Biological Control 25 (2002) 92–98 www.academicpress.com
Systemic component to intrastadial developmental resistance in Lymantria dispar to its baculovirus Kelli Hoover,* Michael J. Grove, and Shengzhong Su Department of Entomology, The Pennsylvania State University, University Park, PA 16802, USA Received 19 December 2001; accepted 28 February 2002
Abstract We investigated intrastadial developmental resistance of the gypsy moth, Lymantria dispar, to its host-specific baculovirus, L. dispar multinucleocapsid nucleopolyhedrovirus (LdMNPV). Susceptibility of the gypsy moth to LdMNPV decreased markedly as the insect aged within the fourth instar and this resistance was systemic because it could not be overcome by bypassing the midgut and injecting the virus directly into the hemocoel. An LD88 dose of polyhedra delivered orally to newly molted fourth instars produced 74, 37, 29, 27, 38, and 60% mortalities in larvae that were orally inoculated at 12, 24, 48, 72, 96, or 120 h post-molt to the fourth instar, respectively. An LD77 dose of budded virus delivered intrahemocoelically to newly molted fourth instars produced 84, 54, 29, 48, 59, and 46% mortalities in larvae that were injected at 12, 24, 48, 72, 96, or 120 h post-molt, respectively. Developmental resistance was also observed in fourth instars fed on oak foliage, no matter whether they were inoculated orally or intrahemocoelically, suggesting that intrastadial developmental resistance in gypsy moth larvae has a systemic component. Also, the host plant did not appear to affect systemic resistance, although it did increase midgut-based resistance. The degree of developmental resistance by intrahemocoelic inoculation was equivalent in oak- and diet-fed insects. In contrast, the degree of resistance by oral inoculation was much greater in oak- than in diet-fed insects, probably because of the combined effects of host-plant inhibition and increasing resistance to viral disease as the larvae aged within the instar. Ó 2002 Elsevier Science (USA). All rights reserved. Keywords: Baculovirus; Nucleopolyhedrovirus; Developmental resistance; Gypsy moth; Lymantria dispar
1. Introduction The gypsy moth, Lymantria dispar (L.) (Lepidoptera: Lymantriidae), is well established as a serious defoliator of forest and shade trees over much of the Northeastern USA (Doane and McManus, 1981) and continues to expand its range. For example, gypsy moth populations are moving into Wisconsin, Indiana, Illinois, and southern West Virginia (http://www.ento.vt.edu/STS/). Among the naturally occurring control agents for gypsy moth is the baculovirus L. dispar multinucleocapsid nucleopolyhedrovirus (LdMNPV). This virus has no effect on non-target species, is a natural component of the gypsy moth ecosystem, and is a sound environmental and ecological pest management tool because of
*
Corresponding author. Fax: +814-865-3048. E-mail address:
[email protected] (K. Hoover).
its safety and compatibility with other forms of pest control (Doane and McManus, 1981). Outbreaks of gypsy moth are frequently terminated by naturally occurring epizootics of LdMNPV (Campbell, 1963; Doane, 1970). Although this virus has been used successfully to reduce gypsy moth populations (Doane, 1970), viral efficacy continues to be highly variable, in part, because of low virulence. For example, in third instars under the same bioassay conditions, the lethal dose that kills 50% (LD50 ) of LdMNPV in its prototype host, L. dispar, is 15 higher (Keating et al., 1989) than the LD50 of Autographa californica MNPV in one of its most susceptible hosts, Trichoplusia ni H€ ubner (Huber and Hughes, 1984). A potential contributor to low virulence of LdMNPV in gypsy moth may be intrastadial developmental resistance. As insects develop from molt to molt, they become increasingly resistant to infection by many types of pathogens, including baculoviruses (Kirkpatrick et al., 1998; Teakle et al., 1986), but only a few studies of
1049-9644/02/$ - see front matter Ó 2002 Elsevier Science (USA). All rights reserved. PII: S 1 0 4 9 - 9 6 4 4 ( 0 2 ) 0 0 0 4 1 - 5
K. Hoover et al. / Biological Control 25 (2002) 92–98
variation in susceptibility to baculoviruses within larval instars have been reported (David, 1978; Engelhard and Volkman, 1995; Kirkpatrick et al., 1998; Merritt, 1977). Herein, we examined intrastadial developmental resistance to LdMNPV in fourth instar gypsy moth.
