| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Molecular Oncology, Markers, Clinical Correlates |
Laboratory of Molecular Medicine, Molecular Pharmacology and Therapeutic Program, Sloan-Kettering Institute [J. O. B., F. L., J. H. C., E. D.], Department of Surgery, Head and Neck Service, [J. O. B], and Thoracic Surgery Service [V. R.], Department of Pathology [F. L., D. K.], and Department of Medicine [E. D.], Memorial Hospital; Memorial Sloan-Kettering Cancer Center, New York, New York 10021
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Clinical and preclinical evidence support the view that pulmonary SCCs derive from progressive changes in the bronchial epithelium beginning with cellular metaplasia or atypia and progressing through varying degrees of dysplasia on to carcinoma in situ and then to invasive carcinomas (2, 3, 4, 5) . Those genetic changes that cause bronchial dysplastic lesions to progress to SCC are poorly understood. LOH involving chromosomes 3p, 5q, 9p, or 17p are reported frequently in selected preneoplastic or malignant lung lesions, and aberrant p53 expression or p53 mutations are also detected in these lesions (6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19) . Of these genetic changes, LOH of chromosomes 3p or 9p are among the most common during lung carcinogenesis (6 , 7 , 8) .
Preneoplastic bronchial epithelial lesions often exhibit LOH with increasing incidence of LOH for chromosomal 3p and 9p microsatellite markers occurring during lung carcinogenesis (7 , 8) . Analysis of preneoplastic and malignant lung lesions derived from the same patient reveal that LOH involving chromosomes 3p and 9p are often from the same allele, which suggests that these lesions are clonally related (7 , 8) . LOH at chromosome 9p often leads to loss of p16 expression and to cell cycle deregulation (8 , 20) . Other reported abnormalities in the preneoplastic and malignant lung include those affecting the FHIT gene located at chromosome 3p14.2 (17 , 18) and the MCC/APC loci that maps to chromosome 5q21 (14 , 15) .
Frequent alterations in bronchial preneoplasia include those affecting cell cycle regulation. Aberrant expression of the G1 cyclins, cyclin D1 and cyclin E, often occurs in bronchial preneoplasia (21) . Aberrant expression of these cyclins is more frequent in bronchial preneoplasia than that observed for p53 or the retinoblastoma (Rb) gene product (21) . Coaberrant expression of p53 and the epidermal growth factor receptor occurs in subsets of bronchial preneoplasia (10) . Comprehensive analysis has not been undertaken for aberrant expression or mutation of p53 and LOH at the 3p, 5q, 9p, or 17p chromosomal sites when multiple high-grade dysplasias are associated with SCC in the same pulmonary resection.
This study was conducted to examine comprehensively whether concordant or discordant LOH changes occur at chromosomes 3p, 5q, 9p, or 17p in synchronous high-grade dysplastic bronchial lesions associated with a resected SCC. Whether aberrant expression or mutations of p53 were present was studied. When histologically normal bronchial epithelium was available, these tissues were examined. This permitted an in-depth analysis of genetic changes present in normal versus preneoplastic or malignant bronchial epithelium. In several pulmonary resections, a second distinct high-grade dysplasia was identified. These were studied for concordant or discordant genetic losses or mutations relating to those detected in the histologically normal bronchus, other preneoplastic lesions, or the SCC.
Several findings are notable from this study. Common regions of chromosomal loss were detected in different high-grade dysplastic bronchial lesions associated with SCC. In other cases, clonal divergence was found in these dysplastic lesions relative to the SCC. This divergence was especially evident when p53 mutations were searched for in the preneoplastic versus malignant bronchial lesions. Histologically normal bronchial epithelium present in some of these pulmonary resections harbored genetic alterations similar to those present in the associated SCC. These findings have implications for lung cancer prevention and for understanding the steps involved in pulmonary squamous cell carcinogenesis.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Immunohistochemistry.
Immunohistochemical staining for p53 was performed on these lung
tissues. Histologically normal bronchial epithelium, high-grade
bronchial dysplasia, and SCC were each evaluated for p53 expression
using established immunohistochemical methods (10
, 21)
,
and formalin-fixed and paraffin-embedded tissue sections were
evaluated using the pAb1801 monoclonal antibody (Oncogene
Sciences, Uniondale, NY).
Microdissection.
