Synthesis and antiviral activity of PB1 component of the influenza A RNA polymerase peptide fragments

Accepted Manuscript Synthesis and antiviral activity of PB1 component of the influenza A RNA polymerase peptide fragments O.V. Matusevich, V.V. Egorov, I.A. Gluzdikov, M.I. Titov, V.V. Zarubaev, A.A. Shtro, A.V. Slita, M.I. Dukov, A.-P.S. Shurygina, T.D. Smirnova, I.V. Kudryavtsev, A.V. Vasin, O.I. Kiselev PII: DOI: Reference:

S0166-3542(14)00308-8 http://dx.doi.org/10.1016/j.antiviral.2014.10.015 AVR 3538

To appear in:

Antiviral Research

Received Date: Revised Date: Accepted Date:

9 July 2014 28 October 2014 29 October 2014

Please cite this article as: Matusevich, O.V., Egorov, V.V., Gluzdikov, I.A., Titov, M.I., Zarubaev, V.V., Shtro, A.A., Slita, A.V., Dukov, M.I., Shurygina, A.-P.S., Smirnova, T.D., Kudryavtsev, I.V., Vasin, A.V., Kiselev, O.I., Synthesis and antiviral activity of PB1 component of the influenza A RNA polymerase peptide fragments, Antiviral Research (2014), doi: http://dx.doi.org/10.1016/j.antiviral.2014.10.015

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Synthesis and antiviral activity of PB1 component of the influenza A RNA polymerase peptide fragments. O.V. Matusevicha, * V.V. Egorovb,d, I.A. Gluzdikova, M.I. Titova, V.V. Zarubaevb, A.A. Shtro b, A.V. Slita b, M.I. Dukovb, A.-P.S. Shuryginab, T.D. Smirnovab, I.V. Kudryavtsev c,e A.V. Vasinb,f, O.I. Kiselevb a

Saint Petersburg State University, Faculty of Chemistry, 26, Universitetskii pr., Petrodvorets, St.

Petersburg, Russia, 198504 b

Research Institute of Influenza, Molecular Virology Department, 15/17, Prof. Popova str., St.

Petersburg, Russia, 197376 c

Institute of Experimental Medicine of the North-West Branch of the Russian Academy of

Medical Sciences (IEM NWB RAMS), 12, Akad. Pavlova str., St. Petersburg, Russia, 197376 d

Petersburg Nuclear Physics Institute, NRC KI, Orlova roscha, Gatchina, Russia, 188300

e

Far East Federal University, 8, Suhanova str., Vladivostok, Russia, 690950

f

State Polytechnical University, 29, Polytecnicheskaya str., St. Petersburg, Russia, 195251

*Corresponding author, [email protected]

Highlights - Peptides corresponding to portions of the amino acid sequence of the PB1 protein was selected and synthesized - Terminal modifications (acetylation and amidation) were shown to increase the activity of the peptides significantly -The most active peptide is able to cross the cell membrane -Mechanism of its action did not involve the inactivation of the virus outside of the cell - Inhibition of influenza virus by the peptide took place during the early stages of viral reproduction

Abstract This study is devoted to the antiviral activity of peptide fragments from the PB1 protein – a component of the influenza A RNA polymerase. The antiviral activity of the peptides synthesized was studied in MDCK cell cultures against the pandemic influenza strain A/California/07/2009 (H1N1) pdm09. We found that peptide fragments 6-13, 6-14, 26-30, 395400, and 531-540 of the PB1 protein were capable of suppressing viral replication in cell culture. Terminal modifications i.e. N-acetylation and C-amidation increased the antiviral properties of the peptides significantly. Peptide PB1 (6 – 14) with both termini modified showed maximum antiviral activity, its inhibitory activity manifesting itself during the early stages of viral replication. It was also shown that the fluorescent-labeled analog of this peptide was able to penetrate into the cell. The broad range of virus-inhibiting activity of PB1 (6-14) peptide was confirmed using a panel of influenza A viruses of H1, H3 and H5 subtypes including those resistant to oseltamivir, the leading drug in anti-influenza therapy. Thus, short peptide fragments of the PB1 protein could serve as leads for future development of influenza prevention and/or treatment agents.

