Rapid changes in desiccation resistance in Drosophila melanogaster are facilitated by changes in cuticular permeability

Journal of Insect Physiology 56 (2010) 2006–2012

Contents lists available at ScienceDirect

Journal of Insect Physiology journal homepage: www.elsevier.com/locate/jinsphys

Rapid changes in desiccation resistance in Drosophila melanogaster are facilitated by changes in cuticular permeability Aimee L. Bazinet, Katie E. Marshall, Heath A. MacMillan, Caroline M. Williams, Brent J. Sinclair * Department of Biology, The University of Western Ontario, London, ON N6A 5B7, Canada

A R T I C L E I N F O

A B S T R A C T

Article history: Received 20 August 2010 Received in revised form 13 September 2010 Accepted 14 September 2010

Insects can improve their desiccation resistance by one or more of (1) increasing their water content; (2) decreasing water loss rate; or (3) increasing the amount of water able to be lost before death. Female Drosophila melanogaster have previously been reported to increase their resistance to desiccation after a desiccation pre-treatment and recovery, but the mechanism of this increased desiccation resistance has not been explored. We show that female, but not male adult D. melanogaster increased their resistance to desiccation after 1 h of recovery from a 3 to 4.5 h pre-treatment that depletes them of 10% of their water content. The pre-treatment did not result in an increase in water content after recovery, and there is a slight increase in water content at death in pre-treated females (but no change in males), suggesting that the amount of water loss tolerated is not improved. Metabolic rate, measured on individual flies with flow-through respirometry, did not change with pre-treatment. However, a desiccation pre-treatment did result in a reduction in water loss rate, and further investigation indicated that a change in cuticular water loss rate accounted for this decrease. Thus, the observed increase in desiccation resistance appears to be based on a change in cuticular permeability. However, physiological changes in response to the desiccation pre-treatment were similar in male and female, which therefore does not account for the difference in rapid desiccation hardening between the sexes. We speculate that sex differences in fuel use during desiccation may account for the discrepancy. ß 2010 Elsevier Ltd. All rights reserved.

Keywords: Desiccation tolerance Cuticular water loss Respiratory water loss Phenotypic plasticity

1. Introduction Desiccation is a significant stress for terrestrial animals, and the success of insects in terrestrial environments has been ascribed, at least in part, to their ability to effectively tolerate desiccation (Chown and Nicolson, 2004). Desiccation stress continues to be a primary factor shaping insect distribution and behaviour, and an inability to respond to desiccation stress may compound the negative consequences of climate change for some insects. For example, Drosophila birchii (Diptera: Drosophilidae) appears to be restricted to rainforest fragments by its inability to survive desiccation, but the limited capacity of some populations to respond to selection suggest that drying of that habitat may lead to extinctions (Hoffmann et al., 2003). Similarly, cycles of desiccation stress can reduce metabolic reserves and egg-laying capacity in the mosquito Culex pipiens (Diptera: Culicidae), and thus impact population dynamics (Benoit et al., 2010).

* Corresponding author at: Department of Biology, The University of Western Ontario, 1151 Richmond St N., London, ON N6A 5B7, Canada. Tel.: +1 519 661 2111x83138; fax: +1 519 661 3935. E-mail address: [email protected] (B.J. Sinclair). 0022-1910/$ – see front matter ß 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.jinsphys.2010.09.002

Through inter-specific comparisons and selection experiments, it is apparent that increased desiccation resistance in insects can be achieved through some combination of (1) reducing the rate of water loss, (2) increasing the bulk amount of water available to lose, and(/ or) (3) increasing the amount of water loss that can be tolerated prior to death (Gibbs et al., 1997). Water loss rate may be reduced by altering respiratory water loss, largely through controlling respiratory patterns (Chown, 2002). For example, insects from more arid environments appear to be more likely to utilise discontinuous gas exchange, which would be expected to decrease respiratory water loss (White et al., 2007). Variation in cuticular water loss (which constitutes the bulk of water loss in insects; Chown and Nicolson, 2004) appears to be driven by the permeability of the epicuticular hydrocarbons (Gibbs, 2002a). Bulk water may be increased by simply increasing haemolymph volume (Hadley, 1994), but in Drosophila, bulk water is increased primarily by accumulating glycogen, which not only provides metabolic water when metabolised, but also has 3–5 times its mass in water hydrogen-bound to the molecule (Gibbs, 2002b). Finally, there is considerable variation in the amount of water loss that insects can tolerate (Hadley, 1994). The mechanisms determining water loss tolerance are not wellunderstood, although the disaccharide trehalose has been implicated in cellular protection of organisms that survive losing very large quantities of water (Watanabe, 2006).

