June, 2001

Journal of Vector Ecology

21

Aedes aegypti: size, reserves, survival, and flight potential

Hans Briegel*, Irne Knsel, and Susanne E. Timmermann
Institute of Zoology, University of Zrich, CH-8057 Zrich, Switzerland * Corresponding author Received 1 May 2000; Accepted 31 August 2000 ABSTRACT: Female Aedes aegypti of small and large body sizes were fed ad libitum from eclosion with eight different concentrations of sucrose from 0.1% to 50%; females were also starved with access to water. For each experiment we determined the survivorship of such populations. The 90%, 50%, 10%, and maximal survivorships followed linear regressions with the logarithm of the sucrose concentration. For each condition we measured the extent of synthesis of glycogen and lipid reserves. There was a critical sucrose concentration of 0.5% for both size classes: lower concentrations were of no nutritive effect, and all higher concentrations extended survivorship and allowed reserve synthesis. With respect to the teneral value, and normalized for body size, the maximal amounts of glycogen increased 2-3-fold within one week, whereas lipogenesis increased 3-5-fold requiring two weeks. Solid sugar cubes could also be utilized as long as drinking water was available, but synthesis of additional reserves failed. Flight mill experiments revealed the temporal flight pattern, its maturation after eclosion, and the maximal flight performances. Flights shorter than 1000 m per female per night were considered as low activities, whereas flights longer than 1000 m represented strong vigorous flights. Maximal distances were from 11-18 km/female/night. Periods of continuous flights lasted between 2-9 hr per female (mean 2.2 hr). Maximal flight performances were gradually reached within the first and third day of eclosion. Mean caloric energy consumption during flight was 33% to 44% of the pre-flight glycogen, accompanied by lipid reductions of 9%. Evidently, feeding on carbohydrates allows extended flight activities of this species and is essential for survival in the absence of blood meals. Journal of Vector Ecology 26(1): 21-31. Keyword Index: Aedes aegypti, reserves, body size, survival, flight potential INTRODUCTION In most mosquitoes, females require additional protein after eclosion to initiate vitellogenesis. For Aedes aegypti, the metabolic relationships between blood meal utilization and fecundity have been quantitatively determined, in relation to size and source of the blood meal, age and size of the female, and in comparison to Anopheles (Briegel 1990a, b). Van Handel (1965) has characterized obesity for the mosquito Ae. sollicitans and conspicuous lipogenesis was also found in several other species, especially in Ae. aegypti (Briegel 1990a, Timmermann and Briegel 1999, Briegel and Timmermann for Ae. albopictus, in prep). Mosquitoes are not ready for reproduction immediately after eclosion. Males require one day for the rotation of their terminalia, and females develop their host-seeking behavior within one to two days (Klowden 1990; Briegel unpublished data). Depending on the availability of a host, females might need to wait a few more days for their first blood meal. In general, seeking carbohydrate sources, i.e. feeding on sugar solutions, is a common behavior for most of them and for both sexes (Foster 1995). Feeding on sugar solutions extends their survival time (Briegel and Kaiser 1973, Nayar and Sauerman 1971, Foster 1995), and improves their reserve conditions, mainly in lipids (Van Handel 1965; Nayar and Sauerman 1971, 1974; Briegel 1990a). So far, only Ae. aegypti was reported to largely abstain from sugar feeding in favor of blood meals (Scott et al. 1997, Costero et al. 1998). In nature, nectar is assumed to be the prevalent source of sugars but probably all sorts of sugars in different concentrations are available (Foster 1995). Van Handel et al. (1972) demonstrated the prevalence of sucrose, besides fructose and glucose, and many other sugars in minor quantities for over 40 different nectar samples and fruit juices. The availability of sugar solutions is not confined to nectar, but a wide variety of carbohydrates may be accessible, with honeydew probably the most important (Foster 1995). Ae. taeniorhynchus is quite tolerant in accepting most sugars, while Ae. aegypti was reported to be more selective (Nayar and Sauerman 1971). Observations of mosquitoes feeding on flowers are relatively rare,

