Antimicrobial activity of antisense peptide–peptide nucleic acid conjugates against non-typeable Haemophilus influenzae in planktonic and biofilm forms

Background

Antisense peptide nucleic acids (PNAs) are synthetic polymers that mimic DNA/RNA and inhibit bacterial gene expression in a sequence-specific manner.

Methods

To assess activity against non-typeable Haemophilus influenzae (NTHi), we designed six PNA-peptides that target acpP, encoding an acyl carrier protein. MICs and minimum biofilm eradication concentrations (MBECs) were determined. Resistant strains were selected by serial passages on media with a sub-MIC concentration of acpP-PNA.

Results

The MICs of six acpP-PNA-peptides were 2.9–11 mg/L (0.63–2.5 μmol/L) for 20 clinical isolates, indicating susceptibility of planktonic NTHi. By contrast, MBECs were up to 179 mg/L (40 μmol/L). Compared with one original PNA-peptide (acpP-PNA1-3′N), an optimized PNA-peptide (acpP-PNA14-5′L) differs in PNA sequence and has a 5′ membrane-penetrating peptide with a linker between the PNA and peptide. The optimized PNA-peptide had an MBEC ranging from 11 to 23 mg/L (2.5–5 μmol/L), indicating susceptibility. A resistant strain that was selected by the original acpP-PNA1-3′N had an SNP that introduced a stop codon in NTHI0044, which is predicted to encode an ATP-binding protein of a conserved ABC transporter. Deletion of NTHI0044 caused resistance to the original acpP-PNA1-3′N, but showed no effect on susceptibility to the optimized acpP-PNA14-5′L. The WT strain remained susceptible to the optimized PNA-peptide after 30 serial passages on media containing the optimized PNA-peptide.

Conclusions

A PNA-peptide that targets acpP, has a 5′ membrane-penetrating peptide and has a linker shows excellent activity against planktonic and biofilm NTHi and is associated with a low risk for induction of resistance.

Introduction

Non-typeable Haemophilus influenzae (NTHi) is a commensal of the human upper respiratory tract, as well as a pathogen in otitis media and other respiratory tract infections. Acute otitis media imposes a large burden on both children and their parents, with 13 million episodes annually at an annual cost of ∼$3 billion in US children aged 0–4 years.¹ The incidence of acute otitis media is 0.5–1.2 episodes per person-year during the first 2 years of life.² Since the introduction of the pneumococcal conjugate vaccines, a shift of otopathogens in otitis media has been observed. Instead of Streptococcus pneumoniae, NTHi has been the leading cause of acute otitis media (30%–65%) and the most common otopathogen detected from middle ear fluids of children with otitis media with effusion (20%–30%).^3–7

Biofilm formation plays a central role in the pathogenesis of otitis media.^8–10 NTHi forms biofilm on middle ear mucosa and in the nasopharynx of children.^11,12 Bacterial biofilm made by NTHi is up to 1000 times more resistant to standard antibiotics than planktonic bacteria.^13–19 Interestingly, the usefulness of antibiotics for the treatment of otitis media is determined by their ability to kill only planktonic bacteria.

Nasopharyngeal colonization is a critical early step in otitis media. Thus, preventing or eradicating nasopharyngeal colonization by otopathogens will prevent otitis media.^20,21 Two strategies to achieve this concept are the use of antimicrobial agents and vaccines. However, prolonged treatment or suppression with currently available antimicrobial agents eradicate the normal upper airway flora and are associated with unacceptable adverse effects. Furthermore, the 10-valent pneumococcal H. influenzae protein D conjugate vaccine prevents only 36% of otitis media episodes caused by NTHi,²² and has little or no effect on nasopharyngeal colonization.²³ Thus, this vaccine will not have a herd effect through reducing NTHi nasopharyngeal colonization. Therefore, new approaches beyond traditional antibiotics to treat and prevent otitis media are urgently needed. The application of antisense technology to bacterial infections has opened a potentially new era of therapeutics.

