The GME team studies the pathogenic and adaptive properties of the spore-forming bacteria of the Bacillus cereus sensu lato group, as well as Clostridioides difficile. To reach our objectives we rely on molecular genetics and biochemistry approaches combined with single-cell imaging technologies and the use of the versatile infection model Galleria mellonella. As a direct result of our fundamental research, biotechnological applications in the fields of human health, food safety and crop protection also form a significant part of the GME team’s activities.
C. difficile (Cd) is an anaerobic and spore-forming entero-pathogen that is able to jump across host species and represents a One Health problem. This opportunist behavior is controlled by the host microbiota: it can be eliminated or asymptomatically carried in a healthy host, but, after an antibiotic treatment, microbiota dysbiosis favors Cd emergence as a pathogen. Its secreted toxins cause diarrhea that are difficult to treat and recurrent. Like those of many pathogens, Cd biofilms could be involved in antibio-tolerance, persistence and infection recurrence.
Bt, Bc and Cd can trigger various developmental pathways that enable them to persist in their niche. In particular, these bacteria can produce biofilms, form spores or enter a state of non-sporulated persistence. These processes are activated when bacteria encounter unfavorable environmental conditions during infection or outside the host.
The different stages of the Bt/Bc infectious cycle (virulence, necrotrophism and sporulation) are controlled by cell-cell communication or quorum-sensing (QS) systems that enable bacteria to coordinate gene expression in response to cell density and environmental stimuli (doi: 10.3390/toxins6082239). Our research has led to the characterization of the RRNPP family of quorum sensors, initially comprising Rap, NprR, PlcR and PrgX (doi: 10.1073/pnas.0704501104). Work is currently underway to study how QS functions in vivo at the single cell level. We are also characterizing molecules that can inhibit QS (via quorum-quenching) and thus reduce bacterial virulence or sporulation in collaboration with Pr Z. Hayouka (The Hebrew University of Jerusalem). In addition, the ecological role of QS systems will be investigated using evolutionary biology approaches in collaboration with Pr. B. Raymond (University of Exeter).
Although quorum-sensing controls different stages of the Bt/Bc infectious cycle, not all bacteria engage in each process, resulting in phenotypic heterogeneity, in biofilms as well as during infection. In addition to spores, considered for many years to be the only mode of survival for these bacteria, a non-sporulating subpopulation capable of persisting through the late stages of infection of G. mellonella was identified (doi: 10.1128/mbio.00371-23). We are interested in characterizing the genetic basis of this state, and its importance for the fitness of the bacteria, using global and targeted approaches (Tn-Seq and single-cell gene expression analysis) combined with genetics methods. This study extends to other members of the B. cereus group such as emetic and clinical Bc, as well as Bacillus anthracis and Bc biovar anthracis, to determine the importance of non-spore-forming forms during infection for these closely related bacteria that present different life cycles.
Canonical Bt strains produce insecticidal parasporal crystals during sporulation in the same cell as the spore. A few rare strains however, can differentiate into two distinct subpopulations of spore-formers and crystal-producers and we showed that this division-of-labor phenotype provides the bacterium with a fitness advantage in competition with a canonical Bt strain (doi: 10.1038/ismej.2014.122). The CpcR transcription regulator was characterized as being responsible for this phenotype (doi: 10.1111/mmi.14439). Ongoing work aims at understanding the mechanisms underlying this phenotypic heterogeneity. This work illustrates the diversity of strategies employed by these bacteria to control their sporulation and enable their dissemination in various ecological niches. These studies were performed in collaboration with the laboratory of Pr. F. Song (IPP CAAS Beijing) and contributed to the creation of the International Associated Laboratory focused on bacteria-plant-insect interactions for disease biocontrol (LIA BIPI).
The B. cereus group includes closely related species that produce virulence factors, such as enterotoxins, hemolysins, phospholipases, proteases and adhesins (doi: 10.1128/microbiolspec.GPP3-0032-2018). The master virulence regulator PlcR, which belongs to the RRNPP family of quorum sensors, controls the expression of most of these factors. However, while some strains are responsible for food-borne gastroenteritis, there is no correlation between the presence and expression level of virulence genes and the pathogenicity of strains. Distinguishing enteropathogenic strains is challenging due to the absence of an available animal model. Our objective is to investigate the regulation of virulence factor production in diarrheal versus environmental strains under conditions similar to those found in the intestine. A better understanding of the regulation of Bc pathogenicity will enable the identification of specific genetic determinants that differentiate enteropathogenic strains from the others.
Our group has identified a new actor of the virulence of Bt: the transcriptional regulator VipR (doi: 10.1128/microbiolspec.GPP3-0032-2018). This regulator is autoregulated and controls the expression of the exported insecticidal protein Vip3A, a toxin highly active towards the insect Spodoptera frugiperda, a devastating crop pest. In addition to Vip3a, VipR activates several other genes at the beginning of the stationary phase, including amidase and Cry toxin genes. Together, these results lead to reconsider the regulation and role of Cry toxins during infection, that 40 years of study had definitively classified as proteins specifically produced during sporulation to form spore-associated crystals. We are currently investigating the role of this regulator and of the amidases, as well as the potentially unexpected functions of the early expressed insecticidal toxins during the infectious cycle of the bacteria in an insect model. Part of this work will be conducted in the frame of the LIA BIPI focused on bacteria-plant-insect interactions for disease biocontrol.
