RNA interference (RNAi) is a cellular process by which an mRNA is targeted for degradation by a small interfering RNA that contains a strand complementary to a fragment of the target mRNA, resulting in sequence specific inhibition of gene expression. The discovery of RNAi enabled the use of loss-of-function analyses in many non-model insects other than Drosophila to elucidate the roles of specific genes. The RNAi approach has been widely used on insects in several fields, including embryogenesis, pattern formation, reproduction, biosynthesis and behavior. The increasing availability of insect genomes has made the RNAi technique an indispensable technique for characterizing gene functions in insects. Here we review the current status of RNAi-based experiments in insects and the applications of RNAi for species-specific insecticides, focusing on non-drosophilid insects. We also identify future applications for RNAi-based studies in Entomology.
The recognition of RNA interference (RNAi) as a method for analyzing the functions of individual genes culminated with the awarding of the Nobel Prize for the discovery of RNAi (Fire et al. 1998) to Andrew Z. Fire and Craig C. Mello in 2006. This discovery has paved an easier way for the use of loss-of-function analyses in many animals, including insects. Before the discovery of RNAi, it was almost impossible to analyze the functions of genes in insects for which genetic analyses are difficult. Now, RNAi allows the rapid and straightforward analysis of gene function in non-model insects. Currently, many RNAi-based studies with various insects are in progress, and several reviews on their results have already been published. In particular, Bellés (2010) comprehensively reviewed RNAi-based studies on insects, covering some 30 species representing nine orders. However, this approach is still relatively new, and this field will continue to expand as the number of insect genome projects increases and as RNAi methods are used to analyze the functions of new genes. Even in Drosophila melanogaster, after the completion of its genomic sequencing, a large number of genes whose functions are unknown were identified, and their functions have been investigated with RNAi methods (Mathey-Prevot & Perrimon 2006). Thus, the combination of genomic sequencing projects with RNAi methods will lead to an increase in our understanding of gene functions. In this review, we focus on two topics: (i) the current status of RNAi-based experiments in non-drosophilid insects and (ii) the applications of RNAi as species-specific insecticides. Finally, we discuss the future of RNAi-based studies in Entomology.
CURRENT STATUS OF RNAI-BASED EXPERIMENTS IN INSECTS
The advent of RNAi represents a new experimental paradigm beyond Drosophila, as it opens the door to the study of gene functions in other species (Bellés 2010). In 1999, Brown et al. (1999) showed that RNAi could be used to phenocopy the mutations of the Deformed orthologue in the flour beetle, Tribolium castaneum. In 2000, Hughes and Kaufman (2000) reported the use of RNAi to dissect gene function in the development of the milkweed bug, Oncopeltus fasciatus, which undergoes hemimetabolic maturation. Since then, many RNAi studies on insects have been reported (for a review, Price & Gatehouse 2008; Posnien et al. 2009; Bellés 2010). In general, when a double-stranded RNA (dsRNA) with a strand complementary to a fragment of the target mRNA is injected into insects, phenotypes appear, depending on the concentration of dsRNAs, the types of targeted genes and other still unknown factors.
To review the RNAi-based experiments reported in non-drosophilid insects, we classify the experiments into four types: parental, embryonic, larval/nymphal/pupal, and regeneration-dependent RNAi.
