Update
TRENDS in Ecology and Evolution
Vol.19 No.7 July 2004
| Research Focus
Superparasitism: a non-adaptive strategy? K. Tracy Reynolds1 and Ian C.W. Hardy2 1 2
Institute of Cell, Animal and Population Biology, University of Edinburgh, Edinburgh, UK, EH9 3JT School of Biosciences, University of Nottingham, Sutton Bonington Campus, Loughborough, Leicestershire, UK, LE12 5RD
Although once thought to be detrimental, superparasitism (where a host is parasitized more than once) by solitary parasitoids is now accepted to be an adaptive strategy. However, a recent study reveals that this might not always be the case. Varaldi et al. show that the superparasitism behaviour of the wasp Leptopilina boulardi is caused by a vertically and horizontally transmitted infectious agent. A reinterpretation of the adaptive significance of superparasitism in this species might therefore be required. Many species of insects, known collectively as parasitoids, are free-living as adults but, as juveniles, develop parasitically on other ‘host’ species, killing the host in the process. Parasitoids are either gregarious, where more than one adult is produced per host, or solitary, where only a single adult is produced. For solitary parasitoids, superparasitism (where two or more eggs are laid into a single host by one or more females) results in larval competition, from which there is only one survivor. Even though it is commonly observed, superparasitism among solitary parasitoids was long regarded as non-adaptive behaviour, wasteful of both eggs and time. During the 1980 s and 1990 s, this view was overturned by the development of functional hypotheses and empirical studies showing that superparasitism could be adaptive [1 – 3]. New work by Varaldi and colleagues [4] with the solitary parasitoid wasp Leptopilina boulardi adds a new twist, suggesting that some instances of superparasitism are not adaptive after all, at least not for the female laying the eggs or, if it is, that the mechanism of its evolution is more complex than was previously thought. According to theoretical models, adaptive superparasitism is expected to be flexible, with the number of eggs laid per host varying according to the proportions of parasitized and unparasitized hosts available. For example, when few hosts are available, many of the hosts encountered by a female parasitoid will already be parasitized and/or subsequently visited by a second female. In this situation, females can gain the greatest payoff by laying eggs into all of the hosts that they encounter, regardless of the number of eggs the hosts already contain. In addition, by laying more than one egg into each host (self-superparasitism), females can increase the probability that the eventual survivor will be one of their own offspring [1]. However, Varaldi and colleagues [4] found that L. boulardi females from some natural populations Corresponding author: K. Tracy Reynolds (
[email protected]). Available online 27 March 2004 www.sciencedirect.com
never superparasitize their Drosophila larvae hosts, whereas, in other populations, individuals vary enormously in their egg-laying behaviour, depositing up to 15 eggs in a host. Furthermore, in the laboratory, superparasitism was variable between, but not within, sibmating lines derived from single females from a behaviourally variable population. Crosses between such lines revealed that the effect was strictly maternally inherited, as did crosses between two natural populations with differing superparasitism behaviours. Varaldi et al. suspected that the variation in superparasitism behaviour in L. boulardi was due to an extrachromosomal factor rather than to nuclear gene differences. To investigate the infectiousness of this putative factor, females from a superparasitizing strain also parasitized hosts containing single eggs laid by females from a non-superparasitizing strain. The emergent adult females were subsequently given hosts, and the parentage of each female was determined by their genotype. Offspring of superparasitizing-strain mothers always superparasitized, but so did 71% of the offspring of non-superparasitizing strain mothers. The tendency to superparasitize was passed on to subsequent generations. Superparasitism behaviour can therefore be both horizontally and vertically transmitted. Infectious behavior It seems that variation in the tendency to superparasitize is caused by a virus: preliminary investigations have located viral particles in the oviducts of superparasitizing but not non-superparasitizing strain females. A question that immediately arises is whether superparasitism behaviour is a purely pathological response of the wasp, or the result of selection on the virus to favour virus transmission. The ability of some parasites to alter host behaviour adaptively is well known. Two examples are the effects of the lancet fluke Dicrocoelium dendriticum on ants [5] and the protist Toxoplasma gondii on mice (reviewed in [6]). Dicrocoelium dendriticum causes ants, the intermediate host of the fluke, to climb blades of grass, where ingestion by sheep, the definitive host, is more likely. Similarly, T. gondii causes a reduction in timidity in its intermediate mouse host, making it more susceptible to predation by cats, the definitive host. Such complex and specific alterations to behaviour strongly suggest that these are examples of evolved host manipulation. In the case of L. boulardi, superparasitism results in the deposition of infected eggs into Drosophila larvae that already contain uninfected eggs, enabling horizontal
348
Update
TRENDS in Ecology and Evolution
