Carcinogenesis

This Article Abstract FREE Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (46) Request Permissions Google Scholar Articles by Xing, E. P. Articles by Yang, C. S. Search for Related Content PubMed PubMed Citation Articles by Xing, E. P. Articles by Yang, C. S. Social Bookmarking          What's this?

© 1999 Oxford University Press

Article

Aberrant methylation of p16^INK4a and deletion of p15^INK4b are frequent events in human esophageal cancer in Linxian, China

Eric Poe Xing, Yan Nie, Li-Dong Wang¹, Guang-Yu Yang and Chung S. Yang²

Laboratory for Cancer Research, College of Pharmacy, Rutgers University, Piscataway, NJ 08854 8020, USA and
¹ Henan Medical University, Zhengzhou, China

	Abstract

Top Abstract Introduction Materials and methods Results Discussion References

 
p16^INK4a and p15^INK4b genes, which encode two functionally related CDK inhibitors, recently emerged as candidate tumor suppressor genes since they were both localized to 9p21, which frequently undergoes hemizygous and homozygous deletion in a variety of tumor types. To determine the mode of inactivation of these two genes in human esophageal squamous cell carcinoma (ESCC), we performed multiple molecular analyses in 60 ESCC specimens from Linxian, China using DNA methylation assay, LOH analysis, deletion screening and SSCP-sequencing. We observed that p16^INK4a inactivation was predominantly associated with aberrant methylation in the CpG island of its promoter region, whereas p15^INK4b frequently had homozygous deletions. Compared with aberrant methylation, which occurred in 17 of 34 cases, homozygous deletion of p16^INK4a and LOH at its nearby D9S942 microsatellite marker were observed at a much lower frequency (17%). Intragenic mutation in p16^INK4a gene was rare. In contrast, homozygous deletion in p15^INK4b and LOH at the nearby D9S171 marker were observed at frequencies of 35 and 47%, respectively, and the two events were significantly associated with each other. On the other hand, aberrant methylation of p15^INK4b was relatively infrequent (6/34) and occurred concomitantly with p16^INK4a methylation. Among the 60 cases, only four contained a continuous homozygous deletion spanning both p15^INK4b and p16^INK4a. Six cases were exclusively deleted at p16^INK4a and 17 exclusively deleted at p15^INK4b. LOH at D9S942 and D9S171 was also found to be mutually exclusive. Our results suggest that the alteration mode at 9p21 was not uniform, and the two genes were inactivated by distinct mechanisms. Altogether, 68% of the samples harbor at least one type of alteration in p16^INK4a gene and 50% of the samples were altered in p15^INK4b gene, indicating that they are the frequent inactivating targets during ESCC development.

Abbreviations: CDK, cyclin-dependent kinase; ESCC, esophageal squamous cell carcinoma; LOH, loss of heterozygosity; SSCP, single strand conformation polymorphism; MI, microsatellite instability.

	Introduction

Top Abstract Introduction Materials and methods Results Discussion References

 
The putative tumor suppressor genes p16(INK4a/MTS1/CDKN2) and p15(INK4b/MTS2) encode two important cyclin dependent kinase (CDK) inhibitors which negatively regulate G₁–S transition of the proliferating cells by contributing to the maintenance of pRb in an active (hypophosphorylated) state (1,2). p16^INK4a binds to and inhibits CDK4/6 activity during G₁ stage (3). p15^INK4b binds to the cyclin D-dependent kinase and prevents p27 association. p27 then binds to, and inactivates, cyclin E-CDK2 complex, thereby blocking the cell cycle at the G₁–S boundary (4). The absence of the inhibitory effect of p16^INK4a or p15^INK4b on the cyclin/CDK enzymes predisposes the cells to a risk of uncontrolled cell growth.

Inactivation of p16^INK4a and p15^INK4b genes has been observed in many types of human cancers including esophageal squamous cell carcinoma (ESCC). Multiple mechanisms seem to be involved in causing gene inactivation (5–18). Some studies showed that germline mutations in p16^INK4a gene might be related to familial melanoma (6), but in general the mutation is relatively infrequent in sporadic melanoma and most other cancers (7). Allelic loss, observed as loss of heterozygosity (LOH) at 9p21 is common in various cancer types and is often an early event (8,9). Homozygous deletions of p16^INK4a and p15^INK4b also occur at significant frequencies and sometimes correlate with a high risk of metastasis (10,11). Hemizygous and homozygous deletion at 9p21 are widely considered to be one of the primary mechanisms of p16^INK4a/p15^INK4b inactivation and may play an important role in the invasiveness of many cancers during their late stage. It was hypothesized that, due to the existence of multiple tumor suppressor genes including p16^INK4a, p15^INK4b and p19^ARF (which is translated from an alternatively spliced product of p16^INK4a gene using a different open reading frame) (12) in this region, deletion of a large piece of genomic material would be more effective over point mutation to eliminate two or more critical genes (13). Recently, aberrant methylation of the CpG island at the promoter regions of p16^INK4a and p15^INK4b genes was reported in many cancers (14) and was associated with loss of transcription (15,16). Interestingly, it was observed that for most epithelial-derived tumors, inactivation via promoter methylation occurs exclusively for p16^INK4a, whereas in most hematopoietic malignancies and gliomas, this mechanism seems to involve the p15^INK4b gene only (17,18). This suggests that promoter hypermethylation could be involved in the selective inactivation of targets at 9p21 region in different types of tumor.

