An orally administrated nucleotide-delivery vehicle targeting colonic macrophages for the treatment of inflammatory bowel disease

Biomaterials 48 (2015) 26e36

Contents lists available at ScienceDirect

Biomaterials journal homepage: www.elsevier.com/locate/biomaterials

An orally administrated nucleotide-delivery vehicle targeting colonic macrophages for the treatment of inflammatory bowel disease Zhen Huang a, Jingjing Gan a, Lixin Jia a, Guangxing Guo a, Chunming Wang a, b, Yuhui Zang a, Zhi Ding a, Jiangning Chen a, Junfeng Zhang a, **, Lei Dong a, * a

State Key Laboratory of Pharmaceutical Biotechnology, School of Life Sciences, Nanjing University, Nanjing 210093, China State Key Laboratory of Quality Research in Chinese Medicine, Institute of Chinese Medical Sciences, University of Macau, Macau SAR, Macau 999078, China

b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 19 August 2014 Accepted 20 January 2015 Available online

Tumor necrosis factor-alpha (TNF-a) plays a central role in the pathogenesis of inflammatory bowel disease (IBD). Anti-TNF-a therapies have shown protective effects against colitis, but an efficient tool for target suppression of its secretion - ideally via oral administration - remains in urgent demand. In the colon tissue, TNF-a is mainly secreted by the colonic macrophages. Here, we report an orallyadministrated microspheric vehicle that can target the colonic macrophages and suppress the local expression of TNF-a for IBD treatment. This vehicle is formed by cationic konjac glucomannan (cKGM), phytagel and an antisense oligonucleotide against TNF-a. It was given to dextran sodium sulfate (DSS) colitic mice via gastric perfusion. The unique swelling properties of cKGM enabled the spontaneous release of cKGM& antisense nucleotide (ASO) nano-complex from the phytagel scaffold into the colon lumen, where the ASO was transferred into colonic macrophages via receptor-mediated phagocytosis. The treatment significantly decreased the local level of TNF-a and alleviated the symptoms of colitis in the mice. In summary, our study demonstrates a convenient, orally-administrated drug delivery system that effectively targets colonic macrophages for suppression of TNF-a expression. It may represent a promising therapeutic approach in the treatment of IBD. © 2015 Elsevier Ltd. All rights reserved.

Keywords: Inflammatory bowel disease Colonic macrophages TNF-a Antisense oligonucleotide Targeting delivery

1. Introduction Inflammatory bowel disease (IBD) is a group of chronic disorders within the gastrointestinal tract, caused by dysregulated immune responses [1]. Notably, tumor necrosis factor a (TNF-a), a versatile inflammatory cytokine involved in many physiological processes, plays a central role in the pathogenesis of IBD. In the colon tissue, TNF-a is mainly produced by colonic macrophages [2]. During the colitic process, the high level of TNF-a promotes the production of other pro-inflammatory cytokines, increases the leukocytes migration via stimulating the expression of adhesion molecules by endothelial cells and induces the formation of granuloma [3]. Current therapeutic strategies, aimed at blocking TNF-a for the treatment of IBD, have shown great promises in cell models but * Corresponding author. Tel.: þ86 25 89681320. ** Corresponding author. Tel.: þ86 25 83592502. E-mail addresses: [email protected] (J. Zhang), [email protected] (L. Dong). http://dx.doi.org/10.1016/j.biomaterials.2015.01.013 0142-9612/© 2015 Elsevier Ltd. All rights reserved.

achieved limited clinical success [4]. A notable example is infliximab, a chimeric monoclonal antibody targeting TNF-a, which has exhibited satisfactory performances in alleviating IBD in clinical trials [5]. However, in systemically, non-selectively blocking TNF-a, this drug also brought about obvious side effects, such as causing immunodeficiency-related infections and generating antibodies against the drugs [6,7]. Therefore, anti-TNF-a therapies have proven effective, but they need to be limited at the specific site of inflammation. To address this unmet medical need, we aimed to develop an orally-administrated vehicle for targeted delivery of anti-TNF-a nucleotides to colonic macrophages. The microspheric vehicle comprises three components - i) cationic konjac glucomannan (cKGM), ii) phytagel, and iii) an antisense nucleotide (ASO) specifically against TNF-a. cKGM can provide three functions. First, it is cationized to conjugate the negatively charged ASO to form the cKGM&ASO nano-complex core. Second, it has high mannose moieties that can be recognized by mannose receptors (MR) that are abundant on the macrophages [8]. The interaction between MR

