The systems biology of microbial infection aims to understand the interaction between hosts and pathogens by the integrative combination of experiment and theory. The focus of this symposium is on the interaction between the immune systems of humans or animals and bacterial or fungal pathogens. We welcome contributions that, for example, focus on
The International Symposium on Systems Biology of Microbial Infection is taking place biennially since the year 2011 at the HKI-Center for Systems Biology of Infection in Jena, Germany.
Due to the corona pandemic the SBMI symposium will take place as a digital meeting this year. Please note that all presentations are directed to CET. (https://time.is/CET)
We are looking forward to welcome you at the Digital SBMI 2021!
The ZOOM link for the symposium has been sent to you by Email. If you do not receive a confirmation Email when trying to register or the ZOOM link within 24 hours after registration please contact email@example.com.
The process of breathing is realized in a highly specialized system, the alveolus. The fragile alveolar homeostasis is challenged during life by numerous respiratory pathogens resulting in epithelial and endothelial damage, barrier disruption, and robust inflammatory response. Nevertheless, various host defense strategies preserve this balance by immune cell migration and the expression of specific substances. However, these processes can be strongly influenced by important comorbidity factors, e.g., obesity and aging.
The host-pathogen crosstalk can be investigated by using different model systems. Considerable progress has been made in this area in recent years; organoids, the alveolus-on-a-chip, and humanized mouse models are available to figure out the pathomechanisms behind the infection.
Own data demonstrate that especially bacterial co-infections in the context of viral disease lead to severe barrier damage and that inhibitors of intracellular signaling processes can result in a sufficient inhibition of virus replication. The current COVID-19 pandemic demonstrates clearly that host factors such as senescence can lead to an increased susceptibility, which, due to the broad organ tropism of SARS-COV-2, can also result in damage outside the lung.
Using sophisticated technologies, system biology can improve the understanding of the interplay between pathogens and the host to establish new solutions against our enemies.
During the global COVID-19 pandemic day care centers (DCCs) were closed to avert transmission. However, children are only mildly affected by the disease and closure of DCC imposes negative effects on children’s health. Therefore, (re)opening DCCs while simultaneously providing continuous surveillance testing might be a feasible alternative. The objective of the study is to design an optimal COVID-19 surveillance protocol by state-based modeling and proof its feasibility.
To evaluate various strategies we developed a virtual infection spread model for DCCs as state-based model (SBM), which simulates the infection spread in a DCC after the introduction of one primary case. For each individual, viral load kinetics is modeled and infection transmission is modeled using an aerosol transmission and infection risk calculator. Using the SBM we evaluate an optimal DCC setting with respect to (i) how many children have to participate in regular testing, (ii) how frequent testing has to be performed, (iii) and can DCCs remain open with continuous surveillance.
To investigate feasibility, we conducted a 12-week longitudinal study starting in October 2020. Nine pre-selected DCCs in Wuerzburg (Germany) were assigned to four different surveillance approaches (modules) for the detection of SARS-CoV-2 by RT-PCR. Asymptomatic children and childcare workers (CCWs) were screened by mid-turbinate nasal swabs twice weekly (module 1, one DCC), once weekly (module 2, one DCC) or by home-sampled saliva twice weekly (module 3, two DCCs). In module 4, symptomatic children, CCWs and the respective household contacts of five DCCs were offered tests on demand.
Consent to surveillance (71%) and successful study participation (68%) was highest in non-invasive home-sampled saliva testing. During the 12-week study period, no SARS-CoV-2 infection was detected in asymptomatic individuals. Simulation results of the SBM show that the expected number of an additional infection is less than 1 provided that twice-weekly testing and children participation of over 50% is realized. Furthermore, it supports the importance of testing on Mondays to minimize the risk of outbreaks.
In conclusion, surveillance of SARS-CoV-2 in DCC by continuous non-invasive sampling is feasible and the SBM provides evidence of a low risk of outbreaks by twice weekly testing.
The availability of high-performance computers, the increasing utilization of Machine Learning, and the ability to observe microbiological objects in three dimensions and over time, open up a wide range of applications for image-based computer simulations. These applications reproduce experimental images or synthetic models to predict biological events based on measured microscopy data. We developed a Godot—based simulation framework, where 3-dimensional datasets serve as the basis for accurate simulations of the behaviour of epithelial and endothelial cells, macrophages and viruses. The physiological response of these components simulate phagocytosis and viral infection in the tissue model under various conditions. Smaller-scale simulations are able to run on standard home computers or even laptops, whereas larger projects will naturally require scaled-up resources.
