A Short Chemically-Modified dsRNA-Binding PNA (dbPNA) Inhibits

Subscriber access provided by EKU Libraries

Article

A Short Chemically-Modified dsRNA-Binding PNA (dbPNA) Inhibits Influenza Viral Replication by Targeting Viral RNA Panhandle Structure Julita Kesy, KIRAN M PATIL, Subaschandrabose Rajesh Kumar, Zhiyu Shu, Hui Yee Yong, Louis Zimmermann, Alan Ann Lerk Ong, Desiree-Faye Kaixin Toh, Manchugondanahalli S. Krishna, Lixia Yang, Jean-Luc Decout, Dahai Luo, Mookkan Prabakaran, Gang Chen, and Elzbieta Kierzek Bioconjugate Chem., Just Accepted Manuscript • DOI: 10.1021/acs.bioconjchem.9b00039 • Publication Date (Web): 05 Feb 2019 Downloaded from http://pubs.acs.org on February 6, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Bioconjugate Chemistry

A Short Chemically-Modified dsRNA-Binding PNA (dbPNA) Inhibits Influenza Viral Replication by Targeting Viral RNA Panhandle Structure

Julita Kesy,1,† Kiran M. Patil,2,† Subaschandrabose Rajesh Kumar,3,† Zhiyu Shu,2 Hui Yee Yong,4,5,6 Louis Zimmermann,7 Alan Ann Lerk Ong,2 Desiree-Faye Kaixin Toh,2 Manchugondanahalli S. Krishna,2 Lixia Yang,2 Jean-Luc Decout,7 Dahai Luo,4,6 Mookkan Prabakaran,3,* Gang Chen,2,* Elzbieta Kierzek1,*

1Institute

of Bioorganic Chemistry, Polish Academy of Sciences, Noskowskiego 12/14, 61-704

Poznan, Poland. 2Division of Chemistry and Biological Chemistry, School of Physical and Mathematical Sciences, Nanyang Technological University, 21 Nanyang Link, Singapore 637371. 3Temasek Life Science Laboratory, 1 Research Link, National University of Singapore, Singapore, 117604. 4Lee Kong Chian School of Medicine, Nanyang Technological University, EMB 03-07, 59 Nanyang Drive, Singapore 636921. 5NTU Institute of Structural Biology, Nanyang Technological University, EMB 06-01, 59 Nanyang Drive, Singapore 636921. 6School

of Biological Sciences, Nanyang Technological University, 60 Nanyang Drive,

Singapore 636921. 7University Grenoble Alpes/CNRS, Département de Pharmacochimie Moléculaire, ICMG FR 2607, UMR 5063, 470 Rue de la Chimie, F-38041 Grenoble, France.

*To whom correspondence should be addressed. Email: [email protected], or [email protected], or [email protected] † These authors contributed equally to this work.

1 ACS Paragon Plus Environment

Bioconjugate Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 47

Abstract RNAs play critical roles in diverse catalytic and regulatory biological processes, and are emerging as important disease biomarkers and therapeutic targets. Thus, developing chemical compounds for targeting any desired RNA structures has great potential in biomedical applications. The viral and cellular RNA sequence and structure databases lay the groundwork for developing RNA-binding chemical ligands through the recognition of both RNA sequence and RNA structure. Influenza A virion consists of eight segments of negative-strand viral RNA (vRNA), all of which contain a highly conserved panhandle duplex structure formed between the first 13 nucleotides at the 5′ end and the last 12 nucleotides at the 3′ end. Here, we report our binding and cell culture anti-influenza assays of a short 10-mer chemically-modified double-stranded RNA (dsRNA)-binding Peptide Nucleic Acid (PNA) designed to bind to the panhandle duplex structure through novel major-groove PNA·RNA2 triplex formation. We demonstrated that incorporation of chemically-modified PNA residues thio-pseudoisocytosine (L) and guanidine-modified 5-methyl cytosine (Q) previously developed by us facilitates the sequence-specific recognition of Watson-Crick G-C and C-G pairs, respectively, at physiologically relevant conditions. Significantly, the chemically modified dsRNA-binding PNA (dbPNA) shows selective binding to the dsRNA region in panhandle structure over a single-stranded RNA (ssRNA) and a dsDNA containing the same sequence. The panhandle structure is not accessible to traditional antisense DNA or RNA with a similar length. Conjugation of the dbPNA with an aminosugar neamine enhances the cellular uptake. We observed that 2-5 µM of the dbPNA-neamine conjugate results in a significant reduction of viral replication. In addition, the 10-mer dbPNA inhibits innate immune receptor RIG-I binding to panhandle structure and thus RIG-I ATPase activity. These findings would provide the foundation for developing novel dbPNAs for the detection of influenza viral RNAs and therapeutics with optimal antiviral and immunomodulatory activities. 2 ACS Paragon Plus Environment

Page 3 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


INTRODUCTION The functional roles of viral RNA structures include the regulation of replication, transcription, splicing, translation and virus packaging.1-11 Thus, viral RNAs may serve as promising targets for developing chemical probes and potential therapeutics.12-22 Since 1918, influenza viral infection has been a significant threat to human health affecting socio-economic status worldwide. Influenza A virus (IAV) is a common human pathogen, and the annual epidemics cause severe illness and about 290,000 to 650,000 deaths every year worldwide according to a World Health Organization (WHO) report (accessed in December 2017).23 Currently, two antiinfluenza protein targets are available in the clinic: the M2 ion channel protein and neuraminidase (NA) responsible for nascent virion release. The emergence of drug-resistant viral strains makes currently available treatment schemes often ineffective. Thus, it is of vital importance to develop novel anti-influenza drugs12-15, 20-22, 24-26 to fight viral infections.

