Clinical Cancer Research CTRC-AACR San Antonio Breast Cancer Symposium
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Xing, E. P.
Right arrow Articles by Yang, C. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Xing, E. P.
Right arrow Articles by Yang, C. S.
Clinical Cancer Research Vol. 5, 2704-2713, October 1999
© 1999 American Association for Cancer Research

Advances in Brief

Mechanisms of Inactivation of p14ARF, p15INK4b, and p16INK4a Genes in Human Esophageal Squamous Cell Carcinoma1

Eric Poe Xing2, Yan Nie2, Yunlong Song, Guang-Yu Yang, Yuyang Christine Cai, Li-Dong Wang and Chung S. Yang3

Laboratory for Cancer Research, College of Pharmacy, Rutgers, The State University of New Jersey, Piscataway, New Jersey 08854-8020 [E. P. X., Y. N., Y. S., G-Y. Y., Y. C. C., C. S. Y.], and Henan Medical University, Zhengzhou, Henan, 457500 China [L-D. W.]


   ABSTRACT
Top
 ABSTRACT
Introduction
 Materials and Methods
 Results
 Discussion
 REFERENCES
 
The 9p21 gene cluster, harboring growth suppressive genes p14ARF, p15INK4b, and p16INK4a, is one of the major aberration hotspots in human cancers. It was shown that p14ARF and p16INK4a play active roles in the p53 and Rb tumor suppressive pathways, respectively, and p15INK4b is a mediator of the extracellular growth inhibition signals. To elucidate specific targets and aberrations affecting this subchromosomal region, we constructed a detailed alteration map of the 9p21 gene cluster by analyzing homozygous deletion, hypermethylation, and mutation of the p14ARF, p15INK4b, and p16INK4a genes individually in 40 esophageal squamous cell carcinomas (ESCCs) and compared the genetic alterations with mRNA expression in 18 of these samples. We detected aberrant promoter methylation of the p16INK4a gene in 16 (40%), of p14ARF in 6 (15%), and of p15INK4b in 5 (12.5%) tumor samples. Most p16INK4a methylations were exclusive, whereas all but one of the p14ARF/p15INK4b methylations were accompanied by concomitant p16INK4a methylation. We detected homozygous deletion of p16INK4a in 7 (17.5%), of p14ARF-E1{beta} in 13 (33%), and of p15INK4b in 16 (40%) tumor samples. Most deletions occurred exclusively on the E1{beta}-p15INK4b loci. Two samples contained p14ARF deletion but with p16INK4a and p15INK4b intact. No mutation was detected in the p14ARF and p16INK4a genes. Comparative RT-PCR showed good concordance between suppressed mRNA expression and genetic alteration for p15INK4b and p16INK4a genes in the 18 frozen samples, whereas 5 of the 13 cases with suppressed p14ARF mRNA expression contained no detectable E1{beta} alteration but aberrations in the p16INK4a locus. Our results show that in human ESCCs, p14ARF is a primary target of homozygous deletion along with p15INK4b, whereas p16INK4a is the hotspot of hypermethylation of the 9p21 gene cluster. The frequent inactivation of the p14ARF and p16INK4a genes may be an important mechanism for the dysfunction of both the Rb and p53 growth regulation pathways during ESCC development.


   Introduction
Top
 ABSTRACT
 Introduction
Materials and Methods
 Results
 Discussion
 REFERENCES
 
The 9p21 chromosomal band is one of the most frequently altered genomic regions in human cancers (1) . Within a short distance of ~50 kb, this region harbors a gene cluster consisting of three genes, p14ARF, p15INK4b, and p16INK4a, all of which have putative tumor suppressor roles (2) . In addition to physical proximity, the genomic structures of these genes are remarkably interrelated (Fig. 1)Citation . p14ARF and p16INK4a transcripts are produced via utilization of a common coding sequence for exons 2 and 3, together with distinct sequences for promoter and exon 1 (3) . However, the resulting proteins are completely different because different reading frames are used for the respective translation processes (4) . The conservation in mammalian genome of this unique gene structure, usually seen in primitive organisms subject to genome size constrains, suggests either the possible ancient origin or the biological essentialness of the unitary inheritance of these two genes. p15INK4b is highly homologous to p16INK4a, particularly in exon 2, where they share 91% sequence identity (5) , indicating their origination by a gene duplication event. The 9p21 gene cluster is the first of its kind identified in the human genome associated with multiple tumor suppressor activities. Besides its functional importance in regulating cell proliferation, which makes this gene cluster a target of selective inactivation during carcinogenic process, there may be a physical basis underlining its frequent disruption in cancer. It was reported that at least two tightly clustered breakpoints exist within the cluster, the sequence context of which potentially facilitates illegitimate V(D)J recombinase activities (6) . Furthermore, the promoter regions of all three genes are highly abundant with CpG islands that are susceptible to hypermethylation (7) . It is intriguing why presumably important genes are clustered in such a manner that renders high susceptibility to genetic alterations.



View larger version (17K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 1. Genomic organization of the 9p21 gene cluster and schematic description of the involvement of p14ARF, p15INK4b, and p16INK4a gene products in the pRb and p53 antitumorigenetic pathways [based on a diagram that appeared in Robertson et al. (7) ]. Solid lines, regulatory steps of each pathway; dashed lines, cross-talk between the pRb and p53 pathways.

 
Recent studies have revealed that p14ARF and p16INK4a play active roles in the p53 and Rb growth-control pathways, respectively (Fig. 1Citation ; Ref. 8 ). p16INK4a is a cyclin-dependent kinase inhibitor functioning upstream Rb. It can negatively regulate cell cycle progression by preventing the phosphorylation (inactivation) of Rb during G1 phase (9) . p14ARF restrains cell growth by abrogating Mdm2 inhibition of the p53 activity, and therefore facilitates p53 mediated cell cycle arrest and apoptosis (10) . It was demonstrated that oncogenic Ras elicits an anti-tumorigenic response mediated by the up-regulation of both p14ARF and p16INK4a, which in turn activate the tumor suppressors p53 and pRb, respectively (11 , 12) . Recent study further showed that p14ARF provides a failsafe mechanism for defective Rb pathway by its inducibility via deregulated E2F-1 activity resulted from Rb inactivation (13) . Compared with p14ARF and p16INK4a, p15INK4b is less prominent as a tumor suppressor. In contrast to p16INK4a, which is activated by intracellular stimuli, p15INK4b suppresses cell growth in response to extracellular stimuli such as TGF-{beta} (5) .

