Novel Polyanions Inhibiting Replication of Influenza Viruses

Abstract

Novel sulfonated derivatives of poly(allylamine hydrochloride) (NSPAHs) and N-sulfonated chitosan (NSCH) have been synthesized, and their activity against influenza A and B viruses has been studied and compared with that of a series of carrageenans, marine polysaccharides of well-documented anti-influenza activity. NSPAHs were found to be nontoxic and very soluble in water, in contrast to gel-forming and thus generally poorly soluble carrageenans. In vitro and ex vivo studies using susceptible cells (Madin-Darby canine kidney epithelial cells and fully differentiated human airway epithelial cultures) demonstrated the antiviral effectiveness of NSPAHs. The activity of NSPAHs was proportional to the molecular mass of the chain and the degree of substitution of amino groups with sulfonate groups. Mechanistic studies showed that the NSPAHs and carrageenans inhibit influenza A and B virus assembly in the cell.

INTRODUCTION

Influenza A virus is a pathogen that causes acute viral infections, most commonly in seasonal outbreaks; however, appearance of new strains may result in pandemics as seen for Spanish flu in 1918, or more recently with the H1N1/09 virus. According to the available literature, influenza virus infections account for 3 to 5 million cases of respiratory disease yearly, of which ∼250,000 to 500,000 cases are fatal (1). The disease is most severe in children under 2 years old, elderly, and individuals with an underlying disease (2, 3).

Currently, vaccination is considered to be the most effective method for prevention of severe influenza illness (4), though it also has its drawbacks. The vaccine is developed yearly for the Northern and Southern hemispheres, but emergence of novel strains may limit the efficacy of the vaccination (3). For severe cases of influenza in vaccinated and/or unvaccinated individuals, there are two approved drugs available, hampering the release of the virus from the infected cell by inhibition of the viral neuraminidase (NA) activity, i.e., oseltamivir and zanamivir, which are most effective within 48 h after exposure/infection (3, 5, 6). Unfortunately, the availability of the vaccine and the two antivirals is not sufficient to fight influenza, so the number of fatal cases is still surprisingly high and influenza viruses are rapidly evolving, leading to the emergence of escape mutants (7). A perfect example of how important it is to develop novel treatment strategies is provided by two older medicinal products hampering influenza virus replication—amantadine and rimantadine—which interact with the M₂ ion channel protein. Even though these drugs proved to be effective, rapid virus evolution resulted in the appearance of escape variants, and therefore, these substances are no longer in use for antiviral therapy (3, 5, 8, 9). These facts altogether drive the research on anti-influenza drugs, and some compounds inhibiting NA activity are already in development or are approved in some countries (e.g., laninamivir, peramivir) (10–12).

In addition to the aforementioned NA, the other essential protein for virus transmission is hemagglutinin (HA), which is responsible for binding of the virus to the receptor and virus entry into susceptible cells (13). Previous efforts to utilize this protein as a target in therapy were successful, as it has been reported that some peptides can interact with viral HA protein and hamper virus internalization into the cell. It was also shown that Fludase (recombinant protein with neuraminidase-like activity) cleaves off the sialic acid from the cell surface and makes the cells nonpenetrable to influenza virus (14).

The concept of utilization of macromolecules for inhibition of influenza virus infection has already been introduced. It was found that the sulfated derivatives of polysaccharides (mainly carrageenans but also dextran sulfate, chondroitin sulfate, heparin sulfate, and sulfate derivatives of cellulose, curdlan, xylan, and fucoidans) can interact directly with the influenza virus particles and other enveloped viruses such as herpesvirus (HSV) or human immunodeficiency virus (HIV), preventing their adsorption and internalization into the cells (15, 16). Unfortunately, the solubility of carrageenans is limited, especially in aqueous solutions containing potassium and calcium ions (17), since in the presence of these ions carrageenans form viscous gels, which is undesirable in pharmaceutical application (18). What is more, another well-known side effect of sulfated polysaccharides is their anticoagulant activity (19). Despite that, ι-carrageenan has passed the clinical trials, and it is currently approved and marketed in some countries as an anti-influenza drug (20–22).

In the current study, we tested a number of different polymeric compounds for their ability to inhibit influenza virus replication. We have confirmed antiviral activity of heparan sulfate and carrageenans, but more importantly, we have demonstrated that the N-sulfonated derivative of poly(allylamine) with a molecular mass of 56 kDa and a degree of substitution (DS) of 98% of amino groups with sulfonate groups (NSPAH-56-98) showed promising antiviral activity, inhibiting virus replication in vitro and ex vivo. Detailed analysis of the mechanism of action showed that the compound stops the release of progeny virions. Surprisingly, the conducted analyses suggest that ι-carrageenan inhibits influenza virus replication utilizing a similar mechanism, different than previously described.

MATERIALS AND METHODS

Polymers.

Poly(allylamine hydrochloride) (PAH-15-0, average molecular mass of ∼15 kDa; PAH-56-0, average molecular mass of ∼56 kDa), poly(allylamine hydrochloride) solution (PAH-65-0, average molecular mass of ∼65 kDa, 20% wt in H₂O), chitosan (CH) (low molecular weight, 75 to 85% deacetylated), sulfur trioxide trimethylamine complex (STTC), fluorescein 5(6)-isothiocyanate (FITC), ι-carrageenan (ι-car),S and κ-carrageenan (κ-car) were obtained from Sigma-Aldrich, Poland. Food carrageenans, i.e., Gelcarin CH 8718 (G-CH), Gelcarin ME 3054 (G-ME), and Viscarin SD 389 (V-SD), were gifts from FMC BioPolymer, USA. For all poly(allylamine hydrochloride) (PAH) derivative compounds [i.e., novel sulfonated derivatives of poly(allylamine hydrochloride) (NSPAHs)], the compounds are named by the following convention: the acronym (PAH or NSPAH), molecular mass (in kilodaltons), and degree of substitution (as a percentage). For unmodified PAHs, the acronym is followed by the molecular mass and zero.

Synthesis of sulfonated polymers. (i) N-sulfonated chitosan.

N-sulfonation of chitosan was conducted using the previously described method, with some modification (23). Briefly, 0.5 g of chitosan (CH) was dispersed in 25 ml of distilled water in a two-necked flask and treated with 1.65 g Na₂CO₃ (POCh, Poland). After 30 min of mixing, 2.16 g STTC was added. The reaction mixture was incubated at 55°C under constant mixing and bubbling with nitrogen for 82 h. The mixture was cooled down and diluted, and unreacted chitosan was separated by centrifugation for 7 min at 2,650 relative centrifugal force (rcf) and discarded. Subsequently, the product was transferred to dialysis tubes (molecular mass cutoff value of 12.8 kDa) and dialyzed exhaustively against distilled water for 7 days. The dialyzed solution was lyophilized to yield a white, fluffy solid of N-sulfonated chitosan (NSCH).

(ii) N-sulfonated poly(allylamine).

