вход по аккаунту



код для вставкиСкачать
180 33-37 (1996)
Sir Alastair Currie CRC Laboratories, Molecular Medicine Centre, University of Edinburgh, Western General Hospital,
Crewe Road, Edinburgh EH4 2XU, U.K.; *Department of Pathology, Ninewells Hospital and Medical School,
Dundee DDl 9SY, U.K.
Four genetic poljmorphisms in the A P C and MCC genes at chromosome 5q21 were analysed for loss of heterozygosity &OH) in 97
primary squamous carcinomas and adenocarcinomas of the lung. LOH was identified in at least two polymorphic loci in 41 percent of
informative cases. There was no significant difference in the frequency of LOH between squamous carcinomas and adenocarcinomas.
Within the adenocarcinoma group, however, LOH appeared to be more common in tumours having a bronchial origin (519; 56 per cent)
than in parenchymal adenocarcinoma (6/21; 29 per cent). All 32 tumours showing LOH at one or more polymorphic sites were examined
for mutations in the mutation cluster region (MCR) of A P C by single-strand conformational polymorphism (SSCP) analysis. Mutations
were not detected in any of these cases. We therefore propose that it is likely that a tumour suppressor gene on 5q other than A P C is
involved in the pathogenesis of lung cancer.
cell lung cancer; adenocarcinoma; squamous carcinoma; A PC; MCC
and relatively well-characterized tumour suppressor
gene, which is implicated in hereditary and acquired
Lung cancer is characterized by a greater heterocolorectal cancer and also in some other neopla~ms.~
geneity of morphological appearances than the other
The function of APC is not fully understood. It
common visceral malignancies. Eight histological
associates with plakoglobin and p-catenin and may
classes of primary pulmonary malignancy are described
be involved in the regulation of epithelial cellLCel1
in the World Health Organisation (WHO) classification
adhesion.lO Through binding to these proteins, it may
of lung turnours.* From the therapeutic and prognostic
also influence a Wnt-I-stimulated cell-signalling pathviewpoint, the prime distinction to be made is between
way, controlling the differentiation fate of epithelial
small cell lung cancer (SCLC) and non-small cell lung
cells.11,'2In vitro APC protein also acts as a microtubule
cancer (NSCLC), with the former having a poorer
binding protein and enhances microtubule polymerizclinical outlook. A variety of molecular lesions have
ation.13,14It is unknown if APC plays a role in lung
been described in lung cancer, some of which show
cancer and there is no increased incidence of lung cancer
correlation with histological classification. Thus, ampliin patients with familial adenomatous polyposis (FAP),
fication of c-myc is often associated with SCLC, while
but we have recently shown that the APC protein is
K-ras mutations and increased expression of c-erbB-2
normally expressed in respiratory epithelium.
are seen in NSCLC, particularly a d e n o c a r ~ i n o m a . ~ ~ ~
The aims of the present study were to determine the
Recently there have been a number of reports of
frequency of 5q21 involvement in NSCLC and in pargenetic losses at chromosome 5q in lung carcinoma^.^-*
ticular to determine the relative occurrence of loss in
Most attention has been paid to SCLC, in which such
squamous carcinoma and adenocarcinoma, the most
losses are described in about 80 per cent of cases, but
common forms of NSCLC, as there is emerging evidence
other groups, including our own, have shown losses at
that there may be significant differences in the allelo5q in NSCLC.5.s Consistent loss of genetic material at a
types of these neoplasms.16 These differences may be of
particular chromosomal arm in a neoplasm is evidence
practical significance, since epidemiological investifor the presence of a tumour suppressor gene at that site.
gations have shown that the relative incidences of these
Although other loci on 5q may be implicated, 5q21 is a
tumours are changing with adenocarcinoma becoming
prime candidate region, as it contains the tumour supmore prevalent.17 This shift in the histological profile of
pressor gene APC (adenomatous polyposis coli) and the
lung cancer has led some authors to re-evaluate the
candidate tumour suppressor gene MCC (mutated in
WHO classification, which recognizes several morphocolorectal cancer). APC in particular is an important
logical subtypes of adenocarcinoma, but has not been
shown to be of prognostic or biological significance. It
Addressee for correspondence: C. A. Cooper, Sir Alastair Currie
has been suggested that a more clinically relevant disCRC Laboratories, Molecular Medicine Centre, University of
tinction can be drawn between adenocarcinomas having
Edinburgh, Western General Hospital, Crewe Road, Edinburgh
origin (like squamous carcinoma and SCLC) from large
EH4 2XU, U.K.