Our goal was to create a versatile program which finds its use in various fields for testing and understanding the details of infection processes that are behind various diseases. In addition, the tool can be used to examine the effects of possible drugs and treatments designed to fight these simulated infections and ailments. By using an open-source engine, the framework is freely accessible, thus ensuring easy acceptance by the community. The software is easy to handle, the script is written inside Godot using a Python-like language, thus providing an easy learning curve for the large community of Python programmers. By easily modifying and rearranging the simulated environment, as well as by being able to utilize real-life cell shapes via 3D surface file imports from e.g. Imaris-based 3D surface reconstructions, our tool provides a versatile solution for a wide range of cell biology—related problems.
Misinformation in social networks are considered to have manifold adverse effects on the individuals using them, e.g., erosion of trust in politicians and traditional newspapers or reduced adherence to pro-environmental behavior and public health measures in a pandemic situation. Acquiring a better understanding of how such misinformation diffuses through a social network and what can be effective countermeasures therefore seems reasonable. Pioneering works on the diffusion of misinformation have revealed that epidemic models constitute a useful framework to characterize the spread of misinformation in social networks. However, relatively little is known about the effects which specific countermeasures social media platforms can implement (e.g., fact-checking or post-deletion) can have on the spread of misinformation. In a recently published work (see https://doi.org/10.1371/journal.pone.0256179), colleagues and I have yielded first evidence about the effectiveness and sufficiency of such countermeasures in containing the spread of misinformation in social networks. We used a simple epidemic SIR (Susceptible-Infected-Removed) model to characterize the spread of a specific conspiracy theory diffusing on Twitter in 2020, and we also created representations of both fact-checking and tweet-deletion in the epidemic model. We found that the spread of the conspiracy theory can be modeled adequately within an epidemiological framework, indicating parallels between misinformation diffusion and the spread of infectious diseases. Furthermore, our subsequent simulations indicate that fact-checking is an effective mechanism in an early stage of misinformation diffusion, while tweet-deletion shows only moderate efficacy but is less time-sensitive. In general, an early response seems critical to gain control over the spread of misinformation through social networks. The further potential and perspectives of epidemic models for characterizing the dynamics and containment of misinformation in social networks will be discussed.
Aspergillus fumigatus is the most important mould pathogen of man. This fungus affects the health of over 20 million people worldwide and is responsible for >350,000 deaths annually. The majority of mortality due to Aspergillus infection is caused by chronic infections that follow tuberculosis, COPD, sarcoidosis or other lung damaging conditions. These infections typically last a period of several years and carry a cumulative year on year mortality of about 8%.
We recently observed that fungi resident in the lung for >5 years had significant changes in their genomes. This adaptation is the end product of an adaptive process that begins with first contact of the fungal spores with the lung epithelium. Specific genome types of A. fumigatus preferentially infect in cases of chronic disease. Phagocytosis and killing of spores are the first steps in compatibility. This is followed by signalling from epithelium to specialised immune cells. Finally, genetic factors in the host play a strong role in receptivity of the host to fungal infection and persistence of the infection.
Fungi that survive this initial interaction, which typically lasts a few hours to days, become resident and begin to establish infections. In CPA these lead to lung cavitation and formation of fungal balls with devastating consequences for the host. Almost all chronic infections become drug resistant – either before or after antifungal treatment as adaptation to the stresses in the lung induces drug resistant phenotypes. Adaptation appears to be driven by large scale genome rearrangements more frequently than by single nucleotide changes.
Understanding features of the A. fumigatus genome that allow survival of early interactions at the lung epithelium and that occur during growth in the lung is critical to our understanding pf pathogenicity and may allow targeted diagnostic approaches to identifying at-risk infections.
Aspergillus fumigatus is an environmental saprobe and opportunistic human fungal pathogen. Despite the more than 300,000 cases of invasive disease globally each year, a comprehensive survey of the genomic diversity present, including the relationship between clinical and environmental isolates, and how this genetic diversity contributes to virulence and antifungal drug-resistance, has been lacking. In this study, we define the pangenome of A. fumigatus using a collection of 300 environmental and clinical genomes from a global distribution, 188 of which were sequenced in this study. We find that of the 10,907 unique orthogroups, 7,563 (69%) are core and found in all isolates, while 3,344 show presence/absence variation, representing 16-22% of each isolate’s genome. Using this large genomic dataset of both environmental and clinical samples, we found an enrichment for clinical isolates in a genetic cluster whose genomes also contain more accessory genes, including more transmembrane transporters, proteins with iron-binding activity, and genes involved in both carbohydrate and amino acid metabolism. Finally, we leverage the power of genome-wide association to identify genomic variation associated with clinical isolates and triazole resistance as well as characterize genetic variation in known virulence factors. This characterization of the genomic diversity of A. fumigatus allows us to move away from a single reference genome that does not necessarily represent the species as a whole and better understand its pathogenic versatility, ultimately leading to better management of these infections.