Five-µm sections were obtained from paraffin-embedded tissue blocks
and were mounted onto glass slides, using standard techniques
(10
, 21)
. Histologically normal bronchial epithelium,
high-grade bronchial dysplasia, and SCC were microdissected
individually using a standard microdissection microscope and routine
techniques. Histologically benign lymph nodes were isolated from each
case and were the source of germ-line genomic DNA that was used to
determine LOH for each informative marker.
H&E staining of representative sections confirmed the precision of the microdissection. Unstained specimens were placed in xylene (Sigma, St. Louis, MO) solutions to dissolve the paraffin. Subsequently, 250 µl of 70% ethanol were added, and specimens were pelleted for 5 min using a microfuge. The xylene was removed, and the pellet was air-dried for 1–2 h. Tissue pellets were digested in 1% SDS with 5 mg/ml proteinase K added to the solution in a 50°C bath for 72 h or until the pellet was completely digested. Fresh proteinase K solution was added every 12 h. This solution was extracted three times with equal volumes of a phenol/chloroform solution saturated with Tris-EDTA buffer (pH 7.9). Genomic DNA was isolated from the aqueous phase by precipitation with ammonium acetate and cold 100% ethanol using glycogen (Boehringer Mannheim, Indianapolis, IN) as a carrier. Genomic DNA was precipitated at -20°C with a microfuge for 20 min. Pellets were washed twice with cold 70% ethanol and air-dried before suspension in Tris-EDTA buffer before storage at 4°C.
LOH Markers.
LOH was evaluated by PCR assays using informative microsatellite
markers and previously established techniques (7
, 8
, 13, 14, 15)
. The microsatellite markers were chosen to reflect those
DNA regions frequently lost in preneoplastic bronchial lesions or SCC.
These LOH markers were located on chromosomes 3p, 5q, 9p, or 17p. These
regions encompass the p16 locus on chromosome 9p, frequently
deleted segments of chromosome 3p, the commonly deleted region
involving the MCC/APC loci on chromosome 5q, and an affected
region of chromosome 17p.
The primer sequences for these markers were available from a microsatellite genomic database (23) , and conditions for the PCR assays were optimized for each LOH marker. PCR assays were performed typically for 25–35 cycles at an annealing temperature of 55–62°C. Prior to annealing, reactions were incubated at 95°C before preceding with the optimal cycle number for each LOH marker. The final extension reactions were for 5 min at 72°C. Amplitaq gold (Perkin-Elmer, Norwalk, CT) was used as the polymerase. Reactions were performed with 32P- or 33P-cytosine nucleotides (Amersham, Piscataway, NJ). The reaction products were size-fractionated on a 6% polyacrylamide gel. Gels were dried and autoradiography was performed using standard techniques. Autoradiographs were analyzed for evidence of LOH and were scored as showing retention of heterozygosity, LOH, or noninformative loci or markers that were not evaluable for examination for technical reasons.
DNA Sequence Analysis of p53.
Exons 5 through 9 of p53 were sequenced using established techniques
(24)
and genomic DNA was derived from
histologically normal bronchial epithelium, high-grade bronchial
dysplastic lesions, or SCC present in the same pulmonary resections.
The desired DNA regions were amplified using PCR-based purification of
two segments, a 500-bp segment spanning exons 5 and 6 and a 750-bp
segment spanning exons 7–9. The desired PCR fragments were
size-fractionated using agarose gel electrophoresis. These DNA
fragments were purified using a Qiaex II gel extraction kit, (Qiagen
Inc., Hilden, Germany). Sequence analysis was performed using a
thermosequenase kit (Amersham, Piscataway, NJ) and
33P- or 32P-labeled
dideoxynucleotide to terminate the PCR reactions. The reaction products
were then electrophoresed on a 6% acrylamide sequencing gel that was
subsequently dried. Autoradiography was performed after exposure to XAR
film (Eastman Kodak, Rochester, NY) at -70°C for an appropriate
length of time. Because of the limited genomic DNA isolated from
the high-grade dysplastic lesions, the exon sequenced was that
identified as mutant in the associated SCC. Mutations were confirmed
independently by repeat sequence analysis after reamplification of the
region of DNA containing the p53 exon of interest.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Results of the LOH analyses for each microsatellite marker appear in
Fig. 1
A. Informativity for each of the 15 LOH markers was greater
than 50% except for the D9S162 marker for which only 2 of 8 examined
cases were informative. Ninety-one loci were informative and 12 loci
were inevaluable because of insufficient genomic DNA isolated from the
high-grade bronchial dysplasias. Fig. 1
A maps
comprehensively LOH for the depicted microsatellite markers in the
histologically normal bronchial epithelium, high-grade dysplastic
bronchial epithelial tissues, and associated SCC. Representative
autoradiographs using radiolabeled PCR products and independent
microsatellite markers for chromosomes 3p, 5q, 9p, or 17p are displayed
in Fig. 1
B. A representative result for an individual
chromosome 3p marker examined in the histologically normal bronchial
epithelium adjacent to two independent high-grade dysplastic lesions
and an associated SCC from the same case is shown in Fig. 1
C. A photomicrograph for p53 immunohistochemical expression
in one representative case is displayed in Fig. 1
D.
