. (2006) Evolutionary evidence suggests lymphoma/leukemia as shown by spectral karyotyping. Hum. that CpG island-associated Alus are frequently unmethylated in Pathol., 31, 771–774. human germline. Hum. Genet., 119, 457–458. 72. Jin,Y., Jin,C., Wennerberg,J., Hoglund,M. and Mertens,F. (2001) 59. Choi,I.S., Estecio,M.R., Nagano,Y., Kim do,H., White,J.A., Cytogenetic and fluorescence in situ hybridization characterization Yao,J.C., Issa,J.P. and Rashid,A. (2007) Hypomethylation of of chromosome 8 rearrangements in head and neck squamous cell LINE-1 and Alu in well-differentiated neuroendocrine tumors carcinomas. Cancer Genet. Cytogenet., 130, 111–117. (pancreatic endocrine tumors and carcinoid tumors). Mod. Pathol., 73. Karpf,A.R. and Matsui,S. (2005) Genetic disruption of cytosine 20, 802–810. DNA methyltransferase enzymes induces chromosomal instability in 60. Toyota,M., Ho,C., Ahuja,N., Jair,K.W., Li,Q., Ohe-Toyota,M., human cancer cells. Cancer Res., 65, 8635–8639. Baylin,S.B. and Issa,J.P. (1999) Identification of differentially 74. Egger,G., Liang,G., Aparicio,A. and Jones,P.A. (2004) Epigenetics methylated sequences in colorectal cancer by methylated CpG in human disease and prospects for epigenetic therapy. Nature, 429, island amplification. Cancer Res., 59, 2307–2312. 457–463.

Journal

Nucleic Acids Research – Oxford University Press

Published: Feb 15, 2008

There are no references for this article.