Predicting bee community responses to land-use changes: Effects of geographic and taxonomic biases.

Land-use change and intensification threaten bee populations worldwide, imperilling pollination services. Global models are needed to better characterise, project, and mitigate bees' responses to these human impacts. The available data are, however, geographically and taxonomically unrepresentative; most data are from North America and Western Europe, overrepresenting bumblebees and raising concerns that model results may not be generalizable to other regions and taxa. To assess whether the geographic and taxonomic biases of data could undermine effectiveness of models for conservation policy, we have collated from the published literature a global dataset of bee diversity at sites facing land-use change and intensification, and assess whether bee responses to these pressures vary across 11 regions (Western, Northern, Eastern and Southern Europe; North, Central and South America; Australia and New Zealand; South East Asia; Middle and Southern Africa) and between bumblebees and other bees. Our analyses highlight strong regionally-based responses of total abundance, species richness and Simpson's diversity to land use, caused by variation in the sensitivity of species and potentially in the nature of threats. These results suggest that global extrapolation of models based on geographically and taxonomically restricted data may underestimate the true uncertainty, increasing the risk of ecological surprises.

Responses of 9 lepidopteran species to Bacillus thuringiensis: How useful is phylogenetic relatedness for selecting surrogate species for nontarget arthropod risk assessment?

To evaluate phylogenetic relatedness as a proxy for susceptibility to Bacillus thuringiensis (Bt) when selecting species to act as surrogates for others in prerelease testing, we examined the responses of 11 laboratory-reared lepidopteran colonies, comprising members of 2 families, 5 genera, and 9 species, to a commercial Bt preparation. Survival, pupal mass, and timing of pupation and adult emergence of 2 noctuids (Spodoptera litura and Helicoverpa armigera) and 7 tortricids (Cnephasia jactatana, Ctenopseustis obliquana, Ctenopseustis herana,Planotortrix octo, Planotortrix notophaea,Planotortrix excessana [2 different laboratory colonies], and Epiphyas postvittana [2 colonies]) were examined after feeding first instar larvae with artificial diet containing 5 µL/100 mL Dipel ES (Bt subsp. kurstaki). Bt caused significant larval mortality in all species except S. litura, in which only pupation was delayed compared with untreated controls. Neither of the noctuid species tested would act as a suitable surrogate for the other in tests of Bt impacts on survival. With the exception of the 2 colonies of E. postvittana, which differed from each other not only in their responses to Bt but also in their development times when not treated with Bt, species within each tortricid genus had similar responses to Bt and thus could act as surrogates for each other. Members of different genera within this family could represent each other only if relatively coarse measurement endpoints (e.g., toxic or not) were considered adequate for assessing risks to nontarget species in the field.

Potential use of an arthropod database to support the non-target risk assessment and monitoring of transgenic plants.

Worldwide, plants obtained through genetic modification are subject to a risk analysis and regulatory approval before they can enter the market. An area of concern addressed in environmental risk assessments is the potential of genetically modified (GM) plants to adversely affect non-target arthropods and the valued ecosystem services they provide. Environmental risk assessments are conducted case-by-case for each GM plant taking into account the plant species, its trait(s), the receiving environments into which the GM plant is to be released and its intended uses, and the combination of these characteristics. To facilitate the non-target risk assessment of GM plants, information on arthropods found in relevant agro-ecosystems in Europe has been compiled in a publicly available database of bio-ecological information during a project commissioned by the European Food Safety Authority (EFSA). Using different hypothetical GM maize case studies, we demonstrate how the information contained in the database can assist in identifying valued species that may be at risk and in selecting suitable species for laboratory testing, higher-tier studies, as well as post-market environmental monitoring.

The use of biotin-binding proteins for insect control.