Influenza viruses contain a conserved panhandle double-stranded RNA (dsRNA) region, which is formed between the 5′ and 3′ termini of all influenza genomic viral RNA (vRNA) segments (see Figure 1A).1-5, 27-30 The panhandle RNA structure is essential for the viral RNA replication, transcription, and packaging. A conserved structural motif present in the vRNA panhandle consists of a U residue opposite to two A residues (Figure 1A), and has been identified as a target site by small molecular weight compounds.4, 12-14 A relatively high concentration (> 40 µM) of the small molecules can induce viral inhibition activities in cell based assays.13,

14

Traditional duplex-forming antisense compounds show antiviral activity20, 21 but may not easily get access to the sequence of the highly conserved and structured vRNA panhandle region (see Figure 1A,B), because the preformed duplex structure with or without protein bound27, 30-34 may significantly weaken the binding of traditional antisense strands.20, 35-37



Page 4 of 47

Figure 1. Influenza panhandle RNA duplex structure and its targeting by dsRNA-binding PNA (dbPNA). (A) A representative panhandle structure found in vRNA segment 8 (vRNA8) of Influenza A/California/04/2009 (H1N1).1 The panhandle sequence and structure are conserved in influenza type A, with the conserved residues boxed. A previous NMR structural study reveals that a U residue on the 3′ side may form a novel structure with two consecutive A residues on the 5′ side,4 which is targeted by one PNA T residue of the dbPNA IR-1 (shown in blue) in this study. The 10-mer dbPNA was designed to cover the major part of the panhandle structure. The terminal three base pairs are not targeted due to the presence of a U·G pair. The remaining large majority of the RNA residues of the vRNA are replaced with the residues shown in gray forming a small RNA hairpin structure (PH-v) for the binding assays. The underlined U residue is replaced with Cy3 dye in PH-v-Cy3. Two traditional antisense DNA strands (DNA1 and DNA2, shown in brown) are expected to show weakened binding due to the fact that the preformed RNA duplex structure needs to be disrupted for binding. (B) A control ssRNA fragment of PH-v (PH-v-ss) used for binding study. PH-v-ss can bind to the traditional antisense DNA1. (C) Chemical structure of dbPNA-neamine conjugate IR-1b. The lysine residue (black) has an L configuration. Neamine is shown in red.


Page 5 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Peptide Nucleic Acid (PNA) is an artificial nucleic acid with the nucleobases anchored on a neutral pseudopeptide backbone via a methylene carbonyl linkage (Figures 1A,C, 2, and S1),38 and are useful for applications in biology, diagnostics, and potential therapeutics.16, 39-46 PNAs are chemically stable and show superior resistance to nucleases and proteases, and are capable of forming duplexes via Watson-Crick pairing (with the complementary strands in a parallel or antiparallel orientation) with significantly enhanced binding affinity and specificity compared to natural DNAs or RNAs.38,

40, 47-49

Unmodified PNAs can form both Watson-Crick and

Hoogsteen pairs, and thus show no selectivity and can bind to both single-stranded RNAs (ssRNAs) and dsRNAs, resulting in the formation of various triplex structures including PNA·dsDNA and PNA·dsRNA triplexes (see C+·G-C and T·A-U triples in Figure 2) or strand invasion complexes of PNA·DNA-PNA and PNA·RNA-PNA triplexes.38, 50-52

An unmodified PNA (containing unmodified C and T bases) can bind to the major groove of an RNA duplex and form a sequence-specific PNA·RNA-RNA triplex at relatively low pH (pH 5.5, stabilized by C+·G-C and T·A-U triples) (see Figure 2).52 It is interesting to note that the relatively short unmodified PNAs show preferential binding to dsRNAs over dsDNAs.52 Modified bases have been incorporated into PNAs to enhance the binding affinity and specificity toward dsRNAs at physiologically relevant conditions.39, 53-55 We have demonstrated that PNAs incorporating modified bases such as novel G-C pair-recognizing thiopseudoisocytosine (L) and C-G pair recognizing guanidine-modified 5-methylcytosine (Q) can selectively bind to dsRNAs over ssRNAs and dsDNAs in a sequence-specific manner at near physiological conditions (Figure 2),39, 55-60 suggesting that our dsRNA-binding PNAs (dbPNAs) are complementary to RNA-binding small molecules and traditional antisense compounds,19, 6163

and have a great potential in targeting the dsRNA regions of viral and cellular RNAs.



