First record of Phlebotomine sandflies (Diptera: Psychodidae) in the Comoros Archipelago with description of Sergentomyia (Vattieromyia) pessoni n. sp. and S. (rondanomyia) goodmani comorensis n. ssp.

No Phlebotomine sandflies had ever been reported in the Comoros Archipelago, including the three islands of the Republic of the Union of Comoros (Grande Comore, Mohéli and Anjouan) and the French oversea department of Mayotte. During three field surveys carried out in 2003, 2007 and 2011, we provided the first record of Phlebotomine sandflies in this area. A total of 85 specimens belonging to three species were caught: a new species S. (Vattieromyia) pessoni n. sp. (two females from Grande Comore), a new subspecies of Sergentomyia (Rondanomyia) goodmani (80 specimens from Grande Comore and one from Anjouan) and Grassomyia sp. (two females from Mohéli). The individualisation of these taxa was inferred both from morphological criteria and sequencing of a part of the cytochrome b of the mitochondrial DNA. These taxa are closely related to Malagasy sandflies.


INTRODUCTION
T o our knowledge, no record of Phlebotomine sandfly exists in the literature and there was no case of leishmaniasis in the Comoros Archipelago. The latter includes the Islands of the Union of the Comoros (Grande Comore, Anjouan and Mohéli) and Mayotte (a French oversea department). In the neighbouring countries, little is known about Phlebotomine sandflies from South-eastern Africa (Abonnenc, 1972;Artemiev, 1985;Davidson, 1987) and from Madagascar (Depaquit et al., , 2004a(Depaquit et al., & b, 2007(Depaquit et al., , 2008Léger et al., 2005). We performed three field works in order to inventory the four islands. A total of 85 adult specimens have been captured. All the specimens have been examined morphologically and some of them have been processed for molecular biology. All of them belong to the genera Sergentomyia and Grassomyia. Two new taxa for Science are described in the present study.

MATERIAL AND METHODS
P hlebotomine sandflies have been caught using CDC miniature light traps and UV miniature light traps in the four main islands of the Comoros Archipelago: Grande Comore, Mohéli, Anjouan and Mayotte (Fig. 1, Table I). The archipelago of the Comoros is located at the northern end of the Mozambique Channel in the Indian Ocean, halfway between the coasts of Madagascar and the African continent. The volcanic chain of the Comoros appears to be a "hot spot trace". The Comoros are made up of volcanic rocks, primarily undersaturated alkali olivine basalts. Phonolitic and small volumes of trachytic lavas have also been Original contribution Parasite, 2012, 19, 195-206 reported (Pavlovsky & de Saint-Ours, 1953;Flower, 1973;Emerick & Duncan, 1982). The volcanic rocks have been differentiated into three volcanic phases, in which basaltic lavas prevail. Scorias and puzzolanic tuffs have been reported from Grande Comore, Mohéli and Anjouan. Phonolitic and trachytic rocks occur on Mayotte (Pavlovsky & de Saint-Ours, 1953). On the geo-chronological plan, the latest information on the age of the three islands is respectively as follows: 1.49, 0.48, 0.36 millions years (My) for Mayotte, Mohéli and Anjouan (Armstrong, 1972: Emerick & Duncan, 1982, 1983: Nougier et al., 1986 and 0.13 My for Grande Comore (Emerick & Duncan, 1982, 1983 Depaquit et al. (2008) and those of S. goodmani have been specially processed in the present study. Sandflies were stored in 96 % ethanol. The head and genitalia were cut off in a drop of ethanol, cleared in boiling Marc-André solution, and mounted between slide and cover slide for species identification. The body related to the specimen was stored dried in a vial at -20 °C before DNA extraction. All the specimens were observed utilizing  a BX50 microscope and measured using the Perfect Image software (Aries Company, Chatillon, France) and a video camera connected to the microscope. Drawings have been done using the camera lucida installed on the microscope. Some specimens have been remounted in Canada balsam after complete processing of washing, dehydration in baths of ethanol 70 to 100, then beech creosote. Genomic DNA was extracted from the thorax, wings, legs and abdomen of individual sandflies using the QIAmp DNA Mini Kit (Qiagen, Germany) following the manufacturer's instructions, modified by crushing the sandfly tissues with a piston pellet (Treff, Switzerland), and using an elution volume of 200 μl, as detailed in Depaquit et al. (2004a). following Esseghir et al. (1997): five cycles of (denaturation at 94 °C for 30 s, annealing at 40 °C for 60 s and extension at 68 °C for 60 s), followed by 35 cycles of (denaturation at 94 °C for 60 s, annealing at 44 °C for 60 s and extension at 68 °C for 60 s). Amplicons were analysed by electrophoresis in 1.5 % agarose gel containing ethidium bromide. Direct sequencing in both directions was performed using the primers used for DNA amplification. The correction of sequences is done using Pregap and Gap softwares included in the Staden Package (Bonfield & Staden, 1996). Molecular analyses are based on the sequence alignment performed using the ClustalW routine included in the bioedit version 5 software (Hall, 1999) and checked by eye. According to the objective of this study, which is not a phylogenetical analysis, a Neighbor-Joining (NJ) analysis was performed using MEGA 5 software (Tamura et al., 2011), with the Kimura-2 parameter model and using uniform rates among sites.