In human ESCC, the functional significance of p16^INK4a alteration is still unclear. Although p16^INK4a gene has been shown to be deleted, mutated or hypermethylated in a number of ESCC cell lines, molecular analysis from primary tumors has not consistently indicated the alterations of this gene in vivo (19–24). For example, frequencies of homozygous deletion in cell lines range from 33 to 92% (19), whereas in primary tumors, only 0–16% have been observed (20,21). This apparent discrepancy could be due to (i) p16^INK4a deletion as a result of in vitro immortalization or selection during culturing; (ii) loss of p16^INK4a during very late stages of ESCC progression or (iii) technical difficulties in detecting homozygous deletion from primary tumor DNA due to contamination of DNA from normal tissue. The frequency of p16^INK4a mutations in primary tumors also vary among reports, from 0 to 50%, and differed significantly between different ethnic groups if not due to experimental variations (19 and references therein). Recent study also showed de novo promoter hypermethylation of p16^INK4a in a group of Japanese ESCC patients (21). However, due to the lack of comparability of works from different groups, our knowledge of the mode of p16^INK4a inactivation in ESCC is still sketchy. Our recent results from immunohistochemical analysis indicated that p16^INK4a expression was present in only three out of 22 ESCC cases (22), suggesting that the p16^INK4a gene is frequently inactivated in primary ESCC. The molecular basis for this p16^INK4a immunohistochemical negativity needs to be established. Furthermore, it has been suggested that p15^INK4b may also be an important target for inactivation in certain type of tumors (17). Recently, Tanaka et al. showed that this gene was co-deleted with p16^INK4a in 53% of the ESCC cell lines they analysed (20). However, the status of p15^INK4b gene in primary ESCC has not been reported.

To better understand the role of both p16^INK4a and p15^INK4b gene alterations in ESCC, we analysed 60 ESCC samples from a high incidence area in Henan, China. Multiple molecular analyses were performed in parallel to examine several potential types of aberrations affecting the p16^INK4a and p15^INK4b genes in the tumor samples. Matched pairs of normal and tumor DNA samples were screened for homozygous deletion using multiplex-PCR, and for frequency of LOH at 9p21 using two microsatellite markers D9S942 and D9S171. Among these samples, 34 were further analysed by PCR-methylation assay to evaluate aberrant methylation at the gene promoter regions and by SSCP-sequencing to examine p16^INK4a mutation.

	Materials and methods

Top Abstract Introduction Materials and methods Results Discussion References

 
Tumor samples and DNA preparation
Sixty surgically-resected esophageal specimens containing both tumor and nearby normal tissues were collected between 1991 and 1993 from patients with primary ESCC in Linxian (now Linzhou City), Henan Province, China, and were fixed in 80% ethanol. Approximately 20 mg of normal tissue from each sample was used directly for DNA extraction. All tumor tissues were embedded in paraffin, cut into 5 µm serial sections and deparaffinized before use. H&E staining and histopathological examination were performed on representative sections from each sample block to determine the tumor region. Tumor tissues were dissected from the neighboring four successive sections. Generally, contamination of the normal cells was ~20–25% mainly due to the fibroblast and inflammatory cells surrounding the cancer nests which were difficult to remove. Genomic DNA was extracted by proteinase K digestion and phenol–chloroform extraction.

PCR amplification
Primer sequences for all PCR amplifications in this study are shown in Table I. Sets 1 (GDB:188218) and 4 (GDB:370738) were obtained from the GDB database through internet, and were used for LOH assay. Sets 2 and 5 (21) were used in the PCR-methylation assay. Set 3 was used for p15^INK4b gene deletion analysis. Sets 6 and 7 were used in deletion assay and single strand conformation polymorphism (SSCP)-sequencing of p16^INK4a gene. All primers not given the source were designed using a primer picking software developed by Whitehead Institute/MIT Center for Genome Research. For PCR reaction, genomic DNA was amplified in a total volume of 25 µl containing 20 pmol of each primer, 250 µmol of dNTP, 1.4–2.0 mM MgCl₂, standard buffer and 0.5 U Taq polymerase (all from Gibco BRL). Mg²⁺ concentration, PCR cycles and annealing temperature (T_m) were optimized for each assay in pilot experiments.

View this table: [in this window] [in a new window]   Table I. Oligonucleotides for determining p16INK4a and p15INK4b aberrations

 
PCR-based gene deletion and LOH analysis
Homozygous deletion at the p16^INK4a and p15^INK4b loci was analysed by comparative multiplex PCR using the D9S171 product or ß-actin fragment as internal standards, and a closely sized PCR fragment of the gene as the primary target. The D9S171 microsatellite marker was originally selected to access LOH at 9p21. It was also used as an internal control in addition to ß-actin because it was never found to be homozygously deleted and is relatively close to the target genes on the same chromosome. This provides a similar exposure rate of both the target and the control to the Taq polymerase, and can also identify LOH at D9S171 simultaneously. However, p15^INK4b deletion was only analysed with the ß-actin internal control, because the p15^INK4b exon 2 has a size overlap with the D9S171 product.

To ensure linear amplification, PCR cycle number for tumor and normal DNA reactions were experimentally determined based on the product intensity versus cycle number gradient curve using a modified protocol originally described by Gonzales-Zulueta et al. (16). For PCR amplification, about 200 ng normal and 60 ng tumor genomic DNA were used. Forward primers of both target gene and control were radiolabeled at the 5' end with [-³³P]dATP. Typically, the following PCR conditions were used: 5 min denaturation at 95°C before adding the Taq polymerase; 2 min at 95°C, 2 min at T_m, 2 min at 72°C for initial cycle; then 27–29 cycles of 90 s at 95°C, 45 s at T_m, 45 s at 72°C; followed by 2 min at 72°C. The PCR products were resolved on 6 or 20% polyacrylamide gel depending on product size difference. After autoradiography for 4 h, the film was developed and the ratio of intensity of the target gene versus control was measured and calculated using a computer imaging system (Image-Pro Plus, Media Cybernetics). A homozygous deletion was scored if the normalized signal intensity (target/ß-actin) in the tumor lane was <25% of that in the normal lane. The threshold was so chosen because of the possible 20–25% non-cancerous cell contamination in our tumor samples. LOH at D9S171 was determined simultaneously with p16^INK4a deletion, whereas LOH at D9S942 was determined separately.

PCR-based methylation assay
Normal DNA (400 ng) and tumor DNA (80 ng) were digested for 4 h with 10 U methylation-sensitive (CfoI, HpaII, SmaI) or methylation-insensitive (MspI) restriction enzymes. After 2 min incubation at 95°C to inactivate the enzymes, the DNAs were subject to a second round of digestion by another 10 U freshly added enzymes to improve specificity and completeness of digestion (all of the enzymes are from Gibco BRL and digestion was performed under manufacturer's instructions). Since p16^INK4a and p15^INK4b promoter regions have common methylation-sensitive restriction sites (i.e. HpaII), the same digested DNA samples were used to determine methylation status of both genes. One hundred nanograms of normal and 20 ng of tumor DNA were amplified by PCR using primer set 2 or 5 which flanks the restriction sites, and with ß-actin as internal control. The PCR conditions were similar to those described in the deletion assay with cycle number fine-tuned to limit the amplification to a linear range. PCR products were electrophoresed on 3% agarose gels. Same amounts of undigested and MspI-digested DNA were used as the controls in every sample examined. Experiments were repeated three times for all samples to ensure reproducibility Signal intensities normalized by the internal control were measured and calculated as described in the deletion analysis. Methylation at a specific restriction site was determined by the ratio of the normalized intensity (target/internal standard) of the samples pretreated with methylation-sensitive enzymes over that of the undigested samples. A ratio >0.5 was defined as aberrant methylation.

