Evidence of two mitochondrial lineages and genetic variability in forensically important Lucilia eximia (Diptera: Calliphoridae) in Colombia

Abstract Lucilia eximia (Wiedemann, 1819) (Diptera: Calliphoridae) is a blowfly with medical and forensic importance that shows genetic and color variation, however, these variations have not justified the description of new species. But in forensic entomology an accurate identification of species and subpopulations is crucial. We explored the genetic variation of L. eximia from eight localities, in five natural regions in Colombia using two mitochondrial fragments, including the standard locus for insect identification COI and the Cytb-tRNA-Ser-ND1 region. We found significant differentiation at COI and Cytb-tRNA-Ser-ND1 level, characterizing two lineages and revealing a deep and significant genetic split. High values of FST and genetic distances supported the two lineages. The origin of the divergence of L. eximia remains to discover. Examining whether the lineages have diverse ecological and biological behaviors could be a significant impact on the use of L. eximia in forensic and medical science. Our results could have relevant implications for the use of post-mortem interval estimation based on insect evidence, as well as our sequences improve the database used in DNA-based methods for identifying forensically important flies.


Introduction
Many of us relate Lucilia eximia (Wiedemann, 1819) (Diptera: Calliphoridae) as an annoying fly, despite its valuable and unexpected uses. Applying controlled myiasis, their larvae remove necrotic tissue of chronic wounds (Wolff et al. 2010). Its developmental rate helps forensic entomologists to estimate the time of death of a corpse (Catts andGoff 1992, Acosta et al. 2021). Also, entomological evidence provides valuable information concerning the circumstances of death, including season, location, movement or storage after death, use of drugs, and even linking child neglect (Benecke andLessig 2001, Amendt et al. 2007). But the use of entomological evidence relies on accurate species identification and subpopulations (Tomberlin et al. 2011).
Despite its value, morphology-based identification of insects shows limitations (Gemmellaro et al. 2019). Taxonomic keys are not always available and can be difficult to use, especially when identifying immature stages or cryptic species (Harvey et al. 2003, Zehner et al. 2004. Lucilia eximia exhibits color patterns and genetic variability when comparing mainland and island specimens (Whitworth 2010(Whitworth , 2014. In one study in Colombia, L. eximia exhibited a high level of variability based on the Cytochrome Oxidase I gene (COI), suggesting the presence of two species within-also called cryptic species (Solano et al. 2013). However, this study suggested including more specimens from a wider distribution range in Colombia (Solano et al. 2013).
Regarding the color patterns of L. eximia, its vestiture varies from silver, silver-gold to gold (Whitworth 2010). Probably this variation is a result of geographical differences. These color variations as well as genetic variation lead to the description of new species in several groups of insects (Guerrero andFernández 2008, Badejo et al. 2020). Most insects show morphological variation, often related to geographical origin, and landscape alterationas a result of urbanization (Theodorou 2022). But this variation increases the difficulty to identify (Carolan et al. 2012) and include them in posterior studies.
Mitochondrial genes help to identify insects, including Calliphoridae (Stevens and Wall 2001, Salem et al. 2015, Fuenteslópez et al. 2020. Specifically, a short fragment of the COI gene is reliable for species identification, even identifying cryptic species (Mathur et al. 2012, Scarpassa andAlencar 2012). The Cytochrome b gene (Cytb) is also used to study intraspecific variability in Diptera (GilArriortua et al. 2013, Pech-May et al. 2018. For instance, to study the L. eximia genetic variability Giraldo et al. (2011) used a fragment of mitochondrial DNA including the Cytb and the NADH dehydrogenase genes (Cytb-tRNA-Ser-ND1). However, this study showed a moderate level of variability, conversely to the high level of variability revealed by COI sequences in Colombian specimens (Solano et al. 2013).
To study the genetic variability of L. eximia, we analyzed specimens collected from eight localities in five Colombia natural areas, using two mitochondrial fragments, the COI barcode region and Cytb-tRNA-Ser-ND1. We expected to find substantial variation between L. eximia populations.

