Comparison of modification of a bacterial uricase with N-hydroxysuccinimide esters of succinate and carbonate of monomethoxyl poly(ethylene glycol)
Chun Zhang † Xiaolan Yang † Ang Gao Xiaolei Hu Jun Pu Hongbo Liu Juan Feng Juan Liao Yuanli Li ∗ Fei Liao
Unit for Analytical Probes and Protein Biotechnology, Key Laboratory of Clinical Laboratory Diagnostics of the Education Ministry, College of Laboratory Medicine, Chongqing Medical University, Chongqing, People’s Republic of China
Abstract Uricase after modification with monomethoxy poly(ethylene glycol) (mPEG) is currently the sole agent to treat refractory gout. For formulating Bacillus fastidious uricase, succinimidyl carbonate of mPEG-5000 (SC-mPEG5k) and succinimidyl succinate of mPEG-5000 (SS-mPEG5k) were compared. SC-mPEG5k possessed higher purity, comparable reaction rate constant with glycine but lower hydrolysis rate, and stronger effectiveness to modify amino groups. The uricase possessed two types of amino groups bearing a 25-fold difference in reactivity with SC-mPEG5k or SS-mPEG5k at pH 9.2. Oxonate and xanthine concentration-dependently protected the bacterial uricase from inactivation during PEGylation. With SC-mPEG5k at a molar ratio of 200 to uricase subunits and
oxonate of 50 µM, the PEGylated uricase (1) retained about 73% of the original activity, (2) displayed about 10% reactivity to rabbit anti-sera recognizing the native uricase, (3) elicited IgG in rats accounting for about 5% of that by the native uricase, (4) exhibited circulation half-life time of about 25 H in cock plasma in vivo, and (5) concurrently maintained uric acid at lowered levels for over 20 H. Hence, PEGylation with SC-mPEG under the protection of a competitive inhibitor was a practical approach to formulation of the bacterial uricase; protection of enzymes by competitive inhibitors during C 2014 International PEGylation may have universal significance. Union of Biochemistry and Molecular Biology, Inc. Volume 61, Number 6, Pages 683–690, 2014
Keywords: uricase, PEGylation, competitive inhibitor, succinimidyl carbonate of mPEG, protection
1. Introduction Abbreviation: BSA, bovine serum albumin; DMF, N-dimethylforamide; ELISA, enzyme-linked immunosorbent assay; mPEG, monomethoxyl poly(ethylene glycol); mPEG5k-uricase, uricase PEGylated with SC-mPEG5k at molar ratio of 200 to uricase subunits; NHS, N-hydroxysuccinimide; SC-mPEG5k, NHS ester of carbonate of mPEG-5000; THF, tetrahydrofurane; TNBS, 2,4,6,-trinitrobenzenesulfonic acid. ∗ Address
for correspondence: Fei Liao, MSc, Unit for Analytical Probes and Protein Biotechnology, Key Laboratory of Clinical Laboratory Diagnostics of the Education Ministry, College of Laboratory Medicine, Chongqing Medical University, Chongqing, People’s Republic of China. Tel.: +86 23 68485240; Fax: +86 23 68485240; e-mail: [email protected]. †These authors contributed equally to this work. Supporting Information is available in the online issue at wileyonlinelibrary.com. Received 22 August 2013; accepted 5 February 2014 DOI: 10.1002/bab.1215 Published online 28 March 2014 in Wiley Online Library (wileyonlinelibrary.com)
Gout is a chronic metabolic disorder requiring treatment with anti-urate drugs like allopurinol throughout the life of patients [1–3]. Refractory gout is characterized by the ineffectiveness/inapplicability of allopurinol, and uricase (EC 1.7.3.3) is the sole agent effective for refractory gout [4–8]. To prolong circulation time in vivo and mask immunogenic sites of uricases as exogenous proteins, their modification with monomethoxyl poly(ethylene glycol) (mPEG), PEGylation, is a plausible approach [9–14]. In general, PEGylation of residues of wide distribution on uricase surfaces is preferred, and Nacylation of amino groups with N-hydroxysuccinimide (NHS) ester of mPEG in excess is practical [14]. However, PEGylated uricases show lower activities [15–17]. To treat refractory gout, PEGylated uricases have to be administered at large doses and short intervals [16, 17], which induce antibodies against mPEG to shorten the circulation time of PEGylated uricases [18–21]. For continuous treatment of refractory gout, biweekly
683
Biotechnology and Applied Biochemistry therapeutic doses of PEGylated uricases should be smaller than 0.2 mg [8]. Hence, the process to PEGylate any uricase should be carefully optimized for the activity of the PEGylated uricase as high as possible. A uricase from Bacillus fastidious ATCC 29604 has high activity and excellent thermostability, supporting the significance of formulation to test its therapeutic potential [15, 22, 23]. NHS ester of succinate of mPEG-5000 (SS-mPEG5k) and NHS ester of carbonate of mPEG-5000 (SC-mPEG5k) are representative agents to PEGylate proteins via N-acylation [14, 15, 24]. SC-mPEG5k and SS-mPEG5k undergo both N-acylation and hydrolysis during PEGylation of proteins, but they still possess some differences related to PEGylation of proteins. For instance, SC-mPEG5k can be easily purified to higher homogeneity, but SS-mPEG5k is usually contaminated with unreactive mPEG [24, 25]. The hydrolysis rate constant of SC-mPEG5k at pH 8.0 and 25 ◦ C is just about half of that of SS-mPEG5k [24–26] (http://laysanbio.com/clientuploads/mPEGSVA final review.doc). Thus, smaller quantities of SC-mPEG5k can be utilized to produce the same degree of PEGylation and thus facilitate purifying the PEGylated products. On the other hand, ligands to active sites of some enzymes protect them from inactivation during modification [27–29]. Oxonate and xanthine are competitive inhibitors of uricases and have negligible reactivity to SC-mPEG5k or SS-mPEG5k [15, 22, 30]. Herein, we compared PEGylation of the bacterial uricase with SC-mPEG5k and SS-mPEG5k, with or without the protection of oxonate and xanthine, and results indicated that SC-mPEG5k was more favorable for PEGylation of the bacterial uricase under the protection of competitive inhibitors.
2. Materials and Methods 2.1. Materials and chemicals Monomethoxy polyethylene glycol 5000 (mPEG5k), 2,4,6,trinitrobenzenesulfonic acid (TNBS), and diethylaminoethyl (DEAE) cellulose were from Sigma–Aldrich (St Louis, MO, USA). Xanthine, NHS, uric acid, and triphosgene were from Alfa Aesar (Tainjing, China). Potassium oxonate was from Shandong Zhongketaidou Chemical Co. (http://en.sdzktd.com/) (Jinan, Shandong, China). N-Dimethylforamide (DMF) was pretreated with acetate anhydride and restilled. Bovine serum albumin (BSA), tributylamine, and other chemicals were domestic reagents. Goat anti-IgG (rabbit) secondary antibody (IgG), antiIgG (rat) secondary antibody (IgG), and anti-IgM (rat) secondary antibody (IgG) were labeled with horseradish peroxidase (Sangon Bioengineering Co., Shanghai, China), and were utilized as received following the standard protocol.
2.2. Preparation of a recombinant bacterial uricase Uricase form B. fastidious ATCC 29604 was expressed in Escherichia coli BL21 (DE3) [15, 31]. After lysis of cells by sonication treatment, soluble uricase was purified by DEAE cellulose chromatography via elution with 0.10 M Tris–HCl at
684
pH 8.0 plus a linear gradient of NaCl from 0.10 to 0.40 M. Eluted portions with activities over 6 kU g−1 were collected and dialyzed against 0.20 M sodium borate buffer at pH 9.2 for 24 H with four changes of the buffer. Other proteins in the preparation were undetectable by SDS-PAGE.
2.3. Uricase activity assay Activities of uricase were determined by quantifying absorbance at 293 nm on a Shimadzu UV 2550 spectrophotometer (Kyoto, Japan) equipped with a thermostatic apparatus at 25 ◦ C in 0.20 M sodium borate buffer at pH 9.2 containing 75 µM uric acid [31]. Absorptivity of uric acid was 11.5 mM−1 cm−1 [32]. The change of absorbance after 30 Sec since the reaction initiation by the addition of a sample was recorded at 5 Sec intervals. One unit of uricase was the quantity able to oxidize 1 µmol uric acid in one Min. To measure uricase in cock serum, a sample of 50 µL was added and the change of absorbance was measured at 60 Sec intervals.
2.4. Preparation of SS-mPEG5k and SC-mPEG5k In a three-necked flask, 10 g mPEG5k was dissolved in 30 mL CH2 Cl2 ; 0.89 g triphosgene was dissolved in 5 mL redistilled toluene and added into mPEG5k solution drop by drop; the mixture was stirred for 18 H at room temperature [24]. Then, the reaction mixture was concentrated to 10 mL under reduced pressure at 40 ◦ C and the targeted intermediate was precipitated by the addition of 100 mL diethyl ether. To further purify the intermediate, its dissolution in 10 mL CH2 Cl2 and precipitation by the addition of 100 mL diethyl ether were repeated twice to give a white powder that was finally dissolved in 30 mL CH2 Cl2 again. To obtain SC-mPEG5k, 0.29 g NHS dissolved in 4 mL tetrahydrofurane (THF) was added to the solution of the intermediate. The resulting mixture was stirred for 8 H at room temperature and then concentrated to 10 mL under reduced pressure at 40 ◦ C. The final product, SC-mPEG5k, was precipitated by the addition of 100 mL diethyl ether. To purify SC-mPEG5k, the resulting powder was dissolved in 10 mL THF and precipitated by the addition of 100 mL diethyl ether for three times to give a white powder bearing 80% yield in the mass of mPEG5k (Fig. S1, Supplementary Data in the Supporting Information). Reactivity of SC-mPEG5K with amino groups was verified via the detection of the fluorescent derivative after reaction with N-(1-naphthyl)-ethylenediamine [33, 34]. SS-mPEG5k was prepared as before [15]. Molar equivalency of NHS ester was estimated via the reaction with ethanolamine and TNBS assay of residual ethanolamine [25].