Z. Huang et al. / Biomaterials 48 (2015) 26e36

and cKGM can mediate the phagocytosis of the nano-complex core into macrophages. Third, cKGM has uniquely strong waterabsorption and swells almost indefinitely [9]. In contrast, phytagel swells poorly. By mixing cKGM and phytagel, we aimed to fabricate a microspheric system that could expectedly collapse in the colon lumen, because of the contrast swelling properties of the two materials. Consequently, the core cKGM&ASO complex can be released and internalized into macrophages in the local tissue. In the present study, we fabricated this microspheric system, tested its drug release and transfection efficiency, and evaluated its therapeutic performance in a dextran sodium sulfate (DSS) induced murine colitis model. Our design offers a new, orally-administrative approach for targeted gene delivery to colonic macrophages, which may open up new avenues for development of patient-friendly gene therapy strategies for the treatment of IBD. 2. Materials and methods 2.1. Synthesis of materials and reagents KGM was obtained from Megazyme (Wicklow, Ireland). cKGM was prepared by incorporating ethylenediamine within the hydroxyl groups of KGM by an N, N'carbonyldiimidazole (CDI) activation method with different ratios between KGM and CDI, as previously report [10]. The cationic degrees were determined by element analysis of nitrogen (CHNeO-Rapid, Hanau, Germany). Besides this, KGM and cKGM were characterized by fourier transform infrared spectroscopy (FT-IR, Nicolet, 170SX, Thermal Fisher Scientific, USA) and NMR spectrometer (Bruker AVANCE DRX-500, Bruker Corporation, USA) to obtain the structure information. Phytagel, lipopolysaccharide (LPS), mannanase and cellulase were purchased from Sigma (St. Louis, MO). Other chemical reagents were purchased from Sangon Biotech (Shanghai, China). The phosphorothioate-modified antisense oligonucleotide against TNF-a, ISIS25302 was synthesized by Life Technologies (La Jolla, CA, USA). The sequence of ISIS25302 is: 50 -AACCCATCGGCTGGCACCAC-3'. The ASO sequences have been proven to be effective in our previous study [11]. As a scrambled control oligonucleotide (SCO), ISIS 18154 (50 -TCAAGCAGTGCCACCGATCC-30 ) was used. Alexa 546labeled ASO (Alexa 546-ASO) was synthesized for in vitro transfection and in vivo cell localization and tissue distribution. 2.2. Preparation and characterization of cKGM&ASO complexes and the gel retardation assay cKGM&ASO complexes were prepared by mixing an aqueous solution of cKGM with ASO according to the indicated weight ratio. Briefly, different weights of cKGM were dissolved in 500 ml saline, mixed with 500 ml saline containing 1000 mg ASO and incubated for 60 min at room temperature to obtain different complexes of cKGM and ASO. The diameters and zeta potential of cKGM&ASO complexes were examined by a 90 Plus Particle Sizer (Brookhaven Instruments, Holtsville, NY, USA). Different complexes were prepared in saline at different N/P ratios. Transmission electron microscope (TEM, JEM-200CX, JEOL, Tokyo, Japan) was applied to observe the morphology of the complexes. To examine whether the binding ability of cKGM to ASO, cKGM with different cationization degree was complexed with ASO. After incubating at room temperature, the cKGM/ASO complex solution of 10 ml was added to 2 ml of loading buffer and applied to a 3% agarose gel in a Trisborate-ethylenediaminetetraacetic acid buffer solution (TBE, pH 8.3) containing 0.1 mg/ml ethidium bromide. The electrophoretic evaluation of the complex was carried out in TBE solution at 100 V for 30 min. The gel was imaged with a UV transilluminator (Gel Doc, 2000, BioRad Laboratories, Hercules, CA, USA). To examine whether the enzyme from gut flora degraded the cKGM&ASO complex, 1 mg/ml nano-complex (McKGM/MASO ¼ 5) was incubated in 100 ml 0.05 M sodium citrate-hydrochloric acid buffer solution (pH ¼ 5.0) containing 6 U/ml mannanase and 5 U/ml cellulase for 24 h at 37  C. Then, the digested cKGM&ASO nano-complex was also performed with gel retardation assay. Moreover, 1 mg/ml cKGM or KGM was also digested by mannanase and cellulase, the total reducing sugar degraded by enzymes at indicated time points was examined by dinitrosalicylic acid colorimetric (DNS) protocol [12]. 2.3. Fabrication cKGM/phytagel microsphere (KPM) containing ASO The cKGM/phytagel microspheres were prepared using water-in-oil (W/O) emulsion method as previously described with minor modification [13]. Briefly, different ratios of phytagel, cKGM and ASO were completely dissolved in deionized water after being heated at 80  C for 15 min. The hot polysaccharide solution (4 ml) containing 8 mg ASO was quickly added to pre-heated 90  C peanut oil (50 ml) under stirring at 600 rpm for 30 min. Then the oil mixture was finally transferred into 200 ml amount of 1% CaCl2 solution for 5 min to enhance the gel strength. The microspheres containing ASO or Alexa 546-ASO were collected by centrifugation and further washed with acetone before microscopic observation. The dried

