elevation of the placental glutathione s-transferase form ... · kidney, pancreas, and lung (18),...

6
[CANCER RESEARCH 49, 5225-5229, September IS, 1989] Elevation of the Placental Glutathione S-Transferase Form (GST-7r) in Tumor Tissues and the Levels in Sera of Patients with Cancer1 Shigeki Tsuchida, Yoshihito Sekine, Ryuzaburo Shineha, Tetsuro Nishihira, and Kiyomi Sato2 Second Department of Biochemistry, Hirosaki University School of Medicine, 5 Zaifu-cho, Hirosaki 036 ¡S.T., K. S.], and the Second Department of Surger); Tohoku University School of Medicine, 1 Seiryo-cho, Sendai 980, Japan ¡Y.S., R. S., T. N.] ABSTRACT The content of glutathione .S-transferase placenta! form (GST-T) in colon and esophageal cancer tissues was determined by single radial immunodiffusion or activity inhibition tests. Average levels in colon adenomas and carcinomas were 163±38 and 143±30 MK/K. respectively, about 6-fold higher than the value of 25 ±4 /ig/g observed for normal colonie mucosa. The content in esophageal cancer was similarly increased at 240 ±160MU/S.about 6-fold higher than the level found in normal mucosa, 42 ±21 /ig/g. The content was significantly higher in highly differentiated carcinoma (331 ±138 itg/g) than in moderately (205 ± 123) or poorly (125 ±61) differentiated carcinomas, suggesting that GST-T content seems to be related to the degree of differentiation of esophageal cancer. GST-T was also expressed in cell lines derived from various cancers, including 1MR 32, TE-9, and Ca Ski cells. The results thus indicate that GST-T may be a useful marker for a wide range of cancers. An enzyme-linked immunosorbent assay developed to determine serum GST-ir content is described. Using this method GST-T could be accu rately measured in the range between 0.7 and 150 ng/ml, without inter ference of other isoenzymes. The serum GST-ir content was 1.3 ±1.2 ng/ml in 35 healthy controls and values of over 3.7 ng/ml (control mean + 2 SD) were found in patients with cancers of the stomach (10 of 23 cases), esophagus (26/43), bile duct (3/9), and colon (3/9), and in some leukemic cases. Although elevation of the serum content was not so often as that of tissue content, the fact that higher serum values of patients with esophageal cancer often reverted to the normal range after surgical removal of the cancer suggested a direct derivation of serum GST-T from tumor tissues. Thus, follow-up of elevated serum GST-T levels may be useful for monitoring cancer patients during the course of treatment. INTRODUCTION Previous studies from our laboratory showed that GST-P1 which increases 30- to 50-fold during rat chemical hepatocar- cinogenesis is localized in hepatic hyperplastic nodules and hepatoma tissues (1, 2). We also demonstrated that GST-ir, which is related to rat GST-P in enzymatic and immunological properties (2, 3), may be a possible immunohistochemical marker for cancers of the colon (4) and uterine cervix (5). Recent reports from other laboratories have further showed strong expression of GST-ir or an anionic GST in gastric cancer (6), malignant melanoma (7), lung cancer (8, 9), and a range of other human tumors (10, for review, see Refs. 11 and 12). The expression of GST-ir has been also noted in multidrug-resistant human breast cancer cells (13, 14) and in a cis-dichlorodiam- mineplatinum(II)-resistant human squamous carcinoma cell line (15, 16, for review see Ref. 17). However, since significant Received 2/3/89; revised 6/5/89; accepted 6/16/89. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. ' This study was supported in part by a Grant-in Aid for Cancer Research from the Ministry of Education, Science and Culture of Japan, and a research grant from the Princess Takamatsu Cancer Research Fund. 2To whom requests for reprints should be addressed. 3The abbreviations used are: GST-P, rat placenta! form of glutathione S- transferase; GST-T, human placenta! form of glutathione 5-transferase; ELISA, enzyme-linked immunosorbent assay; PBS, phosphate-buffered saline consisting of 12 mM phosphate buffer, pH 7.2, and 14S HIM NaCI; BSA, bovine serum albumin. amounts of GST-*- are present in many organs, including the kidney, pancreas, and lung (18), under normal conditions, com parison of expression in cancer and normal tissues is required to allow evaluation of the potential of GST-ir as a tumor marker. Several isoenzymes of GST have been isolated from human tissues (3), and the specific substrates for GST-ir have not been known. However, as GST-ir is immunologically different from other isoenzymes (19), protein content in cancer tissues or cultured cell lines can be accurately determined by immunolog ical methods and compared with corresponding values for nor mal tissues. A particularly sensitive method using enzyme- linked immunosorbent assay has been developed in our labo ratory for determination of GST-?r levels in sera of patients with malignant tumors, as described in the present paper. MATERIALS AND METHODS Reagents. Polybuffer exchanger PBE 94 and Polybuffer 74 were obtained from Pharmacia (Uppsala, Sweden); Tween 20 and goat anti- rabbit IgG conjugated with horseradish peroxidase were from Bio-Rad (Richmond, CA); bovine serum albumin and 2,2'-azinobis(3-ethylben- zothiazoline-6-sulfonic acid) were from Wako Pure Chemicals (Osaka, Japan). The other reagents were of analytical grade. Tissue Samples. Adenomatous polyps of the colon were obtained endoscopically. Cancer specimens for the esophagus and colon were obtained at surgery or at autopsy. Normal tissues adjacent to the cancers were used as controls. After dissection, all tissues were stored at -80°C until use. Cell Culture. Cell lines, TE-1 to 12, established from human squa mous carcinomas of the esophagus (20), were grown in monolayers in a medium consisting of RPMI 1640 supplemented with 7% fetal calf serum and antibiotics. These cell lines have a doubling time of 29-120 h in vitro. Cells in stationary phase were harvested and used for determination of GST-ir content. The other cell lines used were gener ous gifts of Prof. Tomihisa Funyu, Department of Urology, Hirosaki University, or Dr. Takashi Tsuruo, Cancer Chemotherapy Center, Cancer Institute, Tokyo, Japan. Sera from Patients and Healthy Controls. Serum samples were ob tained from healthy volunteers and from patients prior to treatment. Most of the cancer patients were at Stages III or IV. The sera of 29 patients with esophageal cancer were collected before and 10-30 days after operation. Clinical diagnoses, which were confirmed by histopath- ological examinations, were unknown to the authors prior to the assay for GST-ir content. The sera used for the ELISA assay were confirmed to be free from hemolysis by single radial immunodiffusion using anti- human hemoglobin antibody (21). GST Assay. GST activity was assayed with l-chloro-2,4-dinitroben- zene as a substrate according to the method of Habig et al. (22). Antibody Preparations. GST-ir was purified from human placenta as reported previously (5, 19). Anti-GST-ir antisera were raised in rabbits or guinea pigs as described earlier (23). ^-Globulin was prepared from the antisera by ammonium sulfate fractionation (20-33% saturation). Specificity of the rabbit antibody for GST-?r was confirmed by analysis of the cytosolic fraction from human placenta by an immunoblotting method (24). Quantitation of Tissue and Cell GST-T Levels. GST-ir levels in colon or esophageal cancers and normal mucosae were determined by the immunoprecipitation or single radial immunodiffusion method (18). 5225 on April 9, 2020. © 1989 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Upload: others