2. Materials and methods 2.1. Insects Egg masses of L. dispar were obtained from the USDA-Forest Service Insectary at Otis Air National Guard Base, MA and stored at 4–5 °C until needed. Larvae hatched from these egg masses were reared to the fourth instar for all experiments. In an attempt to maximize variability within, but not among, treatment groups with respect to maternal effects on egg size within an egg mass, genetic variability among egg masses, and chronological age at which larvae reached the fourth instar, we used the following rearing scheme. Eight to ten egg masses were disinfected by soaking in 18.5% formaldehyde for 1 h, rinsing under running distilled water for 1 h, and drying in a biological containment hood for 30 min. Disinfected egg masses were divided into quarters by cutting them with scissors, once along their long axis and again along their short axis. This yielded four egg mass pieces that were placed individually into four sterile polystyrene petri dishes, giving a total of 2–2:5 egg mass equivalents per dish. A small piece of artificial diet (Southland Products, Lake Village, AR) was also placed in each dish. Dishes were sealed with parafilm, which was perforated with a #0 insect pin. Individual dishes were placed in a 25 °C growth chamber on four successive days. Neonates were transferred into 240 ml paper cups containing artificial diet with plastic lids (Sweetheart, Chicago, IL) at a density of 40–50 larvae/ cup and reared in a growth chamber at 25 °C, 40% RH, 16L:8D cycle. When third instar larvae became premolts to the fourth instar (identified by head capsule slippage), they were removed from the growth chamber and moved into fresh, empty paper cups in a 4–5 °C refrigerator. Larvae were held for up to 4 days at 4 °C, prior to molting. We have observed that chilling premolt third instar gypsy moths up to 4 days does not affect their ability to molt or their permissiveness to viral infection (data not shown). Molting was initiated in chilled larvae by moving them to a 28 °C growth chamber. Newly molted larvae were placed individually in clear, plastic 30 ml cups with plastic lids (Comet Products, Chelmsford, MA) containing a 2 2 3 cm piece of artificial diet. Plastic cups were labeled with the time at which the larvae were to be injected. Larvae were designated for injection at various times post-molt (e.g., 0, 48, or 120 h 15–30 min post-molt, which we define as
93
40 ; 448 , or 4120 , respectively) (Engelhard and Volkman, 1995). Actual times post-molt used in each experiment are described below. 2.2. Viral inocula Polyhedral occlusion bodies (OBs) from the A21 (Slavicek et al., 1995) isolate of the LdMNPV were amplified in gypsy moth larvae and provided to us as a stock suspension in distilled water by Dr. Suzanne Thiem (Michigan State University). Polyhedra were maintained as a stock suspension in water at 4 °C for 2 year without loss of potency. For bioassays, aliquots of stock polyhedra were diluted under aseptic conditions with filter sterilized glycerol (EM Sciences, Gibbstown, NJ) and water (3:2 v/v) to maintain them in suspension as described in Washburn et al. (1995). For experiments using diet-fed insects, we initially assayed five dilutions of polyhedra, each against 30 insects, to identify a dilution that would cause approximately 80% mortality in 40 larvae (observed mortalities were 88 2:5% across treatments). The standard LD88 dilution of stock polyhedra was 1:10,000. Fresh dilutions were prepared for each bioassay. Hemocytometer counts of diluted polyhedral suspensions ranged from 250 to 400 OBs/ll among trials. For experiments designed to examine developmental resistance in insects fed on oak foliage, a range of viral dosages were used (dosage–mortality experiments) because we had no information on dosages that would produce an approximate LD80 using microinjection of larvae fed on foliage. Budded virus (BV) of LdMNPV A21 was also obtained from Dr. Thiem’s laboratory as the filtered supernatant of a sixth passage through Ld652Y cells (Goodwin et al., 1978) cultured in TC-100 medium plus 10% fetal bovine serum (FBS) (Gibco-BRL Life Technologies, Grand Island, NY). This viral preparation was also assayed against diet-fed 40 L. dispar larvae to estimate an approximate LD80 dilution in 40 larvae (observed mortality was 76:7 4:2% across treatments). However, because virus titer began to drop after 1 year of storage at 4 °C, as determined by an insect bioassay, additional BV was generated in our laboratory by a single passage of the stock virus through LdEItA (Lynn et al., 1988) or Ld652Y cells grown in TC-100 medium plus 10% FBS. BV was harvested by collecting the supernatant from infected cells that had been centrifuged at 100g for 5 min. The supernatant was serially filtered through 0.45 and 0.2 lm syringe filters and stored at 4 °C. We assayed six dilutions of BV, each against 30 insects, to obtain a dilution that would produce an approximate LD80 in 40 larvae. After 6 months of storage at 4 °C, BV stock dilutions were bioassayed again and a new dilution identified that produced equivalent mortality in 40 larvae. These LD77 dilutions were equivalent to 0.027 plaque forming units (PFU)/ll for virus from
94
K. Hoover et al. / Biological Control 25 (2002) 92–98
LdEItA cells and 0.028 PFU/ll from Ld652Y cells, as estimated by end point dilution of the BV against Ld652Y cells using the method of Reed and Muench as described in O’Reilly et al. (1994).
tions among experiments. Thus over time, each treatment group was replicated 3–12 times.