22

Journal of Vector Ecology

June, 2001

despite great efforts by several entomologists in the field (McCrae et al. 1969, McCrae 1989, Haeger 1960). Flight potential is an important consequence of the status of reserves, particularly for Ae. aegypti and its epidemiological potential (Reiter, personal communication). The first movements in their imaginal life take place before any food sources are encountered, and therefore, the energy requirements have to be met by the reserves they carry over from the larval/pupal period, i.e. the teneral reserves. Subsequently, carbohydrate sources need to be found and tapped for as long as no blood donors are available. This study aims at a complete, systematic and quantitative laboratory investigation of the sucrose feeding and its metabolic consequences. While Nayar and Sauerman (1974) studied the regulation of daily intake and its inverse correlation with concentrations, we exclusively followed the effects of increasing concentrations on survival and the absolute synthesis of reserves. Newly eclosed females were provided with sucrose solutions of gradually increasing concentrations from zero up to 50% ad libitum. The survivorships were recorded, live and dead females sampled from these experiments, analyzed for reserves and then compared with the conditions at eclosion. In view of the quantitative analyses, we then addressed the flight performance of Ae. aegypti and its quantitative relationships to the glycogen and lipid reserves present at selected times of life. For that purpose we have utilized a computerized flight mill similar to Rowley et al. (1968) or Nayar and Sauerman (1973) and recorded distances flown by individual mosquitoes, and simultaneously recorded the temporal flight pattern. We did not attempt to imitate field conditions, but rather aimed at a more complete analysis of the physiological interactions between the metabolic background and the maximal flight potential for the yellow fever mosquito, Ae. aegypti. In addition, these data form the background for results on reproductive cycles, which will be reported elsewhere (Briegel et al., in preparation). MATERIALS AND METHODS Aedes aegypti (L.) strain UGAL, was used exclusively for this investigation. Larvae were reared under optimal conditions, i.e. 200 larvae per pan, fed TetraMin with our standard regime, or under moderate crowding conditions, i.e. 400 larvae per pan, fed 1 spoon of standard diet on alternate days, as described in detail by Timmermann and Briegel (1993). The former colonies produced large imagines, the latter small ones. Wing length and its cubic value were used to record

body size (Briegel 1990a). All colonies were kept at 27C, the imagines at a relative humidity of 852%, and with long-day conditions (14 hr photophase) with dim light periods of 45 min to simulate sunset and sunrise. All experimental cohorts (50 each) were kept in 8 l cages and provided with two cotton wicks that were soaked with the sucrose solutions to be offered ad libitum: 0.1%, 0.5%, 1%, 2%, 5%, 10%, 20%, and 50%, or water. Survival was recorded by daily counts and removal of dead insects, which were fixed for analyses. At the same time, in parallel cages of identical treatment, live females were sampled (5-10 per day) and fixed as well, to trace their changes in reserves. For tissue fixation, each female was put into a reacting tube (14x100 mm), together with 100 l ethanol 100%, and heated for 10 min at 90C. These samples were stored at room temperature until analysis, but no longer than two weeks. The flight mills were round containers (15x6 cm for diameter and height). At the tips of two thin arms of 5 cm length, a human hair of 1-2 cm was fixed at which the mosquito was glued with a minimal drop of wax applied to its anterior portion of the mesothoracic tergite. The correct flight position of the mosquito was crucial for optimal performance. The revolutions of the drum were recorded by a photocell and fed into the computer. A program was designed to count and store these revolutions per selected time interval, 30 sec in our experiments. The program then converted these revolutions into absolute distances flown per time interval, which also were summed up to total distances in meters. In our experience, flight durations of more than 18 hr were extremely rare with Ae. aegypti and therefore, this was the arbitrary cut-off point of daily data collections. All flight trials were run at a room temperature of 20-22C and overnight to minimize disturbances by lab personnel. In preliminary tests we found no dependence of the flight activities from lighting regimes in this species. The printout of the computer protocols provided all single spikes of flight activities for the entire test period on absolute time axes, revealing the temporal pattern of flight activities. Spontaneous interruptions or resumptions were visible (exemplified in Figure 3). Movements of less than 1m/min were neglected as noise or background activities. Based on their total flight distances, we defined weak fliers with total distances below 1000 m, and strong fliers exceeding 1000 m per female per night. Females were considered exhausted when they did not move any more on the flight mill, but were still alive; dead females were discarded. A total of 294 females was available for analysis. The analyses for glycogen and lipid were carried

June, 2001

Journal of Vector Ecology

23

Figure 1.