Post on 03-Apr-2020

2 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Elevation of the Placental Glutathione S-Transferase Form ... · kidney, pancreas, and lung (18), under normal conditions, com parison of expression in cancer and normal tissues is

[CANCER RESEARCH 49, 5225-5229, September IS, 1989]

Elevation of the Placental Glutathione S-Transferase Form (GST-7r) in TumorTissues and the Levels in Sera of Patients with Cancer1

Shigeki Tsuchida, Yoshihito Sekine, Ryuzaburo Shineha, Tetsuro Nishihira, and Kiyomi Sato2

Second Department of Biochemistry, Hirosaki University School of Medicine, 5 Zaifu-cho, Hirosaki 036 ¡S.T., K. S.], and the Second Department of Surger); TohokuUniversity School of Medicine, 1 Seiryo-cho, Sendai 980, Japan ¡Y.S., R. S., T. N.]

ABSTRACT

The content of glutathione .S-transferase placenta! form (GST-T) incolon and esophageal cancer tissues was determined by single radialimmunodiffusion or activity inhibition tests. Average levels in colonadenomas and carcinomas were 163 ±38 and 143 ±30 MK/K.respectively,about 6-fold higher than the value of 25 ±4 /ig/g observed for normalcolonie mucosa. The content in esophageal cancer was similarly increasedat 240 ±160 MU/S.about 6-fold higher than the level found in normalmucosa, 42 ±21 /ig/g. The content was significantly higher in highlydifferentiated carcinoma (331 ±138 itg/g) than in moderately (205 ±123) or poorly (125 ±61) differentiated carcinomas, suggesting thatGST-T content seems to be related to the degree of differentiation ofesophageal cancer. GST-T was also expressed in cell lines derived fromvarious cancers, including 1MR 32, TE-9, and Ca Ski cells. The resultsthus indicate that GST-T may be a useful marker for a wide range ofcancers.

An enzyme-linked immunosorbent assay developed to determine serumGST-ir content is described. Using this method GST-T could be accurately measured in the range between 0.7 and 150 ng/ml, without interference of other isoenzymes. The serum GST-ir content was 1.3 ±1.2ng/ml in 35 healthy controls and values of over 3.7 ng/ml (control mean+ 2 SD) were found in patients with cancers of the stomach (10 of 23cases), esophagus (26/43), bile duct (3/9), and colon (3/9), and in someleukemic cases. Although elevation of the serum content was not so oftenas that of tissue content, the fact that higher serum values of patientswith esophageal cancer often reverted to the normal range after surgicalremoval of the cancer suggested a direct derivation of serum GST-T fromtumor tissues. Thus, follow-up of elevated serum GST-T levels may beuseful for monitoring cancer patients during the course of treatment.

