Lab on a Chip View Article Online
Published on 10 March 2014. Downloaded by University of Chicago on 30/10/2014 09:50:24.
PAPER
Cite this: Lab Chip, 2014, 14, 1842
View Journal | View Issue
Microfluidic generation of chitosan/CpG oligodeoxynucleotide nanoparticles with enhanced cellular uptake and immunostimulatory properties Song Chen,*ab Huijie Zhang,b Xuetao Shi,c Hongkai Wu*cd and Nobutaka Hanagata*be Chitosan/cytosine-phosphodiester-guanine oligodeoxynucleotide (CpG ODN) nanoparticles as potential immunostimulatory adjuvants were synthesized by the conventional bulk mixing (BM) method and a novel microfluidic (MF) method. Their size and size distribution, CpG ODN loading efficiency, surface charge, biocompatibility, cellular uptake, and immunostimulatory response were investigated. In the BM method, nanoparticles were synthesized by vortexing a mixture of chitosan solution and CpG ODN2006x3-PD solution. In the MF method, the nanoparticles were synthesized by rapidly mixing a chitosan solution and CpG ODN solution in a poly(dimethylsiloxane) microfluidic device. Our results indicated that particle size and size distribution, CpG ODN loading efficiency, and surface charge could be easily adjusted by using the tuning preparation method and controlling the flow ratio of fluid rates in the different microfluidic channels. Compared with the BM method, the MF method yielded a decrease in particle size and size range, an increase in CpG ODN loading efficiency, and a decrease in surface charge. After the particles were exposed to 293XL-hTLR9 cells, a water-soluble tetrazolium salt assay indicated that the BM and MF-processed nanoparticles had no significant toxicity and were biocompatible. An immunochemical assay indicated that both types of nanoparticles entered 293XL-hTLR9 cells and were located in the endolysosomes. The MF-processed nanoparticles showed much higher cellular uptake
Received 6th January 2014, Accepted 7th March 2014 DOI: 10.1039/c4lc00015c www.rsc.org/loc
efficiency. After the particles were exposed to peripheral blood mononuclear cells, an enzyme-linked immunosorbent assay quantitatively indicated that both types of nanoparticles stimulated the production of interleukin-6 and the MF-processed nanoparticles showed a much stronger immunostimulatory response. These results indicate that the MF method can be used to synthesize nanoparticles with a controllable size and size range for enhancing the biological activity of DNA and other biomolecules.
1. Introduction Chitosan is a linear polysaccharide biopolymer consisting of N-acetyl-D-glucosamine and D-glucosamine and is derived from chitin extracted from sea crustaceans. It has excellent biocompatibility and biodegradability.1,2 It also has a strong positive charge arising from rapid protonation of its two subunits under acidic conditions, or even neutral conditions, and thus has strong affinity to negatively charged materials such as a
JSPS Research Fellow, 8 Ichibancho, Chiyoda-ku, Tokyo 102-8472, Japan Biomaterials Unit, National Institute for Materials Science, 1-2-1 Sengen, Tsukuba, 305-0047, Japan. E-mail: [email protected]; Fax: +81 29 859 2449; Tel: +81 29 859 2000 c WPI Advanced Institute for Materials Research, Tohoku University, Katahira 2-1-1, Aobaku, Sendai, 980-8577, Japan d Department of Chemistry, Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong, China. E-mail: [email protected] e Nanotechnology Innovation Station, National Institute for Materials Science, 1-2-1 Sengen, Tsukuba, Ibaraki 305-0047, Japan. E-mail: [email protected] b
1842 | Lab Chip, 2014, 14, 1842–1849
alginate3 and DNA.4 Numerous studies have demonstrated that chitosan-based materials can effectively pack and condense DNA to produce a chitosan/DNA complex for protecting DNA from nuclease degradation. Employing chitosan-based materials as a delivery system of DNA has thus become a hot research topic in gene delivery systems.4,5 Controlling the size and size distribution of chitosan/DNA nanoparticles will be crucial to achieving the goal of controllable cellular uptake and biological activity. The conventional preparation route involves bulk mixing (BM) of chitosan solution and DNA solution and spontaneous self-assembly into complexes due to electrostatic forces. Such a self-assembly process typically occurs in a reaction platform with a characteristic length on the order of centimeters and results in highly heterogeneous chemical and/or mechanical conditions.6 Thus, it is still a challenge to control the size and size distribution of chitosan/DNA nanoparticles. In contrast to the BM process, a microfluidic (MF) platform manipulates fluids at the microscale level. Since the
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 10 March 2014. Downloaded by University of Chicago on 30/10/2014 09:50:24.
Lab on a Chip
reaction in the microchannels is mainly driven by interfacial forces, heat- and mass-transfers are significantly enhanced,6 and this results in a much better control over the flow rate, reaction time, and concentration of reagents compared with the BM process. It has been demonstrated that the MF method provides a much better control over the size and size distribution of nano/microparticles.7,8 Recent studies have shown that MF-processed gene delivery systems exhibit significantly enhanced biological activity. In particular, MF-processed PEI/pGFP nanocomplexes were found to be much smaller than BM-processed nanoparticles and had much higher transfection efficiency in NIH 3T3 cells.9 Moreover, MF-processed antisense oligodeoxyribonucleotide lipopolyplex nanoparticles had higher levels of Bcl-2 antisense uptake and showed more efficient downregulation of the Bcl-2 protein level in comparison with BM-processed nanoparticles.10 However, the MF platform has rarely been used to synthesize chitosan/DNA nanoparticles with controllable size and size distribution and enhanced biological activity. Cytosine-phosphodiester-guanine oligodeoxynucleotides (CpG ODNs) have been used as immune adjuvants because they can be recognized by Toll-like receptor 9 (TLR9) found in antigen-presenting cells (APCs) and B cells to activate the immune system.11 However, naked CpG ODNs have the drawbacks of low cellular uptake and low immunostimulatory response. Nanoscale delivery systems can effectively deliver various DNA and RNA into cells since they are small enough to be taken up by the cells via endocytosis. Recent studies have indicated that the immunostimulatory response of CpG ODNs can be significantly enhanced with the help of various nanoscale delivery systems including silicon dots,12 boron nitride nanospheres,13 and chitosan–silica nanohybrids.14 Our previous studies have demonstrated that chitosan and its derivatives can condense CpG ODNs and deliver them to immune cells to activate the immunostimulatory response.14 However, the cellular uptake and immunostimulatory response of chitosan/CpG ODN complexes remained very low. Considering the advantages of the MF method over the BM method, we hypothesized that MF-processed chitosan/CpG ODN nanoparticles would be much smaller than BM-processed nanoparticles and might have better cellular uptake efficiency and immunostimulatory activity. To demonstrate this hypothesis, in the present study, we used both methods to synthesize immunostimulatory chitosan/CpG ODN nanoparticles and we compared the resulting nanoparticles in terms of their size and size distribution, CpG ODN loading efficiency, surface charge, biocompatibility, cellular uptake, and immunostimulatory response. In vitro biocompatibility and cellular uptake were evaluated by incubating the nanoparticles with 293XL-hTLR9 cells, while in vitro immunostimulatory response was evaluated by exposing the nanoparticles to peripheral blood mononuclear cells (PBMC) and quantifying the amount of IL-6 produced by the cells via the enzyme-linked immunosorbent assay (ELISA) assay.
This journal is © The Royal Society of Chemistry 2014
Paper
2. Materials and methods 2.1 Fabrication of the microfluidic device The poly(dimethylsiloxane) (PDMS) microfluidic device, schematically depicted in Fig. 1, had multiple flow microchannels (200 μm in width and 50 μm in depth). In this device, the flow stream along the central channel and the flow streams along the two side channels meet at the cross-section position and focus the hydrodynamic flow in the central channel. A silicon master with multiple SU-8 microchannel patterns was first created using a standard photolithographic microfabrication process. A mixture of PDMS prepolymer and curing agent (10 : 1; Silpot 184, Dow Corning Toray Co., Ltd., Tokyo, Japan) was spin-coated on the silicon master. It was then degassed and cured at 70 °C for 2 h. The resulting PDMS sheet with multiple microchannels was removed from the silicon master; the PDMS sheet and a glass slide were separately exposed to O2 plasma in a plasma cleaner (PDC200, Yamato Scientific, Japan). After 45 seconds of exposure, the PDMS sheets were taken out of the cleaner and rapidly sealed to the glass slide to produce a microfluidic device with multiple microchannels. The device was connected to syringes using silicon rubber tubes and the solution of reagents was injected into the channels using programmable syringe pumps (New Era Pump Systems Inc., USA).
2.2 Synthesis of chitosan/CpG ODN nanoparticles Chitosan powders (60–120 kDa, Sigma-Aldrich) were added to an HAc–NaAc solution (pH = 5) and stirred at room temperature for 2 h to produce 0.1% (w/v) chitosan solutions. Before use, the chitosan solution was sterilized using a 0.22 μm filter. Natural phosphodiester CpG ODN2006x3-PD (5′TCGTCGTTTTGTCGTTTTGTCGTTTCGTCGTTTTGTCGTTTTGTCGTTTCGTCGTTTTGTCGTTTTGTCGTT-3′) (72 mer, Fasmac
Fig. 1 Schematic illustration of a microfluidic PDMS device. Chitosan solution was injected from inlet 1 and inlet 2 into the side channels, while CpG ODN solution was injected from inlet 3 into the center channel. The resultant chitosan/CpG ODN nanoparticles were collected from the outlet.
