Wastewater is the receptacle of chemical molecules and microorganisms that reflect our consumption habits, drug use, exposure to pollutants, and infections. Wastewater-based epidemiology, thanks in particular to the OBEPINE project in France, experienced a remarkable boom during the COVID-19 crisis. Selected and financed as part of the France 2030 plan, the OBEPINE+ project aims to build a French national platform for research and development in wastewater-based epidemiology, as part of the national system for preventing future emerging infectious diseases and high-risk pathogens. A great deal of work is currently being carried out in this area, focusing on the monitoring of human respiratory viruses (influenza A/B, respiratory syncytial virus, SARS-CoV-2), emerging infections (poliovirus, measles, mPox), zoonotic infections (avian influenza, hantavirus, leptospirosis) and, more recently, sexually transmitted infections and, in particular, human papillomavirus infections.

The dynamic and rapid evolving landscape of viral pathogens, driven by high mutation rates, adaptive evolution and other factors, highlights the indispensable role of vaccine in preventing the spread of virus infections. As new viral pathogens continuously emerge and existing viruses reemerge, the need for innovative, adaptable, and robust vaccine development platforms has become progressively essential.

To facilitate the discovery of viral vaccines, we have developed a comprehensive platform which can support discovery of various vaccine modalities, such as traditional and mRNA vaccines. This vaccine platform encompasses in vitro and in vivo assay systems for a wide range of viruses, including HBV, RSV, influenza and herpes viruses, among others. Our capabilities include vaccine design, vector adaptation, in vitro assessment, evaluations of immunogenicity, efficacy, and vaccine-induced ADE across animal models. Our novel adjuvants and antigen delivery vectors can be employed to enhance immunogenicity and safety profiles of candidate vaccines. Furthermore, we provide comprehensive supports for IND-enabling studies and bioanalysis for vaccine clinical trials.

A compelling example of the vaccine development platform is the RSV assay system. Candidate vaccine expression is evaluated in cell-based assays using FACS and Western blot. Pre- and post-F specific antibodies are examined in animal models through ELISA. Immunogenicity is assessed in animals by measuring the humoral (IgG, IgA, PCA, DCA) and cellular immune responses (ELISpot, ICS). Importantly, vaccine efficacy is evaluated through protection studies in vaccinated animals by challenging with various viral strains, with outcomes determined by monitoring clinical symptoms and quantifying viral load in lungs.

The potential of RNA-based vaccines to prevent and treat infectious diseases has been demonstrated to a remarkable extent. However, the clinical impact of these agents is often constrained by delivery technologies that compromise cell integrity, limit bioavailability, or trigger unwanted immune responses. The FlashRNA^® platform, offered by RNAlead, represents a novel generation of hybrid nanoparticles that have been engineered to address these limitations. In contrast to conventional lipid nanoparticles (LNPs), FlashRNA^® has been demonstrated to achieve efficient and safe delivery of multiple RNA molecules while maintaining high RNA stability and low immunogenicity. Its unique composition enables efficient cellular uptake and cytoplasmic release, thereby enhancing protein expression. Preclinical data demonstrate the versatility and safety of this approach across several models relevant to vaccine development. By addressing key challenges in RNA encapsulation, biodistribution, and immune modulation, FlashRNA® provides a robust and innovative alternative to LNP-based systems, paving the way for next-generation RNA vaccines and immunotherapies.

The large-scale production of viral vectors, whether for vaccines or gene therapies, faces significant challenges in terms of quality control. Conventional characterization methods are often time-consuming, costly, and require the combination of multiple techniques to access critical information such as viral particle concentration, genome encapsidation rate, and presence of defective or empty capsids. These constraints limit rapid decision-making during production and increase the risk of discarding usable batches due to incomplete or insufficiently sensitive analyses.

To address these limitations, we are developing ViroID, a nanopore-based technology designed to enable rapid, universal, and quantitative characterization of viral particles directly at the point of production. The project originates from academic research conducted at the Physics Laboratory of ENS Lyon (CNRS). It is currently supported by a maturation program with PULSALYS and is progressing towards the creation of a start-up. Beyond bioproduction, we are also exploring applications in environmental monitoring, in particular the surveillance of wastewater to follow viral circulation at the population level.