In a 50-ml two-necked flask, 0.5 g of poly(allylamine) hydrochloride (PAH-15-0, molecular mass of ∼15 kDa, or PAH-56-0, molecular mass of ∼56 kDa) or 2.45 ml of 20% (wt/vol) poly(allylamine hydrochloride) solution (PAH-65-0, molecular mass of ∼65 kDa) was dissolved in 25 ml of distilled water. Subsequently, 1.85 g of Na₂CO₃ (POCh, Poland) was added, and the mixture was stirred for 45 min under constant bubbling with nitrogen to dissolve the polymer and unlock the amino groups. An adequate quantity of STTC was added to obtain the polymers with different degrees of substitution (Table 1). The reaction mixture was incubated at 55°C for 48 h under constant stirring and under bubbling with nitrogen. Then, the mixture was diluted with distilled water, cooled down, and dialyzed (molecular mass cutoff of 14 kDa) against water for 7 days. The obtained polymers were isolated from the solution using the freeze-drying technique.

TABLE 1.

Reaction conditions during anionic modification of the polymers and the degree of substitution with functional groups

Polymer	MMa (kDa)	Molar ratio of STTC and polymer amino groups	DS (%)b
PAH-15-0	15	0	0
PAH-65-0	65	0	0
NSPAH-15-30	15	0.8	30
NSPAH-15-95	15	5.0	95
NSPAH-56-98	56	5.0	98
NSPAH-65-75	65	1.8	75
NSPAH-65-89	65	5.0	89
NSCH	0	3.0	65

^aMM, molecular mass.

^bThe degree of substitution (DS) was calculated based on the results of the elemental analysis. DS is defined as the number of sulfonate groups per allylamine unit (NSPAH) or glucose unit (CH).

(iii) FITC-labeled polymers.

One hundred milligrams of poly(allylamine) hydrochloride (PAH-15-0) or N-sulfonated poly(allylamine) (NSPAH-65-89) was dissolved in 20 ml of distilled water. Five milligrams of fluorescein 5(6)-isothiocyanate (FITC) was dissolved in 1 ml of acetone, the solution was added dropwise to the sample containing PAH-15-0 or NSPAH-65-89 and vigorously mixed for 24 h at 45°C in the dark. Subsequently, the solution was diluted and dialyzed (molecular mass cutoff of 14 kDa) against water for 3 days and dialyzed further against acetone-water mixture (1:4 [vol/vol]) for 1 day and freeze-dried. The efficiency of polymer labeling was checked using UV-visible (UV-Vis) spectroscopy. In the UV-Vis spectra of FITC-labeled NSPAH-65-89 and PAH-15-0, a new absorption band at 450 nm appeared which could be associated with FITC bound to the polymer chain. The degree of substitution, defined as the number of FITC groups per allylamine unit of PAH-15-0 and NSPAH-65-89, was found to be 0.7% and 0.4%, respectively, based on the UV-Vis spectra.

Fourier transform infrared spectroscopy (FTIR)-attenuated total reflectance (ATR) spectra were recorded using a Nicolet IR 200 spectrometer (Thermo Scientific, USA). Elemental analysis was performed using a Vario Micro CHNS elemental analyzer (Elementar, Germany). The zeta potential was measured using a Zetasizer Nano-ZS (Malvern Instruments, England). Samples of polymers (0.5 mg/ml) were measured in two different media, i.e., in 1× phosphate-buffered saline (PBS) buffer at pH 7.4 and in the cell culture medium (0% Dulbecco modified Eagle medium [DMEM], i.e., DMEM containing no fetal bovine serum), by using laser Doppler velocimetry. Gel permeation chromatograms (GPC) were obtained using a high-performance liquid chromatography (HPLC system equipped with a PL Aquagel -OH mixed 8-μm column (Agilent Technologies, USA), 515 pump (Waters, USA), 717 Plus autosampler (Waters, USA), and 2996 photodiode detector (Waters). The eluent was 0.1 M NaNO₃, the sample volume was 30 μl, the sample concentration was 1 mg/ml, and the flow rate was 0.5 ml/min.

Cell culture.

Madin-Darby canine kidney epithelial cells (MDCK) (ATCC CCL-34 cell line) were maintained in DMEM (PAA Laboratories, Germany) supplemented with 3% heat-inactivated fetal bovine serum (FBS) (PAA Laboratories), penicillin (100 U/ml), and streptomycin (100 μg/ml) (i.e., 3% DMEM). The cells were cultured in T₂₅ flasks (TPP, Switzerland) at 37°C with 5% CO₂.

Human airway (tracheobronchial) epithelial (HAE) cells were obtained from airway specimens resected from patients undergoing surgery under protocols approved by the Silesian Center for Heart Diseases. This study was approved by the Bioethical Committee of the Medical University of Silesia in Katowice, Poland (approval no. KNW/0022/KB1/17/10 on 16 March 2010). Written informed consent was obtained from all patients. Primary cells were expanded on plastic to generate passage 1 cells and plated at a density of 3 × 10⁵ cells per well on permeable Transwell insert (6.5-mm-diameter) supports. HAE cultures were generated by provision of an air-liquid interface for 6 to 8 weeks to form well-differentiated, polarized cultures that resemble in vivo pseudostratified mucociliary epithelium.

Influenza A and B viruses.

Four influenza A virus isolates (A/H3N2/EVA obtained within the European Virus Archive EVA project, A/H3N2/Victoria/361/2011, A/H1N1/California/04/2009, and A/H1N1/Georgia/F32551/2012, a gift from BEI Resources) and an influenza B virus isolate (B/Brisbane/22/2008, a gift from BEI Resources) were used in the studies. Virus stock was propagated by infecting MDCK cells in Dulbecco modified Eagle medium (PAA Laboratories, Germany) supplemented with penicillin (100 U/ml) and streptomycin (100 μg/ml), without fetal bovine serum (i.e., 0% DMEM) in the presence of trypsin (1 μg/ml). Cells were lysed at day 2 postinfection (p.i.) by two freeze-thaw cycles, aliquoted, and stored at −80°C. A control MDCK cell lysate containing no virus was prepared and stored in the same way as the virus stock.

Virus yield was assessed by titration of fully confluent MDCK cells in the presence of trypsin (1 μg/ml) in 96-well plates according to the Reed and Muench formula (24). The plates were incubated at 37°C for 2 days, and the occurrence of a cytopathic effect (CPE) was scored using an inverted microscope.

For virus inhibition assays, MDCK cells were seeded on 96-well plates. Virus infection was carried out after 2 days of incubation at 37°C with 5% CO₂, with the virus suspended in the respective polymer solution at a dose of 400 50% tissue culture infectious doses (TCID₅₀)/ml in a total volume of 100 μl. Virus cultures were maintained in 0% DMEM supplemented with trypsin (1 μg/ml).

Mechanism of action.

The methods used to determine the mechanism of action of the active polymers were conducted using previously described procedures with some modifications (25). The assays are detailed described below.

(i) Virus inactivation assay.

Influenza A virus (IAV) samples were incubated with polymers with constant mixing at 22°C for 1 h. Subsequently, the samples were diluted to decrease the polymer concentration below its active range (i.e., <10 μg/ml). Consequently, if the antiviral effect relies on the direct interaction between the virus and the polymer, the antiviral effect should be visible after such treatment. The suboptimal concentration of the polymer during infection does not affect virus replication, as tested experimentally (data not shown). The control sample was prepared in a similar manner, but 0% DMEM was used instead of the polymer solution. Virus yield was assessed by titration of fully confluent MDCK cells on day 2 postinfection, according to the Reed and Muench formula (24).

