Use of MALDI-TOF MS for the Identification of Chad Mosquitoes and the Origin of Their Blood Meal

Abstract.

Matrix-assisted desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) is a clinical microbiology tool for the systematic identification of microorganisms. It has recently been presented as an innovative tool for the rapid and accurate identification of mosquitoes and their blood meal. To evaluate the capacity of this tool to identify mosquitoes collected in a tropical environment and preserved with silica gel, we analyzed 188 mosquitoes of different species collected in Chad, which were preserved with silica gel for 2 months. The MALDI-TOF MS analysis correctly identified 96% of the mosquitoes and 37.5% of their blood meals. Using MALDI-TOF MS and molecular biology, eight mosquito species were identified, including Anopheles gambiae s.l., Anopheles rufipes, Culex quinquefasciatus, Culex neavei, Culex pipiens, Culex perexiguus, Culex rima, and Culex watti. Blood meal identification revealed that mosquitoes fed mainly on humans, birds, and cows. Matrix-assisted desorption/ionization time-of-flight mass spectrometry appears to be a promising, fast, and reliable tool to identify mosquitoes and the origin of their blood meal for samples stored with silica gel.

Introduction

Mosquitoes are the primary arthropod vectors of infectious diseases, posing serious economic and public health problems because of their role in the transmission of numerous human and veterinary pathogens.¹ To human, they are capable of transmitting not only parasitic diseases such as malaria and lymphatic filariasis but also serious arboviruses including yellow fever, dengue fever, chikungunya, Zika virus, and West Nile virus (WNV) infections.² Malaria, caused by several species of Plasmodium parasites (Plasmodium falciparum, Plasmodium vivax, Plasmodium malariae, Plasmodium ovale wallikeri, P. ovale curtisi, Plasmodium knowlesi, and Plasmodium simium), is transmitted to humans by female Anopheles spp. mosquitoes.^3,4 According to the World Health Organization, approximately five million additional cases of malaria in 2016 compared with 2015 and 445,000 deaths were reported.⁵

Despite the availability of antimalarial treatments, vector control measures are needed to control the mosquito vectors.⁶ Long-lasting insecticide-treated bed net, indoor residual spraying, larviciding, and community education to promote vector avoidance are commonly used approaches.⁷ The implementation of vector control and surveillance strategies against mosquitoes requires entomological surveys including correct identification not only of the vectors but also of their blood meal for a better understanding of their biting behavior (endophilic or exophilic and anthropophilic or zoophilic).^8,9

Mosquito identification is most often performed using morphological criteria using identification keys and/or molecular methods.¹⁰ These methods, however, have limitations, which may be the absence of identification keys or specific documentation, expertise in entomology, and inability to differentiate species from the same complex for the morphological method. On the other hand, molecular approaches are time consuming, expensive, and limited by the completeness of online sequence databases.¹⁰ Similarly, the origin of the blood meal is identified by several methods such as precipitin, enzyme immunoassay, and molecular tests.^11,12 However, these methods also have drawbacks, such as the difficulty of obtaining specific antisera against a wide variety of host species, the effect of blood meal digestion and DNA extraction protocol, the high cost, handling time, and the need for bulky equipment.^13,14

Matrix-assisted desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) is a clinical microbiology tool used for the systematic identification of bacteria,^15,16 archaea,¹⁷ fungus, and parasites.¹⁸ Recently, MALDI-TOF MS has been presented as an alternative tool for the rapid identification of many arthropods including mosquitoes^19–22 and the origin of their blood meal.^23,24 This involved different entomological studies using fresh samples from the laboratory and samples collected from the fields which were either frozen, preserved, preserved in alcohol, or crushed on Whatman paper.^23–26

The aim of this study was to evaluate the ability of MALDI-TOF MS to identify mosquitoes and the origin of their blood meal using samples collected in the field in a tropical setting, preserved with silica gel, and sent to a place which has a MALDI-TOF MS device available.

Materials and methods

Mosquito collection.