2.5. Characterization of reaction kinetics of SC-mPEG5k and SS-mPEG5k At pH 9.2 in 0.20 M sodium borate buffer, the bacterial uricase shows excellent thermostability and NHS is reasonably stable. During hydrolysis or reaction with amino groups in the borate buffer at pH 9.2, reaction curves of SS-mPEG5k or SC-mPEG5k were continuously recorded at 1 Sec intervals by the absorbance at 260 nm; background absorbance of the solution in the absence of the NHS ester was corrected [24–26]. To
PEGylation of a Uricase with Active Esters of mPEG
characterize hydrolysis of SS-mPEG5k or SC-mPEG5k, the final levels were 80, 60, and 40 µM; reaction curves were recorded at 20, 25, and 30 ◦ C, separately. To characterize the reaction with amino group of glycine, the final level of SS-mPEG5k or SC-mPEG5k was fixed at 80 µM, whereas the final levels of glycine were 5.0, 10.0, and 20.0 mM to record reaction curves at 20, 25, and 30 ◦ C, separately. All reaction curves were recorded within 10 Min on the Shimadzu UV 2550. Under stated conditions, the concentration of hydroxyl oxide can be assumed to be a constant during hydrolysis and the final levels of glycine can be assumed to be constants during reaction with either NHS ester. The hydrolysis of a NHS ester or its reaction with an amino group of glycine should apparently follow first-order kinetics. The rate constant for hydrolysis served as background in apparent rate constants for reactions of glycine with SS-mPEG5k or SC-mPEG5k. The program for analyzing enzyme reaction curves at substrate levels much lower than the Michaelis–Menten constant was used [35, 36]; the goodness-of-fit was judged by determination coefficients to search for combinations of parameters.
2.6. PEGylation of uricase with SC-mPEG5k or SS-mPEG5k and effects of ligands Uricase was PEGylated with SC-mPEG5k or SS-mPEG5k in 0.20 M sodium borate buffer at pH 9.2 and 6 ± 2 ◦ C. To determine the effects of molar ratios of a NHS ester to uricase amino groups, an indicated quantity below 5.0 mg of a NHS ester of mPEG5k dissolved in DMF was added drop by drop with fine tips into 1.5 mL uricase solutions (1.7 g L−1 uricase, final DMF was below 2%) in sodium borate buffer at pH 9.2 under mild stirring, or its solid of over 5 mg was directly added into uricase solutions. After reaction for 90 Min, 950 µL mixtures were withdrawn to react with 50 µL aqueous TNBS solution for 90 Min to quantify amino groups [15, 25]; the modification degree was the percentage of amino groups lost. Meanwhile, the remaining solution in 550 µL was immediately quenched with 10 µL saturated solution of glycine in the sodium borate buffer at 25 ◦ C to measure residual uricase activity. To examine the effects of ligands on PEGylation, a stock solution of 2.0 mM oxonate potassium, 2.0 mM xanthine, or 6 mM uric acid was made in 0.20 M borate buffer at pH 9.2. To solutions of uricase in 0.20 M sodium borate buffer at pH 9.2 and 6 ± 2 ◦ C, one ligand at different quantities was added, then SC-mPEG5k or SS-mPEG5k in a minimum volume of DMF for a 200 molar ratio to uricase subunit was added to give the final uricase of 1.7 g L−1 in a 1.5 mL solution. After reaction for 90 Min at 6 ± 2 ◦ C, the modification degree and residual uricase activities were quantified as described above. To examine potential immunogenicity of the PEGylated uricase, it was dialyzed against 10 mM Tris–HCl buffer at pH 8.8 and 4 ◦ C for 8 H. BSA was PEGylated with SC-mPEG to prepare the conjugate mPEG5k–BSA in a similar way for the quantification of antibodies specific for mPEG5k, assuming no epitopes of
Biotechnology and Applied Biochemistry
BSA have the same immunoreactivity as those of the bacterial uricase.
2.7. Immunization of rats with uricase and SC-mPEG5k-modified uricase To prepare antibodies specific for the native uricase, rabbit was immunized via subcutaneous multipoint administration of 200 µg of the purified uricase every 10 days. Freund’s complete adjuvant was used the first time, and Freund’s incomplete adjuvant was used in subsequent immunization for the following 40 days. Rabbit serum was collected 3 days after the last intravenous injection of 200 µg of the native uricase and stored at −70 ◦ C until use. To test immunogenicity of each sample, seven male rats were employed in each group. Rats were immunized via subcutaneous administration of 50 µg antigen once a week for four weeks. Rat sera were collected through orbital venous and stored at −70 ◦ C until analysis.
2.8. Detection of antibodies in rat sera Each well was incubated for 2 H at 37 ◦ C with 200 µL/well of a coating antigen at 10.0 mg L−1 in 0.20 M bicarbonate buffer at pH 9.2 (the native uricase, SC-mPEG5k-modified uricase, or SC-mPEG5k-modified BSA); each well was then washed with 300 µL of 0.15 M sodium phosphate buffer at pH 7.2 for three times, and blocked with 300 µL coating solution of 50 g L−1 BSA. A properly diluted rat serum or other sample at 200 µL was added to a well for a capture reaction of 1 H at 37 ◦ C. Then, the sample was discarded and the well was washed three times with 200 µL of 0.15 M sodium phosphate buffer at pH 7.2 containing 0.055% Tween 20; a working solution of horseradish peroxidase-labeled goat anti-rat or anti-rabbit IgG/M antibody at 200 µL was added to each well for incubation of 1 H at 37 ◦ C following the standard operation guidelines. After washing of each well with 0.15 M sodium phosphate buffer at pH 7.2 for five times as described above, a 200 µL 3,3 , 5,5 -tetramethyl benzidine hydrochloride working solution was added for a 30 Min reaction at 37 ◦ C and the reaction was terminated with 50 µL H2 SO4 at 2.0 M. Absorbance at 450 nm was measured in 10 Min with a Biotek ELX 800 microplate reader. Data were expressed as mean ± SD.
2.9. Pharmacological properties of the uricase modified with SC-mPEG5k Cocks inherently bear hyperuricemia for testing pharmacological actions of uricase. Cocks were raised in the animal center of Chongqing Medical University under clean conditions. The native or PEGylated uricase was administered through wing-root intravenous injection at 2 U per kg body weight (measured at pH 7.4 in 50 mM sodium borate buffer, just about 25% activity of that at pH 9.2) to two groups of six male cocks (about 1.7 kg). Blood was obtained from the wing-root vein with a heparinized syringe at different times in 48 H; red blood cells were removed immediately by centrifugation at 430 g for 5 Min to obtain plasma. With 200 µL plasma, proteins were immediately
685
Biotechnology and Applied Biochemistry precipitated via the addition of 20 µL 65% HClO4 and vigorous vortex for 5 Min. Then, 20 µL potassium carbonate solution saturated at 25 ◦ C was slowly added to adjust pH of the mixture to about 7.4; 30 Min later, the mixture was centrifuged at 4,700 g for 10 Min to give a clear supernatant to measure plasma uric acid. Meanwhile, plasma uricase activity was measured at pH 9.2 in sodium borate buffer in 5.0 Min as described above [22]. Plasma uricase activities were analyzed with a biexponential model to eliminate interference of partially modified uricase [23].
2.10. Analysis of plasma uric acid by a direct kinetic uricase method To avoid interference from uricase inhibitors and reductants/ oxidants, a direct kinetic uricase method was used to quantify uric acid in cock plasma [22]. In brief, for a 1.165 mL solution of 0.20 M sodium borate at pH 9.2 plus a 30 µL sample, the absorbance at 293 nm was taken as the initial absorbance after the correction of dilution effects (0.4%); then, a 5 µL concentrated solution of recombinant B. fastidious uricase was added for a final 40 U L−1 to monitor the change of absorbance at 293 nm and at 25 ◦ C. With each sample, a uricase reaction curve was recorded at 5 Sec intervals within 7.0 Min, stored in a text file, and read into memory for analysis by a program. The Michaelis–Menten constant of the uricase was preset at 0.28 mM; the background absorbance at 293 nm after the completion of the uricase reaction was predicted. When the difference between the corrected initial absorbance and the absorbance after the reaction for 7.0 Min was less than 0.050, the absorbance after the reaction for 7.0 Min was directly taken as the background absorbance. The difference between the initial absorbance and the background absorbance indexed the net absorbance of uric acid in the reaction solution. The assay had the detection limit of about 1 µM, whereas the upper limit of about 35 µM.
