Frequent epigenetic inactivation of Rb1 in addition to p15 and p16 in mantle cell and follicular lymphoma

Human Pathology (2007) 38, 1849–1857

www.elsevier.com/locate/humpath

Original contribution

Frequent epigenetic inactivation of Rb1 in addition to p15 and p16 in mantle cell and follicular lymphomaB C.S. Chim MBChB, MD, PhD, FRCP, FACP a,*, K.Y. Wong BSc b , F. Loong, MBBS, FRCPA b , W.W. Lam MBBS, FRCPA c , G. Srivastava PhD b,* a

Department of Medicine, Queen Mary Hospital, The University of Hong Kong, Pokfulam, Hong Kong Department of Pathology, Queen Mary Hospital, The University of Hong Kong, Pokfulam, Hong Kong c Department of Pathology, Princess Margaret Hospital, Hong Kong b

Received 12 January 2007; revised 3 May 2007; accepted 3 May 2007

Keywords: Mantle cell lymphoma; Follicular lymphoma; Cell cycle; p15; p16; Rb1; Hypermethylation

Summary Dysregulation of cell cycle control is an important mechanism in carcinogenesis. Gene promoter hypermethylation is an alternative mechanism of gene inactivation. We analyzed the methylation status of the tumor suppressor components of the INK4/Rb pathway in mantle cell lymphoma and follicular lymphoma by methylation-specific polymerase chain reaction for p15, p16, p18, and Rb1 in 23 mantle cell lymphoma and 30 follicular lymphoma cases and lymphoma cell lines. The methylationspecific polymerase chain reaction results showed that in mantle cell lymphoma, frequent p16 (82%) but infrequent p15 (8.7%) or Rb1 (17.4%) hypermethylation occurred, with p16 and Rb1 hypermethylation being mutually exclusive (P = .01). In follicular lymphoma, frequent hypermethylation of p15 (36.7%), p16 (56.7%), and Rb1 (43.3%) occurred, with p15 and Rb1 hypermethylation being mutually exclusive (P = .05). Concurrent methylation of p15 and p16 occurred in 26.7% of patients with follicular lymphoma and 8.7% of patients with mantle cell lymphoma. Compared with mantle cell lymphoma, there was more frequent p15 (P = .025) hypermethylation but comparable Rb1 (P = .07) and p16 (P = .07) hypermethylation in follicular lymphoma. In a patient with follicular lymphoma with sequential biopsies, Rb1 was unmethylated and expressed at diagnosis but became methylated and down-regulated at relapse. Moreover, methylation analysis of these 4 genes in an additional 8 patients with grade I follicular lymphoma showed that Rb, but not the other genes, was preferentially methylated in grade II (P = .03). In summary, most patients with mantle cell lymphoma and follicular lymphoma had epigenetic aberrations targeting the INK4/Rb pathway. There is more frequent p16 hypermethylation in mantle cell lymphoma and p15 or Rb1 hypermethylation in follicular lymphoma. The role of Rb methylation in disease or histologic transformation in follicular lymphoma warrants further study. © 2007 Elsevier Inc. All rights reserved.

B

There is no prior or duplicate publication or submission elsewhere of any part of the work that has been included in the article. Approval for this research has been obtained from the institutional review board. ⁎ Corresponding authors. C.S. Chim is to be contacted at Department of Medicine, Queen Mary Hospital, Hong Kong. G. Srivastava, University Pathology, Queen Mary Hospital, Hong Kong. E-mail addresses: [email protected] (C. S. Chim), [email protected] (G. Srivastava). 0046-8177/$ – see front matter © 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.humpath.2007.05.009

1. Introduction Mantle cell lymphoma (MCL) and follicular center cell lymphoma (FL) are 2 forms of small B-cell lymphomas derived from neoplastic transformation of their normal counterparts in the mantle zone and germinal center, respectively [1]. MCL is characterized in virtually all cases

