Evaluation of three vacuolar ATPase genes as potential RNAi target in Henosepilachna vigintioctopunctata

Highlights

• Three vATPase genes are highly expressed in epithelial tissues.

• Silencing each vATPase gene at the fourth instar stage causes larval lethality.

• RNAi of vATPase gene at the third instar stage leads to more severe impacts.

• RNAi efficiency is in a gene- and developing stage-dependent manner.

Abstract

Henosepilachna vigintioctopunctata, a major threat to commercial potato production, is sensitive to RNA interference (RNAi), a gene-silencing mechanism triggered by double stranded RNA (dsRNA) matching a specific target gene sequence. Vacuolar ATPases (vATPases) are ATP-driven proton pumps involved in numerous biological processes. A holoenzyme vATPase consists of at least 14 different subunits and accessory proteins. Previous results reveal that silencing vATPaseB decreases survival rate, reduces food consumption and inhibits development in H. vigintioctopunctata. In this report, we identified two other subunit genes (vATPasea and vATPased) and compared the RNAi efficacy of vATPasea and vATPased with that of vATPaseB. We found the highest expression level of HvvATPasea, HvvATPaseB and HvvATPased occurred in the hindgut, followed by those in the foregut, Malpighian tubules and midgut, and lowest in the epidermis and fat body. When the fourth-instar larvae were allowed to ingest corresponding dsRNA for three days, the relative transcript levels of HvvATPasea, HvvATPaseB and HvvATPased were significantly decreased by 79%, 83% and 73%, respectively. Knockdown of HvvATPasea, HvvATPaseB and HvvATPased caused 50%, 85% and 100% larval lethality. After the third-instar larvae had exposed to corresponding dsRNA for three days, the levels of HvvATPasea and HvvATPased significantly reduced by 91% and 96% respectively. All the treated larvae were dead. Our results demonstrate that RNAi efficiencies varied among different vATPase subunit genes and various development stages. Moreover, our data imply the feasibility of RNAi as an alternative method for controlling this critical potato pest.

Introduction

RNA interference (RNAi) is a biological process that involves providing double stranded RNA (dsRNA) matching a specific gene sequence to silence the expression of the target gene at the transcriptional or post-transcriptional level (Christiaens et al., 2020, Liu et al., 2020). The process of gene silencing can be exploited in agriculture to control insect pests. Of the various approaches tested, expression of dsRNA in transgenic crop plants and direct application of dsRNA as an insecticide appear promising for use in the field (Vogel et al., 2019). For example, transgenic corn, which has the commercial name SmartStax PRO, targeting western corn rootworm and northern corn rootworm, was approved by the Canadian Food Inspection Agency (CFIA) in 2016 and the US Environmental Protection Agency (EPA) in 2017. Resistance in SmartStax PRO is mediated by the combined action of a dsRNA construct that specifically targets the sucrose-non-fermenting 7 gene (DvSnf7), together with two insecticidal proteins from Bacillus thuringiensis encoded by the crystal (Cry) genes Cry3Bb1 and Cry34Ab1/Cry35Ab1 (Head et al., 2017). An alternative strategy relies on the use of in vitro synthesized dsRNAs, via foliar application, trunk injection, seed dressing, or soil drench, to trigger RNAi in plant pests (Christiaens et al., 2020). Irrespective of application method, identification of essential genes is the first step for their potential application via RNAi (Liu et al., 2020).

Insect vacuolar-type H⁺-ATPases (vATPases) are the complex membrane-bound rotary molecular motors. They are localized in the apical membranes of nearly all insect epithelial tissues (Muench et al., 2014). By hydrolyzing ATP to ADP and phosphate, vATPases pump protons across membranes (Wieczorek et al., 2009, Wieczorek et al., 1999). The structures of vATPases are conserved in eukaryote species (McGuire et al., 2017, Song et al., 2020). A holoenzyme vATPase consists of two functional subcomplexes, V₁ and V₀. The cytoplasmic (peripheral) V₁ subcomplex is an ATPase and is comprised of eight different subunits (A-H), with stoichiometry of A₃B₃CDE₃FG₃H in Lepidopteran Manduca sexta and yeast Saccharomyces cerevisiae (Kitagawa et al., 2008, Muench et al., 2009). V₀ subcomplex is a membrane-bound, proton-conducting complex. It is composed of 6 different subunits (a₁d₁c_4-5e₁c'₁c“₁) (Kitagawa et al., 2008, Merzendorfer et al., 2000, Song et al., 2020).