ViroID integrates a nanopore system coupled to a highly sensitive optical detection module that enables the real-time detection of individual fluorescently labeled viral particles translocating through nanopores under flow [1]. The device provides access to key parameters in a single assay, including viral particle concentration, loading efficiency, and the presence of unwanted particles. The method is compatible with low sample volumes (<20 µL) and short processing times (<1 h), making it particularly suitable for in-process monitoring.

Because it relies solely on fluorescent labeling, ViroID enables adjustable specificity levels depending on the needs of the application, while providing a rapid and robust quantitative method. For example, by selecting appropriate fluorescent labels, we can target entire viruses through their capsid proteins, viral envelopes, and/or genomes. Unlike molecular techniques such as PCR or ELISA, which assess viral components individually, our method provides information on intact particles, offering a more integrated overview of sample quality. This flexibility makes ViroID compatible with a wide range of viruses used in vaccines such as virus-like particles and viral vectors, including small ones as Adeno-Associated Viruses. Moreover, it offers a significant advantage for concentration quantification in low-concentration samples, as its detection limit is less than 10³ particles/mL, which is 1000 times lower than current whole-virus detection techniques like Nanoparticle Tracking Analysis or interferometric microscopy.

In collaboration with several co-development partners, including the pharmaceutical platform FRIPharm (Lyon) and the Viral Vector Manufacturing Center (Nantes), we have developed a first compact prototype, which has already been tested on bioproduction samples. Efforts are also underway to design an eco-responsible consumable system based on reusable glass microfluidic chips and minimal use of reagents, without the need for cold-chain logistics.

While the technology holds promise for various applications, its primary target is the bioproduction of viral vectors, where it can significantly reduce batch rejection rates, improve process control, and shorten validation timelines.  

We are actively seeking industrial and academic partners to co-develop and validate VIRO-ID in real-world production environments and ensure its alignment with sector-specific needs.

Background: The increasing threat of emerging infectious diseases demands rapid-response vaccine platforms that can be quickly adapted to new pathogens. Current vaccine development timelines of years are inadequate for pandemic preparedness, while traditional approaches often fail to generate robust immune responses against highly pathogenic viruses.

Offer/project description: We developed nanoSFERIC, a breakthrough modular vaccine technology featuring protein scaffolds that self-assemble into oligomeric nanoparticle rings with planar antigen display. The platform enables rapid antigen modification and swapping within weeks and involves efficient and scalable manufacturing with well-defined product. The rigid ring structure enhances immunogenicity through optimized BCR clustering and CD45 exclusion, delivering superior antibody responses after single-dose administration.

Innovative strength & Applications: nanoSFERIC’s key innovation lies in its planar and high-density antigen presentation, combined with rapid plug-and-play antigen swapping capabilities. Nanoparticle population is homogeneous and stable at 4oC. Preclinical studies demonstrated superior immunogenicity compared to soluble proteins and virus-like particles, with sustained antibody responses and compatibility with both intramuscular and mucosal delivery routes. The technology is applicable both for the development of pandemic vaccines and for the development of personalized cancer vaccines, offering significant cost-effectiveness and global deployment advantages.

Conclusion: nanoSFERIC represents a paradigm shift in vaccine development, providing the flexibility and rapid response capabilities essential for addressing current and future infectious disease challenges.

Innovative delivery technologies are redefining the field of immunotherapy, providing versatile platforms to address urgent needs in infectious diseases, epidemic preparedness, oncology, and chronic infections. While passive immunotherapy with purified monoclonal antibodies has demonstrated clinical value, new RNA- and viral vector–based strategies offer opportunities for more sustained in situ expression and improved therapeutic flexibility.