2.11. Analyses of proteins and amino groups Protein was quantified by the Bradford method with BSA as the reference [37]. SDS-PAGE using 8% separation gel was used to analyze compositions of the PEGylated uricases, and protein bands were stained with Coomassie Brilliant Blue R250. Amino groups were quantified by reaction with TNBS to measure the absorbance at 420 nm with glycine as the reference [22, 25].
3. Results and Discussion 3.1. PEGylation of the recombinant uricase In general, PEGylation of a protein via N-acylation with a NHS ester of mPEG prefers alkaline pH as long as the protein is stable (Datasheet 1, Supplementary Data). The bacterial uricase is stable at pH 9.2; the optimum pH for PEGylation with SC-mPEG5k was about 9.2 [24]. Thus, PEGylation of the uricase was tested at pH 9.2 hereafter.
686
FIG. 1
PEGylation of the uricase in 0.20 M sodium borate at pH 9.2 and 6 ◦ C. (a) SC-mPEG5k and (b) SS-mPEG5k.
Activities of the PEGylated uricase as mixtures of PEGylated isoforms decreased in an exponential manner, whereas modification degrees of amino groups followed biphasic increases, with respect to molar ratios of SC-mPEG5k to uricase subunits (Fig. 1a). The modification degrees linearly and rapidly increased to 60% with respect to molar ratios below 100 of SC-mPEG5k to uricase subunits; amino groups reactive to SCmPEG5k at such levels belonged to exposed ones. When final levels of SC-mPEG5k were further increased, activities of the PEGylated uricase became smaller and smaller, whereas the modification degrees grew slowly to about 80% when the molar ratio of SC-mPEG5k to uricase subunits reached 200; then modification degrees became a plateau of below 85% when molar ratios of SC-mPEG5k to uricase subunits were over 250. Amino groups PEGylated only at molar ratios of SC-mPEG5k to uricase subunits over 200 belonged to partially buried ones. From response slopes of modification degrees with respect to molar ratios of SC-mPEG5k to uricase subunits, there was about a 25-fold difference in reactivity with SC-mPEG5k between the exposed and partially buried amino groups. Moreover, TNBS accessed only about 12 amino groups on each subunit of the
PEGylation of a Uricase with Active Esters of mPEG
bacterial uricase. Hence, there should be not more than three partially buried amino groups accessible by SC-mPEG5k in each subunit of the uricase. Similarly, there were biphasic changes of modification degrees of amino groups, whereas there was an exponential decrease in activities of the PEGylated uricase with respect to molar ratios of SS-mPEG5k to uricase subunits (Fig. 1b). Based on response slopes of modification degrees to levels of SS-mPEG5k, the difference in reactivity of two types of uricase amino groups with SS-mPEG5k was comparable to that with SC-mPEG5k. Interestingly, when molar amounts of SS-mPEG5k in its preparation were estimated from molecular weight and mass quantities, there were lower modification degrees but higher activities of the uricase after PEGylation with SS-mPEG5k than those with SC-mPEG5k at the same molar ratios of below 200 to uricase subunits [22]. However, when the final levels of SS-mPEG5k were increased until the modification degrees were the same as those with SC-mPEG5k, activities of the PEGylated uricase were consistent with those modified with SC-mPEG5k. Therefore, (1) activities of the PEGylated uricase depend on modification degrees of amino groups; (2) to mask all epitopes of the bacterial uricase, a NHS ester of mPEG should be in great molar excess to uricase subunits for PEGylation via Nacylation; (3) any approach is desirable to protect the uricase from inactivation during PEGylation.
3.2. Protection of the bacterial uricase by competitive inhibitors Oxonate and xanthine are potent competitive inhibitors of uricases. To this bacterial uricase at pH 9.2 and 6 ◦ C, oxonate, xanthine, and uric acid showed affinities of about 3 µM, 20 µM, and 0.18 mM, respectively [15, 30]. Oxonate, xanthine, or uric acid protected this uricase from inactivation by SC-mPEG5k in a concentration-dependent manner, but caused no significant changes of modification degrees when the molar ratio of SCmPEG5k to uricase subunits was kept at 200 (Fig. 2). Obviously, such protection effects were easily saturated with oxonate and xanthine, but could not be saturated with uric acid owing to strong action of the uricase at the tested levels. Inhibitors bearing higher affinities produced stronger protection effects on the uricase during PEGylation. After PEGylation with SCmPEG5k at the molar ratio of 200 to uricase subunits, the PEGylated uricase retained over 70% of the activity of the native uricase when 50 µM oxonate was used, whereas it retained just about 50% of the activity of the native uricase in the absence of all ligands. When molar ratios of SC-mPEG5k to uricase subunits were 350, activities of the uricase PEGylated even in the presence of 50 µM oxonate decreased to just about 25% versus about 10% in the absence of oxonate, but there were still no detectable differences in the modification degrees. During PEGylation with SS-mPEG5k, similar protection effects of oxonate or xanthine and similar dependence of such protection effects on molar ratios of SS-mPEG5k to uricase subunits were observed (Fig. 3). Hence, the protection effects of competitive
Biotechnology and Applied Biochemistry
FIG. 2
Protection of the uricase during PEGylation at 6 ◦ C in 0.20 M sodium borate at pH 9.2 with SC-mPEG5k at the molar ratio of 200 to uricase subunits. (a) Oxonate, (b) xanthine, and (c) uric acid.
inhibitors on the uricase during PEGylation depend on the final levels of PEGylation agents. To ensure favorable activities of the PEGylated uricase, any PEGylating agent should be restricted to reasonable levels; molecular engineering of the bacterial uricase for site-specific PEGylation via N-acylation to mask
687
Biotechnology and Applied Biochemistry
FIG. 4
FIG. 3
Protection of the uricase during PEGylation at 6 ◦ C in 0.20 M sodium borate at pH 9.2 with SS-mPEG5k at the molar ratio of 200 to uricase subunits. (a) Oxonate and (b) xanthine.
all immunogenic sites while retaining activities as much as possible is preferred [14]. The binding of any ligand caused negligible changes of the molecular weights of conjugates of the uricase and mPEG5k analyzed by SDS-PAGE (Fig. 4). There was a small amount of the native uricase leftover in each preparation of conjugates. The two principal bands of conjugates had molecular weights of about 56 and 66 kDa, respectively, and a minor band of the conjugate had a molecular weight of about 76 kDa, indicating that the number of mPEG5k chains attached to each subunit should be less than 8. Oxonate, xanthine, and uric acid all carried negative charges at pH 9.2; their bindings may involve ionic interactions with some essential amino groups of the bacterial uricase to reduce their reactivity and thus retard PEGylation of such essential amino groups. The protection of enzymes by their competitive inhibitors from inactivation during PEGylation may be universally effective; the investigation of such protection effects of enzyme inhibitors may reinforce PEGylation of common exogenous proteins as potential biodrugs.
688
SDS-PAGE analysis after PEGylation at 6 ◦ C in 0.20 M sodium borate at pH 9.2 with SC-mPEG5k at the molar ratio of 200 to uricase subunits. Lanes 1–5: the use of 30 μM oxonate, the use of 100 μM xanthine, the use of 800 μM uric acid, the use of no ligands, and the native uricase alone.
3.3. Pharmacological properties of the PEGylated uricase The uricase was PEGylated in the presence of 50 µM oxonate with SC-mPEG5k at the molar ratio of 200 to uricase subunits for 60 Min reaction at 6 ◦ C in 0.20 M sodium borate buffer at pH 9.2. The PEGylated uricase retained about 73% activity of that of the native uricase. To test the suitability of the PEGylation process to the uricase, pharmacological properties of the mixture of the PEGylated uricase were directly evaluated. Of the PEGylated uricase, the binding capability in vitro to specific anti-uricase sera of rabbit was just about 10% of that of the native uricase (Fig. 5a); immunogenicity in vivo in rats was decreased to just about 5% of the native uricase based on the quantities of uricase-specific IgG elicited (Fig. 5b). However, at just 1 week after the first immunization with the PEGylated uricase, IgM in rat sera specific for mPEG5k was detected (Fig. 5c). The native uricase had a circulation half-life time in vivo of about 40 Min based on changes of uricase activities in cock plasma. Using a biexponential decay model to analyze the changes of uricase activities [23], the longest circulation half-life time of the PEGylated uricase in cock plasma in vivo
PEGylation of a Uricase with Active Esters of mPEG
FIG. 5
Preliminary evaluation of pharmacological properties of the uricase after PEGylation at 6 ◦ C in 0.20 M sodium borate at pH 9.2 plus 50 μM oxonate with SC-mPEG5k at the molar ratio of 200 to uricase subunits. (a) Reactivity of the PEGylated uricase with rabbit anti-uricase sera. The rabbit anti-uricase sera were diluted to quantify its IgG bound to an antigen of 1.0 μg coated in microplate wells, and quantification of captured IgG was made with a goat anti-IgG (rabbit) secondary antibody (IgG) labeled with horseradish peroxidase. Bovine serum albumin (BSA) was coated to test cross-reactivity. (b) Detection of anti-uricase IgG in rat sera after immunization. The native uricase of 1.0 μg was coated in microplate wells; rat sera after immunization with the uricase and the PEGylated uricase were diluted by 1:90 and 1:1, respectively. Then, each sample serum (200 μL) was added into wells for capture reaction. The bound IgG was quantified with a goat anti-IgG (rat) secondary antibody labeled with horseradish peroxidase following the standard protocol of sandwich enzyme-linked immunosorbent assay (ELISA). (c) Detection of anti-mPEG IgM in rat sera after immunization. Rats were immunized with the PEGylated uricase as described in context. The PEGylated BSA of 1.0 μg was coated in microplate wells and rat sera were diluted by 1:270 and then added into wells for capture reaction. The bound IgM was quantified with a goat anti-IgM (rat) secondary antibody labeled with horseradish peroxidase following the standard protocol of sandwich ELISA. (d) Changes of uricase activities in cock plasma. (e) Changes of uric acid levels in cock plasma.
