Paraphyly of the Subgenus Sintonius (Diptera, Psychodidae, Sergentomyia): Status of the Malagasy Species. Creation of a New Subgenus and Description of a New Species

During an inventory of Phlebotomine sand flies carried out in Madagascar, we have identified some specimens showing morphological characters related to the subgenus Sintonius of the genus Sergentomyia. We started a molecular study based on cytochrome b mtDNA and on D1–D2 and D8 domains of the rDNA. The sampling includes all the Sergentomyia species available and also S. (Sergentomyia) schwetzi, S. (Parrotomyia) magna, and the following species belonging to the subgenus Sintonius: S. clydei, S. christophersi, S. affinis vorax, S. adleri and S. meilloni. The Sintonius subgenus (sensu Theodor) is paraphyletic. The Malagasy specimens morphologically Sintonius-like are never clustered with the continental Sintonius. We propose a new subgenus to include them: Trouilletomyia subg. nov. Due to the lack of mesanepisternal setae, the species huberti is removed from the genus Phlebotomus and we propose here a new combination: Sergentomyia huberti comb. nov. The male of S. huberti is pinpointed and described for the first time. Lastly, a new species for Science is described on one female: Sergentomyia (Trouilletomyia) boironis n. sp.


Introduction
The subgenus Sintonius of the genus Sergentomyia was created in 1931 [1] in an article about a questionable classification of the Phlebotomine sandflies opposed to that previously proposed [2]. Curiously, P. (Euphlebotomus) philippinensis, and the American species Lutzomyia gomezi and Nyssomyia intermedia were included in this group [1]. Nitzulescu's classification has not obtained the approval of any subsequent authors [3][4][5][6][7] who have considered Sintonius as a subgenus of the genus Sergentomyia França and Parrot, 1920, including the species exhibiting a cibarial armature and annealed spermathecae (designed as ''spermathèques crénelées'' by Nitzulescu). Later, Theodor considered this subgenus as an artificial group that does not sufficiently take into account the structures of the pharynx and particularly of the male genitalia. Consequently, he provided a new definition of this group [6,7]: scanty erect hairs on abdominal tergites, style with four spines (either all terminal or two terminal and two subterminal), hooked parameres, pointed aedeagus, segmented spermathecae, buccal cavity of varying forms, pharynx lampglass shaped with a few teeth posteriorly or with ridges only.
During the last decade, we carried out several sandfly inventories in many parts of Madagascar. They included several specimens of males and females sharing the characters defining the subgenus Sintonius. Taking into account the high level of endemism in Madagascar [8] especially within the phlebotomine sandflies [9][10][11][12][13][14][15][16][17] before including these specimens in the subgenus Sintonius, we carried out a study based on two ribosomal and one mitochondrial molecular markers. This study is not a phylogenetic analysis of the subgenus Sintonius due to a limited sampling, nor a phylogeny of the genus Sergentomyia. However it demonstrates that the Sintonius subgenus (sensu Theodor) is paraphyletic. It also permitted new insights in the Malagasy sand flies with i) the creation of a new subgenus in the genus Sergentomyia, ii) the identification of a new species and iii) the correction of the position of one Malagasy species wrongly placed in the genus Phlebotomus.

Ethics Statement
For insect collections, we obtained a license for collecting and transporting zoological material Nu 154/10/MEF/SG/DGF/ DCB.SAP/SLRSE. No endangered or protected species were collected in this study.

Sand Fly Sampling
In total, the molecular sampling encompasses 41 specimens of Sergentomyia from eight countries (Table 1). It includes seven species of Sintonius sensu Theodor, the Sergentomyia known from Madagascar and some other African species. They were collected using CDC miniature light traps (John W. Hock company, Gainesville, FL), ultraviolet miniature light traps, sticky traps, or Malaise traps. The traps usually run overnight from 5 p.m. to 8 a.m. the following morning.
Moreover, several other specimens not processed for molecular biology and the P. huberti holotype and paratype have been examined.

