Selective suppression of influenza A/H5N1 virus replication in vitro using nanocomplexes consisting of siRNA and aminopropylsilanol nanoparticles

Full Text

Introduction

Nucleic acid (NA) therapy offers unique opportunities to influence the genetic material of the cell. However, its effectiveness is limited by the instability of NAs in relation to cellular nucleases and their low ability to penetrate the cytoplasmic membrane, which makes it necessary to use different delivery systems [1].

Small interfering RNA (siRNA) is a promising type of RNA-based therapeutic agents because their mechanism of action is catalytic and each siRNA molecule can inactivate several target RNA molecules. siRNA molecules are intensively investigated as antiviral agents. L. Singh et al. presented a wide range of applications of nanoscale materials for the treatment of common viral infections [2]. For the clinical success of the proposed methods for the delivery of NA fragments into cells, safety and efficiency remain vital requirements. Various approaches have been proposed to address the problem of siRNA delivery, e.g., using viruses, cationic lipids, polymers and transport peptides. The successful use of bioconjugates of siRNA and N-acetylgalactosamines is also worth mentioning [3]. However, all methods have limitations for therapeutic use. A huge number of potential siRNA-based drugs have not undergone clinical trials because many factors (low efficiency of siRNA delivery to target cells, toxicity, degradation of siRNA by nucleases, filtration by kidneys, uptake by immune cells, off-target effects, low efficiency of penetration through hydrophobic cell membrane and release of siRNA from endosomes, etc.) limit the use of siRNA in biomedicine.

One of the most promising approaches to solving the problem of siRNA delivery into cells is the use of non-viral vectors based on nanoparticles (NPs) [2, 4, 5]. Different types of NPs have been used for delivery of siRNA. NPs consisting of cationic polymers (poly-L-lysine, polyamidoamine, polyethylenimine, and chitosan) or lipids are the most studied delivery methods [6, 7]. Given the wide variety of available materials, each with many potential modifications, the composition of the NPs can be optimized to deliver a specific type of RNA [8-10]. Delivery systems should fulfill a number of important requirements: they should increase the ability of RNA penetration into cells, provide effective protection of RNA from degradation by cellular nucleases, and have low toxicity.

Despite certain successes in the development of methods for the delivery of RNA fragments into cells, the problem of their delivery method still remains unsolved. Therefore, it is advisable to search for other methods to deliver siRNA into cells.

We have previously developed delivery systems for oligodeoxyribonucleotides and deoxyribozymes based on the use of titanium dioxide and aminopropylsilanol (APS) NPs for their effect on NA targets. It has been shown that DNA fragments within the created nanocomposites are site-specific and effectively affect target genes in vitro and in vivo [11–14].

The aim of this study is to determine the possibility of using non-agglomerated, water-soluble APS NPs for siRNA delivery into cells in the form of the Si~NH₂/siRNA nanocomplexes to effectively suppress replication of the influenza A/H5N1 virus.

Materials and methods

Reagents used in this study were from commercial suppliers: (3-aminopropyl)triethoxysilane, trypsin, penicillin, streptomycin (Sigma-Aldrich); DMEM medium (Dulbecco's modified Eagle's medium; Biolot); fetal bovine serum (Gibco), MTT ((3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide), NeoFroxx), and dimethyl sulfoxide (Component-Reactiv). Trypsin was used at a concentration of 2 μg/mL, penicillin and streptomycin at a concentration of 100 U/mL. Chicken erythrocytes, MDCK cells, and the influenza virus A/chicken/Kurgan/05/2005 (H5N1) strain were obtained from the collections of Vector.

Oligoribonucleotides and their derivatives were synthesized by the solid-phase method on an automated ASM-800 DNA/RNA synthesizer (Bioset) using an optimized protocol for a synthesis scale of 0.4 mmol. 2'-deoxy-, 2'-O-TBDMS-, and 2'-O-methyl-phosphoramidite (Glen Research) were used as monomers. Sulfurizing reagent II (Glen Research) was used to introduce the thiophosphate group. The concentration of oligonucleotides was determined spectrophotometrically by measuring their optical density in solution using a Shimadzu U-1800 spectrophotometer (Shimadzu).

