Bioconjugate Chem. 2007, 18, 1028−1035
1028
ARTICLES Role of Sulfhydryl Groups in Transfection? A Case Study with Chitosan-NAC Nanoparticles Brigitta Loretz, Marlene Thaler,† and Andreas Bernkop-Schnu¨rch* Department of Pharmaceutical Technology, Institute of Pharmacy, and Institute of Zoology, Leopold-Franzens-University Innsbruck, 6020 Innsbruck, Austria. Received October 4, 2006; Revised Manuscript Received February 20, 2007
This study investigated the use of chitosan-N-acetylcysteine (NAC) as a non-viral gene carrier. In particular, we aimed to elucidate whether the advantage of thiolation was more pronounced in the stabilization of particles or in the effect of nonspecific sulfhydryl reduction of the target cells. Low-viscosity chitosan was modified by covalent binding of NAC. The resulting conjugate displayed 1.35 mM SH/g polymer. Particles produced via self-assembly of chitosan conjugate and pDNA had a mean particle size of 113.7 nm and a positive ζ-potential. Sulfhydryl group content on the particle surface was investigated by Ellman’s test and papain reactivation assay, with the result of about 100 nM SH groups/mL nanoparticle suspension. An oxidation step was performed to stabilize polyplexes via disulfide bonds. The enhanced stability of oxidized particles against both polyanion heparin and alkaline pH was proven by a gel retardation assay. The stabilization was demonstrated to be reversible by treatment with glutathione. Further, the effect of immobilized SH groups and of supplementation with free NAC on transfection efficacy on Caco-2 cells was investigated. The expression of the transgene was raised 2.5-fold and 10-fold with nonoxidized thiomer polyplexes in comparison to polyplexes of unmodified chitosan and oxidized chitosan-NAC, respectively. The impact of sulfhydryl reduction on transfection was assessed via thiol group inactivation with 5,5′-dithiobis-(2-nitrobenzoic acid) (DNTB). This inactivation resulted in a decrease of transfection efficacy. In conclusion, chitosan-NAC conjugate was demonstrated to be beneficial for transfection, either for stabilization via disulfide bonds or for raising the expression of transgene via shifting the redox potential of the target cells.
INTRODUCTION Results in recent clinical trials have renewed interest in nucleic acid-based therapeutics. The transition of these promising molecules into drugs needs to address the still most urgent problem of efficient delivery (1). Development of novel delivery platforms that prevent their degradation and facilitate specific uptake into target tissues is therefore a key issue. Here, in particular, the field of nanotechnology is raising expectations and is a main focus of many recent approaches (2-5). Currently, there is an enormous interest in non-viral vectors in general and in cationic polymer-based gene delivery systems, so-called polyplexes, in particular (2, 6-9). These systems have several advantages, such as ease of production and modification, lack of mutagenesis, and reduced potential of immunogenicity, in comparison to viral vectors. Recently, thiolated polymers have been reported as a promising tool in gene delivery. Reversible cross-linking via disulfide bonds was proposed as a strategy to keep the polyplex structure stable in the extracytoplasmic environment, while inducing the efficient release of entrapped pDNA from the polyplexes after their arrival in the cytoplasmic compartment (10-14). This tuneable release should be triggered by the large concentration * Corresponding author. Phone: +43-512-507-5371. Fax: +43-512507-2933. E-mail:
[email protected]. Address: Dep. of Pharmaceutical Technology, Univ. Innsbruck, Innrain 52c, Josef-Mo¨llerHaus, 6020 Innsbruck, Austria. † Institute of Zoology.
differential of intracellular glutathione compared to extracellular levels. In the intracellular reductive environment, the disulfide bonds should be cleaved. Few references so far have proposed that the reduced sulfhydryl group itself may enhance the transfection (15-17). The present study aims to examine the role and potential of sulfhydryl groups for enhancement of transfection efficiency. The new thiomer chitosan-NAC was chosen because of the comparatively high degree of modification and the low toxicity of the ligand N-acetylcysteine. Particular attention was paid to the quantification of thiol groups displayed by the gene carrier system. Effects of cross-linked and non-cross-linked thiolated particles on stability and transfection efficacy were assessed in Vitro. As a control and to get the first clues about the functional mechanism of thiolation-related improvement of transfection, experiments with thiol-inactivating agent 5,5′-dithiobis-(2nitrobenzoic acid) (DNTB) were performed.