Biotechnology and Applied Biochemistry
689
Biotechnology and Applied Biochemistry was close to 25 H (Fig. 5d), consistent with that in rats after PEGylation with SS-mPEG5k at a comparable level [15]. The PEGylated uricase showed clear indications to lower uric acid levels in cock plasma in vivo (Fig. 5e). Notably, in the first hour after administration, native uricase and the PEGylated uricase rapidly decreased uric acid levels in cock plasma. After that point, however, plasma uric acid levels recovered rapidly with the native uricase, but remained at lowered levels for more than 20 H with the PEGylated uricase. Thus, the use of SCmPEG5k in great molar excess to PEGylate the uricase in the presence of oxonate is effective to enhance its pharmacological properties; it is feasible to obtain a therapeutic formulation after careful purification of the PEGylated uricase. In theory, PEGylation is a classical approach to formulation of proteins as biodrugs. Some new strategies are reported to formulate protein drugs, especially uricases bearing an unfavorable optimum pH of about 9.0 [38–40]. For example, the encapsulation of uricase in an alkalinized liposome also greatly improves the pharmacological actions of uricase at no cost to immunogenicity. However, the stability of liposome in vivo is a concern when long-term administration of formulated uricase to patients of refractory gout is considered. It is expected that the encapsulation of PEGylated uricase in an alkalinized liposome may be an even better way to formulate uricase whose optimum pH is about 9.0.
4. Conclusion The use of a NHS carbonate of mPEG in the presence of a competitive inhibitor for PEGylation of the bacterial uricase via N-acylation is an effective and preferable approach for its formulation. The protection of enzymes by their competitive inhibitors from inactivation during PEGylation may have some universal significance.
5. Acknowledgements The work was supported by the National Natural Science Foundation of China (Nos. 30672009, 81071427), Natural Science Foundation Project of CQ (CSTC2012JJA0057), and the Education Ministry of China (No. 20125503110007).
6. References [1] [2] [3] [4]
Edwards, N. L. (2008) Arthritis Rheum. 58, 2587–2590. Fels, E., and Sundy, J. S. (2008) Curr. Opin. Rheumatol. 20,198–202. Richette, P., and Bardin, T. (2010) Lancet 375, 318–328. Sherman, M. R., Saifer, M. G., and Perez-Ruiz, F. (2008) Adv. Drug. Deliv. Rev. 60, 59–68. [5] Chohan, S., and Becker, M. A. (2009) Curr. Opin. Rheumatol. 21, 143–149. [6] Terkeltaub, R. (2010) Nat. Rev. Rheumatol. 6, 30–38. [7] Schlesinger, N., Yasothan, U., and Kirkpatrick, P. (2011) Nat. Rev. Drug Discov. 10, 17–18.
690
[8] Yang, X. L., Yuan, Y. H., Zhan, C. G., and Fei, L. (2012) Drug Dev. Res. 73, 66–72. [9] Harris, J. M., and Chess, R. B. (2003) Nat. Rev. Drug Discov. 2, 214–221. [10] Veronese, F. M., and Pasut, G. (2005) Drug Discov. Today 10, 1451–1458. [11] Marsac, Y., Cramer, J., Olschewski, D., Alexandrov, K., and Becker, C. F. (2006) Bioconjug. Chem. 17, 1492–1498. [12] Veronese, F. M., and Mero, A. (2008) Biodrugs 22, 315–329. [13] Pasut, G., and Veronese, F. M. (2009) Adv. Drug Deliv. Rev. 61, 1177–1188. [14] Zhang, C., Yang, X. L., Yuan, Y. H., Pu, J., and Liao, F. (2012) Biodrugs 26, 209–215. [15] Zhang, C., Yang, X. L., Feng, J., Yuan, Y. H., Li, X., Bu, Y. Q., Xie, Y. L., Yuan, H. D., and Liao, F. (2010) Biosci. Biotechnol. Biochem. 74, 1298–1301. [16] Sundy, J. S., Ganson, N. J., Kelly, S. J., Scarlett, E. L., Rehrig, C. D., Huang, W., and Hershfield, M. S. (2007) Arthritis Rheum. 56, 1021–1028. [17] Yue, C. S., Huang, W., Alton, M., Maroli, A. N., Waltrip, R. W., Wright, D., and Marco, M. D. (2008) J. Clin. Pharmacol. 48, 708–718. [18] Tsuji, J., Hirose, K., Kasahara, E., Naitoh, M., and Yamamoto, I. (1985) Int. J. Immunopharmacol. 7, 725–730. [19] Ganson, N. J., Kelly, S. J., Scarlett, E., Sundy, J. S., and Hershfield, M. S. (2006) Arthritis Res. Ther. 8, R12. [20] Wang, X. Y., Ishida, T., and Kiwada, H. (2007) J. Control. Release 119, 236–244. [21] Ishida, T., and Kiwada, H. (2008) Int. J. Pharm. 354, 56–62. [22] Zhao, Y. S., Zhao, L. N., Yang, G. Q., Tao, J., Bu, Y. Q., and Liao, F. (2006) Biotechnol. Appl. Biochem. 45, 75–80. [23] Feng, J., Li, X., Yang, X. L., Zhang, C., Yuan, Y. H., Pu, J., Zhao, Y. S., Xie, Y. L., Yuan, H. D., Bu, Y. Q., and Liao, F. (2010) Arch. Pharm. Res. 33, 1761–1769. [24] Zalipsky, S., Seltzer, R., and Menon-Rudolph, S. (1992) Biotechnol. Appl. Biochem. 15, 100–114. [25] Gao, A., Yang, X. L., Zhang, C., Long, G. B., Pu, J., Yuan, Y. H., Long, G. B., Li, Y. L., and Liao, F. (2012) Chem. Cent. J. 6, 142. [26] Mero, A., Pasut, G., Dalla, V. L., Fijten, M. W., Schubert, U. S., Hoogenboom, R., and Veronese, F. M. (2008) J. Control. Release 125, 87–95. [27] Schiavon, O., Caliceti, P., Ferruti, P., and Veronese, F. M. (2000) Farmaco 55, 264–269. [28] Rendina, A. R., and Cheng, D. (2005) Biochem. J. 388, 895–903. [29] Milgrom, E. M., and Milgrom, Y. M. (2012) Biochem. Biophys. Res. Commun. 423, 355–359. [30] Li, X., Feng, J., Zhang, C., Pu, J., Yang, X. L., Yuan, Y. H., and Liao, F. (2011) Chin. J. Appl. Environ. Biol. 17, 120–123. [31] Feng, J., Liu, H. B., Yang, X. L., Gao, A., Liao, J., Feng, L. P., Pu, J., Xie, Y. L., Long, G. B., Li, Y. L., and Liao, F. (2013) Chem. Cent. J. 7, 69. [32] Liao, F., Zhu, X. Y., Wang, Y. M., and Zuo, Y. P. (2005) J. Biochem. Biophys. Methods 62, 13–24. [33] Liao, F., Xie, Y. L., Yang, X. L., Deng, P., Chen, Y. W., Xie, G. M., Zhu, S., Liu, B. Z., Yuan, H. D., Liao, J., Zhao, Y. S., and Yu, M. A. (2009) Biosens. Bioelectron. 25, 112–117. [34] Xie, Y. L., Yang, X. L., Pu, J., Zhao, Y. S., Zhang, Y., Xie, G. M., Zheng, J., Yuan, H. D., and Liao, F. (2010) Spectrochim. Acta A Mol. Biomol. Spectrosc. 77, 869–876. [35] Liao, F., Yang, D. Y., Tang, J. Q., Yang, X. L., Liu, B. Z., Zhao, Y. S., Zhao, L. N., Liao, H., and Yu, M. A. (2009) Clin. Biochem. 42, 926–928. [36] Liao, F., Zhao, L. N., Zhao, Y. S., Tao, J., and Zuo, Y. P. (2007) Anal. Sci. 23, 439–444. [37] Bradford, M. M. (1976) Anal. Biochem. 72, 248–254. [38] Tan, Q. Y., Wang, N., Yang, H., Zhang, L. K., Liu, S., Chen, L., Liu, J., Zhang, L., Hu, N. N., Zhao, C. J., and Zhang, J. Q. (2010) Int. J. Pharm. 384,165–72. [39] Tan, Q., Zhang, J., Wang, N., Li, X., Xiong, H., Teng, Y., He, D., Wu, J., Zhao, C., Yin, H., and Zhang, L. (2012) Eur. J. Pharm. Biopharm. 82, 43–48. [40] Tan, Q. Y., Zhang, J. Q., Wang, N., Yang, H., Li, X., Xiong, H. R., Wu, J. Y., Zhao, C. J., Wang, H., and Yin, H. F. (2012) Int. J. Nanomed. 7, 3929–3938.
