Synthesis, Cellular Transport, and Activity of ... - ACS Publications

We compare two schemes to obtain methylprednisolone (MP)−polyamidoamine dendrimer (PAMAM-G4-OH) conjugate. Glutaric acid (GA) was used as a spacer ...
0 downloads 0 Views 303KB Size
Download PDF
Bioconjugate Chem. 2005, 16, 330−337

330

Synthesis, Cellular Transport, and Activity of Polyamidoamine Dendrimer-Methylprednisolone Conjugates Jayant Khandare,†,‡ Parag Kolhe,†,§ Omathanu Pillai,‡ Sujatha Kannan,‡ Mary Lieh-Lai,‡ and Rangaramanujam M. Kannan*,† Department of Chemical Engineering and Material Science, and Biomedical Engineering, Wayne State University, Detroit, Michigan 48202, and Department of Pediatrics (Critical Care Division), Children’s Hospital of Michigan, Wayne State University, Detroit, Michigan 48201. Received August 19, 2004; Revised Manuscript Received January 26, 2005

Dendrimers have emerged as promising multifunctional nanomaterials for drug delivery due to their well-defined size and tailorability. We compare two schemes to obtain methylprednisolone (MP)polyamidoamine dendrimer (PAMAM-G4-OH) conjugate. Glutaric acid (GA) was used as a spacer to facilitate the conjugation. In scheme A, PAMAM-G4-OH was first coupled to GA and then further conjugated with MP to obtain PAMAM-G4-GA-MP conjugates. This scheme yields a lower conjugation ratio of MP, presumably because of lower reactivity and steric hindrance for the steroid at the crowded dendrimer periphery. In scheme B, this steric hindrance was overcome by first preparing the MPGA conjugate, which was then coupled to the PAMAM-G4-OH dendrimer. The 1H NMR spectrum of the conjugate from scheme B indicates a conjugation of 12 molecules of MP with the dendrimer, corresponding to a payload of 32 wt %. In addition, conjugates were further fluorescent-labeled with fluoroisothiocynate (FITC) to evaluate the dynamics of cellular entry. Flow cytometry and UV/visible spectroscopic analysis showed that the conjugate is rapidly taken up inside the cell. Fluorescence and confocal microscopy images on A549 human lung epithelial carcinoma cells treated with conjugates show that the conjugate is mostly localized in cytosol. MP-GA-dendrimer conjugate showed comparable pharmacological activity to free MP, as measured by inhibition of prostaglandin secretion. These conjugates can potentially be further conjugated with a targeting moiety to deliver the drugs to specific cells in vivo.

INTRODUCTION

Dendrimers represent a relatively new class of welldefined monodisperse macromolecules with striking features such as controlled structure, globular shape, and a high density of “tunable” surface functional groups in their periphery (1-5). A variety of dendrimers have been synthesized and tailored for diverse applications (6). Encapsulation of drugs in PEGylated dendrimers led to increased circulation time (7, 8). Various strategies have been devised to load dendrimers with drug molecules, genetic materials, targeting agents, dyes, and imaging agents either by encapsulation, complexation, or conjugation (9-15). Although investigation of cellular delivery of drugs using dendrimers is at an early stage, these nanomaterials offer several advantages when compared to liposomes and microparticles. Their multiple, surface functional groups can be easily modified to potentially attach a large number of drug molecules, ligands, and antibodies, making them ideal vehicles for targeted drug delivery. They have been shown to be rapidly internalized into cells. Thus, they are promising nanovehicles in several therapeutic areas, where the target is intracellular (16-21). The present investigation demonstrates * To whom correspondence should be addressed. Phone: 313-577-3879, Fax: 313-577-3810; E-mail: rkannan@ chem1.eng.wayne.edu. † Department of Chemical Engineering and Material Science, and Biomedical Engineering. ‡ Department of Pediatrics (Critical Care Division). § Currently at Schering-Plough.

the potential of dendrimers as promising vehicles for delivering steroids to lung epithelial cells. A critical issue in dendritic drug delivery systems is achieving a high drug payload. However, the payload that each polymer can carry typically depends on the total number of reactive groups. For example, a linear PEG molecule with two reactive end groups can be conjugated to two drug molecules, whereas, for the same molecular weight, hyperbranched polymers (HBPs) such as dendrimers with multiple reactive end groups can be conjugated to several drug molecules (17). Genes, targeting agents, and imaging agents have been linked with dendrimers to increase therapeutic efficacy. In most cases, achieving adequate drug loading has been a challenge. It may be possible that some of the dendrimer-drug conjugates show less therapeutic activity compared to free drug, due to relatively low drug payload and the lack of drug release from conjugate. This is especially true for the anticancer agents such as doxorubicin and methotrexate (18). Despite the fact that dendrimers possess a high density of surface functional groups for chemical modification, the nanoscale architecture of a dendrimer (∼10 nm) causes steric hindrance for covalent conjugation of drugs in general, and large steroid molecules in particular. This results in typical drug payloads of less than 10% for high molecular weight drugs (1). Achieving high drug payload for steroidal drugs (Mw ∼ 374.5 Da) is challenging, since both the dendrimer and the steroid exhibit steric hindrance for covalent conjugation. Therefore, the main focus of this study was to (i) synthesize dendrimer-drug