In this study, we performed a head-to-head comparison of three delivery formats for the same neutralizing antibody against SARS-CoV-2: (i) purified antibody, (ii) lipid nanoparticle–formulated mRNA (LNP-RNA), and (iii) recombinant adeno-associated virus (rAAV). Neutralizing antibody constructs were evaluated in C57BL/6 mice to investigate expression kinetics, persistence, functional activity, and immunological responses.

Purified antibody injections led to rapid serum detection but with a short half-life, consistent with expected pharmacokinetics. Although persistence could be extended through Fc engineering, this format remains intrinsically transient. LNP-RNA delivery at the tested dose (5 µg/mouse) did not result in detectable antibody expression, underscoring the limitations of current formulations optimized for vaccination rather than sustained therapeutic protein expression. In contrast, rAAV-mediated delivery supported robust and long-lasting antibody expression, making it the most effective of the three strategies for sustained in vivo production. Ongoing work is focused on evaluating alternative RNA formats (linear mRNA, self-amplifying RNA, and circular RNA) to assess whether they can achieve intermediate persistence between short-lived purified antibodies and long-term rAAV expression.

Beyond persistence, rAAV-derived antibodies showed enhanced Fc effector functions, including stronger ADCC and ADCP activities. This improvement likely stems from host cell–driven glycosylation of antibodies produced in situ, highlighting an additional functional advantage of vectorized delivery.

Finally, the immunological profile of each format revealed notable differences. Purified antibody injections and rAAV delivery were largely immunologically neutral, with minimal induction of inflammatory responses. By contrast, LNP-RNA administration triggered transient cytokine release and innate immune activation, both systemically and in the lungs. While such an effect is desirable in a vaccination context, it may represent a drawback for passive immunotherapy, particularly in patients where hyperinflammation could be detrimental.

Together, these findings underscore the complementary advantages of each strategy: purified antibodies for immediate neutralization in emergency or intensive care settings and rAAV for durable antibody expression in chronic or prophylactic contexts. A combinatorial therapeutic approach leveraging these complementary modalities may therefore represent a powerful and flexible toolbox for future passive immunotherapy against infectious and non-infectious diseases.

Buisson Marlyse ^{1, 2}, Solis Morgane ³, Pedenko Borys ¹, Dergan-Dylon Leonardo ¹, Grossi Laurence ¹, Bally Isabelle ¹, Amen Axelle ^1,4, Pletan Catherine ⁵, Schoehn Guy ¹, Fafi-Kramer Samira ³, Poignard Pascal ^{1, 2}  

1. Institut de Biologie Structurale (IBS), Univ. Grenoble Alpes, CEA, CNRS, Grenoble, France
2. Laboratoire de Virologie, CHU Grenoble Alpes, Grenoble, France
3. Institut de virologie, Hôpitaux Universitaires, Inserm UMR1109, Université de Strasbourg, Strasbourg, France
4. Laboratoire d’Immunologie, CHU Grenoble Alpes, Grenoble, France
5. SpikImm, Biotech company, Paris, France

Kidney transplantation is the most common solid organ transplant and the gold standard treatment for end-stage renal disease (ESRD). However, reactivation of BK polyomavirus (BKPyV), a ubiquitous childhood infection, is a major complication in immunosuppressed recipients. BK virus–associated nephropathy (BKVAN) occurs in 1–10% of kidney transplant patients and is a leading cause of graft dysfunction and loss. No approved antiviral therapies exist, and current management relies on reducing immunosuppression, a nonspecific and often suboptimal strategy. Monoclonal neutralizing antibodies (mAbs) therefore represent a promising targeted therapeutic option.

Here, we report the isolation of 53 unique human mAbs from a kidney transplant recipient with strong serum neutralizing activity against the three most prevalent BKPyV genotypes. Antibodies were generated by single memory B-cell sorting using pentamers of the VP1 capsid protein, the target of neutralizing antibodies. These mAbs, derived from 30 unrelated B-cell lineages, were characterized for binding and neutralization and showed diverse activity profiles. Notably, six mAbs from one lineage demonstrated broad and ultrapotent neutralization of genotypes I, II, III, and IV in the sub-nanomolar range. Importantly, their neutralization potency was not significantly affected by BC loop mutations commonly found in clinical isolates, including D60N, E73K, A72V/E73Q and E82D.