Lab Chip, 2014, 14, 1842–1849 | 1843
View Article Online
Published on 10 March 2014. Downloaded by University of Chicago on 30/10/2014 09:50:24.
Paper
Lab on a Chip
Inc., Kanagawa, Japan) was diluted in sterilized water to a concentration of 300 μg mL−1. Chitosan/CpG ODN nanoparticles were then synthesized by the BM and MF methods. In the BM method, 400 μL of chitosan solution was mixed with 200 μL of CpG ODN solution, and the mixture was vortexed for 1 min and left for 30 min at room temperature to yield chitosan/CpG ODN nanoparticles (coded as BMChCpG). In the MF method, the programmable syringe pumps separately injected the chitosan solution from inlet 1 and inlet 2 into the side channels at a controlled flow rate of 100–200 μL min−1, while they injected the CpG ODN solution from inlet 3 into the central channel at a controlled flow rate of 100–200 μL min−1. The flow ratio of chitosan/CpG ODN was set as 0.5, 1 or 2. The chitosan and CpG ODN solutions met at the cross-junction position and hydrodynamic focusing occurred in the central channel to produce chitosan/CpG ODN nanoparticles, which were collected at the outlet. The resultant chitosan/CpG ODN nanoparticles were coded according to the flow ratio as MFChCpG0.5, MFChCpG1, or MFChCpG2. Table 1 lists the reaction parameters and sample abbreviations.
2.3 Characterization The size and morphology of the resultant nanoparticles were observed under a field emission scanning electron microscope (FE-SEM; JSM 6500, JEOL, Tokyo, Japan). The particle size distribution was measured by dynamic light scattering (DLS-6000AL, Photal Otsuka Electronics, Japan), whereas the zeta potential was measured using a laser electrophoresis zeta-potential analyzer (LEZA-600, Otsuka, Japan). To evaluate the encapsulation efficiency of CpG ODNs, freshly synthesized chitosan/CpG ODN nanoparticles were collected by centrifuging the solution at 12 000 rpm for 5 min. Free CpG ODNs in the supernatant were quantified using a NanoDrop 2000 spectrophotometer at 260 nm. The encapsulation efficiency of CpG ODNs in the chitosan/CpG ODN nanoparticles was calculated using the following equation: Encapsulation efficiency(%) = [(Wo − Ws)/Wo] × 100% Here, Wo is the amount of CpG ODNs before reacting with chitosan and Ws is the amount of CpG ODNs in the supernatant. To evaluate the release behavior of CpG ODNs, chitosan/ CpG ODN nanoparticles collected by centrifugation were re-suspended in an HAc–NaAc buffer solution (pH 5.5) in a 1.5 mL Eppendorf tube and placed in a shaking bath at
37 °C. At predetermined time intervals, the nanoparticles were collected from the suspension by centrifugation at 12 000 rpm at 5 min and re-suspended in a fresh HAc–NaAc buffer solution. Free CpG ODNs in the supernatant were quantified using the above-described method. The percentage of CpG ODNs released from the nanoparticles was calculated using the following equation: Release percentage(%) = (Wr/Wt) × 100% Here, Wr is the amount of CpG ODNs released from the nanoparticles into the supernatant and Wt is the total amount of CpG ODNs in the nanoparticles.
2.4 Cell toxicity 293XL-hTLR9 cells stably expressing human TLR9 (Invivogen) were grown in Dulbecco's Modified Eagle Medium (DMEM) supplemented with 10% FBS, 100 U mL−1 penicillin and 100 μg mL−1 streptomycin in a humidified atmosphere of 5% CO2 and 95% air at 37 °C and used to evaluate the toxicity of the nanoparticles. Cells were detached from the flask using trypsin, seeded in a 96-well culture plate at a density of 1 × 104 cells per well, and grown in DMEM overnight. The culture medium was then replaced with fresh medium containing chitosan/CpG ODN nanoparticles. A water-soluble tetrazolium salt (WST-1) assay was performed to evaluate cell viability. After 24 h and 48 h of culturing, the culture medium was replaced with a WST-1 containing culture medium. After another 4 h of culturing, the absorbance at 450 nm was measured using a microplate reader (MTP-880, Corona Electric, Japan) in order to determine if formazan was formed from the water-soluble tetrazolium salt. Cells without any chitosan/CpG ODN nanoparticles were used as the control. The morphologies of the cells incubated with and without nanoparticles were visualized under an inverted microscope (DM2500, Leica).
2.5 Cellular uptake CpG ODNs are recognized by TLR9 receptors which are located in the endolysosomes.11–14 A successful cellular uptake of chitosan/CpG ODN nanoparticles is thus crucial to activation of the immune response. To monitor the behavior of cellular uptake, CpG ODNs were replaced with 3′-fluorescein isothiocyanate (FITC)-labeled CpG ODNs when the chitosan/FITC–CpG ODN nanoparticles were prepared using either method.
Table 1 Reaction parameters and sample abbreviations
1 2 3 4
Sample abbreviation
Preparation method
Chitosan flow rate (μL min−1)
CpG ODN flow rate (μL min−1)
Flow ratio
BMChCpG MFChCpG0.5 MFChCpG1 MFChCpG2
BM MF MF MF
None 100 100 200
None 200 100 100
None 0.5 1 2
1844 | Lab Chip, 2014, 14, 1842–1849
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 10 March 2014. Downloaded by University of Chicago on 30/10/2014 09:50:24.
Lab on a Chip
Paper
293XL-hTLR9 cells were seeded in a four-compartment glass-bottom dish (35 mm) at a density of 4 × 104 per well and cultured overnight. The culture medium was then replaced with a fresh medium containing chitosan/FITC–CpG ODN nanoparticles. To confirm whether the nanoparticles had entered the cells and endolysosomes, lysosome-associated membrane protein 1 (LAMP-1) was immunochemically stained as an endolysosomal marker. Anti-LAMP-1 antibody (ab24170, Abcam) was used as the primary antibody, while Alexa Fluor® 555 donkey anti-rabbit IgG (H + L) (Invitrogen) was used as the secondary antibody. Cells were visualized under a confocal laser scanning microscope (CLSM; Leica). 2.6 In vitro immunostimulatory response Peripheral blood mononuclear cells (PBMCs, Cellular Technology Limited, US) were used to evaluate the in vitro immunostimulatory response of the nanoparticles. RPMI 1640 medium was supplemented with 10% FBS, 100 U mL−1 penicillin, and 100 μg mL−1 streptomycin. The PBMCs were thawed and suspended in RPMI 1640 medium at a density of 106 cells mL−1. 190 μL of the cell suspension was mixed with 10 μL of chitosan/CpG ODN nanoparticles synthesized via either method. The mixture was put in a 96 well culture plate in a humidified atmosphere of 5% CO2 and 95% air at 37 °C. After 48 h, cell cytokine interleukin-6 (IL-6) secreted by the PMBCs that were stimulated by the nanoparticles in the culture medium was quantified by enzyme-linked immunosorbent assay (ELISA) using a Human IL-6-ELISA Kit (eBiosciences, Vienna, Austria) according to the manufacturer's instructions.
Fig. 2 Phase microscope image of a microfluidic PDMS device (a) and fluorescent microscope images of hydrodynamic focusing of a mixture of fluorescein (green) fluid and rhodamine B (red) fluid at the meeting position of channels. The rhodamine B/fluorescein flow ratios were 0.5 (b), 1 (c), and 2 (d).
be seen that the width of the green region at the crosssection of the device decreased as the flow ratio of the fluid rate in the side channel relative to the central channel increased from 0.5 to 2. This indicates that increasing the flow ratio resulted in a decrease in the mixed amount of fluid coming from the central channel and an increase in the mixed amount of fluid coming from the side channels during hydrodynamic focusing. That is, the mixing behavior of different fluids at the cross-section position could be easily adjusted by tuning the flow ratio.
2.7 Statistical analysis All results were obtained in triplicate, expressed in terms of the mean and standard deviation (SD) and analyzed using one-way analysis of variance (ANOVA) with a significance level of p < 0.05.