1850 by t(11;14), with dysregulation of the cell cycle by upregulation of the cyclin D1 gene located at 11q13 [1,2]. In contrast, FL is characterized by t(14;18), with up-regulation of BCL-2 located at 18q21, which confers resistance to apoptosis on the malignant lymphoid cells [1-3]. Cellular proliferation is mediated by progression through the cell cycle, with 2 major cell cycle checkpoints located at G1S and G2M [4-6]. Rb1 is a nucleoprotein important in the suppression of G1S cell cycle progression. This is mediated by the sequestration of E2F by Rb1 in quiescent cells, a transcription factor for genes important in G1S progression [7]. Quiescent cells in G0 phase contain hypophosphorylated Rb1, which sequesters the transcription factor E2F. Upon activation of the cell by mitogens, up-regulation of D-type cyclins results in activation of cyclin-dependent kinases (CDKs) 4 and 6 and the resulting hyperphosphorylation of Rb1. E2F is then released, inducing the transcription of S1-specific genes and an irreversible commitment to cell cycle progression [5,6]. Regulatory elements inhibitory to cell cycle progression include the INK4 (p15, p16, p18, and p19) and the CIP/KIP (p21CIP, p27KIP1 , p57KIP2 ) families of proteins that give rise to cell cycle arrest [8,9]. The INK4 family are CDK inhibitors (CKIs) that share similar functional domains (ankyrin repeats for protein-protein interactions). These domains allow the INK4 family to compete with cyclin D for binding to CDKs 4 and 6, thus preventing CDKs 4 and 6 activation and cell cycle progression [8]. Therefore, various mechanisms exist to inhibit or inactivate Rb1 function, thereby resulting in unchecked cellular proliferation, in cancers. These include Rb1 inactivation by genetic and epigenetic mechanisms [7], overexpression of cyclin D1, and activating mutations of CDK 4. DNA methylation, catalyzed by DNA methyltransferase, involves the addition of a methyl group to the carbon 5 position of the cytosine ring in a CpG dinucleotide, leading to a conversion to methylcytosine [10,11]. In many hemic cancers, the CpG islands of selected genes are aberrantly hypermethylated, resulting in repression of transcription of these genes. Hypermethylation thus serves, in addition to mutation and deletion, as an alternative mechanism of gene inactivation [10,11]. The role of Rb1 in carcinogenesis has been illustrated by familial retinoblastoma, in which patients inheriting loss of 1 Rb1 allele develop retinoblastoma at an early age by inactivating mutation of the other Rb1 allele. Moreover, sporadic mutation of Rb1 has been demonstrated in cancers other than retinoblastoma [7], indicating that the tumor suppressor role of Rb1 is not restricted to the retina [12]. Moreover, G 1S cell cycle progression may also be dysregulated by inactivation of the members of the INK4 family of CKI [13]. In this study, we investigated if methylation of the INK4 family of CKIs and Rb1 might contribute to lymphomagenesis of MCL and FL. p19 was not included in the analysis because tumor suppressor activity

C. S. Chim et al. has not been demonstrated in transgenic mice deficient in p19 [14].

2. Materials and methods 2.1. Patients and diagnosis Diagnosis of MCL and FL was made according to standard criteria [1,2]. Patients were staged according to the Ann Arbor system. All patients underwent complete staging including full blood count, serum biochemistry, serum lactate dehydrogenase level, computerized tomographic scan of thorax and abdomen, and bilateral bone marrow trephine biopsies. Immunophenotyping was performed on cryostat sections and paraffin sections with standard immunoperoxidase technique. Paraffin sections of formalin- or B5-fixed tissue were stained with hematoxylin-eosin to confirm the diagnosis of lymphoma and examined for the expression of B- and T-cell markers. The panel of antibodies used included CD3 (Leu4; Becton Dickinson, San Jose, CA), CD3 (polyclonal; Dako, Glostrup, Denmark), CD5 (Dako), CD10 (J5; Coulter, Hialeah, FL), CD19 (Leu 12; Becton Dickinson), CD20 (L26; Dako), CD22 (Dako), CD23 (Dako), and cyclin D1 (Zymed, San Francisco, CA). The clinicopathologic features of the patients with MCL have been previously reported [2]. All 23 cases of MCL were classic-variant, with the phenotype of CD5+, CD10−, and CD23−, and immunoreactive for cyclin D1. Five patients were positive for nested polymerase chain reaction (PCR) for BCL-1 gene rearrangement [2]. The 30 patients with FL were FL grade II according to the World Health Organization classification [1]. There were 23 patients (18 male and 5 female) with MCL, with a median age of 66 years (range, 49-75 years), and 30 patients (21 male and 9 female) with FL, with a median age of 57 years (range, 33-88 years).

