Quantitation of Secretory Component Levels in Cyst Fluids, Ascitic Fluids, and Sera From Ovarian Adenocarcinoma Patients 1.2 Jerry L. Klein,
3
Stanley A. Gall,
4
and Jeffrey R. Dawson
ABSTRACT -A secretory component (SC) was detected by radioimmunoassay In the cyst fluids, ascitic fluids, and sera from patients with ovarian adenocarcinomas. Serous cyst fluids and ascitic fluids showed lower levels (expressed as means± SE) of SC (1.37±O.37 and 1.24±0.24 "glml, respectively) than mucinous cyst fluids (181.50±50.40 "g/ml). SC levels In the sera of all patients with ovarian adenocarcinoma were high (12.67±1.43 "glml) when compared to SC levels In the sera of normal Individuals (2.34±0.41 "g/ml). Sera from patients with ovarian cancers diagnosed as serous, mucinous, papillary, and poorly differentiated adenocarcinomas showed SC levels of 9.93±1.68, 22.44±3.24, 7.35±1.13, and 10.10±1.58 >Lg/ml, respectlvely.-J Natl Cancer Inst 61: 57-60, 1978.
Secretory immunoglobulin A is the predominant immunoglobulin found in external secretions. Generally, the structure consists of a dimeric IgA, j-chain, and SC, a glycoprotein with a molecular weight of approximately 70,000 (1). IgA and j-chain are synthesized by specific plasma cells, whereas SC is synthesized by epithelial cells lining the glandular lumen (1-3). Recently, investigators have shown the presence of FSC associated with several forms of epithelial cancer. Huang et al. (4), using immunofluorescence microscopy and radioactive labeling techniques, showed that three cell lines of human colon cancer can synthesize SC in vitro. Harris and co-workers (5), also using immunofluorescence microscopy for FSC, found intense staining of tumor cells and stromal cells in tissue samples from patients with breast cancer. This observation contrasted with the staining of only ductule epithelial cells in sections of normal breast tissue. Abnormal staining patterns were also observed in colon, lung, and bladder carcinomas. Our laboratory has developed a radioimmunoassay for SC and has used it to determine the SC levels in the sera, cyst fluids, and ascitic fluids of patients with ovarian adenocarcinomas. Preliminary studies revealed significantly elevated levels of SC in these patients compared to normal individuals.
hyde-insolubilized human serum albumin (8). The purified FSC gave a single band by sodium dodecyl sulfate polyacrylamide gel electrophoresis (9) (data not shown). Production of antiserum to SC.-Rabbits were given sc injections of 1 mg of FSC in complete Freund's adjuvant weekly for 6 weeks. After a rest period of 4 weeks, they were boosted with 1 mg of FSC in complete Freund's adjuvant. Serum was collected 8 days later and absorbed with 120 mg lyophilized normal human plasma per ml of serum. The antiserum was tested for anti-SC activity by immunodiffusion, aliquoted, and frozen at -20° C. It gave a single line of identity with commercially prepared antiserum (Behring Diagnostics) when tested against purified FSC, sera containing high concentrations of SC, and ovarian cyst fluids (data not shown). Labeling of FSC.-FSC was labeled with 1251 with the use of 100 p.Ci Bolton-Hunter reagent (New England Nuclear, Boston, Mass.) per 85 p.g protein (10). The labeled FSC was separated from free iodine reagent on a Sephadex G-25 column equilibrated with dextrose gelatin-Veronal buffer (Grand Island Biological Co., Grand Island, N.Y.). Radioimmunoprecipitation assay.- The general procedures for radioimmunoassays outlined by Hunter (11) were used to develop a suitable assay system. 6 Antiserum to SC was titrated by immunoprecipitation against 1251_ labeled FSC. The antiserum was serially diluted with PBS, and normal rabbit serum was added to a final serum concentration of 10%. Five microliters of diluted antiserum was added to 25 #,1 of labeled FSC (=75 ng) and incubated at 37° C for 1 hour. Ten microliters of goat anti-rabbit IgG (Miles Laboratories, Inc., Elkhart, Ind.) was added for maximum precipitation, and the mixtures were incubated at 4° C overnight. The total contents of each assay mixture were counted, the precipitates were washed four times with PBS, and the contents were then counted again in a Nuclear-Chicago gamma counting system (1l85 series; Nuclear-Chicago
MATERIALS AND METHODS
ABBREVIATIONS USED: FSC=free secretory component; PBS=phos phate-buHered saline; SC=secretory component.
