Homoptera: Psyllidae - PubAg - USDA

HORTICULTURAL ENTOMOLOGY

Pear Transformed with a Lytic Peptide Gene for Disease Control Affects Nontarget Organism, Pear Psylla (Homoptera: Psyllidae) GARY J. PUTERKA,1 CHRIS BOCCHETTI, PHAT DANG,2 R. L. BELL,

AND

RALPH SCORZA

USDAÐARS, Appalachian Fruit Research Station, 45 Wiltshire Road, Kearneysville, WV 25430

J. Econ. Entomol. 95(4): 797Ð802 (2002)

ABSTRACT The biology and behavior of pear psylla, Cacopsylla pyricola Foerster, on a transgenic clone of ÔBartlettÕ pear, Pyrus communis L., containing a synthetic antimicrobial gene, D5C1, was compared with that of a nontransgenic parental clone to determine whether there were any nontarget effects. The gene construct also contained the marker gene nptII (aminoglycoside 3⬘-phosphotransferase II) that encodes for antibiotic resistance to identify transformed plants. The purpose of the original transformation was to enhance pear resistance to the bacterial disease Þreblight caused by Erwinia amylovora (Burr.) Winslow et al. The biology and behavior of pear psylla on a transgenic clone were compared with a nontransgenic parental pear clone in short- (ⱕ7-d) and long-term (32-d) studies. Short-term studies indicated pear psylla adults preferred to settle and oviposit, and nymphs fed more and developed slightly faster, on transgenic pear compared with nontransgenic pear. In contrast, a long-term study on psylla colony development showed considerably fewer eggs, nymphs, and adults were produced on transgenic pear. Although adults reared on transgenic pear did not have weight affected, females produced fewer eggs and nymphal hatch was signiÞcantly reduced on the transgenic pear clone. Our results suggest that pear psylla biology and behavior are initially enhanced on this transgenic pear clone. However, chronic exposure of psylla populations to transformed pear plants that express the nptII marker and lytic peptide genes had detrimental effects on pear psylla reproductive biology. Field studies would be required to determine the speciÞc effects of each gene on pear psylla biology and behavior and whether these effects would be expressed under natural conditions. The four-fold reduction in psylla population levels that resulted on this disease resistant transgenic pear line would be an added beneÞt to a pear integrated pest management (IPM) program. Overall, this study demonstrates that genetically altering plants to control one particular organism can have unintentional yet beneÞcial effects against other nontarget pest organisms in agricultural crops. KEY WORDS pear psylla, transgenic pear, lytic peptide, nontarget organisms

GENETICALLY MODIFIED PLANTS with foreign genes to impart resistance to insects and disease are gaining momentum. This approach is especially desirable for fruit trees that characteristically take decades to produce new cultivars by combining resistance genes with other desirable traits through conventional breeding techniques. Plants transformed with foreign genes to control one pest species could impact nontarget organisms. One of the few examples is the transformation of corn with the Bacillus thuringiensis Berliner (Bt) gene to control lepidopteran pests. Pollen from transformed corn was shown to potentially affect other nontarget lepidopteran species (Losey et al. 1999). Tobacco transformed with a lytic peptide gene has shown resistance to bacterial wilt caused by Pseudomonas solancearum (Jaynes et al. 1993). Lytic peptide 1

E-mail: [email protected]. U.S. Horticultural Research Laboratory, USDAÐARS, 2001 S. Rock Road, Ft. Pierce, FL 34945. 2

genes are among those genes currently being inserted into pear (Pyrus spp.) to produce resistance to a serious bacterial disease, Þreblight, caused by Erwinia amylovora (Burr.) Winslow et al. (Reynoird et al. 1999, Dang et al. 2002). SpeciÞcally, the lytic peptide gene under our investigation is a synthetic cecropin gene incorporated into a gene construct containing a signal peptide sequence that directs the secretion of the peptide gene product, lytic peptide (Dang et al. 2002). Pear psylla, Cacopsylla pyricola Foerster, is a major pest of pear in the United States that feeds primarily on phloem tissue of pear. This pest is host speciÞc to certain Pyrus spp. (Bell and Stuart 1990). Changes in pear plant chemistry, physiology, or morphology brought on by the expression of foreign genes could impact the biology and behavior of pear psylla. Ideally, cross-resistance to both disease and pear psylla may be expressed in pear transformants expressing lytic peptide genes. The purpose of this study was to determine what nontarget effects pear transformed with the lytic

