Leading Article

Concurrent disruption of p16INK4a and the ARF-p53 pathway predicts poor prognosis in aggressive non-Hodgkin's lymphoma

Article metrics

Abstract

The INK4a/ARF locus at chromosome 9p21 encodes two structurally and functionally distinct molecules with tumor-suppressive properties. p16INK4a controls cell cycle progression by inhibiting phosphorylation of the retinoblastoma protein (Rb), while ARF prevents MDM2-mediated degradation of p53. By using a panel of PCR-based methods, we have examined the status of the p16INK4a, ARF and p53 genes in 123 cases of non-Hodgkin's lymphoma (NHL) at diagnosis. Alterations of one or more of these genes were detected in seven of 36 (19%) cases with low- to intermediate-grade histology, and in 35 of 87 (40%) cases with aggressive histology. For the aggressive lymphomas, the Kaplan–Meier estimate of overall survival for cases with disruption of either p16INK4a or the ARF-p53 pathway was not different from cases with retention of both pathways (5-year survival 45% vs 35%; P = 0.85), suggesting that selective inactivation of one of the pathways does not significantly influence overall survival. By contrast, the 5-year survival was only 7% for cases with concurrent disruption of p16INK4a and the ARF-p53 pathway vs 38% for cases with retention of one or both pathways (P = 0.005). Similar results were obtained when the analysis was confined to diffuse large B cell lymphomas (P = 0.019). On stepwise multivariate regression analysis including factors from the international prognostic index, concurrent disruption of p16INK4a and the ARF-p53 pathway was an independent negative prognostic factor in NHL with aggressive histology (P = 0.006). Our results suggest that the compound status of the p16INK4a and ARF-p53 pathways is a major determinant of outcome in NHL.

Introduction

Non-Hodgkin's lymphoma (NHL) encompasses a broad spectrum of tumors derived from cells of the lymphoid system. Tumors are usually assigned to one of several subtypes on the basis of clinical, immunophenotypical, genetic, and morphological properties.1 Although combination chemotherapy has dramatically improved the overall prognosis in patients with aggressive NHL, marked individual differences in response to treatment still exist within and between different subtypes. Several phenotypic factors have been reported to influence the outcome, including age, tumor stage, performance status, tumor size, tumor burden, serum lactate dehydrogenase concentration (S-LDH), and number of involved extranodal disease sites.2 Unraveling the biological background for these clinical prognostic factors may eventually form the basis for designing and implementing more differentiated therapeutic regimens.

Alterations of several oncogenes and tumor suppressor genes have been demonstrated in NHL.3 As in many other types of cancer, the most commonly altered targets reported in NHL are the genes encoding the tumor suppressors p16INK4a and p53.45678910111213141516171819202122 These proteins are key components of the machinery that controls cell cycle progression.2324 p16INK4a exerts a tumor-suppressive function by specifically interfering with the catalytic activity of complexes between cyclin D and cyclin-dependent kinase 4 (CDK4) or CDK6.25 This inhibitory effect on cyclin D/CDK4(6) complexes prevents phosphorylation of the retinoblastoma protein (Rb) and the subsequent release of transcription factors that are required for passage into and through S phase, in particular members of the E2F transcription factor family. Another potent CDK inhibitor, p21WAF1/CIP1, is a downstream effector of p53 and controls cell cycle progression by inhibiting the activity of a broader range of CDKs.262728

Genetic and epigenetic alterations of p16INK4a and p53 have been found in several subtypes of NHL, in particular in the progressed variants, yet little is known about their prognostic significance.45678910111213141516171819202122 A single study has demonstrated poor prognosis in cases with deletions or rearrangements of p16INK4a,21 and p53 mutations have been associated with adverse outcome in aggressive lymphomas.1013

Recently, a novel putative tumor suppressor, ARF (human p14ARF and murine p19ARF), was identified that may provide an important link between p53 and the p16INK4a-Rb pathway.2930 Expression of ARF, which is induced by hyperproliferative stimuli like the Myc and E1A oncogenes3132 but not by DNA damage,3334 results in apoptosis or cell growth arrest in both G1 and G2.3335 ARF and p16INK4a are encoded by a single genetic locus, designated INK4a/ARF and located at chromosome 9p21.35363738 This locus consists of exon 1β specific for ARF, exon 1α specific for p16INK4a, and exons 2 and 3 common to both genes (Figure 1). Despite the extensive overlap at the gene level, ARF and p16INK4a share no amino acid homology and possess distinct biochemical and regulatory properties. Recent work has shown that the tumor-suppressive functions of ARF may be ascribed to its ability to inhibit MDM2-mediated degradation of p53 in a feedback loop where p53 upregulates its inhibitor, MDM2, and downregulates its activator, ARF.34394041 Specifically, ARF binds to MDM2 and sequesters it into the nucleolus, thereby preventing it from degrading p53 in the nucleoplasm.424344 The existence of this delicate molecular circuitry suggests that loss of ARF function may be functionally equivalent to mutation of p53. This notion is reinforced by the observations that mice lacking ARF exon 1β, but retaining functional p16INK4a, develop tumors that are broadly similar to those reported in mice lacking p53,3345 and that continuously growing mouse embryo fibroblast (MEF) cell lines usually harbor either ARF or p53 loss of function.33 The possible role of ARF in the development of human cancers is, however, largely unknown. Evidence for selective deletion of exon 1β has been demonstrated only in a few cases of melanoma and T cell acute lymphoblastic leukemia.4647 Moreover, inactivation of ARF by hypermethylation or mutation of exon 1β is rare in human tumors,363748495051 including NHL.52 Here, we demonstrate that single disruption of either p16INK4a or the ARF-p53 pathway does not significantly influence the outcome of treatment in aggressive NHL, whereas concomitant disruption of both pathways is associated with shortened survival and is an independent negative prognostic factor.

Figure 1
figure1

 Diagramatic representation of the INK4a/ARF locus, the transcripts encoded by this locus, and the molecular pathways in which they participate. ARF (p14ARF) is encoded by exons 1β and 2 and blocks MDM2-induced p53 degradation. p16INK4a is encoded by exons 1α, 2 and 3 and inhibits CDK4(6), which in turn prevents cells with functional Rb from entering S phase.

