No Impact of Transgenic <I>npt</I>II<I>-leafy Pinus radiata</I> (Pinales: Pinaceae) on Pseudocoremia suavis (Lepidoptera: Geometridae) or Its Endoparasitoid Meteorus pulchricornis (Hymenoptera: Braconidae)

TRANSGENIC PLANTS & INSECTS

No Impact of Transgenic nptII-leafy Pinus radiata (Pinales: Pinaceae) on Pseudocoremia suavis (Lepidoptera: Geometridae) or Its Endoparasitoid Meteorus pulchricornis (Hymenoptera: Braconidae) E.P.J. BURGESS,1 E. I. BARRACLOUGH,1 A. M. KEAN,1 C. WALTER,2

AND

L. A. MALONE1,3

Environ. Entomol. 40(5): 1331Ð1340 (2011); DOI: http://dx.doi.org/10.1603/EN11116

ABSTRACT To investigate the biosafety to insects of transgenic Pinus radiata D. Don containing the antibiotic resistance marker gene nptII and the reproductive control gene leafy, bioassays were conducted with an endemic lepidopteran pest of New Zealand plantation pine forests and a hymenopteran endoparasitoid. Larvae of the common forest looper, Pseudocoremia suavis (Butler), were fed from hatching on P. radiata needles from either one of two nptII-leafy transgenic clones, or an isogenic unmodiÞed control line. For both unparasitized P. suavis and those parasitized by Meteorus pulchricornis (Wesmael), consuming transgenic versus control pine had no impact on larval growth rate or mass at any age, larval duration, survival, pupation or successful emergence as an adult. Total larval duration was 1 d (3%) longer in larvae fed nptII-2 than nptII-1, but this difference was considered trivial and neither differed from the control. In unparasitized P. suavis larvae, pine type consumed did not affect rate of pupation or adult emergence, pupal mass, or pupal duration. Pine type had no effect on the duration or survival of M. pulchricornis larval or pupal stages, mass of cocoons, stage at which they died, adult emergence, or fecundity. Parasitism by M. pulchricornis reduced P. suavis larval growth rate, increased the duration of the third larval stadium, and resulted in the death of all host larvae before pupation. The lack of impact of an exclusive diet of nptII-leafy transgenic pines on the life history of P. suavis and M. pulchricornis suggests that transgenic plantation pines expressing nptII are unlikely to affect insect populations in the Þeld. KEY WORDS np-ENDITALII selectable marker gene, leafy gene, biosafety, risk assessment, insect host-parasitoid

Over the last 15 yr, plantings of genetically modiÞed (GM) annual crops have grown from zero to 148 million ha throughout the world (James 2010), making this one of the most successful new agricultural technologies of recent times. Genetic modiÞcation of plantation trees for enhanced resistance to forest pests and diseases, increased wood Þber strength, reduction of lateral branching and cone production, and resistance to herbicides are seen as fundamental enhancements that would add value to production forests (Walter 2004) and provide a faster alternative to conventional breeding techniques for these long-lived plant species. More recently, forest trees have been identiÞed as a source of providing biomaterials and biofuels, thereby providing an alternative to fossil fuels and also mitigating the effects of climate change (Fenning et al. 2008). Examples of commercially deployed GM trees include virus-resistant papaya (Carica papaya L.) in Hawaii and insect-resistant Bt poplars in China (Fer1 The New Zealand Institute for Plant & Food Research Ltd, Private Bag 92169, Auckland 1142, New Zealand. 2 New Zealand Forest Research Institute Ltd Research (Scion), 49 Sala Street, Rotorua, New Zealand. 3 Corresponding author, e-mail: louise.malone@plantandfood. co.nz.

reira et al. 2002, Ewald et al. 2006, Lin et al. 2006, Walter et al. 2010). However, the prospect of GM trees raises particular public concerns prompted in part by the emotional and cultural connections many people feel to trees and forests, as well as their importance in landscapes and in providing a “sense of place” (Carter et al. 2009). Many people value the recreational opportunities that forests offer in a “natural” setting. The strong values-driven component of human attitudes to trees and forests suggests that public concern will demand a high degree of proof of safety for transgenic plantation forests. Given the longevity of trees and the complexity of the interactions between trees and other components of their surrounding ecosystems, including their provision of habitats and resources for other organisms, comprehensive data are required to assess whether GM trees will have unintended impacts on ecosystems and nontarget organisms before their Þeld release (Henderson and Walter 2006, Farnum et al. 2007, Finstad et al. 2007). To date, despite ⬎700 Þeld studies with trees worldwide, no such impacts have been observed (Walter et al. 2010 and references therein). In spite of this evidence, and the comfort and familiarity with GM crops now felt by citizens of countries such as the United States of America, which

0046-225X/11/1331Ð1340$04.00/0 䉷 2011 Entomological Society of America

1332

ENVIRONMENTAL ENTOMOLOGY

have embraced this technology, others remain unconvinced of its safety. Environmental risk assessment is a process that is informed by scientiÞc evidence but is ultimately values-driven, i.e., the aspects of the environment that we aim to protect from adverse effects are determined by citizensÕ concerns (USEPA 1998). Thus, environmental risk assessment is context- and country-speciÞc. For example, New Zealand has a very conservative stance toward GM crops, with a moratorium on Þeld trials between 2000 and 2003 and none yet approved for commercial release. The European Cooperation in Science and Technology (eCOST) Action FP0905 on the biosafety of forest transgenic trees emphasizes concern that “the potential for unintended consequences of pleiotropic effects following transgene expression may be enhanced in long-lived forest trees” (http://www.cost-actionfp0905.eu/). The citizens of New Zealand and some European countries may demand a standard of proof of transgenic tree safety requiring research that would not be deemed a high priority in the United States. Precedents for prioritizing research based on citizensÕ (rather than scientistsÕ) concerns may be found in ongoing efforts to excise selectable antibiotic resistance marker genes from transgenic plants after transformation and the development of ÔintragenicÕ plants as alternatives to transgenics (Gleave et al. 1999, Wang et al. 2005, Ballester et al. 2008, Chakraborti et al. 2008, Fladung et al. 2010, Rommens et al. 2011, Rosellini 2011). Here we investigate the hypothesis that modifying Pinus radiata D. Don with the selectable antibiotic resistance marker gene neomycin phosphotransferase II (nptII) and the leafy gene involved in the control of reproductive development in plants (Moyroud et al. 2009) will not alter the development or survival of a lepidopteran feeding exclusively on the plant or a parasitoid attacking this lepidopteran host. The pines were modiÞed with leafy as part of a wider study to investigate impacts of altering reproductive genes on ßowering and cone formation, with the longer term goals of inducing earlier ßowering to enable faster breeding and production of nonßowering transgenic pines. Sterile trees would direct more resources into wood production instead of development of reproductive structures and also would circumvent the production and environmental distribution of transgenic pollen. The insect species studied here were the New Zealand endemic common forest looper, Pseudocoremia suavis (Butler) (Lepidoptera: Geometridae) and its self-introduced endoparasitoid Meteorus pulchricornis (Wesmael) (Hymenoptera: Braconidae: Euphorinae). This study complements Þeld studies involving the parent trees from which those in the current study were cloned as well as other transclones containing nptII and modiÞed genes affecting reproductive development. In those Þeld studies, the treesÕ impacts on above-ground invertebrate and below-ground rhizosphere microbial communities associated with the trees were assessed. The transclones did not affect

