Fc-free single-chain antibody mRNA therapy for airway infection of multidrug-resistant Pseudomonas aeruginosa.

Skip to main content An official website of the United States government Here's how you know Here's how you know Official websites use .gov A .gov website belongs to an official government organization in the United States. Secure .gov websites use HTTPS A lock ( Lock Locked padlock icon ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites. Search Log in Dashboard Publications Account settings Log out Search… Search NCBI Primary site navigation Search Logged in as: Dashboard Publications Account settings Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide PERMALINK Copy As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice Nat Commun . 2026 Apr 9;17:2960. doi: 10.1038/s41467-026-71040-8 Search in PMC Search in PubMed View in NLM Catalog Add to search Fc-free single-chain antibody mRNA therapy for airway infection of multidrug-resistant Pseudomonas aeruginosa Mao Kinoshita Mao Kinoshita 1 Department of Anesthesiology, Graduate School of Medical Science, Kyoto Prefectural University of Medicine, Kyoto, Japan Find articles by Mao Kinoshita 1 , Ken Kawaguchi Ken Kawaguchi 1 Department of Anesthesiology, Graduate School of Medical Science, Kyoto Prefectural University of Medicine, Kyoto, Japan Find articles by Ken Kawaguchi 1 , Yuki Mochida Yuki Mochida 2 Medical Research Laboratory, Institute of Integrated Research, Institute of Science Tokyo, Tokyo, Japan 3 Innovation Center of NanoMedicine (iCONM), Kawasaki Institute of Industrial Promotion, Kawasaki, Japan Find articles by Yuki Mochida 2, 3 , Nguyen B T Le Nguyen B T Le 2 Medical Research Laboratory, Institute of Integrated Research, Institute of Science Tokyo, Tokyo, Japan 3 Innovation Center of NanoMedicine (iCONM), Kawasaki Institute of Industrial Promotion, Kawasaki, Japan Find articles by Nguyen B T Le 2, 3 , Atsushi Kainuma Atsushi Kainuma 1 Department of Anesthesiology, Graduate School of Medical Science, Kyoto Prefectural University of Medicine, Kyoto, Japan Find articles by Atsushi Kainuma 1 , Naoko Takeda-Miyata Naoko Takeda-Miyata 4 Department of Pathology, University Hospital, Kyoto Prefectural University of Medicine, Kyoto, Japan Find articles by Naoko Takeda-Miyata 4 , Motohiro Kojima Motohiro Kojima 4 Department of Pathology, University Hospital, Kyoto Prefectural University of Medicine, Kyoto, Japan Find articles by Motohiro Kojima 4 , Teiji Sawa Teiji Sawa 5 University Hospital, Kyoto Prefectural University of Medicine, Kyoto, Japan Find articles by Teiji Sawa 5, ✉ , Satoshi Uchida Satoshi Uchida 2 Medical Research Laboratory, Institute of Integrated Research, Institute of Science Tokyo, Tokyo, Japan 3 Innovation Center of NanoMedicine (iCONM), Kawasaki Institute of Industrial Promotion, Kawasaki, Japan 6 Pandemic Preparedness, Infection and Advanced Research Center (UTOPIA), The University of Tokyo, Tokyo, Japan Find articles by Satoshi Uchida 2, 3, 6, ✉ Author information Article notes Copyright and License information 1 Department of Anesthesiology, Graduate School of Medical Science, Kyoto Prefectural University of Medicine, Kyoto, Japan 2 Medical Research Laboratory, Institute of Integrated Research, Institute of Science Tokyo, Tokyo, Japan 3 Innovation Center of NanoMedicine (iCONM), Kawasaki Institute of Industrial Promotion, Kawasaki, Japan 4 Department of Pathology, University Hospital, Kyoto Prefectural University of Medicine, Kyoto, Japan 5 University Hospital, Kyoto Prefectural University of Medicine, Kyoto, Japan 6 Pandemic Preparedness, Infection and Advanced Research Center (UTOPIA), The University of Tokyo, Tokyo, Japan ✉ Corresponding author. Received 2025 Jun 17; Accepted 2026 Mar 11; Collection date 2026. © The Author(s) 2026 Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/ . PMC Copyright notice PMCID: PMC13066097  PMID: 41957036 Abstract With the growing threat of antimicrobial resistance (AMR), alternatives to conventional antibiotics are urgently needed. Here, we show that mRNA-based therapeutics encoding single-chain variable fragment (scFv) antibodies—small, targeted antibody derivatives—provide robust protection against Pseudomonas aeruginosa , a major multidrug-resistant pathogen. We target the bacterial type III secretion system (T3SS), a needle-like apparatus used by the pathogen to inject toxins into host cells. When delivered intravenously via lipid nanoparticles, the scFv-encoding mRNA prompts sustained protein production, overcoming the typically short half-life of small antibody fragments. This treatment mitigates lung inflammation, reduces bacterial load, and improves survival in clinically relevant models, including immunocompromised mice infected with multidrug-resistant, exoU -positive (highly cytotoxic) clinical isolates. We find that Fc-free scFv antibodies, consisting only of the antigen-binding domain, migrate more efficiently from the bloodstream to the lung epithelium—the primary site of infection—than their larger counterpart conjugated to an Fc domain. This enhanced tissue penetration results in superior therapeutic outcomes. Overall, mRNA-encoded, Fc-free antibody fragments represent a promising and versatile platform for combating life-threatening bacterial infections without relying on traditional antibiotics. Subject terms: Molecular medicine, Translational research, RNA vaccines Here, the authors develop mRNA-based therapeutics encoding single-chain variable fragment antibodies that target the T3SS of Pseudomonas aeruginosa , which shows therapeutic potential in an immunocompromised mouse infection model. Introduction The rapid proliferation of drug-resistant bacteria has emerged as a critical and escalating threat to global public health. In 2024, the World Health Organization identified 15 priority drug-resistant pathogens, among which Pseudomonas aeruginosa was classified as a high-priority pathogen 1 . A global analysis reported in 2019 estimated that antimicrobial resistance (AMR) was directly responsible for 1.27 million deaths annually, approximately 80% of which were attributable to six bacterial species, including Escherichia coli , Staphylococcus aureus , and P. aeruginosa 2 . With antibiotic discovery stagnating, projections suggest that, by 2050, deaths caused by multidrug-resistant infections (10 million) could exceed those from cancer (8.2 million) 3 . Another report, released in September 2024, warns that, over the next 25 years, antibiotic-resistant bacteria may directly cause more than 39 million deaths worldwide, with an additional 169 million deaths resulting from complications associated with such infections 4 . Alarmingly, given the slow development of novel antibiotics, the evolution of bacterial resistance appears to be accelerating, underscoring the urgent need for innovative alternative therapeutic strategies 5 . Among drug-resistant pathogens, P. aeruginosa poses a substantial threat, particularly by causing opportunistic respiratory infections in patients on ventilators, individuals with cystic fibrosis, and immunocompromised hosts 6 , 7 . Monoclonal antibodies represent a promising therapeutic strategy, offering pathogen-specific targeting with minimal off-target effects. Our research has identified the type III secretion system (T3SS) and its associated exotoxins as critical targets for antibody-based intervention, given their strong association with acute lung injury and increased mortality 8 . Notably, in preclinical models, antibodies directed against PcrV—a structural cap protein of the T3SS—markedly reduced lung pathology and improved survival outcomes by effectively blocking the translocation of all T3SS effector toxins into host cells 9 , 10 . These encouraging preclinical results have prompted clinical development, including previous clinical trials involving the monoclonal anti-PcrV antibody 11 , 12 , which we helped develop, as well as trials employing a bispecific monoclonal antibody targeting PcrV and the Psl exopolysaccharide 13 , 14 . Despite promising outcomes in these trials, no immunization strategies targeting P. aeruginosa have received clinical approval 15 , even a quarter-century after the initial proof of concept was demonstrated in animal models 16 . One major barrier to the clinical translation of recombinant antibodies for infectious diseases is the high cost associated with their complex production, purification, and characterization. In this context, mRNA therapy offers a significant advantage through its cell-free, streamlined manufacturing process, which simplifies both production and purification 17 , 18 . Unlike recombinant antibodies, which exhibit diverse physicochemical properties depending on their sequences, mRNAs encoding different antibodies share similar characteristics, enabling standardized manufacturing processes. Following the success of mRNA-based COVID-19 vaccines, there has been growing interest in using mRNA to produce therapeutic proteins in situ within the patient’s body. This approach has shown promise in clinical trials for genome editing and protein replacement therapies 19 – 21 . The application of mRNA is now expanding into antibody-based therapies, driving animal and clinical studies 22 – 28 . Additional advantages of mRNA include the ability to produce antibodies with patient-specific post-translational modifications and the ease of adapting the platform to antibody variants such as fragments and bispecifics. In this context, we evaluated the therapeutic potential of mRNA encoding an anti-PcrV antibody in a murine model of P. aeruginosa infection. Taking advantage of mRNA design flexibility, we prepared two formats: scFvs with and without conjugation to the mouse IgG Fc domain (mFc). Notably, while recombinant Fc-free scFv antibodies typically exhibit a short circulation half-life on the order of tens of minutes, mRNA delivery can extend their retention in the body 29 . Using ionizable lipid nanoparticles (LNPs) for intravenous (i.v.) delivery, which induces protein expression in the liver, both Fc-free and Fc-conjugated scFv mRNAs significantly improved survival following bacterial challenge. This approach proved effective in a clinically relevant model involving immunocompromised mice infected with multidrug-resistant, highly cytotoxic clinical isolates of P. aeruginosa . Structural predictions and supporting studies indicated that mutations in these isolates, as well as other reported mutations, do not impair antibody recognition 30 – 32 , underscoring the robustness of our approach. Furthermore, our study revealed a key advantage of the Fc-free formulation over the Fc-conjugated version in terms of biodistribution. While current mRNA-based antibody therapies typically utilize Fc-conjugated formats 22 , 23 , 25 , 28 , 33 , we found that Fc-free scFv migrates from the bloodstream to the alveolar space, the primary site of P. aeruginosa infection, more efficiently than Fc-conjugated scFv, resulting in better therapeutic outcomes. Results Design of mRNA constructs and characterization of LNPs We constructed mRNA encoding an anti-PcrV scFv antibody (scFv-m166) based on the genetic information of the murine monoclonal anti-PcrV IgG (mAb166) 34 and added a tissue plasminogen activator signal sequence (tPAss) at the amino-terminus to facilitate secretion (Fig. 1A-1 ). To evaluate the influence of the Fc domain, we also prepared mRNA encoding tPAss-scFv conjugated with the mouse hinge region and the CH2 and CH3 Fc-domains at the carboxy-terminus (scFv-m166-mFc, Fig. 1A-2 ). As negative controls, we used mRNA encoding either luciferase or an irrelevant scFv targeting hen egg lysozyme (scFv-control, Fig. 1A-3 ) 35 . All four mRNA constructs were encapsulated in LNPs prior to injection into mice. These LNPs exhibited hydrodynamic diameters of 100–130 nm and polydispersity indices below 0.15 in dynamic light scattering measurements, and demonstrated high mRNA encapsulation efficiency above 90% (Table S1 ). Fig. 1. Prophylactic and therapeutic efficacy in murine models of P. aeruginosa infection. Open in a new tab A Schematic representation of mRNA constructs: 1) Fc-free scFv-m166. 2) Fc-conjugated scFv-m166-mFc mRNA. 3) Fc-free scFv-control mRNA . B Experimental protocol for the prophylactic model. P. aeruginasa PA103 (1 × 10 6 cfu) was administered intratracheally (right). C , D Post-challenge monitoring of body temperature (left), physical activity (center), and survival (right). C Intramuscular injection of scFv-m166 mRNA, luciferase mRNA. mAb166 10 µg premix i.t.: mice received P. aeruginosa PA103 (1 × 10 6 cfu) pre-mixed with 10 µg of anti-PcrV IgG protein (mAb166) intratracheally (i.t.). D Intravenous injection of scFv-m166 mRNA, scFv-m166-mFc mRNA, luciferase mRNA, scFv-control mRNA, or mAb166 (5 µg) i.v.: mice received 5 µg of mAb166 protein. E Extended prophylactic efficacy. Left: Experimental design. scFv-control mRNA, scFv-m166 mRNA, or scFv-m166-mFc mRNA was administered intravenously (i.v.) at (1) 2, (2) 48, or (3) 96 hours prior to challenge with P. aeruginosa PA103 (1 × 10 6 cfu). Mice’s survival was monitored for 1 week. The value of n represents the number of mice in each group. Data for body temperature and physical activity in panels C and D represent the mean ± SD. Statistical comparisons were performed using Kruskal–Wallis nonparametric tests with Bonferroni correction. All tests were performed as two-sided analyses. Significant differences were observed across all groups for all measured parameters. Specifically, the results were as follows: body temperature: H (44) = 385.27, P < 0.001 ( n = 472); physical activity: H (44) = 391.64, P < 0.001 ( n = 510). Survival differences were assessed by log-rank tests with P values adjusted using the Benjamini–Hochberg method. All tests were performed as two-sided analyses. * P < 0.05 vs. uninfected control; † P < 0.05 vs. i.v. scFv-control mRNA; ‡ P < 0.05 vs. i.m. luciferase mRNA. § P < 0.05 vs. the group that received prophylactic treatment 2 hours before infection in ( E ). Source data are provided as a Source Data file. mRNA-encoded anti-PcrV scFv confers prophylactic protection against lethal P. aeruginosa infection We first investigated the therapeutic potential and mode of action of scFv-m166 and scFv-m166-mFc mRNA in a prophylactic setting. In this model, mice received mRNA LNPs either intramuscularly or intravenously 2 h prior to bacterial challenge. As a control, the mAb166 protein was administered intravenously. Based on a previous preclinical study of mRNA-based antibody therapy for infectious disease, we used a dose of 10 µg/mouse for mRNA and 5 or 10 µg/mouse for mAb166 protein throughout this study 23 . For the challenge, mice were intratracheally instilled with a lethal dose (1 × 10⁶ colony-forming units (CFU)/mouse) of P. aeruginosa PA103 (Fig. 1B ). Previous studies have shown that, without treatment, most mice succumb within 24 h due to acute lung injury caused by type III secreted bacterial toxins 8 , 36 . Therefore, the observation period for this experiment was first set to 24 h. In the intramuscular (i.m.) mRNA treatment, scFv-m166 mRNA conferred 100% survival at 24 h post-challenge, whereas the survival rate in the control group receiving luciferase mRNA was below 10% (Fig. 1C ). Consistent with this, scFv-m166 mRNA mitigated the declines in body temperature and physical activity levels following infection, indicating that the antibody produced from the