Aberrant methylation and impaired expression of the

Leukemia (2003) 17, 910–918 & 2003 Nature Publishing Group All rights reserved 0887-6924/03 $25.00 www.nature.com/leu

Aberrant methylation and impaired expression of the p15INK4b cell cycle regulatory gene in chronic myelomonocytic leukemia (CMML) M Tessema1, F La¨nger1, J Dingemann1, A Ganser2, H Kreipe1 and U Lehmann1 1 Institute of Pathology, Medizinische Hochschule Hannover, Carl-Neuberg-Street 1, Hannover, Germany; and 2Department of Hematology/Oncology, Medizinische Hochschule Hannover, Carl-Neuberg-Street 1, Hannover, Germany

The important cell cycle regulatory gene p15INK4b has been shown to be inactivated in acute myeloid leukemia and myelodysplastic syndrome. Little is known about the expression and epigenetic modification of this gene in chronic myelomonocytic leukemia (CMML) that belongs to the myelodysplastic/myeloproliferative disorders (MDS/MPD) with a high proportion of blastic transformation. Analysis of bone marrow trephines in a series of 33 CMML cases showed an aberrant p15INK4b gene methylation in up to 58% of cases. Methylation was analyzed employing different methylation-specific PCR and genomic sequencing protocols. It turned out to be spread over a broad area of the 50 region and exhibited substantial heterogeneity between cases and even in individual patients. The degree of aberrant methylation was correlated with a reduced mRNA as well as reduced protein expression, and was associated with a higher expression of DNA methyltransferase DNMT 3A. We conclude that aberrant gene methylation is a frequent event in CMML that might contribute to the pathogenesis of this MDS/MPD. Leukemia (2003) 17, 910–918. doi:10.1038/sj.leu.2402891 Keywords: chronic myelomonocytic leukemia (CMML); p15INK4b gene; methylation; bone marrow trephines

classification, CMML is placed in a separate subgroup as MDS/myeloproliferative disorder (MDS/MPD).11 The studies analyzing p15INK4b gene methylation in MDS also included some CMML samples, but only in very limited numbers. Altogether the analysis of only 20 CMML samples from different institutions employing different methodologies and analyzing different regions of the gene is reported in the literature.8,9,12–14 In order to clarify the frequency and patterns of p15INK4b gene methylation in CMML, we analyzed a large single center cohort (n ¼ 33) using different methods. Nearly all publications addressing the question of p15INK4b gene methylation in MDS and CMML do not analyze the expression of the p15 gene. Only Preisler et al8 and Daskalakis et al14 reported the expression analysis in two and three CMML samples, respectively. Therefore, we analyzed the p15INK4b mRNA expression level in all 33 CMML samples using quantitative real-time PCR technology and assayed for the protein expression using immunohistochemistry to evaluate the functional consequences of the epigenetic alteration.

Materials and methods Introduction Alteration of methylation patterns is a common event in human neoplasia. In addition to deletion or mutation, it constitutes another complementary acting inactivation mechanism for growth regulatory genes leading to loss of proliferation control.1 Members of the family of cyclin-dependent kinase (CDK) inhibitors, like p15INK4b and p16INK4a, are among the most frequent targets of aberrant methylation. The proteins encoded by these genes induce cell cycle arrest at the G1/S phase transition through inhibition of the kinase activity of the cyclin D/CDK four and six complex.2 Methylation-induced silencing of the p15INK4b and the p16INK4a gene has been documented in a wide variety of epithelial and hematological malignancies.3 In acute myeloid leukemia (AML) hypermethylation of p15INK4b occurs in more than 75% of the cases, which is more prevalent than any of the other genetic abnormalities.4,5 Also for the myelodysplastic syndrome (MDS), a heterogeneous stem cell disorder with a varying probability to progress to AML,6 aberrant methylation of the tumor suppressor gene p15INK4b has been shown (see Quesnel and Fenaux7 and Preisler et al8 and references therein). p15INK4b gene methylation occurred more frequently in high-risk MDS with an increased tendency to progress to blast transformation.9 Chronic myelomonocytic leukemia (CMML) belongs to the MDS according to the French–American–British (FAB) cooperative leukemia group classification.10 In the new WHO Correspondence: Dr U Lehmann, Institute of Pathology, Medizinische Hochschule Hannover, Carl-Neuberg-Street 1, D-30625 Hannover, Germany; Fax: +49 511 532 5799 Received 28 October 2002; accepted 20 December 2002

