Star Structure of Antibody-Targeted HPMA Copolymer-Bound

Institute of Microbiology, Academy of Sciences of the Czech Republic, Vıdenská ... star structure with those of previously described classic antibod...
0 downloads 0 Views 167KB Size
206

Bioconjugate Chem. 2002, 13, 206−215

Star Structure of Antibody-Targeted HPMA Copolymer-Bound Doxorubicin: A Novel Type of Polymeric Conjugate for Targeted Drug Delivery with Potent Antitumor Effect Marek Kova´rˇ,† Jirˇ´ı Strohalm,‡ Toma´sˇ Etrych,‡ Karel Ulbrich,‡ and Blanka R ˇ ´ıhova´*,† Institute of Microbiology, Academy of Sciences of the Czech Republic, Vı´denˇska´ 1083, 142 20 Prague 4, Czech Republic, and Institute of Macromolecular Chemistry, Academy of Sciences of the Czech Republic, Heyrovsky Square 2, 162 06 Prague 6, Czech Republic. Received June 29, 2001; Revised Manuscript Received December 14, 2001

The aim of this study was to compare the properties and antitumor potential of a novel type of antibodytargeted N-(2-hydroxypropyl)methacrylamide (HPMA) copolymer-bound doxorubicin conjugates with star structure with those of previously described classic antibody-targeted or lectin-targeted HPMA copolymer-bound doxorubicin conjugates. Classic antibody-targeted conjugates were prepared by aminolytic reaction of the multivalent HPMA copolymer containing side-chains ending in 4-nitrophenyl ester (ONp) reactive groups with primary NH2 groups of the antibodies. The star structure of antibodytargeted conjugates was prepared using semitelechelic HPMA copolymer chains containing only one reactive N-hydroxysuccinimide group at the end of the backbone chain. In both types of conjugates, B1 monoclonal antibody (mAb) was used as a targeting moiety. B1 mAb recognizes the idiotype of surface IgM on BCL1 cells. The star structure of the targeted conjugate had a narrower molecular mass distribution than the classic structure. The peak in the star structure was around 300-350 kDa, while the classic structure conjugate had a peak around 1300 kDa. Doxorubicin was bound to the HPMA copolymer via Gly-Phe(D,L)-Leu-Gly spacer to ensure the controlled intracellular delivery. The release of doxorubicin from polymer conjugates incubated in the presence of cathepsin B was almost twice faster from the star structure of targeted conjugate than from the classic one. The star structure of the targeted conjugate showed a lower binding activity to BCL1 cells in vitro, but the cytostatic activity measured by [3H]thymidine incorporation was three times higher than that seen with the classic conjugate. Cytostatic activity of nontargeted and anti-Thy 1.2 mAb (irrelevant mAb) modified HPMA copolymer-bound doxorubicin was more than hundred times lower as compared to the star structure of B1 mAb targeted conjugate. In vivo, both types of conjugates targeted with B1 mAb bound to BCL1 cells in the spleen with approximately the same intensity. The classic structure of the targeted conjugate bound to BCL1 cells in the blood with a slightly higher intensity than the star structure. Both types of targeted conjugates had a much stronger antitumor effect than nontargeted HPMA copolymer-bound doxorubicin and free doxorubicin. The star structure of targeted conjugate had a remarkably higher antitumor effect than the classic structure: a single intravenous dose of 100 µg of doxorubicin given on day 11 completely cured five out of nine experimental animals whereas the classic structure of targeted conjugate given in the same schedule only prolonged the survival of experimental mice to 138% of control mice. These results show that the star structure of antibody-targeted HPMA copolymer-bound doxorubicin is a suitable conjugate for targeted drug delivery with better characterization, higher cytostatic activity in vitro, and stronger antitumor potential in vivo than classic conjugates.

