Relevance of Bt toxin interaction studies for environmental risk assessment of genetically modified crops.

In recent years, different Bacillus thuringiensis (Bt) toxin-encoding genes have been combined or 'stacked' in genetically modified (GM) crops. Synergism between Bt proteins may occur and thereby increase the impact of the stacked GM event on nontarget invertebrates compared to plants expressing a single Bt gene. On the basis of bioassay data available for Bt toxins alone or in combination, we argue that the current knowledge of Bt protein interactions is of limited relevance in environmental risk assessment (ERA).

Cross-order and cross-phylum activity of Bacillus thuringiensis pesticidal proteins.

The increasing number of Bacillus thuringiensis proteins with pesticidal activities across orders and phyla raises the question how widespread cross-activities are and if they are of sufficient biological significance to have implications for ecological safety of those proteins in pest control applications. Cross-activity is reported for 27 proteins and 69 taxa and is substantiated by reasonable evidence (mortality estimates) in 19 cases involving 45 taxa. Cross-activities occur in 13 primary rank families across three classes of pesticidal proteins (Cry, Cyt and Vip), and comprise 13 proteins affecting species across two orders, five proteins affecting three orders and one protein affecting four orders, all within the class Insecta. Cross-activity was quantified (LC50 estimates) for 16 proteins and 25 taxa. Compared to toxicity ranges established for Diptera-, Coleoptera-, Lepidoptera- and Nematoda-active proteins, 13 cross-activities are in the low-toxicity range (10-1000µg/ml), 12 in the medium - (0.10-10µg/ml) and two in the high-toxicity range (0.01-0.10µg/ml). Although cross-activities need to be viewed with caution until they are confirmed through independent testing, current evidence suggests that cross-activity of B. thuringiensis pesticidal proteins needs to be taken into consideration when designing and approving their use in pest control applications.

Activity of Bacillus thuringiensis cyt1Ba crystal protein against hymenopteran forest pests.

A crystal-spore suspension of PS201T6 was toxic to larvae of Diprion similis (Hymenoptera: Diprionidae). Toxicity was at least in part attributable to the Cyt1Ba crystal protein, as demonstrated by bioassays of solubilized protein produced by Escherichia coli expressing PS201T6's cyt1Ba gene. PS201T6 reduced survival and growth of D. similis in a 2-week field experiment. In laboratory bioassays, both toxin and parental strain affected Acantholyda erythrocephala (Pamphiliidae), Pikonema alaskensis (Tenthredinidae), and Neodiprion sertifer (Diprionidae), as well as spruce budworm, Choristoneura fumiferana (Lepidoptera: Tortricidae). Affecting insects across at least four orders (Diptera, Coleoptera, Hymenoptera, Lepidoptera), Cyt1Ba has the broadest insecticidal activity spectrum among Bacillus thuringiensis crystal proteins documented to date.

Towards the development of an autocontamination trap system to manage populations of emerald ash borer (Coleoptera: Buprestidae) with the native entomopathogenic fungus, Beauveria bassiana.

