• Open Access

Mating increases starvation resistance and decreases oxidative stress resistance in Drosophila melanogaster females


Jadwiga M. Giebultowicz, Department of Zoology, Oregon State University, 3029 Cordley Hall, Corvallis, OR 97331, USA. Tel.: 541 737 5530; fax: 541 737 0501; e-mail: giebultj@science.oregonstate.edu


Mating stimulates complex physiological changes in females of Drosophila melanogaster. Long-term effects of mating are manifested in increased fecundity and shortened lifespan. It is not clear how mating affects stress resistance in fly females. We addressed this question here and found that mated and highly fecund wild-type D. melanogaster females have significantly higher resistance to starvation throughout their lifetime than age-matched virgin females. Mean survival time under starvation was age dependent with maximum survival time observed in 15-day-old mated females. Mating-induced increase in starvation resistance was associated with significantly higher fat reserves stored as triacylglycerols. While mated females had higher resistance to starvation, their resistance to oxidative stress was significantly lower than in age-matched virgins. Our study revealed that mating leads to an opposing relationship between resistance to starvation and resistance to oxidative stress in Drosophila females. Thus, shortened lifespan of mated females is associated with their high-fat content and greater susceptibility to oxidative stress.

Mating in Drosophila melanogaster leads to many changes in female behavior and physiology, which are induced by male accessory gland proteins (ACP) transferred to females during copulation (Chapman et al., 1995; Soller et al., 1997; Wolfner, 2002; Kubli, 2003). Mated females show reduced sexual receptivity and a substantial increase in egg production. Mating also shortens the lifespan of female flies; this effect is often ascribed to increased reproductive effort, which may carry nutritional and metabolic costs (Reznick, 1985; Salmon et al., 2001; Kirkwood, 2005). A recent report demonstrated that mating substantially increases food consumption in D. melanogaster females (Carvalho et al., 2006), suggesting that reduced lifespan of mated females may not be a consequence of nutrient shortage as previously postulated. To obtain further insights into the physiological basis of differences in longevity between virgin and mated females, we tested their resistance to starvation and to oxidative stress.

In the first experiment, we compared the survival time of Canton-S virgin and mated females exposed to starvation at different ages. At every age tested, mated females were significantly more resistant to starvation than their virgin counterparts (Fig. 1). In virgin females, starvation resistance showed a general declining tendency as females aged. In mated females, starvation resistance increased with age reaching a maximum on Day 15 and then declined in older females. Thus, starvation resistance is a dynamic trait that depends on both mating status and the age of females. Increased starvation resistance in 15-day-old mated females was observed also in other fly strains. Mated w1118 females survived under starvation 74% longer than virgins (n = 114 and 116 for mated and virgins, respectively, P < 0.0001). Mated Oregon-R females displayed the same trend (not shown), suggesting that increased survival under starvation in mated and reproducing females may be a common trait in D. melanogaster.

Figure 1.

Age-dependent profiles of starvation resistance in virgin and mated Canton-S females reared on molasses/cornmeal medium containing 7% yeast. Bars represent mean (± SEM) survival time, which was significantly longer for mated than for virgin females at every age tested (t-test: P < 0.01 on Days 4 and 45; P < 0.001 on Days 8, 15 and 30). Data shown are based on two independent experiments with a total number of 300 virgin and 300 mated females tested for each age group.

The increase in starvation resistance in mated females prompted us to verify that our Canton-S flies show expected trends in fecundity and longevity. We compared these traits in mated and virgin females reared in identical experimental conditions as those tested for starvation resistance. Three cohorts of virgin and mated females had their fecundity recorded for 15 days. Virgin females (n = 140) produced on average 52 ± 6 eggs while mated females (n = 128) produced 141 ± 5 eggs per female over this period; this difference was highly significant (P = 0.0012). Longevity assessment of females used in this experiment showed that the lifespan of virgins was extended by 31%. Mated females had a significantly shorter average lifespan of 45.6 ± 2 days than virgin females (59.7 ± 1.2 days). Thus, mated females used in our study showed an expected decrease in longevity and increase in fecundity; yet, their starvation resistance was significantly higher than in virgins. These data fail to support a previously suggested positive correlation between starvation resistance and longevity, and a negative correlation between starvation resistance and fecundity in D. melanogaster (Service et al., 1985; Chippindale et al., 1993; Simmons & Bradley, 1997).

The ability of mated females to survive longer under starvation may be related to a number of factors, such as lower energy expenditure or higher energy reserves. To probe the physiological bases of increased starvation resistance, we first tested whether mated flies become less active than virgins when starved. Recordings of locomotor activity of mated and virgin females as described by Lee & Park (2004) suggested that both were highly active from the onset of starvation until their death (data not shown). Next, we compared food intake between virgin and mated females using diet containing FDA blue dye no. 1 (McCormick, Hunt Valley, MD, USA), allowing the volume of consumed diet to be estimated (Edgecomb et al., 1994). Mated females had a significantly higher concentration of blue dye in their gut homogenates than did virgin females (data not shown). These data are in agreement with a recent report demonstrating increased food consumption in mated D. melanogaster females (Carvalho et al., 2006).