influenza A; antiviral peptides; influenza A polymerase; PB1 1. Introduction. Influenza A virus has a segmented RNA-genome consisting of 8 RNA minus-strands encoding 16 proteins that can serve as drug targets: hemagglutinin (HA), neuraminidase (NA),

matrix protein 1 (M1), proton channel protein (M2), M42, nucleoprotein (NP), nonstructural protein 1 (NS1), nuclear export protein (NEP or NS2), proteins PB1-F2 and PB1-F3 (N40), PAX, PA-N155 and PA-N182 as well as three proteins that make up the RNA-polymerase – PB1, PB2, and PA (Vasin et al., 2014). Despite the fact that influenza virus life-cycle contains many potential drug targets for directed therapy, few have been used to date. Currently, there is a limited choice of drugs suitable for influenza treatment and prevention. High variability of the influenza virus and the emerging resistance to existing drugs make it necessary to develop new means of influenza treatment and prevention (Du et al., 2012) . Influenza prevention and treatment is based on vaccines, as well as agents acting directly on viral proteins. The latter includes 4 FDA-approved drugs: the adamantan derivatives (amantadine and rimantadine) as well as the neuraminidase inhibitors – oseltamivir (Tamiflu), zanamivir (Relenza) (FDA ) . Despite its wide spectrum of antiviral activity, ribavirin is used to treat influenza only in patients with life-threatening conditions due to a large number of side effects (Crotty et al., 2002) . Viral polymerase inhibitor T-705 (favipavir) is currently in phase III clinical trials (Furuta et al., 2013) .

RNA polymerase, the heterotrimeric complex which catalyzes viral RNA synthesis, is a promising drug target that has not yet been used as such in a clinical setting. The proteins PB1, PB2, and PA, which are parts of the influenza polymerase complex, are conservative among all influenza virus types and subtypes. The PB1 protein is a key component of the polymerase complex since it contains the active site of the polymerase as well as two domains that interact with PB2 and PA, respectively, and are important for the assembly of a functional complex (Torreira et al., 2007) . It was previously shown that the part of the PB1 protein crucial to binding to the PA protein is the 1-15 region. Peptides containing sequence fragments 1-15 and 1-25 of the PB1 protein suppress viral replication in cells. It was suggested that this peptide inhibitor competes with the full-size PB1 protein for binding sites on the PA protein (He et al., 2008) . It is of great practical value to find out which part of the N-terminal PB1 sequence is essential for antiviral activity and if other parts of the PB1 sequence exhibit antiviral properties when synthesized as peptide fragments.

Our goal was to find peptides corresponding to parts of PB1 amino acid sequence that efficiently inhibit influenza A virus replication in cell culture and to evaluate the effect of terminal modifications of the most active peptides on their virus-inhibiting activity.

2. Materials and methods 2.2.Peptide synthesis. Peptides were synthesized manually by solid-phase method on Rink amide resin and 2chlorotrityl chloride resin by Fmoc/t-Bu strategy using both stepwise elongation and convergent synthesis approaches. Reaction efficiency was monitored using the Kaiser test for free amino acids (Kaiser et al., 1970) . The following permanent protecting groups were used: tert-butyloxycarbonyl (BOC) for lysine ε-aminogroup; 2,2,4,6,7-pentamethyldihydrobenzofurane-5-sulfonyl (Pbf) for arginine guanidine group; triphenylmethyl (Trt) for glutamine and asparagine amide groups; tert-butyl (tBu) for threonine, tyrosine, and serine OH-groups; (O-tBu) for aspartic and glutamic acids β- and γcarboxygroups. Peptides were purified by preparative RP-HPLC; their structure was confirmed using mass spectrometry (HRMS(ESI+)). All of the peptides studied and shown in table 1, except PB1(111-130) and PB1(381-400), were synthesized by stepwise chain elongation. Peptides PB1(111-130) and PB1(381-400) were synthesized by convergent synthesis. Peptide PB1(1-25) was synthesized using both strategies combined. The synthesis of the peptides listed in table 1 is covered in more detail in our earlier publication (Matusevich et al., 2011) . The amide of peptide 6-13 (compound 3a) was obtained by amidation using the ammonium salt of hydroxybenzotriazole and diisopropylcarbodiimide: Boc-T1-L2-L3-F4-L5-K6(Boc)-V7-P8-OH 1) DIC/NH4OBt 2) TFA/TIS/H2O H-T1-L2-L3-F4-L5-K6-V7-P8-NH2 Compounds 3c, 4a, 4c, 4d, 7a, 7c, 14a, 14c, 18a, and 18c were all synthesized on Rink amide resin.