2.3. Bioassays
Intrastadial developmental resistance was evaluated over a range of viral doses in insects fed oak foliage, their preferred host plant. For these experiments, larvae were reared through the second instar on Southland Products diet. Premolt third instars were placed in 270 ml cups with field-collected pin oak (Quercus palustris Meunchh) foliage. Foliage was obtained from the same two trees for all experiments and disinfected by immersion in 2% bleach for 2 min followed by two 2 min rinses in deionized water. Larvae were fed foliage until they became premolt fourth instars (approximately 5 days). A series of eight stock dilutions of polyhedra containing 0–10,860 OBs/ll were prepared in 60% glycerol and stored at 4 °C. Developmentally staged cohorts at 40 ; 416 ; 448 , and 472 stages were each bioassayed against six of the eight doses, with the least resistant stages receiving the lower doses and the most resistant stages receiving the higher doses. Because oak foliage has a profound host-plant effect on mortality by LdMNPV in gypsy moths (Keating and Yendol, 1987), we used the 416 stage instead of the 412 stage (as had been used for diet-fed insects described above) to minimize compounding of host-plant and developmental resistance effects. We observed that gypsy moth larvae placed on oak foliage immediately after molting did not feed until approximately 15 h post-molt. Therefore, we did not inject them until the 416 stage to ensure that the gut was filled with foliage. Thirty larvae were used per viral dose from each cohort, and the same number was dosed with 60% glycerol as controls from each cohort. The same number of diet-fed insects was compared concurrently to foliage-fed insects. At each time point post-molt, 30 diet-fed larvae were treated with each viral dose and the same number of control larvae was treated with glycerol only. Following injection, larvae were placed in 110 mm petri dishes with a piece of damp filter paper and an oak leaf. The filter paper was moistened and leaves were replaced as needed. Larvae were observed for 21 days and scored daily for mortality. These experiments were repeated twice. Dose–mortality data were subjected to probit analysis (Statistica, StatSoft, Tulsa, OK) to estimate the LD50 and LD75 for each treatment. Significant differences among treatments were determined by comparing confidence intervals at the 0.05 level. To determine if developmental resistance in foliagefed insects was gut-based or had a systemic component, cohorts of larvae at 40 ; 416 ; 448 , and 472 stages were injected intrahemocoelically with a range of doses from 0.0027 to 2.7 PFU/larva. Thirty larvae were treated with
Larvae that had been developmentally staged as described above were injected with LdMNPV, either orally or intrahemocoelically, using a Pax-100 microapplicator (Burkhard Scientific, Uxbridge, UK) and a sterile 1 cm3 tuberculin syringe with a 32 gauge steel needle (Popper and Sons, NY) (Engelhard and Volkman, 1995). For oral injections, a blunt tipped needle was positioned between the mandibles of a larva and gently inserted through the foregut into the anterior midgut to deliver 1 ll droplet of OBs suspended in 60% glycerol. For intrahemocoelic injections, a sharp needle was inserted through the abdominal body wall at the junction of either the second or third proleg to deliver 1 ll BV in TC100 plus 10% FBS. Control larvae were injected with the same volume of carrier solution without virus. For all experiments following injection, each larva was returned to its respective diet and observed daily over the next 21 days for mortality, molting, and pupation. For all experiments in which mortalities among times of inoculation post-molt were compared at a single viral dosage, mortalities were subjected to arcsine transformation, followed by one-way ANOVA and means separation using PLSD (Zar, 1984). The influence of hours post-molt at the time of inoculation on the probability of dying from viral infection was analyzed using logistic regression of pooled data (Collett, 1994). Data from larvae inoculated orally and intrahemocoelically were analyzed separately. Unique aspects of experiments in which the impact of developmental age on mortality by virus was investigated in diet-fed or foliage-fed insects are described below under their respective subheadings. 2.4. Developmental resistance in diet-fed insects Using a single dosage of polyhedra or BV of LdMNPV, we evaluated susceptibility of gypsy moths as they aged within the fourth stadium by oral and intrahemocoelic inoculations, respectively. For both oral and intrahemocoelic injections, 40 larvae were injected with virus and 20 with control solutions at 0, 12, 24, 48, 72, 96, or 120 h post-molt (designated as 40 ; 412 ; 424 ; 448 ; 472 ; 496 ; or 4120 ). Because it was not possible to inject larvae from all time points in a given day, cohorts from 3 to 4 time points (controls and virus-treated) were injected in a single experiment. For subsequent experiments, cohorts of larvae staged for 3–4 additional time points were injected, always including a time point that had been characterized previously to detect any devia-
2.5. Developmental resistance in foliage-fed insects
K. Hoover et al. / Biological Control 25 (2002) 92–98
95
each dose at each time point and 30 larvae were treated with TC100 + 10% FBS only as controls. These experiments were repeated twice. 2.6. Influence of body weight on susceptibility to LdMNPV To determine if the increasing body weight of gypsy moth larvae influenced resistance to LdMNPV in fourth instars, a range of larval sizes at the 448 stage were weighed before being virally inoculated as described above. A cohort of 30 larvae was inoculated orally with 727 OBs/larva and another cohort intrahemocoelically with 0.027 PFU/larva, as described above. Mortality was recorded daily. These experiments were repeated twice. Data were analyzed by logistic regression (Collett, 1994) to determine if body weight influenced mortality by virus (Statistica, StatSoft).
