Epoxide Hydrolase: A Marker for Experimental Hepatocarcinogenesis

Transcription

1 ANNALS OF CLINICAL AND LABORATORY SCIENCE, Vol. 14, No. 1 Copyright 1984, Institute for Clinical Science, Inc. Epoxide Hydrolase: A Marker for Experimental Hepatocarcinogenesis MARTIN J. GRIFFIN, P h.d., and NAZARETH GENGOZIAN, Ph.D. Cancer Research Program, Oklahoma Medical Research Foundation, Oklahoma City, OK ABSTRACT Rat liver chem ical hepatocarcinogenesis, induced by interrupted feeding of 2-acetylaminofluorene, results in various cellular preneoplastic stages and finally in a hepatoma in about 70 to 90 percent of the rats. The putative precursors of hepatomas, called hyperplastic nodules, appear after 12 weeks of feeding and, after 16 weeks of feeding carcinogen, most of them are persistent. Epoxide hydrolase is a tightly bound endoplasmic reticulum enzyme which is strongly induced in hyperplastic nodules and hepatomas. This enzyme has been purified, high-titre rabbit antiserum prepared to it, and this antiserum used to search for epoxide hydrolase immunodeterminants in the sera from chemically induced nodule or hepatoma bearing rats. An enzyme linked immunosorbent assay (ELISA) assay using this antiserum showed significant titres of circulating microsomal epoxide hydrolase antigen, range 0.01 to 2.50 (mean 1.18 ± 0.30) xg per ml, in all 24 hepatoma bearing rats tested. Eight sera from animals with large hyperplastic nodules were also significantly positive for this antigen, while sera from six normal controls were negative (<0.004 jig per ml serum). A passive hemagglutination inhibition assay with sheep or normal rat red blood cells sensitized with pure rat microsomal epoxide hydrolase was capable of detecting 0.2 xg hydrolase per ml serum. With this assay, sera from four rats with hepatomas were found to contain 0.8 to 1.6 (xg epoxide hydrolase immunodeterminants per ml. Control rat sera had no detectable immunodeterminants. Thus epoxide hydrolase, a marker induced during experimental chemical hepatocarcinogenesis and called the preneoplastic antigen, has been shown to be circulating in the tumor bearing host. Introduction The development of cancer in many organs and tissues of mammals may occur through proliferation of new cell populations that differ from the original target tissues; within these new populations, certain clones are believed to be involved in further cell evolution of neoplasia.2,5 Localized areas of hyperplasia can be induced in an essentially reproducible manner in rat liver by the inter /84/ $00.90 Institute for Clinical Science, Inc.

2 2 8 GRIFFIN AND GENGOZIAN m ittent feeding of ethionine or 2-acetylaminofluorene (2-AAF).1,3 Available evidence suggests that these hyperplastic liver nodules contain cellular clones that may undergo further alteration to malignant neoplasia.3,4 Hyperplastic nodules share some morphological and biochemical properties in common with either norm al liver or hepatocellular carcinoma.4 For example, Okita et al1415 observed the presence of an antigen in hyperplastic nodules that also was present in primary hepatomas induced by 2-AAF, ethionine, 3'methyl-p-dimethylaminoazobenzene, dimethylnitrosamine, or diethylnitrosamine. This preneoplastic antigen could not be dem onstrated by immunodiffusion in normal adult rat liver or in regenerating rat liver.14,15 This liver microsomal membrane protein is released from hyperplastic nodule microsomes as detected by a double diffusion assay using a rabbit anti-nodule microsome antiserum back-absorbed with normal liver microsomes. This preneoplastic antigen was purified from nodule microsomes and was shown to be present in normal microsomes at one quarter the level of nodule m icrosom es.6 The protein was subsequently identified as epoxide hydrolase (EC ).10 The enzyme released by in vitro incubation of hyperplastic nodule microsomes has a reduction in subunit molecular w