• E-mail
• Print
• Comment
• Font Size
• Digg
• del.icio.us
• Discuss article

Frequent Expression of Human Carcinoma-Associated Antigen, a Mucin- Type Glycoprotein, in Cells of Prostatic Carcinoma

Posted on: Wednesday, 22 December 2004, 03:01 CST

Context.-Human carcinoma-associated antigen (HCA) is a mucin glycoprotein recognized by antibodies raised against epiglycanin, the latter having been originally purified from mouse mammary carcinoma cells. Human carcinoma-associated antigen expression is increased in sera of patients with various carcinomas, including prostatic carcinoma. However, to our knowledge, expression of HCA in benign and neoplastic prostatic tissue has not been studied.

Objective.-To compare the expression of HCA in cells of primary and metastatic prostatic carcinomas with its expression in non- carcinoma-associated cells.

Design.-We studied 40 cases of primary and 36 cases of metastatic prostatic carcinomas by immunohistochemical staining with anti-HCA monoclonal antibodies G1 and HAE3. The blocks from primary carcinomas also contained normal prostatic tissue (40 cases), benign prostatic hyperplasia (16 cases), and high-grade prostatic intraepithelial neoplasia (32 cases).

Results.-The 2 antibodies stained carcinomas more frequently than normal prostatic tissue, hyperplasia, and prostatic intraepithelial neoplasia (P < .001). The differences in the staining of low-grade versus high-grade tumors was not statistically significant with either antibody. The staining was present in the cytoplasm and on the luminal membrane surface of the tumor cells and in the luminal secretions. In metastatic prostatic carcinomas, G1 and HAE3 staining was positive in 44% and 67% of the cases, respectively.

Conclusions.-Our results showed that mucin protein HCA is overexpressed in cells of prostatic carcinoma, which may have value in diagnosis and therapy. Its role in carcinogenesis also merits further study.

(Arch Pathol Lab Med. 2004;128:1412-1417)

Human carcinoma-associated antigen (HCA)1 is a mucin-type glycoprotein identified by antibodies raised against epiglycanin (EPGN), the latter having been originally purified from TA3-Ha murine mammary carcinoma ascites cells.2 Two monoclonal antibodies, G1 and HAE3, were raised against EPGN and have been shown to bind HCA in vitro specifically. Interestingly, the body fluids and sera of patients with a variety of carcinomas showed increased levels of HCA.3,4 Preliminary data have shown that serum HCA level may be very useful in identifying patients with prostatic carcinomas.5 However, the tissue expression of HCA has not been studied. We report our results on the immunohistochemical study of primary and metastatic prostate cancers for their expression of HCA.

MATERIALS AND METHODS

Selection of Cases

This study was approved by the Institutional Research Subject Review Board at the University of Rochester, and all the identifiers were deleted to protect patient confidentiality. Forty cases of prostatectomy for adenocarcinoma were selected for the study. Each case was reviewed, and representative blocks (1 per case) were selected. To correlate HCA expression with lumen formation, the tumors were arbitrarily divided into 2 groups: low grade (good lumen formation, Gleason grades 2 and 3) and high grade (poor lumen formation, Gleason grades 4 and 5). Among the blocks used, 39 of 40 had areas of low-grade carcinoma and 26 had areas of high-grade carcinoma. All the sections contained benign prostatic tissue. In addition, 16 cases contained benign prostatic hyperplasia (BPH), and 32 cases contained high-grade prostatic intraepithelial neoplasia (PIN).

Thirty-six cases of metastatic prostatic carcinoma from an equal number of patients were also selected. The hormonal treatment status of the patients was unknown. The metastatic sites included bone (11 cases), bowel (2 cases), liver (1 case), lymph nodes (20 cases), lung (1 case), and omentum (1 case).

Purification of EPGN

Epiglycanin was purified by the method of Codington et al6 from ascitic fluid obtained from C57/B16 mice inoculated with the EPGN- superexpressing murine mammary carcinoma cell line TA3-MM1.

Preparation of Monoclonal Antibodies

Generation of Immunoglobulin (Ig) M Monoclonal Antibody Gl.-Six- week-old female CBL57/6J mice were immunized on day O intraperitoneally (IP) with 35 g of EPGN mixed 1:1 with Ribi adjuvant. Boosts were administered as follows: day 22, 44 g EPGN with Ribi adjuvant, IP; day 48,1 10^sup 3^ TA3-MM1 cells, IP; day 60, 3 10^sup 4^ TA3-MM1 cells, IP; day 67, 1 10s TA3-MM1 cells, IP; day 85, 28 g EPGN with Ribi adjuvant, IP; day 129, 28 g EPGN in phosphate-buffered saline, IP; and day 130, 28 g EPGN, IP. Hybridomas were generated on day 133 by fusing spleen cells from the multiply immunized mice with the BALB/ c-derived NS1 myeloma and screened in a conventional enzymelinked immunosorbent assay using microtiter plates coated with purified EPGN. Cloned G1 cells were grown in a stirred tank bioreactor (Creative Biomolecules, Lebanon, NH).