Preparation of Si~NH₂ aminopropylsilanol nanoparticles and Si~NH₂/siRNA nanocomplexes

APS nanoparticles (Si~NH₂) were synthesized by hydrolysis of (3-aminopropyl)triethoxysilane, which was added dropwise to hot water, and the mixture was stirred at this temperature for 15 h followed by cooling to room temperature [15]. The pH value of the resulting solution (10.6) was adjusted to 7.5 using 1 M HCl. The concentration of the final Si~NH₂ solution (0.26 M) was evaluated by titration of amino groups using 1 M HCl. The reaction yield was 95–97%. Si~NH₂ NPs were studied by physicochemical methods, i.e., dynamic light scattering, ultraviolet, infrared spectroscopy, transmission and atomic force microscopy [15].

RNA molecules were immobilized on APS NPs [16] due to the electrostatic interaction between negatively charged internucleotide phosphate groups in oligoribonucleotides (p) and positively charged amino groups (NH₂) in NPs. The nanocomplexes with single- and double-stranded RNA (Si~NH₂/RNA and Si~NH₂/siRNA, respectively) were prepared by mixing RNA or siRNA with 0.26 M Si~NH₂ in water, provided that the NH₂/p ratio was 50 (we considered the number of phosphate groups in only one strand). The size and zeta potential of the obtained APS nanoparticles and nanocomplexes with RNA molecules were measured by dynamic light scattering on a Zetasizer Nano ZS Plus device (Malvern).

Toxicity analysis of nanoparticles and nanocomplexes in MDCK cell culture

Samples in DMEM medium (Biolot; 0.1 mL at concentrations of 5–50 μM for siRNA or 5–50 mM for Si~NH₂) were added to wells of 96-well plates with MDCK cells. Cells in 0.1 mL of DMEM maintenance medium were used as a control. After incubation of cells for 2 days at 37ºC and 5% CO₂, the culture medium was removed and MTT dye (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide) in PBS-D buffer (Dulbecco's phosphate-buffered saline; 0.075 mL, 1 mg/mL) was added to each well. Cells were incubated for 90 min at 37ºC, after which the dye solution was removed and dimethyl sulfoxide (0.1 mL) was added. After incubation for 10 min, the optical density in each well was measured on an Emax spectrophotometer (Molecular Devices) at a wavelength of 540 nm, which is an indicator of the number of viable cells in the monolayer.

The dependence of optical density on the concentration of the tested sample was presented in semi-logarithmic coordinates, and the 50% cytotoxic concentration (CC₅₀) of each sample was calculated using the SoftMaxPro-4.0 computer program.

Antiviral activity of nanocomplexes

A/chicken/Kurgan/05/2005 (H5N1) virus was grown in the allantois cavity of 10-day-old chicken embryos at 37ºC. Allantois fluid was collected 48 h after virus inoculation and stored at –80ºC. MDCK cells were seeded at a rate of 10⁵ cells/mL in DMEM nutrient medium containing 10% fetal bovine serum (Gibco) into 96-well plates (100 μL/well) and incubated at 37ºC, 5% CO₂, and 100% humidity. After reaching ~80% monolayer, the medium was removed and samples of Si~NH₂, Si~NH₂/RNA, and Si~NH₂/siRNA were added to the wells at a concentration of 0.5 mM (for Si~NH₂, corresponding to 0.5 μM for RNA or siRNA) in 100 μL of DMEM medium. The control sample was the same medium without the nanocomplexes. Oseltamivir at a concentration of 10 μg/mL was used as a reference drug.

The concentration of the Si~NH₂/siRNA nanocomplexes was varied in the range of 0.01–1.00 μM (for siRNA) in experiments that involved the investigation of their dose-dependent antiviral activity.

Cells were incubated in the presence of the samples at 37ºC, 5% CO_2, and 100% humidity for 4 h followed by washing the cells with the same medium. Cells were then infected with the A/H5N1 virus in trypsin-containing (2 μg/mL) DMEM medium (100 μL in each well) at an infection multiplicity of 0.01 50% tissue culture infectious dose (TCID₅₀) per 1 mL. After virus adsorption for 1 h at room temperature, the medium containing virus was removed, cells were washed with trypsin-free DMEM medium, and the same medium containing trypsin was added to each well (100 μL). After incubation for 48 h, serial 10-fold dilutions (10^–1 to 10^–8) of the culture fluid containing virus from each well were applied to MDCK cells with repeated incubation for 48 h for further evaluation of the virus titer. The presence of cytopathic action was recorded under the microscope and in hemagglutination reaction with 1% suspension of chicken erythrocytes. The virus titer was expressed in terms of lg TCID₅₀/mL. To evaluate the dependence of virus inhibition on the concentration of nanocomplexes, the percentage of inhibition of virus production was calculated using the formula: (A – B)/A, where A is the virus titer in control (without sample) in TCID₅₀/mL; B is the virus titer in experiment (with sample) in TCID₅₀/mL.

