SEARCH

SEARCH BY CITATION

Keywords:

  • Drosophila;
  • genetic correlations;
  • metabolism;
  • phenotypic plasticity;
  • stress resistance

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Correlated responses to artificial selection for stress tolerance can provide insight into underlying genetic variation and the physiological basis of stress resistance. Lines of Drosophila melanogaster held in the absence of food or with an unsuitable resource, specifically decomposing lemon, responded to selection by becoming starvation resistant. The lemon-selected lines also adapted by evolving a resource-based induction response. Compared to control lines, the selected lines tended to store more lipid, develop slower and have a larger body size. Additional responses included resistance to desiccation and acetone fumes, suggesting multiple stress resistance is a correlated result of selection for starvation resistance. The specific metabolic rate was lower in the starvation selected lines and enzyme activities changed in response to selection. In particular, enzyme activities indirectly associated with lipid biogenesis increased in both types of selected lines. The correlated responses to the two selection regimes were sufficiently consistent to indicate a common basis for starvation resistance. Specific responses to starvation selection appeared to oppose the short-term phenotypic responses to starvation. Thus, a common response to stress selection may be to ameliorate the immediate physiological impact of the stress factor.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Biological stress may be defined in evolutionary terms (Hoffmann & Parsons, 1991). Sibly & Calow (1989) broadly define stress as ‘an environmental condition that, when first applied, impairs Darwinian fitness’ and similarly Koehn & Bayne (1989) define stress as ‘any environmental change that acts to reduce the fitness of an organism’. Genetic variation in stress tolerance will result in adaptive change to an extent that depends on the frequency of environments faced by the organism and the associated physiological costs (Hoffmann & Parsons, 1991). Unsuitable or insufficient food resources resulting in deprivation of normal nutrients constitutes environmental stress and it has been argued that stress associated with marginal resources impacts populations of most species (White, 1993). In this study, starvation resistance is investigated as a form of stress resistance that is poorly understood with respect to its potential importance as a form of adaptation.

Selection experiments are a useful way of studying changes in populations in response to stress. For instance, a previous starvation resistance selection experiment was conducted on D. melanogaster in which flies from replicate lines were deprived of food until high mortality occurred (Chippindale et al., 1996). The selected lines evolved to become highly resistant and the correlated responses to selection included increased lipid content and slower development time. However, results from selection experiments have not always been consistent from one study to another. For instance, Hoffmann & Parsons (1989a,b, 1993a) found that lines selected for increased desiccation resistance had a lowered resting metabolic rate and a decreased rate of water loss without any change in water content of the flies. In a different set of selected lines, Gibbs et al. (1997) also found a decreased rate of water loss but in contrast to Hoffmann and Parsons they also found a significant change in the amount of water flies contained as well as a change in glycogen level. One possible explanation for the different responses is that the lines used by Hoffmann and Parsons were derived from a sample of flies not far removed from the field, whereas the lines used by Gibbs et al. (1997) are part of a long-standing selection experiment derived from an old laboratory population. Unfortunately, the magnitude and nature of genetic changes associated with laboratory adaptation are largely unknown. While inconsistencies among studies prevent generalizations, the heterogeneity that they suggest may be important.

There is evidence for a relationship between stress, energetics and metabolic rate (Hoffmann & Parsons, 1991). Stress resistance can be achieved by altering energy expenditure to compensate for the physiological impact caused by stress or repair damage caused by exposure to a stressor. Similarly, conserved metabolic resources could be used to withstand stress. At the whole organism level, respiration rate (metabolic rate) represents the level of energy use by organisms. As a corollary, reduced metabolic rate would be expected to be manifest in changes in enzyme activities associated with energy compound use and biosynthesis. Starvation in particular might strongly impact the activity of these enzymes.

The goal of the present study was to investigate starvation resistance using markedly different selection regimes and lines of D. melanogaster that were relatively recently isolated from the field. A pair of populations was subjected to starvation selection in a manner similar to that described in Chippindale et al. (1996). The second selection regime was selection on lemon, a natural but marginal food source. The assumption that lemons are a suboptimal resource is based on the observation that D. melanogaster is relatively scarce on lemon in the field (Prince & Parsons, 1980) and data indicating that lemon covered with Penicillium is a poor habitat (Atkinson, 1981). At the start of the experiment it was anticipated that lemon could select for starvation resistance. Starvation resistance was measured in the lemon selected lines, starvation selected lines and control lines. Chippindale et al. (1996) measured development time and lipid content, and these traits were also investigated here. To evaluate the possibility that stress resistance responses have a common genetic basis (Hoffmann & Parsons, 1989b), all lines were tested for resistance to desiccation and solvent exposure. Metabolic rate was also measured.

