Antisense Phosphorodiamidate Morpholino Oligomers Targeted to an Essential Gene Inhibit Burkholderia cepacia complex

Abstract

Background:

Members of the Burkholderia cepacia complex (Bcc) cause significant morbidity and mortality in patients with chronic granulomatous disease (CGD) and cystic fibrosis (CF). Many Bcc strains are antibiotic resistant requiring the exploration of novel antimicrobial approaches including antisense technologies, such as phosphorodiamidate morpholino oligomers (PMOs).

Methods:

Peptide-conjugated PMOs (PPMOs) were developed to target the acpP gene, encoding an acyl carrier protein thought to be essential for growth. Their antimicrobial activities were tested against different strains of Bcc in vitro and in infection models.

Results:

PPMOs targeting acpP were bactericidal against clinical isolates of Bcc (> 4 log reduction), whereas a PPMO with a scrambled base sequence (Scr) had no effect on growth. Human neutrophils (PMN) were infected with B. multivorans, and treated with AcpP PPMO. AcpP PPMO augmented killing compared to PMN alone ± Scr PPMO. CGD mice infected with B. multivorans were treated with AcpP PPMO, Scr PPMO or water at 0, 3 and 6 hours post-infection. Compared to water treated controls, the AcpP PPMO treated mice showed a ~80% reduction in the risk of dying by day 30 and relatively little pathology.

Conclusions:

AcpP PPMO is active against Bcc infections in vitro and in vivo.

Introduction

The Burkholderia cepacia complex (Bcc) is comprised of seventeen phenotypically similar but genetically distinct species[1]. These Gram-negative bacteria cause opportunistic infections in specific hosts such as those with chronic granulomatous disease (CGD)[2-4] or cystic fibrosis (CF)[3, 5]. Nosocomial outbreaks of Bcc infections have been associated with various contaminated solutions[6-8]. Chronic Bcc infection in CF contributes to progressive pulmonary decline[9, 10] as well as the rapid deterioration in lung function referred to as “cepacia syndrome”[11, 12]. In addition, many strains of Bcc found in patients with CF are intrinsically antibiotic resistant, further complicating therapy in these patients[13]. In one large study of Bcc isolates from patients with CF, 18% of isolates were resistant to all the antimicrobials tested; even the most active agents (minocycline, meropenem and ceftazidime) inhibited only ~20-40% of the isolates[14]. Therefore, the need for new antimicrobials for Bcc infections is urgent.

The phosphorodiamidate morpholino oligomers (PMOs)[15] are modified oligomers, whose nucleotides are coupled to morpholine rings linked by phosphorodiamidate linkages. The morpholine phosphorodiamidate backbone resists degradation by RNases while nucleotide bases bind to mRNA in a sequence-specific manner to prevent translation. The conjugation of various antimicrobial peptides to the PMO (herein called PPMO) has led to dramatically increased cell permeability[16]. These molecules have been used successfully in other bacteria such as Escherichia coli and Salmonella[17, 18]. In the case of E. coli, the gene targeted was acpP (acyl carrier protein), which is essential for bacterial growth[17, 19]. Acyl carrier proteins are important for lipid biosynthesis in both bacteria and plants[20]. We report the first use of PPMOs in members of the Bcc both in vitro and in vivo, by targeting acpP.

Methods

Growth of Bacteria

Isolates used in this study are listed in Table 1 and, except where indicated, were obtained from American Type Culture Collection (ATCC, Manassas, VA). Clinical isolates of Bcc were obtained from the lungs of patients with chronic granulomatous disease (CGD) who are followed longitudinally at the National Institutes of Health under the institutional review board-approved protocol #93-I-0119. B. cenocepacia HI2424 and J2315 were kindly provided by John Lipuma (Ann Arbor, MI). All strains were stored in the Microbank Bacterial storage system (Pro-Lab Diagnostics, Austin, TX) at −80°C until use. A bead was streaked onto a tryptic-soy agar w/5% sheep blood plate (Remel, Lenexa, KS) and grown at 37° C with 5% CO₂ until individual colonies formed. A single colony was inoculated into 10 mL of Luria-Bertani (LB) broth (Quality Biological, Gaithersburg, MD) and grown overnight at 37°C with shaking. The culture was harvested and washed three times in 150 mM NaCl and resuspended as described below.