Vol. 40, no. 5

invertebrate species abundance, diversity, richness, or composition, although there were seasonal differences in these measures (Schnitzler et al. 2010). Small seasonal shifts were detected in rhizosphere microbial communities, as were occasional transient differences between control and transgenic trees, but these were not considered to constitute a meaningful impact (Lottmann et al. 2010). Environmental risk assessments are mandatory before Þeld release of any transgenic plants in New Zealand (HSNO Act 1996). The relevant legislation places particular emphasis on the preservation of all native and valued introduced ßora and fauna. Pinus radiata is a valued introduced plant species, comprising 90% of New ZealandÕs 1.6 million ha of plantation forests. P. suavis is a common native foliagefeeding lepidopteran in these forests (Lavery and Mead 1998), which is thought to be generally kept at subpest levels by a suite of natural enemies (Alma 1977) including M. pulchricornis. Pseudocoremia suavis populations have occasionally reached outbreak sizes in New Zealand P. radiata plantations, causing signiÞcant damage and mortality to trees (White 1974). Meteorus pulchricornis, which lays its eggs within the host larva and emerges from the host as a third-instar larva to spin a cocoon in which to pupate, comprised 97.5% of the parasitoids emerging from P. suavis larvae Þeld-collected over 18 mo from 2008 to 10 in a P. radiata forest north of Auckland, New Zealand (E.P.J.B., unpublished data). Using no-choice tests and a tritrophic system comprising P. radiata, P. suavis, and M. pulchricornis to simulate a “worst-case scenario” we examined the potential impacts of transgenic nptII-leafy pine trees on these nontarget insects. Materials and Methods Plant Material. Three isogenic lines of P. radiata were used; two of these were transgenic, produced from independent transformation events by using biolistic transformation to insert novel genetic material into embryonic tissue. These transclones were assigned the names nptII-1 and nptII-2. The third was an untransformed isogenic control line. The transformation vector was based on the plasmid pUC 19 (Yanisch-Perron et al. 1985) and contained the selectable antibiotic resistance marker gene neomycin phosphotransferase II (nptII) under the control of the CaMV 35S promoter (Walter et al. 1998). The vector also contained the bla ampicillin resistance gene for selection in Escherichia coli (Migula) Castellani and Chalmers, regulated by a bacterial promoter that does not lead to expression in P. radiata. The vector further contained the leafy gene related to the control of reproductive development in Arabidopsis thaliana (L.) Heynh. (Weigel et al. 1992) under the control of a double CaMV 35S promoter. Putatively transgenic tissue was initially isolated through nptII selection by using the aminoglycoside antibiotic Geneticin. The integration and expression of the introduced genes in transgenic tissue, plants, or both were subsequently

October 2011

BURGESS ET AL.: NO IMPACT OF TRANSGENIC PINE ON P. suavis

conÞrmed by polymerase chain reaction (PCR), Southern hybridization, and nptII ELISA assays (Grace et al. 2005). Plantlets produced from these and other transformation events as well as isogenic, untransformed control plants were grown and assessed for 4 yr in a containment greenhouse. In 2003, P. radiata trees representing several transclones and control trees were transferred to a fenced Þeld site to enable investigation of potential environmental impacts (Lottmann et al. 2010, Schnitzler et al. 2010). Plants used in the current study originated from cuttings taken in 2004 from trees studied in this Þeld trial. The cuttings were kept in 1-liter pots within the Þeld trial site for 30 mo. They were then transferred to a containment glasshouse under natural light:dark conditions at ⬇18 Ð 20⬚C for a further 16 mo. Trees were ⬇0.6 m high and 4 yr old (4 yr from cuttings) when used in the current study. Eleven transgenic plants from the nptII-1 transclone, nine plants from the nptII-2, and ten untransformed control plants were used in bioassays. At 4 yr of age, the trees were not sufÞciently mature for normal reproductive development to be initiated and the absence of early ßowering suggests the leafy gene was not functional at this stage. To feed the P. suavis larvae used in this study, young needles attached to fascicles, picked from 2 to 10 cm below the growing tips of branches (i.e., young, but not the newest light-green growth), were sterilized in 0.1% aqueous sodium hypochlorite for 7 min to remove any external pathogens, rinsed well, and dried before use. Needles were taken from several plants of each of the three types of pine on each harvesting occasion. Insects. Pseudocoremia suavis larvae were obtained from a laboratory colony established from larvae and moths Þeld-collected at Woodhill Forest (36o 36⬘32.6⬙ S, 174o 16⬘41.7⬙ E), 45 miles north of Auckland, New Zealand. Individuals were initially line-bred to ensure the colony would be free of disease. The colony was maintained at Plant & Food Research, Auckland, on a diet of surface-sterilized P. radiata foliage. Meteorus pulchricornis were sourced from a laboratory colony established from individuals that emerged from Helicoverpa armigera (Hu¨ bner) (Lepidoptera: Noctuidae) larvae Þeld collected in HawkeÕs Bay, New Zealand. This thelytokous (parthenogenetic diploid female only) strain is likely to have originated from a similar strain in Asia (Fuester et al. 1993, Berry and Walker 2004). The M. pulchricornis colony was maintained using Spodoptera litura (F.) (Lepidoptera: Noctuidae) reared on artiÞcial diet (McManus and Burgess 1995) as the host species. The S. litura colony was established from larvae and moths collected in Queensland, Australia. Meteorus pulchricornis adults were kept en masse in ventilated 4-liter plastic jars and were provided with honey and cotton wool wicks soaked in 10% sucrose solution. Females used in experiments had emerged from cocoons between 2 and 4 wk before experiments and, to gain experience of parasitism, had

