Skip to main content
NCBI home page
Molecular Therapy logoLink to Molecular Therapy
. 2013 Nov 26;22(2):397–408. doi: 10.1038/mt.2013.245

TNF-α Gene Silencing Using Polymerized siRNA/Thiolated Glycol Chitosan Nanoparticles for Rheumatoid Arthritis

So Jin Lee 1, Aeju Lee 1,2, Seung Rim Hwang 3, Jong-Sung Park 4, Jiyeon Jang 4, Myung Sook Huh 5, Dong-Gyu Jo 4, Soo-Young Yoon 2, Youngro Byun 6, Sun Hwa Kim 1, Ick Chan Kwon 1,7, Inchan Youn 1,*, Kwangmeyung Kim 1,*
PMCID: PMC3916041  PMID: 24145554

Abstract

Among various proinflammatory cytokines involved in the pathogenesis of rheumatoid arthritis (RA), tumor necrosis factor (TNF)-α plays a pivotal role in the release of other cytokines and induction of chronic inflammation. Even though siRNA has the therapeutic potential, they have a challenge to be delivered into the target cells because of their poor stability in physiological fluids. Herein, we design a nanocomplex of polymerized siRNA (poly-siRNA) targeting TNF-α with thiolated glycol chitosan (tGC) polymers for the treatment of RA. Poly-siRNA is prepared through self-polymerization of thiol groups at the 5′ end of sense and antisense strand of siRNA and encapsulated into tGC polymers, resulting in poly-siRNA-tGC nanoparticles (psi-tGC-NPs) with an average diameter of 370 nm. In the macrophage culture system, psi-tGC-NPs exhibit rapid cellular uptake and excellent in vitro TNF-α gene silencing efficacy. Importantly, psi-tGC-NPs show the high accumulation at the arthritic joint sites in collagen-induced arthritis (CIA) mice. Treatment monitoring data obtained by the matrix metalloproteinase 3–specific nanoprobe and microcomputed tomography show that intravenous injection of psi-tGC-NPs significantly inhibits inflammation and bone erosion in CIA mice, comparable to methotrexate (5 mg/kg). Therefore, the availability of psi-tGC-NP therapy that target specific cytokines may herald new era in the treatment of RA.

Introduction

Improved understanding of pathogenesis in rheumatoid arthritis (RA) has led to novel therapeutic targets such as proinflammatory cytokines, including tumor necrosis factor (TNF)-α, interleukin-1, and interleukin-6.1 Among these cytokines, TNF-α secreted by activated macrophages is regarded as effective, because it induces the release of other cytokines in RA synovium, bone resorption, and chronic inflammation.2,3 It also has a synergistic effect to release matrix metalloproteinases (MMPs), a class of zinc-dependent enzymes participating in destruction and remodeling of extracellular matrix.4,5 In effect, TNF-α blocker alone or combination with other drug shows a marked inhibition of arthritis and delays joint degradation.6,7 However, even though anti-TNF-α monoclonal antibodies such as infliximab and adalimumab have been successfully developed over the last decade, their application is hindered by the risk of injection site reactions, infusional reactions, or infections including tuberculosis.8

Meanwhile, RNA interference (RNAi) mediated by small interfering RNA (siRNA) is a powerful tool to specifically silence gene expression for the treatment of variety of diseases, including viral infections, cancer, and autoimmune disorders.9 The 21–25 nucleotide, double-stranded siRNA associates with RNA-induced silencing complex in the cytoplasm and cleaves complementary target mRNA.10,11 On the contrary to monoclonal antibodies mainly targeting epitopes on the cell surface or in the blood stream, siRNA can target even mRNAs which their translated proteins are located inside cells.12 Even though siRNAs are powerful agents, they have a challenge to be delivered into the cytosol of target cells because of their poor stability in physiological fluids.13 Thus, siRNA delivery systems such as cationic liposomes, lipids, and polymers have been developed to enhance the therapeutic efficacy of siRNA.14,15,16 In addition, small peptides also can deliver siRNA into target cells and knockdown the specific protein expression.17 Among various systems, nanoparticle-based delivery systems can be accumulated in highly vascularized arthritic synovial tissues, with extended localization.18 However, it is difficult to form condensed nanoparticles via charge-charge interactions between cationic molecules and anionic siRNA because of the low charge density and short chain length of siRNA.19,20

Alternative strategies have been suggested to overcome this limitation. For example, Bolcato-Bellemin et al. showed that sticky siRNAs with short complementary overhangs could enhance complex stability and siRNA-mediated gene silencing.21 Chemically crosslinked siRNAs such as multimerized siRNA also could be prepared through disulfide linkages at both ends of modified siRNA.22 Recently, we presented that polymerized siRNA (poly-siRNA) can form stable complexes with biocompatible thiolated glycol chitosan (tGC) polymers through charge-charge interaction and chemical crosslinking.23

Herein, we designed a nanocomplex of poly-siRNA targeting TNF-α with tGC polymers (psi-tGC-NPs) for the treatment of RA. The physicochemical characteristics of psi-tGC-NPs were studied using dynamic light scattering, transmission electron microscopy, and gel retardation analysis. In vitro cellular uptake and gene silencing efficiency of psi-tGC-NPs were evaluated in murine macrophage cells. Also, the in vivo biodistribution of psi-tGC-NPs at arthritic joints of collagen-induced arthritis (CIA) mice were monitored using optical imaging. Finally, in vivo therapeutic efficacy of intravenously injected psi-tGC-NPs was evaluated using microcomputed tomography (micro-CT) and histological examination.

Results

Preparation and characterization of psi-tGC-NPs

We designed tGC nanoparticles encapsulated with poly-siRNA targeting TNF-α. It was hypothesized that these nanoparticles may be localized at the arthritic joint sites followed by rapid macrophage uptake and release of free siRNAs by degradation of disulfide linkages, leading to knockdown of target mRNA and the treatment of RA (Figure 1a). As shown in Figure 1b, each thiol group at the 5′ end of sense and antisense strand of siRNA was modified to generate poly-siRNA with reducible disulfide linkages. The 5′ end-thiolated siRNA molecules have been used for various siRNA modifications without affecting their RNAi activity.24 Self-polymerized siRNA by disulfide linkages can be degraded to mono-siRNA in the presence of reducing reagent such as glutathione. The primary amine groups of glycol chitosan (GC) polymers used as a carrier for poly-siRNA were partially modified into thiols with sulfosuccinimidyl 6-[3′-(2-pyridyldithio)-propionamido] hexanoate(sulfo-LC-SPDP), and then tGC was obtained as a product (Figure 1c). Synthesized tGC can form complex with poly-siRNA through charge interactions and disulfide crosslinking reaction, resulting in psi-tGC-NPs (Figure 1d).

Figure 1.

Figure 1

Open in a new tab

Schematic diagrams. (a) Uptake of psi-tGC-NPs into macrophage cells leading to tumor necrosis factor (TNF)-α gene knockdown. (b) Formation of poly-siRNA. (c) Synthesis of tGC polymers. (d) Complexation of poly-siRNA with tGC polymers.