Inactivation of the Rb and p53 tumor suppressor pathways is observed in most human cancers (1) . By virtue of its close involvement in both pathways, the p14ARF-p15INK4b-p16INK4a gene cluster at chromosome 9p21 may be a nexus of the cellular growth-control network, the inactivation of which results in collapse of the tumor suppression system. To date, a vast amount of data has demonstrated multiple types of genetic alterations on the 9p21 region, the prevalence of which varies with the type of tumors. For example, large homozygous deletions are common in head and neck tumors, bladder carcinomas, and malignant gliomas (14, 15, 16) . Transcriptional-inactivating promoter methylation was common in breast and colon cancer for the p16INK4a gene (17) and in leukemia for p15INK4b (17 , 18) . Recently, Robertson et al. (7) identified that the promoter of p14ARF gene is a CpG island and observed its hypermethylation in colon cancer cell lines, which was responsible for the gene inactivation (7) . Point mutation and small deletion of the p16INK4a locus was common in pancreatic adenocarcinoma, biliary tract cancers, and hereditary melanoma (reviewed in Ref. 19 ) and affected p16INK4a exclusively or together with p14ARF (4) . Despite the ample evidence of the 9p21 gene cluster as a frequent tumor target, it is not known whether all three genes in the cluster are indiscriminately affected in a tumor and whether all of the genes are uniformly disrupted by the same mechanism.

We reported previously an alteration pattern of the 9p21 region in human ESCC,4 in which p16INK4a was primarily affected by hypermethylation and less frequently subject to deletion, whereas p15INK4b was frequently homozygously deleted and occasionally methylated (20) . This pattern is different from reports on other types of cancers identifying p16INK4a as the main deletion target and p15INK4b as a bystander, which possibly play a small part in the tumor suppressor role (2) . In light of the recent elucidation of the tumor suppressor role of p14ARF, the first exon of which is only 12 kb downstream of the p15INK4b gene, and the presence of evidence of exclusive deletion of p14ARF-E1{beta} with the retention of both p15INK4b and p16INK4a in T-cell acute lymphoblastic leukemia (21) , we suspect that our previous result may indicate p14ARF-E1{beta} as being a primary target for inactivation in ESCCs. In consideration of this point, we thought it is necessary to investigate whether genomic alterations indeed associate with altered gene expressions in ESCCs. In the present study, we constructed a detailed alteration map of the 9p21 gene cluster by analyzing the homozygous deletion and aberrant methylation of the p14ARF, p15INK4b, and p16INK4a genes individually in 40 resected ESCC samples and analyzed the mRNA expression pattern of the respective genes in 18 frozen samples. We show that suppressed mRNA expression occurred at a high frequency for all three genes in primary ESCCs, and homozygous deletion is the primary cause leading to such inactivation for the p14ARF and p15INK4b genes, whereas aberrant methylation is the main event underlining p16INK4a inactivation.


   Materials and Methods
Top
 ABSTRACT
 Introduction
 Materials and Methods
Results
 Discussion
 REFERENCES
 
ESCC Specimen Preparation and RNA/DNA Extraction.
Eighteen primary ESCC specimens, together with their adjacent normal epithelial tissues, were collected from patients in Linzhou City (formerly Linxian) of northern China, a well-recognized high-risk area for ESCC. The samples were frozen in liquid nitrogen right after surgical resection and were kept in -70°C until processing for RNA/DNA extraction. Adjacent normal epithelia accompanying the tumors were identified through pathological examination and dissected directly from the tissue blocks. All tumor tissues were embedded in OTC and cryosected into 15-µm serial slides under -20°C. H&E staining and histopathological examination were performed on a representative section from each tissue block to identify the tumor region. Tumor tissues were then dissected from the neighboring three consecutive sections. Polyadenylated RNA was extracted from both the normal and tumor tissues using the Rneasy Mini kit (Qiagen, Santa Clarita, CA) according to the manufacturer’s specification. To ensure complete comparability of the analyses at the DNA and RNA levels, all leftover remnants from mRNA preparation were reused for genomic DNA extraction by proteinase K digestion and phenol/chloroform extraction. Another 22 pairs of tumor and normal DNAs from the paraffin-embedded ESCC specimens were directly taken from a sample collection (previously analyzed for p16INK4a and p15INK4b status; Ref. 20 ) and were further examined for p14ARF molecular alterations. All of the primers used in this study are listed in Table 1Citation .


View this table:
[in this window]
[in a new window]

  		Table 1 Summary of oligonucleotidesa

 
Methylation-specific PCR and Methylation Status Verification by Sequencing.
Aberrant methylation was detected using the CpG WIZ Amplification kit (Oncor, Gaithersburg, MD), following the protocol provided. Except for p14ARF gene, for which the primers used in the analysis were designed by ourselves (Table 1Citation and Fig. 2Citation ), primers for p16INK4a and p15INK4b analysis came with the kit. The DNA was modified by bisulfite reaction and then underwent two separate PCR reactions using unmethylated-specific and methylated-specific amplifiers. The bisulfite treatment converts the unmethylated cytosine to uracil and thus makes the sequence of unmethylated DNA read differently from that of the methylated DNA after modification. Methylation was determined by the presence of methylated-specific PCR products.



View larger version (7K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 2. Genomic positions of the PCR targets. The regions that were amplified and their genomic positions for hypermethylation analysis (solid lines) and HD (dashed lines) are shown. The alteration numbers and percentages are shown above or below the lines. The sequence of p14ARF was taken from GenBank (accession no. L41934). The sequences of p16INK4a and p15INK4b are as described previously (20) .

 
To verify the methylation status determined by methylation-specific PCR, products from methylated and unmethylated-specific PCR reactions were cloned using the TOPO TA cloning kit (Invitrogen, Carlsbad, CA). For both types of PCR reactions, five clones were randomly picked up, and the cloned fragments were sequenced directly.

Homozygous Deletion Analysis Using Duplex PCR.
HDs of the p16INK4a-E1{alpha}, E2, p14ARF-E1{beta}, and 15INK4b genes were examined using methods established previously in our laboratory (20) . In brief, each gene locus was analyzed by comparative multiplex PCR using the {beta}-actin fragment as internal standard and a closely sized PCR fragment of the gene as the primary target. About 100 ng of DNA were used for the duplex PCR amplification. Forward primers of both target gene and control were radiolabeled at the 5' end with [{gamma}-33P]dATP. After 27–29 cycles of PCR, the products were resolved on 6% denaturing or 8% nondenaturing polyacrylamide gel, depending on product size difference. After autoradiograph 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, Silver Spring, MD). A HD was scored if the normalized signal intensity (target/{beta}-actin) in the tumor sample was <25% of that in the normal epithelium. The threshold was so chosen because of an estimated upper limit of 20–25% noncancerous cell contamination in our tumor samples.

