453 Loss of Expression of the Gene Deleted in Colon Carcinoma (DCC) Is Closely Related to Histologic Differentiation and Lymph Node Metastasis in Endometrial Carcinoma Makoto Saegusa, M.D. Miki Hashimura, B.Ph. Atsuko Hara, M.D. Isao Okayasu, M.D. Department of Pathology, Kitasato University School of Medicine, Kanagawa, Japan. BACKGROUND. Although frequent loss of the tumor suppressor gene deleted in colon carcinoma (DCC) has been demonstrated in endometrial carcinoma, an alteration of the expression during normal menstrual cycle and tumorigenesis from hyperplastic lesions is still unclear. METHODS. A total of 151 endometrial carcinomas (endometrioid type), along with 90 hyperplasias (23 simple, 30 complex, and 37 atypical) and 143 normal endometria (28 atrophic, 44 proliferative, and 71 secretory), were immunohistochemically investigated for expression of DCC as well as for estrogen and progesterone receptors (ER and PR). Analysis for DCC mRNA levels was also performed on 37 endometrial carcinomas and 14 normal endometria. RESULTS. DCC expression was observed in endometrial glandular cells in both proliferative and secretory stages; the immunoreactivity scores were not related to values for either ER or PR. The values for DCC were significantly higher in hyperplasia than in normal endometria, and then decreased in the sequence leading to Grade 3 carcinoma. In endometrial carcinoma, reduction or loss of DCC expression was significantly related to the histologic evidence of malignancy and lymph node metastasis, and this was in keeping with the results of mRNA analysis. The transcripts derived from alternative splicing in the extracellular domain were not observed in any tumor samples. CONCLUSIONS. The findings of this study indicate that DCC expression may be linked to the maintenance of differentiated glandular cells during the normal menstrual cycle without any relation to immunoreactivity for ovarian hormone receptors. Moreover, loss or reduction of expression may be a significant event in the progression of endometrial carcinoma through metastatic features. Cancer 1999;85:453– 64. © 1999 American Cancer Society. KEYWORDS: endometrial carcinoma, endometrial hyperplasia, gene deleted in colon carcinoma (DCC), estrogen receptor, progesterone receptor. T Address for reprints: Makoto Saegusa, M.D., Department of Pathology, Kitasato University School of Medicine, 1-15-1 Kitasato, Sagamihara, Kanagawa 228-8555, Japan. Received March 17, 1998; revision received June 22, 1998; accepted June 22, 1998. © 1999 American Cancer Society he gene DCC (deleted in colorectal carcinoma) has been identified at chromosome 18q21, a common region of allelic loss in human colorectal carcinomas.1 Genomic analysis has revealed that the open reading frame of this gene consists of 29 exons spanning nearly 1.4 megabases, and that it encodes a 1447–amino acid transmembrane protein with four immunoglobulinlike and six fibronectin type III–like extracellular domains.2,3 DCC transcripts and protein are at low levels in most tissues but are abundant in the central and peripheral nervous system, where they may play important roles in cell growth and differentiation.3 454 CANCER January 15, 1999 / Volume 85 / Number 2 FIGURE 1. Immunoreactivity of the gene deleted in colon carcinoma (DCC) in normal and hyperplastic endometria is depicted. Semiserial sections of normal endometrium are shown in proliferative (A,F) and secretory (B,G) stages and hyperplasias with complex (C,H), simple (D,I), and atypical (E,J) types. (A–E) H&E staining; (F–J) DCC immunohistochemistry. (F,G) Moderate immunoreactivity for DCC is observed in endometrial glandular cells. High magnification (⫻200) of areas indicated by arrows is illustrated in the insets. (H,I,J) Note the strong cytoplasmic binding in hyperplastic glandular components. (Original magnification ⫻100.) The DCC gene has been considered as a candidate tumor suppressor gene for a variety of human malignancies. Reduction of transcripts or allelic deletions of this gene are frequently observed in prostate, breast, esophageal, and brain tumors.4 –7 Moreover, it has been demonstrated that full length, but not truncated, DCC can suppress the tumorigenicity of a human squamous carcinoma cell line.8 Endometrial carcinoma is a common malignancy in the