Patients Formalin-fixed paraffin-embedded bone marrow trephines from CMML patients (n ¼ 33) as well as biopsies displaying only nonneoplastic reactive alterations (n ¼ 20) were retrieved from the archive of the Institute of Pathology, Medizinische Hochschule Hannover, Germany. Cases of CMML were defined as outlined by the recently published criteria of the WHO.11 Briefly, a persistent monocytosis (41 109/l), the absence of the Philadelphia chromosome, fewer than 20% blasts in the blood or bone marrow and dysplastic features in at least one cell lineage of the bone marrow were required.

DNA isolation and methylation analysis DNA isolation and bisulfite treatment were performed essentially as described.15 For the qualitative detection of CpGmethylation approx. 25 ng of converted DNA were amplified using the methylation-specific primer pairs 1M50 /30 and 1U50 /30 , as well as 3M50 /30 and 3U50 /30 (see Figure 1a for location and Table 1 for sequences and annealing temperature) in a total reaction volume of 25 ml containing 20 mM Tris-HCl (pH 8.4), 50 mM KCl, 200 mM dNTP, 400 nM each primer, 2.5 mM MgCl2 and 0.625 units Platinumt-Taq (Invitrogen, Karlsruhe, Germany). After an initial denaturation step at 951C for 5 min 35 or 40 cycles, followed with 30 s at 951C, 40 s annealing and 30 s at 721C and a final extension step of 5 min at 721C. Approximately 20 ng of converted DNA were amplified using methylation-independent primer pairs. For direct sequencing of

p15 methylation in CMML M Tessema et al

911

a 59

29

3

direct sequencing MSP- Primer

Primer I

Primer II

KG1a healthy control #4 healthy control #6

1 - 10 %

b bp

11 - 49 %

50 - 89 %

90 - 100 %

co ntr ol #4 co ntr ol #6 KG 1a

0%

64

p15

51

89 80

β-GUS

Figure 1 (a) Schematic representation of the CpG islands encompassing the first exon of the p15INK4b gene. The region covered by the direct sequencing approach is indicated by horizontal lines, the location of the MSP primer pairs is indicated by arrows. The three cytosine residues analyzed by Nguyen et al19 using the MSNuPE-assay are marked by asterisks. Each circle represents one potentially methylated cytosine base (the first CpG corresponds to nucleotide no. 80 in the GenBank sequence S75756, the last one to nucleotide no. 527). (b) RT-PCR analysis of a completely methylated sample (cell line KG1a) and unmethylated samples (biopsies displaying only mild reactive changes). Reference transcript: b-GUS.

PCR products, a nested PCR was performed using primer pairs mod F2/mod R2 and mod F1-t/mod R1-t for the 50 -region and 2I50 /30 and 3I50 -t/30 -t for the 30 -region in a total reaction volume of 25 ml containing 20 mM Tris-HCl (pH 8.4), 50 mM KCl, 200 mM dNTP, 400 nM each primer, 1.5 mM MgCl2 and 0.625 units Platinumt-Taq (Invitrogen, Karlsruhe, Germany). The nested primer pairs contain M13 universal sequencing primer tails. After an initial denaturation step at 951C for 5 min 25 cycles follow with 30 s at 951C, 40 s at 541C and 30 s at 721C, and a final extension step of 10 min at 721C. A measure of 1 ml of the amplification product is added to a new reaction mix containing primer pair mod F1-t/mod R1-t, followed by 25 cycles with an annealing temperature of 521C. A very similar nested PCR was performed for the more 30 located region with the appropriate primers (see Table 1) and annealing temperatures of 60 and 651C, respectively. For quantification of methylation levels at a given CpG site, the intensities of the corresponding C- and Tbands were compared: equal intensities means 50% methylation, three times more C than T means 75% methylated, etc. For sequencing of individual clones, a single-step PCR was performed using primer pair 1I50 /30 for the 50 region and 2I50 /30