Effect of female body size (large versus small) and sugar diet (0.5% versus 5% ad lib.) on the content of glycogen and lipid of Ae. aegypti. Caloric data are related to the teneral conditions, i.e. at eclosion. The absolute calories were converted to SSCC-values and its teneral amount defined as 1.0, marked as a dashed horizontal line in each diagram. The shadowed band represents the respective m.i.a., i.e. the death line. As a further means of comparison, the actual survivorship data of these four cohorts have been plotted. Open circles are values for dead females, and filled circles for live controls.

24

Journal of Vector Ecology

June, 2001

Figure 2.

The 50% survival times (solid lines) of Aedes aegypti fed water (shadowed area) or increasing concentrations of sucrose ad lib. plotted against the logarithmic value of sucrose concentration for females of different body sizes (solid lines). Large females had wing lengths always higher than 3.1 mm, and all small females had wing lengths below 2.8 mm, both groups representing extremes of the possible sizes. For comparison, the 90% (lower), 10% (upper), and maximal survival time regressions have been added (dashed lines). All survival times follow linear regressions (formulas in Table 1).

out by the procedures of Van Handel and Day (1988), which allowed us to quantify both components within the same samples. In certain experiments we also measured the content of total sugar with glucose as a standard. The glycogen and lipid data were expressed in calories per single female. These caloric values were then normalized for size and presented as SSCC-values (size-specific caloric contents; see Timmermann and Briegel 1993). Furthermore, the SSCC-values of small and large teneral females were arbitrarily set as 1.0 in all diagrams, and all SSCC-values converted accordingly. Data for survivorship were plotted as survival curves, and the 90%, 50%, 10% and maximal survival times were determined graphically from these plots. These four survival points from all the diets were then related to the logarithm of sucrose concentrations by plotting against the log value of the sugar concentration; all showed significant linear regressions. All means are accompanied by standard errors (S.E.) and sample sizes (N).

RESULTS Survival times Survivorship curves usually follow an inverse sigmoidal pattern, revealing the maximal survival time for each feeding regime (Figure 1). By graphic interpolation we determined the 90%, 50%, 10%, and maximal survival times. When plotting these selected time points against the increasing dietary sucrose concentrations in a semilogarithmic diagram, they showed linear regressions (Figure 2). Females of the two extreme size classes were analyzed. Females provided with water to prevent desiccation showed the effect of starvation. Their body size clearly affected survival, which was 4.0 1.0 d (N=9) for small females with a mean body size of 14.102.08 mm3 (N=112, range 9.26-20.80), and 9.82.1 d (N=9) for large females with a size of 41.443.20 mm3 (N=48, range 32.77-48.69), a significant difference in survival time (t=3.69, p<0.01). Females with access to 0.1% sucrose showed no substantial difference in their survival times over those that were starved (Table 1). Therefore, this

June, 2001 Table 1.

Journal of Vector Ecology Regression formulas (Y=a+bX) for the various survival times (Y, days) of small and large females with access to water only (starvation), or increasing sucrose concentrations (X, logarithmic value; N=7 for each). Values for starving females and for females offered 0.1% sucrose are excluded from the regressions. See also Fig. 1. survival time a b r2 t starving females
a a b b

