Studies on antigenicity of the polyethylene glycol (PEG)-modified uricase

Studies on antigenicity of the polyethylene glycol (PEG)-modified uricase

Int. J, lmmunopharmac., Vol. 7, No. 5, pp. 725-730, 1985. Printed in Great Britain. 0192-0561/85 $3.00+ .00 •1 1985 International Society for lmmunop...

415KB Sizes 4 Downloads 130 Views

Int. J, lmmunopharmac., Vol. 7, No. 5, pp. 725-730, 1985. Printed in Great Britain.

0192-0561/85 $3.00+ .00 •1 1985 International Society for lmmunopharmacology.

STUDIES ON ANTIGENICITY OF THE POLYETHYLENE GLYCOL (PEG)-MODIFIED URICASE JUN-ICH1 TSUJI and KATSUMI HIROSE Toyobo Research Center, Toyobo Co., Ltd., Katata, Ohtsu, Shiga 520-02, Japan and ETSUKO KASAHARA, MAKI NAITOH and ITARU YAMAMOTO Department of Immunochemistry, Faculty of Pharmaceutical Sciences, Okayama University, Okayama 700, Japan (Received 7 August 1984 and in final form 3 January 1985)

Abstract - - The purified uricase (urate: oxygen oxidoreductase, EC 1.7.3.3) from Candida utilis was

modified to varying degrees with monomethoxypolyethylene glycol (PEG) of different molecular weights using cyanuric chloride as the coupling reagent. The PEG-uricase conjugates were examined on their immunological properties by means of ring test and passive cutaneous anaphylaxis (PCA). As increasing amounts of PEG were attached to uricase, it showed decreasing ability to elicit antibody production in rabbits. When sufficient polymers were attached, the modified uricase was devoid of the capacity to combine in vivo and in vitro with antibodies from guinea pigs injected with the unmodified uricase, however, were still able to react with antibodies to PEG-uricase conjugate. Antibodies against PEG-uricase conjugates also reacted with PEG modified superoxide dismutase (superoxide: superoxide oxidoreductase, EC 1.15.1.1). These results indicate that the coupling of PEG to uricase resulted in the loss of original antigenicity and immunogenicity, but in the appearance of new antigenicity and immunogenicity which never showed any cross-reactions against the native uricase.

Major problems associated with enzyme therapy are the short blood-circulating lives and the immunogenicity of exogenous enzymes. A number of investigators have attempted to improve the clinical effectiveness of enzymes. Some investigators have developed a procedure for modifying proteins by the covalent attachment of monomethoxypolyethylene glycol (PEG). They reported that PEGmodified enzymes showed extended bloodcirculating lives, and levels of immunogenicity so low as to be undetectable by procedures such as gel immunodiffusion, precipitin reaction or complement fixation (Abuchowski, van Es, Palczuk & Davis, 1977; Abuchowski, McCoy, Palczuk, van Es & Davis, 1977; Savoca, Abuchowski, van Es, Davis & Palczuk, 1979; Nishimura, Matsushima & Inada, 1981). Uricase has been employed therapeutically in the treatment of hyperuricemia and gout (London & Hudson, 1957; Kissel, Lamarche & Royer, 1968). Extended treatment resulted in detectable immune 725

responses. Davis, Park, Abuchowski & Davis (1981) recently have reported the application of an adduct of PEG and uricase from Candida utilis to men with advanced haematological malignancies. For clinical use the immunological properties of conjugates must be studied in detail. The present investigation was, therefore, designed to determine whether PEG-modified enzyme showed nonantigenic and non-immunogenic when the experiment was carried out under severe conditions. In the work to be described here, we used the conjugates of uricase of varying degree of substitution with PEG of different molecular weights.

EXPERIMENTAL PROCEDURES

Uricase from Candida utilis and superoxide dismutase (SOD) from bovine erythrocytes were prepared in this (Toyobo) laboratory. Uricase was

726

JUN-ICHI TsuJ1 et al.

examined for homogeneity by disc gel electrophoresis, and its enzymic activity was 14.3 units/mg of protein. Cyanuric chloride (2,4,6trichloro-s-triazine) was obtained from Nakarai Chemicals Ltd. (Kyoto) and was recrystallized twice from anhydrous benzene immediately before use. Monomethoxypolyethylene glycols of 5000, 7500 and 10,000 daltons exhibited narrow molecular weight distribution as determined by high performance liquid chromatography using ShodexA803 column (Showa Denko Co., Ltd., Tokyo). All other chemicals and solvents from commercial sources were of reagent grade quality.