CCC 0022-3417/96/090033-05
0 1996 by John Wiley & Sons, Ltd.
Received 24 January 1996
Accepted 29 March 1996
airways and those arising in the distal lung parenchyma,
from bronchiolar and alveolar lining cells.18A comparison between the bronchial and parenchymal groups of
adenocarcinomas was therefore included in the study.
Our strategy was to determine the frequency of LOH
at four polymorphic loci in APC and MCC in a series of
surgically resected NSCLC cases. LOH at the appropriate genetic locus is not, of itself, sufficient evidence to
implicate a particular tumour suppressor gene in the
pathogenesis of a neoplasm. The widely accepted multistage model of carcinogenesis predicts that development
of neoplasia requires the remaining copy of the gene
to be mutated or otherwise ina~tivated.~
Mutations in
the APC gene have been detected in a number of
tumours, including those of the colon, pancreas, and
s t o m a ~ h . The
~ ~ ' majority
~ ~ ~ ~ of mutations have been
found in the first half of the gene's coding sequence,
two-thirds of somatic mutations in colorectal tumours
being restricted to a region called the mutation cluster
region (MCR), between codons 1286 and 1513.21We
therefore decided to use an established single-strand
conformational polymorphism (SSCP) technique to
screen for mutations in the MCR of the remaining APC
allele of those cases showing LOH at 5q21.
Cases and tissue samples
DNA was extracted from paraffin blocks of 97 surgical resections (lobectomy and pneumonectomy) for primary squamous cell carcinoma or adenocarcinoma of
the lung, using standard methods.22 Separate samples
were taken from each case to include tumour and
non-tumour (usually uninvolved lymph node) tissue.
The tumour samples were dissected from 10 pm sections
under histological guidance, to minimize stromal/
inflammatory cell contamination. Cases included were
chosen as follows: 32 consecutive squamous carcinomas,
58 consecutive adenocarcinomas, and 7 additional nonrandomly selected adenocarcinomas, all of bronchial
type. The latter were included because our experience
has shown that bronchial adenocarcinomas are rather
less common than those arising in the pulmonary paren~ h y m aAdenocarcinomas
were allocated, on the basis
of macroscopic and histopathological appearances, into
parenchymal or bronchial subtypes, with an additional
category for neoplasms of uncertain or mixed histogenesis. Parenchymal carcinomas were defined by lack
of a bronchial origin and by a 'lepidic' growth pattern.
Cytologically, they were composed of cells resembling
type 2 pneumocytes, Clara cells ('tongue-shaped' cells),l7
or mucus cells. The bronchial adenocarcinomas showed
obvious origin from and disruption of bronchial mucosa
and tended to have an acinar or papillary pattern.
LOH in APC and MCC genes
LOH for each tumour was determined using polymerase chain reaction (PCR) amplification of intragenic
sequences containing known polymorphic sites within
the APC and MCC genes. Restriction fragment length
Fig. I-Two samples analysed by PCR amplification of the naturallyoccurring length polymorphism at MCC exon 10. In each case, the
'normal' D N A sample (reactive lymph node) is shown in lane N. Both
patients are heterozygous for this locus, indicated by A1 (175) and A2
(161) base-pairs. The corresponding tumour samples (T) show loss of
A2 in patient 1 (LOH). Heterozygosity is retained in patient 2. Lane M
contains molecular weight markers
polymorphisms (RFLPs) in the 3' untranslated region
(UTR)24and exon 11 of the APC genez5and a RFLP in
the 3'UTR of MCQ6 were analysed using enzymes
Sspl, h u l , and M u e l l l , respectively. A fourth polymorphism, a length polymorphism in MCC exon
was also analysed. Conditions and primers have been
previously publishedz8 with the exception of a new
APC 3' UTR primer. Here the upstream primer was
replaced with a new primer (5' GAAGAGACTGCA
ATGTCTAAGAA 3') to produce a smaller 3 18 bp PCR
product, which was easier to amplify from archival
paraffin-embedded tissues. A control DNA fragment
containing the appropriate restriction site was added to
the APC exon 11 (Rsul)and the MCC 3'UTR ( M u e l I I )
digests to ensure complete restriction enzyme cleavage.