10.1021/bc0498018 CCC: $30.25 © 2005 American Chemical Society Published on Web 02/26/2005

PAMAM Dendrimer−Methylprednisolone Conjugates

conjugates with relatively high steroidal drug payload, (ii) demonstrate the potential of such conjugates to deliver high drug concentrations inside the cells, and (iii) evaluate their pharmacological activity. We have previously investigated PAMAM-G4-NH2 dendrimer-ibuprofen complexes involving ionic interaction between surface amine groups of the dendrimer and carboxyl groups of ibuprofen. The complexes demonstrated rapid cell entry into A549 human lung epithelial carcinoma cell lines and showed antiinflammatory activity by suppressing the COX-2 gene in the in vitro studies (19). But, since complexation or encapsulation of drug is a weak ionic interaction, the concern arises that the drug may decouple in vivo before it reaches the target site. Unlike these complexes, covalently conjugated drugs will be more stable at physiological pH, making them more suitable for in vivo applications. We have earlier reported dendrimer-ibuprofen conjugates with high payload using PAMAM G4-OH-terminated dendrimer (20). The suitability of a dendrimer for delivering drug will depend on its molecular weight, generation number, end functionality, and charge. The polycationic systems show toxicity and higher elimination rate from the circulation. Anionic systems exhibit reduced uptake of molecules in cultured cells and are rapidly removed via the scavenger receptor in animal models (21). Conjugation of drug, targeting moiety, or imaging agent to the dendrimer also depends on the physicochemical properties of the agent to be coupled. This is especially true for a steroidal drug such as methylprednisolone. Covalent and noncovalent conjugations of drugs, genetic materials, and lipids to dendrimers have previously been reported (22). The conjugation of bulkier molecules such as steroids to dendrimers aims to address the issues of drug reactivity, stability, and most importantly the steric hindrance. To synthesize high payload drug-dendrimer conjugates using steroids, it is essential to overcome the steric hindrance and low reactivity of the drug molecule. The methodology reported here is an attempt to increase the payload of MP with dendrimer by increasing reactivity and at the same time minimizing steric hindrance. Methylprednisolone (MP) is a corticosteroid, 6R-methyl-11β,17R,21-trihydroxy-1,4-pregnadiene-3,20-dione. It decreases inflammation by stabilizing leukocyte lysosomal membrane and has a greater antiinflammatory potency than prednisolone. The corticosteroid has three OH groups at 11β, 17R, and 21. The hydroxyl group at C21 is a primary OH and therefore is most favorable for the covalent conjugation with terminal COOH groups of the dendrimer. Previously, MP has been conjugated with dextran by using succinate as a linker (23). The ester bond so formed was enzymatically hydrolyzed in rat blood. However, the rate of hydrolysis was very slow, corresponding to a half-life of ∼25 h of MP in dextranMP conjugates. In this paper, we evaluate synthetic approaches for obtaining MP-dendrimer (PAMAM G4-OH) conjugate with a high drug payload. We report conjugation of MP to PAMAM G2.5-COOH terminal dendrimer and PAMAM G4-OH terminal dendrimer using glutaric acid as a spacer. In addition, fluoroisothiocynate (FITC) was covalently linked with free MP and MP-dendrimer conjugate to investigate the cellular transport of free and dendrimer-conjugated MP. The dynamics of cell entry of the conjugate was studied using A549 human lung epithelial carcinoma cell line. Flow cytometry and UV/ visible spectroscopy were used to follow the time-dependent transport of FITC-labeled MP and dendrimer-MP into lung epithelial cells. Fluorescence and confocal

Bioconjugate Chem., Vol. 16, No. 2, 2005 331

microscopy were used to visualize the cellular localization of the conjugate. In vitro pharmacological activity of the conjugate was demonstrated by measuring the prostaglandin (PGE2) levels in cell supernatant. EXPERIMENTAL SECTION