Chemical coupling o f PEG to enzymes To obtain the conjugate of PEG and enzyme, 2-omethoxypolyethylene glycol-4,6-dichloro-s-triazine (activated PEG) was synthesized, and coupled to enzyme through the amino group using cyanuric chloride essentially as described by Abuchowski et al. (1977). Briefly three batches of PEG, with average molecular weights 5000, 7500 and 10,000 (i.e. PEG-5000, PEG-7500 and PEG-10,000, respectively), were activated using a 1:3 molar ratio PEG and cyanuric chloride. Uricase (2 × 10-8 mol) was dissolved in 1 ml of 0.1 M borate buffer, pH 9.5. The solution was brought to 4°C and each activated PEG-5000, PEG-7500 and PEG-10,000 (2.25 × 10-s tool, an amount in 9.3 fold molar excess of available amino groups) was added. The pH was maintained at 9.5 by means of a pH-stat (Hiranuma Sangyo Co. Ltd., Chiba). After 30 min, unattached PEG was removed in an ultrafiltration cell (UHP-76, Toyo Roshi Co., Ltd., Tokyo) using a UK-50 membrane (Toyo Roshi Co., Ltd.) and 0.1 M borate buffer, pH 8.0 as the dialyzing solution. The dialyzates were lyophilized and stored at 4°C. The amount of PEG attached to uricase was estimated by measuring the reduction in primary amino groups by the method of Habeeb (1966). Each of PEG-5000, PEG-7500 and PEG-10,000 was found to be attached to 44, 45 and 45070 of the amino groups of uricase with retention of 9.4, 6.4 and 7.3°7o of the original activity, respectively. A series of PEG-10,000-uricase preparations were made in which increasing fractions of the amino groups were substituted with PEG-10,000. Treatment of uricase with activated PEG-10,000 in molar amounts equal to 4.2, 7.8 and 9.1 of the available amino groups resulted in products with 28, 37 and 4107o of the amino groups substituted by PEG-10,000, and 32, 11 and 3070 of the original activity, respectively.

The method of preparation of PEG-7500-SOD conjugates was similar to that described for PEGuricase conjugates, i.e. treatment of SOD with activated PEG-7500 in molar amounts equal to 2.5 and 5.0 of the available amino groups resulted in products with 39 and 59°7o of the amino groups substituted by PEG-7500, respectively. PEG-7500-enzyme conjugates were also prepared utilizing the succinylated derivative of PEG and the mixed anhydride method. PEG was first converted to PEG-succinate ester by the method of Joppich & Luisi (1979). Then, this derivative was reacted with N-hydroxy succinimide using dicyclohexylcarbodiimide essentially as described by Anderson, Zimmerman & Callahan (1964). This final product, i.e. PEG-succinimidyl succinate was coupled to uricase or SOD through the amino group. The treatment of uricase and SOD with PEGsuccinimidyl succinate in molar amounts equal to 2.5 and 5.0 of the available amino groups resulted in products with 41 and 6007o of the amino groups substituted by PEG-7500, respectively.

Immunological procedures Four groups of two New Zealand White rabbits were immunized by repeated injections of uricase or PEG-uricase. Uricase or PEG-uricase was dissolved in sterile saline and emulsified with an equal amount of complete Freund's adjuvant. An emulsion (0.25 ml) containing 0.3 mg protein of antigen was injected intracutaneously on the animal's back. The rabbits were boosted once every 2 weeks for a month with corresponding antigen in incomplete Freund's adjuvant. Bleedings were taken from the ear vein 10 days after each injection. Antisera against uricase and PEG-uricase were also prepared in guinea pigs, which were injected intracutaneously 3 times at an interval of 2 weeks with 0.3 mg protein of antigen in combination with complete Freund's adjuvant. The animals were bled by heart puncture 12 days following the final injection. The sera of guinea pigs within each group (n = 4) were pooled for determination of the average antibody titer by the ring test and PCA assay. The separated sera were stored at - 8 0 °. Estimation o f antibodies Sera obtained from the immunized rabbits and guinea pigs were analyzed for IgG and IgM using a ring test which was done by observing the precipitin ring in the test tube (diameter: 3 mm), as the result of reaction between the two-fold serially diluted antisera and the solution of antigens (0.2 mg

