Nasonia Vitripennis B.H

Home , Nasonia, Nasonia vitripennis, Pteromalidae

Blackwell Publishing Ltd Resource quality affects restlessness in the parasitoid wasp Nasonia vitripennis B.H. King & J.H. Ellison Department of Biological Sciences, Northern Illinois University, DeKalb, IL 60115-2861, USA Accepted: 20 September 2005

Key words: dispersal, host quality, locomotor activity, patch leaving, patch residence time, Hymenoptera, Pteromalidae, superparasitism

Abstract Optimal foraging theory, speciﬁcally the marginal value theorem, predicts quicker leaving (shorter residence time) from poorer patches. One proximal mechanism for achieving the leaving is that exposure to lower-quality resources may trigger increased restlessness (proportion of time in locomotion). Which aspects of host quality, if any, affect restlessness was examined in females of the parasitoid wasp Nasonia vitripennis Walker (Hymenoptera: Pteromalidae). Females were individually exposed to a single host. Restlessness was greater both during and after exposure to a host, when the host was externally damaged vs. intact. Other aspects of host quality that affected restlessness were whether the host was parasitized and whether it was dead and unsuitable for offspring development. In contrast, the host’s age and stage did not affect restlessness. Increased restlessness did not make females more willing to launch themselves across an inhospitable environment using their wings.

also affect restlessness? We address these questions, Introduction using adult females of the parasitoid wasp Nasonia vitrip- When an animal encounters a low-quality resource, it can ennis Walker (Hymenoptera: Pteromalidae) when the increase its chances of encountering a better resource by resource is a host. dispersing. For example, on a large scale, when seasonally The hosts of N. vitripennis are ﬂy pupae, which are harsh conditions approach, animals may undertake long- found in groups of one up to hundreds (Werren, 1983) in distance dispersal that takes them to better conditions far carrion, refuse, or birds’ nests (Whiting, 1967; Rueda & from their original site. On a smaller scale, animals may Axtell, 1985). Thus, a host could be within a patch or disperse between patches of resources or even between could itself be considered a patch among other patches of resource units within a patch. multiple hosts. An adult female drills through the host’s Optimal foraging theory (Stephens & Krebs, 1986) puparium to lay multiple eggs on the host pupa’s surface predicts when animals should leave resources, predicting, within the puparium. She may also host feed on the same for example, quicker leaving (shorter residence time) on host or a different host. From a distance, females are poorer patches (Charnov, 1976). Proximal mechanisms attracted by olfaction to substrates in which hosts develop, that have previously been proposed for leaving include whereas discrimination between low-quality and high- repulsion or habituation to the resource or changes in turn quality hosts appears to occur after contact with the host, rates or speed after encountering the resource (reviewed in sometimes not until after drilling into it (Wylie, 1958). The Bell, 1991; van Alphen et al., 2003). Speed, as previously types of hosts from which females produce fewer or lower- measured, depends on speed while in locomotion and quality offspring include already parasitized hosts (Wylie, depends on restlessness (proportion of time in locomo- 1965, 1966), physically damaged hosts, frozen or old hosts tion). However, greater restlessness alone can lead to (Wylie, 1963; King & Skinner, 1991; Milward de Azevedo & shorter residence time (see Appendix). The presence or Cardoso, 1996), and externally damaged hosts (BH King, absence of a resource has been shown to affect speed unpubl.). (reviewed in Bell, 1991; van Alphen et al., 2003), but does First, the effect of host quality on restlessness in the it affect restlessness and does the quality of the resource presence of the host was examined. Then, more extensive experiments examined restlessness after the host was no *Correspondence: E-mail: [email protected] longer present. Restlessness was predicted to increase in

72 King & Ellison response to (1) an already parasitized host relative to an unparasitized host, (2) an externally damaged or dead (freeze-killed) host relative to an intact live host, (3) no host relative to an externally damaged host, and (4) an older host relative to a younger host. The effect of exposure to a host larva relative to a host pupa was also examined. Larval hosts only become suitable for parasitization when they later pupate. Although this study looked primarily at walking, we also examined the effects of host quality and restlessness on a female’s subsequent willingness to disperse across an inhospitable environment using her wings.