A.L. Bazinet et al. / Journal of Insect Physiology 56 (2010) 2006–2012

To date, most comparisons of the mechanisms of desiccation resistance have been made of inter-specific (e.g. Gibbs and Matzkin, 2001), or population-level (e.g. Gibbs et al., 1997) differences. During desiccation, water loss rates may be nonlinear (e.g. Benoit et al., 2007), and insects may also have plastic responses to desiccation. For example, mating increases the desiccation resistance of females of two desert Drosophila species (Knowles et al., 2004), dragonflies alter spiracular patterns as they are dehydrated (Miller, 1964) and Collembola actively synthesise low molecular weight carbohydrates to increase atmospheric water vapour absorption in response to a mild desiccation stress (Sjursen et al., 2001). A prior desiccation event allowed female Drosophila of several species (including D. melanogaster) to survive desiccating conditions for longer than their non-pre-treated counterparts (Hoffmann, 1990, 1991). Acclimation to desiccation has been explored in some other species (e.g. Bayley and Holmstrup, 1999; Hayward et al., 2007), but generally over longer time periods and with a pre-treatment involving slow desiccation at relatively high relative humidities. In these cases, synthesis of protective compounds (e.g. sugars and polyols) appears to underlie an increase in desiccation tolerance. The mechanisms underlying ‘rapid desiccation hardening’ (RDH) observed in Drosophila are unclear. Given the plethora of genetic tools in Drosophila, identifying the mechanisms of RDH may allow the identification of the underlying pathways in this and other insects, and to determine the underpinnings of constraints to plasticity in insect responses to environments with novel or changing desiccation conditions. Here, we explore the physiological mechanisms of RDH in male and female D. melanogaster. We first confirm the presence of the response, and then compare metabolic rate, cuticular and respiratory water loss rate, water content and water content at death among pre-treated (RDH) and non-pre-treated flies with the aim of determining the relative importance of each of these factors in the RDH response.

2007

2.3. Water and glycogen content Water content of individual flies was determined gravimetrically using a 0.5 mg MX5 microbalance (Mettler Toledo, Columbus, OH, USA) by weighing the flies (wet mass), drying in an oven (at 60 8C for at least 24 h), and then weighing them again (dry mass). The difference between wet and dry mass gave the mass of water in each fly. Live flies were killed and immobilised by briefly plunging them (in sealed microcentrifuge tubes) into liquid nitrogen vapour. The flies were allowed to equilibrate to room temperature (ca. 10 min) before initial weighing. Glycogen content of individual flies was determined using methods adapted from Marshall and Sinclair (2010). Briefly, individual flies were homogenized in 1.7 mL microcentrifuge tubes containing 60 mL 0.05% Tween 20 solution and 1.0 mm glass beads using a Next Advance Bullet Blender (Next Advance, Averill Park, NY, USA) for 5 min at setting 8. Another 40 mL of Tween 20 solution was added and homogenized for 1 min at setting 1. The samples were then centrifuged for 1 min at 16,000  g and the supernatant removed and frozen at 80 8C until used for enzymatic assays. Triplicates (10 mL) from each sample were loaded onto 96-well microplates along with 90 mL of glucose assay reagent (Sigma– Aldrich, St. Louis, MO, USA), left for 5 min at room temperature, and then absorbance of each well was read on a spectrophotometer (SpectraMAX 340 pc, Sunnyvale, CA, USA) at 340 nm to determine the amount of free glucose. Once read, 10 mL of a 0.8 mg mL 1 solution of Rhizopus amyloglucosidase (Sigma–Aldrich, St. Louis, MO, USA) was added to each well and left at room temperature overnight to convert glycogen to glucose. After 12 h, the plate was read again at 340 nm, and glycogen (in glucose units) calculated as the difference between glucose concentration before and after amyloglucosidase digestion. Carbohydrate concentrations were determined using standard curves of known glucose concentrations. 2.4. Water loss rate

2. Methods 2.1. Insect rearing and maintenance A Drosophila melanogaster population was derived from isofemale lines collected from the London, Ontario, area in 2007 (Marshall and Sinclair, 2010). Flies were reared on a 14-day cycle at 21.5 8C (13:11 L:D, 50% RH) on a banana-yeast medium in 35 mL vials with around 70–100 individuals per vial, with mass mating and egg collection each generation conducted in 3.8 l plastic population cages. Virgin adult flies to be used in experiments were sexed and sorted individually under CO2 within 12 h of eclosion. Flies were given 48 h to recover from CO2 anaesthesia (Nilson et al., 2006) and used in experiments after another 48 h, so all experiments were conducted on ca. 4-day-old individuals. All experiments were conducted in the laboratory at an average temperature of 21 8C. 2.2. Desiccation Flies were desiccated individually as per Gibbs et al. (1997). Briefly, individual flies were transferred without CO2 to empty 35 mL vials and restricted to the lower half of the vial with a foam stopper. Approximately 2 g of silica gel (4–10 mesh; J.T. Baker, Phillipsburg, NJ, USA) was added on top of the stopper and the vial was covered with parafilm (Pechiney Plastic Packaging, Menasha, WI, USA). Monitoring with an iButton hygrochron (Maxim Integrated Products, Sunnyvale, CA, USA) indicated that relative humidity drops to 5% within 90 min of the closure of the vial.