Statistical analysis

Statistical analysis was performed using the Statistica v. 12 program (StatSoft Inc.). Virus titer in control and experiment (without or with experimental samples, respectively) was calculated using the Spearman–Kerber method and expressed as lg TCID₅₀/mL. Differences between the results with experimental and control samples were considered significant at p ≤ 0.05.

Results and discussion

Silica NPs are considered as promising carriers for delivery of NAs into cells [17]. Most often, amine-modified Si-NPs are used for immobilization of NAs and their fragments. We synthesized non-agglomerated APS NPs (hydrodynamic diameter — ~1 nm, zeta potential — ~10 mV) [15].

The small size of Si~NH₂ particles provide water-soluble preparations. Characterization of APS NPs using physicochemical methods is described in our previous work [15]. It is shown that the obtained NPs are not prone to agglomeration and can be stored for several months.

Si~NH₂/RNA and Si~NH₂/siRNA nanocomplexes were obtained by electrostatic interaction between negatively charged inter-nucleotide phosphate groups in RNA and siRNA and positively charged protonated amino groups in Si~NH₂ NPs. The addition of negatively charged siRNA molecules to the NPs leads to a change in zeta potential from ~(+10 mV) to ~(–30 mV) and particle size from ~1 nm to ~200 nm, thus indicating the formation of nanocomplexes.

As a target for RNA and siRNA, we chose segment 5 of the influenza A virus encoding a nucleoprotein that plays a key role in the incorporation of the viral genome into the cell nucleus of an infected organism, thus facilitating further replication and assembly of viral particles [18]. A region near the 3' end of this segment starting at nucleotide 1496, which is conserved vulnerable to siRNA action [19–21], was selected for siRNA exposure.

We synthesized two native (RNA₁ and RNA₂) and eight modified (RNA₃–RNA₁₀) oligoribonucleotides containing TT at the 3' end (Table 1). RNA₅ and RNA₆ contain, in addition to TT, 2'-OMe groups at the U*A site in the center of the strands. The presence of the 2'-O-Me group in this site increases resistance to serum nucleases, thereby maintaining the interference ability of siRNA [22] and providing long-term suppression of target gene expression [23]. The RNA₇ strand contains 2'-O-Me groups at all positions and 2 internucleotide thiophosphate groups at both ends. The RNA₈ strand differs from RNA₇ by the presence of three consecutive 2'-F groups at positions 9–11 from the 5' end. Oligoribonucleotides RNA₂, RNA₄, RNA₆, and RNA₈ are sense strands, and RNA₁, RNA₃, RNA₅, and RNA₇ are antisense strands directed to (–)RNA and (+)RNA of the viral genome, respectively.

 

Table 1. Oligoribonucleotides used in this study

RNA chain		Nucleotide sequence, 5'-3'
RNA1	Antisense	CUCCGAAGAAAUAAGAUCC
RNA2	Sense	GGAUCUUAUUUCUUCGGAG
RNA3	Antisense	CUCCGAAGAAAUAAGAUCCTT
RNA4	Sense	GGAUCUUAUUUCUUCGGAGTT
RNA5	Antisense	CUCCGAAGAAAU*AAGAUCCTT
RNA6	Sense	GGAUCUU*AUUUCUUCGGAGTT
RNA7	Antisense	C*PSU*PSC*C*G*A*A*G*A*A*A*U*A*A*G*A*U*C*C*PSTPST
RNA8	Sense	G*PSG*PSA*U*U*U#A*U#U#U#C*U*U*C*G*G*A*G*PSTPST

Note. *2'-O-methyl group; ^#2'-fluoro group; PS — internucleotide thiophosphate group; symbol d for deoxyribotimidine has been omitted.

 

The synthesized RNAs were used to obtain the Si~NH₂/RNA nanocomplexes based on non-agglomerated APS NPs. Their biological activity was investigated on the example of suppression of the A/H5N1 virus replication in infected MDCK cells at a multiplicity of infection of 0.01 TCID₅₀/cell. The results show that all single-stranded oligoribonucleotides (both sense and antisense strands) in the Si~NH₂/RNA nanocomplexes suppressed replication of the influenza A/H5N1 virus by 0.7–1.3 orders of magnitude (5–20 times) (Fig. 1, columns 1–8).