Isolation of FSC.-FSC was purified from human colostrum by the procedure of Kobayashi (6), with the use of a combination of ammonium sulfate fractionation, ion exchange chromatography, and gel filtration. Protein fractions were followed spectrophotometrically at 280 nm. Protein concentrations were determined by the method of Lowry et al. (7). Samples containing protein were tested for the presence of FSC by immunodiffusion against rabbit anti-human FSC (Behring Diagnostics, Somerville, N.J.). Lactoferrin, which copurifies with FSC, was removed by two adsorptions with glutaraldeVOL. 61, NO. I, JULY 1978 Downloaded from https://academic.oup.com/jnci/article-abstract/61/1/57/917897 by INSEAD user on 05 March 2018
3. 5
Received September 28. 1977; accepted February 12, 1978. Supported by grant IM-94 from the American Cancer Society and by Public Health Service training grant CA09058 from the National Cancer Institute. S Division of Immunology, Department of Microbiology and Immunology, Duke University Medical Center, Durham, N.C. 27710. • Department of Obstetrics and Gynecology, Duke University Medical Center. 5 Address reprint requests to Dr. Dawson. 6 Klein JL, Dawson JR: Submitted for publication. I
2
57
J
NATL CANCER INST
58 Klein, Gall, and Dawson
Corp., Des Plaines, Ill.). The counts precipitated are expressed as the percent total counts in the assay system. A standard precipitation curve was established with the use of purified FSC samples serially diluted with PBS (concentrations of 10 J.Lg/ml to 0.16 J.Lg/ml). Five microliters of anti-SC (diluted 1/128 with PBS plus 10% normal rabbit serum) was added to 25 J.Ll of a sample containing known concentrations of FSC. Twenty-five microliters of 125I-Iabeled FSC was added, and the mixture treated as described above. Serum samples to be tested were obtained from patients and normal volunteers and stored at -200 C. All samples, prior to use, were reduced with 20 mM 2-mercaptoethanol for 1 hour at room temperature and alkylated with 25 mM iodoacetate for an additional hour to dissociate SC from immunoglobulins and to prevent nonspecific binding of FSC to serum proteins and immunoglobulins (12). Samples (25 J.LI) of serum diluted 1: I with PBS were tested as described below. SC concentrations were determined from the standard precipitation curve. Cyst fluids from ovarian adenocarcinoma and ascitic fluids associated with ovarian adenocarcinoma were centrifuged at 2,OOOXg for 20 minutes to remove cells and solid material and were stored at -200 C. These samples were also reduced and alkylated prior to use. Immunodiffusion studies on serum and fluid samples.-Immunodiffusion studies were done on glass microscope slides coated with 1.5% Difco Noble agar (Difco Laboratories, Detroit, Mich.) in PBS, pH 7.2, containing 0.02% sodium azide as a preservative. Six wells were cut into the agar in a hexagonal pattern equidistant from a seventh central well. Ten microliters
l.-SC levels (expressed in mean±SE) in normal volunteers and ovarian cancer patients. Normal sera2.34±O.41 j.lg/ml; ascitic f1uid1.24±O.24 /lg/ml; serous cyst f1uidI.13±O.37 /lg/ml; mucinous cyst f1uid181.5o±50.40 /lg/ml; ovarian adenocar· cinoma sera-12.67±1.43/lg/ml; serous adenocarcinoma sera-9.93±1.68 /lg/ ml; mucinous adenocarcinoma sera22.44±3.24 /lg/ml; papillary adenocar· cinoma sera-7.35±1.13 j.lg/ml; and poorly differentiated adenocarcinoma sera-IO.IO±1.58 /lg/ml. FSP=free secretory piece, an alternative designation for FSC.
TEXT·FlGURE
of each sample to be tested was placed in one of the six peripheral wells, and 10 J.LI of rabbit anti-SC was placed in the middle well. Diffusion was carried out for 48 hours at room temperature in a moist covered chamber to pn;vent drying. The slides were washed with PBS at room temperature for 48 hours, with a change of PBS at 24 hours, and washed with distilled water for an additional 24 hours. The slides were then air dried for 24-48 hours and stained with amido black. RESULTS SC Levels in Cyst and Ascitic Fluids
The SC levels in cyst and ascitic fluids of ovarian cancer patients are shown in text-figure I. Serous cyst fluids and ovarian adenocarcinoma ascitic fluids of unspecified histology contained low levels (expressed as means±sE) of SC (1.l3±0.37 and 1.24±0.24 J.Lg/ml. respectively). The levels were similar to those found in normal serum (2.34±0.41 J.Lg/ml). In contrast, the SC levels in mucinous cyst fluids were extremely high (181.50±50.40 J.Lg/ml). The SC levels in the sera of patients and normal volunteers are also shown in text-figure I. Normal volunteers had a low SC level in their sera (2.34±0.41 J.Lg/ml) compared to ovarian cancer patients (l2.67±1.43 J.Lg/ml). In the ovarian cancer patients, the levels were the highest when the diagnosis was mucinous adenocarcinoma (22.44±3.24 J.Lg/ml). The SC levels were also elevated in patients with serous, papillary, and poorly differentiated adenocarcinomas (9.93±1.68, 7.35±1.l3, and 1O.10±1.58 J.Lg/ml. respectively).