EXPERIMENTAL PROCEDURES Polymer Synthesis and Characterization. For the synthesis of chitosan-N-acetylcysteine, 0.5 g of low-viscosity chitosan (Fluka, Vienna, Austria, lot number 455080/1) was dissolved in 50 mL 0.1% (v/v) acetic acid. N-Acetylcysteine (Fluka, Vienna, Austria) was dissolved in distilled water to obtain a 1 M solution. The pH of both solutions was adjusted to 5.0. The coupling reaction between the amino group of chitosan and the carboxyl group of N-acetylcysteine was mediated by 1′-ethyl3(3-dimethylaminopropyl)carbodiimide-HCl (EDAC). Initially,
10.1021/bc0603079 CCC: $37.00 © 2007 American Chemical Society Published on Web 06/07/2007
Bioconjugate Chem., Vol. 18, No. 4, 2007 1029
Role of Sulfhydryl Groups in Transfection?
250 mg of EDAC was added to 50 mL of the chitosan solution and stirred shortly before 25 mL of the NAC solution was added. This reaction mix was stirred at room temperature for 2 h. Subsequently, the polymer solution was extensively dialyzed in five steps for 12 h each. The first dialysis was performed against 5 L distilled water containing 0.2 mM HCl, the second against 5 L distilled water, 0.2 mM HCl, and 2 µM EDTA; and finally, three more times against 5 L distilled water containing 0.2 mM HCl. The purified polymer solution was frozen and lyophilized. Freeze-dried polymers were stored at 4 °C. Determination of the Thiol Group Content. The amount of reduced sulfhydryl groups on modified chitosan was determined using the photometrical Ellman’s test as described previously (18). To determinate the total amount of sulfhydryl groups on the polymer, all disulfide bonds were reduced with an excess of sodium borohydride, following Leitner et al. (19). The amount of disulfide bonds was calculated by substracting the quantity of free thiol groups from the totality of thiol moieties present on the polymer and dividing the result by two. Preperation of Nanoparticles. Unmodified chitosan and chitosan-NAC were dissolved in distilled water in final concentrations of 0.02% and 0.1% (w/v), respectively. The pH of both polymer solutions was adjusted to 4.5 with acetic acid, and the solutions were filtered through sterile filters with 0.2 µm pore size. As model plasmid pEGFP-C2 (Clontech, Mountain View, CA) coding for enhanced green fluorescence protein was used. DNA solution was prepared in distilled water with a pDNA content of 100 µg/mL. Polymer solutions and DNA solution were separately heated to 50 °C for 10 min. For formation of chitosan control particles (Cs NP), 400 µL of the 0.02% chitosan solution was added to 250 µL of pDNA solution, followed by immediate vortexing for 30 s. Chitosan-NAC nanoparticles (NAC NP) were produced in the same way with the modification that 250 µL of pDNA solution was diluted with 306 µL of distilled water before 94 µL of 0.1% chitosanNAC solution was added. In the case of oxidized nanoparticles (oxid. NP), immediately after preparation 3 and 6 µL of a 3% hydrogen peroxide solution, respectively, were added to the nanoparticle suspensions and mixed briefly. A higher amount of chitosan-NAC polymer than unmodified chitosan was used in order to obtain particles of comparable size distribution and ζ potential. The calculated ratio of amino groups to phosphate groups was 5:1 for chitosan control particles and 4.7:1 for chitosan-NAC particles. Characterization of Nanoparticles. Photon Correlation Spectroscopy. To determinate the hydrodynamic diameters of the polyplexes, dynamic light scattering measurements were carried out using PSS Nicomp 380 ZLS instrument (Nicomp Particle Sizing Systems, Santa Barbara, CA). At least 6 samples of each nanoparticle suspension were measured for 3 cycles of 