Stability, Intracellular Delivery, and Release of siRNA from Chitosan Nanoparticles Using Different Cross-Linkers

Chitosan (CS) nanoparticles have been extensively studied for siRNA delivery; however, their stability and efficacy are highly dependent on the types of cross-linker used. To address this issue, three common cross-linkers; tripolyphosphate (TPP), dextran sulphate (DS) and poly-D-glutamic acid (PGA) were used to prepare siRNA loaded CS-TPP/DS/PGA nanoparticles by ionic gelation method. The resulting nanoparticles were compared with regard to their physicochemical properties including particle size, zeta potential, morphology, binding and encapsulation efficiencies. Among all the formulations prepared with different cross linkers, CS-TPP-siRNA had the smallest particle size (ranged from 127 ± 9.7 to 455 ± 12.9 nm) with zeta potential ranged from +25.1 ± 1.5 to +39.4 ± 0.5 mV, and high entrapment (>95%) and binding efficiencies. Similarly, CS-TPP nanoparticles showed better siRNA protection during storage at 4˚C and as determined by serum protection assay. TEM micrographs revealed the assorted morphology of CS-TPP-siRNA nanoparticles in contrast to irregular morphology displayed by CS-DS-siRNA and CS-PGA-siRNA nanoparticles. All siRNA loaded CS-TPP/DS/PGA nanoparticles showed initial burst release followed by sustained release of siRNA. Moreover, all the formulations showed low and concentration-dependent cytotoxicity with human colorectal cancer cells (DLD-1), in vitro. The cellular uptake studies with CS-TPP-siRNA nanoparticles showed successful delivery of siRNA within cytoplasm of DLD-1 cells. The results demonstrate that ionically cross-linked CS-TPP nanoparticles are biocompatible non-viral gene delivery system and generate a solid ground for further optimization studies, for example with regard to steric stabilization and targeting.


Introduction
Small interfering RNAs (siRNAs) have the potential for therapeutic application in many diseases, including cancer [1,2]. Despite efficient and reliable gene silencing activity in vitro, only limited siRNA intracellular delivery has been achieved, owing to its rapid enzymatic degradation and poor cellular uptake [3,4]. Therefore, effective delivery systems that can protect USA). Human colorectal adenocarcinoma cells (DLD-1) were obtained from ATCC (Manassas, VA, USA). Phosphate-buffered saline (PBS) was obtained from Sigma-Aldrich (St. Louis, MO, USA). Dulbecco's modified Eagle's medium (DMEM), penicillin-streptomycin (penstrep), and fetal bovine serum (FBS) were purchased from Gibco (New York, USA). Heparin sodium was purchased from Leo (Ballerup, Denmark). Deionized water of resistivity 18.2MOcm was used in all the experiments. Acetic acid and other chemicals were of analytical grade.

Preparation of CS nanoparticles
2.2.1 Ionic gelation. CS nanoparticles were prepared via ionic gelation methods established by Calvo et al. with some modifications [26]. CS solutions (0.1%, 0.2%, 0.3%, and 0.4% w/v) were prepared by dissolving CS in 2% v/v glacial acetic acid. Three cross-linking agents (TPP, DS, and PGA) were investigated. TPP solution (0.1% w/v) was prepared by dissolving TPP in deionized water, and DS and PGA solutions were prepared using the same method. CS nanoparticles were prepared by adding 1.2 mL of cross-linker aqueous solution dropwise into 3 mL of CS solutions (0.1%, 0.2%, 0.3%, and 0.4% w/v) at room temperature, with constant magnetic stirring (MS MP8 Wise Stir Wertheim, Germany) at 700 rpm for 30 min. The nanoparticles were later incubated for another 30 min at room temperature before further analysis. Centrifugation (Optima L-100 XP Ultracentrifuge, Beckman-Coulter, CA, USA) was performed at 13,000 x g at 10°C for 30 min to collect nanoparticles. The supernatants were discarded and pellets of nanoparticles were re-suspended in filtered (0.25-μm Millex GP filter unit, Millipore, Billerica, MA) deionized water.
2.2.2 siRNA entrapment. To associate siRNAs with the CS-TPP, CS-DS, and CS-PGA nanoparticles, 3μL of siRNAs (19 μg/μL) was added to 1.2 mL of cross-linker in aqueous solution (0.1% w/v), and this was added to 3 mL of CS solution (0.1%, 0.2%, 0.3%, and 0.4% w/v) under constant magnetic stirring (700 rpm) at room temperature. The particles were then incubated at room temperature for another 30 min before further analysis.