798

JOURNAL OF ECONOMIC ENTOMOLOGY

peptide gene had on the reproductive biology and feeding behavior of pear psylla through a series of studies conducted under controlled laboratory conditions. Materials and Methods Pear Transformation. ÔBartlettÕ pear was previously transformed with a lytic peptide gene at the USDA Appalachian Fruit Research Station in Kearneysville, WV, (Dang et al. 2002). The synthetic lytic peptide gene used in the transformation, D5C1, was obtained from Demegen Biotechnologies (Pittsburgh, PA). The gene construct also contained the marker gene nptII (aminoglycoside 3⬘-phosphotransferase II) that encodes for antibiotic resistance to identify transformed plants. The D5C1 gene expression was controlled by the ubiquitin promoter that allows increased intercellular expression of the gene when the plant is wounded or when methyl jasmonate is applied. The promoter is also active in high levels in senescing leaves (Garbarino et al. 1995). The transformed pear clone used in this study was selected for its vigorous growth and high rate of lytic peptide expression. The gene in this line was expressed at a constant rate, therefore no inducer such as jasmonic acid or wounding was required (Dang et al. 2002). Psylla Colony. Pear psylla were obtained from the Þeld in August 1999. They were raised on Bartlett pear seedlings that were caged with a Lexan plastic cylinder (1.5 mm in diameter by 5.8 mm in height) topped with a Þne mesh polyester Noseeum netting (Recreational Equipment, Seattle, WA). Pear psylla were reared on nontransgenic pear plants for 30 d before studies began. All colonies and experiments were maintained at a temperature of 27.8 ⫾ 2⬚C and a photoperiod of 16:8 (L:D) h. Settling and Oviposition Choice Study. A choice experiment was conducted by placing shoot cuttings from each pear genotype into a caged arena. A transgenic or nontransgenic shoot cutting trimmed to one leaf and ⬇15 cm in length was put into a 20-ml glass scintillation vial and secured by ParaÞlm (American National Can, Greenwich, CT). One vial each of transgenic and nontransgenic pear shoot cuttings was placed within the same cage to allow psylla a choice between pear types. The cage was infested with six mated pairs for a 3-d period. Choice arenas were replicated Þve times in a completely randomized design. Numbers of adults and eggs present on transgenic or nontransgenic plant material were recorded 3 d after infestation. Adult numbers were recorded Þrst before the experiment disassembled so that eggs could be counted on each leaf under a stereomicroscope. No-Choice Oviposition Study. Shoot cuttings from transgenic and nontransgenic pear were prepared in the same manner as in the above-mentioned study on settling and oviposition. Clip-cages were attached to the second fully expanded leaf and infested with one mated pair of adult psylla per cage. The clip-cages were constructed of 2-cm-diameter plastic tubing attached to the arms of a hair clip. The edge of the plastic

