Abstract
Laboratory selection for resistance to starvation has been conducted under relatively controlled conditions to investigate direct and correlated responses to artificial selection. With regards to starvation resistance, there are three physiological routes by which the trait can evolve: resource accumulation, energy conservation and starvation tolerance. A majority of energetic compounds and macromolecules including triglycerides, trehalose and other sugars, and soluble protein increased in abundance as a result of selection. Movement was additionally investigated with selected males moving less than control males and selected females exhibiting a similar response to selection. Results obtained from this study supported two of the possible evolutionary mechanisms for adaptation to starvation: energy compound storage and conservation. If the response to selection is based on an evolutionarily conserved pattern of genetic correlations (elevated lipid, elevated sugars and reduced movement), then the response to selection is medically relevant and the genetic architecture should be investigated in depth.
Keywords: Drosophila, laboratory selection, starvation, body composition, movement
Introduction
In natural populations, reductions in available food resources are common. The physiology of adaptation to starvation is a subject of general interest for the study of evolution and its potential impacts on human health. A comparative study of starvation and desiccation survival in Drosophila melanogaster analyzed patterns of variation between species, between populations, and within populations (Hoffmann and Harshman 1999). In general, starvation resistance is correlated with relatively long life span, slow development, reduced egg production and large body size. These correlations are based on phenotypic data, but it is reasonable to hypothesize that these associations are evolutionarily derived and have a genetic basis. Experimental evolution of starvation resistance in the laboratory can yield insight into the mechanism of evolution within the laboratory (Rion and Kawecki 2007) and in natural populations.
A series of laboratory selection experiments for starvation resistance have been conducted using D. melanogaster with acute starvation as the selective agent. Six sets of lines have been directly selected for starvation resistance with some maintained at a selection intensity of approximately 50% mortality each generation similar to the selection pressures used within the present study. There has been a significant and often substantial direct response to selection in all six sets, indicating widespread genetic variation for this trait. The correlated (indirect) responses to selection are of particular interest as they can provide insight into mechanisms underlying the response to selection. The first laboratory evolution study of starvation resistance was based on a set of lines selected at the University of California at Irvine (Rose et al. 1992). These lines were investigated based on correlated responses to selection: life history, stress survival, phenotypic plasticity and physiology. Additional laboratory selection experiments for starvation resistance were based on two separate selection experiments (Harshman and Schmid 1998, Harshman et al. 1999). These lines were investigated in terms of life history changes, stress responses, phenotypic plasticity, physiology and enzyme activities. More recent starvation resistance selection experiments included three further sets of starvation selected lines (Bubliy and Loeschke 2005, Hoffmann et al. 2005, Baldal et al. 2006). In two of these studies, the evolution of correlated stress responses was the primary research focus (Bubliy and Loeschke 2005, Hoffmann et al. 2005). In the remaining selection experiment, a range of traits were measured as correlated responses to selection: metabolic rate, oxidative stress survival and lifespan (Baldal et al. 2006).
Correlated responses to selection for starvation resistance allow for investigation of the mechanisms underlying the response to selection and the relationship of natural populations to laboratory-based studies. Insight into the mechanisms underlying selection are of particular interest. In the case of starvation resistance, there are three physiological routes by which traits can evolve: resource accumulation, energy conservation and starvation tolerance (Rion and Kawecki 2007). Starvation selected lines are well-suited for the investigation of each general mechanism in terms of correlated responses to selection. The information generated from addressing these questions of the evolution of starvation resistance can be used to provide insight into adaptation to food deprivation in natural populations as well as investigate human health implications attributed to excess nutrient storage.
The present study is a relatively controlled laboratory selection experiment that measures direct and correlated responses to selection for starvation resistance. The level of selection on females and males was approximately 50% mortality during each generation for 15 generations. The number of breeders was held constant for each population, which increased consistency within the selection experiment. The direct response to selection and all correlated responses were measured on D. melanogaster derived from selection generation 15. A range of body composition measurements were obtained for both starved and unstarved adult flies of both sexes taken from the selected and control lines. This included soluble protein, glycogen, total sugar, trehalose and triglycerides. Using physical methods, the proportion of triacylglycerides, diacylglycerides and free fatty acids was estimated. Movement of flies from the selected and control lines was additionally measured. The breadth of body composition measurements obtained and the inclusion of movement is a unique feature of this study. The lean mass proportion of energetic compounds increased in the selected lines relative to the control lines and decreased as a function of starvation. Movement was reduced in selected line males as a correlated response to selection, with a corresponding trend observed in females.