Measurement of mRNA Expression of the p14ARF, p15INK4b, and p16INK4a Genes.
cDNA was synthesized using the Advantage RT-for-PCR kit (Clontech, Polo Alto, CA) with random priming as recommended in the protocol provided. Comparative PCRs were performed with the GAPDH gene as internal standard and one of the gene in the 9p21 gene cluster as target. PCR cycle numbers (typically 27–30 cycles) were experimentally determined in pilot studies to limit the reaction in the linear stage. PCR products were resolved on 20% nondenaturing polyacrylamide gel or 3% agarose gel, and signal intensities were quantified using a computer image system. The levels of the gene transcripts were quantified by the ratio of the intensity of the target signal over the intensity of the GAPDH internal standard in the same duplex PCR reaction.

Statistical Pattern Recognition and Data Classification.
To establish a reasonable criterion discriminating the different mRNA expression states, expression level data measured from all tumor and normal samples for a single gene were subject to automatic pattern recognition and clustering using the standard kNN method (22) , where the "k" was set at 3. This procedure was performed independently for each gene of interest. A density-distribution curve was calculated and plotted using S-PLUS statistical package on a Sparc-3 Station, and evaluation of normality of the density distribution was done by comparing the distribution with a normal distribution using "qqnorm" plot done by S-PLUS. kNN classifier was coded in the C programming language and run on a Unix Ultra-Sparc 3 system. Detailed mathematical discussion, program description, and general data experiment of this method will be summarized in a separate report.


   Results
Top
 ABSTRACT
 Introduction
 Materials and Methods
 Results
Discussion
 REFERENCES
 
Aberrant Methylation of 5'-CpG Islands in 9p21 Gene Cluster.
The methylation status of the p14ARF, p15INK4b, and p16INK4a CpG islands were investigated in 40 ESCC samples together with their adjacent normal epithelial tissues. Among them, 18 were frozen samples and 22 were from a collection of paraffin-embedded specimens analyzed previously for p15INK4b and p16INK4a alterations (20) ; enough DNA remained for analyses of p14ARF alterations. Using methylation-specific PCR (Fig. 3a)Citation , among the 40 samples, we detected aberrant methylation of the p14ARF gene in 6 samples (15%), of the p15INK4b gene in 5 samples (12.5%), and of the p16INK4a gene in 16 samples (40%). The frequency of the p15INK4b and p16INK4a methylation is consistent with our previous observation (20) . The aberrant methylation of p14ARF in primary tumors, to our knowledge, has not been previously reported. Ten of the 16 p16INK4a-methylation cases were observed exclusively on the p16INK4a promoter region. On the other hand, four of the six cases with p14ARF methylation and four of the five cases with p15INK4b methylation occurred in samples in which all three genes have aberrant methylation. In many cases, in addition to the presence of methylation-specific PCR signal, a PCR signal corresponding to the unmethylated DNA sequence could also be detected from the same samples. The unmethylated signals might come from stromal cells present between the cancer nests, which could not be completely removed in our microdissection procedure. To verify the specificity of the PCR reaction, we sequenced the PCR products that resulted from both methylated-specific amplifiers and unmethylated-specific amplifiers and confirmed that, indeed, only in the unmethylated-specific PCR product was every "C," including those in the CpG dinucleotide, changed to a "T" as a result of chemical modification. In the methylated-specific PCR product, the CpG sequence was not changed, indicating it is indeed amplified from DNA containing methylation (Fig. 3b)Citation .



View larger version (68K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 3. Determination of the methylation status of the CpG islands (using p14ARF analysis as an example). a, methylation-specific PCR analysis. Two pairs of samples, representing presence of CpG island hypermethylation in tumor but not in normal DNA (no. 25), or no such event in either tumor or normal DNA (no. 26), are shown as examples. CpG methylation is determined by the presence of PCR product amplified by methylated-specific primers (M). Lack of methylation is evidenced by the presence of only PCR product amplified by unmethylated-specific primers (U) but not the methylated-specific product. b, confirmation of methylated-specific PCR by sequencing the PCR fragments. In the unmethylated-specific PCR product (25N-U, the cloned fragment as marked under the cloning vector), all of the "C"s were converted to "T"s (arrow), but in the methylated-specific product (25T-M), the "C"s in CpG dinucleotides were unchanged because of methylation, which resisted chemical modification by bisulfite.

 
Homozygous Deletion of the p14ARF, p15INK4b, and p16INK4a Genes.
Gene deletion was examined by differential PCR analysis of the genomic DNA. Duplex PCR amplification was performed to generate a 177-bp fragment covering exon 1{beta} of the p14ARF gene, a 172-bp fragment in the exon 2 of p15INK4b, a 204-bp exon 1{alpha} fragment of p16INK4a, or a 222-bp exon 2 fragment of p16INK4a, together with a 187-bp fragment of the {beta}-actin gene as a reference (Fig. 4)Citation . The {beta}-actin fragment was amplified in all 40 pairs of normal and tumor samples. The p14ARF-E1{beta} product was amplified in all 40 normal samples, but it was either not detected or had a much lower intensity in 13 tumor samples (32.5%). We presumed that these 13 cases contained a homozygous deletion at the first exon of the p14ARF gene, and the low residue signals may come from wild-type stromal and infiltrative cells. Likewise, together with the previously analyzed paraffin samples, homozygous deletion of p15INK4b gene was observed in 16 (40%) of the 40 cases. In contrast, a relatively lower percentage of samples showed homozygous deletions of p16INK4a. Seven of the 40 samples (17.5%) contained homozygous deletion at E2. We performed the homozygous deletion analysis for E1{alpha} on the 18 frozen samples (nos. 25–62) that had enough DNA for the study. Four of them showed homozygous deletion at E1{alpha}. The deletion pattern of E1{alpha} is the same as that of E2, suggesting that E2 was always codeleted with E1{alpha} because of their physical proximity. Only five cases contained homozygous deletion in all three loci, and 13 cases had deletion only at the centrameric end of the cluster (E1{beta}-15INK4b loci), whereas exclusive deletion of the p16INK4a at the telemeric end was observed in only two cases. In many cases, E1{beta} of p14ARF was deleted together with p15INK4b. However, two samples (nos. 38 and 920925) contained homozygous deletion at p14ARF only.



View larger version (83K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 4. HD of the p14ARF-E1{beta}, p15INK4b, and p16INK4a-E1{alpha}, E2 in primary ESCCs determined by duplex PCR. Arrows, band positions of the genes being examined; the remaining band in each gel-panel is the {beta}-actin internal control. In the upper and lower panels, the PCR products were resolved on 8% nondenaturing polyacrylamide gel. HD can be seen in sample 28 for both p15INK4b and p16INK4a. In the middle two panels, the PCR products were resolved on 6% sequencing gel. HD of p14ARF-E1{beta} is apparent in samples 28 and 38. HD of p16INK4a-E1{alpha} can be seen in sample 28 only. P, positive control using human placenta DNA.