3. Results and discussion 3.1 Hydrodynamic focusing Hydrodynamic focusing is a unique feature of the microfluidic platform.6–8 Its occurrence enables the microfluidic device to control the size and size distribution of the resultant nanoparticles.8 To observe hydrodynamic focusing, a solution of rhodamine B was used to mimic the solution of chitosan and was separately injected into the two side channels from inlet 1 and inlet 2, while a solution of fluorescein was used to mimic the solution of CpG ODNs and was injected into the central channel from inlet 3. The microfluidic device was mounted on a fluorescent microscope. Fig. 2 shows that the green fluorescein solution was squeezed by the red rhodamine B solution and clearly reveals hydrodynamic focusing due to rapid mixing of the flow streams. Thus, the present device is a typical microfluidic platform and can be potentially used to synthesize chitosan/CpG ODN nanoparticles of controllable size and size distribution. It can
This journal is © The Royal Society of Chemistry 2014
3.2 Particle characterization CpG ODNs are recognized by TLR9 receptors, which are only expressed in immune cells such as APCs and B cells.11,15 Several nanoscaled delivery systems have been developed and they have been used to deliver CpG ODNs into those immune cells. The size of the nanoparticles has a significant effect on cellular uptake. Zhu et al. reported that polylysine-coated mesoporous silica spheres with a diameter of 500 nm could deliver CpG ODNs into 293XL-hTLR9 cells in order to activate the immune response.16 Zhang et al. reported that chitosancoated boron nitride nanospheres with a diameter of 150 nm delivered CpG ODN2006x3-PD into PBMC cells to stimulate IL-6 production.17 In our previous study, we demonstrated that chitosan–silica nanoparticles with a diameter of 200–300 nm could deliver CpG ODN2006x3-PD into 293XL-hTLR9 cells to activate the production of IL-6.14 In most cases in these studies, the size of the nanoscale delivery systems for delivery of CpG ODNs had to be less than 500 nm. The size and size distribution of BM-processed chitosan/ DNA nanoparticles have been investigated in detail, and it has been found that they are greatly influenced by parameters such as chitosan deacetylation degree, chitosan concentration, chitosan molecular weight, and DNA concentration.18,19 Since the
Lab Chip, 2014, 14, 1842–1849 | 1845
View Article Online
Published on 10 March 2014. Downloaded by University of Chicago on 30/10/2014 09:50:24.
Paper
BM method mainly relies on bulk forces between reactants and provides inhomogeneous reaction conditions, it is hard to control the size and size distribution of the resulting nanoparticles. Our previous study has shown that BM-processed chitosan/CpG ODN nanoparticles exhibited a wide size distribution ranging from several hundred nm to several micrometers due to agglomeration.14 Unlike the BM method, the MF method mainly relies on interfacial forces between fluids, and such forces play a key role in controlling the size and size distribution of the final nanoparticles. Jahn et al. found that the interfacial forces were closely associated with the flow ratio of the fluid rate of the side channel relative to that of the center channel and that they increased as the flow ratio increased, resulting in a decrease in the particle size and size range.6 Thus, it is speculated that the size and size distribution of chitosan/CpG ODN nanoparticles can be tuned by altering the flow ratio. DLS measurements provide direct evidence of the particle size and size distribution. Fig. 3 shows the DLS results for BMChCpG, MFChCpG0.5, MFChCpG1, and MFChCpG2. BMChCpG shows a wide and bimodal size distribution of 398 ± 291 nm, while all of the others show single modal size distribution. The size distributions of MFChCpG0.5, MFChCpG1, and MFChCpG2 were 234 ± 80 nm, 179 ± 56 nm, and 204 ± 45 nm, respectively. In contrast, the MF method resulted in a decrease in the particle size and size range compared with the BM method. Moreover, the size of the MF-processed nanoparticles was less than 500 nm and was thought to be suitable for delivery of CpG ODNs into the immune cells. Furthermore, we compared BM and MF-processed chitosan/ CpG ODN nanoparticles in terms of their size and morphology. Fig. 4 shows typical SEM images of BMChCpG and MFChCpG2. BMChCpG (Fig. 4(a)) was irregular in shape and seriously agglomerated with a size more than 1000 nm (arrows), while
Fig. 3 Size distributions of chitosan/CpG ODN nanoparticles: (a) BMChCpG, (b) MFChCpG0.5, (c) MFChCpG1, and (d) MFChCpG2.
1846 | Lab Chip, 2014, 14, 1842–1849
Lab on a Chip
Fig. 4 SEM images of chitosan/CpG ODN nanoparticles synthesized by the BM method (a) and the MF method (b).
MFChCpG2 had a much better morphology and smaller size of less than 500 nm (Fig. 4(b)). In contrast, the MF method is indeed much better at controlling the size and size distribution of the resultant nanoparticles with a diameter less than 500 nm for delivery of CpG ODNs. The above results indicate that the MF method is a simple way of preparing chitosan/CpG ODN nanoparticles with a small size, narrow size distribution, and good morphology. To the best of our knowledge, this is the first demonstration of the fine control that can be obtained by using microfluidics to manipulate the size and size distribution of chitosan/DNA nanoparticles. In the case of the BM method, the binding of DNA to chitosan is mainly driven by electrostatic interactions. As the BM method works at the centimeter level6 and has difficulty in providing homogenous reaction conditions, the mixing of chitosan and CpG ODNs is not homogenous, resulting in loose and uncontrollable packing and large nanoparticles with a wide size distribution (Fig. 5(a)). In the MF device, enhanced mixing at a high flow rate inside 3D microchannels could be obtained, leading then to smaller nanoparticles.20 The formation of smaller and more monodisperse nanoparticles might be also due to the mixing time being much shorter than the particle's aggregation time.8,21 Thus, the infusion of chitosan and CpG ODNs is much more homogenous and precise. The packing of chitosan and CpG ODNs is much stronger and denser and results in a much smaller size and narrower size distribution (Fig. 5(b)).
Fig. 5 Possible formation mechanisms of chitosan/CpG ODN nanoparticles synthesized by the BM method (a) and the MF method (b).
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 10 March 2014. Downloaded by University of Chicago on 30/10/2014 09:50:24.
Lab on a Chip
Consequently, it was speculated that the CpG ODN loading efficiency of the MF-processed nanoparticles would be much higher than that of BM-processed nanoparticles and the surface charge of MF-processed nanoparticles is smaller than that of the BM-processed ones. Fig. 6(a) and (b) show the CpG ODN loading efficiency and surface charge of the resultant nanoparticles. Indeed, the CpG ODN loading efficiency of MFChCpG0.5, MFChCpG1 and MFChCpG2 was much higher than that of BMChCpG, while the surface charge of BMChCpG was much higher than that of MFChCpG0.5, MFChCpG1 and MFChCpG2. As depicted in Fig. 2, increasing the flow ratio led to a decrease in the mixing amount of CpG ODNs. Therefore, it is reasonable that the CpG ODN loading efficiency decreased and the surface charge increased as the flow ratio increased from 0.5 to 2 due to the difference in the mixing behavior between the BM and MF methods. The cell membrane has a net negative surface charge.22 A positive surface charge is thus advantageous for cellular uptake of chitosan/DNA nanoparticles. For this reason, MFChCpG2, which has a smaller size and larger surface charge, was selected for the biological assessments presented next.
3.3 Cytotoxicity Biocompatibility is one of the most important properties for a nanoscale delivery system to have, since toxicity can alter or disrupt cell membranes.23 Here, the BM- and MF-processed
Paper
Fig. 7 WST-1 results (a), cell number (b), and cell morphology (c) of chitosan/CpG ODN nanoparticles synthesized by BM and MF methods after exposure to 293XL-hTLR9 cells.
chitosan/CpG ODN nanoparticles were exposed to 293XLhTLR9 cells, and cell viability was evaluated by the WST-1 assay. Fig. 7(a) shows the results of the WST-1 assay. The cell viability of both types of nanoparticles was very similar to that of the control cells without any nanoparticles after 24 h and 48 h of culturing, indicating that both types of nanoparticles had no significant toxicity and were biocompatible. Fig. 7(b) shows that the cell number increased as the culturing time increased, indicating that the nanoparticles had no effect on cell proliferation and further confirming their biocompatibility. A similarity in the cell size and morphology was observed between the cells that took up the nanoparticles and the control (Fig. 7(c)), and this is direct evidence of the biocompatibility of both types of nanoparticles.
3.4 Cellular uptake
Fig. 6 CpG ODN loading efficiency (a) and surface charge (b) of chitosan/CpG ODN nanoparticles: BMChCpG, MFChCpG0.5, MFChCpG1, and MFChCpG2.
This journal is © The Royal Society of Chemistry 2014
Previous studies have demonstrated that BM-processed chitosan/CpG ODN nanoparticles could enter 293XL-hTLR9 and were located in the endolysosomes.14 However, their large size and size distribution resulted in low cellular uptake and poor immunostimulatory response. As BMChCpG was much larger than MFChCpG2, it was speculated that the cellular uptake of BMChCpG would be much lower than that of MFChCpG2. To monitor the cellular uptake behavior, FITC– CpG ODNs were used to prepare BMChCpG and MFChCpG2. Fig. 8(a) shows confocal microscope fluorescent images of BMChCpG and MFChCpG2 after exposure to 293XL-hTLR9 cells for 48 h. Chitosan/FITC–CpG ODN nanoparticles exhibited green fluorescence, while endolysosomes marked with LAMP-1 exhibited red fluorescence. The green fluorescence combined with the red fluorescence to make yellow fluorescence. The yellow fluorescent points indicate that BMChCpG and MFChCpG2 were taken up by the cells and were located in the endolysosomes. The number of yellow points was much larger for MFChCpG2, indicating that
Lab Chip, 2014, 14, 1842–1849 | 1847
View Article Online
Published on 10 March 2014. Downloaded by University of Chicago on 30/10/2014 09:50:24.