Structural studies of Fab–VP1 complexes from five different antibodies by cryo-electron microscopy revealed that the neutralizing mAbs recognize quaternary epitopes spanning three VP1 monomers. The binding sites mapped to the variable BC loop and the conserved HI loop, regions critical for receptor interaction. Functional assays confirmed that the most potent mAbs efficiently blocked VP1 binding to receptor-expressing HEK293T cells. Structural comparison of two highly potent somatic variant mAbs showed nearly identical footprints on VP1 but revealed subtle differences in molecular contacts. These findings, supported by mutagenesis and affinity assays, indicate that the two antibodies are not interchangeable.

Five highly potent, broadly neutralizing mAbs were prioritized for preclinical evaluation, and one lead candidate was selected for further development. This antibody, designated SPK004, is being developed by the company SpikImm as a long-acting IgG molecule with an extended half-life, to serve as a prophylactic intervention against BKVAN in kidney transplant recipients.

Background: Identifying pathogen-derived peptides presented by human leukocyte antigens (HLA) is essential for developing targeted immunotherapies. HLA-peptide specificity determines which peptides can trigger T-cell responses and explains inter-individual differences in immune reactivity. However, the extensive polymorphism of HLA molecules and the vast number of potential pathogen-derived peptides make systematic identification challenging. Although computational tools exist to predict HLA class II–peptide binding, their accuracy remains limited. Therefore, novel high-throughput laboratory tools for epitope identification are essential.

Project Description: We applied high-density peptide microarrays to systematically map peptide interactions with HLA class II, focusing on human cytomegalovirus (HCMV) epitopes restricted by HLA-DRB1*1101. The microarrays contained 4,327 overlapping 15-mer peptides covering 24 HCMV proteins. Peptide-HLA binding was detected using biotinylated HLA monomers and fluorescent streptavidin. Selected peptides were further analyzed with FluoroSpot assay to assess their capacity to stimulate T-responses. This approach enables high-throughput identification of HLA class II binders as T-cell epitope candidates.

Innovative Strength & Applications: High-density peptide microarrays provide a scalable, high-throughput platform for mapping peptide–HLA interactions and identifying T-cell epitopes with functional relevance. Applications include selection of peptides effectively presented by HLA class II subtypes, discovery of T-cell epitope candidates for immunotherapy and vaccine design, and assessment of potential T-cell–dependent anti-drug antibody (ADA) responses.

Conclusion: Our study demonstrates that high-density peptide microarrays can systematically identify HLA class II binders and functional T-cell epitopes, complementing computational predictions and accelerating epitope discovery for immunotherapies, vaccine development, and T-cell-dependent anti-drug antibody (ADA) assessment. This approach provides an efficient, scalable tool for immune system analysis and supports the development of targeted interventions against infectious diseases.

Charlotte Primard¹, Mathilde Jarlier¹, Judith Del Campo¹, Séverine Valsesia¹, Florian Gobet², Thomas Courant², Delphine Guyon-Gellin¹, Alexandre Le Vert¹, Florence Nicolas¹

¹Osivax, Lyon, France

²VFI, Lausanne, Suisse

Influenza-A viruses’ evolution and diffusion pose a significant threat to animal and human health. Zoonotic transmission, particularly from avian strains as H5N1, H7N9, or H9N2, remains a major pandemic risk. While neutralizing antibodies targeting the hemagglutinin (HA) surface protein offer effective protection, HA-based vaccines provide limited coverage against antigenically divergent strains. Osivax is developing OVX836, a recombinant vaccine targeting the highly conserved nucleoprotein (NP). In humans, OVX836 elicited strong NP-specific T-cells and antibody responses, with a preliminary efficacy signal of protection of 84%.

Here we evaluated the protection provided by OVX836 alone and with SQ adjuvant (a squalene emulsion with QS21 saponin) against H5N1 and H7N9 in pre-infected mice and ferrets. These models reflect human pre-existing NP immunity from repeated seasonal influenza infections.