Thiol-modified siRNAs were self-polymerized under mild oxidation and basic condition in 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES) buffer (pH 8.0). While monomeric siRNA was shown as a single band on 8% polyacrylamide gel electrophoresis, the synthesized disulfide-linked poly-siRNAs displayed a ladder-like migration pattern indicating molecular weight distribution of poly-siRNA (Figure 2a). The molecular weight distribution of poly-siRNA was from 50 to above 300 base pairs. Each band fraction of poly-siRNA was quantitatively analyzed, and 52.8% of poly-siRNA molecules were above 200 bp. On average, poly-siRNA has 12 times higher molecular weight per one molecule compared to typical monomeric siRNA, indicating an increase in available negative charges and molecule sizes. Although the siRNA molecules were successfully self-polymerized to give a higher molecular weight and charge density, the binding affinity of poly-siRNA and unmodified GC was not sufficient to form stable nanosized complexes. Herein, we used the thiol-modified GC polymers to improve the binding affinity with poly-siRNA. In a gel retardation assay, poly-siRNAs were successfully encapsulated into tGC polymers and produced complexes at weight ratios of 1:5 and 1:10 (poly-siRNA:tGC) (Figure 2a).

Figure 2.

Figure 2

Open in a new tab

Characterization of psi-tGC-NPs. (a) Confirmation of poly-siRNA formation by gel electrophoresis and gel retardation of mono- or poly-siRNA and complex of tGC with poly-siRNA at different weight ratios (1:2, 1:5, and 1:10). Lane 7 shows reduction of poly-siRNA by adding 1,4-dithiothreitol (DTT). (b) Size and morphology of psi-tGC-NPs by dynamic light scattering (DLS) and transmission electron microscopy (TEM).

Interestingly, poly-siRNA in the psi-tGC-NP was completely degraded into monomeric siRNAs under 10 mmol/lof 1,4-dithiothreitol (DTT) in HEPES buffer within 30 minutes (Figure 2a, lane 7). It indicates that the reducible poly-siRNAs are readily degraded under reductive conditions like in the cell cytosol. The mean particle size, determined by dynamic light scattering measurement, was found to be smaller at a weight ratio of 1:10 (poly-siRNA:tGC) than mean sizes at other weight ratios such as 1:5, 1:20, and 1:40 (more than 400 nm, data not shown). At a weight ratio of 1:10, transmission electron microscopy images demonstrated that psi-tGC-NPs had spherical morphology, and the average hydrodynamic diameter and zeta potential of the complexes were about 370 nm and 4.01 ± 0.72 mV, respectively, as measured by dynamic light scattering (Figure 2b).

In vitro cellular uptake and gene silencing efficacy of psi-tGC-NPs

The cellular uptake of psi-tGC-NPs (a weight ratio of poly-siRNA:tGC = 1:10) was monitored in murine macrophage RAW 264.7 cells with fluorescence microscopy. Contrary to faint fluorescence of near-infrared fluorescence (NIRF) dye–labeled poly-siRNA in nonactivated cells, distinct red spots of NIRF dye–labeled poly-siRNAs in the psi-tGC-NPs were clearly observed in lipopolysaccharide-activated RAW 264.7 cells within 1 hour (Figure 3a). It indicates that the complexation of poly-siRNA with tGC enhanced the uptake of poly-siRNA by lipopolysaccharide-activated macrophage cells and the localization of poly-siRNA in the cytoplasm. On the other hand, psi-tGC-NPs were not taken up in nonactivated macrophage cells, because the phagocytosis of nanoparticles is mediated by a nonopsonic attachment to the macrophage surface via scavenger receptors and dependent on macrophage activation status. The abundance of activated macrophages in rheumatoid synovium, especially at the cartilage-pannus junction, significantly correlates with the degree of articular destruction or progression in RA.25

Figure 3.

Figure 3

Open in a new tab

In vitro cellular uptake and gene silencing effect of psi-tGC-NPs in murine macrophage RAW 264.7 cells. (a) Confocal microscopic images of RAW 264.7 cells treated with FPR-675–labeled psi-tGC-NPs. Scale bar = 20 µm. (b) Reverse transcription PCR (RT-PCR) images for tumor necrosis factor (TNF)-α mRNA levels obtained from lipopolysaccharide (LPS)-activated RAW 264.7 cells treated with free poly-siRNA, psi(scramble)-tGC-NPs, psi-LF and psi-tGC-NPs targeting TNF-α (100 nmol/l of poly-siRNA) for 4 hours. (c) Semiquantitative analysis for TNF-α mRNA expression. Graph was plotted after normalizing with the gene expression intensity of β-actin.

To evaluate the relative gene silencing efficiency of psi-tGC-NPs, psi-tGC-NPs targeting TNF-α were treated to RAW 264.7 cells, and the level of TNF-α expression was examined via reverse transcription PCR (RT-PCR). At the optimized weight ratio (1:10), scrambled psi-tGC-NPs (psi(scramble)-tGC-NPs) were also prepared at the same condition with that of psi-tGC-NPs. In Figure 3b, psi-tGC-NPs had an inhibitory effect comparable with poly-siRNA/Lipofectamine complex (psi-LF) on TNF-α gene expression. In quantitative analysis of mRNA by RT-PCR, psi-tGC-NPs treatment (100 nmol/l) efficiently suppressed TNF-α expression up to 78%, comparable to that of psi-LF treatment (81%) (Figure 3c). On the other hand, there were no significant changes in the TNF-α suppression among untreated, free poly-siRNA, and psi(scramble)-tGC-NPs. These data indicate that psi-tGC-NPs silence gene expression in a sequence-specific manner.

Accumulation of systemically administered psi-tGC-NPs in arthritic joints

To demonstrate in vivo efficacy of psi-tGC-NPs, the development of CIA mice was evaluated. As shown in Figure 4a, we monitored disease progression of CIA from 2 to 7 weeks after the first immunization. At 3 weeks after the primary immunization, mild swelling of tarsal joint (score 1) was observed, and swelling and erythema intensified more and more as time passed. At 7th week, severe swelling encompassing the ankle, foot, and digits or ankylosis of the limb was observed in CIA mice (score 4) (Figure 4b). The in vivo accumulation of psi-tGC-NPs was monitored by optical imaging after intravenous injection to CIA mice. At 5 weeks after the first immunization, psi-tGC-NPs containing 50 µg of poly-siRNA labeled with NIRF dye were injected via tail vein into the mice scored about 3.

Figure 4.

Figure 4

Open in a new tab

In vivo imaging of psi-tGC-NPs accumulated in the arthritic joints of collagen-induced arthritis (CIA) mice. (a) The paw swelling in CIA model by naked eyes. (b) Arthritic score in CIA model. (c) In vivo near-infrared fluorescence (NIRF) tomographic images of arthritic joints after intravenous injection of psi-tGC-NPs or free poly-siRNA. (d) Fluorescence intensity of FPR-675–labeled poly-siRNA at hind limbs was recorded at each time point (n = 3 mice per group).(e) NIRF reflectance images of CIA mice 24 hours after intravenous injection of psi-tGC-NPs or free poly-siRNA.(f) Ex vivo NIRF images of dissected organs after 24 hours.