mRNA effectively protected against P. aeruginosa lung infection. As a positive control, the “mAb166 premix i.t.” group received the lethal bacterial dose premixed with mAb166 protein via intratracheal (i.t.) instillation. All mice in this group survived the 24-h period, confirming the antibody’s neutralizing activity. mRNA-based antibody therapy also demonstrated efficacy following i.v. administration. Intravenous injection of mRNA LNPs led to protein expression predominantly in the liver, particularly in hepatocytes (Fig. S1 ). In this setting, both scFv-m166 and scFv-m166-mFc mRNAs significantly improved mouse survival and promoted more effective recovery of body temperature and activity levels compared with those in control groups receiving luciferase mRNA or scFv-control mRNA during the 24-h observation period (Fig. 1D ). Notably, these treatments maintained over 90% pro-survival efficacy when the observation period was extended to 1 week (Fig. 1E-1 ). To further investigate the duration of scFv activity, we evaluated pro-survival efficacy after extending the interval between mRNA administration and bacterial challenge to 48 h and 96 h. Both scFv-m166 and scFv-m166-Fc mRNA retained therapeutic efficacy when administered 48 h prior to challenge, but not 96 h prior to it (Fig. 1 E- 2 , 3 ). These results indicate that scFv expressed from mRNA provides protective activity against bacterial challenge for more than 2 days, regardless of Fc conjugation. Fig. 2. Mode of action of scFv mRNA in the prophylactic model. Open in a new tab Mice were challenged with P. aeruginosa 2 h after mRNA administration. Various parameters were assessed 24 h ( A – E , G ) or 12 h ( F ) post-challenge: A Wet lung weight, reflecting pulmonary edema. B Myeloperoxidase (MPO) activity. C Bacterial burden [colony-forming units (CFU) per lung]. D Interleukin-6 (IL-6) concentration. E Tumor necrosis factor-alpha (TNF-α) concentration in lung homogenates. F Systemic bacterial dissemination. The number of bacteria [CFU per 100 µL blood, 100 g liver, or 10 mg spleen] was quantitatively determined by culturing blood samples or organ homogenates. Data are presented as box plots showing the interquartile range (box), median (centerline), and minimum and maximum values (whiskers). Values outside the range of [LQ – 1.5 × interquartile distance (IQD)] to [UQ + 1.5 × IQD] were treated as outliers. The value of n represents the number of mice in each group, and each circle in the figure shows data from an individual mouse; open circles indicate those who survived for 24 hours, while filled circles indicate those who died within 24 hours. Kruskal–Wallis nonparametric tests, with multiple comparisons adjusted by Bonferroni correction, were performed. All tests were performed as two-sided analyses. Significant differences were observed across all groups for all measured parameters. Specifically, the results were as follows: wet lung weight: H (8) = 71.333, P < 0.001 ( n = 93); IL-6: H (8) = 49.638, P < 0.001 ( n = 91); MPO: H (8) = 58.275, P < 0.001 ( n = 90); bacteria in lung: H (7) = 62.754, P < 0.001 ( n = 84); IL-6: H (8) = 49.638, P < 0.001 ( n = 91); TNF-α: H (8) = 42.678, P < 0.001 ( n = 91). * P < 0.05 vs. uninfected control; † P < 0.05 vs. i.v. scFv-control mRNA; ‡ P < 0.05 vs. luciferase mRNA. G . Representative hematoxylin and eosin (H&E) staining of lung sections. We performed at least two independent experiments using different mice. For key group comparisons, we conducted replicate experiments with five mice, as shown in Fig. S2 , and board‑certified pathologists evaluated the results through quantitative assessment. Left: ×20 magnification (scale bar: 200 µm); right: ×40 magnification (scale bar: 100 µm). Source data are provided as a Source Data file. Fig. 3. Therapeutic efficacy of mRNA treatment against P. aeruginosa infection. Open in a new tab A Experimental protocol. B Survival of mice treated with 5 or 10 µg of mAb166 protein. C Survival of mice treated with scFv-m166 or scFv-m166-mFc mRNA. The value of n represents the number of mice in each group. Survival differences were assessed by the log-rank test with P values adjusted using the Benjamini–Hochberg method. All tests were performed as two-sided analyses. Significant differences were observed across all groups for all measured parameters. * P < 0.05 vs. uninfected control; † P < 0.05 vs. scFv-control mRNA. Source data are provided as a Source Data file. We further explored the mechanism of action underlying the prophylactic effects of scFv-m166 mRNA and scFv-m166-mFc mRNA administration 2 h prior to bacterial challenge. Mice injected with luciferase mRNA or scFv-control mRNA showed significant increases in lung weight, neutrophil myeloperoxidase (MPO) activity, and bacterial load in the lungs compared with uninfected controls at 24 h post-challenge (Fig. 2A–C ). The increase in lung weight indicated edema, while elevated MPO activity reflected inflammatory responses. Notably, both i.m. and i.v. administration of scFv-m166 or scFv-m166-mFc mRNA effectively alleviated these pathological changes and reduced the bacterial load in the lungs. Consistent with these findings, mRNA treatment reduced interleukin-6 (IL-6) and tumor necrosis factor-α (TNF-α) expression in lung homogenates to levels comparable to those in uninfected controls, whereas control mRNA groups showed elevated IL-6 and TNF-α levels (Fig. 2D, E ). Furthermore, scFv mRNA treatment effectively suppressed bacterial dissemination to distal organs. Specifically, i.v. administration of scFv-m166 or scFv-m166-mFc mRNA reduced bacterial loads in the blood, liver, and spleen at 12 h post-infection compared with those in mice treated with scFv-control mRNA (Fig. 2F ). Histological analysis at 24 h post-challenge further confirmed the therapeutic effects (Fig. 2G ). Lungs from mice treated with luciferase mRNA or scFv-control mRNA exhibited signs of acute lung injury, including alveolar hemorrhage and atelectasis. In contrast, treatment with scFv-m166 or scFv-m166-mFc mRNA—administered either intramuscularly or intravenously—alleviated these pathological changes. According to the Lung Injury Scoring System from the American Thoracic Society (Fig. S2A-1 ) 37 , groups intravenously treated with scFv-m166 or scFv-m166-Fc mRNA showed significant mitigation of pathological changes, including hyaline membranes, proteinaceous debris filling the airspaces, and alveolar septal thickening, compared with the findings in the scFv-control mRNA group (Fig. S2A-2, B ). Although proinflammatory changes, such as neutrophil infiltration into alveolar and interstitial spaces, were still observed in the scFv-m166- and scFv-m166-Fc-treated groups, the treatment markedly reduced overall inflammatory reactions. In the scFv-control mRNA intravenous group, the lung tissue exhibited extensive necrotizing inflammation, a hallmark of pneumonia caused by P. aeruginosa strains that secrete the type III phospholipase A 2 toxin ExoU 38 . In contrast, such inflammatory lesions were absent in both the scFv-m166 mRNA i.v. group and the scFv-m166-mFc mRNA i.v. group (Fig. S2A-3 ). These findings further support the tissue-protective activity conferred by scFv mRNA therapy. These experiments using a prophylactic model of P. aeruginosa infection demonstrated the effectiveness of mRNA-encoding anti-PcrV Fc-free and Fc-conjugated antibodies in reducing bacterial load and mitigating lung inflammation, ultimately minimizing mortality (Figs. 1 and 2 ). While both i.m. and i.v. routes were effective for delivering mRNA LNPs, i.v. injection may be more suitable for future applications in larger animals and humans. In larger animals, the ratio of the volume within which intramuscularly injected solutions are distributed to total body volume becomes smaller. In contrast, i.v. injection allows targeting of the liver—a large, accessible organ—and several clinical trials have demonstrated the feasibility of liver-specific delivery using LNPs in humans 19 , 20 . As such, we adopted the i.v. route for subsequent experiments. Anti-PcrV scFv provides therapeutic efficacy against lethal P. aeruginosa infection Next, we evaluated the therapeutic potential of anti-PcrV scFv antibody mRNA by administering the mRNA 30 min after i.t. challenge with P. aeruginosa PA103 (Fig. 3A ). This model more closely mimics a scenario of clinical treatment for P. aeruginosa infection compared with the prophylactic model described above (Figs. 1 and 2 ). Mouse survival was monitored for 1 week. In the control group treated with scFv-control mRNA, all mice died within 24 h (Fig. 3B ). In contrast, i.v. administration of the mAb166 protein improved survival in a dose-dependent manner, reaching up to 80.0% at a dose of 10 µg. Intravenous injection of scFv-m166 mRNA significantly improved mouse survival compared with that in the scFv-control group (0.0%), achieving an 80.0% survival rate at 1 week post-challenge (Fig. 3C ). scFv-m166-mFc mRNA also significantly improved survival compared with scFv-control mRNA, although to a lesser extent than the Fc-free scFv-m166 mRNA, resulting in a 50% survival rate at 1 week post-challenge. scFv-m166 maintains robust binding to diverse PcrV variants found in clinical isolates Antibody therapy offers a promising solution to the growing clinical challenge of antimicrobial-resistant P. aeruginosa , as the mechanisms by which antibodies exert their effects differ from those of antimicrobials. This motivated us to evaluate the feasibility of our approach using clinically isolated antimicrobial-resistant strains (Figs. 4 and S3 ). To this end, we prepared 10 in-hospital isolates 39 , along with laboratory strains PAO1 and PA103. All 10 clinical isolates exhibited resistance to multiple antimicrobials, with varying susceptibility profiles (Fig. 4A ). In contrast, the laboratory strain, PA103, used in the previous experiments (Figs. 1 – 3 ) was susceptible to all tested antimicrobials. Fig. 4. Characterization of various P. aeruginosa clinical isolates. Open in a new tab A Summary of 12 strains, including two laboratory strains (PAO1 and PA103) and 10 clinical isolates: multilocus sequence typing (MLST), exoenzyme genotypes, cytotoxicity, pigment production, non-synonymous SNPs in the pcrV gene, and antimicrobial susceptibility profiles. B eBURST analysis of MLST data. A minimum spanning tree was constructed using PHYLOViZ data. C Pigment production in tryptic soy broth. D Cytotoxicity in BEAS-2B bronchial epithelial cells. Left: Quantitative data are presented as box plots, where the box represents the interquartile range, the centerline indicates the median, and whiskers denote the minimum and maximum values ( n = 3 per group; 2–4 h post-infection). Each open circle shows data from an individual well of the assay culture plate from a single experiment. Upper right: A representative image of the assay plate. Statistical significance was determined using the Kruskal–Wallis test followed by Bonferroni correction for multiple comparisons. All tests were performed as two-sided analyses. Significant differences were observed across all groups for all measured parameters. Specifically, the results were as follows: 2 h: H (12) = 34.705, P = 0.001 ( n = 39); 3 h: H (12) = 36.944, P < 0.001 ( n = 39); 4 h: H (12) = 37.289, P = 0.0002 ( n = 39). * P < 0.05 vs. the corresponding uninfected control at each time point; Lower right: Trypan blue staining at 4 h, visualizing cell death (scale bar: 100 µm). Non-cytotoxic strains ( exoS ⁺/ exoU ⁻): blue. Cytotoxic strains ( exoS ⁻/ exoU ⁺): red. Abbreviations: MLST, multilocus sequence typing; PIPC, piperacillin; CAZ, ceftazidime; IPM/CS, imipenem/cilastatin; MEPM, meropenem; AZT, aztreonam; GM, gentamicin; AMK, amikacin; CPFX, ciprofloxacin. Source data are provided as a Source Data file. Notably, P. aeruginosa possesses multiple virulence factors, and its pathogenicity varies among isolates. In particular, T3SS-associated cytotoxicity, which correlates with lung injury and mortality, is linked to the exoS / exoU genotypes 36 . Among the genes encoding four T3SS toxins produced by P. aeruginosa —ExoS, ExoT, ExoU, and ExoY— exoS encodes a 49-kDa ADP-ribosylating enzyme known as exoenzyme S, while exoU encodes a 72-kDa phospholipase A 2 cytotoxin 40 . The clinical isolates differed in their exoS / exoU genotypes (Fig. 4A and S3 ). Multilocus sequence typing (MLST) analysis revealed that the 12 strains, including the 10 clinical isolates and 2 laboratory strains, were widely distributed across the eBURST phylogenetic tree, which includes 5,331 sequence types (STs), as reported in the PubMLST database at the time of analysis (Fig. 4B and S3A ). The exoS / exoU genotypes of the 12 P. aeruginosa strains used in this study showed a strong correlation with the phylogenetic classification based on MLST data determined by the eBURST algorithm (Fig. S3B, C ). Several isolates, including KN5, KN17, and KN33, exhibited high levels of pigment production in vitro (Fig. 4C ), a trait often associated with increased virulence. The cytotoxicity of the 12 tested strains against lung epithelial cells is shown in Fig. 4D . Six clinical isolates exhibited enhanced cytotoxicity compared with the highly virulent laboratory strain PA103; all of these were of the exoS ⁻/ exoU ⁺ genotype. In contrast, five strains, including PAO1, displayed lower cytotoxicity than PA103 and were all of the exoS ⁺/ exoU ⁻ genotype. To investigate whether mutations in the clinical isolates might affect antibody binding, we sequenced the pcrV genes from all 12 strains and compared the results with previously reported mutations known to disrupt recognition by the mAb166 blocking antibody 41 . Using the PcrV amino acid sequence of the PAO1 strain—which served as the reference in the P. aeruginosa whole-genome project—as a baseline, one to five amino acid substitutions were identified in eight strains, including PA103. The locations of these mutations and their corresponding strains are illustrated in a molecular phylogenetic tree based on PcrV sequences (Fig. 5A ). The amino acid variations in PcrV showed a complete correlation with MLST and exoS /exoU genotypes. In total, six amino acid substitutions were identified, along with two previously reported substitutions (S225K/R) 41 . Bacterial proteins from the 12 P. aeruginosa strains cultured in trypticase soy broth containing 10 mM nitrilotriacetic acid to induce T3SS expression were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and subjected to immunoblotting using the mAb166 IgG. All 12 strains displayed a band corresponding to PcrV at approximately 32.3 kDa, demonstrating that all PcrV variants were successfully recognized by mAb166 IgG (Fig. 5B-1 ). However, considerable variation in band intensity was observed, which may reflect differences in PcrV expression and secretion levels among strains, as well as possible differences in antibody binding to the PcrV variants. To accurately assess binding capacity while excluding the influence of expression and secretion levels, the pcrV gene from each strain was expressed using the same vector transformed into E. coli . Under these conditions, all PcrV variants showed comparable band intensity in immunoblots with mAb166 IgG, suggesting that mutations in PcrV have minimal impact on the binding capability of mAb166 IgG (Fig. 5B-2 ). Fig. 5. Phylogenetic analysis and 3D-structure prediction. Open in a new tab A Phylogenetic tree of PcrV-associated non-synonymous SNPs among 12 P. aeruginosa strains. The anti-PcrV antibody (m166) recognizes the blocking domain of PcrV (aa #144–257). Red: Mutation within the domain. Green: Mutations outside the domain. The distance matrix is calculated using pairwise alignment, and a guide tree is then constructed based on this matrix using the Neighbor-Joining method. B Immunoblot analysis of PcrV protein variants bound by anti-PcrV IgG (mAb166). 