 
Mutational Analysis of E1{beta} of the p14ARF Gene.
Previously, we analyzed a collection of paraffin-embedded ESCC samples for p16INK4a gene mutation and detected no point mutation in the samples included in the present study (20) . We further performed PCR-single strand conformation polymorphism analysis of E1{beta} of the p14ARF gene in the 18 frozen ESCC samples, using methods described previously. We found no E1{beta} mutation in these 18 samples, consistent with the reports from other laboratories (7) .

Characterization of p14ARF, p15INK4b, and p16INK4a mRNA Expression.
Levels of the p14ARF, p15INK4b, and p16INK4a transcripts were determined in 18 frozen ESCC samples by comparative RT-PCR analysis. A 264-bp fragment of the p14ARF gene transcript, a 428-bp fragment of the p15INK4b transcript, or a 176-bp fragment of the p16INK4a transcript were generated, respectively, with a 456-bp fragment of the GAPDH transcript coamplified as the internal standard. Although RT-PCR signals of the genes of interest were detected (at least with a minimal level) in almost all samples, nonparametric classification of the signal intensities (normalized by GAPDH signal) using the kNN method (22) revealed that the mRNA expression level of each of the genes had an apparent two-cluster distribution (Fig. 5)Citation . The two clusters correspond to minimally overlapped groups of data points and were therefore designated as class I and class II gene expression levels. Class I is tightly clustered with a small mean value and is well approximated by a Gaussian distribution. Class II has a significantly greater mean value and a much broader distribution not strictly Gaussian. Statistically, this suggests the presence of nonrandom regulating factors in samples showing class II mRNA levels but no such mechanisms underlining class I level mRNA expressions. On the average, the mRNA level of class II is ~10 times higher than that of class I. We interpret the two classes as representing distinctive states of gene expression; class I corresponds to a basal level, or suppressed state of gene expression, and class II corresponds to active gene expression. We defined the thresholds of suppressed and active p14ARF, p15INK4b, and p16INK4a gene expression as {[0, 0.26], [0.76, {infty}]}, {[0, 0.15], [0.76, {infty}]}, and {[0, 0.16], [0.50, {infty}]}, respectively. The thresholds for the basal levels that correspond to the Gaussian-like class I were simply set to µ + 2{varsigma}, whereas because of the spreadness of the distribution of class II, we take µ-{varsigma} as the lower limit of the active expression state to ensure enough separation between basal and active expression. Between the two thresholds is a rejection region. Any mRNA level that fell into this range was considered as indeterminable of its expression state, or simply, as an intermediate state. From Fig. 6Citation , It can be seen that chance occurrence to misclassify a value as indicating suppressed or active state is very low (strict calculation and mathematical discussion on this problem will appear in a separate report).



View larger version (16K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 5. Classification of the mRNA expression levels of the p14ARF, p15INK4b, and p16INK4a genes. Data points representing the measured expression signal intensities of each gene were classified by kNN classifier and were also analyzed using S-PLUS statistical software. The approximate probabilistic-distribution-density curve of the expression levels was plotted for each gene (X axis, expression level), and all of the values of the measured expression levels were marked by a * under the curve. It is apparent that the expression levels of each gene fall into two distinct classes. Arrow, thresholds for accepting a value as representing active or basal level expressions. Between the two thresholds is the rejection region, values that were considered as indeterminable in the case of expression state.

 


View larger version (70K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 6. mRNA expression pattern of p14ARF, p15INK4b, and p16INK4a genes. Arrows, band positions of the genes being examined; the remaining band in each gel panel is the GAPDH internal standard. RT-PCR products for p15INK4b analysis were resolved on 20% nondenaturing polyacrylamide gels, and products for p14ARF and p16INK4a were resolved on 3% agarose gels. P, positive control using human placenta RNA; M, 100-bp ladder marker. In the upper panel, samples 25, 60, and 36 display loss or very low level of p14ARF transcripts in the tumor. Among them, nos. 25 and 36 have a high level of gene transcripts in the matched normal tissues (pattern I), and no. 60 has a low level of gene transcripts in the normal tissue (pattern II). From the middle panel, samples 56 and 28 show pattern I repressed expression, and sample 25 has pattern II repressed expression in tumor. In the lower panel, samples 40 and 42 exhibit pattern I repressed expression, and samples 28 and 56 have pattern II repressed expression in tumor.

 
As summarized in Table 2Citation , a great variation of gene expression exists in normal tissues adjacent to the ESCCs. Among all 18 normal samples, 9 and 7 actively expressed the p14ARF and p15INK4b genes, respectively, and for both genes an additional three showed intermediate level (close to the active threshold) expression, whereas basal level expressions of these two genes were less frequent. These seemed to suggest a frequent presence of oncogenic and extracellular growth factor stimuli in the cells close to cancer. However, although presumably also inducible by oncogenic stimuli (11) , in most normal cases (14 of 18) the p16INK4a gene displayed only a basal level of expression. Taken together, there was no association in the expression of p14ARF, p15INK4b, and p16INK4a mRNA in the 18 normal samples, suggesting the existence of distinct regulations of each members of the 9p21 gene cluster.


View this table:
[in this window]
[in a new window]

  		Table 2 Alterations of 9p21 gene clustera

 
Although the 456-bp GAPDH signal was invariant in all 18 ESCC tumor samples, suppressed mRNA expression was observed in 12 (67%) tumor samples for p14ARF, in 9 (50%) tumor samples for p15INK4b, and in 12 (67%) tumor samples for p16INK4a (Table 1Citation and Fig. 6Citation ). There is one tumor sample (no. 47) containing p14ARF expression at an intermediate level slightly above the threshold of suppression state. Detailed analyses revealed that for all three genes, the suppression of gene expression in tumors had two patterns. For example, among the 12 tumor cases that failed to actively express p14ARF, 9 cases (nos. 25, 36, 38, 42, 43, 46, 47, 48, and 56) contained a high or intermediate level of p14ARF mRNA in the normal epithelia adjacent to the tumors (pattern I), and the remaining 4 cases (nos. 28, 33, 40, and 60) corresponded to a basal level of p14ARF expression in the accompanying normal epithelia (pattern II). Likewise, among the 9 tumor samples that showed suppressed p15INK4b expression, 5 were categorized as pattern I suppression, and 4 were categorized as pattern II suppression. Among the 12 tumor samples that showed suppressed p16INK4a expression, 3 showed pattern I suppression and 9 showed pattern II suppression. Taken together, only 4 of the 18 ESCC samples maintained an elevated level of mRNA expression for all of the three genes. The remaining 14 (77%) cases either showed suppression of all three genes (8 cases) or at least one of the three genes in the tumor (Fig. 7)Citation . These results suggested that the 9p21 gene cluster expression is frequently suppressed in ESCC cells.



