[The lung microbiota. Review]. / Le microbiote pulmonaire.

In the last 20years, culture-independent DNA-based techniques ("shotgun sequencing") demonstrated that complex microbial communities reside on most epithelial surfaces, including the lower airways. Until the amniotic sac ruptures, a fetus is considered to be essentially sterile. Many factors affect the composition of the lung microbiota: inheritance, mode of delivery, diet, and age-related changes in adults. It interacts with the digestive and oropharyngeal microbiotas. Animal models show that these interactions play a role in innate pulmonary immunity and modulation of the inflammatory response. The microbial composition of the airway microbiota differs between healthy children and those with chronic lung disease. The advances in the comprehension of microbiome changes have resulted in new approaches concerning the microbiota for treatment and prevention of disease.

[Contribution of in vitro models to the management of hematologic risk in pharmacovigilance]. / Contribution des modèles in vitro à la gestion du risque hématologique en pharmacovigilance.

The decisions of health authorities concerning adverse effects of drugs are usually notified following clinical observations, but are rarely associated to experimental data. The haematopoietic tissue is one of the most sensitive to these effects. In order to anticipate and to explain adverse effects, it becomes necessary to carry out in vitro assays on normal human haematopoietic progenitors. We had the opportunity to use human cord blood progenitors which are able to repopulate allogeneic aplastic bone marrow. Many advantages are associated with this model: numerous samples, non-invasive, absence of species bias, possibilities of mechanistic approach. The clonogenic potential of progenitors in soft agar, as well as their ability to expand in liquid medium after stimulation with specific growth factors, have been used. Evidence of dose-related toxicity by inhibition of colony formation or proliferation was analysed in the presence of reference molecules. Results were reproducible despite an intrinsic variability of progenitor density between samples. They were comparable to assays on bone marrow progenitors reported by us and others. Comparison of toxicity thresholds with plasma therapeutic ranges showed the potential risk for some molecules tested.

A cellular model for drug interactions on hematopoiesis: the use of human umbilical cord blood progenitors as a model for the study of drug-related myelosuppression of normal hematopoiesis.

A cellular model of hematopoiesis which would be more convenient than bone marrow (BM) progenitors and directly relevant to human pathology is needed in order to investigate xenobiotic toxicity. Human umbilical cord blood (HCB), previously shown to be able to repopulate BM, provides a powerful in vitro model of normal human hematopoiesis. In order to validate the use of normal HCB progenitors as targets for dose-related myelosuppression, we used clonogenic assays and expansion in a liquid culture of progenitor-enriched cell suspensions from HCB. A series of 8 reference molecules, doxorubicin, cytosine-arabinoside, 5-fluorouracil, 3'-azido-3'-deoxythymidine, acetylsalicylic acid, sodium valproate and two cephalosporin antibiotics, were tested. In vitro 50% inhibition concentrations (IC50) were compared to those observed or reported with BM progenitors, and to the values of plasma concentrations from treated patients. HCB progenitors as in vitro targets for cytotoxic molecules were easy to access and handle, and their use was sensitive, specific and reproducible. They gave results similar to BM progenitors and allowed a qualitative approach to cellular metabolism and toxicity using morphological, flow cytometric and chromatographic methods.