Using the developmental gene bicoid to identify species of forensically important blowflies (Diptera: calliphoridae).

Identifying species of insects used to estimate postmortem interval (PMI) is a major subject in forensic entomology. Because forensic insect specimens are morphologically uniform and are obtained at various developmental stages, DNA markers are greatly needed. To develop new autosomal DNA markers to identify species, partial genomic sequences of the bicoid (bcd) genes, containing the homeobox and its flanking sequences, from 12 blowfly species (Aldrichina grahami, Calliphora vicina, Calliphora lata, Triceratopyga calliphoroides, Chrysomya megacephala, Chrysomya pinguis, Phormia regina, Lucilia ampullacea, Lucilia caesar, Lucilia illustris, Hemipyrellia ligurriens and Lucilia sericata; Calliphoridae: Diptera) were determined and analyzed. This study first sequenced the ten blowfly species other than C. vicina and L. sericata. Based on the bcd sequences of these 12 blowfly species, a phylogenetic tree was constructed that discriminates the subfamilies of Calliphoridae (Luciliinae, Chrysomyinae, and Calliphorinae) and most blowfly species. Even partial genomic sequences of about 500 bp can distinguish most blowfly species. The short intron 2 and coding sequences downstream of the bcd homeobox in exon 3 could be utilized to develop DNA markers for forensic applications. These gene sequences are important in the evolution of insect developmental biology and are potentially useful for identifying insect species in forensic science.

1. Introduction

DNA-based methods in forensic entomology have advanced rapidly since the first successful attempt to discriminate important forensic insect species in 1994 [1]. DNA genotyping is now routinely used in the laboratory [2]. Traditionally, a small number of expert insect taxonomists have identified species based on morphological traits. However, growth rates and habitats of important forensic insects differ from one species to another. Further, insect specimens captured at the crime scene or from corpses are often at immature developmental stages and are outwardly indistinguishable, making them morphologically unsuitable for forensic purposes [3]. Identifying important forensic insects, including blowflies and fleshflies, using DNA-based techniques is promising because it contributes considerably to the estimation accuracy of the postmortem interval (PMI) [4].

Blowfly species (Diptera: Calliphoridae) are among the first to colonize decomposing human bodies, and they play a significant ecological role by infesting a corpse with their larvae. Many blowfly species are distributed globally, while others show more localized distributions. Blowflies have 12 chromosomes, including ten autosomes and X and Y chromosomes [5].

DNA-based techniques for identifying forensic blowfly species have been developed using genes encoded in mitochondrial genomes, such as cytochrome oxidase subunits I and II (COI and COII) [5-11] or nuclear genes such as 28S ribosomal RNA [12, 13]. Since the conventional DNA markers have some limitations, new autosomal-based DNA markers for the identification of forensic insect species may ameliorate the problems associated with conventional markers. Among blowfly species, two Lucilia species (L. illustris and caesar), two tropical Luciliinae species (Lucilia cuprina and Hemipyrellia ligurriens), and two African Chrysomya species (C. putoria and C. chloropyga) show species level paraphylies [14-16]. According to Zaidi et al., identification of closely related species may demand a multigene approach [17].

Bicoid (bcd) is a gene whose transcripts are secreted from maternal nurse cells into the anterior poles of eggs. During embryonic patterning, the protein forms a concentration gradient that determines the anterior-posterior axis, significantly impacting the development of the head and thorax in Drosophila melanogaster [18, 19]. The bcd protein is a transcription and translation factor that contains a homeodomain and that acts as a morphogen in the formation of the anterior-posterior pattern [20]. Embryonic bcd transcripts localize in the anterior regions of blowfly and housefly embryos [21, 22].

Found only in higher dipteran families (suborder: Cyclorrhapha), which include most forensic fly species, bcd genes are evolutionary novel [20, 23]. The gene bcd is an exceptional member of the Hox3 cluster of homeobox genes [23]. Full or partial bcd gene sequences have been determined in many dipteran species, including a primitive cyclorrhaphan fly (Megaselia abdita) [23], a dozen Drosophila species [24], the housefly (Musca domestica), and some blowflies, such as Lucilia sericata and Calliphora vicina [25]. A genomic map of the housefly bcd gene is now available, along with that of D. melanogaster [25]. Though bcd expression in housefly and blowflies may have diverged from that of the fruitfly [21, 22, 26], the structures of the four exons have been largely conserved between the fruitfly and housefly, with some introns expanded in housefly [25] (see supplementary Figure 1 in Supplementary Material available online at http://dx.doi.org/10.1155/2013/538051).

