Cloning, expression and functional characterisation of chitinase from larvae of tomato moth (Lacanobia oleracea): a demonstration of the insecticidal activity of insect chitinase

https://doi.org/10.1016/j.ibmb.2004.06.012Get rights and content

Abstract

Chitinases are vital to moulting in insects, and may also affect gut physiology through their involvement in peritrophic membrane turnover. A cDNA encoding chitinase was cloned from larvae of tomato moth (Lacanobia oleracea), a Lepidopteran pest of crops. The predicted protein contains 553 amino acid residues, with a signal peptide of 20 a.a. Sequence comparison showed 75–80% identity with other Lepidopteran chitinases. L. oleracea chitinase was produced as a functional recombinant enzyme in the yeast Pichia pastoris. A fusion protein containing chitinase joined to the N-terminus of snowdrop lectin (GNA) was also produced, to determine whether GNA could deliver chitinase to the haemolymph of Lepidopteran larvae after oral ingestion. The purified recombinant proteins exhibited similar levels of chitinase activity in vitro. Both proteins were highly toxic to L. oleracea larvae on injection, causing 100% mortality at low dose (2.5 μg/g insect). Injection of chitinase prior to the moult resulted in decreased cuticle thickness. The recombinant proteins caused chronic effects when fed, causing reductions in larval growth and food consumption by up to 60%. The oral toxicity of chitinase was not increased by attaching GNA in the fusion protein, due to degradation in the larval gut, preventing GNA acting as a “carrier”.

Introduction

In insects, chitin is a vital component of the cuticle and peritrophic matrix (PM), where it functions as a dynamic and protective structural polysaccharide. Metamorphosis, and growth through ecdysis, is dependent upon the ability of insects to break down and re-synthesise chitin-containing structures. Chitin synthesis and degradation is achieved through the tissue-specific expression of chitin synthases and chitinolytic enzymes, both of which have been shown to be under tight developmental and hormonal control (reviewed by Merzendorfer and Zimoch, 2003). The importance of chitin for insect development and the absence of chitin polymers in vertebrates has led to interest in chitin biosynthesis and turnover as targets for insecticidal molecules, particularly in the development of novel strategies for the protection of crops against insect pests.

Insect chitinases belong to family 18 of the glycohydrolase superfamily and are characterised by a multi-domain structure with theoretical molecular masses ranging from 40–85 kDa (Kramer and Muthukrishnan, 1997). Chitinases have been cloned from Lepidopteran and Coleopteran insects including tobacco hornworm (Manduca sexta; Kramer et al., 1993), the fall webworm (Hyphantria cunea; Kim et al., 1998), the silkworm (Bombyx mori; Kim et al., 1998), the common cutworm (Spodoptera litura; Shinoda et al., 2001), the yellow mealworm (Tenebrio molitor; Royer et al., 2002), and the spruce budworm (Choristoneura fumiferana; Zheng et al., 2002). The Lepidopteran chitinases all contain a signal peptide, an N-terminal catalytic domain, a PEST-like linker region (enriched in proline, glutamate, serine and threonine), and a cysteine-rich chitin-binding domain (Kramer and Muthukrishnan, 1997). The chitinase from the coleopteran T. molitor has a complex structure, with multiple catalytic and chitin-binding domains. Whilst chitinase activity has been shown to depend on the presence of the catalytic domain alone (Wang et al., 1996, Zhu et al., 2001), the interaction of insect chitinases with chitin in the cuticle and the PM is thought to depend upon the co-ordinated action of the chitin-binding domain and the catalytic domain (Arakane et al., 2003).