2.3.Viruses and cells. Influenza A viruses A/California/07/09(H1N1)pdm09, A/Puerto Rico/8/34 (H1N1), A/Aichi/2/68 (H3N2),

A/Mallard/Pennsylvania/10218/84

(H5N2)

and

oseltamivir-resistant

strain

A/Vladivostok/02/09 (H1N1) previously shown to carry H275Y mutation of oseltamivir resistance were used. The viruses were propagated in the allantoic cavity of 10-12-day-old chicken embryos for 48 hours at 37˚С. Cell culture experiments were performed in MDCK cells (ATCC CCL-34) grown in 96-well plates in alpha-MEM medium with 10% fetal bovine serum. 2.4.Virus titration. A series of 10-fold dilutions (10-1 – 10-7) was prepared from the initial virus-containing sample using MEM medium. 100 µl of each dilution was added into the wells of a 96-well plate and left to incubate for 2 days at 37˚С in a 5% CO2 atmosphere. After 48 hours, aliquots of culture medium was taken from each well and transferred to a microwell plates for hemagglutination assay. 2.5.Hemagglutination assay. To perform the assay, 100 µl of culture fluid containing the virus was taken from the wells of the cell culture plate and added into the wells of a microwell plate. After that equal amounts of 1% chicken red blood cell suspension in normal saline were added into the wells of a microwell plate. The results were evaluated after 30 min incubation at room temperature. The infectious titer of the virus was determined as the maximum dilution that caused complete agglutination of red blood cells. Viral titers were calculated as 50% tissue infectious dose (log10TCID50) per ml using Reed-Muench method (Reed and Muench, 1938). Assessing the toxicity of the peptides in vitro. In order to assess toxicity in MDCK cells, the compounds studied were diluted two-fold serially with concentrations ranging from 500 to 3.9 µg/ml in alpha-MEM maintenance medium. Aliquots of diluted compounds were added into the wells of a 96-well plate. Cells were incubated for 48 hours at 37˚С in a 5% CO2 atmosphere. After that, a microtetrazolium (MTT) assay was performed. Cells were washed twice with normal saline (0.9% NaCl), following which 100 µl per well of MTT (3-(4,5-dimethylthiazol)-2,5-diphenyltetrazolium bromide) was added at a concentration of 0.5 µg/ml in normal saline. Plates were incubated for 1 hour at 37 ˚С, culture medium was then removed, and 0.1 ml of dimethylsulfoxide was added into each well. Optical

density was measured using a spectrophotometer Victor 2 1440 at 535 nm. The results obtained were used to calculate the CTD50 (50% Cytotoxicity Dose), i.e., the concentration of the compound that kills 50% of the cultured cells. In CTD50 measurements during range of virus-inhibiting activity definition experiments the maximal compound 4 concentration used was 500 µM. 2.6.Determining the antiviral activity of the peptides in vitro. A confluent monolayer of MDCK cells was infected as follows: 100 µl of the peptide dissolved in maintenance medium was added into each well of a 96-well plate then incubated for 1 hour at 37˚С in a 5% CO2 atmosphere. After that, 100 µl per well from each of the series of 10-fold dilutions (10-1 – 10 -7) of the virus was added and the cells were left to incubate for 2 days at 37˚С in a 5% CO2 atmosphere. After 48 hours of incubation, 100 µl of culture fluid was taken from each well and added into the wells of a microwell plate to perform a hemagglutination assay. The decrease in viral infectious titer was used to assess the antiviral activity of the compounds tested. On the basis of the data obtained, for each compound it was calculated the concentration that caused the viral titer to drop two-fold (by 0.3 log10TCID50), after which the selectivity index (SI) was calculated as the ratio of CTD50 to ED50 (50% Effective Dose). 2.7.Virucidal activity. The compound at maximum non-toxic dilution was incubated with influenza virus A/California/07/2009 (H1N1) pdm09 diluted 10-fold - for one hour at 37˚С. After that, a series of 10-fold dilutions of the virus were prepared (ranging from 10 -1 to 10-6) and used to infect MDCK cells, followed by incubation for 48 hours at 37˚С in a 5% CO2 atmosphere. The infectivity of the virus was assessed in a hemagglutination assay with chicken red blood cells according to standard method. The decrease in viral titer as compared to control was used to assess the virucidal activity of the compound. 2.8.Time-of-addition experiments. MCDK cells were grown for 24 hours in alpha-MEM medium with 10% fetal bovine serum, were washed with normal saline, and then fresh medium containing trypsin was added (1 µg/ml). Compound tested was added at its maximum non-toxic concentration at the following time points: -2 to -1 hours (before the adsorption of the virus), -1 to 0 hours (adsorption) as well as at several time points post adsorption: 0-1, 0-2, and 4-10 hours. After 10 hours of incubation, the