Filamentous fungi are potent producers of structurally diverse natural products and are a subject of extensive research for the development of novel drugs. Aspergillus fumigatus, a relevant human pathogenic fungus, produces a class of sphingolipid inhibitors, named as sphingofungins. They inhibit the serine palmitoyl transferases (SPTs) that catalyze the first step in the biosynthesis of sphingolipids in eukaryotes and are known for the important structural and signaling role in the cell. These compounds have a potential to be used as therapeutics against human neurodegenerative diseases connected to the deregulation of sphingolipid metabolism. Our localization studies showed that the biosynthesis of sphingofungins is partially compartmentalized in the perinuclear endoplasmatic reticulum (ER) and in independent ER-derived vesicles together with the SPTs. Additionally, we developed high-throughput screening of the fungal hyphae using confocal microscopy for image acquisition and subsequent image processing and analysis in order to confirm the colocalization of these enzymes. The obtained results posed a question what are the mechanisms by which the fungus reduces the poisonous effects of its own product. We identified that the aminotransferase of the cluster plays a dual role – promotes the first crucial step during the biosynthesis of sphingofungin in the cytosol, and in parallel helps in the sphingolipid biosynthesis when localized in the vesicles, thus reducing self-poisoning effects. This work revealed the presence of enzymes in fungi that have different roles based on their cellular localization.
Tuberculosis (TB) is one of the world’s deadliest infectious diseases. Caused by the pathogen Mycobacterium tuberculosis (Mtb), the standard regimen for treating TB consists of treatment with multiple antibiotics for at least six months. There are a number of complicating factors that contribute to the need for this long treatment duration and increase the risk of treatment failure. The structure of granulomas, lesions forming in lungs in response to Mtb infection, create heterogeneous antibiotic distributions that limit antibiotic exposure to Mtb. We can use a systems biology approach pairing experimental data from non-human primates with computational modeling to represent and predict how factors impact antibiotic regimen efficacy and granuloma bacterial sterilization. We utilize an agent-based, computational model that simulates granuloma formation, function and treatment, called GranSim. A goal in improving antibiotic treatment for TB is to find regimens that can shorten the time it takes to sterilize granulomas while minimizing the amount of antibiotic required. We also created a whole host model to study Mtb dynamics within a human host. Overall, we use these models to help better understand TB treatment and strengthen our ability to predict regimens that can improve clinical treatment of TB.
Bacterial clonal growth ensure the efficient colonization of the ecological niches as well as the hosts. To ensure their survival, even a small population of genetically identical bacteria must engage in elaborated social behaviors. This is rendered possible by the reversible expression of alternative phenotypes among isogenic individuals, within the micrometer ranges, a phenomenon termed phenotypic heterogeneity. Pathogens are often acquired from their ecological reservoir. A prime example is the waterborne Gram-negative bacterium and re-emerging pathogen, Legionella pneumophila that causes a severe pneumoniae: the Legionnaires’ disease (LD). L. pneumophila colonizes and persist in virtually all water systems where the bacterium has evolved as facultative intracellular parasite of free living protozoa. True cellular playgrounds, those natural interactions have largely shaped the bacterium behavior and the virulence program that is involved during the infection of lung associated phagocytes. Fostered by an innovative combination of single-cell and Omics technologies, our research unveiled L. phenotypic heterogeneity as a major feature in the eco-pathological life-cycle of this bacterium. Here, I will discuss (i) the community behavior of L. pneumophila, (ii) the mechanisms driving this bacterium cell-to-cell variations and (iii) the contribution of L. pneumophila phenotypic heterogeneity to the infection process and the therapeutic failures.
Striednig B,..., Personnic N. Quorum sensing governs a transmissive Legionella subpopulation at the pathogen vacuole periphery. EMBO Rep. 2021 Sep 6;22(9): e52972. doi: 10.15252/embr.202152972.
Personnic N, et al. Quorum sensing controls persistence, resuscitation, and virulence of Legionella subpopulations in biofilms. ISME J. 2021 Sep 19. doi: 10.1038/s41396-020-00774-0.
Personnic N, et al. Quorum sensing modulates the formation of virulent Legionella persisters within infected cells. Nature Communications. 2019 Nov 18;10(1):5216. doi: 10.1038/s41467-019-13021-8.