The mice injected with psi-tGC-NPs(50 µg of poly-siRNA) showed high fluorescence intensity in the arthritic joints or paws at 1 hour after injection (Figure 4c). The maximum NIRF signal shown at 1 hour after injection persisted for up to 24 hours. In contrast, free poly-siRNA injected into CIA mice was not accumulated in the arthritic joints. In the quantitative analysis of fluorescence intensity, the group treated with psi-tGC-NPs showed significantly higher fluorescence intensity than the group treated with free poly-siRNA even at 1 hour after injection, and the total photon counts of psi-tGC-NPs in the arthritic joints were almost twofold higher than those of free poly-siRNA at 6 hours after injection (Figure 4d). The fluorescence intensity was lower and disappeared more rapidly in CIA mice injected with free poly-siRNA compared with mice injected with psi-tGC-NPs (Figure 4e). Ex vivo biodistribution study also revealed that psi-tGC-NPs exhibited stronger fluorescence in the arthritis joints of front and hind legs compared to free poly-siRNA (Figure 4f). This was caused by arthritic joint specific siRNA delivery of psi-tGC-NPs by the enhanced permeation and retention. The notable retention of psi-tGC-NPs in kidney may originate from its longer blood circulation time compared to free poly-siRNA. However, it may not be a serious problem because foreign materials in kidney are generally excreted from body by urine before 1 day.26 These results suggest that psi-tGC-NPs have prolonged circulation in vivo compared to free poly-siRNA and can be specifically accumulated in the arthritic joints.

In vivo therapeutic efficacy of psi-tGC-NPs in CIA mice

To evaluate therapeutic efficacy of psi-tGC-NPs, CIA mice were treated with psi-tGC-NPs (1.5 mg/kg of poly-siRNA) or methotrexate (MTX, 5 mg/kg) from 3 to 7 weeks after the first immunization, and treatment monitoring data were obtained each week. NIRF signals in the arthritic joints of untreated saline control mice injected with MMP-3–specific nanoprobe increased over the course of RA development, and then became maximal at 5 weeks after the first immunization (Figure 5a). In the previous study, the fluorescence recovered in in vivo imaging using an MMP-3–specific polymeric probe was well correlated with upregulated MMP-3 activity and the arthritis severity score in the joint tissues.27 Thus, MMP-3–specific nanoprobe can be used to predict the total joint damage progression and to assess therapeutic efficacy of antirheumatic drugs using optical imaging system. Meanwhile, substantially reduced NIRF signals were observed in the group treated with psi-tGC-NPs or MTX compared with untreated saline control group from 5 weeks after the first immunization. There was only slight swelling or erythema confined to the peripheral joints in the group treated with psi-tGC-NPs or MTX, whereas untreated saline control group showed severe swelling of entire paw (Figure 5b). These results corresponded with the arthritic scores and hind paw thickness measurements (Figure 5c,d). Universal negative control siRNA loaded psi(sc)-tGC-NPs formulation showed no target gene silencing and therapeutic effects, indicating a sequence-specific RNAi effect (data not shown).

Figure 5.

Figure 5

Open in a new tab

Monitoring of the disease progression in collagen-induced arthritis (CIA) mice. (a) In vivo near-infrared fluorescence (NIRF) images of normal mice (left, top) and CIA mice in each group (untreated, methotrexate (MTX)-treated, and psi-tGC-NP–treated) after intravenous injection of MMP-3 nanoprobe (n = 8) at different stages of disease (at 3, 4, 5, and 7 weeks after the primary immunization). (b) Enlarged paws in the untreated and psi-tGC-NP–treated CIA mice. (c) Mice were scored according to the disease progression. Arthritic score of untreated CIA mice increased rapidly and reached the highest score at 5–7 weeks after the primary immunization. On the other hand, arthritic scores of MTX- and psi-tGC-NP–treated groups were sustained like normal state. (d) The severity of arthritis assessed by measuring hind paw thickness over time (n = 5 mice per group).

Micro-CT analyses were also performed to assess the impact of psi-tGC-NPs treatment on bone destruction in CIA mice. Bone erosions in paws and ankle joints were found in the reconstructed three-dimensional micro-CT images of untreated saline control mice at 6–7 weeks after the first immunization (Figure 6a). Loss of bone integrity in the tibia and femur was also apparent in untreated saline control mice (Figure 6b). On the other hand, the smooth bone surface with only a little bone erosion was shown in micro-CT images of the group treated with psi-tGC-NPs or MTX (Figure 6a,b). Psi-tGC-NP treatment also significantly prevented the loss of bone volume, which was comparable to MTX treatment.

Figure 6.

Figure 6

Open in a new tab

Three-dimensional reconstruction of paws and knee joints and bone volume analysis using Micro-CT. (a) Paw damage was visualized using micro-CT. Computed tomographic images of the paws from untreated, methotrexate (MTX)-treated, and psi-tGC-NP–treated mice with collagen-induced arthritis (CIA) at 4, 5, 6, and 7 weeks (w) after the first immunization are shown (n = 3 per each week). Bone volume of MTX-treated (10.59 ± 1.76) and psi(TNF-α)-tGC-NP–treated (10.76 ± 1.83) knee joint was sustained like normal one (data not shown, 12.10 ± 0.94). (b) Knee joint damage was visualized using micro-CT. Bone volume of MTX-treated (17.95 ± 2.73) and psi(TNF-α)-tGC-NP–treated (16.51 ± 2.91) knee joint was sustained like normal one (data not shown, 19.06 ± 0.73).

Analysis of TNF-α expression in serum and arthritic joints

The histological cross-sections of knee joints revealed more extensive loss of safranin O-stained (red) proteoglycans in untreated saline control mice than in psi-tGC-NP–treated mice (Figure 7a). Knee joint sections from MTX-treated mice showed normal cartilage surfaces with a clear interface. The cartilage was maintained also in knee joint sections from psi-tGC-NP–treated mice except for only a slight decrease in Safranin O staining. Immunohistochemistry for TNF-α showed abundant TNF-α in knee joint tissues from untreated saline control mice. On the other hand, tissues from psi-tGC-NP–treated mice showed the reduction of TNF-α compared to those from saline controls. Especially, the staining signal for TNF-α was hardly visible in MTX-treated mice.

Figure 7.