Emerald ash borer, Agrilus planipennis Fairmaire (Coleoptera: Buprestidae) is an invasive species from Asia that was discovered in North America Canada, in 2002. Herein, we describe studies to develop an autocontamination trapping system to disseminate Beauveria bassiana to control beetle populations. The standard trap for emerald ash borer in Canada is a light green prism trap covered in an insect adhesive and baited with (Z)-3-hexenol. We compared of green multifunnel traps, green intercept panel traps (both with and without fluon coating) and green prism traps for capturing emerald ash borer in a green ash plantation. The coated green multifunnel traps captured significantly more males and more females than any other trap design. We examined the efficacy of two native B. bassiana isolates, INRS-CFL and L49-1AA. In a field experiment the INRS-CFL isolate attached to multifunnel traps in autocontamination chambers retained its pathogenicity to emerald ash borer adults for up to 43 d of outdoor exposure. Conidia germination of the INRS-CFL isolate was >69% after outdoor exposure in the traps for up to 57 d. The L49-1AA isolate was not pathogenic in simulated trap exposures and the germination rate was extremely low (<5.3%). Mean (+/- SEM) conidia loads on ash borer adults after being autocontaminated in the laboratory using pouches that had been exposed in traps out of doors for 29 d were 579,200 (+/- 86,181) and 2,400 (+/- 681) for the INRS-CFL and the LA9-1AA isolates, respectively. We also examined the fungal dissemination process under field conditions using the L49-1AA isolate in a green ash plantation. Beetles were lured to baited green multifunnel traps with attached autocontamination chambers. Beetles acquired fungal conidia from cultures growing on pouches in the chambers and were recaptured on Pestick-coated traps. In total, 2,532 beetles were captured of which 165 (6.5%) had fungal growth that resembled B. bassiana. Of these 25 beetles were positive for the L49-1AA isolate.

Complete rRNA sequence, arrangement of tandem repeated units and phylogeny of Nosema fumiferanae from spruce budworm, Choristoneura fumiferana (Clemens).

We provide molecular systematics of a microporidian species, Nosema fumiferanae, one of the most common natural enemies of spruce budworm, Choristoneura fumiferana. The uncharacterized flanking region upstream of the large subunit (LSU) rRNA and the complete rRNA cistron of N. fumiferanae was 4,769 bp long. The organization of the rRNA gene was 5'-LSU rRNA-ITS-SSU rRNA-IGS-5S-3' and corresponded primarily to most insect (i.e. lepidopteran) Nosema species identified and classified to date. Phylogenetic analysis based on the complete rRNA cistron indicated that N. fumiferanae is closely related to Nosema plutellae and is correctly assigned to the "true" Nosema group. Suggestions were provided on a criterion to delineate the "true" Nosema from other microsporidian species.

Prevalence of Nosema sp. (Microsporidia: Nosematidae) during an outbreak of the jack pine budworm in Ontario.

Microsporidia are believed to play little or no role in outbreaks of the jack pine budworm, Choristoneura pinuspinus Freeman (Lepidoptrera: Tortricidae), because the short duration (2-4 years) of those outbreaks may not permit significant build-up of the pathogen. We conducted the first survey of Nosema sp. (Microsporidia: Nosematidae) over the course of a recent jack pine budworm outbreak in Ontario. Between 2004 and 2010 the outbreak defoliated a cumulative total of 1.78 million ha. Microscopic examination of ~15,000 overwintering larvae collected over 6 years in sites with densities of 3 larvae per branch or more revealed widespread occurrence of Nosema at generally high infection intensities. The pathogen was present in 69.5% of the 518 plots that were monitored. Prevalence of infection was generally low (below 40% in 84% of plots with infected larvae) but reached high levels (80-95%) locally and increased rapidly in most infestations within 1-2 years of onset. We hypothesize that the habit of early-instar larvae to feed on developing male flowers (pollen cones) after spring emergence is critical in allowing rapid build-up of Nosema by increasing efficiency of horizontal transmission (higher density of both infected larvae and egested spores). Nosema infection may contribute to the complexity of jack pine budworm outbreak patterns by affecting egg recruitment and early-instar survival at the stand level in concert with known effects of budworm-induced reductions in pollen cone production on those processes.

Morphology, molecular characteristics and prevalence of a Cystosporogenes species (Microsporidia) isolated from Agrilus anxius (Coleoptera: Buprestidae).