View larger version (30K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 7. Summary of mRNA expression pattern of p14ARF, p15INK4b, and p16INK4a genes and their corresponding molecular status in 18 frozen ESCC samples. As shown in the histogram, only 4 of the 18 samples (nos. 26, 32, 50, and 53) maintain active expression of all three genes in tumor, whereas 8 cases have low expression of all three genes in tumor.

 
Correlation Between Gene Expression and Molecular Alterations of the 9p21 Cluster.
All promoter hypermethylation and homozygous deletion events detected in the p14ARF, p15INK4b, and p16INK4a genes corresponded to an absence or significant repression of the mRNA expression of the respective genes. Taken together, among the 12 frozen tumor samples showing repression of p16INK4a expression, 8 contained hypermethylation and 4 had deletions of the p16INK4a gene. Whereas among the nine samples with repressed p15INK4b expression, three had hypermethylation and six had homologous deletion. For p14ARF, however, an exact correspondence of altered gene expression to alteration at E1{beta} locus was not observed; among the 12 cases showing repressed expression, 2 had promoter hypermethylation, 6 had E1{beta} deletion, and 4 had no detectable genetic alterations at the E1{beta} locus. Sample 47, which had a close-to-suppression intermediate p14ARF expression, also harbored p16INK4a methylation.


   Discussion
Top
 ABSTRACT
 Introduction
 Materials and Methods
 Results
 Discussion
REFERENCES
 
The chromosome 9p21 region harbors three genes: p14ARF, p15INK4b, and p16INK4a, all of which have growth-suppressive activities. On the basis of the unusual genomic organization of this gene cluster and the functional relevance of these genes in both the Rb and p53 pathways, we reasoned that genes within this compact cluster can be differently regulated and are potentially susceptible to distinctive inactivating mechanisms, rather than a single deletion or methylation event occurring across the whole region. In the present study, we elucidated the mechanisms and approximate frequencies of aberrations affecting this subchromosomal region in human ESCCs and observed good correspondence between the molecular alteration and loss of gene expression.

Previous studies have demonstrated that the expression of p14ARF, p15INK4b, and p16INK4a normally undergo elevation in highly proliferative cells such as those affected by oncogenic stimuli (e.g., H-rasV12 or SV40 T-antigen, which activate p14ARF and p16INK4a; Refs. 11 and 23 ), or growth factors (e.g., TGF-{beta}, which activates p15INK4b; Ref. 5 ) if the genes are intact. Upon activation, these gene products can counteract excessive cell proliferation. In our study, elevated levels of gene expressions were frequently observed for p14ARF and p15INK4b, and sometimes for p16INK4a in the morphologically normal epithelia adjacent to tumors, suggesting that abnormal growth stimuli might exist in these tissues. The p14ARF, p15INK4b and p16INK4a possibly trigger an antiproliferation response there. The expression of p14ARF, p15INK4b, and p16INK4a, however, was frequently repressed in ESCC tumor samples; among which, 44% (8 of 18) of the samples showed repression of all three genes.

To date, two mechanisms have been postulated as primary causes of inactivation of the potential tumor suppressor genes on 9p21: homozygous deletion and promoter hypermethylation (24) . p16INK4a has been widely regarded as the major target of 9p21 deletion. But because of the dual coding capacity of the p16INK4a-p14ARF locus, a deletion occurred on p14ARF exons 2 or 3 could still disrupt p16INK4a. There is evidence in a murine system that much of the tumorigenic phenotype associated with p16INK4a deletion may in fact be attributable to disruption of the p14ARF gene (23) . However, whether p14ARF is a primary inactivation target is in debate. Point mutations in E1{beta} are rare. Mutations in E2 of p14ARF and p16INK4a almost exclusively inactivate p16INK4a protein only (25) . Deletion of p14ARF is either accompanied by deletion of p15INK4b or p16INK4a or both. Hypermethylation of p14ARF was observed in several colon cancer cell lines, which also contained hypermethylation on p16INK4a (7) .

In our study, we found that deletion at the p16INK4a locus (7 of 40) was significantly less frequent than at p14ARF E1{beta} and p15INK4b loci (13 of 40 and 16 of 40, respectively). In fact, most of the p16INK4a deletions (five of seven) only occurred in cases harboring complete 9p21 gene-cluster deletion. These results suggest that p14ARF, rather than the p16INK4a gene, is more likely to be a main target of deletion at 9p21. Although mostly deleted together with p15INK4b, possibly due to their physical proximity, p14ARF-specific deletion was seen in two ESCC samples (nos. 38 and 920925), and such event has also been reported previously (21 , 26) . Underlining the structural basis of this type of genomic changes, a recent study revealed the existence of tightly clustered breakpoints close to the E1{alpha} and E1{beta} loci, and possibly also upstream of p15INK4b (21) . There is evidence that the rearrangements in this region may involve illegitimate V(D)J recombinase activity, which could contribute to frequent gene-specific deletions (6) . Recent studies suggested the importance of p14ARF as a potent tumor suppressor. Mice lacking p19ARF (the mouse homologue of p14ARF) develop a cancer phenotype (27) , p14ARF-null embryonic fibroblasts, that exhibit a high rate of spontaneous immortalization and could be transformed by oncogenic Ras (23) . The nature of p14ARF as a bona fide tumor suppressor gene in ESCC also owns to its remarkable frequency of transcriptional inactivation, 12 of 18 ESCCs, among the highest in the 9p21 gene cluster in our study. p14ARF can activate p53 both by neutralizing Mdm2, which destabilizes p53 (10) , and by interacting directly with p53 (28) in response to oncogenic stimuli. Therefore, the loss of p14ARF function in tumor cells as observed in our study potentially compromises the p53-mediated cell cycle arrest and apoptotic process therein.