2.2. Samples and DNA preparation DNA was extracted from frozen lymph node biopsies of 30 patients with grade II FL and 23 patients with MCL. To study the role of methylation of these genes in histologic transformation of FL, DNA was obtained from an additional 8 patients with grade I FL. We also isolated DNA from the mononuclear cells of peripheral blood (PBMC)s of 10 healthy volunteers. Normal lymph node biopsies from 6 patients were used as negative control, whereas methylated control DNA (CpGenome Universal Methylated DNA; Intergen, New York, NY) was used as positive control in all the experiments. We also examined 3 MCL cell lines (Granta-519, Jeko-1, Mino) and 2 FL cell lines (SUDHL6, DHL16). MCL cell lines Granta-519, Jeko-1, and Mino were provided by Raymond Lai (Department of Laboratory Medicine and Pathology, University of Alberta and Cross

Frequent epigenetic inactivation of Rb1 Cancer Institute, Walter MacKenzie Health Sciences Center, Edmonton, Canada). FL cell lines SUDHL6 and DHL16 were obtained from the German National Resource Centre for Biological Material (DSMZ, Braunschweig, Germany).

2.3. Bisulfite treatment of genomic DNA Genomic DNA was extracted from fresh frozen tissues using the proteinase K digestion and phenol-chloroform method [15]. Five micrograms of genomic DNA was denatured by adding 33 μL of 0.3 mol/L NaOH at 37°C for 15 minutes. Denatured DNA was mixed with 333 μL of bisulfite solution and treated in darkness for 4 hours at 55°C. The bisulfite solution was prepared as 2.4 mol/L sodium metabisulfite (pH 5.0-5.2) (Sigma Chemical, Co., St. Louis, MO)/0.5 mmol/L hydroquinone (Sigma). Treated DNA was desalted and purified using the Qiaex II kit (Qiagen, Hilden, Germany). Recovered DNA was dissolved in 100 μL of TE buffer (pH 8.0) and stored at −20°C.

1851

2.4. Methylation-specific polymerase chain reaction The methylation-specific polymerase chain reaction (MSP) for gene promoter methylation was performed as previously described [16-18]. The bisulfite-treated DNA was PCR-amplified with strand-specific primers, and the PCR products were analyzed by electrophoresis on 10% polyacrylamide gels, stained with ethidium bromide, and visualized under ultraviolet illumination. The primer sequences used for p15, p16, p18, and Rb1 were previously reported [18-20]. The Rb1 promoter organization and the location of primers are shown in Fig. 1.

2.5. DNA sequencing The identity of the methylated sequences was confirmed by automated DNA sequencing. The PCR products were gelpurified, sequenced bidirectionally (ABI Prism dRhodamine

Fig. 1 (A) Sequence of the CpG island including the Rb1 promoter and its first exon. MSP primers (Rbmet, Rbunmet, and Rbcom) are underlined with a solid bar. The first exon is defined by the 2 filled triangles. The translation start codon is bold, and the binding sites for Sp1 and E2F are boxed. CpG sites are bold and numbered according to the order used in bisulfite genomic sequencing. Primer sequences are as follows: Rbmet, 5′-GCGTTTTAGTTCGCGTATCGATTAGCGTTTTAG-3′; Rbunmet, 5′-TGGTGGGTTTGGGAGTTTTGTGGATGTGATGTT-3′; and Rbcom, 5′-CCTACCCCRACTCCCRTTACAAAAATAATTTCAAC-3′. R indicates that adenine and guanine were added during synthesis of primer Rbcom in equal amounts to avoid preferentially binding to methylated or unmethylated template. The positions of the 201-bp PCR fragment representing the methylated allele and 154-bp PCR fragment representing the unmethylated allele are indicated. (B) Location of the 61 CpG sites analyzed in Rb1 gene promoter. Each vertical bar represents 1 CpG site.

F1

1852 Terminator Cycle Sequencing Kit, PE Biosystem), and analyzed on an automated DNA sequence analyzer (377 ABI Prism; PE Biosystem, Foster City, CA).

dinucleotide remained as “C,” whereas unmethylated cytosine read as “T” after bisulphite conversion. The methylation status of the cell lines is shown in Table 1.