A microsporidium was isolated from the bronze birch borer, Agrilus anxius Gory (Coleoptera: Buprestidae), collected near Sudbury and Sault Ste Marie, Canada. Light and electron microscopic investigations showed that gross pathology and ultrastructure of the investigated Cystosporogenes species was similar to those characterized and described for other Cystosporogenes species. Small subunit rRNA gene sequence data and comparative phylogenetic analysis confirmed that the microsporidian species from A. anxius is most closely related to the genus Cystosporogenes clade of microsporidia. Infection average in the Sudbury and Sault Ste Marie beetle populations was >80% and relatively stable in 2006-2007 but declined in 2008. Field prevalence of the A. anxius isolate, mechanisms that may potentially be involved in its horizontal (autoinfection) and vertical (transovarial) transmission, and disease dynamics are discussed. The congeneric relationship between Agrilus planipennis and A. anxius makes it imperative to study the virulence of this Cystosporogenes species in A. planipennis.

Interactions between Bacillus thuringiensis subsp. kurstaki HD-1 and midgut bacteria in larvae of gypsy moth and spruce budworm.

We examined interaction between Bacillus thuringiensis subsp. kurstaki HD-1 (Foray 48B) and larval midgut bacteria in two lepidopteran hosts, Lymantria dispar and Choristoneura fumiferana. The pathogen multiplied in either moribund (C. fumiferana) or dead (L. dispar) larvae, regardless of the presence of midgut bacteria. Inoculation of L. dispar resulted in a pronounced proliferation of enteric bacteria, which did not contribute to larval death because B. thuringiensis was able to kill larvae in absence of midgut bacteria. Sterile, aureomycin- or ampicillin-treated larvae were killed in a dose-dependent manner but there was no mortality among larvae treated with the antibiotic cocktail used by Broderick et al. (2006, 2009). These results do not support an obligate role of midgut bacteria in insecticidal activity of HD-1. The outcome of experiments on the role of midgut bacteria may be more dependent on which bacterial species are dominant at the time of experimentation than on host species per se. The L. dispar cohorts used in our study had a microflora, that was dominated by Enterococcus and Staphylococcus and lacked Enterobacter. Another factor that can confound experimental results is the disk-feeding method for inoculation, which biases mortality estimates towards the least susceptible portion of the test population.

Activity of spores and extracellular proteins from six Cry+ strains and a Cry- strain of Bacillus thuringiensis subsp. kurstaki against the western spruce budworm, Choristoneura occidentalis (Lepidoptera: Tortricidae).

We characterized insecticidal activity of previously untested strains of Bacillus thuringiensis kurstaki belonging to two crystal serovars (K-1 and K-73) against the western spruce budworm (Choristoneura occidentalis Freeman 1967). By testing various components, we demonstrated that spores play a critical role in the pathogenesis of each strain. Spore-free crystals caused low mortality and purified spores were generally not toxic. The addition of spores to purified protoxin increased toxicity several hundred-fold, regardless of the parental strain from which the spores or protoxins were derived. The crystal and spore components did not account for full insecticidal activity of whole sporulated cultures owing to the toxicity of soluble proteins that are secreted during cell growth. We observed a marked difference in toxicity of secreted proteins between the K-1 and K-73 type strains, with the K-1 preparations causing much higher mortality, mass reduction, and inhibition of pupation. There was a consistent correlation between relative toxicity of secreted protein preparations and the presence and quantity of the Vip3A protein, suggesting that this protein contributes to the virulence of B. thuringiensis subsp. kurstaki in western spruce budworm larvae. However, other virulence factors have to be invoked to explain the synergizing effect of spores from both K-1 and K-73 strains on Cry protein toxicity.

Insecticidal activity of Bacillus thuringiensis crystal proteins.

Published data on insecticidal activity of crystal proteins from Bacillus thuringiensis are incorporated into the Bt toxin specificity relational database. To date, 125 of the 174 holotype known toxins have been tested in approximately 1700 bioassays against 163 test species; 49 toxins have not been tested at all; 59 were tested against 71 Lepidoptera species in 1182 bioassays; 53 toxins were tested against 23 Diptera species in 233 bioassays; and 47 were tested against 39 Coleoptera species in 190 bioassays. Activity spectra of the tested toxins were summarized for each order. Comparisons of LC(50) values are confounded by high variability of the estimates, mostly due to within-species variation in susceptibility, and errors associated with estimation of toxin protein content. Limited analyses suggest that crystal protein toxicity is not affected by quarternary toxin rank or host used for gene expression, but that pre-ingestion treatment by solubilization or enzymatic processing has a large effect. There is an increasing number of toxin families with cross-order activity, as 15 of the 87 families (secondary rank) that are pesticidal are active against more than one order. Cross-order activity does not threaten environmental safety of B. thuringiensis-based pest control because toxins tend to be much less toxic to taxa outside the family's primary specificity range.