PEGylation of a Uricase with Active Esters of mPEG
Copyright of Biotechnology & Applied Biochemistry is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
Chun Zhang † Xiaolan Yang † Ang Gao Xiaolei Hu Jun Pu Hongbo Liu Juan Feng Juan Liao Yuanli Li ∗ Fei Liao
Unit for Analytical Probes and Protein Biotechnology, Key Laboratory of Clinical Laboratory Diagnostics of the Education Ministry, College of Laboratory Medicine, Chongqing Medical University, Chongqing, People’s Republic of China
Abstract Uricase after modification with monomethoxy poly(ethylene glycol) (mPEG) is currently the sole agent to treat refractory gout. For formulating Bacillus fastidious uricase, succinimidyl carbonate of mPEG-5000 (SC-mPEG5k) and succinimidyl succinate of mPEG-5000 (SS-mPEG5k) were compared. SC-mPEG5k possessed higher purity, comparable reaction rate constant with glycine but lower hydrolysis rate, and stronger effectiveness to modify amino groups. The uricase possessed two types of amino groups bearing a 25-fold difference in reactivity with SC-mPEG5k or SS-mPEG5k at pH 9.2. Oxonate and xanthine concentration-dependently protected the bacterial uricase from inactivation during PEGylation. With SC-mPEG5k at a molar ratio of 200 to uricase subunits and
oxonate of 50 µM, the PEGylated uricase (1) retained about 73% of the original activity, (2) displayed about 10% reactivity to rabbit anti-sera recognizing the native uricase, (3) elicited IgG in rats accounting for about 5% of that by the native uricase, (4) exhibited circulation half-life time of about 25 H in cock plasma in vivo, and (5) concurrently maintained uric acid at lowered levels for over 20 H. Hence, PEGylation with SC-mPEG under the protection of a competitive inhibitor was a practical approach to formulation of the bacterial uricase; protection of enzymes by competitive inhibitors during C 2014 International PEGylation may have universal significance. Union of Biochemistry and Molecular Biology, Inc. Volume 61, Number 6, Pages 683–690, 2014
Keywords: uricase, PEGylation, competitive inhibitor, succinimidyl carbonate of mPEG, protection
1. Introduction Abbreviation: BSA, bovine serum albumin; DMF, N-dimethylforamide; ELISA, enzyme-linked immunosorbent assay; mPEG, monomethoxyl poly(ethylene glycol); mPEG5k-uricase, uricase PEGylated with SC-mPEG5k at molar ratio of 200 to uricase subunits; NHS, N-hydroxysuccinimide; SC-mPEG5k, NHS ester of carbonate of mPEG-5000; THF, tetrahydrofurane; TNBS, 2,4,6,-trinitrobenzenesulfonic acid. ∗ Address
for correspondence: Fei Liao, MSc, Unit for Analytical Probes and Protein Biotechnology, Key Laboratory of Clinical Laboratory Diagnostics of the Education Ministry, College of Laboratory Medicine, Chongqing Medical University, Chongqing, People’s Republic of China. Tel.: +86 23 68485240; Fax: +86 23 68485240; e-mail: [email protected]. †These authors contributed equally to this work. Supporting Information is available in the online issue at wileyonlinelibrary.com. Received 22 August 2013; accepted 5 February 2014 DOI: 10.1002/bab.1215 Published online 28 March 2014 in Wiley Online Library (wileyonlinelibrary.com)
Gout is a chronic metabolic disorder requiring treatment with anti-urate drugs like allopurinol throughout the life of patients [1–3]. Refractory gout is characterized by the ineffectiveness/inapplicability of allopurinol, and uricase (EC 1.7.3.3) is the sole agent effective for refractory gout [4–8]. To prolong circulation time in vivo and mask immunogenic sites of uricases as exogenous proteins, their modification with monomethoxyl poly(ethylene glycol) (mPEG), PEGylation, is a plausible approach [9–14]. In general, PEGylation of residues of wide distribution on uricase surfaces is preferred, and Nacylation of amino groups with N-hydroxysuccinimide (NHS) ester of mPEG in excess is practical [14]. However, PEGylated uricases show lower activities [15–17]. To treat refractory gout, PEGylated uricases have to be administered at large doses and short intervals [16, 17], which induce antibodies against mPEG to shorten the circulation time of PEGylated uricases [18–21]. For continuous treatment of refractory gout, biweekly
683
Biotechnology and Applied Biochemistry therapeutic doses of PEGylated uricases should be smaller than 0.2 mg [8]. Hence, the process to PEGylate any uricase should be carefully optimized for the activity of the PEGylated uricase as high as possible. A uricase from Bacillus fastidious ATCC 29604 has high activity and excellent thermostability, supporting the significance of formulation to test its therapeutic potential [15, 22, 23]. NHS ester of succinate of mPEG-5000 (SS-mPEG5k) and NHS ester of carbonate of mPEG-5000 (SC-mPEG5k) are representative agents to PEGylate proteins via N-acylation [14, 15, 24]. SC-mPEG5k and SS-mPEG5k undergo both N-acylation and hydrolysis during PEGylation of proteins, but they still possess some differences related to PEGylation of proteins. For instance, SC-mPEG5k can be easily purified to higher homogeneity, but SS-mPEG5k is usually contaminated with unreactive mPEG [24, 25]. The hydrolysis rate constant of SC-mPEG5k at pH 8.0 and 25 ◦ C is just about half of that of SS-mPEG5k [24–26] (http://laysanbio.com/clientuploads/mPEGSVA final review.doc). Thus, smaller quantities of SC-mPEG5k can be utilized to produce the same degree of PEGylation and thus facilitate purifying the PEGylated products. On the other hand, ligands to active sites of some enzymes protect them from inactivation during modification [27–29]. Oxonate and xanthine are competitive inhibitors of uricases and have negligible reactivity to SC-mPEG5k or SS-mPEG5k [15, 22, 30]. Herein, we compared PEGylation of the bacterial uricase with SC-mPEG5k and SS-mPEG5k, with or without the protection of oxonate and xanthine, and results indicated that SC-mPEG5k was more favorable for PEGylation of the bacterial uricase under the protection of competitive inhibitors.
2. Materials and Methods 2.1. Materials and chemicals Monomethoxy polyethylene glycol 5000 (mPEG5k), 2,4,6,trinitrobenzenesulfonic acid (TNBS), and diethylaminoethyl (DEAE) cellulose were from Sigma–Aldrich (St Louis, MO, USA). Xanthine, NHS, uric acid, and triphosgene were from Alfa Aesar (Tainjing, China). Potassium oxonate was from Shandong Zhongketaidou Chemical Co. (http://en.sdzktd.com/) (Jinan, Shandong, China). N-Dimethylforamide (DMF) was pretreated with acetate anhydride and restilled. Bovine serum albumin (BSA), tributylamine, and other chemicals were domestic reagents. Goat anti-IgG (rabbit) secondary antibody (IgG), antiIgG (rat) secondary antibody (IgG), and anti-IgM (rat) secondary antibody (IgG) were labeled with horseradish peroxidase (Sangon Bioengineering Co., Shanghai, China), and were utilized as received following the standard protocol.
2.2. Preparation of a recombinant bacterial uricase Uricase form B. fastidious ATCC 29604 was expressed in Escherichia coli BL21 (DE3) [15, 31]. After lysis of cells by sonication treatment, soluble uricase was purified by DEAE cellulose chromatography via elution with 0.10 M Tris–HCl at
684
pH 8.0 plus a linear gradient of NaCl from 0.10 to 0.40 M. Eluted portions with activities over 6 kU g−1 were collected and dialyzed against 0.20 M sodium borate buffer at pH 9.2 for 24 H with four changes of the buffer. Other proteins in the preparation were undetectable by SDS-PAGE.
2.3. Uricase activity assay Activities of uricase were determined by quantifying absorbance at 293 nm on a Shimadzu UV 2550 spectrophotometer (Kyoto, Japan) equipped with a thermostatic apparatus at 25 ◦ C in 0.20 M sodium borate buffer at pH 9.2 containing 75 µM uric acid [31]. Absorptivity of uric acid was 11.5 mM−1 cm−1 [32]. The change of absorbance after 30 Sec since the reaction initiation by the addition of a sample was recorded at 5 Sec intervals. One unit of uricase was the quantity able to oxidize 1 µmol uric acid in one Min. To measure uricase in cock serum, a sample of 50 µL was added and the change of absorbance was measured at 60 Sec intervals.
2.4. Preparation of SS-mPEG5k and SC-mPEG5k In a three-necked flask, 10 g mPEG5k was dissolved in 30 mL CH2 Cl2 ; 0.89 g triphosgene was dissolved in 5 mL redistilled toluene and added into mPEG5k solution drop by drop; the mixture was stirred for 18 H at room temperature [24]. Then, the reaction mixture was concentrated to 10 mL under reduced pressure at 40 ◦ C and the targeted intermediate was precipitated by the addition of 100 mL diethyl ether. To further purify the intermediate, its dissolution in 10 mL CH2 Cl2 and precipitation by the addition of 100 mL diethyl ether were repeated twice to give a white powder that was finally dissolved in 30 mL CH2 Cl2 again. To obtain SC-mPEG5k, 0.29 g NHS dissolved in 4 mL tetrahydrofurane (THF) was added to the solution of the intermediate. The resulting mixture was stirred for 8 H at room temperature and then concentrated to 10 mL under reduced pressure at 40 ◦ C. The final product, SC-mPEG5k, was precipitated by the addition of 100 mL diethyl ether. To purify SC-mPEG5k, the resulting powder was dissolved in 10 mL THF and precipitated by the addition of 100 mL diethyl ether for three times to give a white powder bearing 80% yield in the mass of mPEG5k (Fig. S1, Supplementary Data in the Supporting Information). Reactivity of SC-mPEG5K with amino groups was verified via the detection of the fluorescent derivative after reaction with N-(1-naphthyl)-ethylenediamine [33, 34]. SS-mPEG5k was prepared as before [15]. Molar equivalency of NHS ester was estimated via the reaction with ethanolamine and TNBS assay of residual ethanolamine [25].