Paper
Fig. 8 (a) Confocal laser microscope fluorescent images of 293XLhTLR9 cells after incubation with chitosan/FITC–CpG ODN nanoparticles (green) synthesized via BM and MF methods. Endolysosomes were immunochemically stained (red). Yellow fluorescence indicates that the nanoparticles were taken up by the cells and entered the endolysosomes. (b) Total yellow fluorescence intensities in Fig. 8(a) were calculated using the ImageJ software and quantitatively reflected the cellular uptake efficiency.
MFChCpG2 had a much better cellular uptake. To quantify the difference in cellular uptake, the total yellow fluorescence densities in each photo were quantified using the ImageJ software by following the method described in the literature.24 According to the total yellow fluorescence intensity for both types of nanoparticles (Fig. 8(b)), the cellular uptake efficiency of MFChCpG2 was about three times larger than that of BMChCpG. MFChCpG2 indeed showed enhanced cellular uptake. 3.5 In vitro immunostimulatory response Free CpG ODN2006x3-PD is class-B CpG ODN,25 and it specifically stimulates IL-6 production.14,17 After cellular uptake, the release of free CpG ODN2006x3-PD from chitosan/CpG ODN nanoparticles is an important factor in stimulating IL-6 production. Since the pH value of endolysosomes ranges from 5 to 6, in order to mimic the release of CpG ODN2006x3-PD from chitosan/CpG ODN nanoparticles in the endolysosomes, the nanoparticles were suspended in an HAc–NaAc buffer solution (pH 5.5). Under these conditions, the CpG ODNs might be released from the nanoparticles since chitosan would be swollen because of its pKa value around 6.3–6.4.26 Fig. 9(a) shows the release of CpG ODNs from BMChCpG and
1848 | Lab Chip, 2014, 14, 1842–1849
Lab on a Chip
Fig. 9 (a) Release profile of CpG ODNs from chitosan/CpG ODN nanoparticles synthesized via BM and MF methods and (b) the amount of IL-6 produced by PBMCs after incubation with either type of nanoparticle for 48 h.
MFChCpG2. The release of CpG ODNs from MFChCpG2 was slightly slower than that from BMChCpG. This is similar to the case of a previous study14 in which the release of CpG ODNs from BM-processed chitosan/CpG ODN nanoparticles was slow because the electrostatic forces between chitosan and CpG ODNs were too strong. As MFChCpG2 was denser and smaller than BMChCpG, it is reasonable that the release of CpG ODNs from MFChCpG2 was slower than that from BMChCpG. As a result, it was speculated that the immunostimulatory response of MFChCpG2 was much lower than that of BMChCpG. To evaluate the immunostimulatory response, both types of nanoparticles were exposed to PBMCs. After 48 h, the level of IL-6 produced by PBMC cells was quantified (Fig. 9(b)) and the amount for MFChCpG2 was about two times larger than that for BMChCpG. Moreover, MFChCpG2 showed a much better immunostimulatory response. Note that the difference in cellular uptake efficiency was not consistent with the difference in the amount of cytokine. Ideally, after the nanoparticles were taken up by the cells, the amount of IL-6 for MFChCpG2 should be three times larger than that for BMChCpG. As demonstrated in a previous study,14 the release of CpG ODNs also had an effect on cytokine production. As the release rate of CpG ODNs from
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 10 March 2014. Downloaded by University of Chicago on 30/10/2014 09:50:24.
Lab on a Chip
Fig. 10 Schematic presentation of a possible mechanism for chitosan/ CpG ODN nanoparticles to activate the immunostimulatory response.
MFChCpG was slower than that from BMChCpG, there was a slight decrease in the cytokine production for MFChCpG2. This is why an inconsistency was observed. Fig. 10 shows a possible mechanism of IL-6 production that fits our results. Chitosan/DNA nanoparticles entered cells via the endocytosis route and were swollen in the endolysomes to release CpG ODNs. After they were recognized by the TLR9 receptor, the immune reaction was activated to stimulate the production of IL-6.
Conclusions In summary, a microfluidic platform was used to prepare chitosan-based nanoparticles for delivery of CpG ODNs. Our results clearly indicated that chitosan/CpG ODN nanoparticles synthesized in this way not only had much smaller sizes and a narrower size range, but also exhibited enhanced cellular uptake and immunostimulatory response in comparison with the nanoparticles synthesized via the conventional bulk mixing method. Thus, the present microfluidic platform is applicable to the synthesis of novel chitosan/DNA nanoscaled delivery systems with enhanced biological activity.
Notes and references 1 A. Di Martino, M. Sittinger and M. V. Risbud, Biomaterials, 2005, 26, 5983. 2 R. Jayakumar, D. Menon, K. Manzoor, S. V. Nair and H. Tamura, Carbohydr. Polym., 2010, 82, 227. 3 Z. Li, H. R. Ramay, K. D. Hauch, D. Xiao and M. Zhang, Biomaterials, 2005, 26, 3919.
This journal is © The Royal Society of Chemistry 2014
Paper
4 G. Borchard, Adv. Drug Delivery Rev., 2001, 52, 145. 5 H. Ragelle, G. Vandermeulen and V. Préat, J. Controlled Release, 2013, 172, 207. 6 A. Jahn, W. N. Vreeland, M. Gaitan and L. E. Locascio, J. Am. Chem. Soc., 2004, 126, 2674. 7 J. T. Wang, J. Wang and J. J. Han, Small, 2011, 7, 1728. 8 R. Karnik, F. Gu, P. Basto, C. Cannizzaro, L. Dean, W. Kyei-Manu, R. Langer and O. C. Farokhzad, Nano Lett., 2008, 8, 2906. 9 C. G. Koh, X. Kang, Y. Xie, Z. Fei, J. Guan, B. Yu, X. Zhang and L. J. Lee, Mol. Pharmaceutics, 2009, 6, 1333. 10 C. G. Koh, X. Zhang, S. Liu, S. Golan, B. Yu, X. Yang, J. Guan, Y. Jin, Y. Talmon, N. Muthusamy, K. K. Chan, J. C. Byrd, R. J. Lee, G. Marcucci and L. J. Lee, J. Controlled Release, 2010, 141, 62. 11 N. Hanagata, Int. J. Nanomed., 2012, 7, 2181. 12 S. Chinnathambi, S. Chen, S. Ganesan and N. Hanagata, Sci. Rep., 2012, 2, 534. 13 C. Zhi, W. Meng, T. Yamazaki, Y. Bando, D. Golberg, C. Tang and N. Hanagata, J. Mater. Chem., 2011, 21, 5219. 14 S. Chen, H. Zhang, S. Chinnathambi and N. Hanagata, Mater. Sci. Eng., C, 2013, 33, 3382. 15 D. M. Klinman, Nat. Rev. Immunol., 2004, 4, 249. 16 Y. Zhu, W. Meng, H. Gao and N. Hanagata, J. Phys. Chem. C, 2011, 115, 13630. 17 H. Zhang, S. Chen, C. Zhi, T. Yamazaki and N. Hanagata, Int. J. Nanomed., 2013, 8, 1783. 18 H. Q. Mao, K. Roy, V. L. Troung-Le, K. A. Janes, K. Y. Lin, Y. Wang, J. T. August and K. W. Leong, J. Controlled Release, 2001, 70, 399. 19 K. Corsi, F. Chellat, L. Yahia and J. C. Fernandes, Biomaterials, 2003, 24, 1255. 20 J. S. Sun, Y. L. Xiangyu, M. M. Li, W. W. Liu, L. Zhang, D. B. Liu, C. Liu, G. Q. Liu and X. Y. Jiang, Nanoscale, 2013, 5, 5262. 21 F. S. Majedi, M. M. Hasani-Sadrabadi, S. H. Emami, M. A. Shokgrozar, J. J. VanDersarl, E. Dashtimoghadam, A. Bertsch and P. Renaud, Lab Chip, 2013, 13, 204. 22 S. Patil, A. Sandberg, E. Heckert, W. Self and S. Seal, Biomaterials, 2007, 28, 4600. 23 J. E. Fuller, G. T. Zugates, L. S. Ferreira, H. S. Ow, N. N. Nguyen, U. B. Wiesner and R. S. Langer, Biomaterials, 2008, 29, 1526. 24 G. Gabriely, T. Wurdinger, S. Kesari, C. C. Esau, J. Burchard, P. S. Linsley and A. M. Krichevsky, Mol. Cell. Biol., 2008, 28, 5369. 25 W. Meng, T. Yamazaki, Y. Nishida and N. Hanagata, BMC Biotechnol., 2011, 11, 88. 26 J. Li, J. F. Revol and R. H. Marchessault, J. Colloid Interface Sci., 1996, 183, 365.