cells were scraped off and a series of 10-fold dilutions was prepared from the resulting suspension which was then seeded into the MDCK cell culture and incubated for 48 hours at 37˚С in a 5% CO2 atmosphere. The infectivity of the virus was assessed by hemagglutination assay. 2.9.Flow cytometry. Into the wells of 6-well plates with MDCK cell monolayer 100 µl per well of the peptide dissolved in the maintenance medium (50µg/ml) and incubated at 37˚С in a 5% CO2 atmosphere. Maintenance medium or maintenance medium containing DMSO (3%) 100 µl per well were used as negative controls. Following incubation, the cells were detached from the plate surface with an accutase solution (Sigma), washed twice with phosphate-buffered saline, and then resuspended (0.5×106 cells/ml) in normal saline containing 2% fetal bovine serum. Before the analysis was performed, the cells were additionally stained with 7-AAD (Invitrogen) in order to assess cell viability (dead cells – 7-AAD positive cells – were also excluded from the analysis). The samples were analyzed using a Navios™ (Beckman Coulter, USA) flow cytometer equipped with two diode lasers with 488 nm and 635 nm excitation wavelengths. To correctly exclude from the analysis the cells whose shape and size did not match those of intact MDCK cells, necessary logical restriction were applied to diagrams depicting distribution of particles by forward scatter and side scatter. Not less than 10000 MDCK cells were analyzed in each sample. Mathematical processing of the flow cytometry data was performed using Navios Software v.1.2 and Kaluza™ v.1.2 (Beckman Coulter, USA) software. 2.10.

Fluorescence microscopy.

Efficiency in penetration of fluorescently-labeled compound 4d through cell membrane was assessed in MDCK cell culture grown to a 100% monolayer in a 96-well polystyrene plate (“Nunc”, Denmark). Growth medium was removed from the plate, after which 100 µl per well of the solution containing the labeled peptide was added (30 µg/ml).

The plates were then

incubated at 37˚С in a 5% CO2 atmosphere. At 60 and 120 min the cells were washed to remove any remaining peptide and observed under an inverted luminescence microscope (“Leica DM IL”, Austria). Photographs were taken using a digital camera (“Leica DFC 290”, Austria).

3. Results 3.1.Selection of peptides for synthesis.

Search was restricted to the following PB1 amino acid sequence regions (strain A/Hong Kong/156/97(H5N1)): 1-30, 111-130, 271-290, 381-420, 525-540. These highly conservative regions differ only in a single non-equivalent substitution in the highly pathogenic H5N1 strain and the most recent strain A/California/07/09(H1N1)pdm09.

As a result the following peptides were chosen for synthesis: PB1 protein fragments 1-5, 1-25, 6-13, 6-14, 6-25, 14-25, 26-30, 111-130, 271-290, 381-386, 381-390, 381-400, 391-400, 395-400, 411-420, 525-530, 525-535, and 531-540. Their amino acid sequences and antiviral activities are listed in Table 1 (for antiviral activity see comments in the text below). 3.2.Antiviral activity of the peptides. Antiviral activity of the peptides was assessed in MDCK cell culture against the influenza strain A/California/07/2009 (H1N1)pdm09/. During the experiments 50% cytotoxic dose and effective dose were assessed, using which the selectivity index (SI) was calculated. Peptides with SI >10 were considered active. First, unmodified peptides #1-18 (Table 1) were tested. Two drugs were chosen as reference compounds: rimantadine, to which the strain studied is resistant, and oseltamivir, to which the strain is sensitive. As can be concluded from the data in Table 1, all the peptides synthesized were non-toxic to the cells. Most active against the influenza virus were РВ1 protein fragments 6-13 (compound 3), 6-14 (compound 4), 26-30 (compound 7), 395-400 (compound 14), and 531-540 (compound 18). In order to increase stability and bioavailability, modified analogs of the most active peptides were synthesized: N-terminal acetylation and C-terminal amidation were used. These modifications eliminate terminal charge, thus enabling the peptides to better mimic the corresponding region in the full-size protein (unless the region itself is terminal), increases resistance towards exopeptidases and facilitates penetration into the cell. The test results are presented in Table 2. As is evident from the data in the table, in all cases except for peptide 395-400 and its modified analogs (peptides 14 a, b, c, see Table 2), C-terminally amidated derivatives were more active than free carboxylic derivatives, whereas introduction of the N-terminal acetyl group into C-terminally amidated peptides led to an increase in activity only for fragments 6-13 and 6-14 of

PB1 (compounds 3 and 4). Compound 4c, – a derivative of peptide PB1 6-14, - was the most active with SI > 111. 3.3.Virucidal activity of the peptides. To elucidate the mechanism of antiviral activity of compound 4c, another series of experiments was carried out studying its virucidal properties and cellular localization. Study of virucidal activity revealed no decrease of infectious titer of the virus after incubation with peptide in cell-free system (data not shown). This proves that the compound has no virucidal activity and this mechanism should not be considered for compound 4c.