The digested PCR products were run on agarose or
polyacrylamide gels and stained with ethidium bromide.
LOH at a given locus was defined by a reduction in the
intensity of an allele band of the tumour in comparison
with its corresponding normal (Fig. 1). Cases polymorphic at two or more loci were subjected to statistical
SSCP analysis of the MCR of the APC
All cases showing LOH at 5q21 were screened for
APC mutations. DNA fragments were amplified by
PCR (Table I). Some of the primers used have previously been published.z9 Cycling conditions consisted
of an initial denaturation step of 5 min at 94"C, followed
by 36 cycles of 30s denaturation at 94"C, 30s at the
appropriate annealing temperature (see Table I), and
60s extension at 72"C, with a final extension step of
10 min at 72°C. One microlitre of denaturing solution
(0.5 M NaOH, 10 mM EDTA) was added to 5 pl of PCR
product. Samples were incubated for 5 min at 50"C, then
3pl of stop solution [95 per cent formamide, 10mM
EDTA (pH 8), 0.01 per cent bromophenol blue, 0.01 per
cent xylene cyanol] was added to each sample following
Table I-Sections
of APC MCR for SSCP analysis
size (bp)
37 17-3963
Primers 5'+3'
29 1
441 14614
incubation. Nine microlitres of sample was loaded onto
a non-denaturing MDE gel (AT Biochem, PA, U.S.A.)
containing 5 per cent glycerol in 1X TBE and run at
25°C at 25 W. Gels were subsequently silver-stained
according to the manufacturer's instructions (Bio-rad
Laboratories Ltd., U.K.). Colorectal adenocarcinoma
cases previously shown to harbour mutations in the
MCR of APC were included as positive controls.
A total of 97 NSCLC cases were studied for LOH
within the A P C and MCC genes. Sixty-four of the 90
consecutive tumours were informative at two or more
loci (41 adenocarcinoma, 23 squamous carcinomas).
LOH was seen in 37 per cent (15/41) of adenocarcinomas
and 48 per cent (1 1/23) of squamous carcinomas, giving
an overall LOH of 41 per cent in our informative
NSCLC cases. Statistical analysis indicated no significant difference in the frequency of LOH between
squamous carcinomas and adenocarcinomas h2=0.772,
df= 1; P<0.2).
Within the adenocarcinoma category, which included
the additional non-randomly selected bronchial adenocarcinomas, there appeared to be a difference between
tumours of bronchial and parenchymal origin, with 5/9
(56 per cent) informative adenocarcinomas arising in the
bronchus showing LOH, while parenchymal cancers
showed LOH in 6/21 cases (29 per cent). This difference
falls short of significance at the 5 per cent level. Sixteen
adenocarcinomas could not be clearly described as arising in the bronchus or the parenchyma. LOH was seen
in eight (50 per cent) of these turnours.
There was no relationship between LOH and tumour
stage. In all cases that were informative for polymorphisms in both genes analysed, loss or retention was
always seen in both APC and MCC; there were no cases
showing discordance between the two genes.
Mutation analysis of the APC gene by SSCP
Thirty-two lung cancer cases were screened for
mutations in the MCR of APC by SSCP. This group
consisted of 21 adenocarcinomas and 11 squamous
carcinomas. The adenocarcinoma group included five
non-randomly selected cases of bronchial origin and one
case of parenchymal origin which showed LOH at one
or more polymorphic loci. In no case was there a
difference in the pattern of band mobility between
control and tumour DNA, providing strong evidence
against the presence of APC mutation in any of these
NSCLC cases.