Phenotypic plasticity between two morphological phenotypes, yeast and hyphae, is a key mechanism by which Candida albicans can thrive in many microenvironments and cause disease in a host. Understanding the decision points and key driver genes controlling this important transition and how these genes respond to different environmental signals is critical to understanding how C. albicans causes infections. My group built a mathematical model of the C. albicans yeast to hyphal transition, integrating multiple environmental factors and regulatory mechanisms. We validated the model by a systematic comparison to prior experiments. The discrepancies motivated alternative hypotheses that are testable by follow-up experiments. Analysis of this model (by methods initiated by my group) revealed two time-constrained windows of opportunity that must be met for the complete transition from the yeast to hyphal phenotype, as well as control strategies that can robustly prevent this transition. Our collaborators experimentally validated two of these control predictions in C. albicans strains lacking the transcription factor UME6 and the histone deacetylase HDA1, respectively. We expect this model will serve as a strong base from which to develop a systems biology understanding of C. albicans morphogenesis.
Mathematical modeling of the Candida albicans yeast to hyphal transition reveals novel control strategies, DJ Wooten, JGT Zañudo, D Murrugarra, AM Perry, A Dongari-Bagtzoglou, Reinhard Laubenbacher, Clarissa J Nobile, Réka Albert, PLoS computational biology 17 (3), e1008690 (2021)
Non-alcoholic fatty liver disease (NAFLD) belongs to the most common chronic liver diseases and gained an estimated prevalence of around 25-30% in Europe. Its pathogenesis is characterized by excessive accumulation of fat in hepatocytes leading to a non-alcoholic fatty liver (NAFL). With ongoing fat accumulation, inflammatory processes could set in leading to a non-alcoholic steatohepatitis (NASH). This stage also involves fibrosis, which could lead to complete scarring of liver tissue. As the liver is closely connected to the gut via portal vein blood supply, it is thereby exposed to high-fat diet related implications like intestinal microbiome compositional shifts. As significant bacteriome compositional shifts were identified before in patients with advanced NAFLD in comparison to healthy controls, this may indicate a similar compositional shift in the human intestinal mycobiota.
Materials & Methods
To investigate the human intestinal mycobiome diversity in NAFLD we isolated microbial DNA from stool samples of 188 NAFLD patients as well as 27 healthy controls and used it for ITS1 sequencing. The mycobiome data was then correlated to multiple disease parameters.
A wide variety of 224 fungal species were identified in the stool samples, including Saccharomyces cerevisiae, Candida albicans and Debaryomyces hansenii. First results of our ongoing bioinformatical analysis revealed a significant increase of Candida species as well as Debaryomyces hansenii abundance in samples of patients with advanced liver fibrosis in comparison to patients with early liver fibrosis.
Our data showed that the abundance of various fungal species correlates with liver fibrosis in NAFLD patients. The future analysis will focus on the correlation of mycobiome and bacteriome data, which are also available for all samples of this study. This might reveal possible interaction mechanisms between intestinal fungi and bacteria which influence the progress of NAFLD.
The increasing resistance of several pathogenic yeast species against known drugs is alarming and demands not only development of novel antifungals but also a thorough mechanistic understanding of drug action. Polyene macrolides (short polyenes) are naturally occurring amphiphilic antifungals, synthesized in various Streptomyces bacteria. Polyenes, such as amphotericin B or nystatin are often used as last line of defense, since resistance is relatively low, while other polyenes, like natamycin are used primarily as food preservatives. Polyene macrolides are supposed to target ergosterol in the plasma and vacuole membrane, but despite decades of research, their detailed mechanism of action has not been unequivocally determined. Large polyenes, such as amphotericin B or nystatin can form ion channels in lipid bilayers, while natamycin, a smaller polyene, does not and likely directly interferes with membrane protein function. We have developed an ultraviolet (UV)-sensitive imaging platform to detect the weak intrinsic fluorescence of polyenes and thereby directly observe their interaction with model and cell membranes. For nystatin, we observe that micron-sized aggregates form, which can interact with the plasma membrane (PM) of mammalian cells. Both, nystatin and natamycin bind extensively to the PM of baker’s yeast, Saccharomyces cerevisiae, as inferred from a strong UV signal in this membrane. Loading a yeast mutant, unable to synthesize ergosterol with cholesterol gave much less binding of both polyenes to the yeast PM. Sterol import into these yeast cells and consequently polyene binding required the ABC transporters Aus1/Pdr11, suggesting these transporters as novel drug targets. Binding of natamycin to the yeast PM caused vacuolar targeting of the proton pump Pma1 but did not alter the lateral distribution of the eisosome marker Sur7. Molecular dynamics (MD) simulations, electron paramagnetic resonance spectroscopy (EPR) and nuclear magnetic resonance spectroscopy (NMR) show that natamycin interferes preferentially with ergosterol-dependent acyl chain ordering of membrane phospholipids. Together, these results explain the selective fungicidal effect of the polyenes nystatin and natamycin by their differential interaction with yeast membranes containing ergosterol compared to the mammalian cholesterol. Our label-free imaging platform is currently combined with X-ray microscopy of yeast ultrastructure and deep-learning assisted phenotype classification of vacuole function upon polyene treatment. This multimodal imaging approach will provide a useful tool set for characterizing polyene action in biofilms in the future.