A prospective association between intermediary metabolism enzyme activities, energy storage compound composition (Clark & Keith, 1988; Clark, 1989) and starvation resistance motivated investigation of energy storage compounds and relevant enzyme activities in the selected and control lines. The enzymes assayed are responsible for catabolism or synthesis of compounds used for energetics. These included glyceraldehyde phosphate dehydrogenase, phosphoglucomutase and hexokinase which metabolize glucose, glycogen synthase which synthesizes glycogen, and both glycogen phosphorylase and trehalase which catabolize carbohydrate energy storage compounds. Fatty acid synthase was examined because it directly participates in lipid biosynthesis, while glucose-6-phosphate dehydrogenase, malic enzyme and 6-phosphogluconate dehydrogenase were assayed because they generate NADPH which indirectly contributes to lipid biosynthesis. Finally, the abundance of energy storage compounds was measured. Measurements were made using stressed and unstressed flies so that the selection response could be compared to phenotypic changes induced by starvation.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Thirty isofemale lines of D. melanogaster were initiated from flies collected on fallen fruit in a lemon grove near Orange Cove, Fresno County, California (see Harshman et al., 1991). The flies were only cultured on a standard cornmeal, yeast, sugar, agar Drosophila medium referred to in this study as laboratory medium. The lines were pooled by combining approximately an equal number of offspring from each isofemale line to produce a population of at least 3000 individuals. A mass population of this size was subsequently maintained on a standard laboratory medium for 15 generations. The mass population was then subdivided into eight lines, each with ≈500 individuals. Three of the lines were maintained on laboratory medium serving as unselected controls, three lines were reared on lemon and two lines were subjected to selection for starvation resistance as described in the section below. The lemon selected lines correspond to those described in Harshman et al. (1991) and have the same line type designation (L). The control lines described in this study (C) correspond to the lines designated M in the earlier study. The establishment of these lines is described in Harshman et al. (1991). The starvation selected lines (S) were not described in the previous study. All lines used in this study were assayed between selection generations 55 and 85.

Selection regimes

Lemon

For selection on lemon, ≈150 flies (males and females) were placed in each of two bottles with 10 g of freshly cut pesticide-free (‘organic’) lemon. Generally one piece of lemon was placed on several discs of filter paper to absorb excess moisture although a smaller second piece of cut lemon was sometimes added to bring the total weight to 10 g. In three replicate bottles for each line, a total of ≈450–500 adults were held on cut fruit for around 7–10 days at room temperature until at least 50% mortality was observed. Mortality may have been due to the natural insecticidal activity of lemons (Su & Horvat, 1987), microbes or microbial metabolites. We suspect the latter because heavy mortality was generally observed after microbial growth had become extensive and the fruit was substantially decomposed. In each bottle adult mortality occurred on rotting lemon and 35–70 survivors were transferred to 30 g of cut lemon on a bed of vermiculite in each replicate breeding and rearing bottle. The rearing regime was heterogeneous and larval density was not carefully controlled; to reduce larval crowding, adults were removed from fresh lemon once larvae were observed. In general, 100–200 selected flies were used as the parents of 300 progeny used for selection in the next generation. Three independent lines (L) were subjected to 85 generations of selection in this manner.

Starvation

For selection based on the absence of food, ≈ 150 flies (males and females) were placed in each of two empty bottles per line. The cotton plug for each bottle was saturated with water and thus flies were held in a high-humidity environment in constant proximity to moisture. After at least 50% mortality, the survivors were transferred to laboratory medium to produce the next generation. The survivors were predominantly females. Initially, the flies were selected for 45 h without food. During the generations used for assays in this study, flies were selected for 120 h without food. In general, 100–150 selected flies were used as the parents of 450–500 progeny used for selection in the next generation. Larval overcrowding was reduced in the rearing bottles by controlling the number of days that adults were allowed to lay eggs. Two independent lines (S) were subjected to 85 generations of selection in this manner.

Control lines

Three lines were reared on laboratory medium and not subjected to selection. A similar number of flies were used to initiate the control and selected lines each generation. Three independent lines (C) were maintained for the same number of generations as the selected lines.

Assays

Selected and control line flies were reared for at least one generation without selection on laboratory Drosophila medium prior to assays. Flies used in the assays were 4–7 days old. Except as otherwise noted, flies were tested at room temperature (≈20 °C). Assays reported here were conducted on samples derived from generations 55–85. The initial tests of starvation were conducted at generation 55, while most correlated stress responses and metabolic rates were assayed at around generation 60. One desiccation assay (where flies were scored until they died – see below) was conducted at generation 70, the assays on energy storage compounds and enzymes activities were conducted at around generation 75 and development rate was measured at generation 85.

Starvation resistance

Flies from all populations were tested for survival in the absence of food at room temperature. For the starvation resistance assay, 50 females or 50 males were placed in an empty bottle capped with a water-saturated fibre plug. The number of dead flies was determined at intervals until all flies had died. Three replicate bottles for each sex and two trials were used to test starvation resistance.