2.3
In Vitro Characterization of CS-TPP/DS/PGA siRNA nanoparticles 2.3.1 Particle size, PDI, zeta potential. The mean particle diameter (z-average), polydispersity (PDI), and zeta potential (surface charge) of freshly prepared and CS nanoparticles were determined by photon correlation spectroscopy (PCS) using ZS-90 Zetasizer (Malvern Instruments, Worcestershire, UK). All measurements were performed after the samples were harvested by centrifugation and re-suspended in deionized distilled water. Each sample was assayed in triplicate at 25°C, and data are reported as mean±standard deviation.
2.3.2 Morphological analysis. Morphological characterization of unloaded and siRNAloaded CS-TPP/DS/PGA nanoparticles was carried out using transmission electron microscopy (TEM)(Tecnai Spirit, FEI, Eindhoven,The Netherlands). A drop of nanoparticles dispersion was placed on the copper microgrid that was natively stained by 3% w/v phosphotungstic acid. The stained nanoparticles was incubated for 5-10 min and evaporated at room temperature (25 ± 2°C). Then, it was viewed under the TEM for imaging of samples.
2.3.3 Entrapment efficiency. The entrapment efficiency of siRNAs (% entrapped) for CS-TPP/DS/PGA nanoparticles was measured using a UV-vis spectrophotometer (Shimadzu UV-1800, Shimadzu Scientific Instruments, Japan) at 260 nm. Briefly, the free siRNA in supernatant recovered from centrifugation (13,000 x g at 10°C for 30 min) was quantified by measuring its absorbance at 260 nm wavelength (the maximum absorption of nitrogenous bases of nucleotides) with a dual beam UV-vis spectrophotometer. Concentration of free siRNA was determined using Beer's Law (A 260 ƐCL) where C is the concentration of siRNA, A 260 is the absorbance at 260 nm, Ɛ is the extinction coefficient and L is the path length of the cuvette.
Extinction coefficient of siRNA is 385101Lmol -1 cm -1 and entrapment efficiency was calculated using the following formula: where C sample is the concentration of siRNA added and C supernatant is the concentration of siRNA in the supernatant. All measurements were performed in triplicate, and data are reported as mean ± standard deviation. 2.3.4 Gel retardation assay. The binding efficiency of siRNA to CS-TPP/DS/PGA nanoparticles was determined using 4% w/v agarose gel electrophoresis and SYBR Green (Invitrogen, Carlsbad, CA, USA) staining. Twenty microliters of sample (prepared at various CS concentrations) containing 0.2 μg of siRNA was loaded into the wells. A 10-bp DNA ladder was used as a size reference. Free siRNA and unloaded CS nanoparticles were used as positive and negative controls, respectively. The siRNA bands were viewed using a real-time UV transilluminator at 480 nm as per manufacturer's protocol (Invitrogen, Carlsbad, CA, USA). Moreover, the quantitative analysis was performed by using imageJ software (version 1.45s) developed by National Institutes of Health (NIH, USA).

Stability studies
2.4.1 Serum protection assay. siRNA-loaded CS-TPP/DS/PGA nanoparticles prepared from 0.1% w/v CS were selected for serum protection assay because of their small particle size, net positive charge, and high entrapment efficiency. A volume of 200 μL of siRNA-loaded CS nanoparticles (containing 5 μg of siRNA) was incubated at 37°C with an equal volume of RPMI supplemented with 10% FBS. Naked siRNA acted as a control and was treated in the same manner. At each time interval (0 min, 30 min, 2 h, 4 h, 24 h, and 48 h), 40 μL of the mixture was removed and stored at -20°C in preparation for gel electrophoresis. Free siRNA was used as an internal control. Before performing gel electrophoresis, the samples were incubated in a bath incubator at 60°C for 3 min to terminate serum activity. A volume of 5 μL of heparin (1000 U/mL) was then added to displace siRNA from the CS-TPP/DS/PGA nanoparticles. The integrity of the siRNA displaced from nanoparticles was analyzed by conducting gel electrophoresis with 4% w/v agarose gel stained with SYBR Green (Invitrogen, Carlsbad, CA, USA). Electrophoresis was performed for 30 min at 110 volts and siRNA bands were visualized under a real-time UV transilluminator at 480 nm (Invitrogen, Carlsbad, CA, USA). Moreover, the quantitative analysis was performed by using imageJ software developed by National Institutes of Health (NIH).
2.4.2 Storage stability of CS nanoparticles. CS concentration of 0.1% w/v was used to prepare CS nanoparticles for this and further experiments. CS-TPP/DS/PGA nanoparticles loaded with siRNA were suspended in deionized distilled water andin PBS (pH 7.4), stored at 4°C for 15 days.The storage stability of nanoparticles was determined by measuring the particle size of CS nanoparticles at pre-determined time points.