Organs are delicate 3-D structures with high degrees of functional organization. These structures are affected on the microscopical but also meso- and macroscopical level by pathological conditions. We have been using light sheet fluorescence microscopy (LSFM) to investigate the fine details of organs in unprecedented detail with a focus on blood vessels systems and infiltrating immune cells. In this talk I will briefly introduce the principal method and then demonstrate on several examples how inflammation massively affects a delicately structured environment.
In host-pathogen interactions, often the host (attacked organism) defends itself by some toxic compound and the parasite, in turn, responds by producing an enzyme that inactivates that compound. In some cases, the host can respond by producing an inhibitor of that enzyme, which can be considered as a counter-counter defence. An example is provided by cephalosporins, β-lactamases and clavulanic acid (an inhibitor of β-lactamases). Here, we tackle the question under which conditions it pays, during evolution, to establish a counter-counter defence rather than to intensify or widen the defence mechanisms. We established a mathematical model describing this phenomenon, based on enzyme kinetics for competitive inhibition. We use an objective function based on the area under the curve. The optimal allocation of defence and counter-counter defence can be calculated in an analytical way. The calculation provides a threshold value for the dissociation constant of the inhibitor. Only in the case of strong binding of the inhibitor, it pays to have a counter-counter defence. This theoretical prediction accounts for the observation that not for all defence mechanisms, a counter-counter defence exists. Our results should be of interest for computing optimal mixtures of β-lactam antibiotics and β-lactamase inhibitors such as sulbactam, as well as for other molecular-ecological interactions and to fight antibiotic resistance in general.
In this interplay, a paradox occurs: If the inactivating enzyme is very efficient, the toxin becomes useless. If that is no longer produced, the enzyme becomes useless, so that production of the toxin becomes useful again. The question arises whether this leads to an oscillatory change in strategies or whether a steady state is attained as a sort of compromise, in which both species produce an optimal amount of defence chemical and enzyme, respectively? We tackle that question by using methods from game theory.
Macrophages are important cells of the innate immune system. They play many roles in host defense which is reflected by their different activated subtypes. Depending on their function and phenotype, macrophages are classified into classically activated macrophages (pro-inflammatory M1 cells), alternatively activated macrophages (anti-inflammatory M2 cells), and non-activated (resting M0 cells). Fast characterization of macrophage phenotypes would be important for studying the contribution of the various subtypes to the various stages of the inflammation process.
In this work, the ability of label-free Raman spectroscopic imaging to discriminate between different macrophage phenotypes was investigated. Macrophages were derived from peripheral blood monocytes of three different human healthy donors. 67 Raman images of the three macrophage phenotypes (M0, M1, and M2) were analyzed. The spectra of the images were processed using chemometric methods of unmixing (N-FINDR), correction (Extended Multiplicative Scatter Correction, EMSC), and discrimination (principal component analysis and linear discriminant analysis, PCA-LDA). The discrimination models were validated using leave-one-donor-out cross-validation. The results show that Raman imaging is able to discriminate between pro- and anti-inflammatory phenotypes with good quality. The detected differences also correspond to the biochemical feature of the various phenotypes.
The research of interactions between the pathogens and their hosts is key for understanding the biology of infection. Commencing on the level of individual molecules, these interactions define the behavior of infectious agents and the outcomes they elicit. Discovery of host-pathogen interactions (HPIs) conventionally involves a stepwise laborious research process. However novel computational approaches including machine learning and deep learning allow to significantly accelerate the discovery process, particularly for rich information sources like microscopy. One example of such approaches includes an algorithm we have devised to detect interactions between intracellular Toxoplasma gondii parasites and the host cell innate immune response molecules with high accuracy from micrographs obtained in high-content fashion. In another example, it was possible to detect intracellular and extracellular poxvirus virions in 3D superresolution micrographs without specific immunohistochemical labelling. This was possible through transfer-learning-enabled deep learning model inference from seemingly irrelevant fluorescence channels, allowing to distinguish minute changes in virus particle signal upon internalization. Finally, bringing temporal dimension as a source of information for deep learning algorithms allows predicting infection outcomes in a population of infected and uninfected host cells employing time-lapse microscopy data. Altogether, these examples suggest a great potential for HPI analysis using novel machine learning.