Materials and methods General methods The N. vitripennis were from laboratory colonies, with Calliphora vomitoria (L.) (Diptera: Calliphoridae) as hosts. Experiments utilized mated adult female wasps maintained on honey. Temperature, relative humidity, and wasp age were controlled within but not between experiments.

Behaviour in the presence of a host The first damaged-host experiment examined what females (n = 18) did during exposure to a damaged host relative to what females (n = 18) did during exposure to an intact host. A damaged host’s puparium was cracked open at the base of the head but left attached. Females produce about half as many offspring from such hosts as from intact hosts (BH King, unpubl.). Each female was exposed to her host for 2 h. During this 2 h, instantaneous observations were Figure 1 Female Nasonia vitripennis behaviours seen at 15-min made every 15 min of her behaviour and whether she was interval sampling in the first damaged-host (Calliphora vomitoria) experiment, in which a female was given either (A) an on or off the host. Behaviours are listed in Figure 1; intact live host or (B) an externally damaged host. exploring the host included antennating the host or dragging her ovipositor over its surface. restlessness, defined as the proportion of time that she Restlessness after exposure to a host spent walking, hopping, or flying; however, most wasps In these experiments, each female was exposed to a host only walked, although they roamed widely in the terrarium. of a certain quality (summarized in Table 1). When first In the first long-dead host experiment, after testing, exposed to females, old hosts were covered with bristles females were weighed to the nearest 0.001 mg to determine and had black legs, wings, and antennae; all other hosts did whether female weight might mediate any effect of host not and were 2 days younger. Parasitized hosts contained quality on restlessness, e.g., via host feeding. conspecific larvae. Fresh-dead hosts had been kept in the freezer until 1 h before use to allow complete thawing Motivation for long-range dispersal after exposure to a host without decay or drying. Larval hosts pupated about 1 day The second long-dead host experiment tested whether females after being with the female parasitoid. Long-dead hosts that exhibit greater restlessness disperse more readily across had been in the freezer for 3 days and then stored at room an inhospitable environment using their wings. This experi- temperature (about 23 °C) for 1–38 days prior to use. The ment was similar to the first long-dead host experiment but female was exposed to the host for 2 h in the damaged- after being tested for restlessness, each female was placed in host, parasitized-host, and host-age experiments, and for an uncovered, small Petri dish (3.4 cm in diameter by 1.1 cm 3 h in the dead-host experiments. high) of moistened sand. This small dish was centred in a After exposure, the female was placed alone in a large medium-sized Petri dish (5.6 cm in diameter by 0.9 cm glass terrarium for 10 min, during which we measured her high), which was half filled with water, to create a moat.

Host quality and restlessness in Nasonia vitripennis 73

Table 1 Mean ± SEM (range) female Treatment n Restlessness Hops and ﬂights Nasonia vitripennis restlessness (proportion of time in locomotion) and Second damaged-host experimenta* number of hops and ﬂights during 10 min Intact host 15 0.50 ± 0.06 (0.15–0.86) 0.00 ± 0.00 (0–0) tests after exposure to a host (Calliphora Damaged host 15 0.67 ± 0.05 (0.19–0.93) 0.20 ± 0.11 (0–1) vomitoria) Third damaged-host experimentb*** Damaged host 30 0.24 ± 0.04 (0.00–0.91) 0.07 ± 0.07 (0–2) No host 30 0.72 ± 0.03 (0.29–0.92) 5.83 ± 1.42 (0–35) Parasitized-host experimentc*** Unparasitized host 30 0.47 ± 0.04 (0.11–0.98) 1.20 ± 0.46 (0–10) Parasitized host 30 0.72 ± 0.03 (0.39–0.94) 2.20 ± 0.45 (0–11) Host-age experimentd ns Old host 30 0.39 ± 0.04 (0.08–0.99) 0.13 ± 0.08 (0–2) Young host 30 0.40 ± 0.04 (0.09–0.88) 0.03 ± 0.03 (0–1) Fresh-dead host and larval host experimente ns Live host 30 0.51 ± 0.06 (0.00–0.97) 0.00 ± 0.00 (0–0) Larval host 30 0.56 ± 0.05 (0.15–0.97) 0.13 ± 0.10 (0–3) Dead host 30 0.59 ± 0.05 (0.00–0.96) 0.17 ± 0.14 (0–4) First long-dead host experimentf*** Live host 30 0.30 ± 0.03 (0.05–0.73) 0.13 ± 0.09 (0–2) Dead host 30 0.47 ± 0.03 (0.22–0.79) 0.93 ± 0.71 (0–21) Second long-dead host experimentg* Live host 20 0.42 ± 0.05 (0.01–0.85) 0.05 ± 0.05 (0–1) Dead host 18 0.61 ± 0.06 (0.23–0.93) 1.17 ± 0.79 (0–14) Comparisons of restlessness, *P<0.05, ***P<0.001, ns P>0.05. at = 2.12, d.f. = 28, P = 0.04. bU = 62.00, P<0.001, mean ranks: 43.43 vs. 17.47. ct = 5.30, d.f. = 58, P<0.001. dt = 0.07, d.f. = 58, P = 0.95. eLive vs. dead: t = 1.02, d.f. = 58, P = 0.31; live vs. larval: t = 0.70, d.f. = 58, P = 0.49. ft = 3.86, d.f. = 58, P<0.001. gt = 2.42, d.f. = 36, P = 0.02.