The rates of CO2 production and water loss under desiccating conditions were measured using flow-through respirometry using a method modified from Williams et al. (2010). Briefly, individual flies were placed into 4 cm3 glass chambers in a PELT5 temperature controlled cabinet (0.1 8C; Sable Systems International, Las Vegas, USA) at 30 8C. Dry, CO2-free air was pumped through the chamber at 25 mL min 1 to a LI7000 CO2/H2O infra-red gas analyser (Li-Cor, Lincoln, NE, USA), and data acquired using Expedata software (Sable Systems International). Baselines were recorded on an empty chamber at the beginning and end of each run to provide a zero measurement to correct for instrument drift. Flies were given an acclimation period of 40 min and then data were recorded every second for 40 min. Each fly was weighed before and after each run to ensure gravimetric water loss was consistent with flow-through rates. Data were means extracted from a 5-min section with no evidence of activity or excretion, and showing a relatively high variance in CO2 production to allow partition of respiratory and cuticular water loss rates. Cuticular and respiratory water loss rates were estimated using the method of Gibbs and Johnson (2004). Briefly, water loss rate was regressed against CO2 production rate to obtain a slope (incremental water loss cost of gas exchange) and intercept (cuticular water loss rate). Where the slope of the regression was not significantly different from zero, these two data points (one control and one pre-treated male) were excluded from analysis. Respiratory water loss rate across the period was taken as the difference between total water loss and cuticular water loss rates.

2008

A.L. Bazinet et al. / Journal of Insect Physiology 56 (2010) 2006–2012

2.5. Determination of pre-treatment and rapid desiccation hardening To determine a pre-treatment time, individual flies were desiccated, and groups of 15 removed at hourly intervals and water content measured. A pre-treatment time that resulted in a loss of 10% total water content was used for subsequent experiments. This pre-treatment time was 3.0 h for male flies, and 4.5 h for female flies. To control for a potential effect of starvation, 100 males and 100 females were subjected to the same experimental methods as for determining the pre-treatment conditions, but were given nonnutritive agar instead of an empty vial and were not exposed to the silica gel desiccant. Ten flies were removed every hour for 8 h and water content determined gravimetrically. To ensure that pre-desiccated flies were able to fully regain lost body water, 60 males and 60 females were desiccated for the predetermined pre-treatment time and then transferred without CO2 to a vial with food to recover. Each hour after transfer, groups of 10 flies were removed and water content was measured gravimetrically. The minimum time taken for the flies to recover to 100% of their initial water content was used for subsequent recovery treatments. To determine whether pre-treatment improved desiccation resistance, 40 flies of each sex were subjected to pre-treatment and recovery conditions. These pre-treated flies, and a control group of 40 males and 40 females that were not pre-treated, were then returned to desiccating conditions. Mortality (an inability for the flies to right themselves when disturbed) was scored every 60 min for the first 5 h and then every 30 min for the subsequent 13 h. If a fly was dead during a check in the desiccation survival assay, the fly was removed and water content determined gravimetrically (water content at death).

after 1 h (used as subsequent recovery time). Starvation did not lead to a change in dry mass over time (F1,234 = 0.42, p = 0.516), and post hoc analysis showed that water content during starvation did not change with time in males (F11,106 = 0.9, p = 0.546) and only between a few time points in females (F11,106 = 4.17, p < 0.001, Fig. 1), largely because of variation in body size among groups. 3.2. Effect of desiccation pre-treatment on desiccation resistance The best model for survival of desiccation retained all terms (sex, pre-treatment and their interaction) and had a log-logistic error distribution (Statistics are given in Table 1). Pre-treatment significantly improved desiccation survival by pre-treated (calculated mean survival time of 15.9  0.4 h) over untreated female flies (13.1  0.3 h; Fig. 2). By contrast, male flies survived desiccation for significantly less time than females (Fig. 2 and Table 1). There was no significant difference in desiccation survival time between male flies that had been pre-treated (9.6  0.2 h) and those that has not (9.0  0.2 h). A total of 42 flies (14 females and 28 males) out of 219 survived beyond the time course of the observations, however calculating survival times without these data did not alter the conclusions. The best-fitting model for each sex differed, with male survival curves best described by either a Weibull or extreme value error distribution, and female survival curves best described by a loglogistic error distribution (Tables 2 and 3).

