PhD projects
The DTU includes 16 projects across three public research institutions in Luxembourg – the University of Luxembourg (UL), the Luxembourg Institute of Health (LIH), and the Luxembourg Institute of Science and Technology (LIST) – to cover a complementary scope of timely research topics around causal and mechanistic roles of the microbiome in pathogenesis
Axis 1: host-microbiome interaction
PI: Dirk Brenner, LIH
Microbial metabolites significantly influence the function and differentiation of T cells, particularly regulatory T cells (Tregs), which play a crucial role in maintaining immune homeostasis and preventing autoimmune diseases. One example are short-chain fatty acids (SCFAs), such as butyrate, propionate, and acetate, produced by gut microbiota fermentation of dietary fibres. SCFAs have been shown to enhance the differentiation and function of Tregs, thereby promoting anti-inflammatory responses. Additionally, these metabolites can modulate the expression of genes involved in Treg cell development by applying epigenetic changes such as methylation or acetylation, further underscoring their pivotal role in shaping immune response.
This project aims to identify and characterize microbial metabolites impacting Tregs, including the effects of metabolites on Treg functions and responses to immune challenges. We plan to link metabolites to their microbial origins and correlate them with immune-related diseases using database comparisons. In vivo experiments will confirm their therapeutic potential, offering new strategies for regulating the immune system and treating immune-mediated diseases.
PI: Evan Williams, UL
The gut microbiome plays a key role in the metabolic state of all animals and alters their susceptibility to a wide variety of diseases. Over the past decade, researchers have established diagnostic and predictive bacterial taxa that explain, mitigate, and are relevant in the treatment of many diseases. Clinical implementation has also been performed, for instance with the use of faecal microbiota transplants (FMT) from healthy donors to treat irritable bowel disease (IBD). However, much remains unknown: of the tens of thousands of bacterial taxa, which specifically are critical for understanding disease?
Only around 3000 of the more than 10,000 known species in the microbiota have been studied. In our ongoing aging microbiome study, only 10 of the top 30 predictive taxa for obesity have clearly defined taxonomy. Here, we expect to find dozens of highly predictive and diagnostic taxa for metabolic disease states and future outcomes. The timelines of microbiome–disease links, and the associated gene regulatory networks, will provide critical insight for identifying how the microbiome can treat diseases. Furthermore, by associating previously unannotated microbial taxa with metabolic disease, we can identify a timeline of risk attuned to an individual’s own genotype and environment.
PI: Anupam Sengupta, UL
Host-pathogen interactions are hallmarks of inflammatory bowel disease (IBD) including chronic illnesses like Crohn’s disease and ulcerative colitis. Triggered by microbial dysbiosis, autoimmune response and genetic shifts, the high incidence of these diseases highlights missing mechanistic links in our current understanding. For instance, how the breakdown of stable microbe-epithelial mucosal walls progresses in IBD, as well as the cause-effect interrelations between disruption of the epithelial composition and microbial dysbiosis (or vice versa) are currently unknown. Microbial biophysics, together with machine-learning based quantitative imaging could provide key insights into the underlying biophysical dynamics (active transport, micro-structural changes end emergent feedback), thereby paving the way toward an immuno-mechanical framework for host-pathogen interactions, specifically in the context of IBD management.
This collaboration will provide a unique opportunity to link molecular biology and genomics with biophysics and biomechanics in the context of gut health and disease. Specifically, the research contributions from the Sengupta Lab will reveal how microbial physiology/distribution maps to the topology, mechanical and immune properties of the host environment, offering a key step that will enable data-based mechanistic modelling, advancing our understanding of the mechanistic feedbacks which could be harnessed for the next generation of diagnostics and treatment.
PI: Emma Schymanski, UL
In the MICROH DTU, we explored the dark matter of the microbiome with exposomics and developed analytical and cheminformatics approaches to explore a variety of matrices. The patient-derived cerebrospinal fluid (CSF) results in the context of Alzheimer’s disease (AD) identified several small molecules that indicated disease progression better than classical biomarkers, including bile acids. While faeces contained the most information for microbiome questions, these were not coupled with other exposure samples to assess the influence of exposures. Other cohort studies with dust samples coupled with plasma or urine have revealed several chemicals related to penetrance of LRRK2 mutations in Parkinson’s disease (PD), including the presence of relevant microbiota. MICRO-PATH will build on this strong basis, exploring a broader range of samples and matrices to investigate the microbiome-related pathogenesis in PD and establishing a model to verify exposure-microbiome interactions using NeuroHuMiX.