To evaluate the applicability of bcd gene sequencing to the identification of forensic insect species, partial genomic sequences of bcd genes containing the homeobox, and flanking sequences were obtained by PCR amplification from a number of specimens of 12 blowfly species collected in Korea. The sequences were determined and analyzed for suitability in blowfly species identification.

2. Materials and Methods

2.1. Fly Collection

To collect flies, pork liver bait was used to attract them to traps in several regions of South Korea. After being submerged in 70% ethanol solution, flies were identified under a dissecting microscope. Morphological identification was done under the taxonomic Keys by Kano and Shinonaga [27]. Both male and female flies were subjected to DNA analysis. To trace the transmission of single nucleotide and insertion/deletion polymorphisms, some adult blowfly species were mated in the laboratory and genotyped. A few F1 and several F2 progenies of the same first-generation females were obtained and subjected to subsequent experimental analysis.

2.2. DNA Extraction

After immersion into liquid nitrogen, whole blowfly bodies without heads were ground into powder in 1.5 mL microcentrifuge tubes with piston pellets (Tokken Inc., Japan). The genomic DNA was extracted with phenol-chloroform-isoamyl alcohol (25 : 24 : 1).

2.3. Polymerase Chain Reaction (PCR) and Cloning

The initial PCR was performed with a degenerate primer pair (F1 and R1) based on similarities in the C. vicina (AJ297855), L. sericata (AJ297856), and M. domestica (AJ297853 and AJ297854) coding sequences (see supplementary Table 1). The initial degenerate primer pair amplified–350 bp-long homologous sequences, spanning from the 3′ region of exon 2 to immediately downstream of the bcd homeobox motif. For the initial PCR, touch-down amplification was performed with an initial step of 95°C for 11 min, followed by eight cycles of 95°C for 30 sec, annealing temperatures starting at 42°C for 1 min and decreasing 1°C/cycle, and extension at 72°C for 1 min. These cycles were followed by 35 cycles of 95°C for 30 sec, 42°C for 1 min, and 72°C for 1 min, with a final extension at 72°C for 15 min. DNA fragments amplified with degenerate primers were visualized on agarose gel electrophoresis, eluted using a GeneClean III kit (Q-Biogene, Bio 101 Systems, Carlsbad, CA, USA), and cloned into a TA vector system (Real Biotech Corporation, Banquiao, Taiwan). The plasmids were extracted from the E. coli host strains, and the sequences were determined.

Based on the initial sequences, two new 5′ primers (F2 and F3) were designed to hybridize 60–80 bp downstream of the initial 5′ primer (F1). The new 5′ primers and a 3′ primer (R2), which hybridized more than 200 bp downstream of the initial primer, were used for subsequent PCR amplifications [9, 10]. After analyzing the sequences from the first and second rounds of PCR, specific primers (FD1–9 and RD1–3) were designed (see supplementary Table 2) and used to amplify 500–525 bp targets. The sequences were determined either by direct sequencing or by sequencing after cloning.

2.4. Sequencing

Sequencing was performed with automatic sequencers (ABI PRISM 310 genetic analyzer) [9, 10]. Direct sequencing of amplified products was performed using the BigDye Terminator Sequencing kit (v1.1) (Applied Biosystems, Foster City, CA, USA). The specific forward and reverse primer sets (FD1–9 and RD1–3) were used to sequence the amplified products. Some amplified DNAs were cloned into TA vectors, and multiple colonies were sequenced to delineate haplotypes. All of the sequences obtained with the FD-RD primer pairs were submitted to GenBank (Accession Numbers GU256065–256168 and 979851–979856). Most bcd sequences were heterozygous though homozygous sequences of two L. sericata individuals (GU256104-5 and GU256107-8), one T. calliphoroides (GU256113-4), and two Ch. pinguis (GU256138–141) were also deposited.

2.5. Sequence Analysis and Phylogenetics

mRNA and genomic sequences were downloaded from GenBank (NCBI) and analyzed in order to resolve the bcd genomic structures (see supplementary Figure 1). The XM_00210237 and AF465792 sequences were downloaded to resolve the genomic structure of Drosophila simulans bcd, which has about 97% nucleotide identity to the coding sequence of D. melanogaster. For the common housefly (M. domestica), the AJ297853 and AJ297854 mRNA sequences were compared to partial genomic sequences (AJ297850–52). The bcd sequences of the housefly and the mRNA sequence of C. vicina (AJ297855) were used as references to compare and analyze new blowfly DNA sequences.

