miR-147b-modulated expression of vestigial regulates wing-morph differentiation in the bird cherry-oat aphid Rhopalosiphum padi

The wing polyphenism occurs under crowding and nutrition-deficiency conditions in most aphid species. Although the influence of environmental factors on wing polyphenism of aphids have been extensively investigated, molecular mechanisms underlining wing-morph differentiation has not been fully understood. The expression levels of the twenty genes involved in wing patterning network were examined, and only vestigial (vg) showed significantly different expression levels in both whole-body and wall-body of third nymphal instars, with 5.4- and 16.14- fold higher in winged lines compared to wingless lines, respectively in Rhopalosiphum padi. Moreover, vg expressions were higher in winged aphids compared to that in wingless aphids of third, fourth instar nymphs and adults, and larger difference ratio were observed in third (21.38-fold) and fourth (20.91-fold) instar nymphs relative to adult (3.12-fold) between wing morphs. Suppression of vg using RNAi repressed the wing development of third winged morphs. Furthermore, modulation of miR-147b levels by microinjection of its mimics decreased vg expression levels and repressed wing development. Our findings suggest that vg is essential for wing development and that miR-147b modulates its expression. To our knowledge, our results provide an empirical evidence that miRNA is involved in the regulation of wing morphs in aphids. padi . The role of vg in wing differentiation in R. padi was further investigatedacross the nymphal instars and adult and by vg RNAi. Our results demonstrate that vg is essential for wing development during wing differentiation in R. padi . More importantly, the expression of vg is regulated by miR-147b. These findings provide strong empirical evidence that at least one miRNA is involved in the regulation of wing morphs in aphids. In this study, the expression levels of the 20 genes in the wing patterning network were examined, and only vg showed significantly different expressions in both whole body and the body wall in third instar nymphs between the two wing morphs of R. padi . The role of vg in wing differentiation in R. padi was further investigated across the nymphal instars and adult and by vg RNAi. Our results demonstrate that vg is essential for wing development during wing differentiation in R. padi. The results indicate that vg is involved in wing differentiation and is overexpressed in the winged morphs due to increased transcription rather than vg gene duplication in R. padi. We have shown that vg transcription is post-transcriptionally regulated by miR-147b, which binds to its target sequence present in the vg mRNA. These results provide an empirical evidence that miRNA is involved in the regulation of wing morphs in aphids. the The third instar winged aphids were selected for injection. Each aphid was injected with 13.8 nL of a 40 μM agomir solution, and the control was injected with agomir-NC. At 24 h post-injection, the 20 nymphs in each sample were collected and total RNA was extracted using TRIzol. The relative expression levels of vg and miR-147b were determined using qRT-PCR. The wing morphs were recorded 48 h after injection. All experiments were performed in triplicate.


Abstract Background
The wing polyphenism occurs under crowding and nutrition-deficiency conditions in most aphid species. Although the influence of environmental factors on wing polyphenism of aphids have been extensively investigated, molecular mechanisms underlining wing-morph differentiation has not been fully understood.

Results
The expression levels of the twenty genes involved in wing patterning network were examined, and only vestigial (vg) showed significantly different expression levels in both whole-body and wall-body of third nymphal instars, with 5.4-and 16.14-fold higher in winged lines compared to wingless lines, respectively in Rhopalosiphum padi. Moreover, vg expressions were higher in winged aphids compared to that in wingless aphids of third, fourth instar nymphs and adults, and larger difference ratio were observed in third (21.38fold) and fourth (20.91-fold) instar nymphs relative to adult (3.12-fold) between wing morphs. Suppression of vg using RNAi repressed the wing development of third winged morphs. Furthermore, modulation of miR-147b levels by microinjection of its mimics decreased vg expression levels and repressed wing development.

Conclusions
Our findings suggest that vg is essential for wing development and that miR-147b modulates its expression. To our knowledge, our results provide an empirical evidence that miRNA is involved in the regulation of wing morphs in aphids.

