Introduction

The mechanism responsible for the development of animal body plans is fascinating because of the large diversity in animal morphology. During determination of the proper segment identity, homeotic (or Hox) genes encoding conserved homeodomain transcription factors, have a crucial role in characteristic patterns of segments along the anterior–posterior body axis (McGinnis and Krumlauf, 1992; Hughes and Kaufman, 2002). In arthropods, changes in Hox gene expression domains or in their functions can lead to the diversity of segment morphology, appendage number and pattern (Carroll, 1995; Averof and Patel, 1997).

Study of the molecular mechanism underlying the formation of homologous appendage patterns in insects is required for understanding the evolution of morphological diversity (Carroll, 2000; Jockusch et al., 2004). All insects have three pairs of thoracic legs, the morphology of which is thought to be determined by Hox genes regulating a variety of downstream target genes related to limb development, such as Distal-less (Dll), Wingless (Wg) and Decapentaplegic (Dpp) (Jockusch et al., 2000; Grienenberger et al., 2003; Gebelein et al., 2004). In Drosophila melanogaster, the homeotic Antennapedia (Antp) gene is expressed in the thoracic embryonic epidermis and is required for proper development of the thorax and legs (Abbott and Kaufman, 1986; Carroll et al., 1986; Wirz et al., 1986). Loss-of-function mutations in the embryo result in transformation of the second and third thoracic segments toward the prothoracic segment (Wakimoto and Kaufman, 1981). Ectopic expression of the Antp gene leads to distinct phenotypes, such as the transformation of antennae into second legs and eye reduction (Schneuwly et al., 1987; Plaza et al., 2008; Prince et al., 2008). Moreover, in the spider Achaearanea tepidariorum, a non-insect arthropod, the Antp gene represses the growth of legs in the first segment of the abdomen (Khadjeh et al., 2012). There are reports of extensive studies, but further exploration of the particular role of the Antp gene underlying the development of the thorax and thoracic appendages is needed.

In the silkworm Bombyx mori, >30 different homeotic mutants called the E complex have been identified and most of them show anomalies in proleg development (Tazima, 1964; Banno et al., 2005). The E complex was presumed to be analogous to the D. melanogaster bithorax complex (BX-C; Ueno et al., 1992) and studies on B. mori BX-C genes have suggested that all of them are involved in proleg patterning (Masumoto et al., 2009; Tomita and Kikuchi, 2009). However, the Hox genes are complex and few of these mutants have been characterized at the molecular level.

Wedge eye-spot (Wes) is a new homeotic mutant that was discovered during the preservation and investigation of the characteristics of silkworm resources (Dai et al., 2009). Genetic analysis revealed that it is controlled by a single dominant gene and is located at 24.3 cM on the sixth linkage in the silkworm genetic linkage map. The genetic distance between Wes and EKp, a member of the E complex (Banno et al., 2005; Xiang et al., 2008), was ∼3.0 cM (Dai et al., 2009). In the present study, we identified and characterized the gene responsible for the Wes mutant, which induces lethality in the embryonic developmental stage of homozygotes; thus, the Wes mutant strain is maintained as a heterozygous stock. We demonstrated that the Wes locus encodes the homeodomain-containing protein BmAntp, and conclude that the abnormal BmAntp is responsible for the Wes mutant. Our results suggest that the BmAntp gene is essential for specifying the identify of thoracic segments.

Materials and methods

Silkworm strains

The Wes strain (Wes/+Wes) and wild-type strain Dazao were used as parent strains to produce F1 offspring from a single-pair cross. As there is no recombination in female silkworms, 22 progeny from a single-pair backcross between an F1 female and a Dazao male (BC1F) were used for the linkage analysis and 244 progeny from the cross Dazao female × F1 male (BC1M) were used for the recombination analysis (Figure 2e, right). The Wes strain +Wes/+Wes used as a control was generated from Wes (Wes/+Wes) mutant selfing. Silkworm strains Wes and Dazao were obtained from the Silkworm Gene Bank at Southwest University (Chongqing, China). The developing embryos were incubated at 25 °C with adequate humidity, and the larvae were reared with fresh mulberry leaves at 25 °C with a photoperiod of 12 h light/12 h dark.