Females had to be highly motivated to launch (leave more often off the host (vs. on) during exposure to an contact with the inner dish) because they risked landing externally damaged host than during exposure to a live in the water. They made repeated intention movements intact host (68%, n = 162 vs. 40%, n = 162; χ2 = 26.27, (wing lifts) before launching. How long until the female d.f. = 1, P<0.001), even when females were not exploring launched was recorded (up to 20 min), as was the number the host or drilling/ovipositing (81%, n = 134 vs. 53%, of non-grooming wing lifts. n = 121; χ2 = 23.59, d.f. = 1, P<0.001). By the end of 2 h with an intact host, females were usually just standing still on the host. Results Behaviour in the presence of a host Restlessness after exposure to a host The relative occurrence of observations (across all sampling After exposure to a host with a damaged puparium, females periods and all females) in which females were walking (vs. were more restless than after exposure to an intact host, but still) was greater during exposure to an externally damaged much less restless than after having been without a host host than during exposure to a live intact host (Figure 1) (Table 1). After exposure to a parasitized host, females (47%, n = 131 vs. 32%, n = 113; χ2 = 5.48, d.f. = 1, P = were significantly more restless than after exposure to an 0.019). The percentage of females that were off the host unparasitized host (Table 1). The age of a pupal host had (vs. on) changed little (72–67%) from the start to the end no significant effect on restlessness (Table 1). of the experiment during exposure to a damaged host, but Exposure to a fresh-dead host did not increase restless- decreased steadily (78–17%) during exposure to a live ness (Table 1). However, restlessness was significantly intact host. Thus, over the entire experiment, females were greater for females that produced offspring from such hosts

74 King & Ellison

no effect of host type (β = 0.18 ± 0.59, Exp(β) = 1.20; χ2 = 0.095, d.f. = 1, P = 0.76) or previous restlessness (β = 0.002 ± 0.002, Exp(β) = 1.002; χ2 = 1.33, d.f. = 1, P = 0.25) on launching.