Nucleotide and amino acid sequences were aligned using ClustalW in MEGA4 software [28, 29], which calculated nucleotide distance matrices based on maximum composite likelihoods. Phylogenetic trees were generated using the neighbor-joining method with a bootstrapping of 1,000 replicates.

3. Results

3.1. Coding Sequence Variation

The sequences were analyzed according to a dichotomy of coding (3′ region of exon 2 and 5′ half of exon 3) and noncoding sequences (intron 2) (Table 1). The coding sequences, excluding intron 2, showed a relatively small intraspecific distance. Ch. pinguis and L. ampullacea individuals showed similar variation (0.0083, n = 7 and 0.0081, n = 3; resp.), while A. grahami, Ch. megacephala, L. illustris, and H. ligurriens had virtually no intraspecific variation (Table 1).

Table 1

Intraspecific distances of the partial coding sequences and intron 2 of the bcd gene*.

Species (abbreviation)	Coding sequence	Intron 2	N
(1) Triceratopyga calliphoroides (Cl)	0.003024	0.000000	14*
(2) Aldrichina grahami (Gr)	0.000000	0.000000	7
(3) Calliphora lata (La)	0.001335	0.000000	10
(4) Calliphora vicina (Vi)	0.001661	0.000000	11
(5) Chrysomya megacephala (Mg)	0.000000	0.000000	3
(6) Chrysomya pinguis (Pi)	0.008341	0.055927	7*
(7) Phormia regina (Re)	0.000600	0.000000	8
(8) Lucilia ampullacea (Am)	0.008057	0.017557	3
(9) Lucilia caesar (Ca)	0.001029	0.000000	8
(10) Lucilia illustris (Il)	0.000000	0.000000	22
(11) Hemipyrella ligurriens (Li)	0.000000	0.000000	2
(12) Lucilia sericata (Se)	0.001468	0.013168	8*

Coding sequences = 3′ of exon 2 and 5′ half of exon 3 of bcd.

N: sample size.

*The number of the sequences deposited in GenBank is greater than the sample size.

The interspecific nucleotide differences (distances) in the coding sequences between the 12 blowflies ranged from 0.21  to  0.26 (see supplementary Table 3). Among the 12 blowfly species, C. vicina and Ch. pinguis differed the most (distance = 0.1235) followed by C. vicina and Ch. megacephala (0.1215). On the other hand, L. caesar and L. illustris (distance = 0.0006) showed the highest sequence similarity, followed by Ch. megacephala and Ch. pinguis (0.01646).

Partial amino acid sequences downstream of the bcd homeodomain (about 70 amino acids) displayed comparatively higher inter- and intraspecific variations than did the homeodomains among the 12 blowfly species (Figure 1). These variations reflect some nonsynonymous nucleotide and codon insertion/deletion polymorphisms. The partial protein sequence of exon 2 (about 25 amino acids) showed few amino acid changes among and within species.

Figure 1

Multiple alignments of the amino acid sequences downstream of the bcd homeodomain in blowflies. The sequences–70 amino acids downstream of the bcd homeodomain in 12 blowfly species were aligned to the 73 amino acid sequence of M. domestica (AJ297854). All 12 blowfly species lack the three serine residues present in M. domestica. Other serine residues are also deleted from around the distinct serine-rich region. All blowfly sequences were represented without being accountant for intraspecific variation.

3.2. Intron 2 Variation

Intron 2 from bcd genes showed high interspecific variation in the lengths and nucleotide sequences. The sequence lengths could be classified into two types (Table 2). The first type was similar in size to D. melanogaster (55 bp). bcd intron 2 was either 52 bp or 53 bp in T. calliphoroides, 53 bp in A. grahami, C. vicina, C. lata, L. caesar, and L. illustris, 53–55 bp in L. sericata and 57 bp in H. ligurriens. Intron 2 in L. ampullacea was either 53 bp or 60 bp depending on the presence of an eight-nucleotide insertion. These blowfly species belong to the subfamilies Calliphorinae and Luciliinae. In addition, a multiple alignment showed that intron 2 from Calliphorinae species, such as T. calliphoroides, A. grahami, C. vicina, and C. lata, differed from those of the Luciliinae species, such as H. ligurriens, L. caesar, L. illustris, L. ampullaceal, and L. sericata (Table 2).