Materials and Methods
Selection for starvation resistance
Establishment of the Base Population and Sub-Populations
The base population used for selection was derived from inbred lines produced from mated females collected at the Wolfskill Experimental Orchard in Yolo County, California. Following collection, the females were brought into the laboratory to establish isofemale lines. From the inbred lines, ten were chosen at random to establish an initial population for selection. The base population was established by intercrossing all of the lines in all possible combinations, including reciprocal crosses. Approximately 15,000 flies were used to establish the initial base population, which remained at an average population size of at least 10,000 flies for two years prior to beginning the present selection experiment. The population was maintained using an overlapping generation population regime in which 20 bottles were present in the cage, and each week the four oldest bottles were replaced. This overlapping population regime contributed to the maintenance of relatively long life span and increased stress resistance compared to a batch culture procedure for population maintenance in the laboratory (Hoffmann et al. 2001, Linnen et al. 2001).
Prior to initiation of selection, the base population was divided into eight subpopulations, which were used to produce four replicate lines for selection and four replicate control lines. Each subpopulation was maintained at 25°C and 12:12 L:D at a population size of approximately 4,000 randomly mating adults that were derived from vials seeded with a constant number of eggs (100 eggs per vial). This discreet population regime was continued for four generations prior to conducting selection.
Selection
Selection for starvation resistance was conducted on adult mated males and females that were approximately seven days old upon initiation of selection. For this purpose, 2,000 males and 2,000 females were obtained from each of the eight subpopulations to establish four selected and four control lines. Adult flies of one sex from each line were placed in population cages at the start of selection.
Males and females within the control lines were provided with six petri-plates containing food, while flies in the replicate selected lines received six plates containing solidified agar as a water source. Plates were changed every other day during the morning time-point to provide flies with adequate food or water. Relative humidity within the cages was maintained by placing the cages within a clear plastic bag containing a damp paper towel. The moistened paper towel was replaced in conjunction with removal and replacement of food or agar plates.
During the process of selection for starvation resistance, the response to selection was assessed in each individual cage by tabulating mortality levels at 12 hour intervals as dead flies were removed by aspiration. Upon reaching 50% mortality in the selected lines, flies were removed from the selected line cages containing solidified agar and control line flies were removed from cages with food. Flies were removed from the cages by aspiration following brief exposure to carbon dioxide. D. melanogaster from each line were placed in plastic bottles with fly medium at a density of approximately 150 flies per bottle. The flies were allowed a two day recovery period prior to breeding to produce the next generation. A standard number of 75 males and 75 females in six bottles from each of the replicate selected and control lines were allowed to mate. The bottles were used to harvest eggs and a standard number of eggs was collected per vial. From these eggs, a population of 2,000 males and 2,000 females for each individual replicate line were obtained for each generation of selection.
To determine the direct response to selection, flies from selected and control lines were placed in cages containing petri-plates with only solidified agar. The relative humidity was maintained within the cage by wet paper towels placed in the surrounding clear plastic bags. Using flies two generations removed from selection, the time to total mortality was determined for each line. Mortality levels were tabulated at 12 hour intervals until all the flies were dead. Movement, body composition and weight measurements were obtained from flies that were 3–6 generations removed from selection (relaxed selection).
Movement
The number of movements of individual D. melanogaster from the selected and control lines following either 32 hours of starvation or 32 hours in the presence of food were determined using individuals that were six generations removed from 15 generations of artificial selection for starvation resistance. Mated flies selected for movement analysis were exposed to the presence or absence of starvation at seven to eight days post eclosion prior to placement within glass capillary tubes (5 mm diameter by 65 mm length). Once inside the tubes, the flies were allowed to recover from ethyl ether exposure used to separate the sexes prior to experimentation. Food was provided at both ends of the capillary tube, with one end only partially covered to allow airflow.
The placement of each individual within the 64 total spaces between the two Drosophila activity monitors (TriKinetics) was determined using a statistical randomization scheme generated by SAS 9.2 (SAS, 2009). Implementation of statistical randomization of sample placement within each of the monitors was used to reduce or eliminate positional effects in the acquisition of data. Each capillary tube was centered with respect to the monitor and secured in place with a rubber band.
Detection of the number of movements of individual flies was accomplished using an infrared beam that bisected the glass capillary tube located within each position of the monitors. Each time an individual fly crossed the beam, the computer recorded the movement. The number of movements was quantified for each of the six total replicates obtained from each selected and control line exposed to the presence or absence of starvation. Movement was recorded in 10 minutes intervals for 48 hours. This allowed for the identification of variation within and among lines and treatments with respect to alterations in the light cycle. Environmental conditions were held constant during experimentation at 25°C with a 12:12 L/D cycle. Relative humidity was maintained using moistened cotton balls placed within clear plastic bags surrounding each monitor. The cotton balls were moistened with distilled water daily to prevent desiccation of the experimental subjects.
Body composition assays
Flies used in the body composition assays were allowed to mate prior to flash freezing the samples for extraction. Mated D melanogaster used in body composition analysis were approximately five to nine days old at the time of collection. Each sample of flies were homogenized in an Eppendorf tube using a 5/32 inch stainless steel grinding ball from OPS Diagnostics in a Talboys High Throughput Homogenizer (OPS Diagnostics) at 1,120 rpm for three minutes. Ten flies were homogenized for each of the three biological replicates of each sample. Following chemical treatment of homogenates for each assay, the reactions were read in a microtiter plate reader (VersaMax, Molecular Dynamics). For the assays described by VanHandel (1985), the optical density of the standards and samples were read at two wavelengths to increase the linearity of the readings.
