0 10 20 30 40 50 60 70 Nitrogen eaten (mg)
Figure 2.3 Utilization plots of nutrient output versus intake. (a) Hypothetical plot in which the dotted line represents total elimination of a nutrient, while the solid line shows its utilization up to a certain level, with elimination occurring beyond this level. (b) Utilization plot for nitrogen in fifth instar Locusta migratoria fed four different diets (7: 7 indicates a diet containing 7% protein and 7% digestible carbohydrate). When nitrogen consumption exceeded 30 mg, uric acid excretion was used to remove the excess. Source (a): Regulatory Mechanisms in Insect Feeding, 1995, Simpson et al., pp. 251-278. With kind permission of Kluwer Academic Publishers. Source (b): Zanotto et al. (1993). Physiological Entomology 18, 425-434, Blackwell Publishing.
are the geometrical representation of analysis of covariance (ANCOVA) designs and are statistically preferable to the widely used ratio-based indices (Raubenheimer 1995). Specific examples of both behavioural regulation of food intake and physiological regulation of its utilization are described in Section 2.2.3 below.
Some studies have used both old and new methods, represented by nutritional indices and bicoordinate plots, to assess nutritional performance (Chown and Block 1997; Ojeda-Avila et al. 2003). A plot of growth rate against consumption rate gives a visual assessment of ECI, while growth against absorption gives ECD.
Regulatory mechanisms involved in insect feeding were comprehensively reviewed by Chapman and de Boer (1995). We will briefly examine the inter-relationships between feeding and digestion in caterpillars, the regulation of meal size in fluid feeders, and finally the regulation, especially in locusts, of intake of protein and digestible carbohydrate. These are all relatively short-term aspects, whereas diet switching, as defined by Waldbauer and Friedman (1991), refers to long-term changes in diet that occur between larva and adult or between instars in some herbivores.
Inter-relationships between consumption and digestion have been explored in detail in caterpillars (for excellent reviews see Reynolds 1990; Woods and Kingsolver 1999). The experimental caterpillar is commonly the fifth instar of the tobacco hornworm M. sexta (Lepidoptera, Sphingidae); appropriately, Manduca means 'the chewer'. During its last instar this species increases in mass at the rate of 2.7 g per day, which is faster than the growth rates of similar-sized altricial birds and is achieved without the benefit of endothermy (Reynolds et al.
1985). The gut and its contents form 39 per cent of the mass of the caterpillar throughout the feeding period of the fifth instar. The transformation of leaf protein into insect tissue to achieve such rapid growth can be divided into four steps (Woods and Kingsolver 1999): consumption of leaves, digestion of protein into small peptides and amino acids, absorption of amino acids across the midgut epithelium, and construction of tissue. The last three steps are post-ingestive events which influence, and are influenced by, the rate of consumption.
In the continuous flow digestive system of a caterpillar, gut passage rates are equal to rates of consumption. Gut passage rates involve a trade-off between fast processing and thorough processing. Woods and Kingsolver (1999) used a chemical reactor model to predict the concentration profiles of proteins and their breakdown products along the midgut, and found that an intermediate consumption rate gave the highest rate of absorption. Reynolds (1990) reached the same conclusion using a model of optimal digestion (Sibly and Calow
1986): AD is optimized at the optimal retention time, which then determines the rate of consumption. Caterpillars, therefore, restrain food intake to an optimal level which maximizes the rate of nutrient uptake and the rate of growth (Reynolds et al. 1985). From measurements of food passage rates, midgut dimensions, proteolytic activity in the lumen (Vmax), and the protein concentration giving half-maximal rates (Km), Woods and Kingsolver (1999) predicted that protein is digested rapidly in the anterior midgut but absorption of breakdown products may be a limiting step. However, Reynolds (1990) measured rapid uptake of labelled amino acids in the anterior midgut, which would suggest post-absorptive rather than absorptive constraints on growth. Further studies should emphasize caterpillars eating leaves, because plant proteins differ from those in artificial diets, and undigested protein in leaf fragments will extend further along the midgut (Woods and Kingsolver 1999). Most caterpillars feed by leaf-snipping (their mandibles have cutting but not grinding surfaces) and few of the cells in the ingested tissue are crushed. However, AD values for carbohydrate and protein are surprisingly high, suggesting that nutrients are extracted when the cell walls become porous, and this digestive strategy is apparently as efficient as that of grasshoppers, which crush leaf tissues (Barbehenn 1992).
Evidence for restrained food intake also comes from behavioural observations on temporal patterns of feeding in caterpillars. The combination of meal durations and meal frequencies determines the proportion of time an insect spends feeding. This, combined with the instantaneous feeding rate, gives the overall consumption rate (Slansky 1993). The proportion of time spent feeding by M. sexta larvae is up to 80 per cent on tobacco leaves, compared to 25 per cent on artificial diet, the difference being due to relative water contents, although growth rates are identical on both diets (Reynolds et al. 1986). Bowdan (1988) examined the microstructure of feeding on tomato leaves using an electrical technique, and showed that larger caterpillars ate more by increasing bite frequency and the length of meals, but meal frequency was unchanged. The periods of inactivity even at high rates of consumption, and the compensatory feeding which occurs when artificial diet is diluted with water or cellulose (Timmins et al. 1988), confirm that caterpillars could consume more food than they do.