Peptide nucleic acids (PNAs) are artificially synthesized polymers that mimic DNA or RNA.²⁴ The binding between PNA and DNA is stronger than DNA–DNA binding, allowing the use of short PNAs (typically 11-mers) as antimicrobial agents. PNA oligomers also show high binding specificity with a single mismatch being destabilizing, resulting in highly specific targeting of genes.^24–26 Another important feature of PNAs is that they are not recognized by nucleases or proteases, making them resistant to enzyme degradation. Recently, an innovative approach of conjugating a short peptide to PNAs was developed, facilitating penetration of PNAs through the bacterial cell wall.^27,28 In addition, a linker/spacer [e.g. ethylene glycol linker (egl), 8-amino-3,6-dioxaoctanoic acid] can be added between the PNA and the membrane-penetrating peptide to increase solubility. PNA-peptides have antimicrobial activity by inhibiting expression of genes that are critical for bacterial viability. Because PNA-peptides inhibit gene expression in a sequence-specific manner, these molecules can be designed to eradicate pathogens, without disrupting non-targeted bacteria such as normal, commensal bacteria.

Antisense molecules are active against several bacteria, including Klebsiella pneumoniae,²⁹Escherichia coli,³⁰Salmonella enterica,³¹Burkholderia,³²Staphylococcus aureus³³ and Streptococcus pyogenes.³⁴ PNA-peptides targeted to the mRNAs of essential genes (such as acpP, ftsZ, gyrA and others) show potencies in the low to sub-micromolar range against many Gram-negative bacteria, which is as potent as many standard antibiotics.^{27,33,35–37} The few times that PNA-peptides have been tested in animal models of infection, they have increased survival and reduced the bacterial load in organs.^33,38

By contrast, antisense molecules against NTHi have not yet been studied. Our aim was to assess in vitro antimicrobial activity of PNA-peptides targeting a specific gene that is important for viability and biofilm formation of NTHi. We investigated the activity of PNA-peptides against NTHi in planktonic and biofilm forms as part of a strategy to eradicate NTHi nasopharyngeal colonization.

Materials and methods

Bacterial strains and growth conditions

NTHi strain 86-028NP is a nasopharyngeal isolate from a patient who underwent tube insertion for chronic otitis media and has been extensively characterized.³⁹ Ten nasopharyngeal and 10 middle ear fluid isolates from children with otitis media were evaluated (Table 1). These 20 isolates were collected from various geographical areas. NTHi strains were grown on chocolate agar at 35°C with 5% CO₂ or in supplemented brain heart infusion (sBHI) broth with shaking at 37°C. sBHI consisted of BHI broth supplemented with haemin and NAD at 10 μg/mL each.

acpP gene sequence analysis

The acpP genes and 10 bp upstream were amplified by PCR from genomic DNA of the 20 NTHi clinical isolates and sequences were determined at the Roswell Park Cancer Institute sequencing facility using primers listed in Table S1 (available as Supplementary data at JAC Online). These sequences were aligned and analysed with BLAST (http://blast.ncbi.nlm.nih.gov/Blast.cgi).

Design of PNA-peptides

PNA-peptides were purchased from PNA Bio Inc. (Thousand Oaks, CA, USA). An 11-mer that is complementary to the acpP gene (−4 to +7) of NTHi strain 86-028NP and spans the ATG start codon was designed (acpP-PNA1; Table 2). The acpP gene encodes an acyl carrier protein that is essential for lipid biosynthesis. The PNA was conjugated to a peptide (RFFRFFRFFR on either the 3′ or 5′ end of the PNA), which facilitates penetration through the bacterial cell wall. Another acpP-PNA that is complementary to a region downstream of the start codon but does not include the start codon of acpP (+4 to +14) was also designed (acpP-PNA14, Table 2). Control PNAs are 11-mers that are non-complementary to any genes of NTHi strain 86-028NP. Our PNAs are named by the combination of components: ‘target gene’-PNA-‘design of base sequence’-‘position of penetrating peptide’-‘possession of a linker’. For example, acpP-PNA1-3′N targets the acpP gene (acpP-PNA1-), has a peptide conjugated to the 3′ end of the PNA (-3′) and contains no linker (N).

