Efficacy of a Pseudomonas aeruginosa serogroup O9 vaccine

ABSTRACT There are currently no approved vaccines against the opportunistic pathogen Pseudomonas aeruginosa. Among vaccine targets, the lipopolysaccharide (LPS) O antigen of P. aeruginosa is the most immunodominant protective candidate. There are 20 different O antigens composed of different repeat sugar structures conferring serogroup specificity, and 10 are found most frequently in infection. Thus, one approach to combat infection by P. aeruginosa could be to generate immunity with a vaccine cocktail that includes all these serogroups. Serogroup O9 is 1 of the 10 serogroups commonly found in infection, but it has never been developed into a vaccine, due in part to the acid-labile nature of the O9 polysaccharide. Our laboratory has previously shown that intranasal administration of an attenuated Salmonella strain expressing the P. aeruginosa serogroup O11 LPS O antigen was effective in clearing bacteria and preventing mortality in mice following intranasal challenge with serogroup O11 P. aeruginosa. Consequently, we set out to develop a P. aeruginosa serogroup O9 vaccine using a similar approach. Here, we show that Salmonella expressing serogroup O9 triggered an antibody-mediated immune response following intranasal administration to mice and that it conferred protection from P. aeruginosa serogroup O9 in a murine model of acute pneumonia.

lipid A embedded in the outer membrane, the core oligosaccharide, and the O antigen polysaccharide, which extends out from the surface of the bacterial cell.For P. aeruginosa, 20 different International Antigenic Typing System serogroups are recognized based on the expression of the O antigen portion.All of the O antigen structures have been determined (8), with each serogroup possessing subtype strains with subtle variations, leading to over 30 subtypes (9).The serogroup-specificity of the O antigen suggests that a comprehensive P. aeruginosa LPS-based vaccine would need to encompass all these subtypes.Fortunately, numerous studies have found 10 serogroups to be most common in various types of infections (10)(11)(12)(13).
One attractive approach would be to develop a cocktail of the most common LPS serogroups as a vaccine.However, polysaccharides are generally considered poor immunogens and alone do not elicit a robust immune response [reviewed in reference (14)].Because of this, many of the currently available polysaccharide-based vaccines are polysaccharide-protein conjugates (15).For P. aeruginosa, in the 1980s-1990s, the Swiss Serum and Vaccine Institute developed and tested an octavalent conjugated vaccine with eight different P. aeruginosa serogroups covalently coupled to the exotoxin A antigen of P. aeruginosa (16).While the clinical results of these studies with this octavalent vaccine have not been reported, this vaccine did not contain serogroups O8 or O9 because they contain internal ketosidic linkages (17) and thus cannot be adequately separated from the toxic lipid A component using acid hydrolysis needed to conjugate to the protein carrier.More recently, Nasrin et al. (13) used a "Multiple Antigen Presenting System" based on high molecular weight polysaccharides (18) and targeted eight of the most common P. aeruginosa O antigen serogroups (13).Serogroups O8 and O9 were also missing from this system.
Consequently, we set out to develop a vaccine for one of these "neglected" O antigen serogroups of P. aeruginosa.Our laboratory has previously shown that intra nasal administration of an attenuated Salmonella strain expressing the Pseudomonas aeruginosa serogroup O11 LPS O antigen was effective in clearing and preventing mortality in mice following intranasal challenge with serogroup O11 P. aeruginosa (19).Here, we show that Salmonella-expressed serogroup O9 can trigger an antibody-medi ated immune response following intranasal administration that conferred protection in a murine model of acute pneumonia.