Page 6 of 47

Figure 2. Chemical structures of base triples and weak Watson-Crick-like base pairs formed between dbPNA and RNA. The influenza panhandle region contains Watson-Crick A-U, G-C, and C-G pairs, and non-Watson-Crick A-C, and G-U pairs (see Figure 1A) as revealed by previous NMR structural studies.4, 12, 13 Due to the fact that dbPNA L and Q bases form unstable Watson-Crick-like L-G and Q·C pairs, respectively, dbPNAs containing L and Q bases show sequence specific and selective binding to dsRNAs over ssRNAs.39, 56-58

In this study, we designed a Q- and L-modified short (10-mer, see Figure 1A,C) dbPNA to target the conserved panhandle duplex region of IAV genomic vRNA (Figure 1A). We characterized by non-denaturing polyacrylamide gel electrophoresis (PAGE) and circular dichroism (CD) spectroscopy, and thermal melting the binding of the designed dbPNA to short oligonucleotide hairpins mimicking the viral RNA structure through novel PNA·RNA2 triplex formation (Figure 1A). Traditional antisense DNAs and RNAs, however, show no appreciable binding to the panhandle structure. We conjugated the dbPNA with an aminosugar neamine (Figure 1C) to enhance the cellular uptake.64, 65 For the first time, we demonstrated by cell


Page 7 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


culture assays that a short dbPNA-neamine conjugate targeting influenza panhandle RNA duplex has a potent anti-viral activity.

MATERIALS AND METHODS PNA synthesis All the PNAs used in this study were synthesized following a manual solid phase peptide synthesis (SPPS) protocol. The MBHA resin was used with a loading value of 0.25 mmol/g. All the PNAs were grown from C- to N-terminus on the MBHA resin using Boc chemistry SPPS protocol. The Boc-PNA-L-OH and Boc-PNA-Q-OH monomers were synthesized according to our reported protocols.56, 57 The Boc-PNA-T-OH monomer was purchased from ASM Research Chemicals GmbH.

In order to attach neamine and/or 5(6)-carboxyfluorescein to the PNA on solid support, we coupled Fmoc-Lys (Boc)-OH amino acid to the N-terminus of the PNA. The Boc group was removed, followed by coupling of the neamine monomer to N-terminus of the PNAs as reported previously.64, 65 The coupling of the neamine monomer was done using PyBOP and DIPEA as peptide coupling reagents and DMF as a solvent. The coupling of the neamine monomer to the PNA was done at slightly elevated temperature (45 °C) in the shaker incubator over a period of 12-16 h. To attach the 5(6)-carboxyfluorescein to the lysine- and neamine-attached PNA, the Fmoc group was removed and the peptide coupling reaction was carried out by using a commercially available 5(6)-carboxyfluorescein using DIC (Diisopropylcarbodiimide) and HOBt as coupling reagents and DMF as a solvent. The fluorescein-attached PNA was protected from the light during the coupling reaction and further studies. The synthesized PNAs were



Page 8 of 47

detached from the MBHA resin using a TFA/TFMSA cleavage protocol. The crude PNAs were purified by RP-HPLC and characterized by MALDI-TOF analysis.

Non-denaturing polyacrylamide gel electrophoresis (PAGE) The short RNA oligonucleotides were purchased from Sigma (Singapore). The short Cy3labelled RNA oligonucleotide PH-v-Cy3 was snap cooled from 95 °C, followed by annealing with PNA (with or without DNA) by slow cooling from 65 °C to 4 °C and then kept at 4 °C overnight. The incubation buffer (folding buffer) contains 200 mM NaCl, 0.5 mM EDTA, 20 mM HEPES (pH 7.5). The concentration of the RNA PH-v-Cy3 was constant (0.05 µM). To 20 µl of the samples, 4 µL of 35% glycerol (diluted in the incubation buffer) was added just before the samples were loaded into the wells. 12% non-denaturing PAGE (22 cm × 16.5 cm) was run at 4 °C at 250 V for 5h in 1× TBE buffer. The gel images were obtained with a Typhoon imager. The bands corresponding to the bound and unbound forms of RNA were quantified. The Kd values were obtained by fitting to equation 1,56-58, 66 𝐵

1

𝑌 = 𝑌0 +(2𝑅0)(𝑅0 +𝑋 + 𝐾d ― ((𝑅0 + 𝑋 + 𝐾d)2 ― 4𝑅0𝑋)2),

(1)

where R0 is the RNA hairpin concentration (0.05 µM). X is the total PNA concentration. Y is the fraction of triplex at varied PNA concentration. Y0 and B are the initial and the maximum change of the triplex fraction, respectively. Kd is the dissociation constant.