The long arm of chromosome 5 has recently been the
focus of much attention in tumour biology. Particular
attention has been paid to the 5q21 region, the locus of
APC,the gene responsible for FAP.25 LOH at 5q21 has
been described in a variety of neoplasms including
carcinomas of the colorectum, stomach, pancreas,
oesophagus, and kidney. 19,28,3@32 In the present study,
we have shown that LOH at 5q21 is also a feature of
some 41 per cent of randomly selected NSCLC cases, a
higher percentage than in a recently published study,*
and that NSCLC exhibiting LOH is not associated with
mutation of the recognized MCR of APC.
The APC gene has previously been shown to exhibit
LOH in various tumour types and a significant mutation
rate in the MCR of this gene has been described
in carcinomas of the colorectum, pancreas, and
~ t o m a c h . ~ J 9One
~ 2 ~previous study has looked at a very
limited series of lung cancers, finding no evidence of
mutation in seven neoplasms showing genetic loss at this
10cus.'~Our 32 cases showing LOH were screened using
SSCP and no mutations were detected in the MCR. To
compute the probability that such a result might have
been obtained by chance, we used the following logic. In
sporadic colorectal cancer, more than 60 per cent of
mutations occur in the MCR, but analysis of these
colorectal tumours using the SSCP technique has
detected mutations in only 50 per cent of cases.21
Therefore it appears that SSCP is detecting approximately 80 per cent of mutations. If mutations in the
MCR of the APC gene were detected as frequently in
NSCLC as in colorectal tumours (50 per cent), then the
probability of obtaining a negative result in our SSCP
analysis could be calculated by 0.532=2.33- lo. Similarly, there is 95 per cent probability that the prevalence
of detectable mutations is less than 9 per cent. This is
calculated by taking the statistical confidence limit of
P<0.05 and solving for x in the equation (1 - x ) ~ ~ = O . O ~ ;
x=0.09). This is strong evidence against the involvement
of the gene in pulmonary carcinogenesis, a finding which
is consistent with tumour epidemiology, since there is no
increase in incidence of lung cancer in FAP patients.
The MCC gene at 5q21 has been described as a
potentially important tumour suppressor gene. MCC
was originally identified in colorectal cancer, but more
recent mutation analysis by SSCP has failed to show
mutations in a series of 80 carcinomas of the large
We know of no reports of MCC mutation
outside the colorectum and it therefore seems increasingly likely that the status of MCC as a tumour suppressor gene is in some doubt. Further investigation of this
association will be necessary, as is a detailed mapping of
losses at 5q, to localize any other possible tumour
suppressor genes. Interestingly, a recent study has shown
that LOH at a novel locus (del 27) lying centromeric to
5q21 is seen in a high proportion of lung cancers.33
The few previous reports of genetic losses at 5q21 in
lung cancer have largely concentrated on SCLC.6.7 A
recently published study showed LOH in 29 per cent of
informative NSCLC cases with a marginal but significant difference in the frequency of LOH between
squamous carcinoma and adenocarcinoma, the former
being the more frequent.* Our larger study suggests a
higher incidence of LOH in NSCLC (41 per cent) but no
difference in LOH between squamous carcinoma and
Recently there has been accumulating evidence of a
shift in the histological pattern of pulmonary neoplasia,
with a relative increase in the proportion of adenocarcinomas, particularly when compared with squamous
carcinomas and SCLC9 The recent rise in the incidence
of adenocarcinomas has been noted particularly to
involve those arising in the periphery of the lung.34
Morphological and immunohistochemical studies have
shown that there may be biological differences between
adenocarcinomas of bronchial compared with those of
parenchymal origin. The latter arise on a background
of alveolar atypical hyperplasia, with demonstrable
increase in expression of the products of the p53 and
c-erbB-2 genes.35It is also evident that adenocarcinomas
of bronchial origin have a distinctly poorer clinical
prognosis.I8 Our demonstration that LOH at 5q21
appears to be associated more with bronchial than
parenchymal adenocarcinomas is intriguing and may, if
confirmed in a larger series, provide further biological
support for the proposed subclassification of adenocarcinomas. It may be that LOH at 5q21 is associated with
the common bronchial neoplasms of all types (squamous
carcinoma, adenocarcinoma, and SCLC) and is of lesser
importance in the increasingly frequent parenchymal
We conclude that LOH at 5q21 is more common
in SCLC than in NSCLC, with no difference in the
frequency of LOH between adenocarcinomas and
squamous carcinomas. LOH was detected more frequently in adenocarcinomas of parenchymal rather than
bronchial origin, although these results do not reach
statistical significance. Finally, it seems unlikely that
APC plays an important role in NSCLC.