(ii) Virus adsorption assay.

Fully confluent MDCK cells were overlaid with the polymer at the desired concentration or control sample and incubated at 37°C for 1 h. Following the incubation, medium was removed and cells were rinsed three times with sterile PBS. Subsequently, fresh medium with IAV (400 TCID₅₀/ml) or a mock-treated control sample was added to each well, and the cells were incubated at 37°C for 2 h. Media with nonadsorbed IAV particles were removed, and MDCK cells were rinsed three times with sterile PBS. Cells were overlaid with 0% DMEM (100 μl/well) and cultured at 37°C for 48 h. Virus infection was scored based on the appearance of the cytopathic effect. Cell culture supernatant samples were collected 2 days p.i. and used for real-time PCR analysis. Control samples containing no virus were treated in the same manner. As there was no polymer on the cells during or after infection, inhibition of virus replication may result only from interaction of the polymer with cells (e.g., receptor depletion).

(iii) Virus attachment assay.

Fully confluent MDCK cells were chilled on ice, exposed to cold (4°C) solutions of polymers in cell culture media and incubated at 4°C for 1 h. At this temperature, virions can attach to their receptors on the cell surface, but virus internalization is hampered due to the inhibition of intracellular transport (25). Subsequently, medium was removed, and cells were rinsed three times with ice-cold (4°C) PBS. Cells were overlaid with 0% DMEM (100 μl/well) and cultured at 37°C for 48 h. Briefly, infection was possible only if viruses were able to attach to the receptor/attachment factor in the presence of the polymer. Cell culture supernatant samples were collected 2 days p.i. and used for real-time PCR analysis.

(iv) Virus internalization assay.

The virus internalization assay was carried out as previously described (25). Briefly, fully confluent MDCK cells were overlaid on ice with 100 μl of ice-cold medium containing IVA at 400 TCID₅₀/ml or mock-treated control samples and incubated on ice for 2 h. Subsequently, the medium was discarded, and the cells were rinsed three times with ice-cold PBS to remove unbound IAV particles. Next, the cells were overlaid with medium containing the polymer solutions or control sample, warmed up, and incubated at 37°C for 2 h. Briefly, during the first incubation, virions were expected to attach to their receptors as described above, while virus internalization was inhibited at 4°C. The subsequent 2 h of incubation at 37°C in the presence of the tested polymer allowed virus entry only if virus internalization had not been affected by the polymer tested. Following this step, viruses that were not internalized were inactivated using the acidic wash (pH 3.0; 0.1 M glycine, 0.1 M sodium chloride), as described before. Further, cells were rinsed three times with sterile PBS (pH 7.4), and fresh medium (0% DMEM) was applied. The cells were incubated at 37°C for 48 h, and CPE was evaluated using an inverted microscope. Mock-treated control samples were managed in the same manner.

(v) Virus assembly, packing, and releasing assay.

Fully confluent MDCK cells were inoculated with 100 μl/well IAV (or mock-treated control samples) at 400 TCID₅₀/ml. Following 2 h of incubation at 37°C medium was removed and cell surface was rinsed three times with sterile PBS (pH 7.4). Subsequently, cells were overlaid with polymer-supplemented medium and cultured at 37°C for 48 h. Finally not only cell culture supernatant samples were used for real-time PCR analysis, but also lysed cells (virus replication assay). Mock-treated control samples were managed in the same manner.

(vi) Virus replication assay.

This assay was conducted analogically to the assay described in the previous paragraph. The only difference was that in this assay the tested sample was cell lysate and not cell culture medium as in the assay above.

Quantitative real-time PCR.

Total RNA was isolated from the cell culture supernatant or apical washes from HAE cultures using EZ-10 Spin Column Total RNA Mini-Preps Super kit (Bio Basic Canada Inc., Canada) according to the manufacturer's instructions. Reverse transcription was carried out with a high-capacity cDNA reverse transcription kit (Life Technologies) according to the manufacturer's instructions. Conversion of total RNA to single-stranded cDNA was carried out using a Veriti thermal cycler (Applied Biosystems, USA). Virus yield was determined using real-time PCR. Briefly, 2.5 μl of cDNA was amplified in a 10-μl reaction mixture containing 1× TaqMan Universal PCR master mix, No AmpEraseUnG (Thermo Scientific, Poland), a specific probe labeled with 6-carboxyfluorescein (FAM) and 6-carboxytetramethylrhodamine (TAMRA) (200 nM), and primers (900 nM each). Carboxy-X-rhodamine (Rox) was used as a reference dye. The sequences of primers for A/H3N2/EVA (segment 7) are as follows: Sense primer (IVF [IV stands for influenza virus, and F stands for forward]), 5′ AGA TGA GTC TTC TAA CCG AGG TCG 3′; Antisense primer (IVR [R stands for reverse]) 5′ TGC AAA AAC ATC TTC AAG TCT CTG 3′l and FAM/TAMRA probe, 5′-TCA GGC CCC CTC AAA GCC GA −3′. The reaction was monitored with a 7500 Fast Real-Time PCR machine (Applied Biosystems, USA) with the following settings: 2 min at 50°C, 10 min at 95°C, 40 cycles, with 1 cycle consisting of 15 s at 95°C and 1 min at 60°C. The primers for the other influenza virus strains (Swine Influenza H1 Forward primer, Swine Influenza H1 Reverse primer, and Swine Influenza H1 probe for A/H1N1/California/04/2009 and A/H1N1/Georgia/F32551/2012 subtypes; Influenza A Forward primer, Influenza A Reverse primer, Influenza A probe for A/H3N2/Victoria/361/2011 subtype; and Influenza B Forward primer, Influenza B Reverse primer, and Influenza B probe for B/Brisbane/22/2008 subtype) were obtained from BEI Resources (Manassas, VA, USA). In order to assess the copy number for influenza viruses used in the study, DNA standards were prepared. For influenza A/H3N2/EVA fragment, segment 7 was amplified and cloned into pTZ57R/T plasmids using InsTAclone PCR cloning kit (Thermo Scientific). The resulting plasmid was linearized, and its concentration was assessed using a spectrophotometer. Samples were serially diluted and used as an input for real-time PCR. For other influenza strains, standards were obtained from BEI Resources (Manassas, VA, USA).

In this article, the data from quantitative PCR are presented as log removal values (LRVs) in order to enable comparison of results obtained from different assays. LRV was calculated according to the following formula: LRV = −log (c_i/c₀) where c_i is the number of viral RNA copies per milliliter in the sample in the culture treated with a given polymer and c₀ is the number of viral RNA copies per milliliter in control sample (untreated cells). Furthermore, raw data (number of copies per milliliter) are presented in the files in the supplemental material.

Fetuin binding assay.

The fetuin binding assay (FBA) was based on enzyme-linked immunosorbent assay (ELISA). Each well on the 96-well Nunc Maxisorp plate was coated with 100 μg/ml of fetuin solution (100 μl/well) in sterile 1× PBS. The wells were coated for 24 h at 4°C. The wells on the plate were washed three times with 100 μl of 0.05% Tween 20 in Dulbecco's phosphate-buffered saline (DPBS), and then 100 μl of blocking solution (3% bovine serum albumin, 0.1% Tween 20 in PBS with 1.8 mM magnesium cations and no calcium cations [PBS^+Mg-Ca]) was added to each well. Blocking was carried out for 24 h. On day 3, the blocking buffer was discarded, and wells were washed four times with 100 μl of PBS^+Mg-Ca.

