When administered in artificial diets, the cysteine protease inhibitor (PI) oryzacystatin I (OC-I) induced moderate but significant growth inhibition on the pea aphid (Acyrthosiphon pisum Harris), the cotton/melon aphid (Aphis Gossypii Glover) and the peach potato aphid (Myzus persicae Sulzer). In transgenic oilseed rape plants (Brassica napus L. cv. Drakkar) expressing OC-I under the control of the CaMV-35S RNA promoter, oryzacystatin was detected in both leaves and phloem sap. Transgenic plants from three independent homozygous lines were used to test the effect of in planta-expression of OC-I on M. persicae. Mean adult weight and fecundity were significantly reduced, and aphid biomass produced in 2 weeks was decreased by 25–40% for aphids fed transgenic plants, when compared to those fed control plants. The effects of OC-I on M. persicae were correlated with the decrease of a major cathepsin L/H-type cysteine protease activity, detected in whole insect extracts. Immuno-histological analysis showed OC-I labeling along the gut epithelium, but also its association with aphid bacteriocytes and oenocytes. These results suggest that OC-I affects M. persicae through digestive tract targets, but also by reaching the haemolymph, thereby inhibiting extra-digestive proteolytic activities and interacting with functions related to aphid reproduction. Overall, it appears that PIs can display deleterious effects against phloem-feeding insects in addition to their activity on leaf-feeding insects.

The effects of mustard trypsin inhibitor MTI-2 expressed at different levels in transgenic tobacco, arabidopsis and oilseed rape lines have been evaluated against three different lepidopteran insect pests. 1. Plutella xylostella (L.) larvae were the most sensitive to the ingestion of MTI-2. The inhibitor expressed at high levels in arabidopsis plants caused rapid and complete mortality. High mortality and significantly delayed larval development were also detectable in oilseed rape expressing MTI-2 at lower levels. 2. Mamestra brassicae (L.) larvae were sensitive only at high MTI-2 expression level, as obtained in transgenic tobacco and arabidopsis, whereas no effects were observed for larvae fed on plants showing relatively low expression levels such as those of oilseed rape lines. 3. Feeding bioassays with Spodoptera littoralis (Boisduval) larvae were carried out using the same oilseed rape lines, showing that at these low expression levels no mortality was observed although a delay in larval development did occur. The levels of insect gut proteolytic activities of the larvae still alive at the end of a 7 day feeding bioassay were usually higher than in the controls, but no new proteinases were expressed in any case. The combined results described in this paper demonstrate altogether the relevance of a case-by-case analysis [target insects and proteinase inhibitor (PI) level of expression in planta] in a PI-based strategy for plant protection.

Oilseed rape is an important crop in Europe. Many insect pests, mainly Coleoptera, feed on various parts of this plant at different cultural stages. Creation of genotypes more resistant to these pests via the expression of entomopathogenic proteins in transgenic plants could reduce the need for chemical control. We thus investigated a new crop protection strategy based on the expression of proteinase inhibitors (PI) in transgenic oilseed rape to limit attacks by coleopteran pests. A serine PI (CII from soybean) and a cysteine PI (OCI from rice), able to inhibit the corresponding proteolytic activities of these coleoptera in vitro, were introduced via Agrobacterium in a spring genotype. The expression level of the introduced PIs was determined and homozygous plants were obtained. The impact of ingestion of OCI or/and CII expressed in transgenic plants was evaluated on growth, development and mortality of four pests feeding on oilseed rape (Baris, cabbage stem flea beetle, rape stem weevil and cabbage seed weevil). Although there were no marked deleterious effects on the tested insects, various reactions were observed at the gut proteinase level. The results are discussed along with prospects for improving pest resistance in oilseed rape. Moreover, since this plant is highly attractive to pollinating insects, possible deleterious effects on honeybees were assessed. Honeybees mainly possess serine gut proteinases, and in vitro feeding bioassays allowed us to determine a threshold dose (50-100 mug/g pollen) of a serine PI, beyond which negative effects were observed. Consumption of high doses of this PI (much higher than those produced in leaves) induced protease overproduction. However, oilseed rape expressing serine PI, under control of the CaMV 35S promoter, does not threaten honeybees, since the transgene is not expressed in pollen and nectar.

