Bioconlugate Chem. 1994, 5, 348-351
a48
A Sensitive Assay for Maleimide Groups Rajeeva Singh ImmunoGen, Inc., 148 Sidney St., Cambridge, Massachusetts 02139. Received November 24, 1993"
A sensitive spectrophotometric assay has been developed for maleimide groups incorporated into proteins. The assay involves the reaction of maleimide with an excess of cysteine and quantitation of the remaining cysteine using an enzymatic assay that is based on the stoichiometric conversion of an inactive disulfide form of papain to the active enzyme. The restored enzymatic activity is then measured using a chromogenic substrate. The amount of maleimide is calculated as the difference between the amounts of initial cysteine and assayed remaining cysteine. The limits of detection of this assay are about 0.1 nmol maleimide in a final assay volume of 1.2 mL. This assay is about 100-fold more sensitive than a similar assay using Ellman's reagent.
INTRODUCTION Maleimide groups rapidly and selectively react with thiols, forming stable thioether bonds. This characteristic has led to their widespread use in the formation of stable protein conjugates (1,2). Maleimide groups are typically incorporated in proteins by random modification of their lysine amino groups. Sensitive assays of maleimide and thiol groups are required for the efficient conjugation of proteins that are expensive and available only in small amounts. Maleimides can be directly assayed spectrophotometrically at 302 nm, having an extinction coefficient E302nm of 620 M-' cm-l (3). However, the small €3&'nm renders this assay not very sensitive, and the assay is further complicated by the protein absorbance at the same wavelength. Indirectly, maleimide groups can be assayed by first reacting them with a known amount of thiol present in excess and then assaying the remaining unreacted thiol using Ellman's reagent (2,4,5). The amount of maleimide is calculated as the difference between the initial amount of thiol and the amount of unreacted thiol after complete reaction of all maleimide groups. The sensitivity of this indirect assay is limited by that of the Ellman's assay ( ~ 1 = 14 150 M-' cm-l) (6, 7);this assay reaches its limit of accurate detection for large proteins (e.g., antibodies, M, 160 000) modified with an average of one maleimide group, requiring concentrations of greater than 1 mg/mL to measure a change in the absorbance of 0.1 unit (in a cuvet of 1-cm pathlength). We have previously described an enzymatic assay for thiol groups that is based on the stoichiometric activation of papain-SSCHa and that is about 100-foldmore sensitive than Ellman's assay (8). In this paper we extend the use of this assay to the measurement of maleimide groups incorporated into proteins. EXPERIMENTAL PROCEDURES Materials. Papain (EC 3.4.22.2; 2 X crystallized suspension in 50 mM sodium acetate, pH 4.5, containing 0.01% thymol), avidin (egg white), 4-(N-maleimidomethy1)cyclohexane-1-carboxylicacid N-hydroxysuccinimide ester (SMCC), N;ethylmaleimide, and N-benzoyl-L-arginine-p-nitroanilide, hydrochloride were purchased from Sigma (St.Louis, MO). The murine monoclonal antibody anti-B4 (IgG1, M,160 000;A0.l%1,, = 1.4 a t 280 nm) was purified from hybridoma culture supernatants by affinity ~~
@
Abstract published in Advance ACS Abstracts, May 1,1994. 1o
2
chromatography on protein A followed by ion-exchange chromatography. All other reagents were of commercial reagent grade. The compositions of buffers frequently used were as follows: buffer A (10 mM potassium phosphate, 150 mM NaC1,l mM EDTA, pH 7.2); buffer B (5 mM sodium acetate, 50 mM NaC1,0.5 mM EDTA, pH 4.7); buffer C (50mM sodium phosphate, 1mM EDTA, pH 7.0). Papain-SSCH3 was prepared as described before (8). Commercial papain contains about 90 % inactive mixed disulfide papain-SS-cysteine (8). Briefly, homogeneous samples of papain-SSCH3 were prepared by first reducing commercial papain with cysteine and then treating with methyl methanethiosulfonate. Such preparations of papain-SSCH3 were used in the experiments described below