Listeria monocytogenes (Lmo), a Gram-positive foodborne pathogen, poses a significant threat to public health due to its ability to cause listeriosis, associated with high mortality rates among susceptible populations. Teichoic acids (TAs), including wall-teichoic acid (WTA) and lipoteichoic acid (LTA), are essential components of the cell wall, playing multifaceted roles in bacterial physiology and virulence. Previous studies suggest a potential link between Galactose (Gal)-deficient TAs and the virulence of Lmo, particularly through the modulation of ActA, the actin assembly-inducing protein, function.
Here, we aimed to dissect how Gal-TAs modulate ActA during the cell-to-cell spread process of Lmo. Through construction of in-frame deletion mutants lacking Gal-deficient decoration on WTA, LTA, or both TAs, we assessed their phenotypic traits during infection using immunofluorescence techniques. Furthermore, we purified WTA polymers and recombinant ActA for surface plasmon resonance analysis to explore the molecular interaction between WTA and ActA.
Our findings indicate that only Gal-WTA is responsible for ActA dysfunction, while Gal-LTA does not play a role therein. Additionally, we observed a weak binding affinity between ActA and WTA in the micromolar range, suggesting the involvement of additional host proteins in the ActA-WTA relationship.
Future investigation will focus on evaluating the contribution of various host proteins to the ActA-WTA interaction, providing insights into the structure-function relationship of cell wall-associated macromolecules in Gram-positive pathogens. 

Chemical diversification of sialic acids on surface structures including flagella, capsules or lipopolysaccharides is widespread across bacterial lineages. Despite the plethora of naturally occurring sialic acid derivatives, the genetic basis for derivatization remains unresolved. We discovered a tricistronic module, FlmEFX, that modifies the pseudaminic acid (Pse) moiety present on the six glycosylated flagellins of Caulobacter crescentus, a synchronizable α-proteobacterium assembling a polar flagellum.
The glycan modifications on Caulobacter flagellins from different strains (WT and mutants carrying in frame deletions of flmE, flmF and flmX alone or in combination) were identified by glycopeptide analysis. Mass spectrometry was used to dissect the interaction network of the flagellin glycosylation system. The conservation of the Pse biosynthesis pathway and specifically of the FlmEFX module was investigated by complementing Caulobacter mutant strains with orthologs from different species.
We found that FlmEFX controls the multi-step chemical derivatization of Pse and that balanced FlmEFX expression is important to prevent a block in Pse-dependent flagellin glycosylation, assembly and secretion. Transcription of the flmEFX module is integrated into the Caulobacter developmental regulatory circuit, as it is under control of the cell cycle master regulator CtrA.
Our results indicate that FlmE is a putative methyltransferase promoting a 28 Da modification of the basic Pse unit, whereas FlmX inhibits flagellin export and assembly in the absence of FlmE. Together the FlmEFX unit enhances heterologous bio-conjugation of flagellins with the terminal Pse-derivative in recombinant Escherichia coli producing Pse and the flagellin glycosyltransferase FlmG. FlmEFX co-occur mostly with FlmG-dependent flagellin glycosylation systems, yet orthologs are also encoded in O-antigen gene clusters of pathogens.
Our previously established heterologous flagellin bio-conjugation system in E. coli expressing the Pse biosynthetic pathway from Caulobacter allows the study of multi-step derivatization of Pse. In this work we identified a transferable genetic module, FlmEFX, that terminally modifies Pse. In contrast to the six Pse biosynthesis genes that are widespread, the flmEFX module is less common and primarily present in several clades of the Caulobacterales in the α-proteobacteria, suggesting that it confers diversification of flagellar glycans in these bacteria.

Introduction:
Clostridium perfringens causes enteric diseases in different animal species and in humans. Its pathogenicity relies on the secretion of a large arsenal of virulence factors, many of them belonging to the family of hemolysin-like β-pore forming toxins (β-PFTs). Knowledge about role and action of many of these toxins is still limited. To understand their role in the pathogenesis of bacterial infections we need to determine the molecular and structural basis of their cell, tissue, and species specificity.

Aims:
We aimed to identify the cellular receptor and the pore-structure of Necrotizing Enteritis Toxin F (NetF), a hemolysin-like β-PFT associated with fatal hemorrhagic enteritis in dogs and foals caused by C. perfringens type A strains.

Methods:
Using recombinantly expressed NetF, we performed cell viability assays on 29 different mammalian cell lines. We compared gene expression profiles of susceptible and resistant cell lines. The best receptor candidate was confirmed using CRISPR/Cas9 single gene knockout and ectopic overexpression studies. Mutated and chimeric receptor proteins were expressed in HAP1 cells to determine the receptor specificity of NetF. We used cryo-EM to determine the pore structure of NetF .

Results:
We identified Capillary Morphogenesis Protein 2 (CMG2 or ANTXR2) to be important in NetF-mediated cytotoxicity. CMG2 is also one of the two known receptors for protective antigen (PA) of Bacillus anthracis anthrax toxin. We demonstrated that CMG2 expression on target cells is essential for NetF toxicity and that PA competitively inhibits NetF cytotoxicity. To further investigate the interaction of NetF with its putative membrane protein receptor, we engineered mutant versions of CMG2 lacking parts of its extracellular domain or presenting domains of similar proteins. Using this approach, we showed that the Ig-like domain of CMG2 is crucial for NetF toxicity and that the amino acid region 243-250 of CMG2 is part of NetF binding site. The oligomeric pore-structure of NetF shows cap-, rim-, and stem domains typical for hemolysin-beta-pore forming toxins.

Conclusions:
We identified the cellular receptor and determined the 3D structure of a central virulence factor of C. perfringens strains causing enteric disease in horses and dogs. Furthermore, our results highlight molecular mechanisms and structures that confer receptor-, cell-type and species specificity for clostridial hemolysin-like βPFTs.