Non-target exposomics is still a challenging field due to the immaturity of the methods and large sample fluctuations observed, interfering with statistical calculations. The precious nature of samples such as CSF makes it difficult to obtain sufficient samples. With this project we wish to build on strong foundations established in the MICROH DTU. We will properly assess the most informative samples in a PD context (so far, only mismatched or incomplete sample pairs have been available) to establish which samples are needed for future cohort studies, and establish a HuMiX-based system to verify hypotheses and promising biomarkers in controlled conditions, reducing the need for patient samples.
PI: Christian Vincenot, Alexander Skupin
The complexity of microbiomes and the resulting heterogeneous data represent a major obstacle for dissecting the principles underlying the development of populations in health and disease. Mechanistic data integration and computational modelling can support the understanding of multispecies dynamics and identify key drivers of linked interactions. Current strategies apply either heuristic approaches or focus on very specific pathways and may thereby miss essential interactions. To address this challenge, we will develop and apply new data fusion approaches coupled to agent-based modelling for predictive simulations of gut microbiome dynamics.
The developed mechanistic data analysis tools will be used to integrate different experimental data sets of the MICRO-PATH DTU (like P4 and P12) and provide candidates for mechanistic interactions. This approach will allow to implement data-driven mechanistic models for hypothesis generation and subsequent experimental validation in the MICRO-PATH consortium.
Axis 2: microbiome’s role in a variety of diseases
PI: Paul Wilmes, UL
The human microbiome has been implicated in the initiation and progression of neurodegenerative diseases including Parkinson’s disease and Alzheimer’s disease. In these disease contexts, microbiome-derived molecules, such as amyloidogenic proteins including the bacterial protein curli, have been implicated in triggering and exacerbating disease processes.
This project aims to achieve:
- Advanced annotations of microbiome-borne amyloidogenic proteins.
- Mechanistic insights into cross-seeding between microbiome-borne amyloidogenic proteins and human proteins implicated in neurodegenerative diseases.
PI: Michael Heneka, UL
Neuroinflammatory responses in Alzheimer’s disease (AD) are complex and involve various cellular and molecular players, including amyloid-β (Aβ) deposition-mediated NOD-like receptor protein (NLRP) inflammasome activation in microglia, inflammasome-dependent formation of ASC specks and subsequent release of inflammatory cytokines. Recent evidence suggests that inflammasome activation is connected to seeding and spreading of Aβ in AD and involved in the pathogenesis of tauopathies, supporting the amyloid-cascade hypothesis in AD.
This project aims to gain:
- Insights into the in vivo amyloidogenic potential of microbiome-derived proteins.
- Molecular insights into microbial protein-induced tau pathology and Aβ-induced spread of tau pathology.
PI: Henry-Michel Cauchie, LIST
The gut resistome, i.e. all antibiotic resistance genes (ARGs) circulating in the gut, determine the microbiome dynamic, i.e. how an individual’s gut microbiome will respond to and recover from an antibiotic therapy. Moreover, the gut resistome is also a source of ARGs conferring antimicrobial resistance to virulent or clinically relevant strains. The horizontal gene transfer (HGT) between bacteria can occur in any environment and particularly when bacterial loads are high, like in the gut. Notably, the gut microbiomes of diseased individuals are enriched in facultative pathogens which encode ARGs and might have an increased infective competence.
A risk assessment of ARG excretion caused by gut dysbiosis will be established by combining data obtained about the frequency of ARG transfer events by transduction in case of gut dysbiosis and the survival of AMR bacteria in surface water as well as their competencies to transfer ARGs into the environment.