27

microspheres were obtained by gradient alcohol dehydration. The morphology of microsphere was observed by scanning electron microscope (SEM, Se3400N II, Hitachi, Tokyo, Japan). To evaluate the ASO loading efficiency, 100 mg dried KPM containing Alexa 546-ASO was pulverized by homogenizer, added to 100 ml 0.05 M sodium citrate-hydrochloric acid buffer solution (pH ¼ 5.0) containing 60 U/ml mannanase and 50 U/ml cellulase and was incubated at 37  C until a clear solution was obtained. The concentration of ASO was quantified by examining the fluorescence intensity at 573 nm. 2.4. In vitro drug release and swelling studies In order to simulate the alimentary tract conditions, drug release and swelling studies were preformed in an orbital shaker (Thermo Scientific, Waltham, Massachusetts, USA) at 37  C and 60 rpm. The different compositions of cKGM and phytagel formulated microspheres containing Alexa 546-ASO (dry weight: 100 mg) was first tested in simulated gastric fluid (NaCl/HCl buffer, pH 1.2) for the first 2 h. Then, the dissolution medium was changed to simulated intestinal fluid (K2HPO4/NaOH buffer, pH 7.2) for 4 h and finally replaced by simulated colonic fluid (K2HPO4/NaOH buffer, pH 6.8) for 18 h. The shape of KPM at different stages was also observed by Nikon microscope (TE-2000U, Tokyo, Japan) and its diameter was determined via Nis-element basic research software (Nikon). The dissolution medium was collected at indicated time points to quantify the released ASO by examining the fluorescence intensity at 573 nm. TEM was performed to study the morphology and size of contents released from KPM. The swelling index of KPM was calculated according to the following formula: Swelling indexð%Þ ¼ ðWt  WiÞ=Wi100% where Wi is the weight of initial dried KPM and Wt is the weight of KPM at indicated time points which was removed the free water from its surface. 2.5. Cell culture and transfection experiments Mouse leukemia monocytic macrophage cell line, Raw 264.7 cells and mouse colon cell line, CT-26 cells (ATCC, Manassas, VA, USA) were cultured in RPMI-1640 containing 10% FBS. The cell cultures were incubated in room air with 5% CO2 at 37 C and 95% humidity. Raw 264.7 cells and CT-26 cells were cultured in 24-well plates for the transfection experiment. Before transfection, complete medium was removed, and cells were rinsed once with PBS. The naked ASO, cKGM&ASO complex or microsphere released contents (MRC) containing 1 mg of ASO or Alexa 546-ASO was diluted with 0.3 ml medium and was used to refill the well. After incubation at 37  C for 6 h, the medium containing the complex was removed. The cells were rinsed twice with PBS and refilled with medium. Transfection of the lipofectamine&ASO complex was performed as controls according to the manufacturer's protocol. The transferred cells were examined by Nikon confocal microscope (C2þ, Nikon) and analyzed using Nis-element advanced research software (Nikon). To quantify the transfection efficiency, transferred cells were collected and analyzed via flow cytometer (BD Biosciences, San Jose, CA, USA). To further investigate the suppression effect of MRC on TNF-a production, macrophages transfected with naked ASO, naked ASO with simulated gastric fluid pre-treatment for 2 h &lipofectamine, cKGM&ASO complex, lipofectamine&ASO complex, MRC or cKGM/ mismatch ASO complex were further stimulated by 100 ng/ml LPS for another 6 h. The supernatant was collected to determine TNF-a concentration by ELISA kit (R&D systems, Minneapolis, MN, USA). 2.6. Cell viability assay Cytotoxicity of cKGM was examined using the cell counting kit-8 (CCK-8, Dojindo, Tokyo, Japan). Raw 264.7 cells and CT-26 cells were seeded in a 96-well plate at a number of 5000 cells/well and cultured overnight. Then the cells were incubated in 100 ml serum-free medium containing the set amount of cKGM or cKGM&ASO complex. The concentration of cKGM applied in this experiment is 20, 50, 100, 200, 500 and 1000 mg/ml, respectively and 50 mg/ml PEI was used as a control of cKGM. cKGM&ASO complex (McKGM/MASO ¼ 5:1) at the concentration of 20, 50, 100, 200, 500 and 1000 mg/ml (containing ASO of 4, 10, 20, 40, 100 and 200 mg/ml), respectively were also tested. PEI/ASO complex (MPEI/MASO ¼ 10:1) at the concentration of 50 mg/ml (containing ASO of 5 mg/ml) was used as the control. After 6 h treatment, the medium was removed and the cells were rinsed twice with PBS. The wells were refilled with complete medium and the cells were cultured for another 24 h. Next, 10 ml CCK-8 solution was added into the well and was further incubated for another 1 h at 37  C. Absorbance at 450 nm was measured with a €nnedorf, Switzerland). microplate reader (Tecan Group Ltd., Ma 2.7. Establishment of dextran sodium sulfate (DSS)-induced colitic model Female C57/B6 mice of the same background were obtained from Laboratory Animal Center of Nanjing University (Nanjing, China). All animals received human care according to Chinese legal requirements. Acute DSS colitis was induced by addition of 5% (w/w) DSS (MW 36,000e50,000, MP Biomedicals, Santa Ana, CA, USA) in the drinking water. Calculate 5 ml DSS solution per mouse per day and DSS solutions were replaced every two days. Control mice received the same drinking

28

Z. Huang et al. / Biomaterials 48 (2015) 26e36

On day 1, 3, 5 and 7 during the colitis model induced process, mice was administrated of 0.1 g 3# KPM (2.25% KGM and 0.75% phytagel) containing ASO or scrambled ASO (SCO) (5 mg ASO/kg body weight) in 200 ml saline by oral gavage. DSS group received only saline as a control. The behavior and weight of the mice was observed daily, and the colons were excised at day 8 for macroscopic observation, length measurement, histopathological analysis, myeloperoxidase (MPO) activity and cytokines analysis. The disease activity index (DAI) was used to evaluate the grade of intestinal inflammation. For histopathological examination, histological score of H&E staining sections was graded from 0 to 4 according to a previous report [14].

group in ester bond) existed in the spectrum of cKGM when compared with that of KGM, also indicating that ethylenediamine was successfully connected to KGM (Fig. 1A and B). The electrophoresis of nano-complexes formulated by ASO and cKGM with different ethylenediamine contents at McKGM/MASO ratio of 3.0, 5.0 and 7.0 was shown in Fig. 1C. The migration of ASO toward was gradually retarded with the increase of both ethylenediamine contents and McKGM/MASO ratio. When [CDI]/[OH] ratio was increased to 7.0, ASO was completely retarded at McKGM/MASO ratio of 5.0. As shown in Table 2, either higher [CDI]/[OH] ratio or higher McKGM/MASO ratio could reduce the diameter and increase the zeta potential of cKGM/ASO complex. When cKGM/ASO complex was formulated by cKGM with [CDI]/[OH] ratio of 7.0 modification and at McKGM/MASO ratio of 5.0, the diameter of the complex was about 200 nm and the zeta potential was þ16.7 mV (Table 2). Therefore, cKGM with [CDI]/[OH] ratio of 7.0 and cKGM associated with ASO at McKGM/MASO ratio of 5.0 or higher was further applied in cell transfection and animal treatment. The morphology of cKGM&ASO complex (McKGM/MASO ¼ 5.0) was observed by TEM. Results in Fig. 1D indicated that the diameter of cKGM&ASO complex was about 200 nm and its shape was spherical. The cytotoxicity of cKGM and cKGM&ASO complex was further examined in CT-26 cells and RAW 264.7 cells. Results shown in Fig. 1EeF indicated that both cKGM alone and cKGM&ASO exhibited low toxicity against the two types of cells even at the highest concentration (cell viability> 80%). Meanwhile, PEI/ASO complex of 50 mg/ml already exerted high cytotoxicity (cell viability<40%).