CD spectroscopy The CD spectra were taken using a JASCO model J-1500-150 spectropolarimeter. The concentrations of RNA oligonucleotide hairpins and PNA or DNA are 2 and 4 µM, respectively. The buffer contains 100 mM NaCl, 10 mM sodium phosphate, pH 7.0. RNA hairpins were snap cooled from 95 °C, followed by annealing with PNA or DNA by slow cooling from 65 °C to 8 ACS Paragon Plus Environment

Page 9 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


room temperature, and then kept at 4 °C overnight. The measurement was done at room temperature.

UV-absorbance-detected thermal melting UV-absorbance-detected thermal melting experiments were conducted using the Shimadzu UV-2550 UV-Vis spectrophotometer with the use of an 8-microcell cuvette. The optical path length of the 8-microcell cuvette is 1 cm. The absorbance at 260 nm was recorded with the temperature ramp rate at 0.5 °C/min. The buffer contains 200 mM NaCl, 20 mM HEPES, 0.5 mM EDTA, at pH 7.5. All samples contain 2 µM RNA and 2 µM PNA or DNA in 130 µL buffer. The samples containing the ssRNA and PNA were annealed by slow cooling from 95 °C to room temperature, followed by incubation at 4 °C overnight. Data were normalized at low temperature readings and the melting temperatures were determined based on the Gaussian fits of the first derivatives of the curves.

In vitro translation assay The cell free translation assay was done as previously described.60, 67-69 The DNA sequences for the RNA of interest (rHP1) were inserted between the Renilla luciferase (RLuc) and firefly luciferase (FLuc) reporter genes of p2luc vector. The dual luciferase report system was originally constructed for quantifying −1 ribosomal frameshifting efficiencies. mRNAs for both the experimental construct and a positive control were generated by linearization of the plasmids using restriction enzyme PmlI (NEB) followed by in vitro transcription. PNA with a series of concentrations (0.2, 2, 5, 10, and 20 µM) were added to 0.1 µM mRNA templates and the mixture was incubated at room temperature for 20 minutes. After the incubation, in vitro 9 ACS Paragon Plus Environment


Page 10 of 47

translation was carried out by adding 8.75 µL of nuclease treated Rabbit Reticulocyte Lysate mixture (Promega) into the mRNA/PNA mixture (with a final volume of 12.5 µL). The in vitro translation reaction was run at 30 °C for 90 minutes followed by reaction quenching on ice for 20 minutes.

The luminescence of the translation products was quantified by using the Dual-Glo Luciferase assay system (Promega). The expression levels of RLuc and FLuc were monitored using Tecan infinite M200 microplate reader on Nunclon™ flat-bottom 96-well black microplate with an integration time of 3 seconds at room temperature. FLuc level was measured by adding 50 µL of the corresponding luciferase substrate to 2 μL in vitro translation product, followed by the addition of 50 µL Stop & Glo reagent for the measurement of the RLuc expression.

Cell line and influenza strains All tests concerning cell culture were carried out on Mandin-Darby Canine Kidney (MDCK) cell line. Cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% heat-inactivated fetal bovine serum, 2 mM glutamine and antibiotics (penicillin 100 U/ml and streptomycin 100 µg/ml) in 5% CO2 environment at 37 °C. Experiments were carried on pdm09 H1N1 (A/California/04/2009), pdm09 H1N1 (A/Singapore/TLL01/2009), H7N1 (A/common iora/Indonesia/F89/11/95), and reverse genetics-derived H5N1 (RG-H5N1). RGH5N1 is a reassortant virus previously generated and contains the hemagglutinin and neuraminidase from A/Hubei/1/2010 and the internal six genes from A/Puerto Rico/8/1934.70

Confocal microscopy


Page 11 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


MDCK cells were seeded on CELLview Cell Culture Dishes (Greiner Bio One) with glass bottom at a density of 1.2×105 per compartment and grown in a culture medium at 37 °C for 24 h. Then, the culture medium was replaced with the medium containing 4 µM carboxyfluorescein-labelled PNA-neamine conjugates for 12 h. After the incubation, the cells were washed and observed under confocal microscope.

Cell culture tests MDCK cells were seeded on 96-well plate at a density of 2×104 and grown in the culture medium as described above for 24 h. The culture medium was then replaced with the medium containing PNAs for 12 h. The cell culture was then infected with influenza virus diluted with infection medium (0.3% BSA, 100 U/ml penicillin, 100 µg/ml streptomycin in PBS) at 0.01 multiplicity of infection (MOI). After 1 hour incubation at room temperature (RT) on a gently rocking platform, supernatants were removed and the post-infection medium (0.3% BSA, 100 U/ml penicillin, 100 µg/ml streptomycin, 2 mM glutamine, 1 µg/mL TPCK-treated trypsin in DMEM) was added to the cell culture. Infected cells were kept at 33 °C, 5% CO2 for 24 h. The cell culture supernatants were harvested for indirect immunofluorescence assay (IFA) and the total RNA from the cell monolayer was isolated for real-time PCR analysis. In addition, 2′OMe-modified antisense oligonucleotides were transfected with lipofectamine into MDCK cell line at a final concentration of 4 µM. The cell culture was infected with influenza virus 18 h after transfection as described above. The total RNA was extracted for quantification by real time PCR 24 h postinfection.