--
2!10 200 60
50
~
40
::::l
a:
....
,
'"
E .... 30 Q.
'"
..·
LL
'""-
· ..
20
--.~ .~
10
0
J
f
_'C_
NORMAL SERA
NATL CANCER (NST
Downloaded from https://academic.oup.com/jnci/article-abstract/61/1/57/917897 by INSEAD user on 05 March 2018
ASCITIC FLUID
SEROUS CYST FLUID
MUCINOUS OVARIAN CYST FLUID
.
-.-
...
--'-
SEROUS
MUCINOUS PAPillARY POORLY
CARCINOMA SERA
,
DIFFERENTIATED
VOL. 61, NO. I, JULY 1978
Elevated Secretory Component In Ovarian Adenocarcinoma 59
identical competition curves over a wide range of dilutions. 40
DISCUSSION
30
a w ~
~
U ~
20
Q.
C/l
IZ :::l
o
U
10
O~~
256
__~__~__~~__~__~______________~ 128
64
32
16
8
4
2
II DILUTION
TEXT· FIGURE 2.-Radioimmunoassay compellllon curves. Purified FSC, mucinous cyst fluid, and serum samples were diluted with PBS. The radioimmunoassay was run as described in "Materials and Methods." The counts precipitated are expressed as a percent of the counts in the assay system. All points were done in triplicate with less than I % variance within a group. X----X and 0-----0 indicate mucinous cyst fluids; t:r------6 and ~, serum samples; .... --e, purified FSC.
Immunodiffusion Patterns of Serum and Cyst Fluid Samples
Immunodiffusion patterns of reduced and alkylated serum and cyst fluid samples are shown in figure 1. Two serum samples from ovarian cancer patients containing high levels of SC, two mucinous cyst fluid samples, and purified FSC were used in each study. Rabbit anti-SC was placed in the center wells. A single line of identity was found among all samples tested and purified FSC. Serum samples from patients with all four types of ovarian adenocarcinomas (text-fig. I) gave identical results. Radioimmunoassay Competition Curve of Serum and Cyst Fluid Samples
Radioimmunoassay competition curves of serum and cyst fluid samples are shown in text-figure 2. A competition curve with the use of unlabeled purified FSC is also shown. All samples were serially diluted with PBS for the assay. Two mucinous cyst fluids, two serum samples, and purified FSC gave essentially VOL. 61. NO. I, JULY 1978 Downloaded from https://academic.oup.com/jnci/article-abstract/61/1/57/917897 by INSEAD user on 05 March 2018
The findings of Huang et al. (4), with cultured cell lines of colon cancer, and Harris et al. (5), with speCImens of breast cancer tissue, suggested that levels of SC might be elevated in the sera of patients with ovarian cancer. Preliminary screening of cyst fluids from ovarian cancer patients for SC by immunodiffusion in our laboratory (13) revealed that the detection of SC in the cyst fluids correlated with a diagnosis of mucinous cystadenoma or cystadenocarcinoma. For this reason, we developed a more sensitive radioimmunoassay for detecting Sc. 6 In addition, the specificity of the antisera and the immunologic relationship of reduced and alkylated serum and fluid forms of SC with purified FSC were determined. SC in reduced and alkylated serum and cyst fluid samples was indistinguishable from purified FSC, on the basis of the results shown in figure I and text-figure 2. Immunodiffusion in agar against rabbit anti-SC gave a single fused precipitin band for purified FSC and for patients' sera and cyst fluids containing high levels of this protein. Radioimmunoassay studies gave competition curves of identical slope for purified FSC and for similar serum and cyst fluids. We were not able to test ascitic and serous cyst fluids in the same manner, inasmuch as these materials contained low levels of Sc. However, on the basis of the specificity of the antisera and the above results, we conclude that SC is, in fact, what is measured. We cannot make any conclusions with respect to the molecular form of SC (bound or free) in the various fluid samples. We suspect the ratio of bound to free may vary from sample to sample depending on a number of factors. We chose, therefore, to equalize results by reduction and alkylation of all samples to give a preponderance of FSC and to use an assay system based on the detection of FSC. We were able to show by radioimmunoassay the presence of SC in mucinous cyst fluids (l81.50±50.40 /-Lg/ml) as well as in serous cyst fluids (1.l3±O.37 /-Lg/ml) and in ascitic fluids (1.24±O.24 ~g/ml). Serous cyst fluids and ascitic fluids had always given negative results when the less sensitive immunodiffusion techniques were used. The very high levels of SC found in the mucinous cyst fluids by radioimmune techniques agreed with our earlier data (13) and suggested the screening of serum samples from normal volunteers and patients with ovarian adenocarcinomas. As expected, the SC levels of patients with mucinous adenocarcinomas were elevated when compared to those of the normal volunteers (22.44±3.24 /-Lg/ml vs. 2.34±0.41 /-Lg/ml). An interesting finding was the elevated serum levels in almost all ovarian cancer patients and not just in those patients whose adenocarcinoma was diagnosed as mucinous. Serum samples from patients with serous, papillary, and poorly differentiated adenocarcinomas had elevated levels (9.93±1.68, 7.35±1.l3, and 1O.1O±1.58 /-Lg/ml, respectively). These findings may not be con-