[(Fig._1)TD$IG]

2.6. Statistical analysis Analyses of covariance were used to compare water and glycogen content, water content at death, water loss and CO2 production rates among sexes and treatments using SAS (v. 9.1, SAS Institute, Cary, NC, USA). Analysis of variance was used to compare dry mass among sexes and treatments, also using SAS. Glycogen content was square-root transformed prior to analysis to improve normality. Tukey’s HSD was used for all post hoc comparisons. For ease of visualisation, data analysed by ANCOVA are presented as per-fly values, uncorrected for body mass. Mean  SEM are presented throughout. Accelerated failure time (AFT) models (Swindell, 2009) built in R (R Development Core Team, 2010) using the survival package (http://CRAN.R-project.org/package=survival) were used to determine the effects of sex and pre-treatment on desiccation survival times. Models using exponential, extreme, Gaussian, logistic, Weibull, or log-logistic error distributions were compared, and the best-fitting model was chosen using Akaike’s Information Criterion (AIC). Then a model simplification approach was used, beginning with the saturated model including sex, pre-treatment and their interaction (Crawley, 2005). Model terms were retained on the basis of p-values. To compare survival curves between sexes, separate models were fit for male and female data. The best-fitting model for each sex was chosen using AIC. 3. Results 3.1. Recovery from pre-treatment The pre-treatment times chosen to give approximately 10% water loss were 3.0 and 4.5 h for males and females. After pretreatment, flies given access to food regained initial water content

Fig. 1. Changes in mean (SEM) water content (A) and dry mass (B) in adult male (open symbols) and female (filled symbols) Drosophila melanogaster during starvation (during which the flies had access only to non-nutritive agar for water). Although perfly values are presented for water content, analysis was conducted with mass as a covariate. There were no significant differences in dry mass between times within sexes, or in water content between times for male flies. Letters that differ indicate significant differences in water content of females among time points (Tukey’s HSD, p < 0.05), there were no differences in water content of males, or dry mass of either sex. Error bars that are not visible are obscured by the symbol.

A.L. Bazinet et al. / Journal of Insect Physiology 56 (2010) 2006–2012

2009

Table 1 Model statistics for saturated (all terms included) accelerated failure time models of desiccation survival with and without a desiccation pre-treatment in adult male and female D. melanogaster. The best-fit model (with lowest AIC) is in bold. Scale and all coefficients are indicated as mean  SE, with p-values indicated as *p < 0.05, **p < 0.01, ***p < 0.001; DF indicates degrees of freedom for the model. N = 219 for each model. Coefficients for sex and pre-treatment represent the change in survival relative to female and untreated flies, respectively. Error distribution

x2

DF

AIC

Exponential Extreme Gaussian Logistic Weibull Log-logistic

8.45 208.77 195.57 207.06 212.14 209.85

3 3 3 3 3 3

1301.6 855.4 827.0 825.2 811.2 802.6

Coefficients Intercept

[(Fig._2)TD$IG]

Sex

2.61  0.13*** 14.70  0.25*** 13.32  0.26 *** 13.17  0.27 *** 2.67  0.02*** 2.57  0.02***

Fig. 2. Desiccation survival of Drosophila melanogaster adults with (filled symbols) and without (open symbols) a desiccation pre-treatment. Black line and symbols: male, grey line and symbols, female. N = 40 flies per treatment/sex combination. The curves differ significantly between pre-treated and control females, but not males; see text for statistics.

3.3. Mechanisms of rapid desiccation hardening Water content after pre-treatment and recovery was higher in female than male flies (F1,35 = 23.2, p < 0.001), but did not differ from control levels in either sex (F1,35 = 0.05, p = 0.826, Fig. 3). Pretreatment and recovery resulted in a decrease in total glycogen stores in both male and female flies (F1,35 = 7.86, p = 0.008, Fig. 3), but glycogen content did not differ between males and females when body mass was taken into account (F1,35 = 0.29, p = 0.591). Pre-treatment resulted in a consistent (but non-significant) decrease in mean dry mass (0.003 mg in males, 0.034 mg in females; F1,36 = 2.72, p = 0.108). Dry mass was slightly, but nonsignificantly, lower (by ca. 0.009 mg in males, 0.034 mg in females) in pre-treated flies by the same after pre-treatment (F1,39 = 2.72, p = 0.108) and at death (Fig. 4; F1,67 = 3.22, p = 0.77). Water content at death did not differ in absolute terms between treatments (Fig. 4), but when using dry mass as a covariate, water content per dry mass at death was slightly and significantly higher (i.e. reduced