Scanning electron microscopy

Embryos from developing eggs, 7 days after oviposition, were dissected under a stereomicroscope (C-DSS230, Nikon Corporation, Tokyo, Japan), fixed in 2.5% glutaraldehyde and kept at 4 °C for at least 2 h. The embryos were immersed in a phosphate buffer (0.1 M, pH 7.4) three times for 15 min each, then in 1% osmic acid for 2 h and then in water three times for 15 min each. Finally, the embryos were dehydrated by passage through a graded series of ethyl alcohol (30, 50, 60, 70, 80, 90 and 100%) for 15 min in each, then immersed in 50% 2-methyl-2-propanol, 75% 2-methyl-2-propanol, 100% 2-methyl-2-propanol, 2-methyl-2-propanol:methyl cyanide=2:1, 2-methyl-2-propanol:methyl cyanide=1:1, for 10 min each, then stored in methyl cyanide. The embryos were coated with gold before observation under a scanning electron microscope (S-3000N, Hitachi, Tokyo, Japan).

Genomic DNA preparation

The genomic DNA of Wes/Wes was extracted from embryos dissected from developing eggs at 7 days after oviposition using DNAzol (Invitrogen, Carlsbad, CA, USA). The genomic DNA of parental strains and BC1 individuals was extracted from adults and pupae, respectively. The samples were disrupted in liquid nitrogen and suspended in 100 μg ml−1 proteinase K in DNA extraction buffer (pH 8.0, 10 mM Tris-HCl, 0.1 M EDTA, 0.5% SDS) at 50 °C for 5–8 h before extraction with phenol/chloroform by standard procedures. The DNA was precipitated with cold absolute ethanol and dissolved in TE buffer (pH 8.0, 10 mM Tris-HCl, 1 mM EDTA).

Positional cloning

To construct the Wes linkage map, simple sequence repeat (SSR) markers on chromosome 6 (Miao et al., 2005; Zhan et al., 2009) and newly developed markers were screened for 22 BC1F individuals. The primers for the markers are given in Table 1. The segregation patterns in 244 BC1M individuals (Supplementary data) were analyzed using JoinMap 4.0 software (Van Ooijen, 2006), with a logarithm of the odds (LOD) threshold of 4.0 and the Kosambi mapping function (Kosambi, 1944).

Table 1 The primer sets used in this study
Full size table

According to the results of the preliminary SSR mapping, the candidate genes were analyzed using the silkworm genome database SilkDB (http://silkworm.swu.edu.cn/silkdb/) (Duan et al., 2010). We used the BLASTp program (http://blast.ncbi.nlm.nih.gov/Blast.cgi; Altschul et al., 2005) to find the homologous and orthologous genes associated with determining segment identity during the development of other species.

Cloning of B. mori Antp (BmAntp) fragment

Total RNA was isolated from eight 7-day-old Dazao and eight 7-day-old Wes/Wes embryos using the MicroElute Total RNA Kit (Omega Bio-Tek, Norcross, GA, USA) according to the manufacturer’s instructions. The Wes/Wes homozygotes were distinguished from others by their fused thorax and antenna-like appendages under mouthparts. Embryos were excised with a mini-homogenizer (Polytron PT1600E, Kinematica, Lucerne, Switzerland). cDNA was synthesized from total RNA samples with oligo(dT) primers and a Moloney murine leukemia virus reverse transcriptase (Promega, Madison, WI, USA) according to the manufacturer’s instructions. Bmactin 3 was used as the internal control. We designed primers P1 and P2 according to the sequence of Antp (GenBank accession number: KC165846) in the silkworm (Table 1). The PCR reactions used Ex Taq (Takara, Dalian, China) under the following conditions: initial denaturation at 94 °C for 4 min; 32 cycles at 94 °C for 45 s, Tm (57 °C and 64 °C for P1 and P2, respectively) for 1 min, 72 °C for 1 min; and a final extension at 72 °C for 10 min. The PCR products were cloned into the pMD19-T vector (Takara) and sequenced.