5 min. The data obtained at a detection angle of 90° at 23 °C were analyzed by the Gauss volume-weighted method to calculate the mean diameter. ζ-Potential Measurement. Determination of ζ potential was performed with a 1:5 dilution of nanoparticle suspensions in distilled water to reduce the shielding effect of ions in the suspension. A scattering angle of 14.7° and an electronical field of 5 V/cm were used for analysis. Transmission Electron Microscopy. Nanoparticle suspensions were analyzed after 2 h of incubation at room temperature. Drops of 10 µL particle solution were placed on pioloformcoated grids and air-dried for 10 min. The redundant fluid was removed while the dried nanoparticles remained on the grids. These grids were dried and examined by a transmission electron microscope (ZEISS LIBRA120) with an in-column energy filter
Figure 1. Chemical structure of chitosan-NAC as synthesized within this study. In addition, the possibility of cross-linking via disulfide bonds between two acetyl residues is displayed. Table 1. Particle Characterizationa sample
mean size [nm]
ζ-potential [mV]
Cs NP NAC NP ox. NAC NP
107.1 ( 29.2 113.7 ( 32.3 73.0 ( 6.0
+9.48 ( 1.2 +6.54 ( 0.6 +17.01 ( 3.2
a Particle size and ζ potential of particles in suspension determined by photon correlation spectroscopy. Unmodified chitosan nanoparticles (Cs NP), chitosan-NAC nanoparticles (NAC NP), and disulfide-cross-linked chitosan-NAC particles (ox. NAC NP) all generated at 4.7:1 N/P ratio. Indicated values are means ( SD of at least six experiments.
(EFTEM). Digital micrographs were obtained from a ProScan Slow Scan CCD camera system using iTEM 5.0 Software from Soft Imaging System GmbH. Papain ReactiVation Assay. Papain (Sigma-Aldrich, Vienna, Austria) was inactivated to papain-S-SCH3 following the method of Singh (20). The papain reactivation assay was performed basically in the same way as described by Singh et al. (20) and Wright and Viola (21). Degassing of the solutions was not performed. For standards, a dilution series ranging from 0.9375 to 120 nM NAC was used. To aliquotes of 120 µL nanoparticle suspension, 14 µL of 100 mM sodium acetate buffer pH 4.5 were added. All samples were brought to the same volume of 280 µL by the addition of distilled water. An excess of papain (700 µL, 0.6 mg/mL) was added to each sample. The reaction mixtures were kept at room temperature for 1 h. Subsequently, 1 mL of 3.4 mM N-benzoyl-L-arginine-p-nitroanilide solution was added to all sample tubes. After 1 h further incubation at room temperature, the samples were briefly centrifuged. Finally, the absorbance was measured at 410 nm. The concentration of thiol groups in the test solutions was calculated with the linear function obtained from the calibration curve. Ellman’s Test. Aliquots of 120 µL of nanoparticle suspensions were diluted with 160 µL of 0.5 M PO4 buffer, pH 8, before adding 280 µL of Ellman’s reagent (0.3 mg/mL 5,5′-dithiobis(2-nitrobenzoic acid) (DNTB)) dissolved in phosphate buffer. Following 2 h incubation at room temperature in the dark, the samples were briefly centrifuged, and the absorbance was measured at 450 nm. Stability Test. Stability of various nanoparticle suspensions was investigated under challenging conditions such as 80-fold concentration of polyanion heparin (Sigma-Aldrich, Vienna, Austria) or minimum essential medium (MEM), pH 10, including 20% serum. For these assays, 50 µL aliquots of nanoparticle suspensions were mixed with 50 µL of aqueous heparin solution (16 mg/mL) and MEM medium, respectively. After an incubation period of 2 h at 37 °C and 400 rpm in an Eppendorf shaker, electrophoretic mobility assay on 0.7% agarose gel was performed.