Malagasy Study Sites
Captures have been carried out in the west and the south of Madagascar, which are subject to the trade winds with significant differences in rainfall and temperature explaining differences about climates.
The four prospected localities where sandflies have been processed in the present study are detailed below and on figure 1.
Anjohikely cave. This cave belongs to the Anjohibe caves complex, in the most southern part of the sandy tray Mahamavo, in the North-Western part of Madagascar at 80 km north of Mahajanga (15u33.79S, 46u52.59E, altitude: 100 m a.s.l.). The vegetation cover consists of ''Mokoty'' savannah with some patches of preserved forest on the limestone, building lapiaz. Sand flies were captured using a CDC light trap.
Anjohikinakina cave. It is located in the national park of Bemaraha, at 15.5 km north of Bekopaka in the western part of Madagascar (19u0935.6400S, 44u4693.7200E, altitude: 130 m a.s.l). The forest has dry deciduous type of limestone karst soil. The high plain of Bemaraha is a karst formation (Tsingy), located inside the reserve training pinnacles whose access is extremely difficult. Further north, can be found rolling hills interspersed with limestone rock formations. The river Manambolo limits the reserve to the south. Sand flies were captured using a CDC light trap.
Namoroka. This Special Reserve is located in the North-Western part of Madagascar in a dry deciduous forest on limestone karst ground. It is located within a large region from the Middle Jurassic limestone, similar to the Bemaraha's Tsingy reserve located 250 km to the south. The limestone plateau that occupies the greater part of the reserve is in the form of a lapiaz. The south part of the Reserve is characterized by the presence of many sinkholes, some of which contain permanent pools. Sand flies have been collected using a Malaise trap near the Cave of Ambovonomby (16u28.29S, 45u20.99E, altitude: 200 m a.s.l).
Isalo. This National Park is located in the southern part of the centre of Madagascar. It is composed of Jurassic continental sandstone. The south and east of the massif consists of sandstone layers whose elements are size and resistance varies according to erosion. Sand flies were captured using a Malaise trap (22u37.609S-45u21.499E, altitude: 822 m a.s.l.).

Morphological Analysis
The sand flies collected 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 microscope slide and cover slide for species identification directly in chloral gum or after dehydration in Canada balsam. To allow long-term preservation of the specimens previously mounted in chloral gum, they were remounted in Canada balsam after complete processing of washing and dehydration. The body related to the specimen was dried and stored in a vial at 220uC before DNA extraction. The specimens were observed under a BX50 microscope and measured using the Perfect Image software (Aries Company, Chatillon, France) and a video camera connected to the microscope. Drawings were made using the camera lucida.

Molecular Analysis
Genomic DNA was extracted from the thorax, wings, legs and abdomen of individual sand flies using the QIAmp DNA Mini Kit (Qiagen, Germany) following the manufacturer's instructions, modified by crushing the sand fly tissues with a piston pellet (Treff, Switzerland), and using an elution volume of 50 to 200 ml [9].
All the mtDNA and rDNA amplifications were performed in a 50 ml volume using 5 ml of extracted DNA solution and 50 pmol of each of the primers. The PCR mix contained (final concentrations) 10 mM Tris HCl (pH 8.3), 1.5 mM MgCl 2 , 50 mM KCl, 0.01% Triton X 100, 200 mM dNTP each base, and 1.25 units of 5 prime Taq polymerase (Eppendorf, Germany). The cycle begins with an initial denaturation step at 94uC for 3 min and finishes with a final extension at 68uC for 10 min. PCRs were done with the following temperature profiles. Amplicons were analyzed 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 was done using Pregap and Gap softwares included in the Staden Package [20].
We used three different data sets for phylogenetic analyses: the rDNAs D1-D2 and D8, and the mtDNA cytochrome b.
Consensus sequences were aligned by the Clustal W algorithm [21] from the BioEdit 4.8.10 sequence editor [22], and corrected manually.
Neighbor-Joining (NJ). The NJ method [26] has been analysed by MEGA software version 5 [27]. Genetic distances were corrected according to the transition/transversion rate (Kimura's two-parameter method). Bootstrap confidence values were calculated from 1,000 replications.
Maximum parsimony. Maximum parsimony (MP) analysis were performed using the branch and bound option of MEGA when possible or the heuristic search. The node support was assessed by bootstrapping over 100 replications.
Maximum likelihood. Sequence data were analysed by PhyML [28] based on maximum likelihood. The ML trees were constructed using the substitution models selected by MODEL-