Table 2

The bcd intron 2 sequences of 12 blowfly species: Drosophila melanogaster and Musca domestica.

Single-nucleotide polymorphism: R = A or G; Y = C or T; W = A or T; M = A or C; K = G or T; M = A or C.

The underlined nucleotide(s), for example, AA stand for the two alleles of AA versus deletion (-).

N: sample size.

*For the intron 2 analysis, an additional sequence of the bcd gene in L. sericata (GU979856) was included.

L. ampullacea had an eight-nucleotide insertion in one individual at position 21 (indicated with slash symbol (/)).

The second type of intron was about 10 bp longer, similar in size to that of the housefly (M. domestica) (73 bp). Intron 2 was 63 bp in P. regina, 68-69 bp in Ch. pinguis, and 69 bp in Ch. megacephala. These species belong to the subfamily Chrysomyinae.

The interspecific sequence variation in intron 2 was relatively higher than that of the coding sequences (see supplementary Table 4). The interspecific distance was highest between A. grahami and P. regina (0.9730). L. caesar and L. illustris (distance = 0.0263) showed the highest similarity, with the next highest observed between Ch. megacephala and Ch. pinguis (distance = 0.0319) (see supplementary Table 4).

Six species, Ch. pinguis, L. ampullacea, L. sericata, L. illustris, L. caesar and C. lata, contained intraspecific single nucleotide polymorphisms (Table 2). Single-nucleotide insertion/deletion (indel) polymorphisms were discovered in Ch. pinguis, and T. calliphoroides, while a two-nucleotide indel polymorphism was found in L. sericata. An eight-nucleotide indel polymorphism was also discovered in L. ampullacea. L. sericata, grown in the laboratory, confirmed the stable transmission of these single-nucleotide and indel polymorphisms to progeny (data not shown).

3.3. Phylogenetic Tree

A phylogenetic tree of the 12 blowfly species was constructed based on the partial bcd genomic sequences containing the homeobox and flanking coding and noncoding sequences (Figure 2). With the partial bcd genomic sequences, three subfamilies: Luciliinae, Chrysominae, and Calliphorinae, within the family Calliphoridae, could be classified among the 12 blowfly species collected from Korea. In the subfamily Luciliinae, the taxonomic status of the genus Hemipyrellia (H. ligurriens), which is distinct from the genus Lucilia (L. illustris, L. caesar, L. ampullacea, and L. sericata), was confirmed. In the subfamily Chrysominae, some Ch. pinguis haplotypes fell outside a cluster of other Ch. pinguis and Ch. megacephala sequences. In the subfamily Calliphorinae, the genus Calliphora, namely, C. vicina and C. lata in this study, did not form a monophyletic group, clustering with a species of a different genus, A. grahami. Almost all blowfly species in this study could be discriminated in the phylogenetic tree except the Ch. megacephala and Ch. pinguis species pair (Figure 2).

Figure 2

A neighbor-joining phylogenetic tree of 12 blowfly species. The bcd genomic sequence of M. domestica was used as the outgroup to construct the phylogenetic tree. Most sequences are diploid except some haploid sequences of L. sericata (GU256104-5, GU256107-8), T. calliphoroides (GU256113–GU256114) and Ch. pinguis (GU256138-141).

4. Discussion

The genomic bcd sequence is potentially useful for identifying forensically important insect species. Partial bcd sequences of about 500 bp were amplified to evaluate possible species-specific DNA markers, with promising results.

We observed intra- and interspecific lengths and sequence polymorphisms in intron 2 bcd from 12 blowfly species. The sizes of intron 2 in D. melanogaster, D. simulans, and M. domestica were interesting (Figure 1). Among the three dipteran species, intron 2 showed small-size differences (55 bp, 55 bp, and 73 bp, resp.). In contrast, introns 1 and 3 showed substantial size differences and were expanded in M. domestica, but conserved to a considerable extent in the two Drosophila species. Introns 1 and 3 were expanded 6–8 times in M. domestica (12–13 kbp and 2.5 kbp) compared to those of the two Drosophila species (about 400–500 bp) (Figure 1) [25].