1030 Bioconjugate Chem., Vol. 18, No. 4, 2007
Loretz et al.
Figure 2. TEM micrograph of nonoxidized chitosan-NAC/pDNA nanoparticles at N/P ratio 4.7:1.
µL supernatants were taken after 14 h of incubation and analyzed according to the manufactures protocol from Cytotoxicity Detection Kit (LDH) from Roche (Vienna, Austria). Cytotoxicity was calculated with the equation
cytotoxicity [%] ) (experimental value - low control)/ (high control - low control) × 100
Figure 3. Quantization of thiol groups by two independent spectrophotometric assays. Black bars indicating the results of Ellman’s test and open bars of papain reactivation assay. Investigated samples were 0.1% chitosan-NAC polymer solution, suspension of chitosan-NAC/ pDNA nanoparticles, nanoparticle suspension after 72 h incubation at room temperature, suspension of chitosan-NAC/pDNA particles produced at pH 5.5, and nanoparticle suspension after oxidation step with 3 µL H2O2.
The degree of enzymatic degradation was analyzed by digestion of 50 µL nanoparticle suspension in 1×DNaseI buffer pH 8 (Promega, Madison, WI) with 0.2 units DNaseI (Promega, Madison, WI) at 37 °C for 30 min and subsequent gel electophoresis. Release of pDNA charge from oxidized particles in a reducing environment like in cytoplasm was assessed in MEM medium containing 0 to 10 mM glutathione. Cytotoxicity Tests. LDH. The Caco-2 human colon carcinoma cells (ATCC Nr. HTB-37) were grown in 12-well plates at an initial seeding density of 1 × 105 cells in 1 mL Eagle minimum essential medium (MEM) supplemented with 2.2 g/L sodium bicarbonate, 2 mM L-glutamine, penicillin/streptomycin, and 20% FCS (Gibco, Nr. 26140-079, Invitrogen, Lofer, Austria). After 24 h incubation, cells were washed once with phosphate buffered saline (PBS) before transfection with 260 µL of nanoparticle suspension per milliliter MEM was performed. MEM medium without serum and without phenol red was used during the cytotoxicity test. As a low control, cells were incubated with medium only. The high control was prepared with medium plus 2% Trition-X100. Samples of 100
MTT. Caco-2 cells were plated into 12-well plates at a density of 5 × 104 cells/mL. After 24 h preincubation, culture medium was replaced by culture medium plus 260 µL nanoparticle suspension/well. Cells were incubated for 14 h before the medium with test solution was aspirated. Cells were washed once with PBS to remove traces of the particle solution. Subsequently, 1 mL MEM without phenol red, containing 0.5 mg/mL (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide) (MTT), was added to each well. After 3 h of further incubation, the medium with unreacted dye was aspirated. The converted dye was solubilized with 1 mL acidic isopropanol (0.04 M HCl in absolute isopropanol) by pipetting several times. The dye solution was transferred into tubes and centrifuged 2 min at 15.000 g prior to measurement of OD 570 nm with background subtraction of OD 650 nm. The relative cell viability [%] related to control wells was calculated according to the equation
viability [%] ) (test Abs 570 nm - Abs 650 nm)/ (control Abs 570 nm - Abs 650 nm) × 100 Transfection. 1 × 105 Caco-2 cells in 1 mL MEM medium were seeded into 12-well plates and incubated for 24 h. Medium was aspirated, and cells were washed once with PBS. First, 260 µL nanoparticle suspension, containing 10 µg of pDNA, was added to the cells and overlaid with 740 µL medium including 20% serum. Cells were incubated with the nanoparticles for 14 h. After this transfection step, cells were washed once with PBS, fed with 1 mL fresh medium, and incubated for another 48 h. Fluorescent cells were monitored with a fluorescent microscope using UV light and the appropriate filter for green fluorescence. Cells were washed once with PBS. Thereafter, 100 µL lysis buffer (0.25 M Tris, pH 7.4, 0.25% (v/v) Trition X-100 and 2.5 mM EDTA) were added and the plate shaken at room temperature for 5 min. To enhance cell lysis, a freeze-thaw
Bioconjugate Chem., Vol. 18, No. 4, 2007 1031
Role of Sulfhydryl Groups in Transfection?