INTRODUCTION

Previous studies from our laboratory showed that GST-P1

which increases 30- to 50-fold during rat chemical hepatocar-cinogenesis is localized in hepatic hyperplastic nodules andhepatoma tissues (1, 2). We also demonstrated that GST-ir,which is related to rat GST-P in enzymatic and immunologicalproperties (2, 3), may be a possible immunohistochemicalmarker for cancers of the colon (4) and uterine cervix (5).Recent reports from other laboratories have further showedstrong expression of GST-ir or an anionic GST in gastric cancer(6), malignant melanoma (7), lung cancer (8, 9), and a range ofother human tumors (10, for review, see Refs. 11 and 12). Theexpression of GST-ir has been also noted in multidrug-resistanthuman breast cancer cells (13, 14) and in a cis-dichlorodiam-mineplatinum(II)-resistant human squamous carcinoma cellline (15, 16, for review see Ref. 17). However, since significant

Received 2/3/89; revised 6/5/89; accepted 6/16/89.The costs of publication of this article were defrayed in part by the payment

of page charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

' This study was supported in part by a Grant-in Aid for Cancer Research

from the Ministry of Education, Science and Culture of Japan, and a researchgrant from the Princess Takamatsu Cancer Research Fund.

2To whom requests for reprints should be addressed.3The abbreviations used are: GST-P, rat placenta! form of glutathione S-

transferase; GST-T, human placenta! form of glutathione 5-transferase; ELISA,enzyme-linked immunosorbent assay; PBS, phosphate-buffered saline consistingof 12 mM phosphate buffer, pH 7.2, and 14S HIM NaCI; BSA, bovine serumalbumin.

amounts of GST-*- are present in many organs, including the

kidney, pancreas, and lung (18), under normal conditions, comparison of expression in cancer and normal tissues is requiredto allow evaluation of the potential of GST-ir as a tumormarker.

Several isoenzymes of GST have been isolated from humantissues (3), and the specific substrates for GST-ir have not beenknown. However, as GST-ir is immunologically different fromother isoenzymes (19), protein content in cancer tissues orcultured cell lines can be accurately determined by immunological methods and compared with corresponding values for normal tissues. A particularly sensitive method using enzyme-linked immunosorbent assay has been developed in our laboratory for determination of GST-?r levels in sera of patientswith malignant tumors, as described in the present paper.

MATERIALS AND METHODS

Reagents. Polybuffer exchanger PBE 94 and Polybuffer 74 wereobtained from Pharmacia (Uppsala, Sweden); Tween 20 and goat anti-rabbit IgG conjugated with horseradish peroxidase were from Bio-Rad(Richmond, CA); bovine serum albumin and 2,2'-azinobis(3-ethylben-zothiazoline-6-sulfonic acid) were from Wako Pure Chemicals (Osaka,Japan). The other reagents were of analytical grade.

Tissue Samples. Adenomatous polyps of the colon were obtainedendoscopically. Cancer specimens for the esophagus and colon wereobtained at surgery or at autopsy. Normal tissues adjacent to the cancerswere used as controls. After dissection, all tissues were stored at -80°C

until use.Cell Culture. Cell lines, TE-1 to 12, established from human squa

mous carcinomas of the esophagus (20), were grown in monolayers ina medium consisting of RPMI 1640 supplemented with 7% fetal calfserum and antibiotics. These cell lines have a doubling time of 29-120h in vitro. Cells in stationary phase were harvested and used fordetermination of GST-ir content. The other cell lines used were generous gifts of Prof. Tomihisa Funyu, Department of Urology, HirosakiUniversity, or Dr. Takashi Tsuruo, Cancer Chemotherapy Center,Cancer Institute, Tokyo, Japan.

Sera from Patients and Healthy Controls. Serum samples were obtained from healthy volunteers and from patients prior to treatment.Most of the cancer patients were at Stages III or IV. The sera of 29patients with esophageal cancer were collected before and 10-30 daysafter operation. Clinical diagnoses, which were confirmed by histopath-ological examinations, were unknown to the authors prior to the assayfor GST-ir content. The sera used for the ELISA assay were confirmedto be free from hemolysis by single radial immunodiffusion using anti-human hemoglobin antibody (21).

GST Assay. GST activity was assayed with l-chloro-2,4-dinitroben-zene as a substrate according to the method of Habig et al. (22).

Antibody Preparations. GST-ir was purified from human placenta asreported previously (5, 19). Anti-GST-ir antisera were raised in rabbitsor guinea pigs as described earlier (23). ^-Globulin was prepared fromthe antisera by ammonium sulfate fractionation (20-33% saturation).Specificity of the rabbit antibody for GST-?r was confirmed by analysisof the cytosolic fraction from human placenta by an immunoblottingmethod (24).

Quantitation of Tissue and Cell GST-T Levels. GST-ir levels in colonor esophageal cancers and normal mucosae were determined by theimmunoprecipitation or single radial immunodiffusion method (18).

5225

on April 9, 2020. © 1989 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Page 2: Elevation of the Placental Glutathione S-Transferase Form ... · kidney, pancreas, and lung (18), under normal conditions, com parison of expression in cancer and normal tissues is

ELEVATION OF GST-w IN HUMAN CANCERS

The tissues (0.1-0.5 g) were homogenized with 4 volumes of 10 HIMTris-HCl, pH 7.8, containing 0.2 M NaCI and centrifuged at 105,000x g for 45 min. The supernatant (final 0.06 unit/ml) was then incubatedwith the rabbit anti-GST-*- antibody at 4°Covernight. After incubation,

the GST activity in the supernatant obtained by centrifugation at 16,000x g for 15 min was assayed. In the presence of nonimmune 7-globulinused as controls, the GST activity did not decrease during incubation.GST-*- tissue content was calculated as follows. The activity precipi

tated with the antibody was divided by the known specific activity ofGST-ir (0.12 unit/jig protein) (18). Single radial immunodiffusion wasperformed according to Mancini et al. (25). GST-ir content in culturedcell lines was assayed as above after sonication for 30 s using a ModelUR-20P sonifier (Tomy Seiko Co., Ltd., Tokyo, Japan). Protein content was measured according to Lowry et al. (26) using bovine serumalbumin as a standard.