Table 1.

Minimum Inhibitory Concentrations (MICs) of PPMOs in Various Bcc Isolates

	AcpP PPMO #1 (μM)	AcpP PPMO #2 (μM)	AcpP PPMO #3 (μM)	Scr PPMO (μM)
B. multivorans (CGD clinical isolate #1)	2.5	2.5	20	>80
B. multivorans (CGD clinical isolate #2)	2.5	2.5	40	>80
B. cepacia (CGD clinical isolate #3)	5	5	>40	>80
B. cepacia (CGD clinical isolate #4)	10	5	>40	>80
B. cenocepacia (CGD clinical isolate #5)	5	5	>40	>80
B. cepacia ATCC 25416	5	10	>40	>80
B. cenocepacia ATCC BAA-245	5	10	>40	>80
B. dolosa ATCC BAA-246	2.5	5	>40	>80
B. multivorans ATCC BAA-247	2.5	2.5	40	>80
B. pyrrocinia ATCC 15958	10	10	>40	>80
B. vietnamiensis ATCC BAA-248	>40	5	>40	>80
B. ambifaria ATCC BAA-244	>40	5	>40	>80
B. cenocepacia HI2424	5	5	>40	>80
B. cenocepacia J2315	5	10	>40	>80

Antisense oligomers

PMOs and PPMOs were synthesized as described[17] with the base and amino acid sequences shown in Table 2.

Table 2.

Peptide-PMO (PPMO) Sequences Used for These Studies

Name	AVI number NG-	Target	PMO Base Sequence (5′ to 3′)	Peptide Sequence
AcpP PPMO #1	05-0550	acyl carrier protein bases 4-141	TCG ATG TTG TC	RFFRFFRFFRXB2
Scr3 PPMO	05-0553	Scrambled sequence control	ATC GTT GCA TC	RFFRFFRFFRXB
AcpP PPMO #2	07-0348	acyl carrier protein bases −5 to 6	GTC CAT TAC CC	RFFRFFRFFRXB
AcpP PPMO #3	07-0349	acyl carrier protein	TCG ATG TTG TC	RXRRXRRXRRXRXB

¹Numbering from the first base of the start codon.

²X indicates 6-aminohexanoic acid, and B indicates beta-alanine.

³Scr indicates scrambled base sequence

In vitro and minimal inhibitory concentration (MIC) Studies

For in vitro studies, overnight cultures were washed and final pellets were resuspended in 1 mL of Mueller-Hinton broth. Concentrations were calculated using optical density at 600nm (OD₆₀₀) and previously determined cfu/OD₆₀₀ value for each strain. Cultures were diluted to a final concentration of 5 × 10⁵ cfu/mL in Mueller-Hinton broth. The PPMO or peptide alone was added to a final concentration of 20 μM. Triplicate 100 μl aliquots were transferred to a Costar 96-well flat bottom ultra-low attachment microplate (Corning, Incorporated, Corning, N. Y., United States) and incubated with shaking (200 rpm) at 37°C. Growth was monitored by OD₆₀₀ using a DTX880 Multimode Detector (Beckman Coulter, Fullerton, CA, United States). OD₆₀₀ readings were measured at 1-hour intervals. After 24 hours, bacterial cell viability (cfu/mL) was determined by plating dilutions of each sample in duplicate using a Whitley Automatic Spiral Plater WASP 2 (Microbiology International, Frederick, MD). Plates were incubated for 24-48 hours at 37°C with 5% CO₂, and colonies were enumerated on a ProtoCOL Automated Colony Counter (Microbiology International). Dilutions were plated by hand when the cfu/ml fell below the minimum threshold of the WASP 2 plater. Experiments were repeated at least five times. MIC determinations were performed using established broth microdilution methods[21].