In vitro release studies
The release characteristics of siRNA were studied using PBS (pH 7.4). Samples (4 mL) were centrifuged at 13,000 x g for 30 min at 37°C and pellets were re-suspended in 3 mL of PBS. The re-suspended pellets were stirred at 100 rpm by using a magnetic stirrer for 8 days at 37°C. At different time intervals, the samples were centrifuged at 13,000 x g for 30 min at 25°C. The supernatant recovered from centrifugation was used for further analysis, and an equivalent volume of fresh PBS was added to the sample. The amount of siRNA present in the supernatant was analyzed using a UV-vis spectrophotometer (Shimadzu UV-1800, Shimadzu Scientific Instruments, Japan) at 260 nm.

Cytotoxicity studies
DLD-1 cells (ATCC, Manassas, VA, USA) were cultured in RPMI 1640 medium at a cell density of 4 × 10 4 per well. The cells were supplemented with a medium containing 10% FBS and 1% penstrep, and maintained at 37°C in a humidified 5% CO 2 /95% air atmosphere. After 24 h and 48 h of incubation of untreated cells, naked siRNA and siRNA-loaded CS-TPP/DS/PGA nanoparticles at 37°C, a final dilution of 1/10 per cell volume of alamarBlue reagent was added to the treated cells, followed by incubation for 4 h prior to analysis. The absorbance of each sample at 570 nm (A570, the maximum absorption of alamarBlue product (resorufin)) was measured on a microplate reader (Varioskan Flash, Thermo Scientific, Waltham, MA, USA). The assay was performed in the presence of 10% FBS. Cell viability was determined using the following equation: The amount of FBS proteins in the supernatant recovered after ultracentrifugation was determined via Bradford Protein Assay Kit (Sigma Aldrich St. Louis, MO, USA). The FBS protein adsorption efficiency was calculated by the following equation: where C total represents the total amount of protein in medium and C sup is the concentration of protein in supernatant. 2.6.1 LIVE/DEAD cell viability assay. This assay was performed to measure the functional status of the cell by detecting cytoplasmic esterase activity using the LIVE/DEAD Viability/Cytotoxicity kit for mammalian cells (Invitrogen, Carlsbad, CA, USA), which contains calcein, which fluoresces green in living cells, and ethidium bromide, which fluoresces red in dead cells. This assay was performed in 96-well plates. Briefly, DLD-1 cells were plated at a seeding density of 2× 10 4 per well. The cells were supplemented with a medium containing 10% FBS and 1% penstrep, and maintained at 37°C in a humidified 5% CO 2 /95% air atmosphere. The cells were treated with siRNA-loaded CS-TPP/DS/PGA nanoparticles for 24 h and 48 h. The assay was performed in the presence of 10% FBS. Subsequently, the cells were rinsed twice with PBS, fluorochromes (calcein/ethidium bromide) were added, and cells were incubated for 45 min. Finally, reagents were removed and cells were analyzed with a Floid Cell Imaging Station (Molecular Probes Life Technology, France) for calcein and ethidium bromide flourescence.

Cellular internalization
To determine successful siRNA internalization, DLD-1 cells (3×10 4 ) were seeded in a 96-well tissue culture plate and cultured for 24 h (80% confluence). The cells in each well were incubated for 4 h in a medium with 10% FBS with free 6-FAM-siRNA, or with siRNA carried by CS-TPP nanoparticles. After incubation, the cells were washed twice with PBS and then stained with 1μg/mL of Hoechst stain 33342 for 15 min at 37°C. Studies were performed using a Floid Cell Imaging Station (Molecular Probes Life Technology, France).