female reproductive tract, and its development is closely related to change in ovarian hormone status. Although frequent reduction or loss of DCC has also been demonstrated in endometrial carcinomas,9,10 any association with estrogen and progesterone receptors is still unclear. In addition, to our knowledge there has been no report regarding DCC expression during the normal menstrual cycle and in hyperplastic lesions. In this study, to clarify the significance of DCC gene abnormalities to the development and progression of endometrial carcinomas, we therefore investigated normal, hyperplastic, and malignant endometria, using immunohistochemistry and a combination of the reverse transcriptase–polymerase chain reaction and the Southern blot hybridization (RT-PCR/ SBH) assays to determine expression levels. The results were compared with hormone receptor status and a number of other prognostic factors, such as clinical stage, degree of myometrial invasion, and lymph node metastasis. In addition, it was examined whether DCC inactivation is related to either loss of heterozygosity or aberrant alternative splicing of mRNA. MATERIALS AND METHODS Cases A total of 151 cases of endometrial carcinoma (endometrioid-type), surgically resected at Kitasato University Hospital during the period 1988 –1997, were investigated. Ninety cases of hyperplasia (including 23 simple and 30 complex hyperplasias without atypia, and 37 atypical hyperplasias), obtained by curettage or hysterectomy, were also examined, along with 28 samples of atrophic (postmenopausal) endometrium adjacent to carcinoma samples and 115 normal biopsy DCC Expression in Endometrial Carcinoma/Saegusa et al. 455 FIGURE 2. Immunoreactivity scores for the gene deleted in colon carcinoma (DCC), estrogen receptor (ER), and progesterone receptor (PR) are given for normal (A) and hyperplastic (B) endometria. AE: atrophic endometrium; Pro: proliferative phase; Sec: secretory phase; SH: simple hyperplasia; CH: complex hyperplasia; AH: atypical hyperplasia. The data are mean ⫾ standard deviation values. specimens (44 proliferative and 71 secretory). All tissues were routinely fixed in 10% formalin and processed for embedding in paraffin wax. Histologic diagnoses were made according to the criteria of the International Federation of Gynecology and Obstetrics (1988).11 The carcinoma cases comprised 89 Grade (G) 1, 39 G2, and 23 G3 lesions. Of all carcinoma cases investigated, 143 tumors were available for examination of the correlation between immunohistochemical parameters and clinicopathologic data, including clinical stage, degree of myometrial invasion, and lymph node status. Areas of squamous differentiation within tumors, including squamous metaplasia (Sq-M) and morules, were classified after Zaino et al.12 Of 151 carcinomas, 20 cases with Sq-M foci and 16 tumors with morule lesions were also investigated for whether DCC immunoreactivity was altered in Sq-D areas. In addition, 37 endometrioid carcinomas and 14 normal endometrial samples were further snap-frozen in liquid nitrogen for RNA analysis. Immunohistochemistry Immunohistochemistry was performed using a combination of the microwave oven heating and the standard streptavidin-biotin-peroxidase complex (LSAB kit, Dako, Copenhagen, Denmark) methods. Immunohistochemical detection of DCC protein was per- formed as described by Shibata et al.13 Briefly, slides were heated in 10 mM citrate buffer (pH 6.0) for six 5-minute cycles for DCC and three 5-minute cycles for estrogen and progesterone receptors (ER and PR), using a microwave oven, and were then incubated overnight at 4°C with optimum dilutions of primary antibodies. The antibodies used were antihuman DCC monoclonal antibody (clone G97-449, ⫻100 dilution, Pharmingen, San Diego, CA), anti-ER mouse monoclonal antibody (⫻80 dilution, Novocastra Lab. Ltd., Newcastle, UK), and anti-PR mouse monoclonal antibody (⫻80 dilution, Novocastra Lab. Ltd., Newcastle, UK). To confirm the specificity of binding, normal mouse serum (⫻500 dilution) was supplied instead of primary antibody as a negative control. Scoring for DCC and Immunoreactivity of Hormone Receptors Scoring of the immunohistochemistry results was done as previously reported.14 Briefly, based on the percentages of immunopositive epithelial cells among the