Depletion of vATPase subunits is lethal to insects. Update, most reports focus on V₁ subcomplex subunits. For instance, the vha55 (encoding the B-subunit) mutants (deleted for the entire locus) of Drosophila melanogaster survive to hatching but fail to thrive, whereas point mutant homozygotes can die midway through embryonic development (Allan et al., 2005). Likewise, knockdown of vATPaseB significantly increases the mortality of juveniles in Coleopteran Leptinotarsa decemlineata (Zhu et al., 2011) and Henosepilachna vigintioctopunctata (Lü et al., 2020), Hemipteran Peregrinus maidis (Yao et al., 2013), and Blattarian Periplaneta fuliginosa (Sato et al., 2017). Moreover, depletion of vATPaseB mRNA affects survival and molting in the Chinese mitten crab, Eriocheir sinensis (Hou et al., 2020).

However, relative few attentions have been paid on genes encoding V₀ subcomplex subunits and some results are mutually contradictory. In D. melanogaster, for example, Allan et al. found that the vha100-2 (encoding the a subunit) mutant flies are viable (Allan et al., 2005). Conversely, other workers have shown that P-element insertions in vha100-2 are lethal (Bonin and Mann, 2004). It is important to test whether knockdown of V₀ subcomplex subunit encoding genes causes severe defective phenotypes in a non-Drosophilid insect species.

Moreover, RNAi efficacy varies among genes. This phenomenon has been documented in Coleoptera, Lepidoptera, Hemipteran and Orthoptera insects (Cooper et al., 2019). However, the comparison of the same type of genes has been documented in only one insect species Tribolium castaneum, in which only 11 highly efficient RNAi targets are documented among 111 non-sensory G protein-coupled receptors (GPCRs). Moreover, RNAi of 8 GPCR genes causes > 90% mortality, with dopamine-2 like receptor and latrophilin receptor being best (Bai et al., 2011). We are eager to know the differences in RNAi efficiencies among V₁ and V₀ subcomplex subunit encoding genes.

Furthermore, RNAi efficiency is inconstant among life stages (Cooper et al., 2019, Scott et al., 2013). For example, D. melanogaster larvae are refractory to injection or feeding of naked dsRNA, but RNAi is successful when injection into embryos and adults (Miller et al., 2008). In L. decemlineata, ingestion of dsLdSAHase targeting S-adenosyl-L-homocysteine hydrolase by young larvae causes higher RNAi efficiency than that by old larvae (Guo et al., 2015). In Apis mellifera, when a 504 bp of dsvitellogenin is injected at the preblastoderm stage, 15% workers have strongly reduced levels of vitellogenin mRNA. In contrast, 96% individuals show the mutant phenotype when dsvitellogenin is introduced by intra-abdominal injection in newly emerged bees (Amdam et al., 2003). The findings prompt us to determine whether RNAi efficiencies differ among different larval instars.

Therefore, a comparative study of both V₀ and V₁ subcomplex encoding genes on RNAi effects between larval instars should provide critical insights into how RNAi-based control strategies can be optimized for specific insect pests. In the present paper, bacterially expressed dsRNA fragments from three vATPase genes (vATPasea and d encoding V₀ subcomplex proteins, vATPaseB coding for a V₁ subcomplex protein) were produced. Their RNAi efficiencies were compared among different dsRNAs and between larval stages (feeding at the third- or fourth-instar stages). We found that RNAi efficiencies varied among different vATPase subunit genes and between larval instars. The possible reasons for the difference were also discussed.