RNAi can be applied with particular ease in Caenorhabditis elegans, in which the injection of dsRNA into the body cavity or the application of dsRNA via ingestion leads to gene inactivation in offspring embryos (Fire et al. 1998; Timmons & Fire 1998). This RNAi effect has enabled efficient, genome-wide functional screens in this organism (Fraser et al. 2000). Bucher et al. (2002) found that the injection of dsRNA into the mother's hemocoel results in a knockdown of the zygotic genes in the offspring embryos of T. castaneum (“parental RNAi”, paRNAi). Parental RNAi was found to be highly efficient in Tribolium. Nearly 100% of embryos in the first egg laid after injection had RNAi phenotypes. In 2004, Liu and Kaufman (2004a,b) adopted the technique of paRNAi for use in milkweed bugs, Oncopeltus fasciatus (Hemiptera : Lygaeidae), to determine the role of the gap genes, hunchback and Krüppel, in segmentation. Ronco et al. (2008) and Mito et al. (2008) found that paRNAi also acts in the cricket, Gryllus bimaculatus. Shinmyo et al. (unpublished data) observed the effects of paRNAi on caudal development in Gryllus embryos (Fig. 1), where only the head region is formed in eggs after the injection of dsRNA targeting Gb'caudal in female crickets. This result is consistent with that obtained by emRNAi (Fig. 1) (Shinmyo et al. 2005). Lynch and Desplan (2006) developed a method for paRNAi that entails injecting pupae with dsRNA in the wasp, Nasonia vitripennis. In the two-spotted spider mite, Tetranychus urticae, paRNAi against Distal-less was used, resulting in phenotypes with canonical limb truncation as well as the fusion of leg segments (Khila & Grbic 2007). Together, these findings suggest that paRNAi may function in many metazoan taxa, indicating that the transmission of dsRNA across oocyte membranes is a conserved feature of the RNAi response. Since large numbers of RNAi embryos can be readily obtained with paRNAi and analyzed using standard histochemical and in situ hybridization procedures, the functional characterization of genes has been greatly facilitated. Thus, the use of paRNAi is an important method for analyzing early embryogenesis, even for species where embryonic RNAi is already available, and it is crucial for many insects whose eggs are not accessible or do not survive with microinjection (Bucher et al. 2002). Parental RNAi should allow large-scale RNAi screens in many species and will greatly facilitate functional approaches in the burgeoning field of evolutionary developmental biology (Bucher et al. 2002).
When dsRNA for a target gene is injected into developing eggs, the RNAi phenotypes can appear in embryos (Fig. 1), larvae/nymphs, or even adults. This type of RNAi is called embryonic RNAi (emRNAi). Although using paRNAi is the most efficient way to obtain RNAi phenotypes in embryos, if a target gene is indispensable for egg formation, then no eggs can be obtained. In such cases, emRNAi has been used to analyze gene function during embryogenesis. For instance, Grossmann et al. (2009) studied the role of wingless (wg) in the leg development of T. castaneum using stage-specific staggered emRNAi because the paRNAi experiments lead to either empty eggshells or wild-type cuticles. In this case, emRNAi was able to circumvent the problem of effects exhibited during gonad formation or oogenesis. By staggering the injections, the effects that lead to early embryonic lethality can also be excluded. With emRNAi, they demonstrated the separate functions of Tribolium wg in distal and ventral leg development.
Tomoyasu and Denell (2004) demonstrated the use of RNAi to create pupal and adult loss-of-function phenotypes in T. castaneum by injecting dsRNA into late instar larvae (larval RNAi, laRNAi). The laRNAi technique has been used to analyze gene functions in post-embryonic development to study the molecular basis of adult morphological diversity in various organisms. For example, Tomoyasu et al. (2005) used laRNAi in T. castaneum to determine the function of Ubx/Utx during hindwing/elytron development. Ubx/Utx RNAi induced a complete transformation of hindwing to elytron. Unlike Drosophila Ubx, which normally modifies the development of membranous wings to produce halters, Ubx/Utx in this beetle seems to promote membranous hindwing development by repressing elytron identity. Larval RNAi is also useful for analyzing the functions of enzymes in biosynthesis pathways. Arakane et al. (2005) demonstrated that laRNAi against laccase 2 in T. castaneum resulted in the formation of white and soft bodies in a dsRNA dose-dependent fashion. Ohnishi et al. (2009) found that pupal RNAi against Bombyx mori fatty acid transport protein (Bm'FATP) significantly suppressed the accumulation of cytoplasmic lipid droplets by preventing the synthesis of triacylglycerols, which resulted in a significant reduction in the production of the pheromone bombykol. In this case, they injected dsRNA into the abdominal tip of 1-day-old pupae. After injection, the pupae were maintained under normal conditions until adult emergence.