1) PcrV was obtained from cultures of various PcrV strains. 2) PcrV variants were expressed using the same expression vector transformed into E.coli . Top panel: SDS-PAGE stained with Coomassie Brilliant Blue (CBB). Bottom panel: Immunoblot using anti-PcrV IgG mAb166 conjugated with HRP. Images were independently obtained at least twice, all showing consistent results. C PcrV sequence (PAO1), mutation sites, and blocking epitope recognized by mAb166 (blue). Green dotted lines: Helical structure forming a coiled-coil. Positions a and d among the seven amino acid residues (a to g) that constitute the helix are indicated. Red dotted line: Overlap with the m166-VH-CDR3 region. D AlphaFold3 prediction of PcrV blocking domain. AlphaFold3 prediction of pentameric structure binding to m166 Fab ( E ) and m166 scFv ( F , G ). E , F Left: Side views of PcrV bound to m166 Fab or scFv-m166. Right: Top views of binding structure. G . The positions of non-synonymous SNPs. Source data are provided as a Source Data file. Structurally, PcrV is predicted to form a mushroom-shaped cap comprising a coiled-coil shaft—formed by central and C-terminal helical regions—and a globular domain of pentameric PcrV in complex with the anti-PcrV blocking antibody (Fig. 5C, D ) 30 – 32 . AlphaFold3-based structural modeling with Visual Molecular Dynamics (VMD) revealed that the mAb166 Fab and the engineered scFv-m166 predominantly bind to the globular domain at the periphery of the secretion pore and partly block its outlet (Figs. 5E , F , S4 , and S5 ). Further modeling using AlphaFold3 indicated that the identified amino acid substitutions, including those within the blocking epitope, are located at positions spatially distant from the mAb166 binding interface and do not induce significant conformational changes to the mushroom-like cap structure of PcrV (Figs. 5G and S4 ). These findings support the structural robustness of scFv-m166 antibody binding across genetically diverse P. aeruginosa isolates. Multiple scFv-m166 molecules can bind simultaneously to a single PcrV oligomer with minimal steric hindrance Besides the antibody binding site within the pentameric PcrV, the three-dimensional configuration of the antibody and PcrV is also critical for antibody potency 42 . In this regard, structural modeling showed minimal steric hindrance when a single scFv-m166 molecule binds to a PcrV pentamer (Fig. S6A ). Furthermore, the modeling suggests that two scFv-m166 molecules can simultaneously bind to a single PcrV pentamer with minimal steric hindrance between the two scFv proteins (Fig. S6B ). To test this possibility experimentally, we examined the binding behavior of recombinant scFv-m166 and oligomeric PcrV in vitro. Size-exclusion chromatography (SEC) demonstrated that two or more scFv-m166 molecules can bind simultaneously to a single PcrV oligomer (Fig. S7 ), consistent with the AlphaFold3-based modeling as shown in Fig. S6B . scFv mRNA provides protection in clinically relevant models using immunocompromised mice and antimicrobial‑resistant P. aeruginosa Next, we evaluated the therapeutic potential of anti-PcrV scFv mRNA against P. aeruginosa infections using these clinical isolates. Since P. aeruginosa infections frequently occur in immunocompromised patients, we modeled this clinical scenario using immunocompromised mice. Immunosuppression was induced by intraperitoneal cyclophosphamide administration, given 2 and 4 days prior to bacterial challenge (Fig. 6A ). This regimen successfully induced leukopenia, as confirmed in Fig. 6B . To assess the broad efficacy of our approach, we conducted an initial screen using the 12 P. aeruginosa strains described earlier. In this setup, each group consisted of 12 mice, with each mouse infected with a different strain. Thirty minutes post-infection, mice received an i.v. injection of scFv-control mRNA, scFv-m166 mRNA, scFv-m166-mFc mRNA, mAb166 protein (all at 10 µg), or colistin sulfate (125 µg). Survival was then monitored over 7 days. Given the high cytotoxicity of exoS ⁻/ exoU ⁺ clinical isolates and the use of immunocompromised mice, a lower bacterial dose (2.5 × 10 3 CFU/mouse) was used in this experiment compared with the dose used in previous experiments with the PA103 laboratory strain and immunocompetent mice (Figs. 1 – 3 ; 1 × 10 6 CFU/mouse). Fig. 6. Therapeutic efficacy of treatment in immunocompromised mice infected with 12 clinical isolates of P. aeruginosa. Open in a new tab A Schematic of the experimental protocol. B Effects of cyclophosphamide on peripheral blood leukocyte counts and body weight ( n = 10 mice per group; each data point represents an individual mouse). Body weight changes are presented as mean ± SD. Leukocyte counts are displayed as maximum, 75th percentile, median, 25th percentile, and minimum. Kruskal–Wallis nonparametric tests, with multiple comparisons adjusted by Bonferroni correction, were performed. All tests were performed as two-sided analyses. P < 0.05 vs. untreated controls. C Survival following bacterial challenge. The survival rate in the scFv-control group was significantly lower than that in the uninfected control group ( P = 0.042). The colistin group exhibited significantly improved survival compared with the scFv-control group ( † P = 0.02). The value of n represents the number of mice in each group. Survival differences were assessed by the log-rank test with P values adjusted using the Benjamini–Hochberg method. All tests were performed as two-sided analyses. Significant differences in peripheral blood leukocyte counts were observed across the time points after cyclophosphamide treatment. Specifically, the results were as follows: peripheral blood leukocyte counts: H (2) = 20.703, P < 0.001 ( n = 50). * P < 0.05 vs. uninfected control; † P < 0.05 vs. scFv-control mRNA. D Summary of therapeutic outcomes. Strains were classified into low-virulence ( exoS ⁺/ exoU ⁻) and high-virulence ( exoS ⁻/ exoU ⁺) groups. %Rescue indicates the proportion of strains that were lethal in the scFv-control group but not in the scFv-m166 group. Statistical analyses were performed using Pearson’s chi-square test. N.S .: Not significant ( P ≥ 0.05). Source data are provided as a Source Data file. Among the 12 mice treated with scFv-control mRNA, 10 died within 7 days, corresponding to a survival rate of 16.7% (Fig. 6 C and 6D ). Mortality in this group occurred later than that observed in experiments using the PA103 laboratory strain and immunocompetent mice (Figs. 1 – 3 ). This difference may be attributable to the lower bacterial dose (2.5 × 10 3 CFU/mouse) used in the current experiment compared with that in the previous experiments (1 × 10 6 CFU/mouse, Figs. 1 – 3 ). Furthermore, differences in proinflammatory responses between immunocompromised and immunocompetent models may also affect survival, as excessive inflammation can lead to lung tissue damage. The treatment with colistin sulfate (125 µg) resulted in four deaths (66.7% survival, P = 0.02 vs. scFv-control mRNA), and the administration of mAb166 protein (10 µg) led to five deaths (58.3% survival, P = 0.053 vs. scFv-control mRNA) over the same period. Notably, only 3 of the 12 mice treated with scFv-m166 mRNA died, yielding a significantly higher survival rate of 75.0% ( P = 0.02 vs. scFv-control mRNA). However, scFv-m166-mFc mRNA treatment resulted in seven deaths (41.7% survival, P = 0.053 vs. scFv-control mRNA), indicating reduced efficacy. Of the 10 P. aeruginosa strains that caused fatal infections in mice treated with scFv-control mRNA, treatment with scFv-m166 conferred therapeutic efficacy in infections with seven strains, leading to survival of the infected mice, corresponding to a 70% rescue rate for infections that were otherwise lethal ( P < 0.001) (Fig. 6D ). Similarly, within the 7-day observation period, five mice (50.0%, P = 0.01) treated with mAb166 protein and six mice (60.0%, P = 0.01) treated with colistin sulfate succumbed to infection. Importantly, focusing on infections caused by the seven highly cytotoxic exoU + clinical isolates, all mice treated with scFv-control mRNA died, whereas six of seven mice (85.7%) treated with scFv-m166 mRNA survived ( P = 0.001), highlighting the potent protective effect of scFv-m166 mRNA in severe infection settings. To further evaluate the potential of our approach against highly cytotoxic and lethal exoU + clinical isolates, we conducted therapeutic experiments targeting the KN41 and KP2401 strains. These two strains exhibited the highest cytotoxicity toward BEAS-2B bronchial epithelial cells among the 12 strains tested (Fig. 4D ). KN41 is resistant to piperacillin (PIPC), aztreonam (AZT), gentamicin (GM), and ciprofloxacin (CPFX) (Fig. 4A ). KP2401, which is resistant to PIPC, imipenem/cilastatin (IPM/CS), meropenem (MEPM), AZT, GM, amikacin (AMK), and CPFX, qualifies as a multidrug-resistant P. aeruginosa (MDRP) under international criteria by exhibiting resistance to all three major antibiotic classes: carbapenems, aminoglycosides, and fluoroquinolones. As shown in Fig. 7 , low bacterial doses (5.0 × 10 3 CFU/mouse) were used in this experiment, which involved highly cytotoxic bacterial strains and immunocompromised mice. Following the induction of leukopenia via two injections of cyclophosphamide, mice were challenged with a clinical P. aeruginosa isolate and received i.v. mRNA treatment 30 min post-infection (Fig. 7A ). In this immunocompromised model, all mice infected with KN41 (Fig. 7B ) or KP2401 (Fig. 7C ) and treated with scFv-control mRNA died within 2–3 days. In mice infected with KN41, mRNA antibody treatment significantly improved survival, with 60.0% survival for scFv-m166 mRNA and 36.4% for scFv-m166-mFc mRNA 1 week post-challenge (Fig. 7B-2 ). Colistin sulfate, a curative antimicrobial agent, and mAb166 IgG protein also conferred survival benefits, with survival rates of 40.0% and 50.0%, respectively (Fig. 7B-1 ). However, the therapeutic effects of these agents were limited in the KP2401 model, with survival rates of 18.6% for colistin sulfate and 20.0% for mA166 protein (Fig. 7C-1 ). Notably, in this model, scFv-m166 mRNA achieved a survival rate of 45.5% (Fig. 7C-2 ). These results underscore the therapeutic potential of scFv antibody mRNA in a clinically relevant infection model involving multidrug-resistant, highly cytotoxic strains. Fig. 7. Therapeutic efficacy in immunocompromised mice infected with highly cytotoxic multidrug-resistant P. aeruginosa. Open in a new tab A Experimental design. Immunocompromised mice were intratracheally challenged with lethal doses of P. aeruginosa clinical isolates, followed by therapeutic interventions. B KN41 strain. Survival of immunocompromised mice treated 30 min post-infection with KN41 via intravenous injection of: (1) colistin sulfate (125 µg), mAb166 IgG (10 µg), or scFv-control mRNA; (2) scFv-m166 mRNA or scFv-m166-mFc mRNA. C KP2401 strain. Survival of immunocompromised mice treated 30 min post-infection with KP2401 via intravenous injection of: (1) colistin sulfate (125 µg), mAb166 IgG (10 µg), or scFv-control mRNA; (2) scFv-m166 mRNA or scFv-m166-mFc mRNA; (3) combination of colistin sulfate (125 µg) with scFv-m166 mRNA or scFv-m166-mFc mRNA. D Delayed treatment (KP2401 strain). Survival of immunocompromised mice treated at ( 1 ) 2 hours, ( 2 ) 4 hours, or ( 3 ) 8 hours post-infection with KP2401 via intravenous injection of scFv-m166 mRNA alone, scFv-m166 mRNA + colistin sulfate (125 µg), or colistin sulfate (125 µg) alone. The value of n represents the number of mice in each group. For statistical analysis, survival curves were compared using the log-rank test with P values adjusted by the Benjamini–Hochberg method. All tests were performed as two-sided analyses. * P < 0.05 vs. uninfected control; † P < 0.05 vs. scFv-control mRNA. Source data are provided as a Source Data file. To evaluate the efficacy of scFv mRNA treatment in established lung infections, we extended the interval between bacterial challenge and mRNA administration in a model using the KP2401 strain. Delayed administration of scFv-m166 mRNA at 2, 4, or 8 h post-challenge still significantly prolonged mouse survival compared with that of control mice injected with scFv-control mRNA at 2 h post-challenge (Fig. 7D ). These findings suggest the potential of scFv-m166 mRNA therapy for treating established P. aeruginosa lung infections. Moreover, when compared to colistin sulfate, to which KP2401 is susceptible, m166 mRNA demonstrated comparable or slightly higher pro-survival efficacy. With its robust antibacterial activity, scFv-m166 mRNA represents a promising therapeutic option even against bacteria resistant to existing antimicrobials. Although survival rates gradually declined as the interval between challenge and treatment increased, this may be explained by bacterial proliferation and disease progression prior to treatment. It is also important to note that, in this model, mice received a high bacterial dose immediately, creating harsher conditions than in typical clinical settings. We also evaluated a combination of scFv mRNA and an antimicrobial, as combining alternative treatments with standard therapies represents a realistic and reasonable approach for future clinical applications. In this experiment, scFv-m166 or scFv-m166-mFc mRNA and colistin sulfate (125 µg) were intravenously administered at 30 min, 2 h, 4 h, or 8 h post-challenge with KP2401, which is susceptible to colistin sulfate at 2 µg/mL. Combination therapy achieved 100% 1-week survival, regardless of the interval between bacterial challenge and treatment (Fig. 7C-3, D ). In contrast, 1-week survival rates were below 50% for monotherapy with either scFv-m166 or scFv-m166-mFc mRNA or colistin sulfate. These findings demonstrate the superiority of combination therapy over monotherapy, highlighting the strong potential of scFv-m166 mRNA as an adjuvant treatment. Fc-free scFv effectively migrates from the bloodstream to the lung epithelium Upon closer observation, Fc-free scFv-m166 mRNA tended to yield higher survival rates than the Fc-conjugated scFv-m166-mFc mRNA in both KN41 and KP2401 (Fig. 7 ), as well as in the experiment involving 12 strains (Fig. 6 ). A similar trend was observed in therapeutic experiments using a laboratory strain, PA103 (Fig. 3C ). In contrast, in prophylactic experiments, Fc-free and Fc-conjugated formulations produced comparable outcomes in terms of body temperature, physical activity, survival rates, bacterial loads, and proinflammatory responses in the lungs (Figs. 1 and 2 ). The i.t. bacterial challenge model led to a rapid infection onset, with hypothermia and reduced activity observed as early as 4 h post-challenge (Fig. 1C, D ). Given the acute nature of the disease, differences in the mRNA administration schedule between therapeutic and prophylactic models may critically influence treatment outcomes. Specifically, the therapeutic model may require faster antibody distribution to the infection site—the lung epithelium—than