MERS-CoV is a zoonotic coronavirus that can lead to severe respiratory illness in humans, but only causes mild respiratory illness in dromedary camels, which are a major reservoir host for MERS-CoV. To identify species-specific host factors that may influence disease outcomes in humans and camelids, airway epithelial cell (AEC) culture models were developed for C. bactrianus and L. glama.
Single-cell transcriptomics was performed on human, camel, and llama AEC cultures infected with MERS-CoV, HCoV-229E, and dcCoV-229E. HCoV-229E belongs to the common cold coronaviruses and causes mild symptoms in humans, while dcCoV-ACN4 is a phylogenetically related virus, which was isolated from dromedary camels. To compare the response to CoV infection among different host species, we established an extensive bioinformatics pipeline in R, that is compatible with camelid species and allowed functional analysis.
Using this workflow, we were able to annotate not only the human airway epithelial cells in our samples but also the main cell types of the camelid species. Additionally, we investigated how host cell tropism and viral entry receptor expression presented under uninfected and virus-infected conditions. Intriguingly, we found that viral entry receptor expression changes upon MERS-CoV and dcCoV-ACN4 infection and that these changes might play a major role in the manifestation of the host cell tropism for both viruses. Furthermore, we assessed the differential gene expression patterns between infected and uninfected cells in each species and airway epithelium cell type. We discovered that several genes involved in the unfolded protein response network (UPR), cilium organization and early anti-viral response were upregulated in MERS-CoV-infected secretory and ciliated cells.
Notably, this work represents the first steps in generating a comprehensive framework to analyze and identify the fundamental characteristics of virus-host interactions in the respiratory epithelium of well and less-studied species with single-cell resolution.

Unraveling bacterial gene function drives progress in various areas, such as food production, pharmacology, and ecology. While omics technologies capture high-dimensional phenotypic data, linking them to genomic data is challenging, leaving 40-60% of bacterial genes undescribed. One approach to do this in high throughput is microbial GWAS (mGWAS), but existing solutions are made for analyzing very few traits only.

To overcome this bottleneck, we present Scoary2, an ultra-fast microbial mGWAS software. It automates post-mGWAS workflows using an interactive web-app which can be connected to external tools such as the comparative genomics software OpenGenomeBrowser, enabling fast and integrative data exploration.

Scoary2 is, to the best of our knowledge, the first software that makes it feasible to study large phenotypic multi-omics datasets using mGWAS. As proof of concept, we used Scoary2 to explore the metabolome of 44 yogurts, each produced by combining a different strain of the species Propionibacterium freudenreichii with starter culture and discovered two previously unknown genes which affect the carnitine metabolism.

EUCAST advises screening for extended spectrum beta-lactamases (ESBL), plasmidic AmpC beta-lactamases, and carbapenemases in Enterobacteriaceae. GPT-4, a multi-modal large language model (LLM) by OpenAI, could potentially aid in this screening process.
Aim: We aimed to validate a customized GPT-agent to identify potential resistance mechanisms.
Methods: We customized a GPT-agent (“EUCAST-GPT-expert”) using EUCAST guidelines, expert rules, and breakpoint table (v13.1). We uploaded disk diffusion images of 225 Gram-negative isolates that were analyzed for the following resistance mechanisms: “none”, “ESBL”, “AmpC”, and “carbapenemase”. We compared routine diagnostic output (reference standard) to (i) EUCAST-GPT-expert, (ii) medical microbiologists, and (iii) non-customized GPT-4, thereby evaluating sensitivities, specificities, negative and positive predictive values for detecting these resistance mechanisms.
Results: Three human readers showed concordance in 814/862 (94.4%) phenotypic categories and used in median eight words (IQR 4-11) for reasoning. Median sensitivity and specificity for ESBL, AmpC, and carbapenemase were 98% and 99.1%, 96.8% and 97.1%, and 95.5% and 98.5%, respectively. Three independent prompting rounds of the EUCAST-GPT-expert showed concordance in 706/862 (81.9%) categories but used in median 158 words (IQR 140-174) for reasoning. Median sensitivity and specificity for ESBL, AmpC, and carbapenemase prediction were 95.4% and 69.23%, 96.9% and 86.3%, and 100% and 98.8%, respectively. E. coli (n=132) showed lower median ESBL sensitivities, but higher specificities compared to Klebsiella pneumoniae and K. oxytoca (n=45) with 86.4% vs 100% and 76.9% vs. 61.9%, respectively. In the non-customized GPT-4, only 169/862 (19.6%) categories could be interpreted. Of these, 137/169 (81.1%) categories agreed with routine diagnostic. The non-customized GPT-4 used in median 85 words (IQR 72-105) for reasoning.
Conclusion: Human experts showed higher concordance (94.4% vs. 81.9%) and shorter argumentations (median 8 vs. 158 words) compared to EUCAST-GPT-expert. Human experts showed comparable median sensitivities and higher specificities compared to EUCAST-GPT-expert. GPT-agents more commonly showed unspecific flagging of ESBL and AmpC, potentially, resulting in additional testing and higher costs. GPT-agents are not IVDR/FDA-approved as diagnostic tools, thus in-depth validation of LLMs is critical and in silico datasets for benchmarking are needed.

Background: Knowledge on molecular mechanisms of Ceftazidime-Avibactam (CZA) resistance in Pseudomonas aeruginosa is scarce. In this multiomics study on laboratory-evolved CZA-resistant P. aeruginsa we unravel CZA resistance mechanisms and provide insights into its cross-resistance to meropenem (MEM).

Materials/methods: From six clinical P. aeruginosa, three representative clones of each isolate were anaylzed for resistance selection to sub-inhibitory concentrations (SICs) of CZA (n=3) or MEM (n=3). After three passages of being subjected to the same antibiotic concentration, minimum inhibitory concentrations (MIC) were determined. The final MIC was determined when at least one clone became resistant (EUCAST) and after a maximum of 54 days of consecutive incubation. Whole genome and transcriptome (Illumina platform) as well as proteome (Orbitrap Fusion Lumos Mass spectrometer) analysis of the parental isolates and their resistant derivatives (n=18) were then performed. In parallel, a transposon insertion mutant library of P. aeruginosa was screened, and CRISPR/Cas9 genome editing was performed to target specific mutations.