Isolation of human neutrophils (PMN)

Human blood samples used in these studies were collected after informed consent from normal subjects (NIH Protocol 99-CC-0168) and CGD patients (NIH Protocol 93-I-0119). Venous blood was collected into tubes containing acid citrate dextrose (Solution A, BD Biosciences) to prevent coagulation. Purification of neutrophils was performed as previously described[22]. Briefly, erythrocytes were removed from citrate-anticoagulated blood using 3% dextran sedimentation and the resulting leukocytes subjected to hypotonic lysis to remove the remaining erythrocytes. The leukocytes were fractionated over a discontinuous Percoll gradient and the PMN-enriched pellet washed in HBSS(−). Neutrophils were counted in 3% acetic acid using a hemocytometer. Purity typically was greater than 95%. To isolate serum, blood was collected into SST tubes (BD Biosciences) and processed per manufacturer's instructions.

Neutrophil Killing Assays

Assays were performed in non-pyrogenic polystyrene 96-well plates (BD Falcon Tissue Culture Plates, San Jose, CA) pre-coated with 1.6% human albumin (Talecris Biotherapeutics, Research Triangle Park, NC) for one hour at 37°C with 5% CO₂. PMN and Bcc were incubated at a multiplicity of infection (MOI) of 1:1 in a total volume of 200 μl that contained RPMI 1640 with L-Glutamine (Invitrogen), 25 mM HEPES pH 7.5 (Thermo Scientific HyClone) in the presence or absence of 10% autologous serum. Plates were centrifuged at 500 × g for 8 minutes to synchronize phagocytosis and then incubated at 37°C with 5% CO₂. After one hour of incubation, the designated PPMO (20 μM) or buffer control was added. At the indicated time points, the cells were placed on ice and treated with 0.1% saponin (final concentration) for 10 minutes followed by mechanical shearing with a 28-gauge 0.5cc syringe. Supernatants were diluted and spread on LB plates then cultured and enumerated as described above.

Mouse challenge studies

The mouse model for X-linked CGD was used for infection studies. These mice lack gp91^phox and are susceptible to infections with CGD pathogens including the Bcc. gp91 KO mice (Jackson Lab, Bar Harbor, Maine) backcrossed to the C57BL/6 background [23] were maintained at the NIAID animal facility under specific pathogen-free conditions. The Institute's Animal Care and Use Committee approved all experiments. Mice were 12-24 weeks old, and were sex-matched for each set of experiments. Mice were injected intraperitoneally with 0.1 mL of 5 × 10⁷ cfu/mL of B. multivorans. This was followed immediately by an intraperitoneal dose of H₂O control or PPMO treatment (0.1 mL containing 2 mg/mL PPMO). Further treatments were given at 3 and 6 hours after the initial inoculum, for a total of 3 treatment doses. Mice were followed for signs of morbidity (hunched posture, ruffled fur, reluctance to move) and were euthanized according to protocol. A subset of mice from each group (n=8) had blood drawn for quantitative culture determination 24 hours post-infection and treatment. Three independent mouse challenge experiments were performed. AcpP PPMO and scrambled PPMO groups were included in all 3 experiments. For histopathologic analysis, organs (lymph nodes, spleen, liver) were fixed in 10% phosphate-buffered formalin before sectioning and staining.

Design of PPMOs

acpP is highly sensitive to inhibition by antisense oligomers in a variety of Gram-negative bacteria, including E. coli[19, 26], Salmonella enterica[17, 18], Pseudomonas aeruginosa, Acinetobacter baumannii, and Klebsiella pneumoniae (B. Geller, unpublished). PPMOs were designed with complementary 11-base sequences immediately downstream of, or spanning the start codon of acpP (Table 1). This length and these positions have been shown to be optimum for inhibition, with minimal effect on host gene expression[25]. All oligomers were conjugated to one of two membrane-penetrating peptides, (RFF)3RXB or (RXR)4XB (X is 6-aminohexanoic acid; B is β-alanine). These two peptides are some of the most effective in enabling penetration of PMOs in E. coli[16], S. enterica[17, 18], and other bacteria (B. Geller, unpublished). PPMOs were designed with base sequences that are complementary to regions near the start of acpP (Table 1).