Single strand conformation polymorphism (SSCP) and sequencing analysis
PCR was carried out as described with 1 µCi of -³²P-dATP. The labeled PCR products were denatured and electrophoresed on 6% polyacrylamide gel containing glycerol at 10–25 W for 7–10 h. Shifted bands from the tumor DNA reaction as compared with the adjacent normal sample, as well as the human placental control were eluted and amplified by a second PCR followed by agarose gel electrophoresis. The final PCR products were purified for DNA sequencing using the Wizard PCR Preps DNA Purification System from Promega (Madison, WI) according to manufacturer's instructions. DNA sequencing was performed using the same primer for PCR–SSCP. DNA was sequenced with the Thermal Sequenase Cycle Sequencing Kit from Amersham Life Science (Cleveland, OH) according to the manufacturer's instructions. In cycle sequencing, the thermal cycle consisted of 2 min at 94°C; 20 cycles of 1 min at 94°C, 1 min at 58°C and 1 min at 72°C; 10 cycles of 1 min at 94°C and 1 min at 72°C; and 1 min at 72°C. The reaction mixture was loaded on a 6% polyacrylamide sequencing gel made from the Sequagel DNA Sequencing Solutions (National Diagnostics, Atlanta, GA).

	Results

Top Abstract Introduction Materials and methods Results Discussion References

 
LOH at 9p21 loci in primary ESCC
Genomic DNA obtained from 60 matched pairs of primary tumors and nearby normal tissues were examined for incidence of LOH at 9p21 where both p16^INK4a and p15^INK4b reside. Altogether, 17 of the 51 informative cases were found to harbor LOH at least one of the two loci we analysed (Table II). However, frequency of LOH at individual locus was significantly different. Allelic loss at D9S942 (Figure 1a), which is located close to p16^INK4a in a region up to 35kb long between p16^INK4a and p15^INK4b, was evident in eight of the 47 informative cases. Whereas nine out of the 19 informative cases showed allelic imbalance at D9S171 (Figure 1b), which is ~2–3 Mb centromeric to the p15^INK4b gene. No case was found to contain LOH at both loci.

View this table: [in this window] [in a new window]   Table II. Inactivation of the p16INK4a and p15INK4b genes in ESCCa

 

View larger version (48K): [in this window] [in a new window]   Figure 1. LOH and MI at 9p21 loci in primary ESCC. PCR products of matched pairs of normal (N) and tumor (T) were loaded next to each other for comparison. Hemizygous loss of each marker are indicated by an arrow. (a) Representatives of LOH and MI at microsatellite marker D9S942. Case 910731 showed retention of both alleles at this locus, whereas tumors from cases 910783 and 910743 showed allelic imbalance with loss of a lower or an upper allele, respectively. Case 910804 harbored a MI in tumor as evidenced by the down shift of both alleles caused by a deletion of two trinucleotide repeats in both alleles. (b) LOH at D9S171. Loss of either the upper or the lower allele was seen in tumors from the first four cases. Case 920951 retained both alleles at this locus, and cases 920953 and 910751 were examples of two types of non-informative cases.

 
In all 60 pairs of tumor and normal DNA samples, retention of at least one allele of the D9S171 microsatellite marker was observed. However, there were six cases showing no PCR product for the similar sized D9S942 marker after several attempts. This implies a possible homozygous deletion at the D9S942 locus. Microsatellite instability (MI) at these two loci was also monitored in the same analysis. Only two cases were found to harbor MI in marker D9S942. Interestingly, in one case, variation of the number of nucleotide repeats was seen in both alleles (Figure 1a).

Homozygous deletion of p16^INK4a and p15^INK4b genes in primary ESCC
Comparative multiplex PCR showed that 10 out of the 60 primary ESCC cases had homozygous deletions in the p16^INK4a locus. The deletion was evidenced by the loss of the exon 2 PCR product while retaining the D9S171 product in a multiplex PCR reaction (Figure 2a), and was confirmed by parallel PCR reactions targeting exon 1 with ß-actin as control. A significantly higher frequency of homozygous deletion at the p15^INK4b locus, however, was observed in the same sample collection, with 21 out of 60 devoid of exon 2 product (Figure 2b). Among these 21 cases, 17 contained deletion only at p15^INK4b loci and four had dual deletions at both p16^INK4a and p15^INK4b loci. All of the four cases with dual deletions at both loci were among the six cases that failed to produce the D9S942 product during the LOH study, implying a large and contiguous deletion encompassing the p16^INK4a-p15^INK4b region in these samples. For the remaining two cases with no D9S942 amplification, one was found to have a deletion at p16^INK4a, the other had a p15^INK4b deletion. Altogether, 27 out of the 60 primary ESCC patients contain homozygous deletion at the 9p21 region, with p15^INK4b being the most frequently targeted gene. The remaining 33 samples had no homozygous deletion in either of the genes (Table II).

View larger version (47K): [in this window] [in a new window]   Figure 2. Homozygous deletion (HD) of p16INK4a and p15INK4b genes in primary ESCC detected by multiplex PCR. Deletion was defined by the target/control ratio in the tumor lane being <25% of that of the normal lane. (a) HD of p16INK4a was seen in the tumors from cases 933495, 920976 and 6. All of them showed the absence of the p16INK4a exon 2 product whereas retained the D9S171 band in the same PCR reaction. Case 920928 was a comparative case in which no HD was observed. (b) HD of the p15INK4b gene. c1, control DNA amplified with p15 amplifier only; c2, control DNA amplified with p15 and ß-actin amplifiers. Except for 910673, all other cases contain HD as evidenced by no or very weak exon 2 product in the tumor samples. Case 933945 also contained codeletion at p16INK4a locus, as seen in (a).