Figure 7

Open in a new tab

Analysis of tumor necrosis factor (TNF)-α expression in serum and arthritic joints and histological examination. (a) Safranin O staining (upper; red: cartilage, gray green: cytoplasm) and anti-TNF-α staining (lower; brown: TNF-α, purple: counter stain) of an arthritic knee joint at 7 weeks after the first immunization (n = 3 per each group). Scale bar = 100 µm. (b) Each group of collagen-induced arthritis (CIA) mice blood serum was obtained at 4, 5, and 7 weeks after the first immunization, and serum concentrations of TNF-α were measured. Values are the mean ± SEM. *P < 0.05, **P < 0.01 by one-way analysis of variance. (c) Immunoblot results show the expression of TNF-α and MMP-3 in knee joints at 4, 5, and 7 weeks after the first immunization. The band of TNF-α was detected approximately at 52 kDa and the band of MMP-3 was detected at 57 kDa (pro) and 28 kDa (active). β-actin was used as loading control (n = 3 per each group). Lane 1 (N), lane 2 (U4), lane 3 (U5), lane 4 (U7), lane 5 (M4), lane 6 (M5), lane 7 (M7), lane 8 (S4), lane 9 (S5) and lane 10 (S7) indicates normal control group, untreated saline control groups at 4, 5, and 7 weeks, methotrexate (MTX)-treated groups at 4, 5, and 7 weeks, and psi-tGC-NP–treated groups at 4, 5, and 7 weeks, respectively. (d) Representative histological images of transferase dUTP nick end labeling (TUNEL)–stained sections of small intestinal tissues from the normal and CIA mice treated with saline, MTX and psi-tGC-NPs (n = 3 per each group). Scale bar = 100 µm.

Serum TNF-α level was also determined in the course of inflammatory response. As shown in Figure 7b, serum TNF-α level was elevated corresponding to disease progression in untreated saline control mice, whereas it was significantly reduced in psi-tGC-NP–treated mice at 5 weeks after the first immunization (P < 0.05). It was more significantly reduced in MTX-treated mice (P < 0.01). The expression of TNF-α and MMP-3 in the immunoblot was also increased according to the progression of arthritis in CIA mice (Figure 7c). The signal intensities for TNF-α (detected approximately at 52 kDa) and active form of MMP-3 (detected at 28 kDa) as well as pro form of MMP-3 (detected at 57 kDa) were reduced in psi-tGC-NP–treated mice. They were further reduced in MTX-treated mice. Apoptotic damage in the intestine was also evaluated by terminal deoxynucleotidyl transferase dUTP nick end labeling assay on intestinal sections from each group in CIA model (Figure 7d). In MTX-treated mice, the number of transferase dUTP nick end labeling–positive cells was markedly increased in the small intestinal crypts and villi. Nonspecific action of systemically administered MTX has been associated with a broad array of possible adverse effects such as gastrointestinal, hematologic, pulmonary, and hepatic toxicities. On the contrary, systemically administered psi-tGC-NPs are considered not to show undesirable cytotoxicity and nonspecific targeting at its therapeutic dose.

Discussion

This work describes a new approach for the treatment of RA using TNF-α gene silencing with nanoparticle-based poly-siRNA delivery system. Since polymerized siRNA cannot freely cross the cell membrane, appropriate delivery systems are required to facilitate its access to the desired intracellular sites of action. To develop a poly-siRNA delivery system with stable and condensed nanostructure, the primary amine groups of GC polymers were partially thiolated with sulfo-LC-SPDP and DTT. Charge-charge interactions of poly-siRNA with tGC polymers decrease the distances between thiol groups in tGC, leading to intramolecular chemical crosslinking (Figure 1).

Psi-tGC-NPs with a condensed structure showed the stability against physiological anionic carbohydrates and serum nucleases in the previous study.23 After DTT treatment, poly-siRNA in the nanoparticle was successfully cleaved into monomeric siRNAs within 30 minutes (Figure 2a). Thus, systemically delivered psi-tGC-NPs would be stable in blood while allowing for the release of free siRNAs in the cytosol by intracellular reductive glutathione.

We hypothesized that siRNA mediated by RNAi mechanism could knockdown TNF-α mRNA and inhibit the progression of CIA. The level of TNF-α expression examined by RT-PCR showed that there was no significant difference between untreated group and psi(sc)-tGC–treated group, whereas psi(TNF-α)-tGC-NP efficiently suppressed TNF-α gene expression in 100 nmol/l concentration (Figure 3b,c). It indicates that psi-tGC-NP can silence the target gene in a sequence-specific manner.

As shown in Figure 4c, systemically injected psi-tGC-NPs into CIA mice was circulated in the bloodstream, and then accumulated rapidly and selectively in arthritic joints. The blood-joint barrier in RA joints is characterized by angiogenesis with highly permeable vasculature, and amenable to drug targeting mediated by the enhanced permeation and retention effect.28 The endothelial gaps produced by inflammatory stimulus are in the range of 0.36–0.47 µm, which is large enough to permit the concentration-dependent passage of the spherical psi-tGC-NPs with a mean diameter of 370 nm.29 Poor lymphatic drainage of macromolecules may play an additional role in increased retention of polymeric nanoparticles in the arthritic joints.30 After accumulated in RA joints, nanoparticles can be introduced into activated macrophage cells by the phagocytosis.31

MMPs play an important role in inducing bone resorption and cartilage destruction, and can be synergistically downregulated by inhibition of TNF-α. They are synthesized and secreted as latent pro form and their activation is due to the cleavage and removal of the prodomain by proteolysis. The immunoblot results with knee synovial membrane tissues from CIA mice indicated that psi(TNF-α)-tGC treatment suppressed the expression of pro form and active form of MMP as well as the expression of TNF-α (Figure 7c). The suppressed level of expression compared to the untreated saline control was more prominent from 5 weeks after the first immunization, when the expression of TNF-α was remarkably increased in CIA mice. The expression pattern of active MMP-3 also correlated well with the arthritis severity and in vivo NIRF signal when MMP-3–specific nanoprobe was used in CIA mice. Thus, MMP-3, one of the most promising biomarkers in rheumatology, is useful in predicting the total joint damage progression and to assessing clinical effectiveness of antirheumatic drugs.

We also showed that psi-tGC-NPs did not only suppress paw swelling but also progressive joint destruction. TNF-α can act as a master regulator in the RA microenvironment, because other proinflammatory cytokines are linked into a network with TNF-α. In addition, TNF-α was also reported to be involved in the differentiation and activation of osteoclasts resorbing bone.32 Thus, blockade of TNF-α has been considered to be an attractive target for suppressing both inflammation and bone resorption in RA.33 The protection of bone destruction by psi-tGC-NPs was comparable with that achieved by treatment with MTX, one of the most commonly used disease-modifying antirheumatic drugs.