The D. melanogaster bcd sequences (AF466621-45) did not vary in intron 2 of bcd (Table 2) among 25 isofemale lines collected from Zimbabwe, a representative ancestral population [30]. Intron 2 in M. domestica is larger than that of D. melanogaster and showed intraspecific variation that transmits stably to ensuing generations (data not included).

The interspecific length polymorphisms of intron 2 could be classified into two categories: first, a D type including the subfamilies Calliphorinae and Luciliinae, similar to the intron 2 of D. melanogaster and D. simulans, and second, an M type including the subfamily Chrysomyinae, similar to that of M. domestica (Table 2). Lengthwise, the first type could be further classified into the two subfamilies: intron 2 in Luciliinae tended to be slightly longer than Calliphorinae (Table 2). The nucleotide distance values complement the length polymorphism-based distinction: species within the same subfamily showed below-average pairwise distances, while species of different subfamilies displayed above-average pairwise distances (average distances among the 12 blowfly species = 0.5453) (see supplementary Table 4).

Downstream of the bcd homeobox (about 210 nucleotides), exon 3 also displayed intra- and interspecific variations (Figure 1). The posthomeodomain sequence, which includes exons 3 and 4, is thought to interact with other transcription factors and proteins for translation [20].

There are also interspecific amino acid differences immediately downstream of the homeobox (Figure 1), which may provide the basis for a simple, reliable, and effective DNA-based method for distinguishing forensic insect species. Thus, taxon-specific sequence-characterized amplified region (SCAR) markers [31-33] based on bcd will likely become available for forensic applications.

The SCAR markers can be developed to discriminate between a pair of sister species like L. caesar and L. illustris [15, 27]. In this study, eight L. caesar individuals and 22 L. illustris individuals were subjected to analysis of bcd polymorphisms. A single nucleotide difference between L. caesar and L. illustris (position 42 in intron 2, multiple alignments in Table 2) could distinguish the two closely related species; cytosine (C) is present in this location in L. caesar and adenine (A) in L. illustris. This specific difference can be incorporated into SCAR markers for species identification. These markers should be able to discriminate blowfly species captured in South Korea.

All blowfly species were well clustered according to subfamilies Luciliinae, Chrysominae and Calliphorinae in this study (Figure 2), according to partial bcd sequencing phylogenetic analyses. In the subfamily Luciliinae, the genera Hemipyrellia and Lucilia formed individual clades. The closely related sister species, L. illustris and L. caesar, were also distinguishable. In the subfamily Chrysominae, the analysis of more individuals may be necessary to distinguish between Ch. megacephala and Ch. pinguis. A greater number or complete bcd sequences from more individuals may resolve the apparent species-level paraphyly in Ch. pinguis. The genus-level paraphyly in the subfamily Calliphorinae (genera Aldrichina, Cynomya, Eucalliphora, Triceratopyga, etc.) based on mtDNA phylogenies has been reported previously [10, 34, 35]. C. vicina and C. lata showed similar trends in this study; they did not form a genus-level monophyly, instead A. grahami clustered with the two Calliphora species (Figure 2). Based on the COI phylogeny, Rognes [34] and Whitworth [35] suggested that A. grahami and the genus Eucalliphora should be incorporated into the genus Calliphora. The bcd phylogenetic tree supports this suggestion, calling for a possible future taxonomic revision.

This study showed that even a 500 bp partial bcd genomic sequence containing the homeobox motif can be useful for distinguishing most forensically important blowfly species including previously recognized confused taxa, that is, L. illustris and L. caesar. Additionally, this kind of study may be extended to other confused taxa such as L. cuprina and some Chrysomya species [14-16]. This study is the first to use a renowned developmental gene, bicoid, in forensic science and entomology. Bcd sequences of 12 blowfly species are useful for species identification although many blowfly species uncommon in Korea were not included in this study. However, we expect that other researchers can easily apply our degenerate PCR strategy to amplify bcd sequences of their regional blowfly species.

Supplementary Table 1: The primers used for the initial and second PCR amplification.

Supplementary Table 2: The species-specific primer pairs used.

Supplementary Table 3: Inter-specific distances of the bcd coding sequences.

Supplementary Table 4: Inter-specific distances of the bcd intron 2.

Supplementary Figure 1: The bicoid genomic structures of Drosophila melanogaster, Drosophila simulans and Musca domestica.

Supplementary Figure 2: Multiple alignments of the bcd homeodomain amino acid sequences in dipteran species.

Click here for additional data file.

Click here for additional data file.

Click here for additional data file.