for the 30 region. The reaction conditions were as described above with 40 cycles and an annealing temperature of 56 and 601C, respectively. PCR products were inserted into an appropriate plasmid vector using the Topot-cloning kit (Invitrogen, Karlsruhe, Germany) and sequenced using the CycleReadertAuto DNA sequencing kit from Fermentas (MBI Fermentas, St Leon-Roth, Germany). The sequence analysis was performed on a LI-COR 4200 automatic sequencer (LICOR Inc, Lincoln, Nebraska USA). The regions amplified for the direct sequencing and for the cloning and sequencing are not 100% identical. Owing to slightly different primers, the latter approach does not include the first two CpG sites.

RNA isolation and quantitative real-time RT-PCR Total RNA was isolated from formalin-fixed paraffin-embedded bone marrow trephines exactly as described.16 For cDNA synthesis 1 mg of total RNA was transcribed using 200 U of SuperScript II RNase Reverse Transcriptase (Invitrogen, Karlsruhe, Germany) following the manufacturer’s protocol. For Leukemia


912 Table 1

Primers and probes sequences used in the study Name

Sequencesa

1M5 1M3 1U5 1U3 3M5 3M3 3U5 3U3 mod F2 mod R2 mod F1-t mod R1-t 1I5 1I3 2I5 2I3 3I5-t 3I3-t p15-exf p15-Exr Probe

GCGTTCGTATTTTGCGGTT AGCCAGCAAGCCAATAACATGC CGCGTTCGTATTTTGCGGT AGCCAGCAAGCCAATAACATGCC GAGTGTCGTTAAGTTTACGGT AAACCCTCATCGCTACCG GAAGAGTGTTGTTAAGTTTAT CCAAACCCTCATCACTACCA GGAGTTTAAGGGGGTGG CCTAAAAAACCTAAACTCAACT TGTAAAACGACGGCCAGTGGTTTTTTGGTTTAGTTGAAAA CAGGAAACAGCTATGACCAACTCAACTTCATTACCCTC TTTTTTGTCGGTTGGTTTTTTATTTTG CTCCCCATTACTTCAACTCAAATCCAAA GGGAGGGTAATGAAGTTGAGTTT CCAAAAACTATCRCACCTTCTCCA CAGGAAACAGCTATGACCGGAGGGTAATGAAGTTGAGTTGAGTTTAGGTT TGTAAAACGACGGCCAGTAATATCRCACCTTCTCCACTAATCCC ATCCCAACGGAGTCAACCG CGCTGCCCATCATCATGA FAM-TTCGGGAGGCGCGCGATC-TAMRA

M13–21

Forward

TGTAAAACGACGGCCAGT

M13-REV

Reverse

CAGGAAACAGCTATGACC

50.0

b-GUS

Forward Reverse Probe

CTCATTTGGAATTTTGCCGATT CCGAGTGAAGATCCCCTTTTTA FAM-TGAGTCACCGACGAGAGTGCTGG-TAMRA

60.0


GGTTCTTCCTCCTGGAGAATGTC GGGCCACGCCGTACTG FAM-CCTTCAAGCGCTCCATGGTCCTGAA-TAMRA

60.0


AGGAAGCGCAAGCAGCACCC ATTGGGTAATAGCTCTGAGGCG FAM-ACACGCCAAAGGACCCTGCGGT-TAMRA

60.0


CCAGCCCTGCGGCAG GTTGACGAGGATCGAGTCTTCC FAM-CCGGCGTCCTCCTCTCCATTGAG-TAMRA

60.0

Gene p15INK4B

DNMT1

DNMT3A

DNMT3B

a

58.0 62.0 54.0 54.0 54.0 52.0 58.0 60.0 65.0 60.0

All the sequences are listed in a 50 –30 order.