Larvae of Baris coerulescens Scop. (Coleoptera: Curculionidæ) exhibit a complex array of gut proteinase activities comprising cysteine and serine proteinases. The major cysteine proteinase activity, showing an optimum at pH 6.0, corresponds to at least 4 different proteinases. On the contrary, the minor serine proteinase activity, with an optimum at pH 9.0, seems to be due essentially to a single proteinase. The cysteine proteinase inhibitor oryzacystatin I (OC-I) inhibits completely the cysteine proteinase activity in vitro. However, larval growth and survival were not significantly different on control and transgenic oilseed rape plants expressing high levels of active OC-I. In larvae grown on transgenic plants, cysteine proteinase activity was dramatically decreased, whereas serine proteinase activity was increased by more than 2-fold, when compared to larvae raised on control plants. For both activities, no new proteinase was detected in insects fed plants expressing OC-I. These results suggest that partial compensation of the inhibition of cysteine proteinase activity by the increase in serine proteinase activity allowed the larvae to overcome the effects of OC-I consumption. This case illustrates problems that could arise when trying to achieve high levels of protection for plants against Coleopteran pests possessing a complex digestive proteinase pool.

Plant genetic engineering offers opportunities for the creation of insect-resistant plants by insertion and expression in planta of entomopathogenic proteins. Two main approaches to obtain such plants have been explored. The first one involves the use of delta-endotoxin coding sequences originating from the bacterium Bacillus thuringiensis. The second approach uses plant-derived genes, such as those encoding enzyme inhibitors or lectins. Much work throughout the world is devoted to obtaining plants of different species expressing such genes and showing resistance to insect pests. Research projects under development, and an assessment of the situation and of the problems encountered on the way to a commercial use of such transgenic plants are discussed in this review.

The resistance of a transgenic line of oilseed rape expressing constitutively the cysteine proteinase inhibitor oryzacystatin I (OCI) was assessed against Psylliodes chrysocephala L. (Coleoptera: Chrysomelidae). The levels of OCI expression in the transformed line averaged 0.2% and 0.05% of total soluble protein in leaves and petioles respectively. In vitro analyses showed that P. chrysocephala larvae use both cysteine and serine proteinases for protein digestion, and that all the cysteine proteolytic activity is OCI-sensitive. However, bioassays showed that adults fed identically on leaf discs from control or transformed plants. When larvae were reared on transgenic plants expressing OCI, they showed an increase in weight gain compared to those reared on control plants. Furthermore, those larvae from transgenic plants exhibited a 2-fold increase in both cysteine and serine proteolytic activity as a reponse to the presence of OCI. The plasticity of insect digestive physiology and feeding behaviour are discussed, as well as the relevance of engineering a genotype expressing both types of proteinase inhibitors.

Incorporation of genes encoding proteinase inhibitors into oilseed rape genome could confer resistance to Coleoptera, which are the major pests on rape in Europe. A detailed study of the digestive proteinase of a model cruciferous-feeding Coleoptera, Phaedon cochleariae, showed that this insect relies on a complex proteolytic system including serine, cysteine, aspartyl proteinases, and leucine aminopeptidases. The inhibition of general and specific activities by a range of proteinase inhibitors in vitro suggested that oryzacystatin I (OCI) and Bowman–Birk inhibitor (BBI) would have adverse effects when ingested by the larvae. However, the growth and the feeding of larvae reared on oilseed rape leaf discs treated with a high dose of OCI and/or BBI were not affected. Moreover, the levels and patterns of proteolytic activities were not modified in these larvae. The study of the interactions between P. cochleariae larval proteinases and OCI and BBI revealed that both inhibitors were rapidly cleaved by serine proteinases in association with leucine aminopeptidases, and consequently lost their inhibitory capacity. This mechanism of resistance is very efficient, and may be widespread among Coleoptera. The major implications for insect control using proteinase inhibitor-based strategies are discussed.