ELISA for the Determination of Serum GST-ir Level. ELISA wasperformed according to Engvall (27). Nunc TSP plates (A/S Nunc,Roskilde, Denmark) were incubated with 100 pi of PBS containingguinea pig anti-GST-ir antibody (15 Mgprotein/ml) at room temperature overnight. After washing twice with PBS, the plates were incubatedwith PBS containing 5% BSA and 5% bovine serum in order to blocknonspecific binding. The plates were washed five times with PBScontaining 0.05% Tween 20 (washing buffer), and then incubatedovernight with various amounts of purified GST-T in PBS containing5% BSA or with test sera. After washing five times, they were thenincubated for 2 h with rabbit anti-GST-7r antibody (7.5 ¿¿gprotein/ml)in PBS containing 5% BSA and 1% goat serum. They were againwashed, then incubated for l h with the goat anti-rabbit IgG antibodyconjugated with horseradish peroxidase, diluted 3,000-fold with PBScontaining 1% bovine serum. After washing, the peroxidase substrate,0.3 mivi2,2'-azino-bis(3-ethylbenzothiazoline-6-suIfonic acid) in 0.1 M

citrate buffer, pH 4.0, containing 0.03% H2O2was added and the platesincubated at room temperature for 30-60 min. The color intensity ofeach well was measured at 405 nm using a spectrophotometer formicroplates (MTP-22; Corona Electric Co., Katsuta, Ibaragi, Japan).The absorbance at 405 nm was 93-96% of that at 417 nm. Forintraassay precision, the coefficient of variation was 10% when 30individual measurements were done, while for interassay precision, thecoefficient among five individual measurements was 11%.

GST-a and p were purified from human liver as reported previously(19). GST-p was purified from human red blood cells by a similarprocedure to that for GST-ir.

Statistical Analysis. The statistical evaluations were made by Student's t or by x2 test.

RESULTS

GST-ir Content in Colon and Esophagus Cancer Tissues. Theresults of the immunoprecipitation test using anti-GST-7r antibody revealed that GST-ir accounted for about 60% of totalGST activity in normal colon and about 95% in the cancer; 80-90% in normal esophagus and 75-95% in esophageal cancer.The precipitation patterns of these enzymes were not differentfrom that of purified GST-ir from the placenta (data not shown).It was confirmed by the double immunodiffusion method thatthe major GST form in both colon and esophageal cancers wasimmunologically identical to the placenta! GST-ir (Fig. 1).Therefore, GST-ir content in the respective tissues was calculated as described in "Materials and Methods." Values for GST-

ir content in colon cancers and adenomatous polyps were 149±30 ng/g tissue, and 163 ±38, respectively, about 6-foldhigher than that for normal colon (Table 1). Similar valueswere also obtained by single radial immunodiffusion (data notshown).

In esophageal cancers GST-ir content of 38 specimens was240 ±160 ¿ig/gtissue, a 6-fold higher value than that observedfor normal esophagus tissue (Table 1). The relationship betweenGST-ir content and degree of differentiation was examined in

Fig. 1. Double immunodiffusion of extracts from colonie polyp and carcinomaand esophageal carcinoma against the anti-GST-ir antibody. In each case 30 n\ ofthe supernatant, after centrifugation at 10S.OOOx g for 45 min. was applied tothe well. /, GST-ir purified from human placenta (5 <ig);2, extract from coloniepolyp; 3, from colon cancer; 4 and 5, are from esophageal cancer; 6, frommalignant melanoma; 7, anti-GST-ir antibody ( 15 fig).

the 30 cancer specimens. GST-ir content is less in poorlydifferentiated carcinoma than in highly or moderately differentiated carcinomas (Table l, P < 0.05). The content seems tobe related to the degree of differentiation of esophageal cancer.This finding is similar to that of GST-P, a rat equivalent formof human GST-ir, in that it is expressed in primary hepatomasinduced by several carcinogens, but not expressed in dediffer-entiated transplantable Yoshida ascites hepatomas (2). Métastases in lymph nodes also show a high content, 200 ±150 pg/g tissue, comparable to that of primary cancers. All the esophageal cancer specimens examined (38 cases) had a high GST-7rcontent of over 50 /ug/g tissue. When total GST activity andGST-ir content were expressed per mg protein of cytosol, thesimilar extent of elevation was obtained as that expressed pergram tissue (Table 1). Immunoprecipitation with specific antibodies to GST-o or n revealed that these isoenzymes were alsoelevated in a few cases and accounted for 20-25% of the totalGST activity (data not shown). The results demonstrated a clearincrease of GST-ir content in colon and esophagus cancers aswell as in adenomatous polyps.