MLST

MLST ST was determined as previously described.⁴⁰ The allelic profile and MLST were determined using the MLST.net server (http://haemophilus.mlst.net/). On the same web site, we ran the eBurst analysis to determine clonal complex and to compare genetic diversity of the 20 clinical isolates.

MICs for planktonic bacteria

MICs were determined by microdilution⁴¹ in sBHI, instead of Haemophilus test medium. MICs were defined as the lowest concentration of the PNA-peptide that inhibited growth after 20 h, as detected by turbidity.

Establishment of NTHi biofilm on peg lid

Strain 86-028NP and six clinical isolates were grown as biofilms using a minimum biofilm eradication concentration (MBEC) assay plate (Innovotech Inc., Edmonton, Canada) according to the manufacturer's protocol. Briefly, a 120 μL inoculum (5 × 10⁵ cfu/mL) was added in duplicate to wells of a 96-well plate (Corning Inc., Corning, NY, USA). An MBEC pin lid was placed into the plate and incubated in a humidified incubator at 37°C with shaking at 175 rpm for 24 h.

MBECs for NTHi

Preformed biofilms on the MBEC peg lids were rinsed twice in sBHI (150 μL/well) to remove residual planktonic bacteria. The lid was transferred into a new plate containing 2-fold dilutions of PNA-peptides (total volume 120 μL/well). This plate was incubated for 20 h at 37°C, with shaking at 175 rpm. After 20 h exposure to the PNA-peptides, the MBEC lid was rinsed twice and placed into an original MBEC base plate with sBHI (150 μL/well). The plate was placed in a dry stainless steel insert tray, which sits on the water of the sonicator, then sonicated for 30 min on ‘high’ to remove biofilms from pegs. The broth from the wells was collected, a 20 μL volume was plated and colonies were counted the following day. The MBEC is defined as the minimum concentration of antimicrobial agent that eradicates the biofilm. In this study, MBEC represents the lowest concentration of PNA-peptide that kills 99.9% of the population of NTHi within biofilm.

Selection of acpP-PNA-peptide-resistant strains

To select acpP-PNA-peptide-resistant strains, serial overnight passages of strain 86-028NP and 3113-S in sBHI broth containing a sub-MIC PNA-peptide concentration (1/4–1/8 MIC) were performed.

Determination of genome sequence

Susceptible (86-028NP) and resistant (86-028NP-D15A) strains were 150 bp paired-end sequenced on the Illumina HiSeq 2000 in the UB Next-Generation and Expression Analysis Core at University at Buffalo. The raw sequencing data were trimmed to remove sequencing adaptors and low-quality regions using Btrim.⁴² The sequences were aligned to the H. influenzae 86-028NP genome (GenBank CP000057.2) using BWA.⁴³ The SNPs were called and annotated from the alignments using in-house scripts.

Construction of NTHI0044 mutant

NTHI0044 deletion mutants were constructed in NTHi strains 86-028NP and 3113-S by using overlap extension PCR⁴⁴ and homologous recombination as described previously.^45,46 Briefly, the transforming DNA for the mutant was composed of three overlapping fragments that included 1 kb upstream of the gene being knocked out (fragment 1), the non-polar kanamycin resistance cassette amplified from plasmid pUC18K⁴⁷ (fragment 2) and 1 kb downstream of the gene (fragment 3) using the oligonucleotide primers listed in Table S1. A mutant was constructed by transformation of each strain with a fragment composed of fragments 1, 2 and 3, and selecting on chocolate agar containing 50 μg/mL ribostamycin. The insert and surrounding sequences of each of the mutants were confirmed by sequence determination.