Statistical analysis
Data are presented as mean ± standard deviation (SD), and were analyzed by 1-way ANOVA with Tukey post-hoc tests using SPSS 21.0. p-values <0.05 indicated statistical significance of differences between groups.
Results and Discussion 3.1 Particle size, PDI, and zeta potential The mean particle size of siRNA-loaded CS-TPP/DS/PGA nanoparticles was increased significantly by increasing the CS concentration from 0.1% to 0.4% w/v, as shown in Table 1. This was expected because of the lower viscosity observed at lower concentrations of CS, which resulted in better solubility and interaction between CS and crosslinkers, and thus smaller particle size [27]. Moreover, significant decrease in particle size was observed after loading of siRNA into CS-TPP/DS/PGA nanoparticles in comparison to unloaded CS-TPP/DS/PGA nanoparticles as shown in (Table 2), which was in accordance with previous findings [28]. Among TPP/DS/PGA, CS-TPP-siRNA nanoparticles had the smallest particle size range as compared to the nanoparticles obtained from DS and PGA, as shown in Table 1.TPP produces small particles because it is a small polyanionic molecule that forms strong ionic interactions with the NH 3 + groups of CS [27].
In contrast, CS-PGA-siRNA nanoparticles had significantly larger particle sizes across CS concentrations, as shown in Table 1, which could be due to the fact that PGA is a macromolecule with fewer ionic interactions with CS NH 3 + groups [25]. CS-DS-siRNA also showed a large particle size which could be due to the aggregation of small particles that tend to fuse together and form groups, as shown in Fig 1[29]. All cross-linker formulations showed PDI values from 0.2 to 1, and values increased as CS concentration increased from 0.1% to 0.4% w/v (Table 1). PDI values of formulations using TPP as the cross-linking agent were within the acceptable range, indicating narrow distribution of particle size, while PGA showed high PDI values, implying broad distribution of particle size. The surface charge of unloaded CS nanoparticles ranged from approximately +30 to +85mV ( Table 2) at CS concentrations from 0.1% to 0.4% w/v. In general, the zeta potential of siRNAloaded CS-TPP/DS/PGA increased as the concentration of CS increased (Table 1), because a greater number of excess free positive charges were available that did not counteract the negatively charged siRNA (the amount of siRNA was fixed) [28]. Moreover, the decrease in zeta potential after siRNA loading demonstrated the succesful association of siRNA with the particles. The presence of siRNA phosphate groups would neutralize CS NH 3 + groups, and therefore lower the positive charges of cationic CS [28,30]. TPP produced nanoparticles that exhibited low zeta potential, and DS and PGA produced nanoparticles with high zeta potential (Table 1). This result could be caused by the strong cross-linking between TPP and the NH 3 + groups of CS, which would result in neutralization of NH 3 + groups and low zeta potential. Higher zeta potential was observed for other cross-linkers because of the lower degree of cross-linking between PGA/DS and CS [30].

Morphology
Images of CS nanoparticles loaded with siRNA obtained by TEM are shown in Fig 1. Fig 1(A) shows an assorted morphology of unloaded CS-TPP nanoparticles, consisting mostly spherical and some irregular particles. Similar morphology was observed when nanoparticles were loaded with siRNA as shown in Fig 1(D). However, unloaded CS-DS/PGA nanoparticles showed aggregation and irregular morphology (Fig 1B and 1C). Similarly, CS-DS-siRNA and CS-PGA-siRNA showed aggregation due to inefficient cross-linking (Fig 1E and 1F). Variation in cross-linking ability was expected to contribute to differences in morphology. TPP showed excellent cross-linking ability and successfully neutralized CSNH 3 + groups, resulting in small and more spherical nanoparticles. In contrast, DS and PGA did not cross-link CS as efficiently as TPP at the concentrations studied, which resulted in irregular morphology. Moreover, aggregation was most likely due to lower degree of cross linking between DS or PGA and CS [30], resulted in excess positive charge from the unneutralized CS NH 3 + groups. It has been reported that under acidic conditions, electrostatic repulsion and interchain hydrogen bonding interactions exist in equilibrium when CS concentration below a certain limit. Above this limit, NH 3 + groups tend to cause surface shielding and result in domination of intermolecular hydrogen bonding during cross linking process. This leads to plenty of CS molecules involved in the cross-linking of a single particle and subsequently large particles are formed. The formation of large particles leads to flocculent precipitate because electrostatic repulsion between particles is not sufficient to maintain the stability of these large particles [27].

