I53
Loss of Heterozygosity Analysis on the Long Arm of Chromosome 7 in Gastric Carcinoma
Jintian Li Shijuan Mai Bingjian Feng Qisheng Feng Lixi Huang Xingjuan Yu Zhizhong Pan Youqing Zhan Jianchuan Xia
OBJECTIVE
tr: order to det:ne the common deleted region related to
primary gastric carcinomas he,.terozygos,t (LOH)
in Chinese.
c,'r~ hi;man
~!',e frequency of toss of
chromosome
7q and
its chrficai
sigr, fic;-~r~c:e, were investigated. METHODS
A set of 9 microsateliite markers or~ 7q with an average
genetic distance of !0cM we,re usc~d to identify LOH by mutti--PCR amplification of malched lumor and r:on-tumor DNAs from 70 patien{s with primary gastric carcinoma, The PCR products were separated by eieclrophoresis in poiyacrylamide gels and analysed for LOH by using Genescan and Genotyper software. RESULTS The lotal irequency of LOH at any locus on 7q was 34,3% (247 70) in the tumors, Compared to non--tumor DNA, LOH at D7S486 and D7S798 loci were higher, 24.0%(12/50) and 19.2%(5/26), respectively
Research Department, Cancer Centre, Sun Yat-sen University, Guangzhou 510060,China,
The total frequency of LOH on 7q was markedly higher with an increase in the ciink::al stage (P<0,05). The frequency d L.OH at D7S486 in cases witt~ Iyrnph node m,etastasis was si.gnificanty higher than in cases without lyrnph node metastasis, P=0 015, CONCLUSION
Correspondence to: Jintian Li Telephone: 86-20-8734-3183 E-mail : lijintian006@hotm all.corn
The higher incider~ce of LOH at D7S486 and D7S798 in
primary gastric carcinoma compared to r~ormal tissue suggests that the potentia] tumor suppressor genes (TSGs) involved in the progression of gastric carcinoma mJghl be nearby these 2 loci,
KEYWORDS: gastric carcinomas, chromosome7q, lossof heterozygosity (LOH), tumorsuppressorgene(TSG). This work was supported by the National Natureal Science Foundation of China. (No,30070840)
astdc carcinoma is one of the most fi'equent malignances of the
G digestive system. Over the past 20 years, the incide~ce or" gastric carcinoma has shown a rising tendency and its mortalil:y is the first among malignances in China. Iu The development and progression of cancer are the result of abnom~al alterations in multigenes with multisteps, including activation of oncogenes and disfunction of tumor suppressor genes (TSG). At the present time the genetic alterations playing an important role in the development and progression of gastric carcinoma are still uncertain, Previous genetic studies showed that there were higher frequencies of gain on l q, 2q, 6p, 7q, 8q, 17p, 17q, 20q and of loss on lp, lq, 2p, 3p, 4q, 5q, 6p,
Received April 22, 2004; accepted June 25, 2004, Chinese Journal of Clinical Oncolgy Email:
[email protected] Tel(Fax):86-22-2352-2919
7q, 9p, l lp, 1 lq, 12q, 15q, 17p, 17q, 18q, 19p, 21q. r:st Ii has been suggested that alterations of the above chromosomes might be
154 Chinese Journal of Clinical Oncology 2004/Volume 1/Number 3
involved in the development and progression of
Stage IV, 26 cases.
gastric carcinoma. Using the direct analytical method of fluorescence in situ hybridization
IFISH) on
DNA extraction
decolored G-band chromosomes, fragmlent losses on
Using a standard phenol-chloroforna method, total
7q, 3p, lp and 17p were detected. Losses were more
DNA was extracted from matched gastric carcinoma
flequent on 7q (21/26 cases) in prima~2r gastric
and normal mucosa tissues. The purity and concentra-
carcinoma, It is suggested that TSG related to gastric
tion of total DNA were determined by agarose dec-
carcinoma
trophoresis and ultraviolet spectrophotometry.
might
exist
on
the
long
arm
of
chromosome 7. So in this study, in order to further carcinoma-related TSG, loss of heterzygosity (LOH)
Selection of microsatellite markers and mulfi-PCR amplification
of microsatellite markers on 7q were examined in
A
primary gastric carcinoma samples matched tumor
chromosome 7q were selected from Genethon Human
and non-tumor tissues.
