Wreck This Journal Pages Printable

Figure 2.13 Effect of a soybean trypsin inhibitor (SKTI) on the mean mass (±SE) of surviving tomato moth larvae (Lacanobia oleracea) feeding on (a) artificial diet and (b) transgenic potato plants.

Day 9 Day 15

Day 21

Day 9 Day 15

Day 21

Figure 2.13 Effect of a soybean trypsin inhibitor (SKTI) on the mean mass (±SE) of surviving tomato moth larvae (Lacanobia oleracea) feeding on (a) artificial diet and (b) transgenic potato plants.

Note: The inhibitor was expressed at 2% of total protein in potato leaf-based diet and at 0.5% of total protein in potato plants. Reduction of growth was much more apparent for larvae feeding on artificial diet than for those on SKTI-expressing plants. Source: Reprinted from Journal of Insect Physiology, 45, Gatehouse et al., 545-558. © 1999, with permission from Elsevier.

and rapid adaptation on the part of the insect pest. Insects can compensate for the loss of activity by hyperproduction of endogenous proteinases or by upregulation of new, inhibitor-insensitive protein-ases, but both strategies are expensive in terms of amino acid utilization (Broadway and Duffey 1986; Jongsma and Bolter 1997). We might expect better adaptation in specialist herbivores, but the Colorado potato beetle, Leptinotarsa decemlineata, is only partially able to compensate for the effects of induced proteinase inhibitors in potato leaves (Bolter and Jongsma 1995).

Most proteinase inhibitors have little effect against phloem-feeding insects, whose diet is rich in free amino acids. However, the activity of lectins against homopteran pests is receiving considerable attention. Lectins are a diverse group of anti-nutrient proteins, often accumulated in plant storage tissues, which bind to carbohydrates (Peumans and Van Damme 1995). They have multiple binding sites and may bind directly to glycoproteins in the midgut epithelium, or may bind to and clog the peritrophic matrix. Snowdrop lectin, when expressed in transgenic potato crops, confers partial resistance to aphids (Down et al. 1996). Because lectins bind to the gut epithelium and enter the haemolymph, they have the potential to act as carrier proteins for delivery of insect neuropeptides as insecticides, when oral administration of the peptides alone is ineffective (Fitches et al. 2002).

Coping with plant allelochemicals Apart from behavioural avoidance, insects can deal with allelochemicals by detoxifying and excreting them. Polysubstrate monooxygenases (mixed-function oxidases) are non-specific detoxification enzymes, rapidly induced by the presence of toxins. The terminal component is cytochrome P-450, which catalyses the oxidation of toxins to produce more polar compounds that are excreted or further metabolized. Multiple cytochrome P-450 genes are typically expressed simultaneously, hence the wide range of chemically dissimilar toxins (including pesticides) on which they act. To take a familiar insect as example, the specialist tobacco feeder M. sexta absorbs ingested nicotine into the midgut cells and metabolizes it to cotinine-N-oxide, which is cleared from the haemolymph by the Malpighian tubules (Snyder et al. 1994). Rapid and reversible induction of nicotine metabolism, and the efficient active transport system in the tubules, are major adaptations of M. sexta to high levels of this active alkaloid. The reversibility of the response suggests that detoxification might be costly, but in M. sexta larvae the processing of nicotine does not impose a significant metabolic cost, nor does the processing of toxins from non-host plants, although the latter do have other adverse effects (Appel and Martin 1992).

According to coevolutionary theory, certain insect species have been successful in counteracting plant defences and those defences may then be used as unique feeding stimulants for species which specialize on the plant. Meanwhile the toxic or deterrent effect still works for other, generalist herbivores (for many fascinating examples see Harborne 1993). In such specialist feeders, allelochemicals may be sequestered for chemical protection, as in insects which sequester cardiac glycosides from milkweeds (Asclepiadaceae). Moreover, the allelochemicals become feeding attractants and oviposition stimulants, contributing to the evolution of close insect-plant associations and the enormous diversity of angiosperm feeders (Farrell 1998). Molecular phylogenies of Blepharida (Coleoptera, Chrysomelidae), many of which are monophagous, and Bursera species, which are rich in terpenes, show that host shifts have been strongly influenced by host plant chemistry (Becerra 1997). The role of chemistry in plant-animal interactions is beautifully illustrated by the cactus-microorganism-Drosophila system of the Sonoran desert (reviewed by Fogleman and Danielson 2001). Four species of Drosophila feed and reproduce in the necrotic tissue of five species of columnar cacti, each fly on a specific cactus, the specificity being due to the allelochemistry of the cacti (the presence or absence of certain sterols, alkaloids and terpenoids) and the volatile cues resulting from microbial action. Molecular ecological studies are now focusing on the evolution of the multiple cytochrome P-450 enzymes involved in these Drosophila-cactus relationships.