 
Relationship between the p16^INK4a and p15^INK4b homozygous deletion and 9p21 LOH was studied (Table III). There was no significant association between LOH at D9S942 and homozygous deletion of the p16^INK4a gene (P = 0.488). The incidence of D9S942 LOH, however, was significantly higher in samples with no p15^INK4b deletion (P = 0.023). Nevertheless, LOH at D9S171 is strongly associated with homozygous deletion at p15^INK4b (P = 0.017), but not with that at p16^INK4a.

View this table: [in this window] [in a new window]   Table III. Correlation of D9S942 or D9S171 LOH with homozygous deletion of p16INK4a or p15INK4b in ESCC

 
Aberrant methylation of the CpG island in p16^INK4a and p15^INK4b promoter region
Among the 60 cases of ESCC we analysed, 34 had yielded enough genomic DNA for us to examine the methylation status of the p16^INK4a and p15^INK4b promoter regions depicted in Figure 3a. Aberrant methylation of p16^INK4a in the tumor DNA was detected in 17 of the 34 cases (Figure 3b). Among them, 13 were found to be hypermethylated as manifested by their resistance to all three methylation-sensitive restriction enzymes while being sensitive to the methylation-insensitive enzyme MspI. The remaining four cases contained partial methylation as shown by their sensitivity to less than three of the methylation-sensitive enzymes, a phenomena indicating that methylation in these cases did not cover the entire CpG island and, therefore, exposing some of the sites to enzymatic restriction. Among these four cases, two cases showed abnormal methylation pattern in which the targeted regions were resistant to MspI restriction, but sensitive to one of the methylation-sensitive enzymes. This suggests the existence of an unconventional methylation pattern such as ^mCCGG, rather than the standard C^mCGG pattern in these patients, which would confer their resistance to MspI digestion. Out of the 17 methylated cases of ESCC, 14 cases did not have methylation in the normal DNA. In the remaining three cases, partial methylation, which covered a subset of the sites methylated in the tumors was seen in normal DNA (data not shown). There were 17 cases (50%) in which no PCR product was obtained from the genomic DNA after treatment with any one of the methylation-sensitive enzymes. Among them, four samples did not produce PCR products even when not treated with the enzymes, and they all corresponded to the p16^INK4a homozygous deletion cases found in the deletion analysis. The other 13 cases could undergo PCR normally when no restriction enzymes were applied and, therefore, were believed to contain no methylation at p16^INK4a promoter region.

View larger version (51K): [in this window] [in a new window]   Figure 3. PCR-methylation assay of p16INK4a and p15INK4b genes in primary ESCC samples. Genomic DNAs were treated with methylation-sensitive (C, CfoI; H, HpaII; S, SmaI) or methylation-insensitive (M, MspI) restriction enzymes, respectively, prior to PCR amplification. Aberrant methylation is defined with the criterion described in the Materials and methods. As controls, PCR using methylation-insensitive restriction enzyme (M) treated DNA yields no positive signal, and PCR of undigested DNA (U) always gives positive signal. Matched normal (N) and tumor (T) assays were loaded in adjacent panels for comparison. (a) Schematic representation of the promoter regions being analysed. The p16INK4a target is a 122 bp fragment 4 bp downstream from the start codon with three sensitive sites (21). The p15INK4b target is a 137 bp fragment 171 bp upstream from the start codon with two sensitive sites. The choice of this target region is based on previous description by Herman et al. (17), promoter sequence is available at Genbank with accession no. S75756. (b) Upper, hypermethylation of p16INK4a in tumor of case 910731; lower, no methylation in tumor of case 920934. The upper band in the gel is the ß-actin internal standard. (c) Left, hypermethylation of p15INK4b in tumor of case 910731; right, no methylation in tumor of case 910925.

 
A much lower frequency of aberrant methylation in the p15^INK4b promotor region was observed by analysing the same digested DNA samples used in p16^INK4a methylation assay with amplifiers specific for p15^INK4b promoter region. This region contains 2 HpaII/MspI sites and is 171 bp upstream exon 1. Only six samples out of 34 harbored aberrant methylation (Figure 3c), all of them also contained aberrant methylation in the p16^INK4a promoter. Among the 28 cases where no PCR product was obtained from the HpaII or MspI treated genomic DNA, 11 actually correspond to the cases previously detected with a homozygous deletion at p15^INK4b locus. The remaining 17 samples were determined as having no aberrant methylation of p15^INK4b promotor.

Intragenic mutation of p16^INK4a gene in ESCC
PCR products of exons 1 and 2 of p16^INK4a gene from the same 34 pairs of primary ESCC and normal tissues were analysed by SSCP. Except for four cases, which showed homozygous deletion at p16^INK4a locus, all others had specific PCR products. Only one mobility shift was detected in exon 1 of the 31 tumor samples, whereas all of the other normal and tumor DNAs showed normal SSCP pattern (Figure 4a). Upon sequencing, the exon 1 shifted band was determined to contain a microdeletion of 18 nucleotides starting at codon 20 (Figure 4b). This in frame microdeletion will result in a truncated form of p16^INK4a protein. Sequencing analysis of all the other cases showed normal p16^INK4a sequence.

View larger version (111K): [in this window] [in a new window]   Figure 4. Detection of a rare microdeletion in exon 1 of the p16INK4a gene in primary ESCC. (a) PCR–SSCP analysis of tumor DNA and normal control showed a mobility shift in tumor DNA indicated by an arrow from case 920922. No shifted band was observed from case 920934. (b) Sequencing of the recovered shifted band from case 920922 revealed a microdeletion of 18 bp starting at codon 20 (nucleotide 59) compared with the normal DNA.

 

	Discussion

Top Abstract Introduction Materials and methods Results Discussion References

 
In this study, a detailed analysis of the inactivation patterns of two related genes, p16^INK4a and p15^INK4b, was conducted in primary human ESCC. As summarized in Figure 5, preferential aberrant methylation of the p16^INK4a promoter and homozygous deletion of p15^INK4b gene were found to be frequent events in primary ESCC, along with some other less frequent aberrations.