quantification of p15INK4b, and the DNA methyltransferases DNMT1, DNMT3A and DNMT3B mRNA levels a real-time PCR was performed using a 96-well tray and optical caps (Applied Biosystems, Weiterstadt, Germany) with a 25 ml final reaction mixture containing 250 nM each primer, 150 nM probe, 0.625 units of Platinum Taq (Invitrogen, Karlsruhe, Germany), 200 mM each of dATP, dCTP, dTTP and dGTP in 1 Platinum Taq reaction buffer and 4 ml cDNA. For primer and probe sequences see Table 1. The reaction mixture was preheated at 951C for 5 min, followed by 45 cycles at 951C for 15 s and 601C for 1 min. As a reference transcript for p15INK4b mRNA expression levels the housekeeping gene b-glucuronidase (bGUS) was used, for the normalization of the DNMT mRNA levels, the proliferation-associated cyclinD2 transcript was used.

Immunohistochemistry For immunohistochemical staining, deparaffinized sections were heated for 3 min at 1211C in 10 mM citrate buffer (pH 6) Leukemia

Annealing temperatures (1C)

using a laboratory autoclave. The anti-p15INK4b antibody Ab-6 (Clone 15P06) from NeoMarkers (Westinghouse, Fremont, CA, USA) was used in a 1:100 dilution and the detection of bound antibodies on the sections was performed exactly as described.17 Staining intensities were evaluated by two pathologists (FL, HK) without prior knowledge of the methylation status.

Statistical analysis On the basis of their blast count CMML class I cases, as defined by the WHO classification,11 were divided into two groups: lower blast count (p5%) and higher blast count (5–10%). Similarly, the cases were classified into two groups based on their degree of methylation. Cases with ‘’, ‘(+)’ or ‘+’ MSP band intensities (Figure 2) or methylated CpG sites of 10% or less upon sequencing were classified as ‘low methylation’. Those with ‘++’ or ‘+++’ MSP band intensities or more than 10% CpG sites methylation upon sequencing were classified as ‘high


913 192 184

U-Signal

M-Signal

124

+++

+++

++

(+)

-

+

-

Figure 2 Examples of methylation-specific PCR results. PCR products were separated on a polyacrylamide gel. Below each lane, the semiquantitative evaluation of the band intensities is indicated.

methylation’. The w2 test of independence was used and Pvalues less than 0.05 were regarded as significant.

Results

Methylation of the p15INK4b gene in myeloid cell lines and primary CMML samples The promoter region and Exon 1 of the p15INK4b gene is located in a typical CpG island, which is schematically depicted in Figure 1a. In cell lines and primary lymphocytes from healthy donors, a very close correlation between methylation of this CpG island and transcriptional silencing was observed. Whereas bone marrow trephines displaying only mild reactive changes are devoid of any methylation, the AML cell line KG1a contains a fully methylated p15INK4b island. Accordingly, in this cell line no p15INK4b transcript can be detected (Figure 1b), whereas in the control biopsies p15INK4b mRNA was readily detectable. Nearly all methylation studies of the p15INK4b gene described in the literature employ a methylation-specific PCR assay (MSP, Herman et al18) with the primer pair I. Using this primer pair 33% (11/33) of the CMML samples showed a product in the Mreaction of the conventional MSP (Figure 2). Under these reaction conditions, PBMC fractions from healthy volunteers and bone marrow trephines displaying only mild reactive changes were completely devoid of any M-signal (data not shown). Since Nguyen et al19 analyzed the methylation status of three individual CpG-dinucleotides located more downstream in the coding region (marked with small asterisks in Figure 1a), we designed a new MSP primer pair targeting this 30 region of Exon 1 as indicated in Figure 1a). Using this more 30 located primer pair II, 58% of the CMML samples (19/33) showed a clearly visible M-signal. INK4b

