902
Bioconjugate Chem. 2002, 13, 902−913
Comparison of Yttrium and Indium Complexes of DOTA-BA and DOTA-MBA: Models for 90Y- and 111In-Labeled DOTA-Biomolecule Conjugates Shuang Liu,* John Pietryka, Charles E. Ellars, and D. Scott Edwards Bristol-Myers Squibb Medical Imaging,† 331 Treble Cove Road, North Billerica, Massachusetts 01862. Received October 8, 2001; Revised Manuscript Received January 18, 2002
Yttrium and indium complexes of 1,4,7,10-tetraaza-4,7,10-tris(carboxymethyl)-1-cyclododecylacetylbenzylamine (DOTA-BA) and 1,4,7,10-tetraaza-4,7,10-tris(carboxymethyl)-1-cyclododecylacetyl-R-(+)R-methylbenzylamine (DOTA-MBA) were prepared in order to study solution structures of 90Y- and 111In-labeled DOTA-biomolecule conjugates. 90Y and 111In complexes M(L) (M ) 90Y and 111In; L ) DOTA-BA and DOTA-MBA) were prepared from the reaction of MCl3 with DOTA-BA and DOTAMBA, respectively, in ammonium acetate buffer. A reverse phase HPLC method revealed that both 90 Y and 111In complexes show only one radiometric peak in their radio-HPLC chromatograms. It was also found that 111In(DOTA-BA) and 111In(DOTA-MBA) are more hydrophilic than their corresponding 90Y analogues, suggesting different coordination spheres in 111In and 90Y complexes of the same DOTA conjugate. Complexes M(L) (M ) Y and In; L ) DOTA-BA and DOTA-MBA) were prepared and characterized by HPLC, LC-MS, and NMR (1H and 13C) methods. The HPLC concordance experiments for 90Y(DOTA-MBA)/Y(DOTA-MBA) and 111In(DOTA-MBA)/In(DOTA-MBA) show that the same complex is prepared at both tracer and macroscopic levels. The NMR data (1H and 13C) clearly demonstrates that Y(DOTA-BA) and Y(DOTA-MBA) exist in solution as one predominant isomer. VT NMR data (1H and 13C) show that In(DOTA-BA) and In(DOTA-MBA) are fluxional at room temperature while Y(DOTA-BA) and Y(DOTA-MBA) become fluxional only at elevated temperatures. The fluxionality of these complexes is due to rapid rotation of acetate/acetamide chelating arms and inversion of ethylenic groups of the macrocyclic ring.
INTRODUCTION
There is great current interest in the development of target-specific therapeutic radiopharmaceuticals for the treatment of cancers. Several reviews have appeared recently covering a broad range of topics related to radiolabeled small biomolecules (BM) as target-specific therapeutic radiopharmaceuticals (1-7). Many acyclic and macrocyclic bifunctional chelators (BFCs) have been used for the radiolabeling of biomolecules, including antibodies and small peptides (2, 6, 8-17). While the 90Ylabeled BFC-BM conjugate is used for tumor therapy, the 111In-labeled BFC-BM conjugate is often used as an imageable surrogate for dosimetry determination and therapy monitoring (17, 18-27). The advantage of using 111 In as the imaging surrogate for 90Y is that 111InCl3 is commercially available and its half-life is almost identical to that of 90Y. However, recent studies show biodistribution differences between 90Y- and 111In-labeled antibodies and small peptides (17, 28-30). The difference in biological properties raises some questions about structural differences between 90Y and 111In DOTA chelates. To study solution structures of 90Y- and 111In-labeled DOTA-BM conjugates, we prepared yttrium and indium complexes (Figure 1) of two model compounds: 1,4,7,10tetraaza-4,7,10-tris(carboxymethyl)-1-cyclododecylacetylbenzylamine (DOTA-BA) and 1,4,7,10-tetraaza-4,7,10tris(carboxymethyl)-1-cyclododecylacetyl-R-(+)-R-methyl* To whom correspondence should be addressed. Tel: 978671-8696 (S.L.); FAX: 978-436-7500. E-mail: shuang.liu@ bms.com. † Formerly DuPont Pharmaceuticals Company.
