Advances in oral RNAi for disease vector mosquito research and control

Abstract

Mosquito vectors in the genera Anopheles, Aedes, and Culex transmit a variety of medically important pathogens. Current vector control tools are reaching the limits of their effectiveness, necessitating the introduction of innovative vector control technologies. RNAi, which facilitates functional characterization of mosquito genes in the laboratory, could one day be applied as a new method of vector control. Recent advances in the oral administration of microbial-based systems for delivery of species-specific interfering RNA pesticides to mosquitoes may facilitate translation of this technology to the field. Oral RNAi-based pesticides represent a new class of biorational pesticides that could combat increased global incidence of insecticide resistance and which could one day become critical components of integrated human disease vector mosquito control programs.

Introduction:

The mosquito genome projects facilitated research in new facets of mosquito biology, including functional genetic studies in medically important Aedes (dengue, Zika, chikungunya, and yellow fever vector), Culex (West Nile and lymphatic filariasis vector), and Anopheles (malaria vector) disease vector mosquitoes. Genetic advancements in mosquitoes have made the potential for using gene-centered vector control strategies a reality, challenging researchers to identify potential gene targets for vector control, as well as reliable methods for manipulating mosquito gene function in the laboratory that could one day be extended to the field [1]. RNAi, initially employed in Caenorhabditis elegans, in which microbial systems for oral delivery of interfering RNA were pioneered, is a cellular mechanism that prevents expression of mRNA transcripts when triggered by interfering RNAs such as double-stranded RNA (dsRNA), small interfering RNA (siRNA), and short hairpin RNA (shRNA) [2, 3]. RNAi has revolutionized the field of functional genomics by permitting the interrogation and characterization of candidate genes in a wide variety of insects, including mosquitoes [4]. Silencing efficiency, however, is variable and dependent on a number of parameters including the: (i) interfering RNA species (dsRNA, siRNA or shRNA) administered and its molecular stability, (ii) delivery system, which impacts upon biochemical processing downstream of the anatomical site of entry [5], and (iii) cellular physiology of the arthropod [6]. These factors, either singularly or combined, have historically hampered development of RNAi as a mosquito control technology.

This review focuses on recent advancements in oral delivery of interfering RNA to medically important mosquitoes in the context of its value to fundamental biological research, potential as a biorational vector control intervention, and the commercial and regulatory challenges that must be resolved while developing such a product for the global market.

RNAi-mediated analysis of mosquito gene function in the post-CRISPR-Cas9 era:

RNAi has played a key role in functional genomics, a cross-disciplinary field that has expanded our knowledge of insect evolutionary, developmental, physiological, and molecular biology at a remarkable pace. Although RNAi does not generate heritable germline mutations, an advantage of CRISPR-Cas9 gene editing strategies that have been successfully adapted in mosquitoes [7-10], it offers several advantages that may be of utility. First, RNAi is a conditional method of gene silencing that requires no long-term maintenance of genetically modified mosquito strains. The conditional nature of RNAi also allows researchers to control the stage at which gene silencing initiates, circumventing challenges such as developmental lethality or sterility which can hinder the production and maintenance of strains bearing heritable mutations. Moreover, even with significant CRISPR-Cas9 advancements, genetic engineering of non-model insects is still a relatively labor-intensive and expensive process. Thus, despite concerns for off-site targeting and variations in the level of silencing obtained, which can be dependent on the gene targeted and the tissue in which it is silenced, RNAi is often employed for functional genetics studies in mosquitoes and other insects [1]. Although a majority of laboratories use long dsRNA (~300-400 bp) molecules in their studies, the use of siRNAs and shRNAs enables confirmation of phenotypes using multiple interfering RNAs that target different 21-25 bp sites in a single gene [11], which can alleviate concerns that the phenotype under analysis has arisen from off-site targeting. As discussed below, the use of siRNAs/shRNAs also permits the targeting of short target nucleotide sequences that are conserved in multiple insect species, facilitating comparative studies of gene function [12, 13].

Orally-mediated delivery of interfering RNA:

Microinjection is widely used for delivery of interfering RNA to mosquito embryos, larvae, pupae, and adults [14, 15]. Despite the widespread use of microinjection, this technique is labor-intensive and requires both technical skill and a microinjection setup [16]. The stress of microinjection to the organism can complicate phenotype characterization, particularly for analysis of behaviors, and this technique cannot be extended to the field [16]. To address these concerns, a variety of oral delivery strategies, the subject of this review, have been successfully employed in mosquitoes [4]. These include: direct ingestion [17, 18], larval soaking (that most likely occurs through ingestion) [19-23], nanoparticle-mediated uptake with chitosan or other nanoparticles [15, 16, 24-26], and microbially-expressed vector systems such as bacteria and yeast [1, 12, 13, 20-22, 27]. Recent studies have focused on oral RNAi-mediated inhibition of developmental transcripts that cause defects in the development of mosquito tissues of vector importance or that interfere with reproductive traits. For example, targeting semaphorin-1a results in severe olfactory [28] and visual system [29] defects. Silencing of fasciculation and elongation protein zeta 2, leukocyte receptor cluster, suppressor of actin, offtrack, synaptotagmin, and semaphorin-1a in the larval brain correlate with high levels of mosquito larval mortality [12, 13, 21, 22]. Oral RNAi-mediated silencing of doublesex disrupts fitness and lifespan of Aedes aegypti [15] and Anopheles gambiae [27] female mosquitoes. By functionally characterizing these and other genes that are essential for mosquito survival and reproduction, researchers are identifying potential genetic targets for vector control, as well as potential means of sex-sorting mosquitoes for large-scale male release programs.