Nymphal RNAi (nyRNAi) was reported by Dong and Friedrich (2005) as a systemic RNAi mediated gene knockdown in the juvenile grasshopper, Schistocerca americana. They found that the injection of dsRNA corresponding to the vermilion eye color gene of first instar nymphs triggered a suppression of ommochrome formation in the eye lasting through two instars, equivalent to 10–14 days. More recently, Dong and Friedrich (2010) found that after the long-term arrest of eye development in S. americana nymphs induced by nyRNAi against the homologs of the early retinal genes eyes absent (eya) or sine oculis (so), eye development resumed after knock-down expiry. Taking advantage of the time-limited nature of this systemic RNAi, they discovered the presence of an inherent capacity for the underlying gene regulatory network to accommodate for partitioning visual system development into discrete phases, as in insects which develop adult eyes indirectly in the eye imaginal disc like Drosophila. Martin et al. (2006) found that the injection of dsRNA into the hemocoel of nymphs and adults of the cockroach, Blattella germanica, can be used to silence gene function in vivo. They used nyRNAi to elucidate the function of RXR/USP, which is one component, along with EcR, of the heterodimeric nuclear receptor of 20-hydroxyecdysone (20E). They demonstrated that Bg'RXR knockdown nymphs progressed through the instar correctly, but development was arrested at the end of this stage, making them unable to molt into adults. These results suggested that RXR/USP function, in relation to molting, is conserved across the insect Class. Nakamura et al. (2007) used nyRNAi to study the mechanisms of leg regeneration of the cricket, G. bimaculatus. Bando et al. (unpublished results) showed that white adult crickets could be generated by nyRNAi against laccase 2, when injected at the fifth instar (Fig. 2). The mechanisms of the circadian rhythm have also been studied with use of nyRNAi by Moriyama et al. (2008). They found that a single injection of period (per) dsRNA into the Gryllus abdomen of the final instar nymphs effectively knocked down the mRNA levels in adults to about 50% of control animals. Most of the per dsRNA-injected crickets completely lost their circadian-dependent locomotor activity rhythm under constant darkness for up to 50 days after the injection. To explore the possibility that G. bimaculatus could be a useful model for characterizing the molecular mechanisms of human diseases, Hamada et al. (2009) performed loss-of-function analyses on G. bimaculatus genes homologous to the human genes that are responsible for certain human disorders: fragile X mental retardation 1 (fmr1) and Dopamine receptor (DopR). For Gb'fmr1, they observed three major nyRNAi phenotypes: (i) abnormal wing postures; (ii) abnormal calling song; and (iii) loss of the circadian locomotor rhythm; indicating that the cricket has the potential to become a novel model system to explore human neuronal pathogenic mechanisms and to screen therapeutic drugs with RNAi. The conservation of systemic RNAi in these insects suggests that this pathway can be exploited for the gene-specific manipulation of larvae/nymphs and pupae/adults in a wide range of insects.
Nymphs of hemimetabolous insects, such as cockroaches and crickets, exhibit a remarkable capacity for regenerating complex structures from damaged legs (for review see Nakamura et al. 2008). Until recently, however, approaches to elucidate the molecular mechanisms underlying the process of regeneration in insect legs have not been available. Currently, our group is the only one working on insect leg regeneration with RNAi. Using the cricket, G. bimaculatus, as a model, Nakamura et al. (2008) found that RNAi phenotypes can be observed during regeneration following the amputation of legs. Because no phenotype is induced by nyRNAi in an intact cricket leg, this effect is designated as regeneration-dependent RNAi (rdRNAi). Since that time, the functions of various genes encoding signaling factors and cellular adhesion proteins, such as Fat and Dachsous, have been investigated during Gryllus leg regeneration. Mito et al. (2002) showed that Gryllus orthologues of Drosophila hedgehog (Gb'hh), wingless (Gb'wg) and decapentaplegic (Gb'dpp) are expressed during leg regeneration and play essential roles in the establishment of the proximal-distal axis. To confirm the roles of Gb'wg during regeneration, Nakamura et al. (2007) used rdRNAi and