Host-plant specificity

There are two main benefits to feeding on a mixture of plants: selection of a suitable balance of nutrients, and dilution of allelochemicals, so that levels of particular compounds remain below critical values. Grasshoppers are highly mobile compared to other phytophagous insects and generally polyphagous, and they have exploited the fact that grasses have minimal chemical defence (Joern 1979; Harborne 1993). Increased locomotion presents more opportunities for diet mixing. However, different populations of an oligo-phagous species may be regional specialists, and individual insects may be more specialized than the population as a whole. This variation may be a function of both plant resistance and insect preference (Singer and Parmesan 1993). The latter hypotheses are not supported by Joern's (1979) study of two arid grassland communities in Texas, in which niche breadths for 12 grasshopper species common to both study sites were strikingly similar, or by close field observations of feeding in the polyphagous grasshopper Taeniopoda eques, in which single meals of individual females included up to 11 food items (Raubenheimer and Bernays 1993). The more phylogenetically derived insect orders (with more sedentary larvae) have tended towards diet specialization, which suggests greater efficiency than polyphagy, but there is not much supporting evidence for the idea that increased performance on one plant species is correlated with reduced performance on others, or that there is a physiological advantage to be gained from feeding specialization (Jaenike 1990). In a detailed comparison of 20 species of Lepidoptera larvae of varying degrees of specialization, the feeding generalists had slower consumption and growth rates, but this was mainly because they tended to be tree-feeders (Scriber and Feeny 1979). The advantages may be more ecological than physiological (as when a specialized insect becomes adapted to detoxify that plant's allelochemicals, and may even sequester and use the toxins for its own defence). For a readable account of the complexities of host plant specificity see Schoonhoven et al. (1998). It is also becoming clear that the study of tritrophic interactions may be necessary to explain host plant selection of some insect herbivores that select plants that are apparently suboptimal in nutritional quality (Singer and Stireman 2003).

2.5 Growth, development, and life history

The amount and quality of food consumed by an insect determines its performance; in the larval stage this is measured as growth rate, development time, body mass, and survival, and in the adult as fecundity, dispersal, and survival (Slansky and Scriber 1985). In the previous section we examined some of the variation in food quality experienced by insect herbivores; now we turn to variation in nutritional needs of insects during growth, development and reproduction. Much of this section concerns trade-offs between competing fitness functions, and these become apparent as a result of developmental plasticity (i.e. environmentally caused variation within a single genotype during development; see also Chapter 5). Insects provide excellent opportunities for experimentation on the nutritional basis of life history trade-offs, so it is not necessary to rely on correlations to show causality.

Nutritional factors are important in explaining the success of holometabolous development. Caterpillars have high protein requirements for rapid tissue growth, but are relatively sedentary, while cockroaches or grasshoppers need more carbohydrate to sustain higher activity levels, but their growth rates are lower (Bernays 1986b; Waldbauer and Friedman 1991). Caterpillars have double the consumption rate, double the gut capacity, and ECD values which are 50 per cent higher than do acridids. They also produce and maintain much lighter integuments: The cuticle of acridids is 10 times as heavy, and up to 50 per cent of total dry mass excluding the gut contents (Bernays 1986b).

2.5.1 Development time versus body size

Three traits central to life history theory are closely interrelated: adult size, development time, and growth rate. It is commonly accepted that there is a trade-off between short development time and large adult size (assuming constant growth rates), but an organism that grows at a high rate can achieve both (Arendt 1997; Nylin and Gotthard 1998). These negative associations between traits are exacerbated by stressful conditions, suggesting competition between different organismal demands for limited resources. Physiological studies aimed at elucidating the mechanistic basis of life history trade-offs were reviewed recently by Zera and Harshman (2001).

Extending development time increases the risk of predation, but so can high growth rates which depend on increased feeding rates. Bernays (1997) clearly demonstrated the risks that caterpillars face from parasitoids and predators when they feed. Examination of feeding behaviour throughout the fourth and fifth instars of Helicoverpa armigera caterpillars shows that exponential growth is sustained more by increased ingestion rates than increased time spent feeding, especially during the late fifth instar which is most susceptible to bird predation (Barton Browne and Raubenheimer 2003). The fitness cost of high growth rate in terms of predation risk has been tested experimentally by using photoperiod to manipulate growth rate in the wood butterfly Pararge aegeria (Nymphalidae) (Gotthard 2000). Shorter day length induced slower growth, corresponding to late summer conditions when larvae enter diapause in the pupal stage, and this was accompanied by 30 per cent lower mortality due to a generalist predator, Picromerus bidens (Heteroptera, Pentatomidae), introduced to the cages. In seasonal environments development time is complicated by diapause, which only occurs in certain stages—this situation favours genotypes capable of plasticity in growth rate (Nylin and Gotthard 1998). Butterflies such as P. aegeria can either speed up their development to produce an additional generation before winter, or slow down and enter diapause. Gotthard (2000) distinguishes between the instantaneous mortality risk of the fast-growing caterpillars in his experiment, and their total mortality risk during the larval stage, which might actually be lower because of the shorter development time. Flexible growth strategies have recently been investigated in an alpine beetle, Oreina elongata (Chrysomelidae), and late season light conditions led to an increase in growth rate and shorter development time, but no change in prepupal weight (Margraf et al. 2003). The authors argue that if 'catch-up growth' occurs in this species, in which a short and unpredictable growing season might be expected to select for rapid growth, then it may be common in temperate insects.

Classic examples of the trade-off between body size and development time are seen in male butterflies which emerge first (protandry) and are consequently smaller than females (Lederhouse et al. 1982). However, the assumption that both sexes are growing at the same rate is not always true. Males of Pieris napi which develop directly instead of entering diapause are under severe time constraints and respond to selection for large size and protandry by increasing their growth rate (Wiklund et al. 1991).

Prolonged development can be viewed as a means of increasing food consumption on suboptimal foods (Slansky 1993). If M. sexta are exposed to low dietary protein levels as early instars, low growth rates persist in the fifth instar even after transfer to a better diet (Woods 1999). Fig. 2.14 demonstrates how supernumerary moults by larvae of the African armyworm Spodoptera exempta (Noctuidae) enable them to reach the same final size when they are reared on poor quality grasses (Yarro 1985). Some female insects undergo an additional larval instar in order to store more nutrients for oogenesis (e.g Stockoff 1993). The seed beetle Stator limbatus (Bruchidae) varies by an order of magnitude in adult body size, due to resource competition between multiple larvae in a single seed, but egg sizes do not differ much and a longer

Was this article helpful?

0 0

Post a comment