2.6. Rb1 immunohistochemistry

3.2. MSP in primary lymphoma samples

In brief, after deparaffinization, sections were heated for 10 minutes in 10 nmol/L sodium citrate buffer (pH 6.0) in a scientific microwave oven. The primary Rb1 antibody (Clone G3-245; BD Biosciences Pharmingen, San Diego, CA) was used at 1:50 dilution and incubated with tissue sections for 1 hour. The secondary antibody and peroxidase steps were carried out using a commercially available kit according to the manufacturer's protocol (LSAB2 kit; Dako Ltd, Buckinghamshire, UK). Localization of antibody binding was visualized using 3′,3′-diaminobenzidine as chromogen (Nichirei, Tokyo, Japan) with hematoxylin counterstaining. Negative controls were omission of the primary antibody. Only nuclear positivity was assessed, and cytoplasmic staining was regarded as nonspecific [21].

Methylation of p15, p16, and Rb1 was illustrated in Table 2 and Fig. 3. Most samples harbored methylation of at least 1 of these 3 genes. To study if methylation might be involved in the histologic progression of disease, an additional 8 cases of grade I FL were studied for methylation of p15, p16, p18, and Rb. We showed that p15 was methylated in 1 (12.5%) grade I and 11 (36.7%) grade II FLs (P = .39), p16 in 3 (37.5%) grade I and 17 (56.7%) grade II, and Rb in none of grade I and 43.3% of grade II cases (P = .03) (Fig. 4, Table 3).

2.7. Statistics Association between p15, p16, and Rb1 hypermethylation and (1) categorical variables including sex, stage at diagnosis (limited, stage I/II; advanced, stage III/IV), and complete remission (CR) and (2) continuous variables, such as age, were studied by χ2/Fisher exact tests and Student t test. CR was defined as complete resolution of symptoms and signs and normalization of all imaging studies. Overall survival (OS) was measured from the date of diagnosis to date of death or last follow-up. OS was computed by the KaplanMeier method. The OS of patients with and without p16 methylation was studied. Survival curves were compared by the log-rank test. All P values were 2-sided.

3. Results 3.1. MSP in controls and cell lines

F5

C. S. Chim et al.

None of the 4 genes tested (p15, p16, p18, and Rb1) was methylated in 10 normal PBMCs from healthy volunteers and 6 normal lymph node biopsies from 6 patients. Unmethylated status was demonstrated by positive amplification in unmethylated primers (U-MSPs) and lack of amplification in methylated primers (M-MSP) (Figs. 2A and 3A). Methylated positive control DNA showed complete methylation by positive amplification in M-MSP but not in U-MSP. The specificity of the MSP for Rb1 was shown by DNA sequencing of the methylated MCL cell line, Granta519, a primary FL biopsy (F18), and the normal PBMC sample (Fig. 2B). The DNA sequence of methylated control DNA is aligned and compared with the germline sequence of wild-type DNA. Methylated cytosine residues in CpG

T4

T4 F5

F5 T4

3.3. Rb1 expression in reactive tonsils and primary lymphoma samples In reactive tonsils, intense nuclear Rb1 expression was demonstrated in germinal centers but not in the mantle zone (Fig. 5A). In 1 patient with FL who had serial lymph node biopsy, Rb1 protein was expressed in the lymphoma cells at diagnosis, in which Rb1 was unmethylated. He achieved CR with combination chemotherapy but relapsed 2 years later. Lymph node biopsy at relapse showed low Rb1 expression in the lymphoma cells, which acquired Rb1 hypermethylation at relapse (Fig. 5B and C).

F5

F5

3.4. Association of p15, p16, and Rb1 hypermethylation with clinical parameters and treatment outcome In both MCL and FL, there was no association between p15, p16, and Rb1 hypermethylation with age, sex, and stage of disease (limited, stage I/II; advanced, stage III/IV) (Table 4). Moreover, in MCL, there was no association of p15 (P = .99), p16 (P = .99), and Rb1 (P = .99) hypermethylation with CR rate, and no association of p16 (P = .18) and Rb1 hypermethylation (P = .88) with projected OS. However, this might be limited by the small sample size, and thus, the prognostic value of p16 hypermethylation warrants further analysis in prospective studies with a larger number of patients. Correlation of p15 hypermethylation and OS was not computed because there were only 2 patients with p15 hypermethylation. Similarly, in FL, there was no association of p15 (P = .99), p16 (P = .21), and Rb1 (P = .70) hypermethylation with CR rate, and no association of p15 (P = .30), p16 (P = .57), and Rb1 (P = .17) hypermethylation with projected OS. In the 23 patients with MCL lymphoma, median OS in patients with unmethylated p16 (n = 4) is not reached, and the median OS of patients with methylated p16 (n = 19) is 65.1 months (P = .17).

