Tissue Localization of Zinc Glycinate Marker and Carcinoembryonic Antigen by Immunofluorescence. 11. Immunofluorescence Microscopy 1.2
w. G. Doos, M.D., 3.4 C. A. Saravis, Ph.D., 5 N. Zamcheck, M.D.,
7
G. Pusztaszeri, M.D., and L. S. Gottlieb, M.D. 8.9
ABSTRACT-Preliminary indirect immunofluorescence studies on the zinc glycinate marker (ZGM) were compared with carcinoembryonic antigen (CEA) immunofluorescence. ZGM, detected in 26 of 29 human colon adenocarcinomas, was associated with the epithelial component of the malignant glands. Fluorescence was generally less strong and more granular for ZGM than for CEA and was found in intraglandular spaces, luminal border areas, and cytoplasm. ZGM concentration and tissue. localization appeared to be related to tumor differentiation. ZGM was also detected in benign colon mucosae (adjacent to and distant from the carcinomas) from patients with colon carcinoma, but differed from CEA in that it was present in the deep crypt portion only. Gastric, pancreatic, esophageal, and anal adenocarcinomas, as well as benign gastric pyloric and small bowel mucosae had detectable ZGM. CEA, but not ZGM, was observed in 20 nongastrointestinal carcinomas to date. Studies are under way to determine whether ZGM is a marker associated with colon and gastrointestinal adenocarcinoma specifically or undifferentiated crypt cells of the colon and digestive tract in general.-J Natl Cancer Inst 60: 1375-1382, 1978.
The ZGM of colon carcinoma has been previously reported (1). Each newly described antigen derived from tumors must be localized in tissues to define the site and source of the antigen and to establish its morphologic relationship to the malignant cells. In this paper we present preliminary observations of ZGM tissue localization and compare it with that of CEA, utilizing indirect immunofluorescence techniques (2-4).
6
B. Burke, 6 S. K. Oh, Ph.D.,
5
cence stammg. Tumors from tissues other than the colon were treated in an identical manner. In each instance all slides prepared for the routine diagnostic evaluation were also available for examination. Each tumor was graded for differentiation, as originally defined by Rankin and Broders (5), and for depth of invasion by the Dukes system (6). Paraffin-embedded tissues selected from the Mallory files were also utilized. These tissues had been conventionally fixed in Formalin and embedded in paraffin for routine pathologic slide preparations. Five-micron sections of the blocks were made and deparaffinized in a conventional manner and stored in PBS at pH 7.4 prior to incubation with the antiserum. Antisera.-In initial studies multivalent anti-ZGM antiserum produced in tolerant rabbits was used. In subsequent studies a pooled, functionally monospecific rabbit anti-ZGM antisera was used (rabbit #R-2l) (7). Rabbit anti-CEA antiserum was obtained from Chasma Scientific, Brighton, Massachusetts (lot #CEAQOl) and reacted with all tissues examined for ZGM as an additional control for comparison. Rabbit anti-CEA antiserum was recently obtained from Dr. H. Hansen, Hoffmann-La Roche, Inc., Nutley, New Jersey. Both ABBREVIATIONS USED: CEA = carcinoembryonic antigen; FITC = fluorescein isothiocyanate; H & E = hematoxylin and eosin; NCA = nonspecific cross-reacting antigen; PBS = phosphate-buffered saline; ZGM = zinc glycinate marker.
MATERIALS AND METHODS Received July 20,1977; accepted November 18.1977. Supported by Public Health Service (PHS) grants CA18620-01 and CA04486-17 from the National Cancer Institute (NCI) and general research support grant 4057 from the Division of Research Resources, National Institutes of Health. 3 Mallory Gastrointestinal Research Laboratory, Mallory Institute of Pathology, Boston City Hospital; and Department of Pathology, Boston University School of Medicine, Boston, Mass. 02118. 4 Address reprint requests to Dr. W. G. Doos, Mallory Gastrointestinal Research Laboratory, Boston City Hospital. 5 Mallory Gastrointestinal Research Laboratory and Sears Surgical Laboratory, Boston City Hospital; Department of Surgery, Harvard Medical School; and Department of Pathology, Boston University School of Medicine. 6 Mallory Gastrointestinal Research Laboratory, Boston City Hospital. 7 Mallory Gastrointestinal Research Laboratory, Boston City Hospital; Department of Medicine, Harvard Medical School; and Department of Pathology, Boston University School of Medicine. 8 Mallory Institute of Pathology, Boston City Hospital; and Department of Pathology, Boston University School of Medicine. 9 We thank Drs. P. Burtin, R. Cotran, E. Alpert, and R. McCloskey for their advice and Ms. S. Melngailis for technical assistance. 1
Tissues.-Portions of surgically removed colorectal carcinoma were rapidly frozen either by exposure to isopentane and Dry Ice or by ajet stream of pressurized CO 2 , The frozen tissues were then embedded in O.C.T. compound (an embedding medium for frozen tissue specimens; Lab-Tek Products, Westmont, Ill.) and stored at -600 C until sectioned. The carcinomas were oriented on a cryostat chuck so that they included an immediately adjacent segment of nonmalignant mucosa. "Adjacent mucosa" in this paper refers to this segment of mucosa in contact with the carcinoma and present on the same section within 5 to 10 mm of the tumor. Sections of distant nonmalignant mucosae were also taken, usually from the surgical margins, always more than 5 cm from the tumor mass. Sectioning was done in a cryostat at -20 0 C. Serial sections 4-6 IL thick were prepared, and alternate sections were stained with conventional H & E for comparison with the fluoresVOL. 60. NO. 6, JUNE 1978
2
1375
J NATL CANCER INST
1376
DOOS ET AL.