All mosquitoes were collected as part of an entomological survey made by the French Army Centre for Epidemiology and Public Health in Chad in October 2017. Mosquitoes were collected using BG sentinel traps (Biogents AG, Weißenburgstraße, Regensburg, Germany)²⁷ and CDC light traps (John W. Hock Company, Gainesville, États-Unis),²⁸ which were inspected every day between 6:00 am and 7:00 am and between 6:00 pm and 7:00 pm. The mosquitoes were morphologically identified every day to the genus and at the species level for Anopheles females, using morphological criteria.²⁹

A convenient sample of mosquitoes was selected for the present study. The mosquitoes were stored individually (for Anopheles spp.) or by a pool of 3–10 specimens (for Culex spp.) in a 1.5-mL Eppendorf tube with silica gel (Carl Roth GmbH, Karlsruhe, Germany) before being sent to Marseille, France, on November 3, 2017, for subsequent analysis.

Preparation of samples for MALDI-TOF MS analysis.

Matrix-assisted desorption/ionization time-of-flight mass spectrometry identification of mosquito samples.

Two months after being maintained at room temperature, the legs of each mosquito were removed and placed individually in 1.5 mL Eppendorf tubes with glass powder (Sigma, Lyon, France), 15 μL of 70% (v/v) formic acid (Sigma), and 15 μL of 50% (v/v) acetonitrile (Fluka, Buchs, Switzerland). The samples were ground using a tissue lyzer (Qiagen, Hilden, Germany) over three cycles of 30 ms⁻¹ for 60 seconds.³⁰ The samples were then centrifuged at 10.000 rpg for 1 minute, and 1 μL of supernatant of each homogenate was deposited on a MALDI-TOF MS target plate in quadruplicate (Bruker Daltonics, Wissembourg, France) and covered with 1 μL of α-cyano-4-hydroxycynnamic acid (CHCA) matrix solution composed of saturated α-cyano-4-hydroxycynnamic acid (Sigma), 50% acetonitrile (v/v), 2.5% trifluoroacetic acid (v/v) (Aldrich, Dorset, UK), and high-performance liquid chromatography (HPLC) grade water.^26,30 After drying for several minutes at room temperature, the target was introduced into the MALDI-TOF Microflex LT mass spectrometer (Bruker Daltonics, Breman, Germany) for analysis.

Matrix-assisted desorption/ionization time-of-flight mass spectrometry identification of blood meal sources.

The engorged abdomen of each female mosquito was individually ground in an Eppendorf tube containing 50 μL of HPLC grade water. After centrifugation, 10 μL of the supernatant was used for 10 μL for the MALDI-TOF analysis, as previously described.²³ Ten microliter of the abdomen supernatant was mixed with 20 μL of 70% (v/v) formic acid and 20 μL of 50% (v/v) acetonitrile (Fluka) and then centrifuged at 10. 000 rpm for 20 seconds. One microliter of supernatant from each sample was placed on the MALDI-TOF target plate in quadruple (Bruker Daltonics) and recovered with 1 μL of CHCA matrix solution composed of saturated α-cyano-4-hydroxycynnamic acid (Sigma), 50% acetonitrile (v/v), 2.5% trifluoroacetic acid (v/v) (Aldrich), and HPLC grade water. After drying for several minutes at room temperature, the target was introduced into the MALDI-TOF Microflex LT mass spectrometer (Bruker Daltonics) for analysis.

Spectral analysis.

Protein mass profiles were acquired using a Microflex LT MALDI-TOF mass spectrometer, with detection in the linear positive ion mode at a laser frequency of 50 Hz in a mass range of 2–20 kDa. The acceleration voltage was 20 kV and the extraction time was 200 ns. Each spectrum corresponds to the ions obtained from the 240 laser shots performed in six regions in the same location and acquired automatically using the AutoXecute Flex Control software v.2.4 (Bruker Daltonics). Spectrum profiles obtained from mosquito legs and engorged abdomens were visualized with FlexAnalysis software v.3.3, and low-quality spectra were excluded from the study based on their intensity, reproducibility, and noise. They were then exported to ClinProTools version v.2.2 (Bruker Daltonics) and MALDI-Biotyper v.3.0 (Bruker Daltonics) for data processing.