Biotin-binding proteins (BBPs), expressed in transgenic plants, are insecticidal to a very wide range of insects. The expression levels required are generally low (approximately 100 ppm), and although higher than required for Bacillus thuringiensis (Bt) delta-endotoxins, BBPs are effective across a broader range of insect orders and other invertebrates than the Bt Cry proteins. Avidin and streptavidin, in particular, have been reported as causing death or severe growth reduction in at least 40 species of insects across five insect orders (Lepidoptera, Coleoptera, Orthoptera, Diptera, and leaf-eating Hymenoptera) and mites. In addition, due largely to its rapid dilution in ecosystems, no adverse impacts on nontarget microorganisms or invertebrates have been recorded. Because the target, biotin, cannot itself be modified to prevent it binding to BBPs and remain effective as a vitamin, the major avenue open to insects to develop resistance is unavailable. Two properties of the biotin-avidin complex make it highly suitable for use in transgenic plant crop protection strategies against a large range of insects; its extreme stability and its resistance to proteolysis. However, because the nutritional value of the plant could potentially be compromised in the absence of biotin supplementation, its use in nonfood crops such as fiber, forestry, and biofuel crops is seen as the most suitable initial focus for this technology.

Developing risk hypotheses and selecting species for assessing non-target impacts of GM trees with novel traits: the case of altered-lignin pine trees.

A procedure is presented for developing environmental risk hypotheses associated with the deployment of forest trees genetically modified to have altered wood properties and for selecting non-target species to test these hypotheses. Altered-lignin Pinus radiata trees intended for use in New Zealand are used as a hypothetical case study to illustrate our approach. Firstly, environmental management goals (such as wood production, flood control or preservation of biodiversity) were identified and linked to the forest attributes they require. Necessary conditions for each attribute were listed and appropriate assessment endpoints for them developed. For example, biological control of pests may be one condition necessary for a forest to have healthy trees, and the diversity and abundance of natural enemy species in the forest could be an appropriate assessment endpoint for measuring this condition. A conceptual model describing the relationships between an altered-lignin GM pine tree and potentially affected invertebrates and micro-organisms in a plantation forest was used to develop a set of risk hypotheses describing how the GM trees might affect each assessment endpoint. Because purified lignin does not represent the properties it imparts to wood, maximum hazard dose tests with non-target organisms, as are used to inform toxin risk assessment, cannot be conducted. Alternative experiments, based on current knowledge of the responses of organisms to lignin, must be designed. A screening method was adapted and applied to a database of invertebrate species known to inhabit New Zealand pine forests to identify and prioritize non-target invertebrate species that could be used as experimental subjects for examining these hypotheses. The screening model and its application are presented, along with a set of recommendations for pre-release tests with GM pines and potentially affected invertebrates and micro-organisms.

No unintended impacts of transgenic pine (Pinus radiata) trees on above ground invertebrate communities.

As part of an investigation into the potential unintended ecological impacts of transgenic trees, invertebrates were sampled from a field trial of transgenic Pinus radiata D. Don carrying the expressed antibiotic resistance marker gene neomycin phosphotransferase II (nptII) along with other genes known to affect reproductive development in plants and from nontransformed control trees. Invertebrate species abundance, richness, diversity, and composition were measured and compared among trees of five different transclones and nontransformed isogenic control trees. Invertebrates were sampled at six-monthly intervals over a period of 2 yr. In total, 19,162 individuals were collected comprising 279 invertebrate recognizable taxonomic units. Total invertebrate populations as well as populations of herbivorous lepidopteran larvae and Hemiptera were compared among transgenic and control trees. The results show that the transclones had no significant unintended influence on species abundance, richness, diversity, or composition for all populations investigated.

Developing biosafety risk hypotheses for invertebrates exposed to GM plants using conceptual food webs: a case study with elevated triacylglyceride levels in ryegrass.