Indirect Immunofluorescence assay (IFA)



Page 12 of 47

Supernatants collected from infected cell culture were used to prepare 10-fold serial dilutions of the virus with infection medium. Cell culture monolayers on 96-well plate were infected with serial dilutions of viral supernatants for 1 hour at RT on a gently rocking platform. Then the supernatants were discarded and cells were maintained in post-infection medium for 8-10h at 33 °C, 5% CO2. The supernatants were again removed and the cells were fixed with 4% formaldehyde and then permeabilized with 0.5% Triton X-100 solution in PBS for 20 min at RT. Blocking was done using 3% BSA solution in PBS for 1h at RT. Then the solution was replaced with mouse anti-influenza primary antibody targeting nucleoprotein (NP) (MAB8257 Merck) diluted with 3% BSA solution in PBS (1µg/ml) and incubated for 1-2 h at 37 °C. The detection was carried out with FITC-conjugated secondary rabbit anti-mouse IgG antibody (AP160F Merck) diluted with 3% BSA solution in PBS (1:150) incubated for 30-60 min at 37 °C. Visualization under fluorescent microscope was followed by calculation of fluorescentforming units (FFU/ml).

Real-Time PCR analysis Total RNA from the cell culture was extracted using Trizol reagent.71 500 ng of each isolated sample was treated with RNase-free DNase I (Invitrogen) at 37 oC for 30 min according to manufacturer’s instructions. The DNase I enzyme was inactivated by adding 1 µL 25mM EDTA with a final concentration of 2.3 mM followed by heating at 75 oC for 10 min. The quality of isolated RNA was checked by agarose gel. Reverse transcription was performed with the primer specific for matrix protein 1 (M1) gene (5′-ATGAGTCTTCTAACCGAGGTCG-3′) and SuperScript III Reverse Transcriptase (Invitrogen). A fraction of the sample after DNase I treatment (1 µL) was mixed with 1× First-Strand Buffer, 0.4 µM gene-specific primer, and H2O with a final volume of 5 µL. The solution was kept at 90 oC for 3 min, at 55 oC 10 min, and


Page 13 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


then placed on ice. Then the solution was added with 10 mM DTT, 2.5 mM dNTPs, 1× FirstStrand Buffer, 10U RNasin inhibitor, and 50 U SuperScript III Reverse Transcriptase with a final volume of 10 µL, and incubated at 55 oC for 50 min. Then, the reaction was inactivated by heating at 70 oC for 15 min.

1 µL of the obtained cDNA was subjected to real-time PCR with gene-specific primers (forward 5′-AGACCAATCTTGTCACCTCTGAC-3′,

reverse

AGGGCATTTTGGACAAAGCGTCTACG-3′),

probe

5′(5′-FAM-

TCACCGTGCCCAGTGAGCGAGGACTGC-TAMRA-3′) and 5× HOT FIREPol Probe qPCR Mix Plus (Solis BioDyne) according to the manufacturer’s protocol. The reaction was performed in triplicates at 95 °C for 15 min, 39 cycles of 95 °C for 20s, 60 °C for 1 min, with each followed by plate read. The absolute quantifications allowed the comparison of viral RNA copies in PNA-treated and untreated samples.

Antiviral activity against Influenza A viral subtypes The antiviral activity of IR-1b was tested against three different influenza A subtypes (RGH5N1, pdm H1N1 and H7N1). Subconfluent monolayers of MDCK cells in 96-well culture plates were pre-treated with 5µM IR-1b or neamine alone at 37 °C in 5% CO2 for 12h. Then the cells were washed with PBS and infected with 100µl of 0.05 MOI of pdm H1N1 or RGH5N1 or H7N1 virus for 36 h at 37 °C. After 36 h incubation, the supernatants were collected and subjected to virus titration. The virus supernatants were serially diluted 10-fold and titrated by infecting MDCK culture on a 96-well plate. After 36 h of incubation at 37 °C, the cells were fixed with 4% paraformaldehyde and indirect IFA was performed using anti-N1NA specific monoclonal antibody to determine the viral titer. The viral titer was then calculated as 50% 13 ACS Paragon Plus Environment


Page 14 of 47

tissue culture infectious dose per volume (TCID50/mL) using the Reed and Muench method. The experiment was repeated three times with triplicates.

Cytotoxicity assay MDCK cells were treated with PNAs by following a similar protocol as that for antiviral assays described above. After 12h incubation in the presence or absence of PNAs, the cell culture was subjected to CellTiter 96® Non-Radioactive Cell Proliferation Assay (MTT) (Promega) according to the manufacturer’s protocol. The absorbance measured at reference wavelength 650 nm was deducted from that recorded at 570 nm. The normalized absorbance values of PNAtreated wells and untreated wells were compared.