Protein
Total soluble protein was determined using the Pierce BCA Protein Assay, which is a copper-based assay that is relatively resistant to interference from non-proteinaceous compounds in solution. The microtiter plate protocol was employed.
Glycogen and Total Sugar
Quantification of glycogen and total sugar was performed using methods described by Van Handel (1985). After processing the homogenate, the supernatant containing the sugars was decanted and evaporated down. Glycogen obtained from the sample remained with the fly tissue after precipitation with sodium sulfate. Anthrone reagent was added to both the sugars and the glycogen. Optical densities were determined in a spectrophotometer at wavelengths of 555 and 625 nm.
Trehalose
Trehalose, a common disaccharide in insects, was quantified using methodology described by Van Handel (1985). Homogenates were heated at 90°C for seven minutes, which resulted in hydrolysis of sucrose to glucose and fructose while leaving trehalose intact. Addition of sodium hydroxide heated at 90°C led to the destruction of anthrone reactivity to glucose and fructose, allowing for the quantification of trehalose. Optical densities were measured in a spectrophotometer at wavelengths of 555 and 625 nm.
Triglycerides
The triglyceride abundance was determined using the BioVision Triglyceride kit. Lipase was added to homogenates, resulting in the cleavage of triglycerides to free fatty acids and glycerol for quantification. The optical density was read at 570nm.
Dry Weight and Lean Mass
Dry weight and lean mass measurements were obtained. Flies that were flash frozen in liquid nitrogen from each of the replicate selected and control lines (males and females, starved and unstarved) were used for weight determination. Flies were dried by adding ten flies to an open 1.5ml microfuge tube placed in a 65°C drying oven overnight. Each of the ten flies was weighed using a Sartorius microbalance and the average dry weight was obtained. Lean masses were determined using flies subjected to a Bligh and Dyer (1959) lipid extraction conducted prior to drying the samples overnight. Standardization of the body composition measurements was conducted using lean masses.
Estimates of Lipid Class Proportions
Estimates of the proportion of lipid classes (polar lipids, triacylglyercerides, diacylglycerides and free fatty acids) were conducted at the Kansas Lipidomics Research Center (KLRC) at Kansas State University on lipid samples extracted at the University of Nebraska-Lincoln. Each extraction contained ten mated flies of the same sex at five to nine days post-eclosion. D. melanogaster were subject to a Bligh and Dyer (1959) lipid extraction, and were then dried under a gentle stream of nitrogen in 2ml glass tubes with Teflon caps prior to shipping to the KLRC. Samples and standards used routinely for the Drosophila were introduced into a tandem mass spectrometer by continuous infusion in solvent into the electrospray ionization source. The ion fragments of the lipids were separated in an electric field and sequentially scanned to identify lipids by class with peaks within individual lipid classes corresponding to different lipid species. Quantification of each lipid species occurred by comparison to internal standards.
The KLRC estimated the ratios of triacylglyercides (TAGs) to polar lipids, diacylglyderids (DAGs) to polar lipids and free fatty acids (FFAs) to polar lipids. For example, the normalized signal for total TAGs (nmol) was divided by the total nmol for polar lipids. This parameter was not an exact measurement of concentration, as the estimate of TAGs was not precise due to an inability to determine each of the 3 fatty acids present on each triglyceride molecule. The estimated ratios were not used for statistical analyses.
Statistical analyses
A mixed model analysis was conducted using SAS 9.2 software (SAS, 2009) to analyze body composition, weight and movement. Lines were used as a random factor nested within the selection and environmental treatments.
Results
Responses to selection and selection relaxation
Figure 1 illustrates the mortality curves of flies held in cages with solidified agar with an external source of water vapor to maintain relative humidity. After 15 generations of selection, the selected and control lines were highly differentiated in terms of their survival curves (Figure 1). There was a greater survival difference between selected and control line females than males. Selected lines had an extended period (approximate 100 hours) of very low levels of mortality. Thereafter, the selected line flies had a shallower mortality curve than control line flies, which was especially pronounced in females. Figure 2 displays the mortality curves after six generations of relaxed selection following selection generation 15. This assay was conducted to determine if starvation survival was lost as a function of relaxation while assays were being conducted. In males the loss of starvation resistance is present, but the change is smaller than in females.
Fig. 1.
Acute starvation survival of females and males from selected and control lines after 15 generations of selection.
Fig. 2.
Acute starvation survival six generations after relaxation of selection following selection generation 15.