INTRODUCTION

Cytostatic drugs that are used for conventional chemotherapy of cancer do not show any specific accumulation within the tumor but, due to the higher sensitivity of rapidly dividing cancer cells to these drugs, some preferential killing is achieved. Unfortunately, also tissues containing highly proliferating cells such as bone marrow and epithelia are very often seriously damaged by conventional chemotheraphy, and these toxic sideeffects are limiting factors for the chemotherapy dosage. It has been reported that conjugation of the free drug to * To whom the correspondence should be addressed. Phone: +420 2 475 23 43. Fax: +420 2 472 11 43. E-mail: rihova@biomed. cas.cz. † Institute of Microbiology. ‡ Institute of Macromolecular Chemistry.

a polymeric carrier reduces its nonspecific toxicity (1, 2) and markedly prolongs its blood half-life (3). A conjugate of a synthetic, water-soluble and biocompatible copolymer based on N-(2-hydroxypropyl)methacrylamide (HPMA) and containing doxorubicin bound via the Gly-Phe(D,L)Leu-Gly spacer, called PK1, is undergoing phase I/II clinical study in UK (4), and a conjugate containing doxorubicin and IgG was successfully used in the Czech Republic with a human patient with generalized breast carcinoma (5). The idea of increasing the therapeutic effect and minimizing the side-toxicity of common anticancer drugs is very attractive. This could be achieved by site-specific drug delivery using a targeting moiety specifically recognizing tumor cells. Targeted HPMA copolymer-bound drugs have previously been shown to have antitumor effect in vitro and in vivo (6). These conjugates consist

10.1021/bc010063m CCC: $22.00 © 2002 American Chemical Society Published on Web 02/12/2002

New Type of Polymeric Conjugate for Drug Targeting

of a HPMA copolymer backbone to which the drug and the targeting moiety are conjugated via a biodegradable spacer Gly-Phe(D,L)-Leu-Gly. This spacer ensures a controlled intracellular release of the bound drug. The targeting moieties to be used can be monoclonal or polyclonal antibodies (7), saccharides (8), lectins (9), and hormones (10). The risk of arising of antibodies directed against the targeting moiety after repeated administration, as commonly seen with immunotoxins, is lowered thanks to the decreased immunogenicity of proteins bound to HPMA copolymer (11). Decisive in antibody-targeted drug delivery is the selection of appropriate surface receptor for targeting. In many experimental and preclinical studies, receptors such as carcinoembryonic antigen (12), R-fetoprotein (13), CD5 (14), CD22 (15), and many others have been explored. In this study, we have used as a tumor model system murine BCL1 leukemia. It is B-cell derived malignancy and BCL1 cells thus possess surface IgM. We have used the B1 monoclonal antibody (mAb) recognizing the idiotype of surface IgM on BCL1 cells as targeting moiety. Since the idiotype of each clone of B-cells is a unique structure, we have used a strictly tumor-specific marker for drug targeting. Moreover, the surface IgM on B-cells has been shown as a receptor with a high level of expression and rapid internalization after ligand binding (16, 17). Antibody-targeted or lectin-targeted HPMA copolymerbound doxorubicin conjugates were shown to exhibit binding to target cells, cytostatic activity in vitro, and antitumor effect in vivo (18). These conjugates were prepared by an aminolytic reaction of the multivalent HPMA copolymer containing ONp reactive groups with antibody molecules (the IgG molecule contains more than 70 amino groups of lysine residues). Since there are many reactive groups distributed along the both HPMA copolymer and antibody molecule, two or more molecules of the antibody can be joined together to form a high-molecular weight branched structure (classic structure, Figure 1A). Recently, we have introduced a novel type of antibodytargeted HPMA copolymer-bound doxorubicin of a star structure. In this type of conjugate, HPMA copolymer chains containing only one reactive N-hydroxysuccinimide group at the end of the molecule react with the amino groups of the only one molecule of antibody (Figure 1B). This means that the star structure of HPMA copolymerbound doxorubicin targeted with antibody is a more homogeneous and better characterized conjugate than the classic structure. The purpose of this study was (a) to synthesize and characterize a star conjugate and a classic conjugate of HPMA copolymer-bound doxorubicin targeted with B1 mAb, (b) to compare the binding activity of free B1 mAb, B1mAb in the star conjugate and in classic conjugate, (c) in vitro evaluate cytostatic activity of both types of HPMA copolymer-bound doxorubicin targeted with B1 mAb and compare it to the cytostatic activity of nontargeted and anti-Thy 1.2 mAb containing (irrelevant antibody) conjugate using tumor cell line BCL1, and (d) to determine the binding capacity to target cells and antitumor potential of star and classic structure of HPMA copolymer-bound doxorubicin targeted with B1 mAb in vivo. MATERIAL AND METHODS

Chemicals. 1-Aminopropan-2-ol, glutathione (GSH), ethylenediaminetetraacetic acid (EDTA), 3-sulfanylpropionic acid (MPA), methacryloyl chloride, glycyl-D,L-