Statistical Methods

For the in vitro experiments, cell viability (cfu/mL) at 24 hours was calculated for each species and PPMO treatment condition, including no treatment. The difference in log(count+0.5) between cultures treated with AcpP PPMO versus untreated is analyzed by paired t-test. Values were linearly transformed by the addition of a constant, 0.5, to allow statistical testing of the log counts, as samples were completely sterilized by AcpP PPMO (i.e., zero cfu/ml). For the neutrophil killing experiments, mean (and 95% confidence intervals) for the difference in log10(count) between 1) serum alone versus pmn+serum, 2) pmn+serum versus pmn+serum+AcpP PPMO, and 3) pmn+serum+scrambled PPMO and pmn+serum+AcpP PPMO are calculated at 2 hr, 4 hr and 24 hr. Pairwise differences were analyzed at each time using the paired t-test. For the mouse challenge studies, time to death or euthanization during the first 30 days is recorded. Kaplan-Meier survival curves were calculated for data combined from three experiments. A Cox proportional hazards regression stratified on experiment day was fit to estimate the mortality hazard ratio for each treatment comparison. Treatment differences in the 24-hour blood counts (cfu/ml) were tested using the Wilcoxon rank-sum test. Cox proportional hazards regression models, both unadjusted and adjusted for treatment were fit to assess the association between increases in blood cfu/mL and mortality.

Results

Inhibition of Burkholderia species in culture by PPMOs

For initial growth curves, the following PPMOs were tested: AcpP PPMO #1 (which targets acpP), Scr PPMO (scrambled sequence control) and the (RFF)₃RXB peptide alone. Cultures of B. multivorans (CGD clinical isolate #1), B. cenocepacia (CGD clinical isolate #5), and B. cepacia (CGD clinical isolate #3) were grown for 24 hours in the presence or absence of 20 μM PPMO prior to enumeration of viable CFU(Table1). Regardless of the species, all cultures without PPMO increased by approximately 4 logs after 24 hours of growth (cfu/ml was 5 × 10⁵ at t=0) (Figure 1). There was no appreciable difference in growth of the strains when incubated with or without the Scr PPMO. However, AcpP PPMO #1 against acpP was bactericidal in culture, with a >4-log reduction in cfu at 24 hours (p<0.0001) compared to the 0 h culture. In all samples from cultures treated with AcpP PPMO #1, there were no detectable colonies when plated undiluted (limit of detection 10 cfu/ml).

Figure 1.

The AcpP PPMO is bactericidal in vitro against members of the Bcc. B. multivorans (CGD clinical isolate #1, Panel A); B. cepacia (CGD clinical isolate #3, Panel B); Panel C: B. cenocepacia (CGD clinical isolate #5, Panel C) were grown overnight alone, or in the presence of PPMOs. Scr PPMO = scrambled sequence control; AcpP PPMO #1 = acyl carrier protein PPMO; Peptide only = (RFF)₃RXB peptide without oligomer where X indicates 6-aminohexanoic acid and B indicates beta-alanine. Each bar represents the mean cfu/mL from 5 independent experiments. ** p<0.0001 compared to T0 Burkholderia only (paired t-test). The 24-hour counts for AcpP PPMO were 0 cfu/ml for all observations in panels A-C, with the limit of detection indicated as 10 cfu/ml.

Minimum inhibitory concentrations (MICs) were determined for various PPMOs using a panel of laboratory and clinical isolates of Bcc. As shown in Table 1, Acp PPMO #1 and AcpP PPMO #2 were potent, with broad activity against many species of Bcc. For AcpP PPMO #2, the MICs ranged from 2.5 - 10μM. AcpP PPMO #1 was effective against all strains tested except B. vietnamensis and B. ambifaria. AcpP PPMO #3 that was attached to the (RXR)₄XB peptide was significantly less effective than those attached to (RFF)₃RXB.

The acpP gene is well conserved across members of the Bcc, and the region targeted by the PPMOs is identical in many of the species such as B. cenocepacia, B. multivorans, and B. dolosa (Figure 2). In addition, the gene has high conservation in other members of the Burkholderia genus (Figure 2). However, for B. vietnamensis and B. ambifaria there is a 1-base mismatch in the overlapping sequence targeted by both PPMOs. The mismatch with AcpP PPMO #1 is positioned at the third base from the 3′end of the PPMO, whereas with Acp PPMO #2 it occurs at the 5’end.