A second deletion hotspot is the p15INK4b locus. Although in most cases p15INK4b deletion was accompanied by a concomitant E1{beta} deletion, 4 of the 16 deletion cases exclusively targeted p15INK4b. p15INK4b has often been considered as an innocent bystander of the deletion at 9p21 (24) . Previous evidence of p15INK4b alteration in ESCC has been scarce, and its tumor suppressor role in esophageal carcinogenesis is uncertain. Our observation of frequent and sometimes elusive p15INK4b deletion suggests that p15INK4b may itself be a tumor suppressor gene disrupted during ESCC development. Despite its close linkage and functional similarity to p16INK4a, p15INK4b plays a role of nonredundant cell cycle checkpoint. It is a mediator in the cell cycle control pathway originating from extracellular stimuli such as transforming growth factor {beta} and IFN-{alpha} (5 , 29) . In principle, this pathway is independent of the intracellular pathway mediated by p16INK4a and is equally crucial in maintaining a balanced cell cycle regulation. Tumorigenic transformation in somatic tissues are frequently preceded or accompanied by other cellular abnormalities, such as inflammation or increased proliferation. The growth factors secreted by the inflammatory cells and the increased cellular contact may produce such growth inhibition signal mediated by p15INK4b to preclude cell cycle progression and therefore provide a way of counteracting the transformation tendency. It is possible that inactivation of p15INK4b can desensitize the cell to such extracellular signals and as a result contribute to cancer development. Indeed, evidence of specific inactivation of the p15INK4b gene has been increasing in recent studies in acute lymphoblastic leukemia (30) and other solid tumors (summarized in Ref. 2 ).

Our observation of relatively infrequent homozygous deletion at the p16INK4a locus is consistent with most previous studies on primary ESCC tumors (31 , 32) but disagrees with the results from ESCC cell lines (summarized in Ref. 33 ). Hayashi et al. (34) suggested that simultaneous loss of p16INK4a and p14ARF expression, which was observed in 11 of 18 frozen tumor samples in our study, is an indicator of homozygous deletion of p16INK4a. However, we found that 8 of the 11 cases actually contained hypermethylation on the p16INK4a promoter. Most of the methylation took place exclusively on the p16INK4a promoter. On the basis of the significantly higher frequency (40%) of hypermethylation on p16INK4a compared with p14ARF (15%) and p15INK4b (12.5%), and its relatively low deletion rate, we propose that p16INK4a is a primary target of aberrant hypermethylation in ESCC. Our hypothesis of p16INK4a as a primary methylation target can easily explain observations in the previous report of cases that fail to express p16INK4a but express p14ARF readily (34) , because p14ARF is controlled by a different promoter (3) , which may not be comethylated with p16INK4a. The preferential methylation of the p16INK4a promoter may relate to its special local genomic structure, featured by a cluster of breakpoints located just 5' to E1{alpha}, the sequence content of which bears the hallmark of V(D)J recombinase activity (6) . It was shown that aberrant DNA structures, such as integration intermediates, seem to unleash de novo activity of the mammalian DNA methyltransferase (35) . In our study, the sharp difference of deletion rates between the p16INK4a and E1{beta}-p15INK4b loci suggests possible rearrangement hotspot upstream of the p16INK4a locus. Therefore, frequent p16INK4a promoter methylation may be relevant to the frequent rearrangement events upstream of p16INK4a. Transcriptional silencing promoter-methylation has been rarely observed for p14ARF and p15INK4b in epithelium-derived tumors. Our observation of both events revealed an alternative mode of p14ARF and p15INK4b inactivation in ESCCs, although at a low frequency. Except for one case with methylation exclusively on p14ARF, all methylation events beyond the p16INK4a gene have concomitant p16INK4a methylation, suggesting they have an association with p16INK4a methylation and can also be exclusively targeted.

Although all abnormalities of p16INK4a and p15INK4b expression correspond to an underlining homozygous deletion or hypermethylation of the gene, we were unable to establish a good match between the E1{beta} alterations and inactive p14ARF expression. Specifically, we observed four tumors samples that showed inactive p14ARF expression but contained neither detectable E1{beta} deletion or promoter methylation. Among them, one case (no. 48) can be easily explained by a deletion at p16INK4a because that event also removed part of the p14ARF gene sequence shared with p16INK4a. The remaining three cases (nos. 33, 36, and 40), together with a sample (no. 47) with lower intermediate level of p14ARF expression, all appeared to have intact p14ARF gene, except that they all contained hypermethylation in the downstream p16INK4a locus. Gonzalgo et al. (36) have shown in cell lines that hypermethylation of p16INK4a had no effect on the transcription of p14ARF. It would be very interesting to determine whether other factors, such as p53, which is stabilized by p14ARF and causes the down-regulation of p14ARF (7) , also contribute to p14ARF inactivation.

In summary, our results show that the newly identified growth suppressor p14ARF, together with p15INK4b, is a primary target of homozygous deletion, whereas p16INK4a is the hypermethylation hotspot in human ESCC. Such a polarity may reflect the presence of sequence-specific elements that favor such alteration or the effect of growth selection during cancer formation. Because recent studies have shown that oncogenic stimuli elicit the antitumorigenic response by up-regulating both p14ARF and p16INK4a, which in turn activate the tumor suppressors p53 and pRb, respectively (10 , 11) , the tumor suppressor role of the 9p21 gene cluster becomes apparent. The unique genomic structure and compact organization of these genes as a cluster may be essential for the highly coordinated regulation in maintaining a balanced Rb and p53 pathway function. Our observation of frequent coinactivation of p14ARF and p16INK4a, and even inactivation of the entire 9p21 gene cluster, provides additional evidence for the dysfunction of both Rb and p53 tumor suppression pathways in ESCC development.


   ACKNOWLEDGMENTS
 
We thank Jinshang Zhang of the Department of Statistics at Rutgers University for help in the statistical analysis of our data and in generating the density distribution plots on his workstation. We also thank Dr. Jie Liao for assistance in preparing the frozen tissue samples for our analysis.


   FOOTNOTES
 
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1 Supported by NIH Grant CA65781, facilities from National Institute of Environmental Health Sciences Center Grant ES 05022, National Cancer Institute Cancer Center Supporting Grant CA 72030, and National Natural Science Foundation of China (39840012). Back

2 These authors contributed equally to this study. Back

3 To whom requests for reprints should be addressed, at Laboratory for Cancer Research, College of Pharmacy, Rutgers University, Piscataway, NJ 08854-8020. Phone: (732) 445-5360; Fax: (732) 445-0687. Back

4 The abbreviations used are: ESCC, esophageal squamous cell carcinoma; HD, homozygous deletion; kNN, k-Nearest-Neighbor; GAPDH, glyceraldehyde-3-phosphate dehydrogenase. Back

Received 4/26/99; revised 8/ 2/99; accepted 8/ 2/99.