J
NATL CANCER INST
60 Klein, Gall, and Dawson
sis tent with the hypothesis that the cyst fluid and, therefore, the tumor is the source of the elevated serum levels of SC, but they suggest the existence of a secretory immune response to the tumor or to secondary infection. If we could show that a significant proportion of FSC is in sera from patients with mucinous ovarian tumors, this would argue in favor of the tumor as the source of SC. However, in preliminary studies, we have determined that, after Sephadex G-200 column chromatography of unreduced serum samples from patients with mucinous ovarian tumors, the SC is exclusively associated with IgA (Dawson JR, Gall SA: Unpublished results). Because of these findings, it is not possible to determine if the elevated serum levels of SC in patients with mucinous adenocarcinoma are the result of extremely high levels found at the tumor site. Studies are in progress to determine if the stage of the disease and the prognosis correlate with the serum levels of SC in patients with mucinous ovarian tumors.
REFERENCES (1) HEREMANS JF: Immunoglobulin A. In The Antigens (Sela M, ed), vol 2. New York: Academic Press, 1974, pp 365-522 (2) TOURVILLE DR, ADLER RH, BIENENSTOCK J, et al: The human secretory immunoglobulin system: Immunohistochemical localization of IgA, secretory piece and lactoferrin in normal
FIGURE 1.- Immunodiffusion patterns of sera and cyst fluid samples. Samples (10 #,1) were placed in each of the six wells surrounding the center well. Clockwise from the top: purified FSC, serum, cyst fluid, purified FSC, cyst fluid, and serum. Ten microliters of rabbit anti-SC was placed in the center well. Slides were incubated at room temperature in a humid chamber for 48 hr and then washed with PBS for 48 hr and distilled water for an additional 24 hr. Slides were air dried for 24-48 hr and stained with amido black.
human tissues. J Exp Med 129:411-430, 1969
(3) BRANDTZAEG P: Mucosal and glandular distribution of immu-
noglobulin components. Immunology 26:1101-1114, 1974 (4) HUANG S·W, FOGH J, HONG R: Synthesis of secretory component by human colon cancer celIs. Scand J Immunol 5:263-268, 1976 (5) HARRIS JP, CALEB MH, SoUTH MA: Secretory component in human mammary carcinoma. Cancer Res 35:1861-1864, 1975 (6) KOBAYASHI K: Studies on human secretory IgA: Comparative studies on the IgA-bound secretory piece and the free secretory piece protein. Immunochemistry 8:785-800, 1971 (7) LOWRY OH, ROSEBROUGH NJ, FARR AL, et al: Protein measurement with the Folin phenol reagent. J Bioi Chem 193:265-275, 1951 (8) VAN MUNSTER PJ, STORINGA GB, POELS·Z\NDERS S: Isolation of free secretory component (S.c.) from human milk, determination of its molecular weight. Immunochemistry 8:471477, 1971 (9) MAIZEL JV: Polyacrylamide gel electrophoresis of viral proteins. In Methods in Virology (Maramorosch K, Koprowski H, eds), vol 5. New York: Academic Press, 1971, pp 180-246 (10) BOLTON AE, HUNTER WM: The labeling of proteins to high specific radioactivities by conjugation to a 1251-containing acylating agent. Biochem J 133:529-539, 1973 (11) HUNTER WM: Radioimmunoassay. In Handbook of Experimental Immunology (Weir OM, ed), 2d ed. Oxford: BlackwelI Scientific Publications, 1973, pp 17.1-17.36 (12) GHETIE V, MOTA G: The decrease of human colostral immunoglobulin A resistance to papain action a(ter gradual release of the secretory component. Immunochemistry 10:839-840, 1973 (13) GARCIA JA, KLEIN JL, KUTTEH WH; et al: Immunological studies on the cystic effusions of ovarian epithelial neoplasms. Am J Obstet Gynecol 129:281-284, 1977
(
1
J
NATL CANCER INST
Downloaded from https://academic.oup.com/jnci/article-abstract/61/1/57/917897 by INSEAD user on 05 March 2018
VOL. 61, NO. I, JULY 1978
3
Stanley A. Gall,
4
and Jeffrey R. Dawson
ABSTRACT -A secretory component (SC) was detected by radioimmunoassay In the cyst fluids, ascitic fluids, and sera from patients with ovarian adenocarcinomas. Serous cyst fluids and ascitic fluids showed lower levels (expressed as means± SE) of SC (1.37±O.37 and 1.24±0.24 "glml, respectively) than mucinous cyst fluids (181.50±50.40 "g/ml). SC levels In the sera of all patients with ovarian adenocarcinoma were high (12.67±1.43 "glml) when compared to SC levels In the sera of normal Individuals (2.34±0.41 "g/ml). Sera from patients with ovarian cancers diagnosed as serous, mucinous, papillary, and poorly differentiated adenocarcinomas showed SC levels of 9.93±1.68, 22.44±3.24, 7.35±1.13, and 10.10±1.58 >Lg/ml, respectlvely.-J Natl Cancer Inst 61: 57-60, 1978.
