Development of pea albumin based GNA fusion proteins as novel biopesticides for the control of aphid (Hemiptera: aphidae) crop pests.

Abstract

Novel insecticidal compounds with new modes of action are urgently required as current reliance and overuse of chemical pesticides that target a limited number of sites within insect pests has driven widespread development of resistance in crop pests. Furthermore, the most widely utilised biopesticide Bacillus thuringiensis (Bt) targets lepidopteran, coleopteran and certain dipteran pests but is ineffective against hemipteran sap-sucking pests. Hemipteran sap-sucking pests cause considerable damage to crops through direct damage and transmission of many economically important plant viruses. Further, they have been documented to rapidly develop resistance to novel insecticides. Fusion of invertebrate specific neurotoxins to snowdrop lectin (Galanthus nivalis agglutinin; GNA) has previously been shown to enhance oral activity of peptide toxins by virtue of its ability to bind to the insect gut epithelium and subsequently to transport attached peptides across the gut to the circulatory system. In addition to delivering attached toxins, GNA via binding to insect gut polypeptides, was hypothesised to provide an opportunity to potentiate the efficacy of insecticidal proteins that act at the gut surface. Pea albumin (PA1b), from pea (Pisum sativum) seeds is a proteinaceous inhibitor of V-ATPase proton pumps that are localised to the insect gut epithelium. The PA1 gene cassette is expressed in planta as a preproprotein and cleaved into two mature proteins, PA1b and PA1a. PA1b has previously been shown to cause mortality when supplied in diet to several insect species from different orders including hemipteran pest species. PA1a has no documented insecticidal activity. This project focuses on the development and production of novel recombinant fusion proteins based on linking PA1b to GNA for the control of three different aphid species (pea aphid; Acrythosiphon pisum, peach potato aphid; Myzus persicae and grain aphid; Sitobion avenae).
Production of a highly pure recombinant pea albumin – GNA fusion protein (PA1b/GNA) in Pichia pastoris proved challenging; multiple expression constructs with varying linker regions, orientations and histidine purification tags were typically expressed at relatively low levels and recovery from nickel affinity purification columns was poor. However, the use of a fusion protein expression construct that included a strep-tactin affinity tag allowed the recovery of relatively pure protein samples from fermented P. pastoris cultures. An LC50 ¬ of 7 nmol mL-1 PA1b/GNA against A. pisum was subsequently derived from artificial diet bioassays and this suggested that fusion of PA1b to GNA did indeed enhance toxicity as compared to activity reported for PA1b alone in the literature. However, due to low recovery rates for PA1b/GNA, an alternative approach to generate sufficient quantities of highly pure fusion proteins based upon constructs encoding both PA1b and PA1a linked to GNA (PAF/GNA) was pursued. Expression of PAF/GNA in P. pastoris resulted in higher expression levels, and more importantly, allowed the recovery of greater quantities of highly pure protein samples as compared to PA1b/GNA.
The oral toxicity of PAF/GNA and recombinant PAF was assessed through artificial diet bioassays against three species of aphid pests: A. pisum (LC50 = 5 nmol mL-1), S. avenae (15 nmol mL-1) and M. persicae (34 nmol mL-1). PAF/GNA was significantly more orally toxic towards A. pisum and M. persicae as compared to recombinant PAF or GNA alone. Greater retention of PAF/GNA in the A. pisum gut over time, as compared to controls (PAF, GNA or ovalbumin) was demonstrated using confocal microscopy of A. pisum aphids pulse fed on fluorescently labelled proteins. Significant differences in toxicity were observed between A. pisum and M. persicae, it was hypothesised that this may be due to the stability of PAF/GNA when exposed to insect proteases. PAF/GNA, GNA and PAF were all similarly stable when exposed to salivary secretions from either species for up to 48 hours. Feeding assays were performed to assess the in vivo stability, no intact protein was detected in either species, but the quantity of breakdown products decreased in A. pisum over the 48 hours, possibly suggesting that the quantity of diet consumed was decreased compared to M. persicae. Antifeedant effects were indeed indicated in free choice experiments, A. pisum displayed a clear preference for diets without pea albumin or GNA, whereas M. persicae showed no significant preference except after 48 hours feeding on PAF/GNA where they preferentially fed on control diet.
Whilst PAF/GNA due to its potentially limited host range is unlikely to be utilised in a commercial setting, there may be a case made for the expression of a PAF/GNA fusion protein under the control of a phloem specific promoter as a novel method to control aphids that are one of the most economically damaging groups of agricultural pests.