T4

Frequent epigenetic inactivation of Rb1

4. Discussion The Rb1 gene is frequently methylated in both FL and MCL. Methylation of the Rb1 promoter has been shown, in other cancers, to contribute to carcinogenesis [7]. Rb1 expression is found to be frequently down-regulated in 53% to 59% of MCLs

1853 [22,23] and 48% to 83% of FLs [22,24]. Rb1 is demonstrated to be frequently methylated in both types of lymphoma in this study, so promoter hypermethylation appears to be a potential mechanism leading to down-regulation of the Rb1 gene. Our study also showed frequent methylation of p16 in MCL. The frequent p16 but not p15 hypermethylation in our

Fig. 2 (A) MSP for Rb1 gene promoter methylation. MW indicates molecular weight markers; N1 to N10, DNA from PBMC of 10 healthy volunteers; LN1 to LN6, 6 normal lymph node specimens, showing the absence of amplification for the Rb1 methylated primers M-MSP but positive amplification for the Rb1 U-MSPs; UMD, positive control for methylated DNA, showing amplification for the M-MSP but not the U-MSP Rb1 primers; H2O, blank; M1 to M23, 23 primary MCL cases; F1 to F30, 30 FL cases. F9R is the sequential lymph node biopsy of case F9. (B) DNA sequence of M-MSP band for Rb1 gene promoter methylation of Granta-519 MCL cell line, a primary FL biopsy specimen (F18), and normal PBMC (N1). DNA sequence is aligned and compared with germline sequence of the wild-type DNA (WT). Methylated cytosine residues in CpG dinucleotide remained as ‘‘CC‘‘, whereas unmethylated cytosine read as ‘‘T’’ after bisulfite conversion. Methylated cytosines were underlined.

1854

C. S. Chim et al.

Fig. 3 Representative results of the MSP for p15, p16, and p18 gene promoter methylation: 1 MCL (Granta-519) and 2 FL (SUDHL6 and DHL16) cell lines. 4 primary MCL and 4 FL cases are shown. MW indicates molecular weight markers; N1 to N2, 2 samples of normal PBMC showing the absence of amplification for the Rb1 M-MSPs but positive for the Rb1 U-MSPs; H2O, blank.

patients contrasted with a recent report of frequent p15 but infrequent p16 hypermethylation in MCL [25]. There are several ways to account for this, including technical reasons, specimen source, histologic variation, and ethnic differences. However, the MSP primers used in this study and those used by Hutter et al [25] were the same, and the PCR conditions were similar. In the study by Hutter et al [25], the methylation status of p15 and p16 was determined in 71 patients with MCL, with 36 peripheral blood buffy coat and 35 lymph node specimens. Whether DNA was extracted from frozen tissues or paraffin-embedded specimens was not described. In Table 1 Methylation status of p15, p16, p18, and Rb1 gene promoters in MCL and FL cell lines Cell lines MCL cell lines Granta-519 Mino Jeko-1 FL cell lines SUDHL6 DHL16

p15

p16

p18

Rb1

U U U

del M/U M/U

U U U

M/U U U

U U

M M

U U

U M/U

U indicates unmethylated; del, homozygously deleted; M/U, hemizygously methylated; M, methylated.

contrast, in this study, high-molecular-weight genomic DNA was extracted from frozen lymph node specimens. Moreover, all 23 MCL cases in the current study were classic variants, but the number of histologic variants was not specified in the study by Hutter et al [25]. Furthermore, in the 3 MCL cell lines tested in this study (Table 1), p15 was completely unmethylated, consistent with the infrequent p15 hypermethylation in our primary MCL samples. Whether the disparity of results is due to ethnic difference remains to be