Blind tests for the identification of mosquitoes and blood meals.

To determine the mosquito species and the origin of blood meals, MALDI-TOF MS spectra from the legs and abdominal protein extracts of blood-engorged females were queried against the homemade MS reference spectra database (Table 1) using the MALDI-Biotyper software v3.0. tool (Bruker Daltonics). The level of significance was determined using the log score values (LSVs) provided by the MALDI-Biotyper software v.3.3. corresponding to a matched degree of signal intensities of mass spectra of the query and the reference spectra. Log score values ranged from zero to three. The samples were correctly considered and significantly identified when the spectrum queried had an LSV ≥ 1.8.²³ After molecular identification, the reference spectra of Culex perexiguus (n = 1), Culex watti (n = 1), and Culex rima (n = 1) were added in the homemade MS reference spectra database and a second blind test was made against the new database.

Table 1.

List of the arthropod species present in our homemade MALDI-TOF MS database

Mosquitoes	Imago: Aedes aegypti, Aedes albopictus, Aedes alternans, Aedes australis, Aedes caspius, Aedes cinereus, Aedes dufouri, Aedes flavifrons, Aedes fowleri, Aedes multiplex, Aedes notoscriptus, Aedes polynesiensis, Aedes procax, Aedes vexans, Aedes vigilax, Aedes vittiger, Anopheles annulipes, Anopheles arabiensis, Anopheles claviger, Anopheles coluzzi, Anopheles coustani, Anopheles funestus, Anopheles gambiae Giles, Anopheles hyrcanus, Anopheles maculipennis, Anopheles pharoensis, Anopheles rufipes, Anopheles wellcomei, Anopheles ziemani, Coquillettidia richiardii, Coquillettidia xanthogaster, Culex annulirostris, Culex australicus, Culex insignis, Culex modestus, Culex molestus, Culex neavei, Culex orbostiensis, Culex pipiens, Culex quinquefasciatus, Culex perexiguus, Culex rima, Cx. sitiens, Culex watti, Culiseta longiareolata, Lutzia tigripes, Mansonia uniformis, Ochlerothatus rusticus, O. excrucians, Orthopodomyia reunionensis, and Verralina funerea
	Larvae: Ae. albopictus, Ae. aegypti, An. coluzzi, An. gambiae, Cx. molestus, Cx. pipiens, and Culiseta sp.
Lice	Pediculus humanus, Damalinia bovis, Damalinia caprae, Damalinia ovis, Haematopinus eurysternus, Linognatus vituli, and L. africanus
Fleas	Archaeopsylla erinacei, Ctenocephalides felis, Culex canis, and Xenopsylla chopis
	Pulex irritans, Stenoponia tripectinata, Nosopsyllus fasciatus, and Cx. canis
Ticks	Legs: Amblyomma variegatum, Dermacentor marginatus, D. marginatus–infected with R. slovaca, D. reticulatus, Haemaphysalis concinna, Haemaphysalis punctata, Hyalomma m. rufipes, Ixodes hexagonus, Ixodes ricinus, Rhipicephalus bursa, Rhipicephalus sanguineus, Rh. sanguineus–infected with Rickettsia conorii, Rh. sanguineus–infected with R. massiliae, and Rh. sulcatus
	Amblyomma gemma, Amblyomma cohaerens, Am. variegatum, Argas persicus, Haemaphysalis leachi, Hae. punctata, Haemaphysalis spinulosa, Hyalomma detritum, Hy. m. rufipes, Hyalomma truncatum, I. ricinus, Ornithodoros sonrai, Rhipicephalus annulatus, Rhipicephalus bergeoni, Rh. bursa, Rhipicephalus decoloratus, Rhipicephalus e. evertsi, Rhipicephalus microplus, Rhipicephalus praetextatus, Rhipicephalus pulchellus, and Rh. sanguineus
	Hemolymph: Am. variegatum–infected with Rickettsia africae, D. marginatus, Hy. m. rufipes, Rh. bursa, and Rh. sanguineus
Bed bugs	Cimex lectularius and Cimex hemipterus
Triatominae	Eratyrus mucronatus, Panstrongylus geniculatus, Rhodnius prolixus, Rhodnius pictipes, Rhodnius robustus, and Triatoma infestans
Sand flies	Phlebotomus papatasi, Phlebotomus longicuspis, Phlebotomus perfiliewi, Phlebotomus perniciosus, Phlebotomus sergenti, Sergentomyia minuta
Mite	Leptotrombidium chiangraiensis, Leptotrombidium imphalum, and Leptotrombidium deliense
Blattidae	Supella longipalpa, Periplaneta americana, Blatta orientalis, Blatella germanica, and Blaptica dubia
Flies	Melophagus ovinus and Hippobosca equina
Abdomen of mosquitoes engorged	An. gambiae Giles fed on Homo sapiens, Equus caballus, Ovis aries, rabbit, Balb/C. mouse, Rattus norvegicus, Canis familiaris, Bos taurus, Capra hircus, Gallus gallus, Equus asinus, Tapirus indicus, Tapirus terrestris, Carollia perspicillata, Thraupis episcopus, Erythrocebus patas, and Callithrix pygmaea blood
	Ae. albopictus fed on H. sapiens blood