In order to test the ability of the compound to interfere with the virus-receptor interaction, 100 μl of the test compound at a concentration of 1 mg/ml in 1× PBS^+Mg-Ca was added to each well. 1× PBS^+Mg-Ca was used as a control. The samples were incubated for 2 h at 37°C, and then the plates were washed three times with 0.05% Tween 20 in DPBS. A serially diluted complex of soluble trimerized hemagglutinin (HA), StrepMab-Classic conjugated to horseradish peroxidase (StrepMab-Classic-HRP), and rabbit anti-mouse antibody conjugated to HRP (rabbit-α-mouse-HRP) in 100 μl of blocking buffer was added to each well. The plates were incubated for 30 min on ice and washed three times with 0.05% Tween 20 in DPBS. Signal was developed with 3,3′,5,5′-tetramethylbenzidine (TMB) substrate and scored using a SpectraMAX 250 spectrophotometer (Molecular Devices, USA) at a wavelength (λ) of 480 nm.

Amplex red neuraminidase assay (NA assay).

Neuraminidase activity was measured using an Amplex red neuraminidase assay kit (Molecular Probes) according to the manufacturer's protocol. The protocol describes the NA assay in a total volume of 100 μl per microplate well. First, the polymers were dissolved in 1× reaction buffer. The neuraminidase stock solution was diluted 50-fold with a polymer solution of the proper concentration. For a control, neuraminidase was diluted 50-fold in 1× reaction buffer. Then, 2× working solution (100 μM Amplex red reagent containing 0.2 U/ml HRP, 4 U/ml galactose oxidase, and 500 μl/ml fetuin in 1× reaction buffer) was prepared. The reaction was started by adding 50 μl of the Amplex red reagent (HRP, galactose oxidase, fetuin working solution) to each microplate well containing the samples and controls. The microplate was incubated for 30 min at 37°C and protected from light. The absorbance was measured at 560 nm using a Spectra MAX 250 spectrophotometer (Molecular Devices, USA). The assays were performed in triplicate in a total volume of 100 μl per microplate well.

XTT assay.

MDCK cells were cultured on 96-well plates for 48 h. Subsequently, medium was discarded, cells were washed with sterile PBS, and fresh medium supplemented with studied polymers was added (100 μl per well). Cell viability was evaluated 48 h p.i. using XTT Cell Viability Assay kit (Biological Industries) according to the manufacturer's instructions. Briefly, 25 μl of the activated 2,3-bis-(2-methoxy-4-nitro-5-sulphenyl)-(2H)-tetrazolium-5-carboxanilide (XTT) solution was added to each well. Signal was developed for 2 h at 37°C and scored using a Spectra MAX 250 spectrophotometer (Molecular Devices, USA).

Viability of HAE cultures was determined in a similar manner, though in this case 50 μl of the activated XTT solution was inoculated onto the apical surface of the HAE culture following 3-day incubation with polymers. The signal was measured as described above.

Neutral red uptake (NRU) assay.

MDCK cells were prepared in the same manner as for the XTT assay. Forty-eight hours p.i., media were discarded, cells were rinsed with sterile PBS, and 100 μl of neutral red solution (50 μg/ml) in PBS supplemented with 3% FBS was added to each well. The plate was incubated at 37°C for 2 h, and the cells were washed three times with PBS and lysed (1% aqueous solution of acetic acid in 50% ethanol, 100 μl/well). The absorbance of neutral red absorbed by viable cells was measured at λ = 540 nm using a spectrophotometer (Spectra MAX 250; Molecular Devices). The results obtained were further normalized to the values for control samples where cell viability was set at 100%.

Statistical analysis.

All experiments were performed at least in triplicate, and results are expressed as means ± standard deviations (SD). For all polymers tested, the 50% inhibitory concentration (IC₅₀) values were calculated using the formula function in Graph Pad Prism “Dose response-Inhibition: log(inhibitor) vs. response variable slope.” To determine significance of the results obtained, a comparison between groups was made using the Student t test. P values of <0.05 were considered significant.

RESULTS

Polymers.

A series of sulfonated polymers, i.e., N-sulfonated poly(allylamine)s (NSPAHs) and N-sulfonated derivative of chitosan (NSCH), a natural polymer, was synthesized using sulfur trioxide-trimethylamine complex (STTC) as a sulfonating agent (Table 1 and Fig. 1). For all poly(allylamine hydrochloride) (PAH) derivative compounds (i.e., NSPAHs), the compounds are named by the following convention: the acronym (PAH or NSPAH), molecular mass (in kilodaltons), and degree of substitution (as a percentage). For unmodified PAHs, the acronym is followed by the molecular mass and zero. Unmodified PAHs (PAH-15-0 and PAH-65-0) were included as negative controls, while inclusion of unmodified chitosan was not possible, due to its poor solubility under the experimental conditions.

FIG 1.

Scheme of the synthesis of N-sulfonated poly(allylamine hydrochloride)s, NSPAHs (A) and N-sulfonated chitosan, NSCH (B).

As it has been reported that carrageenans inhibit replication of influenza A virus (17, 26), these natural sulfated polymers were included in the study as reference compounds. In total, five polymers from this group, i.e., κ-carrageenan (κ-car), ι-carrageenan (ι-car), and three carrageenans used as food additives (Gelcarin CH 8718 [G-CH], Gelcarin ME 3054 [G-ME], and Viscarin SD 389 [V-SD]) were used in subsequent analyses as reference compounds. FTIR-ATR spectra of tested carrageenans are presented in Fig. S1 in the supplemental material.

Elemental analysis of the synthesized polymers confirmed the sulfonation of the amino groups and allowed calculation of the degree of substitution of the anionic polymers obtained (Table 1). Substitution with the sulfonate groups was also confirmed using FTIR-ATR. In the FTIR-ATR spectra of NSPAHs, three new bands appeared; these bands were absent in the respective spectra of nonsulfonated PAHs (see Fig. S1 in the supplemental material). The bands, located between 1,180 and 1,140 cm⁻¹, can be attributed to a stretching vibration of the -SO₂-N- group, and the bands that appeared in the range of 1,210 to 1,150 cm⁻¹, 1,060 to 1,030 cm⁻¹, and at 650 cm⁻¹ can be assigned to a stretching vibration of –SO₂– in –SO₃H or –SO₃⁻ groups. In FTIR-ATR spectra of N-sulfonated chitosan, the analogous bands may also be noted; these bands are not present in FTIR spectra of unmodified chitosan (Fig. S1).

To determine the charge of the studied polymeric chains, their zeta potential was measured in two different media, i.e., in 1× PBS buffer (pH 7.4) and in the cell culture medium (0% DMEM). As expected, the zeta potential values for NSPAHs were negative in both media, in contrast to those of unmodified PAHs, and correlate with the density of the anionic charges in the polymers (see Table S1 in the supplemental material). Also, N-sulfonated chitosan and carrageenans showed negative zeta potential.

