A New Method of Metabarcoding Microsporidia and Their Hosts Reveals High Levels of Microsporidian Infections in Mosquitoes (Culicidae)

DNA metabarcoding offers new perspectives, especially with regard to the high-throughput identification and diagnostics of pathogens. Microsporidia are an example of widely distributed, opportunistic and pathogenic microorganisms in which molecular identification is important for both environmental research and clinical diagnostics. We have developed a method for parallel detection of both microsporidian infection and the host species. We designed new primer sets: one specific for the classical Microsporidia (targeting the hypervariable V5 region of small subunit [ssu] rDNA), and a second one targeting a shortened fragment of the COI gene (standard metazoan DNA-barcode); both markers are well suited for next generation sequencing. Analysis of the ssu rDNA data set representing 607 microsporidian species (120 genera) indicated that the V5 region enables identification of >98% species in the data set (596/607). To test the method, we used microsporidians that infect mosquitoes in natural populations. Using mini-COI data, all field-collected mosquitoes were unambiguously assigned to seven species; among them almost 60% of specimens were positive for at least 11 different microsporidian species, including a new microsporidian ssu rDNA sequence (Microsporidium sp. PL01). Phylogenetic analysis showed that this species belongs to one of the two main clades in the Terresporidia. We found a high rate of microsporidian co-infections (9.4%). The numbers of sequence reads for the operational taxonomic units suggest that the occurrence of Nosema spp. in co-infections could benefit them; however, this observation should be retested using a more intensive host sampling. Our results show that DNA barcoding is a rapid and cost-effective method for deciphering sample diversity in greater resolution, including the hidden biodiversity that may be overlooked using classical methodology.