In the next experiment, we asked whether mated females have higher energy reserves, as this could provide a basis for increased starvation resistance (Djawdan et al., 1998). We compared nutrient levels in groups of females collected at 0, 36 and 60 h after the onset of starvation. Triacylglycerols (TAG), glycogen and proteins were separated and measured as described in Van Handel (1965) and Zhou et al. (2004). Before starvation, TAG content in mated females was nearly twice as high as in virgins (Table 1). During food deprivation, TAG levels declined in both mated and virgin females; however, mated females had retained substantial TAG levels after 60 h of starvation in agreement with their longer survival (Fig. 2a). In contrast to fat levels, prestarvation levels of glycogen were not significantly different between mated and virgin females and declined in both by 36 h of starvation (Table 1). Total prestarvation protein content was not significantly different between mated and virgin females (Table 1). Thus, increased starvation resistance in mated females appears to depend on selective accumulation of lipids, which occurred despite a threefold increase in egg production.

Table 1.  Fat, protein and glycogen content in mated and virgin females of Drosophila melanogaster subjected to starvation
NutrientBefore starvationStarved 36 hStarved 60 h
  1. All values are expressed in micrograms/fly; the data are means ± SEM values (n = 5, except n = 3 for virgins starved for 60 h).

  2. The data are pooled from two experiments, in which either 15 or 20 flies were analyzed per sample.

  3. Significant differences (P < 0.01) between the values for virgin and mated females obtained by unpaired t-test are indicated by *.

Triacylglycerols 55.2 ± 1108.4 ± 10* 21.2 ± 0.7 69.8 ± 4.6* 11.1 ± 2 45.2 ± 6*
Proteins199.0 ± 31178.6 ± 34202.9 ± 24165.0 ± 30223.4 ± 34176.2 ± 39
Glycogen 18.2 ± 1 19.0 ± 2  2.1 ± 0.3  1.3 ± 0.6  0.6 ± 0.5  0.4 ± 0.2
Figure 2.

Stress responses of 15-day-old mated and virgin females. (a) Starvation resistance is significantly higher in mated females at P < 0.0001. Median survival of mated and virgin females was 108 and 60 h, respectively. (b) Tolerance to hydrogen peroxide vapor was significantly higher in mated females at P < 0.0001. Median survival was 18 h for virgins and 8 h for mated females. (c) Hyperoxia tolerance (100% O2 at a flow rate of 1 L per min) was higher in virgin females at P < 0.0001. Median survival was 84 h for virgins and 72 h for mated females. In all experiments, Kaplan–Meier survival curves were compared using the log-rank test (GraphPad Prism software).

In parallel to starvation resistance, we tested resistance to oxidative stress in mated and virgin 15-day-old Drosophila females with two experiments. In the first experiment, flies were placed in Petri dishes supplied with 100 µL of 30% hydrogen peroxide (H2O2) on filter paper. Average survival was significantly lower for mated females compared to virgins (Fig. 2b). In the second experiment, flies were placed in vials containing 5% sucrose on filter paper and kept in an air-tight chamber under 100% oxygen with a flow rate of 1 L per min. Average survival under hyperoxia was again significantly lower for mated females compared to virgins (Fig. 2c). Taken together, our experiments demonstrate that the response to two different stressors, starvation and oxidation, changes in opposing directions after mating. These results are surprising as resistance to multiple stressors was shown to change in parallel in flies selected for greater stress resistance or in longevity mutants (Lin et al., 1998; Harshman et al., 1999; Wang et al., 2004). However, most previous experiments were performed on males, and none compared the responses of virgin and mated females to both types of stress in one study.

Our data provide interesting insights into the poorly understood longevity differences between mated and virgin Drosophila females. Virgin females, which have a longer lifespan, appear to be on self-imposed caloric restriction; they are lean and have higher resistance to oxidative stress. Both limited intake of nutrients and high resistance to oxidative stress have been linked to increased longevity in Drosophila and other model organisms (Partridge & Gems, 2002; Sohal, 2002; Lim et al., 2006). Higher fat levels in mated females may increase their susceptibility to oxidative stress; links between lipid oxidative damage and accelerated aging have been demonstrated in Drosophila (Zheng et al., 2005). Interestingly, the increase in body fat during reproduction has been documented in vertebrate females, including mammals (Naismith et al., 1982; Gosman et al., 2006). Our data showing elevated fat reserves in reproducing flies suggest that this phenomenon may be a conserved evolutionary trait. The switch in metabolic pathways that leads to fat accumulation likely evolved to ensure reproductive success. However, this switch carries a cost of decreased resistance to oxidative stress, which is likely to underlie the shortened lifespan of mated females.


We thank Louisa Hooven, Tory Hagen, and Scott Pletcher for reading an earlier version of the manuscript. We thank Dr. Gary Merrill for suggesting a novel method of testing oxidative stress resistance. We are thankful to Natraj Krishnan and James Lester for help with some experiments. This research was supported by National Science Foundation (NSF) grant 0446339 and National Institutes of Health (NIH) R03 AG024647-01 grants to Jadwiga Giebultowicz.