In addition to discovery of target genes that can be manipulated for vector control, the identification of effective oral interfering RNA delivery systems that function well in the laboratory and which might translate to the field is critical. Incongruencies exist in the level of gene silencing that is generated by oral RNAi techniques, with variation of RNAi inefficiency likely resulting from the molecular conformation of RNA, its stability during and after cellular uptake, and transportation to the target site [5, 6, 18, 30-32]. Several recent efforts have been directed toward improvement of chitosan-based strategies and exploration of other nanoparticle interfering RNA delivery systems in mosquitoes and other insects [33-36]. Further work in this area will permit identification of the most efficient nanoparticle-based systems for RNA delivery to mosquitoes. However, the present costs of RNA synthesis are a significant factor for soaking and nanoparticle-based delivery systems, particularly when considering large-scale laboratory and field applications.

The use of microbial interfering RNA expression and delivery systems, which facilitate cost-effective RNA propagation, has attracted attention in recent years. Kumar et al. [37] explored the use of transgenic Chlamydomonas reinhardtii for oral delivery of larvicidal dsRNA to Anopheles stephensi. Although this technique shows promise, the pursuit of regulatory permits for the release of live genetic model organisms, and in some geographical contexts, invasive species, could impact translation of algal delivery systems to the field. Comparisons of the relative efficacy of orally-based delivery mechanisms in A. aegypti and A. gambiae larvae suggest that bacterial and yeast-based strategies for delivery of larvicidal interfering RNA to mosquitoes generate higher degrees of target gene silencing and larval mortality than do other delivery systems [21, 22]. Furthermore, these microbes can be heat-killed with no resulting loss of larvicide activity [21, 38], a likely advantage for field applications, as no live genetically modified organisms would be released into the environment.

Escherichia coli strains that were engineered to produce both long and short dsRNA molecules targeting mosquito genes of interest have facilitated gene silencing in A. aegypti [20, 21, 39] and A. gambiae [22, 27] larvae. Pichia pastoris (yeast) expressing long hairpin RNA (lhRNA) was used to target juvenile hormone acid methyl transferase in A. aegypti larvae [40]. More recently, shRNAs targeting mosquito genes that are required for larval viability were expressed in Saccharomyces cerevisiae (baker’s yeast), in which shRNA was either expressed from plasmids or expression cassettes that had been stably integrated into the yeast genome [1, 11-13, 21, 22]. Following culturing, the S. cerevisiae strains were prepared as dried tablets that were fed to larvae, resulting in silencing of developmental transcripts and high mortality rates (>90%) in multiple mosquito larval species.

The success of microbial-based interfering RNA delivery strategies in mosquitoes suggests that microbe-interfering RNA complexes survive the mosquito gastrointestinal environment. The flour beetle Tribolium castaneum is classed as RNAi-efficient because it processes ingested dsRNA into siRNA via clathrin-dependent endocytosis [41], thus avoiding degradation by gut lumen dsRNases and trapping by cytoplasmic endosomes. StaufenC was recently identified as an obligatory dsRNA binding protein for RNAi initiation in Coleoptera, as well as a potential target for development of RNAi resistance [42]. Although endocytosis is likely the mechanism for microbial-based interfering RNA uptake in mosquitoes, this has not yet been confirmed, and StaufenC has not been identified in mosquitoes, in which RNAi resistance has not yet been reported, but is nevertheless a concern. The mosquito RNAi research agenda should therefore focus on elucidation of the cellular machinery that permits interfering RNA and microbe-interfering RNA transport, how these species are processed and degraded, whether the mechanisms differ between delivery methods, and assessment of the potential for developing resistance to RNAi when utilizing different delivery strategies.

Orally-induced RNAi in vector control applications:

Insecticide resistance is a primary obstacle to global vector control operations, and the development of new classes of environmentally safe pesticides with no cross-resistance to current insecticides is critical [43]. RNAi-based pesticides, promising candidates that satisfy these requirements, have an overwhelmingly good safety profile, particularly with respect to existing chemical pesticides [44]. While recent studies demonstrated a reversal of insecticide resistance by soaking larvae in dsRNAs that targeted genes known to contribute to resistance pathways in A. aegypti [45] and An. stephensi [23], most RNAi-based larvicides are designed to directly kill mosquito larvae. High-throughput screens in A. aegypti and A. gambiae have identified hundreds of genes required for larval viability, several of which have been characterized in detail [12, 13, 21, 22]. An arsenal of interfering RNA pesticides could be developed as a new class of species-specific biorational mosquito larvicides. This would permit rotated use of interfering RNA pesticides, combating the potential for development of point mutations in interfering RNA target sites.