2.5. Characterization of reaction kinetics of SC-mPEG5k and SS-mPEG5k At pH 9.2 in 0.20 M sodium borate buffer, the bacterial uricase shows excellent thermostability and NHS is reasonably stable. During hydrolysis or reaction with amino groups in the borate buffer at pH 9.2, reaction curves of SS-mPEG5k or SC-mPEG5k were continuously recorded at 1 Sec intervals by the absorbance at 260 nm; background absorbance of the solution in the absence of the NHS ester was corrected [24–26]. To
PEGylation of a Uricase with Active Esters of mPEG
characterize hydrolysis of SS-mPEG5k or SC-mPEG5k, the final levels were 80, 60, and 40 µM; reaction curves were recorded at 20, 25, and 30 ◦ C, separately. To characterize the reaction with amino group of glycine, the final level of SS-mPEG5k or SC-mPEG5k was fixed at 80 µM, whereas the final levels of glycine were 5.0, 10.0, and 20.0 mM to record reaction curves at 20, 25, and 30 ◦ C, separately. All reaction curves were recorded within 10 Min on the Shimadzu UV 2550. Under stated conditions, the concentration of hydroxyl oxide can be assumed to be a constant during hydrolysis and the final levels of glycine can be assumed to be constants during reaction with either NHS ester. The hydrolysis of a NHS ester or its reaction with an amino group of glycine should apparently follow first-order kinetics. The rate constant for hydrolysis served as background in apparent rate constants for reactions of glycine with SS-mPEG5k or SC-mPEG5k. The program for analyzing enzyme reaction curves at substrate levels much lower than the Michaelis–Menten constant was used [35, 36]; the goodness-of-fit was judged by determination coefficients to search for combinations of parameters.
2.6. PEGylation of uricase with SC-mPEG5k or SS-mPEG5k and effects of ligands Uricase was PEGylated with SC-mPEG5k or SS-mPEG5k in 0.20 M sodium borate buffer at pH 9.2 and 6 ± 2 ◦ C. To determine the effects of molar ratios of a NHS ester to uricase amino groups, an indicated quantity below 5.0 mg of a NHS ester of mPEG5k dissolved in DMF was added drop by drop with fine tips into 1.5 mL uricase solutions (1.7 g L−1 uricase, final DMF was below 2%) in sodium borate buffer at pH 9.2 under mild stirring, or its solid of over 5 mg was directly added into uricase solutions. After reaction for 90 Min, 950 µL mixtures were withdrawn to react with 50 µL aqueous TNBS solution for 90 Min to quantify amino groups [15, 25]; the modification degree was the percentage of amino groups lost. Meanwhile, the remaining solution in 550 µL was immediately quenched with 10 µL saturated solution of glycine in the sodium borate buffer at 25 ◦ C to measure residual uricase activity. To examine the effects of ligands on PEGylation, a stock solution of 2.0 mM oxonate potassium, 2.0 mM xanthine, or 6 mM uric acid was made in 0.20 M borate buffer at pH 9.2. To solutions of uricase in 0.20 M sodium borate buffer at pH 9.2 and 6 ± 2 ◦ C, one ligand at different quantities was added, then SC-mPEG5k or SS-mPEG5k in a minimum volume of DMF for a 200 molar ratio to uricase subunit was added to give the final uricase of 1.7 g L−1 in a 1.5 mL solution. After reaction for 90 Min at 6 ± 2 ◦ C, the modification degree and residual uricase activities were quantified as described above. To examine potential immunogenicity of the PEGylated uricase, it was dialyzed against 10 mM Tris–HCl buffer at pH 8.8 and 4 ◦ C for 8 H. BSA was PEGylated with SC-mPEG to prepare the conjugate mPEG5k–BSA in a similar way for the quantification of antibodies specific for mPEG5k, assuming no epitopes of
Biotechnology and Applied Biochemistry
BSA have the same immunoreactivity as those of the bacterial uricase.
2.7. Immunization of rats with uricase and SC-mPEG5k-modified uricase To prepare antibodies specific for the native uricase, rabbit was immunized via subcutaneous multipoint administration of 200 µg of the purified uricase every 10 days. Freund’s complete adjuvant was used the first time, and Freund’s incomplete adjuvant was used in subsequent immunization for the following 40 days. Rabbit serum was collected 3 days after the last intravenous injection of 200 µg of the native uricase and stored at −70 ◦ C until use. To test immunogenicity of each sample, seven male rats were employed in each group. Rats were immunized via subcutaneous administration of 50 µg antigen once a week for four weeks. Rat sera were collected through orbital venous and stored at −70 ◦ C until analysis.
2.8. Detection of antibodies in rat sera Each well was incubated for 2 H at 37 ◦ C with 200 µL/well of a coating antigen at 10.0 mg L−1 in 0.20 M bicarbonate buffer at pH 9.2 (the native uricase, SC-mPEG5k-modified uricase, or SC-mPEG5k-modified BSA); each well was then washed with 300 µL of 0.15 M sodium phosphate buffer at pH 7.2 for three times, and blocked with 300 µL coating solution of 50 g L−1 BSA. A properly diluted rat serum or other sample at 200 µL was added to a well for a capture reaction of 1 H at 37 ◦ C. Then, the sample was discarded and the well was washed three times with 200 µL of 0.15 M sodium phosphate buffer at pH 7.2 containing 0.055% Tween 20; a working solution of horseradish peroxidase-labeled goat anti-rat or anti-rabbit IgG/M antibody at 200 µL was added to each well for incubation of 1 H at 37 ◦ C following the standard operation guidelines. After washing of each well with 0.15 M sodium phosphate buffer at pH 7.2 for five times as described above, a 200 µL 3,3 , 5,5 -tetramethyl benzidine hydrochloride working solution was added for a 30 Min reaction at 37 ◦ C and the reaction was terminated with 50 µL H2 SO4 at 2.0 M. Absorbance at 450 nm was measured in 10 Min with a Biotek ELX 800 microplate reader. Data were expressed as mean ± SD.
2.9. Pharmacological properties of the uricase modified with SC-mPEG5k Cocks inherently bear hyperuricemia for testing pharmacological actions of uricase. Cocks were raised in the animal center of Chongqing Medical University under clean conditions. The native or PEGylated uricase was administered through wing-root intravenous injection at 2 U per kg body weight (measured at pH 7.4 in 50 mM sodium borate buffer, just about 25% activity of that at pH 9.2) to two groups of six male cocks (about 1.7 kg). Blood was obtained from the wing-root vein with a heparinized syringe at different times in 48 H; red blood cells were removed immediately by centrifugation at 430 g for 5 Min to obtain plasma. With 200 µL plasma, proteins were immediately
685
Biotechnology and Applied Biochemistry precipitated via the addition of 20 µL 65% HClO4 and vigorous vortex for 5 Min. Then, 20 µL potassium carbonate solution saturated at 25 ◦ C was slowly added to adjust pH of the mixture to about 7.4; 30 Min later, the mixture was centrifuged at 4,700 g for 10 Min to give a clear supernatant to measure plasma uric acid. Meanwhile, plasma uricase activity was measured at pH 9.2 in sodium borate buffer in 5.0 Min as described above [22]. Plasma uricase activities were analyzed with a biexponential model to eliminate interference of partially modified uricase [23].
2.10. Analysis of plasma uric acid by a direct kinetic uricase method To avoid interference from uricase inhibitors and reductants/ oxidants, a direct kinetic uricase method was used to quantify uric acid in cock plasma [22]. In brief, for a 1.165 mL solution of 0.20 M sodium borate at pH 9.2 plus a 30 µL sample, the absorbance at 293 nm was taken as the initial absorbance after the correction of dilution effects (0.4%); then, a 5 µL concentrated solution of recombinant B. fastidious uricase was added for a final 40 U L−1 to monitor the change of absorbance at 293 nm and at 25 ◦ C. With each sample, a uricase reaction curve was recorded at 5 Sec intervals within 7.0 Min, stored in a text file, and read into memory for analysis by a program. The Michaelis–Menten constant of the uricase was preset at 0.28 mM; the background absorbance at 293 nm after the completion of the uricase reaction was predicted. When the difference between the corrected initial absorbance and the absorbance after the reaction for 7.0 Min was less than 0.050, the absorbance after the reaction for 7.0 Min was directly taken as the background absorbance. The difference between the initial absorbance and the background absorbance indexed the net absorbance of uric acid in the reaction solution. The assay had the detection limit of about 1 µM, whereas the upper limit of about 35 µM.