Molecular identification.

Mosquito samples with high-quality spectra and LSV ≥ 1.8 but showing discrepancies between morphological identification and MALDI-TOF MS, and those with high-quality spectra and LSV < 1.8 were all identified at the species level by molecular tools. Quality of spectra was evaluated based on overall intensity of peaks, absence of noise, and reproducibility among each species, and visualized on both FlexAnalysis and ClinProTools software. Besides, some randomly selected well-identified samples (LSV ≥ 1.8 with concordance between morphological identification and MALDI-TOF MS) were also identified at the species level by molecular tools. We used the same workflow for the molecular identification of blood meals.

DNA extractions from individual mosquito heads and thorax samples or 10 μL supernatant of engorged abdomen of females were performed using the EZ1 DNA Tissue kit (Qiagen) according to the manufacturer’s recommendations. To determine the origin of the blood meal, we used primers that specifically amplified the vertebrate cytochrome c oxidase I gene (vCOI) (vCOI_long forward: 5′-AAGAATCAGAATARGTTG-3′; vCOI_long reverse: 5′-AACCACAAAGACATTGGCAC-3′).³¹ As for mosquitoes, a region of the cytochrome c oxidase I gene (mCOI) was amplified using the following primers: (LCO1490 (before): 5′-GGTCAAC AAATCATAAGATATTGG-3′; HC02198 (reverse): 5′-TAAACTTCAGGGTGACCAAAAAATCA-3′,³² and the internal transcribed spacer 2 (ITS2) was amplified using the following primers: forward: 5′-ATCACTCGGCTCATGGATCG-3′; reverse: 5′-ATGCTTAAATTTAGGGGGTAGTC-3′.³³ Positive polymerase chain reaction (PCR) products were then purified and sequenced using the same primers with the BigDye version 1-1 Cycle Sequencing Ready Reaction Mix (Applied Biosystems, Foster City, CA) and an ABI 3100 automated sequencer (Applied Biosystems). The sequences were assembled and analyzed using the ChromasPro software (version 1.34) (Technelysium Pty. Ltd., Tewantin, Australia) and the National Center for Biotechnology Information, Basic Local Alignment Search Tool (NCBI BLAST) website (http://blast.ncbi.nlm.nih.gov).

Results

Mosquitoes collection and morphological identification.

A total of 188 mosquitoes were selected randomly to have a varied number of species and sex, but not as representative of the entire collection during the entomological survey, which will be reported elsewhere. According to the morphological identification, the selected mosquitoes belonged to two genera: Culex spp. represented 112/188 (59.6%) of which 13 were males and 99 were females, and Anopheles spp. represented 76/188 (40.4%) of which 44 were males and 32 were females (Table 2). A total of 62.5% (20/32) of Anopheles females were morphologically identified as belonging to Anopheles gambiae s.l. and 37.5% (12/32) to Anopheles rufipes species.

Table 2.