siRNA entrapment efficiency
An siRNA entrapment efficiency in the range of 85% to 99% was achieved for CS-TPP/DS/ PGA nanoparticles. In general, the entrapment efficiencies of all formulations decreased when CS concentration was increased, as shown in Fig 2. The entrapment efficiency of siRNA-loaded CS-TPP, CS-DS, and CS-PGA formulations was decreased significantly by increasing CS concentration from 0.1% to 0.4% w/v. Higher CS concentrations produced a more viscous solution that hindered the movement of siRNA around the CS chain and led to inefficient siRNA entrapment [31]. The low entrapment efficiency of CS-DS-siRNA and CS-PGA-siRNA may be explained by shielding effects and steric hindrance, which interfered with the interaction between siRNA and CS NH 3 + groups [32].

Binding efficiency of siRNA with CS nanoparticles
To further investigate siRNA binding to CS-TPP/DS/PGA nanoparticles, agarose gel electrophoresis was performed. For CS-TPP-siRNA, complete binding of siRNA to CS nanoparticles was observed (due to the absence of a trailing band), which suggested a strong interaction Comparative Study of CS Nanoparticles Using Different Cross-Linkers occured between CS and siRNA, as shown in Fig 3 (A). However, CS-DS-siRNA and CS-PGA-siRNA showed trailing bands that could indicate some release of siRNA from nanoparticles as shown in Fig 3(B) and 3(C), respectively. The results also showed that the interaction between CS and siRNA in the presence of DS and PGA was comparatively weaker than in the presence of TPP. The findings were further supported by quantitative analysis of relative density of siRNA bands using imageJ software (Table 3).

Stability studies
3.5.1 Serum protection test. The ability of a carrier to protect its payload from nuclease degradation is an important property for efficient gene delivery. siRNA must be protected fromnuclease digestion for maximal activity in the cells. To address this, serum protection test was carried out for CS nanoparticles in 10% FBS. Fig 4 shows trailing bands for siRNA-loaded CS-TPP/DS/PGA nanoparticles, indicating the migration of siRNA from nanoparticles after treatment with heparin. Thus, it was assumed that some siRNA had been unbound from nanoparticles and interacted with components in FBS. The interaction between unbound siRNA and components in FBS had contributed to the appearance of trailing bands due to slower migration of siRNA. The bands were not as intense as the non-treated free siRNA (control) because some siRNA might remain within the nanoparticles (could not be stained by SYBR Green) or be degraded. Naked siRNA started to degrade as early as 0 min with degradation resulting during the mixing of siRNA with serum and freezing steps [28]and most of thes iRNA was degraded after 48 h of incubation (Fig 4(A)). In comparison to naked siRNA, CS-TPP nanoparticles effectively protected siRNA from nuclease degradation up to 48h as brighter and dense band was observed (Fig 4(B)). In contrast, CS-DS and CS-PGA nanoparticles partially protected siRNAafter 48h as lesser dense band was observed compared to CS-TPP  (Fig 4(C)). In case of CS-PGA nanoparticles, the least dense trailing bands indicate the probability of more siRNA was being degraded and interacted with FBS components. In order to support the findings, the quantitative analysis of bands of naked siRNA, siRNA-loaded into CS-TPP/DS/ PGA nanoparticles was performed by using an ImageJ software (as shown in Table 4). The results showed CS-TPP-siRNA provide a better protection for siRNA than CS-DS and CS-PGA nanoparticles (Fig 4(E)). The relative density of siRNA loaded CS-TPP is approximately two-folds higher than CS-DS and three-folds higher than CS-PGA nanoparticles, after 48 h incubation in serum. Naked siRNA showed the least relative density compared to siRNA carried by CS nanoparticles. Thus, CS-TPP nanoparticles effectively protected siRNA from the enzymatic activity of serum components. However, CS-DS and CS-PGA nanoparticles provided partial protection for siRNA when compared with CS-TPP nanoparticles.  Comparative Study of CS Nanoparticles Using Different Cross-Linkers 3.5.2 Storage stability of siRNA loaded CS nanoparticles. Storage stability of siRNAloaded CS-TPP/DS/PGA nanoparticles in deionized distilled water is shown in Fig 5(A).Deionized distilled water was used in order to determine nanoparticles' stability in the medium in which they were suspended as reported by others [33,34]. In cases where the nanoparticles are stable in their suspending medium, additional step to stabilize the nanoparticles such as lyophilisation could be avoided [35]. The previous findings have indicated that cross-linking reactions improve the properties of particulates [29]. The results of this study support the previous finding that CS-TPP-siRNA nanoparticles were stable, and only slight increase in particle size was observed during the experimental period of 15 days.
These results indicate that the cross-linking effect of TPP confers its properties as a stabilizer, and causes polymeric molecules to establish stronger interactions and more stable structures, which are less prone to aggregation [33]. In contrast, CS-DS-siRNA and CS-PGA-siRNA nanoparticles showed significant increase in particle size during the 15-day experimental period, which might be due to collision and adhesion of nanoparticles during storage, and consequent aggregation [36]. Moreover, CS-PGA-siRNA nanoparticles had the largest particle size after 15 days of storage, which indicated the least stable system. This result is in accordance with the previous finding that large cross-linkers poorly penetrate the CS polymer during nanoparticle formation, which leads to poor ionic interaction between the cross-linker and polymer, and thus to unstable nanoparticles that are prone to form aggregates [36]. Moreover, similar findings were observed for storage stability of siRNA-loaded CS-TPP/DS/PGA nanoparticles suspended in PBS as shown in Fig 5(B).