Complementation of NTHI0044 mutant

Complementation was accomplished by using the plasmid pSPEC1.⁴⁸ A fragment containing the NTHI0044 gene and 243 bp upstream including the promoter of the NTHI0044 gene and 165 bp downstream was amplified from genomic DNA of strain 86-028NP and ligated into pSPEC1 at BamHI and EcoRI restriction sites (Table S1). After confirming the insert sequence (pNTHI0044spec), strains were electroporated with pNTHI0044spec that had been methylated with CpG methylase (New England Biolabs, Ipswich, MA, USA) in a 0.1 cm cuvette (200 Ω, 2.5 kV, 25 μF). Cells were plated on chocolate agar containing 200 μg/mL spectinomycin ± 50 μg/mL ribostamycin, incubated overnight and the complemented mutant/resistant strains were obtained. These complemented strains were grown in the presence of spectinomycin for all experiments.

Results

Characterization of clinical isolates

To assess the genetic diversity of the strains used in this study, we determined the MLST of 20 clinical isolates that were isolated from different geographical regions (Table 1). Each of the isolates was a unique ST; 16 (and 4 unclassified) clonal complexes were represented among the 20 isolates. The eBurst analysis indicated these isolates were genetically diverse (data not shown).

Sequence conservation of acpP among strains of NTHi

acpP was chosen as a target for antisense oligomer inhibition because it is an essential gene (encoding an acyl carrier protein required for lipid biosynthesis) that has been successfully targeted by a variety of antisense oligomers in many other bacteria.^{27,31,32,36,49–52} We amplified the acpP genes and 10 bp upstream, determined the sequences and aligned the sequences among 20 clinical isolates of NTHi. The nucleotide sequences of the PNA target sites were identical in all 20 isolates, suggesting that our acpP-PNAs bind the target DNA and/or mRNA and work effectively.

Susceptibility of planktonic NTHi

We determined MICs of six acpP-PNA-peptides and two control-PNA-peptides for the 20 NTHi isolates. Table 1 shows MICs of our original PNA-peptide and our optimized PNA-peptide; our original PNA-peptide (acpP-PNA1-3′N) showed MICs that ranged from 6.6 to 11 mg/L (1.25–2.5 μmol/L) and our optimized PNA-peptide (acpP-PNA14-5′L) showed MICs that ranged from 2.9 to 5.7 mg/L (0.63–1.25 μmol/L). The other four acpP-PNA-peptides showed similar MIC results for NTHi isolates in planktonic form (data not shown). Both control-PNA-peptides showed no activity, with MICs >90 mg/L (>20 μmol/L) for all clinical isolates.

Susceptibility of biofilm-formed NTHi

MBEC is a measure of the capacity of an antimicrobial agent to eradicate a preformed biofilm. The MBEC of our original PNA-peptide (acpP-PNA1-3′N) is 179 mg/L (40 μmol/L) for 86-028NP (Table 3). We assessed several features including the PNA base sequence, the position of the conjugated membrane-penetrating peptide (3′ or 5′ end of the PNA), and the presence of a linker between the PNA and the penetrating peptide. Table 3 shows that a PNA with the base sequence that hybridizes to a region downstream of the acpP start codon (PNA base sequence 14) is more effective than a PNA with the base sequence that spans the start codon (PNA base sequence 1). In addition, a PNA-peptide that has the penetrating peptide conjugated to the 5′ end of the PNA with a linker is the most active in the various designs shown in Table 3. We conclude that the PNA base sequence, the position of the penetrating peptide and the presence of a linker all have an effect on the activity of PNA-peptides against NTHi biofilms.