Survival on decomposing lemon

Flies from all populations were tested for survival on decomposed lemon at room temperature. For each trial, a large quantity of lemon was cut into ≈ 10-g pieces and allowed to rot in a beaker for 2 weeks. Fruit was cut into pieces in an attempt to emulate the decomposition process in bottles used for selection. The decomposed lemon was mixed into a slurry and transferred in 10-g aliquots into vials. Prior to adding flies to each vial, three females and three males aged 4–7 days were held on standard laboratory medium or freshly cut lemon for 4 days. Ten to 16 replicate vials were set up per line for the pre-exposure treatments (laboratory medium vs. fresh lemon). The number of dead flies was determined daily until all flies had died.

Correlated stress assays

Flies from all lines were subjected to desiccation resistance tests. In one test, 10 males or 10 females were placed in an empty vial with no access to water. Mortality was monitored at 12-h intervals until no flies remained alive. In another test, females were subjected to desiccation for 19 h by holding them in sets of 10 in empty vials over silica in a sealed desiccator. The number of females alive was recorded after transfer to fresh medium for an overnight recovery period. Eight replicate vials were used to test each line.

Flies from all lines were tested for resistance to solvent fumes. Ten females or 10 males were placed in vials without food. The vials were each capped with a cotton plug and sealed in a hydration chamber with a standard volume of either 100% ethanol or 1.5% acetone in water. Females were exposed to ethanol fumes for 3 h and males for 1.5 h. Females were exposed to acetone fumes for 24.5 h and males for 20 h. After the chamber seal was broken, the flies in a vial were transferred to a vial with Drosophila medium for an overnight recovery period prior to counting survivors. Eight replicate vials were set up per sex for each line.

Respiration, enzyme activities and energy storage compounds

Respiration rate was measured using females from selected and control lines. A Gilson respirometer was used for each reading on 20 females added to a respirometer tube at 25 °C. The assay was conducted in the dark to minimize activity. The volume of oxygen consumed in a 3-h period was determined for each respirometer tube. At the end of the assay, the females were weighed to determine the specific metabolic (respiration) rate.

Activities of intermediary metabolism enzymes, as well as glycogen, lipid and soluble protein content, were determined by methods described in Clark & Keith (1988). Males were used for this portion of the study because assays on males were previously observed to be more repeatable than those on females. Briefly, the procedure was to weigh and then homogenize groups of five males from each population. The homogenates were centrifuged at 1500 r.p.m. to remove structural and cellular debris. Lipid was resuspended in the homogenate after centrifugation. The homogenates were frozen in microtitre plates which were thawed just prior to assays. Glycogen was determined spectrophotometrically after enzymatic conversion to glucose. Triacylglycerol, the main storage form of storage lipid in insects, was determined spectrophotometrically after enzymatic conversion to glycerol. The sample of flies for the assays was divided so that males were tested unstarved or after 18 h of starvation.

Weight and lipid composition increased in the selected lines and decreased after 18 h of starvation, suggesting that simple standardization to enzyme activities per gram body weight might not be appropriate. It would be appropriate to correct by weight if the weight difference reflected an underlying difference in the number of metabolically active cells. However, in this study, change in weight is at least partially a result of change in amount of energy storage compounds which are not expected to directly affect enzyme activity. Compared to the control lines, protein content was relatively low in the lemon selected lines but not in the starvation selected lines nor in the lines after 18 h of starvation. Protein content is presumably a representative measure of body mass. Considering the options, we decided to analyse the raw data adjusted by protein as a covariate (ANCOVA).

Development rate

This parameter was determined for the selected and control lines by transferring 40 eggs to each of eight replicate vials per line. The vials were held at 22 °C to determine development time. The time of first eclosion from any of the vials was recorded as t0. Thereafter, all vials were checked for adults at intervals of 3–6 h until eclosion was complete in all vials. Development time was tested at generation 85.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Starvation resistance

Figure 1 presents the survival of females and males from the selected and unselected lines in the absence of food. Males and females from the lemon selected lines and the acute starvation selected lines survived longer than did the same sex counterpart from the unselected lines (SAS LIFETEST procedure: χ2< 0.0001). Lemon selection and starvation selection resulted in starvation resistance.

image

Figure 1. Mean survival time (and standard error) without food. For both females and males, line types differ in starvation resistance (SAS LIFETEST procedure: Wilcoxon χ2P<0.0001).

Download figure to PowerPoint

Survival on decomposing lemon

Figure 2 shows survival on decomposed lemon as a function of previous exposure to lemon or laboratory medium. As indicated by the ANOVA described in the legend to Fig. 2, there was no statistically significant effect of previous exposure but the selection treatments did differ significantly. In addition, there was a significant interaction between previous exposure and selection treatment; although the survival of starvation selected lines was not affected, there was an increase in survival when lemon lines were previously exposed to fresh lemon in contrast to a decrease for the control lines. These differences are significant as evident in post hoc protected LSD tests, which indicate increased lemon line survival after pre-exposure to lemon (< 0.001) and decreased control line survival after pre-exposure (< 0.001). The lemon lines have therefore evolved an inducible response that allowed longer survival on decomposing lemon.

image

Figure 2. Mean survival (and standard error) on decomposed lemon after pre-exposure to fresh lemon or medium. From ANOVA: pre-exposure (P=0.4676), line type (P=0.0042), pre-exposure × line type (P=0.0031).