the prophylactic model. In this context, previous studies have reported that the presence of an Fc region influences the microdistribution of antibodies within tissues 43 – 45 . To investigate this, we evaluated the effect of Fc conjugation on the temporal and spatial behavior of scFv antibodies in detail. To quantify antibodies in tissue samples using enzyme-linked immunosorbent assay (ELISA), we first generated standard curves using recombinant scFv-m166 and scFv-m166-mFc proteins derived from Chinese hamster ovary (CHO)-K1 cells transiently transfected with the mammalian expression vectors pcDNA3.1-scFv-m166 and pcDNA3.1-scFv-m166-mFc, each incorporating a carboxyl-terminal c-Myc tag and a hexahistidine ( 6× His) tag (Fig. S8 A, S8B ). The scFv-m166 and scFv-m166-mFc proteins expressed in CHO-K1 cells were subsequently purified in small quantities using immobilized metal affinity chromatography (IMAC) via the C-terminal 6× His tag (Fig. S8 C, S8D ). Immunoblotting analysis under non-reducing SDS-PAGE conditions confirmed that recombinant scFv-m166-mFc was produced as a dimer with an approximate molecular weight of 108.2 kDa (Fig. S8E ). The binding affinities of scFv-m166 and scFv-m166-mFc antibodies to the PcrV antigen were assessed by ELISA, yielding half-maximal effective concentrations (EC₅₀) of 35.0 and 57.2 nM, respectively (Fig. S8F ). We conducted a time-course analysis of anti-PcrV titers in the blood (Fig. 8A ), liver (Fig. 8B ), spleen (Fig. 8C ), and lungs (Fig. 8D ) of mice following i.v. injection of either scFv-m166 or scFv-m166-mFc mRNA, using ELISA on tissue homogenates. Both Fc-free and Fc-conjugated scFv antibodies became detectable in the blood and all tested organs as early as 2 h post-injection. This rapid antibody production is a favorable feature of mRNA in antibody therapy, particularly for targeting acute infections. The blood and tissue levels of Fc-free scFv-m166 antibodies peaked at 3–4 h post-injection. Notably, they remained detectable in the liver, spleen, and lungs for up to 24 h. This persistence of Fc-free scFv may reflect the advantages of mRNA formulations in promoting sustained protein expression. Fc-conjugated antibodies exhibited even greater persistence, maintaining high levels for up to 48 h post-injection. Fig. 8. Temporal and spatial profiling of scFv antibody expression following i.v. mRNA administration. Open in a new tab A – E ELISA of anti-PcrV antibody titers, measured using plates coated with recombinant PcrV protein. A Blood, B Liver, C Spleen, D Lungs. n = 3. E . BALF. n = 16. Box plots show the maximum, 75th percentile, median, 25th percentile, and minimum values. The lower limit of detection is 0.001 nM. The antibody titer is expressed as the equivalent concentration in nanomoles (nM), calculated from the standard curve generated with the recombinant scFv-m166 antibody protein (Fig. S8F ). The value of n represents the number of mice in each group, and each circle in the figure shows data from an individual mouse. Kruskal–Wallis nonparametric tests, with multiple comparisons adjusted by Bonferroni correction, were performed. All tests were performed as two-sided analyses. Significant differences were observed across all experimental groups for all measured parameters. The specific results are as follows: blood titers: H (14) = 41.606, P < 0.001 ( n = 44); Liver titers: H (12) = 34.198, P = 0.001 ( n = 38); Spleen titers: H (38) = 31.806, P = 0.001 ( n = 12); Lu n g titers: H (12) = 28.873, P = 0.004 ( n = 38); BAL titers: H (2) = 31.767, P < 0.001 ( n = 38). * P < 0.05 vs. uninfected control; † P < 0.05 between scFv-m166 and scFv-m166-mFc mRNA groups. F Time-course immunoblot analysis of scFv antibodies in liver homogenates. Cell lysate from BEAS-2B cells transfected with the eukaryotic cell expression vector pDS-scFv-m166, which expresses the scFv-m166 antibody carrying the Igκ secretion signal (Igκ-ss-scFv-m166), was used as a positive control. G Immunoblot analysis of scFv antibodies in liver homogenates collected 3 h post-injection under non-reducing and reducing conditions. Immunoblot images were independently obtained at least twice, all showing consistent results. Source data are provided as a Source Data file. In the lungs, the primary target site for treating i.t. P. aeruginosa infection, the signal from Fc-free scFv-m166 antibodies was lower than that from Fc-conjugated antibodies 3 h post-injection or later. To assess antibody levels on the epithelial side, we performed bronchoalveolar lavage (BAL) 3 h post-injection, as antibody translocation to the airway epithelium is critical for treating airway infections. Intriguingly, Fc-free scFv-m166 antibodies were significantly more abundant in BAL fluid (BALF) than Fc-conjugated antibodies were (Fig. 8E ). This contrasts with the findings from whole lung homogenates, where Fc-conjugated antibodies showed stronger signals (Fig. 8D ). These results indicate that the Fc-free formulation migrates from the bloodstream to the airway epithelium more efficiently than the Fc-conjugated formulation does. To explain the differences in tissue penetration, we evaluated antibody dimerization status in mice using liver homogenates, as the liver showed the strongest signal among the tested organs (Fig. 8A–D ). Time-course immunoblots of liver samples revealed strong signals 3–4 h post-injection for both scFv-m166 and scFv-m166-mFc antibodies (Fig. 8F ), consistent with the ELISA results (Fig. 8B ). Based on these findings, we selected the 3-h post-injection samples for SDS-PAGE under reducing and non-reducing conditions (Fig. 8G ). The scFv-m166-mFc antibodies were larger under non-reducing conditions than under reducing ones, while the size of Fc-free scFv-m166 antibodies remained unchanged between the two conditions. These results suggest that only the Fc-conjugated antibodies dimerize in the body. Although their observed size under non-reducing conditions differed from the theoretical 108.5 kDa, the dimer structure may affect electrophoretic mobility. The dimerized scFv-m166-mFc antibodies (108.5 kDa) are theoretically 3.7-fold larger than the monomeric scFv-m166 ones (29.2 kDa), which may explain the enhanced migration of Fc-free antibodies into the airway epithelium (Fig. 8E ). This efficient migration might be beneficial in controlling acute P. aeruginosa infections (Figs. 3 , 6 , and 7 ). For detailed profiling of antibody expression, we performed immunohistochemical analysis targeting the carboxyl-terminal 6× HIS tag of the expressed scFv antibodies (Fig. 9 ). In the liver, hepatocytes and sinusoidal lining cells expressed scFv-m166 and scFv-m166-mFc antibodies 2–4 h after mRNA injection (Fig. 9A-1 ). Quantitative analysis revealed significant expression in both groups (Fig. 9A-2 and S9 ). In the spleen, scFv-m166 and scFv-m166-mFc expression was observed primarily in immune cells (histiocytes and plasma cells) 3 h after injection (Fig. 9B ). Notably, staining was observed in the cytoplasm of liver and spleen cells, presumably reflecting intracellular antibody prior to secretion. In contrast, extracellular antibodies were undetectable by this method. In the lung, scFv-m166 and scFv-m166-mFc expression was not observed in any of the mice investigated (Fig. 9C ), suggesting low mRNA delivery efficiency to the lungs. However, ELISA confirmed substantial antibody levels in the lungs (Fig. 8D ), likely representing secreted antibodies. These findings suggest that mRNA LNPs induce antibody expression primarily in liver and spleen cells, followed by migration to the lungs via circulation. Importantly, the histological observations revealed no tissue damage in the liver, spleen, or lungs, supporting the safety of the treatment. Fig. 9. Tissue distribution of anti-PcrV scFv antibody expression following i.v. mRNA administration. Open in a new tab Immunohistochemical staining for 6× HIS-tag was performed to assess anti-PcrV scFv expression following i.v. injection of scFv mRNA LNPs. A. (1) Representative liver sections collected 1 to 4 h post-injection. Yellow arrows indicate prominent and focal scFv-m166-mFc expression observed predominantly in zone 2 hepatocytes. Red arrows indicate prominent scFv-m166 expression observed in sinusoidal lining cells. (2) Quantitative analysis of liver sections at 3 h post-injection ( n = 3). Quantification was performed using convolution analysis across five randomly selected microscopic fields per sample. Data are shown as the interquartile range (box), median (centerline), and minimum and maximum values (whiskers). Each data point in the figure represents an individual mouse. Group comparisons were performed using the independent-samples median test (nonparametric), followed by pairwise comparisons. Significance values were adjusted using the Bonferroni correction for multiple testing. All tests were performed as two-sided analyses. Significant differences were observed across all experimental groups, as follows: median = 6.160, H (2) = 6.300, P < 0.001 ( n = 9). * P < 0.05 vs. saline i.v. group. Immunohistochemical images of the spleen ( B ) and lungs ( C ) collected 3 h post-injection. Immunostaining was repeated in multiple mice per group. For the key lung data, quantitative evaluation was performed by board‑certified pathologists using five mice, as shown in Fig. S9 . Yellow arrows indicate positively stained immune cells (histiocyte & plasma cells). Source data are provided as a Source Data file. A more detailed toxicological evaluation was performed first by blood chemistry analysis at baseline (day 0) and at 1, 2, and 4 weeks following i.v. injection of scFv-m166 mRNA or scFv-m166-mFc mRNA. The treatment induced negligible changes in nine serum biochemical parameters, including aspartate aminotransferase, alanine aminotransferase, and lactate dehydrogenase (LDH) (Fig. S10A ). In addition, liver pathology was examined at the same time points, as the liver is a major target organ for intravenously delivered LNPs and exhibited the highest protein expression among the organs (Fig. S1 ). Hematoxylin and eosin (H&E) staining revealed no evidence of acute hepatocellular injury or inflammation, and Sirius Red staining confirmed the absence of fibrosis developed throughout the 4-week observation period (Fig. S10B ). These findings support the safety of scFv mRNA therapy using LNPs. To quantitatively evaluate the tissue distribution of protein expression from mRNA, we intravenously injected LNP-encapsulated mRNA encoding a non-secreted form of luciferase. Among various organs, the liver showed the highest luciferase expression, followed by the spleen (Fig. S1A ). Other organs, including the lungs, exhibited negligible expression. This supports the assertion that mRNA-encapsulating LNPs primarily induce antibody expression in the liver and spleen, with antibodies subsequently migrating to the lungs via circulation. Consistent with the liver histology (Fig. 9A ), luciferase expression was observed in hepatocytes and sinusoidal lining cells (Fig. S1B ). Discussion This study demonstrates the feasibility of antibody mRNA therapy for Pseudomonas aeruginosa lung infection by employing mRNA encoding an scFv antibody targeting the PcrV protein. Notably, our approach demonstrated therapeutic benefits in a clinically relevant model of immunocompromised mice infected with clinically isolated antimicrobial-resistant strains that exhibit enhanced cytotoxicity (Figs. 6 and 7 ). Clinical isolates of P. aeruginosa display high genetic and phenotypic diversity, reflecting the organism’s adaptability to diverse environments. Consequently, each isolate displayed distinct antimicrobial susceptibility and cytotoxicity profiles (Fig. 4 ). In recent years, the emergence of MDRP strains resistant to all three major classes of antibiotics—carbapenems, aminoglycosides, and fluoroquinolones—has become a significant public health threat. Additionally, strains resistant to carbapenems alone are also becoming increasingly prevalent. Regarding cytotoxicity, ExoU, a T3SS-associated cytotoxin, plays a critical role by inducing cell death by disrupting eukaryotic cell membranes through its phospholipase A 2 activity 36 , 46 . ExoU is a key contributor to pulmonary damage in acute lung injury 47 , and the exoU + genotype is associated with bacteremia and poor clinical outcomes 48 – 50 . Intriguingly, anti-PcrV scFv mRNA induced significant therapeutic effects in immunocompromised mice infected with exoU + strains that are highly cytotoxic and multidrug-resistant (Figs. 6 and 7 ), underscoring the potential of this approach in combating AMR (Fig. 10 ). Fig. 10. Schematic overview of this study and the design of anti-PcrV scFv mRNA constructs. Open in a new tab Intravenous administration of LNPs encapsulating scFv mRNA targeting P. aeruginosa PcrV inhibits type III secretion system-mediated virulence, protecting mice from bacterial challenge. This study also explores structural predictions of the PcrV pentamer using AlphaFold3, along with molecular interaction modeling between the PcrV pentamer and the scFv-m166 antibody (Fig. 5 ). The antibody is predicted to interact with the outlet margin at the tip of the needle structure formed by the PcrV pentamer, primarily via its VH-CDR3 domain, effectively capping the secretion pore. Our prediction aligns with recent reports demonstrating that the antibody binds to both the outer region (residues 165–223) and the inner rim of the central pore (residues 225–249) within the mushroom-shaped globular domain at the distal end of the coiled-coil shaft formed by the central and C-terminal helices of PcrV 51 . This suggests two possible mechanisms for the blocking activity of anti-PcrV antibodies: (1) inhibition of toxin extrusion through the needle by binding to the peripheral region of the pore formed by the PcrV pentamer and (2) interference with the interaction between the PcrV pentamer and the pore formed by PopB and PopD in the membrane of the target eukaryotic cell 52 , 53 . The AlphaFold3-based structural interaction predictions and previous reports of amino acid substitutions in PcrV suggest that known mutations do not significantly affect the blocking activity of m166-series anti-PcrV antibodies 41 . This may explain the high therapeutic efficacy of scFv-m166 mRNA against clinical isolates as observed in this study (Figs. 6 and 7 ). Current antibody mRNA therapies generally favor Fc-conjugated formulations 22 , 23 , 25 , 28 , 33 , to prolong antibody retention through neonatal Fc receptor-mediated recycling 29 . In line with this, the present study also demonstrated clear effects of Fc conjugation in increasing scFv concentrations in the blood, liver, spleen, and lungs (Fig. 8 ). This may contribute to the higher pro-survival efficacy observed when Fc-conjugated scFv mRNA was administered 48 h prior to bacterial challenge, compared with that for the Fc-free formulation (Fig. 1E-2 ). These findings suggest that Fc conjugation is beneficial for extending antibody function, particularly in prophylactic settings. In contrast, only a few studies have explored the potential of Fc-free scFv antibodies in mRNA-based therapies, and the advantages of Fc-free formulations over Fc-containing formulations remain largely unclear 54 – 57 . In this context, our study highlights the improved biodistribution of the Fc-free formulation. Fc-free scFv migrated to the lung epithelium, the infection site, more efficiently than Fc-conjugated scFv (Fig. 8E ), which may be advantageous in therapeutic settings involving acute bacterial infection. Notably, translocation of IgG from the bloodstream to the airway epithelium is inefficient; in large animals and humans, serum concentrations of intravenously injected IgG are approximately 500- to 2000-fold higher than those in BALF 58 , 59 . This limits the utility of full-length antibodies in treating bacterial airway infections 60 . In contrast, given their smaller molecular size, Fc-free scFv antibodies can penetrate tissues more efficiently than full-length ones can 43 – 45 , offering a promising solution to this barrier to distribution. Indeed, in our study, the BALF concentration of Fc-free antibodies was more than 6-fold higher than that of Fc-conjugated formulations (Fig. 8E ), despite the blood concentration of the Fc-free formulation being approximately one-third of that of the Fc-conjugated formulations (Fig. 8A ). This enhanced migration of Fc-free antibodies to the lung epithelium may explain their better therapeutic outcomes following bacterial challenge (Figs. 3 , 6 , and 7 ). However, the benefit of Fc-free formulations in accumulating at the site of infection comes at the cost of rapid clearance from circulation, as Fc-free antibodies are typically eliminated within tens of minutes 29 . mRNA therapeutics can overcome this limitation by enabling continuous in vivo antibody production over the course of a day or more 61 – 63 . Indeed, Fc-free antibodies delivered via mRNA remained detectable in the liver, spleen, and lungs for up to 24 h (Fig. 8 ). Furthermore, they exhibited prophylactic effects when administered 48 h prior to bacterial challenge (Fig. 1E ). This suggests that functional Fc-free scFv persisted in mice for over 48 h, even though scFv levels fell below the detection limit at 48 h post-injection (Fig. 8 ). Another requirement for the Fc-free strategy is that the scFv must function independently of Fc-mediated effector functions, such as antibody-dependent phagocytosis. In this context, anti-PcrV antibodies block the transfer of type III secreted toxins into host cells, thereby mitigating lung damage and inflammation (Figs. 2 and S2 ). Furthermore, PcrV blockade can prevent T3SS-mediated damage to alveolar macrophages, preserving their ability to clear P. aeruginosa 16 . These Fc-independent mechanisms likely contribute to the preventive and therapeutic effects observed in this study. Despite the promise of our antibody mRNA therapy in controlling P. aeruginosa , the therapeutic advantage of scFv-m166 mRNA over mAb166 protein was modest. While mRNA-based antibody therapy offers advantages in manufacturing, post-translational modification, and the use of antibody variants, such as scFv, further improvements in therapeutic efficacy are needed. We plan to address this by fine-tuning LNPs and mRNA designs 64 – 67 . Despite these limitations, this study represents a significant step forward in the treatment of P. aeruginosa , demonstrating successful therapy in clinically relevant mouse models created using immunocompromised mice and highly cytotoxic antimicrobial-resistant clinical isolates. It also reveals the benefit of Fc-free formulations in antibody mRNA therapy. Additionally, our approach may be extended to other Gram-negative bacteria that utilize T3SS for their pathogenicity 30 , 31 , contributing to future strategies for combating the growing threat of AMR. Methods Ethical approval and clinical data collection Prior to the collection of clinical data from patients from whom the P. aeruginosa strains were isolated, ethical approval was obtained from the Ethics Committee of Kyoto Prefectural University of Medicine (KPUM) (approval numbers ERB-C-1069-1 and ERB-C-1864-2). The collected data included patient age, sex, hospital ward, date, and source of bacterial isolation, underlying medical conditions, and clinical outcomes. The study was conducted in accordance with the Declaration of Helsinki. The Institutional Review Board of KPUM waived the requirement for informed consent, as the study was a non-interventional, non-invasive, retrospective observational epidemiological investigation. All animal experiments were conducted with the approval of the Animal Care Committee of KPUM (Approval Nos. M2023-544 and M2024-535) or the Animal Care Committee of the Institute of Science Tokyo (Approval No. A2023-191C5). Anti-PcrV scFv mRNA and LNP preparation The mRNA sequence for the single-chain anti-PcrV antibody (scFv-m166) was designed based on genetic information from our previous reports on the murine monoclonal anti-PcrV m166 IgG 9 , 68 . We constructed the scFv to include the essential variable regions of the heavy (H) and light (L) chains from the anti-PcrV blocking antibody m166, linked by a glycine–serine linker (Gly 4 Ser 1 ) 3 . Additionally, we modified the sequence by adding a tissue plasminogen activator signal sequence (tPAss, MDAMKRGLCCVLLLCGAVFVSAR) at the amino-terminus and a cMyc tag along with a hexaihistidine ( 6× HIS) tag at the carboxyl-terminus (Fig. 1A ). To design the mRNA sequence of Fc-conjugated scFv (scFv-m166-mFc) antibodies, the carboxyl-terminus of tPAss-scFv was conjugated with a hinge region and CH2 and CH3 domains of mouse IgG1 followed by a cMyc tag and a 6× HIS tag downstream. Amino acid sequences are provided in Table S2 . mRNA encoding these sequences was prepared with codon optimization, N1-pseudouridine modification, Cap1 structure, and a 100 nt poly(A) tail, which was outsourced to GenScript Japan Co. (Tokyo, Japan). Luciferase mRNA was purchased from Trilink Biotechnologies (San Diego, CA, USA). LNPs were formulated from ALC-0315 (MedChemExpress, Monmouth Junction, NJ, USA), ALC-0159 (MedChemExpress), 1,2-distearoyl-sn-glycero-3-phosphocholine (Fujifilm Wako, Osaka, Japan), and cholesterol (Sigma-Aldrich, St. Louis, MO, USA) at a nitrogen/phosphate ratio of 6 using microfluidics, followed by buffer exchange, in accordance with our previous report 69 . Dynamic scattering measurement (DLS) was performed using a Zetasizer Nano-ZS (Malvern Instruments, Malvern, UK), and RiboGreen assay was performed using the Quant-it RiboGreen RNA Assay Kit (Thermo Fisher Scientific, Waltham, MA, USA), as described previously 70 . Specifically, DLS measurements were conducted using a diode laser (λ = 532 nm) at a scattering angle of 173°. The decay rate of the photon correlation function was analyzed using the cumulant method, and the hydrodynamic diameter of each sample was calculated according to the Stokes–Einstein equation. In the RiboGreen assay, fluorescence from lipid nanoparticles with or without Triton X-100 treatment was measured at an excitation wavelength of 480 nm and an emission wavelength of 520 nm using a plate reader (Microplate Reader Infinite 200PRO F Nano + , Tecan, Männedorf, Switzerland). P. aeruginosa strains Twelve P. aeruginosa strains were used in this study, including two laboratory strains and ten drug-resistant clinical isolates. The laboratory strain PAO1 (multilocus sequence typing (MLST): ST549) is the reference strain for the P. aeruginosa genome project and is a non-cytotoxic strain with the exoS ⁺/ exoU ⁻ genotype 71 , 72 . In contrast, the laboratory strain PA103 (MLST: ST298) is cytotoxic and has the exoS ⁻/ exoU ⁺ genotype 73 . The 10 clinical isolates were drug-resistant strains of P. aeruginosa obtained from hospitalized patients at the Kyoto Prefectural University of Medicine Hospital (Fig. 4A ). Among them, 9 strains—KN5, KN15, KN17, KN27, KN28, KN33, KN34, LN37, and KN41—were isolated between 2005 and 2014 and exhibited resistance to two classes of antibiotics 39 . The clinical isolate KP2401 (MLST: ST5295), collected in 2024, has an exoS ⁻/ exoU ⁺ genotype, carries a class 1 integron cassette, and displays resistance to carbapenems, amikacin, and fluoroquinolones, classifying it as a multidrug-resistant P. aeruginosa strain. Each of the four exoS + clinical isolates had a distinct MLST. Among the six exoU + clinical isolates, two strains—KN27 and KN41—were identified as ST235. Notably, KN27 was positive for the class 1 integron, whereas KN41 was negative. Although KN5, KN17, and KN28 were all classified as ST357, they exhibited distinct antimicrobial susceptibility profiles. KN5 was resistant to piperacillin (PIPC), ceftazidime (CAZ), gentamicin (GM), and amikacin (AMK), but remained susceptible to imipenem/cilastatin (IPM/CS) and meropenem (MEPM). KN17 was resistant to PIPC, CAZ, and IPM/CS, while remaining susceptible to GM and AMK. In contrast, KN28 was resistant to IPM/CS and MEPM, but susceptible to PIPC, CAZ, GM, and AMK. KN41 and KP2401 were susceptible to colistin sulfate at 2 µg/mL. Bacteria from frozen stocks were streaked onto trypticase soy agar plates and cultured in trypticase soy broth supplemented with 10 mM nitrilotriacetic acid (Sigma-Aldrich) at 32 °C for 13 h in a shaking incubator. The cultures were then centrifuged at 8,500 × g for 10 min, and the bacterial pellet was washed twice with saline before being diluted to the desired number of colony forming units (CFUs) per milliliter, as determined by spectrophotometry. The bacterial count was confirmed by plating diluted aliquots onto sheep blood agar and counting CFUs. Genomic DNA extraction For multilocus sequence typing, exoenzyme genotyping, and sequence analysis of the pcrV gene, bacterial genomic DNA was extracted from each clinical isolate using a QIAamp DNA Micro Kit (Cat. No. 65304; Qiagen, Hilden, Germany) according to the manufacturer’s instructions. Multilocus sequence typing (MLST) MLST was performed based on the sequence analysis of internal fragments of seven housekeeping genes ( acsA , aroE , guaA , mutL , nuoD , ppsA , and trpE ). These genes were amplified by polymerase chain reaction (PCR) with high-fidelity DNA polymerase (PrimeSTAR ® HS DNA Polymerase, Cat. No. R010A; Takara Bio, Kusatsu, Japan) using specific primer pairs as previously described 39 . The PCR conditions consisted of 30 cycles of denaturation at 94 °C for 30 s, annealing at 58 °C for 30 s, and extension at 72 °C for 30 s. The resulting amplicons were purified with a PCR Clean-Up Kit (Cat. No. 28104; Qiagen) and sequenced by Eurofins Genomics (Tokyo, Japan) using the designated sequencing primers 39 . Allelic numbers and sequence types (STs) were assigned by comparing the sequences with the P. aeruginosa MLST database ( https://pubmlst.org/paeruginosa/ ) 74 . Exoenzyme genotyping The type III secretion system effector genes, exoS and e xoU , were detected by diagnostic polymerase chain reaction using genomic DNA extracted from clinical isolates as templates. PCR was performed using PrimeSTAR ® HS DNA Polymerase (Cat. No. R010A; Takara Bio) with six sets of specific primers designed from the consensus regions of exoS and exoU in P. aeruginosa 39 . The amplification conditions consisted of 28 cycles of denaturation at 94 °C for 15 s, annealing at 60 °C for 30 s, and extension at 72 °C for 30 s. The resulting genotypes were determined via 2% agarose gel electrophoresis. Molecular and immunological characterization of PcrV variants To identify amino acid substitutions in PcrV, the full-length 884-bp coding region of the pcrV gene was amplified using high-fidelity DNA polymerase (PrimeSTAR ® HS DNA Polymerase; Takara Bio). Polymerase chain reaction was performed with primers flanking the open reading frame, PVOT1-5 (5′-TGCGTGGCTTGTTGATCTGA-3′) and PVOT1-3 (5′-TGCTGGTCGGTGTCGGAA-3′) 41 , under the following conditions: initial denaturation at 94 °C for 5 min; 36 cycles of 94 °C for 30 s, 52 °C for 30 s, and 72 °C for 1 min; and a final extension at 72 °C for 7 min. The resulting pcrV amplicons were cloned into the pCR-Blunt II-TOPO ® vector (Zero Blunt TOPO® Cloning Kit; Thermo Fisher Scientific) and sequenced in both directions (Eurofins Genomics). Phylogenetic analysis based on PcrV amino acid sequences was performed using the EMBL-EBI Clustal Omega tool ( https://www.ebi.ac.uk/jdispatcher/msa/clustalo ), and multilocus sequence typing eBURST analysis was conducted using PHILOViZ 2.0 ( https://www.phyloviz.net ). Structural modeling of the interaction between PcrV and the antibody was carried out using the AlphaFold3 Server ( https://alphafoldserver.com ) 75 and VMD (ver. 1.9.4a57, Biological Modeling, Carnegie Mellon University, https://biologicalmodeling.org/coronavirus/VMDTutorial ), where the complex was visualized as a molecular surface using the Surf rendering method and Fragment coloring scheme. For PcrV immunoblotting, 10 µL of supernatant from an overnight bacterial culture was collected and mixed under reducing conditions (2 × Laemmli sample buffer containing 50 mM dithiothreitol (DTT), Invitrogen NuPAGE 10× Sample Reducing Agent, Cat. No. NP0009; Thermo Fisher Scientific) at 85 °C for 5 min. Then, 10 µL of sample solution (equivalent to an extract from 0.67 mg of liver) was loaded onto a gel for sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) (NuPAGE Bis-Tris Mini Protein Gel 4–12%, Cat. No. NP0321; Thermo Fisher Scientific), and transferred to a poly vinylidene di-fluoride (PVDF) membrane (Invitrogen iBlot™ 3 Transfer Stacks, PVDF, mini 0.2 µm, Cat. No. IB34002; Thermo Fisher Scientific) via a dry blot module (Invitrogen iBlot™ 3, Cat. No. IB31001; Thermo Fisher Scientific). The membrane was blocked with 4% skim milk in PBS/T for 1 h and then incubated with horseradish peroxidase (HRP)-conjugated anti-PcrV antibody (Cat. No. MRO-156MZ-HRP, Lot No. HR1222LY09; Creative BioLab, San Diego, CA, USA) at 1:10,000 dilution in 4% skim milk in PBS with 0.01% Tween-20 (PBS/0.01%T) overnight at 4 °C. Chemiluminescence was induced via an HRP substrate (Western BLoT Quant HRP Substrate, Cat. No. T7103A; Takara Bio), and images were captured via an imaging device (Invitrogen iBright™ CL1500 Imaging System; Thermo Fisher Scientific). PcrV immunoblotting was also performed using purified E. coli -derived recombinant PcrV. For this purpose, the coding region of the pcrV gene from each strain was cloned into the lac operon–based expression vector pQE30, and E. coli M15 strains expressing 6× His-tagged PcrV proteins were generated. Following isopropyl β-D-1-thiogalactopyranoside-induced expression, the resulting E. coli lysates were subjected to SDS-PAGE, as described above. Cytotoxicity assay The cytotoxicity of P. aeruginosa isolates was evaluated using human bronchial epithelial cells (SV40-immortalized BEAS-2B cells; ATCC CRL-3588, Manassas, VA, USA). Upon reaching 95% confluence, the cells were co-cultured with bacteria (5 × 10 7 CFU/mL) in serum-free and antibiotic-free medium. Cell death was monitored over a 2- to 4-h period and quantified using the CytoTox 96 Non-Radioactive Cytotoxicity Assay (Promega, Madison, WI, USA). This assay measures lactate dehydrogenase (LDH) release from the cytosol into the culture supernatant via a colorimetric reaction (absorbance at 490 nm). Cytotoxicity is expressed as the proportion of cell death at 2 h relative to the 4-h time point. Additionally, at 4 h post-infection, cell viability was qualitatively assessed by 0.4% trypan blue staining. Infection of mice with P. aeruginosa strains Certified pathogen-free male ICR mice (7–9 weeks old, body weight 34.0 ± 0.9 g) were obtained from Shimizu Laboratory Supplies Co., Ltd. (Kyoto, Japan). The mice were housed in filter-top cages (5 per cage) under pathogen-free conditions. The animal room maintained a 12-hour light/dark cycle (lights on