Quantitative RT-PCR (qRT-PCR)

Total RNA from the whole body of fifth instar larvae was isolated using TRIzol (Invitrogen) and reverse transcription was carried out as described above. Referring to the full-length cDNA sequence of BmAntp, we designed the gene-specific primer set Antp-C (common) to detect the BmAntp transcripts in both wild-type and Wes heterozygotes, and the primer set Antp-M (mutant) for the mutational transcript in Wes heterozygotes (Table 1). The eukaryotic translation initiation factor 4A (silkworm microarray probe ID sw22934) was used as the internal control. A standard curve was constructed for each target gene using a series of 10-fold dilutions of the plasmid pMD19-T containing the specific fragment. Quantitative RT-PCR (qRT-PCR) experiments were performed using the StepOnePlus real-time PCR system (Applied Biosystems, Foster City, CA, USA) with a SYBR Premix EX Taq_kit (Takara), according to the manufacturer’s recommended procedure. The qRT-PCR conditions were: 95 °C for 4 min, followed by 40 cycles at 95 °C for 15 s and 60 °C for 31 s, 95 °C for 15 s, 60 °C for 20 s and 95 °C for 15 s.

RNA interference (RNAi)

Double-stranded RNAs (dsRNAs) for BmAntp and EGFP were synthesized using the RiboMAX Large Scale RNA Production System-T7 (Promega) based on the PCR templates amplified by the primer set dsBmAntp and dsEGFP, respectively (Table 1) according to the manufacturer’s instructions. The dsRNA precipitate was dissolved in RNase-free water and diluted to a concentration of 5 μg μl−1 for injection. At 2–6 h after oviposition, the eggs were punctured on the ventral side by a micromanipulator (SZX2-STL, Olympus, Tokyo, Japan) using a spiculate metal needle and the injection of dsRNA was done with a thin-walled glass capillary (G-100, Narishige, Tokyo, Japan). The injected eggs were incubated at 25 °C with adequate humidity until dissection. The mutant phenotypes were imaged digitally with a stereomicroscope.

Results

Phenotypes of the Wes mutant

The Wes mutant was discovered and named by Dai et al. (2009) based on the shape of the eye-spot located on the first and second thoracic segment. The eye-spot of the heterozygotic larva (Wes/+Wes) looks like a narrower inverted triangle compared with the wild type (Figures 1a and d), whereas the crescents and stars are normal. The homozygote Wes/Wes is embryonic lethal.

Figure 1
figure 1

Phenotype and appendage morphology of the wild type and Wes mutant. (a) Eye-spot of +Wes/+Wes. Bar=1 cm. The arrowhead indicates the eye-spot. (b) Head of the Dazao embryo. Bar=200 μm. (c) Antenna of Dazao. Bar=50 μm. (d) Eye-spot of Wes/+Wes. Bar=1 cm. The arrowhead indicates the eye-spot. (e) Head of the Wes/Wes embryo. Bar=200 μm. The arrowhead indicates the antennae-like appendages. (f) Enlargement of the antennae-like appendages in (e). Bar=50 μm. (g) Thoracic segments of the Dazao embryo. Bar=500 μm. (h) Enlargement of the same embryo in (g). Bar=100 μm. (i) Enlargement of the thoracic leg on T1. Bar=100 μm. (j) Thorax of the Wes/Wes embryo. Bar=300 μm. (k) Enlargement of the same embryo in (j) as indicated by the the straight line. Bar=100 μm. (l) Enlargement of appendage h in (k). Bar=50 μm. T1, prothoracic segment; T2, mesothoracic segment; T3, metathoracic segment; a, antenna; c, crochet.

Full size image

To further characterize the homozygous mutant embryos, we dissected the embryos in the late developmental stage (6–7 days after oviposition) and observed them under a scanning electron microscope. The normal embryo has three thoracic segments (T1–T3) with one pair of thoracic legs on each (Figure 1g). On the end of each thoracic leg, there is a crochet as a sign of thoracic legs (Figures 1h and i). Amazingly, in the Wes/Wes embryos, the first thoracic legs appeared under the mouthparts and were transformed into antenna-like appendages (Figures 1e and f, compared with Figures 1b and c, respectively). The thorax of the Wes/Wes embryo appeared to be fused (Figure 1j, compared with Figure 1g), and the two remaining pairs of legs did not have the characteristics of thoracic legs, and exhibited partial homeotic transformation to antennae (Figures 1k and l, compared with Figures 1h and i, respectively).

Mapping of the Wes locus

To specify the candidate region responsible for the Wes mutant, we undertook a genetic linkage analysis using the B. mori SSR molecular linkage map and genome sequence. We roughly mapped the Wes mutant using 244 BC1M individuals with the SSR markers on the sixth linkage. The segregation ratio in BC1M individuals was 127 wild type/117 mutant, approaching 1:1. These results indicated that the Wes gene was localized within a 2.0-cM region linked to SSR marker S0609 (Figure 2a), which was on the nscaf2853 scaffold according to the SilkDB database (Figure 2b) (Duan et al, 2010).