These superior results may be attributed to nanoparticle-based delivery system for poly-siRNA using tGC polymers. Reversible disulfide linkages in psi-tGC-NPs are stable in the circulation and extracellular milieu, where a redox potential is relatively high due to a low concentration of glutathione tripeptide. In contrast, the disulfide bonds are prone to rapid cleavage under a reductive environment inside cells. These large difference in reducing potential between the intracellular and extracellular environments can be exploited by psi-tGC-NPs.34 Furthermore, hypoxic status and the depletion of serum reduced glutathione levels in RA also render the reducible psi-tGC-NPs more enhanced stability in serum.35

Although the endocytosed poly-siRNA is depolymerized in the reducing conditions in the cytoplasm, the depolymerization efficacy is unlikely to be 100%. Thus, psi-tGC-NPs should be further explored for their potential immune response. Long double-stranded (ds) RNA activates nonspecific inhibition of gene transcription and production of inflammatory cytokines, which could hypothetically lead to a wide variety of toxicities in human studies. In our previous study, we have already verified that both poly-siRNA and psi-tGC-NPs did not trigger distinct inflammatory responses.23

For the delivery of siRNA, there have been also developed other strong cationic polymers such as polyethylenimine which can interact with negatively charged siRNA. However, the ionic complex between polyethylenimine and siRNA has limitations in that it is hard to form condensed and stable complexes, resulting in release and degradation of siRNA in serum and negligible accumulation in tumor tissue.36

In summary, this study has several implications for the treatment of RA. Firstly, psi-tGC-NP system enhanced the accumulation of poly-siRNA in arthritic joints. After internalized in activated macrophage cells, free siRNAs released from psi-tGC-NPs could efficiently knockdown target mRNA in a sequence-specific manner. Blockade of TNF-α by psi-tGC-NPs could suppress paw swelling but also progressive joint destruction, which was comparable to the efficacy of MTX, the current leading disease-modifying antirheumatic drug.

Materials and Methods

Materials. GC (Mw 250 kDa, degree of deacetylation 82.7%), 4′,6-diamidino-2-phenylindole, N,N,N′,N′-tetramethylazodicarboxamide, ethidium bromide (EtBr), DTT, HEPES, and ethylenediaminetetraacetic acid (EDTA) were purchased from Sigma-Aldrich (Yongin, Korea). Sulfo-LC-SPDP and N-hydroxysulfosuccinimide were obtained from Pierce (Rockford, IL). Lipofectamine 2000 and NIRF dye (Flamma FPR-675) were purchased from Invitrogen (Carlsbad, CA) and BioActs (Incheon, Korea), respectively. For TNF-α gene silencing, thiol-modified siRNA containing thiol groups at the 5′-ends of both sense and antisense strands was synthesized and annealed by Bioneer (Daejeon, Korea) with the following sequence: sense 5′-GACAACCAACUAGUGGUGCdTdT-3′, antisense 5′-GCACCACUAGUUGGUUGUCdTdT-3′. As a control, the scramble siRNA containing thiol groups at the 5′-ends of both sense and antisense strands was also synthesized and annealed with the following sequence: sense 5′-CCUACGCCACCAAUUUCGUdTdT-3′, antisense 5′-ACGAAAUUGGUGGCGUAGGdTdT-3′.

Complexation of poly-siRNA with tGC. First, for the more enhanced stability of siRNA, poly-siRNA was formed as described in the literature.24 Briefly, thiol-modified siRNA (5 mg, 0.38 µmol) for TNF-α gene silencing was dissolved in 500 µl of 10 mmol/l HEPES buffer (pH 8.0, 1 mmol/l EDTA), to which N,N,N′,N′-tetramethylazodicarboxamide (1 mg, 5.8 µmol) was added. The mixture was gently vortexed at room temperature overnight, purified using a PD-10 desalting column (GE Healthcare, Piscataway, NJ), and lyophilized. The poly-siRNA formation was assessed by electrophoresis of the product on a 8% polyacrylamide gel in Tris/Borate/EDTA (TBE) buffer, followed by EtBr staining. Each fractional amount of poly-siRNA base pairs was relatively quantified using Kodak Image Station 4000 MM Digital Imaging System (New Haven, CT), based on the electrophoresis result.

Second, GC polymers as carriers for poly-siRNA were reacted with Sulfo-LC-SPDP to yield tGC, as follows.23 GC (30 mg, 0.12 µmol) was dissolved in phosphate buffer (10 ml, pH 7.4), to which Sulfo-LC-SPDP (6 mg, 11.37 µmol, 10% feed ratio for amine group of GC) was added. The mixture was vigorously vortexed at 25 °C overnight, and then DTT was added for 3 hours with adjustment of pH to 3.5. It was followed by dialysis (MW cutoff: 12–14 kDa) against distilled water for 2 days and lyophilization. The degree of thiolation of tGC polymers was determined by reaction with Ellman's reagent (5,5′-dithiobis-(2-nitrobenzoic acid), DTNB, Pierce, Rockford, IL) and the sample's absorbance was read at 412 nm on a spectrophotometer. Seven percent of amine residues on GC polymers were modified with thiol groups for making psi-tGC. Thus, the ratio of amine:thiolated groups:acetyl groups on the modified chitosan was 74.4: 5.6: 20.

To make the complex of poly-siRNA with tGC, the dispersion of tGC (1 mg) in 0.4 ml of 10 mmol/l HEPES buffer (pH 8.0, 1 mmol/l EDTA) was sonicated at 90 W for 1 minute. Then, poly-siRNA dissolved in HEPES buffer was slowly added to tGC at 37 °C for 1 hour. The effect of tGC polymer concentration on the formation of the poly-siRNA/tGC complexes at mixing ratio from 1:2, 1:5, and 1:10 was confirmed using gel retardation assay using the 8% polyacrylamide gel electrophoresis in 0.5× TBE buffer. The size and surface charge of psi-tGC-NPs were measured using dynamic light scattering (Spectra Physics Laser Model 127-35, Mountain View, CA) and electrophoretic light scattering (ELS-8000, Otsuka Electronics, Japan). The morphology of particles was directly observed with transmission electron microscopy (Philips CM200 and Tecnai G2 F20).

Cellular uptake of psi-tGC-NPs. For the cellular uptake study of psi-tGC-NPs, the murine macrophage cell line, RAW 264.7, was cultivated in RPMI 1640 (Gibco, Grand Island, NY) supplemented with 10% fetal bovine serum (Invitrogen), 100 U/ml penicillin and 100 µg/ml streptomycin at 37 °C and 5% CO2 atmosphere. To assay the different cellular uptakes of psi-tGC-NPs in nonactivated or activated macrophage cells, 1 × 105 cells were seeded onto cover slip in 35 mm dishes and they were incubated without/with lipopolysaccharide (100 ng) for 24 hours. To monitor cellular uptake of psi-tGC-NPs, poly-siRNA was labeled with the NIRF dye, Flamma FPR-675 and the nucleus of macrophage cells was counterstained by 4′,6-diamidino-2-phenylindole. Cells were treated with FPR-675–labeled psi-tGC-NPs (5 µg of poly-siRNA) for 2 hours, fixed with formaldehyde/glutaraldehyde for 10 minutes at room temperature, and visualized using fluorescence microscopy (IX81-ZDC, Olympus, Tokyo, Japan).