3. Results 3.1. Developmental resistance in diet-fed larvae Resistance to mortal infection by oral inoculation of LdMNPV of fourth instar gypsy moths as they aged within the instar was striking. Logistic regression showed that the later in the instar that the larvae were inoculated, the less likely they were to die from viral infection (v2 ¼ 175, df ¼ 1, P < 0:0001, Fig. 1). Mortality dropped from 87:6 8:5% in 40 larvae to 29:4 9:5 and 27:5 6:7% in 448 and 472 larvae, respectively, a decrease of 66 and 69%, respectively (Fig. 1). It was not until the end of the instar that larvae (4120 ) regained significant susceptibility to the virus (59:8 8:4% mortality), but 4120 larvae were still less susceptible than 40 larvae. Interestingly, developmental resistance was not exclusively midgut-based; there was a systemic component. Logistic regression showed that mortality of fourth instars by intrahemocoelic inoculation was a less likely outcome as the insects aged within the instar (v2 ¼ 29.4, df ¼ 1, P < 0:0001, Fig. 1). Following intrahemocoelic inoculation, mortality dropped from 76:7 4:2% in 40 larvae to a low of 29:3 5:3% in 448 larvae, a decrease of 62% (Fig. 1). Mortality of 496 larvae (59:0 3:3%) was similar to that of 40 larvae, but preceded by 24 h the return to greater susceptibility of orally inoculated larvae. In larvae inoculated intrahemocoelically, however, mortality dropped again in 4120 larvae, by 40% (45:7 5:7%), to a level significantly below 40 larvae. Mortality of 412 larvae was significantly lower than 40 larvae by oral but not by intrahemocoelic inoculation. Developmental resistance was not related to larval weight at the time of inoculation, regardless of the in-
Fig. 1. Mean percentage mortality of Lymantria dispar fed on artificial diet and inoculated with LdMNPV at a single dose either orally (open circles) or intrahemocoelically (IH, closed circles) at different times post-molt to the fourth instar. Error bars represent the standard error of the mean. One-way ANOVA was used to compare percentage mortalities at different times post-molt across replicates. For oral inoculation: F ¼ 49.9, df ¼ 6, 37, P < 0:0001; for IH inoculation: F ¼ 10.1, df ¼ 6, 26, P < 0:0001. There were no significant differences among replicates (oral: F ¼ 0.74, df ¼ 12, 31, P ¼ 0:7054; IH: F ¼ 1.55, df ¼ 6, 26, P ¼ 0:2021). Symbols labeled with the same letter indicate mortalities that are not significantly different at the 0.05 level (oral: a– d, IH: w–z). There were 30–40 larvae in each treatment for each replicate and treatments were replicated 3–12 times depending on time point. Dosages were 325 42:1 OBs/larva and 0.027 PFU/larva for oral and intrahemocoelic inoculations, respectively. Best line fits were computer generated to illustrate patterns of resistance, not to imply mechanisms.
oculation method. At the time of inoculation, weights of orally and intrahemocoelically injected 448 larvae ranged from 71.7 to 237.2 and 99.5 to 237.2 mg, respectively, with means of 145:3 5:9 and 123:9 6:3 mg, respectively. Despite this variability in weights, there was no relationship between larval weight and mortality, regardless of the method of inoculation (test that all slopes are zero for oral inoculation: G ¼ 0.034, df ¼ 1, P ¼ 0:854, mortality ¼ 48.6%; test that all slopes are zero for intrahemocoelic inoculation: G ¼ 0.034, df ¼ 1, P ¼ 0:854, mortality ¼ 51.4%). 3.2. Developmental resistance in foliage-fed larvae Developmental resistance in orally inoculated insects was even more evident in fourth instars fed on oak foliage than those fed on artificial diet (Table 1). In foliage-fed insects, the LD50 s of 448 cohorts and 472 cohorts orally inoculated with LdMNPV were 47 and 191 higher than 40 larvae, respectively (Table 1).