Cloning and expression of P. aeruginosa serogroup O9 O antigen on S. typhimurium
Genomic DNA was isolated from P. aeruginosa serogroup O9 strain PAO9 (provided by Gerald B. Pier, Harvard Medical School, Boston, MA), using standard procedures.Genomic DNA was randomly sheared through a syringe needle and was end-repaired and cloned into pWEB::TNC (Epicentre Technologies, Madison, WI), followed by packaging into MaxPlax lambda packaging extracts.The lambda particles were used to infect Escherichia coli EP105.Colonies were absorbed with mouse monoclonal antibodies to P. aeruginosa serogroup O9 (Rougier Bio-Tech Ltd.).Colonies reacting with antisera were separated with anti-mouse antibodies bound to magnetic beads (Dynabeads; Thermo Fisher Scientific), followed by magnetic bead separation using a mini-magnetic particle separator (CPG Inc., Lincoln, Park, NJ).Positive colonies were selected by colony immunoblot using an anti-serogroup O9 mouse monoclonal antibody.The serogroup O9 locus was then cloned from the pWEB::TNC plasmid into the broad host range cosmid vector, pLAFR376 (provided by Laurence Rahme, Massachusetts General Hospital, Boston, MA).To do this, plasmid DNA from a positive colony was digested into approxi mately 20-25 kb fragments with EcoRl and ligated to completely digested pLAFR376.The ligation reactions were packaged as bacteriophage lambda particles with the Stratagene Gigapack XL-11 packaging system and used to infect E. coli HB101.Trans formed cells were selected on Luria broth (LB) media containing [tetracycline (Tet) 10 µg/ mL], where serogroup O9-positive clones were identified by colony immunoblots using the anti-serogroup O9 mouse monoclonal antibody.The positive clone that exhibited the strongest reaction was grown in broth culture for further characterization.This plasmid was isolated and designated pLAFRO9.The construct was confirmed by enzyme digestion and DNA sequencing (Emory Integrated Genomics Core).Plasmid sequences were assembled using PlasmidSPAdes (3.13.1), and the presence of the serogroup O9 locus was confirmed by BLAST against a reference locus accession AF498420.1.Plasmid pLAFRO9 was transferred to Salmonella typhimurium strain SL3261 by P22 transduction using S. typhimurium LB5010 as an intermediate host, as described previously (20), and expression was confirmed by silver-stained gel and Western immunoblot of extracted LPS.

Preparation of bacterial strains used for immunization and infection
S. typhimurium SL3261 containing pLAFRO9 (vaccine) or SL3261 containing the cosmid pLARF376 (vector) were used for immunization.All strains were grown overnight in LB supplemented with 10 µg/mL Tet.Both strains were subcultured and grown to an OD 650 of 0.5.Bacteria were then washed twice and resuspended in phosphate-buffered saline, pH 7.4 (PBS).For infection, P. aeruginosa serogroup O9 was grown on Difco Pseudomo nas Isolation Agar (PIA) overnight at 37°C and suspended in PBS to an OD 600 of 0.5, corresponding to ~10 9 colony-forming units (CFU)/mL.Inocula were adjusted spectro photometrically to obtain the desired immunization or challenge dose in a volume of 20 µL.

LPS extraction, SDS-PAGE, and Western immunoblotting
LPS extracts of P. aeruginosa and Salmonella organisms were prepared as described in Hitchcock and Brown (21) and separated by sodium dodecyl sulfate (SDS)-polyacryla mide gel electrophoresis (PAGE) as described (19) with a Novex X-Cell Surelock mini cell system (Invitrogen, Carlsbad, CA).Tris-bis-polyacrylamide gels (12.5%) were cast in 1.0 mm Invitrogen cassettes.After PAGE separation was completed, lysates were electroblotted onto Trans-Blot 0.2-μm-pore-size pure nitrocellulose membranes (Bio-Rad Laboratories, Hercules, CA) by use of a Bio-Rad Mini Trans-Blot electrophoretic transfer cell.Membranes were blocked and then probed with Pseudomonas serogroup O9-spe cific rabbit polyclonal antibodies (Denka Seiken, Tokyo, Japan), followed by incubation with anti-rabbit secondary antibodies conjugated to alkaline phosphatase (Sigma).Reactions were visualized by the addition of Sigma fast 5-bromo-4-chloro-3-indolylphos phate-nitroblue tetrazolium.

Immunization with S. typhimurium SL3261 and P. aeruginosa challenge
The use of animals in this study was reviewed and approved by the University of Virginia Institutional Animal Care and Use Committee (IACUC) under protocol num ber 2844-02-11.All mice were kept under specific pathogen-free conditions, and all guidelines for humane endpoints were strictly followed.All animal experiments were conducted in accordance with the "Public Health Service Policy on Humane Care and Use of Laboratory Animals" by the NIH, the "Animal Welfare Act and Amendments" by the USDA, and the "Guide for the Care and Use of Laboratory Animals" by the National Research Council (NRC).Female BALB/c mice, five to six weeks old [Harlan Laboratory (Indianapolis, IN) or Jackson Laboratory (Bar Harbor, ME)] were immunized intranasally as described (19).Prior to vaccination, mice were anesthetized by intraperi toneal injection of 0.2 mL of xylazine (1.3 mg/mL) and ketamine (6.7 mg/mL) in 0.9% saline.For intranasal vaccination, mice were given 1 × 10 7 or 1 × 10 9 CFU of either SL3261 (pLAFRO9) (vaccine) or SL3261 (pLAFR376) (vector) intranasally in a 20-µL volume (10 µL per nostril).A booster dose was performed using the same protocol as the initial vaccination, approximately 14 days after the initial vaccination.
For the bacterial challenge, anesthetized mice were given 20 µL of P. aeruginosa strain PAO9 prepared as described above.For in vivo protection experiments, infected mice were observed over five days, and animals that were moribund following infection or in any way appeared to be under acute distress were humanely euthanized and were included as non-survivors in the experimental results.