at 8:00 AM, lights off at 8:00 PM) at a constant ambient temperature of 23 ± 2 °C and relative humidity of 50 ± 10%. All mice had ad libitum access to water and a standard laboratory diet. We excluded female mice to minimize variability in immune responses associated with hormonal cycles. To ensure clear and consistent evaluation of the primary experimental variables, we used male mice to avoid potential fluctuations caused by the estrous cycle. The number of animals in each group is clearly indicated in the corresponding figures, as n values in legends or data points. The experiments were carried out in three distinct settings: a prophylactic setting in immunocompetent mice (Fig. 1A ), a therapeutic setting in immunocompetent mice (Fig. 3A ), and a therapeutic setting in immunocompromised mice (Figs. 6 A and 7A ). For each challenge experiment, the bacterial dose was set to reliably kill nearly all mice. The number of animals in each group is clearly indicated in the corresponding figures, as in values in legends or data points). In the 24-hour acute lung injury experiments, we used 20–24 mice per experimental set (10–12 mice per group, 2 groups). In the 7-day survival experiments, we used 30–36 mice per experimental set (10–12 mice per group, 3 groups). These numbers were established as the upper limits for handling individual animals. We included multiple groups—control groups and prophylactic/therapeutic treatment groups—and repeated the experimental sets on separate days to confirm reproducibility, completing multi-group datasets corresponding to the figures. Prophylactic setting in immunocompetent mice For prophylactic interventions, the mice received i.t. instillation of mAb166 protein (10 µg), an i.m. injection of scFv-m166 mRNA (10 µg), scFv-m166-mFc mRNA (10 µg), or luciferase mRNA (10 µg), or an i.v. injection of mAb166 protein (5 µg), scFv-m166 mRNA (10 µg), scFv-m166-mFc mRNA (10 µg), scFv-control mRNA (10 µg), or luciferase mRNA (10 µg). A formulation of lipid nanoparticle-encapsulated mRNA was administered 2, 48, or 96 h before the tracheal instillation of P. aeruginosa PA103 (1.0 × 10⁶ colony forming unit (CFU), 50 µL). In one group, the PA103 suspension (1.0 × 10⁶ CFU, 50 µL) was premixed with 10 µg of mAb166 IgG protein 34 , 68 , which was purchased from Creative Biolabs (Cat. No. MRO-156MZ, Lot No. CB1222LY09; New York, NY, USA). Body temperature, physical activity, and mouse survival were monitored for 24 h. The physical activity level was assessed as follows: Mice that began moving immediately at the start of observation and exited the circular area (approximately 20 cm in diameter) within 3 s were assigned a score of 4. A score of 3 was assigned to mice that exited within 5 s, a score of 2 to those that exited within 10 s, and a score of 1 to those that required more than 10 s to exit. Mice that had died by the time of assessment were assigned a score of 0. Tracheal infections were induced using an endotracheal needle (modified animal feeding needle, 24 G; Popper & Sons, Inc., New Hyde Park, NY, USA) under brief anesthesia with inhaled sevoflurane (Sevofrane®; Maruishi, Osaka, Japan). Twenty-four hours later, the lungs were harvested, weighed, and homogenized in sterile containers with sterile water. The homogenates were serially diluted and plated on sheep blood agar to quantify the bacterial load in the lungs. The sensitivity limits of the bacteriological assays were 10 CFU/ml for blood and 100 CFU/g for lung tissue. The supernatants from the lung homogenates were stored for measurement of inflammatory cytokine concentrations and myeloperoxidase activity. For each set consisting of three groups (scFv-control mRNA, scFv-m166 mRNA, or scFv-m166-mFc administered intravenously), mice were euthanized 12 h after i.t. bacterial challenge, and bacterial dissemination to the blood, liver, and spleen was quantitatively evaluated. In addition, for another set consisting of three groups (scFv-control mRNA, scFv-m166 mRNA, or scFv-m166-mFc administered intravenously; five mice per group), histopathological evaluation of the lungs was performed at 24 h post-infection, including five animals per group (both surviving and deceased individuals). Furthermore, an additional set of three groups (scFv-control mRNA, scFv-m166 mRNA, or scFv-m166-mFc administered intravenously, 10–12 mice per group) was used to assess survival over 1 week following i.t. bacterial inoculation. Therapeutic setting in immunocompetent mice Mice first received an i.t. instillation of PA103 (1.0 × 10⁶ colony forming unit (CFU), 50 µL), and 30 min later, they were intravenously injected with 100 µL solution of the lipid nanoparticle formulation containing mAb166 protein (5 µg or 10 µg), scFv-m166 mRNA (10 µg), scFv-m166-mFc mRNA (10 µg), or scFv-control mRNA (10 µg), or saline in a control group. The survival of the mice was monitored for 1 week after the intervention. Therapeutic setting in immunocompromised mice In the immunocompromised mouse series, leukopenia was induced by intraperitoneal injections of cyclophosphamide (200 mg/kg; Wako Pure Chemical Industries, Ltd., Tokyo, Japan) administered 4 and 2 days prior to the infection challenge 76 . Blood samples were collected from the tail vein 4 days after the first cyclophosphamide injection. In the designated groups, total leukocyte and differential cell counts were determined via a hemocytometer, and mouse body weights were recorded for 5 days following the cyclophosphamide injections. In the first set of experiments, each of 12 P. aeruginosa isolates (PAO1, PA103, KN5, KN15, KN17, KN27, KN28, KN33, KN34, KN37, KN41, or KP2401; 2.5 × 10³ colony forming unit (CFU) in 50 µL) was intratracheally administered to a pair of mice. Thirty minutes later, the mice were intravenously treated with lipid nanoparticle (LNP)-formulated scFv-control mRNA (10 µg) or scFv-m166 mRNA (10 µg). In the second set of experiments, the mice were intratracheally inoculated with either KN41 or KP2401 (5 × 10³ CFU in 50 µL). Thirty minutes later, they received an i.v. injection (100 µL) of LNP-formulated mRNA-encoding scFv-control (10 µg), scFv-m166 (10 µg), or scFv-m166-mFc (10 µg). Control groups were administered either colistin sulfate (125 µg; 5 mg/kg) or mAb166 protein (10 µg) intravenously. In the third set of experiments, KP2401 (3.0 × 10³ CFU in 50 µL) was intratracheally administered to mice. Two, four, or eight hours later, the mice were intravenously treated with LNP-formulated scFv-control mRNA (10 µg), scFv-m166 mRNA (10 µg), scFv-m166-mFc mRNA (10 µg), colistin sulfate (125 µg), or scFv-m166 mRNA (10 µg) + colistin sulfate (125 µg). Survival was monitored for 1 week. Enzyme-linked immunosorbent assay quantification of anti-PcrV titers After the i.v. injection of a formulation containing lipid nanoparticles encapsulating scFv-m166 or scFv-m166-mFc mRNA, the mice were euthanized at predetermined time points, and blood, liver, spleen, and lungs were collected. The liver, spleen, and lungs were homogenized in tissue lysis solution (15 mL/g tissue; T-PER, Tissue Protein Extraction Reagent, Cat. No. 78510; Thermo Fisher Scientific) and centrifuged to collect the supernatant for antibody titer measurement. Microwell plates (Nunc C96 Maxisorp; Thermo Fisher Scientific) were coated with rePcrV (1.0 μg/mL in 0.05 M NaHCO₃, pH 9.6) for 2 h at 4 °C. The plates were then washed twice with PBS containing 0.05% Tween-20 (Cat. No. P9416; Sigma-Aldrich) (PBS/0.05%T) and blocked with 200 μL of 1% bovine serum albumin/PBS overnight at 4 °C. The samples (100× diluted blood or 1× tissue lysis solution) were applied to the plates (100 μL/well) and incubated for 2 h at 4 °C. After four washes with PBS/0.05%T, horseradish peroxidase-labeled anti-hexahistidine-IgG (Cat. No. HRP-66005; Proteintech, Tokyo, Japan) was added at a 1:10,000 dilution and incubated for 1 h at 37 °C. Following six washes with PBS/0.05%T, 2,2′-azino-bis(3-ethylbenzthiazoline-6-sulfonic acid) (Cat. No. A3219; Sigma-Aldrich) was added to the plates, which were subsequently incubated at room temperature for 30 min. The reaction was stopped by adding 0.5 M H₂SO₄ (100 μL/well), and the optical density (OD) at 450 nm was measured using a microplate reader (Cat. No. MTP-880Lab; Corona Electric Co., Hitachinaka, Japan). Samples with OD values greater than 0.15 at 450 nm were considered positive. Bronchoalveolar lavage fluid collection After euthanasia via the deep inhalation of sevoflurane, a tracheotomy was performed. A total of 2 mL of PBS was injected into the lungs using a syringe, and bronchoalveolar lavage fluid was collected under vibration with a vibration device. The recovery rate was approximately 50–70%. Following centrifugation at 400 × g (himac CT6E with T5SS swing rotor, Hitachi-koki, Tokyo, Japan) for 10 min, the supernatant was used to measure the anti-PcrV titer. Immunoblotting of anti-PcrV scFv The supernatant (100 µL, equivalent to an extract from 6.7 mg of tissue) from the liver homogenate (15 mL T-PER/g liver) of the mice was incubated with nickel‒nitrilotriacetic acid agarose (100 µL of 50% slurry, Cat. No. 30210; Qiagen) for 1 h at room temperature with constant shaking. After two washes with washing buffer, the bound protein components were eluted with elution buffer, mixed under either non-reducing (2× Laemmli sample buffer, Cat. No. 1610737; Bio-Rad Laboratories, Inc., Hercules, CA, USA) or reducing conditions (2× Laemmli sample buffer containing 50 mM dithiothreitol, Invitrogen NuPAGE 10× Sample Reducing Agent, Cat. No. NP0009; Thermo Fisher Scientific) at 85 °C for 5 min. Then, 10 µL of sample solution (equivalent to an extract from 0.67 mg of the liver) was loaded for sodium dodecyl sulfate-polyacrylamide gel electrophoresis (Miniprotean 4–15% TGX gel, Cat. No. 4561081; Bio-Rad Laboratories, Inc.), and transferred to a poly vinylidene di-fluoride or nitrocellulose membrane (Cat. No. IB34002 and Cat. No. IB33002; Thermo Fisher Scientific) via a dry blot module (Invitrogen iBlot™ 3, Cat. No. IB31001; Thermo Fisher Scientific). After blotting on a membrane, the membrane was incubated with 8% acetic acid for 15 min, 3% H₂O₂ in PBS/0.01%Tween-20 for 15 min, blocked with 4% skim milk in PBS/0.01%T for 1 h, and then incubated with a horseradish peroxidase (HRP)-conjugated anti-cMyc antibody (anti-Myc-tag mAb-HRP-DirecT, mouse IgG2bk, Cat. No. M192-7; MBL, Tokyo, Japan) at 1:10,000 dilution in 4% skim milk in PBS/0.01%Tween-20 overnight at 4 °C. Chemiluminescence was induced via an HRP substrate (Western BLoT Quant HRP Substrate, Cat. No. T7103A; Takara Bio), and images were captured via an imaging device (Invitrogen iBright™ CL1500 Imaging System; Thermo Fisher Scientific). Positive control for immunoblotting was prepared from lysates of BEAS-2B cells (human bronchial epithelium immortalized with SV40, Cat. No. CRL-3588; ATCC, Manassas, VA, USA) transfected with the eukaryotic expression vector pDS-m166-cMyc- 6x His via Lipofectamine 2000 Transfection Reagent (Invitrogen Cat. No. 11668027; Thermo Fisher Scientific). Cytokines and myeloperoxidase activity Interleukin-6 and tumor necrosis factor-α concentrations in lung homogenates and plasma were quantified using enzyme-linked immunosorbent assay (ELISA) kits (Cat. No. 550950 and 560478, BD OptEIA ELISA Sets; BD Biosciences, San Jose, CA, USA). To quantify airway inflammation, myeloperoxidase (MPO) activity in lung tissue was measured following the established protocol with minor modifications 77 , 78 . Briefly, frozen lung samples were thawed, minced, and homogenized in 50 mM potassium phosphate buffer (pH 6.0). The homogenate was centrifuged at 10,000 x g for 15 minutes, after which the supernatant was discarded. To release the enzyme from neutrophil granules, the pellet was resuspended in buffer supplemented with 50 mM hexadecyltrimethylammonium bromide (HTAB), homogenized, and further diluted with HTAB-free buffer. The samples then underwent three cycles of sonication and snap-freezing in liquid nitrogen to maximize MPO extraction. After a final centrifugation at 10,000 x g for 10 minutes, the supernatant was collected, diluted, and reacted with 0.167 mg/ml o-dianisidine dihydrochloride and 0.00001% H 2 O 2 . MPO activity was determined by monitoring the change in absorbance at 460 nm over a 2-minute period. Histopathological assay A designated mouse from each group was euthanized 24 h after infection for histological analysis. After euthanasia, the lungs were perfused with 10% buffered formalin phosphate and subsequently embedded in paraffin. Hematoxylin and eosin (H&E)-stained sections were examined by light microscopy. For the three major prophylactic i.v. scFv mRNA/lipid nanoparticle (LNP) treatment groups (scFv-control mRNA/LNP, scFv-m166 mRNA/LNP, and scFv-m166-mFc mRNA/LNP), acute lung injury was quantitatively assessed using H&E-stained paraffin sections of the lung from each mouse challenged with a lethal dose of P. aeruginosa PA103. For each mouse, 20 fields (10 in each of the left and right lobes) were randomly selected at ×40 objective magnification to evaluate the five parameters of the American Thoracic Society Lung Injury Scoring System (Figs. S2A-1, 2 ) 37 . A total Lung Injury Score was calculated, and the median [25th percentile–75th percentile] for each group was determined. As an additional assessment of acute lung injury, the presence of necrotizing inflammatory lesions was evaluated under light microscopy (×40), and each positive finding was scored as 1 point. The total score from 20 fields per mouse was calculated for each of the three groups (Fig. S2A-3 ). For the immunohistochemical analysis of antibody expression, the mice were euthanized after the injection of mRNA-encapsulating LNPs. Mouse liver, spleen, and lungs were excised and fixed in 4% buffered paraformaldehyde. Paraffin sections were stained with horseradish peroxidase-conjugated anti-hexahistidine tag monoclonal antibody (Cat. No. HRP-66005; Proteintech) and developed with 3,3′-diaminobenzidine by Biopathology Institute Co. (Kunisaki, Japan). All tissue sections presented in this study were examined by board‑certified pathologists accredited by the Japan Society of Pathology (N.T.M., Mo.K.). The staining intensity of 3,3′-diaminobenzidine in immunohistochemistry was quantified using the Color Deconvolution plugin within Fiji (an open-source platform for biological image analysis) 79 . Blood chemistry and liver histopathology following scFv mRNA/lipid nanoparticle (LNP) administration Blood samples were collected in small volumes before i.v. administration of scFv‑m166 mRNA/LNP or scFv‑m166‑mFc mRNA/LNP, and at 1, 2, and 4 weeks post‑administration. Nine serum biochemical parameters (AST: aspartate aminotransferase, ALT: alanine aminotransferase, LDH: lactate dehydrogenase, T-Bil: total bilirubin, TP: total protein, ALB: albumin, AMY: amylase, BUN: blood urea nitrogen, Cr: creatinine) were measured. At the same time point after prophylactic i.v. administration of the scFv mRNA, livers were collected from each mouse group, processed into paraffin sections, and subjected to hematoxylin and eosin, and Sirius red staining. Luciferase expression assay Four hours after the i.v. injection of LNPs, organs were collected and homogenized in a Passive Lysis Buffer (Promega). Luminescence was then measured using a Lumat3 LB9508 luminometer (Berthold Technologies, Bad Wildbad, Germany). The luminescence values were standardized based on protein amount. To