Genetic Linkage Map (including D7S515, D7S486~
narrow
the
region
on
7q
containing
gastric
set
D7S530,
MATERIALS AND METHODS
of
9
microsatellite
D7S640,
D7S684,
markers
D7S661,
covering
D7S636,
D7S798 and D7S2465), their heterozygosity being above 75% and their average genetic distance being
General information
10cM. The primers of the rnicrosatellite markers were and
labeled respectively by fluorescent dye FAM, NED or
corresponding normal mucosa tissues were obtained
HEX (PE Co,). According to the molecular weight of
from surgically resected samples in the Cancer
the
Hospital of Sun Yat-sen University from December
fluorescent dyes, these primers were divided into two
2000 to January 2002. The patients had not received
groups (4-5 pairs/group) for PCR. Multi-PCR was
any radio- or chemo-therapy before operation. All
performed in a 51,L1reactive mixture (10x buffer 0.5
tissues were selected immediately after the samples were removed fiom the patient. A portion of the
b~l, 10 mmol/L dNTP 0.2 txl, 25 mmol/L MgCI2 0,3 Ixl, 10 pmol/L primer 0.05 i,l/each, hot star TaqTM
tissues were frozen at - 80C fbr DNA extraction and
DNA polymerase 0,2 U and template DNA 45 rig) by
the other portion routinely treated and sectioned fbr
GeneAmp~PCR System 9700 (PE Co.). The reactive
tt&E staining.
conditions were as fbllows: beginning at 95~ lbr 12
Seventy primary gastric
carcinoma tissues
PCR products
and the
color
of labeling
The 70 patients included 52 males and 18 females
rain tbr pre-denaturing; followed by I0 reactive
with an average age of 52,9 years (range 19-76).
cycles at 94"C 30 s for denaturing. 63~ 60 s for annealing, 72'C 90 s for extending, reducing 0.5~ at
According to the WttO (/990) Histological Isq~ing of Gastric carcinoma, the 70 patients were divided as
each cycle tbr annealing;lbllowed by 26 reactive
follows: papillary adenocarcinoma, 1 case; tubular
cycles at 94~C 30 s for denaturing, 58~
adenocarcinoma, 15 cases; poorly diftbrentiatiated
annealing, 72~C 90 s for extending:final step at 72~
adenocarcinoma, 25cases; mucinous adenocarcinoma,
15 rain for extending,
2 cases;
signet-ring cell carcinoma,
60 s tbr
20 cases;
adenosquamous carcinoma I case. According to the
Genome scanning and poiymorphism analysis of microsatellite marker locus
UICC (1997) New International United TNM Stage of Gastric carcinoma, the 70 patients were divided
A volume of 0.51xl of the PCR products was dilduted in 1.01-tl eletrophoretic loading buffer (containing
into Stage I , 7cases (Stage I A, 3 cases and Stage 1
70% tbrmamide, 3.75 mmol/L EDTA, 7.5 mg/ml
B, 4 cases); Stage i I , 13 cases; Stage III, 24 cases
blue dextran, 1.2 nmol/L Genescan 350 ) and 1.0~xl
(Stage Ill A, 14 cases and Stage HI B, I0 cases) and
of the mixture loaded onto urea-containing denaturing
undifferentiated
adenocarcinoma,
6
cases;
and
Chromosome 7 in gastric carcinomallintian t i e t at 155
Sequagel TM polyacrylamide gel in the presence of
Among Uaem, the frequency of [,OH at D7S486 was
standard markers. The gel image was generated
the highest, 24.0% (12/50 infcwmative cases) and
automatically during eletrophoresis by ABI PRISM
D7S798 was the second, 19.2% (5/26 i~fformafive
377 Sequencer (PE Co.) and was analysed by ABI
cases). I..OH was deteted in 24 of 70 patients, the
sol:tware of Genoscan Version .~.0 -rod Geno vper
total frequency of LOH being 34.3%. In these 24
Version
as
patients, a LOI:t at a single locus was seen in 15 cases
(i.e. infbrmative) when its genotype
(62.5%) and a LOtt at more than I locus was seen in
showed two different peaks of PCR product bands
9 cases (37.5%), and LOI-I at more than 3 loci was
and as homozygosity (i.e., non-infbrmative) when its
seen in 5 cases (25.0%).