Human RIG-I Protein expression and purification Plasmid containing human RIG-I protein in pET-SUMO vector was expressed in Rosetta 2 (DE3) cells.72 Cells were grown in LB broth supplemented with 50 mg/mL kanamycin, 37 mg/mL chloramphenicol, 2.5% v/v glycerol and 50 mM potassium phosphate buffer. Culture was grown at 37 oC until OD600 of 0.8 before cooling it down to 18 oC and inducing it with IPTG for 20 hrs. Cells were harvested by centrifugation at 4000 xg and lysed in buffer containing 25 mM HEPES pH 7.4, 500 mM NaCl, 10% (v/v) Glycerol, and 5 mM βmercaptoethanol. The resuspended cells were passed through the homogenizer (GEA) at 800 bar and clarified by centrfugation at 40 000 rpm at 4 oC. The supernatant were added with NiNTA resin (Thermofisher). The protein of interest were eluted in lysis buffer with 300mM imidazole. The protein was cleaved overnight with SUMO protease at the ratio of 1:40 (w/w). The protein was further polished with Ni-NTA and Heparin HP column and passed through 14 ACS Paragon Plus Environment

Page 15 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


size exclusion chromatography column S200 16/600 pg. The protein was concentrated using vivaspin with 50 000 MWCO and were flash frozen in buffer containing 25 mM HEPES pH 7.4, 150 mM NaCl, 10% (v/v) glycerol and 2 mM DTT.

Fluorescence binding assay of RIG-I protein binding to panhandle RNA The fluorescent RNA used is the hairpin PH-v-Cy3 which is internally labelled with a Cy3 dye. The hairpin PH-v-Cy3 was diluted to a final concentration of 2 nM in assay buffer containing 25 mM MOPS pH 7.4, 50 mM NaCl, 2 mM DTT, 2 mM MgCl2, 0.01% (v/v) Triton X-100. The RIG-I protein was serially diluted to 12 different concentration ranging from 1500 nM to 0.7 nM. The RIG-I protein was incubated with PH-v-Cy3 for 1 hour at room temperature. The assay was carried in a 384-well plate format and the fluorescent intensity was measured in Biotek Cytation 3 multimode reader. The fluorescent intensity was measured using a filter with an excitation of 530/25 nm and emission of 590/35 nm. For the assay in the presence PNA IR-1, 500 nM and 2 µM IR-1 were incubated with 2 nM PH-v-Cy3 for an hour at room temperature before incubating with RIG-I protein for another hour. The assay was carried out in 30 µL volume and the measured intensity was plotted against total protein concentration as described in equation 1 using using GraphPad Prism® version 6. For the fitting, R0 is the RNA hairpin PH-v-Cy3 concentration (2 nM). X is the total protein concentration. Y is the fluorescence intensity at varied protein concentration. Y0 and B are the initial and the maximum change of the fluorescence intensities, respectively. Kd is the dissociation constant.

Panhandle RNA induced RIG-I ATPase assay and effect of IR-1 binding ATPase assay follows a previously established protocol with modifications.73 Briefly, RIG-I was diluted in assay buffer containing 25 mM MOPS pH 7.4, 150 mM KCl, 2 mM DTT, 0.01% 15 ACS Paragon Plus Environment


Page 16 of 47

(v/v) Triton X-100 to a final concentration of 100 nM. Protein was incubated at the final concentration of 1 µM PH-v or IR-1 for an hour with at room temperature. The NADH coupled ATPase assay was carried out in a mixture containing 100 U/mL pyruvate kinase, 20 U/mL lactate dehydrogenase, 500 µM phosphoenol pyruvic acid and 0.2 mM of NADH. Assay was initiated with the addition of ATP: MgCl2 at 1:1 ratio with 8 point serial dilution to a final concentration ranging from 5000 µM to 39 µM. Assay was carried out in 96-well plate format with a final volume of 50 µL. The rate of ATPase activity was determined through the changes of absorbance 340nm at 1 minute interval over 10 minutes in Biotek Cytation 3 multimode reader. For the assay with PNA IR-1, 2 µM IR-1 was incubated with 1 µM PH-v for 1 hour before incubating with RIG-I for another hour. The data were obtained as triplicate and plotted as rate of NADH hydrolysis as a function of ATP concentration. The data were plotted using Michaelis-Menten equation for using GraphPad Prism® version 6.

Table 1. Sequence and MALDI-TOF data of PNAs studied in this papera

PNA

Sequence (Nter-PNA-Cter)

Calculated MW (Da)

Observed MW (Da)

IR-1

NH2-Lys-TLTTTQTLLL

2894.13

2895.00

IR-1a

NH2-ahx-TLTTTQTLLL

2879.11

2886.17

IR-1b

NH2-Lys(Neamine)-TLTTTQTLLL

3397.42

3404.80

cf-Lys(Neamine)-TLTTTQTLLL

3755.46

3763.68

NH2-Lys-TLTLTTTL

2276.86

2277.88

NH2-Lys(Neamine)-TLTLTTTL

2780.16

2781.28

IR-1b-cf P5 P5-b a

“ahx” represents 6-amino hexanoic acid. “Lys” represents a lysine residue with an L configuration. “cf” represents 5(6)-carboxyfluorescein. The detailed chemical structures can be found in Figure S1.