totals of normal or neoplastic cells on a slide section, there were 5 categories of subdivision, as follows: 0, all negative; 1, ⬍10% positive cells; 2, 10 –30%; 3, 30 –50%; and 4, ⬎50%. The immunointensity was also subclassified into 4 groups, as follows: 0, negative; 1⫹, weak; 2⫹, moderate; and 3⫹, 456 CANCER January 15, 1999 / Volume 85 / Number 2 FIGURE 3. Immunoreactivity of the gene deleted in colon carcinoma (DCC) is depicted in endometrial carcinomas. (A–C) Semiserial sections of Grade (G) 1 (A,D), G2 (B,E), and G3 (C,F) endometrioid carcinomas are shown. (A–C) H&E staining. (D–F) Note the strong (D), moderate (E), and negative (F) immunoreactivity. (Original magnification ⫻200.) strong. Immunoreactivity scores were generated by multiplication of the values for the two parameters. The scoring was performed by one of us (M.S.), the evaluation being carried out as triplicate examinations. Reverse Transcriptase–Polymerase Chain Reaction Total cellular RNA was extracted from 37 tumors and 14 normal endometrial frozen tissues using Isogen (Nippon Gene Co., Tokyo, Japan). cDNA was synthesized from 5 g of total RNA using RAV-2 reverse transcriptase (Takara, Shiga, Japan) in the presence of random primers (Takara) and a ribonuclease inhibitor (Takara) in a 20 L reaction volume at 42°C for 60 minutes. One L of cDNA solution was amplified by Taq polymerase (Takara) in a volume of 10 L. For detection of DCC mRNA expression in exons 6 –7, primers used were 5⬘-TTCCGCCATGGTTTTTAAATCA-3⬘ for the sense and 5⬘-AGCCTCATTTTCAGCCACACA-3⬘ for the antisense, published by Fearon et al.1 For analysis of alternative splicing in the extracellular domain (exon 17), primers used were DCK2834S, 5⬘-CCCAGACTAACTGCATCATCATGAG-3⬘ (sense), and DCK3151A, 5⬘-CACCTACTGGTGGGAGCAT-3⬘ (antisense), described by Reale et al.15 The PCR procedure was performed with 35 cycles of denaturation at 94°C for 0.5 minutes, annealing at 55°C for 1 minute, and extension at 72°C for 1 minute, with a final extension time of 7 minutes. As a negative control, water was supplied instead of template DNA for each examination. To examine the quality and quantity of the synthesized cDNA, the ␤-actin specific primers (sense, 5⬘TGCTATCCAGGCTGTGCTAT-3⬘ and antisense, 5⬘GATGGAGTTGAAGGTAGTTT-3⬘) were also ampli- DCC Expression in Endometrial Carcinoma/Saegusa et al. 457 FIGURE 4. (A) Immunoreactivity scores for the gene deleted in colon carcinoma (DCC), estrogen receptor (ER), and progesterone receptor (PR) are given for normal and hyperplastic endometria and endometrial carcinomas. N: normal endometrium, including atrophic, proliferative, and secretory stages; H: hyperplasia; G: grade. The data are mean ⫾ standard deviation (SD) values. (B) Immunoreactivity scores for primary carcinomas (Ca) and areas of squamous differentiation, including squamous metaplasia (SqM) and morules (M), are shown. The data are mean ⫾ SD values. fied.16 PCR assays were performed in duplicate or triplicate. Southern Blot Hybridization A 10 L aliquot of each PCR reaction mixture was electrophoresed in a 3.0% agarose gel and transferred to a Hybond N nylon membrane (Amersham, Tokyo, Japan). After prehybridization using DIG Easy Hyb solution (Boehringer Mannheim, Tokyo, Japan), filters were hybridized overnight with each digoxigenin-labeled exon-specific probe, which corresponded to internal sites between primer sets used. The sequences of oligonucleotide probes for DCC exon 6 –7, DCC exon 17, and ␤-actin were as follows: probe DCC exon 6/7 (5⬘-AATTGGATGAAGAATGGAGATGTGGTCATT3⬘, encoding nucleotides 1087–1116 in the cDNA sequence), probe DCC exon 17 (5⬘-ATGAGTTGGACTCCTCCCTTGAAC-3⬘, nucleotides 2226 –2250), and probe ␤-actin (5⬘-ACTGACTACCTCATGAAGATCCTCACCGAG-3⬘, nucleotides 597– 626). Hybridization signals were detected with a DIG Luminescent Detection Kit (Boehringer Mannheim). The conditions used for hybridization, washing, and detection were in keeping with the manufacturer’s recommendations. Between each hybridization, the filter was stripped before being rehybridized with another probe. RT-PCR/SBH Data Analysis Quantitation of hybridization signals for DCC exon 6/7 and ␤-actin, respectively, was performed by densitometric analysis using NIH Image version 1.58 software. The relative expression level of DCC mRNA was calculated by