/ ANTICANCER RESEARCH 19: 29-34 (1999)
Localization of a Novel Tumor Suppressor Gene Loci on Chromosome 9p21-22 in Oral Cancer 1
1
1
1
HIROSHI NAKANISHI XIAO-LI WANG , FABIANA LICA IMAI , JIRO KAT0 MASASHI SHIIBA 1, 1 TSUNEO MIYA , YUTAKA IMAI2 and HIDEKI TANZAWA 1 1
Department of Oral Surgery, School of Medicine, Chiba University; Department of Oral Surgery, School of Medicine, Dokkyo University, Japan
2
Abstract. Allelic imbalance or loss of heterozygosity (LOH) studies have been u'sed to identify regions on chromosomes that may contain putative tumor suppressor genes. Deletions of chromosome 9 regions have been observed at high frequency in many other types of sporadic tumor, whereas in oral cancer no decisive information about the allelic loss on chromosome 9 has been reported. To provide detailed understanding of the genetic alterations in oral cancer, 24 highly polymorphic markers mapped on chromosome 9 were used to examine 34 cases oforal squamous cell carcinoma (SCC). LOH was detected in 18 (53%) of 34 informative samples at one or more loci examined. On the basis of our results, three commonly deleted regions were identified and a detailed deletion map was constructed. One of the novel regions was on 9p22, where a tumor suppressor gene, interferon a cluster (IFNA) gene, was identified before. Another region was D9SJ57locus at 9p22, telomeric to IFNA Locus and p15!16 genes, and the third was located on 9p21 of the D9S104 locus, centromelic to methylthioadenosine phosphorylase (MTAP) gene and p15!16 genes. Thus, our data suggest that, except for p15!16 and MTAP gene, there were at least two candidate tumor suppressor genes located at chromosome 9p, and that the alteration of these genes is associated with the tumorigenesis oforal sec. Oral squamous cell carcinoma (SCC) is the most common malignant disease of the oral and maxillofacial region, and the 6th most frequent cancer worldwide (1). However, little is compared to known about the molecular basis of oral other human malignancies. Inactivation of tumor suppressor gene (TSG) is considered ·to be associated with carcinogenesis, and alteration in TSGs are widely accepted to
sec
Co"espondence to: Hideki Tanzawa, Department of Oral Surgery, Chiba University School of Medicine, 1-8-1 Inohana, Chuo-ku, Chiba 260-8670, Japan. Fax: +8 1-43-226-2151, E-mail:
[email protected] Key Words: Chromosome 9, oral squamous celi carcinoma (SCC), loss of heterozygosity (LOH), tumor suppressor gene (TSG).
0250-7005/99 $2.00+ .40
be critical events in the multistep process leading to the development of cancer. Many reports have defined tumor suppressor loci on chromosome 9 that encompass several important growth regulatory genes, including the IFNA cluster gene (2) and methylthioadenosine phosphorylase (MTAP) gene on the short arm (3) and nevoid basal cell carcinoma syndrome (NBCCS, golin syndrome) gene (4) on the long arm of the chromosome. These genes have been mapped between markers of D9S171 and IFNA on 9p2122, and between D9S180 and D9S280 on 9q22.3. Additionally, CDKN2 (also called pl61NK4A, MTSJ or CDK41), was a recently identified TSG (5) and mapped on 9p21 (6). Subsequently, the adjacent related gene p15 (CDKN2B!MTS2) was also detected in the downstream (5). Intensive study of the p16 gene and its product has revealed frequent alterations that can inactivate the function of this gene through at least three mechanism: mutation, homozygous deletion, and hyperrnethylation of the promoter region (7, 8). Frequent inactivation of p16 has been observed in lung cancer (9) and melanoma (10), whereas in oral no hypermethylation and a much lower frequency (4%) of mutation was observed in our previous study (11). To determine the potential presence of other TSGs on chromosome 9 in oral an extensive set of highly polymorphic microsateUite markers was used to screen both the long and short arms of chromosome 9 for putative TSG loci. Two candidate regions were identified and a fine deletion map of chromosome 9 harboring putative TSGs was constructed.
sec
sec,
Materials and Methods Patient materials and DNA preparation. Thirty-four primary carcinoma tissues, 5 metastatic tissues and corresponding normal tissues were obtained from 34 patients at the time of surgical resection at Chiba University Hospital. The resected tumor samples were examined microscopically to cbnfirm the presence of tumor, evaluation of tissue morphology and gnide of differentiation. The remaining tissues were carefully trimmed to remove normal tissues and one part of each of these tissues was frozen and stored at for DNA extraction. Genomic DNA was extracted from the frozen tissue samples using Proteinase K (Takara) digestion and phenol extraction according to
-so·c
29
ANTICANCER RESEARCH 19: 29-34 (1999) described previously (11, 12). All cases studied showed at least 70% of tumor cells on morphologic evaluation. Histopathological classification was performed according to the International Classification of Tumors (13). Clinical staging was determined by the UICC TNM staging system (14). Fisher's exact test was used to assess the significance of correlation between LOH and clinical parameters.
