Effects of Chitosan Nanoparticles with Long Synthetic siRNAs Targeting VEGF in Triple-Negative Breast Cancer Cells

Abstract Vascular endothelial growth factor (VEGF) is an essential angiogenic factor in breast cancer development and metastasis. Small interfering RNAs (siRNAs) can specifically silence genes via the RNA interference pathway, therefore were investigated as cancer therapeutics. In this study, we investigated the effects of siRNAs longer than 30 base pairs (bp) loaded into chitosan nanoparticles in triple-negative breast cancer cells, compared with conventional siRNAs. 35 bp long synthetic siRNAs inhibited VEGF gene expression by 51.2% and increased apoptosis level by 1.75-fold in MDA-MB-231 cell lines. Furthermore, blank and siRNA-loaded chitosan nanoparticles induced expression of IFN-γ in breast cancer cells. These results suggest that long synthetic siRNAs can be as effective as conventional siRNAs, when introduced into cells with chitosan nanoparticles.


INTRODUCTION
RNA interference (RNAi) is a potent gene silencing method triggered by double-stranded RNAs (dsRNAs).The effectors of RNAi are small interfering RNAs (siRNAs) that are produced from longer dsRNA precursors by Dicer, which is an RNAse III endonuclease.siRNAs are commonly between 21-23 bp in length with 2-base 3'-overhangs and have sequence-specific gene silencing capability.siRNAs are recognized by the RNA-induced silencing complex (RISC) and the antisense strand of siRNA serves as a guide for degradation of target mRNA (Kim, Rossi, 2008;Ozcan et al., 2015).Although siRNAs have the potential to specifically inhibit the expression of any gene, many challenges including rapid degradation by nucleases in biological fluids and the delivery of siRNA to target cells and tissues, limit the clinical use of RNAi systems.Therefore, the development of appropriate carrier systems is very important for siRNA delivery (Leung et al., 2014).
In vivo, double-stranded RNAs (dsRNAs) that are longer than 23-nucleotides (nt), are cleaved by Dicer into short siRNA products (Amarzguioui et al., 2006).Studies of the effects of different 5' and 3' end structures of dsRNAs, showed that slightly longer siRNAs have higher activity than synthetic 21-bp siRNA targeting the same region (Kim et al., 2005).The use of dsRNAs longer than 30 bp in mammalian systems was initially thought to be unsafe due to activation of interferons and the induction of gene inhibition, independent of the dsRNA sequence (Dahlgren, Wahlestedt, Thonberg 2006;Elbashir et al., 2001).However, it has been reported that multi-targeted long synthetic siRNAs (38 bp) simultaneously suppressed the expression of two genes without triggering interferon response.Synthetic dsRNAs of 34 and 38 bp length specifically inhibited the target gene without an antiviral response (Chang et al., 2009).
Breast cancer is the most common malignancy and the second commonest cause of tumour-related death among women in the world.Mortality rates have

Effects of Chitosan Nanoparticles with Long Synthetic siRNAs Targeting VEGF in
Triple-Negative Breast Cancer Cells Page 2/12 Braz.J. Pharm.Sci.2023;59: e22304 Birnur Cömez, Jülide Akbuğa decreased in recent years because of early diagnosis and advanced treatment methods.Metastasis is the main cause of death from breast cancer.Non-metastatic breast cancer is considered curable as the five-year overall survival is greater than 80%, whereas in metastatic disease there is only a 25% reduction in the five-year overall survival rate (Zarychta, Ruszkowska-Ciastek, 2022).
Angiogenesis plays an important role in cancer growth and metastasis, as it provides the formation of blood vessels that will supply oxygen and nutrients to cancer cells.VEGF is one of the major regulators of vascular growth and angiogenesis.Altered expression levels of VEGF is shown to have significant effects on tumour growth, response to therapy, and survival in breast cancer, which makes VEGF a promising therapeutic target (Ayoub et al, 2022).Both monotherapy and combination therapy applications of siRNA targeting VEGF were shown to inhibit tumour growth and angiogenesis (Chen et al., 2017;Dong et al., 2015;Feng et al., 2014).
Chitosan and its derivatives have been extensively investigated for siRNA delivery due to these polysaccharides being biodegradable, biocompatible, and non-toxic.Chitosan is a cationic polysaccharide with positively charged amines that interact electrostatically with negatively charged nucleic acids to form nano-sized structures (Rudzinski, Aminabhavi, 2010).
This study aimed to investigate the effects of chitosan nanoparticles as a carrier of siRNAs longer than 30 bp and to compare with conventional siRNA in breast cancer cells.For this purpose, we determined VEGF expression, apoptosis, and interferon-gamma (IFN-γ) protein levels after delivery of VEGF-targeted siRNA via chitosan nanoparticles into MDA-MB-231 cell lines.