3.4.Time-of-addition experiments. An experiment was also conducted to study the effect of compound 4c upon various stages of the viral life cycle. For this purpose, the peptide was added to the MDCK cell culture before, simultaneously with, and at various time points after infecting the cells with the virus. After adding the virus, cells were incubated for one hour at 4˚С. In these conditions the virus attaches to the cellular receptors but does not enter the cell. After one hour cells were incubated at 37˚С. The time point when the virus penetrates into the cell was the starting point (designated by 0). The results of evaluation of virus titer are summarized in Fig. 2. As follows from the data presented, maximum decrease in virus titer was observed when the compound was present in the system during the entire experiment. In addition, slight decrease in virus titer was observed at time points (-1)–0 and 0–1 hours . Maximal virus inhibition by compound 4c was achieved at time points 0-2 and 2-4 hours post infection.

3.5.Cellular localization of peptide 4. In order to study the cell-penetrating ability of compound 4c, amidated peptide 4 (compound 4a) was labeled with FITC through the 6-aminohexanoic acid linker. The antiviral activity of the resulting compound (4d, Fig. 3) was compared to that of compound 4c. The fluorescent labeled peptide had a similar activity (CTD50 > > 332,7 µМ, EC50=3,33 µМ SI>100) as compound 4c, which suggests similar mechanisms of action for both compounds 4c and 4d. The results of studying the localization of fluorescently labeled compound 4d in MDCK cells using flow cytometry and fluorescent microscopy indicate that the compound studied

penetrates the cell membrane. Using flow cytometry we showed that this peptide is capable of binding to living MDCK cells (Fig. 4a). Study of stained cells using fluorescent microscopy revealed that staining is due to the peptide penetrating the cell membrane (Fig. 4b).

3.6. Range of virus-inhibiting activity of compound 4. In order to test if the virus-inhibiting activity of compound 4 which has been demonstrated against A/California/07/09 virus is also true for other influenza A viruses, we tested its activity against influenza viruses A/Puerto Rico/8/34 (H1N1) (ED50±SD=8.9±1.0 µM; SI=56), A/Aichi/2/68 (H3N2) (ED50±SD=10.7±0.9 µM; SI=46), A/Mallard/Pennsylvania/10218/84 (H5N2) (ED50±SD=8.5±1.1 µM; SI=58) and oseltamivir-resistant strain A/Vladivostok/02/09 of H1N1 subtype (ED50±SD=1.7±0.8µM; SI=294). The results obtained suggest that the compound 4 is also able to suppress the reproduction different strains of influenza A viruses including that of oseltamivir-resistant phenotype.

4. Discussion. Previous studies using linear peptides directed against the assembly of influenza virus polymerase complexes have shown that this type of viral replication interference is effective and prospective for further development. Wunderlich et al. (Wunderlich et al., 2011) showed that targeting the protein-protein interaction domain between PA and PB1 using peptides derived from the PA N-terminus resulted in strong inhibition in polymerase activity, although no tests have been performed in cell culture. Nicol et al. have demonstrated the virus-inhibiting activity of anti- hemagglutinin peptides both in vitro and in vivo (Nicol et al., 2012) . Nevertheless, the problem of bioavailability remains one of the most important issues both for peptides and low-molecular compounds. In this regard, although active in an enzyme cell-free system, numerous nucleotide analogs lost their virus-inhibiting activity when tested in cell culture (Jochmans, 2008) . Notably, in both cases of study anti-influenza peptides against polymerase complexes in cell culture, the peptides were used indirectly. In particular, Wunderlich et al. conjugated the PB1 1-25 peptide with additional transport membranotropic peptide derived from the Tat protein of human immunodeficiency virus (Wunderlich et al.,, 2009) . Similarly, to increase the solubility of peptides, additional sequences of arginine and lysine (RRRKK) were attached to the peptide by Nicol et al. . Without this sequence, the virusinhibiting properties appeared three orders lower than in its presence (Nicol et al., 2012) .