Materials and Methods. PAMAM-G-2.5-COOH (Mw ∼ 6,267 Da, 32 end groups), PAMAM-G4-OH (Mw ∼ 14,279 Da, 64 end groups) dendrimers, methylprednisolone USP (Mw ∼ 374.5 Da), and dicyclohexylcarbodiimide (DCC) were purchased from Aldrich. Glutaric acid (Mw ∼ 132 Da) (Matheson Coleman and Bell Co.), fluorescent probe fluoroisothiocynate (FITC) (Fluka), Interleukin (IL2β) and lipopolysaccharide (LPS) (Sigma), ELISA kit for prostaglandin estimation (Cayman Chemical), and dialysis membrane of molecular weight cutoff of 3500 Da (Spectra Pore) were purchased. Dimethyl sulfoxide (DMSO), dimethyl formamide (DMF), methanol, and diethyl ether were purchased from Fischer Scientific. All other chemicals used were analytical grade. Synthesis of PAMAM-G2.5-COOH-MP Conjugates. PAMAM-G2.5-COOH (0.088 g, 0.014 mmol) and MP (0.168 g, 0.449 mmol) were dissolved in anhydrous dimethyl sulfoxide (DMSO). (On the basis of the molecular weights, this corresponds to a molar ratio of dendrimer to MP of 1:32, one drug molecule for each dendrimer end group). To this solution was added DCC (0.093 g, 0.45 mmol) as a coupling agent. The reaction was carried out at room temperature for 3 days and filtered to remove dicyclohexylurea (DCU). The solution was further dialyzed (dialysis membrane of molecular weight cutoff ) 3500 Da) against DMSO for 24 h to remove unreacted methylprednisolone. The contents from dialysis bag were dried under vacuum and characterized by 1H NMR (24). Conjugates Were Prepared Using Two Different Synthetic Strategies. (1) GA was first conjugated to PAMAM-G4-OH dendrimer, and then this conjugate was coupled with MP (scheme A). (2) GA was first conjugated to MP, and then the conjugate was coupled with PAMAM-G4-OH (scheme B). Scheme A. A.1: Synthesis of PAMAM-G4-OH-Glutaric Acid Conjugates. PAMAM-G4-OH (0.300 g, 0.021 mmol) and glutaric acid (0.180 g, 1.34 mmol) were dissolved in anhydrous dimethyl sulfoxide (DMSO). This corresponds to a mole ratio of dendrimer to GA of 1:64, with one GA molecule for every OH end group on the dendrimer. DCC (0.276 g, 1.33 mmol) was added as a coupling agent and stirred continuously for 2 days at room temperature. The resulting solution was filtered to remove DCU obtained as a byproduct during the reaction. The filtrate was dialyzed extensively with anhydrous DMSO (dialysis membrane of molecular weight cutoff ) 3500 Da) for 24 h to remove unreacted glutaric acid. The contents from the dialysis bag were dried under vacuum. Repurification was carried out using diethyl ether, and the conjugate was dried under vacuum. 1H NMR revealed that 50 molecules of GA are conjugated per dendrimer. A.2: Synthesis of PAMAM-G4-OH-Glutaric AcidMethylprednisolone Conjugates. PAMAM-G4-glutaric acid conjugate (0.12 g, 0.00574 mmol) and MP (0.1 g, 0.287 mmol) were dissolved in anhydrous dimethyl sulfoxide (DMSO). (The ratio of dendrimer-GA conjugates (50 molecules of GA per dendrimer) to MP was 1:50). To this solution was added DCC as a coupling agent, and the mixture was stirred continuously for 3 days. The DCU obtained was removed by filtration and was then dialyzed extensively with anhydrous DMSO

332 Bioconjugate Chem., Vol. 16, No. 2, 2005

Figure 1. Schematic for dendrimer-glutaric acid-methylprednisolone conjugate synthesis multiplets corresponding to ring structure of methylprednisolone.

(dialysis membrane of molecular weight cutoff ) 3500 Da) for 24 h to remove unreacted MP-glutaric acid and DCC. The contents from the dialysis bag were dried under vacuum to obtain the conjugate. 1H NMR was used to characterize the conjugate. The conjugation ratio was extremely low (one molecule of MP per dendrimer). Therefore, the synthetic approach was changed as described below. Scheme B. B.1: Synthesis of Glutaric Acid-MP Conjugates (Figure 1). MP (0.1 g, 0.265 mmol) and GA (0.036 g, 0.26 mmol) were dissolved in anhydrous DMSO. Dicyclohexylcarbodiimide (DCC) (0.056 g, 0.26 mmol) was added as a coupling agent to the above solution. The reaction was carried out for 24 h at room temperature and filtered to remove DCU obtained as a byproduct

Khandare et al.