2.10. Immunofluorescence staining

3.2. In vitro transfection of cKGM&ASO complex

water without the addition of DSS. At day 8, the mice were sacrificed and tissues were processed as described below. 2.8. Tissue-distribution and cellular localization of ASO cKGM&Alexa 546-ASO nano-complex and different compositions of cKGM and phytagel formulated microspheres containing Alexa 546-ASOin saline solutions were separately given to control healthy mice or DSS-bearing mice via intragastric administration at a dose of 5 mg ASO/kg body weight. Stomach, small intestine and colon at indicated time points were harvested from the experimental colitic mice. To determine the cellular localization of ASO, frozen sections of the colons were harvested 24 h after administration of KPM and were stained with a rat anti-mouse F4/ 80-antibody (eBioscience, San Diego, CA, USA) at 4  C overnight. The secondary antibody, Alexa 488-labeled donkey anti-rat (Life Technologies), was applied at room temperature for 45 min followed by the staining of nuclei with 4, 6-diamidino2-phenylindole (DAPI, Sigma). Frozen sections were imaged using a Nikon confocal microscope (TE2000-U, Nikon). For the quantification of cellular uptake, colonic macrophages and colonic epithelial cells were isolated according to previously described [14]. Alexa 546-ASO in the organs and cells was extracted according to a reported method and quantified by examining the fluorescence intensity at 573 nm [15]. 2.9. Treatment and assessment of colonic inflammatory changes

For mannose receptor expression studies, cells and frozen sections of the colons were stained by a rabbit anti-mouse mannose receptor antibody (Santa Cruz, Dallas, TX USA), rat anti-mouse F4/80 antibody or mouse anti-mouse pan-cytokeratin antibody (eBioscience) at 4  C overnight. To detect TNF-a level in the inflamed colon, the colons cryosections were stained by a rabbit anti-mouse TNF-a antibody (Abcam, Cambridge, MA, USA) at 4  C overnight. The secondary antibody Alexa 488 labeled donkey anti-rabbit, Alexa 546 labeled donkey anti-rat, Alexa 546 labeled donkey anti-mouse or Alexa 546 labeled donkey anti-rabbit (Life Technologies) was applied at room temperature for 45 min followed by the staining of nuclei with DAPI. Cells and frozen sections were photographed using Nikon confocal microscope. 2.11. Statistics Results are expressed as the mean ± standard error of the mean (S.E.M). The differences between groups were analyzed by ManneWhitney U test and, if appropriate, by KruskaleWallis ANOVA test. A value of p  0.05 was considered significant. The survival curves were analyzed by the KaplaneMeyer log-rank test.

3. Results 3.1. Preparation of cKGM&ASO nano-complex and cell viability analysis Elemental analysis indicated that ethylenediamine was successfully introduced into KGM and the modification proportion could be changed by adjusting the amount of CDI added initially (Table 1). The strong signal at d 2.69 clearly indicated the existence of eNHeCH2-CH2-NH2 in Supplementary Fig. 1. The FT-IR spectra indicated the differences at 1708.1 cm1 (indicating the existence of eCONHegroup) and 1261.0 cm1 (indicating the existence of CeO Table 1 cKGM preparation and characterization. KGM molecular weight (Da)

cKGM [CDI]/[OH]

Percentage of ethylenediamine introduced to KGM (%)

200,000

3.0 5.0 7.0

3.81% 4.54% 7.85%

As the mannose receptors can promote phagocytosis, we first examined the mannose receptor expressing cells via immunofluoresence staining. As shown in Fig. 2A-B, mannose receptor was expressed on macrophages but not on colonic epithelial cells both in vitro and in vivo. Alexa 546-labeled ASO was used to observe the transfection efficiency of cKGM&ASO in macrophages and colonic epithelial cells. Results are shown in Fig. 2C. cKGM greatly enhanced the transfection of ASO into RAW264.7 cells. However, its transfect efficiency in CT-26 cells was much lower than lipofectamine and was similar to that of naked ASO. FACS analysis was further used to quantify the transfect efficiency. In consistent with the fluoresce result, about 75% RAW 264.7 cells were transfected with cKGM/Alexa 546-labeledASO but only about 8% CT-26 cells were positive of Alexa 546 (Fig. 2DeE). These results suggested that cKGM may promote ASO transfer partly through mannosemediated endocytosis. 3.3. KPM formulation and in vitro release assay The SEM results in Fig. 3A showed KPM with the composition of 0.5% phytagel and 2.5% cKGM was spherical and its size was about 650 mm, with a rough surface. The bright filed image and fluorescent image of the same KPM in Fig. 3B indicated that ASO was successfully entrapped in KPM and the loading efficiency of ASO in KPM was about 65% of the initial added ASO, with no correlation with KPM compositions (Supplementary Table 1). In vitro studies were performed in a series of simulated digestive fluid to investigate the swelling and release behaviors of different KPMs. As shown in Supplementary Table 2, the maximum swelling index of microspheres which contained high percentage of cKGM occurred at shorter times than microspheres with small amount of cKGM. Four representative KPMs were selected: 1# KPM (0.1% phytagel, 2.9% cKGM); 2# KPM (0.3% phytagel, 2.7% cKGM); 3# KPM (0.5% phytagel, 2.5% cKGM) and 4# KPM (2% phytagel, 1% cKGM). As shown in Fig. 3CeD, 1# KPM rapidly reached the peak point of

Z. Huang et al. / Biomaterials 48 (2015) 26e36

29

Fig. 1. Characterization of cKGM and cKGM/ASO nano-complex. FT-IR spectra of KGM (A) and cKGM (B). (C) Electrophoretic mobility of ASO in cKGM/ASO nano-complexes, cKGM with [CDI]/[OH] ratio of 3.0, 5.0 or 7.0. Lanes 1, 2, 3 and 4 are naked ASO, complexes at McKGM/MASO ratio of 3.0, 5.0 and 7.0, respectively. (D) TEM image of complex self-assembled from cKGM with [CDI]/[OH] ratio of 7.0 and ASO at McKGM/MASO ratio of 5.0. Cell viability of CT-26 and RAW 264.7 cells after incubation with cKGM at different concentrations and PEI at 50 mg/ml (E) and cKGM&ASO complexes at different concentrations with McKGM/MASO ratio of 5.0 and PEI/ASO complexes at 50 mg/ml with MPEI/MASO ratio of 10.0 (F). Five samples were analyzed per condition, and experiments were performed in triplicate.