Figure 1. Structures for yttrium and indium complexes of DOTA-BA and DOTA-MBA.
benzylamine (DOTA-MBA). In this report, we present the synthesis and characterization of yttrium and indium complexes of DOTA-BA and DOTA-MBA at both tracer and macroscopic levels. The focus of this study is to compare yttrium complexes of DOTA-BA and DOTAMBS with their corresponding indium analogues with respect to their structures in solution, hydrophilicity, and solution dynamics. EXPERIMENTAL SECTION
Materials. Acetic acid (ultrapure), ammonium hydroxide (ultrapure), benzylamine, bromoacetyl bromide, chloroacetyl chloride, R-(+)-R-methylbenzylamine, and sodium gentisate were purchased from either Aldrich or
10.1021/bc010134h CCC: $22.00 © 2002 American Chemical Society Published on Web 06/27/2002
Bioconjugate Chem., Vol. 13, No. 4, 2002 903
Sigma Chemical Co. and were used as received. The tributyl ester of 1,4,7,10-tetraazacyclododecane-1,4,7triacetic acid (DO3A(OBu-t)3) was purchased from Macrocyclics Inc, Richardson, TX. 111InCl3 and 90YCl3 (in 0.05 N HCl) was purchased from NEN Life Sciences, N. Billerica, MA. NMR (1H, 13C, 1H-1H COSY, and 1H-13C HMQC) data were obtained using a Bruker DRX 600 MHz FT NMR spectrometer, and chemical shifts as δ are reported in ppm relative to TMS. The LC-MS data were collected using a HP1100 LC/MSD system with an API-electrospray interface. The LC-MS method used a Zorbax C18 column (4.6 mm × 150 mm, 3.5 µm particle size) and a mobile phase gradient starting from 100% solvent A (95% water and 5% TFA) to 100% solvent B (95% acetonitrile and 5% water) at 16 min at a flow rate of 1 mL/min. The HPLC method 1 used a HP-1050 HPLC system with a Rainin Dynamax UV/visible detector (model UVC, λ ) 230 nm) and a Zorbax C18 column (4.6 mm × 250 mm, 80 Å pore size). The flow rate was 1 mL/min with the mobile phase starting 10% of solvent A (0.1% TFA in water) and 90% solvent B (0.1% TFA in acetonitrile) to 20% solvent A and 80% of solvent B at 20 min, followed by an isocratic wash using 40% of solvent A and 60% solvent B from 21 to 26 min. The HPLC method 2 used a HP-1050 HPLC system with a Rainin Dynamax UV/visible detector (model UVC, λ ) 230 nm) and a Zorbax C18 column (4.6 mm × 250 mm, 80 Å pore size). The flow rate was 1 mL/min with the mobile phase starting 30% of solvent A (0.1% TFA in water) and 70% solvent B (0.1% TFA in acetonitrile) to 40% solvent A and 60% of solvent B at 20 min, followed by an isocratic wash using 40% of solvent A and 60% solvent B from 21 to 26 min. The HPLC method 3 used a HP-1100 HPLC system with a UV/visible detector (λ ) 230 nm), an IN-US radiodetector, and a Zorbax C18 column (4.6 mm × 250 mm, 80 Å pore size). The flow rate was 1 mL/min with a gradient mobile phase starting 95% of solvent A (0.025 M ammonium acetate buffer, pH 6.8) and 5% solvent B (acetonitrile) to 8% solvent A and 92% of solvent B at 18 min, followed by an isocratic wash using 40% of solvent A and 60% solvent B from 19 to 25 min. Synthesis of N-(Chloroacetyl)benzylamine. To solution of benzylamine (2.15 g, 0.02 mol) in chloroform (100 mL) was added chloroacetyl chloride (2.26 g, 0.02 mol) in an ice-bath, followed by slow addition of 3 mL of triethylamine. The resulting mixture was stirred at roomtemperature overnight and was then heated to reflux for 2 h. Solvent and excess triethylamine were removed under reduced pressure. The residue was dissolved in 100 mL of chloroform and washed with water (2 × 50 mL). The organic layer was separated and dried over anhydrous sodium sulfate. Removal of chloroform gave a pale yellow waxy solid. Upon addition of a mixture of water and methanol (10:1 ) water:methanol), it became a white solid. The solid was collected and dried under vacuum overnight. The yield was 2.5 g (69%). LC-MS: m/z ) 184.0 for [C9H10ClNO]+. 