Green bottle flies (Calliphoridae, Luciliinae) of Ecuador: geographic distribution, checklist and DNA barcodes

ABSTRACT Green bottle flies (Diptera, Calliphoridae, Luciliinae) comprise a diverse and cosmopolitan taxon, known from at least 1,500 species. They have become crucial elements in forensic investigations, as they spend part of their life cycle in decaying remains. Here, we review the distribution of eleven Luciliinae species in Ecuador: the monotypic Blepharicnema and ten Lucilia species. We identified specimens using morphological characters. Additionally, we DNA barcoded 43 specimens from three species using 658bp segments of the standard Cytochrome Oxidase I (COI) mitochondrial gen. Molecular and morphological identifications presented high correspondence, suggesting COI barcodes are an efficient tool for the identification of these three green bottle flies species. Geographical records are biased towards the northern Andean region, particularly near to large urban settlements. We remark the value to applied forensic research of continuous sampling of necrophagous flies under a variety of habitats and crime conditions.

Forensic entomology provides an emerging tool that combines biological and ecological knowledge of carrion species [4,[7][8][9]15]. For example, the time since colonization by necrophagous flies of dead bodies is used to interpret forensic evidence in criminal cases [7][8][9]. Thus, correct identification of target fly specimens is an essential early step [15][16][17] in legal investigations. However, little research has been done on forensically important flies in Ecuador. Here, we aim to assess the diversity of Luciliinae species associated with carrion by presenting an updated checklist and distribution maps along continental and insular Ecuador. Additionally, we provide DNA COI barcodes for three of these species. Our broader aim is to contribute with geographical data of taxonomically confirmed voucher specimens to the practical use of forensic entomology and faunal inventories in Ecuador.

Checklist
We include in this checklist specimen information from field trip collections, museum specimens, and a review of literature. Field trips were done from 2013 to 2016, covering disturbed and natural habitats at eight sites (ranging from 200 to 2800 m asl) throughout the Andean highlands and the Amazon region of Ecuador. We used modified McPhail, Van Someren- Rydon, and pitfall traps with decaying baits (fish heads, shrimps, meat, chicken viscera), as well as a pig (Sus scrofa) and guinea pig (Cavia porcellus) carcasses, to attract important forensic arthropods. In the laboratory, we sorted and identified approximately 3200 specimens and reared larvae to mature stages when possible. We dissected genitalia for both sexes to verify the identity of those species represented by ten individuals or more (males and females when possible

DNA COI barcoding
To aid in species identification, we barcoded 43 specimens previously identified with morphological characters as Lucilia purpurascens, L. sericata and L. eximia. We amplified and sequenced samples in collaboration with Canadian Centre for DNA Barcoding (CCDB) following the standard protocols of the Biodiversity Institute of Ontario, Guelph University [18], using C_LepFolF and C_LepFolR primers. Sequences are available in GenBank under the accession numbers MF458318 to MF458358, and in the BOLD (www.bold systems.org/) database under the public project FFECU -Forensic Flies of Ecuador. After general verification of congruence between morphological and molecular taxonomic units, each species was given a Barcode Index Number (BIN) by BOLD. BINs can be used as a proxy taxonomic unit [17] and provides a permanent and objective molecular reference for further studies.

Checklist
Examined specimens belong to eleven Luciliinae species (Table 1) occurring along 53 localities from 14 provinces in continental and insular Ecuador ( Figures  2-6). Visual inspection of distribution maps showed the northern Andean highlands as the most sampled area within the country.

DNA COI barcoding
We obtained 43 complete COI sequences from three Lucilia species from Ecuador. Molecular clustering of these 43 specimens within previously barcoded species for the country (sequences available in Genbank) showed high taxonomic concordance ( Figure 7). Molecular clustering of these 43 specimens within a total of 2416 public sequences from 62 putative species is presented in Appendix 1. A BIN analyses of the World dataset suggest that our putative species name for the L. sericata and

Comments
This species is possibly the largest blow fly in the Neotropics [19]. Although most of the natural history of this fly is still unknown, previous studies show that they are strongly attracted to fish bait [19]. Reported for the first time in the Ecuadorian Andes from a specimen collected at Pichincha, Quito by James (1970). Also, B. splendens was reported for Ecuador by [10,[19][20][21]. The material examined from the QCAZ Museum belongs to Imbabura, Zamora and Morona Santiago Provinces.

Distribution
B. splendens is a Neotropical species endemic to the Andes of Peru, Colombia, Bolivia, Ecuador and Venezuela [19].

Comments
Reported for the first time in Ecuador by Whitworth [11] based on specimens collected in Napo and Orellana Provinces. Paratypes are deposited in the LAMC, CNC and UGG Museums. Males are very difficult to collect considering only 3% of specimens obtained from carrion bait traps in French Guiana were L. albofusca in [11].

Comments
First record reported in Isla Española, Galapagos in 1899. Reported for Ecuador by [10,23,24]. No specimens of this species were collected or reviewed.

