Bioconjugate Chem. 1997, 8, 673−679
673
Design and Synthesis of [111In]DTPA-Folate for Use as a Tumor-Targeted Radiopharmaceutical Susan Wang,† Jin Luo,† Douglas A. Lantrip,† David J. Waters,‡ Carla J. Mathias,§ Mark A. Green,§ Philip L. Fuchs,*,† and Philip S. Low*,† Department of Chemistry, Department of Veterinary Clinical Sciences, and Department of Medicinal Chemistry and Molecular Pharmacology, Purdue University, West Lafayette, Indiana 47907. Received January 17, 1997X
Folate-conjugated metal chelates have been proposed as potential imaging agents for cancers that overexpress folate receptors. In a previous study, folic acid was linked through its γ-carboxyl group to deferoxamine (DF), and the 67Ga-labeled complex ([67Ga]DF-folate) was examined for in vivo tumor targeting efficiency in athymic mice with a human tumor cell implant. Although superb tumor-tobackground contrast was obtained, slow hepatobiliary clearance would compromise imaging of abdominal tumors such as ovarian cancer. In the present study, folic acid was conjugated to an alternative chelator, diethylenetriaminepentaacetic acid (DTPA), via an ethylenediamine spacer. The desired DTPA-folate(γ) regioisomer was synthesized by two different approaches, purified by reversed phase column chromatography, and characterized mainly by analytical HPLC, mass spectroscopy, and NMR. In cultured tumor cells, uptake of [111In]DTPA-folate(γ) was found to be specific for folate receptor-bearing cells, and the kinetics of uptake were similar to those of free folate and other folateconjugated molecules. In the normal rat, intravenously administered [111In]DTPA-folate(γ) was found to be rapidly excreted into the urine, giving intestinal levels of radiotracer 10-fold lower than those observed with [67Ga]DF-folate(γ) at 4 h. In a preliminary mouse imaging study, a folate receptorpositive KB cell tumor was readily visualized by γ scintigraphy 1 h following intravenous administration of [111In]DTPA-folate(γ).
INTRODUCTION
The membrane-associated folate receptor is a tumor marker that is overexpressed on a variety of neoplastic tissues, including breast, cervical, ovarian, colorectal, renal, and nasoparyngeal tumors, but highly restricted in most normal tissues (Rettig et al., 1988; Campbell et al., 1991; Coney et al., 1991; Weitman et al., 1992; GarinChesa et al., 1993; Holm et al., 1994; Franklin et al., 1994; Ross et al., 1994; Stein et al., 1991; Li et al., 1996). Previously, it has been shown that the natural receptormediated endocytosis pathway for the vitamin folic acid can be exploited to selectively and nondestructively deliver folate-conjugated small molecules, macromolecules, and drug carriers such as liposomes into cultured tumor cells (Wang et al., 1996; Leamon and Low, 1991, 1993; Lee and Low, 1994). When folate is covalently linked to a molecule via its γ-carboxyl moiety, its affinity for its cell surface receptor (Kd ∼ 10-9 M; McHugh and Cheng, 1979; Antony et al., 1985; Kamen and Capdevila, 1986; Kane and Waxman, 1989; Luhrs et al., 1992; Matsue et al., 1992) remains essentially unaltered. Further, following binding to the cell surface receptor, the conjugated folate is internalized by the cell in much the same manner as the unmodified vitamin (Leamon and Low, 1991; Lee et al., 1996). Recycling of the folate receptor can then lead to further accumulation of the folate conjugates in such target cells. To evaluate the potential of a radiolabeled folate conjugate in tumor diagnostic imaging, the metal chelator
deferoxamine (DF)1 was linked to folic acid, labeled with and imaging was performed in an athymic mouse tumor model using the [67Ga]DF-folate radiopharmaceutical (Wang et al., 1996; Mathias et al., 1996a). The [67Ga]DF-folate conjugate was found to afford excellent tumor/blood, tumor/muscle, and tumor/liver contrast (Mathias et al., 1996a,b). The greatly enhanced tumor/ background contrast (tumor to blood ratio averaged 409 after 4 h and 1500 after 46 h postinjection) indicated the high specificity and efficiency of folate-targeted delivery, probably due to its small size, and therefore the very rapid clearance rate from blood and other nontarget tissues such as liver and muscle. However, it was also shown that approximately 30% of the administered [67Ga]DF-folate was excreted via the hepatobiliary route into the intestines (Mathias et al., 1996b), making the clearance rate through the GI tract a major factor in determining the time frame over which imaging of abdominal tumors would be feasible. Our goal in this study has been to optimize radiotracer performance for tumor imaging by development of a folate receptor-targeted agent that is cleared more rapidly and selectively from the body into the urine. Diethylenetriaminepentaacetic acid (DTPA) was chosen for this purpose because of its high water solubility, relatively simple conjugation chemistry (Hnatowich et al., 1983), and strong metal-chelating interaction with 111In, a 67.4 h half-life γ-emitting radionuclide with nuclear properties desirable for tumor imaging. 67Ga,
EXPERIMENTAL PROCEDURES * To whom correspondence should be addressed. Phone: (317) 494-5273. Fax: (317) 494-0239. † Department of Chemistry. ‡ Department of Veterinary Clinical Sciences. § Department of Medicinal Chemistry and Molecular Pharmacology. X Abstract published in Advance ACS Abstracts, September 1, 1997.