Based on results with strain 86-028NP, we tested six clinical isolates with our original PNA-peptide (acpP-PNA1-3′N) and the PNA-peptide with the highest activity (acpP-PNA14-5′L). The latter, optimized PNA-peptide, acpP-PNA14-5′L was more active, showing MBEC values that were 4–8-fold lower than those for the original PNA-peptide (acpP-PNA1-3′N) (Table 4). We conclude that the optimal design of a PNA-peptide directed at acpP includes a PNA base sequence targeting the region downstream of the acpP start codon with a membrane-penetrating peptide conjugated to the 5′ end of the PNA with a linker.

Induction of resistance to PNA-peptides

After 15 overnight passages of strain 86-028NP with our original PNA-peptide, acpP-PNA1-3′N, at 1/4 MIC (2.8 mg/L, 0.63 μmol/L), a resistant strain was isolated. The resistant strain, 86-028NP-D15A, has three SNPs compared with the original strain 86-028NP based on genome sequence analysis. Of these, a SNP within NTHI0044 gene (ORF44), which is predicted to encode an ATP-binding protein of a conserved ABC transporter, introduced a stop codon (Table 5). The other SNPs were in intergenic regions including opsX-hxuC and NTHI0521-ompP1 (Table 5).

Strain 86-028NP-D15A showed high resistance to our original PNA-peptide acpP-PNA1-3′N, but was fully susceptible to the optimized PNA-peptide acpP-PNA14-5′L (Table 6). Note that the resistant strain 86-028NP-D15A was selected using the original PNA-peptide acpP-PNA1-3′N. We selected another resistant strain from clinical strain 3113-S after 20 overnight passages with the original PNA-peptide acpP-PNA1-3′N at 1/4 MIC (1.4 mg/L, 0.31 μmol/L). The MIC for the strain 3113-S-D20A increased 8-fold (45 mg/L, 10 μmol/L) and has an SNP in the NTHI0044 gene. The point mutation G1286A in the NTHI0044 gene resulted in an amino acid change (Gly429Glu), but did not introduce a stop codon (Table 5). Strain 3113-S-D20A has no SNPs in intergenic regions where we found SNPs in 86-028NP-D15A.

To select a resistant strain to the optimized PNA-peptide acpP-PNA14-5′L, strain 86-028NP was passaged with acpP-PNA14-5′L at 1/4 MIC (1.4 mg/L, 0.31 μmol/L) from days 1 to 15 and at 1/2 MIC (2.9 mg/L, 0.63 μmol/L) from days 16 to 30. Interestingly, the strain 86-028NP-D30A showed high resistance to the original PNA-peptide acpP-PNA1-3′N, but remained fully susceptible to the optimized PNA-peptide acpP-PNA14-5′L that we used to select the strain 86-028NP-D30A. An SNP was present in the NTHI0044 gene of strain 86-028NP-D30A, which introduces a stop codon (Table 5). All three resistant strains have no SNPs in the acpP target gene (Table 5).

Role of NTHI0044 in resistance to PNA-peptides

To assess the role of NTHI0044 in resistance, knockout mutants were constructed in NTHi prototype strain 86-028NP and in the clinical isolate 3113-S. Deletion of the NTHI0044 gene resulted in resistance to the original PNA-peptide acpP-PNA1-3′N in both strains. However, there is little or no effect of knocking out NTHI0044 on susceptibility to the optimized PNA-peptide, acpP-PNA14-5′L (Table 6).

Complementation of the NTHI0044 mutation was accomplished in the 86-028NP-ΔNTHI0044 as well as the resistant strain 86-028NP-D15A. Complementing the NTHI0044 gene in both knockout mutants fully restored susceptibility to acpP-PNA1-3′N (Table 6).

Discussion

In this study, we report that PNA-peptides targeting the acpP gene have antimicrobial effects on NTHi in planktonic and biofilm forms. All 20 NTHi clinical isolates were susceptible to the six different compositions of the acpP-PNA-peptides with MICs that ranged from 2.9 to 11 mg/L (0.63–2.5 μmol/L). There is no significant difference in MICs of the six acpP-PNA-peptides, indicating that all six acpP-PNA-peptides are effective against planktonic NTHi.