found that no regeneration took place when Gb'armadillo (the orthologue of β-catenin) was knocked down. However, no phenotypes were observed when Gb'wg was knocked down, probably due to the redundancy of other Wnt family members. Their results indicated that the canonical Wnt/wg signaling pathway is involved in the process of leg regeneration and determination of the positional information in the leg segment (Nakamura et al. 2008). Similar experiments were performed for Gb'hh by Miyawaki (unpublished results), showing that depletion of Gb'hh mRNA resulted in a lack of normal leg regeneration. These results suggest that Gb'wg and Gb'hh are involved in a distal-to-proximal respecification of the regenerate. Because rdRNAi occurs only during regeneration, one can observe leg phenotypes even for the knockdown of essential genes, which would result in lethal phenotypes in the nymph. Recently, Bando et al. (2009) used a cricket leg model to show that the Dachsous/Fat (Ds/Ft) signaling pathway is essential for leg regeneration. They found that the knock-down of Gb'ft or Gb'ds transcripts by rdRNAi suppressed the proliferation of the regenerating cells along the proximodistal axis concomitantly with remodeling of the pre-existing stump, resulting in regenerated legs that were shorter than normal (Fig. 3). By contrast, the knockdown of the Gb'expanded or Gb'Merlin transcripts induced over-proliferation of the regenerating cells, increasing the length of the regenerated legs. Then, Bando et al. (2009) presented a model to explain their results in which the steepness of the Ds/Ft gradient controls growth along the proximodistal axis of the regenerating leg. This study provided an important insight about how regenerating blastemal cells are aware of both their position and the normal size of the leg. Because the Ds/Ft system is conserved in vertebrates, their results provided clues to the mechanisms of regeneration, which are relevant to vertebrate systems (Bando et al. 2009).
APPLICATIONS OF RNAI FOR DEVELOPMENT OF SPECIES-SPECIFIC DSRNA INSECTICIDES
A serious problem for insecticides is that they can kill non-targeted animals. To address this issue, the possibility of using RNAi to kill only the target animals by down-regulating essential gene functions in insects has been recognized for many years (Price & Gatehouse 2008). Bando et al. (unpublished data) screened Gryllus target genes to develop G. bimaculatus-specific dsRNA insecticides (Fig. 4). However, this method was considered unfeasible because the method relies on the injection of dsRNA into insects, which is not possible for practical application of insecticides. A more effective method may be to use a bait containing dsRNA, as developed in nematodes which can uptake dsRNA through feeding. Because dsRNA may be degraded in the gut, it seemed that the knockdown of a target gene was unlikely. However, Turner et al. (2006) demonstrated that in the horticultural pest, Epiphyas postvittana (Lepidoptera: Tortricidae), RNAi could be triggered by the oral delivery of dsRNA to larvae. Meyering-Vos and Müller (2007) found that treatment of the adult cricket, G. bimaculatus, by injection or ingestion of a dsRNA for sulfakinin, a group of brain/gut neuropeptides, led to a stimulation of food intake, indicating that the uptake of dsRNA in the Gryllus occurs in the gut. Although these results suggest that the oral delivery of dsRNA is feasible, the problem of the continuous feeding needed to use dsRNA, as an insecticide remained unsolved. To circumvent this problem, Baum et al. (2007) made transgenic corn plants engineered to express dsRNAs for the western corn rootworm (WCR). The plants showed a significant reduction in WCR feeding damage in a growth chamber assay, suggesting that the RNAi pathway can be exploited to control insect pests via the in planta expression of a dsRNA. Mao et al. (2007) also made Arabidopsis thaliana and Nicotiana tobacum transgenic plants expressing dsRNA specific to a cytochrome P450 gene (CYP6AE14) of the cotton bollworm (Helicoverpa armigera), which permits this herbivore to tolerate the cotton metabolite, gossypol. When larvae are fed the transgenic plant, larval growth is retarded due to inhibitory effects of gossypol. Thus, they concluded that feeding insects with plant material expressing dsRNA may be a general strategy for the delivery of RNAi and could find applications in entomological research and field control of insect pests (Gordon & Waterhouse 2007).