S1043-1802(97)00129-8 CCC: $14.00
Materials. Folic acid, dimethyl sulfoxide (DMSO), N-hydroxysuccinimide (NHS), carboxypeptidase G2, fluo1 Abbreviations: DMSO, dimethyl sulfoxide; NHS, N-hydroxysuccinimide; FITC, fluorescein isothiocyanate; TFA, trifluoroacetic acid; EDA, ethylenediamine; DTPA, diethylenetriaminepentaacetic acid; DF, deferoxamine; PBS, phosphate-buffered saline; tR, retention time.
© 1997 American Chemical Society
674 Bioconjugate Chem., Vol. 8, No. 5, 1997
rescein isothiocyanate (FITC), tetrabutylammonium phosphate, and cyclic DTPA dianhydride were purchased from Sigma Chemical Co. (St. Louis, MO). The bicinchoninic acid protein assay kit was obtained from Pierce (Rockford, IL). Acetonitrile (HPLC grade), dicyclohexylcarbodiimide, trifluoroacetic acid (TFA), and ethylenediamine (EDA) were purchased from Aldrich (Milwaukee, WI). [111In]Indium(III) chloride was purchased from Mallinckrodt Medical, Inc. (St. Louis, MO). Tissue culture products were obtained from GIBCO (Grand Island, NY), and cultured cells were received as a gift from the Purdue Cancer Center (West Lafayette, IN). Cell Culture. KB cells, a human nasopharyngeal epidermal carcinoma cell line that overexpresses the folate binding protein, and A549 cells, a human lung carcinoma cell line that expresses no detectable folate receptors, were cultured continuously as a monolayer at 37 °C in a humidified atmosphere containing 5% CO2 in folate-deficient modified Eagle’s medium as described previously (Wang et al., 1996). Forty-eight hours prior to each experiment, the cells were transferred to 35 mm culture dishes at 5 × 105 cells per dish and grown to ∼80% confluence. Synthesis and Purification of the EDA-Derivatized Folic Acid. (Method A) Synthesis of a Mixture of EDA-Folate(R) and EDA-Folate(γ). Folic acid (441 mg, 1 mmol) slowly dissolved in 20 mL of DMSO was reacted with 1.2 molar equiv of dicyclohexylcarbodiimide and 2.0 molar equiv of NHS at 50 °C for 6 h. The resulting folate-NHS was mixed with 10 molar equiv of EDA plus 100 µL of pyridine and allowed to react at 25 °C for ∼5 h, by which time the reaction had reached completion by TLC analysis (silica gel plate, 2-propanol/chloroform ) 7/3, Rf values for folic acid and the mixture of EDA isomers were 0.2 and 0.4, respectively). The crude product was precipitated by addition of 20 mL of acetonitrile, centrifuged, and then washed three times with diethyl ether before drying under vacuum. A fine dark yellow powder (390 mg) was obtained. The γ-isomer of EDA-folate (58 mg, 12% yield) was separated from the R-isomer and unreacted folic acid by HPLC on a Microsorb preparative C-18 reversed phase column (250 mm × 21 mm) using a linear gradient (eluant A, water with 0.05% TFA at pH 3.5; eluant B, acetonitrile; gradient, 0 to 15% B over 20 min at a flow rate of 10 mL/min). The elution times for EDA-folate(γ) and EDA-folate(R) were 7.6 and 10.3 min, respectively. The chemical purities and identities of the two EDAfolate monoisomers were examined by analytical HPLC on an Econosphere C-18 reversed phase column (150 mm × 4.6 mm) [EDA-folate(R), tR ) 12.9 min; EDA-folate(γ), tR ) 15.0 min; eluent A, 5 mM phosphate buffer at pH 7; eluent B, acetonitrile; gradient, 1 to 10% B over 15 min at a flow rate of 0.7 mL/min]. Mass spectroscopic analysis and the carboxypeptidase G2 hydrolysis assay were conducted as described previously (Wang et al., 1996). (Method B) Regiospecific Synthesis of EDA-Folate(γ) (Figure 1). To 2.8 g of tetramethylguanidinium L-methyl folate(γ) [5.95 mmol, a new compound of which the synthesis was reported by Luo et al. (1997)] was added 20 mL of ethylenediamine (0.3 mol) with stirring at 25 °C. The solid gradually dissolved as it reacted with the diamine. The reaction was complete in 3 h as indicated by analytical HPLC under the same conditions described in method A, and filtration gave a clear solution which was then transferred to a well-stirred mixture of acetonitrile and diethyl ether (1/1 v/v, 500 mL). The precipitated solid was collected by centrifugation and redissolved in 500 mL of water, followed by addition of 5%
Wang et al.