|
. Histologically normal bronchial epithelial specimens exhibited LOH in
up to 27% of informative markers. High-grade dysplastic bronchial
lesions exhibited LOH in 18–45% of cases using these markers. An even
higher percentage (18–73%) of LOH was detected in the SCC cases. As
lesions advanced from high-grade dysplasia to SCC, there is
increasingly frequent genetic alterations. Notably, when two high-grade
dysplastic lesions were examined for common or discordant mutations,
each lesion was found to harbor LOH at chromosomes 9p and 3p, although
regions that were lost occasionally differed. This discordance was most
evident for p53 mutations. For example, in cases 1 and 3, discordant
p53 mutations were found (see Fig. 1
A and Table 1
).
|
|
on chromosome 9p and the D3S1110
marker on chromosome 3p. All of the preneoplastic and malignant lesions
examined in case 3 had LOH for the D9S171 marker. All of the
preneoplastic and malignant lesions examined in case 4 exhibited LOH
for the D9S1747 marker. When two independent high-grade dysplastic
lesions were analyzed (as in cases 1 through 4), these lesions
displayed LOH for common loci, with as many as three markers having
LOH. In cases 5 through 8, in which only one high-grade dysplastic
lesion was associated with a SCC, these lesions had a similar spectrum
of LOH. For cases 6 and 8, the preneoplastic and malignant lesions
studied had an identical pattern of LOH involving chromosome 3p. These
findings strongly implicate a clonal relationship between the cells
present in these dysplastic lesions and those in the associated SCC. There was heterogeneity in LOH detected in cases with two high-grade dysplastic lesions, as in cases 1 through 4. In each case in which two high-grade dysplastic lesions were present, there was at least one LOH microsatellite marker that did not show LOH concordance when these different dysplastic lesions were compared. Discordance was evident for some chromosomal 3p and 9p markers. Further discordance was noted when the LOH patterns were compared for the dysplastic lesions and the associated SCC.
LOH was identified in the histologically normal bronchial epithelium for chromosome 9p markers in cases 2 and 3. In case 2, the D9S171 marker exhibited LOH in the histologically normal bronchial epithelium and in the associated SCC, but not in either of the examined high-grade dysplastic bronchial lesions. In case 3, the D9S171 marker was lost in the histologically normal bronchial epithelium, in two dysplastic lesions, and in the SCC present in this resection. D9S1747 and D9S1751 markers were lost from chromosome 9p in the histologically normal bronchial epithelium in this case. These findings indicate how genetic alterations can be detected in the bronchus even when there is no histological evidence of preneoplastic changes in the bronchial epithelium. For example, in case 3, the normal bronchial epithelium had more frequent genetic alterations involving chromosome 9p markers than either high-grade dysplastic bronchial lesion present in this pulmonary resection. This is demonstrated in the LOH for D9S171, D9S1747, and D9S1751 chromosome 9p markers.
Analyses of LOH for microsatellite markers associated with the
MCC locus on chromosome 5q reveal that four of the SCC cases
exhibited LOH at this site. Two high-grade dysplastic lesions had LOH
at chromosome 5q. Both were associated with SCCs that harbored this
mutation. LOH for the p53 microsatellite marker was detected in three
of six informative SCC cases. Two of these cases contained p53
mutations, and one contained wild-type p53 sequences within exons 5–9
(Table 1)
. Three high-grade dysplastic bronchial lesions exhibited LOH
for the chromosome 17p marker and were associated with SCC that
exhibited this LOH. In case 3, the SCC and one high-grade dysplastic
bronchial lesion (designated DysA) exhibited LOH
for the p53 microsatellite marker and the same p53 mutation was
identified in these lesions. This mutation was absent in the second
dysplastic bronchial lesion (designated DysB)
that exhibited loss of the chromosome 17p LOH marker (Table 1)
.