Pre-treatment

0.17  0.20 5.09  0.37*** 4.16  0.39*** 4.06  0.37*** 0.41  0.03*** 0.37  0.03***

0.42  0.20* 2.40  0.36*** 2.59  0.38*** 2.83  0.39*** 0.17  0.03*** 0.20  0.03***

Sex  pre-treatment 0.26  0.30 1.87  0.54*** 2.01  0.55*** 2.22  0.53*** 0.12  0.04** 0.13  0.04**

tolerance to desiccation) in female flies after pre-treatment (F1,66 = 11.55, p = 0.001). The rate of carbon dioxide production did not differ between sexes or pre-treatments (sex: F1,37 = 2.74, p = 0.106; treatment: F1,37 = 0.03, p = 0.865; treatment  sex: F1,37 = 0.21, p = 0.652; Fig. 5). Total water loss rate was significantly lower in pre-treated flies than control flies of both sexes (F1,37 = 6.99, p = 0.012), but did not differ between sexes (F1,37 = 0.01, p = 0.924), nor was there a treatment  sex interaction (F1,37 = 0.06, p = 0.811; Fig. 5). Cuticular water loss rate was significantly lower in pre-treated flies (F1,35 = 10.23, p = 0.003), but did not differ between sexes (F1,35 = 0.43, p = 0.517), and there was no treatment  sex interaction (F1,35 = 0.38, p = 0.544; Fig. 5). Lastly, there was no difference in respiratory water loss rate among sexes (F1,35 = 2.86, p = 0.100) or treatments (F1,35 = 0.34, p = 0.561), and there was no sex  treatment interaction (F1,35 = 1.65, p = 0.270; Fig. 5). 4. Discussion D. melanogaster exposed to a mild desiccation stress are able to rapidly reduce their rate of water loss, and this serves to improve desiccation resistance of adult females, but not males. This rapid desiccation hardening was previously described for female D. melanogaster (Hoffmann, 1990), although the previous study did not report results for male flies, and did not explore the mechanisms underlying this increase in resistance. Desiccation resistance in insects can be increased by a combination of increasing the initial water content, reducing the rate of water loss, or increasing the amount of water loss tolerated (Gibbs et al., 2003). Increased desiccation resistance in artificially selected lines of D. melanogaster have resulted in both decreased water loss and an increase in stored water, mediated by increased glycogen stores (Gibbs et al., 1997). We found that the desiccation pre-treatment did not alter the quantity of bulk water and reduced glycogen stores. The desiccation pre-treatment we used is accompanied by starvation, which is the likely cause of the glycogen depletion. The amount of water loss tolerated by the flies did not change significantly with pre-treatment, although when taking dry mass into account, was slightly decreased in pre-treated female flies. Water loss rate was reduced significantly with pre-

Table 2 Model statistics for accelerated failure time models of desiccation survival with and without a desiccation pre-treatment in adult male D. melanogaster. The best-fit model (with lowest AIC) is in bold. Scale and all coefficients are indicated as mean  SE, with p-values indicated as *p < 0.05, ***p < 0.001; DF indicates degrees of freedom for the model. N = 107 for each model. The coefficient for pre-treatment represents the change in survival relative to untreated flies. Error distribution

x2

DF

AIC

Intercept

Pre-treatment

Exponential Extreme Gaussian Logistic Weibull Log-logistic

0.47 208.77 3.73 4.87 2.94 4.64

1 1 1 1 1 1

558.0 322.1 327.2 324.8 322.91 328.41

2.43  0.15*** 9.58  0.15*** 9.05  0.19 *** 9.06  0.18 *** 2.26  0.02*** 2.20  0.02***

0.16  0.23 0.38  0.22 0.53  0.27 (p = 0.051) 0.57  0.39* 0.04  0.03 0.07  0.03*

2010

A.L. Bazinet et al. / Journal of Insect Physiology 56 (2010) 2006–2012

Table 3 Model statistics for accelerated failure time models of desiccation survival with and without a desiccation pre-treatment in adult female D. melanogaster. The best-fit model (with lowest AIC) is in bold. Scale and all coefficients are indicated as mean  SE, with p-values indicated as *p < 0.05, ***p < 0.0001; DF indicates degrees of freedom for the model. N = 112 for each model. The coefficient for pre-treatment represents the change in survival relative to untreated flies. Error distribution

x2

DF

AIC

Intercept

Pre-treatment

Exponential Extreme Gaussian Logistic Weibull Log-logistic

4.21 31.9 30.53 32.31 30.75 31.97

1 1 1 1 1 1

745.6 499.8 482.5 485.0 488.6 479.5

2.61  0.13*** 14.44  0.31*** 13.32  0.31*** 13.20  0.31*** 2.66  0.02*** 2.57  0.02***