2.0.
heterozygosity
The
Ioetls
was
determined
genotype showed a single peak (single allele). Allelic when the ratio of 2 peaks in the tumor DNA (T1/T2)
The relationship between clinicopathological data
was
The
loss tit a heteorzygous locus was determined as LOH obviously
different
fiom
that
of
the
association
of
LOH
LOH
and
frequency
the with
corresponding non-tumor DNA (NI/N2), i.e., T1 ;T2 /
clinicopathological dala is summarized in 'Fable 2.
N I:N2 being higher than 1.5 or lower than 0.67.
The total frequency of LOtt on 7q was signtificantly correlated with the clinical stage, i.e. the higher the
Statistical analysis
clinical stage, the greater the ti'equency of LOt:I, P:-::0.
Comparison of the ratio among groups or between
046. A LOH on 7q was detected in 13 of 26 patients
groups was carried out by usi!lg the Student's test or
with Stage , IV'(50/<,) -~ including 9 microsatellite marker
Fisher's exact test. P<0.05 was considered to be
loci. Among these 13 cases, 6 cases (46.1%) had a
statistically signiticanl.
LOH at the D7S486 locus. Five cases with LOH at more than ~ loci were in Staoe 1\7 In addition, the
RESULTS
frequency of LOH at the D7S486
locus was
signlificantly correlated with lymph node metastasis,
Distribution and frequency of LOH
The frequency of LOH at D7S486 in patients with
Multi-PCR amplification of 9 microsatellite marker
lymph node metastasis was signlificantly higher than
loci on 7q were perfomaed with genomic DNAs
those without lymph node metastasis, P=O.015.
extracted fi'om 70 primary gastric carcinoma and
However, lhe frequency of LOtf was not significantly
matched non-tumor tissues tbllowed b y analysis of
different among various histological t3q~es of gastric
the
9
carcinoma, But it is worth noticing that 4 of 6 patients
microsatellite marker loci could be detected in the 70
with mldifferentiated adenocarcinoma were detected as
primary gastric carcinomas as summarized in Tablel.
having L()H.
microsatellite
marker
loci.