Collection of sera
Sera were collected at four weeks post-immunization as described (19).Briefly, blood was obtained by nicking the lateral tail vein of mice; the blood sat at room temperature for about 4 hours and was then placed at 4°C overnight.Serum was removed from the red blood cell pellet and spun at 1,700 rpm for 10 minutes.Samples were stored at −80°C.

ELISA analysis of P. aeruginosa PAO9 lipopolysaccharide expression from S. typhimurium
Enzyme-linked immunosorbent assay (ELISA) analysis was performed on sera as described (19).Briefly, 96-well microtiter plates were coated with ~5×10 7 CFU/well of P. aeruginosa strain PAO9, incubated overnight at 4°C, washed with PBS plus 0.05% Tween 20 (PBS-T), blocked with PBS supplemented with 2% bovine serum albumin (PBS-B), and then washed with PBS-T again.Serum samples were serially diluted in PBS-B, and 100 µL were placed into each well in PAO9-coated plates in duplicate.Pseudomo nas serogroup O9-specific rabbit polyclonal antibodies were used as positive controls.
After overnight incubation at 4°C, the plates were washed three times with PBS-T and air-dried.Secondary antibodies [anti-mouse total IgG, IgG1, IgG2a, IgG2b, IgG3, or IgM conjugated to alkaline phosphatase (Southern Biotechnology Associates, Inc., Birming ham, AL)] diluted 1:5,000 in PBS-B were then added to individual plates and incubated at 37°C for 1 hour.Plates were developed in the dark for 1 hour with 1 mg/mL 4-nitrophenyl phosphate in substrate buffer (24.5 mg MgCl 2 , 48 mL diethanolamine per 500 mL; pH 9.8); development was stopped by adding 50 µL 3 M NaOH.Plates were read using a plate reader at 405 nm.Data were collected using the SOFTmax PRO software (Molecular Devices Corp., Sunnyvale, CA) and then transferred to the GraphPad Prism version 6.0 software (GraphPad Software, San Diego, CA) for analysis.For the Ig titer determination, total serum IgG or IgM absorbance readings were adjusted by subtraction of values obtained from the blank; the x-intercept defined the endpoint titer and was represented as the reciprocal dilution.IgG subtype quantification for serum samples was based on standard curves that were designed for each antibody isotype by use of GraphPad Prism version 6 software.

Detection of bacterial loads
For bacterial load quantification, mice were euthanized at 24 hours post-infection, and nasal wash and whole lungs were collected aseptically.For the nasal wash, an 18-G catheter was placed at the oropharyngeal opening of the mouse, and 1.0 mL of PBS-B was flushed through the nasal passage and collected.Whole lungs were collected from each mouse, weighed, and homogenized in 1 mL of PBS.Tissue homogenates were serially diluted and plated on PIA, and CFU determination was made 16 to 18 hours later.Final results were expressed as CFU/mL for nasal washes and CFU/g for lung tissues (19).

Passive immunization and infection
All animal procedures were conducted according to the guidelines of the Emory University Institutional Animal Care and Use Committee (IACUC), under approved protocol number DAR-2003421-042216BN. The study was carried out in strict accord ance with established guidelines and policies at Emory University School of Medicine, and recommendations in the Guide for Care and Use of Laboratory Animals of the National Institute of Health, as well as local, state, and federal laws.For passive immu nization, 6-to 8-week-old female BALB/c mice were anesthetized and infected via the intranasal route with ~1 × 10 9 CFU/mouse as described.Antisera collected from PBS-, vector-, or vaccine-immunized mice were delivered to the mice (5 µL /nostril) immediately after infection.Mice were euthanized at 24 hours post-infection and nasal washes and whole organs were collected aseptically.Nasal wash collection was performed as described.Lung, liver, and spleen tissues were collected from each mouse, weighed, and homogenized in 1 mL of PBS.Tissue homogenates were serially diluted and plated on PIA, and CFU determination was made 18 hours later.Final results were expressed as CFU/mL for nasal washes and CFU/g for other organ tissues.