This talk is a collective overview of the following peer reviewed publications:
1) Fisch, D., Yakimovich, A., Clough, B., Wright, J., Bunyan, M., Howell, M., ... & Frickel, E. (2019). Defining host–pathogen interactions employing an artificial intelligence workflow. Elife, 8, e40560.
2) Andriasyan, V., Yakimovich, A., Petkidis, A., Georgi, F., Witte, R., Puntener, D., & Greber, U. F. (2021). Microscopy deep learning predicts virus infections and reveals mechanics of lytic-infected cells. Iscience, 24(6), 102543.
3) Yakimovich, A., Huttunen, M., Samolej, J., Clough, B., Yoshida, N., Mostowy, S., ... & Mercer, J. (2020). Mimicry Embedding Facilitates Advanced Neural Network Training for Image-Based Pathogen Detection. Msphere, 5(5), e00836-20. Furthermore it will outline the future directions for the group I am about to establish at CASUS HZDR.
Bacterial pathogens can alter host cell mechanics to promote their own dissemination through tissues. Borrelia burgdorferi (Bb), the causative agent of Lyme Disease, is an extracellular bacterial pathogen that can spread to distant tissues by travelling in and through the vasculature, lined by a single continuous endothelial cell monolayer. To examine whether Bb infection induces changes in host endothelial cell biomechanics, we used video-microscopy to monitor endothelial cell responses to prolonged exposure to Bb. We found that initially Bb cells homogeneously spread over the endothelial cell monolayers maintaining their spirochete morphology and increased motility, but over time less motile Bb aggregates formed, some of which were intracellular, viable and persisted overtime. We hypothesized that the different forms that Bb attained over the course of infection (spirochetal versus aggregate form) could impact distinctly host cell biomechanics and response to infection. Consistently, we found that host cell motility, cell-extracellular matrix and cell-cell forces in response to infection were transiently reduced during the early stages of infection but reverted to levels similar to those of cells not exposed to infection at later stages when bacterial aggregates formed. To examine how biochemical signaling might regulate the time-dependent biomechanical responses to infection, we performed RNA sequencing and discovered upregulation of multiple host cell innate immune signaling pathways during early but not late infection where differences were minimal compared to cells originating from uninfected wells. Altogether our findings suggest that the changes in host cell innate immune signaling, and biomechanics are tightly regulated and linked to the different morphological forms that Bb cells attain. Our studies also shed light into the processes that may contribute in rendering Bb infection chronic and might facilitate efforts on better understanding Lyme disease.
Streptococcus pneumoniae is a gram-positive human pathogen that colonizes the upper respiratory tract and acts as the causal agent of otitis media, bronchitis and community-acquired pneumonia.
On the bacterial surface, S. pneumoniae harbours several proteins with diverse functions (transport channels, carbohydrate metabolic enzymes, iron receptors) that protrude to the extracellular space. A subset of these proteins is the so-called Choline-Binding Protein (CBP) group.
To allow a precise subcellular localization, single-cell super-resolution Structured Illumination Microscopy (SIM) coupled with batch processing image analysis was used under various conditions.
The super-resolution 3D image stacks were processed using HuygensPro and Imaris. The 2D images were processed using custom-developed workflows written in JIPipe (www.jipipe.org). The protein-specific signals were reconstructed by fitting uniformly small spheres over the bacterial volume. This analysis indicated that the protein distribution is not homogenous along the bacterial surface, but rather concentrated along an equatorial plane of the bacterium. Average distances to the five nearest neighbours showed a similar, equatorial distribution, and revealed a linear relationship between the fluorescence intensity of the clusters and their average distance to their 5-nearest neighbours. The redistribution of peptide clusters was similarly characterized during cell growth using various strains and growth conditions. End-point experiments of peptide distributions were examined using template matching algorithms to reveal the various stages of development during cell division.
The localization of the analysed CBP was not homogeneous along the cell wall which contrasts with the current knowledge presented in the literature. This finding is suggestive of a protective effect exerted by this CBP on designated positions on the cell wall that might correlate with immune attack. The linear relationship between protein cluster intensity and neighbour distance suggests a dynamic and fine-tuned allocation of proteins to the cell wall to avoid areas with simultaneously high protein density and intensity.
Our work presents a novel view and an optimized analytical tool to provide insights into the dynamic spatial positioning of pneumococcal proteins, thus opening new ways to reassess how these pathogens exploit their protein machineries to evade the immune system.