Table 2 Methylation of p15, p16, and Rb1 in primary MCL and FL samples

p15 p16 Rb Absence of methylation of any gene At least 1 gene methylated Concurrent methylation of 2 genes Concurrent methylation of 3 genes

MCL, n (%)

FL, n (%)

P

2 (8.7) 19 (82.6) 4 (17.4) 2 (8.7)

11 17 13 4

.025 .07 .07

21 (91.3) 4 (17.4)

26 (86.7) 13 (43.3)

0 (0)

(36.7) (56.7) (43.3) (13.3)

1 (3.3)

T1

Frequent epigenetic inactivation of Rb1

1855

Fig. 4 MSP for p15, p16, p18, and Rb1 in an additional 8 grade I FLs (F31-38). M-MSP of p15 showed methylation in sample F36 only; M-MSP of p16 showed methylation in samples F33, F35, and F38; M-MSP of p18 and Rb1 showed that all samples were unmethylated. U-MSP of p16 and Rb1 confirmed integrity of DNA samples. MW indicates molecular weight markers; UMD, positive control for methylated DNA, showing amplification for the M-MSP but not the U-MSP Rb1 primers; H2O, blank; M, amplification of methylated allele; U, amplification of unmethylated allele.

clarified. Finally, in view of previous reports that p16 mutation is associated with an inferior prognosis in MCL [26], the OS in patients with and without p16 methylation was studied. Despite that OS in patients with unmethylated p16 appeared superior to that in patients with methylated p16, the difference did not reach statistical significance because of the small number of patients. The prognostic impact of p16 methylation warrants further investigation in a prospective study with a larger number of patients. In 1 of our patients with FL, Rb1 hypermethylation was absent in the diagnostic lymph node, and Rb1 protein was expressed in most tumor cells. At relapse, a low level of Rb1 expression was demonstrated, which was accompanied by the presence of Rb1 hypermethylation in the lymphoma Table 3

Frequency of methylation in grade I and II FL

Histologic grade p15 Grade Grade p16 Grade Grade p18 Grade Grade Rb1 Grade Grade

Gene

P

Unmethylated

Methylated

I II

7 19

1 11

.19

I II

5 13

3 17

.44

I II

8 30

0 0

I II

8 17

0 13

.03

Fig. 5 Rb immunohistochemistry in reactive tonsil. Geminal center B cells are positive for Rb (A), whereas mantle zone B cells show negative staining (original magnification, ×100). (B) FL (case 9) showing strong nuclear reactivity in most neoplastic follicular center cells with no Rb methylation (original magnification, ×200). (C) FL (case F9R) sequential lymph node biopsy of case 9, showing weak nuclear reactivity but with Rb gene methylated (original magnification, ×200).

cells, consistent with acquisition of Rb1 hypermethylation during disease evolution. Moreover, methylation analysis of these 4 genes in 8 grade I FLs showed that Rb, but not the

1856

C. S. Chim et al.

Table 4 Association between p15, p16, and Rb1 gene promoter methylation and clinical parameters in patients with MCL and FL A. Association between p15, p16, and Rb1 gene promoter methylation and clinical parameters in 23 patients with MCL

p15 Age, mean (y) Sex Male Female Stage Advanced Limited p16 Age, mean (y) Sex Male Female Stage Advanced Limited Rb1 Age, mean (y) Sex Male Female Stage Advanced Limited

Methylated

Unmethylated

P

73.5

65.3

.15

2 0

16 5

.99

1 1

17 4

.39

65.0

.78

15 4

3 1

.99

15 4

3 1

.99

65.2

.28

3 1

15 4

.99

3 1

15 4

.99

66.2

69.7

B. Association between p15, p16, and Rb1 gene promoter methylation and clinical parameters in 30 patients with FL

p15 Age, mean (y) Sex Male Female Stage Advanced Limited p16 Age, mean (y) Sex Male Female Stage Advanced Limited Rb1 Age, mean (y) Sex Male Female Stage a Advanced Limited a

Methylated

Unmethylated

P

59.2

57.2

.71

7 4

14 5

.69

8 3

11 5

.99

54.3

.21

13 4

8 5

.44

13 3

6 5

.21

53.7

61.2

.14

10 3

11 6

.69

8 3

11 5

.99

60.7

No data were available for the staging of 3 patients with FL.