Opsonophagocytic killing assay
A luminescent opsonophagocytosis assay was performed according to the method as described (22).Immune sera from mice vaccinated with 10 7 or 10 9 CFU of SL3261 (pLAFRO9) were used, and threefold serial dilutions ranging from 1:100 to 1:59,049 were tested in triplicates.Sera collected from vector-and PBS-treated mice were also tested using the same dilutions.A constitutively bioluminescent P. aeruginosa PAO9 strain was constructed by immobilizing the plasmid pUC18miniTn7T-lux-Tp into the wild-type PAO9 strain to generate PAO9 P1-lux according to the method described (23).Briefly, this assay used 25 µL of each opsonophagocytosis component: P. aeruginosa PAO9 P1-lux, from log phase cultures diluted to 2 × 10 6 CFU/mL; diluted baby rabbit serum (1:10); 2 × 10 7 of differentiated HL-60 cells; and sera collected from PBS-, vector-, or vaccine-immunized mice.For controls, samples that lacked one of the required components were included.When removing a component, the volume was made up with OPK buffer (RPMI 1640 containing glucose + 1% BSA).Killing percentages for vaccine and control (vector or PBS) sera were determined by dividing the relative luciferase units (RLUs) obtained at a particular antibody concentration by the RLU obtained by control wells lacking antibody (25 µL complement + 25 µL HL-60 + 25 µL bacteria + 25 µL OPK buffer).This value is then multiplied by 100 to express the killing percentage.The assay was performed in 96-well plates following a 120-minute incubation at 37°C shaking at 250 RPM.Microtiter plates were read using an Envision Multilabel plate reader (PerkinElmer).

Statistical analysis
All analyses were performed using GraphPad Prism version 10 software.ELISA endpoint titers were calculated using the linear regression of duplicate measurements of adjusted OD 405 and were expressed as the reciprocal dilution.The x-intercept served as the endpoint titer.Antibody titers were compared using the Kruskal-Wallis test for compari son of three groups or the Mann-Whitney U test for two-group analysis.The results of survival studies were represented using Kaplan-Meier survival curves and were analyzed by the log-rank test.

S. typhimurium SL3261 containing pLAFRO9 expresses P. aeruginosa serogroup O9 LPS
We cloned the O antigen locus from the P. aeruginosa serogroup O9 strain, PAO9, into the broad host range cosmid, pLAFR376, using standard techniques.The plasmid was transferred to S. typhimurium SL3261 using P22 transduction.To confirm expression of P. aeruginosa serogroup O9, LPS was purified from PAO9, SL3261, and SL3261 (pLAFRO9) and separated by SDS-PAGE.A silver-stain gel of the LPS fragments is shown in Fig. 1A.The pattern of banding is more similar between the two SL3261 samples (lane 2 and lane 3) and distinct from PAO9 LPS (lane 1); however, banding is seen in all lanes, indicative of LPS expression from both S. typhimurium LPS and P. aeruginosa serogroup O9 LPS.To confirm the expression of P. aeruginosa O antigen, the LPS was analyzed by immunoblotting, and reactivity was detected with polyclonal anti-sera specific for serogroup O9 (Fig. 1B).Sera reacted with LPS from PAO9 (lane 1) and with LPS from the Salmonella containing pLAFRO9 (lane 3); no reactivity to LPS from SL3261 (lane 2) was seen.These data confirm the expression of the P. aeruginosa serogroup O9 LPS in our vaccine strain.