Nomenclatural Acts
The electronic edition of this article conforms to the requirements of the amended International Code of Zoological Nomenclature, and hence the new names contained herein are available under that Code from the electronic edition of this article. This published work and the nomenclatural acts it contains have been registered in ZooBank, the online registration system for the ICZN. The ZooBank LSIDs (Life Science Identifiers) can be resolved and the associated information viewed through any standard web browser by appending the LSID to the prefix ''http://zoobank.org/''. The LSID for this publication is: urn:lsid:zoobank.org:pub:D0300E46-E93C-4CEF-99DE-77D77E5FC442. The electronic edition of this work was published in a journal with an ISSN, and has been archived and is available from the following digital repositories PubMed Central, LOCKSS.

Molecular Analysis
The sequences analysed in the present study have been deposited in Genbank as indicated in table 1. Despite several attempts, the direct sequencing of the S. clydei was not successfully performed for D8 and D1-D2 from the Seychelles specimens, like the sequencing of several markers of the specimens 1, 30 and 33, due to the small quantity of DNA extracts.
A high degree of homology is observed between the topologies obtained by ML and NJ analyses. Consequently, the NJ trees are not shown. The ML and MP trees related to the analysis of the sequences of D1-D2, D8 and cyt b are shown on figures 2, 3, and 4, respectively.
The females from Anjohikinakina cave and Bemaraha (holotype and topotypes) are morphologically similar to those from Anjohikely cave and from Namoroka. Sequences of males and females from these localities show 100% homology regarding D8 marker, a few differences for D1-D2 and cyt b sequences. Regarding the latter, differences are observed between localities, except for the specimens caught in the Bemaraha, divided in two populations. Consequently, we consider that these specimens belong to the same species. The 100% homology between the cyt b sequences of the male specimen number 98 and the female specimen number 1 (P. huberti holotype) allows us to describe the male of this species (see below).
The pairwise distance between the female specimen 146 from Isalo and the S. huberti populations (table 2) is high and is phylogeneticaly isolated (figs 2, 3 and 4). Moreover, this specimen 146 is morphologically very different from the females from Anjohikely, Namoroka and Bemaraha. All these data support the creation of a new species (see below).

Morphological Analysis
The morphological examination of the Malagasy sandflies identified as Phlebotomus huberti (topotypes caught in 2012 and specimens from Namoroka) as well as the reexamination of the female holotype and the paratypes show a lack of mesanepisternal setae, despite the unexplained presence of a group of four Table 1. Cont. mesanepisternal setae in the original description [10]. Consequently, the species huberti cannot belong to the genus Phlebotomus. We here propose a new combination: Sergentomyia huberti comb. nov. based on the existence of two synapomorphies i.e. the absence of pre-apical papilla on the fifth antennal segment and presence of an opened labial furca.

Linking Molecular and Morphological Analysis
The existence of annealed spermathecae and of small teeth on the cibarium is in agreement with the characters of inclusion in the subgenus Sintonius [1,6,7]. However, the molecular data do not support the inclusion of S. huberti comb. nov. in the subgenus Sintonius. The specimen 146 belongs to the sister species of S. huberti. Consequently, we propose a new subgenus including S. huberti comb. nov. and S. boironis n. sp.: Trouilletomyia subg. nov.
Description of Trouilletomyia subg. nov., depaquit and léger According to the monophyly of the Malagasy species Sintoniuslike, we have created for them the Trouilletomyia subg. nov. in the genus Sergentomyia, that we define by i) annealed spermathecae, ii) an armed cibarium in both sexes, iii) a remarkable pharyngeal armature with two types of teeth, like that observed in the subgenera Adlerius, Euphlebotomus or Anaphlebotomus pro parte (Asiatic species) of the genus Phlebotomus. To our knowledge, this pharyngeal armature had never been observed in the genus Sergentomyia.  The description of female [10]   The holotype has been deposited in the department of entomology of the Muséum National d'Histoire Naturelle, Paris.
The male remains unknown.