Detailed sequencing analysis of the p15 CMML bone marrow biopsies

homogenous methylation patterns seen in cell lines. Samples with strong M-signals in the conventional MSP always have several partially or fully methylated sites, such as samples #16 and #18. However, there are also samples with several fully methylated CpG dinucleotides, which are completely devoid of any M-signal in the conventional MSP assay such as samples #7 and #29. In the control samples only occasionally a low-level methylation of a few CpG sites was observed close to the 30 border of the region analyzed (see Figure 3). In addition to the direct sequencing of PCR products, for 15 selected cases, we also sequenced five individual clones per sample. This experimental approach confirmed the interindividual differences of methylation patterns both in proportion as well as location of methylated CpG sites detected by the direct sequencing (Figure 4). In addition, it revealed the marked intraindividual heterogeneity in more detail.

p15INK4b gene methylation and blast count According to the new WHO classification CMML cases with less than 10% blast count are grouped together as ‘CMML, class I’.11 Since nearly all of our samples belonged to CMML class I (31/ 33), we investigated whether the detection of epigenetic events might reveal distinct subgroups of CMML cases in this category. Our series of CMML class I samples was divided into cases with 5% blasts or less and samples with up to 10% blasts. A blast count of more than 5% correlated significantly with methylation in the 50 region (see Figure 1a) as revealed by MSP (w2 test, Po0.018) or sequencing of individual clones (w2 test, Po0.044). No correlation was found between blast count and methylation status of the more 30 located region.

p15INK4b expression in CMML gene in

In order to investigate the methylation patterns in more detail, bisulfite converted DNA samples were amplified with primers flanking the CpG island and the PCR products were sequenced directly following cycle sequencing protocols. The two regions covered by this experimental approach contain altogether 49 potential methylation (CpG) sites in the p15INK4b promoter region and within the first exon (see Figure 1a). The semiquantitative evaluation of the sequencing results are shown in Figure 3. The most obvious result is a high degree of inter- and intraindividual heterogeneity, which is in stark contrast to the

In order to elucidate the functional consequences of aberrant p15INK4b gene methylation expression of p15INK4b mRNA was analyzed using quantitative real-time RT-PCR. Despite the quite high variability of p15INK4b mRNA expression in the control group of 20 bone marrow trephines with only mild reactive changes (see Figure 5a, left column), a general reduction of p15INK4b mRNA expression in CMML is clearly discernible. The subgroup with the highest degree of p15INK4b gene methylation (‘CMML M+’) displayed the lowest expression levels, whereas the CMML samples without any methylation (‘CMML M’) had a median p15INK4b mRNA level more similar to the control group than the total CMML group. Leukemia


914 Control #1

Control #2 Control #3 Control #4 Control #6

#1 #2 #3 #4 #5 #6 #7 #8 #9 #10 #11 #12 #13 #14 #15 #16 #17 #18 #29 #20 #21 #22 #23 #24 #25 #26 #27 #28 #29 #30 #31 #32 #33

0%

1-10%

11-49%

50-89%

90-100%

Figure 3 Results of direct genomic sequencing after bisulfite conversion for all CMML samples and five controls. Each circle represents a potentially methylated cytosine residue. The marked interindividual heterogeneity is obvious. Since the direct bisulfite sequencing approach analyses a mixture of PCR products, for most potential methylation site a fractional methylation indicated by the symbols was detected.

Among the different methods employed for the detection of DNA methylation, the closest correlation between presence of methylated CpG sites and repression of transcription was revealed by the sequencing of individual clones from the more 50 located region. In addition to the measurement of the mRNA level, we assessed p15INK4b protein expression in a subset of samples using immunohistochemistry. All samples with a high degree of methylation showed strongly reduced or nearly complete absence of p15INK4b expression, whereas the samples with no or only very low levels of methylation displayed quite strong p15INK4b protein expression. Figure 5b shows two examples of bone marrow trephines with no methylation and a strong p15INK4b expression (case #25) and high degree of methylation and nearly complete absence of p15INK4b expression (case #16), respectively. Leukemia