Discussion Optimal foraging theory’s prediction of shorter residence times in lower-quality patches (Charnov, 1976) has been particularly well studied among parasitoid wasps (e.g., Wajnberg et al., 2000; van Alphen et al., 2003; Boivin et al., 2004; Outreman et al., 2005), but it also applies to other taxonomic groups (e.g., Persons & Uetz, 1997; Nakashima & Hirose, 2003). The present study contributes to our understanding of leaving times from a mechanistic Figure 2 Restlessness (y) after exposure of Nasonia vitripennis to perspective, showing that in parasitoid wasps, leaving time hosts (Calliphora vomitoria) that had been at room temperature may be related to effects of multiple aspects of host quality 2 for a variable number of days since thawing (x) (R = 0.14, on restlessness. Restlessness of N. vitripennis females was F = 4.55, d.f. = 1, 28, P = 0.042; y = 0.0049x + 0.38). affected not only by whether the host was already parasitized, but also by whether it was dried out or damaged externally. Most previous studies of patch leaving in response to host than those that did not [mean ± SEM: 490 ± 29 s (range: quality have focussed just on whether the hosts are 369–573), n = 6 vs. 316 ± 36 s (range: 0–572), n = 23; parasitized (e.g., Wajnberg et al., 2000; Pierre, 2003). t = 2.38, d.f. = 27, P = 0.025]. Exposure to a single low-quality host was sufficient to Among females exposed to live pupal hosts, restlessness increase restlessness of an N. vitripennis female. The increased was not significantly related to whether the host gave rise restlessness persisted even when no hosts were present, but to parasitoid offspring, an adult fly, or nothing (first long- did not translate into a greater willingness to launch. dead host experiment: F2,27 = 0.51, P = 0.61; fresh-dead Restlessness increased not only after exposure to a poor host and larval-host experiment: F2,26 = 0.52, P = 0.60). host relative to a good host, but also after exposure to no After exposure to a larval host, females were not signifi- host vs. a poor host (an externally damaged host). This cantly more restless than after exposure to a pupal host seems logical as no host represents even less resource than (Table 1). However, restlessness was significantly greater a poor host. Likewise, a previous study showed that females for females whose larval hosts later developed into adult were more restless after exposure to no host vs. an un- flies than those whose larval hosts did not [mean ± SEM: parasitized host (King et al., 2000). 455 ± 39 s (range: 277–562), n = 7 vs. 306 ± 35 s (range: The hosts’ subsequent decay or loss of fluids after death 89–583), n = 22; t = 2.27, d.f. = 27, P = 0.032]. and not its death per se appeared to affect restlessness. Even Exposure to a long-dead host increased restlessness over just 3 days, the weight of a dead host declines (JH (Table 1), and restlessness increased with the number of Ellison, unpubl.). A dehydrated host is of low value days that the dead host had been at room temperature because host feeding is on fluids, as is feeding by offspring (Figure 2). Exposure to a live host vs. a long-dead host had developing within a host (Gerling & Legner, 1968). no significant effect on a female’s weight (mean ± SEM: Exposure to a larval vs. pupal host did not differentially 0.287 ± 0.017 mg vs. 0.284 ± 0.013; t = 0.14, d.f. = 58, P = affect female restlessness. Although larval hosts are 0.89). Parasitoids did not develop on the long-dead hosts. unsuitable for immediate oviposition, females wait for them to pupate (Madej, 1992). They can both host feed Motivation for long-range dispersal after exposure to a host and produce offspring from a single host if it is recently Females exposed to long-dead hosts did not wing-lift more pupated, whereas they cannot if it is older (Wylie, 1958). than females exposed to live hosts (t = 0.67, d.f. = 28, P = Female parasitoids can detect host quality using chemo- 0.50), nor were they more likely to launch (χ2 = 0.05, d.f. = 1, receptors and mechano-receptors on their ovipositors P = 0.83). Females that launched had not previously (Hawke et al., 1973; van Lenteren, 1981). Just detecting exhibited significantly more restlessness than females that host quality may be sufficient to affect restlessness. An did not launch (t = 1.41, d.f. = 36, P = 0.17). Cox’s regression alternative explanation is that females exposed to good (reviewed in van Alphen et al., 2003) likewise showed hosts are heavier as a result of host feeding, and heavier

