A Baculovirus Immediate-Early Gene, ie1, Promoter Drives Efficient Expression of a Transgene in Both Drosophila melanogaster and Bombyx mori

Many promoters have been used to drive expression of heterologous transgenes in insects. One major obstacle in the study of non-model insects is the dearth of useful promoters for analysis of gene function. Here, we investigated whether the promoter of the immediate-early gene, ie1, from the Bombyx mori nucleopolyhedrovirus (BmNPV) could be used to drive efficient transgene expression in a wide variety of insects. We used a piggyBac-based vector with a 3xP3-DsRed transformation marker to generate a reporter construct; this construct was used to determine the expression patterns driven by the BmNPV ie1 promoter; we performed a detailed investigation of the promoter in transgene expression pattern in Drosophila melanogaster and in B. mori. Drosophila and Bombyx belong to different insect orders (Diptera and Lepidoptera, respectively); however, and to our surprise, ie1 promoter-driven expression was evident in several tissues (e.g., prothoracic gland, midgut, and tracheole) in both insects. Furthermore, in both species, the ie1 promoter drove expression of the reporter gene from a relatively early embryonic stage, and strong ubiquitous ie1 promoter-driven expression continued throughout the larval, pupal, and adult stages by surface observation. Therefore, we suggest that the ie1 promoter can be used as an efficient expression driver in a diverse range of insect species.


Introduction
Recent advances in germ-line transformation of non-drosophilid insects have opened up new opportunities for analyzing gene function. However, in non-model insects, a major obstacle to such analyses is the dearth of useful promoters for transgene expression. Reportedly, various promoters from several insect species can be used to express marker genes or foreign genes in transgenic insects (see [1]). The piggyBac vector is the most widely used transformation vector in insects, and many promoters have been used to drive gene expression from this vector. For example, the 3xP3 promoter [2,3], Bombyx mori cytoplasmic actin (BmA3) promoter [4,5,6], Gryllus bimaculatus cytoplasmic actin promoter [7]. Drosophila melanogaster heat shock protein 70 (hsp70) promoter [6,8,9], B. mori nucleopolyhedrovirus (BmNPV) immediate-early gene 1 (ie1) promoter [10], Autographa californica multiple nucleopolyhedrovirus (AcMNPV) ie1 promoter [11,12], Tribolium castaneum a-Tubulin1 promoter [13], B. mori and Antheraea yamamai fibroin H chain promoter [14], B. mori sericin 1 promoter [15], B. mori cecropin B promoter [16], B. mori bombyxin and prothoracicotropic hormone (PTTH) promoter [17], and D. melanogaster polyubiquitin promoter [18,19,20] can each drive gene expression from the piggyBac vector. Unlike most of these promoters, which are derived from naturally occurring, constitutively expressed genes, the 3xP3 promoter was constructed artificially based on the sequence of Pax-6 binding sites; Pax-6 is an evolutionary conserved transcription factor that functions in eye development in metazoans, and the 3xP3 promoter drives expression in visual systems [21,22]. Based on studies of transgenic animals, the eye-specific expression driven by 3xP3 promoter is highly conserved not only in diverse insect orders of Diptera, Lepidoptera and Coleoptera [1], but also in planarians [23].
While various promoters have been used to drive transgene expression in some insects, a more versatile and efficient promoter is necessary for expression of transgenes in a wider variety of insects beyond insect orders. Here, we examined the expression pattern driven by the BmNPV ie1 promoter in fly and silkworm as a candidate for an efficient expression driver for a wide variety of insects. The ie1 gene is present in all lepidopteran nucleopolyhedrovirus (NPV) genomes that have been sequenced; the ie1 gene product acts as an transactivator of some early viral genes and is essential for viral DNA replication ( [24], reviewed in [25,26,27]). The ie1 promoter contains a TATA box and is transcribed via RNA polymerase II by host cells upon baculovirus infection; therefore, the promoter itself can function in uninfected cells that lack other virus-encoded factors [28]. The ie1 promoter is a strong promoter that is used to express foreign genes in cell lines derived from various insect species [29,30]. Therefore, we focused on using the ie1 promoter to express transgenes at high levels in insects, and to this end, we examined the baseline in vivo expression pattern driven by the ie1 promoter. Several viral factors (e.g., IE0, IE1, and IE2) regulate transcription from the ie1 promoter [31,32]; moreover, the ie1 promoter is reportedly activated by hormones of host insects. Treatment of uninfected insect cells in culture or of fifth instar silkworm larvae with ecdysone, a juvenile hormone analogue (JHA), or both resulted in increased ie1 promoter activity [33]. Furthermore, the ecdysone-responsive cis-acting elements have been identified within the BmNPV ie1 promoter by using recombinant BmNPVs [34].