1333

been exposed to S. litura hosts for a period of 2 h before provision of P. suavis hosts in experiments. Bioassays. Excess neonate P. suavis were randomly assigned to be reared either on control, nptII-1, or nptII-2 pine needles. Newly hatched larvae were individually transferred to a ventilated 4-ml plastic cup containing a fascicle with three needles, and kept at 20⬚C. Larvae were kept in darkness for 7 d to encourage settling and feeding behavior, and were then exposed to a photoperiod of 16:8 (L:D) h for the duration of the experiment. After 7 d, most larvae had moulted into the second larval instar. To maximize the potential effects of the pine type on the interaction between host and parasitoid, the second stadium of P. suavis was selected as the earliest stadium likely to give good rates of successful parasitism and one that would provide the greatest period of exposure of the parasitoid to any effects of pine type on the host. All larvae were weighed when 7 d old. Individuals with outlying weights were eliminated and six groups of larvae with similar mean mass and even mass distribution selected. These were randomly assigned to the six treatments consisting of combinations of parasitized or unparasitized larvae with the three pine types (control, nptII-1, or nptII-2). Larvae were parasitized the day after weighing and assignment to treatments, when larvae were 8 d old. Two jars, each containing 45 M. pulchricornis, were used sequentially for parasitizing, with one larva from each of the three parasitized treatments being exposed in rotation, and in random order within each rotation until all hosts were parasitized. Larvae assigned to be parasitized were individually transferred, using a Þne sable brush, into a jar with M. pulchricornis. Each larva was closely observed until a parasitism event was seen and the larva immediately transferred to a new 4-ml clear plastic container with a fresh needle of the appropriate pine treatment type. Larvae assigned to be unparasitized were simply transferred to a new container with fresh food. All larvae were checked daily, fed with fresh needles ad libitum, and transferred into a larger 40-ml clear plastic container when they reached sufÞcient size. Daily records were kept of host larval instar, molting, and survival, parasitoid emergence from the host, and host and parasitoid pupation and adult emergence. The mass of neonate larvae was estimated by determining the mean mass of 10 groups of 10 neonates. Host larvae were individually weighed on the day before parasitism when 7 d old, then when 13 d old, and then regularly every 3 or 4 d until pupation or death. The six treatments were weighed in randomized order on each weigh day. Larval growth rate was determined as the rate of change of mass through time (mg/mg⫺1/d⫺1). Pseudocoremia suavis pupal mass was measured as soon as the chitinous pupal case had hardened and turned a rich brown color. Sex was established in the pupal stage. After the emergence of M. pulchricornis larvae from the host and completion of cocoon spinning, each cocoon was weighed, kept in an individual 4-ml clear plastic container, and weighed again 4 d later.

1334

ENVIRONMENTAL ENTOMOLOGY

Once adult parasitoids emerged, they were individually kept in 75-ml clear plastic containers with screw-top lids, with a 1-cm section of cotton wool dental wick, soaked in 10% honey solution and held in the lid of a 4-ml plastic cup to prevent leakage of solution. Each adult parasitoid was frozen 18 d after emergence from its cocoon, as it was estimated this would coincide with the middle of the plateau of peak fertility (Fuester et al. 1993). These adults were later dissected by placing on a microscope slide, grasping the abdomen and ovipositor with pairs of Þne forceps, and pulling the ovipositor until the ovaries came free. A small droplet of water was then placed on the extracted parts and a coverslip lowered to enable counting of all mature eggs, using a compound microscope at 80⫻ magniÞcation. The total sample size needed to detect meaningful differences in a range of life history parameters was determined by prospective power analysis (Steidl et al. 1997) using the data of Barraclough et al. 2009 as well as data from a range of previous experiments with P. suavis and M. pulchricornis raised on P. radiata in our laboratory. We chose to apply a high statistical power (95%) when designing this study given the need to ensure a low probability of making a type II error (inappropriately rejecting the null hypothesis) when assessing the environmental risks of GM plants (Andow 2003). The availability of extensive previous data enabled us to set the power of the analysis at this high level. Meaningful effects were generally considered to be alterations of 10 Ð15% of the mean of a given parameter. Prospective power analysis showed that at 95% power, a total sample size of 45 individuals per treatment would enable detection of a 11% difference in relative growth rate of unparasitised P. suavis larvae, whereas 75 individuals would be needed to detect a 12% difference in the relative growth rate of parasitized larvae. Further analysis for P. suavis indicated that at 95% power, a sample size of 40 per treatment would detect a 9% difference in larval duration, a 5% difference in pupal duration, and a 7% difference in pupal mass. For M. pulchricornis, at 95% power, 40 individuals per treatment would detect a 10% difference in the time spent within the host larva, a 4% difference in time within the cocoon, a 14% difference in cocoon mass and a 10% difference in fertility. The experiment was repeated three times (three randomized complete blocks), with between 15 and 18 larvae being assigned to each of the six treatments in the Þrst block. Taking into account the mortality observed in the Þrst experimental block, particularly among parasitized P. suavis larvae before parasitoid emergence, as well as the number of parasitized larvae needed to detect a meaningful treatment effect on relative growth rate, the number of larvae assigned to the three parasitized treatments was increased to 30 and set at 15 for the unparasitized treatments in the second and third blocks. The total number of individuals assigned to each of the three unparasitized treatments ranged from 46 to 48 and for the three parasitized treatments N was 77 or 78.

Vol. 40, no. 5

Fig. 1. Survival through time of parasitized (para) and unparasitized (unpara) P. suavis larvae fed control, nptII-1 or nptII-2 pine, from parasitism by Meteorus pulchricornis when larvae were eight d old to pupation or death. Unparasitized larvae were removed from the data when they pupated.

Statistical Analyses. Data were analyzed using SAS 9.1 (SAS Institute 2000 Ð2004), R version 2.90 (R Development Core Team 2009), GenStat (Lawes Agricultural Trust 2007), and Minitab 15 (Minitab 2006). The data for time until an event, such as pupation, molting or death, etc., were analyzed using survival package version 2.35Ð7 in R, with differences in survival curves assessed using Log-Rank tests. Mass data were modeled using the Mixed Procedure in SAS, with overall tests for differences constructed using type 3 sums of squares and the KenwardÐRoger method for degrees of freedom. The relative growth rate model also included an auto-regressive covariance structure of order one to account for the repeated measures on individuals. The proportions of individuals in the different “fate” categories (see Results) were analyzed using a generalized logistic regression in the SAS Logistic procedure. Results Survival and Fates of P. suavis and M. pulchricornis. The majority (89%) of unparasitized P. suavis successfully pupated (126 of 141 larvae), although mortality was higher during the pupal stage with 23% (29 of 126 pupae) dying before emergence as a moth. There was no effect of pine type (control, nptII-1, or nptII-2) on the survival of unparasitized P. suavis larvae or pupae (Fig. 1), or on the ultimate fate of these individuals, i.e., whether they died as a larva or pupa or emerged as an adult (PearsonÕs ␹2 ⫽ 0.76, df ⫽ 4, P ⫽ 0.94) (Fig. 2). Parasitized P. suavis larvae survived for 19 d on average, but all died before pupation. Duration of larval survival was unaffected by pine type (control: N ⫽ 73, median ⫽ 20 d, C.I. ⫽ (19,21); nptII-1: N ⫽ 73, median ⫽ 19 d, C.I. ⫽ (18,20); nptII-2: N ⫽ 70, median ⫽ 19 d, C.I. ⫽ (15,20); ␹2 ⫽ 0.014, df ⫽ 2, P ⫽ 0.99) (Fig. 1). There was no effect of pine type on the fate of M. pulchricornis, i.e., the likelihood of