other 3 genes, is preferentially methylated in grade II FL (P = .03). Therefore, the role of Rb in disease progression or histologic transformation in FL warrants further study. However, because p15 and p16 were frequently involved in disease progression, methylation analysis of these genes warrants more extensive analysis. Overall, only very few patients (2 [8.7%] patients with MCL and 4 [13.3%] patients with FL) did not harbor hypermethylation of any of these 3 genes, suggesting that epigenetic alteration of the tumor suppressor components of the INK4/Rb pathway is very common in both types of lymphoma. MCL is characterized by cyclin D1 overexpression, which predisposes to unchecked cell cycle progression. Data from our study suggest an important role of the disruption of additional components of the cell cycle regulatory machinery in the pathogenesis of disease. The importance of multiple hits of the cell cycle control circuitry in lymphomagenesis of MCL has been illustrated by the presence of concurrent deletion of p15 and p16 or simultaneous p15 and p16 hypermethylation in MCL [27]. In FL, although the pathogenesis of disease is generally associated with the resistance to apoptosis due to upregulation of BCL-2, our data suggest that concurrent dysregulation of cell cycle control by gene hypermethylation is an important component in the pathogenesis of disease. Frequent hypermethylation of p15 and p16 has indeed been demonstrated in FL through methylation array studies [28]. Although tumor suppressor activity has been demonstrated for p18 [29], extensive investigation did not reveal any mutations or chromosomal rearrangements of this gene in hematologic cancers [30]. The absence of p18 methylation in both MCL and FL found here supports a minor role for this protein in the pathogenesis of these 2 lymphoma subtypes. In summary, frequent and concurrent methylation of p15, p16, and Rb1 occurs in both MCL and FL. The resulting epigenetic dysregulation of cell cycle regulation may therefore cooperate with up-regulation of cyclin D1 in MCL and up-regulation of BCL-2 in FL in the pathogenesis of these diseases. The role of Rb hypermethylation in disease progression or histologic transformation in FL warrants further study.

Acknowledgments We thank Dr Raymond Lai, Department of Laboratory Medicine and Pathology, University of Alberta and Cross Cancer Institute, 4B1 Walter MacKenzie Health Sciences Center, 8440 112 St, Edmonton, AB, Canada, for providing us the MCL cell lines Granta-519, Jeko-1, and Mino for this study. In addition, we would like to thank Richard Ford, Department of Hematopathology, University of Texas MD Anderson Cancer Center (Houston, TX) for developing the Mino cell line. We also express our sincere gratitude to all medical and nursing staff in the Department of Medicine,

Frequent epigenetic inactivation of Rb1 Queen Mary Hospital, Pokfulam, Hong Kong, for the provision of expert medical care.

References [1] Harris NL, Jaffe ES, Diebold J, et al. World Health Organization classification of neoplastic diseases of the hematopoietic and lymphoid tissues: report of the Clinical Advisory Committee meeting. J Clin Oncol 1999;17:3835-49. [2] Chim CS, Chan ACL, Choo CK, Kwong YL, Lie A, Liang R. Mantle cell lymphoma in Chinese: clinicopathological features and treatment outcome. Am J Hematol 1998;59:295-301. [3] Sanchez-Beato M, Sanchez-Aguilera A, Piris MA. Cell cycle deregulation in B-cell lymphomas. Blood 2003;101:1220-35. [4] Hann WC, Weinberg RA. Rules for making human tumor cells. N Engl J Med 2002;347:1593-603. [5] Malumbres M, Barbacid M. To cycle or not to cycle: a critical decision in cancer. Nat Rev Cancer 2001;1:222-31. [6] Sherr CJ. The Pezcoller lecture: cancer cell cycles revisited. Cancer Res 2000;60:3689-95. [7] Classon M, Harlow E. The retinoblastoma tumor suppressor in development and cancer. Nat Rev Cancer 2002;2:910-1017. [8] Roussel MF. The INK4 family of cell cycle inhibitors in cancer. Oncogene 1999;18:5311-7. [9] Jeffey PD, Tong L, Pavletich NP. Structural basis of inhibition of CDKcyclin complexes by INK4 inhibitors. Genes Dev 2000;14:3115-25. [10] Herman JG, Baylin SB. Gene silencing in cancer in association with promoter hypermethylation. N Engl J Med 2003;349:2042-54. [11] Chim CS, Liang R, Kwong YL. Hypermethylation of gene promoters in hematologic neoplasia. Hematol Oncol 2002;20:167-76. [12] Weinberg RA. The retinoblastoma gene and gene product. In: Levine AJ, editor. Tumor suppressor genes, the cell cycle and cancer. New York: Cold Spring Harbor Laboratory Press; 1992. p. 43-57. [13] Drexler HG. Review of alterations of the cyclin-dependent kinase inhibitor INK4 family genes p15, p16, p18 and p19 in human leukemia-lymphoma cells. Leukemia 1998;12:845-59. [14] Zindy F, van Deursen J, Grosveld G, Sherr CJ, Roussel MF. INK4ddeficient mice are fertile despite testicular atrophy. Mol Cell Biol 2000; 20:372-8. [15] Maniatis T, Fritsch EF, Sambrook J. Molecular cloning: a laboratory manual. New York (NY): Cold Spring Harbor Laboratory; 1982. p. 280 [Chap 9]. [16] Chim CS, Liang R, Tam C, Kwong YL. Methylation of p15 and p16 genes in acute promyelocytic leukemia: potential diagnostic and prognostic significance. J Clin Oncol 2001;19:2033-40.