In vitro release study
The in vitro release profiles of siRNA-loaded CS-TPP/DS/PGA nanoparticles were investigated for 8 days in PBS at pH 7.4, and results are shown in Fig 6. The release of siRNA was divided into 2 stages on the basis of release rate, which was calculated as the slope of the release profile. In the first stage, siRNA-loaded CS-TPP,-DS, and-PGA nanoparticles showed rapid siRNA release in the first 12 h, resulting in 12%, 18%, and 20% of cumulative release, respectively. The Comparative Study of CS Nanoparticles Using Different Cross-Linkers  Comparative Study of CS Nanoparticles Using Different Cross-Linkers release of siRNA at this stage might involve the diffusion of siRNA bound near the particle surface. CS-TPP nanoparticles showed the lowest initial release of siRNA, which may be attributable to the strong cross-linking of CS-TPP-siRNA nanoparticles [37].The initial release of siRNA from CS-DS-siRNA and CS-PGA-siRNA nanoparticles was greater than from CS-TPP-siRNA nanoparticles, which was expected, due to lower binding efficiency and poor control of diffusion-based release caused by the weak cross-linking abilities of DS and PGA [37]. In the second stage, siRNA was released at a sustained constant rate from CS-TPP/DS/PGA nanoparticles for up to 8 days. TPP contributed the least cumulative siRNA release (33%) in the second stage, followed by DS (41%), and PGA (55%). This relationship was probably due to strong ionic interactions between siRNA and CS in the presence of TPP [36]. In contrast, greater release of siRNA was observed from nanoparticles cross-linked with PGA and DS, which produced a lesser degree of cross-linking [36].