normalization to the hybridization signals for ␤-actin in each case as follows: value for DCC mRNA signal, divided by that for ␤-actin. Based on the average values for relative DCC transcripts in normal endometria, subdivision was into 4 categories, as follows: negative, ⬍10% of normal; ⫾, 10 –50%; 1⫹, 50 – 100%; 2⫹, ⬎100%, according to the methods described by Enomoto et al., with minor modification.9 A 458 CANCER January 15, 1999 / Volume 85 / Number 2 FIGURE 5. The relation between gene deleted in colon carcinoma (DCC) immunoreactivity scores and clinicopathologic factors for endometrial carcinomas is shown. U: upper half; L: lower half; P: positive; N: negative. The data are mean ⫾ standard deviation values. loss or reduction of DCC expression was concluded with scores of either negative or ⫾, and a positive one with 1⫹ or 2⫹. Detection of alternative splicing within DCC exon 17 was performed as described by Reale et al.15 Briefly, DCK2834S and DCK3151A primers generated a product of 341 bp if the more 5⬘ splice acceptor at exon 17 was used (60-nucleotide sequence present) or an alternatively spliced product of 281 bp if the more 3⬘ splice acceptor site was used. Detection of Loss of Heterozygosity (LOH) by PCR Assay Genomic DNAs were extracted using GenomicPrepTM cells and tissue DNA isolation kit (Pharmacia Biotech, Tokyo, Japan). Aliquots of 100 ng DNA were used as templates in a reaction volume of 10 L. For detection of LOH at the DCC locus, a PCR-LOH assay was performed using a primer set recognizing a variable number of tandem repeats (VNTR) within the DCC gene. The sequences of primers were 5⬘-GATGACATTTTCCCTCTAG-3⬘ (sense) and 5⬘-GTGGTTATTGCCTTGAAAAG-3⬘ (antisense),6 and the PCR procedure was carried out with 32 cycles of 0.5 minutes at 94°C, 1 minute at 57°C, and 1 minute at 72°C, along with a predenature step of 2 minutes at 94°C and postextension for 5 minutes at 72°C. The PCR products were electrophoresed on 3% agarose gels and then stained by ethidium bromide. LOH was defined as a more than a 50% decrease in relative intensity of alleles on comparison of amplified alleles between normal and tumor DNAs, as described previously.17 Statistics Statistical analysis of data for immunoreactivity scores was performed using the Mann–Whitney U test and the Pearson’s correlation coefficient. Association of RT-PCR data for DCC transcripts between exon 6/7 and 17 regions were examined by the chi-square test. The cutoff for statistical significance was defined as P ⬍ 0.05. RESULTS DCC Expression in Normal and Hyperplastic Endometria Weak-to-moderate cytoplasmic immunoreactivity for DCC was found in the endometrial glandular cells. The positivity and intensity appeared to be similar in DCC Expression in Endometrial Carcinoma/Saegusa et al. 459 FIGURE 6. Serial sections of morule (A,C) and squamous metaplastic (B,D) areas are shown. (A,B) H&E staining. Immunoreactivity of the gene deleted in colon carcinoma (DCC) is strong in a morule (C) and weak in a squamous metaplastic focus (D). (Original magnification ⫻200.) both proliferative and secretory phases. In some cases, stromal cells also exhibited weak binding, whereas other endometrial components, including myometrium and vessels, lacked immunoreactivity (Fig. 1A,B,F,G). In atrophic (postmenopausal) endometria, DCC positive glandular cells were observed with relatively weak immunointensity. Moderate-to-strong immunoreactivity for DCC was observed in simple or complex hyperplastic lesions, the positive cells being distributed homogeneously with cytoplasmic staining. The immunostaining pattern of atypical hyperplasia was the same. Weak immunoreactivity in stromal cells was also noted in some cases (Fig. 1C,D,E,H,I,J). Antibodies against ER and PR diffusely reacted with several endometrial components, including glandular, stromal, and myometrial cells, with distinct nuclear staining. No alteration of average values for DCC immuno- reactivity scores were observed during normal menstrual cycles, whereas marked reduction was noted for both ER and PR with transition from the proliferative to the secretory phase. The average immunoreactivity scores for atrophic endometria were significantly lower than in either proliferative or secretory values (Fig. 2A). In addition, the average values for DCC were not significantly different among simple, complex, and atypical hyperplasias (Fig. 2B). DCC Expression in Endometrial Carcinomas A variety of immunostaining patterns for DCC were observed in endometrial carcinomas. In some cases, DCC positive tumor cells were diffusely distributed with strong cytoplasmic staining and some membrane binding, whereas completely lacking or sporadic weak-to-moderate positivity were observed in other tumors (Fig. 3). 