Genitalia
Closely resembling that of the specimens of Madagascar. Coxite long of 127-149 μm, exhibiting on its inner face a group of four to six disseminated setae. Sharp style, 56 to 72 μm long with four terminal spines and an accessory seta located at the level of the distal third or quarter. Simple paramere with a hooked extremity. Aedeagus 58 to 79 μm long, straight, sharped and sometimes pointed at the top. Genital filaments 331 to 409 μm long. Genital pump 81 to 112 μm long.  Parasite, 2012, 19, 195-206 Genital filaments/genital pump = 3.73 to 4.68.

DERIVATIO NOMINUM
S. pessoni is dedicated to our colleague Bernard Pesson.
The subspecies comorensis refers to the geographical origin of the studied population.
The holotype and the paratype of S. pessoni, and the holotype and nine paratypes (including four allotypes) of S. goodmani comorensis have been deposited in the Muséum national d'Histoire naturelle of Paris.

MOLECULAR RESULTS
The sequences studied were 497 to 503, depending on species and geographical origin (Table I). An alignment made by ClustalW included in the Bioedit package, and checked by eye, includes 507 positions.
An analysis of the variable sites within the Vattieromyia shows 68 variable sites (Fig. 5). An analysis of the variable sites within the Rondanomyia shows 51 variable sites (Fig. 6). The pairwise distances between and within groups are showed on Table III. The NJ tree based on cytochrome B sequences is given in Fig. 7. Within the Vattieromyia, the species Sergentomyia sclerosiphon, S. namo and S. anka are as strongly supported (bootstrap value: 100 %) as the two specimens caught in the Comoros islands. Concerning the Rondanomyia, we observe a dichotomy between Malagasy and Comorian specimens. The Malagasy specimens are supported by 100 % bootstrap whereas Comorian specimens are supported by 88 % bootstrap. The specimen from Anjouan shows a haplotype differing by many nucleotides from the haplotypes of specimens caught in Grande Comore (Fig. 6). An additional NJ tree (data not shown) carried out without this specimen shows an increasing bootstrap value (96 %).   (Phlebotomus, Idiophlebotomus, Chinius, Spelaeophlebotomus, Grassomyia, Parvidens, Spelaeomyia and Demeillonius). Concerning the Grassomyia, we refer to the position of Abonnenc & Léger (1976) considering this group as a genus. The genus Sergentomyia is presently divided without any phylogenetical argument into six subgenera: Sergentomyia França & Parrot, 1920, Parrotomyia Theodor, 1948, Rondanomyia Theodor, 1948, Sintonius Nitzulescu, 1931, Capensomyia Davidson, 1978and Vattieromyia Depaquit, Léger & Robert 2007(Duckhouse & Lewis, 1980Depaquit, Léger & Robert, 2007). The males being difficult to identify, this classification is mainly based on the morphology of the female spermathecae. Moreover, some species are still ungrouped, due to their atypical spermathecae.
In 2007, Depaquit et al. defined the subgenus Vattieromyia by in the female i) the shape of the spermathecae and the sclerotized parts of their ducts, ii) the cibarial armature palisade-like, iii) an unusual antennal formula without ascoid on AIII. The males also have this original antennal formula shared only by Grassomyia spp. and S. majungaensis. At the light of S. pessoni n. sp., it appears that this species belong to the subgenus Vattieromyia according to the first two characters defining this group. This very original spermathecae has a more important weight than the original antennal formula. Consequently, we now exclude the antennal formula without ascoid on AIII from the definition of the subgenus Vattieromyia.