Ghanem et al. transfected the cells with the peptide-coding plasmid, i.e. the peptide itself was synthesized within the cell thus avoiding the problem of membrane permeability (Ghanem et al., 2007) . One of the factors governing our choice of peptides was the ability of PB1 peptide fragments 1-15 and 1-25 to suppress viral replication. The aim of this study was thus to assess the antiviral activity of smaller N-terminal fragments and to pinpoint more precisely the minimal sequence crucial for antiviral properties. When selecting peptides outside the N-terminus of PB1, various factors were taken into account such as hydrophobicity, solubility, cell-penetrating ability as well as tandem clusters of identical amino acid residues. For instance, the 531-540 fragment of the PB1 protein (compound 18) was selected because it contains a cluster of asparagine residues. Asparagine and glutamine clusters are known to increase the propensity of proteins to form amyloid aggregates (Gasset et al., 1992) , in which β-structural elements predominate. In light of the above it was hypothesized that the PB1 531-540 peptide can interfere with β-structure formation and compromise the stability of the corresponding domain in the full-size protein. Similar reasoning is behind selecting the 111-130 region, which contains a glutamine repeat and a valine repeat. The experiment results showed that compounds 3, 4, 7, 14, and 18 had high SI against influenza virus. Interestingly, compounds 3 and 4 only differ in a single position but the SI of compound 4 is almost three times greater. This difference could be the result of a greater percentage of peptide 4 penetrating into the cell compared to peptide 3 due to a hydrophobic alanine residue in peptide 4. Additionaly, an alanine residue in position 14 in the PB1 protein interacts with glutamine-670 residue in the PA protein (Liu et al., 2009) . We suppose that the antiviral activity of compound 4c – N-acetylated amide of the 6-14 fragment of PB1 is based on interfering with the assembly of the RNA polymerase complex either by peptide 4b binding to the PA protein or by stabilizing the β-conformation of the PB1 Nterminus (Egorov et al., 2013) . We also used simple terminal modifications (amidation and acetylation) of antipolymerase peptides leading to high level of membrane penetration and, therefore, high bioavailability and high efficacy. This approach could become useful for further development of novel peptide-based antivirals as it does not require additional amino acid residues and is not complicated in terms of synthetic procedure.

Time-of-addition experiments showed that the maximal virus inhibition by compound 4c was achieved at time points 0-2 and 2-4 hours post infection, i.e. during early stages of viral life cycle. This suggests that compound 4c affects the early stages of the viral life cycle. In particular, these stages include the synthesis of polymerase subunits and their subsequent assembly into functionally active complex. The peptide we tested could interfere with these processes thus inhibiting virus replication. At the same time, slight decrease in viral titer during stage (-1)–0 hours suggests that compound 4c affect the cells and/or is able to prevent virus’ attachment to the plasma membrane. This could be explained by either a specific mechanism (amino acid similarity to cellular receptors) or non-specific binding of 4c to the membrane. Further studies are therefore needed to elucidate anti- influenza activity of this peptide. In summary, this study included the selecting and synthesizing of peptides corresponding to portions of the amino acid sequence of the PB1 protein, which is a subunit of the influenza A virus RNA polymerase complex, and the determining the antiviral activity of those peptides in cell culture. Terminal modifications (acetylation and amidation) were shown to increase the activity of the peptides significantly. Furthermore, it was shown that the most active peptide was able to cross the cell membrane and that the mechanism of its action did not involve the inactivation of the virus outside of the cell. Inhibition of influenza virus took place during the early stages of viral reproduction. The broad range of virus-inhibiting activity of compound 4 was confirmed using a panel of influenza A viruses of H1, H3 and H5 subtypes including those resistant to oseltamivir, the leading drug in anti-influenza therapy. This compound and its modifications is thus a promising lead for the development of drugs for influenza treatment and prevention.

Acknowledgments The authors wish to thank Dr. D. V. Lebedev (DMRB PNPI NIC KI) for the fruitful discussion of the experiments and Brian Hoettels (SPbSU, University of Wisconsin-Madison) for his help in the manuscript preparation.

Conflict of interests statement The authors declare that there is no conflict of interests regarding the publication of this paper.

Figure and table legends Figure 1. PB1 protein amino acid sequence of the influenza strain A/Hong Kong/156/97(H5N1). Regions chosen for peptide search are highlighted in grey. Table 1. Antiviral activity of unmodified peptides Table 2. Antiviral activity of modified peptides Figure 2. The effect of peptide 4c upon the A/California/07/2009 (H1N1)pdm09/ titer in time-ofaddition experiments. Compound 4c was added at its maximum non-toxic concentration (200 mkg/ml) to the cell-virus system at the following time points: -2 to -1 hours (before the adsorption of the virus), -1 to 0 hours (adsorption) as well as at several time points post adsorption: 0-1, 0-2, and 4-10 hours. Virus titer was determined in supernatants collected after 10 hours of incubation in MDCK cells. Figure 3. Chemical structure of compound 4d. Figure 4. (Upper) – distribution of cells by FITC fluorescence intensity after incubation with compound 4d; flow cytometry results obtained with MDCK cells without compound 4d (white peak), with DMSO (grey peak) or with compound 4d in DMSO (black peak). X-axis is fluorescence intensity of FITC-labeled compound 4d, y-axis is cell count. (Lower) – fluorescence microscopy (up) and light microscopy (down) of MDCK cells after incubation with compound 4d .

References Van der Geer, J., Hanraads, J.A.J., Lupton, R.A., 2010. The art of writing a scientific article. J. Sci. Commun. 163, 51–59.

