- Email: [email protected]
Amplification and expression of the cyclin D1 gene in anal and esophageal squamous cell carcinomas
Amplification and Expression of the Cyclin D I Gene in Anal and Esophageal Squamous Cell Carcinomas I. SHEYN, MD, A. E. NOFFSINGER, MD, S. HEFFELFINGER, MD, PHD, B. DAVIS, BS, M. AI MILLER, BS, AND C. M, FENOGLIO-PREISER, MD Cyclin D1 is a cell-cycle regulator and candidate proto-oncogene implicated in the pathogenesis of numerous tumor types. Amplification of the cyclin D1 gene occurs commonly in esophageal squamous cell carcinomas. However, no studies have examined the role of cyclin D1 in anal carcinogenesis. We examined 20 esophageal squamous cell carcinomas and 24 anal carcinomas for cyclin D1 alterations. Protein expression was evaluated by immunohistochemistry using the cyclin D1GM antibody (Novocastra, Newcastle upon Tyne, UK). Cyclin D1 amplification was examined by fluorescent in situ hybridization (FISH), using a cyclin D1 probe obtained from Toshiya Inaba at St. Jude Children's Research Hospital, Memphis, TN. The FISH sections were analyzed using a Leica (Deerfield, IL) confocal microscope. By immunohistochemistry, 75% of esophageal carcinomas showed evidence of cyclin D1 expression. Cyclin D1 amplification was detected by FISH in 65% of esophageal cancers. There was good
correlation between cyclin D1 protein expression and gene amplification, although some tumors showed protein overexpression in the absence of gene amplification. Among the 24 anal carcinomas studied, 8% showed weak cyclin D1 immunoreactivity in rare tumor cells. None of the anal tumors showed cyclin D 1 amplification. We conclude that cyclin D1 alterations are common in esophageal carcinomas but do not appear to be important in anal carcinogenesis. Immtmohistochemical detection of cyclin D1 protein overexpression is a good predictor of cyclin D1 ampfification. HuM PATHOL 28:270-276. Copyright © 1997 by W.B. 8aunders Company Key words: cyclin, sqnamous cell carcinoma, amplification, FISH, immunohistochemistry. Abbreviations: FISH, fluorescent in situ hybridization; H-E, hematoxylin-eosin; CIS, carcinoma in situ; SSC, sodium chloride sodium citrate.
The PRAD1/cyclin D1 gene lies on c h r o m o s o m e 11q13 and forms part of an amplicon that includes at least two other genes, int-2 and hst-1. This g r o u p of genes is frequently amplified in carcinomas arising in various sites, including the breast, h e a d and neck, lung, and in esophageal squamous cell carcinomas. 124 In addition, overexpression of the cyclin D1 protein may occur in the absence of gene amplification. 25'26 The role of cyclin D1 as a potential o n c o g e n e was suggested when r e a r r a n g e m e n t of the gene was f o u n d in parathyroid a d e n o m a s and some lymphomas. 27-29 Cyclin D1 plays a critical role in cell cycle regulation. W h e n the gene b e c o m e s altered in a cell, one would expect proliferative abnormalities that predispose it to neoplastic transformation. Most investigators have shown cyclin D1 amplification by the use of Southern blots. However, this methodology does not allow one to distinguish between those cells that show gene amplification a n d those that do not. In contrast, fluorescent in situ hybridization (FISH) allows one to precisely localize the amplified gene within cells. In addition, commercially available m o n o c l o n a l antibodies have recently allowed localization of the cyclin D1 protein in cells by immunohistochemistry. Because of the potential i m p o r t a n c e of cyclin D1 alterations in squamous cell carcinogenesis
in a n u m b e r of sites, we elected to use FISH and immunohistochemical localization to d e t e r m i n e (1) the feasibility of FISH analysis in evaluating cyclin D1 amplification in formalin-fixed, paraffin-embedded tissues, (2) the frequency of cyclin D1 amplification a n d protein overexpression in both esophageal and anal carcinomas, and (3) the relationship of i m m u n o h i s t o c h e m i c a l staining for cyclin D1 protein to cyclin D1 gene amplification as detected by FISH analysis.
From the Department of Pathology and Laboratory Medicine, University of Cincinnati, Cincinnati, OH. Accepted for publication June 30, 1996. Supported in part by American Cancer Society Clinical Oncology Career Development Award no. 94-66. Address correspondence and reprint requests to Amy E. Noffsinger, MD, Department of Pathology and Laboratory Medicine, University of Cincinnati, PO Box 670529, Cincinnati, OH 45267-0529. Copyright © 1997 by W.B. Saunders Company 0046-8177/97/2803-000855.00/0
270
METHODS Tissue Selection Formalin-fixed, paraffin-embedded tissues from 20 patients with esophageal squamous cell carcinoma and 24 patients with anal carcinoma were retrieved from the Surgical Pathology files of the University of Cincinnati. An additional esophagectomy specimen without neoplasia was included as a negative control. Five-micron sections were cut from the paraffin blocks and placed on positively charged glass slides (Fisher Scientific, Pittsburgh, PA). One section was stained with hematoxylineosin (H-E), and was evaluated histologically, and the tumor type and grade was determined. In addition, the presence of carcinoma in sire (CIS) and overlying normal squamous epithelium was noted so that these areas could also be examined for cyclin D amplification. The H-E-stained sections were then photographed at low power. The resultant photographs were scanned using a Microtek 35T slide scanner and an Apple Macintosh Quadra computer equipped with Adobe Photoshop software (Apple Computer, Cupertino, CA). The scanned images were printed onto 8.5 X 11 paper on a Hewlett Packard Paint Writer XL color printer (Hewlett Packard, McMinnville, OR). The areas of histological interest were then mapped on these prints for comparison with the fluorescent-labeled sections.
CYCLIN D1 IN ANAL AND ESOPHAGEAL CARCINOMA (Sheyn et al)
FIGURE 2. (A) Hematoxylin-eosin-stained section showing a moderately differentiated squamous cell carcinoma of the esophagus. (B) Immunohistochemical staining for cyclin D1 shows positive nuclear immunoreactivity in many of the neoplastic cells.
FIGURE 1. Diagrammatic representation of the localization of the probes used for the FISH studies.
Immunohistochemistry Five-micron sections applied to positively charged slides were used for immunohistochemical staining with the mouse monoclonal antibody DIGM directed against cyclin D1 (Novocastra Laboratories, Newcastle upon Tyne, UK). The slides
were deparaffinized in xylene and placed in Coplinjars with 75 mL citrate buffer at pH 6.0. They were microwaved on high for 7.5 minutes (to boil), and then microwaved at 60% power for 15 minutes (to simmer). After cooling for 30 minutes, the slides were rinsed in distilled water and processed with a Ventana automated immunohistochemical staining unit (Ventana, Tucson, AZ). The antibody was used at a dilu-
FIGURE 3. (A) Hemtoxylin-eosin-stained section showing a moderately differentiated squamous cell carcinoma arising in the anal canal, (B) The neoplastic cells are nonimmunoreactive with the anti-cyclin D1 antibody.
271
HUMAN PATHOLOGY
Volume 28, No. 3 (March 1997)
tion of 1:15. Paraffin-embedded cell blocks containing A431 cells, an epidermoid carcinoma cell line with known cyclin D1 amplification, served as a positive control.