Table 1. Results of Immunohistochemical Study of Prostatectomy Specimens With G1

Table 2. Results of Immunohistochemical Study of Prostatectomy Specimens With HAE3

Generation of IgM Anti-EPGN Monoclonal Antibody HAE3.-Antibody AE3 has been described previousl,7 and we have decided to use the alternative name HAE3 to avoid confusion with a commonly used anticytokeratin antibody. Six-weekold female C57BL/6 mice were immunized by IP injection of 50 g asialoEPGN on days 0, 14, and 21. On days 14 and 21, the EPGN was administered as an emulsion in complete Freund adjuvant. On day 35, a further intraperitoneal injection of asialoEPGN was administered. On day 38, hyperimmune spleen cells were fused with SP2 mouse myeloma cells. Cloned HAE3 cells were grown in a 500-mL spinner flask in Dulbecco modified Eagle medium supplemented with 10% fetal bovine serum. The resulting HAE3 IgM antibody was purified by affinity chromatography on agarose conjugated with goat anti-mouse IgM (Maine Biotechnology Services, Portland, Me).

Immunohistochemical Study

Immunohistochemistry was performed on formalin-fixed, paraffin- embedded tissue sections using monoclonal antibodies to HAE3 (1:2000) and G1 (1:10000). Paraffin-embedded tissue was sectioned at 5 m and floated on distilled water at a temperature of 45C. Sections were mounted on chemically charged slides, followed by drying at room temperature until opaque, and then were placed in the oven at 57C overnight. sections were deparaffinized according to established procedures and quenched with 3% hydrogen peroxide for 6 minutes. They were then cleared in running water followed by Tris-buffered saline (TBS) (50mM Tris-HCl, 150mM NaCl, and 0.05% Tween 20 at pH 7.6). Antigen unmasking was performed by preheating the slides at 950C to 990C in Dako antigen retrieval solution (citrate buffer, pH 6.1) in a Black and Decker (Shelton, Conn) steamer for 40 minutes followed by a 15-minute cool-down period. Slides were then rinsed with TBS for 5 minutes and mounted in the Dako Autostainer. sections were covered with fresh TBS to prevent drying of sections during mounting. The slides were stained for 60 minutes at the specified titers, followed by 30-minute incubations each in goat anti-mouse IgM (Vector Laboratories, Inc, Burlingame, Calif) and streptavidin- HRP (Jackson Immunoresearch Lab, West Grove, Pa). Slides were developed with AEC+ (DakoCytomation, Carpinteria, Calif) for 10 minutes and rinsed in running distilled water. They were then counterstained in modified Mayer hematoxylin, followed by 0.3% ammonia water and a tap water rinse. Slides were mounted using an aqueous media and viewed with a light microscope.

In the cases of primary prostatic carcinoma, staining in areas of benign prostatic tissue, nodular hyperplasia, PIN, low-grade carcinoma, and high-grade carcinoma was scored separately. In the cases of metastatic carcinoma, only staining of the metastatic tumor was scored. Scoring included percentage of staining and staining intensity (0, 1+ , 2+ , 3+). Results were recorded as positive (≥5% cells staining and staining intensity of 2+ or 3+ ) or negative (<5% cells staining or staining intensity less than 2+). A Fisher exact test was used to generate P values to test statistical significance.

RESULTS

HCA is Overexpressed by Cells of Prostatic Carcinoma

Immunohistochemistry was performed on 40 cases of prostate cancer with monoclonal antibodies G1 and HAE3. The 2 antibodies showed very similar staining patterns in most sections. Generally, cells of prostatic carcinomas stained with both antibodies strongly and diffusely, while staining in normal prostatic tissue, BPH, and high- grade PIN were generally negative or weak. Lower-grade tumors with well-formed glands (Gleason grades 2 or 3) showed more staining than higher-grade tumors with little or no lumen formation (Gleason grades 4 or 5). Although staining was seen in the cytoplasm, there was often concentration of staining on the luminal membrane surface and the luminal secretions.

G1 staining was positive in 35% of normal prostatic tissue, 12.5% of BPH, 40.6% of high-grade PIN, 92.3% of low-grade carcinoma, and 61.5% of high-grade carcinoma cases. HAE3 staining was positive in 40% of normal prostatic tissue, 25% of BPH cases, 37.5% of high- grade PIN cases, 89.7% of low-grade carcinoma cases, and 65.4% of highgrade carcinoma cases. With either antibody, the difference in positive staining in c\arcinoma versus normal prostatic tissue, BPH, and high-grade PIN were statistically significant (P < .001). The difference in positive staining between low-grade and high-grade tumors was not statistically significant (G1, P = .02; AE3, P = .02). The results are summarized in Tables 1 and 2. Figure 1 (A through D) shows representative images of staining by HAE3. Figure 2 (A through D) shows representative images of staining by G1.