Bioconjugate Chem., Vol. 13, No. 2, 2002 207

Figure 1. Schematic drawing of the classic structure of antibody-targeted HPMA copolymer-bound doxorubicin conjugate (A) and the star structure of antibody-targeted HPMA copolymer-bound doxorubicin conjugate (B).

phenylalanine, L-leucyl-glycine, 4-nitrophenol, N-hydroxysuccinimide, 1-hydroxybenzotriazole, triethylamine (TEA), N,N′-dicyclohexylcarbodiimide (DCCI), 2,2′-azobisisobutyronitrile (AIBN), phthalaldehyde (OPA), 5-thio-Dglucose, dimethyl sulfoxide (DMSO), dimethyl formamide (DMF), and doxorubicin hydrochloride (Dox‚HCl) were from Fluka AG (Buchs, Switzerland). All other chemicals and solvents were of analytical grade. The solvents were dried and purified by conventional procedures and distilled before use. Synthesis of Monomers. All monomers were prepared as previously described. Briefly, N-(2-hydroxypropyl)methacrylamide (HPMA) was prepared by methacryloylation of 1-aminopropan-2-ol in methylene chloride (19), methacryloylglycyl-D,L-phenylalanyl-L-leucylglycine 4-nitrophenyl ester (Ma-Gly-Phe(D,L)-Leu-Gly-ONp) by methacryloylation of Gly-Phe-Leu-Gly tetrapeptide following its synthesis from Boc-GlyPheOH and H-LeuGlyOMe (20) (chromatographically pure product, mp ) 134136 °C, amino acid analysis: Gly:L-Phe:D-Phe:L-Leu ) 2.04:0.54:0.46:1.00), and methacryloylglycyl-DL-phenylalanyl-L-leucylglycyl doxorubicin amide (Ma-Gly-Phe(D,L)Leu-Gly-Dox) by the reaction of Ma-Gly-Phe(D,L)-LeuGly-ONp with Dox‚HCl in DMF in the presence of TEA (19) (TLC in chloroform:methanol 8:1 mixture showed two spots at Rf ) 0.46 and Rf ) 0.4 corresponding to the D-Phe and L-Phe isomers; HPLC showed a single peak at 18.22 min (UV detection at 230 and 484 nm); amino

208 Bioconjugate Chem., Vol. 13, No. 2, 2002

acid analysis Gly:L-Phe:D-Phe:L-Leu ) 2.04:0.54:0.46: 1.00). Synthesis of Polymer Precursors and PolymerDoxorubicin Conjugates. Polymer-bound doxorubicin (conjugate 1), classic and star polymer conjugates of doxorubicin with B1mAb antibody (conjugate 2 and conjugate 3), and the classic polymer conjugate of doxorubicin with anti-Thy1,2 antibody (conjugate 4) were prepared from polymer precursors as described (19). Briefly, polymer-Dox conjugate 1 was prepared from the polymer precursor I by its reaction with Dox‚HCl carried out in the presence of TEA in DMSO. Classic polymer conjugates 2 and 4 were prepared by the reaction of the polymer precursor II with Dox‚HCl in the presence of TEA in DMSO, followed by the reaction of such intermediate with the respective antibody in saline solution at controlled pH (7.6 and 8) using pH stat Radiometer Copenhagen. Star conjugate 3 was prepared from the semitelechelic, succinimide group-containing polymer precursor III, by its reaction with the anti-Thy1,2 antibody in phosphate buffer at pH 7.3-7.8. Polymer precursors I and II were prepared by radical precipitation copolymerization of HPMA with Ma-Gly-Phe(D,L)-LeuGly-ONp in acetone (initiator AIBN, 0.6 wt %; concentration of monomers 12.5 wt %; molar ratio HPMA: MaGly-Phe(D,L)-Leu-Gly-ONp 5:1 (precursor I) or 10:1 (precursor II); 60 °C; 24 h) as described (20, 21). Semitelechelic polymer precursor III was prepared by radical precipitation copolymerization of HPMA with Ma-GlyDL-Phe-L-Leu-Gly-Dox carried out in acetone and initiated with AIBN in the presence of 3-sulfanylpropionic acid as a chain-transfer agent. Copolymerization was followed by the reaction of the terminal COOH groupcontaining copolymer with N-hydroxysuccinimide in DMF using DCCI method (19). Characterization of Polymer Precursors and Polymer Conjugates. Polymer precursors I and II were characterized by UV spectrophotometry (the content of oligopeptide side chains terminated in 4-nitrophenoxy groups (ONp)) (20), by amino acid analysis (LDC Analytical, Riviera Beach, FL, precolumn OPA derivatization) and by liquid chromatography (weight- and numberaverage molecular weights, Mw and Mn) after aminolysis of reactive groups with 1-aminopropan-2-ol. PolymerDox conjugates were characterized and tested for the content of free polymer, free drug, or protein by GPC (Sepharose 4B/6B) or FPLC Pharmacia equipped with Superose 6 column (RI and UV detectors) and by electrophoresis (Pharmacia-LKB Phast System, SDS PAGE, gels with gradient 4-15). Molecular weights of all final conjugates and polymer precursors were determined using liquid chromatograph A ¨ KTA-Pharmacia equipped with Superose 6 or TSK 5000 PW columns and RI (refractive index) and multiangle light scattering DAWNDSP-F (Wyatt Technology Corp.) detectors. The protein content in the conjugates was estimated by amino acid analysis (LDC Analytical, Riviera Beach, FL, precolumn OPA derivatization) and Dox content by UV spectrophotometry (22). Neither GPC nor electrophoresis showed significant amounts of free protein in the conjugates and the content of free Dox in the conjugate was less than 0.05% of the total Dox content. Release of Dox from the Conjugates. Release of Dox from polymer conjugates was tested in phosphate buffer pH 6 after their incubation in cathepsin B solution at 37 °C. Incubation medium was 0.1 M sodium phosphate, 0.05 M NaCl, 1 mM EDTA, 5 mM reduced glutathione, and 0.5 µM cathepsin B. The activity of the cysteine proteinase cathepsin B was determined using