Results: Strains exposed to SICs of MEM exhibited a notably high resistance evolution rate, with only 22% (4/18) concurrently displaying cross-resistance to CZA. Conversely, resistance evolution to CZA was significantly slower in strains subjected to CZA. Addition of the efflux pump inhibitor phenyl arginine-β-naphthylamide (PaβN) resulted in a significantly higher CZA-MIC reduction in samples grown in SICs-CZA compared to MEM-MIC reduction in samples grown in SICs-MEM (p=0.0336). Principal components analysis of proteomics and transcriptomics data highlighted sample clustering based on strains. The multiomic approach (genomics, transcriptomics and proteomics) suggests multiple mechanisms contributing in CZA-resistance. CRISPR/Cas9 genome editing identified mutations in dacB directly associated with CZA resistance; ongoing work focuses on genome editing of ampC -which was upregulated in the majority of strains subjected to SICs of CZA - mexR/B, and arnA. .
Conclusions: This ongoing study contributes to a deeper comprehension of the intricate dynamics and molecular mechanisms that regulate CZA resistance in P. aeruginosa.

Phages and bacteria are engaged in a continuous arms race, illustrated by the recent discovery of numerous anti-phage and anti-defense systems. This diversity and prevalence highlight the strong selective pressure of phages on bacteria and vice versa. Yet, in nature, ecosystem functions mostly seem stable. We are largely unaware of how these defense mechanisms are distributed in natural communities and how they adapt. Specifically, do bacteria and their phages persist due to rapid co-evolution (evolutionary response), or is it mainly the selection of preexisting standing diversity (ecological response)? To answer these questions, we analysed mixed cheese starter communities that are continuously subcultured over an entire cheese-making season. In these communities, we have the selection for rapidly growing bacterial strains in milk with possible inflows and changes of microbes and phages. Daily metagenome sequencing and extensive metadata collection over several months at three cheese-making sites revealed the persistence of stable multi-strain bacterial communities. On the contrary, we observed hundreds of active lytic phages, prophages, and satellite phages. While many phages were occasional, others persisted without notable effect on the community function due to the presence of anti-defense systems. While the bacterial strain composition remained stable throughout the season, their genomes continuously acquired novel CRISPR spacers, anti-phage mechanisms and novel putative anti-phage genes. Altogether, these findings suggest that in nature the intricate dance between bacteria and phages is a complex mix of ecological forces (selection of pre-existing diversity) and rapid evolution (de novo) of both the phages and the bacterial hosts.

Aims:
Mapping functional traits to specific organisms is central to the understanding of ecosystem functioning, affecting natural environments, biotechnologically relevant consortia and synthetic communities alike. This heavily relies on the availability of complete, reliable reference genomes. However, despite their importance, complete genomes only account for a minor fraction of available assemblies.

Methods:
To bridge the gap from studying microbiome composition towards the function of key individual isolates, we analyzed the repeat structure and assembly complexity of 37’000 complete RefSeq and 46’500 Genbank genomes, and integrate additional analyses and meta-data to create a rich, publicly available resource for data mining. Nucmer was used to compute repeat pairs as a basis to calculate the overall repeat region content and assembly complexity, two metrics to identify genera with predominantly difficult to assemble genomes. By mining available meta-data, we devise a quality control step that identifies up to 7% of RefSeq assemblies that may contain assembly errors/miss sequence information.

Results:
Complete genomes represent an optimal basis for downstream functional genomics and comparative genomics studies [1,2]. Yet, our analyses revealed that up to 7% of RefSeq assemblies may contain assembly errors/miss sequence information. These mainly comprise cases that used reference-based assembly strategies or assemblies generated from short read datasets. Biologically relevant aspects were also analyzed including 16S rRNA, biosynthetic gene cluster content and their metabolic potential. Tracking the increasing taxonomic spread of complete genomes over time identifies new taxonomic groups, several of which harbor potential for biological applications.
Our meta-analysis of complete prokaryotic genomes allows to identify potential assembly errors in RefSeq genomes, to devise optimal sequencing strategies for difficult to assemble genomes, to improve the resolution of 16S rRNA based amplicon sequencing, to track the increasing taxonomic spread of complete genomes and to link structural and functional traits to distinct taxonomic groups. In silico analysis and comparison of new isolates with this compendium can identify particular interesting strains for various applications in biocontrol, biotechnology and other fields.

The filamentous fungus Rhizopus microsporus engages in endosymbioses with bacteria from the family of Burkholderiaceae. The most prominent example is the obligate interaction with the endosymbiont Mycetohabitans rhizoxinica, which assists the host in defence against predators, facilitates carbon acquisition and is required for its asexual reproduction. Recently, we have introduced bacteria into a naturally endosymbiont-free strain of R. microsporus, which compromised host fitness. Adaptive laboratory evolution increased germination rates and stabilized the artificially induced endosymbiosis. To study the impact of bacteria on host fitness during germination, we used a high-throughput microscopy method with custom-trained U-Net deep learning models. The approach facilitated the segmentation and tracking of individual spores using multipositional time-lapse confocal imaging, allowing us to monitor changes in cell shape and to quantify fungal growth rates by analysing area fluctuations of segmented objects. We also found that the onset of germination can be identified by detection of a decrease in roundness. In addition, our method quantifies fluorescently labeled bacterial endosymbionts and allows for analysis of the relationship between endosymbiont and host growth under different conditions.

Question
Feline infectious peritonitis (FIP) is a fatal feline disease, caused by a feline coronavirus (FCoV), namely feline infectious peritonitis virus (FIPV). FIPV is of great significance in the cat population being a major cause of feline deaths in veterinary practices. As there are neither effective preventive measures nor approved treatment options available, there is an urgent need to identify antiviral drugs against FIPV.

Methods
We produced a baby hamster kidney 21 (BHK) cell line expressing a serotype I FCoV replicon RNA with a green fluorescent protein (GFP) reporter gene (BHK-F-Rep) and used it as an in vitro screening system to test different antiviral compounds.