October 2011

BURGESS ET AL.: NO IMPACT OF TRANSGENIC PINE ON P. suavis

1335

Fig. 2. Proportions of unparasitized P. suavis larvae fed either control, nptII-1 or nptII-2 pine, that died as larvae, died as pupae or emerged as adults.

host death while containing a parasitoid larva, parasitoid death either as a larva or pupa after emergence from the host, or the parasitoid reaching adulthood (Pearson ␹2 ⫽ 4.27, df ⫽ 4, P ⫽ 0.37) (Fig. 3). Over the three treatments, 86 of 216 (40%) host larvae died before emergence of a M. pulchricornis larva from the host body. Only 10 of the remaining 130 M. pulchricornis (8% of those emerging from hosts, 5% of the initial total) died after emergence from the host either as a larva or pupa and 120 (56% of the total) successfully developed to the adult stage.

Growth and Development of Unparasitized and Parasitized P. suavis. Unparasitized P. suavis larvae accumulated mass at a steady rate over 27 d, until metamorphosis began before pupation (Fig. 4). There were no differences on any of the weighing occasions in unparasitized larval mass among groups of larvae raised on the three types of pine. Relative growth rate was calculated as the rate of change of mass through time (mg/mg⫺1/d⫺1) for the period when larvae were aged 13Ð27 d, when the natural log of larval mass plotted against time provided a straight line. Larval mass increased by around 0.23 mg/mg⫺1/d⫺1 and

Fig. 3. Proportions of M. pulchricornis parasitizing Pseudocoremia suavis larvae fed either control, nptII-1 or nptII-2 pine that died within their hosts, died within their own cocoons, or emerged successfully as adults. “Died in host” refers to host death before emergence of a parasitoid larva, which also kills the parasitoid. “Died in cocoon” includes parasitoid larvae that emerged from hosts and failed to spin a cocoon successfully, as well as those that formed a cocoon but died either as a larva or pupa before emerging as an adult wasp.

1336

ENVIRONMENTAL ENTOMOLOGY

Fig. 4. Mass through time of parasitized (para) and unparasitized (unpara) P. suavis larvae fed control, nptII-1 or nptII-2 pine, from hatching to pupation, parasitoid emergence or death. The arrow shown at larval age of 8 d indicates the time at which larvae were exposed to M. pulchricornis for parasitizing.

there were no growth rate differences among the different pine type treatments within the unparasitized larvae (control: N ⫽ 48, mean ⫽ 0.232 mg/mg⫺1/ d⫺1, SE ⫽ 0.0036; nptII-1: N ⫽ 47, mean ⫽ 0.238 mg/mg⫺1/d⫺1, SE ⫽ 0.0038; nptII-2: N ⫽ 46, mean ⫽ 0.228 mg/mg⫺1/d⫺1, SE ⫽ 0.0037; F ⫽ 1.89; df ⫽ 2, 105; P ⫽ 0.16). Parasitized larvae grew more slowly and gained less total mass than unparasitized larvae (Fig. 4). There were no differences in mass among the pine type treatments at any time, and larvae gained mass at around 0.165 mg/mg⫺1/d⫺1 with no effect of treatment on growth rate (control: N ⫽ 78, mean ⫽ 0.164 mg/mg⫺1/d⫺1, SE ⫽ 0.0043; nptII-1: N ⫽ 77, mean ⫽

Vol. 40, no. 5

0.170 mg/mg⫺1/d⫺1, SE ⫽ 0.0043; nptII-2: N ⫽ 76, mean ⫽ 0.160 mg/mg⫺1/d⫺1, SE ⫽ 0.0046; F ⫽ 1.29; df ⫽ 2, 113; P ⫽ 0.28). Unparasitized larvae took 10 d from egg hatch to reach the end of the second stadium, 5 d to complete the third, then spent 5 d in the fourth stadium, seven in the Þfth stadium, and 9 d in the sixth stadium (Table 1). Pine type had no effect on the time taken for unparasitized larvae to develop through each of the stadia (Table 1). Total larval duration was around 31 d, and larvae fed on nptII-1 and nptII-2 pine did not differ in total larval duration from those on control pine (control: N ⫽ 43, mean ⫽ 31.5 d, SE ⫽ 0.27; nptII-1: N ⫽ 41, mean ⫽ 31.05 d, SE ⫽ 0.28; F ⫽ 1.47; df ⫽ 1, 78; P ⫽ 0.23; nptII-2: N ⫽ 40, mean ⫽ 32.15 d, SE ⫽ 0.28; F ⫽ 2.72; df ⫽ 1, 77; P ⫽ 0.10). However, a difference of 1.1 d was detected in the total larval duration from hatching to pupation between the two transgenic pine types (F ⫽ 4.01; df ⫽ 2, 116; P ⫽ 0.021). Parasitized P. suavis larvae took longer than unparasitized larvae to complete the third stadium (seven compared with 5 d) (Table 1) but pine type did not affect the times taken for the parasitized P. suavis to develop through either the Þrst and second, third, or fourth larval stadia (Table 1). Only one parasitized host larva completed a Þfth stadium and the remainder died as earlier instars after the emergence of a M. pulchricornis larva. P. suavis pupae have a mass of around 90 mg and pine type had no effect on this (control: N ⫽ 43, mean ⫽ 92.9 mg, SE ⫽ 2.03; nptII-1: N ⫽ 42, mean ⫽ 89.6 mg, SE ⫽ 1.91; nptII-2 N ⫽ 41, mean ⫽ 92.7 mg, SE ⫽ 1.94; F ⫽ 0.90; df ⫽ 2, 123; P ⫽ 0.41). The duration of the P. suavis pupal stage was 17 d, with very little variation within or between treatments and no effect of pine type on pupal duration (control: N ⫽ 44,

Table 1. Median duration (days) (SE and N) of life stages for unparasitized and parasitized Pseudocoremia suavis feeding as larvae on control, nptII-1, or nptII-2 pine Unparasitized