Page 17 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


RESULTS AND DISCUSSION Characterization of PNA binding properties We designed dbPNA IR-1 for targeting the panhandle stem region conserved in all eight genomic vRNA segments in IAV through PNA·dsRNA triplex formation (Figure 1A and S1, Table 1). Due to the presence of a G-U and an A·C pair in the panhandle region, we utilized PNA L and T bases to form L·G-U and T·A·C triples, respectively, in addition to L·G-C and T·A-U triples (Figures 1 and 2). The conserved structural motif consisting of two A residues opposite to one U residue (Figure 1A) is targeted by one PNA T residue. Such a targeting strategy allows the formation of coaxially stacked PNA·RNA·RNA triplex with the extra A residue expected to be bulged out upon PNA IR-1 binding in the major groove. dbPNAs IR-1 and IR-1a have a lysine residue and a 6-aminohexanoic acid at the N-termini, respectively (Table 1, Figure S1). To facilitate the cell uptake, we further conjugated dbPNA IR-1 at the N-terminus with neamine (designated as IR-1b, Figure 1C and S1, Table 1).64, 65 Labelling of IR-1b with carboxyfluorescein (IR-1b-cf, Table 1, Figure S1) allows the imaging of the intracellular distribution the PNA conjugate.

We quantified the binding affinities of dbPNA IR-1 to a short model dsRNA mimicking panhandle structure (PH-v-Cy3, Figure 1A) by non-denaturing polyacrylamide gel electrophoresis (PAGE). IR-1 binds to the targeted model hairpin PH-v-Cy3 with a Kd value of 0.3 ± 0.1 µM at 200 mM NaCl, pH 7.5 (Figure 3A,B). The binding is not significantly dependent on pH with the Kd value ranging from 0.2-0.4 µM by varying pH from 6.0 to 8.0 (Figure S2A). No binding is observed between dbPNA IR-1 and a control RNA hairpin rHP1 (see rHP1 structure in Figure S3A) or a control PNA P5 (see P5 sequence in Table 1 Figure S3A), suggesting that the binding of dbPNA IR-1 is sequence specific (Figure S2B). A



Page 18 of 47

panhandle structure is also present in the full-length positive-sense complementary RNA (cRNA) (Figure S3C).15 The non-denaturing PAGE data indicate that, IR-1 has a weakened binding to cRNA panhandle (PH-c-Cy3), probably due to the destabilizing effect of two mismatched base triples present for RNA PH-c (Figures S2C and S3C). In addition, PH-c has a relatively weakened stem compared to PH-v, which may cause weakened binding to IR-1.

Figure 3. PAGE assay of the binding of PNA to short model RNA hairpin mimicking panhandle structures (PH-v-Cy3, see Figure 1A). Hairpin PH-v-Cy3 is internally labelled with a Cy3 dye and was loaded at 0.05 µM. The incubation buffer (folding buffer) contains 200 mM NaCl, 0.5 mM EDTA, 20 mM HEPES (pH 7.5). (A,B) Titration of PNA IR-1 into PH-v-Cy3. PH-v-Cy3 binds to IR-1 (Kd = 0.3 ± 0.1 µM). (C,D) Titration of DNA1 or DNA2 (see the DNA sequences in Figure 1A) into PH-v-Cy3. Traditional antisense DNA1 and DNA2 show no binding to PHv-Cy3, suggesting that the stem loop structure of PH-v is not accessible to traditional antisense DNA. (E,F) Titration of DNA1 or DNA2 into the preformed complex of IR-1 and PH-v-Cy3. DNA1 and DNA2 show no binding to preformed complex of IR-1 (with a final concentration of 1 µM) and PH-v-Cy3, suggesting that IR-1 binds to PH-v-Cy3, without disrupting the PH-v stem loop structure.

We further characterized the PNA·dsRNA triplex formation by CD spectroscopy (Figure 4AC). dbPNA IR-1 alone at 4 µM shows no helicity as expected. The positive peak at around 270 18 ACS Paragon Plus Environment

Page 19 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


nm of PH-v (2 µM) has a reduced intensity and is red shifted upon the addition of 4 µM IR-1 (Figure 4A). However, no significant change was observed upon adding a control PNA P5 to PH-v (see P5 sequence in Table 1 and Figure S3), suggesting no triplex formation. PNA P5, but not IR-1 induces the CD signal change of rHP1 (Figure 4B), indicating the triplex formation between P5 and rHP1, in agreement with our previously reported binding data (with a Kd value of around 0.2 µM).56, 57 The data are consistent with the previously reported CD results52, 74, 75 and indicate the sequence-specific formation of PNA·dsRNA triplexes (IR-1·PH-v and P5·rHP1). In addition, the CD data suggest that IR-1 shows no clear evidence of triplex formation with PH-c-Cy3 (Figure 4C, see the structure of PH-c in Figure S3), likely because the CD signal is reduced for a relatively weakened and more flexible triplex.