and in Figure 1. Other experiments were performed with mixtures of papain-SS-cysteine and papain-SSCH3 that had been obtained by reaction of commercial papain with methyl methanethiosulfonate. Although the results were found to be similar using either preparation, we recommend usage of the homogeneous preparation of papain-SSCH3 (8). ~ Modification of Proteins with Maleimide Groups. A 0.6 mM solution of SMCC in ethanoVwater ( l / l , v/v) was freshly prepared by first dissolving SMCC in ethanol using sonication and then adding water. The concentration of maleimide in the solution was determined by measuring its absorbance at 302 nm (€302 = 620 M-l cm-1) (3). A 1.5 molar excess of SMCC (135 p L of the above solution, 81 nmol) was added to a prewarmed antibody solution 1715pL of an antibody solution in 50 mM sodium phosphate, 50 mM NaC1, pH 7; 8.5 mg IgG1,53 nmol] at 30 "C (9). After incubation at 30 OC for 30min, the reaction mixture was purified by gel filtration using a Sephadex G-25 (fine) column equilibrated with buffer A. The concentration of pooled antibody fraction was determined from its absorbance at 280 nm. This modification resulted in an average incorporation of 0.9 maleimide group per antibody molecule (Table 1 1. Assay for Maleimide Groups Using Cysteine and Papain-SSCH3. All solutions were degassed as described before (8). Typically, protein samples containing maleimide in the range 0.7-0.3 nmol (based on estimate from Ellman's assay) were incubated with a 2- to 5-fold molar excess of cysteine (Table 1). In a test sample the maximum theoretical amount of maleimide can also be estimated from the molar excess of SMCC used during modification. In the experiment described below, for the
~ ~ - i a ~ ~ ~ ~ ~ ~ ~0 ~1994 o American ~ - o ~Chemical ~ a ~Society o ~ . ~ o ~ o
Bioconjugate Chem., Vol. 5, No. 4, 1994
Sensitive Assay for Maleimide Groups
smallest measurement of maleimide in this study, the test samples (in quadruplicate) contained 10 pL of the sample (1.66 mg/mL maleimide-modified antibody in buffer A; 17 pg or 0.1 nmol antibody, -0.1 nmol maleimide) and an excess of cysteine (9 pL of a 0.097 mM solution in buffer B; 0.87 nmol). Separate tubes containing standard samples of cysteine (0.29,0.58,0.87,and 1.17 nmol) were prepared by addition of cysteine [3,6,9 (in two tubes, equal to the initial amount of cysteine in test sample), and 1 2 pL, respectively, of a 0.097 mM solution of cysteine in buffer B] followed by addition of 10 pL of buffer A. Blanks (in triplicate) contained only 10pL of buffer A and no cysteine. To all tubes was then added 240 pL of 40 mM sodium phosphate, 2 mM EDTA, pH 7.6. Buffer B was then added to bring the total volume in each tube to 262 pL. The tubes were then incubated for about 40 min. The value of the pH during the initial incubation of maleimide sample with excess cysteine was -7.6. Papain-SSCH3 (0.25 mL of a 1.18 mg/mL solution in 5 mM sodium acetate, 50 mMNaC1, pH 4.5; 0.3 mg papainSSCH3,12.6nmol) (8)was then added, and the tubes were further incubated for about 40 min (pH -7.6). Substrate (0.7 mL of a 3.4 mM N-benzoyl-L-arginine-p-nitroanilide solution in 50 mM bis-Tris-HC1, 1 mM EDTA, 5% v/v DMSO, pH 6.3) was then added at 1-min intervals between successive tubes. The value of pH during the incubation with substrate was -6.3. The absorbance values (410nm) of all samples were measured at about 1h after the addition of substrate, at the same 1min intervals between successive tubes. The value of absorbance for the blank was subtracted from those for the cysteine standards and for the test samples. The AAllOnm values for the cysteine standard solutions were then plotted us the amounts of cysteine initially added. The remaining cysteine in the test samples was then determined using this standard curve. The amount of maleimide in the test samples was calculated as the difference between the amount of initially added cysteine and that of the remaining cysteine (Figure 1). The final substrate concentration was 2 mM; the final volume was 1.2 mL. All volume measurements of