Secretory immunoglobulin A is the predominant immunoglobulin found in external secretions. Generally, the structure consists of a dimeric IgA, j-chain, and SC, a glycoprotein with a molecular weight of approximately 70,000 (1). IgA and j-chain are synthesized by specific plasma cells, whereas SC is synthesized by epithelial cells lining the glandular lumen (1-3). Recently, investigators have shown the presence of FSC associated with several forms of epithelial cancer. Huang et al. (4), using immunofluorescence microscopy and radioactive labeling techniques, showed that three cell lines of human colon cancer can synthesize SC in vitro. Harris and co-workers (5), also using immunofluorescence microscopy for FSC, found intense staining of tumor cells and stromal cells in tissue samples from patients with breast cancer. This observation contrasted with the staining of only ductule epithelial cells in sections of normal breast tissue. Abnormal staining patterns were also observed in colon, lung, and bladder carcinomas. Our laboratory has developed a radioimmunoassay for SC and has used it to determine the SC levels in the sera, cyst fluids, and ascitic fluids of patients with ovarian adenocarcinomas. Preliminary studies revealed significantly elevated levels of SC in these patients compared to normal individuals.
hyde-insolubilized human serum albumin (8). The purified FSC gave a single band by sodium dodecyl sulfate polyacrylamide gel electrophoresis (9) (data not shown). Production of antiserum to SC.-Rabbits were given sc injections of 1 mg of FSC in complete Freund's adjuvant weekly for 6 weeks. After a rest period of 4 weeks, they were boosted with 1 mg of FSC in complete Freund's adjuvant. Serum was collected 8 days later and absorbed with 120 mg lyophilized normal human plasma per ml of serum. The antiserum was tested for anti-SC activity by immunodiffusion, aliquoted, and frozen at -20° C. It gave a single line of identity with commercially prepared antiserum (Behring Diagnostics) when tested against purified FSC, sera containing high concentrations of SC, and ovarian cyst fluids (data not shown). Labeling of FSC.-FSC was labeled with 1251 with the use of 100 p.Ci Bolton-Hunter reagent (New England Nuclear, Boston, Mass.) per 85 p.g protein (10). The labeled FSC was separated from free iodine reagent on a Sephadex G-25 column equilibrated with dextrose gelatin-Veronal buffer (Grand Island Biological Co., Grand Island, N.Y.). Radioimmunoprecipitation assay.- The general procedures for radioimmunoassays outlined by Hunter (11) were used to develop a suitable assay system. 6 Antiserum to SC was titrated by immunoprecipitation against 1251_ labeled FSC. The antiserum was serially diluted with PBS, and normal rabbit serum was added to a final serum concentration of 10%. Five microliters of diluted antiserum was added to 25 #,1 of labeled FSC (=75 ng) and incubated at 37° C for 1 hour. Ten microliters of goat anti-rabbit IgG (Miles Laboratories, Inc., Elkhart, Ind.) was added for maximum precipitation, and the mixtures were incubated at 4° C overnight. The total contents of each assay mixture were counted, the precipitates were washed four times with PBS, and the contents were then counted again in a Nuclear-Chicago gamma counting system (1l85 series; Nuclear-Chicago
MATERIALS AND METHODS
ABBREVIATIONS USED: FSC=free secretory component; PBS=phos phate-buHered saline; SC=secretory component.