In contrast, three factors in the design of PNA-peptides affect susceptibility of biofilm NTHi: (i) base sequence of the PNA; (ii) position of the membrane-penetrating peptide; and (iii) presence of a linker between the PNA and the membrane-penetrating peptide (Tables 3 and 4). The length of the PNA base sequence is typically 10–12 bp, based on uptake properties and sequence specificity.²⁷ Thus, the position of PNA on the target gene is a critical factor affecting PNA efficacy in the design of the PNA base sequence. The ATG start codon is often the best target for many genes. However, this is not the case for the acpP-targeted PNA-peptides in the present study. For some antisense molecules, target positions that are upstream or downstream of the start codon improve activity.⁵³ Comparing the MBEC results in our study, the optimized PNA-peptide (acpP-PNA14, not including the start codon) is more effective than our original, designed PNA-peptide (acpP-PNA1) that includes the start codon (Table 3).

Conjugation of antisense molecules to a membrane-penetrating peptide increases their activity by mediating penetration into bacterial cells.^27,29–34 We evaluated PNAs with a penetrating peptide (RFFRFFRFFR) conjugated to the 3′ and 5′ ends of the PNAs. A PNA-peptide with the penetrating peptide conjugated to the 5′ end of the PNA (acpP-PNA1-5′N) is more active against NTHi biofilms than the same PNA-peptide that differs only by conjugation of the penetrating peptide to the 3′ end of the PNA (acpP-PNA1-3′N) (MIC 45 mg/L versus 179 mg/L, or 10 μmol/L versus 40 μmol/L). These results indicate that conjugation of the membrane-penetrating peptide to the 5′ end of the PNA is preferable to the 3′ conjugation.

Good et al.²⁷ described that egl, a flexible linker, may affect activity. Therefore, we studied the effect of a linker on activity of acpP-targeted PNA-peptides. We demonstrated that egl increased PNA efficacy using two PNAs with different base sequences (Table 3). A PNA-peptide that includes a linker (acpP-PNA1-3′L) is more active than the same PNA-peptide without linker (acpP-PNA1-3′N) (MIC 47 mg/L versus 179 mg/L, or 10 μmol/L versus 40 μmol/L).

We conclude that three characteristics play critical roles in the activity of PNA-peptides against biofilm-formed NTHi: (i) PNA base sequence at a site downstream of the start codon of the acpP gene; (ii) conjugating the membrane-penetrating peptide to the 5′ end of the PNA; and (iii) the presence of a linker between the PNA and membrane-penetrating peptide. Thus, acpP-PNA14-5′L is the most effective PNA-peptide, although we do not see its advantage against planktonic NTHi. The MBECs of our optimized acpP-PNA14-5′L ranged from 11 to 23 mg/L (2.5–5 μmol/L) for six randomly selected clinical isolates. The susceptibility gap between MIC and MBEC is narrow with this PNA-peptide. Therefore, with an appropriately designed PNA-peptide, NTHi biofilms can be eradicated by treatment with relatively low concentrations of PNA-peptide, which can be estimated from the MIC value for planktonic bacteria. Additional adjustments to the PNA-peptide are likely to result in further optimization of activity.

Because antimicrobial resistance causes treatment failure in clinical settings, it is important to investigate mechanisms of resistance in an effort to design countermeasures before the introduction of novel antimicrobials. Ghosal et al.⁵⁴ and Puckett et al.⁵⁵ reported resistance to antisense molecules in E. coli, and both conclude that the protein SbmA transports antisense molecules across the plasma membrane. Ghosal et al.⁵⁴ screened the KEIO collection⁵⁶ of 3985 single gene deletion mutants with their PNA and obtained two PNA-resistant mutants including ΔsbmA and ΔyghF. The NTHI0044 gene that we identified as a responsible gene for resistance to our PNA-peptides has no similarity to genes sbmA and yghF in E. coli, suggesting that NTHi has a different mechanism for PNA resistance compared with that in E. coli.