Figure 5. Toxicity of nanoparticle suspensions and viability of Caco-2 cells after challenge, respectively, were determined by (a) lactate dehydrogenase assay and (b) MTT assay. For both cytotoxicity tests, 14 h incubation was performed. Samples: 260 µL chitosan/pDNA nanoparticle suspension (Cs NP), 260 µL chitosan-NAC/pDNA nanoparticle suspension (NAC NP) 38 µL chitosan solution 0.1% 38 µL chitosan-NAC solution 0.1%.
nm excitation/emission. Fluorescence of untransfected cells was substracted as background. Effect of Thiol Group Modification by DNTB. The dithioloxidizing agent 5,5′-dithiobis-(2-nitrobenzoic acid) (DNTB) was used to inactivate sulfhydryl groups on both cell surface and nanoparticles, to study the impact of active sulfhydryl group on the expression of transgene. A solution of 10 mM DNTB in 0.5 M phosphate buffer pH 8 was freshly prepared before every experiment. Caco-2 cells were treated with 200 µM DNTB in cell culture medium for 5 min followed by washing with PBS immediately before transfection. In a second experiment, DNTB solution was added to the nanoparticle suspension to reach a final concentration of 100 µM DNTB and preincubated for 30 min. The transfection was performed as described in the previous section. Statistical Data Analysis. Results were depicted as mean value ( SD from at least three experiments. Significance among the mean values was calculated using Student’s t test (software package SPSS 11.0). Probability value p < 0.05 was considered significant. Figure 4. (a) Gel retardation assay for particles of unmodified chitosan (Cs), chitosan-NAC nanoparticles, (NAC) and oxidized Cs-NAC nanoparticles treated with 3 and 6 µL H2O2, respectively (ox3 and ox6) (lanes 1-4), and release of pDNA after challenge with 80-fold excess of heparin (lanes 5-8) and at pH 10 (lanes 9-12). (b) Release of pDNA from non-thiolated chitosan particles (Cs NP), thiolated nonoxidized particles (NAC NP), and oxidized chitosan-NAC particles (oxid. NP) achieved by treatment with 10 mM glutathione. (c) Agarose-gel electrophoresis after 30 min DNaseI (0.2 unit) digestion of chitosan control particles (Cs NP), nonoxidized (NAC NP) and oxidized CsNAC particles (oxid.NP).
cycle was performed. Cell lysate was transferred into tubes and cell debris was separated by centrifugation. For fluorescence measurement, 50 µL of cellular extract were transferred into a microtiter plate, and the fluorescence was measured at 485/520
RESULTS N-Acetylcysteine-Modified Chitosan. Chitosan-NAC conjugates, as shown in Figure 1, were synthesized by covalent attachment of N-acetylcysteine to the amino group of the chitosan backbone. The colorimetric quantification of free sulfhydryl groups on the conjugate by Ellman’s test displayed a thiol content of 1.356 mM/g polymer. Investigation of the amount of disulfide bonds present in the conjugate without any oxidation process performed revealed 0.484 mM disulfide bonds per gram polymer. The lyophilized chitosan-NAC conjugate had the appearance of a white, odorless, fibrously structured substance. The conjugate was easily dissolved in 0.1% acetic acid. In aqueous solution of pH 4.5 and stored at 4 °C, it was stable for up to 1 month without detectable loss in transfection efficiency.