antisera were extensively absorbed by extracts of normal human lung and spleen. These antisera were monospecific for CEA when immunoprecipitation techniques were used. Normal goat serum was added to all antis~ra (100 ILl/ml undiluted antiserum) to reduce nonspeClfic staining. Optimal anti-CEA antiserum dilution was 1:64-1: 128 for fresh-frozen tissues and 1: 16-1 :32 for paraffinembedded tissues. Optimal anti-ZGM antiserum dilution was I :32 for fresh-frozen, tissues and 1:4-1:8 for paraffin-embedded tissues. Paraffin-embedded tissues were used as an adjunct to fresh-frozen tissues; they were not used for primary interpretation. FITC-conjugated goat anti-rabbit IgG antiserum was obtained from Behring Diagnostics, American Hoechst Corporation, Somerville, New Jersey (lot #682E). The FITC-labeled antiserum was twice absorbed with mouse liver powder (50 mg/ml for first absorption; 20 mg/ml for second absorption) prior to its use in order to decrease nonspecific staining. The mouse liver po~der was obtained from Baltimore Biological LaboratorIes (a division of Becton, Dickinson and Co.), Cockeysville, Maryland (lot #F8EIAJ). The FITC-Iabeled antiserum was used at a concentration of 1: 10. Staining procedures .-Tissue sections were air dried at room temperature (all subsequent steps were also done at room temperature) for a minimum of several hours. Sections then were immersed in acetone (spectral analyzed) for 10 minutes, rinsed in PBS (pH 7.4), and incubated with the first antiserum in a humid chamber for 30-45 minutes. Subsequently, they were rinsed in PBS and incubated with the FITC-Iabeled goat antirabbit IgG antiserum for 30-45 minutes. After this, they were again rinsed in PBS and m
w. G. Doos, M.D., 3.4 C. A. Saravis, Ph.D., 5 N. Zamcheck, M.D.,
7
G. Pusztaszeri, M.D., and L. S. Gottlieb, M.D. 8.9
ABSTRACT-Preliminary indirect immunofluorescence studies on the zinc glycinate marker (ZGM) were compared with carcinoembryonic antigen (CEA) immunofluorescence. ZGM, detected in 26 of 29 human colon adenocarcinomas, was associated with the epithelial component of the malignant glands. Fluorescence was generally less strong and more granular for ZGM than for CEA and was found in intraglandular spaces, luminal border areas, and cytoplasm. ZGM concentration and tissue. localization appeared to be related to tumor differentiation. ZGM was also detected in benign colon mucosae (adjacent to and distant from the carcinomas) from patients with colon carcinoma, but differed from CEA in that it was present in the deep crypt portion only. Gastric, pancreatic, esophageal, and anal adenocarcinomas, as well as benign gastric pyloric and small bowel mucosae had detectable ZGM. CEA, but not ZGM, was observed in 20 nongastrointestinal carcinomas to date. Studies are under way to determine whether ZGM is a marker associated with colon and gastrointestinal adenocarcinoma specifically or undifferentiated crypt cells of the colon and digestive tract in general.-J Natl Cancer Inst 60: 1375-1382, 1978.
The ZGM of colon carcinoma has been previously reported (1). Each newly described antigen derived from tumors must be localized in tissues to define the site and source of the antigen and to establish its morphologic relationship to the malignant cells. In this paper we present preliminary observations of ZGM tissue localization and compare it with that of CEA, utilizing indirect immunofluorescence techniques (2-4).