96
K. Hoover et al. / Biological Control 25 (2002) 92–98
Table 1 The dose ( SEM) required to kill 50% (LD50 ) of gypsy moth fourth instars fed on diet or oak foliage and inoculated orally at various times post-molt with LdMNPV Hours post-molt
0
48
72
Oak foliage Diet
109 39:0 a 74:3 17:8 a
5168 52:4 d 1805 19:5 c
20; 797 515 e 744 17:4 b
Data were subjected to probit analysis with hours post-molt at the time of inoculation (HPM) and dose entered as independent variables. For diet: probability of dying ¼ )1.390–0.003(HPM) + 0.442(log10 dose); for oak: probability of dying ¼ )1.396 –0.019(HPM) + 0.491 (log10 dose). Units are OBs/larva; LD50 s followed by a different letter were significantly different at the 0.05 level.
In addition, the LD50 s of 448 and 472 larvae orally inoculated with LdMNPV were 2.9 and 28 higher, respectively, than in cohorts of the same age feeding on diet (Table 1). However, the LD50 in 40 larvae fed oak foliage was not significantly different from the LD50 of diet-fed insects. Within each experiment, control mortality across treatments was 5% or less. At a single dosage that produced 90% mortality in 40 larvae inoculated orally (4160 OBs/larva), mortality dropped precipitously in 416 ; 448 ; and 472 cohorts (by 44, 68, and 94%, respectively, Fig. 2). Intrastadial developmental resistance in foliage-fed insects that were inoculated intrahemocoelically was consistent with the level of resistance observed in diet-fed insects. At a single dosage of BV (0.027 PFU/larva), mortality of 448 larvae inoculated intrahemocoelically was 47% lower than mortality of larvae inoculated as 40 s (Fig. 2). Mortality dropped even further (74%) in 472 larvae. Larval mortality of the 416 cohorts was significantly lower in foliage-fed insects that were inoculated orally, but not in insects that were inoculated intrahemocoelically (Fig. 2). In addition, mortality of 416 larvae inoculated intrahemocoelically was statistically equivalent whether the larvae were fed on diet or oak foliage, regardless of the method of inoculation (oral: t ¼ 0.09, df ¼ 2, P ¼ 0:9396; intrahemocoelic (IH): t ¼ 1.72, df ¼ 2, P ¼ 0:2277). Time of inoculation post-molt to the fourth instar in foliage-fed insects negatively influenced mortality by virus, regardless of the method of inoculation (logistic regression: for oral inoculation v2 ¼ 75.2, df ¼ 1, P < 0:0001; for IH inoculation v2 ¼ 36.4, df ¼ 1, P < 0:0001).
4. Discussion Intrastadial developmental resistance in fourth instar gypsy moths clearly has a systemic component but does not begin to be expressed until beyond 12 h post-molt. We observed that mortality of 412 larvae was significantly lower than 40 larvae by oral but not by intrahe-
Fig. 2. Mean percentage mortality of Lymantria dispar fed on oak foliage and inoculated with LdMNPV at a single dose either orally (open circles) or intrahemocoelically (IH, closed circles) at different times post-molt to the fourth instar. Bars represent the standard error of the mean. One-way ANOVA was used to compare percentage mortalities at different times post-molt across replicates. For oral inoculation: F ¼ 113, df ¼ 3, 4, P ¼ 0:0003 ; for IH inoculation: F ¼ 31.0, df ¼ 3, 4, P ¼ 0:0032. There were no significant differences between replicates (oral: F ¼ 0.03, df ¼ 1, 6, P ¼ 0:8761; IH: F ¼ 0.17, df ¼ 1, 6, P ¼ 0:6980). Mortalities of 416 larvae fed on diet were 50:0 6:7 and 81:5 2:1% for oral and intrahemocoelic inoculations, respectively. Symbols labeled with the same letter indicate mortalities that are not significantly different at the 0.05 level (oral: a–d, IH: x–y). There were 30–40 larvae in each treatment for each replicate and treatments were replicated twice. Dosages were 4160 OBs/larva and 0.027 PFU/larva for oral and intrahemocoelic inoculations, respectively. Best line fits were computer generated to illustrate patterns of resistance, not to imply mechanisms.