1H NMR (600 MHz, in CDCl3, chemical shift in ppm relative to TMS): 4.08 (s, 2H, CH2); 4.54 (d, 2H, CH2Ph); 6.97 (bs, 1H, NH); and 7.20-7.85 (m, 5H, benzene). Synthesis of 1,4,7,10-Tetraaza-4,7,10-tris(carboxymethyl)-1-cyclododecyl-acetylbenzylamine (DOTABA). To a solution of DO3A(OBu-t)3 (193 mg, 0.375 mmol) in anhydrous acetonitrile (45 mL) was added N-(chloroacetyl)benzylamine (195 mg, 1.06 mmol). The resulting mixture was stirred at room temperature for 3 days, and was then heated to reflux for 6 h. Solvents were removed
under reduced pressure to give a gummy liquid, which was dissolved in chloroform (50 mL), washed with water, and dried over anhydrous sodium sulfate. Removal of chloroform gave a gummy liquid. To the residue were added dichloromethane (15 mL) and anhydrous TFA (15 mL). The mixture was stirred at room-temperature overnight. Solvents were removed under reduced pressure. The residue was dissolved in concentrated HCl (1.0 mL). Upon addition of acetone (100 mL), a white solid was formed. The solid was collected by filtration, washed with acetone and dried in air. The crude product (150 mg) was dissolved in 1.0 mL of water and was then purified by HPLC (method 1). The fraction at 10-12.5 min was collected. The collected fractions were combined and then lyophilized. The yield was 53 mg. The retention time was 12.5 min with the purity > 98% by HPLC (method 1). LC-MS: m/z ) 494.2 for [C23H36N5O7]+. 1H NMR (600 MHz, in D2O at 40 °C): 3.36-3.46 (m, 16H, CH2, cyclen); 3.85-3.91 (m, 8H, CH2Ph); and 7.46-7.56 (m, 5H, benzene). Synthesis of N-(Bromoacetyl)-R-(+)-r-Methylbenzylamine. To solution of R-(+)-R-methylbenzylamine (2.4 g, 0.02 mol) in chloroform (150 mL) cooled in an ice-bath was added bromoacetyl bromide (4.1 g, 0.02 mol), followed by slow addition of 5 mL of triethylamine. The resulting mixture was stirred at room-temperature overnight. Solvent and excess triethylamine were removed under reduced pressure to give a white solid, which was washed with water and then a mixture of water and methanol (10:1 ) water:methanol). The solid was collected and dried under vacuum overnight. The yield was 3.5 g (∼73%). LC-MS: m/z ) 242.0 for [C10H12BrNO]+. 1H NMR (600 MHz, in CDCl3): 1.60 (d, 3H, CH3); 3.85 (q, 2H, CH2); 5.20 (q, 1H, CH); and 7.20-7.50 (m, 5H, benzene). Synthesis of 1,4,7,10-Tetraaza-4,7,10-tris(carboxymethyl)-1-cyclododecyl-acetyl-(R-(+)-r-methylbenzylamine) (DOTA-MBA). To a solution of DO3A(OBut)3 (193 mg, 0.375 mmol) in anhydrous acetonitrile (45 mL) was added N-(bromoacetyl)-R-(+)-R-methylbenzylamine (103 mg, 0.430 mmol). The resulting mixture was heated to reflux overnight. Solvents were removed under reduced pressure to give a gummy liquid, which was taken up with chloroform (50 mL), washed with water, and dried over anhydrous sodium sulfate. Removal of chloroform gave a liquid residue, to which were added dichloromethane (15 mL) and anhydrous TFA (15 mL). The mixture was stirred at room-temperature overnight. Solvents were removed under reduced pressure. The residue was dissolved in concentrated HCl (1.0 mL). Upon addition of acetone (100 mL), a white solid was formed. The solid was collected by filtration, washed with acetone, and dried in air. The crude product (185 mg) was dissolved in 1.0 mL of water and was then purified by HPLC (method 1). The fraction at 18-20 min was collected. The collected fractions were combined and then lyophilized. The yield was 65 mg. The retention time was 18.5 min with the purity > 98% by HPLC (method 1). LC-MS: m/z ) 508.3 for [C24H38N5O7]+. 1H NMR (600 MHz, in D2O at 40 °C): 1.38 (d, 3H, CH3); 2.80-3.9 (m, 24H, CH2, cyclen and acetate); 4.88 (q, H, CHPh); and 7.20-7.35 (m, 5H, benzene). Synthesis of Y(DOTA-BA). To a 5 mL vial were added DOTA-BA (40 mg, 0.081 mmol) and Y(NO3)3‚6H2O (60 mg, 0.157 mmol), followed by addition of 0.5 mL of 0.5 N ammonium citrate buffer (pH ) 4.8) and 0.5 mL of 0.5 N ammonium acetate buffer (pH ) 7.5). The mixture was heated at 100 °C for 20 min. After being cooled to room temperature, the product was separated by HPLC