This study was supported by a grant from the Scottish
Office Home and Health Department.
1. Dunnill MS, Gatter KC. Cellular heterogeneity in lung cancer. Hi.xtoputhology 1986; 1 0 461-475.
2. World Health Organisation. Histological Typing of Lung Tumours. 2nd
edn. Geneva: WHO, 1981.
3. Wong AJ, Ruppert JM, Eggleton J, Hamilton SR, Baylin SB, Vogelstein B.
Gene amplification of c-myc and n-myc in small cell carcinoina of the lung.
Science 1986; 23: 461464.
4. Tateishi M, Ishida T, Mttsudomi T, Kaneko S. Sugimachi K. Prognostic
value of c-erbB-2 protein expression in human lung adenocarcinoma and
squamous cell carcinoma. Eur J Cuncer 1991; 27: 1372-1375.
5. Ashton-Rickardt PG, Wyllie AH, Bird CC, el al. MCC, a candidate familial
polyposis gene in 5q21, shows frequent allele loss in colorectal and lung
cancer. Oncogene 1991; 6 1881-1886.
6 . Miura I, Graziano SL, Cheng JQ. Doyle LA, Testa JR. Chromosome
alterations in human small cell lung cancer: frequent involvement of 5q.
Cuncer Res 1991; 52: 1322-1328.
7. D’Amico D, Carbone DP, Johnson BE, Meltzer SJ, Minna JD. Polymorphic sites within the MCC and APC loci reveal very frequent loss of
heterozygozity in human small cell lung cancer. Cancer Res 1992; 5 2
1 996- 1999.
8. Fong KM, Zimmerman PV, Smith PJ. Tumor progression and loss of
heterozygosity at 5q and 18q in non-small cell lung cancer. Cuncer Res 1995;
5 5 220-223.
9. Bishop DT, Hall NR. The genetics of colorectal cancer. Eur J Cunwr 1994;
3 0 19461956.
10. Munemitsu S, Albert I, Souza B, Rubinfeld B, Polakis P. Regulation of
intracellular p-catenin levels by the adenomatous polyposis coli (APC)
tumor-suppressor protein. Proc Ntrtl Acud Sci USA 1995: 9 2 30463050.
11. Hinck L, Nelson W, Papkoff J . Wnt-/ modulates celkell adhesion in
mammalian cells stabilising /I-catenin binding to the cell adhesion protein
cadherin. J Celi Biol 1994; 124 729-741.
12. Funayama N, Fagotto F, McCrea P, Gumbiner B. Embryonic axis induction by the armidillo repeat domain of 8-catenin: evidence for intracellular
signalling. J CeN Biol 1995; 128 959%968.
13. Munemitsu S, Souza B, Miiller 0, Albert I, Rubinfeld B, Polakis P. Thc
APC gene product associates with microtubules in v i i w and promotes their
assembly in vitro. Cancer Res 1994; 54: 3676368 I .
14. Smith KJ, Levy DB, Maupin P, Pollard TD, Vogelstein B, Kinzler KW.
Wild-type but not mutant APC associates with the microtubule cytoskeleton. Cancer Res 1994; 5 4 3672-3675.
15. Midgley CA, White S, Howitt R, et a/.APC’ expression in normal human
tissues. (Submitted).
16. Tsuchiya E, Nakamura Y, Weng SY, et al. Allelotype of non-small lung
carcinoma4omparison between loss of heterozygosity in squamous cell
carcinoma and adenocarcinoma. Cuncer Res 1992; 5 2 2478-248 1
17. Gilliland FD, Samet JM. Lung cancer. In: Doll R, Fraumeni JF, Muir CF,
eds. Trends in Cancer Incidence and Mortality. New York: Cold Spring
Harbor Laboratory Press, 1994; 175-195.