1. Crotty, S., Cameron, C., Andino, R., 2002. Ribavirin's antiviral mechanism of action: lethal mutagenesis? J. Mol. Med. (Berl). 80(2), 86-95. 2. Du, J., Cross, T.A., Zhou H.X., 2012. Recent progress in structure-based anti-influenza drug design. Drug Discov. Today. (19-20):1111-20 3. Egorov, V., Matusevich, O., Shaldzhyan, A., Skvortsov, A., Zabrodskaya, Y., Garmay, Y., Landa, S., Lebedev, D., Zarubayev, V., Sirotkin, A., Vasin, A., Kiselev, O., Structural Features of the Peptide Homologous to 6-25 Fragment of Influenza A PB1 Protein, 2013, Int. J. of Pept., 2013, Article ID 370832. 4. FDA: http://www.fda.gov/Drugs/DrugSafety/InformationbyDrugClass/ucm100228.htm 5. Furuta, Y., Gowen, B.B., Takahashi, K., Shiraki, K., Smee, D.F., Barnard, D.L., 2013. Favipiravir (T-705), a novel viral RNA polymerase inhibitor. Antiviral Res. 100(2):446-54. 6. Gasset, M., Baldwin, M. A., Lloyd, D. H., Gabriel, J. M., Holtzman, D. M., Cohen, F., Fletterick, R., Prusiner, S. B., 1992, Predicted alpha-helical regions of the prion protein when synthesized as peptides form amyloid. PNAS 89 (22):10940-10944. 7. Ghanem, A., Mayer, D., Chase, G., Tegge, W., Frank, R., Kochs, G., García-Sastre, A., Schwemmle, M., 2007. Peptide-mediated interference with influenza A virus polymerase. J Virol. 81(14):7801-4 8. He, X., J. Zhou, Bartlam, M. et al., 2008, Crystal structure of the polymerase PAC-PB1N complex from an avian influenza H5N1 virus, Nature, 454(7208):1123–1126. 9. Jochmans, D., 2008, Novel HIV-1 reverse transcriptase inhibitors. Virus Res. 134(1-2):17185. 10. Kaiser, E., Colescott, R.L., Bossinger, C.D., Cook, P.I., 1970. Color test for detection of free terminal amino groups in the solid-phase synthesis of peptides. Anal. Biochem. 34: 595–598

11. Liu, Y., Lou, Z., Bartlam, M., Rao, Z., 2009. Structure-function studies of the influenza virus RNA polymerase PA subunit. Sci. China. C Life Sci. 52(5):450-8. 12. Matusevich, O.V., Gluzdikov, I.A.; Titov, M., 2011. Synthesis of fragments PB1 subunit of RNA polymerase of influenza virus A. Vest. SPb Univer. 4 (2):150. 13. Nicol, M.Q., Ligertwood, Y., Bacon, M.N., Dutia, B.M., Nash, A.A., 2012, A novel family of peptides with potent activity against influenza A viruses. J. Gen. Virol. 93(5):980-6. 14. Reed, L,J., Muench, H., 1938, A simple method of estimating fifty per cent endpoints. Amer. J. Hyg. 27(3): 493-497. 15. Torreira, E., Schoehn, G., Fernández, Y. et al., 2007, Three-dimensional model for the isolated recombinant influenza virus polymerase heterotrimer. NAR, 35 (11):3774–3783. 16. Vasin, A.V., Temkina, O.A., Egorov, V.V., Klotchenko, S.A., Plotnikova, M.A., Kiselev, O.I., 2014. Molecular mechanisms enhancing the proteome of influenza A viruses: An overview of recently discovered proteins. Virus Res. 185C:53-63. 17. Wunderlich, K., Juozapaitis, M., Ranadheera, C., Kessler, U., Martin, A., Eisel, J., Beutling, U,, Frank, R., Schwemmle, M., 2011, Identification of high-affinity PB1-derived peptides with enhanced affinity to the PA protein of influenza A virus polymerase. Antimicrob Agents Chemother 55(2):696-702. 18. Wunderlich, K., Mayer, D., Ranadheera, C., Holle,r A.S., Mänz, B., Martin, A., Chase, G., Tegge, W., Frank, R., Kessler, U., Schwemmle, M., 2009, Identification of a PA-binding peptide with inhibitory activity against influenza A and B virus replication. PLoS One. 4(10):e7517

Table 1. Antiviral activity of unmodified peptides. №

Peptide

Amino acid sequence

1

PB1 (1-5)

MDVNP

2

PB1 (1-25)

MDVNPTLLFL KVPAQNAIST TFPYT

3

PB1 (6-13)

TLLFLKVP

4

PB1 (6-14)

TLLFLKVPA

5

PB1 (6-25)

TLLFLKVPAQ NAISTTFPYT

6

PB1 (14-25)

AQNAISTTFPYT

7

PB1 (26-30)