Expression of DNA methyltransferases Since conflicting reports exist concerning the relation between gene hypermethylation in tumor cells and mRNA expression of DNA methyltransferases (see Robertson20 and references therein), we determined the mRNA expression of DNMT1, DNMT3A and DNMT 3B in all CMML samples (n ¼ 33). For this purpose, we employed a quantitative real-time RT-PCR assay. mRNA levels were normalized to the housekeeping gene b-GUS and the proliferation marker cyclinD2 in order to correct for proliferation-associated upregulation of DNA methyltransferases, which is not specific for chronic myelomonocytic leukemia.20 An upregulation of all three DNA methyltransferases could be observed in all CMML samples with a high degree of p15INK4b gene methylation (‘CMML M+’) when normalizing to the


915

#16 Clone 1 Clone 2 Clone 3 Clone 4 Clone 5





Unmethylated

Methylated

Figure 4 Sequencing analysis of individual clones after bisulfite treatment of genomic DNA. After cloning of the PCR products into an appropriate vector and transformation into bacteria individual clones were picked and sequenced. Therefore, every line represents the analysis of a single allele.

housekeeping gene b-GUS (data not shown). However, after correcting for proliferation-associated unspecific upregulation by normalizing to the proliferation marker cyclinD2, only for DNMT3A a specific and significant upregulation is seen (compare Figure 6b with Figure 6a and c).

Discussion Aberrant methylation of CpG islands is now a well-established epigenetic alteration in the development and progression of malignant neoplasia indicating transcriptional silencing.21 The detection of altered DNA methylation could be used for the detection of malignant transformed cells, and the potential prognostic or predictive values of altered DNA methylation is just emerging.22 However, the use of alterations in DNA methylation patterns in a given gene or a set of genes for diagnostic purposes will require a careful analysis of the methylation patterns in normal tissues and nonmalignant

physiological alterations (like inflammation) as well as a detailed assessment of methylation patterns in patient biopsies. The analysis of nucleic acids extracted from formalin-fixed paraffin-embedded bone marrow trephines enables a direct correlation between morphological assessment, methylation analysis and mRNA or protein measurement. In addition, the use of paraffin-embedded samples provides the opportunity to evaluate a large series of samples in a retrospective manner. With 33 morphologically and clinically confirmed cases of CMML, this is by far the largest study reported so far in the literature. The absence of methylation signals in the conventional MSP of some of the samples that contain clearly visible methylated sites upon sequencing, indicates that the widely used MSP may not be the optimal method to detect methylation, at least in the case of CMML. This could be because of the fact that MSP can only detect few CpG site differences (approx. 3–6 depending on the primers used) between the methylated and unmethylated DNA strands that are found only at the primer binding sites, whereas sequencing reveals all the potential methylation sites in Leukemia


916 400

a

rel. mRNA expression

p15INK4b 300

200

100

0

normal

CMML

CMML M+

CMML M-

b

Figure 5 Expression analysis of the p15INK4b gene in CMML. (a) Measurement of p15INK4b mRNA levels using quantitative real-time PCR methodology. The mean expression level of the control group was set to 100 and all individual expression levels were calculated to this mean using the DDCT-method.25 The horizontal bars indicate the median of each sample group. Transcript levels were normalized to b-GUS. (b) Immunohistochemical detection of the p15INK4b protein. In the right panel, a sample with no detectable methylation shows quite prominent p15INK4b staining, especially in the megakaryocytes. In the left panel, a sample with a high degree of methylation shows clearly reduced or absent p15INK4b staining.

the region. In addition, MSP is not able to detect the high degree of inter- and intraindividual heterogeneity of p15INK4b gene methylation found in this study for CMML. Homogenous methylation patterns as seen in cell lines (see Figure 1a) are completely different from the patterns found in patient biopsies, demonstrating convincingly that cell lines very often are no appropriate model systems for aberrant methylation, also in the case of CMML. Since all studies reporting p15INK4b gene methylation in MDS (including a few CMML samples) used MSP9,13,23 or restriction enzyme-based methods,8,12 which both target only a few potential methylation sites, the heterogeneity of methylation patterns has not yet been addressed in MDS. Only for AML samples, the heterogeneity of p15INK4b gene methylation patterns has been documented.4,5 During the preparation of this manuscript, Daskalakis et al14 also reported the detailed analysis of p15INK4b gene methylation in MDS by sequencing individual clones after bisulfite treatment of genomic DNA, but they examined only the 30 region of the CpG islands (see Figure 1b) and did not show any results for CMML samples. Methylation of the 30 region was more frequent, but it did not correlate with blast count in the new WHO category of CMML Leukemia