eight from 50,000 to 43,000 daltons, and this latter species is not bound tightly to the membrane.7,12 A cytosolic protease, capable of hydrolyzing microsomal epoxide hydrolase in vitro, has been reported by us to be induced 16-fold in nodules and hepatomas.8 This protease is stim ulated by thio compounds and has some specificity for microsomal epoxide hydrolase. Since microsomal hydrolase faces the cytoplasmic side of the endoplasmic reticulum along with other drug metabolizing enzymes, it appears to be restricted from secretion and does not occur in a healthy animal s serum. Therefore, detection of epoxide hydrolase im m unodeterm inants at late and perhaps even early stages of experimental hepatocarcinogenesis might provide a unique sérodiagnostic test for liver cancer. This in turn might be extended to the human disease. Methods A n im a l s Male Sprague Dawley rats, weighing from 100 to 110 g, were purchased.* Rats were fed 0.05 percent 2-AAF in a basal diet that promotes chemical hepatocarcinogenesis.11 The intermittent feeding schedule was used which results in large, p ersisten t hyperplastic liver nodules and, after 16 weeks of feeding, eventually results in 70 to 90 percent hepatomas.1 Control rats were fed basal diet only. Nodules and hepatomas were certified by cold form alin fixed tissue sections using histochemical staining (hematoxylin-eosin) and im m unohistochem ical staining for the induced nodule and hepatoma epoxide hydrolase (rabbit antiepoxide hydrolase serum and goat antirabbit IgG -peroxidase-antiperoxidase complex). I m m u n o l o g i c a l T e c h n i q u e s Epoxide hydrolase antigen was purified by a modification of a previously reported technique.9 The chromatography steps of diethylam inoethyl cellulose (DE-52) and carboxym ethyl cellulose (CM-52) were performed on a Luberol PX solubilized preparation in 20 percent glycerol to maximize yields and sharpen separation of form (B) from form (A). Epoxide hydrolase form (B) was essen * From Charles Rivers Co., Wilmington, MA.

3 tially hom ogeneous after chromatofocusing on P B E -74.t The enzym e appeared homogeneous on analytical SDS/ PAGE, had a very high styrene oxide hydrolase activity (870 nmol per min per mg protein), and had a very similar amino acid composition to a previously reported preparation.9 The protein was a single band on analytical isoelectric focusing and im m unoelectrophoresis. A male, w hite New Zealand rabbit was challenged every two weeks with 100 xg of pure enzyme emulsified in complete Freund s adjuvant until at least four injections were performed. The rabbit antiserum was highly specific as shown by its precipitating only a 50K band (SDS/ PAGE) from an in vitro translation system using radiolabeled amino acids (over 100 microsomal proteins labeled) (13). This antiserum precipitated all of the B-type microsomal styrene oxide hydrolase activity. The ELISA assays were performed using the anti-epoxide hydrolase (B) serum at a dilution of 1/50,000 as the prim ary antibody. The goat antirabbit immunoglobulin conjugated with horseradish peroxidase (second antibody) was purchased, t The albumin blocking solution, the well washing solution, and the color developing reagents were purchased^ Their suggested procedure for second antibody dilution and overall assay technique was followed. Quantitative well colorimetric readings were obtained J Passive hemagglutination inhibition assays were performed using tanned sheep red blood cells and, in some cases tanned, normal rat red blood cells sensitized with the antigen. t Pharmacia Fine Chem ical C o., Uppsala, Sweden. $ From Boehringer Mannheim Biochemicals, Indianapolis, IN. From Kirkegaard and Perry Labs, Gaithersburg, MD. IBy use of the Microelisa Autoreader, purchased from Dynatech Corp., Alexandria, VA. EPOXIDE HYDROLASE 2 9 TABLE I ELISA Assay for Microsomal Epoxide Hydrolase Iromunodeterminants in Sera from Male