View larger version (12K): [in this window] [in a new window]   Figure 5. Schematic representation of the predominant mechanisms inactivating p16INK4a and p15INK4b in ESCC. The first lane shows the location of the genes and polymorphic loci we analysed on chromosome 9p21. The telomeric end is on the left side and the centromeric end is on the right side. ESCC samples are classified into six classes according to their homozygous deletion (HD) pattern as shown in the lower part. The numbers left of the bars indicate the incidence of HD of each class and the deleted region is depicted as the breakage at the corresponding position. Aberrant methylation is represented as an arrow intersecting with a cross. The numbers next to it indicate the methylation incidence among samples analysed for methylation status in the corresponding class. The map is not drawn to scale. Result of methylation status is only available in a subset of samples in each class. It can be seen that aberrant methylation occurred more frequently in p16INK4a gene, whereas homozygous deletion was more frequent in p15INK4b gene. Furthermore, deletions in p16INK4a and p15INK4b were not correlated, but methylation at p15INK4b was always accompanied by a concomitant methylation at p16INK4a.

 
Previously, we have observed frequent absence of p16^INK4a immuno-reactivity in ESCC samples (22). In the present study, among the 28 cases in which we were able to analyse different types of p16^INK4a aberration status, 19 harbored at least one type of molecular alteration within the gene or in its vicinity that potentially inactivates the gene (Table IV). In those samples that had been previously examined for p16^INK4a immunohistochemistry (22), their p16^INK4a staining patterns generally agree with the p16^INK4a gene status. These results suggest that p16^INK4a is a frequent inactivating target comparable with p53 in ESCC (25). However, unlike p53, mutation in p16^INK4a gene is rare in our cases. Although we could not exclude the possibility that some mutations may remain undetected by our SSCP method, our observation is consistent with the recent report that p16^INK4a mutation is in general infrequent in ESCC (19). Among several inactivation modes, aberrant methylation, which can result in transcriptional silencing of the targeted gene, was the most frequent alteration found in p16^INK4a locus. Our frequency of aberrant methylation (50%) is significantly higher than that (19%) reported by Maesawa et al. (21) for a group of Japanese ESCC patients, but is consistent with the observation by Merlo et al. (15) in several other epithelium-derived cancers. To ensure data reliability, efforts were made to avoid incomplete digestion and false amplification using strict quality control procedures including double digestion, linear stage PCR and independent repeats. All results were based on semi-quantitative measurements. In most cases we studied (14/17), aberrant methylation of p16^INK4a was only observed in the tumor, but not in the normal tissues, which implied a transcription-silencing state of the gene only in tumor cells. However, in three cases, a less severe aberrant methylation was also detected in the `normal' samples dissected from regions close to the tumors. Whether this is due to aberrant methylation in the p16^INK4a promoter region of the preneoplastic cells needs further investigation. It has been reported for colon cancer that promoter methylation can be a very early event which precedes many genetic abnormalities such as p53 and Ras mutations (26). Recently, Wong et al. (27) also observed aberrant methylation in early lesions of esophageal adenocarcinoma. It is possible that in ESCC, promoter methylation also plays an important role in p16^INK4a inactivation during the early carcinogenic process.

View this table: [in this window] [in a new window]   Table IV. Summary of inactivation of the p16INK4a and p15INK4b genes by different mechanisms in ESCC

 
The moderate frequency (17%) of homozygous deletion of the p16^INK4a gene in comparison with the frequent aberrant methylation, suggested that deletion is not a major cause of p16^INK4a inactivation in ESCC. Previous studies showed that p16^INK4a homozygous deletions were predominantly observed in ESCC cell line and metastatic lesions (20,21), which imply deletion as a late event involved in the acquisition of invasiveness. We observed that LOH at D9S942, an indication of allelic imbalance of the nearby p16^INK4a gene, had the same frequency to that of the p16^INK4a homozygous deletion, although there was no correlation between them. Glendening et al. (28) have suggested that deletion of a single copy of p16^INK4a could potentially drive cancer development. This is probably because p16^INK4a functions as a CDK inhibitor by direct protein–protein binding. Even a 50% reduction of p16^INK4a expression could weaken its ability to suppress the CDK activity, thereby triggering a proliferation cascade. Hemizygous and homozygous deletions may represent mechanisms for p16^INK4a inactivation in addition to aberrant methylation in ESCC. In particular, since homozygous deletion is far more frequent during metastasis, it might be related to the acquisition of metastatic potential during late stage ESCC development, as suggested by Maesawa et al. (21).

For p15^INK4b, frequency of the aberrant promoter methylation was relatively moderate in contrast to the 50% frequency affecting p16^INK4a. Furthermore, all six cases with methylated p15^INK4b also contained a concomitantly methylated p16^INK4a; no p15^INK4b methylation was observed in samples where p16^INK4a was not methylated or deleted (Figure 5). This suggests that p16^INK4a is the primary target of aberrant methylation on 9p, and p15^INK4b is also methylated in some cases during this process. On the other hand, homozygous deletion was more frequently involved in p15^INK4b inactivation. Observed in 21 out of the 60 ESCC samples, our frequency is higher than many previous reports (13,29,30). The frequent p15^INK4b inactivation in this study may be related to the special patient population, which was from Linxian, China, a region with high incidence of ESCC. Except for four cases, all the p15^INK4b homozygous deletions were observed without a concomitant p16^INK4a deletion, which contrasts previous findings in many cell line studies (20). In studies of ESCC cell lines, Tanaka et al. (20) reported highly frequent deletion (50–60%) of both p16^INK4a and p15^INK4b, but no exclusive homozygous deletion of p15^INK4b was observed. It is possible that cellular immortalization during culturing imposes a selective pressure upon the accumulation of genetic defects. Quick and effective removal of the cell cycle regulatory genes like p16^INK4a and p15^INK4b can provide immediate growth advantage and may quickly accumulate during culturing process. However, in primary ESCC, previous reports showed that p16^INK4a deletion in primary ESCC ranges from 0 to 16% (20,21), which are close to our observation and are significantly lower than the frequency in cell line.