Body composition
Table 1 presents the mean values (standard error) of body composition for selected and control lines under starved and unstarved conditions. All means are expressed in μg per mg of lean mass. Soluble protein in females was not significantly affected by selection (P = 0.1655) or starvation (P = 0.4601). In males, the level of soluble protein was not altered by selection (P = 0.4965), but decreased after starvation (P = 0.0063). There was no significant sex by treatment interaction for protein measurements obtained from either sex (female P = 0.6070, male P = 0.5891). For females, the effect of selection on glycogen was not statistically significant (P = 0.2686), whereas the effect of starvation was substantial (12.4% remaining, P < 0.0001). The same was observed for males: selection (P = 0.2686), starvation (15.4% remaining, P = 0.0002). In neither females nor males was there a significant sex by treatment interaction for glycogen abundance (female P = 0.9359, male P = 0.9922). Total sugars in females were significantly increased by selection (1.4 fold increase, P = 0.0291) and significantly decreased by starvation (47.1% remaining, P = 0.0005). In males, there was a statistically significant increase in total sugars by selection (1.3 fold increase, P = 0.0234) and a significant total sugar decrease after starvation (48.8% remaining, P < 0.0001). There was no statistically significant selection by starvation interactions for total sugars in females (P = 0.9359) or males (P = 0,6262) In general, selection increased total sugars and starvation decreased these sugars by approximately the same amount in females and males. For trehalose in females, selection had no significant effect (P = 0.1615), but starvation significantly decreased the level of this disaccharide (39.9% remaining, P = 0.0269). In males, selection resulted in a significant increase in trehalose (1.59 fold increase, P = 0.0269) and starvation resulted in a significant decrease (21.8% remaining, P < 0.0001). In neither females (P = 0.7981) nor males (P = 0.3946) was there a significant selection by starvation interaction. Generally, there was a trend for selection to increase trehalose, but this was statistically significant only in males, and in both sexes starvation reduced trehalose. For triglycerides measured in females, there was a significant increase following selection (1.68 fold increase, P = 0.0234) and a significant decrease after starvation (16.8% remaining, P < 0.0001). There was no significant selection by starvation interaction for triglycerides in females (P = 0.1152). In males, triglycerides increased markedly as a result of selection (2.33 fold, P = 0.0037) and decreased significantly after starvation (23.2% remaining, P = 0.0002). There was a statistically significant interaction between selection and starvation in males (P = 00422). Overall, selection increased the lean mass concentration of triglycerides and starvation decreased the level of triglycerides markedly in both sexes.
Table 1.
Table values are means (standard errors) for line types (selected and control) under treatments (starved and unstarved) expressed as μg/mg lean mass. PRO= protein, GLY= glycogen, TS= total sugar, TRE= trehalose and TG= triglyceride; F= female, M= male.
| Control
|
Selected
|
|||
|---|---|---|---|---|
| Unstarved | Starved | Unstarved | Starved | |
| (F) PRO | 795.1 (58.2) | 787.7 (57.82) | 764.8 (28.14) | 724.3 (5.97) |
| (M) PRO | 988.3 (42.1) | 1199.9 (17.51) | 1021.2 (60.21) | 1108 (27.70) |
| (F) GLY | 38.4 (1.94) | 3.5 (0.53) | 41.9 (5.58) | 6.5 (1.18) |
| (M) GLY | 28.7 (5.55) | 2.2 (0.78) | 34.1 (7.63) | 7.5 (0.57) |
| (F) TS | 21.3 (3.46) | 9.2 (0.80) | 29.0 (3.87) | 14.5 (1.56) |
| (M) TS | 27.0 (0.19) | 8.4 (1.46) | 28.2 (4.40) | 18.6 (0.94) |
| (F) TRE | 10.3 (1.02) | 3.2 (0.80) | 12.0 (2.07) | 5.7 (0.43) |
| (M) TRE | 6.4 (0.44) | 0.2 (1.21) | 7.6 (0.54) | 2.8 (0.97) |
| (F) TAG | 85.4 (7.64) | 13.0 (2.90) | 139.8 (23.30) | 24.9 (5.72) |
| (M) TAG | 28.7 (2.76) | 6.5 (1.60) | 66.5 (12.40) | 15.6 (0.56) |
Dry weight and lean mass
Table 2 presents the data for dry weight and lean mass measurements. In both sexes, there was a statistically significant increase in dry weight (female P = 0.0090, male P = 0.0240) and lean mass (female P = 0.0280, male P 0.0100) in response to selection and a decrease in dry weight (female P < 0.001, male P < 0.001) and lean mass (female P = 0.0280, male P < 0.0001) following starvation. There were no interactions between selection and treatment (dry weight: female P = 0.3120, male P = 01460; lean mass: female P = 0.5750, P 0.4630).
Table 2.