Mice pre-infected with Mouse-Adapted H3N2-A/Hong-Kong/1/1968, were immunized twice, 21 days apart with either OVX836 (30µg), OVX836+SQ (15µg) or a negative control. Twenty-one days after, mice were challenged with 10^0.6 TCID₅₀ of H5N1-A/Hong-Kong/156/1997 via intranasal instillation. Clinical signs, body weight and survival were monitored daily for 14 days.

Ferrets who were naturally exposed to Influenza and had confirmed haemagglutination inhibiting antibody response against pH1N1 were immunized twice, 21 days apart with either OVX836 (300µg), OVX836+SQ (150µg) or a negative control. NP-specific immune responses were assessed before and after immunizations, by ELISA for anti-NP IgG, and using IFNγ ELISpot for T-cell responses. Twenty-one days after the last immunization, ferrets were challenged with 10⁵ TCID₅₀ of H7N9-A/Anhui/1/2013 via intratracheal instillation, and viral loads were followed on daily-collected nasal swabs. Seven days post-challenge, ferrets were euthanized, then lung examined through gross-pathology/ histopathology.

Pre-exposure to H3N2-MA virus did not protect mice against H5N1 infection. Only 17% of mock-immunized mice survived after losing more than 20% of their initial weight. In contrast, 50% of mice given OVX836 recovered and survived, while 83% of OVX836+SQ vaccinated mice survived the challenge.

In ferrets, intratracheally delivered virus reached the upper respiratory tract, as shown by rising viral loads in daily-collected nasal swabs from mock-immunized animals. In contrast, no viral RNA was detected by RT-PCR during the first 2 days post-challenge in OVX836-immunized ferrets, and throughout 7 days in OVX836+SQ-immunized animals. At scheduled necropsy, lungs gross-pathology scores showed around 50% affected tissue in controls, 25% in the OVX836 group, and only 12.5% in all ferrets immunized with OVX836+SQ, a significantly lower level.

Finally, OVX836 vaccination boosted NP-specific IgG and T-cell responses. NP-specific IgG titers reached 3.56 Log₁₀ (OVX836) and 3.96 Log₁₀ (OVX836+SQ), with a significant difference. Pre-immunization T-cell responses increased from 72 NP-specific IFNγ SFU/10⁶ PBMC to 1372 (OVX836) and 1525 (OVX836+SQ) on average post-vaccination.

To conclude, OVX836, a NP-based vaccine, showed strong efficacy against H5N1 and H7N9 infections, in animals pre-exposed to a seasonal influenza virus. Adding SQ adjuvant allowed antigen dose-sparing and further enhanced protection, improving mouse survival rates, limiting the spread of virus and reducing lung damage in ferrets. These promising results highlight OVX836’s potential as a broadly protective influenza A vaccine candidate, especially valuable for pandemic preparedness.

The development of effective vaccines and immunotherapeutics against infectious diseases faces critical challenges in translating promising in vitro results to clinical success. Traditional immunological assays rely on artificial conditions that fail to predict real-world immune responses, creating significant gaps in our understanding of therapeutic efficacy and safety profiles in physiological environments.

Vixen Bio offers an advanced biosensing service based on the WHITE FOx platform that provides real-time targeted immunological testing directly in complex biological samples. Our platform uses a fluidics-free approach that avoids clogging risks when analyzing whole blood, serum, plasma, or other biofluids, delivering unprecedented access to physiologically relevant immune interactions for researchers and clinicians.

Our platform delivers comprehensive immunological insights critical for infectious disease therapeutics development:

1) Real-time binding kinetics of antibodies, vaccines, and therapeutics directly in serum/plasma, revealing physiologically relevant affinity differences masked by buffer-based assays;

2) Fast Immune cell effector function monitoring including complement activation (CDC), antibody-dependent cellular cytotoxicity (ADCC), and phagocytosis (ADCP) for Mode of action testing, patient selection or lot release testing;

3) Advanced immunogenicity assessment through kinetic profiling of anti-drug antibodies (ADAs), enabling patient stratification and treatment optimization;

Recent applications demonstrate Vixen Bio’s transformative potential: 1) comparative analysis of therapeutic antibodies revealed significant performance differences in human plasma despite identical buffer-based binding affinities, 2) while COVID-19 patient studies provided detailed kinetic profiles of anti-RBD antibodies that correlated with the maturity of immune response more accurately than traditional titer measurements.