during this reaction. The filtrate was evaporated under vacuum to obtain MP-glutarate conjugate. B.2: Conjugation of PAMAM-G4-OH Dendrimer to MP-Glutarate (Figure 1). PAMAM-G4-OH (0.032 g, 0.0024 mmol) and MP-glutaric acid conjugate (Mw ∼506) (0.077 g, 0.152 mM) were dissolved in anhydrous dimethyl sulfoxide (DMSO). (This corresponds to a mole ratio of dendrimer to MP-GA of 1:50). DCC (0.056 g, 0.26 mmol) was added as coupling agent, and the reaction was stirred for 3 days and filtered to remove DCU. The filtrate was dialyzed against anhydrous DMSO for 24 h (dialysis membrane of Mw cutoff ) 3500 Da) to remove unreacted MP-glutaric acid conjugate, free MP, free GA, and DCC. The solvent was removed from the contents of dialysis bag, and the conjugate was evaporated under vacuum. The conjugate was characterized using 1H NMR (Figure 2). Conjugation of FITC to MP and PAMAM-G4Dendrimer-GA-MP. MP (0.1 g, 0.265 mmol) and FITC (0.011 g, 0.28 mmol) or dendrimer-GA-MP (Mw ∼ 20 351 Da) (0.020 g, 0.00078 mmol) and FITC (0.01 g, 0.26 mmol) were dissolved in anhydrous DMSO (This corresponds-to-mole ratio of dendrimer to FITC 1:10). Dicyclohexylcarbodiimide (0.0054 g, 0.265 mmol) was added as a coupling agent for conjugation of FITC to MP and (0.00284 g, 0.014 mmol) for conjugation with dendrimer-GA-MP. The reaction was carried out for 24 h at room temperature and filtered to remove DCU. The FITC-dendrimer-GA-MP conjugate filtrate was dialyzed against anhydrous DMSO for 24 h (dialysis membrane of molecular weight cutoff ) 3500 Da) to remove unreacted FITC and DCC. In addition, free FITC was removed by using excess acetone. The absence of free FITC in the conjugates was verified by thin-layer chromatography (TLC), using chloroform and methanol (1:1) as solvents (25). The conjugates were dried under vacuum to obtain FITC-labeled conjugates. Cell Culture. Human lung carcinoma epithelial cell line A549 was obtained from Children’s Hospital of Michigan Cell Culture facility and used for the cell entry and activity studies. Cells were incubated in RPMI 1640 (Gibco-BRL) cell culture medium supplemented with 10% fetal calf serum (FCS) and 1% penicillin-streptomycin at 37 °C with 5% CO2 in a cell culture incubator. The cells were subcultured every 48 h and harvested from subconfluent cultures (60 to 70%) using 0.05% trypsin (Sigma-Aldrich). Cell Entry of Fluorescent-Tagged Conjugate. A549 cells were seeded at 2 × 105 cells/mL and cultured on 60 × 15 mm culture plates using RPMI 1640 cell culture medium supplemented with 10% fetal calf serum (FCS) and 1% penicillin. When the cells were 60% confluent, they were treated with FITC-labeled MP or FITC-labeled dendrimer-GA-MP conjugate. The supernatant was removed at times 0, 5, 10, 15, 20, 240, and 360 min, and the amount of FITC present in the supernatant was estimated by measuring the UV/visible absorbance at 490 nm for FITC-MP and 498 for FITC-MP-dendrimer conjugate and quantified with a calibration curve (of known concentrations of FITC-MP and FITC-labeled MP-dendrimer conjugates) using appropriate supernatant as blank solution. The pH of the cell supernatant remained unchanged throughout the experiment. After removing the cell supernatant at each time point, the cells were washed with phosphate-buffered saline (PBS), trypsinized, and centrifuged at 1500 rpm for 5 min to obtain a cell pellet. The cells were then rinsed with PBS buffer, spun down twice, resuspended in PBS, and subsequently analyzed using a flow cytometer (FACS

PAMAM Dendrimer−Methylprednisolone Conjugates

Figure 2.

1H

Bioconjugate Chem., Vol. 16, No. 2, 2005 333

NMR spectrum of dendrimer-glutaric acid-methylprednisolone conjugates.

caliber, Becton Dickinson) by counting 10 000 events. (The mean fluorescence intensity of the cells was calculated using the histogram plot). Fluorescence Microscopy. The procedure for cell culture and drug treatment was the same as described in previous section. After treatment with the conjugates for 2 h, the cells were washed with phosphate-buffered saline (pH 7.4). A few drops of the buffer was added before observing under the fluorescence microscope (Leica DM1L inverted microscope) using a magnification of 400×. Images were captured and stored using SensiCamQE 12 bit monochrome camera and Camware 3.1 software, respectively. Confocal Laser Scanning Microscopy. The procedure for cell culture and drug treatment was the same as described in previous section. After treating A549 cells for 2 h with FITC-labeled free dendrimer or MP conjugated dendrimer, the cells were washed with phosphatebuffered saline (pH 7.4). A few drops of the buffer and antifade reagent (Molecular Probes) was added before observing under the confocal microscope (Zeiss LSM 310) using a magnification of 63 × 1.2. The emission and excitation wavelengths were 488 and 518 nm for FITC. Pharmacological Activity of Dendrimer-MP Conjugates. A 549 lung epithelial cells (2 × 105 cells/mL/ well) was seeded in 24-well plates and allowed to grow overnight in RPMI medium supplemented with 10% FCS and 1% penicillin and streptomycin. When the cells were 60% confluent, the medium was removed and washed with serum-free medium (SFM). Each well was treated with 500 µL of SFM, and prostaglandin (PGE2) secretion was induced by addition of lipopolysaccharide (LPS) and interleukin (IL- β2) to each well. After 30 min, either 5 µg of free MP or dendrimer-GA-MP conjugate (5 µg equivalent of MP) in ethanol was added. Control treatments with solvent, polymer, and positive control with no treatment, and negative controls without PGE2 induction, were also studied. The supernatant was removed after 4 h and analyzed for PGE2 concentration using a commercial ELISA kit. Results were represented as percent inhibition of PGE2 compared to positive control. RESULTS AND DISCUSSION

1. Design and Synthesis of Dendrimer-Drug Conjugates. To obtain dendrimer-drug conjugates with high drug payload, it is important to (1) decrease the steric hindrance of the surface functional groups, which