In vitro inhibition of influenza A and B virus replication with NSPAH polymers.

To verify whether NSPAHs show antiviral activity against influenza virus strains, the experiments were conducted using different subtypes of influenza A virus strain (A/H3N2/EVA, A/H1N1/California/2009,A/H1N1/Georgia/F32551/2012, and A/H3N2/Victoria/361/2011) and an influenza B virus strain (B/Brisbane/33/2008). To examine whether synthesized polymers (Table 1) show antiviral activity, two assays were applied using influenza A/H3N2/EVA virus first. The first one was the microscopic evaluation of the cytopathic effect (CPE) 48 h p.i. Visual assessment of CPE indicated that NSPAH-56-98, NSPAH-65-75, and NSPAH-65-89, i.e., polymers with high molecular masses and relatively high degrees of substitution with sulfonic groups, inhibited influenza A/H3N2/EVA virus replication at a concentration of 250 μg/ml (Table 2). NSPAH-15-95 inhibited CPE appearance at 500 μg/ml, while NSPAH-15-30 (the polymer with the lowest molecular mass and a low degree of substitution with sulfonate groups) showed the weakest inhibitory effect for influenza A/H3N2/EVA virus (partial inhibition at 1 mg/ml) (Table 2). Further, all carrageenans tested inhibited CPE caused by influenza A/H3N2/EVA replication at an inhibitory concentration (IC) comparable to those obtained for NSPAH-56-98, NSPAH-65-75, and NSPAH-65-89 (Table 2). The most prominent antiviral effect was observed for κ-car and V-SD, for which no CPE was noted at a concentration of 100 μg/ml. Unmodified PAHs with both low and high molecular masses showed no activity against influenza A/H3N2/EVA virus. Surprisingly, N-sulfonated chitosan, which, beside sulfonate groups, is composed of glucose units like carrageenans, and contains amine groups like poly(allylamine)s, did not exhibit any activity against A/H3N2/EVA virus.

TABLE 2.

Anti-A/H3N2/EVA activity and cytotoxicity of the polymers tested

Polymer	Antiviral activity (μg/ml)		Cytotoxicity (μg/ml)c		Selectivity index (CC50[XTT]/IC50)c
	IC MDCK (CPE)a	IC50 MDCK (qRT-PCR)b	CC50 (XTT)d	CC50 (NRU)e
PA-15-0	—	—	56.3 ± 2.6	49.5 ± 2.7	—
PA-65-0	—	—	93.3 ± 1.7	65.5 ± 2.2	—
NSPAH-15-30	1,000	53.5 ± 1.7	>5,000	>5,000	>93
NSPAH-15-95	500	4.5 ± 2.1	>5,000	>5,000	>1,100
NSPAH-56-98	250	2.0 ± 1.2	>5,000	>5,000	>2,500
NSPAH-65-75	250	0.5 ± 1.2	>5,000	>5,000	>10,000
NSPAH-65-89	250	0.6 ± 1.1	>5,000	>5,000	>8,300
NSCH	—	—	>5,000	>5,000	—
ι-Car	250	4.0 ± 1.0	>5,000*	>5,000*	>1,250*
κ-Car	100	0.2 ± 2.0	>5,000*	>5,000*	>25,000*
V-SD	100	0.2 ± 1.2	>5,000	>5,000	>25,000
G-ME	250	9.5 ± 1.6	>5,000	>5,000	>500
G-CH	250	23.2 ± 1.1	2,054 ± 1	3,200 ± 1	∼88

^aThe lowest compound concentration for which no CPE was noted or in the case of NSPAH-15-30, a partially reduced CPE was noted. —, no antiviral activity.

^bFifty percent inhibitory concentration as assessed by real-time RT-PCR analysis calculated using the formula function in Graph Pad Prism “Dose response-Inhibition: log(inhibitor) vs. response variable slope.” —, no antiviral activity.

^cThe asterisks indicate that ι-carrageenan and κ-carrageenan at a concentration of 5,000 mg/ml form viscous gels, and it was difficult to measure absorbance, but the morphology of the MDCK cells was unchanged. —, no data.

^dCompound concentration required to inhibit cell growth by 50% assessed by the XTT assay.

^eCompound concentration required to inhibit cell growth by 50% assessed by the neutral red uptake (NRU) assay.

To verify the obtained results, quantitative real-time reverse transcription-PCR (RT-PCR) analysis was carried out on the cell culture supernatants. The results were consistent with those obtained based on visual CPE assessment (Table 2 and Fig. 2A; see Fig. S2A in the supplemental material). Inhibitory activity of the synthetic polymers against A/H3N2/EVA virus appeared to be correlated with the molecular mass and degree of substitution, as presented in Fig. 2A and Fig. S2A. Taking into account that for different assays the infection was carried out differently, data are normalized to the values for control samples and presented as log removal values (LRVs) which were calculated according to the formula described in Materials and Methods. The antiviral activity of carrageenans was evaluated concomitantly (Fig. 2B, Fig. S2B, and Table 2). Further, in order to ensure that the compounds limit the production of infectious virus particles, virus titers in cell culture supernatants were determined. Results are presented in Fig. 3 (see also Fig. S3 in the supplemental material).

FIG 2.

(A) Inhibition of influenza A/H3N2/EVA virus replication in MDCK cells by NSPAH-15-30, NSPAH-15-95, NSPAH-56-98, NSPAH-65-75, and NSPAH-65-89. (B) Inhibition of influenza A/H3N2/EVA replication in MDCK cells by ι-carrageenan (ι-car), κ-carrageenan (κ-car), and food carrageenans Gelcarin CH8718 (G-CH), Gelcarin ME 3054 (G-ME), and Viscarin SD 389 (V-SD). Data are presented as log removal values (LRVs), which are normalized to the values for control samples (no polymer added). Before normalization, the data for each concentration (expressed as the number of copies per milliliter) were compared to the value for the control sample (also expressed as the number of copies per milliliter). Values that are significantly different (P < 0.05) from the value for the control are indicated by an asterisk. Values below the detection limit (∼1,000 copies/ml) are indicated by hatched bars. All assays were performed in quadruplicate, and average values with standard deviations (error bars) are presented.

FIG 3.

Polymer-mediated decrease in influenza virus titers. Results of influenza A/H3N2/EVA virus titration carried out on the MDCK cell culture supernatants after 2 days postinfection in the presence of polymeric inhibitors, i.e., NSPAH-15-98, NSPAH-65-89, and ι-car, are shown. The sample that was not evaluated due to the gelling properties of ι-car is indicated by the # symbol. All assays were performed in triplicate, and average values with standard deviations are presented. Data for each concentration are presented as 50% tissue culture infectious doses (TCID₅₀) per milliliter and were compared to control sample (no polymer added). Values that are significantly different (P < 0.05) from the value for the control sample (no polymer added) are indicated by an asterisk.

To verify whether NSPAHs show antiviral activity against other virus strains, the experiments were conducted using A/H1N1/California/2009, A/H1N1/Georgia/F32551/2012, and A/H3N2/Victoria/361/2011 influenza A virus strains and one influenza B virus strain, B/Brisbane/33/2008. The experiments were conducted using NSPAHs showing the highest inhibitory activity against A/H3N2/EVA virus, i.e., NSPAH56-98 and NSPAH-65-89. The results show that both polymers can inhibit replication of all studied viruses. The antiviral activity of both polymers, however, varies remarkably for different subtypes (by up to 6.7 LRV units), being highest against A/H1N1/Georgia/F32551/2012 and B/Brisbane/33/2008 viruses (Fig. 4; also see Fig. S4 in the supplemental material).