18. Edwards CW. Pulmonary adenocarcinoma: a review of 106 cases and
proposed new classification. J Clin Puthol 1987; 4 0 125- 135.
19. Horii A, Nakatsuru S, Miyoshi Y, e t a ) . Frequent somatic mutations o f the
APC gene in human pancreatic cancer. Caricer Re,y 1992; 52: 66966698.
20. Nakatsuru S, Yanagisawa A, lchii S, et ul. Somatic mutations of the APC
gene in gastric cancer: frequent mutations in very well differentiated
adenocarcinoma and signet-ring cell carcinoma. Hum Mol Genet 1992; 1:
21. Miyoshi Y, Nagase H, Ando H, et a/. Somatic mutations of the A PC gene
in colorectal tumours: mutation cluster region in the APC gene. Hum Mol
Genet 1992; 4 229-233.
22. Levi S, Urbono-Ispizua A, Gill R, et a/. Multiple K-rus codon 12 mutations
in cholangiocarcinomas demonstrated with a sensitive polymerase chain
reaction technique. Cancer Res 1991; 51: 3497-3502.
23. Carey FA, Wallace WAH, Fergusson RJ, Kerr KM, Lamb D. Alveolar
atypical hyperplasia in association with primary pulmonary adenocarcinoma: a clinico-pathological study of 10 cases. Thorux 1992; 4 7 1041-1043.
24. Heighway J, Hoban PR, Wyllie AH. Sspf polymorphism in sequence
encoding 3’ untranslated region of the APC gene. Nutleic Acids Re.s 1991;
1 9 6966.
25. Groden J, Thilveris A, Samowitz W, et ul. Identification and characterization of the familial adenomatous polyposis coli gene. Cell 1991; 66: 589-600.
26. Heighway J. Hoban PR, Wyllie AH. Mae111 polymorphism in sequence
encoding 3’ untranslated region of the MCC gene. Nucleic Acids Res 1992;
20: 1160.
27. Miyoshi Y, Nishisho I, Miki Y, et al. Insertionldeletion polymorphism and
other restriction fragment length polymorphisms in the MCC gene. Jpn J
Cancer Res 1992; 8 3 10-14.
28. Curtis LJ,Bubb VJ, Gledhill S, Morris RG, Bird CC, Wyllie AH. Loss of
heterozygosity of MCCis not associated with mutation of the retained alIele
in sporadic colorectal cancer. Hum Mol Genet 1994; 3 443446.
29. Curtis L, Wyllie AH, Shaw J, et ul. Evidence against involvement of APC
mutations in papillary thyroid carcinoma. Eur J Cancer 1994; 7: 984987.
30. Sano T, Tsujino T, Yoshida K, et al. Frequent loss of heterozygosity on
chromosome lq, 5q and 17p in human gastric carcinomas. Cancer Res 1991;
51: 292622931.
31. Boynton RF, Blount PL, Yin J, et al. Loss of heterozygosity involving the
A P C and MCC genetic loci occurs in the majority of human esophageal
cancers. Proc Nut1 Acad Sci USA 1992; 8 9 3385-3388.
32. Morita R, Saito S, Ishikawa J, Ogawa 0, Yoshida 0, Yamakawa K.
Common regions of deletion on chromosomes 5q, 6q and 10q in renal cell
carcinoma. Cuncer Res 1991; 51: 5817-5820.
33. Wieland 1, Bohm M. Frequent allelic deletion at a novel locus on chromosome 5 in human lung cancer. Cancer Res 1994; 54: 1772-1774.
34. Auerbach 0, Garfinkel L. The changing pattern of lung carcinoma. Cancer
1991; 6 8 1973-1977.
35. Kerr KM, Carey FA, King G, Lamb D. Atypical alveolar hyperplasia:
relationship with pulmonary adenocarcinoma, p53 and c-erbB-2 expression.
J Pathol 1994; 114: 249-256.
Без категории
Размер файла
538 Кб
loss, small, non, ofapc, event, cells, 5q21, without, cancer, mutation, evidence, lung, heterozygosity, frequently
Пожаловаться на содержимое документа