Figure 1. Regioselective synthesis of the DTPA-folate conjugate.
hydrochloric acid until pH 7.0, which resulted in precipitation of the product. The solid was collected by centrifugation and washed thoroughly with 3 × 250 mL of water to remove any trace of ethylenediamine (1H NMR was used to assay for the presence of the diamine). A yellow solid (2.1 g, 88% yield) was obtained after washing once with 100 mL of acetonitrile and 3 × 50 mL of diethyl ether and drying for 24 h under vacuum. Analytical HPLC showed a single peak at 15.2 min under the same analysis conditions as above or at 3.0 min using an ionpair agent tetrabutylammonium phosphate (eluant A, 10 mM tetrabutylammonium phosphate buffer at pH 7; eluant B, acetonitrile; gradient, 15 to 50% B over 30 min at a flow rate of 0.7 mL/min). 1H NMR (300 MHz, DMSO-d6/CF3CO2D, ∼10/1 v/v): δ 8.75 (s, 1H, C7-H), 7.66 (d, J ) 8.6 Hz, 2H, Ar), 6.63 (d, J ) 8.6 Hz, 2H, Ar), 4.57 (s, 2H, C9-H2), 4.34 (dd, J ) 4.0, 9.6 Hz, 1H, C19-H), 3.28-3.13 (m, 2H, C25-H2), 2.80 (m, 2H, C26-H2), 2.16 (m, 2H, C22-H2), 2.15-1.85 (m, 2H, C21-H2). 13C NMR (75 MHz, D2O/NaOD): δ 178.8 (C20), 176.0 (C23), 173.1 (C17), 169.5 (C2), 164.0 (C4), 155.5 (C8a), 151.1 (C11), 147.3 (C6), 147.2 (C7), 129.0 (C13 and C15), 128.2 (C4a), 121.4 (C14), 112.5 (C12 and C16), 55.3 (C19), 45.8 (C9), 42.0 (C25), 39.8 (C26), 32.8 (C22), 28.0 (C21) (see the Supporting Information). High-resolution MS (fast atom bombardment): C21H25N9O5 [M + H]+ 484.2057, found 484.2062. Decomposition point: ∼278 °C. Evaluation of FITC-EDA-Folate Affinity for Cell Surface Receptors. Five milligrams of each of the Rand γ-isomers of EDA-folate from method A were dissolved in 1 mL of DMSO and reacted with 3 molar equiv of FITC at room temperature for 3 h. The folateconjugated FITC was precipitated with 10 mL of cold acetone, after which unreacted FITC was removed by washing the pellet three times with cold acetone. Conjugation was confirmed by UV-visible spectroscopy of the product by demonstrating that the resulting spectrum approximated the sum of the spectra of folate and FITC. Affinities of the two folate derivatives for cell surface
[111In]DTPA−Folate as a Tumor-Imaging Agent
folate receptors were determined by incubating KB cells with various concentrations of the conjugates at 4 °C for 30 min. The cells were then washed with 3 × 1 mL of phosphate-buffered saline (PBS, 136.9 mM NaCl, 2.68 mM KCl, 8.1 mM Na2HPO4, and 1.47 mM KH2PO4 at pH 7.4) and dissolved in 2 mL of 1% TX-100. Cell-associated fluorescence was measured on a Perkin-Elmer MPF-44A fluorescence spectrophotometer. Cellular protein was determined by the bicinchoninic acid assay (Lee and Low, 1994). Competition between the folate conjugates and free folic acid for cell surface receptors was examined by including 1 mM folic acid in the incubation medium. Production of DTPA-Folate. One gram of EDAfolate(γ) (2.1 mmol) was dissolved in 50 mL of DMSO by bath sonication overnight. Then the dark yellow solution was slowly added to a stirring suspension of 2.0 g of DTPA dianhydride (5.6 mmol) in 10 mL of anhydrous DMSO. The mixture became homogeneous by the end of addition. Analytical HPLC showed the absence of the starting material EDA-folate after 30 min, at which time the reaction mixture was filtered through a pad of Celite to remove traces of solid residue. After the temperature of the reaction mixture was reduced with an ice bath, 10 mL of 2.5 N NaOH was added to quench the reaction and to neutralize the solution. The resulting precipitate containing the majority of the DTPA-folate(γ) produced was separated by centrifugation from the supernatant. The yellow pellet was then washed with 100 mL of acetonitrile and dissolved in 50 mL of water, and the pH of the solution was adjusted to 7 with concentrated HCl. After filtration to remove the solid residue, the clear orange solution was directly purified on a LiChroprep C-18 reversed phase MPLC column (310 mm × 25 mm, 45-60 µm) using 10 mM ammonium bicarbonate buffer as the eluant (tR ) 20-35 min, flow rate of 10 mL/min). The collected product was concentrated by vaccum and further purified by preparative HPLC on the Microsorb C-18 column with a gradient (eluant A, 10 mM ammonium bicarbonate buffer at pH 7.4; eluant B, acetonitrile; gradient, 0 min at 4% B, 10 min at 12% B, and 15 min at 15% B at a flow rate of 10 mL/min; tR ) 6.3 min) to remove residual bis-conjugated side product (Figure 1) and to obtain 0.85 g of DTPA-folate after lyophilization with a purity above 99% and a yield of 47%. Analytical HPLC on the Econosphere C-18 reversed phase column (150 mm × 4.6 mm) revealed a single peak with a retention time of 11.74 min (eluant, 10 mM tetrabutylammonium phosphate buffer (pH 7) at 75% and acetonitrile at 25%; flow rate of 0.7 mL/min). 1H NMR (300 MHz, D2O): δ 8.46 (s, 1H, C7-H), 7.41 (d, J ) 8.3 Hz, 2H, Ar), 6.34 (d, J ) 8.3 Hz, 2H, Ar), 4.24 (dd, J ) 4.4, 8.4 Hz, 1H, C19-H), 4.15 (s, 2H, C9-H2), 3.53 (s, 4H), 3.47 (s, 2H), 3.42 (s, 2H), 3.18-2.97 (overlap, 14H), 2.28 (m, 2H, C22-H2), 2.28-1.83 (m, 2H, C21-H2) (see the Supporting Information). 