PI: Mahesh S. Desai, LIH
The Western world continues to see increasing rates of food allergy. Given the association of the microbiome with food allergy, it may be possible to alter the disease course by changing the microbiome through targeted interventions. Nevertheless, before such interventions could become a reality, it is imperative to generate precise knowledge about how the microbiome affects the disease. We recently showed that a mucin-degrading gut bacterium, Akkermansia muciniphila, breaks down the oral tolerance in fibre-deprived mice that have a reduced colonic mucus barrier attributed to excessive microbial mucin foraging. Moreover, our unpublished preliminary data in a dietary fibre intervention human cohort study (Luxembourgish Fiber Cohort (LUXFICO)) show that a low-fibre diet fuels activity of colonic mucin-degrading bacteria.
Our translational project will shed light on understanding the functional connection of the human gut microbiome to food allergy, which is expected to aid in the implementation of personalized interventions to engineer the microbiome as both a preventative and therapeutic strategy in food allergy.
PI: Annette Kuehn, LIH
Food allergies are chronic adverse immune responses to otherwise harmless food proteins. Highest rates occur in children who can experience even life-threatening symptoms. Increased permeability of the gut barrier, dysbiosis of the gut microbiome and proinflammatory T2 immune responses are hallmarks of food allergy. Food allergy is a showcase disease to study the crosstalk of gut microbiota and the human immune system at early age. Recently, we showed for the first time that T2 responses to gut microbiota correlated with clinical symptoms of food allergy.
Our study will contribute to the growing evidence of the ‘microbial origin’ of food allergies. The main outcome will be on how reaction phenotypes relate to local anti-commensal T2 responses of the gut. Epithelial barrier damage is a central factor of many inflammatory diseases. By identifying molecular, cellular, and immune aspects, our study features as a blueprint for other diseases, where disrupted epithelia, bacterial translocation, and pathogenic human-microbe crosstalks are key.
Axis 3: microbiome applications
PI: Martyna Szpakowska, LIH
The interplay between the gut microbiome and human G protein-coupled receptors (GPCRs) significantly impacts health, influencing metabolism, immunity, and neurotransmission. Despite the critical role of microbial metabolites in modulating GPCR activity, a comprehensive understanding of these interactions is lacking. Advances in omics such as the development of sensitive tools to map the GPCRome affected by microbiome-derived metabolites are critically needed to address knowledge gaps and identify therapeutic opportunities in this emerging field.
This project will establish an advanced support platform for the functional profiling and quantification of active metabolites or factors secreted by the microbiome in complex biological samples such as serum, stool, or bacteria culture media. This study will shed light on the intricate crosstalk between GPCRs and the microbiome. By examining the interaction of microbial metabolites with GPCRs, we hope to uncover new roles in metabolic regulation, immune response, and disease pathogenesis, highlighting potential therapeutic targets for precision medicine.
PI: Patrick May, UL
In the MICRO-PATH P12 project, we will leverage the vast amount of Parkinson’s disease (PD) microbiome data that we and others have collected over the past few years, including available multi-omic datasets from NCER-PD and the ongoing MiBiPa study. We will be able to integrate not only different types of microbiome omics data, but also clinical, molecular (metabolome, RNA-seq, miRNAs), genetic and phenotypic data, which we will use to build multi-omic signatures for stratification of patient subgroups, combining host genetic and molecular data types with different layers of microbial community data for PD risk prediction.
Based on our ongoing efforts to build polygenic risk scores (PRS) and multi-omic signatures based on host genetic and molecular data alone, we will implement a new framework for a microbial risk score (MRS) that integrates the different microbial multi-omic profiles into a summary risk score that can be used to measure and predict disease susceptibility and patient stratification, which will be further integrated with host PRS and molecular and clinical signatures to increase the power of disease risk prediction for PD.
PI: Elisabeth Letellier, UL
Immunotherapy, in particular immune checkpoint inhibitors (ICI), represent one of the major recent advances in oncology. Over the past years, several studies clearly suggest an important role for the microbiome in mediating response to ICI, potentially via the modulation of the immune system. Currently innovative strategies to harness the microbiome to improve the response rate to ICI are being developed. Diet is a known regulator of the gut microbiome; however dietary guidelines are not yet implemented in today’s cancer treatments.