Response of Lymantria dispar L. (Lepidoptera: Lymantriidae) to Bacillus thuringiensis subsp. kurstaki at different ingested doses and temperatures.

We examined mortality and feeding inhibition response of Lymantria dispar L. (Lepidoptera: Lymantriidae) larvae to ingested doses of Bacillus thuringiensis subsp. kurstaki as a function of dose, instar and temperature. We developed generalized (logistic) linear mixed models and a mixture survival model, commonly used in medical statistics, to analyze the complex data set. We conducted bioassays of Foray 48B with larvae from the NJSS laboratory stock, using droplet imbibing or force-feeding to ensure dose ingestion. The dose causing mortality in 50% of the test population (LD(50)) under standard test conditions (22 degrees C) ranged from 0.019 International Units (IU)/larva for first instar larvae (L(1)) to 1.6 IU/larva for L(4). Temperature affected larval mortality in two ways. Mortality occurred sooner and progressed more rapidly with increasing temperature (13-25 degrees C) at each dose level and instar, while the maximum level of mortality attained by each instar decreased with increasing rearing temperature. The mechanisms underlying this effect are being investigated. Larvae that survived exposure to B. thuringiensis resumed feeding after a period that was dependent on instar, dose, and temperature. The equations describing observed mortality and feeding recovery responses were used to construct a simulation model, which was able to predict both processes, and which forms the basis for a process-oriented model that can be used as a decision support tool in aerial sprays.

Purification of Vip3Aa from Bacillus thuringiensis HD-1 and its contribution to toxicity of HD-1 to spruce budworm (Choristoneura fumiferana) and gypsy moth (Lymantria dispar) (Lepidoptera).

We developed a protocol for obtaining high yields (10-15 mg per 1100 ml of culture supernatant) of highly purified (up to 95%) Vip3Aa protein from HD-1 cultures. The protocol is based on acetone precipitation of supernatant protein, followed by HPLC fractionation (DEAE-5PW column) and several concentration steps. Our protocol resulted in higher yields and purity of Vip3Aa than a previously published method [Estruch, J.J., Warren, G.W., Mullins, M.A., Nye, G.J., Craig, J.A., Koziel, M.G., 1996. Vip3A, a 353 novel Bacillus thuringiensis vegetative insecticidal protein with a wide spectrum of 354 activities against lepidopteran insects. Proc. Nat. Acad. Sci. USA 93, 5389-5394.]. This was achieved by using acetone rather than ammonium sulfate for precipitation of proteins from culture supernatants, and a shallow rather than a steep NaCl gradient for elution of the toxin, and by conducting all the purification steps at low temperature to prevent toxin degradation. In bioassays of the purified protein, Choristoneura fumiferana and Lymantria dispar larvae were less susceptible than Spodopteraexigua (10- and approximately 100-fold, respectively). A B. thuringiensis var. kurstaki strain HD-1 from which the vip3Aa gene had been deleted (EG12414) showed reduced toxicity to S. exigua relative to the unmodified parental strain (EG2001), but not to L. dispar or C. fumiferana. We interpret these results as indicating that the Vip3Aa toxin does not contribute measurably to pathogenicity of HD-1 in these species.

Molecular data and phylogeny of Nosema infecting lepidopteran forest defoliators in the genera Choristoneura and Malacosoma.