13C NMR (75 MHz, D2O): δ 178.9, 178.0, 175.9, 175.7, 173.6, 173.2, 169.0, 164.8 (C4), 154.4 (C2), 152.8 (C8a), 150.2 (C11), 148.3, 148.2, 128.9 (C13 and C15), 126.5 (C4a), 120.7 (C14), 111.7 (C12 and C16), 58.6, 58.2, 57.9, 55.6, 55.4, 51.9, 51.6, 51.1, 50.6, 45.4 (C9), 38.8 (C25), 38.5 (C26), 32.8 (C22), 28.1 (C21) (also see the attached spectrum for assignment). Lowresolution MS (matrix-assisted laser desorption ionization): C35H45N12O14 [M - H]- 857.3, found 857.8. Radiotracer Synthesis. The [111In]DTPA-folate radiopharmaceutical was obtained in high radiochemical yield by ligand exchange from [111In]citrate. Briefly, 111 In3+ (0.2-5.4 mCi) in HCl (0.05 M, 2-55 µL) was transferred to a test tube and buffered by addition of 200 µL of 3% aqueous sodium citrate. The resulting [111In]citrate was mixed with 300-350 µg of DTPA-folate in
Bioconjugate Chem., Vol. 8, No. 5, 1997 675
water (2 mg/mL, pH 7-8). Two to 24 h later, the radiochemical purity of the [111In]DTPA-folate was determined by TLC on C-18 reversed phase plates eluted with methanol and consistently found to exceed 98% ([111In]DTPA-folate Rf ) 0.8; [111In]citrate Rf ) 0.0). The [111In]DTPA-folate product was diluted with normal saline, as needed, prior to use in the animal biodistribution experiments. [67Ga]DF-folate was prepared as described previously (Mathias et al., 1996a). Cellular Uptake of [111In]DTPA-Folate. Cultured KB and A549 cells in 35 mm dishes were incubated with 100 nM [111In]citrate or [111In]DTPA-folate in 1 mL of folate-deficient medium at room temperature for various lengths of time. The cells were then washed with 3 × 1 mL of PBS and suspended in 1 mL of PBS by scraping. The amount of cell-associated radioactivity was determined using an automatic γ counter. Cellular protein content was measured by the bicinchoninic acid assay. In folate competition experiments, the same protocol was used except that 1 mM folic acid was included in the incubation medium. Biodistribution and Imaging Studies with [111In]DTPA-Folate. All animal studies were carried out in accordance with procedures approved by the Purdue Animal Care and Use Committee. The biodistributions of [111In]DTPA-folate and [67Ga]DF-folate were determined in normal male Sprague-Dawley rats following intravenous injection under diethyl ether anesthesia, as described previously (Tsang et al., 1994). In addition, a γ scintillation image of a female athymic mouse Nu/Nu (22 g) was obtained using a PhoGamma 37GP camera fitted with a 300 keV parallel hole collimator. The animal used for this imaging study was maintained for 3 weeks on a folate-free diet (to reduce its serum folate to a level near that of human serum) and had been implanted subcutaneously in the interscapular region with human KB tumor cells, as described previously (Mathias et al., 1996a). The tumor mass at the time of the imaging study was approximately 0.25 g. The animal was imaged for 1 h following intravenous administration of [111In]DTPA-folate (200 µCi in 0.1 mL) via the femoral vein under diethyl ether anesthesia. To promote clearance of the radiotracer from the urinary bladder, 1.5 mL of sterile saline was administered by ip injection immediately following the intravenous radiotracer injection. For imaging, the mouse was re-anesthetized with ketamine (60 mg/kg) and xylazine (6 mg/kg) immediately prior to the image acquisition period. The total administered mass of the DTPA-folate conjugate was 13 µg (0.57 mg/kg). RESULTS
Production of the EDA-Derivatized Folic Acid. Ethylenediamine (EDA) was conjugated to the γ-carboxyl of folic acid either by carbodiimide activation or by regioselective methyl folate synthesis (Figure 1) to serve as a linker between the vitamin and the metal chelator, DTPA. High-resolution FAB mass spectroscopic analysis showed the EDA-folate [M + H]+ parent ion peak at m/e ) 484.2062 (calculated 484.2057), indicating the desired 1/1 conjugation ratio. The chemical purity and identity of EDA-folate(γ) was confirmed by analytical HPLC and NMR spectroscopy (see the Supporting Information). The 1H NMR spectrum shows that one of the amine groups on EDA is conjugated to the γ-carboxylate of folic acid, shifting the protons on the ethylene group to a lower field. Derivatization of the folate at the γ-carboxyl, in contrast to the R-carboxyl, rendered the conjugate a substrate of the hydrolytic enzyme carboxypeptidase G2 (Levy and Goldman, 1967; Fan et al., 1991).
676 Bioconjugate Chem., Vol. 8, No. 5, 1997
Wang et al.
Figure 3. Determination of the affinities of FITC-EDA-folate isomers for cell surface folate receptors. KB cells were incubated with various concentrations of the R-isomer (O) or γ-isomer (b) of the conjugate at 4 °C for 30 min. Cell-associated fluorescence was measured as described in Experimental Procedures. Competition for the receptor binding of FITC-EDA-folate(γ) by excess folic acid (1) was also evaluated.