Vol. 95, no. 4

tubing that contacted the leaf surface was covered by felt cloth to make a good seal. Foam corks were placed on top of the tubing to allow air circulation and prevent escape. The study consisted of 17 replications arranged in a completely randomized design. Egg counts were made 3, 5, and 7 d after infestation. No-Choice Nymphal Feeding Bioassay. Transgenic or nontransgenic pear cuttings were placed in a 20-ml scintillation vial secured by ParaÞlm. A 2-cm-diameter clip-cage was attached to the second fully expanded leaf of a shoot cutting and ⬇20 second-instar psylla were introduced into a cage. Six replications of each plant type were prepared in a completely randomized design. The nymphs were allowed to feed for 3 d. Afterward, the clip-cage was removed from the infested leaf and the nymphs and honeydew were rinsed from the leaf with 7 ml of water two times into a 20-ml scintillation vial that was preweighed (␮g) and labeled. The 20-ml scintillation vial was topped with Noseeum netting to Þlter out the nymphs and exuvia from the rinsate. Vials containing the honeydew solution were allowed to air-dry within a laminar airßow hood for 3 d. Vials were then weighed (␮g) to determine how much honeydew they contained. No-Choice Nymphal Development. A newly hatched nymph ⬍24 h old was placed in 2-cm-diameter clip-cage that was attached to transgenic or nontransgenic pear leaf to observe the time required to the next instar up to adulthood. The clip-cage was attached to the Þrst fully expanded terminal leaf between the leaf margin and mid-vein. The experiment had 15 replications conducted in a completely randomized design. The nymphs were observed daily for the presence of exuvia, which indicated passage to the next instar. The experiment was terminated when nymphs reached adulthood. Chronic Effects on Psylla Population. Studies on the effects of transgenic and nontransgenic pear on the various aspects of pear psylla biology were determined by establishing colonies on transgenic and nontransgenic pear plants and allowing populations to develop over a 32-d period. Four colonies were initiated on each pear genotype by obtaining 10 mated pairs of adult psylla from the stock colony reared on nontransgenic pear and caging them onto individual transgenic or nontransgenic pear seedlings. The adults were allowed to oviposit for 3 d and then removed. The population experiment was replicated three times on nontransgenic and transgenic pear. The forth replicate was initiated to supply mated pairs of adult psylla for an oviposition study. Population composition was determined 32 d after colony establishment. All eggs, nymphs, and adult stages that each colony contained were recorded. Nymphs were recorded by instar. Mature adults were sorted by sex and placed in standard petri dishes so that they could air-dry within a laminar airßow hood for 3 d. Afterward, 10 adults were placed in a preweighed 1.5-␮l microcentrifuge tube and weight (mg) was determined. The oviposition study used adult psylla reared on transgenic or nontransgenic pear 14 d after the colonies were initiated. A mating pair of psylla adults was

August 2002

PUTERKA ET AL.: NONTARGET EFFECTS OF GENETICALLY TRANSFORMED PEAR

Fig. 1. Narrowing of the leaf and reduction in leaf margin serration on transgenic Bartlett pear (left) compared with nontransgenic Bartlett (right).

placed within a 2-cm-diameter clip-cage attached to a fully expanded leaf of a transgenic or nontransgenic plant and numbers of eggs produced were recorded daily over an 8-d period. The experiment was conducted with 17 replications on transgenic and nontransgenic pear. The eggs that resulted after the termination of the oviposition study were further examined for viability. All eggs except 25 freshly deposited white eggs (7Ð 8 d old) were removed from a leaf and the clip-cages were replaced. Initial egg counts per leaf were taken on transgenic and nontransgenic plants and nymphal hatch was recorded daily until it was clear that no more nymphal hatch would occur (nonviable eggs had no embryonic development). There were 10 replications on transgenic and nontransgenic pear. Statistical Analysis. Treatment comparisons for the majority of the studies were made using pairwise ttests, ␣ ⫽ 0.05 (SAS Institute 1999). Data from nochoice nymphal development study were analyzed using analysis of variance (ANOVA) with repeated measures over time and treatment comparisons were made using a pairwise t-tests, ␣ ⫽ 0.05 (SAS Institute 1999). Results and Discussion Effects of Transformation on Pear Growth. Pear transformed with the lytic peptide gene produced some visible structural changes in pear. Leaves of the transformed clone were noticeably narrower, almost devoid of the typical serrated leaf margin, and had chlorotic mottling on the leaf surface in comparison to nontransformed Bartlett pear (Fig. 1). Leaf thickness was randomly measured by a micrometer and no signiÞcant difference (t ⫽ 1.26, df ⫽ 18, P ⫽ 0.25) was found between transgenic (0.198 ⫾ 0.008 mm) and nontransgenic (0.182 ⫾ 0.005) leaves. The structural