Intranasal immunization of mice with SL3261 (pLAFRO9) induces serum antibody responses that confer protection in an acute pneumonia model
Our group has previously demonstrated the safety and efficacy of SL3261 expressing PA103 serogroup O11 O-antigen (19,24).Our studies with P. aeruginosa serogroup O11 expression in SL3261 showed that intranasal administration of 10 7 CFU/mouse elicited a protective immune response (19).Based on these prior observations, we intranasally immunized mice on days 0 and 14 with 10 7 CFU/mouse of the vaccine [SL3261 (pLAFRO9)] or the vector control [SL3261 (pLAFR376)].The kinetics of the Pseudomonas-specific serum IgG and IgM antibody responses in sera collected from vaccine-immunized mice were compared to those immunized with PBS-or vector-only.We compared Pseudomonas-specific antibody responses on days 14 and 28, two weeks after the booster dose.
Interestingly, intranasal immunization using 10 7 CFU/mouse failed to generate a serogroup O9-specific antibody response.We did not observe a substantial difference in serum IgM response in the immunized mice compared to the vector-or PBS-treated mice (Fig. 2A).A similar trend was also observed with the serogroup O9 IgG response (Fig. 2B).Notably, we did not observe a substantial increase in the immune response in any of the immunized mice on day 28, after receiving the booster dose.
To examine the ability of the vaccine strain to confer protection against acute pneumonia in intranasally immunized animals, mice were challenged by intranasal infection with P. aeruginosa PAO9.Prior to this, we tested the virulence of PAO9 delivered via the intranasal route in 20 µL.As shown in Fig. 2C, increased doses decreased the time to when animals became moribund or were under distress, with a 50% lethal dose (LD 50 ) calculated for this strain to be ~2.0× 10 7 CFU, according to Reed and Muench (25).Immunized mice were challenged with ~1× the LD 50 (2.0× 10 7 CFU), and mice were euthanized at 24 hours post-infection to assess colonization in the upper (nasal wash) and lower (lung) respiratory tracts.Livers and spleens were also collected for the determination of bacterial CFU.As seen in Fig. 2D, no statistical difference was seen in FIG 1 Expression of P. aeruginosa serogroup O9 in S. typhimurium.LPS was extracted, applied to SDS-PAGE, and visualized with silver stain (A) or by (B) immunoblot analysis using Pseudomonas serogroup O9-specific rabbit polyclonal antibody, followed by incubation with anti-rabbit secondary antibody conjugated to alkaline phosphatase (Sigma).P. aeruginosa serogroup O9 strain PAO9 (lane 1), S. typhimurium SL321 (lane 2), and S. typhimurium SL3261 containing the plasmid expressing the P. aeruginosa serogroup O9 antigen (pLAFRO9) (lane 3).Molecular weight (MW).Lanes on both gels were spliced and rearranged to match one another for ease of comparison.bacterial loads in the lungs of vaccine-immunized animals and vector-immunized animals, but both were different compared to PBS-immunized mice.Furthermore, dissemination to the liver and spleens of immunized mice was also detected.There was a significant decrease in the bacterial load in the liver and spleen in the vaccine-immu nized mice relative to the CFUs recovered from PBS-immunized mice.However, no statistical difference in bacterial CFU was observed between mice immunized with the vaccine [SL3261 (pLAFRO9)] and mice inoculated with the vector [SL3261 (pLAFR376)] (Fig. 2E).
As we analyzed the ELISA data from this initial experiment, we noted that the IgG levels were not as robust as what we had previously observed for serogroup O11 (19), suggesting that higher doses of the vaccine might be needed.
We subsequently modified our vaccination protocol and immunized mice with a higher dose of the vaccine; we administered 10 7 or 10 9 CFU on days 0 and 14.Serum samples were taken 4 weeks post-vaccination, and we observed a direct correlation between the immunization dose and the level of O9-specific IgG, but not IgM (Fig. 3A  and B), with the mice administered 10 9 CFU of SL3261 (pLAFRO9) exhibiting an increased serogroup O9-specific IgG response after the prime-booster using 10 9 CFU immunization (Fig. 3B).
At 6 weeks post-vaccination, mice were challenged with 4.5 × 10 7 CFU of PAO9 (~2.5× the LD 50 ).After 24 hours, the bacterial colonization in the upper and lower respiratory tracts was determined.There was an inverse correlation between the CFU of PAO9 in either the upper or lower respiratory tract and the level of IgG (Fig. 3C).Notably, we observed a strong correlation between the level of IgG and the subsequent clearance from nasal washes and the lungs, as the bacterial burden tended to decrease with increasing doses of SL3261 (pLAFRO9) administered.
Based on the results from the previous experiment, we repeated the immunization by administering 10 9 CFU of SL3261 (pLAFRO9) vaccine at day 0 and boosted each mouse with the same dose, 10 9 CFU on day 14.ELISA analysis using sera collected four weeks post-first immunization from BALB/c mice that received the vaccine revealed robust P. aeruginosa serogroup O9 LPS-specific IgM and IgG antibody response when compared with sera from the vector-immunized or PBS control mice (Fig. 4A and B).
The IgG subtype responses to P. aeruginosa PAO9 were determined for vaccineimmunized mice.Intranasal immunization elicited significantly higher levels of IgG2a, IgG2b, and IgG3.The levels of IgG2a antibodies were significantly higher compared to IgG1 and IgG3 (P < 0.001) (Fig. 4C).
Given that we observed a significant increase in P. aeruginosa Ig titers following intranasal immunization, we next sought to determine the level of protection against PAO9 challenge in immunized mice.At 6 weeks post-vaccination, mice were challenged with 8.6 × 10 7 CFU of PAO9 (~4× the LD 50 ).Again, we observed a statistically significant difference in the bacterial counts recovered from vaccine-immunized mice when compared to the PBS and vector-immunized controls in either the upper (nasal wash) or the lower (lung) respiratory tract (Fig. 4D).Furthermore, no bacteria were recovered from the spleen or liver tissues of vaccine-immunized mice (Fig. 4E).