Table I. Loss of heterozygosity at 24 microsatellite loci on chromosome 9 in
Analysis of allelic loss. Twenty four microsatellite markers D9S176,
D9S319
D9S180, D9S280, D9S12, D9S287, D9S319, D9S171, D9S162, IFNA, D9S157, D9S104, D9S288, D9S156, D9S43, D9Sl69, D9S54, D9S199, D9S168, D9S200, D9S161, D9S165, D9S\748, D9S1749 and D9S2136 were investigated for microsatellite motifs at chromosome 9. The primer sequences for the chromosome 9 microsatellite markers used in this study were obtained from Research Genetics (Huntsville AL; Kwiatkowski and Diaz, 1993; Genome Data Base) or the Genome Database (Johns Hopkins University, Baltimore, MD). PCR amplification was performed in 201-11 reaction volumes as described previously (15). The forward primers were end-labeled with gamma phosphorus-32-ATP (Amersham, Aylesbury, UK) in the pre-mix step using T4 polynucleotide kinase (Boehringer Mannheim, Indianapolis, IN). The PCR reaction mixture contained 0.2!-lg of sample DNA, 20 pmol of each primer, 10 mM Tris-HCl (pH 8.3), 50mM KCl, 3.0 mM MgClz, 2mM dNTP and 0.5 unit of Taq DNA polymerase (PerkinElmer, Norwalk, CT). Twenty-six to 30 cycles of denaturation at 94 ' C for 1 minute, annealing at 52 ' C-58'C for 1 minute, and extension at 72'C for 1 minute were performed with a DNA Thermal Cycler (PerkinElmer). To detect a homozygous deletions and/or an increase in CO!JY number, we multiplexed two primer sets in one reaction. Primer sets D9S165 and D9S156, D9S161 and D9S199, D9S288 and D9S43 were multiplexed. These sets of primers had identical annealing temperatures and ·the products differed in size by about 30-50bp. If in one sample both sets of primers amplified a product from the normal DNA and only one primer set amplified a p ·Jduct from the corresponding tumor DNA, it was scored as a homozygous deletion. After dilution with an adequate volume of formamide-dye mixture (95 % formamide, 20 mM EDTA, 0.05 % bromophenol blue and 0.05 % xylene cyanol), the PCR products were heat-denatured (95 'C, 10 minute), chilled on ice and electrophoresed on 5% urea-formamide-polyacrylamide gels at SOW for 2 to 3.5 hours, depending on the fragment size. After electrophoresis the gels were vacuum-dried at 80 ' C and exposed to Fuji film for 12 to 48 hours. Assessment of Loss of Heterozygosity (LOH) . LOH for DNA samples was assessed densitometrically, as described by Uzawa et a/ (12). Briefly,
after correction for differences in the amounts of DNA loaded in each lane, the intensities of the signals in tumor DNA were compared to those of the corresponding normal DNA. A reduction in signal intensity of more than 50% was required for LOH. Common deleted regions were defined by considering the loci which most frequently show LOH, together with multiple interstitial deletions. Assessment of Microsatellite Instability (MI). MI was assessed as positive if
there were additional bands in the tumor sample that were not observed in the corresponding normal one or there was a band shift in tumor tissue that contrasted to those of the corresponding normal bands.
Results
24 microsatellite markers on both the long and short arms of chromosome 9 were employer to analyze 34 oral sees with various grades of differentiation and clinical stages. All cases showed heterozygous genotypes in normal tissues at one or more loci on this chromosome. Overall, 53% (18 of 34) of oral sees showed heterozygous loss of one or more 30
oral squamous cell carcinoma.