Figure 2.

Sequence alignment of the acpP gene in species of Bcc. There is high homology of the 5′ acpP gene sequence within the genus Burkholderia. Shown are the acpP sequences for various species of Bcc and other members of Burkholderia. The start ATG is in bold and underlined. AcpP PPMO #1 and AcpP PPMO #2 are two PPMOs designed against the acpP gene. Small letters in red represent a base that has a mismatch compared to the PPMO sequence. An asterisk represents strains where PPMOs were tested.

AcpP PPMO enhances neutrophil killing of B. multivorans

To further test the effect of PPMOs in ex vivo and in vivo infection models, we used a B. multivorans strain (CGD clinical isolate #1) as it was one of the isolates that showed the lowest MIC values. Human neutrophils (PMNs) from healthy blood donors were incubated with serum and B. multivorans for 1 hour followed by the addition of AcpP PPMO #2, Scr PPMO, or tissue culture medium only. A control well included serum and bacteria only. Bacterial cfu from each condition were measured at 2, 4 and 24 hours post-infection. B. multivorans grew 2 logs by 24 hours in the presence of serum alone (Figure 3). At 2 hours post infection, normal PMN in 10% autologous serum had reduced bacterial cfu by 1.35 log (95% CI: 1.16,1.54)(Panel A). AcpP PPMO #2 enhanced PMN killing, and by 2 hours post infection, cfu were further reduced by 0.80 logs (95% CI: 0.67-0.92; p<0.0001)(Panel A). The enhanced killing continued over time, and by 4 hours AcpP PPMO #2 caused an additional 1.17 log (95% CI: 0.98-1.36; p<0.0001) reduction in cfu compared to PMN and serum alone. The difference in killing is most pronounced at 24 hours, where there was a 2.81 log difference (95% CI: 2.70-2.91; p<0.0001) in cfu between PMN in serum vs. the addition of AcpP PPMO #2 (Panel A). The addition of the scrambled control PPMO had no appreciable effect on neutrophil killing compared with serum and PMN alone.

Figure 3.

The AcpP PPMO #2 enhances PMN killing of B. multivorans (CGD clinical isolate #1). The symbols represent time points of cell harvest (2, 4 and 24 hours) and are displayed as the mean cfu/mL. Open circles = B. multivorans alone; Open squares = PMN + B. multivorans; Black circles = PMN + B. multivorans + Scr PPMO; Black squares = PMN + B. multivorans + AcpP PPMO #2. Each symbol represents the mean of 6 normal donors over 3 separate experiments (Panel A) and 3 CGD donors on 2 separate experiments (Panel B) and error bars represents the range.

CGD PMN were defective in their ability to kill B. multivorans (Figure 3, Panel B) with 24 hours counts higher than the starting inoculum. Despite this, the addition of AcpP PPMO #2 enhanced the clearing of B. multivorans in infected PMN from CGD patients by 1.95 logs (95% CI: 1.57,2.33) at 24 hrs.

Acp PPMO protects gp91phox KO mice from B. multivorans infection

To test whether the AcpP PPMO could provide a survival benefit in vivo, challenge experiments were performed using gp91phox KO mice. AcpP PPMO #2 conferred a consistent survival advantage compared to either scrambled PPMO or water treatment (Figure 4). Overall, 55% of AcpP treated mice survived to day 30 (range: 12-100% survival in each of three independent experiments), compared to 25% in the scrambled and 11% in the water treated groups, respectively. A Cox proportional hazards regression model was used to estimate the risk of death as a function of treatment. The mortality hazard ratio for the AcpP PPMO #2 group compared to the water group is 0.21 (95% CI: 0.10,0.43), i.e., a 79% reduction in risk (p<0.0001). The difference between the Scr PPMO and water groups was also significant (p=0.023) with a hazard ratio of 0.44 (95% CI: 0.22,0.89). The difference between the AcpP PPMO #2 and the Scr PPMO groups was also significant (p=0.037) with a hazard ratio of 0.47 (95% CI: 0.23, 0.96). The results were similar when looking at 7-day mortality (data not shown).