quantify protein expression in each liver cell type, a single-cell suspension of liver cells was obtained as described previously, with slight modification to the protocols 80 . Specifically, the liver was perfused through the portal vein with 30 mL of perfusion buffer containing 8.3 g/L sodium chloride, 0.5 g/L potassium chloride, 2.3 g/L 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES), and 2 g/L glucose, followed by 50 mL of dissociation buffer composed of 49.5 mL of perfusion buffer and 0.5 mL of 476 mM calcium chloride supplemented with 1.5 mg of Liberase (Roche, Basel, Switzerland). The livers were then harvested and gently dissociated on ice in preparation buffer consisting of perfusion buffer supplemented with fetal bovine serum (final concentration: 10%) to generate single‑cell suspensions. This cell suspension was separated by centrifugation for 4 min at 50 × g, with the precipitate being used for hepatocyte isolation and the supernatant for Kupffer cell and liver sinusoidal endothelial cell (LSEC) isolation. The precipitate underwent two additional rounds of centrifugation, each for 4 min at 50 × g , to isolate hepatocytes. The supernatant was processed through the MACS cell separation system (Miltenyi, Bergisch Gladbach, Germany) to isolate Kupffer cells using Anti-F4/80 MicroBeads UltraPure and LSECs using CD146 MicroBeads UltraPure. The isolated cells were then used for the luciferase assay, performed as described above. Generation of recombinant anti-PcrV scFv antibody CHO-K1 cells (JCRB9018; National Institutes of Biomedical Innovation, Health and Nutrition, Ibaraki, Japan) were cultured in CH100 serum-free medium (Cat. No. CH100-0005; GMEP Cell Technologies, Kurume, Japan) supplemented with 4 mM L-glutamine. Transient transfection was performed using a mammalian expression vector, pcDNA3.1-scFv-m166, encoding a recombinant scFv-m166 antibody, and the transfection reagent TransIT-CHO (Cat. No. MIR2170; Takara Bio). After 3 days of continued culture, the culture supernatant was collected. The supernatant was buffer-exchanged to 20 mM phosphate buffer (pH 7.0) and concentrated approximately 100-fold using a centrifugal concentrator (Cat. No. 88527, Pierce Protein Concentrators PES, 10 K MWCO; Thermo Fisher Scientific). The concentrated sample was applied to a mini-column packed with Ni Sepharose™ Excel resin (Cat. no. 17371201; Cytiva, Uppsala, Sweden) for immobilized metal affinity chromatography. After washing with wash buffer, bound proteins were eluted using 500 mM imidazole-containing elution buffer, and the eluate was dialyzed against PBS using a cup dialyzer (Slide-A-Lyzer™ MINI Dialysis Device, 10 K MWCO, Cat. No. d69570; Thermo Fisher Scientific). The eluted recombinant scFv-m166 protein was analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, immunoblotting, and enzyme-linked immunosorbent assay. In vitro evaluation of the binding between scFv-m166 and oligomeric PcrV Recombinant scFv-m166 protein, possessing the same amino acid sequence as that expressed from scFv-m166 mRNA used in this study, was produced in CHO cells by GenScript Japan Co. (TurboCHO™ Express). For PcrV protein preparation, the pcrV coding region from P. aeruginosa strain PA103 was subcloned into pQE2, and N-terminally 6× His-tagged PcrV was expressed in E. coli M15 by isopropyl β-D-1-thiogalactopyranoside induction and purified by immobilized metal affinity chromatography and dialysis against PBS to a final concentration of 1 mg/mL. Because recombinant PcrV exists mainly as a monomer in solution, oligomeric PcrV was generated by lowering the pH to unfold the protein, then restoring physiological pH to refold it, following a previous report 81 . Briefly, PcrV was prepared at 2 mg/mL in the buffer containing 25 mM Tris/HCl and 100 mM NaCl (pH 8.0). The pH was reduced to 2.6 with 1 M HCl, followed by the addition of 1 M Tris-HCl (pH 7.5) to raise the pH to approximately 7.5. For isolation of oligomeric PcrV, preparative size-exclusion chromatography (SEC) was performed using an EXTREMA system (JASCO Co., Tokyo, Japan) equipped with a Superdex 200 Increase column equilibrated with buffer containing 25 mM Tris-HCl and 100 mM NaCl (pH 8.0), at a flow rate of 1.0 mL/min, with detection at 280 nm. Before the unfolding–refolding process, PcrV showed a peak at approximately 14.5 min, whereas after the process, the peak shifted to approximately 13 min, with the 14.5-min peak nearly absent (Fig. S7A ), indicating successful preparation of oligomeric PcrV. For analysis of complex formation between scFv-m166 and oligomeric PcrV, both proteins were dialyzed against the SEC mobile phase (25 mM Tris/HCl, 100 mM NaCl, pH 8.0). The PcrV monomer concentration was fixed at 5 μM, and 0, 1, 2, and 5 μM of scFv-m166 was added. SEC was performed under the same conditions as described above, except that an LC-2000 series (JASCO) was used. Control samples of scFv-m166 alone were analyzed at 0.4 mg/mL. Statistical analysis IBM SPSS statistical software (version 29.0.2.0; IBM Corp., Armonk, NY, USA) was used for statistical analyses, except for multiple comparisons of survival curves, which were performed using RStudio (version 2022.7.1, build 554, R version 4.2.1, RStudio PBC). We determined sample sizes using the Sample Size & Power Analysis online service from Prism 10 (version 10.6.0; GraphPad Software, La Jolla, California, USA) ( https://www.graphpad.com/features/power-analysis ). A P -value of <0.05 was considered statistically significant. Before selecting the statistical methods, we performed the Shapiro–Wilk test to assess the normality of the datasets. As shown in an Excel file (Souce_Data_KinoshitaM_etal.xlsx), at least one group in each figure did not meet the assumption of normality, leading us to choose non-parametric tests. Details of the non-parametric tests are also provided in an Excel file (Souce_Data_KinoshitaM_etal.xlsx). Reporting summary Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article. Supplementary information Supplementary Information PDF (35.5MB, pdf) Reporting summary (25.4MB, pdf) Transparent Peer Review file (9.7MB, pdf) Source data Source Data (11.6MB, xlsx) Acknowledgements This study was supported by Leading Advanced Projects for Medical Innovation (LEAP) (21gm0010008h0001 to S.U., T.S.) and Project for Regenerative Medicine and Cell and Gene Therapies (25bm1123069h0001 to S.U., T.S.) from the Japan Agency for Medical Research and Development (AMED), Open Innovation Platform for Industry-Academia Co-Creation (COI-NEXT) Program (JPMJPF2022 to S.U.) from the Japan Science and Technology Agency (JST), Grant-in-Aid for Challenging Research (Pioneering) (23K17480 to S.U., T.S.), Grant-in-Aid for Scientific Research (A) (21H04962 to S.U.), Grants-in-Aid for Scientific Research (B) (23K24435, 22H03176 to T.S.), and Grant-in-Aid for Scientific Research (C) (24K12205 to M.K.) from the Ministry of Education, Culture, Sports, Science and Technology, Japan (MEXT), Multilayered Stress Diseases, Science Tokyo (JPMXP1323015483 to S.U.), Nanken-Kyoten, Science Tokyo (2024-kokusai 1), and Medical Research Center Initiative for High Depth Omics, Science Tokyo. We thank Erika Mochizuki and Reiko Shiratori (Institute of Science, Tokyo) for their technical assistance. T.S. would like to express sincere gratitude to Dr. Jeanine P. Wiener-Kronish for providing the inspiration to pursue research on P. aeruginosa and for her continued support to date. Finally, we thank Edanz ( https://jp.edanz.com/ac ) for editing a draft of this manuscript. Author contributions Conceptualization: S.U., T.S.; Investigation: Ma.K., Y.M., K.K., N.L., A.K., N.T.M., Mo.K.; Project administration: S.U., T.S.; Funding acquisition: Ma.K., S.U., T.S.; Supervision: S.U., T.S.; Writing—original draft: S.U., T.S.; Writing—review & editing: S.U., T.S. Peer review Peer review information Nature Communications thanks Mariette Barbier and Alexander Simonis for their contribution to the peer review of this work. A peer review file is available. Data availability Source data are provided with this paper as a Source Data file (Souce_Data_KinoshitaM_etal.xlsx). All datasets from this study (Source Data File: Source_Data_KinoshitaM_etal.xlsx) are publicly available on Zenodo at the following link: https://zenodo.org/records/18546210 . All the results shown in this work are provided in this file. The nucleotide sequences of the pcrV gene coding regions from the P. aeruginosa clinical isolates identified in this study have been deposited in the DNA Data Bank of Japan (DDBJ) and are available in the GenBank database under the accession numbers: LC880153 [ https://www.ncbi.nlm.nih.gov/nuccore/LC880153 ], LC880154 , LC880155 , LC880156 , LC880157 , LC880158 , LC880159 , LC880160 , LC880161 , and LC880162 . In addition, the sequencing data are presented in the Source Data File submitted. The clinical isolates used in this study have been registered in the PubMLST database ( https://pubmlst.org ) along with their multilocus sequence typing results. A complete list of these isolates is included in the submitted Source Data File. Clinical data used in this study were obtained from patients at Kyoto Prefectural University of Medicine Hospital under approved protocols. Due to ethical and legal restrictions, these data are not publicly available except for the background data presented in the figures of this manuscript, which exclude any personally identifiable patient information. De-identified clinical data may be made available from the corresponding author upon reasonable request and with permission from the Institutional Review Board of Kyoto Prefectural University of Medicine. All other data supporting the findings of this study are included in the article and its Supplementary Information PDF and Source Data Files. Source data are provided with this paper. Code availability The R code used for sample size determination and post-hoc analysis of log-rank tests in survival curve comparisons is provided in the Source Data File for each figure. Competing interests M.K., S.U., and T.S. filed a patent on passive immunotherapy with anti- Pseudomonas single-chain antibody mRNA in Japan (application number: 2024-187725). S.U. is a founder of Crafton Biotechnology. The remaining authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Contributor Information Teiji Sawa, Email: [email protected]. Satoshi Uchida, Email: [email protected]. Supplementary information The online version contains supplementary material available at 10.1038/s41467-026-71040-8. References 1. Sati, H. et al. The WHO Bacterial Priority Pathogens List 2024: a prioritisation study to guide research, development, and public health strategies against antimicrobial resistance. Lancet Infect. Dis. 25 , 1033–1043 (2025). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 2. Murray, C. J. L. & GBD 2021 Antimicrobial Resistance Collaborators Global burden of bacterial antimicrobial resistance in 2019: a systematic analysis. Lancet 399 , 629–655 (2022). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 3. O’Neill, J. Antimicrobial resistance: tackling a crisis for the health and wealth of nations. Review on Antimicrobial Resistance, https://wellcomecollection.org/works/rdpck35v (2014). 4. Naghavi, M. & GBD 2021 Antimicrobial Resistance Collaborators Global burden of bacterial antimicrobial resistance 1990–2021: a systematic analysis with forecasts to 2050. Lancet 404 , 1199–1226 (2024). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 5. MacNair, C. R., Rutherford, S. T. & Tan, M. W. Alternative therapeutic strategies to treat antibiotic-resistant pathogens. Nat. Rev. Microbiol. 22 , 262–275 (2024). [ DOI ] [ PubMed ] [ Google Scholar ] 6. Driscoll, J. A., Brody, S. L. & Kollef, M. H. The epidemiology, pathogenesis and treatment of Pseudomonas aeruginosa infections. Drugs 67 , 351–368 (2007). [ DOI ] [ PubMed ] [ Google Scholar ] 7. Reynolds, D. & Kollef, M. The epidemiology and pathogenesis and treatment of Pseudomonas aeruginosa infections: an update. Drugs 81 , 2117–2131 (2021). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 8. Sawa, T. et al. In vitro cellular toxicity predicts Pseudomonas aeruginosa virulence in lung infections. Infect. Immun. 66 , 3242–3249 (1998). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 9. Baer, M. et al. An engineered human antibody fab fragment specific for Pseudomonas aeruginosa PcrV antigen has potent antibacterial activity. Infect. Immun. 77 , 1083–1090 (2009). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 10. Sawa, T., Ito, E., Nguyen, V. H. & Haight, M. Anti-PcrV antibody strategies against virulent Pseudomonas aeruginosa . Hum. Vaccin Immunother. 10 , 2843–2852 (2014). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 11. Francois, B. et al. Safety and pharmacokinetics of an anti-PcrV PEGylated monoclonal antibody fragment in mechanically ventilated patients colonized with Pseudomonas aeruginosa : a randomized,double-blind, placebo-controlled trial. Crit. Care Med. 40 , 2320–2326 (2012). [ DOI ] [ PubMed ] [ Google Scholar ] 12. Milla, C. E. et al. Anti-PcrV antibody in cystic fibrosis: a novel approach targeting Pseudomonas aeruginosa airway infection. Pediatr. Pulmonol. 49 , 650–658 (2014). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 13. Ali, S. O. et al. Phase 1 study of MEDI3902, an investigational anti- Pseudomonas aeruginosa PcrV and Psl bispecific human monoclonal antibody, in healthy adults. Clin. Microbiol. Infect. 25 , 629 e621–629 e626 (2019). [ DOI ] [ PubMed ] [ Google Scholar ] 14. Chastre, J. et al. Safety, efficacy, and pharmacokinetics of gremubamab (MEDI3902), an anti- Pseudomonas aeruginosa bispecific human monoclonal antibody, in P. aeruginosa -colonised, mechanically ventilated intensive care unit patients: a randomised controlled trial. Crit. Care 26 , 355 (2022). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 15. Priebe, G. P. & Goldberg, J. B. Vaccines for Pseudomonas aeruginosa : a long and winding road. Expert Rev. Vaccines 13 , 507–519 (2014). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 16. Sawa, T. et al. Active and passive immunization with the Pseudomonas V antigen protects against type III intoxication and lung injury. Nat. Med. 5 , 392–398 (1999). [ DOI ] [ PubMed ] [ Google Scholar ] 17. Kis, Z., Kontoravdi, C., Dey, A. K., Shattock, R. & Shah, N. Rapid development and deployment of high-volume vaccines for pandemic response. J. Adv. Manuf. Process 2 , e10060 (2020). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 18. Schlake, T. et al. mRNA: a novel avenue to antibody therapy? Mol. Ther. 27 , 773–784 (2019). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 19. Gillmore, J. D. et al. CRISPR-Cas9 in vivo gene editing for transthyretin amyloidosis. N. Engl. J. Med. 385 , 493–502 (2021). [ DOI ] [ PubMed ] [ Google Scholar ] 20. Koeberl, D. et al. Interim analyses of a first-in-human phase 1/2 mRNA trial for propionic acidaemia. Nature 628 , 872–877 (2024). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 21. Zak, M. M. & Zangi, L. Clinical development of therapeutic mRNA applications. Mol. Ther. 33 , 2583–2609 (2025). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 22. Pardi, N. et al. Administration of nucleoside-modified mRNA encoding broadly neutralizing antibody protects humanized mice from HIV-1 challenge. Nat. Commun. 8 , 14630 (2017). