Nuevas aproximaciones de vigilancia de transmisión de Mycobacterium tuberculosis en entornos comunitarios complejos y de otros patógenos relevantes en el entorno nosocomial

Résumé

The application of molecular strategies has transformed the way in which we approach the analysis of pathogenic microorganisms transmission dynamics. The recent introduction of genomic analysis has caused a second revolution in the ability to discriminate between strains and in the precision in which we can approach epidemiological studies. However, further efforts are needed to define the most suitable way to integrate molecular and genomic approaches and to adapt them to different epidemiological challenges and analytical settings. With this aim, this thesis has been organized in two work packages, focusing on infections i) at the community setting, selecting Mycobacterium tuberculosis as a representative pathogen with a high global relevance, and ii) at the nosocomial context, selecting Mycobacterium chimaera, as a microorganism responsible for the emerging alarm derived from exposure to contaminated health-care devices, and MDR Pseudomonas aeruginosa, a relevant microorganism as responsible for outbreaks involving person-toperson transmission. Any attempt to optimize tuberculosis (TB) transmission surveillance must face the new global scenario for this disease, as a result of international migratory movements. Following this premise, we approach the characterization of high-risk TB, multidrug-resistant TB, transmission in one of these migratory axes, the one involving Latin America and Europe. We started characterizing by means of MIRU-VNTR a sample of MDR circulating strains in Lima, Peru, which revealed a high rate of recent transmission in that population. The comparison of those genotypes with those identified circulating among the Peruvian migrant population in Europe identified shared strains in Italy and Spain. Genomic analysis accurately determined the existence of two intercontinental exportation events of MDR-TB, from Peru to Europe, involving two variants of circulating strains in Lima, one of them being responsible for a subsequent active transmission in Florence. The integrated study between Peru and Europe revealed the need for acquiring genotypic data from circulating MTB strains, which would allow to perform similar transnational studies in other settings. Unfortunately, there are many countries of origin of migrants that lack systematic molecular epidemiology programs. From this limitation, and trying to compensate the knowledge gap between the host countries and the countries of origin of migrants, we designed an alternative strategy based on i) systematic genotyping by MIRU-VNTR of a sample from the population to be analyzed, in order to identify the most prevalent strains, ii) characterization by whole genome sequencing (WGS) of the prevalent clusters to identify strain-marker SNPs and iii) tailoring specific PCRs targeting those strains, which will optimize and simplify their prospective low-cost surveillance. A pilot study for this strategy was run in Panama. A collection of strains from two Panama provinces (Panama and Colon) was genotyped by MIRUVNTR, and revealed a high proportion (50%) of MTB clustered isolates, with an asymmetric distribution of some strains, predominant either in Colon or in Panama. The analysis by WGS of these clusters allowed us to determine that only one of them corresponded to recent transmission, with isolates showing low diversity among them, while the remaining corresponded to prevalent strains. The design of a set of specific PCRs made it possible to ensure prospective on-site surveillance of these strains, which were responsible for a third of all TB cases in the populations studied. One of the prevalent strains in Colon corresponded to the Beijing lineage. Because this lineage has been associated with high virulence and transmission, a rapid update of its presence was carried out in Colon, supported by a specific PCR, revealing that it is responsible for 57% of all incident cases. The integrated study of the WGS data from all the isolates, with those available in global databases, made it possible to determine that it belongs to a modern Beijing sublineage, the African Asian branch 3 (Bmyc13, L2.2.5), being Vietnam the most likely origin, as indicated by the phylogenetically closest isolates. The study of the diversity accumulated by this strain representatives in Panama allowed us to infer that 2000-2012 was the most likely period for its entrance to the country. Given the cost-effectiveness shown by our combined strategy of directed genotyping, WGS and development of strain-specific PCRs, in the study of TB transmission, we pursued to evaluate its transference to the second area of interest of this thesis, nosocomial infections. With this aim, we replicated the work scheme developed for TB, now for the characterization of an XDR P. aeruginosa, involving 14 patients, which had been approaches only by conventional genotyping (pulsed field electrophoresis). The analysis by WGS and the consequent strain-specific PCR targeting strain-marker SNPs identified from the genomic analysis, allowed us to rule out unrelated cases and to identify another new cases that had not been suspected. In addition, it defined the existence of overlapping outbreaks of a smaller magnitude, identified the true index case, and finally, alerted that the outbreak was still active at the time of analysis. Finally, due to our interest in optimizing nosocomial transmission surveillance, and after having focused on an example of person-person transmission, it was appropriate to focus on other events involving exposure to contaminated health-care devices. The recent global alarm caused by the exposure of patients undergoing cardiac surgery to “heater-cooler” (HCU) devices contaminated with Mycobacterium chimaera justified our attention on this second event as a study model. We evaluated the usefulness of applying a real-time PCR. based on the genomic identification of specific M. chimaera sequences, for the fast identification of this bacteria, in order to assure a double response to: i) optimize prospective environmental monitoring, analyzing directly water samples from the HCU devices and ii) retrospectively identify cases infected by this microorganism that would have gone unnoticed. The strategy showed its usefulness for the early identification of contaminated HCU devices, allowing their monitoring and revealed M. chimaera infection in a patient who had undergone cardiac surgery in an institution different from ours. The WGS study of this patient isolate and of those obtained from the contaminated devices, allowed us to demonstrate the involvement of the strain responsible for the global outbreak.