Regulators are acutely aware of the need for meaningful risk assessments to support decisions on the safety of GM crops to non-target invertebrates in determining their suitability for field release. We describe a process for developing appropriate, testable risk hypotheses for invertebrates in agroecosystems that might be exposed to plants developed by GM and future novel technologies. An existing model (PRONTI) generates a ranked list of invertebrate species for biosafety testing by accessing a database of biological, ecological and food web information about species which occur in cropping environments and their potential interactions with a particular stressor (Eco Invertebase). Our objective in this contribution is to explore and further utilise these resources to assist in the process of problem formulation by identifying potentially significant effects of the stressor on the invertebrate community and the ecosystem services they provide. We propose that for high ranking species, a conceptual food web using information in Eco Invertebase is constructed, and using an accepted regulatory risk analysis framework, the likelihood of risk, and magnitude of impact for each link in the food web is evaluated. Using as filters only those risks evaluated as likely to extremely likely, and the magnitude of an effect being considered as moderate to massive, the most significant potential effects can be identified. A stepwise approach is suggested to develop a sequence of appropriate tests. The GM ryegrass plant used as the "stressor" in this study has been modified to increase triacylglyceride levels in foliage by 100% to increase the metabolisable energy content of forage for grazing animals. The high-ranking "test" species chosen to illustrate the concept are New Zealand native species Wiseana cervinata (Walker) (Lepidoptera: Hepialidae), Persectania aversa (Walker) (Lepidoptera: Noctuidae), and the self-introduced grey field slug, Deroceras reticulatum (Müller).

A screening method for prioritizing non-target invertebrates for improved biosafety testing of transgenic crops.

We have developed a screening method that can be used during the problem formulation phase of risk assessment to identify and prioritize non-target invertebrates for risk analysis with any transgenic plant. In previously published protocols for this task, five criteria predominated. These criteria have been combined by our method in a simple model which assesses: (1) the possible level of risk presented by the plant to each invertebrate species (through measurements of potential hazard and exposure, the two principal criteria); (2) the hypothetical environmental impact of this risk (determined by the currently known status of the species' population in the ecosystem and its potential resilience to environmental perturbations); (3) the estimated economic, social and cultural value of each species; and (4) the assessed ability to conduct tests with the species. The screening method uses information on each of these criteria entered into a specially designed database that was developed using Microsoft Access 2003. The database holds biological and ecological information for each non-target species, as well as information about the transgenic plant that is the subject of the risk assessment procedure. Each piece of information is then ranked on the basis of the value of the information to each criterion being measured. This ranking system is flexible, allowing the method to be easily adapted for use in any agro-ecosystem and with any plant modification. A model is then used to produce a Priority Ranking of Non-Target Invertebrates (PRONTI) score for each species, which in turn allows the species to be prioritized for risk assessment. As an example, the method was used to prioritize non-target invertebrates for risk assessment of a hypothetical introduction of Bacillus thuringiensis (Bt) Cry1Ac-expressing Pinus radiata trees into New Zealand.

Distribution and residual activity of two insecticidal proteins, avidin and aprotinin, expressed in transgenic tobacco plants, in the bodies and frass of Spodoptera litura larvae following feeding.

To understand how a major cosmopolitan pest responds to two very different insecticidal proteins and to determine whether herbivorous insects and their frass could be environmental sources of recombinant proteins from transgenic plants, Spodoptera litura (Fab.) (Lepidoptera, Noctuidae) larvae were fed on tobacco leaves expressing either the biotin-binding protein, avidin, or the protease inhibitor, aprotinin. Control larvae received non-transgenic tobacco. Samples of larvae were taken after 5, 6 or 7 days' feeding and frass was collected after two 24-h periods at 6 and 7 days. Insects in all treatments grew significantly during the experiment, but the avidin-fed larvae were significantly smaller than the others on Day 7. Avidin was found in all samples of avidin-fed larvae (7.0+/-0.86 ng mg(-1), n=45), at a lower level than in their frass (31.9+/-5.08 ng mg(-1), n=30), and these frass levels were lower than those of the the leaves fed to the larvae (69.0+/-6.71 ng mg(-1), n=45). All of the avidin detected in these samples was capable of binding biotin. On average, between 10 and 28% of avidin was recovered with the methods used, whereas almost full recovery of aprotinin was effected. Aprotinin levels in larvae (8.2+/-0.53 ng mg(-1), n=45) were also lower than aprotinin levels in frass (77.4+/-6.9 ng mg(-1), n=30), which were somewhat lower than those in the leaves fed to the larvae (88.6+/-2.51 ng mg(-1), n=45). Approximately half the trypsin-binding ability of aprotinin was lost in larvae, and in frass, aprotinin had lost about 90% of its ability to bind trypsin.

Effects of plants genetically modified for insect resistance on nontarget organisms.