Rats Bearing 2-Acetylaminofluorene Induced Liver Tumors Animals* Histology^ Epoxide Hydrolase% Imumodeterminants Experimental Tumor Type (yg/ml) 1 H-I H-II H-II H-II H-II H-II H-II H-II H-II H-II H-III H III H-III H-III H-III H IV H-IV H-IV Ch Ch Ch Ch Ch Ch 0.29 Normals Normal 1 < < < < < <0.004 *All experimental animals received 16 weeks of interrupted feeding of basal diet with 0.05 percent 2-acetylaminofluorene and were then fed basal diet for 16 further weeks. The animals were fasted the night before death by decapitation. Serum was isolated and tumors resected. A small piece of tumor was fixed, sectioned, stained and evaluated for type. fh = Hepatoma (I - relatively similar to normal liver, IV - most neoplastic); Ch = Cholangioma. In contrast to normal liver, these liver tumor sections stained strongly for expoxide hydrolase antigen as determined by use of immunohistochemical staining. $The ELISA assay involved inhibition of color development by antigen in serum. A plate was coated with 0.5 yg per well of chromatofocused pure rat liver microsomal epoxide hydrolase (form B) in 0.1 percent Luberol PX at 4 for 16 hrs. The test sera were incubated with diluted primary rabbit anti-hydrolase (1/50,000) serum for one hour at room temperature which was then used as a source of antibody in the complete ELISA assay. Standard inhibition curves were obtained by diluting pure antigen in 0.1 percent Luberol PX and the range was from 40 to 0.33 ng hydrolase per sample. This tumor appeared necrotic.

4 30 GRIFFIN AND GENGOZIAN T A B L E II Passive Hemagglutination Inhibition Assay for Rat Epoxide Hydrolase in Serum in Normal and Experimental Rats J?at Serum Added 5 Rabbit Anti -Rat Epoxide Hydrolase (Dilutions x 10-1*) \ig Epoxide Hydrolase/ml Serum* Normal ± ± ± Hepatoma «_ ± ± ± *" 1.6 Values estimated from inhibition of dilutions of the anti-epoxide hydrolase antiserum with varying amounts of the enzyme antigen in a checkerboard analysis. Results and Discussion ELISA A s s a y In table 1 are listed the results of ELISA assay for microsomal epoxide hydrolase (form B) immunodeterminants in sera from animals bearing liver tumors induced by 2-AAF. An evaluation of the tumor type and degree of abnormality (H-IV is the most abnormal hepatocellular carcinoma) is also listed. All 24 of the tumor bearing animals had significant titres of microsomal epoxide hydrolase antigen, range 0.01 to 2.50 (mean 1.18 ± 0.30 SD) xg per ml. The six normal strain, sex and age matched control animals had sera with no detectable epoxide hydrolase immunodeterminants (table I) (<0.004 (xg p er ml serum). Sera from eight animals w ith large hyperplastic nodules resulting from interrupted feeding of 2-AAF for 16 weeks followed by three weeks of feeding only basal diet were also all positive for epoxide hydrolase antigen, range 0.46 to 2.25 (mean 1.26 ± 0.28 SD) jlg per ml (not shown in table I). P a s s i v e H In h i b i t i o n A s s a y e m a g g l u t i n a t i o n This test using either tanned sheep or rat red blood cells, pure epoxide hydrolase (B) and rabbit anti-hydrolase serum gave inhibition patterns proportional to (1) the amount of epoxide hydrolase antigen added to the system and (2) to the dilution of antiserum used. The limit of detection of epoxide hydrolase antigen was about 0.2 xg per ml, considerably less sensitive than the ELISA assay (lower limit jxg per ml). This appears to be in the range of sensitivity experienced in other passive hemagglutination inhibition assays. This assay was then used to quantitate epoxide hydrolase immunodeterminants in the serum of rats bearing 2-AAF induced hepatomas. Four hepatoma bearing rats had from 0.8 to 1.6 xg antigen per ml serum (table II). Normal animals were repeatedly negative in the assay. The assay was repeated a minimum of three times, and the biological hepatocarcinogenesis experiment was repeated once, with similar results.

