Anal. Chem. 2007, 79, 9420-9426
Kinetic Stability Studies on Yttrium(III)-1,4,7,10Tetraazacyclododecane-1,4,7,10-tetraacetic Acid by Free-Ion Selective Radiotracer Extraction Denis Jurkin,*,† Franz Josef Gildehaus,‡ and Birgit Wierczinski†
Institut fu¨r Radiochemie, Technische Universita¨t Mu¨nchen, Walther-Meissner-Strasse 3, 85748 Garching, Germany, and Klinik und Poliklinik fu¨r Nuklearmedizin, Ludwig-Maximilians-Universita¨t Mu¨nchen, Marchioninistrasse 15, 81377 Mu¨nchen, Germany
Free-ion selective radiotracer extraction (FISRE) using nocarrier-added 90Y has been applied to assess the dissociation kinetics of yttrium(III)-1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraaceticacid(Y-DOTA),aradiopharmaceutical precursor with a remarkable kinetic stability. In order to extend the operationally defined detection window, a complemental FISRE-based experiment has been succesfully developed. Within a time frame of approximately 10 half-times of 90Y (t1/2 ) 61.4 h), the complex Y-DOTA has been observed to form two kinetically distinguishable species with significantly different kinetic properties. The time-dependent speciation was measured to be sensitive to pH variations even in the neutral pH range (4.5-7.4) whereas the impact of ionic strength changes is negligible. In general, target-specific metalloradiopharmaceuticals are based on bifunctional chelators (BFC) binding the radioactive metal ion as well as metabolizable linkers that are themselves covalently bound to the targeting biomolecule.1-3 In order to ensure an effective transport of the therapeutic dose of radiation to the target tissue, the formation of a thermodynamically and kinetically stable metal-BFC complex under physiological pH is indispensable. This work focuses on the analysis of the kinetic stability of the complex of yttrium with 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid (DOTA). DOTA-modified peptides like [Tyr3]-octreotide and labeled with 90Y or 177Lu are of great clinical interest and are widespread used radiopeptides in the treatment of neuroendocrine tumors, for example, (90Y-DOTATOC).4-6 * To whom correspondence should be addressed. E-mail: jur@ rad.chemie.tu-muenchen.de. Telephone: +49-89-28912206. Fax: +49-89-28912204. † Technische Universita¨t Mu ¨ nchen. ‡ Ludwig-Maximilians-Universita ¨t Mu ¨ nchen. (1) Liu, S.; Edwards, D. S. Bioconjugate Chem. 2001, 12, 7-34. (2) Reichert, D. E.; Lewis, J. S.; Anderson, C. J. Coord. Chem. Rev. 1999, 184, 3-66. (3) Volkert, W. A.; Hoffman, T. J. Chem. Rev. 1999, 99, 2269-2292. (4) Bodei, L.; Cremonesi, M.; Grana, C.; Rocca, P.; Bartolomei, M.; Chinol, M.; Paganelli. G. E. J. Nucl. Med. Mol. Immunol. 2004, 31 (7), 1038. (5) Otte, A.; Jermann, E.; Behe, M. Eur. J. Nucl. Med. 1997, 24, 792. (6) Breeman, W. A.; de Jong, M.; Kwekkeboom, D. J. Eur. J. Nucl. Med. 2001, 28, 1421.