The AcMNPV ie1 promoter fused with homologous region 5 (hr5) enhancer from the same virus was used to drive expression of enhanced green fluorescent protein (EGFP) in the malaria mosquito, Anopheles gambiae [12], and DsRed in the Mediterranean fruit fly, Ceratitis capitata [11]. Furthermore, the BmNPV ie1 promoter has been used to drive expression of the B. mori doublesex gene [10]. However, basal transgenic (i.e., in the absence of baculovirus infection) ie1 promoter activity is still largely uncharacterized in vivo. Here, we examined in detail the expression patterns driven by the BmNPV ie1 promoter in D. melanogaster and B. mori; these species belong to different insect orders (Diptera and Lepidoptera, respectively). Surprisingly, the ie1 promoter drove very similar expression patterns in these two insect species. For both insects, EGFP fluorescence driven by the ie1 promoter was first present during embryogenesis; also for both insects, strong ie1 promoter-driven EGFP fluorescence was evident throughout the body in each developmental stage that followed larval hatching. Thus, we suggest that the ie1 promoter may be useful for efficient expression of transgenes in a wide variety of distantly related insect species.

Materials and Methods
Insects D. melanogaster were reared on a standard medium at 25uC. Mutant embryos, yellow white (y ac w 1118 ), were injected with DNA to generate germ-line transformants and transgenic fly strains.
A bivoltine strain (Daizo) of the silkworm, B. mori, was used for DNA injections and to generate germ-line transformant and transgenic strains. Silkworm larvae were reared on a diet of fresh mulberry at 25uC. The silkworm embryos were staged according to morphological markers as described by Ohtsuki [35] and Morita et al. [36].

Plasmid Construction
We constructed a piggyBac-based vector to use in reporter assays of BmNPV ie1 promoter activity (pBac[BmNPV ie1-EGFP, 3xP3-DsRed]; Fig. 1). To create restriction sites at both ends of the BmNPV ie1 promoter (-631 , -2 from the translational initiation Met codon), polymerase chain reaction (PCR) was performed using BmNPV genomic DNA as a template, a primer set comprising the RE-ie1-1 primer (59-AAGCTTA-GATCTGGCCGGCCGATTTGCAGTTCGGGACAT-39) with Hind III-Bgl II-Fse I site at 59 end and the RE-ie1-2 primer (59-CCATGGTCGTTTGGTTGTTCACGAT-39) with Nco I site at 59 end, and a high-fidelity DNA polymerase (Pyrobest; Takara, Japan). The resultant PCR product was subcloned into the EcoR V site of the pBluescript KS+ vector, and the expected nucleotide sequences were confirmed by DNA sequencing. This recombinant plasmid was then cut with Bgl II and Nco I, and the fragment containing BmNPV sequences were inserted into the BamH I-Nco I site of the pSL[hsp27mp-NLS-EGFP] vector [37]. Since BamH I and Nco I sites were contained at the both ends of the hsp27mp-NLS fragment, the hsp27mp-NLS fragment was removed from pSL[hsp27mp-NLS-EGFP] by cutting with BamH I and Nco I. Thus, the BmNPV ie1 promoter directly connected to the start codon of EGFP. A Asc I-Fse I fragment containing the BmNPV ie1-EGFP sequences was treated with T4 DNA polymerase to create blunt ends and then ligated to the Asc I sites (which had be filled in) of pBac[3xP3-DsRedaf] [1]; the resultant recombinant construct was designated pBac[BmNPV ie1-EGFP, 3xP3-DsRed].

Sequencing and Sequence Analysis
For each final recombinant construct, sequences derived from PCR products or from fragments cut with a restriction enzyme (i.e., any sequence that might have been altered during cloning procedures) were confirmed using the dideoxy chain-termination method by an automatic DNA sequencer (CEQ 2000XL; Beckman Coulter, USA). Sequence analysis was carried out using a DNASIS system (Hitachi Software Engineering, Japan).