For infectious disease applications, Vixen Bio’s platform enables critical assessments including pathogen-host interaction studies in physiological conditions, real-time vaccine response monitoring, prediction of therapeutic antibody efficacy against viral escape variants, and optimization of combination therapies through simultaneous multi-target binding analysis.

The platform’s ability to work directly with patient samples during clinical trials enables real-time safety monitoring, efficacy assessment, and adaptive patient selection strategies – particularly valuable for rapidly evolving pathogens where traditional development timelines are inadequate.

Vixen Bio’s technology represents a paradigm shift towards physiologically relevant immunological assessment, providing the mechanistic insights from antibody to immune cells necessary to accelerate the development of more effective vaccines and therapeutics against infectious diseases. By bridging the gap between preclinical promise and clinical reality, our service offers pharmaceutical companies, research institutions, and clinicians the tools needed to make more informed decisions throughout the drug development process.

Background: The rise of multidrug-resistant (MDR) bacterial strains has made many conventional antibiotics ineffective, posing a significant global health threat. To combat this crisis, innovative strategies are urgently needed to accelerate the discovery and development of novel antimicrobial therapies. Monoclonal antibodies (mAbs) have emerged as a novel therapeutic approach for combating MDR pathogens, offering both specificity and efficacy in targeting bacterial infections.

Offer/project description: Understanding the binding kinetics of mAbs to bacterial surface antigens is crucial for optimizing their therapeutic potential. Traditional methods often fall short in providing real-time, dynamic insights into these interactions on living bacterial cells. This study introduces an innovative approach that enables the real-time detection of ligand interactions on live bacterial cells, facilitating precise measurement of mAbs binding kinetics. By developing new real-time cell binding assays (RT-CBA), we continuously monitor the association and dissociation rates of mAbs with bacterial antigens in their native environment.

Innovative strength & Applications: In this study, we extended the applications of certain RT-CBAs used in oncology for measuring interactions of antibodies with a specific target on bacterial cells. Such technology requires that cells be firmly attached to a Petri dish, and for that reason, the development of new immobilization methods for bacteria was needed. For that purpose, the gram-positive strains E. coli CJ236 and BL21, as well as the gram-negative strain S. carnosus TM300, were immobilized to plastic Petri dishes using antibody capture. This method provided us with a system where the bacteria remain viable for at least 16 hours, making it possible to evaluate kinetic binding traces of fluorescently labelled antibodies directed against surface-displayed bacterial proteins for up to 10-15 hours.

Conclusion: This methodology not only enhances our understanding of mAb affinity and specificity but also informs the rational design and optimization of antibody-based therapeutics. Furthermore, it contributes to the broader field of antimicrobial research by providing a tool to evaluate and develop novel biologics that aim to overcome the challenges posed by MDR bacterial infections.

Inicure AB, founded in 2024, develops novel small-molecule FPR1 agonists that boost the innate immune system to combat antibiotic-resistant infections. Their lead compounds, which have shown potent antibacterial and anti-inflammatory effects in preclinical models, target chronic infections such as UTIs and infected wounds. Inicure is supported by GU Ventures, SciLifeLab, and the INCATE network, and is currently seeking partners and investors to advance toward clinical development.