Antibody response induced by intranasal vaccination mediates opsonic killing of P. aeruginosa PAO9 in vitro
We previously demonstrated that immunization of Salmonella carrying pLPS2 results in the production of O11 O-antigen-specific antibodies and induces opsonic antibodies (19).Here, we examined the efficiency of sera collected from mice immunized and boosted with 10 7 or 10 9 CFU of the vaccine, which promoted opsonophagocytic killing of P. aeruginosa PAO9.As a control, we performed the assay in the absence of vaccineimmune sera to further confirm the role of O-antigen-specific antibodies in protection.Pooled antisera from vaccine-immunized mice mediated significant killing (>80%) at 1:2,700 and 1:900 antiserum dilutions for the 10 7 and 10 9 CFUs, respectively.Killing was not observed using pooled sera from vector-and PBS-immunized mice (Fig. 5).

Passive transfer of antisera from immunized animals to naïve mice provides protection from P. aeruginosa lung infection
To identify whether mice could be protected from pneumonia caused by infection with P. aeruginosa, undiluted antisera from PBS-, vector-, or vaccine-immunized mice were transferred intranasally to naive BALB/c mice at the same time that they were infected with a sublethal dose of PAO9.Briefly, female BALB/c mice 8-10 weeks old were divided into three groups, five mice each.Each group received 10 µL (5 µL/nostril) of sera collected from mice immunized with PBS, 10 7 CFU [SL3261 (pLAFR376)], or 10 9 CFU [SL3261 (pLAFRO9)].Immediately after serum administration, mice were intranasally infected with 1.1 × 10 7 CFU/mouse (~0.5× the LD 50 ) of P. aeruginosa PAO9.Mice were euthanized 24 hours post-infection, and the nasal wash, lung, liver, and spleen from each mouse were removed and homogenized to assess bacterial counts.
The number of viable bacteria recovered from the nasal wash and lung tissue of mice receiving vaccine-immune sera was significantly reduced compared to those recovered from mice receiving PBS-and vector-immune sera.We observed a statistically significant difference in the bacterial counts recovered from mice receiving antisera from vaccine-immunized mice when compared to those treated with antisera from PBS-and vector-immunized controls (Fig. 6A and B) indicating that vaccine-immune sera provided protection from infection.