class I nor did it correlate with the disease progression (data not shown). Also the correlation between methylation status and transcriptional silencing was much better for the 50 region than for the 30 region. These data clearly indicate the importance of selecting the correct region for methylation studies. The correlation between blast count and methylation in the 50 region of the p15INK4b gene may indicate that the existing, primarily morphological classification could be improved or substantiated by including epigenetic alterations. The clinical significance of these findings has to be addressed in future studies. In ongoing studies we analyze the alteration of p15INK4b gene methylation during course of disease. Preliminary data demonstrate an interesting correlation (data not shown). The results presented in here concerning the heterogeneous methylation patterns have to be considered if fast and simple, but reliable assays for the detection of aberrant methylation in a clinical diagnostic setting will be developed. The heterogeneity of methylation patterns found in our CMML series supports the model put forward by Turker24 that silencing is a gradual and evolving process that involves promoter methylation as a rather late event creating variegated methylation patterns.


917

a

900

DNMT1


800 700

References

600 500 400 300 200 100 0

normal

b

CMML

CMML M+

CMML M-

900

DNMT3A

800


700 600 500 400 300 200 100

normal

c

CMML

CMML M+

CMML M-

900

DNMT3B

800 700


Grant numbers and sources of support: Deutsche Krebshilfe, Grant 10-1842-Le I.

600 500 400 300 200 100 0

normal

CMML

CMML M+

CMML M-

Figure 6 mRNA expression of DNA methyltransferases DNMT1 (a), DNMT3A (b) and DNMT3B (c) in CMML employing quantitative real-time PCR methodology. The mean expression level of the control group was set to 100 for every transcript and all individual expression levels were calculated to this mean using the DDCT-method.25 The horizontal bars indicate the median of each sample group. The expression levels shown were normalized to the proliferationassociated transcript cyclinD2, thereby correcting for proliferationinduced changes in DNA methyltransferase expression.

Acknowledgements We thank Britta Hasemeier for expert assistance in preparing the figures as well as Henriette Bruchhardt and Reinhard von Wasielewski for assistance and advice, respectively, concerning the immunohistochemical detection of the p15INK4b protein.