Figure 2. Purification of DTPA-folate(γ) by HPLC. Fifty milligrams of the reaction crude product dissolved in 0.5 mL of H2O was loaded onto a Microsorb preparative C-18 reversed phase column and eluted with the gradient shown above the chromatogram. The desired DTPA-folate(γ) conjugate was eluted at 6.3 min, while the peak at 9.2 min was found to be the DTPA-bis-folate(γ,γ′) side product. Other peaks were the impurities present in tetramethylguanidinium L-γ-methyl folate.
Synthesis of DTPA-Folate. The amide-linked DTPA-folate(γ) conjugate (Figure 1) was obtained by reaction of cyclic DTPA anhydride with the purified EDA-folate(γ), followed by chromatographic purification to remove the major side product DTPA-bis-folate(γ,γ’) and other impurities (Figure 2). The chemical structure of purified DTPA-folate was characterized by MS and NMR (see the Supporting Information), and the purity was confirmed by analytical HPLC. The ion-pair agent in the mobile phase was found to be essential for a quantitative HPLC analysis, probably due to the complete neutralization of the multiple charges on DTPAfolate(γ) which allowed adequate interaction between the sample compound and the hydrocarbon stationary phase. Affinities of FITC-EDA-Folate for Cell Surface Receptors. To determine the affinities of EDA-folate for cell surface folate receptors, the R- and γ-isomers were labeled with the fluorophore FITC and incubated with KB cells overexpressing the receptor. As shown in Figure 3, 1.6 nM FITC-EDA-folate(γ) isomer was required to reach 50% maximal binding, similar to that of folic acid and DF-folate(γ). Excess folic acid in the incubation medium effectively competed with the receptor binding of FITC-EDA-folate(γ). On the other hand, the R-isomer of FITC-EDA-folate had virtually no affinity for the cell surface receptors. The low level of nonspecific uptake observed with both isomers was probably due to the hydrophobicity of the FITC conjugates. Cellular Uptake of [111In]DTPA-Folate. The kinetics of the cellular uptake of the [111In]DTPA-folate complex were evaluated by a time-dependent binding assay. As shown in Figure 4, the time needed to reach 50% saturation of available folate receptors was ∼3 min at room temperature, similar to that observed for folic acid and folate conjugates such as [67Ga]DF-folate (Wang
Figure 4. Cellular uptake of 111In-labeled imaging agents. KB and A549 cells were treated with 100 nM [111In]citrate (O) or [111In]DTPA-folate(γ) with (1) or without (b) competition by 1 mM folic acid for various lengths of time. Cell-associated radioactivity was measured by automated γ counting as described in Experimental Procedures.
et al., 1996) and [125I]BSA-folate (Leamon et al., 1991). Coincubation with 1 mM free folic acid completely blocked conjugate uptake (Figure 4, inverted triangles), further confirming that cell association of the radioactivity was mediated by folate receptors. A549 cells which lack folate receptors did not show significant [111In]DTPA-folate uptake at any time point. Nonspecific uptake of the complex was also not observed in either cell line, probably due to the sufficient hydrophilicity of the entire molecule. Cellular uptake of [111In]citrate, on the other hand, was not saturable and virtually identical in the cell line expressing (KB cell) and not expressing (A549 cell) the folate receptor. Biodistribution of [111In]DTPA-Folate. Following intravenous administration to rats, the [111In]DTPAfolate radiotracer was found to be efficiently cleared from the blood and primarily excreted into the urine (Table 1). Only 3.7 ( 1.4% of the injected 111In dose was found in the intestines at 4 h postinjection (Table 1), a value 10-fold lower than the intestinal radioactivity observed with [67Ga]DF-folate in the rat model at the same time
[111In]DTPA−Folate as a Tumor-Imaging Agent
Bioconjugate Chem., Vol. 8, No. 5, 1997 677
Table 1. Biodistribution of [111In]DTPA-Folate in Normal Male Rats Following Intravenous Injection percentage of injected dose per organ (tissue)a
111In
blood heart lungs liver spleen kidneys (two) intestines and contents bladder and contents
4h
24 h
0.078 ( 0.005 0.021 ( 0.002 0.024 ( 0.003 0.25 ( 0.01 0.019 ( 0.004 13.5 ( 1.4 3.7 ( 1.4 0.39 ( 0.63
0.029 ( 0.006 0.017 ( 0.002 0.022 ( 0.001 0.17 ( 0.01 0.017 ( 0.003 12.1 ( 1.3 1.2 ( 0.3 0.09 ( 0.14
a Values shown represent the mean ( standard deviation of data from three animals (188 ( 7 g). Blood was assumed to account for 7% of the total body mass. The DTPA-folate dose was 0.268 ( 0.021 mg/kg.