FIG 4.

Inhibition of a variety of influenza strains, i.e., influenza A virus (A/H3N2/Victoria/361/2011, A/H1N1/California/04/2009, and A/H1N1/Georgia/F32551/2012) and influenza B virus (B/Brisbane/22/2008) replication in MDCK cells by NSPAH-56-98 and NSPAH-65-89 at a concentration of 1,000 μg/ml. Data are presented as log removal values (LRVs), which are normalized to the values for control samples (no polymer added). Before normalization, the data for each concentration (expressed as the number of copies per milliliter) were compared to the value for the control sample (also expressed as the number of copies per milliliter) (*, P < 0.05). Values below the detection limit (1,000 copies/ml) are indicated by hatched bars. All assays were performed in triplicate, and average values with standard deviations are presented.

Cytotoxicity in vitro.

The cytotoxicity of the polymers can be expressed as the 50% cytotoxic concentration (CC₅₀), i.e., the concentration at which cell viability is reduced by 50% relative to the cell viability of the untreated cell cultures. Two assays were performed in order to determine the cytotoxicity of the polymers tested, i.e., the XTT assay and NRU assay. The results of both tests were consistent. In the studied concentration range of the polymers (i.e., up to 5,000 μg/ml), CC₅₀ was not reached for most of the tested polymers (Table 2), except for G-CH for which the CC₅₀ is 2,054 ± 1 μg/ml (determined using the XTT assay) or 3,200 ± 1 μg/ml (determined using the NRU assay). The selectivity index (SI), defined as the ratio of CC₅₀ to IC₅₀, could not be calculated for the polymers tested because of unknown CC₅₀ value; however, it can be estimated that for G-CH, it falls within the range from 88 (for the XTT assay) to 139 (for the NRU assay).

In contrast, unmodified polyallylamines PAH-15-0 and PAH-65-0 showed relatively high cytotoxicity. CC₅₀ values equal to 56.3 ± 2.6 μg/ml (XTT assay) or 49.5 ± 2.7 μg/ml (NRU assay) for PAH-15-0 and 93.3 ± 1.7 μg/ml (XTT) or 65.5 ± 2.2 μg/ml (NRU) for PAH-65-0 were found (Table 2). The high cytotoxicity observed for nonsulfonated PAH can be explained by considering the differences in the physicochemical properties of these polymers, mainly in their zeta potential values (see Table S1 in the supplemental material), resulting in differences in their interactions with the cell surface.

Ex vivo inhibition of influenza A virus replication.

For further experiments, an ex vivo culture model of fully differentiated respiratory epithelium (HAE) was used to verify the data obtained. This multilayered, ciliated, and fully differentiated model morphologically, structurally, and functionally mimics human respiratory epithelium, providing the optimal conditions for respiratory pathogens. Moreover, these cultures allow infection and replication of several respiratory viruses, including influenza A viruses (27, 28). In order to test the antiviral potential of NSPAH-65-89, i.e., the polymer with the highest inhibitory activity against influenza A/H3N2/EVA virus determined by in vitro tests, the virus (TCID₅₀ = 400/ml) was applied in the presence of NSPAH-65-89 (50 μg/ml or 1,000 μg/ml in 1× PBS) on the apical surfaces of HAE cell cultures and incubated at 37°C for 2 h. Subsequently, the excess unbound virus was removed by washing the cultures three times with sterile 1× PBS, and the cultures were overlaid with 150 μl of sterile 1× PBS supplemented with the tested polymer. Seventy-two hours p.i., the apical surfaces of HAE cell cultures were again overlaid with 150 μl of sterile 1× PBS, and after 10-min incubation at 37°C, apical washes were collected and used for total RNA isolation. Virus quantification showed that influenza A/H3N2/EVA virus replication had been severely affected, and in the sample treated with 1,000 μg/ml of the polymer, no virus was detected (Fig. 5A; see Fig. S5 in the supplemental material). Further, visual inspection of the HAE culture showed that in the presence of the polymer, no CPE development could be spotted (Fig. S6).

FIG 5.

Inhibition of influenza A/H3N2/EVA replication in HAE cultures. (A) Inhibition of virus replication in the HAE culture, as determined by real-time PCR in the presence of NSPAH-65-89 during 3-day infection time. Data on virus replication are presented as log removal values (LRVs) which are normalized to the values for control samples (no polymer added). Before normalization, the data for each concentration (expressed as the number of copies per milliliter) were compared to the value for the control sample (also expressed as the number of copies per milliliter) (*, P < 0.05). The value below the detection limit (∼1,000 copies/ml) is indicated by the hatched bar. (B) Cytotoxicity of NSPAH-65-89 on HAE cultures. Cell viability was assessed by the XTT assay. The data for each concentration were compared to the value for the control sample (no polymer added) (*, P < 0.05). All assays were performed in triplicate, and average values with standard deviations are presented.

The XTT assay was also used to verify the cytotoxicity of the NSPAH-65-89 polymer on HAE cell cultures. No cytotoxicity of the polymer could be noted on HAE cultures following the 72 h of incubation at given concentrations (Fig. 5B; see Fig. S6 in the supplemental material).

Mechanism of action.

We aimed to investigate the mechanism of action of the studied polymers, and the following options were considered. (i) The polymers interact directly with the viral particles, making them unable to attach to the respective receptors on the cell surface. (ii) The polymers interact with the receptors on the cell surface, making them inaccessible to virions. (iii) The polymers do not interact with the virions or receptors directly, but they inhibit virus internalization and nuclear translocation. (iv) The polymers interfere with virus replication, release, or other postentry steps. For these studies, influenza A/H3N2/EVA virus and the polymers with the highest antiviral activity (i.e., NSPAH-15-95, NSPAH-56-98, NSPAH-65-89, ι-car, and κ-car) were used. Detailed descriptions of the methods used to determine the mechanism of action of the active compound are provided in Materials and Methods.

First, we determined whether the polymers interact directly with virions, e.g., by irreversibly coating them, as previously reported for carrageenans (20, 29, 30). Briefly, influenza A/H3N2/EVA virions were incubated with polymers for 1 h at 22°C with constant mixing. Subsequently, samples were diluted 200 times to decrease the polymer concentration below its active range (i.e., <10 μg/ml) and overlaid on MDCK cells in decreasing concentrations (serial dilutions). The control sample was prepared in a similar manner, i.e., the sample was diluted, and 0% DMEM was used instead of the polymer solution. Virus titer (based on CPE) was assessed on day 2 postinfection, according to the Reed and Muench formula (24). No decrease in virus titer could be seen for NSPAH or carrageenans compared to the control samples. This shows that the polymers do not irreversibly interact with the virions.