0.42  0.20* 2.64  0.46*** 2.67  0.46*** 2.81  0.46*** 0.18  0.03*** 0.20  0.03***

treatment, and seems to be the primary mechanism for increased desiccation resistance in pre-treated flies. We were able to partition water loss between respiratory and cuticular water loss. Respiratory water loss and CO2 production rates did not change with pre-treatment, which suggests that there is no modulation of metabolic rate or spiracular opening that can account for the reduced water loss rate. This contrasts with observations among Drosophila species (including D. melanogaster), where metabolic rate is reduced with increasing desiccation resistance, and water loss rate is correlated to metabolic rate

[(Fig._3)TD$IG]

(Gibbs et al., 2003). However, cuticular water loss rate (estimated according to Gibbs and Johnson, 2004) did decrease significantly in both males and females following a desiccation pre-treatment. Much of the waterproofing of insect cuticle is provided by cuticular hydrocarbons (Chown and Nicolson, 2004). Gibbs et al. (1997) found that although the absolute quantity of cuticular lipids did not change with desiccation selection in D. melanogaster, desiccation-selected flies had increased quantities of longer lipids (27 and 29 carbons) compared to control lines. Differences in cuticular hydrocarbons are usually compared among insects on an evolutionary timescale (e.g. Gibbs et al., 1997), although cuticular hydrocarbons used to signal behavioural or reproductive status can rapidly change in Drosophila (e.g. Everaerts et al., 2010), and Gibbs et al. (1998) show that cuticular hydrocarbon profiles change

[(Fig._4)TD$IG]

Fig. 3. Mean (SEM) total water (A) and carbohydrate (B) content of male and female Drosophila melanogaster adults without pre-treatment (closed bars) and after exposure to (and recovery from) a desiccation pre-treatment (open bars). Pre-treatment did not alter water content of flies, but did result in a significant decrease in carbohydrate content in flies of both sexes (indicated by asterisks; see text for statistics).

Fig. 4. Mean (SEM) dry mass at death (A) and water content at death (B) of male and female Drosophila melanogaster adults with (open bars) and without (filled bars) a desiccation pre-treatment. When accounting for dry mass, pre-treatment significantly increased water content at death in both sexes although dry mass at death did not differ significantly between the treatments (see text for statistics).

[(Fig._5)TD$IG]

A.L. Bazinet et al. / Journal of Insect Physiology 56 (2010) 2006–2012

2011

Fig. 5. Water loss rate. Mean (SEM) total water loss rate (A), cuticular water loss rate (B), Respiratory water loss rate (C) and carbon dioxide production rate (D) by Drosophila melanogaster adults at 29 8C with (open bars) and without (filled bars) a desiccation pre-treatment. Total water loss rate and cuticular water loss rate decreased significantly with pre-treatment, but pre-treatment did not alter the rates of respiratory water loss or CO2 production. Note that values presented are per fly, but statistical analysis took individual masses into account.

rapidly during maturation of adult D. mojavensis. The regulation of cuticular permeability is likely hormonal (Treherne and Willmer, 1975); and the observation that the gene encoding Fatty acyl CoA transferase is upregulated in association with reduced water loss rate of diapausing eggs of Aedes aegypti (Urbanski et al., 2010) begins to hint at the underlying molecular mechanisms. It is possible that a rapid change in cuticular hydrocarbon profile, perhaps regulated by stress–response pathways, could account for the change in cuticular water loss rate we observed. The amount of water that insects can tolerate losing varies considerably among taxa and habitats (Gibbs and Matzkin, 2001; Hadley, 1994). However, although several populations of D. melanogaster have been successfully artificially selected for increased desiccation resistance, none of these experiments have changed the amount of water loss tolerated (e.g. Gibbs et al., 1997). The amount of water loss insects can tolerate is likely related, at least in part, to the haemolymph volume and the ability of the insect to sequester or excrete ions (Folk and Bradley, 2003), but the cause of death from water loss is not known. In our study, the absolute water content at death did not change with pretreatment. However, pre-treated flies (particularly females) had lost a portion of their body mass (largely as a result of glycogen depletion) during the pre-treatment, and thus female flies had greater water content at death when body mass was accounted for. Thus, our results are consistent with other observations that