GST-ir Content in Cultured Cell Lines. Values for GST-ircontent in cultured cell lines derived from human cancers andleukemia are summarized in Table 2. IMR 32, TE-9, and CaSki cells showed very high values and other cell lines includingG 401, HT 1376, and TE-8 and those derived from leukemiasalso showed high values. GST-^r content in doxorubicin-resist-ant subclones of K562 cells (28) was 1.4-fold higher than inwild clones (Table 2). GST-ir was not detectable in ZR 75-1cells.

Quantitation of GST-ir Content by ELISA. Using the ELISAmethod described in the present paper, GST-ir purified fromthe human placenta could be measured at levels between 0.7and 150 ng/ml, whereas other GST isoenzymes, GST-a and n,were not detected up to 150 ng/ml (Fig. 2). GST-p from redblood cells gave the same standard curve as GST-ir and thusthe result confirmed that GST-ir does not differ immunologically from GST-p (29, 30). To examine whether or not thismethod is applicable to serum samples, sera with high GST-irlevels were diluted over a wide range, and their absorbance wascompared with that of purified GST-ir (Fig. 3). The slopesgained for the two sera were similar to that of a standard curve,indicating immunological identity of GST TT-related antigen(s)in serum with GST-ir from the placenta.

Serum GST-ir Content in Patients. Data for serum GST-ir

5226

on April 9, 2020. © 1989 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Page 3: Elevation of the Placental Glutathione S-Transferase Form ... · kidney, pancreas, and lung (18), under normal conditions, com parison of expression in cancer and normal tissues is

ELEVATION OF GST-* IN HUMAN CANCERS

Table 1 GST activities and GST-v content in cancer tissuesTotal GST activity and GST-?r content were measured in the supernatants obtained by centrifugation at 105.000 x g for 45 min as described in "Materials and

Methods."

Tissue

Total GST activity"

(units/g tissue)

GST-ir content"

(^g/g tissue)

ColonNormal (4)"

Adenoma (4)Carcinoma (4)

EsophagusNormal (28)Carcinoma (38)

Highly differentiated carcinoma ( 12)Moderately differentiated (12)Poorly differentiated (6)

Metastatic lymph nodes (24)

Malignant melanoma (1)

4.6 ±0.6(0.03 ±0.0If19.6 ±4.6(0.14 ±0.03)17.9 ±3.6 (0.13 ±0.03)

5.7 ±2.6 (0.04 ±0.02)31.3 ±21.8 (0.23 ±0.16)47.3 ±22.2 (0.35 ±0.16)27.6 ±17.5(0.21 ±0.14)16.5 ±7.0(0.12 ±0.05)27.3 ±21.4 (0.20 ±0.16)

80.1 (0.64)

25 ±4(0.17 ±0.03)''163±38(1.16 + 0.27)143±30 (1.06 ±0.22)

42 ±21(0.30±0.15)240±160(1.78±1.19)330±140(2.44+1.04)205+ 120(1.58+ 0.92)125±60(0.93±0.44)200±150(1.48±1.11)

670 (5.33)°Values are expressed as mean + SD.* Number of samples examined.e Total GST activity, units/mg protein of cytosol.d GST-jT content, /jg/mg protein of cytosol.

Table 2 GST activity and GST-w content in cultured cell lines

The assay was carried out as described in Table 1. Each value is the mean ofduplicate assays.

Cell lineGST activity

(units/mg protein)GST-jr content(»ig/mgprotein)

Esophageal cancerTE-8TE-9

NeuroblastomaIMR32

Breast cancerZR 75-1

Cervical cancerCaSki

Wilms' tumor

G 401Bladder cancer

HT 1376Acute lymphocytic leukemia

WildMethotrexate-resistantCytosine arabinoside-resistant

Chronic myelocytic leukemiaK 562 WildDoxorubiein-resistant

Acute promyelocytic leukemiaHL 60

0.080.52

1.00

0.01

0.38

0.18

0.17

0.110.100.01

0.210.34

0.25

0.704.25

6.67

0.08

3.15

1.50

1.39

0.92 (0.25)"

0.830.08

1.231.78

2.06°GST-ir content in mononuclear cells of peripheral blood.

0.8

0.6

0.4 -

0.2

07 1.0 1.6 10GST, ng/ml

25 60 150

Fig. 2. A calibration curve for ELISA of GST-ir. Each point is the mean oftriplicate assays. The procedures are described in the text. •¿�,GST-?r purifiedfrom human placenta; O, GST-p from red blood cells; A, GST-a and A, GST-ji.GST-a and ^ were purified from human liver.

610

content in healthy controls, cancer patients, and pregnantwomen are summarized in Fig. 4. As red blood cells containGST-p which does not immunologically differ from GST-ir(Fig. 2), only sera free from hemolysis were used (see "Materialsand Methods"). Serum GST-ir content was 1.3 ±1.2 (mean ±

5227

Fig. 3. Effect of dilution on absorbance of serum sample in the ELISA. Twosera with high GST-T content, A (A, 60 ng/ml) and B (A, 16 ng/ml) were dilutedup to 610-fold with PBS containing 5% BSA, and then assayed by the ELISA.The change in absorbance of serum samples was compared with that of thepurified GST-ir (•).Sera A and B were obtained from patients with gastric andbile duct cancers, respectively.