Patients and methods

Patients and tissue samples

A total of 123 newly diagnosed NHLs referred to the Departments of Hematology at Herlev Hospital, Rigshospitalet and Odense University Hospital, Denmark, during the period 1984–1994 were included in this study. Routinely processed histological samples were available from all patients. Specimens were stained with hematoxylin-eosin and examined by immunohistology as described,53 and then classified according to the Revised European–American Lymphoma (REAL) classification.1 The following histological subtypes were included: DLC-B (n = 68); follicle center cell lymphoma (FCC) (n = 24); mucosa-associated lymphoid tissue (MALT)-type lymphoma (n = 2); immunocytoma (n = 2); mantle cell lymphoma (MCL) (n = 4); Burkitt's lymphoma (n = 3); peripheral T cell lymphoma (PTL), unspecified (n = 16); angioimmunoblastic T cell lymphoma (AILD) (n = 3); and T cell chronic lymphocytic leukemia (T-CLL) (n = 1). In all cases, frozen tissue blocks were also available. These samples had been frozen immediately after surgery in either liquid N2 or a mixture of 2-methyl butane and dry ice and stored at −80°C until use. Clinical data were obtained retrospectively from the patient files. All patients were treated according to standard regimens. Almost all aggressive tumors were treated with combination chemotherapy consisting of cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP). Some of the cases had previously been examined for aberrations of ARF, p53 and p16INK4a.12 Approval of the study was obtained from the local ethical committee.

DNA isolation and deletion analysis

DNA was extracted from frozen tissue sections by proteinase K digestion and phenol-chloroform extraction, according to standard procedures, or by using the Puregene DNA Isolation Kit (Gentra Systems, Minneapolis, MN, USA). Deletions of the INK4a/ARF locus were detected by semi-quantitative duplex PCR, essentially as described.53 Oligonucleotide primers for INK4a/ARF exons 1β, 1α and 2 and for the internal control sequence (STS) on chromosome 9q were as described.465455 PCR was performed with the GeneAmp PCR System 9600 (Perkin-Elmer Cetus, Emeryville, CA, USA) and was maintained within the exponential range (20–24 cycles). α-33P dATP was included in all reactions for labeling. The amplification products were separated on a 2% agarose gel, which was subsequently dried, exposed to a Storage Phosphor Screen (Molecular Dynamics, Sunnyvale, CA, USA) and inspected using the ImageQuant software (Molecular Dynamics). A case was considered to harbor deletion when the ratio of target sequence to reference sequence was less than 0.2 of that observed in DNA from healthy individuals, and when the results were reproducible in at least three independent experiments. DNA from the DOHH2 lymphoma cell line, which sustains homozygous deletion of the INK4a/ARF locus,6 was used as positive control when mixed with normal control DNA.

Methylation-specific PCR (MSP) and bisulfite denaturing gradient gel electrophoresis (DGGE)

Methylation of the CpG island in the promoter region of the p16INK4a gene56 was characterized by MSP57 and bisulfite-DGGE.58 Both methods are based on initial treatment of genomic DNA with sodium bisulfite to convert unmethylated cytosine to uracil.59 The bisulfite reaction was performed essentially as described.60 Briefly, approximately 1 μg of genomic DNA was denatured in 0.3 mol/l NaOH, followed by the addition of sodium bisulfite (Sigma Chemical, St Louis, MO, USA) to a final concentration of 3.1 mol/l, and hydroquinone (Sigma Chemical) to a final concentration of 2.5 mmol/l. After incubation at 55°C for 16 h, the DNA was recovered by using the GeneClean II Kit (Bio 101, Vista, CA, USA), desulfonated in 0.3 mol/l NaOH, and ethanol-precipitated. For MSP, two sets of primers57 were used to selectively amplify methylated and unmethylated alleles, respectively. PCR was performed in volumes of 25 μl, containing 1 × PCR buffer (10 mmol/l Tris-HCl (pH 8.3), 50 mmol/l KCl, 1.5 mmol/l MgCl2, 0.002% gelatin), 0.2 mmol/l cresol red, 12% sucrose, 10 pmol/l of each primer, 100 μmol/l each dNTP, approximately 100 ng of DNA, and 0.8 units of AmpliTaq polymerase (Perkin-Elmer Cetus). The reaction was hot-started, followed by 40 cycles at 94°C for 20 s, 60°C for 20 s, and 72°C for 30 s. Amplification products were analyzed in a 2% agarose gel. For bisulfite-DGGE, a set of previously described primers58 was used to collectively amplify methylated and unmethylated alleles. PCR was performed as described,58 using the Expand High Fidelity System (Boehringer-Mannheim, Mannheim, Germany). The ‘GC-clamped’ PCR products were analyzed in a 6% polyacrylamide/10–60% denaturing gradient gel (100% denaturant = 7 mol/l urea and 40% formamide).61 The gel was run at 160 V for 4 h in 1 × TAE buffer kept at a constant temperature of 54°C, stained with ethidium bromide, and photographed under UV transillumination.

Mutation detection and direct sequencing

Exons 2–11 of the p53 gene were scanned for mutations by PCR in combination with DGGE, using the primers and electrophoresis conditions described in a previous report.62 Exons 1β, 1α, 2 and 3 of the INK4a/ARF locus were examined for mutations by PCR in combination with single-stranded conformation polymorphism (SSCP) analysis, using previously described primers.4655 Abnormal DGGE and SSCP bands were excised from the gels, and their DNA was eluted in water, reamplified, and sequenced using a 33P-end-labeled primer and the ThermoPrime Cycle Sequencing Kit (Amersham Life Science, Cleveland, OH, USA).

Statistical methods

Probability of survival was estimated according to the method of Kaplan and Meier63 from date of diagnosis to death or last follow-up. Last follow-up was January 1998. The median follow-up time of patients alive (n = 26) was 107 months (range 28–147). The equality between survival curves was tested with the log rank test. The simultaneous relationships of multiple prognostic factors to survival were assessed using Cox's proportional hazard model.64 Variables that met the 0.10 significance criteria in the univariate analysis were included in the multivariate analysis using forward selection. P values less than 0.05 were considered significant. All analyses were performed using the SAS software (Statistical Analysis System Institute, Cary, NC, USA).