The feeding rhythms of caterpillars vary greatly because they depend on ecological factors as well as digestive processes. The caterpillar's task is to maximize growth rate while avoiding risk, and the risk from predators and parasitoids can be great. Bernays (1997) quantified the risk during continuous observation of two caterpillar species, and mortality was so high during feeding that there must be strong selection for rapid food intake. Predation will also increase considerably on nutritionally poor plants. Caterpillars undergo spectacular changes in body size with growth, with major effects on feeding ecology, behaviour, and predator assemblages (Reavey 1993; Gaston et al. 1997). Minimizing risk can involve changes in feeding habit as individuals grow: as body size increases there is a general trend from concealed feeding (leaf miners or gall formers, which suffer from space constraints) to spinning or rolling leaves, to external feeding in late instars. In bigger caterpillars the surface area of gut for absorption is relatively less in relation to the volume of gut contents. Rates tend to decrease with increasing body size but efficiencies do not (Slansky and Scriber 1985). AD does not vary with size in M. sexta, remaining about 60 per cent throughout the fifth instar, but retention time increases. Correction of the body mass component in nutritional indices for the presence of food in the gut reduces their value only slightly (Reynolds et al. 1985).
2.2.2 Regulation of meal size: volumetric or nutritional feedback
The physiological regulation of meal size can potentially include sensory stimuli (either positive or negative), volumetric feedback via stretch receptors in the gut or body wall, haemolymph composition (osmolality or the concentration of individual nutrients), available reserves, and neuropeptides, many of which are known to affect contractile activity of the gut (Gade et al. 1997). Unravelling causal relationships is far from simple. For chewing insects, the most detailed information comes from acridids, in which volumetric feedback from stretch receptors in the gut is important in terminating a meal (reviewed by Simpson et al. 1995). These receptors are located in both the crop and ileum, those in the latter being stimulated by the remains of the previous meal. It is also likely that rapid changes in haemolymph osmolality and nutrient concentration inhibit further feeding. Locusts fed high-protein diets exhibited much greater increases in haemolymph osmolality and amino acid concentrations during a meal than those on low-protein diets, and the result was a longer interval until the next meal (Abisgold and Simpson 1987). Feeding stops when inhibitory feedbacks force excitation below the feeding threshold, and increasing inhibition during a meal is reflected in declining ingestion rates (Simpson et al. 1995). Volumetric feedbacks are less obvious in caterpillars, which lack the capacious crop of acridids, and meal size in Manduca may depend on feedback from nutrients in the gut lumen. Injection of soluble diet extract into the midgut lumen inhibited feeding, while an injection of xylose solution of the same osmolality did not (Timmins and Reynolds 1992).
Carbohydrate feeding is best understood in Diptera, although it is also fundamental to the aerial success of adult Lepidoptera and Hymen-optera, all three orders depending on a variety of liquid carbohydrate resources as immediate energy for flight (Stoffolano 1995). These insects have evolved an expandable and impermeable crop (diverticular in Diptera and Lepidoptera, linear in the Hymenoptera) located in the abdomen. The blowfly Phormia regina (Calliphoridae) has been used as an experimental model, and it is clear that information from abdominal stretch receptors ends the meal. Not surprisingly, the regulation of feeding behaviour has also been thoroughly investigated in blood feeders, especially mosquitoes, where nectar meals are directed to the crop and blood meals to the midgut but both kinds of meal are terminated by abdominal distension (reviewed by Davis and Friend 1995).
Meal quality and feeding regime also influence crop filling. The Australian sheep blowfly Lucilia cuprina ingests greater volumes of dilute glucose solutions by taking larger and more frequent meals (Simpson et al. 1989). Crop volume at the end of a meal was similar in L. cuprina ingesting 0.1 and 1.0 M glucose, due to volumetric inhibition. However, the flies maintained on 1.0 M glucose had fuller crops at the beginning of a meal because dilute solutions empty from the crop more rapidly (see Section 2.3.3) and can be ingested in greater quantity. Evaporative losses caused by bubbling behaviour in Rhagoletis pomonella (Diptera, Tephritidae) reverse the volumetric inhibition, permitting feeding to continue on dilute solutions (Hendrichs et al. 1992). Some confusion in the literature has arisen because researchers have used insects in very different nutritional states. Feeding behaviour varies greatly between insects fed ad libitum and those which are deprived of food and then offered single meals, as elegantly demonstrated by Edgecomb et al. (1994) for Drosophila melanogaster feeding on sucrose-agar diets. Flies fed ad libitum maintained much smaller crop volumes than food-deprived flies fed a single meal, and responded differently to sucrose concentrations up to 0.5 M (Fig. 2.4). In general, the volumes of sugar solution
Was this article helpful?