mocoelic inoculation, suggesting that developmental resistance at this stage is strictly midgut-based. In addition, intrastadial developmental resistance is not the result of a ‘‘dilution effect’’ of increasing larval size because there was no relationship between mortality by virus and larval weight. We observed a marked host-plant effect on the oral LD50 of fourth instar gypsy moths, in agreement with previous reports (Keating and Yendol, 1987; Keating et al., 1989). The combined effects of inhibition by the host plant and increasing resistance as the larvae aged within the instar markedly reduced viral efficacy in foliage-fed insects. However, the LD50 in 40 larvae fed oak foliage was not significantly different from the LD50 of diet-fed insects. This was expected because newly molted fourth instars did not begin feeding until approximately 12–14 h post-molt (our observation). Both diet-fed and foliagefed insects had empty guts when they were inoculated as 40 larvae. Thus, despite being fed as third instars on oak foliage, the host-plant effect on virally induced mortality did not appear to carry over to the next instar.
K. Hoover et al. / Biological Control 25 (2002) 92–98
Interestingly, 416 larvae fed on diet experienced equivalent mortality as 416 larvae fed on foliage, regardless of inoculation method. First, this suggests that although viral inoculum came into contact with oak foliage in orally inoculated insects, there was no host plant effect. Accordingly, the host plant effect on mortality by virus may require a period of gut conditioning. Second, because developmental resistance was not observed in orally inoculated 412 cohorts but was observed in orally inoculated 416 s (regardless of diet), it appears that developmental resistance in 416 larvae is strictly midgut-based and unrelated to the host plant effect on mortality by virus. There are only a few reports of intrastadial developmental resistance to baculovirus infection. David (1978) found that Pieris brassicae (L.) displayed differential susceptibility to orally administered granulovirus as larvae aged within the second instar; resistance was most marked at 24 h post-molt. Merritt (1977) showed that susceptibility to Spodoptera exigua NPV decreased as larvae aged within the fourth instar. More recently, developmental resistance as larvae aged within the fourth instar was reported in Heliothis virescens (F.) and T. ni treated with AcMNPV (Engelhard and Volkman, 1995; Washburn et al., 1998). Larvae of these species, differing in age by no more than a few hours, became increasingly resistant to infection. Thus, our finding of intrastadial developmental resistance is not unique to gypsy moth treated with LdMNPV. What is novel about the gypsy moth is that the basis of this resistance is, at least in part, systemic, and occurs before the penultimate instar. Other reports of developmental resistance of lepidopteran larvae to baculoviruses observed a systemic component only in the penultimate instar but not in earlier instars (Kirkpatrick et al., 1998; Teakle et al., 1986). Recent reports of midgut-based developmental resistance within an instar suggest that the mechanism may involve an increased sloughing rate of primary-infected midgut cells; the cells appear to be jettisoned before the virus has an opportunity to spread systemically (Hoover et al., 2000; Kirkpatrick et al., 1998; Washburn et al., 1995; Washburn et al., 1998). This form of resistance is not systemic because it can be overcome by artificially delivering virus directly into the hemocoel. In the gypsy moth-LdMNPV system, however, developmental resistance cannot be overcome by intrahemocoelic viral inoculation, and thus, has a systemic component. In addition, this phenomenon is biologically relevant because it occurs not only for diet-fed insects, but also occurs in insects fed on their natural host, oak. Intrastadial developmental resistance in the gypsy moth to its host-specific baculovirus, LdMNPV, may be due to a systemic immune response. In at least two studies, a portion of gypsy moth larvae having confirmed infections with LdMNPV recovered from the
97