Studies on Antigenicity protein/ml of 0.01 M phosphate buffered saline, pH 7.3). Ring test titer was expressed by maximum dilution number of antisera by which precipitin ring was observed until 3 h. Antisera collected from immunized guinea pigs were also tested for IgG using the homologous passive cutaneous anaphylaxis (PCA). Guinea pigs were passively sensitized with a cutaneous injection of 0.1 ml of the two-fold serially diluted antisera, at eight sites on both sides of the clipped trunk. Three hours later, the animals were intravenously injected with 1 ml/animal of 0.5070 Evans blue in physiological saline containing 0.5 mg protein of modified or unmodified enzyme. Thirty minutes following the injection, the animals were sacrificed, and the longitudinal and transverse diameters of the area of dye leakage was measured. The end point of the titration was taken as the reciprocal of the highest dilution of each serum resulting in a reaction of 5 mm in diameter.

RESULTS

Effect on antibody production attachment o f PEG to uricase

of

covalent

Uricase was modified to varying degrees (28, 37 and 41070) with PEG-10,000 using cyanuric chloride as the coupling reagent. These preparations were used to immunize rabbits in combination with complete and incomplete Freund's adjuvant. The titers of antibodies produced in animals were

727

measured by the precipitin analysis, i.e. ring test, which is sensitive in/ag range. As shown in Table 1, whereas native uricase elicited strong immune responses after secondary and tertiary injections, the injection of each PEG-uricase conjugate resulted in a very marked decrease of the ability of the animals to mount anti-PEG-uricase responses. As increasing amounts of PEG were attached to uricase, the conjugate showed decreasing ability to elicite antibody production. However, the 41070 modified uricase still acted as an immunogen, i.e. it was capable of causing antibody production. Each PEGuricase conjugate did not react with antibodies obtained after the third immunizing injection of native uricase (data were not shown).

Characterization of antigenic determinants on PEGuricase conjugate Three batches of PEG, with average molecular weights of 5000, 7500 and 10,000, were coupled to uricase using cyanuric chloride. Antisera to PEGuricase conjugates were prepared in guinea pigs. The specificity of antisera to PEG-uricase conjugates was tested in guinea pigs with the aid of the PCA assay. As shown in Table 2, the unmodified uricase elicited a strong PCA reaction in animals which were sensitized with antisera to the same species of uricase. By contrast, animals sensitized with antiPEG-uricase, which had not shown any PCA reaction when challenged with unmodified uricase, gave reduced but significant reaction on injection with PEG(5000, 44070)-uricase, PEG(7500, 4507o)uricase or PEG(10,000, 45%)-uricase. These results indicate that the coupling of PEG to uricase resulted

Table 1. Effect of modification ratio of PEG-10,000 on antibody production in rabbits Ring test titers Primary Secondary Tertiary

Antiserum to

Antigen

Uricase

Uricase

PEG(10,000, 28%)-uricase

PEG(10,000,28%)-uricase

PEG(10,000, 37%)-uricase

PEG(10,000,37%)-uricase

16 8 2 2 0

PEG(10,000, 41%)-uricase

PEG(10,000,41%)-uricase

0

128 128 16 16 8 16 4

256 256 32 32 8 16 8

1

4

4

1

Each rabbit pair received uricase or PEG-uricase in the presence of complete Freund's adjuvant on day 0 and of incomplete Freund's adjuvant on days 14 and 28. The primary, secondary and tertiary responses were assayed 10 days after each injection of antigen. Parentheses after PEG indicate the molecular weights of PEG attached to the enzyme and the percentage of amino groups substituted by PEG.

728

JUN-ICHI TSUJI et al.

in the loss of original antigenic determinants of uricase, but in the appearance of new determinants which never showed any cross-reactions against the unmodified uricase. The experiments described below were designed to provide further evidence for the PEG-uricase conjugates to have gained the unique antigenic determinants. Table 3 shows the results of the ring test. Although PEG-uricase did not show a precipitin ring when tested against antisera to the unmodified uricase, it acted as a good immunogen, i.e. it was capable of causing antibody production, as evidenced by the reaction of PEG-uricase against antisera to itself. The failure to obtain precipitin rings on either reaction of unmodified uricase with anti-PEG-uricase sera or PEG-uricase with antiunmodified uricase provides the unequivocal evidence for the unavailability of the immunogenic determinants of the original antigens in PEGmodified derivatives. The results of these in vitro experiments are in full agreement with the in vivo findings in Table 2 and reinforce the conclusion that conjugation of PEG to uricase resulted in masking of the antigenic determinants of uricase, but that these conjugates had obtained new antigenic determinants which were different from the original one. Further direct supportive evidence for this conclusion was obtained with the aid of the ring test utilizing antisera to modified or unmodified uricase against superoxide dismutase (SOD) and PEG(7500, 39%)-SOD which was prepared by a method similar to that described for PEG-uricase conjugates, i.e. using cyanuric chloride as the coupling reagent. Thus, as can be seen from the data listed in Table 3, slight but definite reactions were seen between antisera of PEG(7500, 34%)-uricase and PEG(7500, 39%)-SOD, but there were no evidence of any reaction between other combinations.