Fluorescent In Situ Hybridization (FISH) Tissues used for FISH analysis included paraffin-embedded tissues from anal and esophageal resection specimens, paraffin-embedded cell blocks consisting of A431 cells, and metaphase spreads from A431 cells. Paraffin-embedded tissues. Slides were heated to 65°C overnight and then dewaxed in xylene and rehydrated through a series of graded alcohols, then allowed to air dry. They were then incubated at 45°C for 45 minutes in Oncor's (Galthersburg, MD) pretreatment solution. The slides were washed in 2× sodium chloride sodium citrate (SSC), and then subjected to proteinase K digestion (250 # g / m L in 2× SSC, prewarmed to 45°C). The slides were incubated in a humid chamber for 20 minutes, washed in 2× SSC, dehydrated through a series of alcohols, and allowed to alrdry. To evaluate the adequacy of proteinase digestion, each of the slides was stained with propidium iodide/Antifade (2.4 # g / m L ; Oncor, Gaithersburg, MD), coverslipped, and evaluated under a fluorescent microscope. If the tissue was underdigested, the proteinase K digestion step was repeated and the slides reevaluated until digestion was optimal. The slides were then dehydrated through a series of alcohols and air-dried. All paraffin-embedded tissues were dual labeled using a digoxigenin-labeled cyclin D1 probe (kindly provided by Toshiya Inaba, St. Jude Children's Research Hospital, Memphis, TN) and a commercially available biotinylated chromosome 11 c~ satellite probe (Oncor, Gaithersburg, MD). The chromosomal locations of the probes and their relationships to the 11q13 amplicon are shown in Fig 1. FISH hybridization buffer (50% formamide, 2× SSC, 40 m m o l / L sodium phosphate buffer p H 7, 1× Denhardt's solution, 0.1% sodium dodecyl sulfate, 10% dextran sulfate) containing the ~ satellite probe (0.5 n g / # L ) and the cyclin D1 probe (1.5 n g / # L ) was added to each slide. The slides were coverslipped, and heated to 90°C for 10 minutes. The sections were incubated in a humid chamber overnight at 37°C. After hybridization the slides were subjected to washes of 2× SSC, 50% formamide (43°C for 15 minutes) and 2× SSC (37°C for 8 minutes). The slides were next washed in 1 X phosphate-buffered detergent (PBD) (phosphate-buffered saline + 0.4% bovine serum albumin + 0.1% Tween 20) at room temperature for at least 2 minutes. Thirty microliters each of fluorescein isothiocyanate labeled anti-digoxigenin and Texas r e d - l a b e l e d avidin were added to each slide and allowed to incubate at 37°C for 5 minutes. The slides were again washed in 1 × PBD. Amplification of the detection signal was achieved through incubating with 60 #L rabbit anti-sheep antibody (15 minutes at 37°C) followed by 60 #L FI4abeled anti-rabbit antibody (15 minutes at 37°C), 60 #L anti-avidin antibody (5 minutes at 37°C) and 60/zL Texas r e d - l a b e l e d avidin (5 minutes at 37°C). The slides were washed in 1× PBD for 2 minutes after incubation with each new antibody. The tissue sections were counterstained with 20 #L 4',6-diamino-2-phenylindole dihydrochloride (DAPI)/Antifade, and coverslipped. Slides were stored at 4°C in dark, humid chambers. Metaphase spreads. Metaphase 3~ spreads were prepared from A431 cells by standard methods. If the metaphase preparations were less than 2 weeks old, they were artificially aged by incubating in 70 mL 2× SSC preheated to 37°C for 30 minutes. If the preparations were more than 2 weeks old, this step was omitted from the protocol. Slides were then dehydrated through a graded series of alcohols and allowed
272
to air dry. After air-drying, the tissues were denatured by treating with 70% formamide, 2?< SSC (pH 7.0) at 70°C for 2 minutes. The slides were again dehydrated by washing in a graded series of alcohols cooled to -20°C, and allowed to airdry. All metaphase spreads were dual labeled using the digoxigenin-labeled cyclin D1 probe described previously, and the commercially available biotinylated Coatosome whole chromosome 11 probe (Oncor, Gaithersburg, MD). The chromosome 11 probe was prewarmed to 37°C for 5 minutes, then denatured at 70°C for 10 minutes (20 #L per slide). The probe was then incubated at 37°C for 2 hours to preanneal. The cyclin D1 probe was denatured at 70°C for 5 minutes (7.5 #L of stock at 6 n g / # L per slide). The two probes were combined and placed on ice. Next, each slide was covered with 27.5 /~L of the probe solution and coverslipped. Slides were incubated overnight at 37°C in a humid chamber, using a standard humid chamber solution (50% formamide, 2× ssc). Posthybridization washes consisted of 50% formamide, 2× SSC at 43°C for 15 minutes followed by 0.1× SSC at 43°C for 15 minutes. The tissue sections were counterstained with 20 #L DAPI/Antifade, and coverslipped. Slides were stored at 4°C in dark, humid chambers. Detection of the hybridization reaction was as described previously for paraffin-embedded tissues.
FISH Evaluation All of the FISH-labeled slides were screened using a standard fluorescent microscope. Screening included evaluation of the quality of the hybridization reaction, and selection of areas of normal epithelium, CIS, and invasive carcinoma for subsequent analysis on a Leica laser confocal microscope (Deerfield, IL). Areas of histological interest were identified by comparison with scanned and printed maps of the corresponding hematoxylin-eosin-stained section. All cases were examined for the presence of cyclin D1 amplification (defined as more than two foci of hybridization per nucleus) and chromosome 11 duplication. At least 50 cells from each tumor were examined, and the pattern of cyclin D1 hybridization compared with adjacent nonneoplastic stromal or lymphoid cells. A431 cells, known to have amplification of the cyclin D1 gene served as positive controls. An esophageal resection specimen from a patient with a nonneoplastic esophageal disease (heterotopic pancreas) served as a negative control. A431 cell metaphases were examined to localize the amplified cyclin D1 genes.
RESULTS Histological Examination The formalin-fixed, paraffin-embedded carcinom a s e x a m i n e d in this study were all s q u a m o u s cell carcin o m a s . O f t h e 20 e s o p h a g e a l t u m o r s , two were well d i f f e r e n t i a t e d , 12 m o d e r a t e l y d i f f e r e n t i a t e d , a n d six poorly differentiated. One of the poorly differentiated c a r c i n o m a s was a s p i n d l e cell c a r c i n o m a . I n a d d i t i o n , o n e e s o p h a g e c t o m y s p e c i m e n f r o m a p a t i e n t with heterotopic pancreas and an inflammatory pseudotumor involving t h e distal e s o p h a g u s was i n c l u d e d in t h e s t u d y as a n o n n e o p l a s t i c c o n t r o l . T h e g r o u p o f 24 a n a l carcin o m a s c o n s i s t e d o f f o u r w e l l - d i f f e r e n t i a t e d , 13 m o d e r ately d i f f e r e n t i a t e d , five p o o r l y d i f f e r e n t i a t e d , a n d two b a s a l o i d s q u a m o u s cell c a r c i n o m a s .
CYCLIN D1 IN ANAL AND ESOPHAGEAL CARCINOMA (Sheyn et al) TABLE 1,
Results of Cyclin D1 Analyses in Paraffin-Embedded Esophageal Carcinomas Cyclin D1 Amplification
Cyclin D1 I H C Case 1 2
3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Histological Grade
Pathological Stage
SCC
NL
SCC
PD MD MD MD-PD MD PD MD WD MD MD WD M-PD MD S p i n d l e cell PD MD MD MD MD MD
ND T4N 1 T4NX T3NX T4N1 T3N1 T4N0 T4NX ND T1NX T2N0 T3N1 T4N0 T2NX ND T4N 1 T3N1 T3N1 T4N1 ND
ND Rare ND Rare Rare + + Rare ND . + . Rare + + + + +
ND ND ND Rare ND Rare
+
-
+ + + + + +
ND ND ND ND ND
+ +
-
+ + + +
ND ND -
.
.
.
.
. Rare ND Rare ND
NL
.
-
ND
A b b r e v i a t i o n s : SCC, s q u a m o u s cell c a r c i n o m a ; CIS, c a r c i n o m a in situ; NL, n o r m a l ; I H C , i m m u n o h i s t o c h e m i s t r y ; WD, well d i f f e r e n t i a t e d ; MD, m o d e r a t e l y d i f f e r e n t i a t e d ; PD, p o o r l y d i f f e r e n t i a t e d ; ND, n o d a t a available.