Variates are means (standard errors) for line types (selected and control) under treatments (starved and unstarved) expressed as μg. F= female, M= male.
| Control
|
Selected
|
|||
|---|---|---|---|---|
| Unstarved | Starved | Unstarved | Starved | |
| (F) Dry Weight | 461.8(13.00) | 359.8(13.50) | 533.5 (8.67) | 411.0 (7.62) |
| (M) Dry Weight | 293.0 (6.98) | 210.8 (4.72) | 336.5 (3.86) | 249.0 (5.18) |
| (F) Lean Mass | 321.0 (4.49) | 298.8 (9.66) | 349.0 (7.44) | 325.3 (6.20) |
| (M) Lean Mass | 213.8 (3.35) | 173.0 (4.10) | 231.3 (4.72) | 195.5 (3.80) |
Estimates of the proportion of lipids
Table 3 presents estimates of the proportion of neutral lipids (triacylglycerides, diacylgylcerides and free fatty acids) in relationship to the amount of polar lipids (phospholipids). The ratio is the mean (SE) normalized neutral lipid signal to the normalized polar lipid signal. For triglycerides, in the unstarved selected lines there is 4 fold increase (males) to 5 fold increase (females) in the level of triglyceride to polar lipids, whereas in unstarved control lines the ratios are relatively low (1.33 in males, 2.79 in females). Starvation decreased the TAG ratio approximately 33% in the selected lines and approximately 45% to 50% in the control lines. For diacylglyercides, selection did not increase the ratio in selected females compared to control lines, but in males the ratio of diacylglycerides to polar lipids increased approximately 36% as a result of selection. Starvation decreased the diacylglyceride level to a greater degree in the control lines than in the selected lines, which was additionally observed for the triglyceride ratio. Selection does not increase the free fatty acid ratios in the selected lines as compared to the control lines. The free fatty acid to polar lipid ratio tends to decrease after starvation, but the effect was not substantial and was not observed in selected males.
Table 3.
Lipid class ratios to the amount of polar lipids were estimated by the Kansas Lipidomics Research Center using mass spectroscopy evaluation of lipid peak responses and internal standards. PL= polar lipid, TAG= triglyceride, DAG= diglyceride and FFA= free fatty acids.
| Control
|
Selected
|
|||
|---|---|---|---|---|
| Unstarved | Starved | Unstarved | Starved | |
| (F) TAG/PL | 2.8 (0.14) | 1.6 (0.07) | 5.1 (0.58) | 3.1 (0.79) |
| (M) TAG/PL | 1.8 (0.15) | 0.89 (0.02) | 3.8 (0.69) | 2.5 (0.61) |
| (F) DAG/PL | 0.86 (0.14) | 0.36 (0.05) | 0.77 (0.18) | 0.44(0.10) |
| (M) DAG/PL | 0.39(0.49) | 0.24 (0.03) | 0.53 (0.13) | 0.39(0.07) |
| (F) FFA/PL | 0.31 (0.02) | 0.30 (0.04) | 0.26 (0.02) | 0.25(0.03) |
| (M) FFA/PL | 0.39 (0.06) | 0.32 (0.05) | 0.29 (0.04) | 0.36(0.08) |
Movement
Table 4 indicates the average movement for males from selected or control lines, the effect of starvation, the effect of light, and the level of interaction between main effects. There was a statistically significant effect of selection on males (P = 0.0006), which, in general, move less. No significant effect of starvation was recorded (P = 0.2351), but the trend was for starvation to increase movement. There was an effect of light such that males moved more when the lights were off (P < 0.0001). A statistically significant interaction (P = 0.0067) between light and selection was detected and this was due to a greater increase in movement in the control lines as compared to the selected lines when the lights were on. Overall for males, selected line flies moved less than the controls, starvation tended to increase movement, and the effect of light was to reduce movement, which was most pronounced in the control lines.
Table 4.
Male mean movement (S.E.) is the average number of times a laser beam in a tube is broken in a ten minute interval. For each line type (selected or control) movement is the mean of the values for each replicate line.
| Selection | Starved | Light | Mean (SE) | |
|---|---|---|---|---|
| Control | 5.7 (0.30) | |||
| Selected | 4.1 (0.30) | |||
| Light | Off | 5.7 (0.21) | ||
| Light | On | 4.0 (0.21) | ||
| Selection*Light | Control | Off | 6.8 (0.32) | |
| Selection*Light | Control | On | 4.6 (0.32) | |
| Selection*Light | Selected | Off | 4.7 (0.32) | |
| Selection*Light | Selected | On | 3.5 (0.32) | |
| Starved | Starved | 5.1 (0.28) | ||
| Starved | Unstarved | 4.6 (0.28) | ||
| Selection*Starved | Control | Starved | 5.9 (0.40) | |
| Selection*Starved | Control | Unstarved | 5.4 (0.40) | |
| Selection*Starved | Selected | Starved | 4.3 (0.40) | |
| Selection*Starved | Selected | Unstarved | 3.9 (0.40) | |
| Starved*Light | Starved | Off | 6.0 (0.30) | |
| Starved*Light | Starved | On | 4.2 (0.30) | |
| Starved*Light | Unstarved | Off | 5.5 (0.30) | |
| Starved*Light | Unstarved | On | 3.8 (0.30) |
Table 5 reports female movement data. In females, there was no statistically significant effect of selection for starvation resistance on movement (P = 0.4018), but the trend was reduced movement in the selected lines. There was no significant effect of starvation (P = 0.3252), but the trend was for reduced female movement. There was a statistically significant effect of light (P < 0.0001), which reduced movement. There were no significant interactions between factors, but there was a marginally significant interaction between selection and light (P = 0.0597). Light reduced movement to a greater extent in the control lines than in the selected lines. In general, there was an effect of selection which reduced movement in both sexes. Additionally, the presence of light reduced movement in both sexes, especially within the control lines.