Figure 2
figure 2

Mapping of Wes locus on B. mori chromosome 6. (a) SSR markers and genetic analysis. Three SSR markers and the Wes locus are shown above the map. Distances in cM are shown below the map. (b) Genomic scaffolds on chromosome 6. Dark gray boxes represent the assembled scaffolds and the names of the scaffolds are shown below. (c) Scaffold map and gene model of the nscaf2853 scaffold on chromosome 6. The SSR marker S0609 and the Wes locus are shown above the map. The genetic distance between S0609 and Wes locus is 2.0 cM. The gray and brown boxes (part of the Hox cluster) represent the ∼1 Mb region scanning by using bioinformatics analysis and chromosome walking. Distances are shown below in Mb. pb, proboscipedia; Dfd, Deformed; Scr, Sex combs reduced; Antp, Antennapedia. (d) The marker P1 shows polymorphism between Dazao and Wes. The sizes of the DNA markers (up to down) are 5000, 3000, 2000, 1000, 750, 500 and 250 bp. (d) Mapping of the Wes locus using P1. Linkage analysis (left) and recombination analysis (right) by using P1. There is no recombination between P1 and the Wes locus.

Full size image

Identification of the Wes candidate gene

On the basis of preliminary mapping, we biologically analyzed the available downstream genomic sequences of S0609 around a 1 Mb region (Figure 2c), and used chromosome walking. Notably, within ∼300 kb downstream of the S0609 marker, we found the homeotic gene cluster that has a critical role in the specification of different anatomical segment identities. We designed primer sets based on the sequences of the Hox genes in B. mori to develop new markers between Wes and Dazao. We found that one pair of primers from BmAntp (P1) displayed a much larger amplification product in the Wes homozygote than in the wild-type Dazao (Figure 2d). Thereafter, when mapping with the new marker P1, we found no recombination between P1 and the Wes locus (Figure 2e). Comparison to the genome assembly data showed that the sequence amplified by P1 belonged to the third exon of BmAntp. Further, BmAntp, a homolog of the Antp gene in D. melanogaster, was reported to be expressed mainly in the thorax and to have an important role in identifying thorax segments (Hughes and Kaufman, 2002). Therefore, we concluded that the BmAntp gene was the candidate gene for the Wes mutant.

Sequence analysis of the candidate gene

Characterization of the genomic organization of BmAntp revealed that the transcription unit spanned a region of ∼44 kb in the nscaf2853 scaffold on the sixth chromosome and consisted of three exons (Figure 3b). We obtained cDNA from Wes homozygous embryo and found two coexisting transcripts amplified by primer set P1 (Figure 3a). Cloning of these two transcripts and comparing the sequences with the wild type in Dazao indicated that the smaller (type 1) was the same as the normal type, whereas the larger transcript (type 2) contained a 1570 bp insertion (GenBank accession number KC165845) (Figure 3b) that destroyed the homeodomain in BmAntp. Interestingly, the expression level of the abnormal transcript was much higher than that of the wild type in the Wes/Wes; meanwhile, the normal transcript was expressed at a much lower level in Wes/Wes compared with the wild type (Dazao) (Figure 3a).

Figure 3
figure 3

Schematic diagram of BmAntp sequences of the wild type (Dazao) and mutant (Wes/Wes). (a) BmAntp transcripts amplified by P1 in Dazao and Wes/Wes. The Bmactin 3 was used as an internal control. (b) Comparison of the structure of BmAntp between Dazao and Wes/Wes. White boxes, exons; diagonal lines, introns; coarse horizontal lines, untranslated regions; black box, the insertion; gray boxes, target site duplication (TSD); gray arrows below Type 1, the primer set Antp-C; and black arrows below Type 2, the primer set Antp-M. (c) Diagram of a LINE-like element with a 5′ truncation insertion in exon 3 of the BmAntp gene. The BmAntp gene spans 1 551 bp and contains three exons. The 1570 bp insertion is flanked by a 15-base TSD of BmAntp cDNA sequence (nucleotides 1 076–1 090). Nucleotides 3 533–5061 represent the BMC1 element.