Generation of Germline Transformants
To obtain transgenic fruit flies and transgenic silkworms that carried the BmNPV ie1 promoter construct, we injected plasmid mixtures containing 500 ng/ml of the BmNPV ie1 promoter construct and 300 ng/ml of a helper plasmid (pshp-pBac, [18]) in injection buffer (5 mM KCl, 0.1 mM K 2 HPO 4 , pH 7.8), into the posterior pole of Drosophila embryos or into the posterior-ventral midline of Bombyx embryos during the syncytium stage of development (2 h and 12 h after oviposition at 25uC, respectively). Microinjections were performed under a dissection microscope (Stemi 2000, Carl Zeiss, Germany) using a micromanipulator (Narishige, Japan) and FemtoJet (Eppendorf, Germany), and injections into Bombyx embryos were performed with a special glass needle (uMPm-02, Daiwa Union, Japan) [38]. Transgenic flies were generated as described by Rubin and Spradling [39]. To identify Drosophila transformants, G 1 adults were screened for expression of the 3xP3-driven DsRed transformation marker in the eyes. To identify B. mori transformants, late stage G 1 embryos were screened for DsRed expression in the central nervous system (CNS). After G 2 generations, we screened EGFP positive embryos at each generation to maintain the transgenic B. mori lines. All B. mori transgenic lines showed similar patterns and intensities of EGFP expression under the control of the ie1 promoter. The transgenic D. melanogaster lines and the transgenic B. mori lines have stably transmitted the transgene for more than 80 and 10 generations, respectively. Inverse PCR Genomic DNA was extracted from adult flies or diapausing silkworm eggs using the QIAamp Tissue Kit (Qiagen, Germany). The following procedures for inverse PCR were performed as described in a previous study [40].

Construction and Transformation of the BmNPV ie1
Reporter Construct Using the piggyBac Vector To characterize the expression profile of the BmNPV ie1 promoter in transgenic insects, we constructed a piggyBac-based vector (pBac[BmNPV ie1-EGFP, 3xP3-DsRed]; Fig. 1) to use in reporter assays. A BmNPV ie1 promoter fragment containing sequences from -631 to -2 upstream of the codon encoding the ie1 translational start site was used to drive expression of EGFP. This promoter region from the BmNPV ie1 gene contained promoter regions that have been well characterized [33,34] and showed 96.2% identity to the promoter region from the AcMNPV ie1 gene [41], which has been used for expression of foreign genes [11,12]. We used DsRed driven by the 3xP3 element as a transformation marker. The 3xP3 element drives transgene expression in Drosophila ocelli, ommatidia, Bolwig organs, and CNS [2,37,42]; this element also drives expression in the Bombyx stemmata, CNS, and peripheral nervous system (PNS) beginning on the fifth day of embryonic development [3].
We generated transgenic flies and transgenic silkworms with this piggyBac-based construct. In D. melanogaster, of 401 injected embryos, 153 larvae hatched into 74 eclosed adults, 50 of which were fertile. These adults produced G 1 progenies when individually crossed with yw adults. One G 0 cross gave rise to several DsRed exressing progenies, resulting in a transformation efficiency of 2.0%. One line is homozygous viable and used for subsequent EGFP expression analysis.
In B. mori, of 275 injected embryos, 23 larvae hatched into 14 eclosed fertile adults. These G 0 adults were intercrossed and their progenies were subjected to screen transformants. One G 0 adult gave DsRed positive progenies, resulting in a transformation efficiency of 7.1%. At least 3 lines were recovered from G 1 progenies because of the multiple insertions occurred at the transgenic G 0 . These multiple insertions of a piggyBac vector are frequently occurred in B. mori [4]. As we did not detect any  difference in DsRed expression pattern except the intensity of DsRed expression, we did not detect any difference in EGFP expression patterns among the transgenic lines. Therefore, we maintain one transgenic line for further analysis.
We examined the genomic DNA sequences surrounding the piggyBac insertion site of D. melanogaster and B. mori by inverse PCR. As shown in Fig. 2, the TTAA target site of piggyBac was found at both ends of the piggyBac insertion site for all individuals, and the characteristic duplication of TTAA, which occurs following piggyBac-mediated transposition, was also observed. This result indicates that piggyBac had precisely integrated into the D. melanogaster and B. mori genome.