Cytotoxicity
To investigate the cytotoxic effect of siRNA-loaded CS-TPP/DS/PGA nanoparticles on DLD-1 cells, an alamarBluecell viability assay was performed in the presence of 10% FBS. Depending on the concentrations and cross-linkers used, all formulations caused a loss of DLD-1 cell viability after 24 and 48h incubation, as shown in Fig 7(A) and 7(B), respectively. Naked siRNA produced an 8% loss of cell viability. Generally, cell viability decreased as CS concentration increased from 0.1% to 0.4% w/v (Fig 7(A) and 7(B)). A 15-25% loss in cell viability was observed for CS-TPP-siRNA nanoparticles in comparison to untreated cells. Cell viability losses of 8-30% and 7-35% were observed for CS-DS-siRNA and CS-PGA-siRNA nanoparticles, respectively, depending on the concentrations used.  The loss of cell viability after 24 h incubation was possibly due to initial interaction of siRNA loaded CS-TPP/DS/PGA with the cells (Fig 7(A)). Higher positive zeta potential of CS-DS-and CS-PGA-siRNA nanoparticles possibly resulted in a stronger interaction with the negatively charged cell membrane, leading to greater loss of cell viability. However, after 48 h incubation, CS-DS-siRNA and CS-PGA-siRNA nanoparticles showed significant increase in cell viability while loss of cell viability was observed for CS-TPP-siRNA nanoparticles (Fig 7(B)).This could be explained by the fact that cell viablity is also influenced by the degree of protein adsorption on the particles' surfaces. No significant difference in protein adsorption was observed between unloaded and loaded CS-TPP/DS/PGA nanoparticles as shown in Fig 8(A) and 8(B) after 24 h incubation. However, increased protein adsorption was observed after 48 h incubation for Comparative Study of CS Nanoparticles Using Different Cross-Linkers unloaded and siRNA-loaded CS-DS and CS-PGA nanoparticles. In comparison to CS-TPP-siRNA, CS-DS-and CS-PGA-siRNA were found to adsorp more proteins on their surface after 48 h incubation in FBS and therefore, protect the cells from direct interaction with the nanoparticles. This phenomenon led to decreased cytotoxicity as previously demonstrated [38,39]. In contrast, CS-TPP-siRNA showed lower protein adsorption and resulted in loss of cell viability after 48 h incubation. However, further investigation is needed to determine this effect on the cells.
3.7.1 LIVE/DEAD cell viability assay. Untreated cells did not produce any loss of cell viability while some loss was observed for the cells treated with siRNA in the presence of 10% FBS, as shown in Fig 9(A) and 9(B). These results are in accordance with cytotoxicity assays that showed increased cell death (red color) after 24 h in cells treated with CS-DS-siRNA and CS-PGA siRNA (Fig 9 (D) and 9 (E), respectively),but recovery from this effect after 48 h incubation (Fig 9(I) and 9(J)). However, CS-TPP-siRNA showed increase cell death (red color) after 48 h in cells (Fig 9(H)) in comparison to 24 h incubation (Fig 9(C)).

siRNA internalization/cellular uptake
This study was performed to determine whether CS-TPP nanoparticles facilitated the delivery of siRNA into DLD-1 cells. Fluorescein-labelled siRNA (6-FAM-siRNA) was used in this study. 6-FAM-siRNA-loaded CS-TPP nanoparticles were incubated with DLD-1 cells for 4h in the presence of 10% FBS. CS-TPP-siRNA nanoparticles infiltrated cells and were primarily distributed in the cytoplasm, which is in agreement with previous observations [40] (Fig 10D-10F). However, naked 6-FAM-siRNA was not detected (Fig 10A-10C). Fluorescence from 6-FAM-siRNA-loaded CS-TPP nanoparticles was detected in the cytoplasm suggesting that siRNA was dissociated from CS-TPP nanoparticles.
Two factors may contribute to the dissociation of the siRNA from the nanoparticles. First, the stability of nanoparticles might be affected by dilution in cytoplasm, resulting in the dissociation of nanoparticles, which subsequently cause the release of siRNA. Secondly, CS can be degraded in the presence of enzymes, resulting in deprotonation and loss of siRNA binding ability, and thus siRNA release into the cytoplasm [40]. Comparative Study of CS Nanoparticles Using Different Cross-Linkers Conclusions CS nanoparticles loaded with siRNA were successfully prepared by ionic gelation method susing three cross-linkers: TPP, DS, and PGA. TPP produced the smallest particle size, with high entrapment and binding efficiencies. Of the cross-linkers studied, TPP created the most stable system, leading to slow burst release of siRNA. Cytotoxic effects were determined to be dependent on concentration of CS used to prepare nanoparticles because it affects particle size and surface charge. Cellular uptake studies confirmed the successful delivery of siRNA into the cytoplasm from siRNA-loaded CS-TPP nanoparticles. The CS-TPP nanoparticles with high binding affinity for siRNA are expected could provide the ideal balance between sufficient protection and efficient intracellular release of siRNA that subsequently could produce remarkable in vivo antitumor effects. Based on the findings, siRNA-loaded CS-TPP nanoparticles demonstrate a great potential for clinical applications in siRNA-based cancer therapies.