Locus symbol
Chromosomal location
Number of informative cases(%)
Frequency of LOH: % (LOH/inform ative cases)
p
27 (81.8)
0 (0/27)
D9S199
p23
33 (100 0)
0 (0/33)
D9S288
pter-22
32 (97.0)
D9S168
p23-22
33 (100.0)
D9S156
p23-22
31 (93.9)
3.2 (1/31)
D9S157
p23-22
17 (51.5)
52.9 (9/17)
D9S162
p23-22
24 (72.7)
0 (0/24)
IFNA
p22
32 (97.0)
D9S1749
p21
11 (33.3)
0 (0/11)
D9S2136
p21
22 (66.7)
0 (0/22)
D9S1748
p21
16 (48.5)
6.3 (1!16)
D9Sl71
p21
31 (93.9)
9.7 (3/31)
D9S169
p21
18 (54.5)
5.6 (1/18)
D9S161
p21
33 (100.0)
6.1 (2/33)
D9S104
p21
D9S165
p21-q21
33 (100.0)
3 (1/33)
D9S200
p13
33 (100.0)
12.1 (4/33)
D9S54
pter-p22
26 (78.8)
D9S43
q21-ter
33 (100.0)
6.1 (2/33)
D9S176
q22.3-31
31 (93.9)
6.5 (2/31)
D9S180
q22.3
27 (81.8)
0 (0/27)
D9S280
q22.3-31
20 (60.6)
0 (0/20)
D9S12
q22.3
24 (72.7)
0 (0/24)
D9S287
q22.3-31
25 (75.8)
0 (0/25)
236 (78.8)
3.1 (1132) 3 (1 /33)
40.6 (13/32)
34.6 (9/26)
0 (0/26)
Infomation of chromosomal location was obtained from the Genome Data Base, Baltimore.
microsatellite markers on chromosome 9. Location of markers and frequencies of LOH are summarized in Table · I. The frequency of alterations on the short arm (9p) and long arm (9q) of chromosome 9 were 44% (15 of 34) and 12% (4 of 34), respectively. 9p showed a significantly higher incidence of alterations than that of 9q (p=0.003). The results of LOH analysis in tumor samples showing allelic deletion and/or MI on chromosome 9 are summarized as
Nakanishi et al: A Detailed Deletion Map on Chromosome 9 in Oral Cancer
24 23 22 21 13
12 11
11 12 13 21.1 21.2 21.3 22.1
~~j
31 32 33
34. 1 34 .2 34.3
Case No. 1 2 4 5 6 7 9 10 1314 1518 19 20 23 27 29 31 33 34 095319 D- O D ODDDDDDODDDDDDDD 095 199 - DODDDDDDDDDDDDDCJDDD 095288 • DDCJDDDODDDDDDL:JDDDDD 095 168 • ooOODDDDDDDDDDODDDC 0951 ~6 • DDODDDDDO- DDDDDDDD D o951s7 - D 095162 - 0 0 0 0 -DDDD-D DDDD -DDD tFNA
•• -••••••ooo-• o-1* •••••••••••••oooooool*
0951749 0 - - D O - - - - - - - - - 0 - - -DO 0952 136 - -ODD-D- C- - D D D D 0 - 0 0 0 0951748 - - -DODD- - D O - - - D · Ill - 0 095171 • •ooDDDDDDDDDDDDDODD 095 169 • o - D D - - 0 - D - - O DD - - -DO 095161 DDDDDDDDD - -DDDDDDDDil o951 0 4 • • • • •o o - o • •o - • •o o 095165 DD. DDDDDDDDDDD DDDDDD o9s2oo • o • o o oo o o • D o o o oo oo • o 09554 DDDDD- DO D D -DODD - DOD D 09543 o • o • o oD - ODODDDDDDDDD 095176 • o • DDDDD B DDD- D D I!IIDDD B 095180 DD ~D- DDDDD- DDDDDDDD 095280 - DODD- DODD - DO- DOD - 09512 DODD- DO- - ODD - DDDDDD 09528 7 - - Od::::JDD- - DDDDDDDDD-
o•l*
Figure 1. Schematic representation of deletion map of chromosome 9 in oral SCC. Tumors with LOH at three distinct regions on chromosome 9. The approximate locations of markers used are shown 0 11 the left and tumor numbers are shown above. • LOH;-D heterozygosity; • instability; and- not informative. Asterisks represeflt the commonly deleted regions.
the deletion map in Figure 1. Three common regions of deletion were identified and centered at the loci of D9S157 (52.9%), IFNA (40.6%) and D9S104 (34.6%) on chromosome 9p21-22. Other loci of the chromosome showed significantly lower frequencies (0 to 12%) of allelic losses (p