are very close to each other. (This could be complicated by the potential back folding of end groups as explained below), and (2) increase the reactivity of the steroidal drug molecule and the dendrimer surface functional groups. Recently we have reported a high payload (on an average 53 molecules of drug per hyperbranched polymer) for the highly reactive and conformationally stable ibuprofen (20). The hydrodynamic radius (Rh) of generation 4-poly(glycerol succinic acid) (PGLSA-OH) dendrimers was estimated by quasielastic light scattering (QELS) method to be 7 nm (6). The authors reported the encapsulation of one Reichardt’s dye per dendrimer molecule. As a consequence, Rh of the dendrimer reduced from 7 to 4 nm, suggesting the collapse of aliphatic dendrimer around the dye. Therefore, the factors that affect drug loading in dendrimers include (1) potential collapse of aliphatic groups and lowering of Rh in organic solvents, and (2) reduced distance between the two peripheral reactive groups. In addition, the reactivity of dendrimers may not be significantly high for direct covalent conjugation with drugs, especially with conformationally unstable and bulkier steroidal drugs such as MP. Such drugs have low reactivity, higher molecular weight, and steric hindrance. The objective of present work was to obtain high payloads of MP with PAMAM dendrimers using DCC as a coupling agent. Therefore, we designed two synthetic approaches for conjugating MP with the dendrimer (schemes A and B). For comparison, the drug (MP) was conjugated directly to PAMAM G2.5-COOH dendrimer. The resulting conjugate had one drug molecule per dendrimer. In the first approach (scheme A), glutaric acid was incorporated as a flexible spacer, first by coupling it to OH end groups of PAMAM G4 dendrimer. This would present the highly reactive COOH group of the flexible GA spacer at the dendrimer surface, potentially relieving part of the crowding effect at the surface. The resulting dendrimer-glutarate conjugate was then linked to 21OH group of MP. The drug loading to dendrimer molecule was estimated using 1H proton integration method (24, 25). However, 1H NMR revealed the incorporation of only one MP molecule per mole of dendrimer. This suggests the presence of steric hindrance and low reactivity of the GA-modified dendrimer conjugate with MP. Therefore, the synthetic approach was slightly modified. In the second approach, the spacer molecule (GA) was first attached to MP, in an attempt to enhance the reactivity

334 Bioconjugate Chem., Vol. 16, No. 2, 2005

of MP (scheme B). The reaction of glutaric acid to MP (Figure 1) is more favorable to primary hydroxyl group at 21, compared to C9 (secondary) and C11 (tertiary) positions (23). MP was first linked to GA and further conjugated with OH groups of G4 dendrimer. 1H NMR spectrum (Figure 2) shows the multiplets between δ 2.0 to 4.0 ppm, which correspond to the presence of 985 protons of CH2 of PAMAM-G4-OH. Multiplets between δ 5.0 to 8.2 ppm correspond to four protons of the aromatic ring of methylprednisolone. The integration ratio of PAMAM-G4-OH to multiplets of methylprednisolone is 0.047, suggesting that 46.78 protons (0.047 × 985 ) 46.295) of methylprednisolone are present. Since four protons correspond to one molecule, on an average, 12 molecules of MP were incorporated per dendrimer molecule. The high payload of MP (32%) per dendrimer indicates that conjugation of a small spacer molecule such as glutaric acid increases the reactivity of MP molecule. In addition the MP-spacer conjugate overcomes the steric hindrance at the dendrimer surface for the large steroid. The resulting ester bond in the conjugate can potentially be hydrolyzed in the cell by lysosomal enzymes. To study the cellular transport of the conjugates, FITC was conjugated to free MP and MP-GA-dendrimer conjugate using DCC as a coupling agent. Unreacted FITC was removed by membrane dialysis using DMSO for 24 h. Any remaining trace amount of free FITC was removed by an additional repurification step using excess acetone. The absence of free FITC in the conjugate was verified by thin-layer chromatography (TLC), using chloroform and methanol (1:1) as solvents (25). The proton integration ratio of MP-GA-dendrimer conjugate was compared to the additional protons arising from FITC molecules in MP-GA-dendrimer-FITC conjugate. 1 H NMR reveals that on an average of six FITC molecules are conjugated to MP-GA-dendrimer conjugate (1H NMR not shown). 2. In Vitro Evaluation of Dendrimer-GA-MP Conjugates. When the conjugates were dissolved in DMSO and kept in a dialysis bag, no drug release was evident even after several hours, as verified by HPLC and UV/visible spectroscopy. To demonstrate the therapeutic potential of dendrimer-GA-MP conjugates, we evaluated these conjugates in A549 human lung epithelial carcinoma cell lines. The cells were treated with FITC-labeled-MP or dendrimer-GA-MP-FITC conjugate. Cell supernatant was analyzed for a period of 4 h by UV/visible spectroscopy, to estimate FITC concentration and the cellular uptake. As seen in Figure 3a,b, based on UV/visible analysis of the cell supernatant, ∼85% of MP-FITC and 35% of dendrimer-GA-MPFITC conjugate entered the cell within 1 h. The cellular entry of free and dendrimer-conjugated MP was further confirmed by flow cytometry by measuring the intracellular fluorescence intensity. By flow cytometry (Figure 3a,b), it was found that ∼50% of free MP and 65% of dendrimer-GA-MP was inside the cell. The discrepancy between the UV/visible spectroscopy measurements (for extracellular concentration) and flow cytometry (for intracellular concentration) can be attributed to the adsorption of drug molecules on the cell surface. When the extracellular concentration is measured by UV/visible spectroscopy, it does not take into account the molecules that are adsorbed onto the cell and those that are internalized. Since the cells were repeatedly washed with PBS (to remove any adsorbed molecules on the cell surface) after removing the cell supernatant, flow cytometry would be more indicative of measuring fluorescence

Khandare et al.