Results

Deletions of the INK4a/ARF locus

Exons 1β, 1α and 2 of the INK4a/ARF locus were examined for deletions by semi- quantitative duplex PCR analysis of genomic DNA from the group of 123 NHLs. Homozygous deletions were detected in three of 36 (8%) low- to intermediate-grade lymphomas and in 12 of 87 (14%) high-grade lymphomas. Exon 1β was selectively deleted in two DLC-Bs (3%), deletion of exon 2 alone was detected in one MCL (25%) and one DLC-B (1.5%), and the remaining 11 cases harbored deletion of all three exons. These included one of 24 FCCs (4%), one of four MCLs (25%), six of 68 DLC-Bs (9%), one of three Burkitt's lymphomas (33%), and two of 16 PTLs, unspecified (13%) (Table 1).

Table 1  Status of the p16INK4a, ARF and p53 genes in 42 non-Hodgkin's lymphomas

p16INK4a promoter hypermethylation

The methylation status of a 150-bp region of the p16INK4a promoter was examined by PCR analysis of bisulfite-reacted DNA (MSP57), using primers designed to specifically amplify methylated alleles. Hypermethylation was detected in two of 36 (6%) low- to intermediate-grade lymphomas, including one of two MALT lymphomas (50%) and one of 24 FCCs (4%). Thirteen of 87 high-grade tumors (15%) showed p16INK4a hypermethylation, including 11 of 68 DLC-Bs (16%), one of three Burkitt's lymphomas (33%), and one of 16 PTLs, unspecified (6%) (Table 1).

We have recently demonstrated that methylation of another CDK inhibitor, p15INK4b, can be detected in a small fraction of lymphocytes from normal blood by using the highly sensitive MSP method.65 To exclude that positive MSP signals for p16INK4a methylation in NHLs were due to such background amplification, we further evaluated the samples by the newly developed bisulfite-DGGE method.5865 This method separates methylated alleles from their unmethylated counterparts after bisulfite treatment because of differences in thermal stability, which forms the basis for quantitating and isolating clonotypic epigenotypes present in complex cell populations. For all NHLs tested positive for p16INK4a methylation by MSP, a distinct band was observed in the denaturing gradient gel at a position corresponding to the migration distance of fully methylated p16INK4a alleles from cell lines with known methylation status (Figure 2a). The intensities of the methylated bands were comparable to those of the unmethylated bands from the same samples, suggesting that a large proportion of cells in the tumors sustained p16INK4a methylation. Sequence analysis of bands excised from the gel confirmed that all CpG sites of the target sequence were methylated (Figure 2b).

Figure 2
figure2

 Bisulfite DGGE analysis of p16INK4a methylation status in two NHLs and four hematopoietic cell lines. (a) DGGE analysis of amplification products generated from bisulfite-reacted DNA. Lanes 1 and 3, normal control PBLs. Lane 2, DLC-B 55. Lane 4, RAJI. Lane 5, MALT 145. Lane 6, U937. Lane 7, RPMI 8226. Lane 8, CA 46. (b) Direct sequence analysis of the lower (methylated) and upper (unmethylated) bands from DLC-B 55. Unreacted CpG sites are indicated by arrowheads.

Mutations of the p16INK4a, ARF and p53 genes

Analysis of all exons and exon–intron boundaries of the INK4a/ARF locus by PCR/SSCP analysis revealed no point or frameshift mutations in the 123 NHLs. The same samples were subjected to systematic PCR/DGGE-based scanning of exons 2–11 and all splice sites of the p53 gene (Figure 3a). Enrichment and direct sequence analysis of DNA from bands with an altered mobility led to the identification of mutations in three of 36 (8%) low- to intermediate-grade cases, including one of four MCLs (25%) and two of 24 FCCs (8%). Among the high-grade cases, 17 of 87 (20%) showed mutations, including 14 of 68 DLC-Bs (21%), two of three Burkitt's lymphomas (66%), and one of 16 PTLs, unspecified (6%) (Table 1). The mutations included 18 missense mutations within exons 5–8, one G to A transition at position −1 of the acceptor splice site at the intron 4/exon 5 boundary, and one 19-bp frameshift deletion in exon 5. Three residues, P151, R248 and R273, were found to be altered in two, four and three of the samples, respectively. This is in keeping with previous observations that these sites are mutation hot spots in lymphomas and other human cancers.6667 One of the missense variants, I254N, was due to a tandem mutation (TC to AT) (Figure 3b). To prove that the two base changes occurred in cis, individual alleles were recovered from the denaturing gradient gel and sequenced (data not shown).

Figure 3
figure3

 PCR-DGGE patterns and sequence analysis of p53 in NHL. (a) DGGE analysis of five NHLs harboring mutations in p53 exon 7. Lane 1, FCL 18 (I254N). Lane 2, DLC-B 60 (N235S). Lane 3, DLC-B 111 (R248W). Lane 4, FCL 155 (R248W). Lane 5, DLC-B 159 (R248Q). Lane 6, PBL. (b) Direct sequence analysis of heteroduplex molecules (sample FCL 18) eluted from the denaturing gradient gel shown in (a).

Correlations among aberrations of the p16INK4a, ARF and p53 genes

The distribution of p16INK4a, ARF and p53 aberrations among individual NHL cases is given in Table 1. One or more of these genes were altered in 42 of 123 cases (34%), including 35 of 87 aggressive lymphomas (40%). There was no difference between the frequencies of these aberrations in B and T cell lymphomas (P = 0.14). Our data suggest a positive selection for tumor cells with aberrations of both p16INK4a and the ARF-p53 pathway. Both concomitant p53 mutation and p16INK4a methylation (P = 0.02; chi-square test) and concomitant p53 mutation and deletion of INK4a/ARF exon 2 (P = 0.01) occurred at significantly higher frequencies than expected. By contrast, the frequency of concurrent mutation of p53 and deletion of the entire INK4a/ARF locus was not higher than expected (P = 0.87).