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 23. Kose, N. et al. A lipid-encapsulated mRNA encoding a potently neutralizing human monoclonal antibody protects against chikungunya infection. Sci. Immunol. 4 , eaaw6647 (2019). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 24. Rybakova, Y. et al. mRNA delivery for therapeutic anti-HER2 antibody expression in vivo. Mol. Ther. 27 , 1415–1423 (2019). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 25. August, A. et al. A phase 1 trial of lipid-encapsulated mRNA encoding a monoclonal antibody with neutralizing activity against Chikungunya virus. Nat. Med. 27 , 2224–2233 (2021). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 26. Deng, Y. Q. et al. Lipid nanoparticle-encapsulated mRNA antibody provides long-term protection against SARS-CoV-2 in mice and hamsters. Cell Res. 32 , 375–382 (2022). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 27. Wu, L. et al. Intravenous delivery of RNA encoding anti-PD-1 human monoclonal antibody for treating intestinal cancer. J. Cancer 13 , 579–588 (2022). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 28. Tkaczyk, C. et al. In vivo mRNA expression of a multi-mechanistic mAb combination protects against Staphylococcus aureus infection. Mol. Ther. 32 , 2505–2518 (2024). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 29. Unverdorben, F. et al. Pharmacokinetic properties of IgG and various Fc fusion proteins in mice. MAbs 8 , 120–128 (2016). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 30. Mueller, C. A. et al. The V-antigen of Yersinia forms a distinct structure at the tip of injectisome needles. Science 310 , 674–676 (2005). [ DOI ] [ PubMed ] [ Google Scholar ] 31. Broz, P. et al. Function and molecular architecture of the Yersinia injectisome tip complex. Mol. Microbiol. 65 , 1311–1320 (2007). [ DOI ] [ PubMed ] [ Google Scholar ] 32. Blocker, A. J. et al. What’s the point of the type III secretion system needle? Proc. Natl. Acad. Sci. USA 105 , 6507–6513 (2008). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 33. Stadler, C. R. et al. Elimination of large tumors in mice by mRNA-encoded bispecific antibodies. Nat. Med. 23 , 815–817 (2017). [ DOI ] [ PubMed ] [ Google Scholar ] 34. Faure, K. et al. Effects of monoclonal anti-PcrV antibody on Pseudomonas aeruginosa -induced acute lung injury in a rat model. J. Immune Based Ther. Vaccines 1 , 2 (2003). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 35. Tsumoto, K. et al. Effect of the order of antibody variable regions on the expression of the single-chain HyHEL10 Fv fragment in E. coli and the thermodynamic analysis of its antigen-binding properties. Biochem. Biophys. Res Commun. 201 , 546–551 (1994). [ DOI ] [ PubMed ] [ Google Scholar ] 36. Finck-Barbancon, V. et al. ExoU expression by Pseudomonas aeruginosa correlates with acute cytotoxicity and epithelial injury. Mol. Microbiol. 25 , 547–557 (1997). [ DOI ] [ PubMed ] [ Google Scholar ] 37. Matute-Bello, G. et al. An official American Thoracic Society workshop report: features and measurements of experimental acute lung injury in animals. Am. J. Respir. Cell Mol. Biol. 44 , 725–738 (2011). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 38. Sawa, T. The molecular mechanism of acute lung injury caused by Pseudomonas aeruginosa : from bacterial pathogenesis to host response. J. Intensive Care 2 , 10 (2014). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 39. Kainuma, A. et al. An outbreak of fluoroquinolone-resistant Pseudomonas aeruginosa ST357 harboring the exoU gene. J. Infect. Chemother. 24 , 615–622 (2018). [ DOI ] [ PubMed ] [ Google Scholar ] 40. Hauser, A. R. The type III secretion system of Pseudomonas aeruginosa : infection by injection. Nat. Rev. Microbiol 7 , 654–665 (2009). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 41. Lynch, S. V. et al. Polymorphisms in the Pseudomonas aeruginosa type III secretion protein, PcrV - implications for anti-PcrV immunotherapy. Micro. Pathog. 48 , 197–204 (2010). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 42. Desveaux, J. M. et al. Neutralizing human monoclonal antibodies that target the PcrV component of the type III secretion system of Pseudomonas aeruginosa act through distinct mechanisms. eLife 14 , RP105195 (2025). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 43. Yokota, T., Milenic, D. E., Whitlow, M. & Schlom, J. Rapid tumor penetration of a single-chain Fv and comparison with other immunoglobulin forms. Cancer Res . 52 , 3402–3408 (1992). [ PubMed ] [ Google Scholar ] 44. Thiel, M. A. et al. Penetration of engineered antibody fragments into the eye. Clin. Exp. Immunol. 128 , 67–74 (2002). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 45. Baylet, A. et al. Transcutaneous penetration of a single-chain variable fragment (scFv) compared to a full-size antibody: potential tool for atopic dermatitis (AD) treatment. Allergy Asthma Clin. Immunol. 17 , 73 (2021). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 46. Sato, H. et al. The mechanism of action of the Pseudomonas aeruginosa -encoded type III cytotoxin, ExoU. EMBO J. 22 , 2959–2969 (2003). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 47. Pankhaniya, R. R. et al. Pseudomonas aeruginosa causes acute lung injury via the catalytic activity of the patatin-like phospholipase domain of ExoU. Crit. Care Med. 32 , 2293–2299 (2004). [ DOI ] [ PubMed ] [ Google Scholar ] 48. Schulert, G. S. et al. Secretion of the toxin ExoU is a marker for highly virulent Pseudomonas aeruginosa isolates obtained from patients with hospital-acquired pneumonia. J. Infect. Dis. 188 , 1695–1706 (2003). [ DOI ] [ PubMed ] [ Google Scholar ] 49. Sawa, T., Shimizu, M., Moriyama, K. & Wiener-Kronish, J. P. Association between Pseudomonas aeruginosa type III secretion, antibiotic resistance, and clinical outcome: a review. Crit. Care 18 , 668 (2014). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 50. Pena, C. et al. Influence of virulence genotype and resistance profile in the mortality of Pseudomonas aeruginosa bloodstream infections. Clin. Infect. Dis. 60 , 539–548 (2015). [ DOI ] [ PubMed ] [ Google Scholar ] 51. Simonis, A. et al. Discovery of highly neutralizing human antibodies targeting Pseudomonas aeruginosa . Cell 186 , 5098–5113 e5019 (2023). [ DOI ] [ PubMed ] [ Google Scholar ] 52. Goure, J. et al. The V antigen of Pseudomonas aeruginosa is required for assembly of the functional PopB/PopD translocation pore in host cell membranes. Infect. Immun. 72 , 4741–4750 (2004). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 53. Kundracik, E., Trichka, J., Diaz Aponte, J., Roistacher, A. & Rietsch, A. PopB-PcrV interactions are essential for pore formation in the Pseudomonas aeruginosa type III secretion system translocon. mBio 13 , e0238122 (2022). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 54. Perche, F. et al. Improved brain expression of anti-amyloid beta scFv by complexation of mRNA including a secretion sequence with PEG-based block catiomer. Curr. Alzheimer Res. 14 , 295–302 (2017). [ DOI ] [ PubMed ] [ Google Scholar ] 55. Thran, M. et al. mRNA mediates passive vaccination against infectious agents, toxins, and tumors. EMBO Mol. Med. 9 , 1434–1447 (2017). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 56. Chen, B. et al. A single dose of anti-HBsAg antibody-encoding mRNA-LNPs suppressed HBsAg expression: a potential cure of chronic Hepatitis B virus infection. mBio 13 , e0161222 (2022). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 57. Basardeh, E. et al. Anti- Acinetobacter baumannii single-chain variable fragments provide therapeutic efficacy in an immunocompromised mouse pneumonia model. BMC Microbiol. 24 , 55 (2024). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 58. Hart, T. K. et al. Preclinical efficacy and safety of mepolizumab (SB-240563), a humanized monoclonal antibody to IL-5, in cynomolgus monkeys. J. Allergy Clin. Immunol. 108 , 250–257 (2001). [ DOI ] [ PubMed ] [ Google Scholar ] 59. Dall’Acqua, W. F., Kiener, P. A. & Wu, H. Properties of human IgG1s engineered for enhanced binding to the neonatal Fc receptor (FcRn). J. Biol. Chem. 281 , 23514–23524 (2006). [ DOI ] [ PubMed ] [ Google Scholar ] 60. Borrok, M. J. et al. Enhancing IgG distribution to lung mucosal tissue improves protective effect of anti-Pseudomonas aeruginosa antibodies. JCI Insight 3 , e97844 (2018). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 61. Uchida, S. et al. In vivo messenger RNA introduction into the central nervous system using polyplex nanomicelle. PLoS One 8 , e56220 (2013). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 62. Pardi, N. et al. Expression kinetics of nucleoside-modified mRNA delivered in lipid nanoparticles to mice by various routes. J. Control Release 217 , 345–351 (2015). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 63. An, D. et al. Systemic messenger RNA therapy as a treatment for methylmalonic acidemia. Cell Rep. 21 , 3548–3558 (2017). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 64. Inagaki, M. et al. Cap analogs with a hydrophobic photocleavable tag enable facile purification of fully capped mRNA with various cap structures. Nat. Commun. 14 , 2657 (2023). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 65. Iwakawa, K. et al. Cubic phase-inducible zwitterionic phospholipids improve the functional delivery of mRNA. Adv. Sci. 12 , e2413016 (2025). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 66. Zhang, C., Fu, X., Liu, Y., Zhao, H. & Wang, G. Burden of infectious diseases and bacterial antimicrobial resistance in China: a systematic analysis for the global burden of disease study 2019. Lancet Reg. Health West Pac. 43 , 100972 (2024). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 67. Zhao, S. et al. Acid-degradable lipid nanoparticles enhance the delivery of mRNA. Nat. Nanotechnol. 19 , 1702–1711 (2024). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 68. Frank, D. W. et al. Generation and characterization of a protective monoclonal antibody to Pseudomonas aeruginosa PcrV. J. Infect. Dis. 186 , 64–73 (2002). [ DOI ] [ PubMed ] [ Google Scholar ] 69. Tockary, T. A. et al. Comb-structured mRNA vaccine tethered with short double-stranded RNA adjuvants maximizes cellular immunity for cancer treatment. Proc. Natl. Acad. Sci. USA 120 , e2214320120 (2023). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 70. Abbasi, S. et al. Carrier-free mRNA vaccine induces robust immunity against SARS-CoV-2 in mice and non-human primates without systemic reactogenicity. Mol. Ther. 32 , 1266–1283 (2024). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 71. Holloway, B. W. Genetic recombination in Pseudomonas aeruginosa . J. Gen. Microbiol. 13 , 572–581 (1955). [ DOI ] [ PubMed ] [ Google Scholar ] 72. Stover, C. K. et al. Complete genome sequence of Pseudomonas aeruginosa PAO1, an opportunistic pathogen. Nature 406 , 959–964 (2000). [ DOI ] [ PubMed ] [ Google Scholar ] 73. Liu, P. V. The roles of various fractions of Pseudomonas aeruginosa in its pathogenesis. 3. Identity of the lethal toxins produced in vitro and in vivo. J. Infect. Dis. 116 , 481–489 (1966). [ DOI ] [ PubMed ] [ Google Scholar ] 74. Jolley, K. A. & Maiden, M. C. BIGSdb: Scalable analysis of bacterial genome variation at the population level. BMC Bioinforma. 11 , 595 (2010). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 75. Abramson, J. et al. Accurate structure prediction of biomolecular interactions with AlphaFold 3. Nature 630 , 493–500 (2024). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 76. Oishi, K. et al. Therapeutic effects of a human antiflagella monoclonal antibody in a neutropenic murine model of Pseudomonas aeruginosa pneumonia. Antimicrob. Agents Chemother. 37 , 164–170 (1993). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 77. Katoh, H. et al. IV immunoglobulin for acute lung injury and bacteremia in Pseudomonas aeruginosa pneumonia. Crit. Care Med. 44 , e12–e24 (2016). [ DOI ] [ PubMed ] [ Google Scholar ] 78. theLabRats. Protocol for assay of myeloperoxidase in frozen Lung tissue in the LabRat.com ., http://www.thelabrat.com/protocols/myeloperoxidaseassay.shtml (2005). 79. Schindelin, J. et al. Fiji: an open-source platform for biological-image analysis. Nat. Methods 9 , 676–682 (2012). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 80. Uchida, S. et al. An injectable spheroid system with genetic modification for cell transplantation therapy. Biomaterials 35 , 2499–2506 (2014). [ DOI ] [ PubMed ] [ Google Scholar ] 81. Gebus, C., Faudry, E., Bohn, Y. S., Elsen, S. & Attree, I. Oligomerization of PcrV and LcrV, protective antigens of Pseudomonas aeruginosa and Yersinia pestis . J. Biol. Chem. 283 , 23940–23949 (2008). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Supplementary Materials Supplementary Information PDF (35.5MB, pdf) Reporting summary (25.4MB, pdf) Transparent Peer Review file (9.7MB, pdf) Source Data (11.6MB, xlsx) Data Availability Statement Source data are provided with this paper as a Source Data file (Souce_Data_KinoshitaM_etal.xlsx). All datasets from this study (Source Data File: Source_Data_KinoshitaM_etal.xlsx) are publicly available on Zenodo at the following link: https://zenodo.org/records/18546210 . All the results shown in this work are provided in this file. The nucleotide sequences of the pcrV gene coding regions from the P. aeruginosa clinical isolates identified in this study have been deposited in the DNA Data Bank of Japan (DDBJ) and are available in the GenBank database under the accession numbers: LC880153 [ https://www.ncbi.nlm.nih.gov/nuccore/LC880153 ], LC880154 , LC880155 , LC880156 , LC880157 , LC880158 , LC880159 , LC880160 , LC880161 , and LC880162 . In addition, the sequencing data are presented in the Source Data File submitted. The clinical isolates used in this study have been registered in the PubMLST database ( https://pubmlst.org ) along with their multilocus sequence typing results. A complete list of these isolates is included in the submitted Source Data File. Clinical data used in this study were obtained from patients at Kyoto Prefectural University of Medicine Hospital under approved protocols. Due to ethical and legal restrictions, these data are not publicly available except for the background data presented in the figures of this manuscript, which exclude any personally identifiable patient information. De-identified clinical data may be made available from the corresponding author upon reasonable request and with permission from the Institutional Review Board of Kyoto Prefectural University of Medicine. All other data supporting the findings of this study are included in the article and its Supplementary Information PDF and Source Data Files. Source data are provided with this paper. The R code used for sample size determination and post-hoc analysis of log-rank tests in survival curve comparisons is provided in the Source Data File for each figure. Articles from Nature Communications are provided here courtesy of Nature Publishing Group ACTIONS View on publisher site PDF (6.0 MB) Cite Collections Permalink PERMALINK Copy RESOURCES Similar articles Cited by other articles Links to NCBI Databases Cite Copy Download .nbib .nbib Format: AMA APA MLA NLM Add to Collections Create a new collection Add to an existing collection Name your collection * Choose a collection Unable to load your collection due to an error Please try again Add Cancel Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top