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Clonal divergence was found in some dysplasias as compared with the associated SCC. This divergence was evident when p53 mutations were searched for in different preneoplastic and malignant lesions. Although concordant genetic alterations were identified in different high-grade dysplasias, discordance was also observed. Notably, histologically normal bronchial epithelium can exhibit genetic alterations similar to those present in an associated SCC. This indicates how some individuals having histologically normal bronchial epithelium may still be at risk for lung cancer development. This study confirmed that pulmonary SCCs exhibit LOH at chromosomes 3p, 5q, 9p, or 17p. Aberrant p53 expression or p53 mutations are evident in some of these carcinomas. LOH for chromosomes 3p, 5q, 9p, or 17p and p53 alterations are reported in selected preneoplastic and malignant bronchial lesions (6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19) . This study advances prior work by conducting a comprehensive analysis investigating the frequency and nature of these genetic changes in synchronous preneoplastic lesions and SCC present in the same pulmonary resections. LOH and mutational results were compared with findings obtained using germ-line DNA isolated from histologically benign lymph nodes present in each pulmonary resection.
Findings indicate that progressive LOH occurs for chromosomal markers
on 3p, 5q, 9p, or 17p when high-grade dysplastic bronchial lesions are
compared with SCC present in the same pulmonary resection.
Histologically normal bronchial epithelium exhibited LOH for some
markers. More frequent LOH was detected in associated bronchial
preneoplasia. SCC exhibited the highest incidence of LOH for examined
markers, as displayed in Fig. 2
. When a second high-grade dysplasia was
present, discordant LOH or DNA mutations were often detected. This was
apparent when mutations for p53 were examined, as shown in Table 1
.
Some histologically normal bronchial epithelial specimens exhibited
similar LOH patterns as in associated SCC, whereas LOH patterns for
other high-grade dysplasias differed from those in SCC. This is shown
in analyses of cases 1, 3, 5, and 6 in Table 1
. This demonstrates how
discordant p53 mutations are frequent in synchronous preneoplastic and
malignant pulmonary lesions. These findings are consistent with
discordant p53 immunostaining and mutational results, as previously
reported in NSCLC (25)
.
Loss of genetic material on chromosomes 3p and 9p are early events in
squamous cell carcinogenesis (7
, 8)
. This chromosomal loss
leads to inactivation of several known or potential tumor suppressors.
Aberrant expression of G1 cyclins is reported in
bronchial preneoplasia (21)
. These cyclins have been
proposed as therapeutic targets for lung cancer prevention
(26, 27, 28)
. These genetic changes can lead to cell cycle
deregulation and uncontrolled proliferation that results in genomic
instability and further abnormalities. Genomic instability is reflected
in the findings displayed in Fig. 1
. Clones of altered bronchial
epithelial cells may acquire additional genetic changes that confer a
growth advantage to these cells.
This study reports that clonal divergence is frequent because
discordant p53 mutations are detected in preneoplastic
versus malignant bronchial tissues (Table 1)
. Aberrant p53
expression was reported previously at late stages of bronchial
preneoplasia (10
, 21)
. These findings and that reported in
this study indicate that p53 alterations are late events in squamous
cell carcinogenesis. Aberrant expression of G1
cyclins (21)
and the epidermal growth factor receptor is
more frequent in bronchial preneoplasia than is altered p53 expression
(10)
. How these changes in gene expression cooperate with
chromosomal LOH is the subject of future work. Specific cassettes of
dominant or recessive genetic events might be required to transform
these preneoplastic lung lesions.
The pattern of genetic changes identified in these dysplastic lesions was often similar to those present in associated SCC. For example, LOH at chromosomes 5q or 17p present in dysplastic lesions was also detected in the associated SCC. Whereas discordant p53 mutations were evident in some SCC as compared with high-grade dysplasia, concordant mutations were detected in two cases. This reveals how a close genetic relationship can exist between SCC and associated preneoplastic lesions. Thus, multiple pathways may exist for malignant progression of preneoplastic lesions present in the same carcinogen-exposed bronchial epithelial field (29) .