799

differences alone could affect pear psylla preference and oviposition in the choice test. However, studies on no-choice oviposition, nymphal feeding, and nymphal development conÞned insects to a small area just off the mid-vein of the leaf to avoid the effect of narrow leaves and lack of serrated leaf margins. Therefore, results between choice and no-choice studies could be compared with determine whether the leaf mid-vein and margin were factors affecting oviposition. Settling and Ovipositional Preference. Pear psylla exhibited a distinct preference to settle (t ⫽ 3.0, df ⫽ 6, P ⫽ 0.02) and oviposit (t ⫽ 3.47, df ⫽ 6, P ⫽ 0.01) on transgenic plant material. Approximately twice as many adults settled on leaves of transgenic pear (55.5 ⫾ 7.9) compared with nontransgenic pear (28.1 ⫾ 4.5). Psylla also oviposited substantially more on transgenic (177.8 ⫾ 9.7) versus nontransgenic (116.0 ⫾ 14.9) pear leaves. Psylla adults had a choice between a single leaf of transgenic or nontransgenic pear and lower numbers of eggs were expected on transgenic leaves because they were ⬇60% of the size of nontransgenic pear leaves, yet, the opposite resulted. Adults mainly oviposited on the upper surfaces of both transgenic and nontransgenic pear leaves. The few eggs that were oviposited on the lower surface of the leaves cannot account for the differences in oviposition. Studies have indicated that oviposition is inßuenced by both chemical and structural cues that adult psylla obtain from the surface of the leaf (Horton and Krysan 1990). Although there were no obvious differences in upper leaf surface structure between the two pear types, there may have been minute differences that could have provided more favorable ovipositional sites (Fig. 1). Another possibility is that the two pear types differed in leaf surface chemistry. No-Choice Oviposition. Nearly three times as many eggs were deposited by pear psylla adults on leaves of transgenic pear than on nontransgenic pear 3 d after infestation (16.9 ⫾ 3.1 and 6.5 ⫾ 2.0, respectively; t ⫽ 2.10, df ⫽ 32, P ⫽ 0.04). This effect continued until the 7 d after infestation (t ⫽ 2.40, df ⫽ 32, P ⫽ 0.02) where nearly three times as many eggs were oviposited on leaves of transgenic pear (63.1 ⫾ 7.8) than on nontransgenic pear leaves (20.5 ⫾ 3.7). These results support the oviposition data obtained from the aforementioned choice study that suggests preference alone cannot account for the differences we found in oviTable 1. Time required (d ⴞ SE) for pear psylla nymphs to pass from one instar to the next on transgenic and nontransgenic pear Instar 1 to 2 2 to 3 3 to 4 4 to 5 5 to adult Total time to adult

Time Nontransgenic

Transgenic

4.3 ⫾ 0.5a 4.7 ⫾ 0.5a 4.3 ⫾ 0.7a 5.1 ⫾ 0.7a 12.9 ⫾ 3.1a 25.9 ⫾ 2.8a

5.3 ⫾ 0.4a 3.2 ⫾ 0.4b 4.9 ⫾ 0.9a 4.1 ⫾ 0.7a 7.0 ⫾ 1.1a 23.5 ⫾ 1.5a

Means within a row are not signiÞcantly different if followed by the same letter (pairwise t-test, ␣ ⫽ 0.05).

800 Table 2.


Vol. 95, no. 4

Mean ⴞ SE number of eggs, nymphs, and adult psylla produced 32 d after establishing 10 females on plants

Plant type Transgenic Nontransgenic

Composition of Populations Eggs

Nymphs

Adults

Total no. insects

Leaves/ plant

1174.5 ⫾ 15.5b 4396.0 ⫾ 261.8a

277.0 ⫾ 67b 1714.7 ⫾ 195.9a

49.0 ⫾ 26.5b 162.0 ⫾ 15.5a

1501.0 ⫾ 56.0b 6272.7 ⫾ 155a

80.6 ⫾ 6.6a 68.3 ⫾ 12.6a

Means within a column are not signiÞcantly different if followed by the same letter (pairwise t-test, ␣ ⫽ 0.05).

position. Clip-cages were positioned between the midrib and leaf margin on the upper surface of both pear types to avoid major leaf structural inßuences while these structures were available in the choice study. Therefore, it is unlikely that the leaf midrib and margins inßuenced psylla oviposition in this study. These results further support the idea that there may be biochemical or minute leaf surface structural differences that account for increased oviposition on transgenic pear leaves. Regardless, transgenic pear is preferred and provides a better ovipositional substrate than the nontransgenic genotype. Nymphal Feeding Bioassay. Nymphal feeding activity was signiÞcantly increased (t ⫽ 2.17, df ⫽ 10, P ⫽ 0.05) on transgenic pear (0.2 ⫾ 0.05 mg/nymph) versus nontransgenic pear leaves (0.1 ⫾ 0.02 mg/nymph). Butt et al. (1988) reported that successful initiation of nymphal feeding and honeydew production were strong indicators of host suitability. Studies showed that honeydew production by nymphs decreased on psylla-resistant pear because of an antibiotic mechanism. Increased levels of honeydew production by nymphs that fed on transgenic pear indicated that transgenic pear is a more suitable feeding host than nontransgenic pear. Nymphal Development Time. There was a signiÞcant time effect (F ⫽ 148.59, df ⫽ 4, 28; P ⬍ 0.0001) and time by plant type interaction (F ⫽ 3.31, df ⫽ 4,