Lab Chip, 2014, 14, 1842–1849 | 1849
Published on 10 March 2014. Downloaded by University of Chicago on 30/10/2014 09:50:24.
PAPER
Cite this: Lab Chip, 2014, 14, 1842
View Journal | View Issue
Microfluidic generation of chitosan/CpG oligodeoxynucleotide nanoparticles with enhanced cellular uptake and immunostimulatory properties Song Chen,*ab Huijie Zhang,b Xuetao Shi,c Hongkai Wu*cd and Nobutaka Hanagata*be Chitosan/cytosine-phosphodiester-guanine oligodeoxynucleotide (CpG ODN) nanoparticles as potential immunostimulatory adjuvants were synthesized by the conventional bulk mixing (BM) method and a novel microfluidic (MF) method. Their size and size distribution, CpG ODN loading efficiency, surface charge, biocompatibility, cellular uptake, and immunostimulatory response were investigated. In the BM method, nanoparticles were synthesized by vortexing a mixture of chitosan solution and CpG ODN2006x3-PD solution. In the MF method, the nanoparticles were synthesized by rapidly mixing a chitosan solution and CpG ODN solution in a poly(dimethylsiloxane) microfluidic device. Our results indicated that particle size and size distribution, CpG ODN loading efficiency, and surface charge could be easily adjusted by using the tuning preparation method and controlling the flow ratio of fluid rates in the different microfluidic channels. Compared with the BM method, the MF method yielded a decrease in particle size and size range, an increase in CpG ODN loading efficiency, and a decrease in surface charge. After the particles were exposed to 293XL-hTLR9 cells, a water-soluble tetrazolium salt assay indicated that the BM and MF-processed nanoparticles had no significant toxicity and were biocompatible. An immunochemical assay indicated that both types of nanoparticles entered 293XL-hTLR9 cells and were located in the endolysosomes. The MF-processed nanoparticles showed much higher cellular uptake
Received 6th January 2014, Accepted 7th March 2014 DOI: 10.1039/c4lc00015c www.rsc.org/loc
efficiency. After the particles were exposed to peripheral blood mononuclear cells, an enzyme-linked immunosorbent assay quantitatively indicated that both types of nanoparticles stimulated the production of interleukin-6 and the MF-processed nanoparticles showed a much stronger immunostimulatory response. These results indicate that the MF method can be used to synthesize nanoparticles with a controllable size and size range for enhancing the biological activity of DNA and other biomolecules.
1. Introduction Chitosan is a linear polysaccharide biopolymer consisting of N-acetyl-D-glucosamine and D-glucosamine and is derived from chitin extracted from sea crustaceans. It has excellent biocompatibility and biodegradability.1,2 It also has a strong positive charge arising from rapid protonation of its two subunits under acidic conditions, or even neutral conditions, and thus has strong affinity to negatively charged materials such as a
JSPS Research Fellow, 8 Ichibancho, Chiyoda-ku, Tokyo 102-8472, Japan Biomaterials Unit, National Institute for Materials Science, 1-2-1 Sengen, Tsukuba, 305-0047, Japan. E-mail: [email protected]; Fax: +81 29 859 2449; Tel: +81 29 859 2000 c WPI Advanced Institute for Materials Research, Tohoku University, Katahira 2-1-1, Aobaku, Sendai, 980-8577, Japan d Department of Chemistry, Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong, China. E-mail: [email protected] e Nanotechnology Innovation Station, National Institute for Materials Science, 1-2-1 Sengen, Tsukuba, Ibaraki 305-0047, Japan. E-mail: [email protected] b
1842 | Lab Chip, 2014, 14, 1842–1849
alginate3 and DNA.4 Numerous studies have demonstrated that chitosan-based materials can effectively pack and condense DNA to produce a chitosan/DNA complex for protecting DNA from nuclease degradation. Employing chitosan-based materials as a delivery system of DNA has thus become a hot research topic in gene delivery systems.4,5 Controlling the size and size distribution of chitosan/DNA nanoparticles will be crucial to achieving the goal of controllable cellular uptake and biological activity. The conventional preparation route involves bulk mixing (BM) of chitosan solution and DNA solution and spontaneous self-assembly into complexes due to electrostatic forces. Such a self-assembly process typically occurs in a reaction platform with a characteristic length on the order of centimeters and results in highly heterogeneous chemical and/or mechanical conditions.6 Thus, it is still a challenge to control the size and size distribution of chitosan/DNA nanoparticles. In contrast to the BM process, a microfluidic (MF) platform manipulates fluids at the microscale level. Since the
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 10 March 2014. Downloaded by University of Chicago on 30/10/2014 09:50:24.
Lab on a Chip
reaction in the microchannels is mainly driven by interfacial forces, heat- and mass-transfers are significantly enhanced,6 and this results in a much better control over the flow rate, reaction time, and concentration of reagents compared with the BM process. It has been demonstrated that the MF method provides a much better control over the size and size distribution of nano/microparticles.7,8 Recent studies have shown that MF-processed gene delivery systems exhibit significantly enhanced biological activity. In particular, MF-processed PEI/pGFP nanocomplexes were found to be much smaller than BM-processed nanoparticles and had much higher transfection efficiency in NIH 3T3 cells.9 Moreover, MF-processed antisense oligodeoxyribonucleotide lipopolyplex nanoparticles had higher levels of Bcl-2 antisense uptake and showed more efficient downregulation of the Bcl-2 protein level in comparison with BM-processed nanoparticles.10 However, the MF platform has rarely been used to synthesize chitosan/DNA nanoparticles with controllable size and size distribution and enhanced biological activity. Cytosine-phosphodiester-guanine oligodeoxynucleotides (CpG ODNs) have been used as immune adjuvants because they can be recognized by Toll-like receptor 9 (TLR9) found in antigen-presenting cells (APCs) and B cells to activate the immune system.11 However, naked CpG ODNs have the drawbacks of low cellular uptake and low immunostimulatory response. Nanoscale delivery systems can effectively deliver various DNA and RNA into cells since they are small enough to be taken up by the cells via endocytosis. Recent studies have indicated that the immunostimulatory response of CpG ODNs can be significantly enhanced with the help of various nanoscale delivery systems including silicon dots,12 boron nitride nanospheres,13 and chitosan–silica nanohybrids.14 Our previous studies have demonstrated that chitosan and its derivatives can condense CpG ODNs and deliver them to immune cells to activate the immunostimulatory response.14 However, the cellular uptake and immunostimulatory response of chitosan/CpG ODN complexes remained very low. Considering the advantages of the MF method over the BM method, we hypothesized that MF-processed chitosan/CpG ODN nanoparticles would be much smaller than BM-processed nanoparticles and might have better cellular uptake efficiency and immunostimulatory activity. To demonstrate this hypothesis, in the present study, we used both methods to synthesize immunostimulatory chitosan/CpG ODN nanoparticles and we compared the resulting nanoparticles in terms of their size and size distribution, CpG ODN loading efficiency, surface charge, biocompatibility, cellular uptake, and immunostimulatory response. In vitro biocompatibility and cellular uptake were evaluated by incubating the nanoparticles with 293XL-hTLR9 cells, while in vitro immunostimulatory response was evaluated by exposing the nanoparticles to peripheral blood mononuclear cells (PBMC) and quantifying the amount of IL-6 produced by the cells via the enzyme-linked immunosorbent assay (ELISA) assay.
This journal is © The Royal Society of Chemistry 2014
Paper
2. Materials and methods 2.1 Fabrication of the microfluidic device The poly(dimethylsiloxane) (PDMS) microfluidic device, schematically depicted in Fig. 1, had multiple flow microchannels (200 μm in width and 50 μm in depth). In this device, the flow stream along the central channel and the flow streams along the two side channels meet at the cross-section position and focus the hydrodynamic flow in the central channel. A silicon master with multiple SU-8 microchannel patterns was first created using a standard photolithographic microfabrication process. A mixture of PDMS prepolymer and curing agent (10 : 1; Silpot 184, Dow Corning Toray Co., Ltd., Tokyo, Japan) was spin-coated on the silicon master. It was then degassed and cured at 70 °C for 2 h. The resulting PDMS sheet with multiple microchannels was removed from the silicon master; the PDMS sheet and a glass slide were separately exposed to O2 plasma in a plasma cleaner (PDC200, Yamato Scientific, Japan). After 45 seconds of exposure, the PDMS sheets were taken out of the cleaner and rapidly sealed to the glass slide to produce a microfluidic device with multiple microchannels. The device was connected to syringes using silicon rubber tubes and the solution of reagents was injected into the channels using programmable syringe pumps (New Era Pump Systems Inc., USA).
2.2 Synthesis of chitosan/CpG ODN nanoparticles Chitosan powders (60–120 kDa, Sigma-Aldrich) were added to an HAc–NaAc solution (pH = 5) and stirred at room temperature for 2 h to produce 0.1% (w/v) chitosan solutions. Before use, the chitosan solution was sterilized using a 0.22 μm filter. Natural phosphodiester CpG ODN2006x3-PD (5′TCGTCGTTTTGTCGTTTTGTCGTTTCGTCGTTTTGTCGTTTTGTCGTTTCGTCGTTTTGTCGTTTTGTCGTT-3′) (72 mer, Fasmac
Fig. 1 Schematic illustration of a microfluidic PDMS device. Chitosan solution was injected from inlet 1 and inlet 2 into the side channels, while CpG ODN solution was injected from inlet 3 into the center channel. The resultant chitosan/CpG ODN nanoparticles were collected from the outlet.