The findings reported in this study have implications for understanding
squamous cell carcinogenesis and for developing lung cancer prevention
strategies. Genetic deletions, especially involving chromosomes 3p and
9p, are not uncommon in histologically normal bronchial epithelium and
dysplastic lesions of individuals exposed to tobacco-derived
carcinogens. These deletions are often similar or identical to those in
the associated SCC but can diverge from genetic changes in a second
dysplasia. These and other findings are summarized in Fig. 3
, which emphasizes how many carcinogenic steps precede and
follow development of high-grade bronchial dysplasias. Solid
arrows in Fig. 3
indicate how sequential alterations in
preneoplastic bronchial lesions accumulate genetic changes in tissues
before invasive SCC arise. A clinical implication of these findings
relating to lung cancer prevention is depicted by dashed
arrows in this figure.
|
A.
Distinct subsets of clonal bronchial epithelial cells can exist in the
epithelium (13)
. The fate of these genetically altered
cells is uncertain, as depicted by the dashed arrows in Fig. 3
. It is not known whether these altered epithelial cells ultimately
contribute to the development of a malignant lung tumor. Yet, evidence
for similar genetic changes in the carcinogen-exposed normal bronchial
epithelium and associated preneoplasia or SCC indicate that these
histologically normal bronchial epithelial cells may place the affected
epithelium at risk for neoplastic transformation. On the basis of these
findings and others reported in the bronchial epithelium of current or
former smokers (30)
and in the bronchial epithelium
adjacent to small cell lung cancers (31)
or NSCLC
(13)
, future clinical trials might target genetically
altered but histologically normal bronchial epithelium for treatment
with chemopreventive agents to inhibit or block neoplastic progression. In summary, this study demonstrates that concordant and discordant mutations exist in histologically normal bronchial epithelium and high-grade bronchial dysplasias associated with SCC. Progressive LOH is found for microsatellite markers of chromosomes 3p, 5q, 9p, or 17p. Aberrant expression or mutations of p53 are late events in bronchial preneoplasia. Discordant mutations (especially involving p53) are evident in some preneoplastic lesions when compared with associated SCC. Because similar mutations can occur in histologically normal and preneoplastic or malignant bronchial epithelium, this should be considered in the design of lung cancer prevention trials.
|
FOOTNOTES |
|---|
1 This work was supported by USPHS Grants
RO1-CA887546 (to E. D.) and T32 CA 09685 (to J. O. B.); National
Cancer Institute, NIH, Department of Health and Human Services; and in
part by American Cancer Society Grant RPG-90-019-10-DDC (to E. D.) and
the American Society of Clinical Oncology Young Investigator Award (to
J. O. B.). 
2 Present address: Harper Hospital, Department of
Pathology, Wayne State University, Detroit, MI 48201-2097.
3 Present address: Department of Pharmacology and
Toxicology, Dartmouth Medical School, Hanover, NH 03755.
4 To whom requests for reprints should be
addressed, at Departments of Medicine and of Pharmacology and
Toxicology, 7650 Remsen, Dartmouth Medical School, Hanover, NH
03755-3835. Phone: (603) 650-1667; Fax: (603) 650-1129.
5 The abbreviations used are: NSCLC, non-small
cell lung cancer; SCC, squamous cell cancer; LOH, loss of
heterozygosity.
Received 7/14/00; revised 11/ 7/00; accepted 11/13/00.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  K. H. Dragnev, D. Stover, and E. Dmitrovsky Lung Cancer Prevention: The Guidelines Chest, January 1, 2003; 123(1_suppl): 60S - 71S. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F.R. Hirsch, D.T. Merrick, and W.A. Franklin Role of biomarkers for early detection of lung cancer and chemoprevention Eur. Respir. J., June 1, 2002; 19(6): 1151 - 1158. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Sasatomi, S. D. Finkelstein, J. D. Woods, A. Bakker, P. A. Swalsky, J. D. Luketich, H. C. Fernando, and S. A. Yousem Comparison of Accumulated Allele Loss between Primary Tumor and Lymph Node Metastasis in Stage II Non-small Cell Lung Carcinoma: Implications for the Timing of Lymph Node Metastasis and Prognostic Value Cancer Res., May 1, 2002; 62(9): 2681 - 2689. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. H. Banham, N. Beasley, E. Campo, P. L. Fernandez, C. Fidler, K. Gatter, M. Jones, D. Y. Mason, J. E. Prime, P. Trougouboff, et al. The FOXP1 Winged Helix Transcription Factor Is a Novel Candidate Tumor Suppressor Gene on Chromosome 3p Cancer Res., December 1, 2001; 61(24): 8820 - 8829. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