5 Since this preneoplastic antigen is now established in the circulation of rats bearing preneoplastic as well as neoplastic livers, it will be of interest to expand this study to include: (a) the earliest time significant titre of this antigen can be d etected when feeding 2-AAF; (b) w hether or not all hepatocarcinogens elicit this circulating antigen; (c) whether or not feeding non-carcinogenic hepatotoxins result in serum epoxide hydrolase; (d) whether or not rodent hosts with virally induced hepatomas have this antigen in circulation; and (e) whether or not this system can be extended to the clinical situation. Acknowledgments Thanks are extended to Dave Phillips and Ron Clark for their excellent technical assistance. This work was supported by DHHS grants CA 24459, AI 19495, and March of Dimes grant # References 1. E p s t e i n, S. M., I t o, M., M e r k o w, L., and F a r b e r, E. : Cellular analysis of liver carcinogenesis: The induction of large hyperplastic nodules in the liver with 2-fluorenylacetam ide or ethionine and some aspects of their morphology and glycogen m etabolism. Cancer Res. 27: , F a r b e r, E.: Ethionine carcinogenesis. Adv Cancer Res. 7: , F a r b e r, E.: Hyperplastic liver nodules. Methods Cancer Res. 7: , F a r b e r, E.: Carcinogenesis: Cellular evolution as a unifying thread Presidential address. Cancer Res. 33: , F o u l d s, L.: Neoplastic Development, vol. 1. New York, Academic Press, Inc., 1969, pp EPOXIDE HYDROLASE G r i f f i n, M. J. a n d K i z e r, D. E.: Purification and quantitation of preneoplastic antigen from hyperplastic nodules and normal liver. Cancer Res. 38: , G r i f f i n, M. J. a n d N o d a, K.: Quantitation of epoxide hydrolase released from hyperplastic nodule and hepatoma microsomes. Cancer Res. 40: , G r i f f i n, M. J., V a z, A. D., a n d N o d a, K.: Release of epoxide hydrolase from 2-AAF induced rat liver hyperplastic nodule and hepatoma microsomes by proteolytic activity. Proc. 13th Internatl. Cancer Cong., 1982, p G u e n g e r i c h, F. P., W a n g, P., M i t c h e l l, M. R., and M a s o n, P. S.: Rat and human liver microsomal epoxide hydrolase: Purification and evidence for the existence of multiple forms. J. Biol. Chem. 254: , L e v i n, W., L u, A. Y. H., T h o m a s, P. E., R y a n, D., K i z e r, D. E., and G r i f f i n, M. J.: Identification of epoxide hydrolase as the preneoplastic antigen in rat liver hyperplastic nodules. Proc. Natl. Acad. Sci. USA 75: , M e d e s, G., F r i e d m a n, B., and W e i n h o u s e, S.: Fatty acid metabolism VIII. Acetate metabolism in vitro during hepatocarcinogenesis by p-dimethylaminoazobenzene. Cancer Res. 16:5 7-66, N o d a, K., V a z, A. D., K i z e r, D. E., and G r i f f i n, M. J.: Sequential changes in histochemical localization and reduction in subunit size of epoxide hydrolase during experimental hepatocarcinogenesis. Carcinogenesis (under consideration). 13. O k a d a, Y., F r e y, A. B., G u e n t h n e r, T. M., O e s c h, F., S a b a t i n i, D. D., and K r e i b i c h, G. : Studies on the biosynthesis of microsomal membrane proteins. Eur. J. Biochem. 122: , O k it a, K. and F a r b e r, E.: An antigen common to preneoplastic hepatocyte populations and to liver cancer induced by iv-2-fluorenylacetamide, ethionine, or other carcinogens. Gann Monograph 17: , O k it a, K., K l i g m o n, L. H., and F a r b e r, E.: A new common marker for premalignant and malignant hepatocytes induced in the rat by chemical carcinogens. J. Natl. Cancer Inst. 54: , 1975.