Figure 3. Decrease in FITC concentration in the cell supernatant (from UV/visible spectroscopy shown with open squares on the left-hand y-axis) and increase in the FITC intensity inside the cell (calculated from mean fluorescence intensity using flow cytometry and is shown with closed squares on the right-hand y-axis) for (a) MP-FITC and (b) dendrimer-GA-MP-FITC conjugate.

Figure 4. Fluorescence activated cell sorter analysis of the cell entry dynamics of methylprednisolone-glutaric acid-dendrimerFITC conjugate in A549 lung epithelial cell line shows rapid internalization of the conjugates within 5 min. The log of FITC absorption intensity (FL1-H on x-axis) is plotted against the number of cells (counts on y-axis). Key: violet, 0 min; green, 5 min; red, 15 min; light blue, 60 min; black, 240 min.

intensity of only those molecules that are internalized into the cell. This is further demonstrated in Figure 4, where rapid internalization of the MP-GA-dendrimer conjugate is noted within 5 min and the fluorescence intensity gradually increases over a period of 4 h. From the present findings, it appears that the measurement of intracellular fluorescence intensity using flow cytometry might be a better method to follow the transport of free drug or drug-polymer conjugates. At the end of 2 h, the cells were observed under fluorescence microscopy to identify the cellular localization of the free and drug conjugated dendrimer (Figure 5a-c). The cells without FITC showed no fluorescence

PAMAM Dendrimer−Methylprednisolone Conjugates

Figure 5. Fluorescence microscopic images (magnification 400×) of A549 cells 4 h of treatment with (a) control (no treatment), (b) FITC-labeled methylprednisolone, and (c) FITClabeled dendrimer-GA-methylprednisolone conjugate.

as expected (Figure 5a). Both free and drug conjugated dendrimer were mainly found to be localized in the cell cytoplasm (Figure 5b,c). Further, this was substantiated from confocal microscopic images in Figure 6b,c, where the fluorescence is seen mostly in the cytoplasm. This suggests that the free dendrimer and MP-dendrimer conjugate are mainly localized in the cytosol in the cell. Epithelial cells are known to possess anionic charge, and charged dendrimers such as PAMAM-NH2-terminated dendrimers can be transported across the cell by electrostatic interactions (14, 26). However, PAMAM-G4OH terminal dendrimers are not charged and hence may be transported by selective absorptive endocytosis mechanism. It appears that the free MP enters the cells faster than the MP-dendrimer conjugate. From the results, it appears that the efficiency of the endocytosis mechanism is influenced by the molecular size and further studies are required to understand the cellular uptake mechanism. Simple passive diffusion of drug molecules (which is the case with free MP) across the cell membrane is a less efficient means to deliver drugs into the cells, as the drug molecules can diffuse in and out of the cell depending on the concentration gradient. Unlike passive diffusion, the endocytotic uptake of cells (in case of MPdendrimer conjugate) is unidirectional; hence, the drug is not only more efficiently transported but is also retained at the site of action inside the cell. For MP, cytoplasmic delivery is required, since it exerts the pharmacological action by binding to the glucocorticoid receptors in the cytoplasmic compartment (27).

Bioconjugate Chem., Vol. 16, No. 2, 2005 335

Figure 6. Confocal microscopic images (magnification 63 × 1.2) of A549 cells after 2 h of treatment with (a) control (no treatment), (b) FITC-labeled dendrimer, and (c) FITC-labeled dendrimer-methylprednisolone conjugate. The conjugates appear to be mainly localized in the cytoplasm.

Antiinflammatory activity of free and dendrimerconjugated MP was evaluated in vitro using A549 cells, which were pretreated with lipopolysaccharide and interleukin (IL-2β) to induce prostaglandin synthesis. Neither the free dendrimer nor the solvent showed any PGE2 suppression. Dendrimer-GA-MP conjugate inhibited prostaglandin synthesis to the same extent (p > 0.05) as free MP (Figure 7) after 4 h of treatment. The results indicate that the conjugate was able to enter the cell and produce the desired pharmacological action. From these results, it appears that high drug payload in the dendrimer conjugate produces a high local drug concentration inside the cell to elicit a significant therapeutic response. At this point, it is unclear whether the MP is released from the dendrimer conjugate inside the cell or if the drug is effective even in the conjugated form. However, once the conjugate enters the cell, it is conceivable that the acidic pH and the enzymes in the endosomes would eventually hydrolyze the ester bond in the conjugate to release the free drug in the cytosol (28). Based on the previous report on dextran-MP conjugate, the hydrolysis process is likely to take much longer than the time frame used in current studies (23). Therefore, we may tentatively conclude that the conjugate may not have released the drug appreciably in 4 h. Nevertheless, the high payload drug conjugate show significant therapeutic activity. It may be possible that the conjugate may slowly release the drug over a sustained period providing therapeutic effect over longer times. Further studies are underway to investigate the stability of dendrimer-drug

336 Bioconjugate Chem., Vol. 16, No. 2, 2005

Khandare et al.