Table 5.
Female mean movement (S.E.) is the average number of times a laser beam in a tube is broken in a ten minute interval. For each line type (selected or control) movement is the mean of the values for each replicate line.
| Selection | Starved | Light | Mean ( SE) | |
|---|---|---|---|---|
| Control | 3.5 (0.20) | |||
| Selected | 3.3 (0.20) | |||
| Light | Off | 3.7 (0.15) | ||
| Light | On | 3.1 (0.15) | ||
| Selection*Light | Control | Off | 4.0 (0.22) | |
| Selection*Light | Control | On | 3.1 (0.22) | |
| Selection*Light | Selected | Off | 3.5 (0.22) | |
| Selection*Light | Selected | On | 3.1 (0.22) | |
| Starved | Starved | 3.3 (0.18) | ||
| Starved | Unstarved | 3.5 (0.18) | ||
| Selection*Starved | Control | Starved | 3.3 (0.25) | |
| Selection*Starved | Control | Unstarved | 3.8 (0.26) | |
| Selection*Starved | Selected | Starved | 3.3 (0.26) | |
| Selection*Starved | Selected | Unstarved | 3.2 (0.25) | |
| Starved*Light | Starved | Off | 3.5 (0.20) | |
| Starved*Light | Starved | On | 3.1 (0.20) | |
| Starved*Light | Unstarved | Off | 3.9 (0.20) | |
| Starved*Light | Unstarved | On | 3.1 (0.20) |
Discussion
The present study reports data on two classes of measurements (body composition and movement) using Drosophila melanogaster selected for starvation resistance. Almost all compounds and macromolecules used for energy decreased as a result of starvation with a majority increasing as a result of selection. Movement tended to decrease as a result of selection for starvation resistance. Thus, two of the mechanisms proposed for adaptation to starvation, energy storage and energy conservation (Rion and Kawicki 2007) were observed in the present study. In general, there was a negative genetic correlation between storage of energy-containing compounds and decreased movement.
Body composition
Body composition measurements obtained from prior D. melanogaster artificial selection experiments for starvation have mainly focused on total lipid quantification, but additional measurement of glycogen and total carbohydrate have been conducted. Total lipid is often recorded as the difference between lean mass and dry weight within previous studies (Zwaan et al. 1995, Chippindale et al. 1996, Chippindale et al. 1998, Djawdan et al. 1998, Harshman et al. 1999, Baldal et al. 2006). These studies have established that selection for starvation resistance resulted in an increase in total lipid levels. In terms of a specific class of lipids, triacylglyceride levels were observed to increase in response to selection for D. melanogaster starvation resistance (Harshman et al. 1999). However, it is generally unknown how selection affects a range of different energetic compounds under unstarved and starved conditions in laboratory lines selected for starvation resistance. In the present study, numerous energy-containing molecules were measured including trigycerides, glycogen, total sugars and trehalose. The level of soluble protein was additionally measured because under extreme starvation conditions protein can be used for energy by some vertebrates. In addition, the ratio of triaclgylcerides, diacylglycerides and free fatty acids to polar lipids (phospholipids) was estimated.
A diversity of energy-containing compounds and macromolecules were altered in abundance as a response to selection and following a sublethal period of starvation in the present study. Triglycerides and total sugars increased in both sexes as a result of selection. In males, trehalose and diacylglyerides also increased as a response to selection. A broad spectrum of macromolecules and compounds decreased after starvation with the exception of free fatty acids in both sexes and protein in females. It is not apparent why free fatty acids are relatively invariant. In excess, fatty acids become increasingly toxic, which constrains maximum concentrations, whereas minimum levels could reflect the necessity of fatty acids to provide energy for life functions. Soluble protein decreased in starved males versus females. Male triglyceride stores are appreciably lower than in females. Males might reach a threshold of lipid depletion sooner than females resulting in males metabolizing proteins faster under starvation conditions.
The hypothesis of counter-impact selection (Harshman and Schmid 1998) was not entirely supported in the present study as the effect of selection was not to oppose the effect of starvation for each class of energetic compound or macromolecule. Glycogen was especially noticeable in this regard because it markedly decreased in response to starvation, but did not increase as a correlated response to selection for starvation resistance. The same phenotypic perturbation versus selection pattern for glycogen was observed in Harshman et al. (1999).