Figure 4.

AcpP PPMO increases survival of B. multivorans (CGD clinical isolate #1) infected mice. Mice were age- and sex-matched for all experiments. Mice were injected with 5 × 10⁶ cfu intraperitoneally (ip) followed by an immediate ip dose of 200μg (0.1 mL) of PPMO treatment (Black squares = AcpP PPMO #2 tx; Black circles = Scrambled PPMO tx) or Water alone (Black diamonds). **AcpP PPMO treatment had a reduction in 30-day mortality compared to water treatment with a hazards ratio (HR) of .21 (p<0.0001); and to scrambled PPMO treatment with a HR of .47 (p=0.037). *Scr PPMO treatment had a reduction in 30-day mortality compared to water treatment with a HR of .44 (p=0.023).

Blood was collected 24 hours post-infection from a subset of mice in each treatment group (n=8) and quantitatively analyzed for bacteria. The median cfu/mL was 9900 for the water group, 10,700 for the Scr PPMO group and 620 for the AcpP PPMO #2 group. The water group had higher 24-hour blood counts compared to AcpP (p=0.0009; Wilcoxon rank-sum). The scrambled group had higher counts than the AcpP group, but the difference was only marginally significant (p=0.082). There was no difference in 24-hour blood counts between the water and scrambled groups (p=0.798) A Cox proportional hazards regression model shows an association between increases in blood cfu/mL and mortality. A 1-log increase in the blood cfu/mL was associated with a 2.35 fold increase in the hazard for death (95% CI 1.76,3.14). When adjusted for treatment group, 24-hour blood counts were still a significant predictor of mortality risk (p<0.0001), with a 1-log increase in counts being associated with a hazard ratio of 2.31 (95% CI: 1.77,3.02).

A pathologist assessed the histopathologic changes in twenty-seven mice representing all three-treatment groups in a single-blind analysis. Severe splenitis was characterized by the replacement of the normal hemapoietic elements of the white pulp by many mature, degenerating, and/or necrotic neutrophils admixed with numerous macrophages and few plasma cells. In addition, severely affected spleens often had fibrin thrombi. Mildly affected spleens had normal red pulp architecture with a few small areas of degenerating or necrotic neutrophils and no thrombi. Thirteen of 27 mice had severe suppurative to necrosuppurative splenitis; 10/27 had mild acute splenitis and 4/27 had no splenitis. These pathologic changes were then correlated to the treatment group from which the particular mouse had come. Seventy-percent (9/13) of the animals with severe splenic disease were in either the scrambled or water treatment groups. In addition, seventy-percent (7/10) of the animals with mild disease were in the scrambled or water-treatment groups. However, seventy-five percent (3/4) of the mice with no splenitis were in the AcpP-treatment group. Figure 5 shows sections of spleen and lymph nodes from a surviving mouse compared with a mouse that died rapidly. The spleen from a mouse treated with AcpP PPMO #2 that survived to day 30 had normal splenic architecture, while the spleen of a water-treated mouse that survived until day 4 showed suppurative splenitis. Similar findings are seen in the mesenteric lymph nodes.

Figure 5.

Histopathologic findings in mice infected with B. multivorans. Spleen and lymph nodes were fixed, sectioned and stained with hematoxylin and eosin. An AcpP PPMO-treated mouse that survived until day 30 has normal splenic architecture (Panel A; 2.5x magnification). The spleen of a water-treated mouse that survived until day 4 has expansion of red pulp with suppurative splenitis, which effaces normal hematopoietic elements (Panel B; 2.5x magnification). Panel C shows an aggregate of neutrophils at 40x. A section of the mesenteric lymph node from the same AcpP PPMO-treated mouse (Panel D) with normal architecture. The mesenteric lymph node from the water-treated mouse has multifocal suppurative lymphadenitis in the outer cortex of the node (Panel E; 2.5x magnification). Panel F shows a focus of suppurative inflammation from the affected lymph node at 40x.