Download figure to PowerPoint

Correlated stress assays

Two different experiments were conducted to assay the resistance of flies to desiccation. Figure 3 presents a summary of the survival of females and males from selected and unselected lines under water deprivation conditions. Flies from the lemon lines survived longer than those from control lines (SAS LIFETEST procedure: χ2< 0.0001). Similarly, flies from the starvation lines survived longer than flies from control lines (SAS LIFETEST procedure: χ2< 0.0001). Table 1 presents the proportion of females that died after 19 h without access to water. The ANOVA presented in Table 1 shows that there is a statistically significant effect of selection treatment (< 0.001). It is clear that the selected lines were less impacted by desiccation.

image

Figure 3. Mean survival time (and standard error) without access to water. For both females and males, line types differ in resistance to water deprivation (SAS LIFETEST procedure = Wilcoxon χ2P<0.0001).

Download figure to PowerPoint

Table 1.  Desiccation resistance of selected and control lines. Mean proportion (standard deviation) adult mortality when females are held without access to water for 19 h. C = unselected lines, L = lemon lines, S = starvation lines. Thumbnail image of

The proportion of females or males surviving after exposure to 100% ethanol or 1.5% acetone is given in Table 2. For each of the tests, there are statistically significant differences between line types (selection treatments) except for the case of male exposure to ethanol fumes. Males and females from the starvation lines are more resistant to acetone fumes than are the other lines. In the case of ethanol exposure, the starvation selected and control lines are similar in resistance whereas females from the lemon selected lines are more susceptible. Flies from the lemon selected lines generally appear to have lost ethanol tolerance.

Table 2.  Resistance of selected and control lines to solvent fumes. Mean proportion (standard deviation) adult mortality after exposure to ethanol or acetone vapour. C = unselected lines, L = lemon lines, S = starvation lines. Thumbnail image of

Metabolism and body composition

Table 3 presents the total and specific metabolic rates (respiration measured by oxygen consumption) for selected and control line females. For total consumption, there is no significant difference between selection treatment and control populations. For the specific metabolic rate (total oxygen consumption adjusted by weight), there is a statistically significant difference between line types. The starvation selected lines have the lowest specific metabolic rate.

Table 3.  Mean (SD) oxygen consumption by females from selected and control lines. C = control lines, L = lemon lines, S = starvation lines. Thumbnail image of

Table 4 presents the proportional mean of weight, protein, glycogen, triacylglycerol and enzyme activities standardized to the value of unstarved control line males. Table 4 also presents the statistical analysis of body composition and enzyme activities. When the ANCOVA is subjected to a sequential Bonferroni correction, only those P values less than the critical value of 0.001 remain statistically significant. Weight decreased in response to starvation and increased in response to selection. The amount of triacylglycerol was higher in the starvation selected lines. A statistically significant decrease in amount of glycogen was observed after 18 h of starvation and there is a suggestion of a decrease in triacylglycerol content.

Table 4.  Relative mean values of starvation selected and lemon selected lines as a proportion of the control line means. The phenotypic treatments (TREAT) result from holding flies unstarved (UNST) on medium or starved (ST) away from food for 18 h. The statistical analysis of line types (LINE) and effect of 18 h of starvation is based on an ANCOVA using total protein as the covariate. Thumbnail image of

There were changes in enzyme activities in response to selection and starvation. Relevant to lipid biogenesis, G6PD (a pentose shunt regulatory enzyme) exhibited higher activity in the selected lines, whereas PGD activity (another pentose shunt regulatory enzyme) decreased in response to starvation. Relevant to glycogen use, GP activity decreased in the selected lines in response to starvation, but not in the control lines. The significant line by treatment interaction term argues that the plasticity of the enzyme changed as a result of selection. Relevant to sugar metabolism, hexokinase activity decreased in response to starvation and GPDH activity increased in response to selection.

Development rate

Figure 4 presents the mean eclosion times for females and males from the lines. Both female and male eclosion times differ significantly between lines (SAS LIFETEST procedure: Wilcoxon χ2< 0.0001). Lemon selected males develop more slowly than males from the other two line types and females from the selected lines take longer to develop than control line females. LSD post hoc tests provide statistical support for both the slower development time of lemon line males compared to starvation line males (= 0.0076) and for the slower development time of lemon line males compared to control line males (= 0.0009). Post hoc tests also indicate that lemon line females do not develop slower than starvation line females (= 0.2063) but they do develop more slowly than the control line females (= 0.0004). There is a suggestion that the starvation selected females develop more slowly than control line females (= 0.0316). In general, selection resulting in starvation resistance has tended to increase the time required for egg-to-adult development.