Whyard et al. (2009) harnessed the sequence specificity of RNAi to design orally-delivered dsRNAs that selectively killed target insects. They found that D. melanogaster, T. castaneum, pea aphids (Acyrthosiphon pisum), and tobacco hornworms (Manduca sexta) were selectively killed when fed species-specific dsRNA targeting vacuolar-type ATPase transcripts. For the aphid nymphs and beetle and moth larvae, dsRNA could simply be dissolved into their diets. However, to induce RNAi in the drosophilid species, the dsRNAs needed to be encapsulated in liposomes to help facilitate the uptake of the dsRNA (Whyard et al. 2009). This method may lead to higher throughput RNAi screens and the development of a new generation of species-specific insecticides. In addition to liposomes, polycation polymers, such as cyclodextrin or dynamic polyconjugates, may be another promising delivery method for dsRNAs into cells (Tiemann & Rossi 2009). Overcoming the specific delivery of dsRNA or siRNA into the cytoplasm of target cells is still an important issue for the use of RNAi for insecticides.
EFFICIENCY OF RNAI DEPENDS ON DEVELOPMENTAL STAGES AND SPECIES
Bellés (2010) listed the insect species studied with RNAi in vivo in his comprehensive review and pointed out that the efficiency of systemic RNAi in vivo depends on the species; less-derived species are generally more sensitive than more-derived species. Tomoyasu et al. (2008) performed a genome-wide survey to compare genes involved in the machinery of RNAi between Tribolium and C. elegans, both of which show a robust systemic RNAi response. They found significant differences between the genes involved in the RNAi machinery of these organisms. Thus, they concluded that insects might use an alternative mechanism for the systemic RNAi response. Because over-expression of dsRNAs within cells using hairpin RNAs readily triggers RNAi in D. melanogaster tissues resistant to systemic RNAi, Bellés (2010) speculated that the poor sensitivity is not related to the RNAi core machinery, but rather to the penetration and transmission of the interfering signal through cells and tissues and to the occurrence of degradation mechanisms that are able to remove alien RNA. Huvenne and Smagghe (2010) reviewed the mechanisms of dsRNA uptake. They list two major types of mechanisms: (i) the transmembrane channel-mediated uptake mechanism, in which orthologs of C. elegans systemic RNAi defective mutant gene (sid) may be involved; and (ii) the endocytosis-mediated uptake mechanism, in which vacuolar H+-ATPase, clathrin, scavenger receptors, etc. may be involved. This uptake mechanism may be linked to innate antiviral immunity responses, like those found in Drosophila (Saleh et al. 2009). Thus, the dependence of RNAi efficiency on developmental stages or species is attributed to differences in the expression patterns of the sid orthologs and/or differences in the sensitivity of dsRNA detection systems for endocytosis in insect cells. If this is the case, one of the ways to circumvent the problem of low RNAi efficiency is to use chemically modified siRNAs or carriers which may stabilize siRNAs and transport them into cells, as investigated for human RNAi therapy (Tiemann & Rossi 2009). Using such methods in insects could help to render other insects amenable to systemic RNAi and may influence pest control approaches (Tomoyasu et al. 2008).
FUTURE PROSPECTS FOR RNAI-BASED EXPERIMENTS IN INSECTS
RNAi-based experiments have provided several interesting results and shed light on gene functions in non-drosophilid insects. However, the mechanisms underlying RNAi phenomena, such as species or tissue-dependent changes in the sensitivity to dsRNAs, are not fully understood. Thus, a more precise elucidation of RNAi mechanisms is an important issue in Entomology. The potential application of RNAi techniques to any gene and any species could lead to comparative studies for the function of a gene, or gene network, in species covering a large spectrum of insect orders (Bellés 2010). Moreover, RNAi can facilitate comparative studies and evolutionary insight into other processes, such as social behaviors, reproductive strategies, and host-parasite interactions (Bellés 2010). Systemic RNAi-based genome-wide screening is particularly useful to identify genes involved at the whole animal level, e.g. in determination of life span and size, metabolic controls, circadian clock systems, ecdyses, etc.
With the increase in the number of genome projects, the RNAi method will be more useful for genome-wide analyses of gene functions (Fig. 5). We will then have access to a large pool of information on various insect genes to clarify what insects are and how to treat insects in the real world. However, the facilities and manpower to analyze such a large body of information are currently unavailable. In the future, a worldwide forum about RNAi should be established within the Entomological Society.
We thank members of our laboratory for providing unpublished data for this review. A portion of the work described was supported by a grant from the Ministry of Education, Culture, Sports, Science and Technology of Japan to T.M and S.N.