The morphological differences observed between the three Malagasy species belonging to the subgenus Vattieromyia and the two Comorian females appear to be of specific level. The females of S. pessoni n. sp. differ from all the Malagasy species by the presence of ascoids on the third antennal segment. Moreover, they are different from S. sclerosiphon by the number of cibarial teeth, from S. namo by the lack of cibarial ring and from S. anka by the shape of the cibarial armature and by the number of denticle ranges. The individualization of S. pessoni n. sp. is reinforced by the molecular data, also individualizing this new species.
Concerning S. goodmani, a strong molecular dichotomy is observed between the Malagasy specimens and the Comorian ones. However, there is little morphological divergence between Comorian and Malagasy specimens. This suggests an occurring spe-ciation event linked with the insularity rather than a speciation fully achieved. Consequently, we consider S. goodmani comorensis n. ssp. as a subspecies of S. goodmani and not as a new species because both their molecular and morphological characters are more discreet and, linked to insularity, perfectly correspond to the concept of subspecies sensu Mayr et al. (1953). Although we are unable to test the interfecondity which could exist between Malagasy and Comorian populations. To individualize this subspecies, we consider i) in females the smaller number of cibarial teeth (15 to 21) and some morphometrical characters of the wing and the antennae and ii) in males, mainly the very small number of coxal setae in the Comorian specimens (four to six without real tuft versus about 15 in specimens from Madagascar). We can't identify the Grassomyia females caught during the present study at a specific level. The two specimens have been mounted for molecular biology processing (Table III) and consequently, their thoraces have been crushed for DNA extraction. The main characters used for the identification of the species of the genus Grassomyia are the absence/presence (and number) of setae on the mesanepimerum and the absence/presence of spines on the anterior and medium femurs (G. dreyfussi Parrot). According to the well developed pharyngeal armature of our specimens from Mohéli, their number of cibarial teeth (34 and 39), and to the characters revised by Abonnenc (1969Abonnenc ( , 1972, it cannot belong to G. inermis Theodor, 1938 (lack of pharyngeal armature, 19 to 25 cibarial teeth), or to G. squamipleuris Newstead, 1912 (48 to 55 cibarial teeth). Even though we have not observed the femurs of the specimens from Mohéli, the latter cannot be G. dreyfussi Parrot, 1933 (42 to 55 cibarial teeth and femurs with spines). Consequently, the Comorian specimens could be G. madagascarensis Abonnenc, 1969(36 to 43 cibarial teeth), or G. ghesquierei Parrot, 1929 (27 to 37 cibarial teeth, lack of setae on the mesanepimerum). A complete revision of this group should be carried out in the future. Nothing is known about the vectorial competence of the three species found in Comoros islands. Although no leishmaniasis cases have been reported, further studies are needed to evaluate the risk of autochtonous transmission of Leishmania. Original contribution Parasite, 2012, 19, 195-206