9420 Analytical Chemistry, Vol. 79, No. 24, December 15, 2007
The thermodynamic stability of yttrium chelates is usually well-studied; quantitative information on kinetic properties, especially at physiological pH, is, however, rare.7-9 Earlier misleadingly referred to as the “stability” constant, the thermodynamic equilibrium constant K alone fails to predict the time scale a complex remains intact in vivo (with the exception of complexes with a Eigen mechanism-based kinetic behavior).10,11 Therefore, appropriate analytical techniques are required to assess the kinetic properties. For this purpose, the free-ion selective radiotracer extraction (FISRE), a method combining common metal ion extractions with the use of radiotracers renouncing competing extrinsic metals or ligands, proved to be suitable in the case of copper and cobalt polyaminocarboxylates, predominantly complexes of environmental interest.12-14 The application of FISRE on metal chelates for medical applications is rather new. Despite the evidently short acquisition time frame of 120 s, first studies with Ho-DTPA, HoDTPA-bispropylamide, Lu-DOTA, and Lu-DOTATOC were promising.14,15 In order to quantify dissociation behavior of kinetically stable complexes as 90Y-DOTA properly, the considered time frame has to be extended significantly up to hours, days, and even weeks. THEORY Thermodynamic Stability. Given a simple (pseudo-) firstorder complex formation reaction of an metal complex (M-L)m+n, where Ln represents the ligand with charge n, according to eq 1, (7) Clarke, E. T.; Martell, A. E. Inorg. Chim. Acta 1991, 190, 37. (8) Martell, A. E.; Smith, R. M. Critical Stability Constants; Plenum Press; New York, 1974. (9) Martell, A. E.; Smith, R. M. NIST Critically Selected Stability Constants of Metal Complexes Database; NIST Standard Reference Database 46; Version 4.0; NIST: Gaithersburg, MD 1997. (10) Ugur, O.; Kostakoglu, L.; Hui, E. T.; Fisher, D. R.; Garmestani, K.; Gansow, O. A.; et al. Nucl. Med. Biol. 1996, 23, 1. (11) Eigen, M. Pure Appl. Chem. 1963, 6, 97. (12) van Doornmalen, J.; Wolterbeek, H. Th.; de Goeij, J. J. M. Anal. Chim. Acta 2002, 464, 141. (13) van Doornmalen, J. Ph.D. Thesis, Delft University of Technology, 2002. (14) Wierczinski, B.; Denkova, A. G.; Peters, J. A.; Wolterbeek, H. Th. In Application of Radiotracers in Chemical, Environmental and Biological Sciences; Lahiri, S., Nayak, D., Mukhopadhyay, A., Eds.; Saha Institute of Nuclear Physics: Kolkata, India, 112; 2006; Vol. 1, p 112. (15) Denkova, A. G. Diploma thesis, Delft University of Technology, 2005. 10.1021/ac701786w CCC: $37.00
© 2007 American Chemical Society Published on Web 11/15/2007
K)
[(M - L)m+n] m
n
)
[M ][L ]
ka kd
(1)
the thermodynamic equilibrium constant under steady-state conditions K is defined as the ratio of equilibrium activities of the metal complex and the product of its components kinetic rate constants ka and kd (s-1). In theory, activities are used instead of concentrations (or radioactivities). Since, however, the ionic strengths of all examined solutions are constant and only relative values are considered, activity and radioactivity can be considered equal. Dissociation Kinetics. The dissociation rate constant kd (s-1) represents the primary parameter to evaluate the kinetic stability, viz. the ability of a complex to remain intact within an experimental time scale. Considering a (pseudo-) first-order reaction, the dissociation is determined by its activity [(M-L)m+n] as well as kd:
d[(M - L)m+n] ) kd[(M - L)m+n] dt
-
(2)
The solution of the differential equation is following first-order exponential function
[(M - L)m+n]t ) [(M - L)m+n]0e-kdt
(3)
with t as the contact time of the complex with the extracting agent, [(M-L)m+n]0 the initial and [(M-L)m+n]t the complex activity at a given contact time. Plotting the logarithm of the relative activity [(M-L)m+n]t/ [(M-L)m+n]0 against t therefore reveals the dissociation rate constant. If two or more kinetically distinguishable complex species formed by parallel reactions are present, the function has to be augmented with i terms that account for the contribution of i species to the dissociation characteristics:
[(M - L)m+n]t )
∑ [(M - L)
m+n i
]0ekdit
(4)
i