Immunohistochemistry Immunohistochemical staining with the anti-cyclin D1 antibody was performed in 17 of 20 esophageal carcinomas and all of the 24 anal carcinomas. Three esophageal cases had insufficient tissue for immunohistochemistry after the FISH studies were completed. Strong nuclear cyclin D1 immunoreactivity was observed in 8 of 17 cases examined (Fig 2). In these tumors, the number of immunoreactive neoplastic cells varied from 40% to 75% of the total cell population. The immunoreactive cells were diffusely distributed within the tumor cell nests. An additional five cases contained rare (<5%), diffusely distributed, immunoreactive tumor cells. In most of these cases, the staining for cyclin D1 was weak. One case that showed strong cyclin D1 staining in the invasive carcinoma also showed staining in areas of carcinoma in situ. Carcinoma in situ was not present in any of the other sections that showed strong cyclin D1 staining. In most cases, the nonneoplastic squamous mucosa showed no immunoreactivity with the anti-cyclin antibody. Two cases, however, showed rare positive cells in the basal portion of the epithelium. Only those cases that showed cyclin D1 immunoreactivity in greater than 5% of neoplastic cells were considered to be overexpressing the cyclin D1 protein. The results of the cyclin D1 immunohistochemistry in esophageal carcinomas are summarized in Table 1. Twenty-three of the 24 anal carcinomas were negative for cyclin D1 expression by immunohistochemistry (Fig 3). In two cases, rare weakly immunoreactive cells were present in the basal layer of the nonneoplastic epithelium overlying the invasive carcinoma. In one of these cases, weak staining was also present in rare, diffusely distributed t u m o r cells ( < 5 % ) . None of the cases showed strong staining with the anti-cyclin D1 antibody.
273
Fluorescent In Situ Hybridization Cyclin D1 amplification was consistently observed in A431 cells. In addition, most of these cells appeared to contain four copies of c h r o m o s o m e 11. In many cases, cyclin D1 was observed both on c h r o m o s o m e 11, and in extrachromosomal locations. This was best visualized in the metaphase spread preparations (Fig 4). The results of the FISH in the formalin-fixed, paraffin-embedded esophageal squamous cell carcinomas are summarized in Table 1. Amplification of cyclin D1 was identified in 13 of 20 (65%) cases of esophageal carcinoma (Fig 5). C h r o m o s o m e 11 duplication occurred in 6 of 14 (43%) cases, and was usually associated with amplification of the cyclin D1 gene. Cyclin D1 amplification was also observed in areas of CIS in the two cases in which CIS was present in the section, but was never observed in normal squamous epithelium (Fig 6). In many cases, apparent endoduplication o f cyclin D1 was observed on c h r o m o s o m e 11. As was observed in the A431 cells, m u c h of the amplified cyclin D1 was not localized to c h r o m o s o m e 11 but app e a r e d to be present in scattered foci t h r o u g h o u t the nucleus. Among the esophageal carcinomas that showed immunohistochemical evidence of cyclin D1 overexpression ( > 5 % positive nuclei), all but one also showed evidence of gene amplification by FISH. T h r e e cases with amplified cyclin D1, however, showed either negative or weak staining in rare cells with tile cyclin D1 antibody, suggesting that the cyclin D1 protein was not overexpressed in these tumors despite gene amplification. T h e r e was no correlation between cyclin D1 amplification and either tumor grade or stage. Cyclin D1 amplification was not observed in any of the 24 anal squamous cell carcinomas examined.
HUMAN PATHOLOGY
Volume 28, No. 3 (March 1997)
B
FIGURE 4. FISH using probes for cyclin D1 (green) and chromosome 11 (red). YeLlow dots represent areas of green cyclin D1 signal superimposed on the red chromosome 11 signal. (A) Paraffin-embedded cell block preparation using A431 cells, Note the presence of numerous copies of the cyclin D1 gene in each cell, in some cells the gene is localized to chromosome 11, but is also present in extrachromosomal locations. Each of the cells is aneuploid, containing more than two copies of chromosome 11. (B) Metaphase spread preparation from A431 celts. Note the presence of four copies of chromosome 11, and multiple copies of the cyclin D1 gene localized to both chromosome 11 and extrachromosomal sites.
Y.
J~
~"
k
"L
FIGURE 5. Cyclin [)1 amplification in an invasive squamous cell carcinoma. Each of the neoplastic nuclei shows cyclin D1 amplification, and many cells contain more than two copies of chromosome 11.
FIGURE 6. FISH in nonneopiastic squamous mucosa of the esophagus. Note that each nucleus contains no more than two copies of the cyclin D1 gene, and two copies of chromosome 11.
274
CYCLIN D1 IN ANAL AND ESOPHAGEAL CARCINOMA (Sheyn et al)
DISCUSSION
carcinomas. 42 The mechanisms of carcinogenesis in the cervix and vulva are thought to parallel those in the anal region in that infection with the h u m a n papillomavirus is an important initiating event in all of these sites. Our findings, however, suggest that in contrast to the findings in carcinomas of the lower female genital tract, cyclin D1 alterations do not play a significant role in tumorigenesis in the anal region. The reason that we detected amplification in a higher percentage of esophageal carcinomas cases than has been previously reported is most likely related to a difference in techniques used for evaluation of gene amplification. Previous studies have used DNA transfer and hybridization techniques such as dot, slot, and Southern blotting to detect gene amplification. These techniques require that DNA be extracted from either fixed or fresh whole tissues. These tissues contain not only tumor cells, but variable amounts of normal tissue as well. As a result, contamination of tumor DNA with sufficient DNA from normal cells that lack amplification of the gene in question, may obscure gene amplification. This is especially true in cases in which only a small a m o u n t of tumor is present. The use of in situ techniques such as FISH, however, allows one to evaluate individual cells in their histological context. For this reason, amplification can be easily identified, even when only small numbers of neoplastic cells are present in the specimen. This is especially true when one couples FISH with confocal microscopy to generate high-resolution images of the hybridization product in the cell nucleus. In addition, as shown in the current study, FISH analysis can be readily applied to formalin-fixed, paraffin-embedded tissues. DNA transfer and hybridization techniques such as Southern blotting cannot be performed on such materials, but instead require fresh whole tissues, which are not always readily available. For this reason, FISH analysis is well suited for use in retrospective studies using archival tissues. Finally, in situ techniques allow one to separately characterize areas of normal epithelium and stroma, areas of dysplasia and carcinoma in situ, and areas of invasive carcinoma. With the use of FISH, we were able to show the presence of cyclin D1 amplification within foci of CIS in two esophageal cancer patients. In addition, the use of dual labeling and hybridization with multiple probes allows one to localize the gene to its specific chromosome. In the current study, we were able to observe that the amplified cyclin D1 was not always present in its normal location on c h r o m o s o m e 11. In many cases, the amplified DNA appeared to be extrachromosomal, perhaps within double minutes. Such observations in situ may eventually allow one to make predictions regarding the role of cyclin D1 in carcinogenesis.