28; P ⫽ 0.016) that suggested developmental time from one instar to the next was dependent on the plant type. Comparisons of instar developmental times revealed that only the time taken for nymphs to pass from the second to the third instar signiÞcantly differed between plant types (Table 1). Total developmental time from nymph to adult was not signiÞcantly affected by transgenic pear. Evidently, the differences in nymphal feeding rates between plant types were not substantial enough to have a major effect on psylla development. Chronic Population Exposure. Rearing psylla colonies over a 32-d period produced results that distinctly contrasted with the short-term studies. Considerably fewer eggs, nymphs, and adults were produced on transgenic pear compared with nontransgenic pear even though both plant types did not signiÞcantly differ in leaf number (Table 2). Neither male nor female adult psylla weights varied signiÞcantly (t ⫽ 2.22, df ⫽ 10, P ⫽ 0.05) between transgenic (male ⫽ 2.2 ⫾ 0.5; female ⫽ 3.3 ⫾ 0.4) and nontransgenic pear (male ⫽ 2.3 ⫾ 0.1; female ⫽ 3.4 ⫾ 0.1). However, when mated pairs were taken from these colonies and allowed to oviposit, the adults reared on transgenic pear progressively laid fewer eggs over time and produced signiÞcantly fewer eggs on 9 d (t ⫽ 2. 32, df ⫽ 32, P ⫽ 0.03) than those adults reared on nontransgenic pear (Fig. 2). The eggs produced by

Fig. 2. Cumulative number of eggs (mean ⫾ SE) produced over time by adults conÞned on transgenic or nontransgenic pear leaves. Adults originated from colonies that were established for 30 d on respective transgenic and nontransgenic pear seedlings before tests were conducted.

August 2002

PUTERKA ET AL.: NONTARGET EFFECTS OF GENETICALLY TRANSFORMED PEAR

801

Fig. 3. Cumulative nymphal hatch (percent ⫾ SE) from eggs produced by adult pear psylla reared on transgenic and nontransgenic pear.

adults reared on transgenic pear were also slightly less fertile (t ⫽ 1.97, df ⫽ 32, P ⫽ 0.06) than those eggs produced by adults reared on nontransgenic pear on 16 d and these differences in fertility remained constant over time (Fig. 3). Puterka et al. (1993) found that less oviposition typically occurs on resistant pear cultivars and that reduced oviposition correlated with nymphal feeding preference. Furthermore, rearing psylla on resistant pear over a 30-d period resulted in a population age structure that was skewed toward having more early instars because nymphal development had been severely retarded (Puterka 1997). In addition, nymphal survival was greatly decreased by the antibiotic affects of resistant pear. The negative impact that transgenic pear had in the long-term population study is probably not related to plant morphological differences (Fig. 1) or host suitability because psylla oviposited and fed better on transgenic pear in short-term studies (Tables 1 and 2). The detrimental effects on pear psyllaÕs reproductive biology that occurred when the adults were reared for a longer period on transgenic pear (Figs. 2 and 3) suggest that an antibiotic factor is responsible. There are several possibilities that might explain how psylla reproduction is being affected. The lytic peptide may be interfering with protein biosynthesis linked to reproduction or the antibiotic properties of the lytic peptide may be affecting the intracellular symbiotes of pear psylla. The function(s) of intracellular symbiotes that reside in tissues associated with the psyllaÕs digestive tract (Chang and Musgrave 1970) is not well known but they have been shown to provide aphids with products needed for biosynthesis and essential nutrients needed for survival (Houk and GrifÞths 1980). In addition, De Turck et al. (2002) have shown that transgenic potato that contained only the marker genes, nptII-gus, increased Colorado potato beetle, Leptinotarsa decemlineata (Say), Þtness in short-term experiments. Our study did not have a transgenic pear line that contained only the nptII marker gene avail-