EGFP Expression Driven by the BmNPV ie1 Promoter in Transgenic Flies
We examined EGFP expressed from the BmNPV ie1 promoter in transgenic flies by surface observation (Fig. 3). EGFP expression was detected during late embryogenesis. In stage 16 embryos (13-16 h after fertilization), EGFP fluorescence was mainly detected in midgut and the PNS (Fig. 3A). In wandering 3rd instar larvae, strong EGFP expression was evident throughout the body (Fig. 3B). In contrast, DsRed expression was observed in the Bolwig organ, CNS, hindgut, and anal plate as previously described ( Fig. 3B; [37,42]). In pupa, EGFP expression was evident throughout the whole body; punctate, strong expression was also evident in some EGFP was also detected along tracheoles that attached to the midgut. (C) Central nervous system and ring gland. EGFP expression was strong in the prothoracic gland region of the larval ring gland. EGFP signals were detected in the peripheral nerves emanating from the abdominal neuromere. DsRed driven by the 3xP3 promoter was expressed in the ventral nerve cord and throughout the brain except within the optic lobe. The lower right panel shows the merged images of transmitted light, EGFP fluorescence, and DsRed fluorescence. In all panels, anterior is to the left. Abbreviations: br, brain; fg, foregut; gc, gastric caecum; hg, hindgut; mg, midgut; mt, Malpighian tubule; pn, peripheral nerves; pv, proventriculus; rg, ring gland; VNC, ventral nerve cord. Scale bars = 100 mm. doi:10.1371/journal.pone.0049323.g004 cells within the body, and the expression along the dorsal vessel was stronger than the expression in other tissues (Fig. 3C, D).
To identify more precisely the tissue that expressed EGFP, we dissected 3rd instar larvae and observed the EGFP fluorescence using a fluorescent microscope (Fig. 4). EGFP expression was evident in the foregut, gastric caecum, and a section of Malpighian tubule (Fig. 4A, B). Also EGFP expression was evident along tracheoles that attached to the midgut, and in the peripheral nerves within the abdominal neuromere (Fig. 4A, B, C). Interestingly, EGFP expression was strong in the prothoracic gland region within the larval ring gland (Fig. 4C). DsRed was expressed in the ventral nerve cord and throughout the brain except for the optic lobe (Fig. 4C).
We also compared the expression pattern driven by the ie1 promoter with that of a widely used, constitutive D. melanogaster promoter derived from the act5C gene. This act5C promoter was used to drive expression of S65T-GFP and is widely used as a GFP marker on a balancer chromosome [43]. A precise comparison of the strength of the ie1 promoter with that of the act5C promoter as it is in TM3-pAct-GFP flies was difficult because the two GFP variants fluoresce with different intensities and because the transgenes were inserted at different genomic locations; nevertheless, this comparison could provide a clue to the utility of ie1 promoter because the TM3-pAct-GFP flies are so widely used [43]. EGFP fluorescence driven by the BmNPV ie1 promoter was much stronger than S65T-GFP fluorescence driven by the D. melanogaster act5C promoter in D. melanogaster third instar larva (Fig. 5).
EGFP fluorescence was also observed in adult flies and strong expression was detected in abdomen (data not shown).

EGFP Expression Driven by the BmNPV ie1 Promoter in Transgenic Silkworms
In stage 17 embryos (appendage development in the gnathal and thoracic segments), no EGFP expression was detected (Fig. 6A); EGFP expression was first detected in some yolk cells beginning in stage 18 (Fig. 6B). By stage 20 (gradual gathering of gnathal appendages), EGFP expression was detected in the lateral and dorsal regions, although the dorsal region was not completely formed in these stage 20 embryos (Fig. 6C). In stage 21B (middle blastokinesis), additional fluorescence was evident in the thoracic legs, and this fluorescence extended from the dorsal to the ventral region (Fig. 6D). Additionally, expression of DsRed, the transformation marker, was evident in the developing ventral nerve cord of stage 21B embryos (Fig. 6E). In stage 23-26 embryos, EGFP expression was evident throughout the whole embryo (Fig. 6F, G), and DsRed was detected only in the stemma and CNS (Fig. 6G'; [3]). Beginning in stage 27 embryos, EGFP and DsRed fluorescences were obscured during surface observation because the body, especially the head capsule, became pigmented (Fig. 6H, H').
