A retrospective cross-sectional quantitative molecular approach in biological samples from patients with syphilis

Syphilis is the sexually transmitted disease caused by Treponema pallidum, a pathogen highly adapted to the human host. As a multistage disease, syphilis presents distinct clinical manifestations that pose different implications for diagnosis. Nevertheless, the inherent factors leading to diverse disease progressions are still unknown. We aimed to assess the association between treponemal loads and dissimilar disease outcomes, to better understand syphilis. We retrospectively analyzed 309 DNA samples distinct anatomic sites associated with particular syphilis manifestations. All samples had previously tested positive by a PCR-based diagnostic kit. An absolute quantitative real-time PCR procedure was used to precisely quantify the number of treponemal and human cells to determine T. pallidum loads in each sample. In general, lesion exudates presented the highest T. pallidum loads in contrast with blood-derived samples. Within the latter, a higher dispersion of T. pallidum quantities was observed for secondary syphilis. T. pallidum was detected in substantial amounts in 37 samples of seronegative individuals and in 13 cases considered as syphilis-treated. No association was found between treponemal loads and serological results or HIV status. This study suggests a scenario where syphilis may be characterized by: i) heterogeneous and high treponemal loads in primary syphilis, regardless of the anatomic site, reflecting dissimilar duration of chancres development and resolution; ii) high dispersion of bacterial concentrations in secondary syphilis, potentially suggesting replication capability of T. pallidum while in the bloodstream; and iii) bacterial evasiveness, either to the host immune system or antibiotic treatment, while remaining hidden in privileged niches. This work highlights the importance of using molecular approaches to study uncultivable human pathogens, such as T. pallidum, in the infection process.This work was supported by the Portuguese Foundation for Science and Technology (EXPL/BIA-MIC/0309/2013).