Next, the ability of polymers to interfere with the receptor-HA interaction was tested. In order to determine whether the tested compounds are able to perturb the receptor-virus binding, we conducted an ELISA-based fetuin binding assay (FBA). Fetuin is a blood glycoprotein mimicking receptors on the surface of the host cells to which viral surface glycoprotein (hemagglutinin [HA]) may attach (31). The results of the FBA showed that there was no difference in HA binding by fetuin in samples treated with NSPAH-65-89 and samples not treated with NSPAH-65-89, which indicated that NSPAH-65-89 does not inhibit interaction between the viral HA protein and the cellular receptor (data not shown). Surprisingly, similar results were obtained for ι-car, suggesting that it has a different mechanism of action than previously reported (29, 30, 32). In order to confirm these observations, the experiment with the intact virus was carried out (see “Virus attachment assay” in Materials and Methods). Briefly, samples containing polymer solutions and influenza A/H3N2/EVA or control samples (i.e., samples not containing the polymer) were mixed on ice and inoculated onto precooled fully confluent MDCK cells; cultures were incubated at 4°C for 1 h. At this temperature, virions can attach to their receptors on the cell surface, but virus internalization and consequently infection are hampered (25). Subsequently, media containing polymers and unbound virions were removed, the cells were rinsed three times with cold (4°C) sterile 1× PBS, and the cells were incubated at 37°C which enabled internalization of receptor-bound virions and subsequent virus replication. Virus replication was assessed based on appearance of the cytopathic effect and the quantitative RT-PCR analysis of culture supernatants collected on day 2 p.i. No difference in CPE development relative to the control sample was observed, and in quantitative RT-PCR, only a slight decrease (<2 log units) in virus yield was found, and it was disproportionate to the concentrations of the polymers. These observations indicate that the polymers do not interfere with the influenza virus interaction with its receptor (30) (Fig. 6).

FIG 6.

(A and B) Inhibition of influenza A/H3N2/EVA virus attachment to MDCK cells by NSPAH-15-95, NSPAH-56-98, and NSPAH-65-89 (A) and by ι-carrageenan (ι-car) and κ-carrageenan (κ-car) (B). Data are presented as log removal values (LRVs) which are normalized to the values for the control samples (no polymer added). Before normalization, the data for each concentration (expressed as the number of copies per milliliter) were compared to the value for the control sample (also expressed as the copies per milliliter) (*, P < 0.05). All assays were performed in quadruplicate, and average values with standard deviations are presented. The detection limit was ∼1,000 copies/ml.

In order to fully understand the mechanism of action, virus internalization assays were conducted. Briefly, fully confluent MDCK cells were inoculated on ice with 100 μl of ice-cold influenza A/H3N2/EVA (or mock infected) at 400 TCID₅₀ per ml and incubated on ice to block entry of virus particles. Following incubation (4°C, 2 h), media were removed, and cells were rinsed three times with ice-cold PBS to remove unbound influenza A/H3N2/EVA particles. Next, the polymer solutions were applied to the cells at room temperature, and the plate was incubated at 37°C for 2 h in order to enable virus internalization into the cells. The polymer solutions were removed, and cells were briefly rinsed with pH 3.0 buffer solution (0.1 M glycine, 0.1 M sodium chloride, pH value corrected using hydrochloride acid) in order to trigger HA conformational changes under low pH so that the residual viruses will no longer be able to enter the cells. Further, cells were rinsed three times with sterile PBS (pH 7.4), and fresh medium (0% DMEM) was applied. Cells were incubated at 37°C, and CPE was evaluated using an inverted microscope. Mock-treated control samples were managed in the same manner. One may expect that inhibition of virus replication should be observable for compounds blocking endocytosis; indeed, in samples in which we blocked endocytosis (by 4°C incubation), vast inhibition of virus replication was observed. However, relevant inhibition was observed only in the presence of NSPAH-56-98 and NSPAH-65-89 at the highest concentration tested (Fig. 7; see Fig. S8 in the supplemental material).

FIG 7.

(A and B) Inhibition of influenza A/H3N2/EVA virus internalization to MDCK cells by NSPAH-15-95, NSPAH-56-98, and NSPAH-65-89 (A) and by ι-carrageenan (ι-car) and κ-carrageenan (κ-car) (B). Data are presented as log removal values (LRVs) which are normalized to the values for control samples (no polymer added). Before normalization, the data for each concentration (expressed as the number of copies per milliliter) were compared to the value for the control sample (also expressed as the number of copies per milliliter) (*, P < 0.05). All assays were performed in quadruplicate, and average values with standard deviations are presented. The detection limit was ∼1,000 copies/ml.

Finally, the inhibition of replication/release of virions was tested (see “Virus assembly, packing, and releasing assay” and “Virus replication assay” in Materials and Methods). MDCK cells were infected with influenza A/H3N2/EVA at 400 TCID₅₀ per ml or mock infected in the absence of the polymers. Following 2 h of incubation, supernatants were discarded, and cultures were rinsed three times with sterile 1× PBS, and subsequently, fresh media containing (or not containing, for the control samples) polymeric compounds were added. The emergence of CPE was evaluated daily. For this protocol, the inhibitory action of the polymers before or during the virus internalization phase could be excluded and was possible only at the replication/release stage. Visual assessment revealed inhibition of the emergence of CPE in the presence of NSPAH-15-95, NSPAH-56-98, and NSPAH-65-89 at concentrations similar to those effective in cultures where polymers were present during the course of infection. It should be noted that analysis of CPE development for carrageenans at high concentration was not possible due to high viscosity of their solutions. The results of this experiment indicate that the studied polymers inhibit the infection at the replication/release stage.

In order to confirm that the polymers inhibited influenza A/H3N2/EVA virus replication or release, the amount of viral RNA in supernatant and in cells was assessed using quantitative RT-PCR. Despite the fact that virus yield in supernatants was reduced in the presence of polymers, virus replication per se is not affected (Fig. 8 and Table 3; see Fig. S9 in the supplemental material), as the intracellular virus yield is not different from that in nontreated cells, suggesting that virus assembly, and not its replication, is inhibited by the polymer. Consequently, we do not observe the release of new viral particles. These results were confirmed using Amplex red neuraminidase assay in order to compare NA activity in the presence and absence of the tested polymers. The results obtained (Fig. S10) indicated that NSPAH-65-89 and ι-car at a concentration of 1,000 μg/ml decreased the NA activity by only about 20% compared to the value for the control sample, while at lower concentrations, they did not influence NA activity. We also examined whether the aggregation of virus on the virus-infected cells was seen in the presence of NSPAH-65-89 and ι-car by electron microscopic experiment. Transmission electron microscopy (TEM) images show no accumulation of virions inside the host cells or on the cell surface (data not shown). On the basis of these findings, we suggest that the mechanism of antiviral activity of the polymers is based on inhibition of virus assembly.

FIG 8.

(A to D) Inhibition of influenza A/H3N2/EVA virus replication and assembly/release from infected MDCK cells by NSPAH-15-95, NSPAH-56-98, and NSPAH-65-89 (results obtained from cells [A] or supernatant [B]) and by ι-carrageenan (ι-car) and κ-carrageenan (κ-car) (results obtained from cells [C] or supernatant [D]). Data are presented as log removal values (LRVs) which were normalized to the values for control samples (no polymer added). Before normalization, the data for each concentration (expressed as the number of copies per milliliter) were compared to the value for the control sample (also expressed as the number of copies per milliliter) (*, P < 0.05). Values below the detection limit (∼1,000 copies/ml) are indicated by hatched bars. All assays were performed in quadruplicate, and average values with standard deviations are presented.