Movement
The effect of selection on movement is a novel measurement in the context of laboratory selection for starvation resistance. Movement has been measured in D. melanogaster selected for desiccation resistance, in which case there was no reduction in movement in the selected lines (Williams et al. 2004). In the present study, movement was statistically significantly reduced in males selected for starvation resistance and the trend was similar in females, which implies that flies are conserving energy for starvation survival by moving less. Food was present within the tubes used for the movement assay, but absent during starvation selection. Perhaps D. melanogaster selected for starvation resistance in this study have a general propensity to reduce movement.
Metabolic rate is related to the degree of movement and has been thought to be a basis for adaptation to stress. It has been argued that decreased metabolic rate contributes to adaptation to environmental stress (Hoffman and Parsons 1991). Studies conducted on a range of species have noted a reduction in metabolic rate in animals with high levels of stress resistance (for example, Lighton and Bartholomew 1988). However, when the non-metabolizing mass (chiefly lipid) was removed from data normalization, evolved elevated stress resistance was not associated with increased metabolic rate (Djawdan et al. 1997) in a Drosophila artificial selection experiment. Moreover, the per fly metabolic rate in starvation resistant lines was not lower in selected lines compared to control lines (Harshman and Schmid 1998) in D. melanogaster investigated in a laboratory selection experiment for starvation resistance. The implication of these studies is that energy conservation is not a mechanism for adaptation to starvation in laboratory-selected lines. However, the present study indicated that energy conservation may underlie a portion of the response to selection based on the observation of decreased movement in the selected lines.
The effect of starvation on movement is additionally of interest. In other studies, starvation resulted in increased movement (Rion and Kawicki 2007). In the present study, there was non-significant tendency for an increase in movement as a result of starvation. One aspect of the response to selection might be to block this propensity. Selection in an environment in which food cannot be found by moving elsewhere (a container) resulted in reduced movement under starvation conditions.
Starvation resistance and human biology
The joint occurrence of obesity, type 2 diabetes, hyperlipidemia and hypertension has increased in prevalence throughout the world resulting in increased mortality. This has led to the widespread view that concurrence of these diseases contributes to one syndrome, classified medically as the metabolic syndrome (Wilson and Grundy 2003). In general, D. melanogaster selection for laboratory starvation resistance results in the accumulation of increased concentrations of lipid. If it is observed that a series of correlated responses to selection parallels that of the metabolic syndrome, then this would provide genetic evidence that this syndrome is phylogenetically conserved.
In the present study, compounds with energetic value increased and movement decreased as a result of selection. Overall, the abundance of energy-containing compounds in the body was negatively correlated with movement. Elevated levels of lipid and sugar, and decreased movement were in line with measurements obtained in patients clinically diagnosed with metabolic syndrome. If the response to selection is based on an evolutionarily conserved pattern of genetic correlations (elevated lipid, elevated sugar, reduced movement), then the response to selection is medically relevant and the genetic architecture should be investigated in depth.
Conclusions
Within this study, selection for starvation resistance increased the abundance of a range of energy-containing compounds macromolecules used in metabolism. Decreased movement was a correlated response to selection for starvation resistance, presumably as a means to conserve energy for survival during periods of starvation. The negative genetic correlation between body composition traits and movement has potential human health implications if the genetic correlations are evolutionarily conserved.
Acknowledgments
Brian Becker and Courtney Peretto (University of Nebraska Lincoln) provided capable laboratory assistance in the course of this study. Tim Carr provided useful discussion and insight into metabolism. Sergey Nuzhdin (University of Southern California) provided the inbred lines used to start the base population used for selection. This research was supported by a grant from the NIH (RO1 DK074136).