Methods
Lucilia eximia adults were collected in eight localities from five natural areas in Colombia defined by IGAC (1997), including highly disturbed areas, semi-rural areas, and secondary forests (Table 1).
We used Van Someren-Rydon traps (DeVries 1987) using a mixture of rotten fish heads and chicken viscera as bait. In each locality, we placed four traps at 1.5 m above the ground for 96 h and separated the traps 200 m from each other. Every 48 h, we replaced the baits and collected the specimens and identified the samples using taxonomic keys for the Neotropical species of Lucilia (Whitworth 2010(Whitworth , 2014. From one specimen leg of the 61 collected specimens, we extracted DNA using the DNeasy Blood & Tissue Handbook (Qiagen) kit. A 650 pb fragment from the COI gene was amplified using primers developed by Folmer et al. (1994) and PCR amplification following the Solano et al. (2013) protocols. A second fragment of 492 pb was amplified from the Cyb-tRNA-Ser-ND1 region, using primers developed by Ready et al. (1997). The PCR products were visualized on 0.8% agarose gel using Ezvision (AMRESCO) under UV light and sequenced bi-directionally (Macrogen Inc. Korea). Sequences were checked against available records in the National Center for Biotechnology Information (NCBI) using the BLAST algorithm. Finally, we verified possible mitochondrial copies at the nucleus (NUMT) by BLASTN search and analyzing codons as suggested by Hlaing et al. (2009) and assembled the sequences using the genome of L. sericata (NC_009733) in Geneious v.8.0.4.
Based on the concatenation of the COI and the Cytb-tRNA-Ser-ND1 genes, we built a tree using Bayesian Inference Analysis and applied the evolutionary model GTR + T based on the Akaike criterion (Akaike 1974), through MrBayes v. 3.1.2 (Ronquist and  (Excoffier et al. 2005) and used the Mantel test to analyze the correlation between geographic distances and F ST values using XLSTAT v.3.9.

Distribution and frequency of haplotypes
We found a total of 16 COI haplotypes, their number varied per locality, from one in Playa Huina and Leticia to seven in Puerto Gaitán. H1 (COI haplotype1) had the highest frequency (28.3%) followed by H2 (17.4%) both shared among Copacabana, Caldas, Pajarito, and Cola del Zorro, the four closest localities in Antioquia (Fig. 1).
We found a total of 8 Cytb-RNAt-Ser-ND1haplotypes, their number varied per locality, from one haplotype in Playa Huina, Pajarito, and Leticia to three in Copacabana and Puerto Gaitán (Fig.  2). H1b (Cytb-RNAt-Ser-ND1 haplotype 1) was the most frequent haplotype (60.6%) and was found in Cola del Zorro, Copacabana, Caldas, Pajarito, and Leticia-the first four localities are close to each other but far from the last Leticia. The second most frequent haplotype H6b (21.3%) was distributed in Playa Huina, Puerto Gaitán, and Santa Marta.

Lucilia eximia genetic lineages
The two lineages of COI and Cytb-tRNA-Ser-ND1 showed nonoverlapping geographic distribution. The lineage I clustered specimens from Santa Marta, Playa Huina, and Puerto Gaitán, whereas, the lineage II clustered specimens from Caldas, Cola del Zorro, Copacabana, Leticia, and Pajarito. The Bayesian concatenated analysis revealed the same clustering (Fig. 4).
Likewise, the F ST values also showed a significant degree of genetic differentiation between the two lineages (COI F ST = 0.9334 P < 0.001; Cytb-tRNA-Ser-ND1 F ST = 0.6685 P < 0.001) and the Mantel test indicated the absence of correlation between F ST and geographic distances (COI, r = 0.0450, P = 0.9580 and Cytb-tRNA-Ser-ND1, r = 0.2720, P = 0.1350). Whereas genetic distance ranged from 0.2% to 6.2% between the two lineages. Lineage I showed higher genetic diversity than the lineage II ( Table 2). The genetic distance within the lineage I ranged from 0.2% to 1.2%, while within the lineage II ranged from 0.2% to 2.4%.