Results
Two inhibitors of the FCoV main protease (Mpro), namely GC376 and Nirmatrelvir, as well as the nucleoside analog Remdesivir proved to be effective in inhibiting the replicon system. Different combinations of these compounds also proved to be potent inhibitors, having an additive effect when combined. Remdesivir, GC376, and Nirmatrelvir all have a 50% cytotoxic concentration (CC50) more than 200 times higher than their half-maximal inhibitory concentrations (IC50), making them important candidates for future in vivo studies as well as clinically implemented drug candidates. Additionally, results were acquired with a virus infection system, where Felis catus whole fetus 4 (Fcwf-4) cells were infected with a previously described recombinant GFP-expressing FIPV (based on the laboratory-adapted serotype I FIPV strain Black) and treated with the most promising compounds. Results acquired with the replicon system were comparable to the results acquired with the virus infection system, demonstrating that we successfully implemented the FCoV replicon system for antiviral screening.

Conclusions
We expect that this system will greatly facilitate future screens for anti-FIPV compounds and provide a non-infectious system to study and evaluate drug resistance mutations that may emerge in the FIPV genome.

Aim

Antibiotic resistance is an increasing threat for human health. Antimicrobial peptides (AMPs) have been proposed as an alternative to antibiotics against resistant bacteria. AMPs are naturally occurring peptides, part of the first immune defence of a wide variety of organisms. AMPs are assumed to cause low rate of resistance emergence. TAT-RasGAP317-326 is a peptide with both anticancer and antimicrobial activities. We performed in vitro resistance selection on Escherichia coli and could observe development of specific resistance to TAT-RasGAP317-326, without cross-resistance to other antimicrobial agents. We aimed here at determining whether this was also the case for other Gram-negative pathogens such as the important nosocomial pathogen Acinetobacter baumannii.

Methods

We performed in vitro resistance selections using A. baumannii strains and determined whether resistant strains became cross-resistant to other antimicrobial agents. We then used whole-genome sequencing and CRISPR-based targeted mutagenesis to identify mutations causing resistance.

Results

We observed, in some cases the emergence of cross-resistance to polymyxin B in strains that were selected with TAT-RasGAP317-326 only. Whole genome sequencing allowed us to determine that cross-resistant strains gained mutations in the pmrAB operon, mutations that were already described to cause resistance to polymyxin B. We could confirm by targeted mutagenesis that these mutations were indeed causing cross-resistance.


Conclusion

We thus show here that contact of A. baumannii with a peptide that is structurally very different from polymyxin B can induce the emergence of resistance to polymyxin B. This indicates that the clinical use of AMPs should be taken with caution, since unexpected cross-resistance could emerge.

Introduction
Antimicrobial resistance is a serious threat to modern medicine. In Gram-positive bacteria, beta-lactam resistance is mainly driven by the production of low affinity penicillin binding protein (PBPs) belonging to the structural class B1 that includes among others PBP2a from Staphylococcus aureus and PBP5 from Enterococcus faecium conferring respectively resistance to methicillin and ampicillin. We aimed to decipher the biochemical and biophysical proprieties of class B1 PBPs.

Methods
Class B1 PBPs from S. aureus (PBP2a) and E. faecium (PBP5) were cloned in pET24a without their N-terminal membrane anchoring domain, recombinant proteins were expressed in E. coli and purified. Point mutation variants and chimeric proteins were created by overlap extension PCR and purified following the same protocol. Competition experiments using bocillin-FL were performed to determine the beta-lactam IC50. Protein stability was investigated by differential scanning fluorimetry using Sypro Orange.

Results
S. aureus PBP2a showed pronounced unfolding transition initiating at physiological temperatures that led to irreversible precipitation and complete loss of activity. Mutations responsible for ceftaroline resistance did not further exacerbate the instability. In E. faecium, stability of the ampicillin resistant variant PBP5 was dramatically decreased compared with WT (Δtm -13.5°C). Analysis of chimeric recombinant proteins showed that both resistance and stability are driven by the penicillin binding domain. In both PBP2a and PBP5 the loop gating the active site was key to explain the transition.

Conclusion
S. aureus PBP2a has an intrinsic thermal instability over a physiological temperature range. In E. faecium PBP5, acquisition of mutations reducing ampicillin affinity is associated major destabilization. In both species, the loop gating the active site played a key role in both the beta-lactam resistance and the thermal stability profile. Therefore, beta-lactam resistance is associated with a stability cost for class B1 PBPs that could be exploited to restore susceptibility.

Aims

Antimicrobial peptides (AMPs) are promising alternatives to classical antibiotics against multidrug resistant pathogens. TAT-RasGAP317-326 is an AMP with broad range antibacterial activity, but its mode of action is unknown. We recently determined that Escherichia coli can become partially resistant to TAT-RasGAP317-326 when passaged with increasing concentrations of this peptide in vitro via mutations of the two-component system EnvZ/OmpR. We aimed at understanding better how TAT-RasGAP317-326 interacts with E. coli by investigating further mechanisms of resistance to this peptide. To do this, we used an E. coli strain we isolated that is resistant to TAT-RasGAP317-326 but not to other antimicrobial peptides.

Methods

We used whole-genome sequencing to identify mutations involved in resistance. We then performed in silico modelling to predict how the discovered mutations may affect the efficacy of TAT-RasGAP317-326. Finally, we confirmed the role of these mutations using CRISPR-Cas9 based targeted mutagenesis.

Results

We found that the E. coli isolate specifically resistant to TAT-RasGAP317-326 had an additional mutation in bamA, an essential gene encoding an insertase involved in insertion of outer membrane proteins in Gram-negative bacteria. This mutation is predicted to modify the charge in a negatively charged loop (Q495-T505) at the surface of the BamA protein. In silico docking simulations predicted that binding affinity between TAT-RasGAP317-326 and BamA varies depending on the charge of the Q495-T505 loop. Using CRISPR-Cas9 based targeted mutagenesis, we showed that other mutations lowering the negative charge of the Q495-T505 loop decreased sensitivity of E. coli to TAT-RasGAP317-326 exposure. Interestingly, BamA activity was not affected by TAT-RasGAP317-326, indicating that BamA may be a potential specific receptor for the AMP TAT-RasGAP317-326.

Conclusion

Our study provides insight into mechanisms of AMP activity on Gram-negative bacteria. These types of studies are important as they contribute knowledge that is critical for the potential use of AMPs as alternatives to classical antibiotics for the treatment of bacterial pathogens.