In our study, different frequencies of LOH were observed at D9S942 and D9S171, and no cases harbored LOH at both loci, suggesting a non-uniform alteration mode on 9p21. The LOH frequency at the 9p21 region centromeric to p15^INK4b (D9S171) is 47%, much higher than that of D9S942 in the region between p16^INK4a and p15^INK4b (17%), and has a statistically significant association with deletion at p15^INK4b (P = 0.017). LOH on chromosome 9p region has been suggested as an early alteration in cancers such as head and neck SCC and esophageal adenocarcinoma (8,9,31). It is possible that the non-uniformity of LOH at 9p21 may also have originated during the early stage of ESCC development in our samples. The p15^INK4b gene may be lost through an early, followed by a late, deletion of one copy each of the gene. Since the D9S171 locus where frequent LOH was observed is ~2 Mb upstream p15^INK4b, we cannot exclude the possibility of the existence of additional critical gene(s) in the vicinity of p15^INK4b, such as p19^ARF (12,32), being co-targeted. The frequent p15^INK4b alteration (Table IV) suggests that it is an important factor involved in ESCC development. The importance of p15^INK4b inactivation in the neoplastic transformation may be directly linked to its role in controlling proliferation in normal cells. p15^INK4b can displace p27 from the cyclin D-CDK4 complex, thereby promoting p27 binding and inactivation of cyclin E-CDK2 (5). This non-constitutive CDK/CDI/pRb pathway can be induced by TGF-ß and plays an essential role of growth inhibition in normal cells (33). An early loss of p15^INK4b may cause the cell to be insensitive to the growth inhibition signal.

The observation that different mechanisms are involved in the inactivation of closely located genes like p16^INK4a and p15^INK4b is not unusual. Herman et al. (17) has reported the preferential methylation of p15^INK4b gene along with selective deletion of p16^INK4a gene in leukemia and glioma. Gene specific selective inactivation of p16^INK4a and p15^INK4b by different mechanisms might be due to differences of the local chromosomal structure of these two loci.

In summary, consistent with the previous finding that 9p21 genetic alteration is closely related to multiple types of tumor development, we showed that both p16^INK4a and p15^INK4b genes were subject to frequent inactivation in human primary ESCC. Our results imply that (i) inactivation of p16^INK4a and p15^INK4b involves different mechanisms, with p16^INK4a predominantly affected by aberrant methylation and p15^INK4b by deletion; (ii) the non-uniform alteration pattern at 9p21 in our ESCC samples may originate during early ESCC stage with the preferential LOH at the p15^INK4b-D9S171 loci where other critical genes may also exist and be targeted. Loss of function of p16^INK4a or p15^INK4b genes could have profound consequences, such as releasing the tight control of CDK/CDI/pRb pathway over cell cycle progression. To fully understand the role of p16^INK4a and p15^INK4b inactivation in ESCC development, it is important to analyse these inactivating events in precancerous lesions and their relationship with other gene alterations.

	Notes

 
² To whom correspondence and reprint requests should be addressed Email: csyang{at}rci.rutgers.edu

	Acknowledgments

 
This work was supported by NIH grant CA65871.

	References

Top Abstract Introduction Materials and methods Results Discussion References

 