Locality records
Ecuador ( Individuals of L. eximia can develop on rotten vegetables and fruits, and are known as a secondary myiasis producer [28]. According to COI data in previous phylogenetic studies, L. eximia might be a species complex [11].

Comments
Reported for the first time in Ecuador by Whitworth [11] based on material housed on the CNC Museum collected at Nayon in 2013. The material examined from the QCAZ Museum belonged to Pichincha and Napo Provinces. It is a relatively common Andean fly, attracted by cow and dog feces and by fish and liver bait. Their role as pollinators is suggested due to their common visitation frequency to flowers [26].

Distribution
The distribution of L. ibis is only known from the Andes highlands in Peru, Bolivia, Ecuador and Argentina [11].
In Ecuador this species is reported in the Andean and Amazonian region from 1900-2900 m asl (this study).

Comments
Reported in Ecuador for the first time in 2014 in Napo province during this study.

Locality records
Ecuador (

Comments
Described by Walker based on a male specimen collected by Charles Darwin in 1835. Reported for Ecuador by [10,11,22,25] based on material collected in Pinta, Fernandina, Genovesa, Española and Santa Cruz Islands. There is no material deposited in the QCAZ Museum for this species.

Locality records
Ecuador (

Comments
Reported from Ecuador by Whitworth [11] based on material housed at CNC Museum collection collected in Guayas and Santo Domingo Provinces in 1963. There is no material deposited in the QCAZ Museum for this species.

Distribution
This species is known from Colombia, Costa Rica, Ecuador, Honduras and Panama [11].

Locality records
Ecuador (

Discussion
We provide evidence for the presence of eleven Luciliinae species in Ecuador: the monotypic Blepharicnema and ten species of Lucilia, from which three species, L. deceptor, L. setosa, and L. piona are endemic to the Galapagos Archipelago, and L. nitida is   The evolutionary history was inferred by using the Maximum Likelihood method and Kimura 2-parameter model [43]. The tree with the highest log likelihood (−1517.85) is shown. The percentage of trees in which the associated taxa clustered together is shown next to the branches. Initial tree(s) for the heuristic search were obtained automatically by applying Neighbor-Join and BioNJ algorithms to a matrix of pairwise distances estimated using the Maximum Composite Likelihood (MCL) approach, and then selecting the topology with superior log likelihood value. A discrete Gamma distribution was used to model evolutionary rate differences among sites (5 categories (+G, parameter = 0.4209)). The rate variation model allowed for some sites to be evolutionarily invariable ([+I], 76.10% sites). The tree is drawn to scale, with branch lengths measured in the number of substitutions per site. This analysis involved 55 nucleotide sequences. Codon positions included were 1st+2nd+3rd+Noncoding. There were a total of 654 positions in the final dataset. Evolutionary analyses were conducted in MEGA X (44)(45).
reported for the first time in continental Ecuador. Previous studies have reported that L. cuprina [11,[27][28][29][30] and L. vulgata [11,28] (both widespread species) may be present in Ecuador according to their current distribution; however, we were not able to review any specimens for Ecuador. Aguirre [21] reported the presence of Lucilia elongata (Shannon 1924) in Ecuador, a species that was later on included in Salazar and Donoso [20] catalogue of species of forensic value for Ecuador. We exclude L. elongata from the present checklist because there are no previous reports in the region for this species [10,11] and it was likely a misidentification by Aguirre [21]. To this date, Ecuadorian Calliphoridae have not been appropriately sampled, hence the diversity and distribution of Luciliinae will likely be found to be much wider than reported here. In this study, the geographical records are biased to the Andean region (particularly near to large urban settlements), while other regions in the southern Andes, Coastal region, Amazon Forest, and the Galapagos Archipelago are poorly represented ( Figure 2). This is a pattern commonly found in Ecuadorian invertebrates [31,32]. For this reason, multiregional sampling in urban (eusynanthropic), seminatural (hemysinanthropic), and natural (asynanthropic) habitats including Ecuadorian highlands and islands, is mandatory to improve available information on diversity, distribution, and ecology of these forensically important insects.
Recent research has incorporated DNA barcoding as a trustworthy method for the identification and discovery of species [33,34] including forensic important flies [16]. Molecular identity is important to confirm species status and to contribute with a faster and more reliable analysis of the entomological evidence [7,16,35] specially when available samples of the crime scene are immature stages, which are difficult to identify using taxonomic keys. Consequently, voucher specimens from Museum collections are highly important to confirm old records, assess taxonomic variation, and fill out gaps on distribution data [28,31]. Therefore, an optimal local reference collection, scientists can rely on, is hardly needed to continue with the development of forensic entomology in the country. Even though previous studies on molecular ID of forensically important flies have shown to be successful when using mitochondrial barcodes, particularly the cytochrome c oxidase subunit I-COI [16,24,[35][36][37], other studies have reported problems when running identity assessments [38,39]. Barcoding, together with Museum collections, contributes with a variety of new opportunities in the analysis and interpretations of forensic importance specimens [16] as well as to the biodiversity inventories and DNA data bulk of the planet [34]. Finally, we expect this study to enhance the collaboration between entomologists and the legal community [4,7,16,[40][41][42].