Parasitized

Life stage

Control

nptII-1

nptII-2

Combined

Control

nptII-1

nptII-2

Combined

First and second instars

10 (0.18) 48 5 (0.10) 46 5 (0.15) 46 7 (1.4) 43 9 (0.23) 22 31 (0.60) 43 17 (0.381) 35

10 (0.27) 44 5 (0.11) 43 5 (0.12) 42 8 (0.35) 42 9 (0.26) 14 30 (0.32) 42 17 (0.306) 31

10 (0.19) 44 5 (0.10) 43 5 (0.18) 42 6 (0.59) 42 9 (0.41) 20 32 (0.45) 41 17 (0.325) 31

10 (0.12) 136 5 (0.06) 132 5 (0.08) 130 7 (0.63) 127 9 (0.16) 56 31 (0.41) 126 17 (0.18) 97

10 (0.20) 68 7 (0.20) 53 5 (0.17) 40 (-)

11 (0.16) 56 7 (0.29) 44 6 (0.22) 32 (-)

(-)

10 (0.21) 66 7 (0.17) 54 5 (0.15) 39 7 (-) 1 (-)

(-)

10 (0.12) 190 7 (0.12) 151 5 (0.11) 111 7 (-) 1 (-)

(-)

(-)

(-)

(-)

(-)

(-)

(-)

(-)

Third instar Fourth instar Fifth instar Sixth instar Hatching to pupation Pupa

Late larval instar and pupal data are unavailable for parasitized larvae because of mortality following emergence of the parasitoid.

October 2011

BURGESS ET AL.: NO IMPACT OF TRANSGENIC PINE ON P. suavis

mean ⫽ 16.8 d, SE ⫽ 0.22; nptII-1: N ⫽ 42, mean ⫽ 16.8 d, SE ⫽ 0.22; nptII-2: N ⫽ 42, mean ⫽ 16.8 d, SE ⫽ 0.18; F ⫽ 0.04; df ⫽ 2, 92; P ⫽ 0.96). Growth, Development, and Fertility of M. pulchricornis. After oviposition into the host P. suavis larva, M. pulchricornis took around 21 d to develop from egg hatch through the larval stadia before emerging from the host as a third-instar larva to spin a cocoon in which to pupate. There was no pine type effect on the time between oviposition by M. pulchricornis and the emergence of the parasitoid larva from the host (control: N ⫽ 48, mean ⫽ 21.2 d, SE ⫽ 0.42; nptII-1: N ⫽ 46, mean ⫽ 20.8 d, SE ⫽ 0.43; nptII-2: N ⫽ 36, mean ⫽ 22.0 d, SE ⫽ 0.58; F ⫽ 1.56; df ⫽ 2, 127; P ⫽ 0.22). The mass of cocoons at the time of spinning was unaffected by pine type (control: N ⫽ 46, mean ⫽ 7.1 mg, SE ⫽ 0.23; nptII-1: N ⫽ 45, mean ⫽ 7.5 mg, SE ⫽ 0.24; nptII-2: N ⫽ 36, mean ⫽ 7.4 mg, SE ⫽ 0.21; F ⫽ 0.65; df ⫽ 2, 124; P ⫽ 0.52). There were no differences among cocoon masses 4 d after spinning, although mass had diminished by around 0.7 mg in all treatments (control: N ⫽ 46, mean ⫽ 6.4 mg, SE ⫽ 0.21; nptII-1: N ⫽ 45, mean ⫽ 6.8 mg, SE ⫽ 0.24; nptII-2: N ⫽ 36, mean ⫽ 6.6 mg, SE ⫽ 0.21; F ⫽ 0.76; df ⫽ 2, 124; P ⫽ 0.47). The parasitoid pupal phase (time in cocoon) lasted 10 d and there was no impact of pine type on the duration of this life stage (control: N ⫽ 43, mean ⫽ 10.0 d, SE ⫽ 0.05; nptII-1: N ⫽ 43, mean ⫽ 9.9 d, SE ⫽ 0.09; nptII-2: N ⫽ 34, mean ⫽ 10.1 d, SE ⫽ 0.14; F ⫽ 0.85; df ⫽ 2, 117; P ⫽ 0.43). Egg counts made from dissected female M. pulchricornis 18 d after emergence from the cocoon showed that each female contained around 40 eggs. The pine type on which the host had fed had no effect on the number of eggs in the parasitoid ovary (control: N ⫽ 42, mean ⫽ 39.6 eggs, SE ⫽ 1.85; nptII-1: N ⫽ 42, mean ⫽ 41.2 eggs, SE ⫽ 2.20; nptII-2: N ⫽ 34, mean ⫽ 38.0 eggs, SE ⫽ 1.71; F ⫽ 0.64; df ⫽ 2, 115; P ⫽ 0.53), strongly suggesting no effect on of pine type on parasitoid fertility. Discussion NptII is the most widely used selectable marker gene in commercial crops (Miki and McHugh 2004). Expression of nptII produces a 25 kDa enzyme that catalyzes the phosphorylation of aminoglycoside antibiotics, making them nontoxic to bacteria that would otherwise be susceptible. Biosafety issues raised about the use of nptII relate to the potential transfer of resistance genes via horizontal gene ßow to intestinal or soil micro-organisms, effects on the efÞcacy of antibiotics used in medicine or animal husbandry, toxicity or allergenicity of the expressed protein, vertical gene ßow via out-crossing to wild or weedy relatives, and resulting environmental impacts (Ramessar et al. 2007). These concerns have led to an ongoing consideration and debate, regulatory constraints, and trade barriers (Strauss et al. 2009) and a continuing research effort to examine the biosafety of such plants (Jelenic´ 2003; Bennett et al. 2004; Anonymous 2008; Craig et al. 2008; Demaneche et al. 2008; Lemaux 2008,