Our thermal melting results suggest that dbPNA IR-1 has no binding to a ssRNA fragment (PHv-ss, Figure 1B) of panhandle region, in contrast to a traditional antisense DNA strand, DNA1 (Figure 1A,B), which shows binding to the ssRNA fragment (Figure 4D). Note that DNA1 is designed to be fully complementary with the 5’ side of the panhandle, containing two instead of one DNA T residue for the targeting of the two A residues involved in the conserved (AA)·U structural motif (Figure 1A). Our non-denaturing PAGE data suggest that neither DNA1 nor DNA2 (Figure 1A) shows binding to PH-v-Cy3 (Figure 3C,D), suggesting that the panhandle duplex structure is not accessible to traditional antisense oligonucleotides. The fact that the preformed complex of IR-1 and PH-v-Cy3 does not show binding to DNA1 or DNA2PNA indicates that IR-1 binds to PH-v-Cy3 without disrupting the preformed panhandle stem structure (Figure 3E,F). Taken together, dbPNA IR-1 is advantageous in selectively recognizing a dsRNA over a ssRNA, as has been observed for our other L- and Q-modified PNAs.39, 55-58



Page 20 of 47

Figure 4. Characterization of potential PNA·dsRNA triplex formation by CD and potential duplex formation by UV-absorbance detected thermal melting. (A-C) CD spectroscopy. PNAs IR-1 and P5 bind to PH-v and rHP1 (see rHP1 structure in Figure S2A), respectively, through the formation of PNA·dsRNA triplexes, as evidenced by a red shift of the peak at around 270 nm upon the addition of PNAs. The data shown in panel C suggest that IR-1 shows no binding to PH-c-Cy3 (see the structure of PH-c in comparison to PH-v in Figure S2B,C). (D) UVabsorbance detected thermal melting. Antisense DNA1 binds to the ssRNA PH-v-ss as evidenced by a thermal melting transition. dbPNA IR-1 shows no binding to the ssRNA PHv-ss.


Page 21 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


To test if PNA IR-1 has potential off-target binding to ribosomal RNAs, the most abundant RNAs in the cell,17 we carried out cell-free translation assay using a dual luciferase (Renilla and firefly luciferase) report system (Figure S4). We have previously shown that PNA P5 can bind to rHP1 stem region of the mRNA and stimulate the translation of firefly luciferase.60 However, with the concentration of PNA IR-1 below 10 µM, both RLuc and FLuc expression levels in the experimental and control constructs do not change significantly. The data suggest that PNA IR-1 shows no significant nonspecific inhibition of translation through off-target binding to ribosomal RNAs, which is in agreement with the fact that we observe minimal cellular toxicity for dbPNA IR-1b (see below).

Cellular uptake We next tested the cellular uptake of the dbPNA. Unmodified PNA oligomers exhibit low cell permeability in the absence of transfection agents or cell-penetrating moieties42,

57, 76-79

or

electroporation methods.46 Our confocal microscopy studies (Figure 5) show that the dbPNA conjugated with neamine, IR-1b-cf, penetrates through the cell membrane of MDCK cells without the aid of a transfection agent, which is consistent with the previous studies for unmodified PNAs attached with neamine.64,

65

The dbPNA conjugate IR-1b-cf, upon

internalized inside cells, is homogenously distributed within the cell including nuclei and mitochondria, although not all the cells have successfully taken up IR-1b-cf (Figure 5). Thus, our confocal imaging studies confirm that efficient cellular uptake of PNAs without endosomal trapping may be facilitated through the conjugation of PNAs with neamine as previously reported.64, 65



Page 22 of 47

Figure 5. Confocal microscope images of MDCK cells treated with 4 µM IR-1b-cf for 12 h. (A) Carboxyfluorescein labelled PNA-neamine conjugate IR-1b-cf is uptaken by MDCK cells. A1: IR-1b-cf (Ex.: 495 nm, Em.: 520 nm). A2: bright field. A3: DAPI staining of nuclei (Ex.: 350 nm, Em.: 470 nm). A4: merged image. (B) Zoom-in view of intracellular distribution of IR-1b-cf. B1: IR-1b-cf. B2: MitoTracker mitochondria staining (Ex.: 580 nm, Em.: 600 nm). B3: DAPI staining of nuclei. B4: merged image.


Page 23 of 47 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 6. Antiviral effect of the dbPNA IR-1b on viral protein expression, RNA replication, viral growth. (A) IFA assay suggests that viral protein expression is inhibited by the dbPNAneamine conjugate IR-1b. (B) Real-time PCR data reveal the inhibitory effect of IR-1b on viral RNA replication. No viral inhibition was observed for IR-1a (see Table 1) without or with neamine added in trans. (C) MTT assay reveals no cellular toxicity of IR-1b. Error bars represent standard deviations derived from three independent experiments. The unpaired twotailed student’s t-test was performed for statistical comparisons (**P

A Short Chemically-Modified dsRNA-Binding PNA (dbPNA) Inhibits

Recommend Documents