SD) ng/ml in 35 controls, 19 males and 16 females, aged 25-66 years (averages 45 years). There was no significant differencebetween the sexes and in age. Assuming the upper limit ofnormal serum content to, therefore, be 3.7 ng/ml (mean + 2SD), abnormally high levels were found in 10 of 23 patientswith gastric cancer, 26 of 43 with esophagus cancer, three ofnine with cancer of the bile duct or gall bladder, three of ninewith colon cancer, and in a small number of leukemic patients.None of the patients with breast cancer (four cases) or pancreatic cancer (6) showed values higher than 3.7 ng/ml. Thepositive rates in gastric and esophageal cancers were significantly higher than that in control (x2 = 7.79, P < 0.01; and

19.60, P < 0.001, respectively), but the values of other cancerswere not significant (P > 0.05). While nine of 10 patients withhepatocellular carcinoma exhibited normal levels, eight of 10with chronic hepatitis, and five of 10 with liver cirrhosis demonstrated values over 3.7 ng/ml. Of 29 pregnant women, 26had over 3.7 ng/ml. Serum GST-ir level was elevated at the10th week of pregnancy, the earliest time point examined, andcontent was not correlated with stage of pregnancy (data notshown).

GST-ir levels in sera from 29 patients whose esophagealcancers were surgically removed, were examined before andafter the operation. Of 19 cases whose GST-ir levels were highbefore the operation, nine showed marked decreases within thenormal range 10-30 days after the operation, and two other

on April 9, 2020. © 1989 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Page 4: Elevation of the Placental Glutathione S-Transferase Form ... · kidney, pancreas, and lung (18), under normal conditions, com parison of expression in cancer and normal tissues is

ELEVATION OF GST-ir IN HUMAN CANCERS

GST- -level in serum, ng/ml

10 20 30

Control (35)

1.3--12Gastric

cancer(23)Gall

bladder(9)Esophageal

(43)Colo-rectal

(9)Lung

(11)Breast

(4)Pancreatic

(6)ALL

(7)AML

(8)CML

(3)Myeloma

(6)Hepatoma

(10)Chronic

hepatitis(10)Liver

cirrhosis(10)Prr•Ij»!

*.?~|

i£I-M- •¿�i.j....

:*-••u.V-•1-*•••

•¿�t...4-•¿�•

* •¿�•¿�••¿�••¿�•¿�1

4MB* •¿� •¿� •¿�•¿�,

- , - 'O 84 160

Fig. 4. GST-ir level in sera of patients with various malignant tumors, patientswith nonmalignant diseases and pregnant women determined by ELISA. Thehatched area denotes the normal range of serum GST-ir defined as 0-3.7 ng/ml(mean + 2 SD of 35 healthy controls). Numbers in parentheses, numbers ofpatients examined. ALL, acute lymphocytic leukemia; AML. acute myelocyticleukemia; and CML, chronic myelocytic leukemia.

cases showed the lower postoperative values than the preoperative by 3 ng/ml or more. These results suggest that the highserum levels in patients are derived from the cancers. In 10cases whose preoperative levels were within the normal range,the level did not change significantly after the operation.

DISCUSSION

The present investigation revealed a 6-fold elevation of GST-Trcontent in colon and esophageal cancer tissues, as comparedto normal tissues (Table 1). Thus, the quantitative results givedirect support to the previous positive immunohistochemicalstaining of GST-Tr in colon cancer (4). The high content in mostcell lines derived from various cancers (Table 2) also indicatesa general localization of GST-Tr in cancer cells themselves, inline with recent immunohistochemical studies which showedexpression of GST-7r at high frequencies in adenocarcinomasof the pancreas, bile duct and breast, in squamous cell carcinoma of the skin, and in malignant melanoma.4 Thus, GST-Tr

appears to be a useful marker for a wide range of cancers.Overexpression of GST-7r was reported in multidrug-resistanthuman cancer cells (13-15). However, GST-ir content of dox-orubicin-resistant K 562 cells was only 1.4-fold higher than that

of the wild cells (Table 2).An ELISA has been developed to examine serum GST-Tr level

in cancer patients. This method allowed detection of GST-Trlevels higher than 0.7 ng/ml (Fig. 2), and was found to beapplicable to assessment of GST-Tr levels within a wide rangein sera (Fig. 3). Although GST-p of red blood cells, with acontent of 12-20 ng/mg of hemoglobin, is immunologicallyidentical with GST-ir (Fig. 2), contribution of GST-p to thevalues was calculated to be under 0.1-0.2 ng/ml in the sera andunder 0.15 /zg/g in the tissues used for this experiment. Sixty% of patients with esophageal cancer, 44% of those with gastriccancer, and 33% of those with bile duct and colon cancers

4 Manuscript in preparation.

demonstrated values over 3.7 ng/ml (mean + 2 SD of normalcontrol, Fig. 2). Although proportionally not as high as forlesions from the same organs showing positive immunohistochemical staining (4, 5), the results suggest that the high serumlevels are derived from the cancers. Furthermore, in some casesof esophageal cancer elevated serum levels reverted to thenormal range after surgical removal of the tumor. Follow-up ofserum GST-Tr levels may thus be useful for monitoring patientsduring the course of treatment.