Material
A core sequence consisting of 19-nt was selected against VEGF-A mRNA (Accession No. NM_001025366.2) using BLOCK-iT RNAi Designer (Invitrogen, CA, USA) to design siRNAs.The core sequence was extended to the 3' end of the target gene such that it contains 33-nt.The sequences of 21-bp VEGF-A siRNA were 5'-GCAGCUACUGCCAUCCAAUdTdT-3' (sense) and 5'-AUUGGAUGGCAGUAGCUGCdTdT -3' (antisense).The sequences of 35-bp VEGF-A siRNA were 5'-GCAGCUACU GCCAUCCAAUCGAGACCCUGGUGGdTdT-3'(sense) and 5'-CCACCAGGGUCUCGAUUGGAUGGCAGUAGCUG CdTdT-3' (antisense).5′-FLO siRNA was labelled with fluorescein and used to monitor cellular uptake.All siRNA sequences were analysed by BLAST to verify the absence of significant homology with other human genes.All siRNAs were synthesised by Invitrogen (Carlsbad, CA, USA).
Low molecular weight chitosan (70 kDa, 75-85% degree of deacetylation) and pentasodium tripolyphosphate (TPP) were obtained from Sigma-Aldrich (St. Louis, MO, USA).DMEM, fetal bovine serum (PAN Biotech, Aidenbach, Germany) was used in the cell culture studies.VEGF and IFN-γ ELISA kits were purchased from Invitrogen (Carlsbad, CA, USA).ApoTarget™ Caspase-3 protease assay kit was obtained from Life Technologies (Carlsbad, CA, USA).Other chemicals and reagents were of the analytical and molecular grade.

Preparation of siRNA-loaded chitosan nanoparticles
Chitosan nanoparticles containing siRNA were prepared by ionic gelation using TPP as the polyanionic agent (Calvo et al., 1997).60 µg of siRNA was added to 0.25% (w/v) TPP solution.This mixture was dropped into 0.25% (w/v) chitosan solution in 2% glacial acetic acid while stirring with a magnetic stirrer.After the dropping process ended, it was stirred for 30 minutes at room temperature.The nanoparticles that formed were washed with double distilled water three times and separated by centrifugation at 12.000 x g, then lyophilised (Leybold-Lyovac, Germany).

Characterisation of nanoparticles
The encapsulation efficiency of the siRNAs was determined by measuring the supernatants collected after centrifugation (12.000 x g, 15 min) at 260 nm with a UV-vis spectrophotometer (Shimadzu Biospec 1601, Shimadzu Scientific Instruments, Japan).siRNA concentrations were calculated from the supernatants using the standard curves of 21-bp and 35-bp siRNAs.The encapsulation efficiency of both siRNAs was calculated by the equation (Katas, Alpar, 2006): Concentration of siRNA added -Concentration of siRNA in supernatant Concentration of siRNA added ×100 Morphological characterisation of siRNA-loaded chitosan nanoparticles was performed visually by transmission electron microscopy (TEM, JEM 1011; Jeol, Tokyo, Japan).The size and zeta potential of nanoparticles were measured by the Zetasizer Nano ZS (Malvern Instruments, UK).The samples were dispersed in double distilled water, and each measurement was performed in triplicate and at 25°C.

Gel retardation assay
The binding of siRNAs to chitosan nanoparticles was determined by 2% w/v agarose gel electrophoresis.20 μl of siRNA-loaded chitosan nanoparticle suspension was loaded into agarose gel with a 5:1 dilution of loading dye.Electrophoresis was performed at 80mA for 30 minutes.Free siRNA was used as a control and siRNA bands were visualised by a UV transilluminator (Vilber Lourmat, France).