1337

2009). Further, widespread effort continues to be directed to excising selectable marker genes from transgenic plants after transformation (Gleave et al. 1999, Wang et al. 2005, Ballester et al. 2008, Chakraborti et al. 2008, Fladung et al. 2010). Much effort has gone into reviewing the safety of nptII in terms of mammalian consumption and whether its use could lead to increased antibiotic resistance in the environment (Nap et al. 1992, Read 2000). Few studies have examined whether nptII expression can directly affect insects, but consumption of nptII transgenic potato foliage by Colorado potato beetle was no different from that of untransformed control plants, whereas expression of the marker gene uidA was associated with increased consumption (Lecardonnel et al. 1999). For both unparasitized and parasitized P. suavis in the current study, consuming transgenic pine versus control pine had no impact on the larval growth rate or mass at any age, the duration of the larval stadia, larval survival, or ultimate fate. Total larval duration was 1.1 d (3%) longer in unparasitized P. suavis feeding on nptII-2 than on nptII-1 pine, but neither differed from the nontransgenic control. In unparasitized larvae, pine type did not affect the rate of pupation, pupal mass, duration of the pupal stage, or the rate of adult emergence. There was no effect of the pine type consumed by the host on the duration or survival of the M. pulchricornis larval or pupal stages, the mass of cocoons, the ultimate fate of individuals, or the number of eggs in the adult parasitoid ovary. In summary, feeding on nptII-leafy transgenic P. radiata compared with control pine had no measurable impact on either the herbivore or its parasitoid. It is difÞcult to envisage a biological mechanism underlying the small but statistically signiÞcant difference in total larval duration of unparasitized P. suavis observed between the nptII-1 and nptII-2 lines, particularly given the absence of differences in all other life history parameters between these two lines and the fact that the development time in neither of these transgenic lines differed from that observed on the control line. Expression of the nptII gene measured in trees of both of these transgenic lines grown in a Þeld trial showed no signiÞcant differences between the transclones (C. W., unpublished data). Longer developmental times, reduced pupal sizes, or both are common effects of reduced food quality in Lepidoptera (Greenberg et al. 2001, Saeed et al. 2010, Barraclough et al. 2009) whether a toxin is present or nutritional content is different. The quality of the food consumed by a lepidopteran host may also result in longer development times, reduced Þnal sizes of parasitoids, or both (Uckan and Ergin 2002, Walker et al. 2007). However, the small increase in time to pupation of unparasitized P. suavis fed on the nptII-2 compared with nptII-1 transgenic foliage in this study (1.1 d) did not translate into any other effect in unparasitized or parasitized hosts or in the parasitoid. Given the many developmental characteristics measured for host and parasitoid in the current study and the absence of any other signiÞcant effects, the 1-d difference found in larval duration is likely to be a statistical artifact that

1338

ENVIRONMENTAL ENTOMOLOGY

has arisen by chance, as would be expected when multiple signiÞcance tests are performed (Bland and Altman 1995). Further, when conducting prospective power analysis during the design of experiments we set the level of a biologically meaningful difference for a life history parameter at 10 Ð15% of the mean (3Ð5 d), thus the 3% (1-d) difference detected between the two transgenic lines is not considered meaningful. Rejection of the null hypothesis in this case would seem likely to represent a type I error. Although the leafy gene was introduced under the control of a double constitutive 35S promoter and expected to express, the 4-yr-old plants used in this study had not reached an age where reproductive development is normally initiated in P. radiata. There was no indication that early initiation of ßowering had occurred either in these plants or in the Þeld-grown trees from which they were cloned, suggesting the leafy gene was not functional at this point in the development of the plants. Two homologues of leafy, i.e., prfll and needly, that are thought to control the formation of reproductive primordia, have since been isolated from P. radiata (Mellerowicz et al. 1998, Mouradov et al. 1998). The nonfunctionality of the leafy gene from A. thaliana introduced into P. radiata may result from sequence differences that have arisen during the evolutionary divergence between gymnosperms and angiosperms. This study provides an example of a nontarget risk assessment of impacts of transgenic metabolically altered plants, P. radiata with nptII and altered ßowering genes, although the introduced leafy gene appears not to be functional in these 4-yr-old trees. We chose to study an herbivore-parasitoid relationship that is signiÞcant in the ecosystem where P. radiata is grown and found no impacts of the transgenic plants on any aspect of the development or survival of either species. These results coincide with those of Schnitzler et al. (2010), who found no impacts of the transclones used in the current study, or others modiÞed with constructs expressing nptII and the leafy, constans, or apetala1 ßowering gene, on the abundance, richness, diversity, or composition of communities of aboveground invertebrate species on P. radiata in the Þeld. Similarly, Lottmann et al. (2010) reported no meaningful impacts of the transgenic trees on soil microorganisms. The diversity of transgenes and the traits that they will confer on novel GM crops is likely to increase, extending the availability of commercial GM varieties beyond the current suite of herbicide-tolerant and insect-resistant GM row crops. Transgenes conferring the ability to withstand abiotic stress, such as drought, frost and salinity, sterility via altered reproductive structures, and other alterations in plant metabolic pathways, are likely to be represented among the “second wave” of commercial GM crops (Hoenicka and Fladung 2006, Wilkinson and Tepfer 2009). Developers of new crops will screen out any novel phenotypes that are economically disadvantageous, but transgenic varieties demonstrating economically positive or neutral phenotypic changes will pose complex

Vol. 40, no. 5

risk assessment challenges for regulators and researchers. Our relatively poor knowledge of the life history parameters of many nontarget species could limit our ability to construct realistic scenarios of exposure to hazards that underpin effective, science-based risk assessment and management. Studies, such as the present one, with key nontarget species and novel GM plants in the laboratory, provide valuable data that will help to reduce uncertainties about the environmental impacts of new GM crops. Acknowledgments Thanks are due to Hancock Natural Resource Group Inc. and their representative Andrea Collings for allowing us to collect insects in the Woodhill Forest. Mark Wohlers of Plant & Food Research provided statistical advice and analyses. Graham Walker and Frances MacDonald of Plant & Food Research supplied Meteorus pulchricornis. New ZealandÕs Foundation for Research Science and Technology funded this research under contract C06X0801.

References Cited Alma, P. J. 1977. Selidosema suavis (Butler) (Lepidoptera: Geometridae). Forest and timber insects in New Zealand, vol. 11., M. J. Nuttal (ed.). Forest Research Institute, New Zealand Forest Service, Rotorua, New Zealand. Andow, D. A. 2003. Negative and positive data, statistical power, and conÞdence intervals. Environ. Biosafety Res. 2: 75Ð 80. Anonymous. 2008. Safety and nutritional assessment of GM plants and derived food and feed: the role of animal feeding trials. Report of the EFSA GMO Panel Working Group on Animal Feeding Trials. Food Chem. Toxicol. 46: Suppl. 1, 70 pp. Ballester, A., M. Cervera, and L. Pena. 2008. Evaluation of selection strategies alternative to nptII in genetic transformation of citrus. Plant Cell Rep. 27: 1005Ð1015. Barraclough, E. I., E.P.J. Burgess, B. A. Philip, M. W. Wohlers, and L. A. Malone. 2009. Tritrophic impacts of Bt-expressing transgenic pine on the parasitoid Meteorus pulchricornis (Hymenoptera: Braconidae) via its host Pseudocoremia suavis (Lepidoptera: Geometridae). Biol. Control 49: 192Ð199. Bennett, P. M., C. T. Livesey, D. Nathwani, D. S. Reeves, J. R. Saunders, and R. Wise. 2004. An assessment of the risks associated with the use of antibiotic resistance genes in genetically modiÞed plants: report of the Working Party of the British Society for Antimicrobial Chemotherapy. J. Antimicrob. Chemother. 53: 418 Ð 431. Berry, J. A., and G. P. Walker. 2004. Meteorus pulchricornis (Wesmael) (Hymenoptera: Braconidae: Euphorinae): an exotic polyphagous parasitoid in New Zealand. N.Z.J. Zool. 31: 33Ð 44. Bland, M. J., and D. G. Altman. 1995. Multiple signiÞcance tests: the Bonferroni method. Br. Med. J. 310: 170. Carter, C., A. Lawrence, R. Lovell, and L. O’Brien. 2009. The Forestry Commission Public Forest Estate in England: social use, value and expectations. Final report, pp. 88. Social and Economic Research Group, Forest Research, Farnham, United Kingdom. Chakraborti, D., A. Sarkar, H. A. Mondal, D. Schuermann, B. Hohn, B. K. Sarmah, and S. Das. 2008. Cre/lox system to develop selectable marker free transgenic tobacco plants conferring resistance against sap sucking homopteran insect. Plant Cell Rep. 27: 1623Ð1633.