image

Figure 4. Mean time (and standard error) of eclosion after the first observation of adult emergence. The line types differ in female and male eclosion time (SAS LIFETEST procedure: Wilcoxon χ2P<0.0001).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

In the present study, body weight, energy storage compound abundance, and enzyme activities were observed to respond to selection (Table 4). Body weight and lipid abundance increased in the selection treatments. In addition, selection for starvation resistance was correlated with slower development time. In the study by Chippindale et al. (1996), selection for starvation resistance using lines derived from a long-standing laboratory population resulted in an increase in lipid content and slower development time. In contrast to that study, the lines used here were relatively recently derived from the field. Thus, there is corroborative evidence for a general association between starvation resistance, increased lipid content and delayed development time. The association between increased starvation resistance and lipid content is what one would expect and is supported by other studies (Service, 1987; Da Lage et al., 1989; Zwaan et al., 1991) whereas the association with developmental delay is less obvious. To explain this observation, Chippindale et al. (1996) proposed a trade-off between larval development time and adult storage compound acquisition arguing that slower development allows accumulation of storage lipid for the adult stage.

We have also demonstrated the presence of an evolved inducible response to a food resource. Importantly, the lemon selected lines survived longer on decomposing fruit after first being exposed to fresh lemon. Pre-exposure to lemon must have induced a change in these flies that allowed them to survive longer on a marginal substrate. This induction effect appears to be specific to the lemon selection lines because prior exposure of the starvation selected lines or control lines did not improve survival on rotting lemon. In fact, prior exposure of control lines to lemon resulted in relatively lower survival. In an earlier study, Harshman et al. (1991) showed that pre-exposure to lemon induced an increase in a substrate-specific glutathione-S transferase (a detoxification enzyme) activity only in the lemon selected lines which may be a mechanism for increased survival on rotting lemon.

Phenotypic plasticity has previously been demonstrated to have a genetic basis and it has also been demonstrated that phenotypic plasticity can respond to selection or impact a selection response (Frey, 1964; Arboleda-Rivera & Compton, 1974; Berenbaum, 1983; Yamasaki & Matsuo, 1984; Schlichting, 1986; Cohen et al., 1992; Hoffmann & Watson, 1993; Prapaipong et al., 1994; Pigliucci et al., 1995; Berenbaum et al., 1996; Watson & Hoffmann, 1996; Chippindale et al., 1997). In D. melanogaster, a condition dependent relationship between body size and temperature was alterable in one selection experiment (Scheiner & Lyman, 1991). In another study on D. melanogaster, Wang & Clark (1995) observed that one response to selection by high sucrose medium was to buffer the effect of exposure to such medium. Our results provide another example of phenotypic plasticity having a genetic basis and being associated with a selection response.

In the present study, both types of selected lines were relatively resistant to desiccation and the starvation selected lines were resistant to acetone fumes. However, starvation resistance was not positively correlated with ethanol fume tolerance. There was evidence for multiple stress resistance, but selection for starvation resistance did not result in positive genetic correlation among all of the stress resistance traits studied. In general, positive genetic correlations among stress resistance traits appear to be relatively common (Hoffmann & Parsons, 1991). However, there are exceptions exemplified by the demonstration that different measures of heat resistance are unrelated (Hoffmann et al., 1997). Nevertheless, the following examples of positive genetic correlations emphasize the relationship between other stress resistance traits and starvation resistance. In earlier studies using D. melanogaster it was determined that lines selected for desiccation or delayed female reproduction were also starvation resistant (Service et al., 1985; Hoffmann & Parsons, 1989a; 1993a,b). Moreover, selection for ethanol knockdown resistance using D. persimilis resulted in resistance to starvation (Cohan & Hoffmann, 1989). An association between lipid content and general stress resistance has been suggested (Hoffmann & Parsons, 1991) and such an association could provide an explanation for the observations of a correlation between starvation resistance and other forms of stress resistance. Alternatively, there may be an underlying stress resistance ‘pathway’ that confers multiple stress resistance. For example, in Caenorhabditis elegans the dauer pathway can confer stress resistance and control metabolism in response to environmental stress (Jazwinski, 1996).

Our lines selected for starvation resistance had a lower specific metabolic rate. A variety of studies indicate an association between reduced metabolic rate and stress resistance (Hoffmann & Parsons, 1991), but there are contradictory Drosophila results. Specifically, selection for delayed female reproduction resulted in lines that had increased stress resistance and a lower metabolic rate in young flies (Service, 1987). Selection for desiccation resistance in D. melanogaster produced lines that were resistant to different stresses and had a relatively low metabolic rate (Hoffmann & Parsons, 1989a,b). Also supporting the hypothesis of a reduction in metabolism as a result of exposure to stressful environments, the study by Wang & Clark (1995) documented an evolved reduction in intermediary metabolism enzyme activities in D. melanogaster cultured on 10% sucrose which is a suboptimal medium (Wang & Clark, 1995). However, Djawdan et al. (1997) found no differences in metabolic rate between desiccation selected and control lines of D. melanogaster after line differences were corrected for metabolic reserves. Approximately 25% of the dry weight of D. melanogaster females comprises lipid and glycogen (Chippindaleet al., submitted). In our study, assuming selected and control line differences in female energy storage compounds similar to those observed in males (Table 4), it is not clear that the differences in female specific metabolic rates would hold up after correction for metabolic reserves.