Survival in relation to status of the p16INK4a, ARF and p53 genes in aggressive lymphomas

On the basis of the observed status of the p16INK4a, ARF and p53 genes and the molecular pathways in which the individual genes participate (cf. Figure 1), each of the 87 patients with aggressive lymphoma histology (DLC-B, Burkitt's lymphoma, or PTL, unspecified) was assigned to one of four groups: (1) cases sustaining inactivation of the p16INK4a pathway by hypermethylation of the p16INK4a gene with no concomitant alterations of the ARF and p53 genes (n = 8); (2) cases with disruption of the ARF-p53 pathway by mutations in the p53 gene (n = 10) or selective deletion of INK4a/ARF exon 1β (n = 2) and apparent retention of p16INK4a; (3) cases sustaining dual pathway disruption, either by deletion of the entire INK4a/ARF locus (n = 8), concomitant deletion of INK4a/ARF exon 2 and p53 mutation (n = 2), or concurrent p16INK4a hypermethylation and p53 mutation (n = 5); and (4) cases with no detectable aberrations of the three genes (n = 52).

According to the Kaplan–Meier estimates, the survival of the eight cases with selective inactivation of p16INK4a by hypermethylation was not statistically different from that of the 12 cases with either p53 mutation or selective deletion of INK4a/ARF exon 1β (5-year survival 50% vs 42%; P = 0.78). Furthermore, no difference in survival was observed between these 20 cases harboring aberrations of either p16INK4a or the p53-ARF pathway and the 52 cases with apparent retention of all three genes (5-year survival 45% vs 35%; P = 0.85), suggesting that selective inactivation of one of these pathways does not significantly influence overall survival (Figure 4). By contrast, the Kaplan–Meier estimate of survival was significantly lower for the 15 patients with concomitant loss of p16INK4a and the ARF-p53 pathway than for those with retention of one or both pathways (5-year survival 7% vs 38%; P = 0.005). Specifically, the five patients with tumors harboring targeted alterations of p16INK4a and p53 died 1, 3, 6, 9 and 14 months after diagnosis, respectively. Similarly, Kaplan–Meier curves for overall survival of patients with lymphomas of the single largest histological subtype, DLC-B, showed that patients with disruption of both pathways had a significantly shorter survival than patients with disruption of one or no pathways (P = 0.019) (Figure 5). Even within the group of cases with low- to intermediate-grade histology, all cases with disruption of both pathways died within the first year from diagnosis.

Figure 4
figure4

 Kaplan–Meier curves for overall survival for patients with aggressive tumors harboring concurrent disruption of p16INK4a and the ARF-p53 pathway (dotted line; n = 15), hypermethylation of p16INK4a, deletion of INK4a/ARF exon 1β or mutation of p53 (dashed line; n = 20), or apparent retention of all three genes (solid line; n = 52).

Figure 5
figure5

 Kaplan–Meier curves for overall survival for patients with DLC-B lymphomas with concurrent disruption of p16INK4a and the ARF-p53 pathway (dotted line; n = 10), hypermethylation of p16INK4a or selective abrogation of the ARF-p53 pathway (dashed line; n = 16), or apparent retention of all three genes (solid line; n = 42).

The clinical and histological features of patients with aggressive lymphomas grouped according to the status of p16INK4a and the ARF-p53 pathway are given in Table 2. Univariate survival analysis including factors from the international prognostic index showed that age, clinical stage, dual p16INK4a and ARF-p53 pathway disruption, and elevated S-LDH were adverse prognostic factors. On stepwise multivariate regression analysis, age, dual p16INK4a and ARF-p53 pathway disruption, and clinical stage retained independent prognostic significance for survival, while elevated S-LDH reached borderline significance (Table 3).

Table 2  Aggressive NHL: patient characteristics at diagnosis in relation to the status of p16INK4a and the ARF-p53 pathway
Table 3  Aggressive NHL: multivariate analysis of prognostic factors for survival

Discussion

Mounting evidence exists that cancer-related genes participate in pathways that control a wide range of cellular functions, and that genetic aberrations along the same pathways may be functionally equivalent. Considering molecular pathways rather than single genes may, therefore, dramatically increase the diagnostic and prognostic value of molecular analyses in cancer. In the present study, we have examined the status of p16INK4a, ARF, and p53 in 123 cases of NHL at diagnosis and, in accordance with current knowledge about the molecular pathways in which these tumor suppressors participate, correlated these results with histology and survival.

In agreement with previous studies,45678910111213141516171819202122 alterations of p16INK4a and p53 were primarily associated with aggressive histology. Among the 87 cases with aggressive histology, mutation of p53, hypermethylation of p16INK4a, and deletion of INK4a/ARF were found in 20%, 15% and 14%, respectively. No mutations were detected by analysis of the entire coding regions of the INK4a/ARF locus, which is in keeping with previous studies showing that p16INK4a mutations are rare in NHL.722 We identified two cases of DLC-B with selective deletion of INK4a/ARF exon 1β, suggesting that independent aberration of ARF function may occur in lymphomagenesis. Recent work by Della Valle et al68 has demonstrated that ARF protein is absent in some human hematopoietic cell lines that abundantly express the ARF transcript. Whether disruption of the ARF-p53 pathway by post-transcriptional inactivation of ARF is a frequent event in lymphoid tumors will be subject to future studies.

One case of DLC-B showed concomitant deletion of the entire INK4a/ARF locus and mutation of the p53 gene. This finding corroborates previous observations in human tumor specimens69 and cell lines3450 that inactivation of ARF and p53 may not always be mutually exclusive. As proposed by Stott et al,34 concurrent deletion of ARF and mutation of p53 may be a simple reflection of the order of events during tumor progression. If deletion of the region encompassing the INK4a/ARF locus on 9p21 occurs early in tumorigenesis, there would be little selection against p53. Conversely, if p53 mutation is an early event, a strong selection pressure would still exist against p16INK4a, which could result in the coincidental co-deletion of ARF.