LOH of all
Table 1. LOH d[stributiq)n a . d frequency of 9 mierosatellite m a r k e r loci in 70 p r i m a r y g a s t r k carcinomas
Micm,'s.aieltile marker loct, s
Chiomosome locatio~
LOH cases
Informative cases
I.X)tt fi-equency (%)
D7S515
[email protected]
6
53
11.3
l)7S486
7q3 l. I
12
50
2-1.0
D7S530
7q32, I
4
50
8.0
D7SM0
7q33
3
63
4.8
D7S684
7@4
4
58
6.9
D7S661
7q35
3
27
t l,l
D7S636
[email protected]
3
(~0
5.0
D7S79R
7q3,~ 2
5
2,q
19.2
D7S2465
7q36.3
5
52
!).6
156 ChineseJournal of Clinical Oncology 2004/Volume l/Number 3
Table 2. Relationship between LOH frequency and clinical-pathological data Clinical and pathological information
LOH frequency on 7q cases(%)
X~
P
LOH frequency at D7S486 cases (%)
X:
P
Male
t7/52(32.7)
0.22
0.636
7/35(20,0)
1.00
0.3/7
Female
7/18(38.91
0.25
0.618
2.63
0.268
0.68
0,4tl
-
0.015
-
0.379
0.12
0.989
Sex
5/15(33.31
Age >53
14/37(37.8)
~53
10t33(30.3)
0.43
0.510
7/26(26.91 5/24t20.8 )
TNM stage I-I[
3/2(1(15,01
lII
8/24(33,3)
6,16
0.046
5/18(27.81
1/13(7,7)
IV
t 3/26(50)
6/19(31.6)
T stage TI-2
3/13(23.ll
T3-4
2t/57(36.81
0.88
0.349
1/8(12.5) l 1/42t26.21
N stage NO
4/19(21.11
N 1-3
20/51 (39.2)
2.(X)
r
0/13(0.01 t 2/37(32.4)
M stage M0
20/60(33.31
Ml
4/10(40.0)
0,17
0.683
9/42(21.4 I} 3/8(37.5)
Histo-typing TAC
5/15(33.3)
2.79
0,425
2/9{22,2)
PI)AC
9/25(36.01
6/22127.3)
SllCC
6/19(31.6)
3/12(25.0)
U DA C
4/6(66.7.)
1/4(25.01
TAC: tubular adenocarcinoma, PDAC: poorly dilferentiatiated adenocarcinoma, SRCC: signet-ring cell carcinoma, UDAC: undifferentiated adenocarcinoma.
DISCUSSION
the possibility" of narrowing the located region of
Previous studies have detected a high t?equency of
-related TSGs at above two loci ,especially at D7S486.
fragment loss on 7q and seeking gastric carcinoma fragment loss on 7q in primary gastric carcinoma by
The development and progression of cancer are the
using direct FISH analysis on decolored G-band
result of abnormal alternations in multigenes with
chromosomes, t61 In our study, genome scanning and
muir/steps. Different genetic alternations may occur in
polymorphic analysis of microsatellite marker locus
the same cancer. The same genetic alternations may
were Used to examine matched tumor and non-tumor
play a role in different cancers, but their significance
samples from 70 primary gastric carcinoma sanlples.
might be different. Previous studies have detected that
The results showed that LOH of all 9 microsatellite
gains or losses on many chromosomes are involved in
marker loci on 7q could be detected in 70 primary
the process of gastric carcinoma development and
gastric carcinomas. So, our results strengthened the
progression, fz~ Chromosome 7q, especially the 7q31
previous findings. In addition, the results also showed
region, has become a point of intense research interest
that the frequency of LOH at D7S486 and D7S798 were
related to various cancers, including cancer of the
higher than the other 7 microsatallite loci, suggesting
head-neck, colon, prostate, breast, kidney, pancreas, stomach, and so on. i;-~IAs to research on EOH on 7q in
Chromosome 7 in gastric carcinoma/]intian Li et al. /57 gastric carcinomas, Kuniyasu et al. ~l showed that the
candidate TSGs related to gastric carcinoma. In
total frequency of LOH on 7q was 32% (26/82 cases) in
addition, the frequency o f LOH at the D7S798 locus
98 gastric carcinomas by using a 5 microsatellite
located in 7q36,2 was high, 19,2% (5/26 cases). But
marker loci assay, and that the highest frequency (43%,
the number of the informative cases was insufficient,
13/30 cases) of LOH among these loci was at the D7S95
requiring an increase in the number of cases for fut'ther
locus located in 7q31-3 5.Nishizuka et al.':3t-tshowed that
research,
the total frequency of LOH on. 7q was 34% (18/53 cases) in gastric carcinomas by using a 14 microsatellite
REFERENCES
marker loci assay, and that the frequency o f LOH at the
1
Li LD. Lu FZ, Zhang SW, et ~fl.Analysis of variation trend and short term detection of Chinese malignant tumor m~)rtality during twen.tyyears. Chin .I Oncol. 1997;19:3-0.
