Identification of phlebotomine sand flies using one MALDI-TOF MS reference database and two mass spectrometer systems

Alexander Mathis,Christian Kaufmann,Sonia Zapata,Vit Dvořák,Holly Tuten,Petr Volf,Valentin Pflüger,Denis Sereno,Jorian Prudhomme,Kristýna Hlavačková,Jérôme Depaquit,Francis Schaffner,Véronique Lehrter,Anne-Laure Bañuls,Petr Halada

doi:10.1186/s13071-015-0878-2

Abstract

BackgroundRapid, accurate and high-throughput identification of vector arthropods is of paramount importance in surveillance programmes that are becoming more common due to the changing geographic occurrence and extent of many arthropod-borne diseases. Protein profiling by MALDI-TOF mass spectrometry fulfils these requirements for identification, and reference databases have recently been established for several vector taxa, mostly with specimens from laboratory colonies.MethodsWe established and validated a reference database containing 20 phlebotomine sand fly (Diptera: Psychodidae, Phlebotominae) species by using specimens from colonies or field-collections that had been stored for various periods of time.ResultsIdentical biomarker mass patterns (‘superspectra’) were obtained with colony- or field-derived specimens of the same species. In the validation study, high quality spectra (i.e. more than 30 evaluable masses) were obtained with all fresh insects from colonies, and with 55/59 insects deep-frozen (liquid nitrogen/-80 °C) for up to 25 years. In contrast, only 36/52 specimens stored in ethanol could be identified. This resulted in an overall sensitivity of 87 % (140/161); specificity was 100 %. Duration of storage impaired data counts in the high mass range, and thus cluster analyses of closely related specimens might reflect their storage conditions rather than phenotypic distinctness. A major drawback of MALDI-TOF MS is the restricted availability of in-house databases and the fact that mass spectrometers from 2 companies (Bruker, Shimadzu) are widely being used. We have analysed fingerprints of phlebotomine sand flies obtained by automatic routine procedure on a Bruker instrument by using our database and the software established on a Shimadzu system. The sensitivity with 312 specimens from 8 sand fly species from laboratory colonies when evaluating only high quality spectra was 98.3 %; the specificity was 100 %. The corresponding diagnostic values with 55 field-collected specimens from 4 species were 94.7 % and 97.4 %, respectively.ConclusionsA centralized high-quality database (created by expert taxonomists and experienced users of mass spectrometers) that is easily amenable to customer-oriented identification services is a highly desirable resource. As shown in the present work, spectra obtained from different specimens with different instruments can be analysed using a centralized database, which should be available in the near future via an online platform in a cost-efficient manner.

Highlights

Rapid, accurate and high-throughput identification of vector arthropods is of paramount importance in surveillance programmes that are becoming more common due to the changing geographic occurrence and extent of many arthropod-borne diseases
We evaluated whether raw mass spectra obtained with the instrument of one company (Ultraflex III, Bruker, Germany) allow species identification by using the software and the reference database established on an instrument of another company (Axima Confidence, Shimadzu, Japan), whereby the two most widespread instrument series were involved
In the case of Ph. barguesae, the partial sequence of the c oxidase subunit I gene (COI) gene obtained from the field-collected specimens best matched with the two GenBank sequences ascribed to this species

Summary

Introduction

Accurate and high-throughput identification of vector arthropods is of paramount importance in surveillance programmes that are becoming more common due to the changing geographic occurrence and extent of many arthropod-borne diseases. Most spectacular is the emergence of invasive species, e.g. Aedes mosquitoes [1, 2], with associated new disease threats [3] Other arthropod vectors such as ticks (Ixodida: Ixodidae) [4, 5], biting midges (Diptera: Ceratopogonidae) [6] and phlebotomine sand flies (Diptera: Psychodidae) [7,8,9] are gradually expanding their distribution ranges in Europe. Because these changes in vector distribution are associated with changes in the risk of exposure to the pathogens they transmit, monitoring the presence and abundances of arthropod vectors is of growing importance in many regions. Updated distribution maps for Europe are provided by ECDC (www.ecdc.europa.eu) for a number of vectors (eight sand fly species, five Aedes mosquito species, four tick species)

Methods

Results

Discussion

Conclusion