Isolation of FSC.-FSC was purified from human colostrum by the procedure of Kobayashi (6), with the use of a combination of ammonium sulfate fractionation, ion exchange chromatography, and gel filtration. Protein fractions were followed spectrophotometrically at 280 nm. Protein concentrations were determined by the method of Lowry et al. (7). Samples containing protein were tested for the presence of FSC by immunodiffusion against rabbit anti-human FSC (Behring Diagnostics, Somerville, N.J.). Lactoferrin, which copurifies with FSC, was removed by two adsorptions with glutaraldeVOL. 61, NO. I, JULY 1978 Downloaded from https://academic.oup.com/jnci/article-abstract/61/1/57/917897 by INSEAD user on 05 March 2018
3. 5
Received September 28. 1977; accepted February 12, 1978. Supported by grant IM-94 from the American Cancer Society and by Public Health Service training grant CA09058 from the National Cancer Institute. S Division of Immunology, Department of Microbiology and Immunology, Duke University Medical Center, Durham, N.C. 27710. • Department of Obstetrics and Gynecology, Duke University Medical Center. 5 Address reprint requests to Dr. Dawson. 6 Klein JL, Dawson JR: Submitted for publication. I
2
57
J
NATL CANCER INST
58 Klein, Gall, and Dawson
Corp., Des Plaines, Ill.). The counts precipitated are expressed as the percent total counts in the assay system. A standard precipitation curve was established with the use of purified FSC samples serially diluted with PBS (concentrations of 10 J.Lg/ml to 0.16 J.Lg/ml). Five microliters of anti-SC (diluted 1/128 with PBS plus 10% normal rabbit serum) was added to 25 J.Ll of a sample containing known concentrations of FSC. Twenty-five microliters of 125I-Iabeled FSC was added, and the mixture treated as described above. Serum samples to be tested were obtained from patients and normal volunteers and stored at -200 C. All samples, prior to use, were reduced with 20 mM 2-mercaptoethanol for 1 hour at room temperature and alkylated with 25 mM iodoacetate for an additional hour to dissociate SC from immunoglobulins and to prevent nonspecific binding of FSC to serum proteins and immunoglobulins (12). Samples (25 J.LI) of serum diluted 1: I with PBS were tested as described below. SC concentrations were determined from the standard precipitation curve. Cyst fluids from ovarian adenocarcinoma and ascitic fluids associated with ovarian adenocarcinoma were centrifuged at 2,OOOXg for 20 minutes to remove cells and solid material and were stored at -200 C. These samples were also reduced and alkylated prior to use. Immunodiffusion studies on serum and fluid samples.-Immunodiffusion studies were done on glass microscope slides coated with 1.5% Difco Noble agar (Difco Laboratories, Detroit, Mich.) in PBS, pH 7.2, containing 0.02% sodium azide as a preservative. Six wells were cut into the agar in a hexagonal pattern equidistant from a seventh central well. Ten microliters
l.-SC levels (expressed in mean±SE) in normal volunteers and ovarian cancer patients. Normal sera2.34±O.41 j.lg/ml; ascitic f1uid1.24±O.24 /lg/ml; serous cyst f1uidI.13±O.37 /lg/ml; mucinous cyst f1uid181.5o±50.40 /lg/ml; ovarian adenocar· cinoma sera-12.67±1.43/lg/ml; serous adenocarcinoma sera-9.93±1.68 /lg/ ml; mucinous adenocarcinoma sera22.44±3.24 /lg/ml; papillary adenocar· cinoma sera-7.35±1.13 j.lg/ml; and poorly differentiated adenocarcinoma sera-IO.IO±1.58 /lg/ml. FSP=free secretory piece, an alternative designation for FSC.
TEXT·FlGURE
of each sample to be tested was placed in one of the six peripheral wells, and 10 J.LI of rabbit anti-SC was placed in the middle well. Diffusion was carried out for 48 hours at room temperature in a moist covered chamber to pn;vent drying. The slides were washed with PBS at room temperature for 48 hours, with a change of PBS at 24 hours, and washed with distilled water for an additional 24 hours. The slides were then air dried for 24-48 hours and stained with amido black. RESULTS SC Levels in Cyst and Ascitic Fluids
The SC levels in cyst and ascitic fluids of ovarian cancer patients are shown in text-figure I. Serous cyst fluids and ovarian adenocarcinoma ascitic fluids of unspecified histology contained low levels (expressed as means±sE) of SC (1.l3±0.37 and 1.24±0.24 J.Lg/ml. respectively). The levels were similar to those found in normal serum (2.34±0.41 J.Lg/ml). In contrast, the SC levels in mucinous cyst fluids were extremely high (181.50±50.40 J.Lg/ml). The SC levels in the sera of patients and normal volunteers are also shown in text-figure I. Normal volunteers had a low SC level in their sera (2.34±0.41 J.Lg/ml) compared to ovarian cancer patients (l2.67±1.43 J.Lg/ml). In the ovarian cancer patients, the levels were the highest when the diagnosis was mucinous adenocarcinoma (22.44±3.24 J.Lg/ml). The SC levels were also elevated in patients with serous, papillary, and poorly differentiated adenocarcinomas (9.93±1.68, 7.35±1.l3, and 1O.10±1.58 J.Lg/ml. respectively).