6
B. Burke, 6 S. K. Oh, Ph.D.,
5
cence stammg. Tumors from tissues other than the colon were treated in an identical manner. In each instance all slides prepared for the routine diagnostic evaluation were also available for examination. Each tumor was graded for differentiation, as originally defined by Rankin and Broders (5), and for depth of invasion by the Dukes system (6). Paraffin-embedded tissues selected from the Mallory files were also utilized. These tissues had been conventionally fixed in Formalin and embedded in paraffin for routine pathologic slide preparations. Five-micron sections of the blocks were made and deparaffinized in a conventional manner and stored in PBS at pH 7.4 prior to incubation with the antiserum. Antisera.-In initial studies multivalent anti-ZGM antiserum produced in tolerant rabbits was used. In subsequent studies a pooled, functionally monospecific rabbit anti-ZGM antisera was used (rabbit #R-2l) (7). Rabbit anti-CEA antiserum was obtained from Chasma Scientific, Brighton, Massachusetts (lot #CEAQOl) and reacted with all tissues examined for ZGM as an additional control for comparison. Rabbit anti-CEA antiserum was recently obtained from Dr. H. Hansen, Hoffmann-La Roche, Inc., Nutley, New Jersey. Both ABBREVIATIONS USED: CEA = carcinoembryonic antigen; FITC = fluorescein isothiocyanate; H & E = hematoxylin and eosin; NCA = nonspecific cross-reacting antigen; PBS = phosphate-buffered saline; ZGM = zinc glycinate marker.
MATERIALS AND METHODS Received July 20,1977; accepted November 18.1977. Supported by Public Health Service (PHS) grants CA18620-01 and CA04486-17 from the National Cancer Institute (NCI) and general research support grant 4057 from the Division of Research Resources, National Institutes of Health. 3 Mallory Gastrointestinal Research Laboratory, Mallory Institute of Pathology, Boston City Hospital; and Department of Pathology, Boston University School of Medicine, Boston, Mass. 02118. 4 Address reprint requests to Dr. W. G. Doos, Mallory Gastrointestinal Research Laboratory, Boston City Hospital. 5 Mallory Gastrointestinal Research Laboratory and Sears Surgical Laboratory, Boston City Hospital; Department of Surgery, Harvard Medical School; and Department of Pathology, Boston University School of Medicine. 6 Mallory Gastrointestinal Research Laboratory, Boston City Hospital. 7 Mallory Gastrointestinal Research Laboratory, Boston City Hospital; Department of Medicine, Harvard Medical School; and Department of Pathology, Boston University School of Medicine. 8 Mallory Institute of Pathology, Boston City Hospital; and Department of Pathology, Boston University School of Medicine. 9 We thank Drs. P. Burtin, R. Cotran, E. Alpert, and R. McCloskey for their advice and Ms. S. Melngailis for technical assistance. 1
Tissues.-Portions of surgically removed colorectal carcinoma were rapidly frozen either by exposure to isopentane and Dry Ice or by ajet stream of pressurized CO 2 , The frozen tissues were then embedded in O.C.T. compound (an embedding medium for frozen tissue specimens; Lab-Tek Products, Westmont, Ill.) and stored at -600 C until sectioned. The carcinomas were oriented on a cryostat chuck so that they included an immediately adjacent segment of nonmalignant mucosa. "Adjacent mucosa" in this paper refers to this segment of mucosa in contact with the carcinoma and present on the same section within 5 to 10 mm of the tumor. Sections of distant nonmalignant mucosae were also taken, usually from the surgical margins, always more than 5 cm from the tumor mass. Sectioning was done in a cryostat at -20 0 C. Serial sections 4-6 IL thick were prepared, and alternate sections were stained with conventional H & E for comparison with the fluoresVOL. 60. NO. 6, JUNE 1978
2
1375
J NATL CANCER INST
1376
DOOS ET AL.
antisera were extensively absorbed by extracts of normal human lung and spleen. These antisera were monospecific for CEA when immunoprecipitation techniques were used. Normal goat serum was added to all antis~ra (100 ILl/ml undiluted antiserum) to reduce nonspeClfic staining. Optimal anti-CEA antiserum dilution was 1:64-1: 128 for fresh-frozen tissues and 1: 16-1 :32 for paraffinembedded tissues. Optimal anti-ZGM antiserum dilution was I :32 for fresh-frozen, tissues and 1:4-1:8 for paraffin-embedded tissues. Paraffin-embedded tissues were used as an adjunct to fresh-frozen tissues; they were not used for primary interpretation. FITC-conjugated goat anti-rabbit IgG antiserum was obtained from Behring Diagnostics, American Hoechst Corporation, Somerville, New Jersey (lot #682E). The FITC-labeled antiserum was twice absorbed with mouse liver powder (50 mg/ml for first absorption; 20 mg/ml for second absorption) prior to its use in order to decrease nonspecific staining. The mouse liver po~der was obtained from Baltimore Biological LaboratorIes (a division of Becton, Dickinson and Co.), Cockeysville, Maryland (lot #F8EIAJ). The FITC-Iabeled antiserum was used at a concentration of 1: 10. Staining procedures .-Tissue sections were air dried at room temperature (all subsequent steps were also done at room temperature) for a minimum of several hours. Sections then were immersed in acetone (spectral analyzed) for 10 minutes, rinsed in PBS (pH 7.4), and incubated with the first antiserum in a humid chamber for 30-45 minutes. Subsequently, they were rinsed in PBS and incubated with the FITC-Iabeled goat antirabbit IgG antiserum for 30-45 minutes. After this, they were again rinsed in PBS and m
