MALDI-ToF detection of Leishmania infantum infection in Lutzomyia longipalpis and Nyssomyia neivai

Abstract

BackgroundMatrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-ToF MS) is widely used for sand fly identification, but its potential to detect Leishmania infections in vectors remain underexplored. This pilot study evaluated whether MALDI-ToF MS protein profiles of lab-reared Lutzomyia longipalpis and Nyssomyia neivai can discriminate Leishmania infantum-infected from uninfected females. MethodologyColonies were experimentally infected with L. infantum using membrane feeding, and females were collected at different days post-blood meal. Thoraces and legs were processed individually for MALDI-ToF MS, and spectra were analysed using both Bruker software and custom R pipelines. Principal findingsUnsupervised approaches (MSP dendrograms, PCA) showed limited or inconsistent separation of infection status for Lu. longipalpis. In contrast, supervised machine-learning models built on peak-intensity matrices achieved excellent discrimination between infected and uninfected specimens for both species, with several algorithms reaching near-perfect performance on an external test set not used for training. Variable-importance analysis highlighted sets of m/z peaks, mainly showing decreased intensity in infected sand flies, as putative infection biomarkers. ConclusionThis proof-of-concept study highlights that L. infantum infection induces reproducible, species-specific alterations in sand-fly MALDI-TOF profiles, supporting further development of high-throughput, MS-based screening of infected vectors. Author summaryLeishmania infantum is a parasite responsible for visceral leishmaniasis, a severe neglected tropical disease. It is transmitted to humans by sandfly vectors. This study explored whether the MALDI-ToF mass spectrometry technique can detect infection by the L. infantum parasite in the two main sandfly vectors in Brazil: Lutzomyia longipalpis and Nyssomyia neivai. The method has already been tested to identify sandfly species, but its ability to detect infected insects had not been well studied. We infected laboratory-reared sandflies and analyzed their protein profiles to see whether infected and uninfected individuals could be distinguished. We found that infection changes the molecular fingerprints of both sandfly species. Machine-learning models were able to distinguish infected from uninfected specimens with very high accuracy. A small part of the most informative signal was shared between both species, while most of the peaks were species-specific, suggesting that infection affects each vector in a slightly different way. These results show that MALDI-ToF has promise as a rapid, low-cost tool for screening sandflies for Leishmania infection. With further validation, this approach could complement existing surveillance methods and help monitor disease transmission in endemic areas.