suggest that water content at death is not plastic in D. melanogaster. A desiccation pre-treatment resulted in a significant decrease in cuticular water loss rate in adult D. melanogaster. To allow accurate measurement of individual water loss and CO2 production rates (necessary to partition cuticular and respiratory water loss), we conducted respirometry at 29 8C, ca. 7 8C warmer than the temperature at which the other assays were done. The increased temperature will increase water loss rates and particularly metabolic rate, which likely means that we have overestimated the respiratory water loss rate. The respiratory water rate we observed was 16–23% of total water loss rate, which is at the upper end of the range of respiratory water loss rates reported elsewhere (Chown, 2002). D. melanogaster are regularly found in habitats where the temperature exceeds 29 8C, so our measurements of cuticular water loss rate are unlikely to have been affected by temperature-related transitions in cuticular permeability. Using our measurements of water content, water content at death and (with some caution) water loss rate, it is possible to develop a simple water budget, following Gibbs et al. (1997). Our simple budget predicts that pre-treatment will increase time to death by 27% in males, and 18% in female flies. The high temperatures used for respirometry make this value difficult to compare to measured values, but it is important to note that the key prediction is an improvement in survival of both male and

2012

A.L. Bazinet et al. / Journal of Insect Physiology 56 (2010) 2006–2012

female flies, whereas we observed a shift in the survival curve only for females. The difference in Lt50 (time for 50% of flies to be killed) in female flies (from 12.5 to 16 h = 28%) is broadly consistent with the budget, whereas the male flies (from 9 to 9.5 h = 6%) showed considerably less improvement in survival than would be predicted from the reduced water loss rate. The survival curves of male and female flies are very different shapes – concave in females and convex in males (Fig. 2, Tables 2 and 3), and it is possible that this difference reflects different underlying processes determining mortality. Male flies are considerably smaller, and begin the desiccation process with less water than females (reflected in the very different lengths of time for which males and females tolerate desiccation). In addition, male flies have fewer carbohydrate reserves than females, even after depletion due to the pre-treatment (Fig. 3). It is possible that the inability to increase desiccation resistance after pre-treatment may be associated with either the depletion of those carbohydrate reserves, or sex differences in the way in which energy reserves are utilised and thereby liberate metabolic water. Future work could include an investigation of energy use during desiccation, and how it differs between males and females. The role of phenotypic plasticity (and the limits thereof) is extremely important for understanding organisms’ responses to their abiotic environment, and therefore to predict responses to anthropogenic and other changes (Angilletta, 2009; Chown and Gaston, 2008). In particular, desiccation resistance is an important determinant of insect distribution, so rapid, inducible changes like those displayed here are of particular relevance to predicting larger-scale insect responses to the environment. Hoffmann (1991) demonstrated rapid desiccation hardening in females of a further four species of Drosophila, and it would be particularly valuable to determine the taxonomic scope of this response (both within and outside Drosophila) and whether the mechanism (a rapid change in water loss rate) is conserved. We suggest that the rapid, inducible, changes in water loss rate in D. melanogaster that we observe may provide a useful system in which the mechanisms underlying variation in desiccation resistance in insects may be readily investigated under highly controlled conditions. Acknowledgements Thanks to Joshua Farhi, Greg Watkinson and Joel Shen for assistance in the laboratory. This research was supported by an NSERC Discovery grant, the Canadian Foundation for Innovation and an Early Researcher Award from the Ontario Ministry for Research and Innovation to BJS. We thank two anonymous referees for their constructive comments on an earlier draft of the manuscript. References Angilletta, M.J., 2009. Thermal Adaptation. Oxford University Press, New York. Bayley, M., Holmstrup, M., 1999. Water vapor absorption in arthropods by accumulation of myoinositol and glucose. Science 285, 1909–1911. Benoit, J.B., Lopez-Martinez, G., Michaud, M.R., Elnitsky, M.A., Lee, R.E., Denlinger, D.L., 2007. Mechanisms to reduce dehydration stress in larvae of the Antarctic midge, Belgica antarctica. Journal of Insect Physiology 53, 656–667. Benoit, J.B., Patrick, K.R., Desai, K., Hardesty, J.J., Krause, T.B., Denlinger, D.L., 2010. Repeated bouts of dehydration deplete nutrient reserves and reduce egg production in the mosquito Culex pipiens. Journal of Experimental Biology 213, 2763–2769.