Host quality and restlessness in Nasonia vitripennis 75 females move around less. However, this was not supported: Boivin G, Fauvergue X & Wajnberg E (2004) Optimal patch female weight was unaffected by host quality. residence time in egg parasitoids: innate vs. learned estimate of Becoming restless after exposure to poor conditions patch quality. Oecologia 138: 640–647. may be a simple mechanism for leaving among other Brzek P & Konarzewski M (2001) Effect of food shortage on the organisms besides parasitoid wasps, e.g., bugs (Caldwell, physiology and competitive abilities of sand martin (Riparia riparia) nestlings. Journal of Experimental Biology 204: 3065– 1974; Rankin & Riddiford, 1977; Saks et al., 1988), honey 3074. bees (Skalicki et al., 1988), blowflies (Ashworth & Wall, Caldwell RL (1974) A comparison of the migratory strategies of 1995), bumblebees (Keaser et al., 1996), and birds (Brzek two milkweed bugs, Oncopeltus fasciatus and Lygaeus kalmii. & Konarzewski, 2001). That restlessness contributes to dis- Experimental Analysis of Insect Behaviour (ed. by LB Browne), persal from seasonally poor sites has long been recognized pp. 304–316. Springer-Verlag, Berlin. in birds, hence the term migratory restlessness (Farner, Charnov EL (1976) Optimal foraging, the marginal value theorem. 1955). Theoretical Population Biology 9: 129–136. As noted, other aspects of movement besides restless- Farner DS (1955) The animal stimulus for migration: experi- ness can also contribute to leaving. Turn rates and speed mental and physiologic aspects. Recent Studies in Avian while moving were not measured in the present study and Biology (ed. by A Wolson), pp. 198–237. University of Illinois have not been examined previously in N. vitripennis. The Press, Champaign, Illinois. Gerling D & Legner EF (1968) Developmental history and repro- greater restlessness associated with poor hosts may not duction of Spalangia cameroni, parasite of synanthropic flies. lead to quicker leaving if turn rate is greater and speed is Annals of the Entomological Society of America 61: 1436– slower for poor than for good hosts. However, such an 1443. effect of resource quality on speed and turn rate is unlikely, Hassell MP & Southwood TRE (1978) Foraging strategies of given that the opposite effect is seen in studies of other insects. Annual Review of Ecology and Systematics 9: 75–98. insects (references in Hassell & Southwood, 1978; Hunter, Hawke SD, Farley RD & Greany PD (1973) The fine structure of 1978; Bell, 1991). sense organs in the ovipositor of the parasitic wasp, Orgilus Future experiments might address what period of time lepidus Muesebeck. Tissue and Cell 5: 171–184. or type of subsequent host encounters it takes to return Hunter KW (1978) Searching behavior of Hippodamia convergens restlessness to pre-encounter levels. Future experiments larvae (Coccinellidae: Coleoptera). Psyche 85: 249–253. might also examine what conditions are necessary to Keaser T, Shmida A & Motro U (1996) Innate movement rules in foraging bees – flight distances are affected by recent rewards motivate longer-distance dispersal, e.g., do females have and are correlated with choice of flower type. Behavioural to reach a threshold of restlessness and does it require Ecology and Sociobiology 39: 381–388. encounters with a certain number of poor hosts or a King BH, Grimm KE & Reno H (2000) Effects of mating on certain period of time without successful oviposition or female locomotor activity in the parasitoid wasp Nasonia host feeding? vitripennis (Hymenoptera: Pteromalidae). Environmental Entomology 29: 927–933. King BH & Skinner SW (1991) Proximal mechanisms of the sex Acknowledgements ratio and clutch size responses of the parasitoid wasp Nasonia Thanks to C. Cowan, J. D’Souza, and D. Thomas for vitripennis to parasitized hosts. Animal Behaviour 42: 23– laboratory assistance and R. King for comments on this 32. manuscript. This research was supported by Northern van Lenteren JC (1981) Host discrimination by parasitoids. Semiochemicals: their Role in Pest Control (ed. by DA Illinois University’s Department of Biological Sciences Nordlund, RL Jones & WJ Lewis), pp. 153–179. John Wiley and complies with current United States laws on animal and Sons, NY, USA. protection and welfare. Madej RF (1992) Assessing local mate competition in the parasitoid wasp Nasonia vitripennis. MS Thesis, Indiana University, References Bloomington, Indiana. Milward de Azevedo EMV & Cardoso D (1996) Frozen pupae of van Alphen JJM, Bernstein C & Driessen G (2003) Information Chrysomya megacephala (Diptera, Calliphoridae) – prelimi- acquisition and time allocation in insect parasitoids. Trends in nary study as breeding substrate for Nasonia vitripennis Ecology and Evolution 18: 81–87. (Walker) (Hymenoptera, Pteromalidae). Arquivos de Biologia Ashworth JR & Wall R (1995) Effects of ovarian development and e Tecnologia 1: 89–98. protein deprivation on the activity and locomotor responses of Nakashima Y & Hirose Y (2003) Sex differences in foraging the blowfly, Lucilia sericata, to liver odour. Physiological behaviour and oviposition site preference in an insect predator, Entomology 20: 281–285. Orius sauteri. Entomologia Experimentalis et Applicata 106: Bell WJ (1991) Searching Behaviour. Chapman & Hall, London. 79–86.

