342
Bioconlugate Chem. 1992, 3, 342-345
Spectrophotometric Method for the Determination of a Bifunctional DTPA Ligand in DTPA-Monoclonal Antibody Conjugates C. Greg Pippin,' Tammy A. Parker,+ Thomas J. McMurry, and Martin W. Brechbiel Chemistry Section, Radiation Oncology Branch, National Cancer Institute, National Institutes of Health, Building 10, Room B3-B69, Bethesda, Maryland 20892. Received January 13,1992 A simple spectrophotometric method was developed to quantitate micromolar concentrations of a bifunctional DTPA ligand in DTPA monoclonal antibody (mAb) conjugates. Titration of a brightly colored 1:2 yttrium(II1) complex of arsenazo I11 with the ligand 1B4M-DTPA obeyed Beer's law over the concentration range 0-2.0 pM 1B4M-DTPA a t 652 nm. From a calibration plot of absorbance versus 1B4M molarity, concentrations of 1B4M-DTPA conjugated to mAb were determined. Mole ratios of 1B4M-DTPA tomAb agreed satisfactorilywiththe ratios obtained by a radioanalytical technique using carbon-14-labeled 1B4M-DTPA and a binding assay using "'In. The spectrophotometric method was applied successfully to the preparation of 1B4M-DTPA mAb anti-TAC, a mAb conjugate used in clinical trials of W Y radioimmunotherapy.
INTRODUCTION Radiolabeled monoclonal antibodies (mAbs) are being developed and evaluated for the diagnosis and treatment of cancer in many clinical laboratories (1-4). Efforts to develop practical radiolabeled mAbs usually involve the conjugation of a bifunctional ligand to the mAb, followed by labeling the ligand-mAb conjugate with a radioactive metal ion (5). In order to maintain optimal and reproducible behaviors of radiolabeled mAbs in clinical investigations, it is necessary to establish the relationship between targeting efficacy of the mAb and the number of ligands per mAb molecule. It has been amply documented that the targeting efficacyand catabolism of mAbs should not be altered by conjugation or radiolabeling procedures (for review see refs 1 and 2). Several analytical methods can quantitate the number of strong metal binding sites in ligand-mAb conjugates. For the amino polycarboxylate class of ligands (e.g. bifunctional EDTA,' DTPA, etc.), these methods include lllIn (6, 7) and 57C0(8) binding assays, T b fluorescence titrations (9),and use of 14C-labeledligands (10). In this laboratory, mAb conjugates with 14Clabeled ligands were prepared for preclinical studies, but these conjugates have been restricted from use in patients because of the 14C label. This restriction and our desire to use rapid, simple procedures to quantitate bifunctional ligands led us to develop a spectrophotometric method to quantitate a bifunctional DTPA ligand currently used in clinical trials of radioimmunotherapy. The spectrophotometric method is based on the reaction between a 1B4M-DTPA ligand (11)mAb conjugate and a yttrium(II1) complex of araenazo I11(see Chart I). Arsenazo I11is a highly sensitive colorimetricreagent for yttrium, the lanthanides, and other metal ions (12, 13). Because of ita sensitivity, several authors (14-16) have studied coordination complexes with arsenazo I11 even at micromolar concentrations. In this study our prime objectives were (a) to determine whether the spectrophotometric method used in this laboratory would be quantitative for the 1B4M-DTPA ligand and, if so, (b) to compare the results of the method with previously documented analytical techniques for ligand quantitation, especially the 14C radioanalytical
* Author to whom correspondence should be addressed.
Student Intern at NIH during Summer 1991 from San Jose State University. +
Not subJectto US. Copyright.
Chart I
C02H
&02H C 0 2 H
lB4M-DTPA
H03S
S03H
Arsenazo I11
technique (10). Because the 14Cradioanalytical technique measures the total binding sites and metal ion titrations measure the available binding sites, a secondary objective was to measure these two distinct physical properties of 1B4M-DTPA-mAb conjugates. EXPERIMENTAL PROCEDURES Reagents. A stock solution of 0.100 M Y(II1) in 10-3 M HCl was prepared from YC13-6H20 (Aldrich, 99.9%) and standardized by titration with EDTA using xylenol orange as indicator. Arsenazo I11 (ca. 98%, < 0.01 pmol Ca2+/mg) was purchased from Sigma Chemical Co. Caution: Arsenazo 111is a possible carcinogen! Avoid skin contact and inhalation of the solid material. All mAb conjugate stocks were prepared as described earlier (17)in concentrations that ranged from 9.6 to 45.4 mgl 'Abbreviations used: DTPA, diethylenetriaminepentaacetate; lB4M-DTPA, 2-methyl-6-@-nitrobenzyl)diethylenetriamineN,N,N'~JV"-pentaacetate;EDTA, ethylenediaminetetraacetate; arsenazo 111, 3,6-bis[(2-arsenophenyl)azol-4,5-dihydroxy2,7-naphthalenedisulfonicacid.