The NTHI0044 protein is predicted as an ABC transporter ATP-binding protein and may recognize and mediate uptake of specific PNAs, such as our original PNA-peptide (acpP-PNA1-3′N). It is difficult to explain why selection on our optimized PNA-peptide (acpP-PNA14-5′L) introduced resistance to the original PNA-peptide (acpP-PNA1-3′N), but not to the PNA-peptide on which the strain was passaged. Because selection with both PNA-peptides individually introduced an SNP in the same NTHI0044 gene, we speculate that the PNA-base-structure itself caused a mutation in the NTHI0044 gene, independent of the linker, membrane-penetrating peptide and base sequence. Alternatively, one might speculate that the NTHI0044 protein recognizes and mediates uptake of the membrane-penetrating peptide ‘RFFRFFRFFR’ regardless of its position. Our original PNA-peptide (acpP-PNA1-3′N) and optimized PNA-peptide (acpP-PNA14-5′L) have the same membrane-penetrating peptide, and thus introduced an SNP in the NTHI0044 gene. The NTHI0044 protein may be the sole system for uptake of the membrane-penetrating peptide conjugated to the 3′ end of the PNA. However, NTHi may possess more than one system for uptake of acpP-PNA14-5′L, suggesting that the 5′ positioning of the membrane-penetrating peptide and/or egl are associated with recognition and uptake by the different system of NTHi. Uptake by more than one system may also partially explain the result that our optimized PNA-peptide (acpP-PNA14-5′L) showed the lowest MICs and MBECs of our six PNA-peptides.

The MIC of acpP-PNA1-3′N against the resistant strain 86-028NP-D15A is 179 mg/L (40 μmol/L). The genome sequence of this strain differs in three SNPs compared with the susceptible starting strain before induction of resistance. Strain 86-028NP-ΔNTHI0044 is intermediate in resistance (MIC of 45 mg/L, 10 μmol/L) compared with the starting strain and the resistant strain, suggesting that the NTHI0044 gene induces only intermediate resistance, not full resistance. We speculate that the difference in MIC value may be caused by one or both of the other two SNPs. However, the other two SNPs in 86-028NP-D15A are in intergenic regions and not detected in 86-028NP-D30A, which also showed high resistance to acpP-PNA1-3′N (Tables 5 and 6). An alternative explanation is that serial passage may change gene expression without causing SNPs in the genome.

In conclusion, PNA-peptides have antimicrobial activity against NTHi in planktonic and biofilm forms. The optimized PNA-peptide (acpP-PNA14-5′L) shows greater activity and lower risk of resistance induction compared with several other acpP-targeted PNA-peptides. PNA-peptides represent a novel approach to eradicate selectively otopathogens from the nasopharynx and thus prevent otitis media. This approach has the important advantage over standard antimicrobial agents of sequence-specific antimicrobial activity for otopathogens, thus leaving normal flora undisturbed. Future work will focus on the effects of the PNA-peptides against intracellular NTHi and in vivo experiments.

Funding

This work was supported by National Institute on Deafness and Other Communications Disorders grant R21DC012917 (T. F. M. and B. L. G.), National Institute of Allergy and Infectious Diseases grant R01AI19641 (T. F. M. and M. M. P.), National Center for Advancing Translational Sciences award UL1TR001412 to University at Buffalo, and the Uehara Memorial Foundation (T. O.). The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Transparency declarations

None to declare.

Supplementary data

Table S1 is available as Supplementary data at JAC Online (http://jac.oxfordjournals.org/).

Acknowledgements

Our thanks to the following people for providing strains: Howard Faden and Diane Dryja, Buffalo, NY, USA; Janet Casey and Michael Pichichero, Rochester, NY, USA; Janet Gilsdorf, Ann Arbor, MI, USA; Aino Ruohola, Turku, Finland; and Ron Dagan, Beer Shiva, Israel.

References