We expect to identify new microbiome-based interventions (pre- and pro-biotics) for cancer patients to enhance ICI efficacy. Altogether this project will unravel new intervention points targeting host-microbiome crosstalk and help establish dietary guidelines for patients undergoing ICI treatment.
PI: Jochen Schneider, UL
Gut dysbiosis is among the significant features of Long COVID Syndrome (LCS), an obscure condition that has been associated with a variety of symptoms causing burden for the individual and the society. Studies reveal alterations in the gut microbiota of COVID-19 patients with reduced microbiota diversity and appearance of specific bacterial taxa associated with LCS. Recovery of the gut microbiome profile post-infection correlates with decreased LCS risk. Dysfunction of the intestinal barrier may thus trigger prolonged immune responses. Additionally, gut dysbiosis may affect neurotransmitter production and metabolism, possibly contributing to neurological and neuropsychiatric symptoms observed in LCS.
We expect to detect changes in gut microbiome composition in response to fasting by comprehensive and systematic analysis. Our study will reveal associations between dietary, metabolic, and immunological changes induced by fasting in association with alterations in gut microbiome populations. These microbial changes are expected to impact neurological function and behaviour, mediated by the gut-brain axis. The latter will be attempted studied in an in vitro approach.
PI: Leslie Ogorzaly, LIST
Faecal microbiota transplantation (FMT), which permanently modifies the recipient’s intestinal microbiota, is today a very promising therapeutic alternative for certain pathologies, such as recurrent Clostridioides difficile infections. Despite FMT’s apparent simplicity, concerns about short and long-term safety, particularly regarding viral risks, may have hampered its widespread adoption. Future efforts should focus on establishing robust risk assessment protocols and addressing these safety concerns to unlock FMT’s full therapeutic potential.
We pioneered a novel aptamer-based method for identifying enteric viruses in clinical samples. Aptamers, small single-stranded nucleic acids (DNA or RNA), possess unique three-dimensional structures and demonstrate exceptional specificity and affinity towards their targets. They offer significant advantages over traditional antibody-based assays and genomic amplification techniques commonly employed in similar diagnostic approaches.
Firstly, we seek to deepen our understanding of the human gut virome and its implications as a potential reservoir for human pathogenic viruses within the framework of FMT. Secondly, we endeavour to pioneer the development of a novel diagnostic tool aimed at enhancing the safety profile of FMT procedures. The new insights, expertise and technologies gleaned from this project will support the actors of the medical sector in their efforts to promote FMT and to foster innovative therapeutic approaches in addressing medical challenges.
PI: Carole Devaux, LIH
The Pseudomonas aeruginosa (P. aeruginosa) bacterium and the Aspergillus fumigatus (A. fumigatus) fungus are associated with persistent and chronic, or recurrent and acute, respiratory infections. P. aeruginosa can thrive as a biofilm at the surface of bronchial airways easing the colonization of the host and resulting in a chronic infection, primarily observed in patients suffering from cystic fibrosis (CF) or chronic obstructive pulmonary disease (COPD). The exopolysaccharide Psl contributes to biofilm formation, immune evasion, inflammation, and damage to the lung. In CF patients, P. aeruginosa is frequently associated with A. fumigatus fungi (≈ 60% cases of over 12 year-old subjects) resulting in a worse prognosis, as it has been observed in COVID-19 patients in intensive care units. Both pathogens display resistance to a variety of antibiotics or antifungals and have now been classified as critical “priority one” pathogens by the WHO.
Antibody-based therapeutics will offer novel perspectives for addressing chronic respiratory infections and counteract antimicrobial/antifungal resistance. We have developed multimeric complexes (CoMiX) that activate the complement pathways at the surface of P. aeruginosa using a scFv targeting the exopolysaccharide Psl, as well as trispecific Natural Killer (NK) cell engagers targeting two surface receptors of NK cells, NKG2D and SLAMF-7, as well as Psl from the bacterium. Both approaches induce killing of bacteria and inhibit acute infection.
We aim to validate the pre-clinical efficacy of therapeutic approaches against lung’s invasive and chronic diseases. We expect to foster innovative pharmaceutical breakthroughs that could be directly translated into patients in the near future.