Figurel. Immunohistochemical study of human carcinoma-associated antigen expression in benign ans neoplastic prostatic tissue and metastatic prostate carcinoma by antibody HAE3. A, Negative staining in an area of benign prostatic hyperplasia. B, Focal positive staining in an area of high-grade PIN, C, Diffuse and strong staining in an area of low-grade carcinoma. D, Focal staining in an area of high-grade carcinoma. E, Positive staining of metastatic prostate carcinoma in lymph node. F, Positive staining of metastatic prostate carcinoma in bone (HAE3 immunohistochemistry, original magnification x400).

Figure 2. Immunohistochemical study of human carcinoma- associated antigen expression in benign and neoplastic pmstatic tissue and metastatic prostatic carcinoma by antibody G1. A, Negative staining in an area of benign pmstatic hyperplasia. B, Focal positive staining in an area of high-grade pmstatic intraepithelial neoplasia. C, Diffuse and strong staining in an area of low-grade carcinoma. D, Focal staining in an area of high-grade carcinoma. E, Positive staining of metastatic prostate carcinoma in lymph node. F, Positive staining of metastatic prostate carcinoma in bone (G1 immunohistochemistry, original magnification x400).

Table 3. Results of Immunohistochemical Study of Metastatic Prostatic Carcinoma With Monoclonal Antibodies G1 and HAE3

Frequent Expression of HCA by Cells of Metastatic Prostatic Carcinoma

Thirty-six cases of metastatic prostatic carcinoma were also studied for the expression of HCA by immunohistochemistry with antibodies G1 and HAE3. Sixty-seven percent and 44% of cases showed positive staining with HAE3 and G1, respectively (Table 3). Representative images of staining by AE3 are shown in Figure 1 (E and F). Representative images of staining by Gl are shown in Figure 2 (E and F).

COMMENT

Prostate cancer has a high prevalence among older men, and screening with a noninvasive test such as the serum level of prostate-specific antigen (PSA) has played an important role in early diagnosis.8 Prostate-specific antigen is a member of the tissue kallikrein family of proteases and possesses serine protease activity. It is equally produced by both benign and malignant prostatic epithelial cells, so that it is organ-specific but not cancer-specific. A serum PSA test is a sensitive test for prostate cancer, but its specificity is very low with many benign conditions, particularly BPH, which also causes an increase in the serum level of PSA. In general, a PSA level of 2.5 to 10 (or 4 to 10) ng/mL is considered a gray zone, since this group consists of both cancer patients and patients with benign conditions, such as BPH. As a result, many patients receive unnecessary biopsies, leading to increased health care costs and patient discomfort. Therefore, assays with better specificity than serum PSA levels are sorely needed to complement PSA test.

It is well known that carcinoma cells of various organs produce mucin proteins, and the mucin is secreted into serum and body fluid, which can be used for diagnostic purposes.9-11 Using serum HCA assay with HAE3 to distinguish between prostate cancer and benign conditions has shown promise.5 In this study, we have provided histologic evidence that HCA is preferentially expressed in cells of prostatic carcinoma in comparison with benign prostatic tissue, including BPH. The histologic pictures show that HCA is produced in the cytoplasm of the tumor cells, concentrated on the cell membrane, and released into extracellular spaces. In lower-grade tumors with wellformed lumens (Gleason 2 or 3 patterns), the luminal secretions stain diffusely and strongly with G1 and HAE3, while in higher- grade tumors without lumen formation (Gleason 4 or 5 patterns) the staining is often focal. We hypothesize that all tumor cells produce and secrete HCA, but when HCA enters stroma, it finds its way into the bloodstream easily, so there is no accumulation of HCA on the cell membrane and extracellular spaces. As a result, the staining is often negative in high-grade rumors and on the side opposite to the lumen in cells of lower-grade tumors. However, once HCA is secreted into the luminal space of lower-grade tumors, the protein is trapped there and can be easily stained with antibodies because these lumens, unlike the lumens of benign prostate, are part of the invasive tumor and are not connected to the normal ductal system of the prostate. As a result, there may not always be a perfect correlation between tissue staining of HCA and serum HCA levels in patients with prostate cancer. In other words, patients with high- grade tumors may not necessarily have lower serum HCA levels than those with low-grade tumors, as would be suggested by the results of the immunohistochemical study. Unfortunately, we do not have the serum HCA data on our study population, and further study correlating serum HCA level with immunohistochemical staining is needed to test this hypothesis.