Published 1992 by Amerlcan Chemical Soclety
Technical Notes
mL. Each conjugate was stored at 4 "C in solutions of 0.020 mM MES buffer (CALBIOCHEM, passed through a column of Chelex-100 resin, Na+ form), 0.15 M NaC1, 0.05% NaN3, pH 6.2. Weighed quantities of the ligand lB4M-DTPA, available from previous studies (111,were dissolved in deionized water with 3 equiv of NaOH. Indium(II1) nitrate (Specpure, Johnson Matthey) was obtained as a lo00 pglmL atomic absorption standard. The radionuclide lllIn chloride was obtained from New England Nuclear. All solutions were prepared with highpurity deionized water (Hydroservices Picosystem, >18 SI cm). Procedures. A 500-mL stock solution of the Y(II1)arsenazo I11 complex (hereafter designated as Y(AA1II)z) was prepared with the following composition: 5.0 pM AAIII, 1.6 pM Y(III), 0.15 M sodium acetate buffer, pH 4.00. Among the factors which led to the selection of this composition, the three prime factors were as follows: (1) the maximum absorbance of Y(AA1II)z at 652 nm, 0.111, was sufficient for quantitative measurements, (2) Beer's law was obeyed when Y(II1) was equal to or less than 1.6 pM in this medium, and (3) the 0.15 M acetate buffer adequately maintained pH 4.00 when mAb conjugates in 0.02 M MES buffer were added to Y(AA1II)z. To determine the mole ratio of 1B4M to mAb, 3.00 mL of Y (AAIII)2 was added to a cuvette and the absorbance measured at 652 nm in a diode-array spectrophotometer (Hewlett-Packard8450A). The 0.15 M acetate buffer, pH 4.00, served as the reference solution for all measurements. From four to six 10 pL additions of 0.123 mM 1B4MDTPA ligand were added serially to the cuvette. Absorbance values were recorded after each addition of ligand and corrected for volumetric dilution, e.g. corrected back to the initial volume of 3.00 mL. Data from three independently prepared stock solutions of 1B4M-DTPA and Y(AAI1I)z were used to construct a calibration plot of versus [1B4M-DTPAIm A similar procedure was followed for a 14C-labeled 1B4M-DTPA mAb anti-TAC conjugate, i.e. four 20-pL aliquots were added to 3.00 mL of Y(AA1II)z. However, in this experiment, a minimum of 10 min elapsed between addition of the mAb conjugate and measurement of the absorbance of Y (AA1II)z in order to ensure equilibrium. From the previously constructed calibration plot, the 1B4M-DTPA molarity was calculated for each of the four 20-pL additions of mAb conjugate. After correction for mAb dilution, the mole ratios of 1B4M-DTPA to mAb were calculated as the quotient [lB4M-DTPAI/ [mAbl. These experiments were done in triplicate. One mAb anti-TAC conjugate served as a secondary standard for subsequent determinations of the mole ratios of 1B4M-DTPA to mAb. Specifically, determination of [lB4M]/[mAb anti-TACI before a new measurement ensured reproducibility2 and also obviated additional calibrations with the 1B4M-DTPA ligand. Beer's law was observed for each mAb conjugate determination (see sample data in Figure 2 inset). In addition, few metal ion contaminants or strong metal binding sites were found in the mAb anti-TAC or IgG stocks. For example, when 12.0 pM concentrations of mAb anti-TAC or IgG were maintained in Y (AAIII)z for 10min, the absorbance of Y (AAIIIh 2The stock solutions of Y(AAIII)2 and arsenazo I11 showed reproduciblebehavior over a 5 month period when their containers were kept in the dark. By contrast, very dilute concentrations (