460 CANCER January 15, 1999 / Volume 85 / Number 2 TABLE 1 Correlation of Immunoreactivity Scores among DCC, ER, and PR Em Ca G1 G2 G3 Hyperplasia Simple Complex Atypical Normal epithelium Atrophic P-phase S-phase n DCC vs. ER r (P value) DCC vs. PR r (P value) ER vs. PR r (P value) 89 39 23 0.32 (0.002) 0.39 (0.015) 0.42 (0.05) 0.13 (0.2) 0.33 (0.04) 0.28 (0.2) 0.41 (⬍0.0001) 0.13 (0.45) 0.39 (0.06) 23 30 37 0.06 (0.79) 0.27 (0.16) 0.02 (0.89) 0.02 (0.92) 0.24 (0.21) 0.03 (0.85) 0.17 (0.4) 0.72 (⬍0.0001) 0.19 (0.26) 28 44 71 0.08 (0.69) 0.004 (0.98) 0.26 (0.03) 0.44 (0.02) 0.26 (0.09) 0.09 (0.45) 0.08 (0.69) 0.32 (0.03) 0.86 (⬍0.0001) DCC: gene deleted in colon carcinoma; ER: estrogen receptor; PR: progesterone receptor; n: no. of cases; r: Pearson’s correlation coefficient; Em Ca: endometrial carcinoma; G: grade; P-phase: proliferative phase; S-phase: secretory phase. The average values for DCC scores decreased in the sequence leading from hyperplasias to Grade 3 carcinomas, the difference being significant, as well as ER and PR status, whereas the scores in hyperplastic lesions were significantly higher than in normal endometrium at any stage (Fig. 4A). As shown in Figure 5, the DCC scores in the negative lymph node metastasis group were significantly higher than in the positive category, but there was no correlation between the scores and either the clinical stage or the degree of myometrial invasion. The DCC antibody also bound to areas of squamous differentiation within tumors, showing strong immunoreactivity in morules and weak or negative staining in squamous metaplastic (Sq-M) lesions (Fig. 6). Average immunoreactivity scores in morules were significantly higher than in carcinomatous areas, in contrast to the Sq-M lesions (Fig. 4B). No correlation was evident between values for DCC and either ER or PR in normal, hyperplastic, and malignant endometrial lesions (Table 1). transcripts and the immunoreactivity scores was observed for endometrial carcinomas (Fig. 8). As shown in Table 2, reduction (⫾) or loss (⫺) of DCC expression was observed in 13 (35.1%) of 37 carcinomas, when subdivision was into 4 categories on the basis of the average values for normal endometria. With primer sets for alternative splicing sites within DCC exon 17, only a 341 bp and not a 281 bp fragment was amplified in 26 (70.3%) of 37 carcinomas (Fig. 7 and Table 2), the positivity being significantly related with that for DCC exon 6/7 transcripts (Table 3). Loss of Heterozygosity of DCC To determine whether alteration of DCC mRNA expression is related to DNA abnormalities, the PCRLOH assay was performed. With a primer set for the VNTR region in the DCC locus, although 11 of 31 (35.5%) endometrial carcinomas were informative, no LOH was identified (Table 2). DISCUSSION RT-PCR/SBH Assay Detection of mRNAs for DCC and ␤-actin genes was performed using a combination of RT-PCR and SBH assays. RNAs obtained from 14 normal endometria and 37 endometrial carcinomas could all be successfully amplified by ␤-actin specific primer sets, the products showing the expected molecular weight of 446 bp (Fig. 7). With primer sets corresponding to DCC exons 6 to 7, amplicons were observed with a molecular weight of 233 bp (Fig. 7). Positive correlation (r ⫽ 0.567, P ⫽ 0.0007) between the relative amounts of DCC exon 6/7 The current study immunohistochemically demonstrated that DCC expression is maintained from proliferative though to secretory stages totally, the immunoreactivity scores being not statistically related to the ovarian hormone receptor status. Hedrich et al.3 proposed that, rather than physically holding cells together, DCC may be important in transmitting signals from the cell’s exterior environment to its interior, thereby playing a role in controlling differentiation. Our findings therefore suggest that DCC may play some role in the differentiative