able to determine its biological or behavioral effects on pear psylla. The contrasting results from the short- and longterm studies underscore the importance of conducting both types studies to determine what affects transgenic plants have on insects. Further study on other pear lines that express the nptII marker and lytic peptide genes is needed to determine which genes are affecting pear psylla biology and whether these effects are consistent with these gene insertions. Ultimately, Þeld studies are needed on pear lines expressing these genes to conÞrm our results in the laboratory and greenhouse. The fourfold reduction in psylla population levels that resulted on this disease resistant transgenic pear line would be an added beneÞt to a pear IPM program. Overall, this research demonstrates that genetically altering plants to control one particular organism can have unintentional yet beneÞcial effects against other nontarget pest organisms in agricultural crops. Acknowledgments We thank Demegen Biotechnologies, Pittsburgh, PA, for providing the lytic peptide D5C1 gene construct used in the genetic transformation of pear.

References Cited Bell, R. L., and L. C. Stuart. 1990. Resistance in eastern European Pyrus germplasm to pear psylla nymphal feeding. HortScience 25: 789 Ð791. Butt, B. A., L. C. Stuart, and R. L. Bell. 1988. Feeding behavior of pear psylla (Homoptera: Psyllidae) nymphs on susceptible and resistant Pyrus germplasm. J. Econ. Entomol. 81: 1394 Ð1397. Chang, K. P., and A. J. Musgrave. 1970. Histochemistry and ultrastructure of the mycetome and its ÔsymbiotesÕ in the pear psylla, Psylla pyricola Foerster. Tissue Cell 1: 597Ð 608.

802


Dang, P. M., R. Scorza, and R. L. Bell. 2002. Transformation of ÔBartlettÕ pear with a lytic peptide. Biotechnology (in press). De Turck, S., P. Giordanengo, A. Cherqui, C. Ducrocq-Assaf, and B. S. Sangwan-Norreel. 2002. Transgenic potato plants expressing the nptII-gus marker genes affect survival and development of the Colorado potato beetle. Plant Sci. 162: 373Ð380. Garbarino, J., T. Oosumi, and W. Belknap. 1995. Isolation of a polyubiquitin promoter and its expression in transgenic potato plants. Plant Physiol. 109: 1371Ð1378. Horton, D. R., and J. L. Krysan. 1990. Probing and oviposition-related activity of summerform pear psylla (Homoptera: Psyllidae) on host and nonhost substrates. Environ. Entomol. 19: 1463Ð1468. Houk, E. J., and G. W. Griffiths. 1980. Intracellular symbiotes of the Homoptera. Annu. Rev. Entomol. 25: 161Ð187. Jaynes, J., P. Nagpala. L. Desefan-Beltran, J. Hong Huang, J. Kim, T. Denny, and S. Cetiner. 1993. Expression of a Cecropin B lytic peptide analog in transgenic tobacco confers enhanced resistance to bacterial wilt caused by Pseudomonas solanacearum. Plant Sci. 89: 43Ð53.

Vol. 95, no. 4

Losey, J. E., L. S. Rayor, and M. E. Carter. 1999. Transgenic pollen harms monarch larvae. Nature (Lond.) 339: 214. Puterka, G. J. 1997. InterspeciÞc variation in pear psylla (Psyllidae: Homoptera) nymphal survival and development on resistant and susceptible pear. Environ. Entomol. 26: 552Ð558. Puterka, G. J., R. L. Bell, and S. K. Jones. 1993. Ovipositional preference of pear psylla (Homoptera: Psyllidae) for resistant and susceptible pear. J. Econ. Entomol. 86: 1297Ð 1302. Reynoird, J. P., F. Mourgues, J. Norelli, H. S. Aldwinckle, M. N. Brisset, and E. Chvreau. 1999. First evidence for improved resistance to Þreblight in transgenic pear expressing the attacin E gene from Hyalophora cercopia. Plant Sci. 149: 23Ð31. SAS Institute. 1999. SAS/STAT. The SAS system for Windows, version 8.0. SAS Institute, Cary, NC. Received for publication 7 June 2001; accepted 12 February 2002.