Subversion of IMPDH2 oncoprotein activity by Legionella effectors

Legionella pneumophila is a ubiquitous environmental bacterium and opportunistic human pathogen. Inhalation of contaminated aerosols facilitates infection of alveolar macrophages, and can cause Legionnaires’ disease, a severe pneumonia. During infection, L. pneumophila delivers ca. 300 different “effector proteins” to the host cell using the Icm/Dot type IV secretion system. These effectors target various host cellular pathways to favor bacterial survival and growth, and to establish a specialized replication compartment, the “Legionella-containing vacuole” (LCV).
Recently, we identified the oncoprotein inosine 5’-monophosphate dehydrogenase type 2 (IMPDH2) as a novel target of Legionella effectors. IMPDH2 is crucial for cell metabolism and catalyzes the conversion of inosine 5’-monophosphate (IMP) to xanthosine 5’-monophosphate (XMP), the first and rate-limiting step of the de novo guanine nucleotide synthesis pathway. The homotetrameric IMPDH2 enzyme assembles into filaments under certain cellular conditions such as low GTP levels or treatment with inhibitors, affecting the regulation of its activity. We observed that L. pneumophila infection caused disassembly of IMPDH2 filaments induced in host cells with the IMPDH2 inhibitor mycophenolic acid (MPA). Intriguingly, this effect was dependent on a bacterial effector protein, which covalently modified IMPDH2 and caused disassembly of ATP-induced IMPDH2 filaments in vitro. Accordingly, covalent modification by the effector might alter IMPDH2 activity and/or its distribution in the host cell and thereby modulate local GTP levels. To assess the effects of the modification on IMPDH2 activity, we employ activity assays.
Current studies aim to clarify the cellular localization of IMPDH2 and bacterial effector proteins during L. pneumophila infection by analysing infected cells and isolated LCVs by confocal microscopy, western blot, and mass spectrometry. By studying this novel covalent modification of IMPDH2 and its functional consequences, we hope to improve our understanding of the complex regulation of this crucial enzyme, and its implications for infection biology, cell biology, and cancer biology.

Chlamydia trachomatis and Waddlia chondrophila are pathogenic intracellular bacteria, with complex life cycles composed of infectious elementary body (EB) and replicating reticulate body (RB). Nucleoside diphosphate Kinases (Ndks) exhibit intriguing moonlighting activities and multifunctionality, necessitating investigation in Chlamydiae. Ndks play a crucial role Chlamydia metabolism and its interactions with host cells. This study focuses on characterizing the chlamydial ndk and investigating its role in chlamydial development and pathogenicity.
In C. trachomatis, the ndk gene (Ctndk) exists as a single copy, while in W. chondrophila, two copies are present (Wcndk1 and Wcndk2). Notably, WcNdk2 in W. chondrophila is associated with a signal sequence at its N-terminal. WcNdk2 protein localization was observed in both the host cell nucleus and Golgi apparatus. The localization of WcNDK2 in the Golgi apparatus aligns with its secretion outside the host cell and its involvement in purinergic function. Additionally, we employed Azidothymidine (AZT) to inhibit Ndk function in C. trachomatis or W. chondrophila-infected cells. Since only W. chondrophila growth was affected by AZT treatment, we hypothesized that AZT inhibits the DNA-binding ability of WcNdk2. Indeed, Ndks has been reported to bind and regulate c-myc in other organisms.
The role of ndk genes in development and pathogenicity of Chlamydiae could be further investigated by generation of ndk null mutants and uncovering their target genes. By disrupting Ndk function, it becomes conceivable to target chlamydia development, therefore opening new avenues for drug development aimed at combating chlamydia infections.

Aims
Genome annotations frequently miss small protein-coding genes which play vital roles in cellular processes and hold potential for novel antibiotics development. Our study employs proteogenomics to uncover such overlooked genes in six clinical strains of Mycobacterium tuberculosis, one of the most severe infectious disease agents worldwide, comparing two lineages with different levels of virulence.

Methods
We de novo assembled complete genomes from PacBio long-read sequencing data, creating an optimal foundation for functional genomics. By hierarchically integrating reference genome annotations, ab initio gene predictions and a modified six-frame translation that considers alternative start codons, we created comprehensive but minimally redundant integrated proteogenomics search databases (iPtgxDBs). Total cell extracts were analyzed with state of the art parallel accumulation-serial fragmentation (PASEF) mass spectrometry (MS). We prioritized novel small protein candidates based on functional predictions and conservation and validated peptides by parallel reaction monitoring.

Results
Compared to our complete genomes, 1 to 4 percent of all annotated genes were partially or completely absent in pre-existing fragmented Illumina assemblies. Comparative genomics revealed a large core genome between the six strains, containing 91 percent of all annotated genes. PASEF MS allowed us to identify over two thirds of all annotated proteins per strain without requiring sub-cellular fractionation, and in combination with iPtgxDBs we were able to identify a total of 30 novel proteins shorter than 100 amino acids. These included proteins that were shared between all six strains as well as lineage and strain specific proteins, with a notable prevalence of toxin-antitoxin system proteins.

Conclusion
We illustrate the successful application of our proteogenomics framework for prokaryotes, which is available as a public web server (https://iptgxdb.expasy.org), to six clinical reference strains of a major human pathogen and leverage state of the art tandem MS to identify novel small proteins even from unfractionated samples.