DISCUSSION
Lipopolysaccharide is an immunodominant protective antigen for P. aeruginosa.The O antigen portion of P. aeruginosa LPS confers serogroup specificity, and P. aeruginosa has 20 different serogroups that vary by sugar composition (26); however, 10 are most commonly associated with infection (10)(11)(12)(13).
Several groups have attempted to develop vaccines to prevent infection based on LPS, and these have been met with limited success; there is currently no P. aeruginosa licensed vaccine (27,28).Recently, Nasrin et al. ( 13) reported a "Multiple Antigen Presenting System" for P. aeruginosa based on high molecular weight polysaccharides purified by the method of Pier and Thomas (18) and have targeted eight of the most common O antigen serogroups (13); however, serogroups O8 and O9 are missing from this system.Both serogroups O8 and O9 are acid-labile in nature and, as such, have never been included in a conjugate vaccine cocktail (29).
Live-attenuated vaccines have been extensively used to deliver heterologous antigens from various pathogens and have been shown to induce protective responses against a wide array of diseases (30)(31)(32).The use of live attenuated Salmonella strains over subunit or conjugate vaccines has grown considerably, mainly due to their ability to mimic natural infections by directly targeting and replicating in lymphoid organs where antigens are presented more efficiently, and hence leading to the activation of both mucosal and systemic immune responses of the host, together with long-term protec tion after mucosal delivery (33)(34)(35).In addition, live attenuated Salmonella strains have been shown to serve as an effective natural adjuvant because of immunostimulatory molecules, such as LPS and flagella, are located on their cell surface (36,37).Our laboratory previously characterized a vaccine that confers serogroup-specific protection against the P. aeruginosa challenge (19).The vaccine consists of Salmonella typhimurium strain SL3261, an attenuated aroA mutant (38), expressing the entire O-antigen locus from a P. aeruginosa serogroup O11 strain (20).Intranasal vaccination with this Salmonella strain conferred complete protection in mice with challenge doses of 5× the LD 50 of both cytotoxic and noncytotoxic P. aeruginosa serogroup O11 strains.Moreover, administration of antibodies from vaccinated mice directly into the nasal passageway and lungs of infected mice was able to confer protection when administered up to 6 hours after infection.
A caveat to this vaccine is that protection is only directed to serogroup O11 strains.P. aeruginosa serogroup O9 is another serogroup commonly found in infection, although it has been "neglected" in these previous P. aeruginosa vaccine cocktails.Herein, we used a similar approach as our previous work and showed that after cloning the serogroup O9 locus, the recombinant Salmonella strain retained all the genes required for native LPS synthesis, resulting in a vaccine candidate that expressed both homologous S. typhimu rium and heterologous P. aeruginosa serogroup O9 LPS (Fig. 1).We further confirmed the sequence corresponded to the serogroup O9 locus as reported by Raymond et al. (39).
Prior to the initiation of immunization studies, we needed to identify a strain to assess protection.We had not been able to find any reports of using serogroup O9 strains in models of infection; therefore, we determined the virulence of our PAO9 strain in our murine intranasal acute pneumonia model.Interestingly, we noted that this strain was not very pathogenic, and therefore, we needed to give large doses to mice to monitor vaccine-mediated protection.This was also the case for additional serogroup O9 strains obtained from collaborators.Supporting this, upon analyzing the population structure of P. aeruginosa, Ozer et al. noted that P. aeruginosa consists of two major groups, Group A and Group B. In their survey, they found that the majority of the serogroup O9 isolates that they characterized (14/15) were in Group A, and interestingly, most Group A strains were lacking the exoU gene (12).Similarly, Faure et al. determined that only one of the four serogroup O9 clinical isolates that they examined secreted ExoU (10).ExoU is a marker for highly virulent strains, especially those associated with acute lung infections (40); thus, if the serogroup O9 strains we tested were lacking exoU, it could explain why they are less able to cause severe infections.
Studies by Faure et al. (10) noted in a survey of clinical isolates that 4% (4/99 total) were serogroup O9 (2% in chronic infections and 2% in acute infections), but that these were not associated with mortality.Using in silico serotyping, Thrane et al. (11) found 1.25% of the 1120 genomes they surveyed were serogroup O9.Interestingly, when looking specifically at the genomes of isolates from cystic fibrosis patients, this number was much lower (0.38% from 529 genomes) (11).Ozer et al. (12) found serogroup O9 strains accounted for 2% of the isolates they sequenced from diverse sources.And more recently, Nasrin et al. (13) surveyed 413 invasive P. aeruginosa from 10 different countries worldwide and found ~3% of them were serogroup O9.It is important to note that isolates from chronic lung infections in cystic fibrosis are often considered LPS-rough due to a lack of LPS O antigen (41); therefore, they are non-typable and would not be expected to be targeted by any O serogroup-specific vaccine.
We noted a number of subtle but interesting differences between our original experiments with serogroup O11 (19) and those performed here with serogroup O9.We needed to modify our vaccination protocol and increase the immunization dose of Salmonella to 10 9 CFU to elicit a robust and protective immune response.This may have reflected low expression of the plasmid-borne genes of the serogroup O9 locus, a reduced amount of LPS expressed on Salmonella, and/or the lability of the O antigen itself.However, our optimized vaccination with 10 9 CFU was found to induce a robust protective response.
Many of the current available bacterial vaccines are based on either purified polysaccharides (pneumococcal polysaccharide vaccine) or polysaccharide-protein conjugates [pneumococcal conjugate vaccines, meningococcal conjugate (menACWY) vaccines, and Hib vaccines].Polysaccharides are known for their poor immunogenicity, as they elicit T-cell-independent (TI) immune responses and therefore fail to achieve isotype switching from IgM to IgG antibody and memory cell production.Conversely, immuni zation with polysaccharides conjugated to immunogenic carrier protein promotes a T-cell-dependent (TD) response, thus resulting in long-term IgG responses (42).LPSassociated O-antigen in live bacteria is naturally linked to the lipid A-core in the outer membrane, which contains highly immunogenic proteins, providing great potential for eliciting TD immune responses, particularly Th1-mediated responses that induce high levels of complement-fixing IgG2a antibodies (43).Using recombinant attenu ated Salmonella for heterologous expression has the advantage of ease of expression of polysaccharide antigens and generating an immune response more typical of a polysaccharide-protein conjugate (44).Supporting this, we found that our vaccine induced high levels of IgG response, particularly IgG2a, IgG2b, and IgG3, in serogroup O9 (Fig. 4C).
We have also found that vaccination with SL3261 (pLAFRO9) induced antibodies that were capable of mediating opsonophagocytic killing (Fig. 5).Protection in our model correlates with the presence of serogroup O9-specific antibodies, particularly IgG.Murine IgG2a and IgG2b are known for their high affinity to bind and engage FcγRI, which subsequently result in complement activation and opsonophagocytic killing activity (45).The predominance of IgG2a and IgG2b suggests that induction of serogroup O9-specific IgG capable of mediating opsonophagocytic killing activity correlates with protective activity in our model.Importantly, we could transfer protection to naïve mice using passive immunization.We found that protection was only observed in mice receiving vaccine antisera rather than vector antisera (Fig. 6), suggesting again that protection in our model is likely to be dependent on the presence of P. aeruginosa serogroup O9-specific IgG antibodies.
To our knowledge, there has only been one report of a serogroup O9-specific epidemic, which was an outbreak of dermatitis in a whirlpool at a hotel in Atlanta, Georgia, in 1981 (46).Whether the lack of detection of outbreaks caused by serogroup O9 isolates is because of avirulence due to the lack of exoU and/or the acid-labile nature of the O antigen itself is not known but is tempting to speculate.However, that is not to assume that all serogroup O9 P. aeruginosa isolates are innocuous.In 2022, there was a fatal case of community-acquired P. aeruginosa pneumonia caused by a serogroup O9 strain in an otherwise immunocompetent individual following the SARS-CoV-2 infection (47).
We have previously shown that we could express P. aeruginosa serogroup O11 on the human licensed typhoid vaccine strain, Salmonella Typhi Ty21a (48), when it contained the plasmid pLPS2 (49) suggesting the feasibility of this approach for human use.Since that time, methods for stable integration of O antigen genes have been developed (34) as well as techniques for the expression of multiple serogroups in the same S. Typhi Ty21a strain (50).
Our results suggest the possibility that this vaccination protocol could be used to develop a passive immunotherapy specific for the "neglected" P. aeruginosa serogroup O9.Such an approach has been used for P. aeruginosa serogroup O11.There is a fully human anti-LPS IgM monoclonal antibody (Panobacumab) that has been tested in patients with nosocomial Pseudomonas pneumonia due to serogroup O11.Treat ment with Panobacumab resulted in a shorter time to clinical resolution compared to untreated patients (51), but this is only effective for patients infected with a serogroup O11 P. aeruginosa strain.The advantage is that passive immunotherapy would be applicable to all patients, including the immunocompromised, who cannot mount an immune response and who are particularly at risk for P. aeruginosa hospital infections.The long-term goal of this research is ultimately the development of a cocktail including all prominent serogroups that could be given to people to protect them against P. aeruginosa infection by any of the typically encountered strains.