2.11. Analyses of proteins and amino groups Protein was quantified by the Bradford method with BSA as the reference [37]. SDS-PAGE using 8% separation gel was used to analyze compositions of the PEGylated uricases, and protein bands were stained with Coomassie Brilliant Blue R250. Amino groups were quantified by reaction with TNBS to measure the absorbance at 420 nm with glycine as the reference [22, 25].
3. Results and Discussion 3.1. PEGylation of the recombinant uricase In general, PEGylation of a protein via N-acylation with a NHS ester of mPEG prefers alkaline pH as long as the protein is stable (Datasheet 1, Supplementary Data). The bacterial uricase is stable at pH 9.2; the optimum pH for PEGylation with SC-mPEG5k was about 9.2 [24]. Thus, PEGylation of the uricase was tested at pH 9.2 hereafter.
686
FIG. 1
PEGylation of the uricase in 0.20 M sodium borate at pH 9.2 and 6 ◦ C. (a) SC-mPEG5k and (b) SS-mPEG5k.
Activities of the PEGylated uricase as mixtures of PEGylated isoforms decreased in an exponential manner, whereas modification degrees of amino groups followed biphasic increases, with respect to molar ratios of SC-mPEG5k to uricase subunits (Fig. 1a). The modification degrees linearly and rapidly increased to 60% with respect to molar ratios below 100 of SC-mPEG5k to uricase subunits; amino groups reactive to SCmPEG5k at such levels belonged to exposed ones. When final levels of SC-mPEG5k were further increased, activities of the PEGylated uricase became smaller and smaller, whereas the modification degrees grew slowly to about 80% when the molar ratio of SC-mPEG5k to uricase subunits reached 200; then modification degrees became a plateau of below 85% when molar ratios of SC-mPEG5k to uricase subunits were over 250. Amino groups PEGylated only at molar ratios of SC-mPEG5k to uricase subunits over 200 belonged to partially buried ones. From response slopes of modification degrees with respect to molar ratios of SC-mPEG5k to uricase subunits, there was about a 25-fold difference in reactivity with SC-mPEG5k between the exposed and partially buried amino groups. Moreover, TNBS accessed only about 12 amino groups on each subunit of the
PEGylation of a Uricase with Active Esters of mPEG
bacterial uricase. Hence, there should be not more than three partially buried amino groups accessible by SC-mPEG5k in each subunit of the uricase. Similarly, there were biphasic changes of modification degrees of amino groups, whereas there was an exponential decrease in activities of the PEGylated uricase with respect to molar ratios of SS-mPEG5k to uricase subunits (Fig. 1b). Based on response slopes of modification degrees to levels of SS-mPEG5k, the difference in reactivity of two types of uricase amino groups with SS-mPEG5k was comparable to that with SC-mPEG5k. Interestingly, when molar amounts of SS-mPEG5k in its preparation were estimated from molecular weight and mass quantities, there were lower modification degrees but higher activities of the uricase after PEGylation with SS-mPEG5k than those with SC-mPEG5k at the same molar ratios of below 200 to uricase subunits [22]. However, when the final levels of SS-mPEG5k were increased until the modification degrees were the same as those with SC-mPEG5k, activities of the PEGylated uricase were consistent with those modified with SC-mPEG5k. Therefore, (1) activities of the PEGylated uricase depend on modification degrees of amino groups; (2) to mask all epitopes of the bacterial uricase, a NHS ester of mPEG should be in great molar excess to uricase subunits for PEGylation via Nacylation; (3) any approach is desirable to protect the uricase from inactivation during PEGylation.
3.2. Protection of the bacterial uricase by competitive inhibitors Oxonate and xanthine are potent competitive inhibitors of uricases. To this bacterial uricase at pH 9.2 and 6 ◦ C, oxonate, xanthine, and uric acid showed affinities of about 3 µM, 20 µM, and 0.18 mM, respectively [15, 30]. Oxonate, xanthine, or uric acid protected this uricase from inactivation by SC-mPEG5k in a concentration-dependent manner, but caused no significant changes of modification degrees when the molar ratio of SCmPEG5k to uricase subunits was kept at 200 (Fig. 2). Obviously, such protection effects were easily saturated with oxonate and xanthine, but could not be saturated with uric acid owing to strong action of the uricase at the tested levels. Inhibitors bearing higher affinities produced stronger protection effects on the uricase during PEGylation. After PEGylation with SCmPEG5k at the molar ratio of 200 to uricase subunits, the PEGylated uricase retained over 70% of the activity of the native uricase when 50 µM oxonate was used, whereas it retained just about 50% of the activity of the native uricase in the absence of all ligands. When molar ratios of SC-mPEG5k to uricase subunits were 350, activities of the uricase PEGylated even in the presence of 50 µM oxonate decreased to just about 25% versus about 10% in the absence of oxonate, but there were still no detectable differences in the modification degrees. During PEGylation with SS-mPEG5k, similar protection effects of oxonate or xanthine and similar dependence of such protection effects on molar ratios of SS-mPEG5k to uricase subunits were observed (Fig. 3). Hence, the protection effects of competitive
Biotechnology and Applied Biochemistry
FIG. 2
Protection of the uricase during PEGylation at 6 ◦ C in 0.20 M sodium borate at pH 9.2 with SC-mPEG5k at the molar ratio of 200 to uricase subunits. (a) Oxonate, (b) xanthine, and (c) uric acid.
inhibitors on the uricase during PEGylation depend on the final levels of PEGylation agents. To ensure favorable activities of the PEGylated uricase, any PEGylating agent should be restricted to reasonable levels; molecular engineering of the bacterial uricase for site-specific PEGylation via N-acylation to mask
687
Biotechnology and Applied Biochemistry
FIG. 4
FIG. 3
Protection of the uricase during PEGylation at 6 ◦ C in 0.20 M sodium borate at pH 9.2 with SS-mPEG5k at the molar ratio of 200 to uricase subunits. (a) Oxonate and (b) xanthine.
all immunogenic sites while retaining activities as much as possible is preferred [14]. The binding of any ligand caused negligible changes of the molecular weights of conjugates of the uricase and mPEG5k analyzed by SDS-PAGE (Fig. 4). There was a small amount of the native uricase leftover in each preparation of conjugates. The two principal bands of conjugates had molecular weights of about 56 and 66 kDa, respectively, and a minor band of the conjugate had a molecular weight of about 76 kDa, indicating that the number of mPEG5k chains attached to each subunit should be less than 8. Oxonate, xanthine, and uric acid all carried negative charges at pH 9.2; their bindings may involve ionic interactions with some essential amino groups of the bacterial uricase to reduce their reactivity and thus retard PEGylation of such essential amino groups. The protection of enzymes by their competitive inhibitors from inactivation during PEGylation may be universally effective; the investigation of such protection effects of enzyme inhibitors may reinforce PEGylation of common exogenous proteins as potential biodrugs.
688
SDS-PAGE analysis after PEGylation at 6 ◦ C in 0.20 M sodium borate at pH 9.2 with SC-mPEG5k at the molar ratio of 200 to uricase subunits. Lanes 1–5: the use of 30 μM oxonate, the use of 100 μM xanthine, the use of 800 μM uric acid, the use of no ligands, and the native uricase alone.
3.3. Pharmacological properties of the PEGylated uricase The uricase was PEGylated in the presence of 50 µM oxonate with SC-mPEG5k at the molar ratio of 200 to uricase subunits for 60 Min reaction at 6 ◦ C in 0.20 M sodium borate buffer at pH 9.2. The PEGylated uricase retained about 73% activity of that of the native uricase. To test the suitability of the PEGylation process to the uricase, pharmacological properties of the mixture of the PEGylated uricase were directly evaluated. Of the PEGylated uricase, the binding capability in vitro to specific anti-uricase sera of rabbit was just about 10% of that of the native uricase (Fig. 5a); immunogenicity in vivo in rats was decreased to just about 5% of the native uricase based on the quantities of uricase-specific IgG elicited (Fig. 5b). However, at just 1 week after the first immunization with the PEGylated uricase, IgM in rat sera specific for mPEG5k was detected (Fig. 5c). The native uricase had a circulation half-life time in vivo of about 40 Min based on changes of uricase activities in cock plasma. Using a biexponential decay model to analyze the changes of uricase activities [23], the longest circulation half-life time of the PEGylated uricase in cock plasma in vivo
PEGylation of a Uricase with Active Esters of mPEG
FIG. 5
Preliminary evaluation of pharmacological properties of the uricase after PEGylation at 6 ◦ C in 0.20 M sodium borate at pH 9.2 plus 50 μM oxonate with SC-mPEG5k at the molar ratio of 200 to uricase subunits. (a) Reactivity of the PEGylated uricase with rabbit anti-uricase sera. The rabbit anti-uricase sera were diluted to quantify its IgG bound to an antigen of 1.0 μg coated in microplate wells, and quantification of captured IgG was made with a goat anti-IgG (rabbit) secondary antibody (IgG) labeled with horseradish peroxidase. Bovine serum albumin (BSA) was coated to test cross-reactivity. (b) Detection of anti-uricase IgG in rat sera after immunization. The native uricase of 1.0 μg was coated in microplate wells; rat sera after immunization with the uricase and the PEGylated uricase were diluted by 1:90 and 1:1, respectively. Then, each sample serum (200 μL) was added into wells for capture reaction. The bound IgG was quantified with a goat anti-IgG (rat) secondary antibody labeled with horseradish peroxidase following the standard protocol of sandwich enzyme-linked immunosorbent assay (ELISA). (c) Detection of anti-mPEG IgM in rat sera after immunization. Rats were immunized with the PEGylated uricase as described in context. The PEGylated BSA of 1.0 μg was coated in microplate wells and rat sera were diluted by 1:270 and then added into wells for capture reaction. The bound IgM was quantified with a goat anti-IgM (rat) secondary antibody labeled with horseradish peroxidase following the standard protocol of sandwich ELISA. (d) Changes of uricase activities in cock plasma. (e) Changes of uric acid levels in cock plasma.