At each larval instar, strong EGFP signal was detected throughout the whole body by surface observation (Fig. 7A, B, C, D', E). We never observed EGFP signal in wild type control larvae (Fig. 7D, D'). In first instar larva, EGFP signal was evident in stemmata, in addition to throughout the body (Fig. 7A), and DsRed signal was evident in the ventral nerve cord and stemmata (Fig. 7A'). Interestingly, EGFP signal did not overlap with DsRed signal within the ventral nerve cord.
To identify more precisely the organs that expressed EGFP, 4th and 5th instar larvae were dissected and individual organs were examined. In 4th instar larva, EGFP expression was strong in the anterior and the posterior midgut (Fig. S1B). To our surprise, EGFP expression was also strong in the pair of prothoracic glands (Fig. S1B, C). EGFP was evident in the middle and posterior part of the silk gland (Fig. S1D, E) and in the tracheole attached to the silk gland (Fig. S1F).
In 5th instar, like in 4th instar, EGFP expression was so strong in the anterior and the posterior of the midgut (Fig. 8B') that the fluorescence was detectable event through the larval cuticle (Fig. 7E). After removing the midgut, EGFP fluorescence was detected in the pair of prothoracic glands (Fig. 8C), the suboesophageal body (Fig. 8D), and in pericardial cells along dorsal vessel and on the alary muscle (Fig. 8J'). EGFP expression was also evident in peritracheal athrocytes, but not in fatbody (Fig. 8J'). EGFP expression was not evident in either the testis (data not shown) or ovary (Fig. 8H'), but it was evident in tracheolar cells that were attached to the ovary (Fig. 8H'). Furthermore, EGFP signal was particularly strong in the trichogen (or trichogen and tormogen) cells in the epidermis (Fig. 8F').
In pupa, BmNPV ie1 promoter-driven expression of EGFP was evident throughout the body, especially in wings and abdomen by surface observation (Fig. 9A', B, C). DsRed expression was evident in the compound eyes, but EGFP expression was not (Fig. 9D).  After eclosion, EGFP fluorescence was evident, but detection of fluorescence was difficult because thick scales covered the entire body (data not shown).

Discussion
In this study, we examined the expression patterns driven by the BmNPV ie1 promoter in D. melanogaster and B. mori. We found that, for both insects, strong EGFP fluorescence driven by the ie1 promoter was evident throughout the whole body and throughout all developmental stages that followed larval hatching. These findings indicated that the ie1 promoter was an efficient and strong promoter in both insect orders and, therefore, that it may be useful for heterologous gene expression or for expressing a transformation marker.
Currently, the cytoplasmic actin promoters are some of the most widely used constitutive promoters in transgenic insects. The BmA3 promoter was first used as a marker of stable germ-line transformation in B. mori [4]; moreover, the BmA3 promoter functions in the pink bollworm, Pectinophora gossypiella [5] and even in the sawfly, Athalia rosae, a Hymenoptera species [6]. Although the BmA3 promoter successfully drives marker gene expression in these two heterologous insect species, the usefulness of this actin promoter may yet be species dependent. Reportedly, the promoter from the cytoplasmic actin (GbA3/4) gene of the two-spotted cricket (G. bimaculatus) was approximately three-fold and 80-fold more active than the promoters from the D. melanogaster act5C and the D. melanogaster hsp70 genes, respectively, in a transient expression assay in G. bimaculatus embryos [44]. Therefore, cytoplasmic actin promoters in heterologous species may be less efficient than that in original species. The BmA3 promoter has been used as a transformation marker in some insects (Lepidoptera and Hymenoptera). However, GFP driven by the BmA3 promoter is evidently not expressed in B. mori embryos [4], and BmA3 promoter-driven GFP expression does not begin until the pharate larval stage in P. gossypiella [5]. In contrast, EGFP fluorescence driven by the G. bimaculatus cytoplasmic actin promoter is detected from the early blastoderm stage onward in transgenic crickets [7]. Although it would be interesting to know what causes the difference between the initiation timing of the expression driven by the BmA3 promoter and that driven by G. bimaculatus cytoplasmic actin promoter, this is another example of a species specificity of the cytoplasmic actin promoter. A promoter that drives gene expression late in development is not particularly useful as a marker of transformation because of the effort necessary to screen numerous hatched larvae to detect a transformant. Here, we found that, in D. melanogaster and in B. mori, the BmNPV ie1 promoter drove strong EGFP expression from late and middle embryogenesis, respectively, throughout adulthood; therefore, we believe that this promoter may be useful for driving expression of transformation markers in a range of distantly related insect species.