Additional experiments were designed to define where the antigenic determinants existed. For this purpose, PCA assays were carried out using modified enzymes prepared with a different method, i.e. PEG-s.s.-uricase and PEG-s.s.-SOD, which were prepared by attaching PEG-succinimidyl succinate and enzymes, against antiserum to PEG-c.c.-uricase. As is evident from the results illustrated in Table 4, animals sensitized with antiserum to PEG-c.c.uricase showed reduced PCA reaction when challenged with PEG-s.s.-uricase and PEG-s.s.SOD, but the reactions were not undetectable. These results indicate that the antigenic determinants may partially exist on the coupling site of PEG and enzyme.

DISCUSSION As increasing amounts of PEG were attached to uricase, the uricase showed decreasing ability to elicit antibody production in rabbits. PEG-modified uricase also showed extended blood-circulating lives. Male SD rats received a single intravenous injection of the native or PEG(10,000, 41o70)-uricase and the enzymic activity was measured over a period of 12 days. The enzymic activity of native uricase became undetectable 10 h after injection, while that of the modified uricase retained 18070 of the original activity at 9 days. Recently, Chen, Abuchowski, van Es, Palczuk & Davis (1981) reported that the covalent attachment of polyethylene glycol (PEG) to uricase rendered this enzyme both non-antigenic and non-immunogenic as determined by Ouchterlony, i.e. no detectable reaction was seen between antiserum to the unmodified uricase and the PEG-uricase, nor was there evidence of any reaction between antisera to

Table 2. Homologous PCA using uricase, and uricase modified with three batches of PEG, against guinea pig antisera to each antigen Antiserum to

Antigen

Uricase PEG(5000, 44%)-uricase

Uricase Uricase PEG(5000, 44%)-uricase Uricase PEG(7500, 45 %)-uricase Uricase PEG(10,000, 45%)-uricase

PEG(7500, 45%)-uricase PEG(10,000, 45%)-uricase

PCA titers 2560 0 256 0 16 0 16

Parentheses after PEG indicate the molecular weights of PEG attached to the enzyme and the percentage of amino groups substituted by PEG.

729

Studies on Antigenicity Table 3. Precipitin analysis using uricase, SOD, PEG-uricase and PEG-SOD, against guinea pig antisera to uricase and PEG-uricase Antiserum to

Antigen

Ring test titers

Uricase

Uricase PEG(7500, 34%)-uricase

64 0

PEG(7500, 34%)-uricase Uricase PEG(7500, 34%)-uricase

0 16

(a)

Uricase

SOD PEG(7500, 39%)-SOD

0 0

PEG(7500, 34%)-uricase SOD PEG(7500, 39%)-SOD

0 4

(b)

Parentheses after PEG indicate the molecular weights of PEG attached to the enzyme and the percentage of amino groups substituted by PEG.

PEG-uricase and either the unmodified uricase or PEG-uricase. Our present study, using the same method of preparing PEG-uricase, demonstrated that PEG-modification of uricase were devoid of the capacity to combine in vivo and in vitro with antibodies induced by immunization with the unmodified uricase, however were still able to react with antibodies to PEG-uricase conjugates (Tables 2 and 3). These differences may be due to insufficient sensitivity and specificity of their criteria. Immunogenicity of our PEG-uricase conjugates is never associated with a non-uricase contaminant, since their contaminants have been removed by the chromatographic procedure, then the uricase was examined for homogeneity by disc gel electrophoresis. A series of PEG in our studies exhibit narrow molecular weight distribution as determined by high performance liquid chromatography using ShodexA803 column. We rationalize that the covalent

attachment of our uniform PEG to uricase may provide a shell around the enzyme that covers antigenic determinants, resulting in the loss of original antigenic determinants. The fact that antisera to PEG-uricase conjugates were specific not only to PEG-uricase conjugate also to PEG-SOD conjugate (Table 3) is probably due to the new antigenic determinants associated with the covalent attachment of PEG. This hypothesis is supported by the fact that animals sensitized with antiserum to PEG-c.c.-uricase showed reduced PCA reaction when challenged with a different method of preparing PEG-enzyme, i.e. PEG-s.s.-enzyme (Table 4). Richter & Akerblom (1983) have recently demonstrated that PEG of mol. wt. 2000-10,000 acquired immunogenicity by covalent coupling to protein, and that the antigenic determinant of PEG might be a sequence of 6-7-CH2CH20-units. This and our results suggested that PEG-conjugated