pathway of endometrial glandular cells through proliferative to secretory DCC Expression in Endometrial Carcinoma/Saegusa et al. 461 FIGURE 7. A reverse transcriptase– polymerase chain reaction/Southern blot hybridization assay, for the gene deleted in colon carcinoma (DCC) and ␤-actin in endometrial carcinomas, is depicted. FIGURE 8. The correlation between gene deleted in colon carcinoma (DCC) immunoreactivity scores and relative amounts of DCC mRNA in endometrial carcinomas is shown. phases without any relation to the expression of ovarian hormone receptors. The relative down-regulation of DCC expression observed for atrophic endometrium may be related to the associated reduction in their proliferative activity. Kurman et al.18 asserted that the majority of hyperplasias, including simple, complex, and atypical types, regress or remain stable rather than progressing. In this study, DCC expression was increased rather than decreased in hyperplastic lesions com- pared with normal endometrium, independent of the histologic subtype. With regard to colorectal lesions, DCC expression appears to be similarly increased in hyperplastic polyps, with retention in early and intermediate stage adenomas but loss in some late adenomas with significant dysplasia.3 Goi et al.19 also reported that normal colonic and adenoma tissues significantly expressed the DCC protein, in contrast to colonic carcinomas showing poor expression. Our results indicate that loss of DCC expression is rare event 462 CANCER January 15, 1999 / Volume 85 / Number 2 TABLE 2 Data on the Relation between DCC mRNA Expression and Immunoreactivity in Endometrial Carcinomas DCC exon 6/7 Case no. Grade C1 G1 C2 G2 C3 G2 C4 G1 C5 G1 C6 G1 C7 G3 C8 G1 C9 G1 C10 G1 C11 G1 C12 G2 C13 G1 C14 G1 C15 G3 C16 G2 C17 G1 C18 G1 C19 G1 C20 G1 C21 G2 C22 G1 C23 G1 C24 G2 C25 G2 C27 G1 C29 G1 C31 G2 C32 G2 C35 G2 C36 G1 C38 G2 C39 G2 C40 G3 C41 G1 C42 G1 C43 G1 Normal (n ⫽ 14) DCC exon 17 Stage MI LN Ratio mRNA* 281 bp 341 bp IHC score for DCC DCC LOH (VNTR) I I III I I I I II I I I I I IV II I III II II I I I I I III III II II II II II III II I I I I U L L L U L U L L U U L L U L L U U U U U U U U L L L L U L L U U L U L L N N P N N N N P N N N N N P N N N N N N N N N N N P N N N N NE NE N P N N P 0.95 0.01 0.54 0.72 0.98 0.9 0.9 1.4 0.96 0.45 0.12 0.51 0.48 0.93 0.86 0.97 1.15 0.23 0 0.16 0.7 0.36 1.6 1.31 1.12 0.27 1.15 0.31 1.29 0.95 0 0.41 0.72 0.28 1.56 1.48 0.75 0.97 ⫾ 0.38 1⫹ ⫺ 1⫹ 1⫹ 2⫹ 1⫹ 1⫹ 2⫹ 1⫹ ⫾ ⫾ 1⫹ ⫾ 1⫹ 1⫹ 1⫹ 2⫹ ⫾ ⫺ ⫾ 1⫹ ⫾ 2⫹ 2⫹ 2⫹ ⫾ 2⫹ ⫾ 2⫹ 1⫹ ⫺ ⫾ ⫹ ⫾ 2⫹ 2⫹ ⫹ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫹ ⫺ ⫹ ⫹ ⫺ ⫹ ⫹ ⫹ ⫹ ⫹ ⫺ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫺ ⫺ ⫺ ⫹ ⫺ ⫹ ⫹ ⫹ ⫹ ⫹ ⫺ ⫹ ⫹ ⫺ ⫺ ⫹ ⫺ ⫹ ⫹ ⫹ ⫹ ND 4 ND 4 6 8 8 12 12 9 4 3 4 6 8 9 4 4 2 4 2 2 ND 1 4 4 8 4 8 6 2 ND ND 2 8 8 6 6.0 ⫾ 3.1 N H H H H H H N N H H H N N N H H H H N N H H H H N H N H N H ND ND ND ND ND ND DCC: deleted in colon carcinoma gene; MI: degree of tumor myometrial invasion; U: upper half of the myometrium; L: lower half of the myometrium; LN: lymph node status; N: negative; P: positive; NE: not examined; *: relative abundance of DCC mRNA; (⫺: ⬍10% of average value for normal endometrium; ⫾: 10–50%; 1⫹: 50–100%; 2⫹: ⬎100%); IHC: immunohistochemistry; ND: not done; LOH: loss of heterozygosity; VNTR: variable number of tandem repeats; N: negative; H: homozygous (not informative). in hyperplasias, suggesting that up-regulation of the expression in these lesions may be associated with accelerated proliferation and differentiative features. Frequent reduction or loss of DCC expression has been documented in a variety of human malignancies, such as colorectal, pancreatic, and prostatic adenocarcinomas.4,20,21 In the colorectal case, DCC expression is markedly reduced during tumor progression from intramucosal to invasive tumors with loss detected in 100% of tumors with liver metastasis, suggesting that DCC inactivation is a late event in tumor progression and that the molecule may act as a suppressor of metastasis.21,22 In tumors of the female reproductive tract, Enomoto et al.9 demonstrated impaired DCC expression in 4 (50%) of 8 endometrial and 12 (55%) of 22 ovarian epithelial tumors, particularly tumors with higher histologic grade, suggesting an association with an aggressive phenotype. Gima et al.10 