GDPPY

8

PB1(111-130)

MEVVQQTRMD KLTQGRQTYD

9

PB1(271-290)

LPVGGNEKKA KLANVVRKMM

10

PB1(381-386)

FNESTR

11

PB1(381-390)

FNESTRKKIE

12

PB1(381-400)

FNESTRKKIE KIRPLLVEGT

13

PB1(391-400)

KIRPLLVEGT

14

PB1(395-400)

LLVEGT

15

PB1(411-420)

MFNMLSTVLG

16

PB1(525-530)

IGVTVI

17

PB1(525-535)

IGVTVIKNNMI

18

PB1(531-540)

KNNMINNDLG

Reference compounds rimantadine oseltamivir carboxylate

M r, g/mo l

574.6 5 2782. 26 930.2 0 1001. 28 2225. 62 1313. 42 547.5 7 2427. 75 2183. 69 752.7 7 1251. 39 2358. 75 1125. 38 630.7 4 1112. 38 600.7 5 1201. 48 1132. 26

Purity

Activity indices CTD50, μM, ±SD, at ED50, μM ±SD, at least 3 independent least 3 experiments independent experiments

SI

95.1

1300±10

185±2

7.0

96.3

360±5

122±3

2.9

98.6

1075±10

102±3

10.5

98.8

999±8

34±2

29.8

97.5

449±4

129±2

3.5

98.1

761±4

380±2

2.0

96.4

1828±15

115±3

15.9

94.5

206±3

58±2

3.6

97.2

274±3

33±1

8.3

95.7

1329±10

664±3

2.0

95.2

400±8

400±8

1.0

96.3

140±2

23±1

6.1

96.5

622±8

71±3

8.7

96.9

1585±10

91±2

17.4

94.9

899±8

369±3

2.4

95.3

1667±10

833±6

2.0

94.7

416±3

416±3

1.0

94.5

883±4

51±3

17.2

284±3

57±1

5

1000±10

1±0,5

1000

Table 2. Antiviral activity of modified peptides №

Peptide

Amino acid sequence

M r, g/mol

Purity

Δ

Activity indices CTD50, μM, ±SD, at ED50, μM, least 3 independent ±SD, at least 3 experiments independent experiments

3

PB1 (6-13)

Δ SI *

SI

H-TLLFLKVP-OH

930.20

98.6

1075±10

102±3

10.5

-

3а

H-TLLFLKVP-NH2

929.20

93.5

1076±10

35±3

31.2

+20.7

3b

Ac-TLLFLKVP-OH

972.24

94.8

1029±10

33±2

31.2

+20.7

3c

Ac-TLLFLKVP-NH2

971.26

97.2

513±5

12±2

43.1

+32.6

H-TLLFLKVPA-OH

1001.28

98.8

999±8

34±2

29.8

-

4а

H-TLLFLKVPA-NH2

1000.30

97.3

500±5

6±1

83.3

+53.5

4b

Ac-TLLFLKVPA-OH

1043.32

93.6

958±10

82±3

11.8

-18

4c

Ac-TLLFLKVPA-NH2

1042.33

95.2

477±5

4±1

111

+81.2

H-GDPPY-OH

547.57

96.4

1828±15

115±3

15.9

-

7а

H-GDPPY-NH2

546.58

97.7

1830±15

100±5

18.2

+2.3

7b

Ac-GDPPY-OH

589.60

98.1

1696±15

95±3

17.9

+2

7c

Ac-GDPPY-NH2

588.62

96.2

849±8

254±3

3.3

H-LLVEGT-OH

630.74

96.9

1585±10

91±2

17.4

14а

H-LLVEGT-NH2

629.76

98.7

794±4

794±4

1

-16.4

14b

Ac-LLVEGT-OH

672.78

98.0

743±5

297±3

2.5

-14.9

14c

Ac-LLVEGT-NH2

671.79

97.3

744±4

744±4

1

-16.4

H-KNNMINNDLG-OH

1132.26

94.5

883±4

51±3

17.2

-

18а

H-KNNMINNDLG-NH2

1131.28

95.4

884±5

44±1

20

+2.8

18b

Ac-KNNMINNDLG-OH

93.5

852±5

71±3

12.0

-5.2

18c

Ac-KNNMINNDLG-NH2

93.7

852±5

109±4

7.8

-9.4

284±3

57±1

5

4

PB1 (6-14)

7

PB1 (26-30)

14

PB1(395-400)

18

PB1(531-540)

1174.30 1173.32

Reference compounds rimantadine

-12.6 -

oseltamivir carboxylate

*- Difference in SI between the modified peptide and its unmodified analog.

1000±10

1±0,5

1000

Figure 1

Figure 2

Figure 3

Figure 4b

Figure 4a