improve the stability and reactivity of MP. The conjugate entered the cell rapidly and achieved a high local drug concentration to elicit the desired pharmacological action. It appears that, at least over the time period of this study, the conjugates are localized primarily in the cytosol, as imaged through fluorescence and confocal microscopy. The conjugate showed comparable therapeutic activity to the free drug, even at short times, where the drug may not have released from the dendrimer. LITERATURE CITED

Figure 7. Box plot showing prostaglandin (PGE2) concentration in the cell supernatant for different treatments (n ) 4). Normal indicates the cells with no induction of prostaglandin synthesis. Control represents cells with prostaglandin induction with LPS+IL1β but with no drug treatment. MP and MPG indicate the cells, which were induced with LPS+IL1β to secrete prostaglandin and were treated for 4 h with free methylprednisolone and dendrimer-GA-MP conjugate, respectively. The cells treated with MP/MPG-suppressed prostaglandin secretion, and the PGE2 levels were comparable to normal cells. Induced cells treated with dendrimer and solvent alone were similar to that of control (not shown). This indicates that the pharmacological activity of the conjugated MP has remained similar to that of the free drug.

conjugate at various pHs and in the presence of enzymes to understand the intracellular hydrolysis of the conjugate. MP is used in the treatment of asthma, but systemic therapy is associated with severe side effects such as gastrointestinal hemorrhage, hypertension, endocrine, metabolic disturbances, immune suppression, and neuropsychiatric disturbances. (29). However, when both MP and a targeting ligand are conjugated to dendrimer, it may not only deliver the drug efficiently to the target site inside the cell, but also reduce the systemic side effects. By achieving a high local drug concentration at the target site, one could overcome the systemic adverse effects of free MP and improve the therapeutic efficacy significantly with a reduced dose. Current studies are ongoing to deliver the dendritic conjugate ‘directly’ to the lung through intranasal administration, in an in vivo lung inflammation model in mice (30). Furthermore, as the conjugates are expected to be stable in blood, it would achieve more sustained drug levels leading to reduce dosing frequency. Hence, the advantages of using dendrimers for delivery of MP would include (i) a high local concentration of MP in the lung through high drug payload (example, 12 MP molecules in a ∼6 nm diameter spherical dendrimer), (ii) superior residence time in the lung enabled by the dendrimer (unpublished data), and (iii) slow hydrolysis of the ester bond enabling sustained delivery inside the cell. CONCLUSIONS

A dendrimer-MP conjugate with a relatively high payload (12 steroid molecules per dendrimer) was prepared by first attaching glutaric acid to MP, then conjugating GA-MP to PAMAM-G4-OH dendrimers. This allowed us to (1) overcome the steric hindrance at the dendrimer surface for the large steroid molecule, and (2)

(1) Kono, K., Liu, M., Frechet, J. M. J. (1999) Design of Dendritic Macromolecules Containing Folate or Methotrexate Residues. Bioconjugate Chem. 10, 1115-1121. (2) Ihre, H. R., Padilla De Jesu´s, O. L., Szoka, F. C., Jr., Fre´chet, J. M. J. (2002) Polyester Dendritic Systems for Drug Delivery Applications: Design, Synthesis, and Characterization. Bioconjugate Chem. 13, 443-452. (3) Grainger, D. W., Okano, T. (2003) Biomedical Micro- and Nano-Technology. Adv. Drug Delivery. Rev. 55, 311-313. (4) Liu, M. J., Frechet, J.M. J. (1999) Designing Dendrimers for Drug Delivery. Pharm. Sci. Technol. Today 2, 393-401. (5) Newcome, G. R., Moorfileld, C. N., Keith, J. N., Baker, G. R., Escamilla, G. H. (1994) Chemistry Within a Unimolecular Micelle Precursor: Boron Superclusters by Site- and DepthSpecific Transformations of Dendrimers. Angew. Chem. Int. Ed. Engl. 33, 2413-2420. (6) Morgan, M. T., Carnahan, M. A., Immoos, C. E., Ribeiro, A. A., Finkelstein, S., Lee, S. J., Grinstaff, M. W. (2003) Dendritic Molecular Capsules for Hydrophobic Compounds. J. Am. Chem. Soc. 125, 15485-15489. (7) Liu, M. J., Kono, K., and Frechet, J. M. J. (2002) Water Soluble Dendrimer-Poly (ethylene glycol) Star Like Conjugates as Potential Drug Carriers. J. Controlled Release 65, 121-131. (8) Yang, H., Morris, J. J., and Lopina, S. T. (2004) Polyethylene Glycol-Polyamidoamine Dendritic Micelle as a Solubility Enhancer and the Effect of the Length of Polyethylene Glycol Arms on the Solubility of Pyrene in Water. J. Controlled Release 273, 148-154. (9) Patri, A. K, Majoros, I. J., and Baker, J. R., Jr. (2002) Dendritic Polymer Macromolecular Carriers for Drug Delivery. Curr. Opin. Chem. Biol. 6, 466-471. (10) Kojima, C., Kono, K., Maryama, K., and Takagishi, T. (2000) Synthesis of Polyamidoamine Dendrimers having Poly(ethylene glycol) Grafts and Their Ability to Encapsulate Anticancer Drugs. Bioconjugate. Chem. 11, 910-917. (11) Rmaileh, R. A., Attwood, D., and D’Emanuele, A. (2003) Dendrimers in Cancer Therapy. dds & s. 3, 65-70. (12) Twyman, L. J., Beezer, A. E., Esfand, R., Hardy, M. J., and Mitchell, J. C. (1999). The Synthesis of Water Soluble Dendrimers and Their Application as Possible Drug Delivery Systems. Tetrahedron Lett. 40, 1743-1746. (13) Tomalia, D. A., Baker, H., Dewald, J. R., Hall, M., Kallos, G., Martin, S., Roeck, J., and Ryder, J. P. Smith. (1985) A New Class of Polymers-Starburst-Dendritic Macromolecules. Polym. J. 17, 117-132. (14) Tajarobi, F., El-Sayed, M., Rege, B. D., Polli, J. E., and Ghandehari, H. (2001) Transport of Polyamidoamine Dendrimers across Madin-Darby Canine Kidney Cells. Int. J. Pharm. 215, 263-267. (15) Malik, N., Wiwattanapatapee, R., Klopsch, R., Lorenz, K., Frey, H., Weener, J. W., Meijer, E. W., Paulus, W., and Duncan, R. (2000) Dendrimers: Relationship Between Structure and Biocompatibility in vitro, and Preliminary Studies on the Biodistribution of I-Labeled Polyamidoamine Dendrimers in vivo J. Controlled Release 65, 133-148. (16) Nigavekar, S. S., Sung, L. Y., Llanes, M., El-Jawahri, A., Lawrence T. S., Becker, C. W., Balogh, L., and Khan, M. K. (2004) 3H Dendrimer Nanoparticle Organ/Tumor Distribution. Pharm. Res. 21, 476-483. (17) Henry, C. M. (2002) Drug Delivery. Chem. Eng. News 80 (34), 39-47.