Derivatio Nominum
The subgenus Trouilletomyia subg. nov. is dedicated to our colleague Jean Trouillet. The species Sergentomyia boironis n. sp. is dedicated to our colleague Pascal Boireau.
To meet the criteria of availability, the authors Randrianambinintsoa & Depaquit are responsible of the name Sergentomyia boironis n. sp. and the authors Depaquit, Léger & Randrianambinintsoa are responsible of the name Trouilletomyia subg. nov. and should be cited as the sole authority of these taxa, according to the Article 50(1) of the lnternational Code of Zoological Nomenclature, 4th edition, 2000.

Discussion
Within the Phlebotominae of the Old World, the genus Sergentomyia França & Parrot, 1920 appears to be a catch fall group, including all the Old World species excluded from all other genera (Phlebotomus, Idiophlebotomus, Chinius, Spelaeophlebotomus, Grassomyia, Parvidens, Spelaeomyia and Demeillonius) [24,25,32]. Species of the genus Sergentomyia share the following characters: a mesanepisternum without setae, abdominal tergites 2-6 carrying usually all or most recumbent hairs, an usual 1/III-XV antennal formula in the males and 2/III-XV in the females with some exceptions, a cibarium with an armature of teeth and/or denticled more developed in females than in males (beyond exceptions), a single paramere, a style with four terminal spines (or often 2 terminal and 2 subterminal) and an accessory spine.
The genus Sergentomyia is regularly mentioned as probable vector of leishmaniases [33,34] and arboviruses [35,36], it is important that the systematics of this group is well assessed.Currently, mainly based on the spermathecal morphology, the genus Sergentomyia is divided in seven subgenera: Sergentomyia The species S. huberti and S. boironis n. sp. show the characters of inclusion in the subgenus Sintonius [1,6,7]: i) scanty erect hairs on the abdominal segments II to VI, ii) annealed spermathecae, iii) a pointed aedeagus and iv) a style with two terminal and two subterminal spines. However, the markers and molecular analyses show that S. huberti and S. boironis n. sp. never cluster with the Sintonius included in the present study. Moreover, the latter group is paraphyletic. Regarding the different analyses, it appears the consistency index of cyt b analysis is low (0.41) whereas those calculated from the ribosomal markers D1-D2 and D8 are higher (0.75 and 0.72), respectively. Consequently, the ribosomal markers are more reliable than cyt b. This mitochondrial marker includes many homoplasic characters and the cyt b trees could be considered as doubtful. According to D1-D2 (   is proven but a more extensive study comparing morphological and molecular analyses is needed in order to revise the group. In our opinion, the main morphological traits characterising the subgenus Sintonius could be symplesiomorphies. For example, many species of Phlebotomus or American sandflies have annealed spermathecae. These data did not allow us to include S. huberti and S. boironis n.sp. in the subgenus Sintonius. The specific value of S. boironis n. sp. is supported by both morphological and molecular data. Its cibarial armature is very different from that of S. huberti and we note the head of the spermathecae is rounded and included in the most distal ring, differing from that of S. huberti and Sintonius. Moreover, phylogenetical analyses show S. boironis n. sp. is the sister species of S. huberti. The pairwise distances are high between S. boironis n. sp and S. huberti: 9.5 to 12.8% for cyt b and 12 to 19% for D1-D2 (table 2). The low values (0.3 to 0.5%) observed for D8 are explained by the MEGA algorithm not taking into account the positions including indels, which support the variability between S. boironis n. sp. and S. huberti. We observe important molecular cyt b variability between the different populations of S. huberti comb. nov. However there is not enough material available from different populations and genders in order to assess if they belong to cryptic species or not. The observation of specimens from Namoroka shows little differences in females (figure 7) and males (figure 8) especially regarding the higher number of cibarial teeth. However, these observations do not seem significant enough at the present time to justify the creation of a new species. The morphometric descriptive statistics (table 3) do not emphasise significant differences between these populations. Despite the existence of two sympatric mitochondrial populations in the Anjohikinakina cave-Bemaraha (specimens 1 and 98 on the one hand, and specimens 605, 606, 612 and 618 on the other hand) will encourage new investigations related to this species.