Kova´ˇr et al.

Bz-Arg-NAp as a substrate (20). In all cases, final Dox concentration was 0.5 mM. At predetermined time intervals the samples were tested for the content of free Dox. The amount of Dox released from the polymer conjugate was determined after its extraction from the incubation media into chloroform. A mixture of 0.1 mL of sample solution and 0.3 mL of buffer (0.2 M Na2CO3/ NaHCO3, pH 9.8) was extracted with 0.6 mL of chloroform. HPLC (Tessek gel C18 column, methanol-water gradient 10-90% methanol, fluorescent detector with excitation at 488 nm and emission at 560 nm) was used to determine the amount of free Dox. The calibration (yield of extraction) was carried out using Dox‚HCl. All drug-release data are expressed as the amount of free Dox relative to the total Dox content in the conjugate. Monoclonal Antibody (mAb). Monoclonal antibody directed against the idiotype of surface IgM of BCL1 cells, clone B1, is of the IgG1 isotype (23). For these studies, B1 mAb was purified from Balb/c mouse ascitic fluid by a combination of 45% saturated ammonium sulfate precipitation and protein A affinity chromatography. The concentration was determined by spectrophotometry (λ ) 280 nm) and the purity was checked by SDS-PAGE. Tumor Lines and Tissue Culture Condition. The murine B cell leukemia BCL1 was purchased from the American Type Culture Collection (ATCC, Rockville, MD). It is a well-described cell line that grows rapidly and consistently in immunocompetent Balb/c mice. Routine culture and all in vitro experiments with BCL1 cell line were conducted in RPMI 1640 medium with extra L-glutamine (300 µg/mL), sodium pyruvate (110 µg/mL), 2-mercaptoethanol (5 × 10-5M), HEPES (10mM), penicillin (100 units/mL), streptomycin (100 µg/mL), and 10% v/v 0.1 µm filtered fetal bovine serum. Cells in exponential growth with viability greater than 95% were used for all experiments. Cultures were maintained at 37 °C in 5% CO2. The murine B cell lymphoma 38C13 is a cell line with characteristics similar to the BCL1 cell line, and both lines express membrane IgM but do not secrete it. For this reason we decided to use the 38C13 cell line as control cells for in vitro binding studies. Routine culture and all experiments with 38C13 cell line were conducted in the same medium and under the same conditions as with BCL1 cell line. Mice. Mice of inbred strain Balb/c were obtained from the breeding colony of the Institute of Physiology (Academy of Sciences of the Czech Republic) and kept in standard housing. Animals 9-15 weeks of age and in the 19-22 g weight range were used in in vivo experiments. All experiments were approved by the Animal Welfare Committee of the Institute of Microbiology Academy of Sciences of the Czech Republic. Biotinylation of Free mAb and Targeted Conjugates Containing mAb. B1 mAb was biotinylated using EZ-Link Sulfo-NHS-LC-Biotin (Pierce). Briefly, 2 mg of B1 mAb in 1 mL of PBS (pH 8.0) was mixed with 100 µL of 1.5 mg/mL Sulfo-NHS-LC-Biotin solution. After a 45min incubation at room temperature, unreacted biotin was removed by dialysis. The degree of biotinylation was determined using HABA (Pierce), and about 4 biotins per molecule of IgG were found. Conjugates containing B1 mAb as a targeting moiety were biotinylated under the same conditions as free B1 mAb, but the degree of biotinylation was not determined because of the presence of doxorubicin. Fluorescence-ActivatedCellSorterAnalysis(FACS). For in vitro binding studies, BCL1 cells were collected from an exponential growth phase culture. They were centrifuged for 5 min at 300 g, washed twice with PBS