In vitro gene silencing test in activated macrophage culture system. To evaluate the in vitro gene silencing efficiency of psi-tGC-NPs, RAW 264.7 cells were seeded in 12-well plates and cultured with 100 ng of lipopolysaccharide for 24 hours. Then, the activated cells were treated with free poly-siRNA, psi-LF, psi(scramble)-tGC-NPs, and psi-tGC-NPs targeting TNF-α (100 nmol/l of poly-siRNA) in each well for 4 hours, and the transfection medium was changed to fresh culture medium containing fetal bovine serum. After 24 hours, total RNA was extracted from treated cells using a RNA isolation kit (Quiagen, Valencia, CA), and subjected to reverse transcription using the High Capacity RNA-to-cDNA Kit (Applied Biosystems, Foster city, CA) to generate the first-strand cDNA. The cDNAs for TNF-α were amplified by RT-PCR under the following conditions: an initial denaturation for 3 minutes at 95 °C, 25 cycles of 30 seconds each at 95 °C (denaturation), 56 °C (annealing) and 72 °C (extension), and a final extension for 3 minutes at 72 °C. The specific primers were synthesized by Bioneer (Daejeon, Korea) with the following sequences: TNF-α (forward or F) 5′-AGTGGAGGAGCAGCTGGAGT-3′ and (reverse or R) 5′-TCCCAGCATCTTGTGTTTCTG-3′); β-actin as a scrambled-sequence for control (F) 5′ AGAGGGAAATCGTGCGTGAC-3′ and (R) 5′ GCGTCCACGTAGTAGTAGCC-3′). The PCR products were subjected to electrophoresis on 2% agarose gels, followed by EtBr staining (triplicate).

Preparation of CIA model. All the animal experiments were performed in accordance with guidelines of the Institutional Animal Care and Use Committee of the Korea Institute of Science and Technology. Murine CIA was achieved according to the published procedure, with minor modifications. Bovine type II collagen (Sigma) was dissolved in 0.1 mol/l acetic acid at a concentration of 2 mg/ml by stirring for 4 hours at 4 °C. To prepare complete Freund's adjuvant, 3 mg heat-inactivated Mycobacterium tuberculosisH37Ra (Difco, Detroit, MI) was added to 1 ml of incomplete Freund's adjuvant (Sigma). Collagen was then added drop by drop to complete Freund's adjuvant (1:1 v/v), followed by emulsification with a homogenizer in ice bath. Thirty male DBA/1J mice (Jungang Lab Animal, Seoul, Korea) at 5 weeks of age were immunized with a 50 µl intradermal injection of the emulsion into the tail. Booster immunization was performed at two weeks after the primary immunization, and then the mice were evaluated for disease progression three times a week as follows.37 Each limb was scored according to the severity of inflammation, where 0 = no evidence of redness or swelling, 1 = redness and slight swelling confined to the tarsal joints or ankle, 2 = redness and mild swelling extending from the ankle to the tarsal joints, 3 = redness and moderate swelling extending from the ankle to the metatarsal joints, and 4 = severe swelling encompassing the ankle, foot, and digits, or ankylosis of the limb.

In vivo imaging of psi-tGC-NPs accumulated in the arthritic joints. To confirm whether psi-tGC-NPs could be accumulated in the arthritic joints, CIA mice were monitored by in vivo NIRF imaging. At 5 weeks after the first immunization, the mice scored higher than two were injected via tail vein with FPR-675–labeled poly-siRNA and psi-tGC-NPs (50 µg of FPR-675–labeled poly-siRNA, n = 3), respectively. After 1, 6, 12, and 24 hours after injection, different poly-siRNA accumulation in the arthritic joints were evaluated using the eXplore Optix system (Advanced Research Technologies, Montreal, Canada). The laser power and count time were set to 13 µW and 0.3 second per point, respectively. The total NIRF intensity in the arthritic joints was quantified by using the region of interest function of Analysis Workstation software. After 24 hours, fluorescence signals in mice injected with poly-siRNA or psi-tGC-NPs were monitored using the Kodak Image Station 4000 MM Digital Imaging System equipped with a special C-mount lens and a near infrared emission filter (600–700 nm; Omega Optical).

In vivo therapeutic efficacy of psi-tGC-NPs in CIA mice. At 3 weeks after the first immunization, CIA mice were randomly allocated into three groups receiving intravenous injections of saline, psi-tGC-NPs (1.5 mg/kg siRNA) or MTX(5 mg/kg) two times a week until 7 weeks. Hind limbs of treated mice were scored for severity of arthritis. The arthritis scores were evaluated using a scale of 0 to 4 for each paw. Hind paw thickness was measured with an electric caliper placed across the ankle joint at the widest point. For monitoring treatment response, mice in each group were intravenously injected with the MMP-3–specific nanoprobe (10 µg in 100 µl of PBS, once a week) at 3 to 7 weeks after the first immunization (n = 7 per each group).27 Briefly, the MMP-3–specific polymeric probe was prepared by conjugating FPR-675, MMP substrate peptide (Gly-Val-Pro-Leu-Ser-Leu-Thr-Met-Gly-Lys-Gly-Gly) and dark quencher (Black Hole Quencher-3) to self-assembled chitosan nanoparticles. After 1 hour after injection of MMP-3–specific nanoprobe, the mice were anaesthetized and fluorescence images of the hind limbs were taken using the eXplore Optix system.

Also, hind limbs (n = 3 per each group) were assessed by micro-CT. After the mice were sacrificed every week, their hind limbs were immediately collected and stored in saline at −80 °C. Hind limb specimens were placed with the implant axis perpendicular to the scanning section, and cross-section images of the specimens were acquired at an isotropic resolution of 35 µm using a micro-CT system (Skyscan 1076, Skyscan, Kontich, Belgium). Scanning parameters were set at voltage of 100 kV, a current of 100 µA, an exposure time of 475 ms and a rotation step of 0.4° with a 1-mm–thick aluminum filter. The volume of knee subchondral bone and middle toe bone were also measured from reconstructed three-dimensional images.

Analysis of TNF-a expression in serum and arthritic joints. Blood was collected from mice in each group at 4, 5, and 7 weeks after the primary immunization, and serum TNF-α levels were determined by sandwich enzyme-linked immunosorbent assay according to the manufacturer's instructions (Uscn Life Science, Wuhan, China). In addition, CIA mice were euthanized and tissues from knee synovial membranes (n = 3 per each group) were ground in liquid nitrogen. The obtained tissue powder was incubated for 1 hour in ice with extraction buffer (50 mmol/l Tris–HCl, pH 7.5, containing 0.2 mol/l NaCl, 5 mmol/l CaCl2 and 1% [w/v] Triton X-100), then centrifuged at 13,000× g for 5 minutes at 4 °C. The supernatant was collected and assayed for protein concentration according to the Bradford method. 10 µg protein of each sample was mixed with a fivefold excess of sample buffer (0.25 mol/l Tris, 0.8% sodium dodecyl sulfate, 10% glycerol, and 0.05% bromphenol blue) and heated to 100 °C for 10 minutes. Reaction products were loaded onto 7.5% polyacrylamide gel, and then transferred to nitrocellulose membrane (Invitrogen). The membrane was blocked with 5% skim milk in Tris-buffered saline containing Tween 20 (TBST, 50 mmol/l Tris–HCl, pH 7.5, 150 mmol/l NaCl, supplemented with 0.2% Tween 20). The membrane was probed with the primary antibody (anti-TNF-α at 1:1,000 dilution, anti-MMP-3 at 1:3,000 dilution, β-actin at 1:3,000 dilution in TBST containing 5% dry milk; Calbiochem). After an overnight incubation at 4 °C, the membrane was washed three times with TBST, and incubated with a peroxidase-conjugated goat antimouse immunoglobulin G antibody (1:5,000 dilution in TBST; Waters, Milford, MA) at room temperature for 1 hour. Bands were visualized by using electrochemiluminescence detection system (GE Healthcare), and scanned with a HP Scanjet G3010.