swelling index and the erosion of 1# KPM has already taken place at 2 h, meanwhile 2# KPM began its degradation during the 2e6 h period. 3# KPM swelled to the top at 6 h after incubation, whereas 4# KPM still didn't reach its peak point until 24 h. The size of 3# KPM at 24 h was significantly larger than 4# KPM (Fig. 3E). In vitro release studies also revealed that more than 60% ASO were released from 3# KPM between 6 h and 24 h, implying 3# KPM may be the most appropriate vector for colon targeting (Fig. 3F). 3.4. Microspheres released contents (MRC) characterization As shown in Fig. 4A, TEM indicated that MRC from 3# KPM were nano-complexes and had the same morphology and size as directly formulated cKGM&ASO complexes. To examine whether MRC could still enter macrophages with high efficacy, transfection experiment was performed. Both the fluorescent images and FACS analysis demonstrated that the transfect efficiency of MRC was equivalent to that of cKGM&ASO and lipofectamine&ASO (Fig. 4BeC). To examine whether ASO is affected by the processing conditions to form KPM and the acidic environment in the stomach, MRC

Table 2 Effective mean diameter and zeta potential of cKGM&ASO complexes. cKGM [CDI]/[OH]

McKGM/MASO

Effective mean diameter (nm)

3.0

3 5 7 3 5 7 3 5 7

982.7 783.7 699.5 653.6 348.6 245.1 296.4 205.1 196.2

5.0

7.0

± ± ± ± ± ± ± ± ±

78.2 67.4 57.3 85.3 71.4 43.9 69.7 32.2 21.5

Zeta potential (mV) 13.14 9.32 2.35 1.98 2.53 5.21 8.4 15.2 16.7

± ± ± ± ± ± ± ± ±

2.61 1.83 0.73 0.16 0.06 0.36 1.13 2.25 1.72

from KPM with a series of simulated digestive fluid treatment and naked ASO pre-incubated with acidic buffer then complexed with lipofectamine was transfected into LPS activated RAW 264.7 cells. ASO with acidic buffer pre-treatment had very weak suppressing effect on TNF-a secretion, whereas MRC could still block TNF-a production (Fig. 4D). As KGM may be degraded by the enzymes which were produced by colon flora, we examined the degrading ability of glycosidase against cKGM. We observed few cKGM could be digested at 24 h points, meanwhile large amounts of KGM could be degraded into monosaccharide (Fig. 4E). The electrophoresis pattern also showed cKGM&ASO with enzyme digestion still stucked in the sample well just like cKGM&ASO without any treatment (Fig. 4F).

3.5. Tissue distribution and cellular localization The tissue distribution of ASO in mice at 2 h, 6 h and 24 h after intragastric administration was shown in Fig. 5AeC. As shown in Fig. 5A, D, significant accumulation of ASO in the stomach from mice with cKGM&ASO nanoparticle or 1# KPM treatment at 2 h. Fluorescence imaging and quantification analysis performed at 6 h after ASO administration indicated that 2# KPM dramatically increased the level of ASO in the small intestine, whereas 3# KPM efficiently enhanced the accumulation of ASO in the colon at 24 h after drug given (Fig. 5B, C and E). As a control, the amount of ASO in all three organs from mice with 4# KPM treatment still maintained at low level for at least 24 h (Fig. 5AeF). Frozen sections prepared from the colons of mice with 3# KPM injection at 24 h were stained with F4/80 antibody. As shown in Fig. 6A, ASO accumulated in the lamina propria of colon mucosa, with predominant uptake by colonic macrophages (F4/80þ cells). To further quantify cellular uptake of ASO, colonic macrophages were isolated 24 h after ASO injection. Other cells obtained during

30 Z. Huang et al. / Biomaterials 48 (2015) 26e36 Fig. 2. Immunofluorescence analysis and cell transfection assay. Dual-immunofluorescence staining with different cell markers and mannose receptor were used to investigate mannose receptor localization in cells (A) (scale bar, 10 mm) and DSS colitic colon (B) (scale bar, 50 mm) (red, F4/80 or pan-cytokeratin (PCK); green, mannose receptor (MR); blue, DAPI nuclear staining). Scale bar, 10 mm. In vitro transfection results of RAW 264.7 cells and CT-26 cells were photographed by fluorescent microscopy (C) and analyzed by FACS (D, E). (Red, ASO; blue, DAPI nuclear staining). Scale bar, 100 mm. Five samples were analyzed per condition, and experiments were performed in triplicate. *P  0.05. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Z. Huang et al. / Biomaterials 48 (2015) 26e36

31

Fig. 3. The observation of KPM morphology and in vitro release assay. (A) SEM images of the full view and the surface of KPM. (B) Bright field and fluorescent images of KPM containing Alexa 546 labeled ASO. Scale bar, 300 mm. The images (C), swelling index (D), diameter (E) and accumulative released ASO (F) of different composition formulated KPMs in stimulated digestive juice at indicated time points. Five samples were analyzed per condition, and experiments were performed in triplicate. *P  0.05.

the separation process were considered as non-macrophages. Obviously, the ASO level in colonic macrophages was rather higher than that in colonic epithelial cells in colitic mice after 3# KPM treatment (Fig. 6B). Meanwhile, few ASO could be transferred into both macrophages and epithelial cells in healthy control mice with 3# KPM administration. (Fig. 6B). 3.6. Reduction of mucosal TNF-a production by 3# KPM Immunofluorescence staining of TNF-a in lamina propria in colitic mice was clearly enhanced compared with control mice. Meanwhile, little TNF-a expression was detected in lamina propria of 3# KPM treated mice (Fig. 6C). ELISA was also used to determine the expression levels of TNF-a. We observed that 3# KPM was more efficient in reducing mucosal TNF-a than naked ASO in DSS colitic mice (Fig. 6D). 3.7. 3# KPM protected mice against DSS-induced acute colitis Mice subjected to 5% DSS solution feeding developed a severe pan colitis characterized by bloody diarrhea, extensive wasting

syndrome, a sustained weight loss and a high mortality. Mice treated with 3# KPM containing ASO at the dose of 50 mg/kg rapidly recovered the lost body weight, had a significant higher survival rate and experienced significantly lower DAI levels (Fig. 7AeC). Increased colonic MPO activity, correlating with mucosal inflammation, was significantly reduced by treatment with ASO&3# KPM (Fig. 7D). Macroscopic examination of colon obtained 8 days after colitis induction showed striking hyperemia, inflammation and shorter colon length. With ASO&3# KPM treatment, the colons of mice showed no significant sign of macroscopic inflammation and longer colon length (Fig. 7EeF). H&E staining showed that DSS-induced colitis affected all layers of colon with submucosal edema, muscle thickening and strong granulocyte infiltration (Fig. 7G). After mice treated with ASO&3# KPM, a striking improvement of histological signs and score became apparent (Figs. 7GeH). We further assessed the effect of ASO&3# KPM on the secretion of colonic cytokines which are mechanistically linked to experimental colitis. As shown in Fig. 7I, DSS colitic mice exhibited high levels of inflammatory cytokines (Interleukin1b (IL-1b), IL-6, IL-12p70 and IL-23). All of these cytokines were significantly suppressed by ASO&3# KPM treatment.