Table 2. Biodistribution of [67Ga]Deferoxamine-Folate ([67Ga]Df-Folate) in Normal Male Rats Following Intravenous Injection percentage of injected dose per organ (tissue)a
67Ga
blood heart lungs liver spleen kidneys (two) intestines and contents bladder and contents
5 min
4h
24 h
8.7 ( 0.2 0.20 ( 0.01 0.51 ( 0.07 8.1 ( 0.4 0.099 ( 0.012 8.63 ( 0.35 10.3 ( 1.1
0.072 ( 0.004 0.020 ( 0.002 0.023 ( 0.004 0.23 ( 0.03 0.007 ( 0.001 11.5 ( 1.0 39.3 ( 1.3
0.045 ( 0.019 0.017 ( 0.001 0.016 ( 0.001 0.203 ( 0.002 0.0075 ( 0.0017 10.2 ( 0.7 2.7 ( 3.5
12.6 ( 3.2
12.8 ( 3.2
0.02 ( 0.02
a Values shown represent the mean ( standard deviation of data from three animals (186 ( 7 g). Blood was assumed to account for 7% of the total body mass. The Df-folate dose was 0.262 ( 0.012 mg/kg.
point (Table 2). Since it is clinically desirable to be able to initiate imaging as rapidly as possible following radiotracer injection, [111In]DTPA-folate may offer advantages over [67Ga]DF-folate for imaging abdominal tumors, due to a more favorable pattern of clearance from nontarget tissues. The [67Ga]DF-folate and [111In]DTPAfolate tracers also differ in the amount of radiotracer found in the “bladder and contents” at 4 h postinjection; however, this difference is of questionable significance, since no provision was made to prevent spontaneous voiding of the urinary bladder. By 24 h postinjection, both the [111In]DTPA-folate and [67Ga]DF-folate radiolabels are largely cleared from the body, with only the kidneys exhibiting substantial levels of radioactivity (Tables 1 and 2). This prolonged renal retention of a fraction of both agents probably reflects radiopharmaceutical binding to the folate receptor known to be present in the proximal tubules (Selhub and Franklin, 1984; Selhub et al., 1987a,b). The ability of [111In]DTPA-folate to target folate receptors in vivo also appears to be demonstrated in a preliminary imaging study with an athymic mouse bearing a subcutaneous folate receptor-positive human KB cell tumor. The KB cell tumor is readily apparent in the γ image obtained 1 h following intravenous administration of the [111In]DTPA-folate radiopharmaceutical (Figure 5). While radiotracer uptake is also apparent in the kidneys, the tumor/background tissue contrast is otherwise quite good, even at this relatively short time postinjection. DISCUSSION
The development of new and improved tumor-selective radiopharmaceuticals is clinically desirable as a means
Figure 5. Whole body γ image (dorsal view; 200 000 counts) of a female athymic mouse with a subcutaneous folate receptorpositive tumor in the right shoulder 1 h following intravenous administration of [111In]DTPA-folate(γ). Radiotracer uptake is apparent in both the tumor (upper right) and kidneys.
of (i) detecting and/or confirming the presence and location of primary and metastatic lesions, (ii) probing biochemical features of neoplastic tissue that have implications for tumor staging and/or subsequent treatment, and (iii) monitoring tumor response to treatment. If sufficiently high tumor specificity can be achieved, radiopharmaceuticals labeled with an appropriate R- or β-emitting nuclide could also provide an attractive means for site-selective delivery of radiation therapy. It has been demonstrated that only those folate conjugates containing the adduct attached at the γ-carboxyl of folic acid retain the ability to bind to cell surface folate receptors with the same affinity as free folic acid (Wang et al., 1996). In this paper, two synthetic methods were employed to conjugate the EDA linker to folic acid. Although method A represents a relatively simple and fast synthetic route, its lack of regioselectivity yields both an inactive R-conjugate and the active γ-conjugate. Also, the requirement for HPLC purification to separate the two isomers and the poor water solubility of the EDAfolate intermediate severely limit large scale production of the desired product. Regiospecific functionalization of the γ-carboxylate on folic acid was achieved, however, by converting folic acid to pteroyl azide and then attaching the commercially available 5-methyl glutamate to form γ-methyl folate (Luo et al., 1997). In this procedure (method B), production of large amounts of EDA-folate(γ) could be accomplished simply by dissolving the γ-methyl folate in EDA. The overall yield from folic acid to EDA-folate(γ) was 52%, and a chemical purity of above 90% was achieved without column purification. Importantly, this
678 Bioconjugate Chem., Vol. 8, No. 5, 1997
unique regiospecific activation of the γ-carboxyl on folic acid can potentially be applied in the preparation of many folate-targeted anticancer agents. Our previous imaging study with [67Ga]DF-folate(γ) in the athymic mouse KB cell tumor model demonstrated the feasibility of targeting tumor folate receptors in vivo with simple folate-chelate conjugates (Mathias et al., 1996a). However, roughly 30% of the [67Ga]DF-folate(γ) radiopharmaceutical was found to be excreted via the GI tract, where its presence could interfere with clear and rapid visualization of abdominal cancers. In order to overcome this obstacle, in the present study, [111In]DTPA-folate(γ) was synthesized and shown to have affinity for the folate receptor in vitro. In preliminary rat and tumor-bearing mouse studies, the [111In]DTPAfolate(γ) appears to afford targeting to folate receptors in vivo, while also providing reasonably selective clearance of unbound radiotracer into the urine. This secondgeneration folate-chelate imaging agent would appear to offer advantages over the previously reported folatedeferoxamine conjugate for imaging tumors in the abdominal region, where the intestinal levels of [67Ga]DFfolate(γ) are expected to contribute undesirably high background activity. While further experiments will clearly be required to better define the suitability of [111In]DTPA-folate(γ) as a radiopharmaceutical for imaging folate receptor-positive tumors and to prove the role of the folate receptor in mediating tumor uptake of the 111In radiolabel, the reported imaging results with the athymic mouse tumor model (Figure 5) support the conclusion that this agent can provide tumor-selective radionuclide delivery in vivo. ACKNOWLEDGMENT