Twenty-three percent of the aggressive NHL cases showed disruption of either p16INK4a or the ARF-p53 pathway, and 17% showed concomitant loss of p16INK4a and the ARF-p53 pathway, either by deletion of the INK4a/ARF locus, concurrent hypermethylation of p16INK4a and mutation of p53, or concurrent deletion of INK4a/ARF exon 2 and mutation of p53. Cases with p53 mutation in combination with p16INK4a hypermethylation or with p53 mutation in combination with deletion of exon 2 of the INK4a/ARF locus occurred more frequently than expected, probably reflecting a selection process. These observed patterns of p16INK4a and ARF-p53 alterations and their correlation with outcome in NHL patients may have clinical as well as biological significance. Notably, we have shown a significant difference in survival between aggressive NHL cases with targeted inactivation of p16INK4a and cases with concomitant inactivation of p16INK4a and ARF. Although we cannot fully exclude the possibility that silencing of p16INK4a by hypermethylation is a reversible process that may be influenced by conventional combination chemotherapy, a more likely explanation for the observed discrepancy in survival between p16INK4a deficient vs p16INK4a- and ARF-deficient cases is that ARF directly inhibits lymphomagenesis. This hypothesis is supported by previous observations that mice selectively inactivated for ARF are prone to developing lymphomas,33 and that INK4a/ARF exon 1β is selectively deleted in a small fraction of chemically induced murine lymphomas.70 The association of INK4a/ARF deletions with adverse prognosis in NHL is in agreement with the results reported by Garcia-Sanz et al,21 who found a median survival of 10 months among seven cases with deletions or rearrangements of INK4a/ARF. Furthermore, a recent study of superficial bladder cancer showed that deletion of the INK4a/ARF locus, but not hypermethylation of p16INK4a, is correlated with poor prognosis.71

Our data suggest the existence of an intimate relationship between p16INK4a and the ARF-p53 pathway in the control of lymphoproliferation. The finding that selective disruption of either p16INK4a or the ARF-p53 pathway does not significantly influence the outcome of treatment in aggressive NHL suggests the existence of a circuitry in which disruption of one of the pathways elicits a compensatory response from the retained pathway. This notion was reinforced by recent work delineating how loss of Rb function may trigger a biochemical fail-safe mechanism to protect against deregulated cell growth. According to the proposed model, loss of functional Rb results in deregulation of the E2F-1 transcription factor, which in turn causes stabilization of p53,7273 probably by activating the transcription of ARF.74 A similar, as yet unknown, mechanism may be elicited from the p16INK4a-Rb pathway that compensates for the selective loss of functional p53.

Our study may have certain limitations. First, not all known components of the p16INK4a and ARF-p53 pathways were examined. Even though genetic alterations of many of these components are considered to be rare in NHL, including mutation and deletion of the RB1 gene75 and hypermethylation of ARF,52 their inclusion might have improved the significance of our results. Second, deletions of regions at 9p21 in NHL may encompass more than 500 kb and eliminate several genes in addition to ARF and p16INK4a, including the gene encoding the cyclin-dependent kinase inhibitor p15INK4b, the interferon gene cluster, and the methylthioadenosine phosphorylase (MTAP) gene.76 Therefore, we cannot unambiguously ascribe a particular phenotype to the specific loss of the INK4a/ARF locus. However, the five patients with tumors harboring targeted alterations of p16INK4a and p53 all died shortly after diagnosis (1, 3, 6, 9 and 14 months, respectively), and hence exhibit a clinical course similar to patients with tumors harboring INK4a/ARF deletions, strongly implicating these tumor suppressors and their respective pathways in the control of lymphoproliferation.

In conclusion, we have shown that virtually all patients with dual disruption of p16INK4a and the ARF-p53 pathway die within the first year of diagnosis, irrespective of histological subtype, while selective inactivation of either pathway does not influence overall survival. These data suggest that p16INK4a and ARF-p53 mediate two cellular pathways that individually suppress lymphoid tumorigenesis and monitor loss of each other in a compensatory fashion. The identification of patients with tumors harboring abrogation of both pathways may identify subgroups of patients who are refractory to conventional therapy and who may therefore be candidates for alternative regimens. Whether the inclusion of possible alterations of the remaining components of the same pathways, including MDM2, CDK4, cyclin D1, E2F-1 and Rb, will further improve the diagnostic and prognostic information is a subject for further studies.

References

  1. 1

    Harris NL, Jaffe ES, Stein H, Banks PM, Chan JKC, Cleary ML, Delsol G, De Wolf-Peeters C, Falini B, Gatter KC, Grogan TM, Isaacson PG, Knowles DM, Mason DY, Muller-Hermelink HK, Pileri SA, Piris MA, Ralfkiaer E, Warnke RA . A revised European–American classification of lymphoid neoplasms: a proposal from the international lymphoma study group Blood 1994 84: 1361–1392

  2. 2

    The International Non-Hodgkin's Lymphoma Prognostic Factors Project . A predictive model for aggressive non-Hodgkin's lymphoma New Engl J Med 1993 329: 987–994

  3. 3

    Howard OM, Shipp MA . The cellular and molecular heterogeneity of the aggressive non-Hodgkin's lymphomas Curr Opin Oncol 1998 10: 385–391

  4. 4

    Gaidano G, Ballerini P, Gong JZ, Inghirami G, Neri A, Newcomb EW, Magrath IT, Knowles DM, Dalla-Favera R . p53 mutations in human lymphoid malignancies: association with Burkitt lymphoma and chronic lymphocytic leukemia Proc Natl Acad Sci USA 1991 88: 5413–5417

  5. 5

    Villuendas R, Sanchez-Beato M, Martinez JC, Saez AI, Martinez-Delgado B, Garcia JF, Mateo MS, Sanchez-Verde L, Benitez J, Martinez P, Piris MA . Loss of p16/INK4A protein expression in non-Hodgkin's lymphomas is a frequent finding associated with tumor progression Am J Pathol 1998 153: 887–897

  6. 6

    Stranks G, Height SE, Mitchell P, Jadayel D, Yuille MA, De Lord C, Clutterbuck RD, Treleaven JG, Powles RL, Nacheva E . Deletions and rearrangement of CDKN2 in lymphoid malignancy Blood 1995 85: 893–901