Amplification of the 11q13 region occurs commonly in squamous cell carcinomas arising in the lung, head and neck, as well as the esophagus. This region is known to contain an amplicon that includes the genes int-2, hst-1 and PRAD1/cyclin D1. The int-2 and hst-1 genes are known to be members of the fibroblast growth factor gene family, and have been implicated as proto-oncogenes in the genesis of mouse mammary carcinomas. ~1s4 Cyclin D1 plays a key role in the regulation of the cell cycle s5 and therefore is considered an oncogene. Studies of h u m a n breast carcinomas have shown that despite the frequent amplification of int-2 and hst-1, these genes are not overexpressed, suggesting that they do not1 p 7lay an imp ortant role in carcino genesis in this site. 4,3
In esophageal carcinomas, the reported frequency of cyclin D1 amplification ranges from 24% to 54% of c a s e s . 19'20'22'23'40 Similar results have been reported in laryngeal carcinomas. 2 We were unable to find any data regarding cyclin D1 alterations in anal carcinomas in the English literature. In a study of cervical and vulvar squamous neoplasia, cyclin D1 amplification was frequent, occurring in three of three vulvar carcinoma cell lines and in 4 of 10 cell lines derived from cervical
REFERENCES 1. Ali IU, Merlo G, Callahan R, et al: The amplification unit on chromosome llq13 in aggressive primary human breast tumors entails the bcl-1, int-2 and hst loci. Oncogene 4:89-92, 1989 2. Jares P, Fernandez PL, Campo E, et al: PRAD-1/cyclin D1
275
HUMAN PATHOLOGY
Volume 28, No. 3 (March 1997)
gene amplification correlates with messenger RNA overexpression and tumor progression in human laryngeal carcinomas. Cancer Res 54:4813-4817, 1994 3. Callender T, Elnaggar AK, Lee MS, et al: PRAD-1 (CCND1) cyclin D1 oncogene amplification in primary head and neck squamous cell carcinoma. Cancer 74:152-158, 1994 4. Gillett C, Fantl V, Smith R, et al: Amplification and overexpression of cyclin D1 in breast cancer detected by immunohistochelnical staining. Cancer Res 54:1812-1817, 1994 5. Zhou DJ, Casey G, Cline MJ: Amplification of human int-2 gene in breast cancers and squamous carcinomas. Oncogene 2:279282, 1988 6. Berenson JR, Koga H, Yang J, et al: Frequent amplification of the bcl-1 locus in poorly differentiated squamous cell carcinoma of the lung. Oncogene 5:1343-1348, 1990 7. Somers K, Cartwright S, Schechter G: Amplification of the int-2 gene in human head and neck squamous cell carcinomas. Oncogene 5:915-920, 1990 8. Leonard JH, Kearsley JH, Chenevix-Trench G, et al: Analysis of gene amplification in head and neck squamous cell carcinomas. IntJ Cancer 48:511-515, 1991 9. BerensonJR, YangJ, Mickel R: Frequent amplification of the bcl-1 locus in head and neck squamous cell carcinomas. Oncogene 4:1111-1116, 1989 10. Lidereau R, Callahan R, Dickson C, et al: Amplification of the int-2 gene in primary human breast tumors. Oncogene Res 2:285291, 1988 11. VarleyJM, Walker RA, Casey G, et al: A common alteration to the int-2 protooncogene in DNA from primary breast carcinomas. Oncogene 3:87-91, 1988 12. Theillet C, AdnaneJ, Szepetowski P, et al: BCL-1 participates in the llq13 amplification found in breast cancer. Oncogene 5:147149, 1990 13. Tsuda H, Hirohashi S, Shimosato 5(, et al: Correlation between long-term survival in breast cancer patients and amplification of two putative oncogene-coamplification units: hst-1/int-2 and c-erbB2/ear-l. Cancer Res 49:3104-3108, 1989 14. Fantl V, Richards MA, Smith R, et al: Gene mnplification on chromosome band llq13 and oestrogen receptor status in breast cancer. EurJ Cancer 26:423-429, 1990 15. Borg A, Sigurdsson H, Clark GM, et al: Association of INT2/ HST1 coamplifieation in primary breast cancer with hormone-dependent phenotype and poor prognosis. BrJ Cancer 63:136-142, 1991 16. Saint-Ruf C, Malfoy B, Scholl S, et al: Gst~r gene is frequently co-amplified with INT2 and HSTF1 proto-oncogenes in human breast cancers. Oncogene 6:403-406, 1991 17. Schuuring E, Verhoeven E, Mooi WJ, et al: Identification and cloning of two overexpressed genes U21B31/PRAD1 and EMS1, within the amplified chromosome llq13 region in human carcinomas. Oncogene 7:355-361, 1992 18. Speicher MR, Howe C, Crotty P, et al: Comparative genomic hybridization detects novel deletions and amplifications in head and neck squamous cell carcinomas. Cancer Res 55:1010-1013, 1995 19. Wagata T, Ishizald K, Imamura M, et al: Deletion of 17p and amplification of the int-2 gene in esophageal carcinomas. Cancer Res 51:2113-2117, 1991 20. Adelaide J, Monges G, Derderian C, et al: Oesophageal cancer and amplification of the human cyclin D gene CCND1/PRAD1. BrJ Cancer 71:64-68, 1994 21. Jaing W, Zhang Y-J, ILahn SM, et al: Altered expression of the cyclin D1 and retinoblastoma genes in human esophageal cancer. Proc Natl Acad Sci U S A 90:9026-9030, 1993 22. Mori M, Tokino T, Yanagisawa A, et al: Association between
276
chromosome llq13 amplification and prognosis of patients with oesophageal carcinomas. EurJ Cancer 28A:755-757, 1992 23. KitagawaY, Ueda M, Ando N, et al: Significance of int-2/hst1 coamplification as a prognostic factor in patients with esophageal squamous carcinoma. Cancer Res 51:1504-1508, 1991 24. Gramlieh TL, Fritsch CR, Maurer D, et al: Differential polymerase chain reaction assay of cyclin D1 gene amplification in esophageal carcinoma. Diagn Mol Pathol 3:255-259, 1994 25. Zukerberg LB, Yang W-I, Gadd M, et al: Cyclin D1 (PRAD1) protein expression in breast cancer: Approximately one-third of infiltrating mammary carcinomas show overexpression of the cyclin D1 oncogene. Mod Pathol 8:560-567, 1995 26. Buckley MF, Sweeney KJE, Hamilton JA, et al: Expression and amplification of cyclin genes in human breast cancer. Oncogene 8:2127-2133, 1993 27. Motokura T, Bloom T, Kim HG, et al: A novel cyclin encoded by a bc/l-linked candidate oncogene. Nature 350:512-515, 1991 28. Arnold A, Kim HG, Gaz RD, et al: Molecular cloning and chromosomal mapping of DNA rearranged with the parathyroid hormone gene in a parathyroid adenoma. J Clin Invest 83:2034-2040, 1989 29. Tsujimoto Y, YunisJ, Onorato-Showe L, et al: Molecular cloning of the chromosomal breakpoint of B-cell lymphomas and leukemias with the t ( 11; 14) chromosome translocation. Science 224:14031406, 1985 30. Freshney RI: Culture of Animal Cells: A Manual of Basic Technique. New York, NY, Wiley-Liss, 1987, pp 169-185 31. Dickson C, Peters G: Potential oncogene product related to growth factors. Nature 326:823, 1987 (letter) 32. Casey G, Smith R, McGillivray D, et al: Characterization and chromosome assignment of human homolog of int-2, a potential proto-oncogene. Mol Cell Biol 6:502-510, 1986 33. Yoshida T, Miyagawa K, Odagiri H, et al: Genomic sequence of hst, a transforming gene encoding a protein homologous to fibroblast growth factor and the int-2 encoded protein. Proc Natl Acad Sci U S A 84:7305-7309, 1987 34. Peters G, Brookes S, Smith R, et al: The mouse homolog of the hst/K-FGF gene is adjacent to int-2 and is activated by proviral insertion in some virally induced mammary tumors. Proc Natl Acad Sci U S A 86:5678-5682, 1989 35. Scherr CJ: Mammalian G1 cyclins. Cell 73:1059-1065, 1993 36. Theillet C, Roy XL, Lapeyriere OD, et al: Amplification of FGF-related genes in human tumors: Possible involvement of HST in breast carcinomas. Oncogene 4:915-922, 1989 37. Liscia DS, Merlo GR, Garrett C, et al: Expression of int-2 mRNA in human tumors amplified at the int-2 locus. Oncogene 4:1219-1224, 1989 38. Laramie GA, Fantl V, Smith R, et al: DllS287, a putative oncogene on chromosome 1l q l 3, is amplified and expressed in squamous cell and mammary carcinomas and linked to BCL-1. Oncogene 6:439-444, 1991 39. BartkovaJ, LukaJ, Muller H, et al: Cyclin D1 protein expression and function in human breast cancer. IntJ Cancer 57:353-361, 1994 40. Jiang W, Kahn SM, Tomita N, et al: Amplification and expression of the human cyclin D gene in esophageal cancer. Cancer Res 52:2980-2983, 1992 41. Michalides R, van Veelen N, Hart A, et al: Overexpression of cyclin D1 correlates with recurrence in a group of forty-seven operable squamous cell carcinomas of the head and neck. Cancer Res 35:975-978, 1995 42. Kuzrock R, Ku S, Talpaz M: Abnormalities in the PRAD1 (cyclin D1/bcl-1) oncogene are frequent in cervical and vulvar squamous cell carcinoma cell lines. Cancer 75:584-590, 1995
correlation between cyclin D1 protein expression and gene amplification, although some tumors showed protein overexpression in the absence of gene amplification. Among the 24 anal carcinomas studied, 8% showed weak cyclin D1 immunoreactivity in rare tumor cells. None of the anal tumors showed cyclin D 1 amplification. We conclude that cyclin D1 alterations are common in esophageal carcinomas but do not appear to be important in anal carcinogenesis. Immtmohistochemical detection of cyclin D1 protein overexpression is a good predictor of cyclin D1 ampfification. HuM PATHOL 28:270-276. Copyright © 1997 by W.B. 8aunders Company Key words: cyclin, sqnamous cell carcinoma, amplification, FISH, immunohistochemistry. Abbreviations: FISH, fluorescent in situ hybridization; H-E, hematoxylin-eosin; CIS, carcinoma in situ; SSC, sodium chloride sodium citrate.