infection (Bakhvalov et al., 1982; Yu et al., 1992); these reports suggest that gypsy moths possess a mechanism for clearing virus from the hemolymph. Thus, it is possible that systemic developmental resistance in gypsy moths is due to an active immune response to viral infection. Alternatively, intrastadial resistance to baculoviral disease may be a result of host tissues becoming refractory to infection. Whatever the mechanism, the fact that this resistance is related to developmental age suggests that it may be hormonally controlled. In particular, the hormones ecdysone and juvenile hormone are the key regulators of insect growth, metamorphosis, and molting and are implicated as modulators of immune function in vitro and in vivo. Susceptibility of ovarian cells of the silkworm Bombyx mori (L.) to BmNPV was inhibited by adding b-ecdysone to culture medium (Su et al., 1989). In vivo, fatal infection by BmNPV in silkworm larvae was suppressed by b-ecdysone (Hou and Yang, 1990). Also in the silkworm, hemopoiesis, which normally occurs during the cellular immune response to bacteria, was induced by Juvenile Hormone III or 20-hydroxyecdysone; these stimulated the development of granulocytes and oenocytoids (Han et al., 1995). In particular, we suspect that high ecdysone titers reduce susceptibility of gypsy moth to LdMNPV by mediating immune responses to viral infection. Baculoviruses encode a number of proteins giving them control over a host’s endocrine system (O’Reilly, 1995). For example, the egt gene of LdMNPV encodes UDPglucosyltransferase, which effectively interferes with larval development by altering a host’s hormonal milieu (Park et al., 1993). Expression of this virally encoded enzyme leads to a marked reduction in ecdysone levels in infected gypsy moth larvae, but not in fourth and fifth instars infected late in the given instar (Park et al., 1993, 1996). It is believed that egt expression is not sufficient at these times to offset increased production of ecdysone by the insect’s prothoracic gland. By blocking host development, the egt gene of LdMNPV increases yield of progeny virus compared to egt-deletion mutants by one order of magnitude (Slavicek et al., 1999). Levels of polyhedra synthesis decrease substantially in fourth and fifth instars. In addition, the later in the instar the infective dose is received, the lower the efficiency with which the virus blocks host development. In future experiments, we plan to explore the mechanism(s) of developmental resistance in gypsy moth and ask if these mechanisms involve hormonal immunoregulation, especially by ecdysone. Further research into the basis of this resistance may lead to methods to improve the field efficacy of LdMNPV, which would greatly increase its cost-effectiveness as a pest management tool for gypsy moth control. In addition, pursuit of the role of hormones in modulating systemic, intrastadial developmental resis-
98
K. Hoover et al. / Biological Control 25 (2002) 92–98
tance in the gypsy moth to LdMNPV may lead to a better understanding of how immunity in insects is mediated by hormones.
Acknowledgments We thank Laura Behrendt, Maria Geleski, Linda Bianchi, Andrea Dreager, Christopher O’Connor, Jason Rosenzweig, Tony Pomicter, Owen Thompson, Djamila Harouka, Po-Yu Chen, Becky Pundiak, Heather Emminger, and Silvia Montero for technical assistance and Nancy Ostiguy for statistical advice. This work was supported in part by the National Science Foundation, Ecology and Evolutionary Physiology Program (Award No. IBN-0077710) and the following Penn State funding programs: the Biotechnology Seed Grant Program of the Life Sciences Consortium, Women in Science and Engineering Research Program, and the Undergraduate Research Support Program of the College of Agricultural Sciences. We thank Suzanne Thiem for participating in stimulating discussions about gypsy moth resistance to LdMNPV and Gary Felton, Glenn Holbrook, and Diana Cox-Foster for comments on previous versions of the manuscript. References Bakhvalov, S.A., Larionov, G.V., Bakhvalov, V.N., 1982. Recovery of Lymantria dispar L. (Lepidoptera: Lymantriidae) after experimental virus infection. Entomol. Obozrenie 61, 755–758. Campbell, R.W., 1963. The role of disease and desiccation in the population dynamics of the gypsy moth Porthetria dispar L. (Lepidoptera: Lymantriidae). Can. Entomol. 95, 426–434. Collett, D., 1994. Modelling Survival Data in Medical Research. Chapman & Hall, London. David, W.A.L., 1978. The granulosis virus of Pieris brassicae (L.) and its relationship with its host. Adv. Virus Res. 22, 111–161. Doane, C.C., 1970. Primary pathogens and their role in the development of an epizootic in the gypsy moth. J. Invertebr. Pathol. 15, 21–33. Doane, C.C., McManus, M.M. (Eds.), 1981. The Gypsy Moth: Research Toward Integrated Pest Management. US Department of Agriculture, Washington, DC. Engelhard, E.K., Volkman, L.E., 1995. Developmental resistance within fourth instar Trichoplusia ni orally incoulated with Autographa californica M nuclear polyhedrosis virus. Virology 209, 384–389. Goodwin, R.H., Tompkins, G.J., McGawley, P., 1978. Gypsy moth cell lines divergent in viral susceptibility. I. Culture and identification. In Vitro 14, 485–494. Han, S.S., Lee, M.H., Yun, T.Y., Kim, W.K., 1995. Haemopoiesis in in vitro haemopoietic organ culture of Bombyx mori larvae. Korean J. Entomol. 25, 281–290. Hoover, K., Washburn, J.O., Volkman, L.E., 2000. Midgut-based resistance of Heliothis virescens to baculovirus infection mediated by phytochemicals in cotton. J. Insect Physiol. 46, 999–1007.