904 Bioconjugate Chem., Vol. 13, No. 4, 2002
(method 2). The fraction at 10.0-13.0 min was collected. Lyophilization of collected fractions gave a white powder. The yield was 27 mg. The purity was > 98% by HPLC (method 3). LC-MS: m/z ) 580.3 for [M + H]+ ([C23H36N5O7Y]+) and 290.7 for [M + 2H]2+. Synthesis of Y(DOTA-MBA). To a 5 mL vial were added DOTA-MBA (56 mg, 0.110 mmol) and Y(NO3)3‚ 6H2O (70 mg, 0.183 mmol). To the mixture were added 0.5 mL of 0.5 N ammonium citrate buffer (pH ) 4.8) and 0.5 mL of 0.5 N ammonium acetate buffer (pH ) 7.5). The resulting solution was heated at 100 °C for 20 min. After being cooled to room temperature, the product was purified by HPLC (method 2). The fraction at 12-15 min was collected, and the collected fractions were evaporated to dryness. The residue was dissolved in water (20 mL), and the resulting solution was lyophilized to give a white powder. The yield was ∼35 mg. The purity was > 98% by HPLC (method 3). LC-MS: m/z ) 594.2 for [M + H]+ ([C24H35N5O7Y]+) and 297.7 for [M + 2H]2+. Synthesis of In(DOTA-BA). To a 5 mL vial were added DOTA-BA (50 mg, 0.081 mmol) and anhydrous InCl3 (30 mg, 0.157 mmol), followed by addition of 0.5 mL of 0.5 N ammonium citrate buffer (pH ) 4.8) and 0.5 mL of 0.5 N ammonium acetate buffer (pH ) 7.5). The mixture was heated at 100 °C for 30 min. After being cooled to room temperature, the product was separated by HPLC (Method 2). The fraction at 10.0-13.0 min was collected. Lyophilization of collected fractions gave a white powder. The yield was 37 mg. The purity > 98% by HPLC (method 3). LC-MS: m/z ) 606.3 for [M + H]+ ([C23H36N5O7In]+) and 303.7 for [M + 2H]2+. Synthesis of In(DOTA-MBA). To a 5 mL vial were added DOTA-MBA (56 mg, 0.110 mmol) and anhydrous InCl3 (50 mg, 0.183 mmol). To the mixture were added 0.5 mL of 0.5 N ammonium citrate buffer (pH ) 4.8) and 0.5 mL of 0.5 N ammonium acetate buffer (pH ) 7.5). The resulting solution was heated at 100 °C for 30 min. After being cooled to room temperature, the product was purified by HPLC (method 2). The fraction at 12-15 min was collected, and the collected fractions were evaporated to dryness. The residue was dissolved in water (20 mL), and the resulting solution was lyophilized to give a white powder. The yield was ∼40 mg. The purity was > 99% by HPLC (method 3). LC-MS: m/z ) 620.2 for [M + H]+ ([C24H35N5O7In]+) and 310.6 for [M + 2H]2+. Synthesis of 90Y(DOTA-MBA) and 90Y(DOTA-BA). To a clean sealed 5 mL glass vial were added 1.0 mL of the DOTA-BA or DOTA-MBA solution (50 µg/mL in 0.5 M ammonium acetate buffer, pH 6.0) and 20-30 µL of 90 YCl3 solution (20 mCi) in 0.05 N HCl. The resulting mixture was heated at 100 °C for 30 min. After being cooled to room temperature, a sample of the resulting solution was first diluted 20-40-fold with 2.0 mM DTPA solution (pH ) 5.0) and was then analyzed by radioHPLC (method 3; injection volume ) 5 µL). For the ITLC, a sample of the resulting solution was used without further dilution. The ITLC method used Gelman silica gel paper strips and a 1:1 (v:v) mixture of acetone and saline as eluant. Using this method, both 90Y(DOTAMBA) and 90Y(DOTA-BA) migrate to the solvent front while 90Y-colloid and unchelated free 90Y remain at the origin. Synthesis of 111In(DOTA-BA) and 111In(DOTAMBA). To a sealed 5 mL vial were added 1.0 mL of the DOTA-BA or DOTA-MBA solution (200 µg/mL in 0.5 M ammonium acetate buffer, pH 6.0) and 100 µL of 111InCl3 solution (5 mCi) in 0.05 N HCl. The resulting mixture was heated at 100 °C for 30 min. After being cooled to room temperature, a sample of the resulting solution was