76 King & Ellison

Outreman Y, Le Ralec A, Wajnberg E & Pierre JS (2005) Effects Whiting AR (1967) The biology of the parasitic wasp Mormo- of within- and among-patch experiences on the patch-leaving niella vitripennis (= Nasonia brevicornis) (Walker). Quarterly decision rules in an insect parasitoid. Behavioral Ecology and Review of Biology 42: 333–406. Sociobiology 58: 208–217. Wylie HG (1958) Factors that affect host finding by Nasonia Persons MH & Uetz GW (1997) Foraging patch residence time vitripennis (Walker) (Hymenoptera: Pteromalidae). Canadian decisions in wolf spiders – is perceiving prey as important as Entomologist 90: 597–608. eating prey? Ecoscience 4: 1–5. Wylie HG (1963) Some effects of host age on parasitism by Pierre JS (2003) Patch leaving decision rules in parasitoids: do Nasonia vitripennis (Walk.) (Hymenoptera: Pteromalidae). they use sequential decisional sampling? Behavioral Ecology Canadian Entomologist 95: 881–886. and Sociobiology 54: 147–155. Wylie HG (1965) Effects of superparasitism on Nasonia vitripennis Rankin MA & Riddiford LM (1977) Hormonal control of (Walk.) (Hymenoptera: Pteromalidae). Canadian Entomolo- migratory flight in Oncopeltus fasciatus: the effects of the gist 97: 326–331. corpus cardiacum, corpus allatum, and starvation on migration Wylie HG (1966) Some effects of female parasite size on repro- and reproduction. General and Comparative Endocrinology duction of Nasonia vitripennis (Walk.) (Hymenoptera: 33: 309–321. Pteromalidae). Canadian Entomologist 98: 196–198. Rueda LM & Axtell RC (1985) Guide to common species of pupal parasites (Hymenoptera: Pteromalidae) of the house fly and other muscoid flies associated with poultry and livestock Appendix: Effect of restlessness on leaving time manure. Technical Bulletin 278. North Carolina Agricultural Imagine two wasps, one with a good host (g) and one with Research Service, North Carolina State University, Raleigh, NC. a poor host (p). Each wasp travels a path of distance d in Saks ME, Rankin MA & Stinner RE (1988) Sexually differentiated leaving its host and reaching the next host. On the path, flight responses of the Mexican bean beetle to larval and adult each moves for duration m at speed v and is still for dura- nutrition. Oecologia 75: 296–302. tion s. Thus, leaving time, t, is m plus s. If restlessness, r, is Skalicki N, Heran H & Crailsheim K (1988) Water budget of the defined as m/t and if rp is greater than rg, then leaving time honeybee during rest and flight. Biona report 6. The Flying is shorter for the wasp that is with the poor host, i.e., t < t : Honeybee – Aspects of Energetics (ed. by W Nachtigall), p g pp. 103–118. Gustav Fischer, Stuttgart, Germany, and NY, USA. Given v = v , then d /m = d /m and since d = d , then Stephens DW & Krebs JR (1986) Foraging Theory. Princeton g p g g p p g p University Press, Princeton, New Jersey. mg = mp Wajnberg E, Fauvergue X & Pons O (2000) Patch leaving decision If rp > rg, then mp/tp > mg/tg and since mg = mp, then rules and the Marginal Value Theorem: an experimental 1/tp > 1/tg

analysis and a simulation model. Behavioural Ecology 11: 577– Thus tp < tg 586. Werren JH (1983) Sex ratio evolution under local mate competi- This model addresses leaving an individual host, which in tion in a parasitic wasp. Evolution 37: 116–124. turn can inﬂuence patch leaving.