Thus far, the artificial 3xP3 promoter has been the most versatile promoter for expression of transformation markers in a wide variety of insect taxa (Diptera, Lepidoptera and Coleoptera). In B. mori, 3xP3-driven expression is evident much earlier than is BmA3 promoter-driven expression; 3xP3-driven expression begins on the fifth day of embryonic development at 25uC [3]. However, the BmNPV ie1 promoter constructed for this study had advantages in the following respects. BmNPV ie1 promoter-driven expression was evident beginning at embryonic stage 18 (3.0,3.5 days after oviposition), which is nearly two days earlier than the first discernable 3xP3-driven expression. More- over, expression from the ie1 promoter was not evident in the nervous system, which is the site of 3xP3-driven expression; thus, the ie1 promoter-driven transformation marker is compatible with 3xP3-driven transformation marker. Furthermore, the BmNPV ie1 promoter functioned not only in B. mori, which is the host of BmNPV, but also in D. melanogaster.
Transformation markers expressed from the AcMNPV ie1 promoter with its hr5 enhancer were assessed in A. gambiae [12] and in C. capitata [11]. Although the expression patterns of this transformation marker were not described thoroughly for either species, expression of this marker in A. gambiae was evident throughout the larval thorax and abdomen and was particularly strong in the anterior stomach and salivary glands [12]; moreover, strong expression was evident throughout the larval body in C. capitata [11]. Here, both in D. melanogaster and in B. mori, BmNPV ie1 promoter-driven expression was evident throughout the body of larvae and pupae, especially in the prothoracic gland, the midgut, and the tracheole. Although these two insects belong to different insect orders (Diptera and Lepidoptera) that diverged at about 240 million years ago [45], ie1 promoter-driven expression was detected in several of the same tissues in both species (Fig. 4,  8). It is astonishing that this promoter, which is approximately 600 bp, has evolutionarily conserved tissue-specific regulatory elements. Since NPVs infect not only Lepidoptera, but also Diptera and Hymenoptera [46], the evolutionary conserved expression pattern of the BmNPV ie1 driven EGFP marker in flies and silkworms may be concerned with something common in regard to infection by NPVs.
This ie1 promoter may facilitate identification of cis-elements that function in multiple insect orders. Reportedly, the ie1 promoter responds to ecdysone (E) and 20-hydroxyecdysone (20E) [33,34]. Furthermore, Kojima et al. [34] identified an ecdysone-responsive element in the ie1 promoter (ie1EcRE, 59-GTGTTATCGACCT-39), and this element is homologous to the ecdysone response element of D. melanogaster (DmEcRE). The active form of ecdysteroid is 20E; ecdysteroid is first synthesized as E in the prothoracic gland or ring gland, secreted into hemolymph, and then enzymatically converted into 20E in peripheral tissues [47]. Our finding, specifically that the ie1 promoter drove EGFP expression in prothoracic gland of B. mori and the ring gland of D. melanogaster, might be related to E synthesis. Our results might provide a clue to the identification of the transactivators that bind to the ie1 promoter. Furthermore, the high expression of EGFP driven by the BmNPV ie1 promoter in the prothoracic glands indicated that the transgenic silkworms with the BmNPV ie1-EGFP reporter can be useful for automated sorting of the cells of prothoracic glands, which are small organs and difficult to isolate by dissection.
In conclusion, the BmNPV ie1 promoter will be useful especially in silkworms for a strong expression driver not only for functional analysis tools such as reverse genetics, tetracycline-controlled transcriptional activation system and in vivo RNA interference but also for a tool for future application as bioresources. Moreover, the ie1 promoter can be used as an efficient expression driver in a diverse range of insect species and it will be useful for future basic and applied research studies.