The number of mosquitoes collected in Chad preserved on silica gel by sex and gender, with the legs, those with the good matrix-assisted desorption/ionization time-of-flight mass spectrometry spectra, and the percentage of identification after the first blind test

Mosquitos species or genus	Number collected	Number with legs	Number of spectra of good quality	Percentage of identification to the species level
Anopheles sp. ♂	44	38 (62.3%)	18 (47%)	94.4% (17/18)
Anopheles gambiae ♀	22	18 (81.8%)	11 (61.1%)	100% (11/11)
Anopheles rufipes ♀	10	5 (50%)	1 (20%)	100% (1/1)
Culex sp. ♂	13	11 (84.6%)	7 (63.6%)	100% (7/7)
Culex sp. ♀	99	97 (97.9%)	67 (69%)	91% (61/67)
Total	188	169 (89.9%)	104 (61.5%)	93.3% (97/104)

Matrix-assisted desorption/ionization time-of-flight mass spectrometry and molecular identification of mosquitoes.

Among the 188 mosquitoes that were preserved with silica gel, 169/188 (89.9%) had at least four legs and were selected for MALDI-TOF MS analysis (Table 2). Of these 169 mosquitoes, 104 (61.5%) provided good-quality MS spectra and were included for further MS analysis.

The spectra obtained from the legs of these 104 mosquitoes were then queried against the in-lab MS arthropod database. A total of 93.3% (97/104) were identified with LSVs ranging from 1.84 to 2.427 (average: 2.119). The other including six Culex spp. (females only) and one Anopheles spp. male had LSV less than 1.8 (Table 3).

Table 3.

Molecular identification of mosquitoes collected in Chad in October 2017 randomly selected for confirmation of MALDI-TOF MS identification

Number of samples	Morphological identification	Log score value	MALDI-TOF MS identification	Molecular identification (COI gene)	Molecular identification (internal transcribed spacer 2 gene)
632	Anopheles sp. ♂	1.79	/	Anopheles rhodesiensis/Anopheles rufipes*	An. rufipes
595	Anopheles sp. ♂	1.84	An. rufipes	An. rufipes	An. rufipes
657	Anopheles sp. ♂	1.84	An. rufipes	An. rufipes	–
693	Anopheles sp. ♂	1.854	An. rufipes	An. rhodesiensis/An. rufipes*	An. rufipes
599	Anopheles sp. ♂	1.869	An. rufipes	An. rufipes	–
698	Anopheles sp. ♂	1.879	An. rufipes	An. rhodesiensis/An. rufipes*	An. rufipes
621	Anopheles sp. ♂	1.89	An. rufipes	An. rufipes	–
677	Anopheles sp. ♂	1.91	An. rufipes	An. rufipes	–
587	Anopheles sp. ♂	1.977	An. rufipes	An. rufipes	–
585	Anopheles sp. ♂	2.005	An. rufipes	An. rhodesiensis/An. rufipes*	–
583	Anopheles sp. ♂	2.023	An. rufipes	An. rufipes	–
563	Anopheles sp. ♂	2.065	An. rufipes	An. rufipes	–
487	Anopheles sp. ♂	2.11	An. rufipes	An. rufipes	–
491	Anopheles gambiae ♀	2.126	An. gambiae	An. gambiae	–
593	An. rufipes ♀	2.128	An. rufipes	An. rufipes	–
297	An. gambiae ♀	2.23	An. gambiae	An. gambiae	–
572	An. gambiae ♀	2.261	An. gambiae	An. gambiae	–
533	Culex sp. ♀	1.037	/	Culex watti	–
694	Culex sp. ♀	1.452	/	Culex rima	–
618	Culex sp. ♀	1.513	/	Culex perexiguus	–
374	Culex sp. ♀	1.635	/	Cx. perexiguus	–
556	Culex sp. ♀	1.679	/	Cx. perexiguus	–
720	Culex sp. ♀	1.73	/	Cx. watti	–
624	Culex sp. ♀	1.956	Culex pipiens	Cx. pipiens	–
521	Culex sp. ♂	2.131	Culex quinquefasciatus	Cx. quinquefasciatus	–
522	Culex sp. ♂	2.139	Cx. quinquefasciatus	Cx. quinquefasciatus	–
740	Culex sp. ♀	2.145	Cx. pipiens	Cx. pipiens	–
705	Culex sp. ♀	2.161	Cx. pipiens	Cx. pipiens	–
715	Culex sp. ♀	2.161	Cx. pipiens	Cx. pipiens	–
568	Culex sp. ♀	2.253	Cx. quinquefasciatus	Cx. quinquefasciatus	–
682	Culex sp. ♀	2.267	Cx. quinquefasciatus	Cx. quinquefasciatus	–
523	Culex sp. ♂	2.314	Cx. quinquefasciatus	Cx. quinquefasciatus	–
520	Culex sp. ♀	2.337	Cx. pipiens	Cx. pipiens	–
592	Culex sp. ♀	2.427	Cx. quinquefasciatus	Cx. quinquefasciatus	–