Integrative and conjugative elements (ICEs) are widespread autonomous mobile DNA occurring within bacterial chromosomes. ICEs contain the genes necessary for the excision of their own DNA from the chromosome, the consequent conjugative transfer to a new recipient cell and the chromosomal reintegration. The elements can also carry additional genes that are non-essential for their transfer, but that can confer adaptive phenotypes to the host. Examples of known adaptive genes include antibiotic or heavy metal resistance, or the ability to metabolize specific carbon sources. As such they are thought to provide important genetic variation among closely related strains. In this work, we characterized the presence, distribution, and variation of ICEs related to the well-described ICEclc among Pseudomonas aeruginosa clinical isolates within a geographically restrained environment to understand the factors contributing to their evolution. We examined a total of 181 P. aeruginosa genome sequences obtained from patient or hospital environment isolates, most of which were obtained from a single hospital during 20 years of sampling. More than 90% of the isolates carried one or more ICEclc-like elements, with different degrees of conservation to the known ICEclc lifestyle and transfer genes. ICE clones closely matched their host clonal phylogeny, but not exclusively, indicating that both clonal evolution and ICE horizontal transfer are occurring in the hospital environment. Variable gene regions among the clinical P. aeruginosa ICEclc-type elements were notably enriched for heavy metal resistance genes, toxin-anti-toxin systems, potential efflux systems and multidrug resistance proteins, a metalloprotease and for a variety of regulatory systems, but not for specific recognizable antibiotic-resistance cassettes. Clonal persistence suggests adaptive benefits of these functional categories, and micro-patterns of gene gain and loss indicate ongoing ICE evolution within the P. aeruginosa hosts. We next aimed at studying the activation and transfer process of a representative subset of the ICEclc-type elements identified in P. aeruginosa. We found that ICE excision could be induced in P. aeruginosa by ectopically expressing the homologs of the known master regulator of ICEclc activation, BisDC, pointing to a similar regulatory cascade. Some elements were successfully transferred to a P. putida host, in which they conferred increased tolerance to specific heavy metals.

Many copiotrophic bacteria experience a feast-famine lifestyle in heterogeneous environments, where they experience starvation in between infrequent encounters with rich nutrient hotspots. Motility increases the encounters with such hotspots by 100-1000 fold, but is also energy demanding to a starving cell, giving rise to a risk-reward trade-off. Here, we used videomicroscopy and particle tracking to quantify the behavioral response of 26 strains from 18 different species of marine bacteria over two days of carbon starvation. Our results reveal a dichotomy in the motility behavior in response to carbon starvation: there are species that cease motility within hours (`limostatic'), whereas others retain motility for at least two days (`limokinetic'). The biomass during starvation remained constant for species that ceased motility but decreased approximately 10 % per day for the species that remained motile, of which several species accumulated energy storage compounds (PHB and polyphosphate) before starvation. This shows strains need to convert biomass into energy to fuel motility, but doing so extends their search behavior and thus increases the chance of future biomass gain. Using machine-learning classifiers we identified a genetic component associated with this dichotomy, sufficiently robust to predict the response of an additional set of marine strains with >80% accuracy. Overall, these results expand our understanding of foraging strategies in bacteria, revealing a dichotomy among copiotrophs: risk-prone foragers that retain motility during starvation, and risk-averse foragers that rapidly give up motility during starvation.

Microbes live in complex communities, where they continuously engage in a range of interactions with other microbes. One class of interactions is the exchange of metabolites, with microbes performing different reactions that constitute a collective, distributed metabolism. While it is known that microbial communities from many environments exhibit distributed metabolism, the structure and features of these distributed metabolic networks is not well understood. To gain insights into the relationships between microbes in a community, we developed a consumer-resource model describing the evolution of metabolic dependencies and collective metabolism. Based on this model, we predicted conditions under which auxotrophy is most likely to evolve. To test these predictions, we are analysing the amino acid and vitamin auxotrophy profiles of genomes of isolates as well as of metagenome-assembled genomes (MAGs) collected across environments. Through this approach, we aim to describe the distribution of biosynthetic machinery in microbial communities, to understand the factors driving the evolution of distributed metabolism, and to identify any universal features of distributed metabolism that appear across environments.

Antibiotic resistance is a growing concern for global health, demanding the development of innovative and efficient strategies to counteract pathogenic bacteria. Iron is crucial for bacterial proliferation during pathogenesis. In polymicrobial infections and within the host, competition for iron is mediated by siderophores – secondary metabolites that scavenge iron with high affinity. Since siderophores are often species specific, they have two opposing effects: they facilitate iron availability for community members possessing compatible receptors while depriving iron from competitors lacking matching receptors. Here we exploited the iron-depriving properties of siderophores produced by non-pathogenic environmental bacteria as an antibacterial strategy to induce iron starvation and growth arrest in pathogens. In a screen involving 320 environmental Pseudomonas strains, we identified five chemically distinct siderophores (pyoverdines) with broad-spectrum antibacterial activity. Purified pyoverdines completely stalled Acinetobacter baumannii and Staphylococcus aureus growth, while showing intermediate activity against Escherichia coli and Klebsiella pneumoniae. Treating infected insect (Galleria mellonella) larvae with pyoverdine significantly increased host survival in the case of A. baumannii and K. pneumoniae infections. Furthermore, pyoverdines exhibited low toxicity against mammalian cells at effective concentrations. Resistance evolution was low and mutations involved in partial resistance were both species and strain specific. Overall, we established siderophore-induced iron starvation as a new strategy to combat bacterial pathogens, including members of the ESKAPE group that are often difficult to treat and resistant to antibiotics.

Acinetobacter baumannii is currently a major threat to human health. With the spread of multidrug-resistant (MDR) and extensively drug-resistant (XDR) strains, the development of complementary strategies is needed. A promising complimentary and realistic strategy could be phage therapy, which uses bacteriophages (phages), i.e viruses that specifically kill bacterial cells during their life cycle.
We designed a two-phage cocktail highly efficient against an XDR A. baumannii isolate collected from a patient with burn wound infection at CHUV (termed Ab125). A first in vitro screen of our collection of 50 different phages identified only phage vB_AbaM_3098 as capable of lysing Ab125. However, quick (ca. 2 h) selection of phage-resistant clones (termed Ab139) occurred. Comparative genomics between Ab125 and Ab139 revealed a Single Nucleotide Polymorphism (SNP) in an intergenic region, currently under investigation.
Very interestingly, we observed that Ab139 became susceptible to six different phages in the collection, otherwise inactive on Ab125. Phage-resistance was also selected when Ab139 was challenged with either of the six phages, with bacterial regrowth observed between 14 h and 16 h. However, combination of vB_AbaM_3098 and phage vB_AbaM_3014 led to a two-phage cocktail capable of totally inhibiting the growth of Ab125. Treatment with the phage cocktail led to 90% survival after 5 days in the in vivo Galleria Mellonella model of infectious diseases, compared to 0% in the non-treated group.
We show that the combination of a phage that only slightly shifted the in vitro bacterial growth curve with an “inactive phage” led to the formulation of a highly bactericidal phage cocktail against Ab125. The therapeutic potential of the assembled cocktail is currently tested in synergy with antibiotics. This work highlights the complexity sometimes involved in the assembly of potent phage cocktail.