Shigella and enteroinvasive Escherichia coli, human-restricted pathogens, can enter epithelial cells of the colon, multiplicate within them, and move between adjacent cells using the type 3 secretion system (T3SS) encoded by the pINV virulence plasmid. A large family of T3SS effectors, E3 ubiquitin-ligases encoded by the ipaH genes, plays a key role in the Shigella pathogenicity via modulation of the cellular ubiquitination leading to the degradation of host proteins. Nevertheless, ipaH gene repertoire in the genus and its impact on strain virulence is still unclear. Recently, the presence of T3SS and ipaH genes in Escherichia marmotae, a potential marmot pathogen, was declared. We performed comparative genomic analysis of the ipaH genes in Shigella and enteroinvasive Escherichia from human and non-human hosts. Shigella ipaH genes as well as their regulatory elements are highly conserved but less than half of Shigella genomes held a complete set of ipaH genes; gene losses and duplications are not consistent to species tree and nomenclature. In contrast to Shigella, non-human-host IpaH proteins reveal two different types of NEL
C-terminal domain and a diverse set of subtract-binding domains; only ipaH9.8 gene was found both in human-host and non-human-host Escherichia. These results provide a framework for understanding of host-pathogens interactions and the role of effectors’ composition.
Opportunistic infections, such as invasive aspergillosis, pose life-threatening risks for patients undergoing hematopoietic cell transplantation (HCT). Therefore, we investigate first-line host defense mechanisms to protect from pulmonary A. fumigatus infection. Here we addressed whether cytokines, including M-CSF and IL-34, affect alveolar macrophages (AMs) function after allogeneic HCT to improve protection from invasive aspergillosis.
To this end we employed dynamic confocal and light sheet fluorescence microscopy, flow cytometry and functional in vitro assays in combination with in vivo mouse models of allo-HCT (8 Gy TBI, C57BL/6> Balb/c) and invasive A. fumigatus lung infection.
HCT recipients survived A. fumigatus infection when infected 6 days but not 4 days after HCT. AMs showed the highest frequency, proliferation and phagocytic activity in the lung, when compared to neutrophils and monocytes, suggesting that AMs were responsible for protecting mice from A. fumigatus infection. This observation was confirmed by selective AM depletion, which rendered mice vulnerable to infection. Subsequently, we investigated whether cytokines could boost AMs to achieve earlier protection. In vitro, M-CSF but not IL-34, increased AMs migration speed (0.6, 0.3 µm/min respectively) and versatility (diffusion co-efficient of 0.78, 0.52 µm2/min, respectively). Next, we treated HCT-recipients with M-CSF and subsequently infected them intratracheally with A. fumigatus 4 days after HCT. M-CSF boosted myelopoiesis by 2-fold and, importantly expanded tissue-resident AMs by 1.5-fold. Additionally, M-CSF improved AMs killing capacity and protected 90% of HCT recipients from lethal A. fumigatus infection on day 4 after HCT. Notably, M-CSF treatment did not protect from lethal aspergillosis upon local depletion of AMs in HCT recipients. Thus, M-CSF holds great potential for clinical application by fostering AMs responsiveness to protect from early A. fumigatus infections in HCT recipients.
Alveolar macrophages (AM) constitute the first line of immune cells in the lung and are responsible for clearing all types of microbial invaders, such as the pathogenic fungus Aspergillus fumigatus. If not efficiently cleared, intruding conidia can form hyphae within hours leading to life-threatening infections like invasive aspergillosis. However, the absolute number of AM, i.e. in human and mice, is still controversially discussed. In the human lung, a range of four up to 46 AM per alveolus is covered [1, 2, 3, 4]. Similarly, AM number for the murine lung varies [3, 5].
In previous studies, we developed a hybrid spatio-temporal agent-based-model (hABM) [6, 7], which allowed to simulate virtual infection scenarios of A. fumigatus in a single alveolus. This model considers a realistic to-scale representation of the alveolus, consisting of a ¾ sphere, with alveolar epithelial cells (AEC) of type 1 and 2 as well as pores of Kohn (PoK). In the model, the AEC, on which a conidium is located, secretes chemokines that diffuse on the inner surface of the alveolus. AM are able to sense the chemokine gradient, which directs their migration towards the conidium. Previously, this model was applied to compare infection dynamics in human and mice  and to investigate the role of PoK . This allows to accurately simulate host-pathogen interactions in the human and murine lung, to compare these hosts and to gain quantitative insight.
In this study, we use the hABM to investigate the impact of the number of AM on infection clearance during A. fumigatus lung infections for various chemokine parameters for the human and murine system. As a measure of infection clearance, we compute the infection score, which is the fraction of simulations in which the fungus was not detected before onset of germination, i.e. before six hours. Fitting joined analytical functions to the in silicio results enables us to derive a surrogate model that aggregates quantitative infection probabilities for given AM, chemokine parameters and the number of conidia per alveolus for the human and murine system, respectively. Furthermore, associated AM domains of the human and murine system can be normalized making them comparable such that systematic characteristics can be explained with the aid of its analytical model parameters. This allows to draw conclusions about differences between the two host environments. Depending on these parameters we gain quantitative understanding about the transferability of experimental observations in mice to the human system.