TABLE 3.

Anti-A/H3N2/EVA activity of the tested polymers after viral infection

Polymer	Antiviral activity after viral infection (μg/ml)
	IC MDCK (CPE)a	IC50 MDCK (qRT-PCR)b
		Supernatant	Cell
NSPAH-15-95	500	0.3 ± 1.8	269.7 ± 1.4
NSPAH-56-98	500	0.1 ± 2.5	153.8 ± 1.7
NSPAH-65-89	500	0.6 ± 1.1	60.3 ± 1.5
ι-Car	5,000c	0.2 ± 1.0c	1.1 ± 1.7
κ-Car	5,000c	7.9 ± 1.1c	4.6 ± 1.2

^aLowest compound concentration for which no CPE was noted.

^bFifty percent inhibitory concentration as assessed by real-time RT-PCR analysis calculated using the formula function in Graph Pad Prism “Dose response-Inhibition: log(inhibitor) vs. response variable slope.”

^cPartial inhibition of CPE was supposed, but carrageenan at this concentration forms a dense gel, so the results may be unreliable.

DISCUSSION

The first anti-influenza compounds developed were M2 ion channel inhibitors (amantadine and rimantadine), which were effective if given during the first phase of the disease. Even though at first they were promising, subsequent studies identified severe adverse reactions related to the drug administration (6). In addition, neither amantadine nor rimantadine can inhibit replication of influenza B virus (7, 9). The emergence of escape mutants further minimized the usefulness of these antiviral substances, and currently, they are not recommended for clinical use (7, 33). The second class of influenza inhibitors encompasses compounds interfering with viral neuraminidase (NA) activity, hampering virus release, and hindering viral colonization of the upper respiratory tract. These compounds include two globally approved compounds (oseltamivir and zanamivir), laninamivir approved in Japan, and peramivir approved in Japan, South Korea, China, and United States (6, 34). However, the antiviral potency of NA inhibitors has been recently questioned and considering that the drug-resistant strains have already emerged (35–38), there is a need for novel and effective anti-influenza drugs. Accordingly, several research groups made an effort to develop novel active compounds targeting different steps of virus replication (39–42).

It has been reported previously that sulfated polysaccharides derived from marine algae, mainly carrageenans (17, 19, 25, 26, 29, 43, 44) but also alginate, fucans, ulvan, and heparin-like structures, e.g., polymannuronic acid propyl sulfate (16, 30), show strong antiviral properties; however, the exact mechanisms of action and structural determinants for these compounds are not fully elucidated. In 1987, Gonzalez et al. (26) hypothesized that natural polysaccharides exert their antiviral activity by direct interaction with virus particles at an early stage of viral infection. Leibbrandt et al. (20) showed that ι-carrageenan possesses antiviral activity not only due to direct interaction with influenza A virus but also due to coating of cellular structures, hindering receptor binding sites. Wang et al. (29) concluded that the high molecular weight prevents carrageenans from penetrating the cell. In their study, they used oligosaccharides obtained by chemical or enzymatic hydrolysis of natural polysaccharides. They reported that the low-molecular-weight carrageenans can effectively inhibit influenza A replication in vitro and weaken lung inflammation in mice infected with IAV. On the other hand, Yamada et al. (45) suggested that the primary mechanism for suppression of viral replication using dextran sulfate is due to blocking progeny virus release from host cells. They speculated that dextran sulfate inhibits NA activity by directly binding to NA, thereby suppressing its ability to recognize sialic acid (45). It is noteworthy that ι-carrageenan has been already marketed by Boehringer Ingelheim in some European countries, as it passed the relevant clinical studies (46–48).

In the current study, new sulfonated poly(allylamine)-based polymers were designed, synthesized, and tested to determine their antiviral potential and compared to the well-known anti-influenza virus activity of different sulfated polysaccharides (carrageenans and chitosan based). The results obtained show that these compounds have strong antiviral activity, which is correlated with their molecular mass and with the degree of substitution with sulfonic groups. Therefore, it can be assumed that long, flexible chains with simple hydrocarbon-based structures and bearing negative charge constitute a desired combination for effective interference with viral infection. Furthermore, cytotoxicity of the polymers decreases with an increase in the degree of polymer substitution with sulfonic groups.

Among the tested compounds, NSPAH-65-89 was the most active, and its antiviral potential was evaluated in vitro (MDCK cells) and ex vivo (fully differentiated human airway epithelium cultures). Visual inspection of the cultures has already revealed that the polymer tested fully blocked development of the cytopathic effect on influenza A virus-infected cultures at 50 μg/ml. Quantitative RT-PCR confirmed these observations and showed a drastic decrease in the number of viral RNA copies in infected cultures. Tested polymers were found to be nontoxic for HAE cultures and MDCK cells at concentrations up to 5 mg/ml. Furthermore, the polymers inhibit replication of different strains of influenza virus type A (H3N2, H1N1) and influenza virus type B.

In order to understand the mechanism of action, we used methods previously described by Harden et al. (25) with some modifications. The aim of these experiments was to identify at which stage of the infection process the polymer blocks virus replication. Because of the structural similarity between NSPAHs and carrageenans, we primarily assumed that the mechanism of action would be similar for NSPAHs and carrageenans and that newly developed compounds would hinder virus-receptor interaction (44). To our surprise, the mechanism of action indeed appeared to be similar, though the inhibition of virus replication by NSPAHs and carrageenans mainly resulted from hampered virus assembly and/or release, and not effectively hindered entry. It is worth noting that mild inhibition of virus entry was observed for both groups of compounds, and some inhibition of virus replication for carrageenans was observed. However, this most likely may be attributed to inhibition of the RNA purification/PCR by the polymers, which was observed for higher polymer concentrations especially for carrageenans (up to 1.5 log units; data not shown).

In conclusion, the N-sulfonated derivatives of high-molecular-mass poly(allylamine) with a high degree of substitution with the sulfonic groups show strong inhibitory effects on replication of influenza A virus in vitro and ex vivo mostly at later stages of the infection. The mechanism of action is similar for NSPAHs and carrageenans and may be attributed to the inhibition of virus release from infected cells. Further, even though NSPAHs show efficacies similar to those of carrageenans, these newly developed anti-influenza compounds present far better physicochemical properties, such as high solubility and lack of gelling properties (18, 49).

Funding Statement

Foundation for Polish Science provided funding to Justyna Ciejka under grant number VENTURES/2013-11/1 cofinanced by the EU European Regional Development Fund. K.P. acknowledges a networking contribution by the COST Action CM1407 (“Challenging organic syntheses inspired by nature - from natural products chemistry to drug discovery”). Parts of this research were carried out with equipment purchased thanks to the financial support of European Union structural funds (grants POIG.02.01.00-12-064/08 and POIG.02.01.00-12-167/08). The Faculty of Biochemistry, Biophysics and Biotechnology of the Jagiellonian University is a partner of the Leading National Research Center supported by the Ministry of Science and Higher Education of the Republic of Poland. The funders had no role in study design, data collection and analysis, the decision to publish, or preparation of the manuscript.