analyzed by radio-HPLC (method 3) and ITLC. The ITLC method for 111In(DOTA-BA) and 111In(DOTA-MBA) is the same as that for 90Y(DOTA-BA) and 90Y(DOTA-MBA). Using this method, 111In(DOTA-BA) and 111In(DOTAMBA) migrate to the solvent front while 111In-colloid and unchelated free 111In remain at the origin. RESULTS
Synthesis of DOTA-BA and DOTA-MBA. The tributyl ester of DOTA-BA was prepared from the reaction of N-(2-chloroacetyl)benzylamine with DO3A(OBu-t)3 in the presence of excess triethylamine. Hydrolysis of the tributyl ester of DOTA-BA in a mixture of anhydrous TFA and dichloromethane produced DOTA-BA as its TFA salt. The crude product was purified by HPLC. Lyophilization of collected fractions afforded DOTA-BA in high purity (>98%). DOTA-MBA was prepared in a similar fashion from the reaction of N-(bromoacetyl)-R-methylbenzylamine with DO3A(OBu-t)3. Purification by HPLC and lyophilization of collected fractions gave DOTA-MBA with purity > 98%. Both DOTA-BA and DOTA-MBA have been characterized by HPLC, NMR, and LC-MS methods. It is important to note that DOTA-BA and DOTA-MBA have to be extremely pure for the preparation of their 90 Y and 111In complexes. Otherwise, it would be very difficult to distinguish radioimpurities from possible isomers of 90Y and 111In complexes of DOTA-BA and DOTA-MBA in their radio-HPLC chromatograms. Synthesis of M(L) (M ) Y and In; L ) DOTA-BA and DOTA-MBA). We prepared yttrium and indium complexes M(L) (M ) Y and In; L ) DOTA-BA and DOTA-MBA) by reacting DOTA-BA and DOTA-MBA, respectively, with excess yttrium(III) nitrate hexahydrate or indium(III) chloride in ammonium acetate buffer (0.5 M, pH ∼ 6) in the presence of ammonium citrate, which was used to prevent metal hydroxide formation. The product was separated from the reaction mixture by HPLC purification. Many attempts were made to grow crystals for these complexes using various solvents or solvent combinations, but were unsuccessful. In(DOTAMBA) tends to give long needlelike microcrystals, which are unsuitable for X-ray diffraction studies. All four complexes, M(L) (M ) Y and In; L ) DOTA-BA and DOTA-MBA), have been characterized by HPLC, LC-MS, and NMR (1H and 13C). Synthesis of M(L) (M ) 90Y and 111In; L ) DOTABA and DOTA-MBA). 90Y and 111In complexes of DOTAMBA and DOTA-BA were prepared by reacting DOTABA and DOTA-MBA, respectively, with MCl3 (M ) 90Y and 111In) and in ammonium acetate buffer (pH ) 6.0) at 100 °C for 30 min. We typically used 50 µg of DOTABA or DOTA-MBA for 20 mCi of 90YCl3 and 100 µg of DOTA-BA or DOTA-MBA for 5.0 mCi of 111InCl3. The use of excess DOTA-conjugate is to compensate for the presence of other trace metal contaminants in the reaction mixture and to achieve the high-yield radiolabeling. 90 Y(DOTA-BA), 90Y(DOTA-MBA), and 111In(DOTA-MBA) can be prepared in high yield with the radiochemical purity of > 98% while 111In(DOTA-BA) was prepared with the RCP of ∼90% under similar radiolabeling conditions. [90Y]- and [111In]-colloid were usually minimal (99.0 >99.0 >99.5 >99.0 ∼90 >99.0 >98.5 >98.5
11.2 10.5 17.3 16.5 7.5 6.7 11.4 10.5
Radiochemical purity in percentage.
Figure 2. Typical radio-HPLC chromatograms (method 3) of 90Y(DOTA-MBA) (top) and 90Y(DOTA-BA) (bottom).