This work was supported in part by a grant from the National Cancer Institute (R01-CA70845). The Purdue Athymic Mouse Facility and Cell Culture Facility are partially supported by Cancer Center (Core) Support Grant P30-CA23168 awarded by the National Cancer Institute. Supporting Information Available: 1H and 13C NMR spectra of EDA-folate(γ), 1H and 13C NMR spectra of DTPA-folate(γ), analytical HPLC chromatograms of the two compounds with the conditions described in Experimental Procedures, and a table showing the biodistribution of [111In]DTPA-folate and [67Ga]DF-folate in rats calculated as a percentage of the injected dose per gram (%ID/g) of tissue wet weight (7 pages). Ordering information is given on any current masthead page. LITERATURE CITED Antony, A. C., Kane, M. A., Portillo, R. M., Elwood, P. C., and Kolhouse, J. F. (1985) Study of the role of a particulate folatebinding protein in the uptake of 5-methyltetrahydrofolate. J. Biol. Chem. 260, 14911-14917. Campbell, I. G., Jones, T. A., Foulkes, W. D., and Trowsdale, J. (1991) Folate-binding protein is a marker for ovarian cancer. Cancer Res. 51, 5329-5338. Coney, L. R., Tomassetti, A., Carayannopoulos, L., Frasca, V., Kamen, B. A., Colnaghi, M. I., and Zurawski, V. R., Jr. (1991) Cloning of a tumor-associated antigen: MOv18 and MOv19 antibodies recognize a folate-binding protein. Cancer Res. 51, 6125-6132. Doerr, R. J., Abdel-Nabi, H., Krag, D., and Mitchell, E. (1991) Radiolabeled antibody in the management of colorectal cancer. Results of a multicenter study. Ann. Surg. 214, 118124. Fischman, A. J., Khaw, B. A., and Strauss, H. W. (1989) Quo vadis radioimmune imaging. J. Nucl. Med. 30, 1911-1915.
Wang et al. Franklin, W. A., Waintrub, M., Edwards, D., Christensen, K., Prendegrast, P., Woods, J., Bunn, P. A., and Kolhouse, J. F. (1994) New anti-lung-cancer antibody cluster 12 reacts with human folate receptors present on adenocarcinoma. Int. J. Cancer 8 (Suppl.), 89-95. Fritzberg, A. R., Berninger, R. W., Hadley, S. W., and Wester, D. W. (1988) Approaches to radiolabeling of antibodies for diagnosis and therapy of cancer. Pharm. Res. 5, 325-334. Garin-Chesa, P., Campbell, I., Saigo, P., Lewis, J., Old, L., and Rettig, W. (1993) Trophoblast and ovarian cancer antigen LK26. Sensitivity and specificity in immunopathology and molecular identification as a folate-binding protein. Am. J. Pathol. 142, 557-567. Goldenberg, D. M., Goldenberg, H., Higginbotham-Ford, E., Shochat, D., and Ruoslahti, E. (1987) Imaging of primary and metastatic liver cancer with 131I monoclonal and polyclonal antibodies against alphafetoprotein. J. Clin. Oncol. 5, 18271835. Griffiths, G. L., Goldenberg, D. M., Jones, A. L., and Hansen, H. J. (1992) Radiolabeling of monoclonal antibodies and fragments with technetium and rhenium. Bioconjugate Chem. 3, 91-99. Halpern, S. E. (1986) The advantages and limits of indium-111 labeling of antibodies. Experimental studies and clinical applications. Appl. Radiat. Isot. 13, 195-201. Hnatowich, D. J., Layne, W. W., Childs, R. L., and Davis, M. A. (1983) Radioactive labeling of antibody: a simple and efficient method. Science 220, 613-615. Holm, J., Hansen, S. I., Hoier-Madsen, M., Sondergaard, K., and Bzorek, M. (1994) Folate receptor of human mammary adenocarcinoma. APMIS 102, 413-419. Kairemo, K. J., Hopsu, E. V., Melartin, E. J., and Heikkila, P. S. (1991) Imaging of a parapharyngeal hemangiopericytoma. Radioimmunoscintigraphy (SPECT) with indium-111-labeled anti-CEA antibody, and comparison to digital subtraction angiography, computed tomography, and immunohistochemistry. Cancer 67, 61-66. Kamen, B. A., and Capdevila, A. (1986) Receptor-mediated folate accumulation is regulated by the cellular folate content. Proc. Natl. Acad. Sci. U.S.A. 83, 5983-5987. Kane, M. A., and Waxman, S. (1989) Role of folate binding proteins in folate metabolism. Lab. Invest. 60, 737-746. Lamki, L. M. (1995) Tissue characterization in nuclear oncology: its time has come. J. Nucl. Med. 36, 207-210. Laufer, I., Keenan, A. M., and Dinsmore, B. (1990) Advances in the diagnosis and imaging of gastrointestinal cancer. Curr. Opin. Oncol. 2, 711-717. Leamon, C. P., and Low, P. S. (1991) Delivery of macromolecules into living cells: a method that exploits folate receptor endocytosis. Proc. Natl. Acad. Sci. U. S. A. 88, 5572-5576. Leamon, C. P., and Low, P. S. (1993) Membrane folate-binding proteins are responsible for folate-protein conjugate endocytosis into cultured cells. Biochem. J. 291, 855-860. Lee, R. J., and Low, P. S. (1994) Delivery of liposomes into cultured KB cells via folate receptor-mediated endocytosis. J. Biol. Chem. 269, 3198-3204. Lee, R. J., Wang, S., and Low, P. S. (1996) Measurement of Endosomal pH Following Folate Receptor-Mediated Endocytosis. Biochim. Biophys. Acta 1312, 237-242. Levy, C. C., and Goldman, P. (1967) The enzymatic hydrolysis of methotrexate and folic acid. J. Biol. Chem. 242, 2933-2938. Li, P. Y., Del Vecchio, S., Fonti, R., Carriero, M. V., Potena, M. I., Botti, G., Miotti, S., Lastoria, S., Menard, S., Colnaghi, M. I., and Salvatore, M. (1996) Local concentration of folate binding protein GP38 in sections of human ovarian carcinoma by in vitro quantitative autoradiography. J. Nucl. Med. 37, 665-672. Luo, J., Smith, M., Lantrip, D. A., Wang, S., and Fuchs, P. L. (1997) Efficient syntheses of pyrofolic acid and pteroylazide, reagents for the production of carboxyl differentiated derivatives of folic acid. J. Am. Chem. Soc. (submitted for publication). Mathias, C. J., Wang, S., Lee, R. J., Waters, D. J., Low, P. S., and Green, M. A. (1996a) Tumor selective radiopharmaceutical targeting via receptor-mediated endocytosis: evaluation of a gallium-67 labeled deferoxamine-folate conjugate. J. Nucl. Med. 37, 1003-1008.