1. Hartwell,L.H. and Kastan,M.B. (1994) Cell cycle control and cancer. Science, 266, 1821–1828.[Abstract/Free Full Text]
2. Morgan,D.O. (1995) Principles of CDK regulation. Nature, 374, 131–133.[Medline]
3. Xiong,Y., Zhang,H. and Beach,D. (1993) Subunit rearrangement of the cyclin-dependent kinases is associated with cellular transformation. Genes Dev., 7, 1572–1583.[Abstract/Free Full Text]
4. Reynisdottir,I. and Massague,J. (1997) The subcellular locations of p15(ink4b) and p27(kip1) coordinate their inhibitory interactions with cdk4 and cdk2. Genes Dev., 11, 492–503.[Abstract/Free Full Text]
5. Kamb,A., Gruis,N.A., Weaver-Feldhaus,J., Liu,Q., Harshman,K., Tavtigian, S.V., Stockert,E., Day,R.S.,III, Johnson,B.E. and Skolnick,M.H. (1994) A cell cycle regulator potentially involved in genesis of many tumor types. Science, 264, 436–440.[Abstract/Free Full Text]
6. Platz,A., Hansson,J., Mansson-Brahme,E., Lagerlof,B., Linder,S., Lundqvist,E., Sevigny,P., Inganas,M. and Ringborg,U. (1997) Screening of germline mutations in the CDKN2A and CDKN2B genes in Swedish families with hereditary cutaneous melanoma. J. Natl Cancer Inst., 89, 697–702.[Abstract/Free Full Text]
7. Cairns,C., Mao,L., Merlo,A., Lee,D.J., Schwab,D., Eby,Y., Tokino,K., van der Riet,P., Blaugrund,J.E. and Sidransky,D. (1994) Rates of p16 (MTS1) mutations in primary tumor with 9p loss. Science, 265, 415–416.[Free Full Text]
8. Shigemasa,K., Hu,C., West,C.M., Clarke,J., Parham,G.P., Parmley,T.H., Korourian,S., Baker,V.V. and O'Brien,T.J. (1997) p16 overexpression: a potential early indicator of transformation in ovarian carcinoma. J. Soc. Gynecol. Investig., 4, 95–102.[Web of Science][Medline]
9. Barrett,M.T., Sanchez,C.A., Galipeau,P.C., Neshat,K., Emond,M. and Reid,B.J. (1996) Allelic loss of 9p21 and mutation of the CDKN2/p16 gene develop as early lesions during neoplastic progression in Barrett's esophagus. Oncogene, 13, 1867–1873.[Web of Science][Medline]
10. Walker,D.G., Duan,W., Povovic,E.A., Kaye,A.H., Tomlinson,F.H. and Lavin,M. (1995) Homozygous deletion of the multiple tumor suppressor gene in the progression of human astrocytomas. Cancer Res., 55, 20–23.[Abstract/Free Full Text]
11. Okamoto,A., Hussain,S.P., Hagiwara,K. et al. (1995) Mutations in the p16INK4A/MTS1/CDKN2, p15INK4B/MTS2, and p18 genes in primary and metastatic lung cancer. Cancer Res., 55, 1448–1451.[Abstract/Free Full Text]
12. Quelle,D.E., Zindy,F., Ashmun,R.A. and Sherr,C.J. (1995) Alternative reading frames of the INK4a tumor suppressor gene encode two unrelated proteins capable of inducing cell cycle arrest. Cell, 83, 993–1000.[Web of Science][Medline]
13. Flores,J.F., Walker,G.J., Glendening,J.M. et al. (1996) Loss of the p16INK4a and p15INK4b genes, as well as neighboring 9p21 markers, in sporadic melanoma. Cancer Res., 56, 5023–5032.[Abstract/Free Full Text]
14. Herman,J.G., Merlo,A., Mao,L., Lapidus,R.G., Issa,J.P., Davidson,N.E., Sidransky,D. and Baylin,S.B. (1995) Inactivation of the CDKN2/p16/MTS1 gene is frequently associated with aberrant DNA methylation in all common human cancers. Cancer Res., 55, 4525–4530.[Abstract/Free Full Text]
15. Merlo,A., Herman,J.G., Mao,L., Lee,D.J., Gabrielson,E., Burger,P.C., Baylin,S.B. and Sidransky,D. (1995) 5' CpG island methylation is associated with transcriptional silencing of the tumor suppressor p16/CDKN2/MTS1 in human cancers. Nat. Med., 1, 686–692.[Web of Science][Medline]
16. Gonzalez-Zulueta,M., Bender,C.M., Yang,A.S., Nguyen,T., Beart,R.W., Van Tornout,J.M. and Jones,P.A. (1995) Methylation of the 5' CpG island of the p16/CDKN2 tumor suppressor gene in normal and transformed human tissues correlates with gene silencing. Cancer Res., 55, 4531–4535.[Abstract/Free Full Text]
17. Herman,J.G., Jen,J., Merlo,A. and Baylin,S.B. (1996) Hypermethylation-associated inactivation indicates a tumor suppressor role for p15INK4B. Cancer Res., 56, 722–727.[Abstract/Free Full Text]
18. Herman,J.G., Civin,C.I., Issa,J.J., Collector,M.I., Sharkis,S.J. and Baylin, S.B. (1997) Distinct pattern of inactivation of p15INK4B and p16INK4A characterize the major types of hemotological malignancies. Cancer Res., 57, 837–841.[Abstract/Free Full Text]
19. Chan,W.C., Tang,C.M.C., Lau,K.W. and Lung,M.L. (1995) p16 tumor suppressor gene mutations in Chinese esophageal carcinomas in Hong Kong. Cancer Lett., 115, 201–206.
20. Tanaka,H., Shimada,Y., Imamura,M., Shibagaki,I. and Ishizaki,K. (1997) Multiple types of aberrations in the p16(INK4a) and the p15(INK4b) genes in 30 esophageal squamous-cell-carcinoma cell lines. Int. J. Cancer, 70, 437–442.[Web of Science][Medline]
21. Maesawa,C., Tamura,G., Nishizuka,S., Ogasawara,S., Ishida,K., Terashima,M., Sakata,K., Sato,N., Saito,K. and Satodate,R. (1996) Inactivation of the CDKN2 gene by homozygous deletion and de novo methylation is associated with advanced stage esophageal squamous cell carcinoma. Cancer Res., 56, 3875–3878.[Abstract/Free Full Text]
22. Yang,G.-Y., Zhang,Z., Jie,L., Seril,D., Wang,L.-D., Goldstein,S. and Yang,C.S. (1997) Immunohistochemical studies of Waf1p21, p16, pRb, and p53 in human esophageal carcinoma and neighboring epithelia from a high-risk area in northern China. Int. J. Cancer, 72, 1–6.[Web of Science][Medline]
23. Shibagaki,I., Shimada,Y., Wagata,T., Ikenaga,M., Imamura,M. and Ishizaki,K. (1994) Allelotype analysis of esophageal squamous cell carcinoma. Cancer Res., 54, 2996–3000.[Abstract/Free Full Text]
24. Muzeau,F., Flejou,J.F., Thomas,G. and Hamelin,R. (1997) Loss of heterozygosity on chromosome 9 and p16 (MTS1, CDKN2) gene mutations in esophageal cancers. Int. J. Cancer, 72, 27–30.[Web of Science][Medline]
25. Gao,H., Wang,L.-D., Zhou,Q., Hong,J.-Y., Huang,T.-Y. and Yang,C.S. (1994) p53 tumor suppressor gene mutation in early esophageal precancerous lesions and carcinoma among high-risk populations in Henan, China. Cancer Res., 54, 4342–4346.[Abstract/Free Full Text]
26. Issa,J., Baylin,S.B. and Herman,J.G. (1997) DNA methylation changes in hematologic malignancies—biologic and clinical implications. Leukemia, 11, Suppl. S7–S11.
27. Wong,D.J., Barrett,M.T., Stoger,R., Emond,M.J. and Reid,B.J. (1997) p16ink4a promoter is hypermethylated at a high frequency in esophageal adenocarcinomas. Cancer Res., 57, 2619–2622.[Abstract/Free Full Text]
28. Glendening,J.M., Flores,J.F., Walker,G.J., Stone,S., Albino,A.P. and Fountain,J.W. (1995) Homozygous loss of the p15INK4B gene (and not the p16INK4 gene) during tumor progression in a sporadic melanoma patient. Cancer Res., 55, 5531–5535.[Abstract/Free Full Text]
29. Gonzalez,M.V., Pello,M.F., Lopez-Larrea,C., Suarez,C., Menendez,M.J. and Coto,E. (1997) Deletion and methylation of the tumour suppressor gene p16/CDKN2 in primary head and neck squamous cell carcinoma. J. Clin. Pathol., 50, 509–512.[Abstract/Free Full Text]
30. Yamada,Y., Hatta,Y., Murata,K. et al. (1997) Deletions of p15 and/or p16 genes as a poor-prognosis factor in adult T-cell leukemia. J. Clin. Oncol., 15, 1778–1785.[Abstract/Free Full Text]
31. Califano,J., Riet,P., Westra,W. et al. (1996) Genetic progression model for head and neck cancer: implication for field cancerization. Cancer Res., 56, 2488–2492.[Abstract/Free Full Text]
32. Haber,D.A. (1997) Splicing into senescence: the curious case of p16 and p19^ARF. Cell, 91, 555–558.[Web of Science][Medline]
33. Hannon,G.J. and Beach,D. (1994) p15INK4B is a potential effector of TGF-ß-induced cell cycle arrest. Nature, 371, 257–261.[Medline]

Received May 21, 1998; revised August 14, 1998; accepted August 25, 1998.