Histological examination. At 7 weeks after the first immunization, the knee joint tissues treated with/without psi-tGC-NPs were isolated from each mouse (n = 3 per each group), immediately fixed in Fixative/Decalcifier (Thermo Fisher Scientific, Waltham, MA) for three days, dehydrated in graded ethanol series, and embedded in paraffin. Sections of the tissues (5-µm thick) were stained with safranin O/fast green. For the immunohistochemical analysis, the decalcified knee joints were frozen in OCT compound. Frozen tissue blocks were cut into 5-µm–thick sections using a cryostat (Leica, Nussloch, Germany), and then sections were mounted on SuperFrost Plus microscope slides (Fisher Scientific, Fair Lawn, NJ). Immunostaining was performed with the mouse monoclonal anti-TNF-α antibody (1:1,000 diluted), peroxidase-conjugated goat antimouse immunoglobulin G (H+L) secondary antibody (1:5,000 diluted), and Dako REAL EnVision Detection System, Peroxidase/DAB+ (Dako, Glostrup, Denmark), followed by counterstaining with hematoxylin (Sigma).

Statistical analysis. Statistical analyses were performed using analysis of variance to compare each week after RA induction and serum TNF-α levels. P values less than 0.05 were considered statistically significant.

Acknowledgments

This study was funded by the Global Research Laboratory (GRL; 2010-0019863) Project, the Fusion Technology Project (2010-50201), Regional Technology Innovation Program of the Ministry of Knowledge Economy (MEK; RTI04-01-01), National R&D Program for Cancer Control of Ministry for Health and Welfare (1020260), M.D.-Ph.D. Program (2010-0019864), Global Innovative Research Center (GiRC) program, Company-Researcher cowork program of Small & Medium Business Administration (SD122946) and the Intramural Research Program (Theragnosis, Global RNAi Carrier Initiative, and KIST Young Fellow) of KIST. The authors declared no conflict of interest.