Bioconjugate Chem., Vol. 16, No. 2, 2005 337

PAMAM Dendrimer−Methylprednisolone Conjugates (18) Quintana, E., Raczka, L., Piehler, I., Lee, A., Myc, I., Majoros, A. K., Patri, T., Thomas, J., Mule´, J., and Baker (2002) Design and function of a dendrimer-based therapeutic nanodevice targeted to tumor cells through the folate receptor. Pharm. Res. 19, 1310-1316. (19) Kolhe, P., Misra, E., Kannan, R. M., Kannan, S., and LiehLai, M. (2003) Drug complexation, in vitro release and cellular entry of dendrimers and hyperbranched polymers. Int. J. Pharm. 259, 143-148. (20) Kolhe, P., Khandare, J., Pillai, O, Kannan, R. M., Kannan, S., and Lieh-Lai, M. (2004). Hyperbranched Polymer Drug Conjugates With High Drug Payload for Enhanced Drug Delivery. Pharm Res. 21 (12), 2185-2195. (21) Meijer, D. K., and Molema, G. (1995) Targeting of Drugs to The Liver. Semin. Liver Dis. 15, 202-256. (22) Hong, S., Bielinska, A. U., Mecke, A., Keszler, B., Beals, J. L., Shi, X., Balogh, L., Orr, B. G., Baker, J. R., Jr., and Banaszak H., M. M. (2004) Interaction of Poly (amidoamine) Dendrimers with Supported Lipid Bilayers and Cells: Hole Formation and the Relation to Transport., Bioconjugate Chem. 15 (4), 774-782. (23) Mehvar, R., Dann R. O., and Hoganson D. A. (2000) Kinetics of Hydrolysis of Dextran-Methylprednisolone Succinate, a Macromolecular Prodrug of Methylprednisolone, In Rat Blood and Liver Lysosomes. J. Controlled Release 68, 5361.

(24) Wiwattanapatapee, R., Lomlim, L., and Saramunee, K. (2003) Dendrimers Conjugates for Colonic Delivery of 5-Aminosalicylic Acid. J. Controlled Release 88, 1-9. (25) Jevprasesphant, R., Penny, J., Attwood, D., McKeown, N. B., and Emanuele, A. D. (2003) Engineering of Dendrimer Surfaces to Enhance Transepithelial Transport and Reduce Cytotoxicity. Pharm. Res. 20, 1543-1550. (26) Kannan, S., Kolhe, P., Raykova, V., Gilbatec, M., Kannan, R. M., Leih-Lai, M., Basset, D. (2004) Dynamics of Cellular Entry and Drug Delivery by Dendritic Polymers into Human Lung Epithelial Carcinoma Cells J. Biomaterial Sci.: Polym. Ed. 15, 311-330. (27) Adcock, I. M., (2000) Molecular Mechanisms of Glucocorticosteroid Actions. Pulm. Pharmacol. Ther. 13, 115-126. (28) Mcgraw, T. E., Maxifield, F. R., Juliano, R. L., (1991) Targeted Drug Delivery, pp 11-41, Springer-Verlag , New York. (29) Physician’s Desk Reference (2002) pp 2486-2488, Medical Economics Company, Inc., Montvale, NJ. (30) DeLorme, M. P., Gao, X., Doyon R. N., Barraclough- M. B., and Bassett, D. (2002) Inflammatory Effects of Inhaled Endotoxin-Contaminated Metal Working Fluid Aerosols in Rats. J. Toxicol. Environ. Health. Part A, 65, 101-118.

BC0498018