The aim of this study was to assess the potential effect of a transgenic line of oilseed rape expressing oryzacystatin I (OCI) on two strains of cabbage seed weevil. The level of OCI expression in seeds was approximately 0.05% of total soluble proteins. The insects were field-collected in two different locations, and their progeny was analyzed after a 3 week-development in pods. Both strains showed a similar pattern of proteolytic activity, and similar levels of OCI-sensitive proteinase activity in vitro. However, the larvae showed differential susceptibility to the transgenic plants. Despite inhibition of digestive proteinases in vitro by OCI in both strains, one strain showed an increased growth rate when fed the transgenic seeds, while the other strain remained unaffected. While suggesting the importance of studying individuals from different populations when assessing the effect of proteinase inhibitor-expressing plants on insect growth, our results also point out the necessity of studying the biochemical interactions taking place in vivo between the recombinant inhibitors and their target proteinases.

Genetic transformation has been used to improve poplar tolerance toward Chrysomela tremulae, a coleoptera causing severe damages in nurseries and young poplar plantations. We have shown in an in vitro assay that cysteine proteinase represent die major proteinase activity in the midgut of C. tremulae. Moreover, in this system the cysteine proteinase inhibitor OCI effectively inhibits most of the digestive proteinase activity. This proteinase inhibitor and the Bacillus thuringiensis delta-endotoxin CRY IIIA, also known to be active against coleoptera, were both evaluated for their toxicity against C. tremulae. Transgenic poplars expressing either ocI or cry IIIA gene were produced. Insects feeding on this transgenic poplars exhibit reduced. larval growth, altered development and increased mortality when compared to the control.

The biochemical interactions between two cystatins from rice seeds, oryzacystatin I (OCI) and oryzacystatin II (OCII), and the cysteine proteinases from three plant parasitic nematodes, Meloidogyne hapla. M. incognita and M. javanica , were assessed using standard protease assays and mildly denaturing gelatin/polyacrylamide gel electrophoresis (gelatin/PAGE). Activity detected in extracts of preparasitic second‐stage larvae (J2) from M. hapla was optimal at pH 5.5 and was inhibited in vitro by the cysteine proteinase inhibitors trans ‐epoxysuccinyl‐L‐leucylamido‐(4‐guanidino) butane, hen egg cystatin, OCI, and OCII. As demonstrated by class‐specific activity staining, all the activity measured between pH 3.5 and pH 7.5 was accounted for by a major proteinase form, Mhp1, and two minor forms, Mhp2 and Mhp3. Mhps were also detected in extracts and excretions of parasitic J2 and adult females, indicating their continuous expression throughout development of M. hapla , and their possible involvement in the extracellular degradation of proteins. Interestingly, the two plant cysteine proteinase inhibitors OCI and OCII showed different degrees of affinity for the major proteinase form, Mhp1. Both inhibitors almost completely inactivated this proteinase in native conditions but, unlike OCII, OCI conserved a high affinity for Mhp1 during mildly denaturing gelatin/PAGE, showing the differential stabilities of the OCI/Mhp1 and OCII/Mhp1 complexes. In contrast to Mhp1, the major cysteine proteinases detected in the two closely related species M. incognita and M. javanica were strongly inhibited by OCII, while the inhibition by OCI was partly prevented during electrophoresis. This species‐related efficiency of plant cystatins against nematode cysteine proteinases could have practical implications when planning their use to control nematodes of the genus Meloidogyne .

The aim of this study was to test the potential of proteinase inhibitors to controlChrysomela tremulae, a beetle that causes severe damage in young plantations and in short-rotation intensive culture (SRIC) of poplar. As a first step, cysteine proteinases were determined to be the major digestive proteinases ofC. tremulae and oryzacystatin OCI, a cysteine proteinase inhibitor, was shown to inhibit this activityin vitro. The gene encoding OCI was introduced into poplar (Populus tremula ×P. tremuloides) and transgenic plants expressing OCI at a high level were selected. Feeding tests on these transgenic plants demonstrate the toxicity of OCI-producing poplar leaves againstC. tremulae larvae.