   REFERENCES
Top
 ABSTRACT
 Introduction
 Materials and Methods
 Results
 Discussion
 REFERENCES
 

  1. Sherr C. J. Cancer cell cycles. Science (Washington DC), 274: 1672-1677, 1996.[Abstract/Free Full Text]
  2. Carnero A., Hannon G. J. The INK4 family of CDK inhibitors. Curr. Top. Microbiol. Immunol., 227: 43-55, 1998.[Medline]
  3. Stone S., Jiang P., Dayananth P., Tavtigian S. V., Katcher H., Parry D., Peters G., Kamb A. Complex structure and regulation of the P16 (MTS1) locus. Cancer Res., 55: 2988-2994, 1995.[Abstract/Free Full Text]
  4. Quelle D. E., Zindy F., Ashmun R. A., Sherr C. J. Alternative reading frames of the INK4a tumor suppressor gene encode two unrelated proteins capable of inducing cell cycle arrest. Cell, 83: 993-1000, 1995.[Medline]
  5. Hannon G. J., Beach D. p15INK4B is a potential effector of TGF-{beta}-induced cell cycle arrest [see comments]. Nature (Lond.), 371: 257-261, 1994.[Medline]
  6. Cayuela J. M., Gardie B., Sigaux F. Disruption of the multiple tumor suppressor gene MTS1/p16(INK4a)/CDKN2 by illegitimate V(D)J recombinase activity in T-cell acute lymphoblastic leukemias. Blood, 90: 3720-3726, 1997.[Abstract/Free Full Text]
  7. Robertson K. D., Jones P. A. 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., 18: 6457-6473, 1998.[Abstract/Free Full Text]
  8. Chin L., Pomerantz J., DePinho R. A. The INK4a/ARF tumor suppressor: one gene–two products–two pathways. Trends Biochem. Sci., 23: 291-296, 1998.[Medline]
  9. Serrano M., Hannon G. J., Beach D. A new regulatory motif in cell-cycle control causing specific inhibition of cyclin D/CDK4 [see comments]. Nature (Lond.), 366: 704-777, 1993.[Medline]
  10. Pomerantz J., Schreiber-Agus N., Liegeois N. J., Silverman A., Alland L., Chin L., Potes J., Chen K., Orlow I., Lee H. W., Cordon-Cardo C., DePinho R. A. The Ink4a tumor suppressor gene product, p19Arf, interacts with MDM2 and neutralizes MDM2’s inhibition of p53. Cell, 92: 713-723, 1998.[Medline]
  11. Serrano M., Lin A. W., McCurrach M. E., Beach D., Lowe S. W. Oncogenic ras provokes premature cell senescence associated with accumulation of p53 and p16INK4a. Cell, 88: 593-602, 1997.[Medline]
  12. Palmero I., Pantoja C., Serrano M. p19ARF links the tumour suppressor p53 to Ras. Nature (Lond.), 395: 125-126, 1998.[Medline]
  13. Bates S., Phillips A. C., Clark P. A., Stott F., Peters G., Ludwig R. L., Vousden K. H. p14ARF links the tumour suppressors RB and p53. Nature (Lond.), 395: 124-125, 1998.[Medline]
  14. Hirama T., Koeffler H. P. Role of the cyclin-dependent kinase inhibitors in the development of cancer. Blood, 86: 841-854, 1995.[Free Full Text]
  15. Orlow I., Lacombe L., Hannon G. J., Serrano M., Pellicer I., Dalbagni G., Reuter V. E., Zhang Z. F., Beach D., Cordon-Cardo C. Deletion of the p16 and p15 genes in human bladder tumors [see comments]. J. Natl. Cancer Inst., 87: 1524-1529, 1995.[Abstract/Free Full Text]
  16. Cairns P., Polascik T. J., Eby Y., Tokino K., Califano J., Merlo A., Mao L., Herath J., Jenkins R., Westra W., Rutter L., Buckler A., Gabrielson E., Tockman M., Cho K. R., Hedrick L., Bova G. S., Issacs W., Koch W., Schwab D., Sidransky D. Frequency of homozygous deletion at p16/CDKN2 in primary human tumours. Nat. Genet., 11: 210-212, 1995.[Medline]
  17. Herman J. G., Merlo A., Mao L., Lapidus R. G., Issa J. P., Davidson N. E., Sidransky D., Baylin S. B. Inactivation of the CDKN2/p16/MTS1 gene is frequently associated with aberrant DNA methylation in all common human cancers. Cancer Res., 55: 4525-4530, 1995.[Abstract/Free Full Text]
  18. Herman J. G., Civin C. I., Issa J. P., Collector M. I., Sharkis S. J., Baylin S. B. Distinct patterns of inactivation of p15INK4B and p16INK4A characterize the major types of hematological malignancies. Cancer Res., 57: 837-841, 1997.[Abstract/Free Full Text]
  19. Pollock P. M., Pearson J. V., Hayward N. K. Compilation of somatic mutations of the CDKN2 gene in human cancers: non-random distribution of base substitutions. Genes Chromosomes Cancer, 15: 77-88, 1996.[Medline]
  20. Xing E. P., Nie Y., Wang L. D., Yang G. Y., Yang C. S. Aberrant methylation of p16INK4a and deletion of p15INK4b are frequent events in human esophageal cancer in Linxian, China. Carcinogenesis, 20: 77-84, 1999.[Abstract/Free Full Text]
  21. Gardie B., Cayuela J. M., Martini S., Sigaux F. Genomic alterations of the p19ARF encoding exons in T-cell acute lymphoblastic leukemia. Blood, 91: 1016-1020, 1998.[Abstract/Free Full Text]
  22. Fukunaga K. Statistical Pattern Recognition 2 Academic Press, Inc. San Diego, CA 1990.
  23. Kamijo T., Zindy F., Roussel M. F., Quelle D. E., Downing J. R., Ashmun R. A., Grosveld G., Sherr C. J. Tumor suppression at the mouse INK4a locus mediated by the alternative reading frame product p19ARF. Cell, 91: 649-659, 1997.[Medline]
  24. Kamb A. Cyclin-dependent kinase inhibitors and human cancer. Curr. Top. Microbiol. Immunol., 227: 139-148, 1998.[Medline]
  25. Quelle D. E., Cheng M., Ashmun R. A., Sherr C. J. Cancer-associated mutations at the INK4a locus cancel cell cycle arrest by p16INK4a but not by the alternative reading frame protein p19ARF. Proc. Natl. Acad. Sci. USA, 94: 669-673, 1997.[Abstract/Free Full Text]
  26. Larsen C. J. p16INK4a: a gene with a dual capacity to encode unrelated proteins that inhibit cell cycle progression. Oncogene, 12: 2041-2044, 1996.[Medline]
  27. Serrano M., Lee H., Chin L., Cordon-Cardo C., Beach D., DePinho R. A. Role of the INK4a locus in tumor suppression and cell mortality. Cell, 85: 27-37, 1996.[Medline]
  28. Kamijo T., Weber J. D., Zambetti G., Zindy F., Roussel M. F., Sherr C. J. Functional and physical interactions of the ARF tumor suppressor with p53 and Mdm2. Proc. Natl. Acad. Sci. USA, 95: 8292-8297, 1998.[Abstract/Free Full Text]
  29. Sangfelt O., Erickson S., Einhorn S., Grander D. Induction of Cip/Kip and Ink4 cyclin dependent kinase inhibitors by interferon-{alpha} in hematopoietic cell lines. Oncogene, 14: 415-423, 1997.[Medline]
  30. Iravani M., Dhat R., Price C. M. Methylation of the multi tumor suppressor gene-2 (MTS2, CDKN1, p15INK4B) in childhood acute lymphoblastic leukemia. Oncogene, 15: 2609-2614, 1997.[Medline]
  31. Igaki H., Sasaki H., Tachimori Y., Kato H., Watanabe H., Kimura T., Harada Y., Sugimura T., Terada M. Mutation frequency of the p16/CDKN2 gene in primary cancers in the upper digestive tract. Cancer Res., 55: 3421-3423, 1995.[Abstract/Free Full Text]
  32. Maesawa C., Tamura G., Nishizuka S., Ogasawara S., Ishida K., Terashima M., Sakata K., Sato N., Saito K., Satodate R. 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, 1996.[Abstract/Free Full Text]
  33. Chan W. C., Tang C. M., Lau K. W., Lung M. L. p16 tumor suppressor gene mutations in Chinese esophageal carcinomas in Hong Kong. Cancer Lett., 115: 201-206, 1997.[Medline]
  34. Hayashi K., Metzger R., Salonga D., Danenberg K., Leichman L. P., Fink U., Sendler A., Kelsen D., Schwartz G. K., Groshen S., Lenz H. J., Danenberg P. V. High frequency of simultaneous loss of p16 and p16{beta} gene expression in squamous cell carcinoma of the esophagus but not in adenocarcinoma of the esophagus or stomach. Oncogene, 15: 1481-1488, 1997.[Medline]
  35. Laird P. W., Jaenisch R. The role of DNA methylation in cancer genetic and epigenetics. Annu. Rev. Genet., 30: 441-464, 1996.[Medline]
  36. Gonzalgo M. L., Hayashida T., Bender C. M., Pao M. M., Tsai Y. C., Gonzales F. A., Nguyen H. D., Nguyen T. T., Jones P. A. The role of DNA methylation in expression of the p19/p16 locus in human bladder cancer cell lines. Cancer Res., 58: 1245-1252, 1998.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 aacredbookHome page
I. Irminger-Finger
Cellular and Molecular Crossroads of Cancer and Aging
Am. Assoc. Cancer Res. Educ. Book, April 12, 2008; 2008(1): 185 - 189.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Epidemiol. Biomarkers Prev.Home page
R. Sailasree, A. Abhilash, K.M. Sathyan, K.R. Nalinakumari, S. Thomas, and S. Kannan
Differential Roles of p16INK4A and p14ARF Genes in Prognosis of Oral Carcinoma
Cancer Epidemiol. Biomarkers Prev., February 1, 2008; 17(2): 414 - 420.
[Abstract] [Full Text] [PDF]