Cattle are the mammalian species with most global biomass associated with a huge impact on Earth. In line with the 3R principles, we developed an ex vivo platform using fresh primary bovine blood cells to characterize and dissect host-pathogen interactions. This procedure employs a novel multiparameter flow cytometry assay (measuring maturation / activation of most immune cell subsets) and a multiplex immunoassay (monitoring chemokine / cytokine secretions). We hypothesize that Bos taurus and Bos indicus genetic background are likely to show different immune responses and susceptibilities towards vector-borne diseases (VBDs) because of their different evolutionary trajectory and origin of domestication. Here we tested our hypothesis using two vector- borne viruses, namely Bluetongue virus (BTV) and Schmallenberg virus (SBV), both expected to trigger increasing numbers of outbreaks in Europe in the future due to climate change. When only considering Bos taurus, we found an ex vivo response towards SBV very moderate compared to BTV; this clearly indicates a fine-tuning of the bovine immune response depending on pathogen. The most striking finding was the genetic-borne differential response towards BTV. Bos taurus exhibited an enhanced activation of monocytes, dendritic cells, CD8 and gamma delta T cells, whereas Bos indicus relied mostly on CD4 T cells. Overall, we provide novel insights in the immune responses of cattle with markedly different genetic background to VBDs. Our platform can be applied to test immune responses of different cattle breeds and pinpoint immune responses that might confer protection and assist breeding programmes and vaccine development.

Background
The skin barrier is vital for protection against environmental threats including insults caused by skin-resident microbes. Dysregulation of this barrier is a hallmark of atopic dermatitis (AD) and ichthyosis, with variable consequences for host immune control of colonizing commensals and opportunistic pathogens. While Malassezia is the most abundant commensal fungus of the skin, little is known about the host control of this fungus in inflammatory skin diseases.
Methods
In this experimental study, MC903-treated mice were colonized with Malassezia spp. to assess the host-fungal interactions in atopic dermatitis. Additional murine models of AD and ichthyosis, including tape stripping, K5-Nrf2 overexpression and flaky tail mice, were employed to confirm and expand the findings. Skin fungal counts were enumerated. High parameter flow cytometry was used to characterize the antifungal response in the AD-like skin. Structural and functional alterations in the skin barrier were determined by histology and transcriptomics of bulk skin. Finally, differential expression of metabolic genes in Malassezia in atopic and control skin was quantified.
Results
Malassezia grows excessively in AD-like skin. Fungal overgrowth could however not be explained by the altered immune status of the atopic skin. Instead, we found that by upregulating key metabolic genes in the altered cutaneous niche, Malassezia acquired enhanced fitness to efficiently colonise the impaired skin barrier.
Conclusions
This study provides evidence that structural and metabolic changes in the dysfunctional epidermal barrier environment provide increased accessibility and an altered lipid profile, to which the lipid-dependent yeast adapts for enhanced nutrient assimilation. Our findings reveal fundamental insights into the implication of the mycobiota in the pathogenesis of common skin barrier disorders.

Multi-locus strain typing (MLST) is a pivotal genomic typing method, utilizing 6-10 housekeeping genes to define strain types (ST) referenced in a specific database[1]. MLST offers high discriminatory power and concise information exchange among hospitals [2]. However, current methods suffer from high cost, workload, and turnaround time (TAT), limiting routine use in outbreaks and local surveillance [3]. To address these limitations, we propose leveraging nanopore technologies, which offer improved TAT and cost-efficiency while maintaining accuracy.
We use adaptive sampling to expedite sequence acquisition, enhancing efficiency [4]. The results will be compared to Whole Genome Sequencing (WGS) sequences acquired under similar conditions. Ultimately, the consensus sequences obtained are evaluated using the PubMLST database [5].
Benchmarking against WGS reveals that while WGS yields quality consensus sequences, adaptive sampling excels in sequencing higher sample volumes. Our findings suggest the scalability of this method across various settings, from clinical isolates to outbreak investigations. In less than two hours, consensus sequences can be obtained, highlighting the rapidity of the approach, in line with previous studies [4].
Future steps involve implementing live monitoring tools for real-time decision-making during sequencing. This innovation promises to revolutionize microbial typing, enabling swift and accurate outbreak management.

Aims: Infective endocarditis (IE) is as a rare yet complex condition, with notable mortality rates and an increasing occurrence in Europe. Diagnosis is challenging, particularly when blood culture results are negative, making it difficult to precisely identify the causative agents. Our objective was to compare the diagnostic efficacy of metagenomic next-generation sequencing (mNGS) with conventional clinical laboratory tests using surgical samples from cardiac valves of 19 suspected IE patients.
Methods: We extracted DNA in duplicate from each resected cardiac valve, using two bacterial DNA enrichment protocols designed to remove host DNA. Metagenomic libraries underwent 2X250 sequencing on an Illumina MiSeq instrument, generating an average of 1.5 million quality-filtered reads per sample. Taxonomic assignments were carried out using read-based approaches.
Results: The consistency between the pathogens identified using mNGS and those from blood cultures was evident. Pathogens identified via valve mNGS but absent in blood cultures are likely genuine causative agents of IE rather than false positives. In three specific cases, the inability of mNGS to identify IE causative agents could be attributed to pathogen clearance, as the valves were resected more than two weeks after blood culture positivity. The prevalence of negative findings in valve cultures underscored the reduced diagnostic efficacy of this method compared to valve mNGS. Additionally, antibiotic-resistance genes were detected in six samples, and complete multi-locus sequence typing (MLST) profiles of suspected IE pathogens were obtained in three samples.
Conclusion: Our results underscore the significance of mNGS as a diagnostic resource for IE, showing its potential to complement blood culture testing.