MALDI-TOF MS = Matrix-assisted desorption/ionization time-of-flight mass spectrometry.

* Non-discriminative results (identical cover and identity values).

Regarding the molecular identification, the seven mosquitoes that had LSVs less than 1.8 and a high-quality spectra, and 26 mosquitoes that had LSVs greater than 1.8 and discrepancies with morphological identification (or obtained from specimen with identification to the genus only) were subjected to standard PCR and sequencing. Among the mosquitoes identified with LSVs greater than 1.8, 22/26 (84.6%) were definitively confirmed by molecular biology with unambiguous similarities with the COI gene of the corresponding species (Table 3). For 4/26 (15.4%) morphologically identified as male Anopheles spp. and as An. rufipes by MS, they showed 99.7% identity with Anopheles sp. M36YA (GenBank accession number: KU187107.1), and 98.8% identity with An. rhodesiensis and An. rufipes, (GenBank accession numbers: KU187106.1 and KJ522838.1), with the COI gene. Samples that could not be identified using the COI gene (not enough divergence between An. rhodesiensis and An. rufipes) were sequenced using the ITS2 gene and revealed 100% identity with An. rufipes reference sequences (GenBank accession number: KJ522822.1). Therefore, MALDI-TOF identification of selected mosquitoes has been confirmed by molecular biology.

Partial COI gene sequences were obtained from the seven mosquitoes that had LSVs less than 1.8: one had 99% identity with An. rufipes (GenBank accession number: KJ522838.1), three were identified as Cx. perexiguus (100% identity; GenBank accession number: KU380423.1), two as Cx. watti (99.2%; GenBank accession number: KU187063.1), and one as Cx. rima (99.6%; GenBank accession number: KU187034.1). Our homemade MS reference spectra database did not contain any of the Cx. rima, Cx. perexiguus, and Cx. watti spectrum. The spectra of the new species not present in our in-lab database before this study (Cx. rima, Cx. watti, and Cx. perexiguus) have been added.

After the molecular identification and upgrade of the database with the reference spectra of Cx. rima, Cx. perexiguus, and Cx. watti from Chad, the second blind test against this updated database identified another Cx. perexiguus that had not been identified during the first blind test with LSV greater than 1.8. After the second blind test analysis, the percentage of MS correct identification reached 96.4%.

Identification of blood meals.

A total 59 abdomens of engorged female mosquitoes (58 Culex quinquefasciatus and one An. gambiae s.l.) were used for MALDI-TOF MS analysis. Of these, 24/59 (40.7%) samples had good quality MS spectra. A total of 9/24 (37.5%) were identified as human blood with LSVs between 1.901 and 2.308. The remaining 15 samples were not realibily identified (LSVs between 1.222 and 1.681, Because of low LSVs, these identifications were considered unreliable²³ (Table 4). The abdomens of engorged mosquitoes with MS spectra of good quality were subjected to sequencing using the COI gene vertebrate to determine the origin of the blood meal (Table 4). Of these, nine that had already been identified as human blood by MALDI-TOF MS with LSVs greater than 1.8 were confirmed by molecular biology with identities ranging from 99.7% to 100% (GenBank accession numbers: MF621085.1, MG970575.2, MG272704.1, MH161386.1, and MF696131.1). Among the 15 samples with low LSVs, sequencing showed that eight had identities ranging from 99.8% to 100% with human blood reference sequences (GenBank accession numbers: MF621085.1, MG970575.2, MG272704.1, MH161386.1, and MF696131.1), two had identities ranging from 95.3% to 96.4% with the European roller (Coracias garrulous) blood reference sequences (GenBank accession number: GQ481616.1), and one had 98.2% identity with domestic goat (Capra hircus) blood reference sequence (GenBank accession number: KX845672.1). No sequence was obtained for three samples.