[111In]DTPA−Folate as a Tumor-Imaging Agent Mathias, C. J., Wang, S., Waters, D. J., Low, P. S., and Green, M. A. (1996b). Ga-67 and In-111 labeled folate-chelate conjugates for targeting tumor-associated folate binding protein. J. Nucl. Med. 37, 347P-348P (abstract). Matsue, H., Rothberg, K. G., Takashima, A., Kamen, B. A., Anderson, R. G. W., and Lacey, S. W. (1992) Folate receptor allows cells to grow in low concentrations of 5-methyltetrahydrofolate. Proc. Natl. Acad. Sci. U.S.A. 89, 6006-6009. McHugh, M., and Cheng, Y. C. (1979) Demonstration of a high affinity folate binder in human cell membranes and its characterization in cultured human KB cells. J. Biol. Chem. 254, 11312-11318. Rettig, W., Garin-Chesa, P., Beresford, H., Oettgen, H., Melamed, M., and Old, L. (1988) Cell-surface glycoproteins of human sarcomas: differential expression in normal and malignant tissues and cultured cells. Proc. Natl. Acad. Sci. U.S.A. 85, 3110-3114. Ross, J. F., Chaudhuri, P. K., and Ratnam, M. (1994) Differential regulation of folate receptor isoforms in normal and malignant tissues in vivo and in established cell lines. Physiologic and clinical implications. Cancer 73, 2432-2443. Selhub, J., and Franklin, W. A. (1984) The folate-binding protein of rat kidney. J. Biol. Chem. 259, 6601-6606. Selhub, J., Emmanouel, D., Stavropoulos, T., and Arnold, R. (1987a) Renal folate absorption and the kidney folate binding protein. I. Urinary clearance studies. Am. J. Physiol. 252, F750-F756. Selhub, J. Nakamura, S., and Carone, F. A. (1987b) Renal folate absorption and the kidney folate binding protein. II. Microinfusion studies. Am. J. Physiol. 252, F757-F760.
Bioconjugate Chem., Vol. 8, No. 5, 1997 679 Serafini, A. N., Ed. (1993) Monoclonal antibody imaging: crossing the research/clinical barrier. Seminars in Nuclear Medicine, Vol. 23, pp 87-179, Saunders, Orlando, FL. Stein, R., Goldenberg, D. M., and Mattes, J. (1991) Normal tissue reactivity of four anti-tumor monoclonal antibodies of clinical interest. Int. J. Cancer 47, 163-169. Tsang, B. W., Mathias, C. J., Fanwick, P. E., and Green, M. A. (1994) Structure-distribution relationships for metal-labeled myocardial imaging agents: comparison of a series of cationic gallium(III) complexes with hexadentate bis(salicylaldimine) ligands. J. Med. Chem. 37, 4400-4406. Wang, S., Lee, R. J., Cauchon, G., Gorenstein, D. G., and Low, P. S. (1995) Delivery of antisense oligonucleotides against the human epidermal growth factor receptor into cultured KB cells with liposomes conjugated to folate via polyethyleneglycol. Proc. Natl. Acad. Sci. U.S.A. 92, 3318-3322. Wang, S., Lee, R. J., Mathias, C. J., Green, M. A., and Low, P. S. (1996) Synthesis, purification, and tumor cell uptake of 67Ga-deferoxamine-folate, a potential radiopharmaceutical for tumor imaging. Bioconjugate Chem. 7, 56-62. Weitman, S. D., Lark, R. H., Coney, L. R., Fort, D. W., Frasca, V., Zurawski, V. R., Jr., and Kamen, B. A. (1992) Distribution of the folate receptor (GP38) in normal and malignant cell lines and tissues. Cancer Res. 52, 3396-3401. Wilbur, D. S. (1992) Radiohalogenation of proteins: an overview of radionuclides, labeling methods, and reagents for conjugate labeling. Bioconjugate Chem. 3, 433-470.
BC9701297