Biotechnology and Applied Biochemistry
689
Biotechnology and Applied Biochemistry was close to 25 H (Fig. 5d), consistent with that in rats after PEGylation with SS-mPEG5k at a comparable level [15]. The PEGylated uricase showed clear indications to lower uric acid levels in cock plasma in vivo (Fig. 5e). Notably, in the first hour after administration, native uricase and the PEGylated uricase rapidly decreased uric acid levels in cock plasma. After that point, however, plasma uric acid levels recovered rapidly with the native uricase, but remained at lowered levels for more than 20 H with the PEGylated uricase. Thus, the use of SCmPEG5k in great molar excess to PEGylate the uricase in the presence of oxonate is effective to enhance its pharmacological properties; it is feasible to obtain a therapeutic formulation after careful purification of the PEGylated uricase. In theory, PEGylation is a classical approach to formulation of proteins as biodrugs. Some new strategies are reported to formulate protein drugs, especially uricases bearing an unfavorable optimum pH of about 9.0 [38–40]. For example, the encapsulation of uricase in an alkalinized liposome also greatly improves the pharmacological actions of uricase at no cost to immunogenicity. However, the stability of liposome in vivo is a concern when long-term administration of formulated uricase to patients of refractory gout is considered. It is expected that the encapsulation of PEGylated uricase in an alkalinized liposome may be an even better way to formulate uricase whose optimum pH is about 9.0.
4. Conclusion The use of a NHS carbonate of mPEG in the presence of a competitive inhibitor for PEGylation of the bacterial uricase via N-acylation is an effective and preferable approach for its formulation. The protection of enzymes by their competitive inhibitors from inactivation during PEGylation may have some universal significance.
5. Acknowledgements The work was supported by the National Natural Science Foundation of China (Nos. 30672009, 81071427), Natural Science Foundation Project of CQ (CSTC2012JJA0057), and the Education Ministry of China (No. 20125503110007).
6. References [1] [2] [3] [4]
Edwards, N. L. (2008) Arthritis Rheum. 58, 2587–2590. Fels, E., and Sundy, J. S. (2008) Curr. Opin. Rheumatol. 20,198–202. Richette, P., and Bardin, T. (2010) Lancet 375, 318–328. Sherman, M. R., Saifer, M. G., and Perez-Ruiz, F. (2008) Adv. Drug. Deliv. Rev. 60, 59–68. [5] Chohan, S., and Becker, M. A. (2009) Curr. Opin. Rheumatol. 21, 143–149. [6] Terkeltaub, R. (2010) Nat. Rev. Rheumatol. 6, 30–38. [7] Schlesinger, N., Yasothan, U., and Kirkpatrick, P. (2011) Nat. Rev. Drug Discov. 10, 17–18.
690
[8] Yang, X. L., Yuan, Y. H., Zhan, C. G., and Fei, L. (2012) Drug Dev. Res. 73, 66–72. [9] Harris, J. M., and Chess, R. B. (2003) Nat. Rev. Drug Discov. 2, 214–221. [10] Veronese, F. M., and Pasut, G. (2005) Drug Discov. Today 10, 1451–1458. [11] Marsac, Y., Cramer, J., Olschewski, D., Alexandrov, K., and Becker, C. F. (2006) Bioconjug. Chem. 17, 1492–1498. [12] Veronese, F. M., and Mero, A. (2008) Biodrugs 22, 315–329. [13] Pasut, G., and Veronese, F. M. (2009) Adv. Drug Deliv. Rev. 61, 1177–1188. [14] Zhang, C., Yang, X. L., Yuan, Y. H., Pu, J., and Liao, F. (2012) Biodrugs 26, 209–215. [15] Zhang, C., Yang, X. L., Feng, J., Yuan, Y. H., Li, X., Bu, Y. Q., Xie, Y. L., Yuan, H. D., and Liao, F. (2010) Biosci. Biotechnol. Biochem. 74, 1298–1301. [16] Sundy, J. S., Ganson, N. J., Kelly, S. J., Scarlett, E. L., Rehrig, C. D., Huang, W., and Hershfield, M. S. (2007) Arthritis Rheum. 56, 1021–1028. [17] Yue, C. S., Huang, W., Alton, M., Maroli, A. N., Waltrip, R. W., Wright, D., and Marco, M. D. (2008) J. Clin. Pharmacol. 48, 708–718. [18] Tsuji, J., Hirose, K., Kasahara, E., Naitoh, M., and Yamamoto, I. (1985) Int. J. Immunopharmacol. 7, 725–730. [19] Ganson, N. J., Kelly, S. J., Scarlett, E., Sundy, J. S., and Hershfield, M. S. (2006) Arthritis Res. Ther. 8, R12. [20] Wang, X. Y., Ishida, T., and Kiwada, H. (2007) J. Control. Release 119, 236–244. [21] Ishida, T., and Kiwada, H. (2008) Int. J. Pharm. 354, 56–62. [22] Zhao, Y. S., Zhao, L. N., Yang, G. Q., Tao, J., Bu, Y. Q., and Liao, F. (2006) Biotechnol. Appl. Biochem. 45, 75–80. [23] Feng, J., Li, X., Yang, X. L., Zhang, C., Yuan, Y. H., Pu, J., Zhao, Y. S., Xie, Y. L., Yuan, H. D., Bu, Y. Q., and Liao, F. (2010) Arch. Pharm. Res. 33, 1761–1769. [24] Zalipsky, S., Seltzer, R., and Menon-Rudolph, S. (1992) Biotechnol. Appl. Biochem. 15, 100–114. [25] Gao, A., Yang, X. L., Zhang, C., Long, G. B., Pu, J., Yuan, Y. H., Long, G. B., Li, Y. L., and Liao, F. (2012) Chem. Cent. J. 6, 142. [26] Mero, A., Pasut, G., Dalla, V. L., Fijten, M. W., Schubert, U. S., Hoogenboom, R., and Veronese, F. M. (2008) J. Control. Release 125, 87–95. [27] Schiavon, O., Caliceti, P., Ferruti, P., and Veronese, F. M. (2000) Farmaco 55, 264–269. [28] Rendina, A. R., and Cheng, D. (2005) Biochem. J. 388, 895–903. [29] Milgrom, E. M., and Milgrom, Y. M. (2012) Biochem. Biophys. Res. Commun. 423, 355–359. [30] Li, X., Feng, J., Zhang, C., Pu, J., Yang, X. L., Yuan, Y. H., and Liao, F. (2011) Chin. J. Appl. Environ. Biol. 17, 120–123. [31] Feng, J., Liu, H. B., Yang, X. L., Gao, A., Liao, J., Feng, L. P., Pu, J., Xie, Y. L., Long, G. B., Li, Y. L., and Liao, F. (2013) Chem. Cent. J. 7, 69. [32] Liao, F., Zhu, X. Y., Wang, Y. M., and Zuo, Y. P. (2005) J. Biochem. Biophys. Methods 62, 13–24. [33] Liao, F., Xie, Y. L., Yang, X. L., Deng, P., Chen, Y. W., Xie, G. M., Zhu, S., Liu, B. Z., Yuan, H. D., Liao, J., Zhao, Y. S., and Yu, M. A. (2009) Biosens. Bioelectron. 25, 112–117. [34] Xie, Y. L., Yang, X. L., Pu, J., Zhao, Y. S., Zhang, Y., Xie, G. M., Zheng, J., Yuan, H. D., and Liao, F. (2010) Spectrochim. Acta A Mol. Biomol. Spectrosc. 77, 869–876. [35] Liao, F., Yang, D. Y., Tang, J. Q., Yang, X. L., Liu, B. Z., Zhao, Y. S., Zhao, L. N., Liao, H., and Yu, M. A. (2009) Clin. Biochem. 42, 926–928. [36] Liao, F., Zhao, L. N., Zhao, Y. S., Tao, J., and Zuo, Y. P. (2007) Anal. Sci. 23, 439–444. [37] Bradford, M. M. (1976) Anal. Biochem. 72, 248–254. [38] Tan, Q. Y., Wang, N., Yang, H., Zhang, L. K., Liu, S., Chen, L., Liu, J., Zhang, L., Hu, N. N., Zhao, C. J., and Zhang, J. Q. (2010) Int. J. Pharm. 384,165–72. [39] Tan, Q., Zhang, J., Wang, N., Li, X., Xiong, H., Teng, Y., He, D., Wu, J., Zhao, C., Yin, H., and Zhang, L. (2012) Eur. J. Pharm. Biopharm. 82, 43–48. [40] Tan, Q. Y., Zhang, J. Q., Wang, N., Yang, H., Li, X., Xiong, H. R., Wu, J. Y., Zhao, C. J., Wang, H., and Yin, H. F. (2012) Int. J. Nanomed. 7, 3929–3938.
PEGylation of a Uricase with Active Esters of mPEG
Copyright of Biotechnology & Applied Biochemistry is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