described loss of DCC expression in 14 (50%) of 28 endometrial carcinomas but did not correlate with histologic differentiation and progression (clinical stage). In our large series, reduction or loss of DCC protein was signifi- DCC Expression in Endometrial Carcinoma/Saegusa et al. 463 TABLE 3 Relation between Exon 6/7 and Exon 17 DCC mRNA Expression in Endometrial Carcinomas DCC exon 17 expression DCC exon 6/7 expression Positive Negative Positive Negative P value 21 (56.8) 5 (13.5) 3 (8.1) 8 (21.6) 0.0019 DCC: gene deleted in colon carcinoma. cantly related to the histologic malignancy and lymph node metastasis in endometrial carcinomas, in line with the results of mRNA assay. Considering no association with FIGO stage or degree of myometrial invasion, the alteration of expression may be closely linked with the tumor progression through metastatic potential. Further studies in this area are clearly warranted. In most human malignancies, inactivation of a tumor suppressor gene commonly appears to be due to a point mutation of one allele and loss of the remaining allele. For DCC inactivation, aberrant splicing and allele-specific loss of transcripts have also been proposed to be important mechanisms.15,23 Reale et al.15 have identified two alternatively spliced regions in the DCC gene, involving extracellular (exon 17) and cytoplasmic (exon 26) domain sequences. In our study, however, no alternative splicing transcripts in extracellular domain sequences could be found in endometrial carcinomas. Although we did not examine alternative splicing in cytoplasmic domain sequences, the results thus suggest that in endometrial carcinomas, alternative splicing of DCC transcripts may be less important than inactivation at the gene level. Loss of heterozygosity (LOH) of the DCC gene has been demonstrated in tumors of the female reproductive tract, for example, in 1 (13%) of 8 and 4 (31%) of 13 endometrial carcinomas by Okamoto et al.24 and Imamura et al.,25 respectively, using the probe OLVII E10. LOH of chromosome 18q detected by the three probes, pHH64, OLVII E10 and OS-4, was also reported for 16 (26%) of 61 endometrial carcinomas.10 However, in the current study using the PCR-LOH assay, no LOH on the VNTR region of the DCC locus was found, even in carcinomas demonstrating loss of mRNA expression. The possible reasons for anomalous results include the following: 1) inevitable contamination of noncancerous cells in the tumor samples may mask any genetic deletion; 2) other regions of the DCC gene may be affected by the deletion or rearrangement. Recently, Kong et al.26 demonstrated that in neuroblastomas LOH is not necessary for DCC gene inactivation and that other mechanisms, such as mutations in the coding regions or alterations in initiator or promoter areas, can contribute. Similar findings have also been reported for osteosarcomas and endometrial carcinomas.9,16 Another interesting finding in the current study was the alteration in DCC expression in morules and squamous metaplastic (Sq-M) lesions within tumors. We previously demonstrated that the morule foci may be intermediate in phenotype between primary tumors and Sq-M lesions, using bcl-2 immunohistochemistry,27 but the current findings suggest a more complex scenario. Clearly, further studies of the relation between DCC expression and cell differentiation are warranted. In conclusion, the current study demonstrated that DCC expression is maintained during the normal menstrual cycle without any relation to expression of ovarian hormone receptors. In addition, loss or reduction of expression may be a specific event in the progression of endometrial carcinomas through metastatic features. REFERENCES 1. 2. 3. 4. 5. 6. Fearon ER, Cho KR, Nigro JM, Kern SE, Simons JW, Ruppert JM, et al. Identification of a chromosome 18q gene that is altered in colorectal cancers. Science 1990;247:49 –56. Cho KR, Oliner JD, Simons JW, Hedrick L, Fearon AC, Presinger P, et al. The DCC gene: structural analysis and mutations in colorectal carcinomas. Genomics 1994;19:525–31. Hedrick L, Cho KR, Fearon ER, Wu T-C, Kinzler KW, Vogelstein B. The DCC gene product in cellular differentiation and colorectal tumorigenesis. Genes Dev 1994;8:1174 – 83. Gao X, Honn KV, Grignon D, Sakr W, Chen YQ. Frequent loss of expression and loss of heterozygosity of the putative tumor suppressor gene DCC in prostatic carcinomas. Cancer Res 1993;53:2723–7. Cropp CS, Lidereau R, Campbell G, Champene MH, Callahan R. Loss of heterozygosity on chromosomes 17 and 18 in breast carcinoma: two additional regions identified. Proc Natl Acad Sci U S A 1990;87:7737– 41. Huang Y, Boynton RF, Blount PL, Silverstein RJ, Yin J, Tong Y, et al. Loss of heterozygosity involves multiple tumor suppressor genes in human esophageal cancers. Cancer Res 1992;52:6525–30. 