Chown, S.L., 2002. Respiratory water loss in insects. Comparative Biochemistry and Physiology A 133, 791–804. Chown, S.L., Gaston, K.J., 2008. Macrophysiology for a changing world. Proceedings of the Royal Society B 275, 1469–1478. Chown, S.L., Nicolson, S.W., 2004. Insect Physiological Ecology. Mechanisms and Patterns. Oxford University Press, Oxford. Crawley, M.J., 2005. Statistics: An Introduction Using R. Wiley, Chichester. Everaerts, C., Farine, J.P., Cobb, M., Ferveur, J.F., 2010. Drosophila cuticular hydrocarbons revisited: mating status alters cuticular profiles. PLoS One 5. Folk, D.G., Bradley, T.J., 2003. Evolved patterns and rates of water loss and ion regulation in laboratory-selected populations of Drosophila melanogaster. Journal of Experimental Biology 206, 2779–2786. Gibbs, A.G., 2002a. Lipid melting and cuticular permeability: new insights into an old problem. Journal of Insect Physiology 48, 391–400. Gibbs, A.G., 2002b. Water balance in desert Drosophila: lessons from non-charismatic microfauna. Comparative Biochemistry and Physiology A 133, 781–789. Gibbs, A.G., Chippindale, A.K., Rose, M.R., 1997. Physiological mechanisms of evolved desiccation resistance in Drosophila melanogaster. Journal of Experimental Biology 200, 1821–1832. Gibbs, A.G., Fukuzato, F., Matzkin, L.M., 2003. Evolution of water conservation mechanisms in Drosophila. Journal of Experimental Biology 206, 1183–1192. Gibbs, A.G., Johnson, R.A., 2004. The role of discontinuous gas exchange in insects: the chthonic hypothesis does not hold water. Journal of Experimental Biology 207, 3477–3482. Gibbs, A.G., Louie, A.K., Ayala, J.A., 1998. Effects of temperature on cuticular lipids and water balance in a desert Drosophila: is thermal acclimation beneficial? Journal of Experimental Biology 201, 71–80. Gibbs, A.G., Matzkin, L.M., 2001. Evolution of water balance in the genus Drosophila. Journal of Experimental Biology 204, 2331–2338. Hadley, N.F., 1994. Water Relations of Terrestrial Arthropods. Academic Press, San Diego. Hayward, S.A.L., Rinehart, J.P., Sandro, L.H., Lee, R.E., Denlinger, D.L., 2007. Slow dehydration promotes desiccation and freeze tolerance in the Antarctic midge Belgica antarctica. Journal of Experimental Biology 210, 836–844. Hoffmann, A.A., 1990. Acclimation for desiccation resistance in Drosophila melanogaster and the association between acclimation responses and genetic variation. Journal of Insect Physiology 36, 885–891. Hoffmann, A.A., 1991. Acclimation for desiccation resistance in Drosophila – species and population comparisons. Journal of Insect Physiology 37, 757– 762. Hoffmann, A.A., Hallas, R.J., Dean, J.A., Schiffer, M., 2003. Low potential for climatic stress adaptation in a rainforest Drosophila species. Science 301, 100–102. Knowles, L.L., Hernandez, B.B., Markow, T.A., 2004. Exploring the consequences of postmating-prezygotic interactions between the sexes. Proceedings of the Royal Society of London B 271, S357–S359. Marshall, K.E., Sinclair, B.J., 2010. Repeated stress exposure results in a survivalreproduction trade-off in Drosophila melanogaster. Proceedings of the Royal Society B 277, 963–969. Miller, P.L., 1964. Factors altering spiracle control in adult dragonflies – water balance. Journal of Experimental Biology 41, 331–343. Nilson, T.N., Sinclair, B.J., Roberts, S.P., 2006. The effects of carbon dioxide anesthesia and anoxia on rapid cold-hardening and chill coma recovery in Drosophila melanogaster. Journal of Insect Physiology 52, 1027–1033. R Development Core Team, 2010. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing. . Sjursen, H., Bayley, M., Holmstrup, M., 2001. Enhanced drought tolerance of a soildwelling springtail by pre-acclimation to a mild drought stress. Journal of Insect Physiology 47, 1021–1027. Swindell, W.R., 2009. Accelerated failure time models provide a useful statistical framework for aging research. Experimental Gerontology 44, 190–200. Treherne, J.E., Willmer, P.G., 1975. Hormonal control of integumentary water-loss: evidence for a novel neuroendocrine system in an insect (Periplaneta americana). Journal of Experimental Biology 63, 143–159. Urbanski, J.M., Benoit, J.B., Michaud, M.R., Denlinger, D.L., Armbruster, P., 2010. The molecular physiology of increased egg desiccation resistance during diapause in the invasive mosquito, Aedes albopictus. Proceedings of the Royal Society B 277, 2683–2692. Watanabe, M., 2006. Anhydrobiosis in invertebrates. Applied Entomology and Zoology 41, 15–31. White, C.R., Blackburn, T.M., Terblanche, J.S., Marais, E., Gibernau, M., Chown, S.L., 2007. Evolutionary responses of discontinuous gas exchange in insects. Proceedings of the National Academy of Sciences of the United States of America 104, 8357–8361. Williams, C.M., Pelini, S.L., Hellmann, J.J., Sinclair, B.J., 2010. Intra-individual variation allows an explicit test of the hygric hypothesis for discontinuous gas exchange in insects. Biology Letters 6, 274–277.