1857 [17] Chim CS, Wong KY, Loong F, Srivastava G. Absence of ATM hypermethylation in mantle cell and follicular lymphoma. Leukemia 2005;19:880-2. [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 U S A 1996;93:9821-6. [19] Roncalli M, Bianchi P, Bruni B, et al. Methylation framework of cell cycle gene inhibitors in cirrhosis and associated hepatocellular carcinoma. Hepatology 2002;36:427-32. [20] Zeschnigk M, Lohmann D, Horsthemke B. A PCR test for the detection of hypermethylated alleles at the retinoblastoma locus. J Med Genet 1999;36:793-4. [21] Geradts J, Kratzke RA, Crush-Stanton S, Wen SF, Lincoln CE. Wildtype and mutant retinoblastoma protein in paraffin sections. Mod Pathol 1996;9:339-47. [22] Kiviniemi M, Sauroja I, Rajamaki A, Punnonen K, Soderstrom KO, Salminen E. Cell cycle regulators p27 and pRb in lymphomas— correlation with histology and proliferative activity. Br J Cancer 2000; 83:1161-7. [23] Jares P, Campo E, Pinyol M, et al. Expression of retinoblastoma gene product (pRb) in mantle cell lymphomas. Correlation with cyclin D1 (PRAD1/CCND1) mRNA levels and proliferative activity. Am J Pathol 1996;148:1591-600. [24] Nguyen PL, Zukerberg LR, Benedict WF, Harris NL. Immunohistochemical detection of p53, BCL-2, and retinoblastoma proteins in follicular lymphoma. Am J Clin Pathol 1996;105:538-43. [25] Hutter G, Scheubner M, Zimmermann Y, et al. Differential effect of epigenetic alterations and genomic deletions of CDK inhibitors [p16 (INK4a), p15(INK4b), p14(ARF)] in mantle cell lymphoma. Genes Chromosomes Cancer 2006;45:203-10. [26] Pinyol M, Cobo F, Bea S, et al. p16(INK4a) gene inactivation by deletions, mutations, and hypermethylation is associated with transformed and aggressive variants of non-Hodgkin's lymphomas. Blood 1998;91:2977-84. [27] Gronbaek K, Nedergaard T, Andersen MK, et al. Concurrent disruption of cell cycle associated genes in mantle cell lymphoma: a genotypic and phenotypic study of cyclin D1, p16, p15, p53 and pRb. Leukemia 1998;12:1266-71. [28] Guo J, Burger M, Nimmrich I, et al. Differential DNA methylation of gene promoters in small B-cell lymphomas. Am J Clin Pathol 2005; 124:430-9. [29] Guan KL, Jenkins CW, Li Y, et al. Growth suppression by p18, a p16INK4/MTS1- and p14INK4B/MTS2-related CDK6 inhibitor, correlates with wild-type pRb function. Genes Dev 1994;8: 2939-52. [30] Hatta Y, Spirin K, Tasaka T, et al. Analysis of p18INK4C in adult T-cell leukemia and non-Hodgkin's lymphoma. Br J Haematol 1997;99:665-7.