There is evidence for an association between genetic variation for intermediary metabolism enzyme activity and energy compound storage. Chromosomes isolated from natural populations of D. melanogaster have been used to demonstrate genetic variation for these enzyme activities and body content of storage lipids and carbohydrates (Clark & Keith, 1988; Clark, 1989). In Clark & Keith (1988), second chromosome replacement lines derived from a Pennsylvania population revealed a positive correlation between triacylglycerol abundance and a fatty acid synthesis (FAS) enzyme activity as well as the activity of two enzymes that generate NADPH (G6PD and PGD) which is required for lipid biosynthesis. In Clark (1989), second chromosome replacement lines from a California population demonstrated an association between triacylgycerol abundance and activity of NADPH generating enzymes (G6PD, PGD and ME) but not FAS activity. Finally, a larger set of relevant enzyme activities were altered by laboratory mutagenesis, which suggests the size of the potential pool of genes that may influence intermediary metabolism and storage compound related enzyme activity (Clark et al., 1995a,b).

Other studies have demonstrated that starvation can affect enzyme activities in D. melanogaster (Pecsenye et al., 1996) and that allozyme variation is associated with starvation resistance (Oudman et al., 1994). The starvation treatment changed enzyme activities in our study. PGD activity decreased in response to starvation, particularly in the starvation selected lines. HEX and GP activity also decreased in response to starvation, especially in the lemon lines. Selection for starvation resistance changed the activity of an enzyme involved in generating NADPH (G6PD) which is indirectly associated with lipid biogenesis and there was a suggestion of an increase in activity of two other NADPH generating enzymes (PGD and ME). However, an enzyme activity directly involved in lipid synthesis, FAS, did not respond to selection. Co-regulation of G6PD and PGD is indicated by data showing that they covary in natural populations (Cavener & Clegg, 1981; Wilton et al., 1982; Miyashita & Laurie-Ahlberg, 1984; Clark & Keith, 1988). Furthermore, these enzyme activities have been observed to coordinately change among related species of Drosophila (Clark & Wang, 1994) which indicates that they tend to coevolve. However, there is no evidence for coregulation from phenotypic manipulation nor strong evidence for coevolution after selection in our study.

We have shown that starvation resistance can arise from two different selection regimes. Parsons (1983) argued that starvation is more likely to be a significant agent of selection in higher-latitude Drosophila populations where the food sources are presumably less predictable. This argument can be extended to a wide variety of organisms that experience substantial temporal or spatial variation in resource availability or quality. The lemon selection regime may select for starvation resistance because decomposing lemon is nutritive poor or toxic. We observed similar correlated responses to selection for starvation resistance regardless of the difference in selection regimes. This consistency in response to selection extends across a range of phenotypes (enzyme activities, body size, lipid storage, metabolic rate and desiccation resistance). Starvation resistance is potentially an important form of stress resistance and there is a suggestion of an underlying consistency among correlated traits.

The correlated selection responses were not necessarily in the same direction as the changes arising from starvation. The enzyme activities that responded to starvation (HEX and PGD) did not significantly change in activity as a function of selection. Conversely, two enzyme activities that responded to selection (G6PD and GPDH) did not respond to starvation. Weight decreased in abundance and lipid marginally decreased in abundance in response to starvation whereas starvation selection appeared to have counteracted this response, resulting in increased weight and lipid. Exposure of the unselected lines to lemon decreased survival on decomposed lemon which was counteracted by the adaptive induction response to selection. PGD decreased in activity in response to starvation and an observed countertrend may be the selected increase in G6PDH activity which also acts to control flux through the pentose shunt pathway.

Some of the responses to starvation selection acted in opposition to the phenotypic response to starvation. For example, the phenotypic impact of starvation was to reduce weight and the selection response was to increase body size which would counter the loss of weight caused by starvation conditions. This pattern is similar to counter-gradient selection as it was originally described for montane Rana clamitans (Berven et al., 1979). The generality suggested by our work is that selection may act to counter (buffer) the immediate phenotypic impact of stress or other agents of selection.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