Phage therapy has emerged as a promising alternative to existing antimicrobial strategies. However, a wide array of antiphage defense systems protects bacteria against phages, reducing the potential of this treatment option. Better knowledge of the mechanisms of phage-bacteria interactions is essential for the development of new phage-based therapies. RNAs have emerged as key components of antiphage defense systems including CRISPR RNAs within CRISPR-Cas and antitoxin RNAs within type I and type III Toxin-Antitoxin or abortive infection (Abi) systems. Clostridioides difficile (CD) is an important human enteropathogen and the major cause of nosocomial infections associated with antibiotic therapy. The treatment failure rate is high with the current therapeutic arsenal that should be urgently reinforced with new strategies specifically targeting CD. We characterized the RNA-based antiphage defense systems in CD that are frequently associated with prophages providing efficient defense against other invading phages or contributing to these regions stability. We recently identified a new Abi-like system within a prophage of an hypervirulent CD strain at the origin of the RNA-targeting CRISPR-Cas13 systems. This work is shedding new light on the coordination of bacterial defense strategies pertinent for the development of phage and genome editing, epidemiological monitoring and new therapeutic strategies. 

The global rise of multidrug-resistant (MDR) bacterial pathogens poses a major public health challenge and underscores the need for innovative preclinical models to accelerate antimicrobial discovery. Conventional in vitro assays, based on immortalized cell lines or broth cultures, fail to reproduce the complexity of the human respiratory tract, while animal models are costly, time-consuming, and raise ethical concerns. To address these limitations, we developed a physiologically relevant infection model combining human primary airway epithelia cultured at the air–liquid interface (ALI) with bioluminescent MDR bacterial strains.

ALI cultures, generated from basal cells isolated from human lobectomy healthy margins, differentiate into a pseudostratified epithelium that reproduces mucociliary function, epithelial polarity, and tight-junction integrity. The epithelial architecture, differentiation and cell type proportions were characterized by confocal microscopy, histology, and flow cytometry. We applied this system to bacterial pathogens of high clinical relevance. Specifically, we focused on Acinetobacter baumannii (ABA) and Klebsiella pneumoniae (KP), two WHO priority pathogens frequently associated with nosocomial pneumonia and poor therapeutic outcomes. Recently isolated MDR clinical strains were genetically engineered to stably express the lux operon, enabling real-time, non-destructive monitoring of bacterial proliferation through bioluminescence.

Upon apical infection of fully differentiated ALI epithelia with varying bacterial loads (ABA and KP), we observed progressive tissue colonization by bioluminescence and CFU counting (apical compartment) followed by epithelial barrier disruption beginning at 18–21 hours post-infection. This was reflected by a decline in transepithelial electrical resistance (TEER) measurement correlated to the detection of viable colony-forming units (CFU) in the basolateral compartment, confirming bacterial translocation. Longitudinal bioluminescence closely paralleled CFU counts, demonstrating that bioluminescence provides a reliable, non-destructive readout of bacterial growth, reducing reliance on CFU enumeration.

To demonstrate the ability of the model for drug screening, we tested the efficacy of tigecycline against ABA at four concentrations (0.5, 1, 4, and 16× MIC). Antibiotic treatment was applied in the basal compartment, mimicking systemic administration, 15 hours after infection to allow tissue colonization. We observed bacterial stabilization at 0.5 and 1× MIC, while higher doses (4 and 16× MIC) induced a significant decrease in bioluminescence and CFU in both compartments (apical and basal sides), demonstrating the efficacy of the antibiotic treatment. This proof-of-concept confirms the ability of the model to support quantitative drug evaluation.

Beyond bacterial load, the ALI platform supports multiparametric readouts, including barrier integrity, cytokine release, host stress and apoptosis markers, and transcriptomic or proteomic profiling. These complementary analyses provide a holistic view of pathogen behavior and host response, enhancing the translational relevance of the model.

In conclusion, the ALI–bioluminescence infection model offers a robust, dynamic, and predictive in vitro tool for studying respiratory bacterial infections and evaluating antimicrobial candidates. By bridging the gap between classic in vitro assays and animal models, and integrating host-pathogen readouts, this approach has the potential to accelerate discovery of effective therapies against respiratory pathogens.

Acknowledgment: We thank C. Kemmer and V. Trebosc (BioVersys) for their contribution to developing the bioluminescent bacterial strains used in this work.