October 2011

BURGESS ET AL.: NO IMPACT OF TRANSGENIC PINE ON P. suavis

Craig, W., M. Tepfer, G. Degrassi, and D. Ripandelli. 2008. An overview of general features of risk assessments of genetically modiÞed crops. Euphytica 164: 853Ð 880. Demaneche, S., H. Sanguin, J. Pote, E. Navarro, D. Bernillon, P. Mavingui, W. Wildi, T. M. Vogel, and P. Simonet. 2008. Antibiotic-resistant. soil bacteria in transgenic plant Þelds. Proc. Natl. Acad. Sci. U.S.A. 105: 3957Ð3962. Ewald, D., J. J. Hu, and M. S. Yang. 2006. Transgenic trees in China? AFZ/Der Wald, Allgemeine Forst Zeitschrift fur Waldwirtschaft und Umweltvorsorge 61: 231-. 233: 236. Farnum, P., A. Lucier, and R. Meilan. 2007. Ecological and population genetics research imperatives for transgenic trees. Tree Genet. Genomes 3: 119 Ð133. Fenning, T. M., C. Walter, and K.M.A. Gartland. 2008. Forest biotech and climate change. Nat. Biotechnol. 26: 615Ð 617. Ferreira, S. A., K. Y. Pitz, R. Manshardt, F. Zee, M. Fitch, and D. Gonsalves. 2002. Virus coat protein transgenic papaya provides practical control of papaya ringspot virus in Hawaii. Plant Dis. 86: 101Ð105. Finstad, K., A. C. Bonfils, W. Shearer, and P. Macdonald. 2007. Trees with novel traits in Canada: regulations and related scientiÞc issues. Tree Genet. Genomes 3: 135Ð139. Fladung, M., T.M.H. Schenk, O. Polak, and D. Becker. 2010. Elimination of marker genes and targeted integration via FLP/FRT recombination system from yeast in hybrid aspen (Populus tremula L. x P. tremuloides Michx.). Tree Genet. Genomes 6: 205Ð217. Fuester, R. W., P. B. Taylor, H. Peng, and K. Swan. 1993. Laboratory biology of a uniparental strain of Meteorus pulchricornis (Hymenoptera: Braconidae), an exotic larval parasite of the gypsy moth (Lepidoptera: Lymantriidae). Ann. Entomol. Soc. Am. 86: 298 Ð304. Gleave, A. P., D. S. Mitra, S. R. Mudge, and B.A.M. Morris. 1999. Selectable marker-free transgenic plants without sexual crossing: transient expression of cre recombinase and use of a conditional lethal dominant gene. Plant Mol. Biol. 40: 223Ð235. Grace, L. J., J. A. Charity, B. Gresham, N. Kay, and C. Walter. 2005. Insect-resistant transgenic Pinus radiata. Plant Cell Rep. 24: 103Ð111. Greenberg, S. M., T. W. Sappington, B. C. Legaspi, T. X. Liu, and M. Setamou. 2001. Feeding and life history of Spodoptera exigua (Lepidoptera: Noctuidae) on different host plants. Ann. Entomol. Soc. Am. 94: 566 Ð575. Henderson, A. R., and C. Walter. 2006. Genetic engineering in conifer plantation forestry. Silvae Genet. 55: 253Ð262. Hoenicka, H., and M. Fladung. 2006. Biosafety in Populus spp. and other forest trees: from non-native species to taxa derived from traditional breeding and genetic engineering. Trees (Berl. West) 20: 131Ð144. [HSNO Act] Hazardous Substances and New Organisms Act. 1996. Public Act No 30. Ministry of the Environment, Wellington, New Zealand. (http://www.legislation.govt. nz/act/public/1996/0030/latest/viewpdf.aspx?search⫽ts_ act_hazardous_resel). James, C. 2010. Global status of commercialized Biotech/GM crops: 2010. ISAAA Brief No. 42, ISAAA, Ithaca, New York. Jelenic´ , S. 2003. Controversy associated with the common component of most transgenic plants - Kanamycin resistance marker gene. Food Technol. Biotechnol. 41: 183Ð 190. Lavery, P. B., and D. J. Mead. 1998. Pinus radiata-a narrow endemic of North America takes on the world, pp. 432Ð 449. In D. M. Richardson (ed.), Ecology and biogeogra-