This work was partially supported by an NSF-EPSCOR Project Grant 92–555225 at the University of Nebraska-Lincoln, a NSF grant to A. G. Clark (DEB 9419631) and a grant from the Australian Research Council to A. A. Hoffmann.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  • 1
    Arboleda-Rivera, F. & Compton, W.A. 1974. Differential response of maize (Zea mays L) to mass selection in diverse selection environments . Theor. App. Genet. 44: 77 81.
  • 2
    Atkinson, W.D. 1981. An ecological interaction between citrus fruit, Penicillium moulds and Drosophila Immigrans Sturtevant (Diptera: Drosophilidae). Ecol. Entomol. 6: 339 344.
  • 3
    Berenbaum, M.R. 1983. Coumarins and caterpillars: a case for coevolution. Evolution 37: 163 179.
  • 4
    Berenbaum, M.R., Favret, C., Schuler, M.A. 1996. On defining ‘key innovations’ in an adaptive radiation. Cytochrome P450S and Papilionidae. Am. Nat. 148: S139 S155.
  • 5
    Berven, K.A., Gill, D.E., Smith-Gill, S.J. 1979. Countergradient selection in the green frog, Rana clamitans. Evolution 33: 609 623.
  • 6
    Cavener, D.R. & Clegg, M.T. 1981. Evidence for biochemical and physiological differences between enzyme genotypes of Drosophila melanogaster. Proc. Nat. Acad. Sci. USA 78: 4444 4447.
  • 7
    Chippindale, A.K., Chu, T.J.F., Rose, M.R. 1996. Complex trade-offs and the evolution of starvation resistance in Drosophila melanogaster . Evolution 50: 753 766.
  • 8
    Chippindale, A.K., Gibbs, A.G., Sheik, M., Yee, K., Djawdan, M., Bradley, T.J. Physiology, development, and the evolution of desiccation resistance in Drosophila melanogaster. Evolution in press.
  • 9
    Chippindale, A.K., Leroi, A.M., Borash, D.J., Saing, H., Rose, M.R. 1997. Phenotypic plasticity and selection in Drosophila life history evolution. 2. Diet, mates, and the cost of reproduction. J. Evol. Biol.10: 269–293.
  • 10
    Clark, A.G. 1989. Causes and consequences of variation in energy storage in Drosophila melanogaster . Genetics 123: 131 144.
  • 11
    Clark, A.G. & Keith, L.E. 1988. Variation among extracted lines of Drosophila melanogaster in triacylglycerol and carbohydrate storage. Genetics 119: 595 607.
  • 12
    Clark, A.G., Wang, L. &, Hullenberg, T. 1995a. Spontaneous mutation rate of modifiers of metabolism in Drosophila . Genetics 139: 767 779.
  • 13
    Clark, A.G., Wang, L., Hullenberg, T. 1995b. P- element-induced variation in metabolic regulation in Drosophila. Genetics 139: 337 348.
  • 14
    Clark, A.G. & Wang, L. 1994. Comparative Evolutionary analysis of metabolism in nine Drosophila species . Evolution 48: 1230 1243.
  • 15
    Cohan, F.M. & Hoffmann, A.A. 1989. Uniform selection as a diversifying force in evolution: evidence from Drosophila . Am. Nat. 134: 613 637.
  • 16
    Cohen, M.B., Schuler, M.A., Berenbaum, M.R. 1992. A host-inducible P450 from a host-specific caterpillar: molecular cloning and evolution. Proc. Nat. Acad. Sci. USA 89: 10920 10924.
  • 17
    Da Lage, J.L., Capy, P., David, J.R. 1989. Starvation and desiccation tolerance in Drosophila melanogaster adults: effects of environmental temperature. J. Insect Physiol. 35: 453 457.
  • 18
    Djawdan, M., Rose, M.R., Bradley, T.J. 1997. Does selection for stress resistance lower metabolic rate? Ecology 78: 828 837.
  • 19
    Frey, K.J. 1964. Adaptation reaction of oat strains selected under stress and non-stress environmental conditions. Crop Sci. 4: 55 58.
  • 20
    Gibbs, A.G., Chippindale, A.K., Rose, M.R. 1997. Physiological mechanisms of evolved desiccation resistance in Drosophila melanogaster . J. Exp Biol. in press.
  • 21
    Harshman, L.G., Ottea, J.A., Hammock, B.D. 1991. Evolved environment-dependent expression of detoxication enzyme activity in Drosophila melanogaster . Evolution 45: 791 795.
  • 22
    Hoffmann, A.A., Dagher, H., Hercus, M., Berrigan, D. 1997. Comparing different measures of heat resistance in selected lines of Drosophila melanogaster . J. Ins. Physiol. 43: 393 405.
  • 23
    Hoffmann, A.A. & Parsons, P.A. 1989a. An integrated approach to environmental stress tolerance and life-history variation. Dessication tolerance in Drosophila . Biol. J. Linn. Soc. 37: 117 136.
  • 24
    Hoffmann, A.A. & Parsons, P.A. 1989b. Selection for increased desiccation resistance in Drosophila melanogaster: additive genetic control and correlated responses for other stresses. Genetics 122: 837 845.
  • 25