Home page
 CarcinogenesisHome page
W. K. Kwok, M.-T. Ling, H. F. Yuen, Y.-C. Wong, and X. Wang
Role of p14ARF in TWIST-mediated senescence in prostate epithelial cells
Carcinogenesis, December 1, 2007; 28(12): 2467 - 2475.
[Abstract] [Full Text] [PDF]

Home page
 Arch Otolaryngol Head Neck SurgHome page
J. K. Stephen, L. E. Vaught, K. M. Chen, V. Shah, V. G. Schweitzer, G. Gardner, M. S. Benninger, and M. J. Worsham
An Epigenetically Derived Monoclonal Origin for Recurrent Respiratory Papillomatosis
Arch Otolaryngol Head Neck Surg, July 1, 2007; 133(7): 684 - 692.
[Abstract] [Full Text] [PDF]

Home page
 Hum Exp ToxicolHome page
T. de Almeida Simao, G. L. De Bonis Almeida Simoes, F. S. Ribeiro, D. A. de Paula Cidade, N. A. Andreollo, L. R. Lopes, J. M. B. Macedo, R. Acatauassu, A. M. R. Teixeira, I. Felzenszwalb, et al.
Lower expression of p14ARF and p16INK4a correlates with higher DNMT3B expression in human oesophageal squamous cell carcinomas
Human and Experimental Toxicology, September 1, 2006; 25(9): 515 - 522.
[Abstract] [PDF]

Home page
 Clin. Cancer Res.Home page
M. Guo, J. Ren, M. G. House, Y. Qi, M. V. Brock, and J. G. Herman
Accumulation of Promoter Methylation Suggests Epigenetic Progression in Squamous Cell Carcinoma of the Esophagus
Clin. Cancer Res., August 1, 2006; 12(15): 4515 - 4522.
[Abstract] [Full Text] [PDF]

Home page
 Arch Otolaryngol Head Neck SurgHome page
M. J. Worsham, K. M. Chen, N. Tiwari, G. Pals, J. P. Schouten, S. Sethi, and M. S. Benninger
Fine-mapping loss of gene architecture at the CDKN2B (p15INK4b), CDKN2A (p14ARF, p16INK4a), and MTAP genes in head and neck squamous cell carcinoma.
Arch Otolaryngol Head Neck Surg, April 1, 2006; 132(4): 409 - 415.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Res.Home page
J. Carr, E. Bell, A. D.J. Pearson, U. R. Kees, H. Beris, J. Lunec, and D. A. Tweddle
Increased Frequency of Aberrations in the p53/MDM2/p14ARF Pathway in Neuroblastoma Cell Lines Established at Relapse
Cancer Res., February 15, 2006; 66(4): 2138 - 2145.
[Abstract] [Full Text] [PDF]

Home page
 J. Mol. Diagn.Home page
W.-C. Xue, K. Y.K. Chan, H.-C. Feng, P.-M. Chiu, H. Y.S. Ngan, S.-W. Tsao, and A. N.Y. Cheung
Promoter Hypermethylation of Multiple Genes in Hydatidiform Mole and Choriocarcinoma
J. Mol. Diagn., November 1, 2004; 6(4): 326 - 334.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Cancer Res.Home page
P. Granelli, M. Cattaneo, S. Ferrero, L. Bottiglieri, S. Bosari, G. Fichera, and I. Biunno
SEL1L and Squamous Cell Carcinoma of the Esophagus
Clin. Cancer Res., September 1, 2004; 10(17): 5857 - 5861.
[Abstract] [Full Text]