 Wallace et al. (1992) Thorax
 Hume et al. (2020) American Journal of Respiratory and Critical Care Medicine
 Stone et al. (1992) American Journal of Respiratory Cell and Molecular Biology
 Crapo et al. (1982) American Review of Respiratory Disease
 Gibbings et al. (2017) American Journal of Respiratory Cell and Molecular Biology
 Pollmächer et al. (2014) PLOS ONE
 Pollmächer et al. (2015) Frontiers in Microbiology
 Blickensdorf et al. (2019) Frontiers in Immunology
 Blickensdorf et al. (2020) Frontiers in Microbiology
Invasive pulmonary aspergillosis is a great threat to immunocompromised patients as treatment options are limited and only successful upon early diagnosis which leads to high mortality rates. Infectious agents are conidia of the mold Aspergillus fumigatus that enter the lung alveoli but, in immunocompetent humans, are cleared by innate immune cells. In immunocompromised patients, however, conidia can germinate and grow into filamentous bodies (hyphae) leading to tissue destruction and invasion of blood vessels. To date, complex human cell-based models to investigate invasive aspergillosis are rare and often lack the quantification of growth parameters of individual hyphae. Our novel “invasive aspergillosis-on-chip” model is composed of human lung epithelial cells at an air-liquid-interface and medium-perfused human endothelial cells separated by a porous membrane. Models were infected by FITC-labelled A. fumigatus conidia on the epithelial side and fungal growth was detected by confocal microscopy. Three-dimensional (3D) data were analysed using automated and semi-automated systems biology methods. The structure of the models was reconstructed by the fluorescence of Calcofluor White-labelled hyphae, the reflected light image identifying the membrane pores and the FITC fluorescence identifying the conidia. Reconstructed hyphae allowed us to compute the morphometric measures of the individual hyphae (length, number of branches and branching levels) in 3D. Additionally, computational data analysis of the reconstructed membrane pores and hyphae revealed an interesting behaviour of invading hyphal branches under physiologic and diseased conditions as well as under influence of antifungal drugs. The development of this versatile “invasive aspergillosis-on-chip” system is very promising in respect to its potential applications in understanding the pathophysiology in invasive aspergillosis and providing a much-needed tool for animal-free drug screening.
The oral commensal Streptococcus oralis is known to increase the virulence of Candida albicans infections in murine oral candidiasis and epithelial cell models through mechanisms which promote the formation of tissue-damaging biofilms. Many probiotic Lactobacillus species are known to inhibit the growth of Streptococcus species. We used an iterative collaborative process between experimentation and modeling to create and test a three-dimensional agent-based model containing L. paracasei and S. oralis in a biofilm. The model included growth, division, decay, and mechanical interactions of each individual bacterium in the biofilm. Through this process, the mostly unexplored relationship between S. oralis and L. paracasei in biofilm growth has been investigated to reveal the inhibitory nature of
L. paracasei on S. oralis biofilms.
Dysbiosis of the intestinal bacterial microbiota caused by antibiotic treatment can initiate overgrowth of commensal Candida species – a predisposing factor for disseminated candidiasis. The gut colonizing bacterium and probiotic Lactobacillus rhamnosus can antagonize C. albicans, the most frequent cause of disseminated candidiasis. To uncover the molecular mechanisms of bacterial antagonism, we investigated the interplay between
C. albicans, L. rhamnosus, and intestinal epithelial cells (IECs) by integrating transcriptional and metabolic profiling, and reverse genetics. Untargeted metabolomics combined with in silico modelling of L. rhamnosus metabolism not only identified antivirulence and antifungal metabolites, but also suggested that IECs foster bacterial growth by releasing specific metabolites such as citric acid, gamma-glutamylalanine and carnitine, which was confirmed experimentally. Bacterial growth on IECs modifies the metabolic environment, including the removal of carbon sources energetically favoured by C. albicans. Consequently, C. albicans is forced to transcriptionally rewire its metabolism and cell biology, which is associated with the reprogramming of specific genes that play a role in virulence. Deletion mutants of these genes exhibit a reduced damage potential, providing a causal explanation for the antagonistic effects driven by metabolic alterations in our model. Altogether, our research suggests that intestinal colonization with bacteria can antagonize C. albicans by reshaping the metabolic environment, forcing metabolic adaptations and consequently reducing fungal pathogenicity.