32 Z. Huang et al. / Biomaterials 48 (2015) 26e36 Fig. 4. Characterization of the contents released from microspheres. (A) TEM image of microspheres released contents (MRC). Sale bar, 400 nm. In vitro transfection of MRC into RAW 264.7 cells was performed. The results were photographed by confocal microscopy (B) and analyzed by FACS (B, C). (Red, ASO; blue, DAPI nuclear staining). Scale bar, 100 mm. (D) The TNF-a level of LPS activated RAW 264.7 cells with different treatments. (E) The degradation ratio of KGM and cKGM with enzyme digestion for 18 h. (F) Electrophoretic mobility of ASO. Lanes 1, 2 and 3 are naked ASO, cKGM&ASO complexes and cKGM&ASO complexes with enzyme digestion. Five samples were analyzed per condition, and experiments were performed in triplicate. *P  0.05. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Z. Huang et al. / Biomaterials 48 (2015) 26e36

33

Fig. 5. The bio-distribution of ASO in the digestive organs. Fluorescence images of the stomach, small intestine and colon at 2 h (A), 6 h (B) and 24 h (C) after cKGM&ASO nanocomplexes or different KPMs containing Alexa 546-labeled ASO administration. (Red, ASO; blue, DAPI nuclear staining). Scale bar, 100 mm. Tissue level of ASO in mice at 2 h (D), 6 h (E) and 24 h (F) after intragastric administration of nano-complexes or KPMs. n ¼ 8e11 mice per group. n.d., no detected. Values are expressed as the mean (SEM). *p  0.05. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

4. Discussion During the colitic process, the elevated level of TNF-a directly drives the damage to the colon tissue via enhancing the apoptosis of colonic epithelial cells and promoting the infiltration and activation of immune cells. Therefore, the key role of TNF-a in IBD makes it to be a promising therapeutic target. Antibodies against TNF-a achieved significant successes in clinical applications [4]. However, unlike rheumatoid arthritis and other systematically autoimmune disorders, IBD is mainly restricted within the intestinal tract, so the systematically neutralization of TNF-a via antibody will disturb the function of whole immune system and resulted in some serious side effects [16]. A more ideal way is to treat the IBD promoting TNF-a in the pathologically functional cells - colonic macrophages. Our previous study suggested that directly intracolonic delivering TNF-a ASO targeting colonic macrophage had significant therapeutic effects on experimental colitis and little influences on the physiological function of the immune system [17]. However, the inconvenience and suffering of enema brought to the patients and the complexity of the operation is a limiting obstacle to its further application. As a convenient route which allows nucleic acid drug easily access to the mucosal surface of the inflamed colon, oral administration can ensure the high local concentration of nucleic acid drug in the alimentary tract. In the present study, we developed an orally administration system based on cKGM and phytagel formulated microsphere, which can specially delivery ASO into the colonic macrophages, suppresses the mucosal

TNF-a production and finally alleviates the symptoms of experimental colitis. The successful application of nucleic acid drugs is highly dependent on delivery systems because of their instability in body fluids, poor cellular uptake and the demand of entering into specific cells for their efficacy [18]. Moreover, because of the constantly changing and sometime extreme environments of the digestive tract, it is even more difficult to orally deliver nucleic acid drugs. To deliver an IBD-therapeutic nucleic acid via digestive tract, the delivery system must have at least two basic functions. Firstly, to protect the structural and functional integrity of the nucleic acids to get through the digestive juices in stomach and intestinal tract and precisely release the drugs into colon; secondly, to deliver the nucleic acids into specific pathological cells. Although there were some attempts to deliver ASO or siRNA into colon via oral administration for the treatment of IBD, there is still no delivery system that can meet the criterion of both colonic sustained release and macrophages targeting delivery at the same time [19,20,21]. To develop such a delivery system, a well-designed functional material is needed. People may usually endow different functions to a system by using different materials [22], which will complicate the whole preparation process and make the in vivo behaviors of the delivery system more difficult to be under control. So, in our present study, we used a multifunctional material, cKGM to solve the problems. As a soluble fiber extracted from Amorphophallus konjac, KGM has been widely used as food stabilizer, gelling agent, weight

34

Z. Huang et al. / Biomaterials 48 (2015) 26e36

Fig. 6. Cellular localization of ASO and colonic TNF-a assay. (A) Frozen sections of colon from mice 24 h after 3# KPM administration. Red, ASO; green, macrophages (F4/80); blue, DAPI nuclear staining. Scale bar, 20 mm. (B) Macrophages and colonic epithelial cells were examined for ASO cellular uptake. The colonic TNF-a level of mice that received different treatments was examined by immunofluorescence staining (C) and by ELISA (D). (Red, TNF-a; blue, DAPI nuclear staining). Scale bar, 100 mm n ¼ 8e11 mice per group. Values are expressed as the mean (SEM). *p  0.05. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