Lab Chip, 2014, 14, 1842–1849 | 1843
View Article Online
Published on 10 March 2014. Downloaded by University of Chicago on 30/10/2014 09:50:24.
Paper
Lab on a Chip
Inc., Kanagawa, Japan) was diluted in sterilized water to a concentration of 300 μg mL−1. Chitosan/CpG ODN nanoparticles were then synthesized by the BM and MF methods. In the BM method, 400 μL of chitosan solution was mixed with 200 μL of CpG ODN solution, and the mixture was vortexed for 1 min and left for 30 min at room temperature to yield chitosan/CpG ODN nanoparticles (coded as BMChCpG). In the MF method, the programmable syringe pumps separately injected the chitosan solution from inlet 1 and inlet 2 into the side channels at a controlled flow rate of 100–200 μL min−1, while they injected the CpG ODN solution from inlet 3 into the central channel at a controlled flow rate of 100–200 μL min−1. The flow ratio of chitosan/CpG ODN was set as 0.5, 1 or 2. The chitosan and CpG ODN solutions met at the cross-junction position and hydrodynamic focusing occurred in the central channel to produce chitosan/CpG ODN nanoparticles, which were collected at the outlet. The resultant chitosan/CpG ODN nanoparticles were coded according to the flow ratio as MFChCpG0.5, MFChCpG1, or MFChCpG2. Table 1 lists the reaction parameters and sample abbreviations.
2.3 Characterization The size and morphology of the resultant nanoparticles were observed under a field emission scanning electron microscope (FE-SEM; JSM 6500, JEOL, Tokyo, Japan). The particle size distribution was measured by dynamic light scattering (DLS-6000AL, Photal Otsuka Electronics, Japan), whereas the zeta potential was measured using a laser electrophoresis zeta-potential analyzer (LEZA-600, Otsuka, Japan). To evaluate the encapsulation efficiency of CpG ODNs, freshly synthesized chitosan/CpG ODN nanoparticles were collected by centrifuging the solution at 12 000 rpm for 5 min. Free CpG ODNs in the supernatant were quantified using a NanoDrop 2000 spectrophotometer at 260 nm. The encapsulation efficiency of CpG ODNs in the chitosan/CpG ODN nanoparticles was calculated using the following equation: Encapsulation efficiency(%) = [(Wo − Ws)/Wo] × 100% Here, Wo is the amount of CpG ODNs before reacting with chitosan and Ws is the amount of CpG ODNs in the supernatant. To evaluate the release behavior of CpG ODNs, chitosan/ CpG ODN nanoparticles collected by centrifugation were re-suspended in an HAc–NaAc buffer solution (pH 5.5) in a 1.5 mL Eppendorf tube and placed in a shaking bath at
37 °C. At predetermined time intervals, the nanoparticles were collected from the suspension by centrifugation at 12 000 rpm at 5 min and re-suspended in a fresh HAc–NaAc buffer solution. Free CpG ODNs in the supernatant were quantified using the above-described method. The percentage of CpG ODNs released from the nanoparticles was calculated using the following equation: Release percentage(%) = (Wr/Wt) × 100% Here, Wr is the amount of CpG ODNs released from the nanoparticles into the supernatant and Wt is the total amount of CpG ODNs in the nanoparticles.
2.4 Cell toxicity 293XL-hTLR9 cells stably expressing human TLR9 (Invivogen) were grown in Dulbecco's Modified Eagle Medium (DMEM) supplemented with 10% FBS, 100 U mL−1 penicillin and 100 μg mL−1 streptomycin in a humidified atmosphere of 5% CO2 and 95% air at 37 °C and used to evaluate the toxicity of the nanoparticles. Cells were detached from the flask using trypsin, seeded in a 96-well culture plate at a density of 1 × 104 cells per well, and grown in DMEM overnight. The culture medium was then replaced with fresh medium containing chitosan/CpG ODN nanoparticles. A water-soluble tetrazolium salt (WST-1) assay was performed to evaluate cell viability. After 24 h and 48 h of culturing, the culture medium was replaced with a WST-1 containing culture medium. After another 4 h of culturing, the absorbance at 450 nm was measured using a microplate reader (MTP-880, Corona Electric, Japan) in order to determine if formazan was formed from the water-soluble tetrazolium salt. Cells without any chitosan/CpG ODN nanoparticles were used as the control. The morphologies of the cells incubated with and without nanoparticles were visualized under an inverted microscope (DM2500, Leica).
2.5 Cellular uptake CpG ODNs are recognized by TLR9 receptors which are located in the endolysosomes.11–14 A successful cellular uptake of chitosan/CpG ODN nanoparticles is thus crucial to activation of the immune response. To monitor the behavior of cellular uptake, CpG ODNs were replaced with 3′-fluorescein isothiocyanate (FITC)-labeled CpG ODNs when the chitosan/FITC–CpG ODN nanoparticles were prepared using either method.
Table 1 Reaction parameters and sample abbreviations
1 2 3 4
Sample abbreviation
Preparation method
Chitosan flow rate (μL min−1)
CpG ODN flow rate (μL min−1)
Flow ratio
BMChCpG MFChCpG0.5 MFChCpG1 MFChCpG2
BM MF MF MF
None 100 100 200
None 200 100 100
None 0.5 1 2
1844 | Lab Chip, 2014, 14, 1842–1849
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 10 March 2014. Downloaded by University of Chicago on 30/10/2014 09:50:24.
Lab on a Chip
Paper
293XL-hTLR9 cells were seeded in a four-compartment glass-bottom dish (35 mm) at a density of 4 × 104 per well and cultured overnight. The culture medium was then replaced with a fresh medium containing chitosan/FITC–CpG ODN nanoparticles. To confirm whether the nanoparticles had entered the cells and endolysosomes, lysosome-associated membrane protein 1 (LAMP-1) was immunochemically stained as an endolysosomal marker. Anti-LAMP-1 antibody (ab24170, Abcam) was used as the primary antibody, while Alexa Fluor® 555 donkey anti-rabbit IgG (H + L) (Invitrogen) was used as the secondary antibody. Cells were visualized under a confocal laser scanning microscope (CLSM; Leica). 2.6 In vitro immunostimulatory response Peripheral blood mononuclear cells (PBMCs, Cellular Technology Limited, US) were used to evaluate the in vitro immunostimulatory response of the nanoparticles. RPMI 1640 medium was supplemented with 10% FBS, 100 U mL−1 penicillin, and 100 μg mL−1 streptomycin. The PBMCs were thawed and suspended in RPMI 1640 medium at a density of 106 cells mL−1. 190 μL of the cell suspension was mixed with 10 μL of chitosan/CpG ODN nanoparticles synthesized via either method. The mixture was put in a 96 well culture plate in a humidified atmosphere of 5% CO2 and 95% air at 37 °C. After 48 h, cell cytokine interleukin-6 (IL-6) secreted by the PMBCs that were stimulated by the nanoparticles in the culture medium was quantified by enzyme-linked immunosorbent assay (ELISA) using a Human IL-6-ELISA Kit (eBiosciences, Vienna, Austria) according to the manufacturer's instructions.
Fig. 2 Phase microscope image of a microfluidic PDMS device (a) and fluorescent microscope images of hydrodynamic focusing of a mixture of fluorescein (green) fluid and rhodamine B (red) fluid at the meeting position of channels. The rhodamine B/fluorescein flow ratios were 0.5 (b), 1 (c), and 2 (d).
be seen that the width of the green region at the crosssection of the device decreased as the flow ratio of the fluid rate in the side channel relative to the central channel increased from 0.5 to 2. This indicates that increasing the flow ratio resulted in a decrease in the mixed amount of fluid coming from the central channel and an increase in the mixed amount of fluid coming from the side channels during hydrodynamic focusing. That is, the mixing behavior of different fluids at the cross-section position could be easily adjusted by tuning the flow ratio.
2.7 Statistical analysis All results were obtained in triplicate, expressed in terms of the mean and standard deviation (SD) and analyzed using one-way analysis of variance (ANOVA) with a significance level of p < 0.05.