Table 4.

MALDI-TOF MS and molecular identification of blood meal origin from abdomens of engorged females

Sample numbers	Log score value	MALDI-TOF MS identification	Molecular identification	% Identity	GenBank accession number
762	1.222	/	No sequence	/	–
605	1.267	/	Coracias garrulus	95.27	GQ481616.1
506	1.271	/	Coracias garrulus	96.40	GQ481616.1
681	1.290	/	Capra hircus	98.17	KX845672.1
623	1.309	/	Homo sapiens	100	MF621085.1
671	1.334	/	H. sapiens	99.68	MH378688.1
646	1.339	/	H. sapiens	100	MH378688.1
691	1.343	/	No sequence	/	–
670	1.377	/	No sequence	/	–
714	1.421	/	H. sapiens	99.07	MF621085.1
550	1.438	/	No sequence	/	–
673	1.453	/	H. sapiens	99.68	MG936624.1
761–3	1.457	/	H. sapiens	99.69	MF621085.1
505	1.509	/	H. sapiens	99.37	MF621085.1
716	1.681	/	H. sapiens	99.69	MH378688.1
695	1.901	Crushed abdomen of Aedes albopictus containing human blood	H. sapiens	100	MH378688.1
663	1.929	Crushed abdomen of Ae. albopictus containing human blood	H. sapiens	99.84	MH161386.1
717	1.953	Crushed abdomen of Ae. albopictus containing human blood	H. sapiens	99.68	MH378688.1
761-7	2.019	Crushed abdomen of Ae. albopictus containing human blood	H. sapiens	99.84	MF696131.1
644	2.078	Crushed abdomen of Ae. albopictus containing human blood	H. sapiens	100	MG272704.1
680	2.102	Crushed abdomen of Ae. albopictus containing human blood	H. sapiens	99.37	MH378688.1
727	2.147	Crushed abdomen of Ae. albopictus containing human blood	H. sapiens	100	MF621085.1
627	2.148	Crushed abdomen of Ae. albopictus containing human blood	H. sapiens	100	MH378688.1
537	2.308	Crushed abdomen of Ae. albopictus containing human blood	H. sapiens	100	MH378688.1

MALDI-TOF MS = Matrix-assisted desorption/ionization time-of-flight mass spectrometry. Results below the threshold of reliable identification (< 1.8) are not reported.

Discussion

Matrix-assisted desorption/ionization time-of-flight mass spectrometry, a widely used tool for the identification biomolecules, is based on the acidic extraction and ionization of the proteins of an organism of interest. The extract is deposited on a steel target, covered with a MALDI matrix, and then dried at room temperature until co-crystallization. The crystallized target is then introduced into the apparatus, where the crystal is irradiated with laser pulses, performing desorption and “soft” ionization.¹⁰ These desorbed and ionized molecules are accelerated in an electric field and separated by a flight tube in the linear or reflectron mode according to their mass–load ratio until they reach a detector.³⁴ Thus, the mass/charge values and intensities, that is, the so-called mass fingerprint of a generated sample, are then compared with a database containing species reference mass fingerprints for species identification.³⁵ Matrix-assisted desorption/ionization time-of-flight mass spectrometry has revolutionized clinical microbiology by its use in the systematic identification of bacteria,^15,16,36 archaea,¹⁷ parasites, and fungi.¹⁸ Recently, it has been introduced in medical entomology as a tool for the rapid and accurate identification of arthropods, detection of the origin of their blood meal, and detection of associated microorganisms.^{23,24,26,37–41} In entomology, adjustments such as arthropod-based body selection and sample crushing protocol are required for proper MALDI-TOF MS identification of arthropods or detection of associated pathogens.^10,26,38,39