1032 Bioconjugate Chem., Vol. 18, No. 4, 2007
Figure 6. (a) Impact of N/P ratio on the transfection efficancy. Transfection of sub-confluent Caco-2 cells was performed for 14 h with nanoparticle suspension containing 10 µg pDNA each. (b) Gene expression of pEGFP after transfection of Caco-2 cells with suspension of chitosan nanoparticles (Cs NP), chitosan nanoparticle suspension supplemented with 50 mM NAC solution, nonoxidized (NAC NP) and oxidized chitosan-NAC nanoparticle (oxid. NAC NP) suspension. Dose of pDNA was 10 µg/well in all cases. *Significance level of p < 0.05. (c) Qualitative analysis of green fluorescence protein expression. Fluorescence microscopy image of chitosan-NAC transfected Caco-2 cells without and with overlaid light microscopy image.
Chitosan-N-Acetylcysteine Nanoparticles. Particles produced at N/P ratio 5:1 for chitosan and 4.7:1 for chitosanNAC, respectively, were characterized. The measured mean size of 113.7 nm ((32.3 nm) for chitosan-NAC nanoparticles was in a comparable range to that of the control particles of unmodified chitosan at 107.1 nm ((29.2 nm). Disulfide bond stabilization caused a decrease in size down to 73.0 nm ((6.0 nm). Chitosan-NAC particles as well as chitosan control particles displayed positive ζ potentials of +6.54 mV and +9.48 mV, respectively (Table 1). The moderate decline in ζ potential was based in the slightly lower N/P ratio. Transmission electron microscopy images, shown in Figure 2, supported the nanoscale size of the particles. Sedimentation and subsequent removal of the solution favored the small fraction of bigger particles in the suspension. This effect might also explain the bigger size of particles in TEM micrographs compared to the measured mean size with photon correlation
Loretz et al.
spectroscopy. Particles displayed spherical shape and high density, indicating tight packing of polymer and nucleic acid chains. Evaluation of the thiol group content was done with two methods, papain reactivation assay and Ellman’s test, respectively. The results, as presented in Figure 3, obtained with both tests are in good accordance. The advantage of papain activation assay and possibly the reason for higher values determined with this assay was the incubation at pH 7.5 instead of pH 8 in Ellman’s buffer. Moreover, the papain assay should measure just the thiol groups displayed on the particle surface, as the size of the enzyme does not allow its penetration into inner parts of the particles. Freshly prepared non-oxidized chitosanNAC nanoparticle suspension displayed about 100-120 nM/ mL SH groups. During 3 days storage at room temperature, the suspension lost about two-thirds of its functional SH groups. Preparation with polymer solution above pH 5 resulted in a loss of SH groups in the nanoparticle suspension. The comparatively milder oxidation processes was sufficient to oxidize almost all sulfhydryl groups present on the polymer. Reversible Stabilization via Disulfide Bonds. To investigate the enhanced stability of oxidized chitosan-NAC/DNA particles, an excess of the polyanion heparin and pH 10 was used. In these extreme environments, the stabilizing effect of disulfide bonds was clearly demonstrated (Figure 4a). Further, the triggered release of payload in the reducing milieu of 10 mM glutathione was shown in the gel retardation assay illustrated in Figure 4b. The agarose gel shown in Figure 4c demonstrated the stabilization against nuclease challenge. Oxidized chitosanNAC particles were more stable in comparison to nonoxidized chitosan-NAC particles, but not more stable than control particles of unmodified chitosan. Moderate Toxicity of Chitosan-NAC Nanoparticles. Since chitosan-NAC is a new polymer, we were also interested in the possible toxic effects of this polymer and the resulting chitosan-NAC/pDNA nanoparticles. The level of plasma membrane damage was investigated with lactate dehydrogenase assay (Figure 5a). In addition, the viability of Caco-2 cells after overnight exposure at the same concentrations and conditions as in transfection was assessed with MTT assay (Figure 5b). Both toxicity tests revealed moderate