Serum stability of the nanoparticles
siRNA-loaded chitosan nanoparticles were suspended in 100 μl PBS (pH 7.4) buffer with 10% fetal bovine serum and incubated at 37°C.During the incubation, aliquots were taken at determined time intervals (0, 1, 4, 6, 24, and 48 h) and stored at -20°C until agarose gel electrophoresis analysis.The samples were incubated in a water bath at 80°C for 5 minutes. 1 μl heparin (5000 IU/ml) was added to each aliquot to displace the siRNA from the chitosan nanoparticles.The integrity of the siRNA was analysed by 2% agarose gel electrophoresis.Electrophoresis was carried out at a constant voltage of 200V for 30 min in 0.5x TBE buffer.siRNA bands were viewed under a UV transilluminator.

In vitro release study
Nanoparticles containing 10 μg siRNA were suspended in 1 ml of RNAse-free PBS (pH 7.4) and incubated in a shaking water bath at 37 °C.The samples were centrifuged at 14.000 x g for 30 min at different time intervals.The supernatants were removed for analysis and replaced with an equivalent volume of fresh PBS.The concentration of siRNA released in the supernatant was measured by a UV-vis spectrophotometer (Shimadzu UV1800, Shimadzu Scientific Instruments, Japan) at 260 nm (Raja, Katas, Wen, 2015).Each sample was studied in triplicate.
The release profiles of each nanoparticle were determined by plotting the percentage of cumulative siRNA release at different times.The release profiles of both nanoparticles were compared with the difference factor (f 1 ) and the similarity factor (f 2 ) by equations ( 1) and (2) (Cano-Vega, Deng, Pinal, 2021): (1) (2) (n = number of timepoints; R t = the released value of the reference at time t; T t = the released value of the test at time t) A model-dependent approach was used to determine the release kinetics of nanoparticles.The release data was applied to the mathematical equations of zero-order, firstorder, Higuchi, Hixon-Crowell, and Korsmeyer-Peppas models (Wojcik-Pastuszka et al., 2019).The most suitable model for the release profile of nanoparticles was selected according to the determination coefficients (R 2 ).

In vitro transfection studies
Cells were seeded into 6-well plates at a density of 2 × 10 6 cells per well and incubated overnight.Transfections were performed on cells that were approximately 70% confluent.Before transfection, the medium was removed, and cells were rinsed with PBS (pH 7.4).Nanoparticle formulations and both free siRNAs (2µg of siRNA per well) in serum-free medium, were added to the cells.After the cells were incubated for 6 hours, the serumcontaining medium was supplemented to the wells.After 48 hours, the culture medium was collected for VEGF ELISA assay.

In vitro cellular uptake studies
Cells (2×10 6 ) were seeded in a 6-well plate and incubated until 70% confluence.Then, transfection was performed with the nanoparticles containing fluorescein (FLO)-labelled siRNAs as mentioned before.Cellular uptake of nanoparticles was observed using a fluorescence microscope (Olympus CKX41, Japan) at different time intervals.

Determination of VEGF levels
VEGF protein that accumulated in the culture medium and recombinant human VEGF protein standards were analysed by sandwich ELISA.ELISA was performed for the cell culture medium according to the manufacturer's instructions (Invitrogen, CA, USA).Absorbances were measured spectrophotometrically at 450 nm.The percentage inhibition of VEGF expression was calculated in siRNA-transfected samples compared with untreated cells.Each test was repeated in triplicate, and standard deviations (SD) were calculated.

Apoptosis assay
The apoptosis study was performed with Caspase 3 colorimetric protease assays according to the manufacturer's protocol.Upon cleavage of the substrate by caspases, the absorbance of free pNA was measured by a microplate reader (Epoch, BioTek Instruments, USA) at 400 nm.
Caspase activity was determined by comparing of the absorbance of apoptotic samples with an uninduced control.

Determination of IFN-γ levels
IFN-γ expression in cell culture medium was determined to observe whether siRNA-loaded nanoparticles induce IFN-γ in vitro, using ELISA assay according to the manufacturer's instructions.The absorbances of the standards and samples were measured spectrophotometrically at 450 nm.IFN-γ protein concentrations were determined according to the standard curve prepared from the standard dilutions.All measurements were performed in triplicate.

Statistical analyses
The results were expressed as the mean ± SD.The treatment groups were analysed using one-way ANOVA with Tukey's multiple comparisons test.All analyses were performed using GraphPad Prism 8.0.1 (GraphPad Software, San Diego, CA).P<0.05 was statistically significant.