Discussion

Burkholderia cepacia complex infections are frequently seen in CGD and CF[2, 3, 5] and are challenging in part because of resistance to antimicrobials. We set out to target acpP in Bcc, assuming this was likely to be essential for survival of the pathogen. Using PPMOs, AcpP has been shown to be essential for growth of E. coli and Salmonella enterica serovar Typhimurium in vitro[17]. AcpP is also essential for growth in Pseudomonas aeruginosa[24] and has recently been suggested as a potential target for antimicrobials in Bcc species[25]. Our results suggest that the acpP gene in Bcc is essential for growth in vitro. The effect of the AcpP PPMO was sequence-specific, as a scrambled sequence PPMO did not inhibit growth of Bcc, nor did incubation with the peptide alone. In vitro activity was seen in 5/7 species of Bcc with AcpP PPMO#1 and 7/7 species of Bcc with AcpP PPMO#2, as expected from the sequence conservation in this region of the acpP gene.

The difference in efficacy between AcpP PPMO #1 and Acp PPMO #2, both of which target acpP, using B. vietnamensis or B. ambifaria as an indicator, appears to be the result of the position of the 1-base mismatch between PPMO and target. The alignment of the target region (Figure 2) suggests that a 1-base mismatch can be tolerated at the extreme 5′end of the PPMO, but not between the ends or perhaps within 3 bases of the 3′end.

In Bcc, the (RFF)₃RXB peptide is much more efficient than the (RXR)₄XB peptide in promoting the antisense effect of the PPMOs. Presumably this is related to differing abilities of the peptides to penetrate the outer membrane of Bcc. In E. coli, both of these peptides efficiently promote entry of PPMOs[16]. Our results with Bcc suggest that the outer membrane of Bcc may differentially affect which peptide can successfully penetrate the organism.

Antimicrobials can demonstrate potency in vitro and then fail in cell culture models or in vivo studies. We used isolated PMN from both normal and CGD patients to determine if the addition of PPMO could augment ex vivo neutrophil killing of bacteria. The AcpP PPMO augmented PMN killing in a time-dependent manner, and this augmentation increased over time up to 24 hours post-treatment. Interestingly, CGD PMN, which are less able to kill Bcc, appreciably reduced B. multivorans only when the AcpP PPMO was added. This effect was probably driven by the presence of the PPMO itself, as indicated by the increase of cfu in control cultures and the decrease of cfu in the AcpP PPMO-treated culture at 24 hours post-treatment.

AcpP PPMO provided a survival advantage in CGD mice following B. multivorans challenge. Although there was variability from experiment to experiment, the AcpP PPMO provided a consistent survival benefit, although mice treated with the scrambled PPMO showed a mild survival advantage as well. However, no activity of scrambled PPMO was seen in in vitro assays or PMN killing experiments. When infected mice were treated with the peptide alone, none survived (data not shown). Importantly, surviving mice treated with the AcpP PPMO showed few pathological changes. The majority of mice showing the most severe pathologic changes had received either water or scrambled PPMO, a finding that further supports the survival differences.

Although antisense molecules have been around for years, dramatic increases in efficacy have been conferred by covalently attaching membrane-penetrating peptides[17, 26]. Tilley et al. demonstrated that the attachment of a peptide to the PMO increased the efficacy of a given PMO by ~50-100 times in mice infected with E. coli[17, 26]. The reason for this improved efficacy is likely due to better bacterial uptake of the oligomers[17]. PPMOs also have efficacy against viruses such as influenza[27], West Nile virus[28], Ebola[29] and members of the Picornaviridae[30]. Antisense molecules also show activity against other bacteria such as Salmonella enterica[17, 18], Mycobacterium smegmatis[31] and Klebsiella pneumoniae[32].

Antisense molecules such as PPMOs have potential advantages as therapeutics for infectious diseases. They can target specific genes of interest, which can help identify genes essential for survival or virulence factors required for infection in vivo. The short length required for inhibition of bacterial genes is insufficient for inhibition of eukaryotic genes[33], thus preventing off-target effects on host tissue. Antisense molecules could also target antibiotic resistance genes, making resistant organisms susceptible[34]. Further studies must address routes of PPMO delivery, including oral, intravenous, and pulmonary administration. PPMOs can be gene-specific inhibitors of the Bcc, and acpP may be a reasonable target for future therapeutic development.