1 Jones PA, Laird PW. Cancer epigenetics comes of age. Nat Genet 1999; 21: 163–167. 2 Obaya AJ, Sedivy JM. Regulation of cyclin-Cdk activity in mammalian cells. Cell Mol Life Sci 2002; 59: 126–142. 3 Esteller M. CpG island hypermethylation and tumor suppressor genes: a booming present, a brighter future. Oncogene 2002; 21: 5427–5440. 4 Dodge JE, List AF, Futscher BW. Selective variegated methylation of the p15 CpG island in acute myeloid leukemia. Int J Cancer 1998; 78: 561–567. 5 Aggerholm A, Guldberg P, Hokland M, Hokland P. Extensive intraand interindividual heterogeneity of p15INK4B methylation in acute myeloid leukemia. Cancer Res 1999; 59: 436–441. 6 Heaney ML, Golde DW. Myelodysplasia. N Engl J Med 1999; 340: 1649–1660. 7 Quesnel B, Fenaux P. P15INK4b gene methylation and myelodysplastic syndromes. Leukemia Lymphoma 1999; 35: 437–443. 8 Preisler HD, Li B, Chen H, Fisher L, Nayini J, Raza A et al. P15INK4B gene methylation and expression in normal, myelodysplastic, and acute myelogenous leukemia cells and in the marrow cells of cured lymphoma patients. Leukemia 2001; 15: 1589– 1595. 9 Tien HF, Tang JH, Tsay W, Liu MC, Lee FY, Wang CH et al. Methylation of the p15(INK4B) gene in myelodysplastic syndrome: it can be detected early at diagnosis or during disease progression and is highly associated with leukaemic transformation. Br J Haematol 2001; 112: 148–154. 10 Bennett JM, Catovsky D, Daniel MT, Flandrin G, Galton DA, Gralnick HR et al. Proposals for the classification of the myelodysplastic syndromes. Br J Haematol 1982; 51: 189–199. 11 Harris NL, Jaffe ES, Diebold J, Flandrin G, Muller-Hermelink HK, Vardiman J et al. The World Health Organization classification of hematological malignancies report of the Clinical Advisory Committee Meeting, Airlie House, Virginia, November 1997. Mod Pathol 2000; 13: 193–207. 12 Uchida T, Kinoshita T, Nagai H, Nakahara Y, Saito H, Hotta T et al. Hypermethylation of the p15INK4B gene in myelodysplastic syndromes. Blood 1997; 90: 1403–1409. 13 Quesnel B, Guillerm G, Vereecque R, Wattel E, Preudhomme C, Bauters F et al. Methylation of the p15(INK4b) gene in myelodysplastic syndromes is frequent and acquired during disease progression. Blood 1998; 91: 2985–2990. 14 Daskalakis M, Nguyen TT, Nguyen C, Guldberg P, Kohler G, Wijermans P et al. Demethylation of a hypermethylated P15/ INK4B gene in patients with myelodysplastic syndrome by 5-Aza20 -deoxycytidine (decitabine) treatment. Blood 2002; 100: 2957– 2964. 15 Lehmann U, Hasemeier B, Lilischkis R, Kreipe H. Quantitative analysis of promoter hypermethylation in laser-microdissected archival specimens. Lab Invest 2001; 81: 635–638. 16 Bock O, Kreipe H, Lehmann U. One-step extraction of RNA from decalcified and archival biopsies. Anal Biochem 2001; 295: 116– 117. 17 von Wasielewski R, Mengel M, Gignac S, Wilkens L, Werner M, Georgii A. Tyramine amplification technique in routine immunohistochemistry. J Histochem Cytochem 1997; 45: 1455– 1459. 18 Herman JG, Graff JR, Myohanen S, Nelkin BD, Baylin SB. Methylation-specific PCR: a novel PCR assay for methylation status of CpG islands. Proc Natl Acad Sci USA 1996; 93: 9821– 9826. 19 Nguyen C, Liang G, Nguyen TT, Tsao-Wei D, Groshen S, Lubbert M et al. Susceptibility of nonpromoter CpG islands to de novo methylation in normal and neoplastic cells. J Natl Cancer Inst 2001; 93: 1465–1472. 20 Robertson KD. DNA methylation, methyltransferases, and cancer. Oncogene 2001; 20: 3139–3155. Leukemia


918 21 Wajed SA, Laird PW, DeMeester TR. DNA methylation: an alternative pathway to cancer. Ann Surg 2001; 234: 10–20. 22 Widschwendter M, Jones PA. The potential prognostic, predictive, and therapeutic values of DNA methylation in cancer. (Commentary re: J Kwong et al, Promoter hypermethylation of multiple genes in nasopharyngeal carcinoma. Clin Cancer Res 2002; 8: 131–137, and H-Z Zou et al, Detection of aberrant p16 methylation in the serum of colorectal cancer patients. Clin Cancer Res 2002; 8: 188–191, Clin Cancer Res 2002; 8: 17–21.

Leukemia

23 Aoki E, Uchida T, Ohashi H, Nagai H, Murase T, Ichikawa A et al. Methylation status of the p15INK4B gene in hematopoietic progenitors and peripheral blood cells in myelodysplastic syndromes. Leukemia 2000; 14: 586–593. 24 Turker MS. Gene silencing in mammalian cells and the spread of DNA methylation. Oncogene 2002; 21: 5388–5393. 25 Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) method. Methods 2001; 25: 402–408.