Nosema isolates from five lepidopteran forest defoliators, Nosema fumiferanae from spruce budworm, Choristoneura fumiferana; a Nosema sp. from jack pine budworm, Choristoneura pinus pinus and western spruce budworm, Choristoneura occidentalis (Nosema sp. CPP and Nosema sp. CO, respectively); Nosema thomsoni from large aspen tortrix, Choristoneura conflictana; and Nosema disstriae, from the forest tent caterpillar, Malacosoma disstria were compared based on their small subunit (SSU) ribosomal RNA (rRNA) gene sequences. Four of the species sequenced, N. fumiferanae, Nosema sp. CPP, Nosema sp. CO, and N. disstriae have a high SSU rDNA sequence identity (0.6%-1.5%) and are members of the "true Nosema" clade. They all showed the reverse arrangement of the (large subunit [LSU]-internal transcribed spacer [ITS]-SSU) of the rRNA gene. The fifth species, N. thomsoni has the usual (SSU-ITS-LSU) arrangement and is not a member of this clade showing only an 82% sequence similarity. We speculate, therefore, that a genetic reversal may have occurred in the common ancestor to the "true Nosema" clade. Although, the mechanism for rearrangement of the rRNA gene subunits is not known we provide a possible explanation for the localization. N. fumiferanae, Nosema sp. CPP, and Nosema sp. CO clustered together on the inferred phylogenetic tree. The high sequence similarities, the reverse arrangement in the rRNA gene subunits, and the phylogenetic clustering suggest that these three species are closely related but separate species.

Vertical transmission of Nosema fumiferanae (Microsporidia: Nosematidae) and consequences for distribution, post-diapause emergence and dispersal of second-instar larvae of the spruce budworm, Choristoneura fumiferana (Clem.) (Lepidoptera: Tortricidae).

We examined vertical transmission of Nosema fumiferanae in the eastern spruce budworm, Choristoneura fumiferana (Clem.) (Lepidoptera: Tortricidae), and how it affects overwintering distribution and survival and spring emergence and dispersal of second-instar larvae in outbreak populations. Females containing 5.0 x 10(5) spores or more consistently produced 100% infected progeny. Transmission efficiency was still 50% at burdens as low as 0.2 x 10(5) spores per moth. Infection intensity in offspring increased with maternal spore load but became highly variable above 25 x 10(5) spores per female. Nosema multiplied in second instars for at least 1 month after they entered dormancy, regardless of temperature (2 degrees C versus 21 degrees C). Infection did not affect the distribution of overwintering larvae in a white spruce canopy. Dormancy survival between late-summer and the following spring was lower in families from infected females and was negatively correlated with larval infection intensity. Infection delayed larval emergence from hibernacula in the spring and resulted in delayed dispersal of emerged larvae, at least when parasite prevalence and infection intensities were high. Infected larvae were less successful in establishing feeding sites after dispersal. Our results underscore the potential of Nosema infection to negatively affect processes early in the budworm life cycle.

Bacillus thuringiensis subsp. kurstaki aerial spray prescriptions for balsam fir stand protection against spruce budworm (Lepidoptera: Tortricidae).

Although commercial formulations of Bacillus thuringiensis subsp. kurstaki (Btk) are being widely used in forest protection against lepidopteran defoliators, optimal application prescriptions have often yet to be worked out in detail. We conducted field experiments over a 6-yr period (1996-2001) in southwestern Québec to determine application prescriptions for optimal protection of balsam fir, Abies balsamea (L.), healthy stands against the spruce budworm, Choristoneura fumiferana (Clemens) (Lepidoptera: Tortricidae). At moderate larval densities (<30 larvae per 45-cm branch tip), similar foliage protection was achieved with one or two Btk applications of 30 billion international units per hectare (BIU/ha). When larval densities exceeded 30 larvae per branch tip, two successive applications of 30 BIU/ha significantly increased foliage protection. Whether the second application took place 5 or 10 d after the first spray did not affect treatment efficacy. Increasing the application dosage from 30 to 50 BIU/ha did not lead to better foliage protection against high larval densities, but the current standard dosage of 30 BIU/ha saved more foliage than 15 BIU/ha against moderate populations. The recommended dosage of 30 BIU can be applied in lower application volumes (1.5 liters/ha) by using a high-potency product (20 BIU/liter), because we did not observe a reduction in efficacy compared with the application of a lower potency product (12.7 BIU/liter) in 2.37 liters/ha. We also demonstrated that Btk can be applied much earlier in the season without compromising spray efficacy: there was no difference in treatment efficacy of double applications at 30 BIU/ha when the first spray was timed for early third, peak third, or early fourth instars.