Background
All organisms show phenotypic plasticity depending on the environmental conditions they experienced [ 1 ]. Polyphenism is an extreme case of phenotypic plasticity in which discrete multiple phenotypes are produced based on a single genome [ 2 ]. Most aphid species exhibit wing polyphenism occurring during the asexual portion of the aphid life cycle, in which winged and wingless female are produced depending on environmental stimulus, e.g. population density and host nutrition [ 3 ]. The wingless morph specializes in reproduction, allowing rapid colony growth. In contrast, the winged form specializes in dispersal which enable aphids to seek out new habitats, mates, and food resources [ 4 ].
The bird cherry-oat aphid, Rhopalosiphum padi (L.), is one of the most globally abundant cereal aphid pests. In addition to directly feeding on plants, R. padi damages cereal crops by transmitting Barley yellow dwarf virus, which causes economically important crop losses [ 5 , 6 ]. R. padi, like most aphids, can produce wing dimorphism when experiencing the crowding and poor nutrition conditions [ 7 , 8 ]. It is easy to produce winged lines owing to its short life cycles and high reproductive rate [ 9 ]. Winged aphids are able to travel long distances and carry viruses in autumn which are considered as a major epidemiological factor for determining the disease incidence[ [10][11][12]. To date, the control of R. padi remains to rely on the application of chemical insecticides, which have leaded to insecticide resistance and environmental pollution [ 13 ].The fight abilities of winged forms and high fecundity of unwinged forms have made aphids control more difficult. Therefore, understanding the molecular mechanisms of wing morphs is important for controlling R. padi effectively.
Generally, wing morphs include determination and differentiation processes that occur at completely different time during aphid development. Mostly, morph determination occurs during embryogenesis in the maternal ovarian cavity and morph differentiation (i.e. wing development/degeneration) occurs during postembryonic development [ 14 ]. Nowadays, the influence of external cues on wing dimorphism have been extensively investigated, especially environmentally regulated maternal hormone in aphids can mediate phenotype production of next generation in the wing-morph determination. Recently, molecular mechanisms of ecdysone signaling controlling wing morph determination was discovered [ 15 ]. However, definitive molecular mechanisms in wing-morph differentiation, especially the roles of wing patterning genes at early wing development stages have been less reported.
Gene networks underlying wing patterns of Drosophila melanogaster have been well investigated. Based on the genomic sequences, the principal wing development gene homologs are largely conserved across holometabolous and hemimetabolous insect [ 16 ]. Brisson et al. [ 16 ] examined the expression levels of 11 genes involved in wing pattering, in which 6 genes showed significantly stage-specific expression level effects and only apterous1(ap1) exhibited different expression levels (< 4-fold) during the first and second nymphal instars in Acyrthosiphon pisum. This is the first and only report about the wing development gene repertoire in aphids, and did not consider the expression levels of wing patterning genes in the body wall, where the wing buds extend in wing differentiation. We depicted a gene network involved in major wing patterning events for R. padi, including anterior-posterior (A-P) patterning genes such as engrailed ( en), . In this study, the expression levels of the 20 genes in the wing patterning network were examined, and only vg showed significantly different expressions in both whole body and the body wall in third instar nymphs between the two wing morphs of R.
padi. The role of vg in wing differentiation in R. padi was further investigatedacross the nymphal instars and adult and by vg RNAi. Our results demonstrate that vg is essential for wing development during wing differentiation in R. padi. More importantly, the expression of vg is regulated by miR-147b. These findings provide strong empirical evidence that at least one miRNA is involved in the regulation of wing morphs in aphids.

Results
Expression profiles of wing patterning genes in wing morphs.
To determine which genes may be involved in wing differentiation during post-embryonic development in R. padi, we evaluated the expression levels of 20 known wing patterning genes ( Fig. 1) between wingless and winged lines in the third instar nymphs using qRT-PCR. Most of genes that contribute to wing patterning and development had similar expression levels in individuals between wingless and winged lines except for vg, in which expression was 5.4-fold higher in the winged line than in the wingless line (Fig. 2). Also, owing to wing bud extends from the body wall of thoracic part in winged line, these genes expression levels were also examined in body wall. Expression levels of vg, sal , omb and srf were significantly higher in body-walls. They were 16.14-, 3.16-, 4.07-and 2.77-fold higher in body-walls of winged lines, respectively, relative to body-wall of wingless lines.
Altogether, our results suggest that vg, sal, omb, and srf might play roles in wing differentiation in R. padi. The VG protein contains the Vg_Tdu domain, which is highly conserved among holometabolous and hemimetabolous insects (Additional file 1).
The expression patterns of vg were determined in different tissues, developmental stages, and wing morphs using qRT-PCR. The results showed that the relative levels of vg transcripts were the highest in the body wall of the winged lines (Fig. 3A) and the lowest in the body wall of the wingless lines (Fig. 3B). This strongly indicates that vg plays an important role in early wing development in R. padi.
Wing development in the winged lines is associated with various developmental stages, and the expression levels of vg were stable from the first to the second nymph stage, then increased sharply from the third nymph to the adult stage in the wingless morphs (Fig.   3C). In contrast, vg expression increased from the first to the third instar nymphs and then decreased in the adult stage in the winged morphs (Fig. 3D). Altogether, the highest expression of vg was found in the third nymphal instar, and it was 9.58-fold higher relative to the first instar nymphs, during winged nymph development.
The relative expression of vg was higher in winged aphids than in wingless aphids in the third and fourth instar nymphs as well as in adults (Fig. 3E), and higher difference ratios were observed in third (21.38-fold) and fourth (20.91-fold) instar nymphs compared with the adult (3.12-fold) between wing morphs. However, we observed no difference in the first and second instar nymphs between wing morphs. In addition, we also investigated the VG protein expression levels, and found that there were higher levels of the protein in the body wall of third nymphal winged lines compared to that in third nymphal wingless morphs (Fig. 3F). These results suggest that the third nymphal stage is the key period of aphid wing dimorphism and vg might be involved in wing differentiation.
RNAi knockdown of vg suppresses wing development.
RNAi experiments were performed to understand the relationship between wing development and vg gene expression. Third nymphal instar aphids of the winged lines were injected with dsRNA. At 24 h after injection with vg dsRNA, the mRNA levels of vg decreased significantly by 44% compared to control insects injected with dsEGFP ( Fig. 4A).
After 48 h, all aphids molted into adults, and injection of vg dsRNA resulted in 68% underdeveloped wing aphids compared to the dsEGFP control aphids, which were 100% normal ( Fig. 4B). The RNAi aphids showed under-developed wings (Fig. 4B). These results further demonstrate that vg plays an important role in wing development in R. padi.