    Hoffmann, A.A. & Parsons, P.A. 1991. Evolutionary Genetics and Environmental Stress. Oxford University Press, Oxford.
  • 26
    Hoffmann, A.A. & Parsons, P.A. 1993a. Selection for adult desiccation resistance in Drosophila melanogaster: Fitness components, larval resistance, and stress correlations. Biol. J. Linn. Soc. 48: 43 54.
  • 27
    Hoffmann, A.A. & Parsons, P.A. 1993b. Direct and correlated responses to selection for desiccation resistance: a comparison of Drosophila melanogaster and D. simulans . J. Evol. Biol. 6: 643 657.
  • 28
    Hoffmann, A.A. & Watson, M. 1993. Geographical variation in the acclimation responses of Drosophila to temperature extremes. Am. Nat. 142: S93 S113.
  • 29
    Jazwinski, S.M. 1996. Longevity, genes, and aging. Science 273: 54 59.
  • 30
    Koehn, R.K. & Bayne, B.L. 1989. Towards a physiological and genetical understanding of the energetics of the stress response. Biol. J. Linn. Soc. 37: 157 171.
  • 31
    Miyashita, N. & Laurie-Ahlberg, C.C. 1984. Genetical analysis of chromosomal interaction effects on the activities of the glucose-6-phosphate and 6- phosphogluconate dehydrogenases in Drosophila melanogaster . Genetics 106: 655 668.
  • 32
    Oudman, L., Delden, W.V., Kamping, A., Biljsma, R. 1994. Starvation resistance in Drosophila melanogaster in relationship to the polymorphisms at the Adh and alpha-Gpdh loci. J. Insect Physiol. 40: 709 713.
  • 33
    Parsons, P.A. 1983. The Evolutionary Biology of Colonizing Species. Cambridge University Press, Cambridge.
  • 34
    Pecsenye, K., Lefkovitch, L.P., Giles, B.E., Saura, A. 1996. Differences in environmental temperature, ethanol and sucrose associated with enzyme activity and weight changes in Drosophila melanogaster . Insect Biochem. Molec. Bio. 26: 135 145.
  • 35
    Pigliucci, M., Schlichting, C.D., Whitton, J. 1995. Reaction norms of Arabidopsis. II. Response to stress and unordered environmental variation. Funct. Ecol. 9: 537 547.
  • 36
    Prapaipong, H., Hung, C.F., Berenbaum, M.R., Schuler, M.A. 1994. The role of P450s in insect–plant interactions. In: Proc. Cytochrome P450 8th International Conference (M. C. Lechner and J. Libbey, eds), pp. 249–255. Eurotext, Paris.
  • 37
    Prince, G.J. & Parsons, P.A. 1980. Resource utilization specificity in three cosmopolitan Drosophila species . J. Nat. Hist. 14: 559 565.
  • 38
    Scheiner, S.M. & Lyman, R.F. 1991. The genetics of phenotypic plasticity. II. Response to selection. J. Evol. Biol. 4: 23 50.
  • 39
    Schlichting, C.D. 1986. The evolution of phenotypic plasticity in plants. Ann. Rev. Ecol. Syst. 17: 667 693.
  • 40
    Service, P.M. 1987. Physiological mechanisms of increased stress resistance in Drosophila melanogaster selected for postponed senescence. Physiol. Zool. 60: 321 327.
  • 41
    Service, P.M., Hutchinson, E.W., MacKinley, M.D., Rose, M.R. 1985. Resistance to environmental stress in Drosophila melanogaster selected for postponed senescence . Physiol. Zool. 58: 380 389.
  • 42
    Sibly, R.M. & Calow, P. 1989. A life-cycle theory of responses to stress. Biol. J. Linn. Soc. 37: 101 116.
  • 43
    Su, H.C.F. & Horvat, R. 1987. Isolation and characterization of four major components from insecticidally active lemon peel extract. J. Agric. Food Chem. 35: 509 511.
  • 44
    Wang, L. & Clark, A.G. 1995. Physiological genetics of the response to a high-sucrose diet by Drosophila melanogaster . Biochemical Genetics 33: 19951995.
  • 45
    Watson, M.J.O. & Hoffmann, A.A. 1996. Acclimation, cross-generation effects, and the response to selection for increased cold resistance in Drosophila. Evolution 50: 1182 1192.
  • 46
    White, T.C.R. 1993. The Inadequate Environment: Nitrogen and the Abundance of Animals. Springer-Verlag, Berlin.
  • 47
    Wilton, A.N., Laurie-Ahlberg, C.C., Emigh, T.H., Curtsinger, J.W. 1982. Naturally occurring enzyme variation in Drosophila melanogaster. II. Relationship among enzymes. Genetics 102: 207 221.
  • 48
    Yamasaki, T. & Matsuo, Y. 1984. Genetic analysis of natural populations of Drosophila melanogaster in Japan. III. Genetic variability of inducing factors of amylase and fitness. Genetics 108: 223 235.
  • 49
    Zwaan, B.J., Bijlsma, R., Hoekstra, R.F. 1991. On the developmental theory of ageing. I. Starvation resistance and longevity in Drosophila melanogaster in relationship to pre-adult breeding conditions. Heredity 66: 29 39.