464 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. CANCER January 15, 1999 / Volume 85 / Number 2 Reyes-Mugica M, Rieger-Christ K, Ohgaki H, Ekstrand BC, Helei M, Kleinman G, et al. Loss of DCC expression and glioma progression. Cancer Res 1997;57:382– 6. Klingelhutz AJ, Hedrick L, Cho KR, McDougall JK. The DCC gene suppresses the malignant phenotype of transformed human epithelial cells. Oncogene 1995;10:1581– 6. Enomoto T, Fujita M, Cheng C, Nakashima R, Ozaki M, Inoue M, et al. Loss of expression and loss of heterozygosity in the DCC gene in neoplasms of the human female reproductive tract. Br J Cancer 1995;71:462–7. Gima T, Kato H, Honda T, Imamura T, Sasazuki T, Wake N. DCC gene alteration in human endometrial carcinomas. Int J Cancer 1994;57:480 –5. Creasman WT. Announcement, FIGO stages: 1988 revision. Gynecol Oncol 1989;35:125–7. Zaino RJ, Kurman R, Herbold D, Gliedman J, Bundy N, Voet R, et al. The significance of squamous differentiation in endometrial carcinoma: data from a Gynecologic Oncology Group study. Cancer 1991;68:2293–302. Shibata D, Reale MA, Lavin P, Silverman M, Fearon ER, Steele G Jr., et al. The DCC protein and prognosis in colorectal cancer. N Engl J Med 1996;335:1727–32. Saegusa M, Kamata Y, Isono M, Okayasu I. Bcl-2 expression is correlated with a low apoptotic index and associated with progesterone receptor immunoreactivity in endometrial carcinomas. J Pathol 1996;180:275– 82. Reale MA, Hu G, Zafar AI, Getzenberg RH, Levine SM, Fearon ER. Expression and alternative splicing of the deleted in colorectal cancer (DCC) gene in normal and malignant tissues. Cancer Res 1994;54:4493–501. Horstmann MA, Posl M, Scholz RB, Anderegg B, Simon P, Baumgaertl K, et al. Frequent reduction or loss of DCC gene expression in human osteosarcoma. Br J Cancer 1997;75: 1309 –17. Saegusa M, Takano Y, Kamata Y, Okayasu I. Bcl-2 expression and allelic loss of the p53 gene in gastric carcinomas. J Cancer Res Clin Oncol 1996;122:427–32. 18. Kurman RJ, Kaminski PF, Norris HJ. The behavior of endometrial hyperplasia: a long-term study of “untreated” hyperplasia in 170 patients. Cancer 1985;56:403–12. 19. Goi T, Yamaguchi A, Nakagawara G, Urano T, Shiku H, Furukawa K. Reduced expression of deleted colorectal carcinoma (DCC) protein in established colon cancers. Br J Cancer 1998;77:466 –71. 20. Hohne MW, Halatsch M-E, Kahl GF, Weinel RJ. Frequent loss of expression of the potential tumor suppressor gene DCC in ductal pancreatic adenocarcinoma. Cancer Res 1992; 52:2616 –9. 21. Itoh F, Hinoda Y, Ohe M, Ban T, Endo T, Imai K, et al. Decreased expression of DCC mRNA in human colorectal cancers. Int J Cancer 1993;53:260 –3. 22. Kikuchi-Yanoshita R, Konishi M, Fukunari H, Tanaka K, Miyaki M. Loss of expression of the DCC gene during progression of colorectal carcinomas in familial adenomatous polyposis and non-familial adenomatous polyposis patients. Cancer Res 1992;52:3801–3. 23. Ekstrand BC, Mansfield TA, Bigner SH, Fearon ER. DCC expression is altered by multiple mechanisms in brain tumours. Oncogene 1995;11:2393– 402. 24. Okamoto A, Sameshima Y, Yamada Y, Teshima S, Terashima Y, Terada M, et al. Allelic loss on chromosome 17p and p53 mutations in human endometrial carcinoma of the uterus. Cancer Res 1991;51:5632– 6. 25. Imamura T, Arima T, Kato H, Miyamoto S, Sasazuki T, Wake N. Chromosomal deletions and K-ras gene mutations in human endometrial carcinomas. Int J Cancer 1992;51:47–52. 26. Kong X-T, Choi SH, Inoue A, Xu F, Chen T, Takita J, et al. Expression and mutational analysis of the DCC, DPC4, and MADR2/JV18-1 genes in neuroblastoma. Cancer Res 1997; 57:3772– 8. 27. Saegusa M, Okayasu I. Down-regulation of bcl-2 expression is closely related to squamous differentiation and progesterone therapy in endometrial carcinomas. J Pathol 1997; 182:429 –36.