  7. 7

    Pinyol M, Cobo F, Bea S, Jares P, Nayach I, Fernandez PL, Montserrat E, Cardesa A, Campo E . p16(INK4a) gene inactivation by deletions, mutations, and hypermethylation is associated with transformed and aggressive variants of non-Hodgkin's lymphomas Blood 1998 91: 2977–2984

  8. 8

    Herman JG, Civin CI, Issa J-PJ, Collector MI, Sharkis SJ, Baylin SB . Distinct patterns of inactivation of p15INK4B and p16INK4A characterize the major types of hematological malignancies Cancer Res 1997 57: 837–841

  9. 9

    Wilson WH, Teruya-Feldstein J, Fest T, Harris C, Steinberg SM, Jaffe ES, Raffeld M . Relationship of p53, bcl-2, and tumor proliferation to clinical drug resistance in non-Hodgkin's lymphomas Blood 1997 89: 601–609

  10. 10

    Ichikawa A, Kinoshita T, Watanabe T, Kato H, Nagai H, Tsushita K, Saito H, Hotta T . Mutations of the p53 gene as a prognostic factor in aggressive B-cell lymphoma New Engl J Med 1997 337: 529–534

  11. 11

    Koduru PR, Zariwala M, Soni M, Gong JZ, Xiong Y, Broome JD . Deletion of cyclin-dependent kinase 4 inhibitor genes P15 and P16 in non-Hodgkin's lymphoma Blood 1995 86: 2900–2905

  12. 12

    Møller MB, Ino Y, Gerdes AM, Skjødt K, Louis DN, Pedersen NT . Aberrations of the p53 pathway components p53, MDM2 and CDKN2A appear independent in diffuse large B cell lymphoma Leukemia 1999 13: 453–459

  13. 13

    Greiner TC, Moynihan MJ, Chan WC, Lytle DM, Pedersen A, Anderson JR, Weisenburger DD . p53 mutations in mantle cell lymphoma are associated with variant cytology and predict a poor prognosis Blood 1996 87: 4302–4310

  14. 14

    Pinyol M, Hernandez L, Cazorla M, Balbin M, Jares P, Fernandez PL, Montserrat E, Cardesa A, Lopez Otin C, Campo E . Deletions and loss of expression of p16INK4a and p21Waf1 genes are associated with aggressive variants of mantle cell lymphomas Blood 1997 89: 272–280

  15. 15

    Koduru PR, Raju K, Vadmal V, Menezes G, Shah S, Susin M, Kolitz J, Broome JD . Correlation between mutation in P53, p53 expression, cytogenetics, histologic type, and survival in patients with B-cell non-Hodgkin's lymphoma Blood 1997 90: 4078–4091

  16. 16

    Klangby U, Okan I, Magnusson KP, Wendland M, Lind P, Wiman KG . p16/INK4a and p15/INK4b gene methylation and absence of p16/INK4a mRNA and protein expression in Burkitt's lymphoma Blood 1998 91: 1680–1687

  17. 17

    Sander CA, Yano T, Clark HM, Harris C, Longo DL, Jaffe ES, Raffeld M . p53 mutation is associated with progression in follicular lymphomas Blood 1993 82: 1994–2004

  18. 18

    Elenitoba-Johnson KS, Gascoyne RD, Lim MS, Chhanabai M, Jaffe ES, Raffeld M . Homozygous deletions at chromosome 9p21 involving p16 and p15 are associated with histologic progression in follicle center lymphoma Blood 1998 91: 4677–4685

  19. 19

    Gombart AF, Morosetti R, Miller CW, Said JW, Koeffler HP . Deletions of the cyclin-dependent kinase inhibitor genes p16INK4A and p15INK4B in non-Hodgkin's lymphomas Blood 1995 86: 1534–1539

  20. 20

    Otsuki T, Clark HM, Wellmann A, Jaffe ES, Raffeld M . Involvement of CDKN2 (p16INK4A/MTS1) and p15INK4B/MTS2 in human leukemias and lymphomas Cancer Res 1995 55: 1436–1440

  21. 21

    Garcia-Sanz R, Gonzalez M, Vargas M, Chillon MC, Balanzategui A, Barbon M, Flores MT, San Miguel JF . Deletions and rearrangements of cyclin-dependent kinase 4 inhibitor gene p16 are associated with poor prognosis in B cell non-Hodgkin's lymphomas Leukemia 1997 11: 1915–1920

  22. 22

    Uchida T, Watanabe T, Kinoshita T, Murate T, Saito H, Hotta T . Mutational analysis of the CDKN2 (MTS1/p16ink4A) gene in primary B-cell lymphomas Blood 1995 86: 2724–2731

  23. 23

    Levine AJ . p53, the cellular gatekeeper for growth and division Cell 1997 88: 323–331

  24. 24

    Sherr CJ . Cancer cell cycles Science 1996 274: 1672–1677

  25. 25

    Serrano M, Hannon GJ, Beach D . A new regulatory motif in cell-cycle control causing specific inhibition of cyclin D/CDK4 Nature 1993 366: 704–707

  26. 26

    Xiong Y, Hannon GJ, Zhang H, Casso D, Kobayashi R, Beach D . p21 is a universal inhibitor of cyclin kinases Nature 1993 366: 701–704

  27. 27

    El-Deiry WS, Tokino T, Velculescu VE, Levy DB, Parsons R, Trent JM, Lin D, Mercer WE, Kinzler KW, Vogelstein B . WAF1, a potential mediator of p53 tumor suppression Cell 1993 75: 817–825

  28. 28

    Harper JW, Adami GR, Wei N, Keyomarsi K, Elledge SJ . The p21 Cdk-interacting protein Cip1 is a potent inhibitor of G1 cyclin-dependent kinases Cell 1993 75: 805–816

  29. 29

    Sherr CJ . Tumor surveillance via the ARF-p53 pathway Genes Dev 1998 12: 2984–2991

  30. 30

    Prives C . Signaling to p53: breaking the MDM2-p53 circuit Cell 1998 95: 5–8

  31. 31

    de Stanchina E, McCurrach ME, Zindy F, Shieh SY, Ferbeyre G, Samuelson AV, Prives C, Roussel MF, Sherr CJ, Lowe SW . E1A signaling to p53 involves the p19(ARF) tumor suppressor Genes Dev 1998 12: 2434–2442