transmission of the virus. Horizontal transmission will benefit the virus when the ultimate winner of the larval competition among the wasps is not the individual that was initially infected by the virus. When the wasp that was initially infected by the virus wins, the virus will be passed on by vertical transmission. Furthermore, the deposition of multiple infected eggs by an infected wasp female should increase the chance of the progeny of an initially infected wasp emerging and, thus, enhance vertical transmission rates of the virus. However, the deposition of too many eggs in a host could result in all of them dying, and there might be as yet undiscovered costs to the wasp in terms of life-history traits that could also limit the success of the virus. There are many questions that remain regarding this host – parasite interaction. For example, why is horizontal transmission imperfect, and how efficient is vertical transmission? Not all non-infected wasp eggs became infected when the host was superparasitized by virally infected wasps and, although the females used in the tests appeared to transmit the virus faithfully, a laboratory population generated from a single female was heterogeneous for the infection. Is this because some individuals exhibit resistance to infection, or do some females carry a lower viral load? Are other parasitoids of Drosophila infected by this virus via horizontal transfer [7]? Perhaps they are, but has the infection been overlooked because the factors that enhance parasite transmission are also predicted to increase wasp fitness? The cost or benefit to the wasp in terms of the number of surviving offspring remains to be quantified. Furthermore, Varaldi et al. [4] do not indicate whether infected females exhibit flexibility in their superparasitism behaviour. For example, do they continue to exhibit superparasitism when hosts are plentiful, or do they behave according to the predictions of adaptive superparasitism models and reduce superparasitism under these conditions? Future directions The incidence and effects of microbial infections on parasitic wasp reproductive biology have received much research attention during recent years [8 – 14]. Much of this work has focused on bacteria that enhance their transmission by altering reproduction. For example, at least two types of bacteria [Wolbachia and an undescribed bacterium in the Cytophaga – Flexibacter– Bacteriod (CFB) group [15]] can induce thelytokous parthenogenesis (where diploid females are produced) in their respective wasp hosts. Interestingly, the CFB bacterium can also alter the oviposition behaviour of its host in a way that appears to be
www.sciencedirect.com
Vol.19 No.7 July 2004
advantageous for both the wasp and the bacteria [15]. There is also much interest in the horizontal transfer of cytoplasmic microbial infections amongst and between species, particularly in parasitoids where the use of a host to rear offspring provides a potential transmission route [7,10]. Varaldi et al.’s findings are a further demonstration of the potential ability of reproductive parasites to be transmitted horizontally. They also show that behavioural manipulations might be more common than suspected. However, whether these factors prove to be beneficial for either the fitness of the wasp or the transmission of the virus remains to be seen. References 1 van Alphen, J.J.M. and Visser, M.E. (1990) Superparasitism as an adaptive strategy for insect parasitoids. Annu. Rev. Entomol. 35, 59 – 79 2 Visser, M.E. (1993) Adaptive self-superparasitism and conspecific superparasitism in the solitary parasitoid Leptopilina heterotoma (Hymenoptera: Eucolidae). Behav. Ecol. 4, 22 – 28 3 Yamada, Y.Y. and Kazuma, S. (2003) Evidence for adaptive selfsuperparasitism in the dryinid parasitoid Haplogonatopus atratus when conspecifics are present. Oikos 103, 173 – 181 4 Varaldi, J. et al. (2003) Infectious behavior in a parasitoid. Science 302, 1930 5 Carney, W.P. (1969) Behavioural and morphological changes in carpenter ants harbouring Dicrocoelid metacercariae. Am. Mid. Nat. 82, 605 – 611 6 Webster, J.P. (2001) Rats, cats, people and parasites: the impact of latent toxoplasmosis on behaviour. Microbes Infect. 3, 1037– 1045 7 Huigens, M.E. et al. (2000) Infectious parthenogenesis. Nature 405, 178 – 179 8 Stouthamer, R. et al. (1999) Wolbachia pipientis: microbial manipulator of arthropod reproduction. Annu. Rev. Microbiol. 53, 71 – 102 9 Pannebakker, B.A. et al. Genetic diversity and Wolbachia infection of the Drosophila parasitoid Leptopilina clavipes in western Europe. Mol. Ecol. (in press) 10 Vavre, F. et al. (1999) Phylogenetic evidence for horizontal transfer of Wolbachia in host – parasitoid associations. Mol. Biol. Evol. 16, 1711 – 1723 11 Fleury, F. et al. (2000) Physiological cost induced by the maternallytransmitted endosymbiont Wolbachia in the Drosophila parasitoid Leptopilina heterotoma. Parasitology 121, 493– 500 12 Hunter, M.S. (1999) The influence of parthenogenesis-inducing Wolbachia on the oviposition behaviour and sex-specific developmental requirements of autoparasitoid wasps. J. Evol. Biol. 12, 735 – 741 13 West, S.A. et al. (1998) Wolbachia in two insect host – parasitoid communities. Mol. Ecol. 7, 1457– 1465 14 Schoenmaker, A. et al. (1998) Symbiotic bacteria in parasitoid populations: coexistence of Wolbachia-infected and uninfected Trichogramma. Oikos 81, 587 – 597 15 Zchori-Fein, E. et al. (2001) A newly discovered bacterium associated with parthenogenesis and a change in host selection behavior in parasitoid wasps. Proc. Natl. Acad. Sci. U. S. A. 98, 12555 – 12560 0169-5347/$ - see front matter q 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.tree.2004.03.021