miR-147b putatively regulates the expression of vg.
There was no significant difference in vg DNA expression levels between wing morphs from body walls of third instar nymphs (Additional file 2). Bioinformatic analysis predicted that miR-147b targets vg. In the vg ORF, there is one potential binding site that shows high sequence complementarity with miR-147b (Fig. 5A). The transcriptional levels of miR-147b in winged lines were significantly lower than in wingless lines, while its predicted target gene vg showed higher expression in winged lines relative to expression in wingless lines ( Fig. 5A).
Because aphid wing polyphenism is associated with colony density, we examined the effect of density on the expression levels of vg and miR-147b in third instar nymphs of the wingless morph. No significant differences in vg and miR-147b expressions were observed between the body walls of third nymphal instar wingless lines from LD and HD conditions (Additional file 3).
To determine whether miR-147b can bind to vg, the predicted target sequences of vg were inserted into the pmirGLO vector to construct the recombinant vector pmirGLO-miR-147b.
Firefly luciferase activity normalized against Renilla luciferase was significantly reduced when pmirGLO-miR-147b was co-transfected with the miR-147b agomir (mimic). However, the luciferase activity levels of the pmirGLOmiR-147b-mut construct were not dramatically affected by the miR-147b agomir compared with the unmutated constructs (Fig. 5C).
These results suggest that miR-147b can regulate the expression of vg by binding to the target sequence in the mRNA. miR-147b can modulate wing development.
To verify that the expression of vg is regulated by miR-147b, miR-147b agomir was injected into the third nymphal instars winged nymph larvae of R. padi, and we then examined the expressions of miR-147b and vg after 24 h, respectively. Compared with control group, expression levels of vg was decreased by 47% after injection for 24 h. Wing development was dramatically repressed in the group injected with the miR-147b agomir, which exhibited two types of phenotypes at rates of 75% and 25% (Fig. 6D); however, wing development in the control group injected with the dsRNA negative control was normal at rates of 100% after 48 h (Fig. 6C). These results demonstrated that miR-147b can affect vg expression and modulate wing development.

Discussion
The growth of the Drosophila wing has been well studied, and it is regulated by signaling . Optomotor-blind ( omb), the downstream target of dpp, is required for distal wing development in Drosophila [18]. Omb is expressed at higher levels in third nymphal instars of winged aphids compared to wingless aphids, while the genes from dpp signaling (i.e. dpp) show no significant differences between the two morphs. Whether the increased expression of omb results from higher levels of vg in winged aphids needs further verification. The other 16 genes had no significant morph effect in third nymphal instars.
Similar results were also observed by Brisson et al. (2010), who reported that the expression of en, hh, dpp, ubx, ap,wg, hth, and dll showed no significant differences between wing morphs of third instar nymphs in Acyrthosiphon pisum, but they did not examine expressions of vg, sal, omb, or srf. Although the expression levels of en, hh, sal, wg, exd, and Ubx were found to be significantly different between macropterous (migratory) and brachypterous forms of Nilaparvata lugens [ 29 ], our study showed that there are no significant differences in the expressions of these genes between aphid morphs. There is the possibility that brachypterous adults still have short wings, while wingless aphids have no wings because they degenerate by the between the wing morphs in this study (Fig. 2). Therefore, we hypothesized that vg We used bioinformatics to predict that miR-147b could potentially regulate the expression of vg. In humans, miR-147b regulates some cellular effects including proliferation, migration, and apoptosis [ 38 ]. Importantly, miR-147b is involved in endothelial barrier function and is a potent inducer of intestinal epithelial cell differentiation [ 39 , 40 ]. We found that vg expression is reduced and wing development is repressed after injecting the miR-147b mimic into R. padi. This is consistent with the target experiments in which the co-transfection of miR-147b mimics with the corresponding target plasmids significantly decreased the relative luciferase activity. Our results are the first to provide direct evidence that miR-147b-meditated regulation of vg expression controls wing development in aphids.
Although we determined here that vg plays an important role in wing differentiation in R.
padi, wing polyphenism is involved in both initial determination and subsequent differentiation [ 41 ]. Physical contacts (tactile stimulation) caused by crowding (high density) or poor nutrition can increase aphid dispersal [ 4 ]. In Nilaparvata lugens, two insulin receptors regulate wing bud development by responding to an insulin-like peptide secreted by the brain, and produce long-winged or short-winged forms [ 42 ]. Recently, the molecular mechanisms of ecdysone signaling in the control of wing morph determination were also determined in A. pisum [ 15 ]. High density has no effect on the expression of either vg or miR-147b in third instar nymphs (Additional file 3). Therefore, we propose a hypothesis to explain wing polyphenism in aphids that includes four processes; 1) environmental factors cause endocrine changes, 2) the increase in the hormone signal results in increased expression of miR-147b in wing primordia, 3) miR-147b negatively regulates expression of vg by binding to the mRNA, and 4) wing discs degenerate in the wingless lines owing to the lack of vg expression (Fig. 7). The opposite occurs in the winged lines, where vg is expressed at high levels in the wing primordia.