  32. 32

    Zindy F, Eischen CM, Randle DH, Kamijo T, Cleveland JL, Sherr CJ, Roussel MF . Myc signaling via the ARF tumor suppressor regulates p53-dependent apoptosis and immortalization Genes Dev 1998 12: 2424–2433

  33. 33

    Kamijo T, Zindy F, Roussel MF, Quelle DE, Downing JR, Ashmun RA, Grosveld G, Sherr CJ . Tumor suppression at the mouse INK4a locus mediated by the alternative reading frame product p19ARF Cell 1997 91: 649–659

  34. 34

    Stott FJ, Bates S, James MC, McConnell BB, Starborg M, Brookes S, Palmero I, Ryan K, Hara E, Vousden KH, Peters G . The alternative product from the human CDKN2A locus, p14(ARF), participates in a regulatory feedback loop with p53 and MDM2 EMBO J 1998 17: 5001–5014

  35. 35

    Quelle DE, Zindy F, Ashmun RA, Sherr CJ . Alternative reading frames of the INK4a tumor suppressor gene encode two unrelated proteins capable of inducing cell cycle arrest Cell 1995 83: 993–1000

  36. 36

    Stone S, Jiang P, Dayananth P, Tavtigian SV, Katcher H, Parry D, Peters G, Kamb A . Complex structure and regulation of the p16 (MTS1) locus Cancer Res 1995 55: 2988–2994

  37. 37

    Mao L, Merlo A, Bedi G, Shapiro GI, Edwards CD, Rollins BJ, Sidransky D . A novel p16INK4A transcript Cancer Res 1995 55: 2995–2997

  38. 38

    Duro D, Bernard O, Della VV, Berger R, Larsen CJ . A new type of p16INK4/MTS1 gene transcript expressed in B-cell malignancies Oncogene 1995 11: 21–29

  39. 39

    Zhang Y, Xiong Y, Yarbrough WG . ARF promotes MDM2 degradation and stabilizes p53: ARF-INK4a locus deletion impairs both the Rb and p53 tumor suppression pathways Cell 1998 92: 725–734

  40. 40

    Pomerantz J, Schreiber-Agus N, Liegeois NJ, Silverman A, Alland L, Chin L, Potes J, Chen K, Orlow I, Lee HW, Cordon-Cardo C, DePinho RA . The Ink4a tumor suppressor gene product, p19Arf, interacts with MDM2 and neutralizes MDM2's inhibition of p53 Cell 1998 92: 713–723

  41. 41

    Kamijo T, Weber JD, Zambetti G, Zindy F, Roussel MF, Sherr CJ . Functional and physical interactions of the ARF tumor suppressor with p53 and Mdm2 Proc Natl Acad Sci USA 1998 95: 8292–8297

  42. 42

    Zhang Y, Xiong Y . Mutations in human ARF exon 2 disrupt its nucleolar localization and impair its ability to block nuclear export of MDM2 and p53 Mol Cell 1999 3: 579–591

  43. 43

    Tao W, Levine AJ . P19ARF stabilizes p53 by blocking nucleo-cytoplasmic shuttling of Mdm2 Proc Natl Acad Sci USA 1999 96: 6937–6941

  44. 44

    Weber JD, Taylor LJ, Roussel MF, Sherr CJ, Bar-Sagi D . Nucleolar Arf sequesters Mdm2 and activates p53 Nat Cell Biol 1999 1: 20–26

  45. 45

    Donehower LA, Harvey M, Slagle BL, McArthur MJ, Montgomery CAJ, Butel JS, Bradley A . Mice deficient for p53 are developmentally normal but susceptible to spontaneous tumours Nature 1992 356: 215–221

  46. 46

    Kumar R, Sauroja I, Punnonen K, Jansen C, Hemminki K . Selective deletion of exon 1β of the p19ARF gene in metastatic melanoma cell lines Genes Chromosomes Cancer 1998 23: 273–277

  47. 47

    Gardie B, Cayuela JM, Martini S, Sigaux F . Genomic alterations of the p19ARF encoding exons in T-cell acute lymphoblastic leukemia Blood 1998 91: 1016–1020

  48. 48

    Gonzalgo ML, Hayashida T, Bender CM, Pao MM, Tsai YC, Gonzales FA, Nguyen HD, Nguyen TT, Jones PA . The role of DNA methylation in expression of the p19/p16 locus in human bladder cancer cell lines Cancer Res 1998 58: 1245–1252

  49. 49

    Robertson KD, Jones PA . The human ARF cell cycle regulatory gene promoter is a CpG island which can be silenced by DNA methylation and down-regulated by wild-type p53 Mol Cell Biol 1998 18: 6457–6473

  50. 50

    Markl ID, Jones PA . Presence and location of TP53 mutation determines pattern of CDKN2A/ARF pathway inactivation in bladder cancer Cancer Res 1998 58: 5348–5353

  51. 51

    Brenner AJ, Paladugu A, Wang H, Olopade OI, Dreyling MH, Aldaz CM . Preferential loss of expression of p16(INK4a) rather than p19(ARF) in breast cancer Clin Cancer Res 1996 2: 1993–1998

  52. 52

    Baur AS, Shaw P, Burri N, Delacretaz F, Bosman FT, Chaubert P . Frequent methylation silencing of p15(INK4b) (MTS2) and p16(INK4a) (MTS1) in B-cell and T-cell lymphomas Blood 1999 94: 1773–1781

  53. 53

    Grønbæk K, Nedergaard T, Andersen MK, thor Straten P, Guldberg P, Møller P, Zeuthen J, Hansen NE, Hou-Jensen K, Ralfkiaer E . 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: 1266–1271

  54. 54

    Ueki K, Ono Y, Henson JW, Efird JT, von Deimling A, Louis DN . CDKN2/p16 or RB alterations occur in the majority of glioblastomas and are inversely correlated Cancer Res 1996 56: 150–153