Hou, R.F., Yang, J.C., 1990. Suppression of nuclear polyhedrosis virus infection by beta-ecdysone in the silkworm, Bombyx mori L. Bull. Inst. Acta Zool. Sinica 29, 21–28. Huber, J., Hughes, P.R., 1984. Quantitative bioassay in insect pathology. Bull. Entomol. Soc. Am. 30, 31–34. Keating, S.T., Yendol, W.G., 1987. Influence of selected host plants on gypsy moth (Lepidoptera: Lymantriidae) larval mortality caused by a baculovirus. Environ. Entomol. 16, 459–462. Keating, S.T., McCarthy, W.J., Yendol, W.G., 1989. Gypsy moth (Lymantria dispar) larval susceptibility to a baculovirus affected by selected nutrients, hydrogen ions (pH), and plant allelochemicals in artificial diets. J. Invertebr. Pathol. 54, 165–174. Kirkpatrick, B.A., Washburn, J.O., Volkman, L.E., 1998. AcMNPV pathogenesis and developmental resistance in fifth instar Heliothis virescens. J. Invertebr. Pathol. 72, 63–72. Lynn, D.E., Dougherty, E.M., McClintock, J.T., Loeb, M., 1988. Development of cell lines from various tissues of Lepidoptera. In: Mitsuhashi, J. (Ed.), Invertebrate and Fish Tissue Culture. Japan Scientific Societies Press, Tokyo, pp. 239–242. Merritt, D.L., 1977. Factors Influencing the Susceptibility of the Beet Armyworm, Spodoptera exigua (Hubner), to a Nuclear Polyhedrosis Virus. University of California, Berkeley. O’Reilly, D.R., 1995. Baculovirus-encoded ecdysteroid UDP-glucosyltransferases. Insect Biochem. Mol. Biol. 25, 541–550. O’Reilly, D.R., Miller, L.K., Luckow, V.A., 1994. Baculovirus Expression Vectors A Laboratory Manual. Oxford University Press, New York. Park, E.J., Burand, J.P., Yin, C.M., 1993. The effect of baculovirus infection on ecdysteroid titer in gypsy moth larvae (Lymantria dispar). J. Insect Physiol. 39, 791–796. Park, E.J., Yin, C.M., Burand, J.P., 1996. Baculovirus replication alters hormone-regulated host development. J. Gen. Virol. 77, 547– 554. Slavicek, J.M., Hayes-Plazolles, N., Kelly, M.E., 1995. Rapid formation of few polyhedra mutants of Lymantria dispar multinucleocapsid nuclear polyhedrosis virus during serial passage in cell culture. Biol. Contr. 5, 251–261. Slavicek, J.M., Popham, H.J.R., Riegel, C.I., 1999. Deletion of the Lymantria dispar multicapsid nucleopolyhedrovirus ecdysteroid UDP-glucosyl transferase gene enhances viral killing speed in the last instar of the gypsy moth. Biol. Contr. 16, 91–103. Su, S.Y., Hsieh, F.K., Hou, R.F., 1989. Effect of insect hormones on infection of nuclear polyhedrosis virus primary cultured cells of Bombyx mori L. Bull. Inst. Acta Zool. Sinica 28, 275–280. Teakle, R.E., Jensen, J.M., Giles, J.E., 1986. Age-related susceptibility of Heliothis punctiger to a commercial formulation of nuclear polyhedrosis virus. J. Invertebr. Pathol. 47, 82–92. Washburn, J.O., Kirkpatrick, B.A., Volkman, L.E., 1995. Comparative pathogenesis of Autographa californica M nuclear polyhedrosis virus in larvae of Trichoplusia ni and Heliothis virescens. Virology 209, 561–568. Washburn, J.O., Kirkpatrick, B.A., Haas-Stapleton, E., Volkman, L.E., 1998. Evidence that the stilbene-derived optical brightener M2R enhances Autographa californica M nucleopolyhedrovirus infection of Trichoplusia ni and Heliothis virescens by preventing sloughing of infected midgut epithelial cells. Biol. Contr. 11, 58–69. Yu, Z., Podgwaite, J.D., Wood, H.A., 1992. Genetic engineering of a Lymantria dispar nuclear polyhedrosis virus for expression of foreign genes. J. Gen. Virol. 73, 1509–1514. Zar, J.H., 1984. Biostatistical Analysis. Prentice-Hall, Englewood Cliffs, NJ.