management and shows little toxicity and pathogenicity to human body [9]. More importantly, the two characteristics - unique chemical composition and excellent water-absorption property enable KGM to be an ideal agent to overcome the obstacles of orally gene targeting delivery system. First, KGM is composed of mannose and b-glucan. Macrophages, which are known to express mannose receptors in abundance, can recognize the invasive foreign objects containing terminal mannose residues and mediate the phagocytosis of these foreign organism [23]. Moreover, b-glucan, the other composition of KGM, may also bind to b-glucan receptor which is also especially expressed on macrophages [24]. KGM was cationized to bind the negative charged ASO. The TEM results showed that cKGM can combine with ASO to form a stable nano-complex. In vitro transfection experiment indicated that the transfection efficiency of the cKGM&ASO complex was much higher in macrophages than that in colonic epithelial cells. This implies that mannose receptor or b-glucan receptor-mediated endocytosis could effectively targeting delivery ASO into macrophages. Second, the nearly unlimited swelling ability of KGM in water solution can assure the precisely release of nucleic acid drug in the inflamed colon. It has been proven that orally administration of nano-complex alone led to the clearance of most particles by the epithelial cells of stomach and small intestine during the passage process [25,26]. The retention of nucleic acid drug may greatly affect the therapeutic efficacy. Microspheres containing cKGM and phytagel were used to solve this problem. One side, due to its large size, microsphere itself cannot be internalized by the epithelial cells

of stomach and small intestine, and can reduce the drug absorption during gastric intact passage. On the other side, cKGM absorbs large amounts of water and leads to the volume expansion of microspheres. Then, the uncontrolled swelling of cKGM causes the burst of microsphere and releases the inside contents. Phytagel was used to improve the strength of microsphere and counteract the expand of cKGM. The manipulation of the ratio between cKGM and phytagel can lead to the collapse of microsphere at indicated time points. The in vitro swelling studies and in vivo distribution assay both indicated that KPM contained higher ratio of cKGM broke at earlier time points, whereas higher portion of phytagel led to the collapse at later time points. TEM examination observed the erosion of microsphere didn't release naked ASO but nanocomplexes and these complexes could also delivery nucleic acid drug into macrophages just like directly formed cKGM&ASO nanocomplex. Therefore, it is speculated that appropriate composition of microsphere collapse in the colon lumen and it was resulted the release of large amounts of nano-complex which is composed of cKGM and ASO. To ensure the nucleic acid can exert therapeutic effect in vivo, the delivery system should not only targeting delivery the drug into the diseased site, but also maintain the integrity and function of nucleic acid drug. We observed that MRC with different digestive fluid treatments could still exert significant suppression effect of TNF-a secretion in LPS activated macrophage. Meanwhile, naked ASO with simulated gastric fluid pre-treatment showed little effect on inhibiting TNF-a production. We speculate the short retention

Z. Huang et al. / Biomaterials 48 (2015) 26e36

35

Fig. 7. Anti colitis activity of 3# KPM. The therapeutic effect of 3# KPM in DSS colitic mice was observed through body weight changes (A), survival analysis (B), disease activity index (C), MPO activity determination (D), colon photographs (E) and colon length (F). Colon sections at day 8 from DSS colitic mice that received different treatments were examined by H&E staining and histopathological scoring (G, H). Scale bar, 100 mm. The level of colonic inflammatory cytokines (IL-1b, IL-6, IL-12p70 and IL-23) from DSS colitic mice with different treatments was determined (I). n ¼ 8e11 mice per group. Values are expressed as the mean (SEM). *p  0.05.

period of KPM in the gastric fluid may prevent the acidic buffer completely permeates into the microsphere. Moreover, KPM contains high amount of cKGM and the amine contents of cKGM may act as a buffer region in acidic environment and delay the hydrolysis of nucleic acid drug by gastric acid. Besides this, the existence of enzyme form the colon flora may degrade cKGM and may lead to the destabilization of cKGM&ASO complex which is existed in the colon lumen [27,28]. But we observed that the degradation percentage of cKGM was much lower than that of KGM with cellulase and mannase digestion and cKGM&ASO complex could still strand in the well after long term digestion. It is supposed that cationic modification makes the enzyme from the colon flora cannot realize the glucosidic bond in cKGM. Therefore, the particle released from microspheres are resistant to the enzyme digestion and could still remain stable in the colon lumen for a long period. Then, the nanocomplexes have enough time to pass through the defect of colonic epithelium and are phagocytized by the underlying macrophages via receptors binding. Theoretically, KPM which works as a dualfunctional system can meet the criterion of both delayed release and targeting delivery in IBD treatment. The in vivo bio-distribution assay approved that 3# KPM specially released cKGM&ASO nano-

complex in the colon site and the nano-complex was predominantly taken up by colonic macrophages, which finally resulted in the reduction of TNF-a and the alleviation of colitis. 5. Conclusion We described here an orally colon macrophage targeted nucleic acid delivery system based on the water-absorption properties of cKGM and macrophage-specific monosaccharide residues receptormediated phagocytosis and tested its anti-inflammation activity in an experimental colitic model. The results demonstrated that targeting delivery of ASO into colonic macrophages effectively suppressed TNF-a expression and achieved favor therapeutic effect. This orally nucleic acid drug delivery system targeting colonic macrophage represents a potential therapeutic approach that may be valuable for IBD treatment. Acknowledgments This work was supported by the National Science Fund for Distinguished Young Scholars (81025019), the National Basic

36

Z. Huang et al. / Biomaterials 48 (2015) 26e36

Research Program of China (2012CB517603), the National High Technology Research and Development Program of China (2014AA020707), the Program for New Century Excellent Talents in University (NCET-13-0272), the National Natural Science Foundation of China (31271013, 31170751, 31200695, 31400671, 51173076, 91129712 and 81102489), the Key Project of the Chinese Ministry of Education (108059), the Ph.D. Programs Foundation of the Ministry of Education of People's Republic of China (20100091120020, 20130091110037), Nanjing University State Key Laboratory of Pharmaceutical Biotechnology Open Grant (KF-GN-201409), Jiangsu Planned Projects for Postdoctoral Research Funds (1302009B), China Postdoctoral Science Foundation funded project (2014M551555). C Wang acknowledges the funding supports from the Macao Science and Technology Development Fund (048/2013/ A2) and the matching grant of the University of Macau (MRG006/ WCM/2014/ICMS).

Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.biomaterials.2015.01.013.