The PRAD1/cyclin D1 gene lies on c h r o m o s o m e 11q13 and forms part of an amplicon that includes at least two other genes, int-2 and hst-1. This g r o u p of genes is frequently amplified in carcinomas arising in various sites, including the breast, h e a d and neck, lung, and in esophageal squamous cell carcinomas. 124 In addition, overexpression of the cyclin D1 protein may occur in the absence of gene amplification. 25'26 The role of cyclin D1 as a potential o n c o g e n e was suggested when r e a r r a n g e m e n t of the gene was f o u n d in parathyroid a d e n o m a s and some lymphomas. 27-29 Cyclin D1 plays a critical role in cell cycle regulation. W h e n the gene b e c o m e s altered in a cell, one would expect proliferative abnormalities that predispose it to neoplastic transformation. Most investigators have shown cyclin D1 amplification by the use of Southern blots. However, this methodology does not allow one to distinguish between those cells that show gene amplification a n d those that do not. In contrast, fluorescent in situ hybridization (FISH) allows one to precisely localize the amplified gene within cells. In addition, commercially available m o n o c l o n a l antibodies have recently allowed localization of the cyclin D1 protein in cells by immunohistochemistry. Because of the potential i m p o r t a n c e of cyclin D1 alterations in squamous cell carcinogenesis
in a n u m b e r of sites, we elected to use FISH and immunohistochemical localization to d e t e r m i n e (1) the feasibility of FISH analysis in evaluating cyclin D1 amplification in formalin-fixed, paraffin-embedded tissues, (2) the frequency of cyclin D1 amplification a n d protein overexpression in both esophageal and anal carcinomas, and (3) the relationship of i m m u n o h i s t o c h e m i c a l staining for cyclin D1 protein to cyclin D1 gene amplification as detected by FISH analysis.
From the Department of Pathology and Laboratory Medicine, University of Cincinnati, Cincinnati, OH. Accepted for publication June 30, 1996. Supported in part by American Cancer Society Clinical Oncology Career Development Award no. 94-66. Address correspondence and reprint requests to Amy E. Noffsinger, MD, Department of Pathology and Laboratory Medicine, University of Cincinnati, PO Box 670529, Cincinnati, OH 45267-0529. Copyright © 1997 by W.B. Saunders Company 0046-8177/97/2803-000855.00/0
270
METHODS Tissue Selection Formalin-fixed, paraffin-embedded tissues from 20 patients with esophageal squamous cell carcinoma and 24 patients with anal carcinoma were retrieved from the Surgical Pathology files of the University of Cincinnati. An additional esophagectomy specimen without neoplasia was included as a negative control. Five-micron sections were cut from the paraffin blocks and placed on positively charged glass slides (Fisher Scientific, Pittsburgh, PA). One section was stained with hematoxylineosin (H-E), and was evaluated histologically, and the tumor type and grade was determined. In addition, the presence of carcinoma in sire (CIS) and overlying normal squamous epithelium was noted so that these areas could also be examined for cyclin D amplification. The H-E-stained sections were then photographed at low power. The resultant photographs were scanned using a Microtek 35T slide scanner and an Apple Macintosh Quadra computer equipped with Adobe Photoshop software (Apple Computer, Cupertino, CA). The scanned images were printed onto 8.5 X 11 paper on a Hewlett Packard Paint Writer XL color printer (Hewlett Packard, McMinnville, OR). The areas of histological interest were then mapped on these prints for comparison with the fluorescent-labeled sections.
CYCLIN D1 IN ANAL AND ESOPHAGEAL CARCINOMA (Sheyn et al)
FIGURE 2. (A) Hematoxylin-eosin-stained section showing a moderately differentiated squamous cell carcinoma of the esophagus. (B) Immunohistochemical staining for cyclin D1 shows positive nuclear immunoreactivity in many of the neoplastic cells.
FIGURE 1. Diagrammatic representation of the localization of the probes used for the FISH studies.
Immunohistochemistry Five-micron sections applied to positively charged slides were used for immunohistochemical staining with the mouse monoclonal antibody DIGM directed against cyclin D1 (Novocastra Laboratories, Newcastle upon Tyne, UK). The slides
were deparaffinized in xylene and placed in Coplinjars with 75 mL citrate buffer at pH 6.0. They were microwaved on high for 7.5 minutes (to boil), and then microwaved at 60% power for 15 minutes (to simmer). After cooling for 30 minutes, the slides were rinsed in distilled water and processed with a Ventana automated immunohistochemical staining unit (Ventana, Tucson, AZ). The antibody was used at a dilu-
FIGURE 3. (A) Hemtoxylin-eosin-stained section showing a moderately differentiated squamous cell carcinoma arising in the anal canal, (B) The neoplastic cells are nonimmunoreactive with the anti-cyclin D1 antibody.
271
HUMAN PATHOLOGY
Volume 28, No. 3 (March 1997)
tion of 1:15. Paraffin-embedded cell blocks containing A431 cells, an epidermoid carcinoma cell line with known cyclin D1 amplification, served as a positive control.