M(L) (M ) 90Y and 111In; L ) DOTA-BA and DOTAMBA), show only one radiometric peak in the region of interest. There are some small peaks due to radioimpurities in their HPLC chromatograms (Figure 2). Since each radioimpurity is less than 0.5%, no further characterization was performed. HPLC retention times of M(DOTA-BA) (M ) 90Y and 111In) are ∼5 min shorter that those of M(DOTA-MBA) (M ) 90Y and 111In) under the same chromatographic conditions, indicating their difference in lipophilicity due to the presence of the extra methyl group. HPLC retention times of M(L) (M ) 111In and In; L ) DOTA-BA and DOTA-MBA) are 4-6 min shorter than those of M(L) (M ) 90Y and Y; L ) DOTABA and DOTA-MBA). The HPLC concordance experiments for M(DOTA-MBA) (M ) 90Y/Y and 111In/In; L ) DOTA-BA and DOTA-MBA) showed that the same complex was prepared at both tracer and macroscopic levels. 1 H NMR Spectra of M(L) (M ) Y and In; L ) DOTA-BA and DOTA-MBA). Figure 3 shows the room temperature 1H NMR spectra of Y(DOTA-MBA) (top) and Y(DOTA-BA) (bottom) in a mixture of 90% D2O and 10% H2O. In the 1H NMR spectrum of Y(DOTA-MBA), there is only one doublet at ∼1.5 ppm from hydrogens of the methyl group and one multiplet at 4.9 ppm from methylene hydrogen of the benzyl group. In the spectrum of Y(DOTA-BA), the two methylene hydrogens of the benzyl group are inequivalent and give an AB quartet at ∼4.5 ppm. The hydrogens of the methylene group of each acetate or acetamide chelating arm are also inequivalent and give four sets of AB quartets in the region of 2.0-
4.0 ppm. Each hydrogen atom in Y(DOTA-MBA) has its distinctive resonance signal. For both Y(DOTA-MBA) and Y(DOTA-BA), there are several superimposed AA′BB′ resonance signals from ethylenic hydrogens of the macrocyclic ring. NMR signals at ∼9.40 ppm are from the amide-hydrogen. The coupling between the amidehydrogen and methylene hydrogen(s) of the benzyl group is obvious. A close look at the 1H NMR spectrum of Y(DOTA-MBA) shows that there are some small satellite peaks on the right or left side of each resonance, but these peaks are so small that they are barely detectable. At temperatures 45 °C). For In(DOTA-BA) (Figure 12),
Bioconjugate Chem., Vol. 13, No. 4, 2002 909
however, the presence of the free carbonyl-oxygen makes DOTA-BA much more flexible. Rotation of the acetamide group and acetate chelating arms and inversion of ethylenic groups are relatively fast at the NMR timescale. Therefore, 1H NMR spectra of both In(DOTA-BA) and In(DOTA-MBA) show broad resonance signals at the room temperature (Figures 6 and 7). This also explains why most 13C resonance signals of In(DOTA-BA) and In(DOTA-MBA) are not seen until the temperature is higher than 65 oC (Figures 10 and 11). If indium and yttrium were to share the same coordination geometry, In(DOTA-BA) and In(DOTA-MBA) should have had a much tighter coordination sphere and should have been more rigid than their yttrium analogues, Y(DOTA-BA) and Y(DOTA-MBA), due to the smaller size of indium. The rotation of the acetamide and acetate chelating arms and inversion of ethylenic groups make In(DOTABA) to become symmetrical. As a result, methylene hydrogens from the acetamide group and those from the opposite acetate chelating arm become equivalent and show two singlets at 3.37 and 3.64 ppm, respectively, in the 1H NMR spectrum of In(DOTA-BA) (Figure 6). However, methylene hydrogens of the remaining two acetate chelating arms remain inequivalent and give two doublets at 3.06 and 3.31 ppm. For In(DOTA-MBA), rotation of the acetamide group and acetate chelating arms and inversion of ethylenic groups do not change the symmetry of the molecule due to the presence of the extra methyl group. All aliphatic hydrogens are distinctive from each other, resulting in four sets of AB quartets for hydrogens from the acetamide group and those from three acetate chelating arms (Figure 6). Isomerism in M(L) (M ) 90Y and Y; L ) DOTA-BA and DOTA-MBA). Radio-HPLC is a powerful tool for separation of different diastereomers of radiolabeled compounds. In this study, we used a reverse phase radioHPLC method to analyze 90Y(DOTA-BA) and 90Y(DOTAMBA). The advantage of radio-HPLC is its capability to determine the radiochemical purity of the 90Y-labeled DOTA-conjugate and to separate different 90Y species in the reaction mixture. A slow gradient mobile phase was