References
- Baldal EA, Brakefield PM, Zwaan BJ. Multitrait evolution in lines of Drosophila melanogaster selected for increased starvation resistance: The role of metabolic rate and implications for the evolution of longevity. Evolution. 2006;60:1435–1444. [PubMed] [Google Scholar]
- Bligh EG, Dyer WJ. A rapid method for total lipid extraction and purification. Can J Biochem Physiol. 1959;37:911–917. doi: 10.1139/o59-099. [DOI] [PubMed] [Google Scholar]
- Bubliy OA, Loeschcke V. Correlated responses to selection for stress resistance and longevity in a laboratory population of Drosophila melanogaster. J Evol Biol. 2005;18:789–803. doi: 10.1111/j.1420-9101.2005.00928.x. [DOI] [PubMed] [Google Scholar]
- Chippindale AK, Chu TJF, Rose MR. Complex trade-offs and the evolution of starvation resistance in Drosophila melanogaster. Evolution. 1996;50:753–766. doi: 10.1111/j.1558-5646.1996.tb03885.x. [DOI] [PubMed] [Google Scholar]
- Chippindale AK, Gibbs AG, Sheik M, Yee KJ, Djawdan M, Bradley TJ, Rose MR. Resource acquisition and the evolution of stress resistance in Drosophila melanogaster. Evolution. 1998;52:1342–1352. doi: 10.1111/j.1558-5646.1998.tb02016.x. [DOI] [PubMed] [Google Scholar]
- Djawdan M, Chippidale AK, Rose MR, Bradley TJ. Metabolic reserves and evolved stress resistance in Drosophila melanogaster. Physiol Zool. 1998;71:584–594. doi: 10.1086/515963. [DOI] [PubMed] [Google Scholar]
- Djawdan M, Sugiyama TT, Schlaeger LK, Bradley TJ, Rose MR. Metabolic aspects of the trade-off between fecundity and longevity in Drosophila melanogaster. Physiol Zool. 1996;69:1176–1195. [Google Scholar]
- Harshman LG, Schmid JL. Evolution of starvation resistance in Drosophila melanogaster: aspects of metabolism and counter-impact selection. Evolution. 1998;52:1679–1685. doi: 10.1111/j.1558-5646.1998.tb02247.x. [DOI] [PubMed] [Google Scholar]
- Harshman LG, Hoffmann AA, Clark AG. Selection on starvation resistance in D. melanogaster: physiological correlates, enzyme activities and multiple stress responses. J Evol Biol. 1999a;12:370–379. [Google Scholar]
- Harshman LG, Moore KM, Sty MA, Magwire MM. Stress resistance and longevity in selected lines of Drosophila melanogaster. Neurobiol Aging. 1999b;20:521–529. doi: 10.1016/s0197-4580(99)00091-3. [DOI] [PubMed] [Google Scholar]
- Hoffmann AA, Harshman LG. Desiccation and starvation resistance in Drosophila: patterns of variation at the species, population and intrapopulation levels. Heredity. 1999;83:637–643. doi: 10.1046/j.1365-2540.1999.00649.x. [DOI] [PubMed] [Google Scholar]
- Hoffmann AA, Hallas R, Sinclair C, Partidge L. Rapid loss of stress resistance in Drosophila melanogaster under adaptation to laboratory culture. Evolution. 2001;55:436–438. doi: 10.1111/j.0014-3820.2001.tb01305.x. [DOI] [PubMed] [Google Scholar]
- Hoffmann AA, Hallas R, Anderson AR, Telonis-Scott M. Evidence for a robust sex-specific trade-off between cold resistance and starvation resistance in Drosophila melanogaster. J Evol Biol. 2005;6:643–657. doi: 10.1111/j.1420-9101.2004.00871.x. [DOI] [PubMed] [Google Scholar]
- Hoffman AA, Parsons PA. Evolutionary Genetics and Environmental Stress. Oxford University Press; 1991. p. 284. [Google Scholar]
- Lighton JRB, Bartholomew GA. Standard energy metabolism of a desert harvester and, Pogonomymrmex rugosus: effects of temperature, body mass, group size, and humidity. Proceedings of the National Academy of Sciences (USA) 1988;85:4767–4769. doi: 10.1073/pnas.85.13.4765. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Linnen C, Tatar M, Promislow D. Cultural artifact: a comparison of senescence in natural, laboratory-adapted and artificially selected lines of Drosophila melanogaster. Evolutionary Ecology Research. 2001;3:877–888. [Google Scholar]
- Rion S, Kawecki TJ. Evolutionary biology of starvation resistance: what have we learned from Drosophila. J Evol Biol. 2007;20:1655–1654. doi: 10.1111/j.1420-9101.2007.01405.x. [DOI] [PubMed] [Google Scholar]
- Rose MR, Sung LN, Park SU, Graves JL. Selection on stress resistance increases longevity in Drosophila melanogaster. Experimental Gerontology. 1992;27:241–250. doi: 10.1016/0531-5565(92)90048-5. [DOI] [PubMed] [Google Scholar]
- Service PM. Physiological mechanisms of increased stress resistance in Drosophila melanogaster selected for postponed senescence. Physiol Zool. 1987;60:321–326. [Google Scholar]
- Stearns SC, Ackermann M, Doebeli M, Kaiser M. Proceedings of the National Academy of Sciences of the United States of America. 7. Vol. 97. National Academy of Sciences; Experimental evolution of aging, growth, and reproduction in fruitflies; pp. 3309–3313. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Van Handel E. Rapid determination of glycogen and sugars in mosquitoes. Journal of American Mosquito Control Association. 1985;1:299–301. [PubMed] [Google Scholar]
- Wilson PWF, Grundy SM. The metabolic syndrome: practical guide to origins and treatment: part 1. Circulation. 2003;108:1422–1425. doi: 10.1161/01.CIR.0000089505.34741.E5. [DOI] [PubMed] [Google Scholar]
- Zwaan B, Bulsma R, Hoekstra RF. Direct selection on life span in rosophila melanogaster. Evolution. 1995;49:649–659. doi: 10.1111/j.1558-5646.1995.tb02301.x. [DOI] [PubMed] [Google Scholar]