Fluorescent In Situ Hybridization (FISH) Tissues used for FISH analysis included paraffin-embedded tissues from anal and esophageal resection specimens, paraffin-embedded cell blocks consisting of A431 cells, and metaphase spreads from A431 cells. Paraffin-embedded tissues. Slides were heated to 65°C overnight and then dewaxed in xylene and rehydrated through a series of graded alcohols, then allowed to air dry. They were then incubated at 45°C for 45 minutes in Oncor's (Galthersburg, MD) pretreatment solution. The slides were washed in 2× sodium chloride sodium citrate (SSC), and then subjected to proteinase K digestion (250 # g / m L in 2× SSC, prewarmed to 45°C). The slides were incubated in a humid chamber for 20 minutes, washed in 2× SSC, dehydrated through a series of alcohols, and allowed to alrdry. To evaluate the adequacy of proteinase digestion, each of the slides was stained with propidium iodide/Antifade (2.4 # g / m L ; Oncor, Gaithersburg, MD), coverslipped, and evaluated under a fluorescent microscope. If the tissue was underdigested, the proteinase K digestion step was repeated and the slides reevaluated until digestion was optimal. The slides were then dehydrated through a series of alcohols and air-dried. All paraffin-embedded tissues were dual labeled using a digoxigenin-labeled cyclin D1 probe (kindly provided by Toshiya Inaba, St. Jude Children's Research Hospital, Memphis, TN) and a commercially available biotinylated chromosome 11 c~ satellite probe (Oncor, Gaithersburg, MD). The chromosomal locations of the probes and their relationships to the 11q13 amplicon are shown in Fig 1. FISH hybridization buffer (50% formamide, 2× SSC, 40 m m o l / L sodium phosphate buffer p H 7, 1× Denhardt's solution, 0.1% sodium dodecyl sulfate, 10% dextran sulfate) containing the ~ satellite probe (0.5 n g / # L ) and the cyclin D1 probe (1.5 n g / # L ) was added to each slide. The slides were coverslipped, and heated to 90°C for 10 minutes. The sections were incubated in a humid chamber overnight at 37°C. After hybridization the slides were subjected to washes of 2× SSC, 50% formamide (43°C for 15 minutes) and 2× SSC (37°C for 8 minutes). The slides were next washed in 1 X phosphate-buffered detergent (PBD) (phosphate-buffered saline + 0.4% bovine serum albumin + 0.1% Tween 20) at room temperature for at least 2 minutes. Thirty microliters each of fluorescein isothiocyanate labeled anti-digoxigenin and Texas r e d - l a b e l e d avidin were added to each slide and allowed to incubate at 37°C for 5 minutes. The slides were again washed in 1 × PBD. Amplification of the detection signal was achieved through incubating with 60 #L rabbit anti-sheep antibody (15 minutes at 37°C) followed by 60 #L FI4abeled anti-rabbit antibody (15 minutes at 37°C), 60 #L anti-avidin antibody (5 minutes at 37°C) and 60/zL Texas r e d - l a b e l e d avidin (5 minutes at 37°C). The slides were washed in 1× PBD for 2 minutes after incubation with each new antibody. The tissue sections were counterstained with 20 #L 4',6-diamino-2-phenylindole dihydrochloride (DAPI)/Antifade, and coverslipped. Slides were stored at 4°C in dark, humid chambers. Metaphase spreads. Metaphase 3~ spreads were prepared from A431 cells by standard methods. If the metaphase preparations were less than 2 weeks old, they were artificially aged by incubating in 70 mL 2× SSC preheated to 37°C for 30 minutes. If the preparations were more than 2 weeks old, this step was omitted from the protocol. Slides were then dehydrated through a graded series of alcohols and allowed
272
to air dry. After air-drying, the tissues were denatured by treating with 70% formamide, 2?< SSC (pH 7.0) at 70°C for 2 minutes. The slides were again dehydrated by washing in a graded series of alcohols cooled to -20°C, and allowed to airdry. All metaphase spreads were dual labeled using the digoxigenin-labeled cyclin D1 probe described previously, and the commercially available biotinylated Coatosome whole chromosome 11 probe (Oncor, Gaithersburg, MD). The chromosome 11 probe was prewarmed to 37°C for 5 minutes, then denatured at 70°C for 10 minutes (20 #L per slide). The probe was then incubated at 37°C for 2 hours to preanneal. The cyclin D1 probe was denatured at 70°C for 5 minutes (7.5 #L of stock at 6 n g / # L per slide). The two probes were combined and placed on ice. Next, each slide was covered with 27.5 /~L of the probe solution and coverslipped. Slides were incubated overnight at 37°C in a humid chamber, using a standard humid chamber solution (50% formamide, 2× ssc). Posthybridization washes consisted of 50% formamide, 2× SSC at 43°C for 15 minutes followed by 0.1× SSC at 43°C for 15 minutes. The tissue sections were counterstained with 20 #L DAPI/Antifade, and coverslipped. Slides were stored at 4°C in dark, humid chambers. Detection of the hybridization reaction was as described previously for paraffin-embedded tissues.
FISH Evaluation All of the FISH-labeled slides were screened using a standard fluorescent microscope. Screening included evaluation of the quality of the hybridization reaction, and selection of areas of normal epithelium, CIS, and invasive carcinoma for subsequent analysis on a Leica laser confocal microscope (Deerfield, IL). Areas of histological interest were identified by comparison with scanned and printed maps of the corresponding hematoxylin-eosin-stained section. All cases were examined for the presence of cyclin D1 amplification (defined as more than two foci of hybridization per nucleus) and chromosome 11 duplication. At least 50 cells from each tumor were examined, and the pattern of cyclin D1 hybridization compared with adjacent nonneoplastic stromal or lymphoid cells. A431 cells, known to have amplification of the cyclin D1 gene served as positive controls. An esophageal resection specimen from a patient with a nonneoplastic esophageal disease (heterotopic pancreas) served as a negative control. A431 cell metaphases were examined to localize the amplified cyclin D1 genes.
RESULTS Histological Examination The formalin-fixed, paraffin-embedded carcinom a s e x a m i n e d in this study were all s q u a m o u s cell carcin o m a s . O f t h e 20 e s o p h a g e a l t u m o r s , two were well d i f f e r e n t i a t e d , 12 m o d e r a t e l y d i f f e r e n t i a t e d , a n d six poorly differentiated. One of the poorly differentiated c a r c i n o m a s was a s p i n d l e cell c a r c i n o m a . I n a d d i t i o n , o n e e s o p h a g e c t o m y s p e c i m e n f r o m a p a t i e n t with heterotopic pancreas and an inflammatory pseudotumor involving t h e distal e s o p h a g u s was i n c l u d e d in t h e s t u d y as a n o n n e o p l a s t i c c o n t r o l . T h e g r o u p o f 24 a n a l carcin o m a s c o n s i s t e d o f f o u r w e l l - d i f f e r e n t i a t e d , 13 m o d e r ately d i f f e r e n t i a t e d , five p o o r l y d i f f e r e n t i a t e d , a n d two b a s a l o i d s q u a m o u s cell c a r c i n o m a s .
CYCLIN D1 IN ANAL AND ESOPHAGEAL CARCINOMA (Sheyn et al) TABLE 1,
Results of Cyclin D1 Analyses in Paraffin-Embedded Esophageal Carcinomas Cyclin D1 Amplification
Cyclin D1 I H C Case 1 2
3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Histological Grade
Pathological Stage
SCC
NL
SCC
PD MD MD MD-PD MD PD MD WD MD MD WD M-PD MD S p i n d l e cell PD MD MD MD MD MD
ND T4N 1 T4NX T3NX T4N1 T3N1 T4N0 T4NX ND T1NX T2N0 T3N1 T4N0 T2NX ND T4N 1 T3N1 T3N1 T4N1 ND
ND Rare ND Rare Rare + + Rare ND . + . Rare + + + + +
ND ND ND Rare ND Rare
+
-
+ + + + + +
ND ND ND ND ND
+ +
-
+ + + +
ND ND -
.
.
.
.
. Rare ND Rare ND
NL
.
-
ND
A b b r e v i a t i o n s : SCC, s q u a m o u s cell c a r c i n o m a ; CIS, c a r c i n o m a in situ; NL, n o r m a l ; I H C , i m m u n o h i s t o c h e m i s t r y ; WD, well d i f f e r e n t i a t e d ; MD, m o d e r a t e l y d i f f e r e n t i a t e d ; PD, p o o r l y d i f f e r e n t i a t e d ; ND, n o d a t a available.