Insect resistance, based on Bacillus thuringiensis (Bt) endotoxins, is the second most widely used trait (after herbicide resistance) in commercial genetically modified (GM) crops. Other modifications for insect resistance, such as proteinase inhibitors and lectins, are also being used in many experimental crops. The extensive testing on nontarget plant-feeding insects and beneficial species that has accompanied the long-term and wide-scale use of Bt plants has not detected significant adverse effects. GM plants expressing other insect-resistant proteins that have a broader spectrum of activity have been tested on only a limited number of nontarget species. Little is known about the persistence of transgene-derived proteins in soil, with the exception of Bt endotoxins, which can persist in soil for several months. Bt plants appear to have little impact on soil biota such as earthworms, collembolans, and general soil microflora. Further research is required on the effects of GM plants on soil processes such as decomposition. Assessment of nontarget impacts is an essential part of the risk assessment process for insect-resistant GM plants.

The expression of a mammalian proteinase inhibitor, bovine spleen trypsin inhibitor in tobacco and its effects on Helicoverpa armigera larvae.

The cDNA for bovine spleen trypsin inhibitor (SI), a homologue of bovine pancreatic trypsin inhibitor (BPTI), including the natural mammalian presequence was expressed in tobacco using Agrobacterium tumefaciens-mediated transformation. Stable expression required the N-terminal targeting signal presequence although subcellular localization was not proven. SI was found to exist as two forms, one coinciding with authentic BPTI on western blots and the second marginally larger due to retention of the C-terminal peptide. Both were retained on a trypsin-agarose affinity gel and had inhibitory activity. Newly emergent leaves contained predominantly the large form whereas senescent leaves had little except the fully processed form present. Intermediate-aged leaves showed a gradual change indicating that a slow processing of the inhibitor peptide was occurring. The stability of SI was shown by the presence of protein at high levels in completely senescent leaves. Modifications to the cDNA (3' and 5' changes and minor codon changes) resulted in a 20-fold variation in expression. Expression of modified SI in transgenic tobacco leaves at 0.5% total soluble protein reduced both survival and growth of Helicoverpa armigera larvae feeding on leaves from the late first instar. In larvae surviving for 8 days, midgut trypsin activity was reduced in SI-tobacco fed larvae, while chymotrypsin activity was increased. Activities of leucine aminopeptidase and elastase-like chymotrypsin remained unaltered. The use of SI as an insect resistance factor is discussed.

Avidin expressed in transgenic tobacco leaves confers resistance to two noctuid pests, Helicoverpa armigera and Spodoptera litura.

Fertile transgenic tobacco plants with leaves expressing avidin in the vacuole have been produced and shown to halt growth and cause mortality in larvae of two noctuid lepidopterans, Helicoverpa armigera and Spodoptera litura. Late first instar H. armigera larvae and neonate (< 12-h-old) S. litura larvae placed on leaves excised from T0 tobacco expressing avidin at 3.1-4.6 microM (micromoles/kg of fresh leaf tissue) had very poor growth over their first 8 days on the leaves, significant numbers had died by days 11 or 12 and all were dead by day 22 (H. armigera) or day 25 (S. litura). Similar results were obtained when late first instar H. armigera larvae were placed on leaves from T1 plants expressing avidin at six different average concentrations, ranging from 3.7 to 17.3 microM. Two larvae on the lowest expressing leaves survived to pupation, but there was total mortality among the other groups and no relationship between avidin concentration and the effects on the larvae. Synergistic effects between avidin-expressing tobacco plants and a purified Bt toxin, Cry1Ba, were demonstrated. Late instar H. armigera larvae fed with leaves from T2 plants expressing avidin at average concentrations of either <5.3 or > 12.9 microM, and painted with Cry1Ba protein at a rate equivalent to an expression level of 0.5% of total leaf protein, died significantly faster than larvae given either of the two treatments alone. Larvae fed with avidin-expressing leaves painted with the protease inhibitor, aprotinin, at a rate equivalent to 1% of total leaf protein had mortality similar to those given avidin-leaves alone. There was no evidence of antagonism between these two proteins.