3. Results and discussion 3.1 Hydrodynamic focusing Hydrodynamic focusing is a unique feature of the microfluidic platform.6–8 Its occurrence enables the microfluidic device to control the size and size distribution of the resultant nanoparticles.8 To observe hydrodynamic focusing, a solution of rhodamine B was used to mimic the solution of chitosan and was separately injected into the two side channels from inlet 1 and inlet 2, while a solution of fluorescein was used to mimic the solution of CpG ODNs and was injected into the central channel from inlet 3. The microfluidic device was mounted on a fluorescent microscope. Fig. 2 shows that the green fluorescein solution was squeezed by the red rhodamine B solution and clearly reveals hydrodynamic focusing due to rapid mixing of the flow streams. Thus, the present device is a typical microfluidic platform and can be potentially used to synthesize chitosan/CpG ODN nanoparticles of controllable size and size distribution. It can
This journal is © The Royal Society of Chemistry 2014
3.2 Particle characterization CpG ODNs are recognized by TLR9 receptors, which are only expressed in immune cells such as APCs and B cells.11,15 Several nanoscaled delivery systems have been developed and they have been used to deliver CpG ODNs into those immune cells. The size of the nanoparticles has a significant effect on cellular uptake. Zhu et al. reported that polylysine-coated mesoporous silica spheres with a diameter of 500 nm could deliver CpG ODNs into 293XL-hTLR9 cells in order to activate the immune response.16 Zhang et al. reported that chitosancoated boron nitride nanospheres with a diameter of 150 nm delivered CpG ODN2006x3-PD into PBMC cells to stimulate IL-6 production.17 In our previous study, we demonstrated that chitosan–silica nanoparticles with a diameter of 200–300 nm could deliver CpG ODN2006x3-PD into 293XL-hTLR9 cells to activate the production of IL-6.14 In most cases in these studies, the size of the nanoscale delivery systems for delivery of CpG ODNs had to be less than 500 nm. The size and size distribution of BM-processed chitosan/ DNA nanoparticles have been investigated in detail, and it has been found that they are greatly influenced by parameters such as chitosan deacetylation degree, chitosan concentration, chitosan molecular weight, and DNA concentration.18,19 Since the
Lab Chip, 2014, 14, 1842–1849 | 1845
View Article Online
Published on 10 March 2014. Downloaded by University of Chicago on 30/10/2014 09:50:24.
Paper
BM method mainly relies on bulk forces between reactants and provides inhomogeneous reaction conditions, it is hard to control the size and size distribution of the resulting nanoparticles. Our previous study has shown that BM-processed chitosan/CpG ODN nanoparticles exhibited a wide size distribution ranging from several hundred nm to several micrometers due to agglomeration.14 Unlike the BM method, the MF method mainly relies on interfacial forces between fluids, and such forces play a key role in controlling the size and size distribution of the final nanoparticles. Jahn et al. found that the interfacial forces were closely associated with the flow ratio of the fluid rate of the side channel relative to that of the center channel and that they increased as the flow ratio increased, resulting in a decrease in the particle size and size range.6 Thus, it is speculated that the size and size distribution of chitosan/CpG ODN nanoparticles can be tuned by altering the flow ratio. DLS measurements provide direct evidence of the particle size and size distribution. Fig. 3 shows the DLS results for BMChCpG, MFChCpG0.5, MFChCpG1, and MFChCpG2. BMChCpG shows a wide and bimodal size distribution of 398 ± 291 nm, while all of the others show single modal size distribution. The size distributions of MFChCpG0.5, MFChCpG1, and MFChCpG2 were 234 ± 80 nm, 179 ± 56 nm, and 204 ± 45 nm, respectively. In contrast, the MF method resulted in a decrease in the particle size and size range compared with the BM method. Moreover, the size of the MF-processed nanoparticles was less than 500 nm and was thought to be suitable for delivery of CpG ODNs into the immune cells. Furthermore, we compared BM and MF-processed chitosan/ CpG ODN nanoparticles in terms of their size and morphology. Fig. 4 shows typical SEM images of BMChCpG and MFChCpG2. BMChCpG (Fig. 4(a)) was irregular in shape and seriously agglomerated with a size more than 1000 nm (arrows), while
Fig. 3 Size distributions of chitosan/CpG ODN nanoparticles: (a) BMChCpG, (b) MFChCpG0.5, (c) MFChCpG1, and (d) MFChCpG2.
1846 | Lab Chip, 2014, 14, 1842–1849
Lab on a Chip
Fig. 4 SEM images of chitosan/CpG ODN nanoparticles synthesized by the BM method (a) and the MF method (b).
MFChCpG2 had a much better morphology and smaller size of less than 500 nm (Fig. 4(b)). In contrast, the MF method is indeed much better at controlling the size and size distribution of the resultant nanoparticles with a diameter less than 500 nm for delivery of CpG ODNs. The above results indicate that the MF method is a simple way of preparing chitosan/CpG ODN nanoparticles with a small size, narrow size distribution, and good morphology. To the best of our knowledge, this is the first demonstration of the fine control that can be obtained by using microfluidics to manipulate the size and size distribution of chitosan/DNA nanoparticles. In the case of the BM method, the binding of DNA to chitosan is mainly driven by electrostatic interactions. As the BM method works at the centimeter level6 and has difficulty in providing homogenous reaction conditions, the mixing of chitosan and CpG ODNs is not homogenous, resulting in loose and uncontrollable packing and large nanoparticles with a wide size distribution (Fig. 5(a)). In the MF device, enhanced mixing at a high flow rate inside 3D microchannels could be obtained, leading then to smaller nanoparticles.20 The formation of smaller and more monodisperse nanoparticles might be also due to the mixing time being much shorter than the particle's aggregation time.8,21 Thus, the infusion of chitosan and CpG ODNs is much more homogenous and precise. The packing of chitosan and CpG ODNs is much stronger and denser and results in a much smaller size and narrower size distribution (Fig. 5(b)).
Fig. 5 Possible formation mechanisms of chitosan/CpG ODN nanoparticles synthesized by the BM method (a) and the MF method (b).
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 10 March 2014. Downloaded by University of Chicago on 30/10/2014 09:50:24.
Lab on a Chip
Consequently, it was speculated that the CpG ODN loading efficiency of the MF-processed nanoparticles would be much higher than that of BM-processed nanoparticles and the surface charge of MF-processed nanoparticles is smaller than that of the BM-processed ones. Fig. 6(a) and (b) show the CpG ODN loading efficiency and surface charge of the resultant nanoparticles. Indeed, the CpG ODN loading efficiency of MFChCpG0.5, MFChCpG1 and MFChCpG2 was much higher than that of BMChCpG, while the surface charge of BMChCpG was much higher than that of MFChCpG0.5, MFChCpG1 and MFChCpG2. As depicted in Fig. 2, increasing the flow ratio led to a decrease in the mixing amount of CpG ODNs. Therefore, it is reasonable that the CpG ODN loading efficiency decreased and the surface charge increased as the flow ratio increased from 0.5 to 2 due to the difference in the mixing behavior between the BM and MF methods. The cell membrane has a net negative surface charge.22 A positive surface charge is thus advantageous for cellular uptake of chitosan/DNA nanoparticles. For this reason, MFChCpG2, which has a smaller size and larger surface charge, was selected for the biological assessments presented next.
3.3 Cytotoxicity Biocompatibility is one of the most important properties for a nanoscale delivery system to have, since toxicity can alter or disrupt cell membranes.23 Here, the BM- and MF-processed
Paper
Fig. 7 WST-1 results (a), cell number (b), and cell morphology (c) of chitosan/CpG ODN nanoparticles synthesized by BM and MF methods after exposure to 293XL-hTLR9 cells.
chitosan/CpG ODN nanoparticles were exposed to 293XLhTLR9 cells, and cell viability was evaluated by the WST-1 assay. Fig. 7(a) shows the results of the WST-1 assay. The cell viability of both types of nanoparticles was very similar to that of the control cells without any nanoparticles after 24 h and 48 h of culturing, indicating that both types of nanoparticles had no significant toxicity and were biocompatible. Fig. 7(b) shows that the cell number increased as the culturing time increased, indicating that the nanoparticles had no effect on cell proliferation and further confirming their biocompatibility. A similarity in the cell size and morphology was observed between the cells that took up the nanoparticles and the control (Fig. 7(c)), and this is direct evidence of the biocompatibility of both types of nanoparticles.
3.4 Cellular uptake
Fig. 6 CpG ODN loading efficiency (a) and surface charge (b) of chitosan/CpG ODN nanoparticles: BMChCpG, MFChCpG0.5, MFChCpG1, and MFChCpG2.
This journal is © The Royal Society of Chemistry 2014
Previous studies have demonstrated that BM-processed chitosan/CpG ODN nanoparticles could enter 293XL-hTLR9 and were located in the endolysosomes.14 However, their large size and size distribution resulted in low cellular uptake and poor immunostimulatory response. As BMChCpG was much larger than MFChCpG2, it was speculated that the cellular uptake of BMChCpG would be much lower than that of MFChCpG2. To monitor the cellular uptake behavior, FITC– CpG ODNs were used to prepare BMChCpG and MFChCpG2. Fig. 8(a) shows confocal microscope fluorescent images of BMChCpG and MFChCpG2 after exposure to 293XL-hTLR9 cells for 48 h. Chitosan/FITC–CpG ODN nanoparticles exhibited green fluorescence, while endolysosomes marked with LAMP-1 exhibited red fluorescence. The green fluorescence combined with the red fluorescence to make yellow fluorescence. The yellow fluorescent points indicate that BMChCpG and MFChCpG2 were taken up by the cells and were located in the endolysosomes. The number of yellow points was much larger for MFChCpG2, indicating that
Lab Chip, 2014, 14, 1842–1849 | 1847
View Article Online
Published on 10 March 2014. Downloaded by University of Chicago on 30/10/2014 09:50:24.