Occurrence of Cystosporogenes sp. (Protozoa, Microsporidia) in a multi-species insect production facility and its elimination from a colony of the eastern spruce budworm, Choristoneura fumiferana (Clem.) (Lepidoptera: Tortricidae).

We have isolated a microsporidium from a laboratory colony of the eastern spruce budworm, Choristoneura fumiferana (Clem.) (Lepidoptera: Tortricidae). Light and electron microscopic investigations showed that gross pathology and ultrastructure of our isolate are similar to those described for Cystosporogenes legeri from the European grape vine moth, Lobesia botrana. Comparative phylogenetic analysis of the small subunit rDNA using maximum likelihood, maximum parsimony, and neighbour joining distance methods revealed perfect homology with the C. legeri sequence. The microsporidian was infectious to other Choristoneura species, as well as Malacosoma disstria, Lymantria dispar, and Lambdina fiscellaria. Incubation of infected egg masses at 41 degrees C for 20 min followed by 30 min in 33% formaldehyde did not reduce disease incidence in larval offspring. Exposure of one or two generations to fumagillin at 6000 ppm or higher eliminated infection in adult moths, but also reduced colony fitness. A clean colony was established by conducting individual matings and selecting disease-free offspring.

The Bacillus thuringiensis Cry1Aa toxin: effects of trypsin and chymotrypsin site mutations on toxicity and stability.

The objective of the present work was to create an active Cry1Aa toxin showing enhanced resistance to degradation by spruce budworm (Choristoneura fumiferana) midgut proteases by mutating potential chymotrypsin and trypsin sites. Fourteen Cry1Aa mutants were created in an Escherichia coli-Bacillus shuttle vector and expressed in a crystal minus Bacillus thuringiensis host. Using spruce budworm gut juice, commercial bovine trypsin and chymotrypsin we performed protease resistance assays with Cry1Aa wild type and mutant toxins. Although many mutants showed little or no change, several mutants showed a > 2-fold increase (R543S, R566G, and F570S) up to a > 4-fold increase in toxicity (F576S), in bioassay studies against C. fumiferana. The in vitro protease resistance assay results indicated a possible involvement of other gut juice components in toxin overdigestion.

Interaction between Cry9Ca and two Cry1A delta-endotoxins from Bacillus thuringiensis in larval toxicity and binding to brush border membrane vesicles of the spruce budworm, Choristoneura fumiferana Clemens.

A genetically altered variant of Cry9Ca from Bacillus thuringiensis shows high potency against the spruce budworm, Choristoneura fumiferana Clemens. Its activity, as measured by feeding inhibition in frass-failure assays, is estimated to be four to seven times greater than B. thuringiensis subsp. kurstaki HD-1, the strain currently used in commercial products to control this insect. Bioassays against budworm of mixtures of the modified Cry9Ca and two of the Cry1A endotoxin proteins produced by HD-1 show neither synergism nor antagonism. Experiments with brush border membrane vesicles from budworm midgut revealed that Cry9Ca and the Cry1A toxins share a common binding site and that bound Cry9Ca can be displaced from the membrane to some extent by the Cry1A toxins. However, it is uncertain whether the binding site is actually the receptor molecule or a membrane protein associated with pore formation.