  55. 55

    Hussussian CJ, Struewing JP, Goldstein AM, Higgins PAT, Ally DS, Sheahan MD, Clark WH, Tucker MA, Dracopoli NC . Germline p16 mutations in familial melanoma Nat Genet 1994 8: 15–21

  56. 56

    Merlo A, Herman JG, Mao L, Lee DJ, Gabrielson E, Burger PC, Baylin SB, Sidransky D . 5′ CpG island methylation is associated with transcriptional silencing of the tumor suppressor p16/CDKN2/MTS1 in human cancers Nature Med 1995 1: 686–692

  57. 57

    Herman JG, Jen J, Merlo A, Baylin SB . Hypermethylation-associated inactivation indicates a tumor suppressor role for p15INK4B1 Cancer Res 1996 56: 722–727

  58. 58

    Guldberg P, Grønbæk K, Aggerholm A, Platz A, thor Straten P, Ahrenkiel V, Hokland P, Zeuthen J . Detection of mutations in GC-rich DNA by bisulphite denaturing gradient gel electrophoresis Nucleic Acids Res 1998 26: 1548–1549

  59. 59

    Wang RY, Gehrke CW, Ehrlich M . Comparison of bisulfite modification of 5-methyldeoxycytidine and deoxycytidine residues Nucleic Acids Res 1980 8: 4777–4790

  60. 60

    Zeschnigk M, Lich C, Buiting K, Doerfler W, Horsthemke B . A single-tube PCR test for the diagnosis of Angelman and Prader–Willi syndrome based on allelic methylation differences at the SNRPN locus Eur J Hum Genet 1997 5: 94–98

  61. 61

    Abrams ES, Stanton VP . Use of denaturing gradient gel electrophoresis to study conformational transitions in nucleic acids Meth Enzymol 1992 212: 71–104

  62. 62

    Guldberg P, Nedergaard T, Nielsen HJ, Olsen AC, Ahrenkiel V, Zeuthen J . Single-step DGGE-based mutation scanning of the p53 gene: application to genetic diagnosis of colorectal cancer Hum Mutat 1997 9: 348–355

  63. 63

    Kaplan EL, Meier P . Nonparametric estimation from incomplete observations J Am Stat Assoc 1958 53: 457–481

  64. 64

    Cox DR . Regression models and life tables J Roy Stat Soc 1972 34: 187–220

  65. 65

    Aggerholm A, Guldberg P, Hokland M, Hokland P . Extensive intra- and interindividual heterogeneity of p15INK4B methylation in acute myeloid leukemia Cancer Res 1999 59: 436–441

  66. 66

    Greenblatt MS, Bennett WP, Hollstein M, Harris CC . Mutations in the p53 tumour suppressor gene: clues to cancer etiology and molecular pathogenesis Cancer Res 1994 54: 4855–4878

  67. 67

    Cho Y, Gorina S, Jeffrey PD, Pavletich NP . Crystal structure of a p53 tumor suppressor–DNA complex: understanding tumorigenic mutations Science 1994 265: 346–355

  68. 68

    Della Valle V, Duro D, Bernard O, Larsen CJ . The human protein p19ARF is not detected in hemopoietic human cell lines that abundantly express the alternative β transcript of the p16INK4a/MTS1 gene Oncogene 1997 15: 2475–2481

  69. 69

    Gazzeri S, Della VV, Chaussade L, Brambilla C, Larsen CJ, Brambilla E . The human p19ARF protein encoded by the β transcript of the p16INK4a gene is frequently lost in small cell lung cancer Cancer Res 1998 58: 3926–3931

  70. 70

    Zhuang SM, Schippert A, Haugen-Strano A, Wiseman RW, Soderkvist P . Inactivations of p16INK4a-α, p16INK4a-β and p15INK4b genes in 2′,3′-dideoxycytidine- and 1,3-butadiene-induced murine lymphomas Oncogene 1998 16: 803–808

  71. 71

    Orlow I, LaRue H, Osman I, Lacombe L, Moore L, Rabbani F, Meyer F, Fradet Y, Cordon-Cardo C . Deletions of the INK4A gene in superficial bladder tumors. Association with recurrence Am J Pathol 1999 155: 105–113

  72. 72

    Tsai KY, Hu Y, Macleod KF, Crowley D, Yamasaki L, Jacks T . Mutation of E2f-1 suppresses apoptosis and inappropriate S phase entry and extends survival of Rb-deficient mouse embryos Mol Cell 1998 2: 293–304

  73. 73

    Pan H, Yin C, Dyson NJ, Harlow E, Yamasaki L, Dyke TV . Key roles for E2F1 in signaling p53-dependent apoptosis and in cell division within developing tumors Mol Cell 1998 2: 283–292

  74. 74

    Bates S, Phillips AC, Clark PA, Stott F, Peters G, Ludwig RL, Vousden KH . p14ARF links the tumour suppressors RB and p53 Nature 1998 395: 124–125

  75. 75

    Dreyling MH, Bullinger L, Ott G, Stilgenbauer S, Muller-Hermelink HK, Bentz M, Hiddemann W, Dohner H . Alterations of the cyclin D1/p16-pRB pathway in mantle cell lymphoma Cancer Res 1997 57: 4608–4614

  76. 76

    Dreyling MH, Bohlander SK, Le Beau MM, Olopade OI . Refined mapping of genomic rearrangements involving the short arm of chromosome 9 in acute lymphoblastic leukemias and other hematologic malignancies Blood 1995 86: 1931–1938

Download references

Acknowledgements

This study was supported by grants from the Danish Cancer Society, the E Danielsen Foundation, the Kaarsen Foundation, the Danish Medical Research Council, the Danish Cancer Research Foundation, and the Novo Nordisk Foundation.

Author information

Correspondence to P Guldberg.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Grønbæk, K., de Nully Brown, P., Møller, M. et al. Concurrent disruption of p16INK4a and the ARF-p53 pathway predicts poor prognosis in aggressive non-Hodgkin's lymphoma. Leukemia 14, 1727–1735 (2000) doi:10.1038/sj.leu.2401901

Download citation

Keywords

  • ARF
  • p16
  • p53
  • lymphoma
  • prognosis

Further reading