Full 6 FIG 2
FIG 2 Serum antibody response of BALB/c mice following intranasal immunization with S. typhimurium SL3261 expressing P. aeruginosa serogroup O9 and bacterial loads in organs of intranasally immunized BALB/c mice after intranasal lethal challenge with P. aeruginosa PAO9.Mice were immunized intranasally on days 0 and 14 with 10 7 CFU/mouse of the vector control [SL3261 (pLAFR376)] or the vaccine [SL3261 (pLAFRO9)].Sera were collected two and four weeks post-vaccination, and the data were analyzed by the Mann-Whitney U test.(A) Serum IgM and (B) IgG responses to P. aeruginosa PAO9 whole antigen.(C) Survival rates for naïve mice after intranasal challenge with various doses of P. aeruginosa PAO9 (n = 4 mice/group).(D) Bacterial load in the nasal wash and lungs.(E) Liver and spleens of intranasally immunized mice 24 hours post-challenge with 2 × 10 7 CFU of PAO9.All samples were plated for viable CFU on Pseudomonas isolation agar (PIA).P values equal to or less than 0.05 were displayed.*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.Error bars represent the mean and SEM.Each point represents a single mouse.Dashed line represents the limit of detection.

FIG 3 9 FIG 4
FIG 3 Immune response and bacterial burden in response to immunizing BALB/c mice with various doses of S. typhimurium SL3261 expressing P. aeruginosa serogroup O9.Anti-P.aeruginosa serogroup O9 serum (A) IgM and (B) IgG antibody response after intranasal vaccination of BALB/c mice to various doses of vector or vaccine.(C) Bacterial load in the nasal wash and lungs of intranasally immunized mice 24 hours post-challenge with 4.5 × 10 7 CFU of PAO9.All samples were plated for viable CFU on PIA.Each point represents a single mouse.Data were analyzed by one-way ANOVA.P values equal to or less than 0.05 were displayed.*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.Error bars represent the mean and SEM.Dashed line represents the limit of detection.

FIG 6
FIG 6 Passive antisera transfer from immunized animals provides protection against acute P. aeruginosa pneumonia in naïve mice.Bacterial loads in (A) nasal wash and lungs, (B) liver and spleens of BALB/c mice after passive intranasally transferred antisera immediately followed by challenge with 1 × 10 7 CFU of PAO9.Mice were euthanized 24 hours post-infection.All samples were plated for viable CFU on PIA.Each point represents a single mouse.Data were analyzed by one-way ANOVA.P values equal to or less than 0.05 were displayed.*P < 0.05, **P < 0.01, ****P < 0.001.Each point represents a single mouse.Error bars represent the mean and SEM.Dashed line represents the limit of detection.