used to maximize the separation of different 90Y-containing species. We also tried several other reverse phase radio-HPLC methods using various reverse phase columns or different solvent combinations. In all the cases, 90 Y(DOTA-BA) and 90Y(DOTA-MBA) show only one radiometric peak in their radio-HPLC chromatogram. DOTA-BA is a DOTA-monoamide with no chiral center. In principle four isomers (A-D in Figure 13) are possible for Y(DOTA-BA). The interconversion within each enantiomeric pair (A S B or C S D) occurs through inversion of the ethylenic groups and concerted or stepwise rotation of the acetate arms (6, 17, 38). On the other hand, the exchange between diastereomers (A S C or B S D) takes place either by rotation of the acetate/acetamide chelating arms while maintaining the same conformation of the macrocyclic ring or by inversion of the ethylenic groups without changing the orientation of the acetate/acetamide chelating arms. If the interconversion between diastereomers (M and m isomers) is sufficiently slow, they should be readily separable by HPLC, particularly using a slow gradient mobile phase. There are two possible explanations for the presence of only one radiometric peak in the HPLC chromatogram of 90Y(DOTA-BA). The lipophilicity of two diastereomers is so similar that they could not be separated by the HPLC method used in this study. Alternatively, the interconversion between two diastereomers is so fast at the HPLC time-scale that the observed single peak is an
Figure 13. Four possible isomeric forms of the complex Y(DOTA-monoamide).
averaged signal from two diastereomers. To eliminate the first possibility, we prepared DOTA-MBA and introduced a chiral center into the DOTA-monoamide conjugate. Unlike those in Y(DOTA-BA), isomers A and B or C and D in Y(DOTA-MBA) are no longer enantiomers but diastereomers, if the exchange between isomers is sufficiently slow at room temperature. We hoped that the conjugation of a chiral amine would make these diastereomers more separable by HPLC. To our surprise, 90Y(DOTA-MBA) shows only one radiometric peak in its radio-HPLC chromatogram, suggesting that it exists in solution as one predominant isomer. It should be noted that one radiometric peak in the radio-HPLC-chromatogram does not necessarily mean only one isomer for any radiolabeled small biomolecule. It needs further confirmation using other analytical methods. Therefore, we prepared Y(DOTA-BA) and Y(DOTA-MBA) and used NMR methods (1H and 13C) to study the solution structure of Y(DOTA-BA) and Y(DOTAMBA). At the NMR time-scale, the rate of exchange between different isomers is relatively slow so that the presence of different isomers in solution is readily detectable. In the 1H NMR spectrum of Y(DOTA-BA), the two methylene hydrogens of benzyl group are inequivalent and give an AB quartet at ∼4.5 ppm (Figure 3). The methylene hydrogens from each acetate/acetamide chelating arm are also inequivalent, and four sets of AB quartets are seen in the region of 2.0-5.0 ppm. These observations suggest that Y(DOTA-BA) is chiral with DOTA-BA being prochiral and that the coordinated DOTA-BA is rigid and nonfluxional at room temperature. Thus, there should be no rapid interconversion within each enantiomeric pair (A S B or C S D) or between diastereomers (A S C, B S D, A S D, or B S C). For Y(DOTA-BA), isomers A and B or C and D are enantiomers and indistinguishable by NMR. For Y(DOTAMBA), however, all isomers (A, B, C, and D) are diastereomers and should be distinguishable by NMR. The presence of only one doublet from the methyl group
910 Bioconjugate Chem., Vol. 13, No. 4, 2002
strongly suggests that Y(DOTA-MBA) exist in solution as one predominant isomer. This conclusion supported by observations of only one 1H NMR signal from the amide group and 13 aliphatic 13C signals from 14 different aliphatic carbon atoms in Y(DOTA-MBA) (Figure 9) and is also consistent with the presence of only m-isomer of Y(DOTA-D-Phe-NH2) in the solid state (17). If all four diastereomers were to exist in solution, a total of 56 aliphatic 13C signals would have been detected in the 13C NMR spectrum of Y(DOTA-MBA). It should be noted that there are some small satellite peaks on both sides of each main peak of the 1H NMR spectrum of Y(DOTA-MBA). This is particularly true in the 1H NMR spectra of Y(DOTA-MBA) at lower temperatures (