Immunohistochemistry Immunohistochemical staining with the anti-cyclin D1 antibody was performed in 17 of 20 esophageal carcinomas and all of the 24 anal carcinomas. Three esophageal cases had insufficient tissue for immunohistochemistry after the FISH studies were completed. Strong nuclear cyclin D1 immunoreactivity was observed in 8 of 17 cases examined (Fig 2). In these tumors, the number of immunoreactive neoplastic cells varied from 40% to 75% of the total cell population. The immunoreactive cells were diffusely distributed within the tumor cell nests. An additional five cases contained rare (<5%), diffusely distributed, immunoreactive tumor cells. In most of these cases, the staining for cyclin D1 was weak. One case that showed strong cyclin D1 staining in the invasive carcinoma also showed staining in areas of carcinoma in situ. Carcinoma in situ was not present in any of the other sections that showed strong cyclin D1 staining. In most cases, the nonneoplastic squamous mucosa showed no immunoreactivity with the anti-cyclin antibody. Two cases, however, showed rare positive cells in the basal portion of the epithelium. Only those cases that showed cyclin D1 immunoreactivity in greater than 5% of neoplastic cells were considered to be overexpressing the cyclin D1 protein. The results of the cyclin D1 immunohistochemistry in esophageal carcinomas are summarized in Table 1. Twenty-three of the 24 anal carcinomas were negative for cyclin D1 expression by immunohistochemistry (Fig 3). In two cases, rare weakly immunoreactive cells were present in the basal layer of the nonneoplastic epithelium overlying the invasive carcinoma. In one of these cases, weak staining was also present in rare, diffusely distributed t u m o r cells ( < 5 % ) . None of the cases showed strong staining with the anti-cyclin D1 antibody.
273
Fluorescent In Situ Hybridization Cyclin D1 amplification was consistently observed in A431 cells. In addition, most of these cells appeared to contain four copies of c h r o m o s o m e 11. In many cases, cyclin D1 was observed both on c h r o m o s o m e 11, and in extrachromosomal locations. This was best visualized in the metaphase spread preparations (Fig 4). The results of the FISH in the formalin-fixed, paraffin-embedded esophageal squamous cell carcinomas are summarized in Table 1. Amplification of cyclin D1 was identified in 13 of 20 (65%) cases of esophageal carcinoma (Fig 5). C h r o m o s o m e 11 duplication occurred in 6 of 14 (43%) cases, and was usually associated with amplification of the cyclin D1 gene. Cyclin D1 amplification was also observed in areas of CIS in the two cases in which CIS was present in the section, but was never observed in normal squamous epithelium (Fig 6). In many cases, apparent endoduplication o f cyclin D1 was observed on c h r o m o s o m e 11. As was observed in the A431 cells, m u c h of the amplified cyclin D1 was not localized to c h r o m o s o m e 11 but app e a r e d to be present in scattered foci t h r o u g h o u t the nucleus. Among the esophageal carcinomas that showed immunohistochemical evidence of cyclin D1 overexpression ( > 5 % positive nuclei), all but one also showed evidence of gene amplification by FISH. T h r e e cases with amplified cyclin D1, however, showed either negative or weak staining in rare cells with tile cyclin D1 antibody, suggesting that the cyclin D1 protein was not overexpressed in these tumors despite gene amplification. T h e r e was no correlation between cyclin D1 amplification and either tumor grade or stage. Cyclin D1 amplification was not observed in any of the 24 anal squamous cell carcinomas examined.
HUMAN PATHOLOGY
Volume 28, No. 3 (March 1997)
B
FIGURE 4. FISH using probes for cyclin D1 (green) and chromosome 11 (red). YeLlow dots represent areas of green cyclin D1 signal superimposed on the red chromosome 11 signal. (A) Paraffin-embedded cell block preparation using A431 cells, Note the presence of numerous copies of the cyclin D1 gene in each cell, in some cells the gene is localized to chromosome 11, but is also present in extrachromosomal locations. Each of the cells is aneuploid, containing more than two copies of chromosome 11. (B) Metaphase spread preparation from A431 celts. Note the presence of four copies of chromosome 11, and multiple copies of the cyclin D1 gene localized to both chromosome 11 and extrachromosomal sites.
Y.
J~
~"
k
"L
FIGURE 5. Cyclin [)1 amplification in an invasive squamous cell carcinoma. Each of the neoplastic nuclei shows cyclin D1 amplification, and many cells contain more than two copies of chromosome 11.
FIGURE 6. FISH in nonneopiastic squamous mucosa of the esophagus. Note that each nucleus contains no more than two copies of the cyclin D1 gene, and two copies of chromosome 11.
274
CYCLIN D1 IN ANAL AND ESOPHAGEAL CARCINOMA (Sheyn et al)
DISCUSSION
carcinomas. 42 The mechanisms of carcinogenesis in the cervix and vulva are thought to parallel those in the anal region in that infection with the h u m a n papillomavirus is an important initiating event in all of these sites. Our findings, however, suggest that in contrast to the findings in carcinomas of the lower female genital tract, cyclin D1 alterations do not play a significant role in tumorigenesis in the anal region. The reason that we detected amplification in a higher percentage of esophageal carcinomas cases than has been previously reported is most likely related to a difference in techniques used for evaluation of gene amplification. Previous studies have used DNA transfer and hybridization techniques such as dot, slot, and Southern blotting to detect gene amplification. These techniques require that DNA be extracted from either fixed or fresh whole tissues. These tissues contain not only tumor cells, but variable amounts of normal tissue as well. As a result, contamination of tumor DNA with sufficient DNA from normal cells that lack amplification of the gene in question, may obscure gene amplification. This is especially true in cases in which only a small a m o u n t of tumor is present. The use of in situ techniques such as FISH, however, allows one to evaluate individual cells in their histological context. For this reason, amplification can be easily identified, even when only small numbers of neoplastic cells are present in the specimen. This is especially true when one couples FISH with confocal microscopy to generate high-resolution images of the hybridization product in the cell nucleus. In addition, as shown in the current study, FISH analysis can be readily applied to formalin-fixed, paraffin-embedded tissues. DNA transfer and hybridization techniques such as Southern blotting cannot be performed on such materials, but instead require fresh whole tissues, which are not always readily available. For this reason, FISH analysis is well suited for use in retrospective studies using archival tissues. Finally, in situ techniques allow one to separately characterize areas of normal epithelium and stroma, areas of dysplasia and carcinoma in situ, and areas of invasive carcinoma. With the use of FISH, we were able to show the presence of cyclin D1 amplification within foci of CIS in two esophageal cancer patients. In addition, the use of dual labeling and hybridization with multiple probes allows one to localize the gene to its specific chromosome. In the current study, we were able to observe that the amplified cyclin D1 was not always present in its normal location on c h r o m o s o m e 11. In many cases, the amplified DNA appeared to be extrachromosomal, perhaps within double minutes. Such observations in situ may eventually allow one to make predictions regarding the role of cyclin D1 in carcinogenesis.