Characterisation of nanoparticles
The weight ratio of chitosan to TPP for the preparation of chitosan nanoparticles was selected to be 1:1 at which the best efficiency was observed.The results regarding the characterisation of nanoparticles are given in Table I.In our study, siRNA-loaded chitosan nanoparticles were positively charged and had a size between 268.2-311.8nm.The success of tumour-targeting drug delivery systems is based on the nanoparticles' ability to accumulate in the desired tissues or cells.The enhanced permeability and retention (EPR) effect allows molecules of certain sizes (usually 100-1000 nm) to accumulate in tumour tissues rather than in normal tissues (Xin et al., 2017).It has been reported that 100-300 nm sized particles proposed for therapeutic use, can pass through the endothelial barriers given that the pore cut-off size in many tumours was between 380 and 780 nm (Hobbs et al., 1998).
The morphology of the nanoparticles was observed with TEM.The images showed that some nanoparticles fell one by one, while others began to form aggregates (Figure 1).TEM results showed that the nanoparticles were irregularly shaped.Due to the shielding effect of NH groups of chitosan and the predominance of intramolecular hydrogen bonds during crosslinking, more chitosan molecules are involved in crosslinking, leading to the formation of larger particles and aggregates like flocculant structures (Fan et al., 2012).

Binding efficiency of siRNA to chitosan nanoparticles
21-bp and 35-bp siRNAs efficiently associated with chitosan nanoparticles were observed by agarose gel electrophoresis.Both siRNAs bound with chitosan were detained within the well (Figure 2).We also found that siRNAs were completely bound to chitosan nanoparticles due to the absence of a trailing band.We determined that the zeta potential values of all prepared nanoparticle formulations were positive (+) and close to neutral.Although a low-charge density contributes to low cytotoxicity and might facilitate the intracellular release of siRNA from the nanoparticles, this could also lead to particle aggregation and low stability (Malmo et al., 2012).

Serum stability of siRNA-loaded nanoparticles
siRNA must be protected against nucleases to ensure high gene silencing efficiency in cells (Braasch et al., 2003).Therefore, the serum stability of chitosan/siRNA nanoparticles was examined in 10% FBS to determine whether the prepared formulations protect siRNAs from nuclease degradation (Figure 3).Degradation of the free siRNAs was observed after 1 hour.The 35-bp siRNA was fully degraded after 48 hours, but the complete degradation of 21-bp siRNA did not occur even after 48 hours (Figure 3A, 3B).The bands for chitosan/siRNA nanoparticles showed the migration of siRNA out from the nanoparticles after treatment with heparin (Figure 3C, 3D).siRNAs recovered from the nanoparticles appear to be protected from degradation for up to 48 hours.When free 21bp and 35-bp siRNA were incubated with 10% FBS, changes in their structure occurred and degradation increased over time.Consequently, chitosan nanoparticles significantly protected both siRNAs from nuclease activity compared to free siRNAs.
According to the R 2 values, the zero-order kinetic model fits the release profile of both nanoparticles (Table II).In the zero-order kinetic model, the R 2 values for chitosan/ 21-bp siRNA and chitosan/ 35-bp siRNA nanoparticles were 0.9061 and 0.9578, respectively.Zero-order kinetics describes constant drug release from the drug delivery system independent of the concentration (Bhasarkar, Bal, 2019).The difference factor (f 1 ) and the similarity factor (f 2 ) calculated to compare the release profiles of both nanoparticles were found to be 10.28 and 59.64, respectively.The two release profiles are similar when the f 1 value is in the range of 0-15 and the f 2 value is in the range of 50-100 (Cano-Vega, Deng, Pinal, 2021).

Release properties of siRNA-loaded chitosan nanoparticles
The 21-bp and 35-bp siRNA released completely from chitosan nanoparticles in approximately 30 days (Figure 4).21-bp siRNA and 35-bp siRNA were released from chitosan nanoparticles within the first 24 hours, resulting in 15% and 17% of cumulative release, respectively.Chitosan nanoparticles were found to have a burst effect in the release profile.The substance confined to the surface layer of the particle is dispersed immediately upon contact with the release medium and leads to a burst effect (Agnihotri, Mallikarjuna, Aminabhavi, 2004).
Comparison of the release profiles of chitosan nanoparticles with both 21-bp and 35-bp siRNAs revealed a similar release profile similar during 24 days.The release rate of siRNAs from the nanoparticles was rapid in the first step followed by a slower release due to the swelling of the polymer, with siRNAs continuously released at a constant rate but then released at an increased rate once again.It is thought that erosion and disintegration of nanoparticles might cause this rapid increase.Degradation of the polymer might cause subsequent physical erosion because of bond breaking, which shows that the degradation process correlates with the erosion of the polymer (Mohammed et al., 2017).