Conclusions
In this study, the expression levels of the 20 genes in the wing patterning network were examined, and only vg showed significantly different expressions in both whole body and the body wall in third instar nymphs between the two wing morphs of R. padi. The role of vg in wing differentiation in R. padi was further investigated across the nymphal instars and adult and by vg RNAi. Our results demonstrate that vg is essential for wing development during wing differentiation in R. padi. The results indicate that vg is involved in wing differentiation and is overexpressed in the winged morphs due to increased transcription rather than vg gene duplication in R. padi. We have shown that vg transcription is post-transcriptionally regulated by miR-147b, which binds to its target sequence present in the vg mRNA. These results provide an empirical evidence that miRNA is involved in the regulation of wing morphs in aphids.

Methods
Insects and cell line culture. Small RNAs were isolated from aphids using the miRNeasy Mini Kit (Qiagen, Germany) following the manufacturer's protocol. First-strand cDNA was synthesized from 2 μg of total RNA using the miScript II RT kit (Qiagen) as directed by the manufacturer. Genomic DNA was extracted from body wall of third nymphal wing morphs, and performed as previously described methods [ 44 ].
qRT-PCR was performed on an ABI 7500 Fast Real-Time PCR System (Applied Biosystems) using SYBR ® Premix Ex Taq™ II (Tli RNaseH Plus) kit (Takara, Japan). The cycling program for qRT-PCR assays for miRNA or mRNA was as follows: initial incubation at 50°C for 2 min and then at 95°C for 2 min, followed by 40 cycles of 95°C for 15 s and 60°C for 30 s according to the manufacturer's protocol. Analysis of the qRT-PCR data was carried out using the 2 − ∆∆Ct method of relative quantification. As an endogenous control, the EF-1α and U6 snRNA transcripts were used to normalize the expression level of mRNA (or DNA) and miRNA, respectively [ 44 , 45 ]. Three biological replicates over at least two days and two dish aphid samples were carried out for each test. All primers were designed based on information from the transcriptome library of R. padi and were listed in additional file 4.
Cloning and sequence analysis of vg.
Total RNA from a mixed sample consisting of 60 aphids from various developmental stages and morphs was isolated as described above. For amplification of a partial vestigial cDNA sequence, PCR primers were designed based on information from the transcriptome library Quantification of miR-147b and vg expression. miR-147b was predicted to target vg by the two miRNA target prediction programs miRanda (http://www.microrna.org/microrna/getDownloads.do) and RNAhybrid (http://bibiserv.techfak.uni-bielefeld.de/rnahybrid/welcome.html) using miRNA library of R.
padi. The expression of both miR-147b and vg were validated in the body walls of third nymphal wing morphs using qRT-PCR. A total of 20 individuals were used as a biological replicate for total RNA extraction and qRT-PCR, and three replicates were performed.
The agomir (mimic) of miR-147b was designed and synthesized by GenePharm Co. Ltd  The presumptive wing-patterning network in Drosophila [16,48]       The effect of miR-147b on wing development. The expression levels of miR-147b (A) and vg (B) in third instar nymph winged lines after injection of miR-147b agomir for 24 h, resepectively. Phenotypes of third nymphal winged aphid after injecting with agomir-NC (C) and miR-147b agomir (D) for 48h, D (i) and [43] phenotypes are at rates of 75% and 25%, respectively. *Significant difference according to Student's t-test (P < 0.05).