Assuming a consecutive dissociation reaction of two complex species including a rate-determining step, as shown in eq 5, +n
(M - L)Am
kD(A)
+n
98 (M - L)Bm
kD(B)
98 Mm + Ln
(5)
a term has to be added that accounts for the formation of (ML)m+nB from (M-L)m+nA: -kd(A)t [(M - L)m+n]t ) [(M - L)m+n + A ]0e -kd(B)t [(M - L)m+n + B ]0e
kd(A)[(M - L)m+n A ]0 ‚(e-kd(A)t kd(B) - kd(A) e-kd(B)t) (6)
The FISRE technique generally involves the charge-dependent extraction of free ions, in this case, the extraction of free yttrium
ions by a Chelex cation exchanger as extracting agent. Two different modes can be applied to assess the dissociation kinetics. The continuous FISRE mode is based on varying the contact time of metal complex and extracting agent by adjusting the sample flow rate through the column. A buffered aqueous solution of free inactive Y3+ ionssthe mobile phasesis delivered to the exchange column by a HPLC pump. Besides, a sample consisting of a carefully equilibrated mixture of inactive Y3+ ions, the corresponding ligand, and high-specific activity 90Y3+ is introduced into the system by a loop injection. In order to prevent significant amounts of free ligand ions, inactive Y3+ is here applied in molar excess in relation to the ligand. Once a metal complex dissociates during the contact time with the Chelex column, the free metal ion is retarded on the column. Since nonradioactive Y3+ ions are present in large molar excess in comparison with the radiotracer as well, reassociation of dissociated 90Y-DOTA species is statistically disfavored. The retention of the 90Y3+ on the Chelex column as a function of contact time (flow rate) consequently allows a quantification of dissociated species. In the batch dissociation mode, the ligand is equilibrated with 90Y3+ first, and then nonradioactive Y3+ is added followed by Chelex extractions at different time intervals after Y3+ addition at a constant mobile-phase flow rate. Hence, long-term dissociation can be observed, while the time frame is primarily limited by the half-time of the applied radiotracer. By using equal final concentrations of metal, ligand, NaCl, and buffer, the continuous and batch method can be adjusted to complement each other. EXPERIMENTAL SECTION Materials. All chemicals were of analytical reagent grade. Milli-Q-Plus (MQ) water (Millipore, Schwalbach, Germany) was used for solution preparation. Yttrium(III)-chloride hexahydrate (99,99%), β-morpholinoethanesulfonic acid monohydrate (MES), 4-(2-hydroxyethyl)piperazine-1-ethanesulfonic acid (HEPES, g99,5%), piperazine (g99%), 1-methylpiperazine (g99,5%), nitric acid (65%), hydrochloric acid (37%), and sulfuric acid (96%) were purchased from Sigma-Aldrich (Munich, Germany). DOTA, sodium chloride, and sodium hydroxide were obtained from Fluka Chemie (Buchs, Switzerland). Ion extracting agents Chelex-100 (200-400 mesh) and Dowex 50 WX 8 as well as the Poly-Prep columns were purchased from BioRad Laboratories (Munich, Germany). ITLC-SG strips were obtained from Pall (Dreieich, Germany). Radiotracer Production. 90Y3+ (no carrier added) was obtained from a 90Sr/90Y generator in form of an R-hydroxyisobutyrate complex. Ion-exchange chromatography was applied in order to remove Sr traces as well as to transform yttrium-Rhydroxyisobutyrate into yttrium chloride. For this purpose, an unfilled Poly-Prep column was rinsed 30 min with 4 M HCl; 1 g of Dowex 50 WX 8 was suspended in 3 mL of 6 M HCl solution and poured into the column. The column material was then purified by subsequent washing with H2O, 0.1, 3, and 6 M HCl, and H2O (15 mL each). After addition of 0.5 mL of HCl (0.1 M) to 10 µL of 90yttrium(III)-R-hydroxyisobutyrate (18.5-30 MBq), the radioactive complex was added onto the column and washed with 50 mL of 0.1 M HCl and 7 mL of 3 M HCl. With 6 mL of 6 M HCl, the yttrium-90 was finally eluted and evaporated to dryness at 120 °C in a nitrogen stream. In order to remove eventual traces of R-hydroxyisobutyric acid, the residue was dissolved in 0.5 mL of Analytical Chemistry, Vol. 79, No. 24, December 15, 2007
9421
Figure 1. FISRE dissociation profiles of Y-DOTA in dependence of pH at I ) 0.01 and the corresponding fitted curve functions. The error bars indicate the standard deviation of three measurements.