Soma et al. (19) reported that GST-Tr content was elevatedin hepatic tumors, including metastatic tumors from the colonand stomach. However, GST-ir was found not be expressed atelevated levels in human primary hepatomas and serum GST-Trvalues were within the normal range in all but one of 10patients investigated in the present study (Fig. 4). Recently,Mouelhi et al. (31) reported that total GST activities are decreased in human hepatoma tissues. The abnormally high serumlevels were obtained for several patients with chronic hepatitisor liver cirrhosis (Fig. 4), and hepatic tissues in these cases wereheavily stained for GST-Tr by immunohistochemistry (data not

shown). It is well known that patients with liver cirrhosis are ahigh-risk group for the development of hepatocellular carcinoma (32). Thus the expression of GST-Tr in liver cirrhosis,colonie adenomatous polyps (Table 1) or dysplasia of the uterine cervix (5) indicated that it may be a useful marker not onlyfor some cancers but also for high risk precancerous lesions.

ACKNOWLEDGMENTS

We thank Professors Yutaka Yoshida and Tomihisa Funyu, of theFirst Department of Internal Medicine and the Department of Urology,Hirosaki University School of Medicine, Dr. Tadashi Aisawa, Department of Internal Medicine, Aomori Prefectura! Central Hospital forsupplying the sera or cultured cells, and Dr. Takashi I suruo. CancerChemotherapy Center, Cancer Institute, Tokyo, for providing K 562cells.

REFERENCES

1. Sato, K . Kitahara, A., Satoh, K.. Ishikawa, T., Tatematsu, M.. and Ito, N.The placenta! form of glutathione 5-transferase as a new marker protein forpreneoplasia in rat chemical hepatocarcinogenesis. Gann, 75:199-202,1984.

2. Satoh, K., Kitahara, A., Soma, Y., Inaba, Y., Hatayama, I., and Sato, K.Purification, induction, and distribution of placenta! glutathione transferase:a new marker enzyme for preneoplastic cells in the rat chemical hepatocarcinogenesis. Proc. Nati. Acad. Sci. USA, 82: 3964-3968, 1985.

3. Mannervik, B. The isoenzymes of glutathione transferase. Adv. Enzymol.Relat. Area Mo!. Biol., 57: 357-417, 1985.

4. Kodak-, C., Fukushi, A., Narita, T., Kudo, H., Soma, Y., and Sato, K. Humanplacenta! form of glutathione 5-transferase (GST-jr) as a new immunohistochemical marker for human colonie carcinoma. Gann, 77: 226-229, 1986.

5. Shiratori, Y., Soma, Y., Maruyama, H., Sato, S., Takano, A., and Sato, K.Immunohistochemical detection of glutathione S-transferase placenta! form(GST-ir) in dysplastic and neoplastic human uterine cervix lesions. CancerRes., 47: 6806-6809, 1987.

6. Tsutsumi, M., Sugisaki, T., Makino, T., Miyagi, N., Nakatani, K., Shiratori,T., Takahashi, S., and Konishi, Y. Oncofetal expression of glutathione Stransferase placenta! form in human stomach carcinomas. Gann, 78: 631-633, 1987.

7. Mannervik, B., Castro, V. M., Danielson, U. H., Tahir, M. K., Hansson, J.,and Ringborg, U. Expression of class Pi glutathione transferase in humanmalignant melanoma cells. Carcinogenesis (Lond.), 8: 1929-1932, 1987.

8. Di Ilio, C., Del Boccio, G., Aceto, A., Casaccia, R., Mudili. F., and Federici,G. Elevation of glutathione transferase activity in human lung tumor. Carcinogenesis (Lond.), 9: 335-340, 1988.

9. Eimoto, H., Tsutsumi, M., Nakajima, A., Yamamoto, K., Takashima, Y.,Maruyama, H., and Konishi, Y. Expression of the glutathione 5-transferaseplacenta! form in human lung carcinomas. Carcinogenesis (Lond.), 9: 2325-2327, 1988.

10. Shea, T. C., Kelley, S. L., and Henner, W. D. Identification of an anionicform of glutathione transferase present in many human tumors and humantumor cell lines. Cancer Res., 48: 527-533, 1988.

11. Sato, K. Glutathione 5-transferases and hepatocarcinogenesis. Gann, 79:556-572, 1988.

5228

on April 9, 2020. © 1989 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Page 5: Elevation of the Placental Glutathione S-Transferase Form ... · kidney, pancreas, and lung (18), under normal conditions, com parison of expression in cancer and normal tissues is

ELEVATION OF GST-ir IN HUMAN CANCERS

12. Sato. k. Glutathione transferases as markers of preneoplasia and neoplasia.Adv. Cancer Res., 52: 205-255, 1989.

13. Batist, G., Tulpule, A., Sinha. B. K., Katki. A. G., Myers. C. E.. and Cowan,K. H. Overproduction of a novel anionic glutathione transferase in multidrug-resistant human breast cancer cells. J. Biol. Chem., 261:15544-15549,1986.

14. Cowan, K. H., Batist, G., Tulpule, A.. Sinha, B. K., and Myers. C. E. Similarbiochemical changes associated with multidrug resistance in human breastcancer cells and carcinogen-induced resistance to xenobiolics in rat. Proc.Nati. Acad. Sci. USA, S3: 9328-9332. 1986.