Full size image

To characterize the insertion, the abnormal 1570 bp fragment containing a 29-nucleotide tract of A residues was analyzed with the BLASTN program at the National Center for Biotechnology Information. Based on BLAST results, we found that the inserted sequence was similar to a long interspersed nuclear element (LINE)-like element, BMC1, a non-LTR retrotransposon on the W chromosome of the silkworm (Abe et al., 1998). The level of identity was 93% between the insertion and the 3′ end of the BMC1 as estimated by BLASTN (Zhang et al., 2000). In addition, we found a duplication of 15 nucleotides as the inserted target site in BmAntp from nucleotides 1076 to 1090 of its cDNA sequence (Figure 3c). Consequently, the insertional event in the Wes mutant was consistent with the hallmarks of the retrotransposition process, including 5′ truncations, the presence of an oligo(dA)-rich tail at the 3′ end and target site duplications of 2–20 bp in length (Cordaux and Batzer, 2009).

Expression level of the BmAntp gene

To assess the relationship between the mutant phenotype (wedge eye-spot) and the BmAntp mutated state (heterozygotes), we used qRT-PCR analysis to quantify BmAntp levels using cDNA prepared from the whole body of fifth instar larvae. The amplified sequences of primer set Antp-C were shared by the two types of BmAntp cDNA, whereas the primer set Antp-M was only for the mutated BmAntp transcript (Figure 3). Interestingly, the result of the qRT-PCR showed that the expression level of Antp-C in the Wes mutant (Wes/+Wes) was markedly higher than that in the wild type (+Wes/+Wes). Considering the expression level of Antp-C is hundreds of times higher than that of Antp-M in the heterozygotes, the elevated Antp-C expression levels in the mutant (heterozygotes) should be due to the overexpression of wild-type Antp (Figure 4).

Figure 4
figure 4

Quantitative analysis of BmAntp in +Wes/+Wes and Wes/+Wes larvae. The bars indicate mean±s.d. (n=3). **P<0.01, Student’s t-test.

Full size image

Functional validation, RNAi of BmAntp

The B. mori larval thorax is composed of prothoracic, mesothoracic and metathoracic segments, and there is a pair of thoracic legs on each segment. The Wes homozygote has a deformed thorax. To validate whether a decrease of the level of BmAntp type 1 (wild type) mRNA is largely responsible for determining the thoracic patterning of B. mori, we synthesized dsRNA and injected it into eggs at 2–6 h after oviposition to suppress the expression of the BmAntp gene in the wild type (Dazao). We observed that very few eggs hatched successfully after the injection of dsBmAntp into eggs. By contrast, most of the eggs injected with doubly distilled water (ddH2O) and dsEGFP under the same conditions (control samples) hatched after incubation. In the BmAntp dsRNA-treated group, a few embryos survived until the late embryonic stage and 20.11% of these exhibited morphological alteration (Table 2). Some of them showed fused thoracic segments and the lack of a mesothoracic leg (Figure 5b), whereas others exhibited three reduced thoracic legs on one side, only remaining visible vestiges (Figure 5c). qRT-PCR revealed that the expression of BmAntp in the morphologically changed individuals injected with BmAntp dsRNA was significantly lower compared with that in the controls (Figure 5d).

Table 2 Phenotypic analysis of BmAntp RNAi experiment
Full size table
Figure 5
figure 5

Effect of RNAi on BmAntp. (a) Wild type (Dazao) embryo injected with doubly distilled water (ddH2O). Bar=0.5 mm. (b and c) RNAi effects on identities of thoracic segments in embryos. After injection of dsBmAntp into the embryo, thoracic segments fused and thoracic legs were reduced partly. Bar=0.5 mm. (d) Expression levels of BmAntp in the embryo injecting ddH2O, dsEGFP and dsBmAntp. ddH2O and dsEGFP-injected controls are shown for comparison. The bars indicate mean±s.d. **P<0.01, Student’s t-test.

Full size image

Discussion

Wes is a newly discovered mutant located on the sixth linkage that exhibits wedge eye-spots in the larvae and causes homeotic changes and lethality in homozygous embryos. These characteristics of Wes mutant provide excellent subjects for studying the developmental effects of mutant genes. An attractive strategy for identifying the gene corresponding to dramatic variations is positional cloning using B. mori genome information. Here, on the basis of our data, we confirmed that variations of BmAntp were responsible for the Wes phenotype.