CiteULike    Connotea    Del.icio.us    What's this?

This article has been cited by other articles:

				J. Wang, A. J. Sasco, C. Fu, H. Xue, G. Guo, Z. Hua, Q. Zhou, Q. Jiang, and B. Xu Aberrant DNA Methylation of P16, MGMT, and hMLH1 Genes in Combination with MTHFR C677T Genetic Polymorphism in Esophageal Squamous Cell Carcinoma Cancer Epidemiol. Biomarkers Prev., January 1, 2008; 17(1): 118 - 125. [Abstract] [Full Text] [PDF]

				Y.-Z. Deng, P.-P. Chen, Y. Wang, D. Yin, H. P. Koeffler, B. Li, X.-J. Tong, and D. Xie Connective Tissue Growth Factor Is Overexpressed in Esophageal Squamous Cell Carcinoma and Promotes Tumorigenicity through -Catenin-T-cell Factor/Lef Signaling J. Biol. Chem., December 14, 2007; 282(50): 36571 - 36581. [Abstract] [Full Text] [PDF]

				M. Fang, D. Chen, and C. S. Yang Dietary Polyphenols May Affect DNA Methylation J. Nutr., January 1, 2007; 137(1): 223S - 228S. [Abstract] [Full Text] [PDF]

				M. Z. Fang, D. Chen, Y. Sun, Z. Jin, J. K. Christman, and C. S. Yang Reversal of Hypermethylation and Reactivation of p16INK4a, RAR{beta}, and MGMT Genes by Genistein and Other Isoflavones from Soy Clin. Cancer Res., October 1, 2005; 11(19): 7033 - 7041. [Abstract] [Full Text] [PDF]

				M. Z. Fang, C. Liu, Y. Song, G.-Y. Yang, Y. Nie, J. Liao, X. Zhao, Y. Shimada, L.-D. Wang, and C. S. Yang Over-expression of gastrin-releasing peptide in human esophageal squamous cell carcinomas Carcinogenesis, June 1, 2004; 25(6): 865 - 871. [Abstract] [Full Text] [PDF]

				Y. Wang, M. Z. Fang, J. Liao, G.-Y. Yang, Y. Nie, Y. Song, C. So, X. Xu, L.-D. Wang, and C. S. Yang Hypermethylation-Associated Inactivation of Retinoic Acid Receptor {beta} in Human Esophageal Squamous Cell Carcinoma Clin. Cancer Res., November 1, 2003; 9(14): 5257 - 5263. [Abstract] [Full Text] [PDF]

				Y. Nie, J. Liao, X. Zhao, Y. Song, G.-y. Yang, L.-D. Wang, and C. S. Yang Detection of multiple gene hypermethylation in the development of esophageal squamous cell carcinoma Carcinogenesis, October 1, 2002; 23(10): 1713 - 1720. [Abstract] [Full Text] [PDF]

				G. D. Stoner and A. Gupta Etiology and chemoprevention of esophageal squamous cell carcinoma Carcinogenesis, November 1, 2001; 22(11): 1737 - 1746. [Abstract] [Full Text] [PDF]

				Y. Nie, G.-y. Yang, Y. Song, X. Zhao, C. So, J. Liao, L.-D. Wang, and C. S. Yang DNA hypermethylation is a mechanism for loss of expression of the HLA class I genes in human esophageal squamous cell carcinomas Carcinogenesis, October 1, 2001; 22(10): 1615 - 1623. [Abstract] [Full Text] [PDF]

				L. Y. Y. Fong, V. T. Nguyen, J. L. Farber, K. Huebner, and P. N. Magee Early Deregulation of the p16ink4a-Cyclin D1/Cyclin-dependent Kinase 4-Retinoblastoma Pathway in Cell Proliferation-driven Esophageal Tumorigenesis in Zinc-deficient Rats Cancer Res., August 1, 2000; 60(16): 4589 - 4595. [Abstract] [Full Text]

				T. J.King, L. H.Fukushima, T. A.Donlon, A.D. Hieber, K. A.Shimabukuro, and J. S.Bertram Correlation between growth control, neoplastic potential and endogenous connexin43 expression in HeLa cell lines: implications for tumor progression Carcinogenesis, February 1, 2000; 21(2): 311 - 315. [Abstract] [Full Text] [PDF]

				E. P. Xing, Y. Nie, Y. Song, G.-Y. Yang, Y. C. Cai, L.-D. Wang, and C. S. Yang Mechanisms of Inactivation of p14ARF, p15INK4b, and p16INK4a Genes in Human Esophageal Squamous Cell Carcinoma Clin. Cancer Res., October 1, 1999; 5(10): 2704 - 2713. [Abstract] [Full Text] [PDF]

				E. P. Xing, G.-Y. Yang, L.-D. Wang, S. T. Shi, and C. S. Yang Loss of Heterozygosity of the Rb Gene Correlates with pRb Protein Expression and Associates with p53 Alteration in Human Esophageal Cancer Clin. Cancer Res., May 1, 1999; 5(5): 1231 - 1240. [Abstract] [Full Text] [PDF]

This Article Abstract FREE Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (46) Request Permissions Google Scholar Articles by Xing, E. P. Articles by Yang, C. S. Search for Related Content PubMed PubMed Citation Articles by Xing, E. P. Articles by Yang, C. S. Social Bookmarking          What's this?

Online ISSN 1460-2180 - Print ISSN 0143-3334

Copyright © 2009 Oxford University Press

Oxford Journals Oxford University Press

• Site Map
• Privacy Policy
• Frequently Asked Questions

Other Oxford University Press sites: Oxford University Press Oxford Journals China Oxford Journals Japan American National Biography Booksellers' Information Service Children's Fiction and Poetry Children's Reference Corporate & Special Sales Dictionaries Dictionary of National Biography Digital Reference English Language Teaching Higher Education Textbooks Humanities International Education Unit Law Medicine Music Online Products Oxford English Dictionary Reference Rights and Permissions Science School Books Social Sciences Very Short Introductions World's Classics