The developmentally and hormonally controlled expression patterns of insect chitinase genes have been relatively well characterised, in relation to the role of the enzyme in moulting and metamorphosis. However, the potential for chitinase to be used as an insecticide, via disruption of the endogenously controlled turnover of chitin in the PM or cuticle of exposed insects, has not been extensively studied. Wang et al. (1996) reported that a truncated 46 kDa M. sexta chitinase, purified from transgenic tobacco, was toxic to larvae of the merchant grain beetle (Oryzaephilus mercator), causing 100% larval mortality after 6 days when administered orally at a level of 2% (w/w), although this result was based on a single assay of seven larvae. Ding et al. (1998) tested transgenic tobacco expressing the truncated 46 kDa M. sexta chitinase for effects against tobacco hornworm (M. sexta) and budworm (Heliothis virescens) larvae. The survival and growth of budworm larvae, but not hornworm larvae, was significantly reduced when reared on plants expressing chitinase at a level of 0.02–0.03% of total protein. However, larval growth of both species was significantly reduced (compared to controls) when fed on chitinase-expressing plants coated with sub-lethal concentrations of Bacillus thuringiensis toxin. These results from diet and transgenic plant bioassays are suggestive of the potential of insect chitinase as an insecticide, but have not provided full evidence to demonstrate its effectiveness. The present paper reports data confirming the toxicity of insect chitinases, both when injected and when administered orally, which establish that chitinase can be an effective insecticide.

An additional aim in this study was to investigate the potential use of snowdrop lectin (GNA) to deliver an attached chitinase enzyme to the haemolymph of orally exposed insects. In previous studies, we have shown that GNA is resistant to gut proteolysis and is able to deliver attached peptides and proteins to the blood of insects exposed to diets containing recombinant fusion proteins (Fitches et al., 2002, Fitches et al., 2004). Using this approach an insect neuropeptide (Manse-AS) and an insect specific spider neurotoxin (SFI1), both non-toxic when ingested in isolation, have been shown to result in toxic effects upon Lepidopteran larvae when ingested as GNA-fusion proteins. Delivery to the haemolymph would allow orally ingested chitinase to act on the cuticle, whereas chitinase in the gut could only affect the peritrophic membrane, and thus the fusion protein might be expected to be more effective as an insecticide than chitinase when delivered orally. Experiments to test this hypothesis are described.

Section snippets

Insect culture

Lacanobia oleracea were reared continuously on artificial diet (Bown et al., 1997) at 25 °C under a 16:8 h light:dark regime.

Materials and general cloning methods

Oligonucleotide primers were synthesised by Sigma-Genosys Ltd. (www.genosys.co.uk). Sub-cloning was carried out using the TOPO cloning kit (pCR2.1 TOPO vector) purchased from Invitrogen (www.invitrogen.com). Restriction endonucleases and T4 DNA ligase were from Promega (www.promega.com), and plasmid DNA was prepared using Promega Wizard miniprep kits. Standard GNA was

Isolation of a cDNA encoding a chitinase gene from L. oleracea larvae

Degenerate primers corresponding to conserved residues of aligned insect chitinase sequences were used in RT-PCR reactions to amplify a 750 bp product from pre-pupal L. oleracea RNA. The amplified product contained an open reading frame (ORF), which was compared to proteins in the GenBank database using BlastP. The amplified sequence had greatest similarity to insect chitinase enzymes. Primers designed from regions of the 750 bp amplification product were used in rapid amplification of cDNA

Discussion

Recombinant chitinase and a chitinase/GNA fusion protein were produced using a Pichia expression system and purified by hydrophobic interaction chromatography (Fig. 2). Previous studies using tobacco transformed with a M. sexta chitinase gene with a predicted molecular weight of 85 kDa (Wang et al., 1996, Ding et al., 1998) showed that plants expressed a truncated, less active 46 kDa chitinase as a result of proteolysis in planta. Similarly considerable problems were encountered with cleavage

Acknowledgements

The authors thank Dr. Subbaratnam Muthukrishnan, Department of Biochemistry, Kansas State University, Manhattan, KS 66506, USA for the kind provision of anti-chitinase antibodies, and Mrs. Christine Richardson for assistance with microscopy. This work was supported by the Pesticides Safety Directorate of the Department for Environment, Food and Rural Affairs.

References (27)

Cited by (0)

View full text