Our results show that most low-grade tumors are positive for HCA, while HCA is less frequently produced in cells of BPH, suggesting that this marker can distinguish between the 2 conditions and will probably have important clinical utility, particularly since serum PSA level performs poorly in this area. It is interesting that normal prostatic tissue expresses HCA more frequently than BPH, suggesting the possibility of a high rate of false positivity if serum HCA is used as a screening test. However, since benign prostatic epithelial secretory cells are separated from stroma by a basal layer, the HCA produced by these cells may not have easy access to stroma and subsequently the bloodstream. In prostate cancer, the basal layer is absent and the secretory products may enter the stroma and bloodstream much more easily. Preliminary studies in serum samples from patients have produced very encouraging results.5 It is conceivable that the combination of serum PSA and HCA testing will achieve much greater specificity, thus eliminating many unnecessary prostatic biopsies.

The molecular structure and function of HCA is unknown. The mechanism by which cells of prostatic carcinoma overexpress HCA is not understood. Nonetheless, since HCA is much more frequently expressed in lowgrade carcinoma than BPH, measuring serum HCA level may help to distinguish between these 2 conditions that cannot be separated by a PSA test alone. Since HCA is also expressed in certain metastatic prostate cancers, monitoring serum HCA may be useful in following patients for recurrence or metastasis. Finally, HCA may be exploited as a target of immunotherapy with humanized antiHCA monoclonal antibodies. The epitope on HCA, once determined with certainty, may be used as a potential vaccine for prostate cancer.

This work was partially supported by a grant from Egenix, Inc, Milibrook, NY.

References

1. Thingstad T, Haavik S, Hansen K, Sletten K, Codington JF, Barsett H. Human carcinoma-associated antigen (HCA), isolated from the endometrial carcinoma cell line KLE-1 and ascitic fluid of a patient with ovarian carcinoma: comparison with epiglycanin. Eur I Pharm Sci. 1998;6:121-129.

2. Codington JF, Linsley KB, Jeanlot RW. Immunochemical and chemical investigations of the structure of glycoprotein fragments obtained from epiglycanin, a glycoprotein at the surface of the TA3- Ha cancer cell. Carbohydr Res. 1975; 40:171-182.

3. Codington JF, Haavik S, Nikrui N, et al. Immunologic quantitation of the carcinoma specific human carcinoma antigen in clinical samples. Oncer. 2002; 94:803-813.

4. Codington JF, Bhavanandan VP, Bloch KJ, et al. Antibody to epiglycanin and radioimmunoassay to detect epiglycanin-related glycoproteins in body fluids of cancer patients. I Natl Cancer lnst. 1984;73:1029-1037.

5. Taylor JA, Goluboff ET, Holub W, et al. The human carcinoma antigen: a new biomarker for prostate cancer. J Urol. 1999;! 61 (4, suppl):209.

6. Codington JF, Cooper AG, Miller DK, et al. Isolation and partial characterization of an epiglycanin-like glycoprotein from a new non-strain-specific subline of TA3 murine mammary adenocarcinoma. I Natl Cancer lnst. 1979:63:153-161.

7. Haavik S, Nilsen M, Thingstad T, et al. Specificity studies of an antibody developed against a mucin-type glycoprotein. Glycoconj J. 1999;16:229-236.

8. Cretzer MB, Partin AW. PSA markers in prostate cancer detection. Urol Clin North Am. 2003:30:677-686.

9. Fradet Y, Lockhart C, and Immunocyt(TM) trialists. Performance characteristics of a new monoclonal antibody test for bladder cancer: Immunocyt(TM). Can J. Urol. 1997;4:400.

10. Dodd I, Tyler JP, Crandon A), et al. The value of the monoclonal antibody (cancer antigen 125) in serial monitoring of ovarian cancer: a comparison with circulating immune complexes. Br I Obstet Gynaecol. 1985:92:1054-1060.

11. Rasoul-Rockenschaub S, Zielinski CC, Kubista E, et al. Diagnostic value of mucin-like carcinoma-associated antigen (MCA) in breast cancer. Eur J Cancer Clin Oncol. 1989;25:1067-1072.

Rongshan Li, MD, PhD; Jorge L. Yao, MD; Patricia A. Bourne, AAS, HT; P. Anthony di Sant'Agnese, MD; Jiaoti Huang, MD, PhD

Accepted for publication August 4, 2004.

From the Department of Pathology, University of Rochester, Rochester, NY.

The authors have no relevant financial interest in the products or companies described in this article.

Reprints: liaoti Huang, MD, PhD, Department of Pathology, University of Rochester Medical Center, 601 Elmwood Ave, Box 626, Rochester, NY 14642 (e-mail: Jiaoti_Huang@urmc.ro\chester.edu).

Copyright College of American Pathologists Dec 2004

Source: Archives of Pathology & Laboratory Medicine

More News in this Category