1339

phy of Pinus. Cambridge University Press, Cambridge, United Kingdom. Lawes Agricultural Trust. 2007. GenStat Release 10.1. VSN International Limited, Oxford, United Kingdom. Lecardonnel, A., G. Prevost, A. Beaujean, R. S. Sangwan, and B. S. Sangwan-Norreel. 1999. Genetic transformation of potato with nptII-gus marker genes enhances foliage consumption by Colorado potato beetle larvae. Mol. Breed. 5: 441Ð 451. Lemaux, P. G. 2008. Genetically engineered plants and foods: a scientistÕs analysis of the issues (Part I). Annu. Rev. Plant Biol. 59: 771Ð 812. Lemaux, P. G. 2009. Genetically engineered plants and foods: a scientistÕs analysis of the issues (Part II). Annu. Rev. Plant Biol. 60: 511Ð559. Lin, S. Z., Z. Y. Zhang, Q. Zhang, and Y. Z. Lin. 2006. Progress in the study of molecular genetic improvements of poplar in China. J. Integr. Plant Biol. 48: 1001Ð1007. Lottmann, J., M. O’Callaghan, D. Baird, and C. Walter. 2010. Bacterial and fungal communities in the rhizosphere of Þeld-grown genetically modiÞed pine trees (Pinus radiata D.). Environ. Biosafety Res. 9: 25Ð 40. McManus, M. T. and E.P.J. Burgess. 1995. Effects of the soybean (Kunitz) trypsin inhibitor on growth and digestive proteases of larvae of Spodoptera litura. J. Insect Physiol. 41: 731Ð738. Mellerowicz, E. J., K. Horgan, A. Walden, A. Coker, and C. Walter. 1998. PRFLL - a Pinus radiata homologue of FLORICAULA and LEAFY is expressed in buds containing vegetative shoot and undifferentiated male cone primordia. Planta 206: 619 Ð 629. Miki, B., and S. McHugh. 2004. Selectable marker genes in transgenic plants: applications, alternatives and biosafety. J. Biotechnol. 107: 193Ð232. Minitab. 2006. Minitab statistical software, version 15.1.0.0. Minitab, State College, PA. Mouradov, A., T. Glassick, B. Hamdorf, L. Murphy, B. Fowler, S. Maria, and R. D. Teasdale. 1998. NEEDLY, a Pinus radiata ortholog of FLORICAULA/LEAFY genes, expressed in both reproductive and vegetative meristems. Proc. Natl. Acad. Sci. U.S.A. 95: 6537Ð 6542. Moyroud, E., G. Tichtinsky, and F. Parcy. 2009. The LEAFY ßoral regulators in angiosperms: conserved proteins with diverse roles. J. Plant Biol. 52: 177Ð185. Nap, J.-P., J. Bijvoet, and W. J. Stiekema. 1992. Biosafety of kanamycin-resistant transgenic plants. Transgenic Res. 1: 239 Ð249. R Development Core Team. 2009. R: a language and environment for statistical computing computer program, version 2.9. R Development Core Team, Vienna, Austria. Ramessar, K., A. Peremarti, S. Gomez-Galera, S. Naqvi, M. Moralejo, P. Munoz, T. Capell, and P. Christou. 2007. Biosafety and risk assessment framework for selectable marker genes in transgenic crop plants: a case of the science not supporting the politics. Transgenic Res. 16: 261Ð280. Read, D. 2000. Use of antibiotic resistance marker genes in genetically modiÞed organisms. Environmental Risk Management Authority of New Zealand, Wellington, New Zealand. Rommens, C. M., A. Conner, H. Yan, and Z. Hanley. 2011. Intragenic vectors and marker-free transformation tools for a greener biotechnology, pp. 93Ð107. In C. N. Stewart, A. Touraev, V. Citovsky, and T. TzÞra (eds.), Plant transformation technologies. Wiley-Blackwell, Ames, IA. Rosellini, D. 2011. Selectable marker genes from plants: reliability and potential. In Vitro Cell. Dev. Biol. Plant 47: 222Ð233.

1340

ENVIRONMENTAL ENTOMOLOGY

Saeed, S., A. H. Sayyed, and I. Ahmad. 2010. Effect of host plants on life-history traits of Spodoptera exigua (Lepidoptera: Noctuidae). J. Pest Sci. 83: 165Ð172. SAS Institute. 2000 –2004. SAS version 9.1.3. SAS Institute, Cary, NC. Schnitzler, F.-R., E.P.J. Burgess, A. M. Kean, B. A. Philip, E. I. Barraclough, L. A. Malone, and C. Walter. 2010. No unintended impacts of transgenic pine (Pinus radiata) trees on above ground invertebrate communities. Environ. Entomol. 39: 1359 Ð1368. Steidl, R. J., J. P. Hayes, and E. Schauber. 1997. Statistical power analysis in wildlife research. J. Wildl. Manage. 61: 270 Ð279. Strauss, S. H., H. M. Tan, W. Boerjan, and R. Sedjo. 2009. Strangled at birth? Forest biotech and the convention on biological diversity. Nat. Biotechnol. 27: 519 Ð527. Uckan, F., and E. Ergin. 2002. Effect of host diet on the immature developmental time, fecundity, sex ratio, adult longevity, and size of Apanteles galleriae (Hymenoptera: Braconidae). Environ. Entomol. 31: 168 Ð171. [USEPA] U. S. Environmental Protection Agency. 1998. Guidelines for ecological risk assessment. (EPA/630/ R095/002F. http://cfpub.epa.gov/ncea/raf/recordisplay. cfm?deid⫽12460). Walker, G. P., P. J. Cameron, F. M. MacDonald, V. V. Madhusudhan, and A. R. Wallace. 2007. Impacts of Bacillus thuringiensis toxins on parasitoids (Hymenoptera: Braconidae) of Spodoptera litura and Helicoverpa armigera (Lepidoptera: Noctuidae). Biol. Control 40: 142Ð151. Walter, C. 2004. Genetic engineering in conifer forestry: technical and social considerations. In Vitro Cell. Dev. Biol. Plant 40: 434 Ð 441.

Vol. 40, no. 5

Walter, C., M. Fladung, and W. Boerjan. 2010. The 20-year environmental safety record of GM trees. Nat. Biotechnol. 28: 656 Ð 658. Walter, C., L. J. Grace, A. Wagner, D.W.R. White, A. R. Walden, S. S. Donaldson, H. Hinton, R. C. Gardner, and D. R. Smith. 1998. Stable transformation and regeneration of transgenic plants of Pinus radiata D. Don. Plant Cell Rep. 17: 460 Ð 468. Wang, Y., B. J. Chen, Y. L. Hu, J. F. Li, and Z. P. Lin. 2005. Inducible excision of selectable marker gene from transgenic plants by the Cre/lox site-speciÞc recombination system. Transgenic Res. 14: 605Ð 614. Weigel, D., J. Alvarez, D. R. Smyth, M. F. Yanofsky, and E. M. Meyerowitz. 1992. LEAFY controls ßoral meristem identity in Arabidopsis. Cell 69: 843Ð 859. White, T.C.R. 1974. A hypothesis to explain outbreaks of looper caterpillars, with special reference to populations of Selidosema suavis in a plantation of Pinus radiata in New Zealand. Oecologia 16: 279 Ð301. Wilkinson, M., and M. Tepfer. 2009. Fitness and beyond: preparing for the arrival of GM crops with ecologically important novel characters. Environ. Biosafety Res. 8: 1Ð14. Yanisch-Perron, C., J. Vieira, and J. Messing. 1985. Improved M13 phage cloning vectors and host strains: nucleotide sequences of the M13mp18 and pUC19 vectors. Gene 33: 103Ð119.

Received 9 May 2011; accepted 19 July 2011.