 

Fig. 1. Titers of A/chicken/Kurgan/05/2005 (H5N1) virus in MDCK cells after their incubation of with Si~NH₂/RNA and Si~NH₂/siRNA nanocomplexes.

C, virus control without samples. Si~NH₂/RNA nanocomplexes containing RNA₁–RNA₈ (1–8), concentration of RNA in nanocomplexes is 0.5 µM; Si~NH₂/siRNA nanocomplexes containing siRNA_1/4, siRNA_3/4, siRNA_5/4 and siRNA_7/4 (9–12), concentration of siRNA in nanocomplexes is 0.5 µM per one strand; ozeltamivir (13), 10 µg/mL or 32 µM; Si~NH₂ (14), 0.5 mM; Si~NH₂/siRNA_9/10, nanocomplex containing unspecific siRNA_9/10 (15). MOI, 0.01 TCID₅₀/cell. The presented average values, standard deviations and differences in the titer values of the virus are calculated using the Spearman–Kerber method. Asterisks designate the difference between the control and the titer values of the virus obtained under the action of the studied series of nanocomplexes, at p < 0.05.

 

From the four studied sense strands, we chose RNA₄ containing deoxydinucleotide TT at the 3' end for protection from exonucleases, as the most active one and formed four siRNA molecules with all possible antisense strands (RNA_1/4, RNA_3/4, RNA_5/4, and RNA_7/4). All siRNAs are duplexes with the same nucleotide sequence and different modification of nucleoside units. The efficiency of the investigated siRNAs was significantly higher compared with single-stranded oligoribonucleotides (Fig. 1). The Si~NH₂/siRNA nanocomplexes suppressed the A/H5N1 virus replication by 2.3–2.8 orders of magnitude (200–630 times).

The main reason for the difference in the effectiveness of siRNA and oligoribonucleotides is the different mechanism of their action. It is known that oligodeoxynucleotides and siRNAs suppress the functions of target RNAs through complementary interactions with the target RNA with its subsequent degradation by cellular RNases (RNase-H1 and AGO2, respectively) [24–26], which eventually leads to the loss of target RNA functions. Single-stranded RNA fragments can form complementary complexes with the target RNA but do not cause degradation of the target RNA. The siRNA molecule first binds to the RISC complex, then the sense (passenger) strand is removed. AGO2 nuclease and the remaining antisense strand within the RISC complex find the target RNA, and AGO2 cleaves it. AGO2 retains the antisense strand for some time as part of the RISC complex for further reactions [27–29].

It is known that siRNAs are rapidly hydrolyzed by cellular nucleases, and various modifications are used to protect against them. It should be noted that even minimally modified siRNA_1/4 and siRNA_3/4 delivered into the cells as part of nanocomplexes with NPs were very effective in suppressing virus replication (by 2.3 orders of magnitude, ~200 times). This implies that APS NPs protect siRNA from cellular nucleases. The most active Si~NH₂/siRNA_5/4 nanocomplex suppressed virus replication by ~3 orders of magnitude.

The antiviral activity of the studied Si~NH₂/siRNA nanocomplexes was comparable to that of oseltamivir (the most commonly used comparison drug in studies of effects on the influenza A virus) but at a much lower concentration of the active component (0.5 μM for siRNA and 32 μM for oseltamivir). The Si~NH₂ NPs did not inhibit virus replication, as would be expected.

The most active Si~NH₂/siRNA_5/4 nanocomplex is characterized in more detail. We evaluated the effect of siRNA, Si~NH₂ NPs in the free state and as part of the Si~NH₂/siRNA_5/4 nanocomplex on the survival of uninfected MDCK cells (Fig. 2). Unbound siRNA_5/4, as expected, was nontoxic in the concentration range investigated. The cytotoxicity of the Si~NH₂/siRNA_5/4 nanocomplex coincides with the toxicity of Si~NH₂ NPs. Consequently, it can be concluded that the cytotoxicity of nanocomplexes is determined by the toxicity of their constituent NPs. The cytotoxic concentration of the drug, at which 50% of cells in the uninfected monolayer are destroyed, determined from the data in Fig. 2, was 38 mM per Si~NH₂ and 38 μM per siRNA.