Amplification of the 11q13 region occurs commonly in squamous cell carcinomas arising in the lung, head and neck, as well as the esophagus. This region is known to contain an amplicon that includes the genes int-2, hst-1 and PRAD1/cyclin D1. The int-2 and hst-1 genes are known to be members of the fibroblast growth factor gene family, and have been implicated as proto-oncogenes in the genesis of mouse mammary carcinomas. ~1s4 Cyclin D1 plays a key role in the regulation of the cell cycle s5 and therefore is considered an oncogene. Studies of h u m a n breast carcinomas have shown that despite the frequent amplification of int-2 and hst-1, these genes are not overexpressed, suggesting that they do not1 p 7lay an imp ortant role in carcino genesis in this site. 4,3
In esophageal carcinomas, the reported frequency of cyclin D1 amplification ranges from 24% to 54% of c a s e s . 19'20'22'23'40 Similar results have been reported in laryngeal carcinomas. 2 We were unable to find any data regarding cyclin D1 alterations in anal carcinomas in the English literature. In a study of cervical and vulvar squamous neoplasia, cyclin D1 amplification was frequent, occurring in three of three vulvar carcinoma cell lines and in 4 of 10 cell lines derived from cervical
REFERENCES 1. Ali IU, Merlo G, Callahan R, et al: The amplification unit on chromosome llq13 in aggressive primary human breast tumors entails the bcl-1, int-2 and hst loci. Oncogene 4:89-92, 1989 2. Jares P, Fernandez PL, Campo E, et al: PRAD-1/cyclin D1
275
HUMAN PATHOLOGY
Volume 28, No. 3 (March 1997)
gene amplification correlates with messenger RNA overexpression and tumor progression in human laryngeal carcinomas. Cancer Res 54:4813-4817, 1994 3. Callender T, Elnaggar AK, Lee MS, et al: PRAD-1 (CCND1) cyclin D1 oncogene amplification in primary head and neck squamous cell carcinoma. Cancer 74:152-158, 1994 4. Gillett C, Fantl V, Smith R, et al: Amplification and overexpression of cyclin D1 in breast cancer detected by immunohistochelnical staining. Cancer Res 54:1812-1817, 1994 5. Zhou DJ, Casey G, Cline MJ: Amplification of human int-2 gene in breast cancers and squamous carcinomas. Oncogene 2:279282, 1988 6. Berenson JR, Koga H, Yang J, et al: Frequent amplification of the bcl-1 locus in poorly differentiated squamous cell carcinoma of the lung. Oncogene 5:1343-1348, 1990 7. Somers K, Cartwright S, Schechter G: Amplification of the int-2 gene in human head and neck squamous cell carcinomas. Oncogene 5:915-920, 1990 8. Leonard JH, Kearsley JH, Chenevix-Trench G, et al: Analysis of gene amplification in head and neck squamous cell carcinomas. IntJ Cancer 48:511-515, 1991 9. BerensonJR, YangJ, Mickel R: Frequent amplification of the bcl-1 locus in head and neck squamous cell carcinomas. Oncogene 4:1111-1116, 1989 10. Lidereau R, Callahan R, Dickson C, et al: Amplification of the int-2 gene in primary human breast tumors. Oncogene Res 2:285291, 1988 11. VarleyJM, Walker RA, Casey G, et al: A common alteration to the int-2 protooncogene in DNA from primary breast carcinomas. Oncogene 3:87-91, 1988 12. Theillet C, AdnaneJ, Szepetowski P, et al: BCL-1 participates in the llq13 amplification found in breast cancer. Oncogene 5:147149, 1990 13. Tsuda H, Hirohashi S, Shimosato 5(, et al: Correlation between long-term survival in breast cancer patients and amplification of two putative oncogene-coamplification units: hst-1/int-2 and c-erbB2/ear-l. Cancer Res 49:3104-3108, 1989 14. Fantl V, Richards MA, Smith R, et al: Gene mnplification on chromosome band llq13 and oestrogen receptor status in breast cancer. EurJ Cancer 26:423-429, 1990 15. Borg A, Sigurdsson H, Clark GM, et al: Association of INT2/ HST1 coamplifieation in primary breast cancer with hormone-dependent phenotype and poor prognosis. BrJ Cancer 63:136-142, 1991 16. Saint-Ruf C, Malfoy B, Scholl S, et al: Gst~r gene is frequently co-amplified with INT2 and HSTF1 proto-oncogenes in human breast cancers. Oncogene 6:403-406, 1991 17. Schuuring E, Verhoeven E, Mooi WJ, et al: Identification and cloning of two overexpressed genes U21B31/PRAD1 and EMS1, within the amplified chromosome llq13 region in human carcinomas. Oncogene 7:355-361, 1992 18. Speicher MR, Howe C, Crotty P, et al: Comparative genomic hybridization detects novel deletions and amplifications in head and neck squamous cell carcinomas. Cancer Res 55:1010-1013, 1995 19. Wagata T, Ishizald K, Imamura M, et al: Deletion of 17p and amplification of the int-2 gene in esophageal carcinomas. Cancer Res 51:2113-2117, 1991 20. Adelaide J, Monges G, Derderian C, et al: Oesophageal cancer and amplification of the human cyclin D gene CCND1/PRAD1. BrJ Cancer 71:64-68, 1994 21. Jaing W, Zhang Y-J, ILahn SM, et al: Altered expression of the cyclin D1 and retinoblastoma genes in human esophageal cancer. Proc Natl Acad Sci U S A 90:9026-9030, 1993 22. Mori M, Tokino T, Yanagisawa A, et al: Association between
276
chromosome llq13 amplification and prognosis of patients with oesophageal carcinomas. EurJ Cancer 28A:755-757, 1992 23. KitagawaY, Ueda M, Ando N, et al: Significance of int-2/hst1 coamplification as a prognostic factor in patients with esophageal squamous carcinoma. Cancer Res 51:1504-1508, 1991 24. Gramlieh TL, Fritsch CR, Maurer D, et al: Differential polymerase chain reaction assay of cyclin D1 gene amplification in esophageal carcinoma. Diagn Mol Pathol 3:255-259, 1994 25. Zukerberg LB, Yang W-I, Gadd M, et al: Cyclin D1 (PRAD1) protein expression in breast cancer: Approximately one-third of infiltrating mammary carcinomas show overexpression of the cyclin D1 oncogene. Mod Pathol 8:560-567, 1995 26. Buckley MF, Sweeney KJE, Hamilton JA, et al: Expression and amplification of cyclin genes in human breast cancer. Oncogene 8:2127-2133, 1993 27. Motokura T, Bloom T, Kim HG, et al: A novel cyclin encoded by a bc/l-linked candidate oncogene. Nature 350:512-515, 1991 28. Arnold A, Kim HG, Gaz RD, et al: Molecular cloning and chromosomal mapping of DNA rearranged with the parathyroid hormone gene in a parathyroid adenoma. J Clin Invest 83:2034-2040, 1989 29. Tsujimoto Y, YunisJ, Onorato-Showe L, et al: Molecular cloning of the chromosomal breakpoint of B-cell lymphomas and leukemias with the t ( 11; 14) chromosome translocation. Science 224:14031406, 1985 30. Freshney RI: Culture of Animal Cells: A Manual of Basic Technique. New York, NY, Wiley-Liss, 1987, pp 169-185 31. Dickson C, Peters G: Potential oncogene product related to growth factors. Nature 326:823, 1987 (letter) 32. Casey G, Smith R, McGillivray D, et al: Characterization and chromosome assignment of human homolog of int-2, a potential proto-oncogene. Mol Cell Biol 6:502-510, 1986 33. Yoshida T, Miyagawa K, Odagiri H, et al: Genomic sequence of hst, a transforming gene encoding a protein homologous to fibroblast growth factor and the int-2 encoded protein. Proc Natl Acad Sci U S A 84:7305-7309, 1987 34. Peters G, Brookes S, Smith R, et al: The mouse homolog of the hst/K-FGF gene is adjacent to int-2 and is activated by proviral insertion in some virally induced mammary tumors. Proc Natl Acad Sci U S A 86:5678-5682, 1989 35. Scherr CJ: Mammalian G1 cyclins. Cell 73:1059-1065, 1993 36. Theillet C, Roy XL, Lapeyriere OD, et al: Amplification of FGF-related genes in human tumors: Possible involvement of HST in breast carcinomas. Oncogene 4:915-922, 1989 37. Liscia DS, Merlo GR, Garrett C, et al: Expression of int-2 mRNA in human tumors amplified at the int-2 locus. Oncogene 4:1219-1224, 1989 38. Laramie GA, Fantl V, Smith R, et al: DllS287, a putative oncogene on chromosome 1l q l 3, is amplified and expressed in squamous cell and mammary carcinomas and linked to BCL-1. Oncogene 6:439-444, 1991 39. BartkovaJ, LukaJ, Muller H, et al: Cyclin D1 protein expression and function in human breast cancer. IntJ Cancer 57:353-361, 1994 40. Jiang W, Kahn SM, Tomita N, et al: Amplification and expression of the human cyclin D gene in esophageal cancer. Cancer Res 52:2980-2983, 1992 41. Michalides R, van Veelen N, Hart A, et al: Overexpression of cyclin D1 correlates with recurrence in a group of forty-seven operable squamous cell carcinomas of the head and neck. Cancer Res 35:975-978, 1995 42. Kuzrock R, Ku S, Talpaz M: Abnormalities in the PRAD1 (cyclin D1/bcl-1) oncogene are frequent in cervical and vulvar squamous cell carcinoma cell lines. Cancer 75:584-590, 1995