New Type of Polymeric Conjugate for Drug Targeting

with 1% of calf albumin and 0.1% of sodium azide (FACS solution), resuspended in FACS solution, and counted in trypan blue. Cell density was adjusted to 1 × 107 cells/ mL and 0.1 mL of this cell suspension was added to each test tube. From this moment on, all test tubes with cells were maintained on ice (0 °C), and all solutions added were adjusted to this temperature. Cells were again centrifuged for 5 min at 300 g and 100 µL of tested samples in different concentrations was then added. The cells were incubated for 45 min and then washed three times with FACS solution. Streptavidine-FITC conjugate at appropriate dilution was added for 30 min. Finally, the cells were washed three times and then resuspended in 0.5 mL of FACS solution. For in vivo binding studies, Balb/c mice 30 days after intraperitoneal administration of 5 × 105 BCL1 cells were intravenously injected with tested samples (250 µg of B1 mAb). At 45 min after the administration the mice were sacrificed, and spleen and 100 µL of blood from each mouse were placed in precooled test tubes containing 5 IU of heparine. Spleen was homogenized, and both the homogenate and blood were transferred to lysis solution to eliminate erythrocytes. After lysis, splenocytes and peripheral blood leucocytes were washed twice with FACS solution, centrifuged for 5 min at 300 g and streptavidin-FITC conjugate at appropriate dilution was added for 30 min. Finally, the cells were washed three times and resuspended in 0.5 mL of FACS solution. At least 7500 live cells were analyzed for each measurement. Proliferation Assay in Vitro. BCL1 cells were collected from an exponential growth phase culture. They were centrifuged for 5 min at 300 g, washed, resuspended in fresh culture medium, and counted in trypan blue. Cell density was adjusted to 5 × 104 cells/mL, and 0.2 mL of this cell suspension was loaded by multichannel pipet into each well of Nunc 96-well flat-bottomed plates. The tested samples were then added to the wells to achieve the desired concentrations; the final well volume was 0.25 mL and cell density 1 × 104 cells/well, with five wells for each test condition. The plates were cultured in 5% CO2 for 72 h at 37 °C, after which 18.5 kBq of [3H]thymidine in 50 µL was added per well, followed by another 6 h incubation. The cells were then collected using a cell harvester (Tomtec, Orange, CT) onto glass fiber filters (filtermat, Wallac, Finland). When the mixture was dried, a sheet of solid scintillator-Meltilex (Wallac) was placed together with a filtermat containing 96 samples in a sample bag and run together through a heat sealer (Microsealer, Wallac). Counting was performed in 1450 MicroBeta TriLux (Wallac). Treatment of Established BCL1 Leukemia in Vivo. BCL1 cells were collected from an exponential growth phase culture. They were centrifuged for 5 min at 300 g, washed, resuspended in fresh culture medium, and counted in trypan blue. An amount of 5 × 105 BCL1 cells in 0.5 mL culture medium were inoculated intraperitoneally on day 0 to Balb/c mice. Treatment with targeted HPMA copolymer-bound doxorubicin was given intravenously either as a single or as three injections and always began on the 11th day. The second injection was given on the 14th day and the third injection on the 17th day. One dose contained either 50 µg of free doxorubicin or 100 µg of HPMA copolymer-bound doxorubicin. Control mice were injected with only physiological solution. Nine or 10 mice were used for each experimental group. Statistical Analysis. The Wilcoxon’s statistics (multiple comparison method) and Student’s t-test were used to evaluate the differences among experimental groups. P values