Effects of nanoparticles on apoptosis
Dysregulated cell proliferation and suppression of apoptosis cause tumorigenesis and tumour growth.Anticancer therapeutics are thought to cause the death of tumour cells by inducing apoptosis (Peng et al., 2013).Ge et al. (2009) reported that the inhibition of VEGF with siRNA increased apoptosis in MCF-7 breast cancer cell lines.The caspase protease activity assay was used to analyse apoptosis (Figure 7).Chitosan/21bp and chitosan/35-bp siRNA nanoparticles increased caspase activity by 2-fold and 1.75-fold compared with the control group, respectively.Both siRNA-loaded chitosan nanoparticles significantly increased caspase activity compared to the control group in MDA-MB-231 cells.The difference between chitosan/21-bp and chitosan/35bp siRNA nanoparticles for the increase in caspase activity was not statistically significant (p= 0.4652).

Effects of nanoparticles on IFN-γ expression
siRNAs lead to activate the mammalian innate immune system.Synthetic siRNAs can promote high level expression of inflammatory cytokines and interferons (Judge, MacLachlan, 2008).Interferons are cytokines that are responsible for activating the immune system and have antiviral and antitumor effects (Castro et al, 2018).Also, treatment of IFN-γ has been shown to suppress the growth of breast and ovarian cancer cells, both in vitro and in vivo and induce apoptosis (Treeck et al., 2005).Production of IFN-γ is mostly seen in T lymphocytes and NK cells.The expression of IFN-γ mRNA in other cell types is very low (Castro et al, 2018).However, Xue, Firestone, and Bjeldanes (2005) were the group to report that IFN-γ gene transcription could be induced in human breast cancer cell lines such as MCF-7 and MDA-MB-231.We examined whether chitosan/21-bp and chitosan/35-bp siRNA nanoparticles induce IFN-γ expression in MDA-MB-231.After transfection, IFN-γ levels were determined by ELISA.IFN-γ expression was not observed in the control groups, but only induced in the cells treated with chitosan nanoparticles (Figure 8).As the free 21-bp and 35-bp siRNA did not generate IFN-γ responses, they are not shown in Figure 8. siRNAs induced cytokine activation in a dose-dependent and sequence-independent manner by applying siRNAs with transfection agents, not only siRNA (Lee et al., 2012).All chitosan nanoparticles caused a significant increase in IFN-γ protein levels compared to the control group (p<0.01).The blank chitosan nanoparticles also induced IFN-γ expressions in vitro, suggesting that this effect is related to chitosan.Chitosan and its derivatives are widely investigated as adjuvant and vaccine delivery systems due to their immunostimulatory effects (Vasiliev, 2014).It was reported that chitosan induced expression of IFN-γ at low levels in lung epithelial cells when administered alone and intranasally to mice (Kumar et al., 2003).Chitosan/35-bp siRNA nanoparticles induced IFN-γ at a greater level than chitosan/21-bp siRNA nanoparticles (p<0.01).

CONCLUSION
In summary, we demonstrated that chitosan nanoparticles prepared with long siRNAs consisting of 35 bp inhibited VEGF gene expression and induced apoptosis in MDA-MB-231 cells which was also seen with 21-bp siRNAs.We also observed that the characteristics and release profiles of chitosan nanoparticles prepared with both siRNAs were similar.The release kinetics of siRNAs from chitosan nanoparticles were found to best fit the zero-order kinetic model for both nanoparticles.Moreover, we observed that chitosan nanoparticles stimulated IFN-γ expression in vitro, when used as a delivery system.This feature of chitosan nanoparticles is thought to positively contribute to siRNA-related anticancer effect through VEGF gene inhibition.

FIGURE 1 -
FIGURE 1 -Transmission electron microscopy images of chitosan nanoparticles at 75kx magnification.

TABLE II -
The kinetic parameters of siRNA release from the chitosan nanoparticles