Correction: Vector mapping and bloodmeal metabarcoding demonstrate risk of urban Chagas disease transmission in Caracas, Venezuela.

[This corrects the article DOI: 10.1371/journal.pntd.0010613.].

Vector mapping and bloodmeal metabarcoding demonstrate risk of urban Chagas disease transmission in Caracas, Venezuela.

Chagas disease is a significant public health risk in rural and semi-rural areas of Venezuela. Triatomine infection by the aetiological agent Trypanosoma cruzi is also observed in the Metropolitan District of Caracas (MDC), where foodborne T. cruzi outbreaks occasionally occur but active vector-to-human transmission (infection during triatomine bloodmeal) is considered absent. Citizen science-based domiciliary triatomine collection carried out between 2007 and 2013 in the MDC has advanced understanding of urban T. cruzi prevalence patterns and represents an important public awareness-building tool. The present study reports on the extension of this triatomine collection program from 2014 to 2019 and uses mitochondrial metabarcoding to assess feeding behavior in a subset of specimens. The combined, thirteen-year dataset (n = 4872) shows a high rate of T. cruzi infection (75.2%) and a predominance of Panstrongylus geniculatus (99.01%) among triatomines collected in domiciliary areas by MDC inhabitants. Collection also involved nymphal stages of P. geniculatus in 18 of 32 MDC parishes. Other collected species included Triatoma nigromaculata, Triatoma maculata, Rhodnius prolixus, and Panstrongylus rufotuberculatus. Liquid intestinal content indicative of bloodmeal was observed in 53.4% of analyzed specimens. Dissection pools representing 108 such visually blooded P. geniculatus specimens predominantly tested positive for human cytochrome b DNA (22 of 24 pools). Additional bloodmeal sources detected via metabarcoding analysis included key sylvatic T. cruzi reservoirs (opossum and armadillo), rodents, and various other synanthropic and domesticated animals. Results suggest a porous sylvatic-domiciliary transmission interface and ongoing adaptation of P. geniculatus to the urban ecotope. Although P. geniculatus defecation traits greatly limit the possibility of active T. cruzi transmission for any individual biting event, the cumulation of this low risk across a vast metropolitan population warrants further investigation. Efforts to prevent triatomine contact with human food sources also clearly require greater attention to protect Venezuela's capital from Chagas disease.

Optimized multilocus sequence typing (MLST) scheme for Trypanosoma cruzi.

Trypanosoma cruzi, the aetiological agent of Chagas disease possess extensive genetic diversity. This has led to the development of a plethora of molecular typing methods for the identification of both the known major genetic lineages and for more fine scale characterization of different multilocus genotypes within these major lineages. Whole genome sequencing applied to large sample sizes is not currently viable and multilocus enzyme electrophoresis, the previous gold standard for T. cruzi typing, is laborious and time consuming. In the present work, we present an optimized Multilocus Sequence Typing (MLST) scheme, based on the combined analysis of two recently proposed MLST approaches. Here, thirteen concatenated gene fragments were applied to a panel of T. cruzi reference strains encompassing all known genetic lineages. Concatenation of 13 fragments allowed assignment of all strains to the predicted Discrete Typing Units (DTUs), or near-clades, with the exception of one strain that was an outlier for TcV, due to apparent loss of heterozygosity in one fragment. Monophyly for all DTUs, along with robust bootstrap support, was restored when this fragment was subsequently excluded from the analysis. All possible combinations of loci were assessed against predefined criteria with the objective of selecting the most appropriate combination of between two and twelve fragments, for an optimized MLST scheme. The optimum combination consisted of 7 loci and discriminated between all reference strains in the panel, with the majority supported by robust bootstrap values. Additionally, a reduced panel of just 4 gene fragments displayed high bootstrap values for DTU assignment and discriminated 21 out of 25 genotypes. We propose that the seven-fragment MLST scheme could be used as a gold standard for T. cruzi typing, against which other typing approaches, particularly single locus approaches or systematic PCR assays based on amplicon size, could be compared.