In insects, the Antp gene is expressed mainly in the thorax and is associated with development of the thoracic legs (Hughes and Kaufman, 2002). Earlier studies in the beetle Tribolium castaneum showed that mutations in ptl, the Antp homolog, cause the transformation of the three pairs of thoracic legs to antennae (Beeman et al., 1989). Similarly, loss-of-function mutations in Antp in D. melanogaster produced a homeotic transformation from the meso- and metathorax to prothorax (Wakimoto and Kaufman, 1981), and Antp is required for leg identity (Emerald and Cohen, 2004). In insects, the Antp gene is highly conserved in its morphological homeotic function, specifying segment identity. In a non-insect arthropod, the spider A. tepidariorum, the Antp gene represses the development of legs in the first segment of the abdomen (Khadjeh et al., 2012). Our data showed that homozygous mutation in BmAntp produced the fused thorax and antennae-like appendages between the head and the thorax (Figure 1). BmAntp RNAi leads to fused thoracic segments and defects in the development of thoracic legs (Figure 5). Given that Wes homozygotes and BmAntp-dsRNAi embryos displayed similar mutant phenotypes, these should be owing to the loss-of-function of BmAntp. Our results show that BmAntp has an important role in thoracic segmentation. Considering the reduced thoracic legs in BmAntp-dsRNAi embryos and antennae-like appendages in the Wes homozygotes, we speculate that BmAntp is required for both development and identity of thoracic leg in B. mori.

Notably, qRT-PCR analysis revealed that the expression level of wild-type BmAntp gene was increased markedly in the heterozygous larvae compared with the wild type (Figure 4). This observation could reflect the fact that a negative feedback loop mechanism for Antp regulation has been disrupted with the disruption of a functional Antp protein. However, given that we do not know the protein level of BmAntp in the mutated state (heterozygotes), whether the mutant phenotype is owing to the BmAntp overexpression state still needs further research. Subsequently, the overexpression state of BmAntp mRNA in heterozygotes raises the question of why heterozygotic embryos do not show any mutant phenotype. One possibility is that the negative feedback loop has not worked in the early embryonic stage, because the thoracic legs are developing as early as ∼48 h after oviposition. Another possibility is that the amount of BmAntp protein is not sufficient to cause visible developmental defects at this stage.

A previous study reported that BmAntp lacked the homeodomain and its downstream region in the B. mori Nc mutant, showing homeotic changes of prothoracic legs into antennae and compression between the prothoracic and mesothoracic segments (Nagata et al., 1996). However, the transformations in Wes homozygotes, such as the appendages under the mouthparts (Figure 1e) and the fused thoracic segments (Figure 1j) are different, to some extent, from that in Nc homozygotes. Why does mutation of the same gene produce different extents of homeotic transformation? Most likely, the normal transcript in the Wes strain is still present, even at a very low level, but not in the Nc mutant. As in the previous reports that transposable element insertions in exons can function as new introns and be spliced out from pre-mRNA (Giroux et al., 1994), here, we suppose that two different Antp transcripts in a Wes/Wes homozygote mutant may also be due to the alternative splicing. As discussed above, an alternative explanation is that the abnormal transcript might simultaneously encode a new protein. Given that the speculative aberrant protein does not encode the homeodomain, resulting in inability to bind original target genes of BmAntp, we tend to favor the earlier explanation for the difference of phenotype between Wes and Nc mutants. In summary, our data imply that precise expression of the Antp gene is required for organization of thoracic segments.

Although Hox genes have been shown to regulate many downstream genes, only a limited number of Hox-regulated target genes have been identified (Choo and Russell, 2011). An earlier report showed that the Ubx gene regulated pigmentation in the hindwings of the butterfly Precis coenia (Weatherbee et al., 1999). In addition, an intriguing case in which Hox protein abdominal-B (abd-B) directly activated expression of the yellow pigmentation gene in the D. melanogaster, demonstrated the relevance between Hox genes and pigmentation genes (Jeong et al., 2006). Consideration of either phenomenon should be focused on the relation between pigmentation and other Hox genes. As mentioned above, the Wes eye-spot area had become smaller compared with that in the wild type. These results give rise to the hypothesis that BmAntp might participate in the thoracic pigmentation directly or through regulation of its downstream target genes. On the basis of these results, our future work will concentrate on the regulation of BmAntp and its target genes, as well as their function in pigmentation.

Data Archiving

Sequence data have been submitted to GenBank: accession numbers KC165845–KC165846.