a 1:1 mixture of concentrated H2SO4 and HNO3 and evaporated at 310 °C in a nitrogen stream. The residue was dissolved in 0.4 mL of 0.05 M HCl, and the radioactivity was measured with an Isomed 501 Activimeter (MED, Nuklearmedizintechnik Dresden GmbH). The yield of 90Y per sample ranged from 81 to 99%, which equals specific activitiy concentrations of 37-74 MBq/mL of the samples after purification. Radiochemical purity was tested via instant thin-layer chromatography (ITLC) in 0.1 M NaHCO3 solution and exceeded 96% in each sample.16 Sample Preparation. Prior to the preparation of the solution, the yttrium speciation in thermodynamic equilibrium was calculated using the software CHEAQSPro in order to avoid precipitates in the mobile phase and the samples under study.17 Thermodynamic equilibrium constants not present in the database were adopted from the NIST database 46.9 Stock solutions were prepared by dissolving NaCl and buffer (MES, HEPES, piperazine, N-methylpiperazine) in Millipore water. The pH was adjusted by a 10 M NaOH solution to the values given in Figure 1. In order to prepare the mobile phase, yttrium(III) chloride was dissolved in stock solution, the ligand solution by dissolving DOTA in stock solution (pH checked). The final samples are prepared by mixing ligand solution, mobile-phase, and stock solution and spiking the mixture with 90YCl3. The final total concentrations of yttrium, DOTA, and buffer were 1 × 10-6, 7.5 × 10-7, and 5 × 10-3 mol‚L-1. Each sample was heated mildly at 60 °C for 18 h and equilibrated at least 2 h at room temperature prior to the start of the measurements. (16) Malja, S.; Schomacker, K.; Malja, E. J. Radioanal. Nucl. Chem. 2000, 245 (2), 403. (17) Verweij, W. CHEAQSPro: A Program for Calculating Chemical Equilibria in Aquatic Systems; Version 2006.1; RIVM: Bilthoven, 2006.
9422 Analytical Chemistry, Vol. 79, No. 24, December 15, 2007
System Setup. Custom Chelex-100 columns were prepared by filling an empty PEEK guard column cartridge (4.6 mm i.d., Upchurch Scientific, Oak Harbor, WA) with yttrium-saturated Chelex-100. A Class-LC10Ai HPLC pump (Shimadzu Corp., Kyoto, Japan) was used for solvent delivery. All system parts (pump heads, tubing, injection valve, etc.) were constructed of inert materials (PEEK, Tefzel). Prior to the measurements, each column was rinsed with the corresponding mobile phase at a constant flow rate of 0.1 mL‚min-1 for 1 h. The dissociation kinetics of the yttrium chelates was analyzed in two ways. For the analysis of the short-term dissociation reactions (continuous mode), the radiotracer was first equilibrated with all yttrium species present in the sample followed by a series of extractions at different flow rates. In contrast, in batch mode, the extractions started immediately after addition of the inactive yttrium solution to the equilibrated tracer-ligand mixture. The extractions were performed at a constant flow rate of 1 mL‚min-1 at different time intervals. For all extractions, an aliquot of 20 µL was withdrawn from the sample and introduced in the system. The eluents with a total volume of 1 mL were collected, and their Cˇ erenkov radiation was measured with a Perkin-Elmer Tri-Carb 2800 TR LSC-counter. Each measurement was conducted three times to prove the reproducibility. For every test series, the procedure was repeated in the absence of the column to determine the total radioactivity per sample. All measurements were performed at room temperature (20 °C). Data Processing. Radioactivity of each eluate was expressed as a fraction of the total radioactivity present in a 20-µL aliquot introduced to the system. This fraction was plotted against either contact time with the extracting agent or time elapsed after
Table 1. Kinetic Parameters Determined by the Continuous FISRE Mode at Different pH Values and Constant Ionic Strength (I ) 0.01)a pH
cY-DOTA(A) (10-7 mol/L)
cY-DOTA(A) (%)
kobs(A) (10-4 s-1)
cY-DOTA(B) (10-7 mol/L)
cY-DOTA(B) (%)
kobs(B) (10-1 s-1)
7.4 7.0 6.8 6.5 6.25 6.0 5.75 5.5 5.25 5.0 4.75 4.5
7.53 ( 0.03 7.40 ( 0.03 6.92 ( 0.06 6.32 ( 0.13 5.93 ( 0.13 5.53 ( 0.10 5.46 ( 0.02 5.04 ( 0.04 5.07 ( 0.01 4.70 ( 0.07 4.70 ( 0.03 4.60 ( 0.02
100.4 ( 0.4 98.7 ( 0.4 92.3 ( 0.8 84.3 ( 1.7 79.1 ( 1.7 73.7 ( 1.3 72.8 ( 0.3 67.2 ( 0.5 67.6 ( 0.1 62.7 ( 0.9 62.7 ( 0.4 61.3 ( 0.3