15. Teicher, B. A., Holden, S. A.. Kelley, M. J., Shea, T. C., Cucchi, C. A..Rosowsky, A.. Henner, W. D., and Frei. E., III. Characterization of a humansquamous carcinoma cell line resistant to cis-diamminedichloroplatinum (II).Cancer Res., 47: 388-393, 1987.

16. Nakagawa, K., Yokota, J., Wada, M., Sasaki, Y., Fujiwara. Y., Sakai, M.,Muramatsu, M., Terasaki, T.. Tsunokawa, Y.. Terada, M., and Saijo, N.Levels of glutathione S-transferase mRNA in human lung cancer cell linescorrelate with the resistance to cisplatin and carboplatin. Gann, 79: 301-304. 1988.

17. Moscow, J. A., and Cowan, K. H. Multidrug resistance. J. Nati. Cancer Inst.,80: 14-20, 1988.

18. Tateoka. N., Tsuchida, S., Soma, Y., and Sato, K. Purification and characterization of glutathione .S-transferases in human kidney. Clin. Chim. Acta.766:207-218, 1987.

19. Soma, Y., Satoh, K.. and Sato, K. Purification and subunit structural andimmunological characterization of five glutathione 5-transferases in humanliver, and the acidic form as a hepatic tumor marker. Biochim. Biophys.Acta, «69:247-258, 1986.

20. Nishihira, T., Kasai, M., Mori. S., Watanabe, T., Kuriya. Y., Suda, M.,Kitamura. M., Hirayama, K., Akaishi, T., and Sasaki, T. Characteristics oftwo cell lines (II : i and TE-2) derived from human squamous cell carcinomaof the esophagus. Gann, 70:575-584, 1979.

21. Saito, H., Tsuchida, S., Nakaji. S., Kakizaki. K., Aisawa. T., Munakata, A.,and Yoshida. Y. An immunologie test for fecal occult blood by counterimmunoelectrophoresis: higher sensitivity and higher positive reactions incolorectal cancer than single radial immunodiffusion and hemoccult test.

Cancer (Phila.), 56: 1549-1552, 1985.22. Habig, \V. H.. Pabst. M. J.. and Jakoby. W. B. Glutathione 5-transferase.

The first enzymatic step in mercapturic acid formation. J. Biol. Chem.. 249:7130-7139. 1974.

23. Satoh, K., Kitahara. A., and Sato. K. Identification of heterogeneous andmicroheterogeneous subunits of glutathione 5-transferase in rat liver cytosol.Arch. Biochem. Biophys.. 242: 104-111. 1985.

24. Towbin, H., Staehelin. T., and Gordon. J. Electrophoretic transfer of proteinsfrom polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Nati. Acad. Sci. USA. 76:4350-4354. 1979.

25. Mancini, G.. Carbonara. A. O., and Heremans, J. F. Immunochemiculquantification of antigens by single radial immunodiffusion. Immunochem-istry, 2: 235-254. 1965.

26. Lowry, O. H., Rosebrough. N. J.. Farr, A. L., and Randall. R. J. Proteinmeasurement with the folin phenol reagent. J. Biol. Chem.. 193: 265-275,1951.

27. Engvall, E. Enzyme immunoassay ELISA and EMIT. Methods Enzymol.,70:419-439, 1980.

28. Tsuruo, T., lida-Saito. H., Kawabata, H.. Oh-hara. T.. Hamada. H.. andUtakoji, T. Characteristics of resistance to Adriamycin in human myelogen-ous leukemia K 562 resistant to Adriamycin and the isolated clones. Gann,77:682-692, 1986.

29. Marcus. C. J.. Habig, W. H., and Jakoby, W. H. Glutathione transferasefrom human erythrocytes. Nonidentity with the enzymes from liver. Arch.Biochem. Biophys.. 188: 287-293. 1978.

30. Guthenberg, C.. and Mannervik, B. Glutathione .S'-transferase (transferase ir)

from human placenta is identical or closely related to glutalhione .C-transfer-ase (transferase p) from erythrocytes. Biochim. Biophys. Acta, 66/: 255-260,1981.

31. Mouelhi, M. E., Didolkar. M. S., Elias, E. G., Guengerich. F. P.. andKauffman. F. C. Hepatic drug-metabolizing enzymes in primary and secondary tumors of human liver. Cancer Res., 47: 460-466. 1987.

32. Miyazi, T. Association of hepatocellular carcinoma with cirrhosis amongautopsy cases in Japan during 14 years from 1958 to 1971. Gann Monogr.Cancer Res.. IS: 129-149. 1976.

5229

on April 9, 2020. © 1989 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Page 6: Elevation of the Placental Glutathione S-Transferase Form ... · kidney, pancreas, and lung (18), under normal conditions, com parison of expression in cancer and normal tissues is

1989;49:5225-5229. Cancer Res   Shigeki Tsuchida, Yoshihito Sekine, Ryuzaburo Shineha, et al.   with Cancer

) in Tumor Tissues and the Levels in Sera of Patientsπ(GST--Transferase FormSElevation of the Placental Glutathione

  Updated version

  http://cancerres.aacrjournals.org/content/49/18/5225

Access the most recent version of this article at:

   

   

   

  E-mail alerts related to this article or journal.Sign up to receive free email-alerts

  Subscriptions

Reprints and

  [email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

  Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/49/18/5225To request permission to re-use all or part of this article, use this link

on April 9, 2020. © 1989 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from