--
2!10 200 60
50
~
40
::::l
a:
....
,
'"
E .... 30 Q.
'"
..·
LL
'""-
· ..
20
--.~ .~
10
0
J
f
_'C_
NORMAL SERA
NATL CANCER (NST
Downloaded from https://academic.oup.com/jnci/article-abstract/61/1/57/917897 by INSEAD user on 05 March 2018
ASCITIC FLUID
SEROUS CYST FLUID
MUCINOUS OVARIAN CYST FLUID
.
-.-
...
--'-
SEROUS
MUCINOUS PAPillARY POORLY
CARCINOMA SERA
,
DIFFERENTIATED
VOL. 61, NO. I, JULY 1978
Elevated Secretory Component In Ovarian Adenocarcinoma 59
identical competition curves over a wide range of dilutions. 40
DISCUSSION
30
a w ~
~
U ~
20
Q.
C/l
IZ :::l
o
U
10
O~~
256
__~__~__~~__~__~______________~ 128
64
32
16
8
4
2
II DILUTION
TEXT· FIGURE 2.-Radioimmunoassay compellllon curves. Purified FSC, mucinous cyst fluid, and serum samples were diluted with PBS. The radioimmunoassay was run as described in "Materials and Methods." The counts precipitated are expressed as a percent of the counts in the assay system. All points were done in triplicate with less than I % variance within a group. X----X and 0-----0 indicate mucinous cyst fluids; t:r------6 and ~, serum samples; .... --e, purified FSC.
Immunodiffusion Patterns of Serum and Cyst Fluid Samples
Immunodiffusion patterns of reduced and alkylated serum and cyst fluid samples are shown in figure 1. Two serum samples from ovarian cancer patients containing high levels of SC, two mucinous cyst fluid samples, and purified FSC were used in each study. Rabbit anti-SC was placed in the center wells. A single line of identity was found among all samples tested and purified FSC. Serum samples from patients with all four types of ovarian adenocarcinomas (text-fig. I) gave identical results. Radioimmunoassay Competition Curve of Serum and Cyst Fluid Samples
Radioimmunoassay competition curves of serum and cyst fluid samples are shown in text-figure 2. A competition curve with the use of unlabeled purified FSC is also shown. All samples were serially diluted with PBS for the assay. Two mucinous cyst fluids, two serum samples, and purified FSC gave essentially VOL. 61. NO. I, JULY 1978 Downloaded from https://academic.oup.com/jnci/article-abstract/61/1/57/917897 by INSEAD user on 05 March 2018
The findings of Huang et al. (4), with cultured cell lines of colon cancer, and Harris et al. (5), with speCImens of breast cancer tissue, suggested that levels of SC might be elevated in the sera of patients with ovarian cancer. Preliminary screening of cyst fluids from ovarian cancer patients for SC by immunodiffusion in our laboratory (13) revealed that the detection of SC in the cyst fluids correlated with a diagnosis of mucinous cystadenoma or cystadenocarcinoma. For this reason, we developed a more sensitive radioimmunoassay for detecting Sc. 6 In addition, the specificity of the antisera and the immunologic relationship of reduced and alkylated serum and fluid forms of SC with purified FSC were determined. SC in reduced and alkylated serum and cyst fluid samples was indistinguishable from purified FSC, on the basis of the results shown in figure I and text-figure 2. Immunodiffusion in agar against rabbit anti-SC gave a single fused precipitin band for purified FSC and for patients' sera and cyst fluids containing high levels of this protein. Radioimmunoassay studies gave competition curves of identical slope for purified FSC and for similar serum and cyst fluids. We were not able to test ascitic and serous cyst fluids in the same manner, inasmuch as these materials contained low levels of Sc. However, on the basis of the specificity of the antisera and the above results, we conclude that SC is, in fact, what is measured. We cannot make any conclusions with respect to the molecular form of SC (bound or free) in the various fluid samples. We suspect the ratio of bound to free may vary from sample to sample depending on a number of factors. We chose, therefore, to equalize results by reduction and alkylation of all samples to give a preponderance of FSC and to use an assay system based on the detection of FSC. We were able to show by radioimmunoassay the presence of SC in mucinous cyst fluids (l81.50±50.40 /-Lg/ml) as well as in serous cyst fluids (1.l3±O.37 /-Lg/ml) and in ascitic fluids (1.24±O.24 ~g/ml). Serous cyst fluids and ascitic fluids had always given negative results when the less sensitive immunodiffusion techniques were used. The very high levels of SC found in the mucinous cyst fluids by radioimmune techniques agreed