Discussion
This is the first approach to the genetic variability of L. eximia in Colombia using two mitochondrial genes. Lucilia eximia showed significant levels of genetic variability with two mitochondrial lineages, revealing a deep and significant genetic split. The mutational steps observed in the haplotype networks supported the idea of L. eximia in Colombia could be a cryptic species complex. Hart and Sunday (2007) found strong evidence that DNA sequences from single species typically stick together in a single haplotype network. Changes from these patterns are usually consistent with hybridization or cryptic species diversity (Motoki et al. 2021, Scarpassa et al. 2021. Studies using the COI barcode region for delimiting Calliphoridae showed intraspecific distance values between 0.0%-1.487% in Colombia (Solano et al. 2013) and 0.0%-0.612% in Australia (Nelson et al. 2007). We found genetic distance values higher than these previous reports in Colombia and similar to those reported from the Caribbean Region specimens (Yusseff-Vanegas and Agnarsson 2017). Conversely, as observed in the last study, our findings do not appear to be explained by geographical origin specimens. Our results confirmed a substantial intraspecific variation of L. eximia in Colombia using the COI barcode and Cytb-tRNA-Ser-ND1 region.
The genetic differences we found in L. eximia in Colombia may be influenced by the landscape and its characteristics. The idea that anthropization affects the molecular variability of L. eximia has not been yet proposed. Ecological conditions, such as the ones resulting from anthropogenic impact, changed the genetic structure of other insects (Ashley et al. 2003, Huber et al. 2008. For instance, two populations of Aedes aegypti (Diptera: Culicidae) from the forest and urban areas showed genetic differences attributed to environmental factors (Huber et al. 2008). In our study, we did not choose the localities based on the level of anthropization. However, we noticed that the localities which presented the lineage I showed low anthropization levels (semi-rural areas). While the localities which presented the lineage II were urban areas. We hypothesize that the presence and distribution of the lineages of L. eximia may be related to the level of anthropization. However, this explanation, it is beyond our purpose and we need additional research to confirm it.
We detected high levels of genetic variability of L. eximia based on mitochondrial regions. However, given the lack of information on the nuclear variability of L. eximia, further population studies using nuclear genes are crucial to determine the variation at this level. Studies on other species have shown that mitochondrial genetic variation does not necessarily indicate species differentiation. For example, two populations of the blowfly Phormia regina (Diptera: Calliphoridae) showed a substantial level of variation in mitochondrial genes but a low variation in nuclear genes and nonmorphological variation. This study concluded that the variation between P. regina populations was related to intraspecific mitochondrial divergence rather than species-level differentiation (Jordaens et al. 2013).
Our results could have important implications for the use of L. eximia in forensic and other medical fields. In other species, fertility, life cycle, emergence, and adult longevity varied between populations and this variation was related to noteworthy molecular variation (Justiniano et al. 2004, Scarpassa and Alencar 2012, Jordaens et al. 2013. For instance, the forensically important blowfly P. regina shows developmental rate variation, which was explained by mitochondrial divergence between populations (Picard and Wells 2009). These findings highlighted the importance of taking into account this variation when using insects in forensic casework. The lineages we found in L. eximia may differ biologically, studying the lineages of L. eximia in more detail is necessary. Differences in the developmental rate between lineages in L. eximia would affect the use of L. eximia in forensic entomology applications.
We found significant results when analyzing a small sample of the population, we suggest that future research include more specimens, as well as nuclear genes. In conclusion, L. eximia exhibits considerable genetic variability and two mitochondrial lineages in Colombia. It remains to elucidate the origin of this divergence and explore if the lineages exhibit differences in ecological/biological behavior, which may have important implications for the use of L. eximia in the forensic field.