References [1] Abraham C, Cho JH. Inflammatory bowel disease. N Engl J Med 2009;361: 2066e78. [2] Heinsbroek SE, Gordon S. The role of macrophages in inflammatory bowel diseases. Expert Rev Mol Med 2009;11:e14. [3] Targan SR, Hanauer SB, van Deventer SJ, Mayer L, Present DH, Braakman T, et al. A short-term study of chimeric monoclonal antibody cA2 to tumor necrosis factor alpha for Crohn's disease. Crohn's disease cA2 study group. N Engl J Med 1997;337:1029e35. [4] Ben-Horin S, Chowers Y. Tailoring anti-TNF therapy in IBD: drug levels and disease activity. Nat Rev Gastroenterol Hepatol 2014;11:243e55. [5] Travassos WJ, Cheifetz AS. Infliximab: use in inflammatory bowel disease. Curr Treat Options Gastroenterol 2005;8:187e96. [6] Ricart E, Ordas I, Panes J. Anti-TNF antibody therapy in Crohn's disease: the risk of a switch. Gut 2012;61:169e70. [7] Schneeweiss S, Korzenik J, Solomon DH, Canning C, Lee J, Bressler B. Infliximab and other immunomodulating drugs in patients with inflammatory bowel disease and the risk of serious bacterial infections. Aliment Pharmacol Ther 2009;30:253e64. [8] Linehan SA, Martinez-Pomares L, Gordon S. Mannose receptor and scavenger receptor: two macrophage pattern recognition receptors with diverse functions in tissue homeostasis and host defense. Adv Exp Med Biol 2000;479: 1e14. [9] Keithley J, Swanson B. Glucomannan and obesity: a critical review. Altern Ther Health Med 2005;11:30e4.

[10] Huang Z, Yang Y, Jiang Y, Shao J, Sun X, Chen J, et al. Anti-tumor immune responses of tumor-associated macrophages via toll-like receptor 4 triggered by cationic polymers. Biomaterials 2013;34:746e55. [11] Huang Z, Zhang Z, Zha Y, Liu J, Jiang Y, Yang Y, et al. The effect of targeted delivery of anti-TNF-alpha oligonucleotide into CD169þ macrophages on disease progression in lupus-prone MRL/lpr mice. Biomaterials 2012;33: 7605e12. [12] Chua M, Chan K, Hocking TJ, Williams PA, Perry CJ, Baldwin TC. Methodologies for the extraction and analysis of konjac glucomannan from corms of Amorphophallus konjac K. Koch. Carbohydr Polym 2012;87:2202e10. [13] Wang C, Gong Y, Lin Y, Shen J, Wang DA. A novel gellan gel-based microcarrier for anchorage-dependent cell delivery. Acta Biomater 2008;4:1226e34. [14] Huang Z, Shi T, Zhou Q, Shi S, Zhao R, Shi H, et al. miR-141 Regulates colonic leukocytic trafficking by targeting CXCL12beta during murine colitis and human Crohn's disease. Gut 2014;63:1247e57. [15] Huang Z, Zhang Z, Jiang Y, Zhang D, Chen J, Dong L, et al. Targeted delivery of oligonucleotides into tumor-associated macrophages for cancer immunotherapy. J Control Release 2012;158:286e92. [16] Stallmach A, Hagel S, Bruns T. Adverse effects of biologics used for treating IBD. Best Pract Res Clin Gastroenterol 2010;24:167e82. [17] Zuo L, Huang Z, Dong L, Xu L, Zhu Y, Zeng K, et al. Targeting delivery of antiTNFalpha oligonucleotide into activated colonic macrophages protects against experimental colitis. Gut 2010;59:470e9. [18] Petros RA, DeSimone JM. Strategies in the design of nanoparticles for therapeutic applications. Nat Rev Drug Discov 2010;9:615e27. [19] Kriegel C, Amiji M. Oral TNF-alpha gene silencing using a polymeric microsphere-based delivery system for the treatment of inflammatory bowel disease. J Control Release 2011;150:77e86. [20] Wilson DS, Dalmasso G, Wang L, Sitaraman SV, Merlin D, Murthy N. Orally delivered thioketal nanoparticles loaded with TNF-alpha-siRNA target inflammation and inhibit gene expression in the intestines. Nat Mater 2010;9: 923e8. [21] Kriegel C, Attarwala H, Amiji M. Multi-compartmental oral delivery systems for nucleic acid therapy in the gastrointestinal tract. Adv Drug Deliv Rev 2013;65:891e901. [22] Laroui H, Viennois E, Xiao B, Canup BS, Geem D, Denning TL, et al. Fab'-bearing siRNA TNFalpha-loaded nanoparticles targeted to colonic macrophages offer an effective therapy for experimental colitis. J Control Release 2014;186: 41e53. [23] Xiao B, Laroui H, Ayyadurai S, Viennois E, Charania MA, Zhang Y, et al. Mannosylated bioreducible nanoparticle-mediated macrophage-specific TNFalpha RNA interference for IBD therapy. Biomaterials 2013;34:7471e82. [24] Brown GD, Taylor PR, Reid DM, Willment JA, Williams DL, MartinezPomares L, et al. Dectin-1 is a major beta-glucan receptor on macrophages. J Exp Med 2002;196:407e12. [25] Roy K, Mao HQ, Huang SK, Leong KW. Oral gene delivery with chitosaneDNA nanoparticles generates immunologic protection in a murine model of peanut allergy. Nat Med 1999;5:387e91. [26] Bhavsar MD, Amiji MM. Gastrointestinal distribution and in vivo gene transfection studies with nanoparticles-in-microsphere oral system (NiMOS). J Control Release 2007;119:339e48. [27] Sood N, Baker WL, Coleman CI. Effect of glucomannan on plasma lipid and glucose concentrations, body weight, and blood pressure: systematic review and meta-analysis. Am J Clin Nutr 2008;88:1167e75. [28] Albrecht S, van Muiswinkel GC, Xu J, Schols HA, Voragen AG, Gruppen H. Enzymatic production and characterization of konjac glucomannan oligosaccharides. J Agric Food Chem 2011;59:12658e66.