Bacterial infections often involve more than one pathogen. While it is known that polymicrobial infections can impact disease outcomes, we poorly understand how pathogens interact with each other, what molecular mechanisms are involved and whether these mechanisms are standard or tailored to specific opponents. Here, we explored how Pseudomonas aeruginosa reacts to six other opportunistic human pathogens that often co-occur in polymicrobial infections: Acinetobacter baumannii, Burkholderia cenocepacia, Escherichia coli, Enterococcus faecium, Klebsiella pneumoniae, and Staphylococcus aureus. We first used time-lapse fluorescence microscopy to follow interaction patterns and fitness of species in growing micro-colonies over time on agarose pads. We identified a broad spectrum of species-specific interactions including mutualism and antagonism. Second, we used a library of gene-expression reporters and deletion mutants to investigate the molecular mechanisms that P. aeruginosa deploys in interactions with the six other pathogens. Using a combination of flow cytometry and fluorescence microscopy, we observed both general and tailored responses of P. aeruginosa to specific pathogens but also environmental conditions. For example, a general response involved the upregulation of the production of the siderophore pyoverdine, an important agent for iron competition. We further observed differential and opponent-specific expression changes of various regulators including stress response, quorum sensing, and two-component systems. Overall, our insights improve our understanding of pathogen-pathogen interactions at both the ecological and molecular levels, highlighting that P. aeruginosa can distinguish between different opponents and mount tailored responses. This could help in predicting outcomes in polymicrobial infections.

Viticulture requires intensive application of pesticides. In Switzerland, vineyards consume the highest amount of fungicides per unit of surface area in the country (23 kg/ha), while in France, vineyards represent 3% of cultivated areas and consume 20% of the total pesticides used in the country. Biocontrol solutions have been used for years, however, they have not been able to replace pesticides effectively. We believe that this lack of results comes from the almost systematic use of strains isolated from soils to treat the aerial parts of plants. Since these strains are not fully adapted to the phyllosphere conditions and cannot develop properly, they are unlikely to reach their full potential regarding their protective capacity. In this project, we used phyllosphere bacteria as biocontrol agents to limit the use of pesticides. We investigated the protective effect of bacteria isolated from grapevine leaves against Botrytis cinerea and Plasmopara viticola, two pathogens affecting the aerial parts of the plant. Such strains are expected to provide a higher level of protection in comparison to bacteria from other environments, since they are believed to successfully contribute to their host’s resistance. Our results show that several strains were able to reduce the spore fitness of the two pathogens, to significantly reduce symptoms in planta and to trigger plant immunity through defense genes upregulation and stilbene production. Experimentations on whole plants have also shown that our strains could provide a systemic effect, since when only a part of the plant was inoculated with bacteria, even uninoculated leaves showed a significant reduction in symptoms after infection. We believe that our solution will provide a long-term and environmentally friendly protection against diseases and will help to enhance the sustainability of viticulture.

Aims: Mammaliicoccus species, which are members of Staphylococcaceae, are part of the natural skin microbiota of mammals, but they are able to acquire virulence and antibiotic resistance genes making some of them difficult to treat opportunistic pathogens. Sheep and dromedaries in Algeria were found to harbour methicillin-resistant Mammaliicoccus lentus. The aim of the study was to characterize the genetic element containing the methicillin-resistant genes in M. lentus from both animals using a WGS-based comparative analysis.
Methods: M. lentus strains were isolated from the nasal cavity of sheep and dromedaries using selective media. The species were identified using MALDI-TOF and the presence of mec genes was confirmed using PCR. The mec-positive M. lentus were sequenced using Illumina short-read and MinION long-read technology and hybrid assembled (Unicycler) to generate a fully resolved circular genome and a complete staphylococcal cassette chromosome mec (SCCmec). The mec containing element was identified by comparative analysis with known SCCmec using BLAST. The final visualization of the complete element was illustrated using Clinker v.0.0.28 81.
Results: A total of nineteen methicillin-resistant M. lentus were isolated from dromedaries (n=5) and sheep (n=14). They all harboured both mecA and mecC located two distinct fragments: the mecA containing fragment was situated downstream of the orfX gene and was most closely related to the class A mec complex (IS431-mecA-mecR1-mecI) of SCCmec type VII of S. pseudintermedius strain KM241; the mecC containing fragment was situated downstream of the mecA fragment and contains a class E mec complex (mecI, mecR, mecC, blaZ) of S. aureus LGA251. This hybrid element lacked the recombinase genes ccr gene complex classifying it as a pseudo-SCCmec hybrid element (ΨSCCmecA-mecC). At the 3' end, it contained an arsenical resistance operon (arsABD). Phylogenetic analysis showed that the ΨSCCmecA-mecC was present in genetically related strains indicating transfer of methicillin-resistant M. lentus between dromedaries and sheep.
Conclusion: A novel ΨSCCmecA-mecC element consisting of a hybrid of mecA and mecC fragments was present in M. lentus strains from different animals in Algeria. Absence of ccr genes and presence of ΨSCCmecA-mecC in genetically related strains suggest that a clone of methicillin-resistant M. lentus is disseminating among sheep and dromedaries.

Aims
Antimicrobial resistance (AMR) poses a significant global threat to public health, and livestock farming is recognized as a contributing factor to the emergence and spread of antibiotic-resistant bacteria. This study investigates the impact of antibiotic use in pigs on the microbial communities present in pig manure. Drug-resistant bacteria present in manure may reach the human food chain through the use of manure as fertilizer.

Methods
We collected 24 manure samples from 13 pig farms in Switzerland that differed in the amount of antibiotics used in the preceding months and years. DNA was extracted using the Zymo Quick-DNA HMW MagBead kit. Libraries were prepared with the SQK-LSK114 kit and sequenced on an R10.4 flow cell (Oxford Nanopore Technologies) with Guppy SUP basecalling.

Results
Nanopore long-read metagenomic sequencing revealed a significantly higher abundance of AMR genes in samples originating from farms with high antibiotic use. In all samples, resistance genes associated with antibiotic classes that inhibit protein synthesis were strongly overrepresented relative to other classes. Genes associated with resistance to important last-resort antibiotics in human medicine were rarely detected. In some samples, specific highly prevalent mobile genetic elements contributed substantially to the overall AMR gene abundance.

Conclusion
Our research offers insights into the dynamics of AMR in pig manure and provides guidance for promoting responsible antibiotic use in veterinary practices.