The method of conservation may also be a limiting factor of this method. Generally, arthropods are collected in the field, far from the laboratories. Therefore, they are usually stored dry with silica gel, in 70% ethanol at +4°C or −20°C for transport to laboratories.²⁵ It is easier and less expensive to transport samples preserved with silica gel or in 70% ethanol than samples preserved at −20° C, and these methods are widely used in African countries.²⁵ However, it was reported that storage for a long time in 70% alcohol may impact MALDI-TOF MS profiles, resulting in lower intensity and lower overall quality than those of fresh or frozen samples.^25,30,42 No MALDI-TOF MS study has yet been conducted on arthropods stored with silica gel despite the advantages of this method such as limited cost and simplicity.

In this study, 96% of the mosquito legs with good spectra were correctly identified with LSV ≥ 1.8 by the MALDI-TOF MS. This confirms the reliability of this method for mosquito identification, as long as the spectra are of good quality. This allowed us to confirm the morphological identification of the female Anopheles spp. up to the species level and of the Culex spp. up to the genus level. It helped us to identify not only male Anopheles spp. but also minority species such as Cx. rima, Cx. perexiguus, and Cx. watti which were not included in our database before this work. The results of this study can be considered robust and reliable as they have been confirmed by molecular biology.

The two species of Anopheles identified in this study, that is, An. gambiae s.l. and An. rufipes, have already been reported in Chad.⁴³ Anopheles gambiae s.l. were the most abundant mosquitoes included in our study and considered the main vector of malaria in Chad.⁴³ Anopheles rufipes is a mosquito that rests frequently in human habitation but feeds on domestic animals and accidentally on humans.²⁹ In this study, we have also identified several Culex species such as Cx. quinquefasciatus, Culex pipiens, Cx. perexiguus, Cx. watti, and Cx. rima. Many of these species have been implicated or suspected in the transmission of parasitic or viral pathogens. The Cx. pipiens species consists of two morphologically identical subspecies, that is, Cx. pipiens pipiens and Cx. pipiens molestus, with distinct trophic preferences: Cx. pipiens pipiens feeds on birds, whereas Cx. pipiens molestus prefers mammals.^44,45 Culex pipiens has been identified as the most important vector species of WNV in the United States because of their vectorial competence and summer abundance.⁴⁶ Presently, only one study has reported the presence of Cx. pipiens in Chad.⁴⁷ Culex quinquefasciatus, known vector of lymphatic filariasis and arboviruses including WNV, has been reported in some African countries.^48–50 Culex perexiguus, considered a potential vector for WNV transmission in birds and horses, has recently been reported in Mali.²⁴ Culex watti has been reported in Madagascar and is believed to have played a role in the WNV transmission.⁵¹ As for Cx. rima, its presence has already been reported in some African countries,^52–54 but its vector role remains unknown to this day. We found that the fragment of the COI gene amplified in this study could not distinguish An. rufipes and An. rhodesiensis. This limitation of molecular biology had already been reported.⁵⁵ Therefore, MALDI-TOF MS could be an alternative tool to meet this challenge because this tool has been proven relevant to discriminate cryptic mosquito species.³⁰

Only 40.7% of abdomens engorged had good quality MS spectra and 37.5% were correctly identified as human blood. Niare et al.⁵⁶ had shown that identification of the origin of the blood meal was relevant up to 24 hours because of blood digestion altering the resulting spectrum. We support the idea that this low percentage of good spectra could be explained by the fact that the Sella score to describe the digestive state of blood, from engorged mosquitoes from zero (non-nourished mosquitoes) to seven (females without visible blood and fully developed eggs in their abdomen), was poorly appreciated visually.¹⁴

Conclusion

The present study demonstrated that MALDI-TOF MS appears to be a promising tool for identifying mosquitoes stored in silica gel and moderately the origin of their blood meal. Although the number of samples used in our study is relatively small, the results obtained are robust and reliable. Thus, it opens the way for future studies with a large number of samples to confirm these preliminary results.