Paper
Fig. 8 (a) Confocal laser microscope fluorescent images of 293XLhTLR9 cells after incubation with chitosan/FITC–CpG ODN nanoparticles (green) synthesized via BM and MF methods. Endolysosomes were immunochemically stained (red). Yellow fluorescence indicates that the nanoparticles were taken up by the cells and entered the endolysosomes. (b) Total yellow fluorescence intensities in Fig. 8(a) were calculated using the ImageJ software and quantitatively reflected the cellular uptake efficiency.
MFChCpG2 had a much better cellular uptake. To quantify the difference in cellular uptake, the total yellow fluorescence densities in each photo were quantified using the ImageJ software by following the method described in the literature.24 According to the total yellow fluorescence intensity for both types of nanoparticles (Fig. 8(b)), the cellular uptake efficiency of MFChCpG2 was about three times larger than that of BMChCpG. MFChCpG2 indeed showed enhanced cellular uptake. 3.5 In vitro immunostimulatory response Free CpG ODN2006x3-PD is class-B CpG ODN,25 and it specifically stimulates IL-6 production.14,17 After cellular uptake, the release of free CpG ODN2006x3-PD from chitosan/CpG ODN nanoparticles is an important factor in stimulating IL-6 production. Since the pH value of endolysosomes ranges from 5 to 6, in order to mimic the release of CpG ODN2006x3-PD from chitosan/CpG ODN nanoparticles in the endolysosomes, the nanoparticles were suspended in an HAc–NaAc buffer solution (pH 5.5). Under these conditions, the CpG ODNs might be released from the nanoparticles since chitosan would be swollen because of its pKa value around 6.3–6.4.26 Fig. 9(a) shows the release of CpG ODNs from BMChCpG and
1848 | Lab Chip, 2014, 14, 1842–1849
Lab on a Chip
Fig. 9 (a) Release profile of CpG ODNs from chitosan/CpG ODN nanoparticles synthesized via BM and MF methods and (b) the amount of IL-6 produced by PBMCs after incubation with either type of nanoparticle for 48 h.
MFChCpG2. The release of CpG ODNs from MFChCpG2 was slightly slower than that from BMChCpG. This is similar to the case of a previous study14 in which the release of CpG ODNs from BM-processed chitosan/CpG ODN nanoparticles was slow because the electrostatic forces between chitosan and CpG ODNs were too strong. As MFChCpG2 was denser and smaller than BMChCpG, it is reasonable that the release of CpG ODNs from MFChCpG2 was slower than that from BMChCpG. As a result, it was speculated that the immunostimulatory response of MFChCpG2 was much lower than that of BMChCpG. To evaluate the immunostimulatory response, both types of nanoparticles were exposed to PBMCs. After 48 h, the level of IL-6 produced by PBMC cells was quantified (Fig. 9(b)) and the amount for MFChCpG2 was about two times larger than that for BMChCpG. Moreover, MFChCpG2 showed a much better immunostimulatory response. Note that the difference in cellular uptake efficiency was not consistent with the difference in the amount of cytokine. Ideally, after the nanoparticles were taken up by the cells, the amount of IL-6 for MFChCpG2 should be three times larger than that for BMChCpG. As demonstrated in a previous study,14 the release of CpG ODNs also had an effect on cytokine production. As the release rate of CpG ODNs from
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 10 March 2014. Downloaded by University of Chicago on 30/10/2014 09:50:24.
Lab on a Chip
Fig. 10 Schematic presentation of a possible mechanism for chitosan/ CpG ODN nanoparticles to activate the immunostimulatory response.
MFChCpG was slower than that from BMChCpG, there was a slight decrease in the cytokine production for MFChCpG2. This is why an inconsistency was observed. Fig. 10 shows a possible mechanism of IL-6 production that fits our results. Chitosan/DNA nanoparticles entered cells via the endocytosis route and were swollen in the endolysomes to release CpG ODNs. After they were recognized by the TLR9 receptor, the immune reaction was activated to stimulate the production of IL-6.
Conclusions In summary, a microfluidic platform was used to prepare chitosan-based nanoparticles for delivery of CpG ODNs. Our results clearly indicated that chitosan/CpG ODN nanoparticles synthesized in this way not only had much smaller sizes and a narrower size range, but also exhibited enhanced cellular uptake and immunostimulatory response in comparison with the nanoparticles synthesized via the conventional bulk mixing method. Thus, the present microfluidic platform is applicable to the synthesis of novel chitosan/DNA nanoscaled delivery systems with enhanced biological activity.
Notes and references 1 A. Di Martino, M. Sittinger and M. V. Risbud, Biomaterials, 2005, 26, 5983. 2 R. Jayakumar, D. Menon, K. Manzoor, S. V. Nair and H. Tamura, Carbohydr. Polym., 2010, 82, 227. 3 Z. Li, H. R. Ramay, K. D. Hauch, D. Xiao and M. Zhang, Biomaterials, 2005, 26, 3919.
This journal is © The Royal Society of Chemistry 2014
Paper
4 G. Borchard, Adv. Drug Delivery Rev., 2001, 52, 145. 5 H. Ragelle, G. Vandermeulen and V. Préat, J. Controlled Release, 2013, 172, 207. 6 A. Jahn, W. N. Vreeland, M. Gaitan and L. E. Locascio, J. Am. Chem. Soc., 2004, 126, 2674. 7 J. T. Wang, J. Wang and J. J. Han, Small, 2011, 7, 1728. 8 R. Karnik, F. Gu, P. Basto, C. Cannizzaro, L. Dean, W. Kyei-Manu, R. Langer and O. C. Farokhzad, Nano Lett., 2008, 8, 2906. 9 C. G. Koh, X. Kang, Y. Xie, Z. Fei, J. Guan, B. Yu, X. Zhang and L. J. Lee, Mol. Pharmaceutics, 2009, 6, 1333. 10 C. G. Koh, X. Zhang, S. Liu, S. Golan, B. Yu, X. Yang, J. Guan, Y. Jin, Y. Talmon, N. Muthusamy, K. K. Chan, J. C. Byrd, R. J. Lee, G. Marcucci and L. J. Lee, J. Controlled Release, 2010, 141, 62. 11 N. Hanagata, Int. J. Nanomed., 2012, 7, 2181. 12 S. Chinnathambi, S. Chen, S. Ganesan and N. Hanagata, Sci. Rep., 2012, 2, 534. 13 C. Zhi, W. Meng, T. Yamazaki, Y. Bando, D. Golberg, C. Tang and N. Hanagata, J. Mater. Chem., 2011, 21, 5219. 14 S. Chen, H. Zhang, S. Chinnathambi and N. Hanagata, Mater. Sci. Eng., C, 2013, 33, 3382. 15 D. M. Klinman, Nat. Rev. Immunol., 2004, 4, 249. 16 Y. Zhu, W. Meng, H. Gao and N. Hanagata, J. Phys. Chem. C, 2011, 115, 13630. 17 H. Zhang, S. Chen, C. Zhi, T. Yamazaki and N. Hanagata, Int. J. Nanomed., 2013, 8, 1783. 18 H. Q. Mao, K. Roy, V. L. Troung-Le, K. A. Janes, K. Y. Lin, Y. Wang, J. T. August and K. W. Leong, J. Controlled Release, 2001, 70, 399. 19 K. Corsi, F. Chellat, L. Yahia and J. C. Fernandes, Biomaterials, 2003, 24, 1255. 20 J. S. Sun, Y. L. Xiangyu, M. M. Li, W. W. Liu, L. Zhang, D. B. Liu, C. Liu, G. Q. Liu and X. Y. Jiang, Nanoscale, 2013, 5, 5262. 21 F. S. Majedi, M. M. Hasani-Sadrabadi, S. H. Emami, M. A. Shokgrozar, J. J. VanDersarl, E. Dashtimoghadam, A. Bertsch and P. Renaud, Lab Chip, 2013, 13, 204. 22 S. Patil, A. Sandberg, E. Heckert, W. Self and S. Seal, Biomaterials, 2007, 28, 4600. 23 J. E. Fuller, G. T. Zugates, L. S. Ferreira, H. S. Ow, N. N. Nguyen, U. B. Wiesner and R. S. Langer, Biomaterials, 2008, 29, 1526. 24 G. Gabriely, T. Wurdinger, S. Kesari, C. C. Esau, J. Burchard, P. S. Linsley and A. M. Krichevsky, Mol. Cell. Biol., 2008, 28, 5369. 25 W. Meng, T. Yamazaki, Y. Nishida and N. Hanagata, BMC Biotechnol., 2011, 11, 88. 26 J. Li, J. F. Revol and R. H. Marchessault, J. Colloid Interface Sci., 1996, 183, 365.
Lab Chip, 2014, 14, 1842–1849 | 1849