Table 4. Homologous PCA using uricase, SOD, PEG-c.c.-uricase or SOD, and PEG-s.s.-uricase or SOD, against guinea pig antiserum to PEG-c.c.-uricase Antiserum to

Antigen

Uricase PEG(7500, 45°70)-c.c.-uricase PEG(7500, 41%)-s.s.-uricase PEG(5000, 44%)-c.c.-Uricase SOD PEG(7500, 59%)-c.c.-SOD PEG(7500, 60%)-s.s.-SOD

PCA titers 0 640 80 0 320 80

PEG-c.c.-enzyme and PEG-s.s.-enzyme were prepared by a cyanuric chloride method and a succinimidyl succinate method, respectively.

730

JUN-ICHI TSUJI et aL

enzyme or allergen induce a new immunogenicity to a part o f P E G or the coupling site o f P E G and protein.

Acknowledgement - - The authors are grateful to Mr. Yoshihisa Yasuda, Mrs. Keiko Yamashita, Miss. Miyuki Akai and Miss. Hiroko Okano for help in carrying out the studies.

REFERENCES

ABUCHOWSKI, A., VAN ES, T., PALCZUK,N. C. & DAVIS, F. F. (1977). Alteration of immunological properties of bovine serum albumin by covalent attachment of polyethylene glycol. J. biol. Chem., 252, 3578-3581. ABUCHOWSKI, A., McCoY, J. R., PALCZUK,N. C., VAN ES, T. & DAVIS, F. F. (1977). Effect of covalent attachment of polyethylene glycol on immunogenicity and circulating life of bovine liver catalase. J. biol. Chem., 252, 3582 - 3586. ANDERSON, G. W., ZIMMERMAN,J. E. & CALLAHAN,F. M. (1964). The use of esters of N-hydroxysuccinimide in peptide synthesis. J. Am. Chem. Soc., 86, 1839-1842. CHEN, R. H. L., ABUCHOWSKI,A., VAN ES, T., PALCZUK,N. C. & DAVIS, F. F. (1981). Properties of two urate oxidases modified by the covalent attachment of poly(ethylene glycol). Biochim. biophys. Acta., 660, 293- 298. DAVIS, S., PARK, Y. K., ABUCHOWSKI,A. & DAVIS, F. F. (1981). Hypouricaemic effect of polyethyleneglycol modified urate oxidase. Lancet, 281 - 283. HABEEB, A. F. S. A. (1966). Determination of free amino groups in proteins by trinitrobenzensulfonic acid. Analyt. Biochem., 14, 328- 336. JOPPICH, M. & LUISI, P. L. (1979). Synthesis of glycyl-L-tryptophylglycinesubstituted by poly(ethylene oxide) at both the carboxy and the amino end groups. Makromolec. Chem., 180, 1381 - 1384. KISSEL, P., LAMARCHE,M. & ROYER, R. (1968). Modification of uricaemia and the excretion of uric acid nitrogen by an enzyme of fungal origin. Nature, 217, 7 2 - 74. LONDON, M. & HUDSON, P. B. (1957). Uricolytic activity of purified uricase in two human beings. Science, 125, 937 - 938. NISHIMURA, H., MATSUSH1MA,A. & INADA, Y. 0981). Improved modification of yeast uricase with polyethylene glycol, accompanied with nonimmunoreactivity towards anti-uricase serum and high enzymic activity. Enzyme, 26, 4 9 - 53. RICHTER, A. W. & AKERBLOM,E. (1983). Antibodies against polyethylene glycol produced in animals by immunization with monomethoxy polyethylene glycol modified proteins. Int. Archs Allergy appl. Immun., 70, 124- 131. SAVOCA, K. V., ABUCnOWSKI,A., VAN ES, T., DAVIS,F. F. & PALCZUK,N. C. (1979). Preparation of a non-immunogenic arginase by the covalent attachment of polyethylene glycol. Biochim. biophys. Acta., 578, 47 - 53.