with our earlier data (13) and suggested the screening of serum samples from normal volunteers and patients with ovarian adenocarcinomas. As expected, the SC levels of patients with mucinous adenocarcinomas were elevated when compared to those of the normal volunteers (22.44±3.24 /-Lg/ml vs. 2.34±0.41 /-Lg/ml). An interesting finding was the elevated serum levels in almost all ovarian cancer patients and not just in those patients whose adenocarcinoma was diagnosed as mucinous. Serum samples from patients with serous, papillary, and poorly differentiated adenocarcinomas had elevated levels (9.93±1.68, 7.35±1.l3, and 1O.1O±1.58 /-Lg/ml, respectively). These findings may not be con-
J
NATL CANCER INST
60 Klein, Gall, and Dawson
sis tent with the hypothesis that the cyst fluid and, therefore, the tumor is the source of the elevated serum levels of SC, but they suggest the existence of a secretory immune response to the tumor or to secondary infection. If we could show that a significant proportion of FSC is in sera from patients with mucinous ovarian tumors, this would argue in favor of the tumor as the source of SC. However, in preliminary studies, we have determined that, after Sephadex G-200 column chromatography of unreduced serum samples from patients with mucinous ovarian tumors, the SC is exclusively associated with IgA (Dawson JR, Gall SA: Unpublished results). Because of these findings, it is not possible to determine if the elevated serum levels of SC in patients with mucinous adenocarcinoma are the result of extremely high levels found at the tumor site. Studies are in progress to determine if the stage of the disease and the prognosis correlate with the serum levels of SC in patients with mucinous ovarian tumors.
REFERENCES (1) HEREMANS JF: Immunoglobulin A. In The Antigens (Sela M, ed), vol 2. New York: Academic Press, 1974, pp 365-522 (2) TOURVILLE DR, ADLER RH, BIENENSTOCK J, et al: The human secretory immunoglobulin system: Immunohistochemical localization of IgA, secretory piece and lactoferrin in normal
FIGURE 1.- Immunodiffusion patterns of sera and cyst fluid samples. Samples (10 #,1) were placed in each of the six wells surrounding the center well. Clockwise from the top: purified FSC, serum, cyst fluid, purified FSC, cyst fluid, and serum. Ten microliters of rabbit anti-SC was placed in the center well. Slides were incubated at room temperature in a humid chamber for 48 hr and then washed with PBS for 48 hr and distilled water for an additional 24 hr. Slides were air dried for 24-48 hr and stained with amido black.
human tissues. J Exp Med 129:411-430, 1969
(3) BRANDTZAEG P: Mucosal and glandular distribution of immu-
noglobulin components. Immunology 26:1101-1114, 1974 (4) HUANG S·W, FOGH J, HONG R: Synthesis of secretory component by human colon cancer celIs. Scand J Immunol 5:263-268, 1976 (5) HARRIS JP, CALEB MH, SoUTH MA: Secretory component in human mammary carcinoma. Cancer Res 35:1861-1864, 1975 (6) KOBAYASHI K: Studies on human secretory IgA: Comparative studies on the IgA-bound secretory piece and the free secretory piece protein. Immunochemistry 8:785-800, 1971 (7) LOWRY OH, ROSEBROUGH NJ, FARR AL, et al: Protein measurement with the Folin phenol reagent. J Bioi Chem 193:265-275, 1951 (8) VAN MUNSTER PJ, STORINGA GB, POELS·Z\NDERS S: Isolation of free secretory component (S.c.) from human milk, determination of its molecular weight. Immunochemistry 8:471477, 1971 (9) MAIZEL JV: Polyacrylamide gel electrophoresis of viral proteins. In Methods in Virology (Maramorosch K, Koprowski H, eds), vol 5. New York: Academic Press, 1971, pp 180-246 (10) BOLTON AE, HUNTER WM: The labeling of proteins to high specific radioactivities by conjugation to a 1251-containing acylating agent. Biochem J 133:529-539, 1973 (11) HUNTER WM: Radioimmunoassay. In Handbook of Experimental Immunology (Weir OM, ed), 2d ed. Oxford: BlackwelI Scientific Publications, 1973, pp 17.1-17.36 (12) GHETIE V, MOTA G: The decrease of human colostral immunoglobulin A resistance to papain action a(ter gradual release of the secretory component. Immunochemistry 10:839-840, 1973 (13) GARCIA JA, KLEIN JL, KUTTEH WH; et al: Immunological studies on the cystic effusions of ovarian epithelial neoplasms. Am J Obstet Gynecol 129:281-284, 1977
(
1
J
NATL CANCER INST
Downloaded from https://academic.oup.com/jnci/article-abstract/61/1/57/917897 by INSEAD user on 05 March 2018
VOL. 61, NO. I, JULY 1978