toxicity. The membrane damage was higher with chitosan-NAC compared to unmodified chitosan, although overall in the moderate range of less than 4% cytotoxicity. The increase in acetylation by attaching N-acetylcysteine should not have caused the higher damage, since other research groups reported a higher toxicity of chitosan with a high degree of deacetylation (22, 23). The difference in cell viability between exposure to chitosan nanoparticles and to chitosan-NAC nanoparticles assessed with MTT assay was less pronounced. Transfection Enhancement with Chitosan-NAC Nanoparticles. To maximize transfection efficiency, the optimum ratio of polymer/pDNA was determined. At N/P ratio of 4.7:1, chitosan-NAC particles resulted in the highest expression of transgene as demonstrated in Figure 6a. Particles suitable for transfection were prepared at ratios from 1.9:1 to 9.4:1, indicating the broad range of loading capacity. Figure 6b illustrates the effect of chitosan-NAC for transfection. These thiomer particles enhanced transfection 2.5-fold in comparison to unmodified chitosan. The dilution of nanoparticle suspension in cell culture medium reached a final concentration of 31 nmol SH /mL. The addition of 50 nmol/ mL free N-acetylcysteine to control chitosan nanoparticles also led to an increase in transgene expression, but not to the same extent as immobilized NAC. Oxidized chitosan-NAC nano-
Role of Sulfhydryl Groups in Transfection?
Bioconjugate Chem., Vol. 18, No. 4, 2007 1033
Figure 7. Partial reversibility of transfection enhancement by inhibition of thiol groups with DNTB. Nanoparticle suspension was supplemented with DNTB solution to achieve a final concentration of 100 µM and preincubated 30 min prior to transfection. Cells were pretreated with 200 µM DNTB for 5 min prior to transfection. Hatched bars indicate suspension of unmodified chitosan nanoparticles and black bars chitosan-NAC nanoparticles.
particles were apparently overstabilized for in Vitro transfection. Their efficiency was decreased below the efficiency of control particles. As a qualitative control of transfection, UV microscopy images overlaid with light microscopy images are displayed in Figure 6c. Reversal of the Transfection Enhancement by DNTB Treatment. If SH groups acted as functional groups to enhance transgene expression, the inactivation of these SH via the quantitative formation of S-S bonds performed with DNTB should inhibit the enhancement. Results of sulfhydryl group inactivation are shown in Figure 7. Because DNTB is a cellimpermeable agent, treatment of Caco-2 cells inactivated just the accessible SH groups of proteins on the outside of the plasma membranes. This inhibition caused a decrease in transfection efficiency. The decrease was more pronounced when the SH groups on the nanoparticles were inactivated. Addition of DNTB to control chitosan nanoparticles caused an increase in transfection. Since unmodified chitosan did not carry any sulfhydryl groups to which DNTB could bind, the reagent remained in an active form in solution and resulted in a comparable effect to free NAC.
DISCUSSION Chitosan-N-acetylcysteine represents an interesting new thiomer, as it exhibits a comparatively high amount of immobilized thiol groups and low toxicity. The relative reactivities of thiols with oxidants were demonstrated to be inversely related to the pK of the thiol group (24). The thiol group of Nacetylcysteine has a pKa of 9.5 and is therefore a compromise between highly reactive thiols like glutathione (pKa ) 8.8) and cysteine (9.2) and more stable thiols like thioglycolic acid (pKa ) 10.4). This should facilitate the production by preventing excessive loss of SH at alkaline conditions and further grant a moderate but effective action in physiologic pH range. The formulation process of nanoparticles demonstrated in this study is simple and feasible for a broad range of polymer/DNA ratios. The obtained nanoparticles were of a size which facilitates endocytosis (25). Polyplexes have been demonstrated to be taken up by two mechanisms, clathrin-coated pits and caveolaemediated endocytosis, depending on particle size (26). Particles
