Martin Fenk (Berlin / DE), Sirak Petros (Leipzig / DE), Bastian Pasieka (Leipzig / DE), Felix M. Key (Berlin / DE)
Alarmingly, many commensal bacteria that peacefully reside in the human microbiome are also able to cause acute opportunistic infections. Thought to be triggered randomly in patients with impaired immunity, intriguing evidence suggests that within-host evolution can promote infection. The major barrier to studying the progression of opportunistic pathogens from carriage to acute infections is the challenge of sampling human patients prospectively. Here we test whether adaptive de novo mutations in asymptomatically colonizing opportunistic pathogen lineages facilitate translocation within the patient's body and promote acute infections. Therefore, we have enrolled 14 critically ill patients who are longitudinally sampled at four microbiome niches before the onset of healthcare-associated infection (HAI). Among four patients, we observe twelve HAIs caused by ten different opportunistic pathogens. Using culture-based sequencing we address within-patient evolution by generating over 600 whole genomes of the pathogen populations to characterize acquisition timing, spatial translocation, and whether adaptive de novo mutations promote acute infection. We find that in 70% of HAI cases the pathogen lineages can be identified within the microbiome prior to or at the time of disease onset. Moreover, using a molecular clock we infer that the pathogen lineages are regularly acquired already before hospitalization of the patient. In a single patient we observe a short-lived mutation in a fimbriae gene associated with HAI. Phenotypic tests reveal the mutation upregulates fimbriae production increasing cell adhesion and leads to higher virulence in a Drosophila infection model. In summary, our prospective approach allowed us to explore the hidden evolutionary paths leading up to HAI and underlines the importance to investigate microbiome dynamics at a population-wide whole genome level during disease.