Most emetic strains of Bc harbor a megaplasmid called pCER270. This plasmid carries the ces locus involved in the production of the emetic toxin, in addition to numerous genes of unknown function or coding for regulators. We have shown that pCER270 is involved in phenotypes specific to emetic strains, such as high spore resistance to heat or the formation of atypical biofilms, and we hypothesize that these specific phenotypes make the emetic strains well-adapted to human food and therefore contribute to their pathogenicity. As a consequence, transfer of pCER270 to non-pathogenic strains could result in the emergence of new pathogens. Although pCER270 is not a conjugative plasmid, we succeeded in transferring it to Bt and B. weihenstephanensis by a conduction process and found that transfer of the plasmid impacted the transcriptome of the new host (doi: 10.1016/j.resmic.2023.104074). We also found that pCER270 impacts both sporulation and biofilm formation and we identified pCER270-encoded regulators involved in the pCER270-induced phenotypes. We are currently determining the effect of pCER270 on the fitness of its host in various environments.
Some insects are considered to be a new sustainable source of protein for human and animal consumption among which are the mealworm Tenebrio molitor and the Black Soldier Fly larvae, Hermetia illucens. Galleria mellonella is included as a model to study interactions of microbiota, pathogen and the host responses. Our research is dealing with two main aspects: 1) the microbial safety of insect-based food/ feed, and 2)the role of gut microbiota and feeding substrate quality in insect resistance against pathogens.
In order to propose an alternative to chemical pesticides, we have developed a new type of biopesticide strains for use against insect pests, such as crop pests and disease vectors (BIOSAFE, MOSKO, BT-VIP). With financial support from SATT Paris-Saclay and INRAE, these strains were developed using the results generated by our work on stationary-phase regulators such as CpcR and VipR, mentioned above, as well as patents (EP17305011 and PCT/EP2023/061003).
Bt is largely used in crop protection because of its massive production of insecticidal Cry proteins, stored in large crystalline parasporal bodies. We have used Bt molecular tools to create a new platform for heterologous proteins production and in vivo-crystallization. Using GFP as a proof-of-concept for our platform, we found that this protein was stored in fluorescent crystals and represented up to 60% of the total bacterial protein content. We are currently improving the process to make it usable with a large range of proteins. The outcomes of this project are multiple. In vivo crystallization will facilitate the 3D-protein structure determination by avoiding the long and tedious in vitro protein crystallization process. In addition, proteins in crystals are stable, and display low intramolecular motion, thus allowing the production of unstable or cytotoxic proteins. Finally, crystals can be used as a drug-delivery vehicle for slow release of pharmaceuticals. This technology is based on patented results (PCT/EP2023/061003 and PCT/EP2023/061001).
Although most strains of Bc are innocuous and largely distributed in the environment, a number of strains are involved in food poisoning. Indeed, Bc is one of the most frequently isolated microorganisms in foodborne outbreaks. However, molecular markers are available only for emetic strains producing the cereulide toxin or for strains carrying the cytK1 gene coding for cytotoxin K1, which both make less than 1/3 of the food poisoning cases. All the other virulence factors are found in environmental as well as in pathogenic strains. As a consequence, almost all isolates inducing diarrheal symptoms cannot be currently detected or identified. We hypothesize that the amount of toxins produced, rather than the presence of a specific virulence factor, is the one of the critical factors responsible for the pathogenicity of diarrheal strains. We are therefore engaged in setting up immune-enzymatic assays in order to quantify the production of virulence factors by suspicious strains of Bc. We already successfully developed two ELISA assays, one directed against the phospholipase sphingomyelinase, the second against the hemolysin HlyII.
The larval stages of the lepidopteran insect G. mellonella are increasingly used in many laboratories as an infection model for bacterial and fungal pathogens. As it can be used for preclinical studies, it meets the 3R concept (reduce, replace, optimize) in order to limit the use of mammal models, notably for screening novel antibiotics but also for pathogen-host interaction studies. This insect has been reared in our insectarium for many years and has been used to identify genes involved in virulence, adaptation and survival and to visualize their expression in situ, particularly for Bc and Bt (see results in Axes 1, 2 and 3). Host-pathogen-microbiota studies were also undertaken as we are able to rear axenic larvae and conduct infection by oral force-feeding. We produce and sell last instars G. mellonella to several scientific laboratories in France. We set up collaborations to notably study bacteria-digestive tract interactions with respect to the role of gut microbiota, and the peritrophic matrix, using dissection and histology approaches. For instance, by using axenic larvae we showed that the gut microbiota played no role in the growth of G. mellonella on its natural feed (bees wax) and that the larva was not able to assimilate polyethylene plastic (PE) as shown in collaboration with Synchrotron using infrared microscopy FTIR on sections of whole larva fed with labelled PE (see figure) (doi: 10.1021/acs.est.1c03417).
Selected publications since 2018. For an exhaustive list since the creation of the Micalis UMR in 2010, click here