References

  1. McInnes IB, Schett G. Cytokines in the pathogenesis of rheumatoid arthritis. Nat Rev Immunol. 2007;7:429–442. doi: 10.1038/nri2094. [DOI] [PubMed] [Google Scholar]
  2. Arend WP. Physiology of cytokine pathways in rheumatoid arthritis. Arthritis Rheum. 2001;45:101–106. doi: 10.1002/1529-0131(200102)45:1<101::AID-ANR90>3.0.CO;2-7. [DOI] [PubMed] [Google Scholar]
  3. Feldmann M, Maini RN. Lasker Clinical Medical Research Award. TNF defined as a therapeutic target for rheumatoid arthritis and other autoimmune diseases. Nat Med. 2003;9:1245–1250. doi: 10.1038/nm939. [DOI] [PubMed] [Google Scholar]
  4. Catrina AI, Lampa J, Ernestam S, af Klint E, Bratt J, Klareskog L, et al. Anti-tumour necrosis factor (TNF)-alpha therapy (etanercept) down-regulates serum matrix metalloproteinase (MMP)-3 and MMP-1 in rheumatoid arthritis. Rheumatology (Oxford) 2002;41:484–489. doi: 10.1093/rheumatology/41.5.484. [DOI] [PubMed] [Google Scholar]
  5. Zeisel MB, Druet VA, Wachsmann D, Sibilia J. MMP-3 expression and release by rheumatoid arthritis fibroblast-like synoviocytes induced with a bacterial ligand of integrin alpha5beta1. Arthritis Res Ther. 2005;7:R118–R126. doi: 10.1186/ar1462. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Elliott MJ, Maini RN, Feldmann M, Kalden JR, Antoni C, Smolen JS, et al. Randomised double-blind comparison of chimeric monoclonal antibody to tumour necrosis factor alpha (cA2) versus placebo in rheumatoid arthritis. Lancet. 1994;344:1105–1110. doi: 10.1016/s0140-6736(94)90628-9. [DOI] [PubMed] [Google Scholar]
  7. Airò P, Antonioli CM, Vianelli M, Toniati P. Anti-tumour necrosis factor treatment with infliximab in a case of giant cell arteritis resistant to steroid and immunosuppressive drugs. Rheumatology (Oxford) 2002;41:347–349. doi: 10.1093/rheumatology/41.3.347. [DOI] [PubMed] [Google Scholar]
  8. Gardam MA, Keystone EC, Menzies R, Manners S, Skamene E, Long R, et al. Anti-tumour necrosis factor agents and tuberculosis risk: mechanisms of action and clinical management. Lancet Infect Dis. 2003;3:148–155. doi: 10.1016/s1473-3099(03)00545-0. [DOI] [PubMed] [Google Scholar]
  9. Kole R, Krainer AR, Altman S. RNA therapeutics: beyond RNA interference and antisense oligonucleotides. Nat Rev Drug Discov. 2012;11:125–140. doi: 10.1038/nrd3625. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Martinez J, Patkaniowska A, Urlaub H, Lührmann R, Tuschl T. Single-stranded antisense siRNAs guide target RNA cleavage in RNAi. Cell. 2002;110:563–574. doi: 10.1016/s0092-8674(02)00908-x. [DOI] [PubMed] [Google Scholar]
  11. Elbashir SM, Harborth J, Lendeckel W, Yalcin A, Weber K, Tuschl T. Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured mammalian cells. Nature. 2001;411:494–498. doi: 10.1038/35078107. [DOI] [PubMed] [Google Scholar]
  12. Oh YK, Park TG. siRNA delivery systems for cancer treatment. Adv Drug Deliv Rev. 2009;61:850–862. doi: 10.1016/j.addr.2009.04.018. [DOI] [PubMed] [Google Scholar]
  13. Bumcrot D, Manoharan M, Koteliansky V, Sah DW. RNAi therapeutics: a potential new class of pharmaceutical drugs. Nat Chem Biol. 2006;2:711–719. doi: 10.1038/nchembio839. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Han SE, Kang H, Shim GY, Suh MS, Kim SJ, Kim JS, et al. Novel cationic cholesterol derivative-based liposomes for serum-enhanced delivery of siRNA. Int J Pharm. 2008;353:260–269. doi: 10.1016/j.ijpharm.2007.11.026. [DOI] [PubMed] [Google Scholar]
  15. Akinc A, Zumbuehl A, Goldberg M, Leshchiner ES, Busini V, Hossain N, et al. A combinatorial library of lipid-like materials for delivery of RNAi therapeutics. Nat Biotechnol. 2008;26:561–569. doi: 10.1038/nbt1402. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Hong CA, Kim JS, Lee SH, Kong WH, Park TG, Mok H, et al. Reductively dissociable siRNA-polymer hybrid nanogels for efficient targeted gene silencing. Adv Funct Mater. 2013;23:316–322. [Google Scholar]
  17. Ye C, Bhan AK, Deshpande V, Shankar P, Manjunath N. Silencing TNF-a in macrophages and dendritic cells for arthritis treatment. Scand J Rheumatol. 2013;42:266–269. doi: 10.3109/03009742.2013.777779. [DOI] [PubMed] [Google Scholar]
  18. Crielaard BJ, Lammers T, Schiffelers RM, Storm G. Drug targeting systems for inflammatory disease: one for all, all for one. J Control Release. 2012;161:225–234. doi: 10.1016/j.jconrel.2011.12.014. [DOI] [PubMed] [Google Scholar]
  19. Jere D, Kim JE, Arote R, Jiang HL, Kim YK, Choi YJ, et al. Akt1 silencing efficiencies in lung cancer cells by sh/si/ssiRNA transfection using a reductable polyspermine carrier. Biomaterials. 2009;30:1635–1647. doi: 10.1016/j.biomaterials.2008.12.005. [DOI] [PubMed] [Google Scholar]
  20. Wiradharma N, Khan M, Tong YW, Wang S, Yang YY. Self-assembled cationic peptide nanoparticles capable of inducing efficient gene expression in vitro. Adv Funct Mater. 2008;18:943–951. [Google Scholar]
  21. Bolcato-Bellemin AL, Bonnet ME, Creusat G, Erbacher P, Behr JP. Sticky overhangs enhance siRNA-mediated gene silencing. Proc Natl Acad Sci USA. 2007;104:16050–16055. doi: 10.1073/pnas.0707831104. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Mok H, Lee SH, Park JW, Park TG. Multimeric small interfering ribonucleic acid for highly efficient sequence-specific gene silencing. Nat Mater. 2010;9:272–278. doi: 10.1038/nmat2626. [DOI] [PubMed] [Google Scholar]
  23. Lee SJ, Huh MS, Lee SY, Min S, Lee S, Koo H, et al. Tumor-homing poly-siRNA/glycol chitosan self-cross-linked nanoparticles for systemic siRNA delivery in cancer treatment. Angew Chem Int Ed Engl. 2012;51:7203–7207. doi: 10.1002/anie.201201390. [DOI] [PubMed] [Google Scholar]
  24. Lee SY, Huh MS, Lee S, Lee SJ, Chung H, Park JH, et al. Stability and cellular uptake of polymerized siRNA (poly-siRNA)/polyethylenimine (PEI) complexes for efficient gene silencing. J Control Release. 2010;141:339–346. doi: 10.1016/j.jconrel.2009.10.007. [DOI] [PubMed] [Google Scholar]
  25. Mulherin D, Fitzgerald O, Bresnihan B. Synovial tissue macrophage populations and articular damage in rheumatoid arthritis. Arthritis Rheum. 1996;39:115–124. doi: 10.1002/art.1780390116. [DOI] [PubMed] [Google Scholar]
  26. Onishi H, Machida Y. Biodegradation and distribution of water-soluble chitosan in mice. Biomaterials. 1999;20:175–182. doi: 10.1016/s0142-9612(98)00159-8. [DOI] [PubMed] [Google Scholar]
  27. Ryu JH, Lee A, Chu JU, Koo H, Ko CY, Kim HS, et al. Early diagnosis of arthritis in mice with collagen-induced arthritis, using a fluorogenic matrix metalloproteinase 3-specific polymeric probe. Arthritis Rheum. 2011;63:3824–3832. doi: 10.1002/art.30628. [DOI] [PubMed] [Google Scholar]
  28. Paleolog EM. Angiogenesis in rheumatoid arthritis. Arthritis Res. 2002;4 suppl 3:S81–S90. doi: 10.1186/ar575. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Baluk P, Hirata A, Thurston G, Fujiwara T, Neal CR, Michel CC, et al. Endothelial gaps: time course of formation and closure in inflamed venules of rats. Am J Physiol. 1997;272:L155–L170. doi: 10.1152/ajplung.1997.272.1.L155. [DOI] [PubMed] [Google Scholar]
  30. Yuan F, Quan LD, Cui L, Goldring SR, Wang D. Development of macromolecular prodrug for rheumatoid arthritis. Adv Drug Deliv Rev. 2012;64:1205–1219. doi: 10.1016/j.addr.2012.03.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Whitehead KA, Langer R, Anderson DG. Knocking down barriers: advances in siRNA delivery. Nat Rev Drug Discov. 2009;8:129–138. doi: 10.1038/nrd2742. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Kobayashi K, Takahashi N, Jimi E, Udagawa N, Takami M, Kotake S, et al. Tumor necrosis factor alpha stimulates osteoclast differentiation by a mechanism independent of the ODF/RANKL-RANK interaction. J Exp Med. 2000;191:275–286. doi: 10.1084/jem.191.2.275. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Howard KA, Paludan SR, Behlke MA, Besenbacher F, Deleuran B, Kjems J. Chitosan/siRNA nanoparticle-mediated TNF-alpha knockdown in peritoneal macrophages for anti-inflammatory treatment in a murine arthritis model. Mol Ther. 2009;17:162–168. doi: 10.1038/mt.2008.220. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Saito G, Swanson JA, Lee KD. Drug delivery strategy utilizing conjugation via reversible disulfide linkages: role and site of cellular reducing activities. Adv Drug Deliv Rev. 2003;55:199–215. doi: 10.1016/s0169-409x(02)00179-5. [DOI] [PubMed] [Google Scholar]
  35. Hassan MQ, Hadi RA, Al-Rawi ZS, Padron VA, Stohs SJ. The glutathione defense system in the pathogenesis of rheumatoid arthritis. J Appl Toxicol. 2001;21:69–73. doi: 10.1002/jat.736. [DOI] [PubMed] [Google Scholar]
  36. Huh MS, Lee SY, Park S, Lee S, Chung H, Lee S, et al. Tumor-homing glycol chitosan/polyethylenimine nanoparticles for the systemic delivery of siRNA in tumor-bearing mice. J Control Release. 2010;144:134–143. doi: 10.1016/j.jconrel.2010.02.023. [DOI] [PubMed] [Google Scholar]
  37. Brand DD, Latham KA, Rosloniec EF. Collagen-induced arthritis. Nat Protoc. 2007;2:1269–1275. doi: 10.1038/nprot.2007.173. [DOI] [PubMed] [Google Scholar]

Articles from Molecular Therapy are provided here courtesy of The American Society of Gene & Cell Therapy

RESOURCES