2
was 34.3%(24/70 cases) in gastric carcinomas and the
Et-Rrifai W, tfarper JC, Cummings OW, et al. Consistent genetic alleratiot~s in xenografts of proximal stomach and
frequency o f LOH at the D7S486 locus was 24.0%
gastro-esophageal junction adenocarcinomas. Cancer Res.
(12/50 cases). The results in our study conform with
1~X)8;58:34-37.
D7S486 locus located in 7q31was 20%
(6/30 cases).
Our study showed that the total frequency o f LOH on 7q
those o f the 2 studies cited.
3
Regarding the relationship between LOI-{ on 7q and clinical thctors, our results did not con~bnn with other
4
studies. In a previous report, Kuniyasu et al. t~01showed that the frequency o f LOH at the D7S95 locus located in 7q31-35 significantly rose in Stage IV patients. In
5
Oncok 1999:21:345-349. 7
gastric carcinoma and prognosis of patients. Our study
8
chromosome 7q.31.I neighboring the Oncogene. 2000;19:1729-1733.
loci were in stage IV, and that the frequency o f LOH at
metastasis, P-=01015. These results further demonstrate that TSGs involved in progression o f gastric carcinoma might exist in the long arm of chromosome 7, especially at the D7S486 locus. The average genetic distance o f microsatellite
Zenklusen JC, Hodges LC, LaCava.M, el al. I)efinitive fut]ct:ionai[ evidence for a tumor suppressor gene on human
(I3/26 cases) and all 5 cases with LOH at more than 2
signlificantly higher than those without lymph node
Gunduzl M,Ouehida M,Fukushimal K, el al. Allelic loss and reduced expression of the ING3, a candidate tumor suppressor gene at 7q31, in humap head and neck cancers. Oncogene. 2002;21:4462-4470.
showed that the total frequency of LOH on 7q rose with
D7S486 in patients with lymph node metastasis was
Xia JC, Xiao S, Zhang JH, el al. Direct chromosome analysis aad FISH sttidy of primary gastric cancer. Chin J
than that o f non-LOH patients. It was confirmed that
frequency o f LOH in Stage IV patients was up to 50%
(LOH) on chromosome 119 in gastric
402-403, 6
patients with Stage llI-IV gastric carcinoma was lower
the clinical stage o f gastric carcinoma,/'=-0.046. The
Chin J O~col. 1998:20:11.6-118. Wang Q, Cheng H, Wang BC, et al. Analysis ~f h.,ss of carcinoma by gene scanning. Chin J Med Genet. 2(~)1; 18:
the abdominal cavity was higher in LOH patients than
LOH at D7S95 heavily influences the progression of
Wang YX, Zheng E, Yang K, et al. Studies of loss of heterozygosity(LOH) in Chinese human gastric cancer tissues.
helerozygosity
addition, the proportion of cancer which difthsed into in non-LOH patients, and that the survival rate of LOH
Yustein AS, Harper .IC, Petroni GR, et al, Allelotype of gastric adenocarcinoma. Cancer Res. 1~)99;59:1437-1441.
9
Fra7G
site.
Zenklusen JC, Conti CJ, Green ED. Mutatkmal and functional analyses reveal that ST7 is a highly coaserved tumor-suppressor gene on human chromosome 7q31. Nat Genet. 2001:27:392-3q8.
t0 Kuniyasu H. Yasui W, Yokozaki H, et al. Frequent loss of heterozygosity of the long arm of chromosome 7 is closly associated with pwgression of human gastric carcinomas. Int J Cancer. 1094;59:597-600.
markers selected in this study was quite long, reaching
11 Nishizuka S, Tamura G, Terashima M, et al. Commonly
10cM. So, more microsatellite markers nearby the
deleted region on the long arm of chromosome 7 in
D7S486 locus should be selected for further research
differentiated ademx'arcinoma of the stomach. Br J Cancer. 1997;76: ] 567-1571.
on LOH in order to narrow the region containing