Larvicide screens also led to the discovery of a number of interfering RNAs with target sites that are conserved in multiple mosquito species, but which are not found in humans or other non-target organisms [12, 13]. Yeast-mediated delivery of shRNAs targeting conserved sites in the synaptotagmin (syt) [12] and semaphorin-1a [13] genes effectively silenced target gene expression and induced high levels of larval mortality in A. aegypti, Aedes albopictus, A. gambiae, and Culex quinquefasciatus, but was not toxic to non-target arthropods. Yeast, which can be used directly for RNA production, as a delivery system, and as larval bait, is a strong odorant attractant for larvae [28] and gravid A. aegypti females, which are attracted to deposit eggs in treated containers [21]. These advantages, as well as the potential for greater acceptance of yeast vs. bacteria among consumers, make it an attractive delivery system for potential development as a mosquito control mechanism [11].

Novel RNAi-based control strategies can also be directed toward adult mosquitoes, a strategy that requires the identification of effective delivery methods to adults. Although reliable methods for topical applications of interfering RNAs to mosquitoes have not yet been identified, sugar-baited oral delivery systems are of increasing interest. Attractant Toxic Sugar Baits (ATSBs) are a simple concept designed to attract an insect to a baited-sugar source that induces mortality upon ingestion [46]. Current ATSB technologies, which often employ insecticides that are not specific to mosquitoes [46], could potentially be enhanced through the use of species-specific interfering RNA pesticides. Studies with A. aegypti have shown potential for incorporating interfering RNA into sugar solutions [17], and sugar-baited delivery of siRNAs targeting neural genes have induced 100% A. aegypti morbidity in laboratory trials (MDS, unpublished results). In addition to ATSBs, viral RNA expression systems present another potential method of lethal RNAi pesticide delivery (see discussion in [4]). The use of symbiotic microorganisms as a means of delivering lethal interfering RNA is also gaining traction [47], but may face greater regulatory hurdles than the use dead microbial interfering RNA delivery systems. Finally, in addition to killing adult mosquitoes, it may also be possible to use RNAi to induce pathogen resistance in mosquitoes, a concept investigated in two recent studies [48, 49], or to impact behaviors that contribute to the spread of disease, for example human host location and blood feeding behavior (see [4] for detailed discussions).

Conclusions and future directions:

In the two decades since it was identified, RNAi has advanced as a critical research tool with the potential to translate innovative bench technology towards operational vector control strategies (Fig. 1). In addition to proof of concept outdoor semi-field trials, which were successfully conducted for yeast interfering RNA larvicides [12, 13], the development and evaluation of commercial formulations, followed by analysis of the potential for scaled production of these formulations, are critical. Careful consideration must be given to the impacts of scaled production on interfering RNA stability. The stability of the commercial product over time is also crucial, particularly given that vector control operations often require long-term storage and operational use at high temperatures [11]. The cost of production, which can be challenging to estimate for novel technologies, will ultimately drive commercialization decisions, and input from specialists who can accurately estimate costs associated with development and manufacture of RNAi-based products is crucial. Evaluation of stakeholder acceptance, a critical component of the successful development of new vector control strategies, is also essential.

Fig. 1. Oral RNAi-based vector control product commercialization process.

The pathway for commercialization, initiating with target gene discovery and concluding with product registry, is summarized. See text for further discussion.

Following the development and scaled production of commercially-ready pesticide formulations, further toxicology and expanded field testing of these formulations are requisite for registry applications through the United States Environmental Protection Agency (EPA) and comparable entities in other countries. The use of heat-inactivated microbial delivery systems, which retain interfering RNA activity [21, 38], may support these applications. Registry of mosquito interfering RNA insecticides by the EPA, which following extensive satisfactory risk assessment [50] recently approved an RNAi-based pesticide targeting an insect agricultural pest, could increase the likelihood of gaining approval for these technologies at additional sites across the globe. However, some nations lack a regulatory body equivalent to the EPA, which will complicate global deployment of RNAi-based interventions. Although further research and regulatory hurdles remain, RNAi-based pesticides represent a new class of biorational insecticides that could combat increased global incidence of pesticide resistance and which could one day become critical components of integrated mosquito control programs.

Highlights:

• Oral RNA interference (RNAi) permits functional characterization of mosquito genes.

• Microbial-based oral RNAi methods are highly effective in mosquitoes.

• Interfering RNAs are a new class of pesticides that can combat resistance.

• Oral RNAi may facilitate species-specific biorational mosquito control.

• Further development and evaluation of RNAi control strategies is critical.