 

Fig. 2. Viability of MDCK cells when treated with siRNA and Si~NH₂ samples in the free state and as part of the Si~NH₂/siRNA nanocomplex.

1, siRNA_5/4; 2, Si~NH₂; 3, Si~NH₂/siRNA_5/4. The x-axis shows the concentration of siRNA in the free state or as part of a nanocomplex (µM) and Si~NH₂ nanoparticles in the free state or as part of a nanocomplex (mM). The y-axis shows optical density of the MTT solution.

 

Table 2 summarizes the results of suppression of the influenza virus production depending on the concentration of the Si~NH₂/siRNA_5/4 nanocomplex in the cell culture.

 

Table 2. Dependence of antiviral activity of Si~NH₂/siRNA_5/4 nanocomplex on siRNA concentration with MOI 0.01 TCID₅₀/cell

Concentration of siRNA5/4 in nanocomplex, µM	Infection titer of influenza A virus lgTCID50/mL	Inhibition of influenza A virus replication, %
1,00	4,50	99,99
0,50	5,75	99,82
0,10	7,00	97
0,05	7,50	90
0,01	7,50	90
Virus control	8,50	–

 

It was shown that in the concentration range of siRNA_5/4 in the Si~NH₂/siRNA_5/4 nanocomplex from 0.01 to 1.00 μM, the suppression of the influenza virus production was 90.00–99.99%.

Conclusion

The results indicate that APS NPs can be used to deliver siRNA into cells as part of the Si~NH₂/siRNA nanocomplexes. The cytotoxicity of the Si~NH₂/siRNA_5/4 nanocomplex is determined by the toxicity of Si~NH₂ NPs. The cytotoxic concentration of the drug, at which 50% of cells in the uninfected monolayer are destroyed, was 38 mM per Si~NH₂ and 38 μM per siRNA. The proposed siRNA-containing nanocomplexes targeting a selected region of the 5th segment of the viral genome mRNA were successfully used to suppress the A/H5N1 virus production in a cellular system. The most effective Si~NH₂/siRNA_5/4 nanocomplex reduced the replication of the influenza A virus in cell culture by ~3 orders of magnitude.

About the authors

Marina N. Repkova

Institute of Chemical Biology and Fundamental Medicine

Email: zarytova@niboch.ncs.ru
ORCID iD: 0000-0002-7108-9036

Cand. Sci. (Chem.), researcher, Laboratory of nucleic acids

Россия, Novosibirsk

Asya S. Levina

Institute of Chemical Biology and Fundamental Medicine

Email: zarytova@niboch.ncs.ru
ORCID iD: 0000-0003-2423-3805

Cand. Sci. (Chem.), senior researcher, Laboratory of nucleic acids

Россия, Novosibirsk

Oleg Yu. Mazurkov

State Research Center of Virology and Biotechnology “Vector”

Email: zarytova@niboch.ncs.ru
ORCID iD: 0000-0001-8164-4091

Cand. Sci. (Biol.), researcher, Department of prevention and treatment of especially dangerous infections

Россия, Koltsovo

Elena V. Makarevich

State Research Center of Virology and Biotechnology “Vector”

Email: zarytova@niboch.ncs.ru
ORCID iD: 0000-0002-5146-8979

researcher, Department of prevention and treatment of especially dangerous infections

Россия, Koltsovo

Ekaterina I. Filippova

State Research Center of Virology and Biotechnology “Vector”

Email: zarytova@niboch.ncs.ru
ORCID iD: 0000-0001-9554-4462

Cand. Sci. (Biol.), researcher, Department of prevention and treatment of especially dangerous infections

Россия, Koltsovo

Natalya A. Mazurkova

State Research Center of Virology and Biotechnology “Vector”

Email: zarytova@niboch.ncs.ru
ORCID iD: 0000-0002-1896-2684

D. Sci. (Biol.), leading researcher, Department of prevention and treatment of especially dangerous infections

Россия, Koltsovo

Valentina F. Zarytova

Institute of Chemical Biology and Fundamental Medicine

Author for correspondence.
Email: zarytova@niboch.ncs.ru
ORCID iD: 0000-0002-9579-9972

D. Sci. (Chem.), leading researcher, Laboratory of nucleic acids

Россия, Novosibirsk

References

Supplementary files