25

sugar-fed 0.1% 1.5 2.0 3.5 4.0 3.8 7.4 10.0 10.5

SMALL

90 % 50 % 10 % maximal 90 % 50 % 10 % maximal
b

10.54 17.16 22.11 25.15 12.99 21.58 27.06 32.94

12.98 18.32 22.15 29.41 10.32 21.13 30.57 35.50

0.932 0.913 0.878 0.824 0.806 0.909 0.912 0.893

8.29 7.23 5.98 4.84 4.56 7.06 7.19 6.46

1.3 2.0 2.5 4.0 5.7 7.1 8.4 9.8

LARGE

b a a b

p<0.001,

p<0.01

concentration was omitted from the linear regressions that were computed for all the other sucrose diets and corresponding survival times (Table 1). With 10% sucrose, the conventional concentration in laboratories, the 50% survival time reached 40 days for large females and 35 d for small females. The maximal survival time is a less reliable parameter because of its high and unpredictable variability; occasionally we have observed females living up to 105 d (Briegel and Kaiser 1973). Thus the 50% survival appears to be the most reliable parameter. For comparison, however, the 90% and 10% survival times and their linear regressions have been added to Figure 2. All these regressions have their Xintercepts between 0.1% and 0.5% concentrations. Therefore, the 0.5% sucrose concentration represents a threshold above which survival times are gradually improved. Sucrose below 0.5% shows survival times similar to water, thus it is of no nutritive value. To estimate the possible sucrose consumption per female in absolute terms, we have assumed an average volume for crop filling of 0.5 l. Based on this, a 0.5% sucrose solution would correspond to 0.01 cal of sugar, whereas with a 50% solution it would amount to 1 cal. Mobilization or synthesis of glycogen and lipids To compare the changes in reserves of glycogen and lipids from live and dead females, we chose the diets of 0.5% and 5% sucrose (Figure 1). The absolute calories per female of these substances have all been normalized for size, i.e. converted to SSCC-values. Furthermore, the SSCC-value of teneral females is

arbitrarily set as 1.0, thus forming the basis for all other SSCC-values. This provides meaningful comparisons of the fate of the teneral reserves during the imaginal life; either they are reduced through starvation and mobilization, or improved through additional synthesis from dietary sucrose. Females starved to death in the presence of drinking water however, reveal the minimal irreducible amounts (m.i.a.) of lipids and glycogen (Van Handel 1984), thus representing the critical threshold for survival. These values are also included in sizenormalized SSCC-units in Figure 1 (shaded band). Ad libitum access to 0.5% or lower sucrose leads to a quick mobilization of the teneral reserves. Obviously, this caloric input is too small to allow new synthesis or even to save the reserves for an extended period. However, with access to 5% or higher sucrose, additional glycogen can be synthesized, exceeding the teneral values, and small females are even able to accumulate additional lipid reserves. Large females, however, only maintained their lipid content. These data do explain the extended survivorships in Figure 2, also mentioned in Table 1. With all diets of 10-50% sucrose, the lipid levels always remained above teneral values, and glycogen levels fell below teneral values only, once the 50% survival time has been passed. The points of maximal synthesis of these reserve components are compared with the teneral status in Table 2, all expressed as many-folds of teneral SSCCvalues. Again, and in both size classes, only sugar concentrations higher than 0.5% provided sufficient caloric input to allow the synthesis of additional

26 Table 2.

Journal of Vector Ecology

June, 2001

Reserve synthesis by female Ae. aegypti when fed increasing concentrations of sucrose. The table shows the days at which maximal many-folds of the teneral SSCC-value occur (average values). Data from females fed less than 1% sucrose are omitted because none allowed additional synthesis of glycogen nor lipids. A dash indicates that these values consistently fell below the teneral SSCCvalue. SMALL FEMALES GLYCOGEN LIPIDS LARGE FEMALES GLYCOGEN LIPIDS day 3 1-3 1-5 1-4 * * x-fold 1.9 2.1 3.2 3.6 >2.0 >2.0 day 2-8 1-12 * * x-fold 1.2 2.8 >2.0 >2.0

SF 1% SF 2% SF 5% SF 10% SF 20% SF 50%

day 1-3 2-14 1-10 1-8 * *

x-fold 1.1 2.6 2.3 2.8 >2.0 >2.0

day 3 2-4 2-6 2-10 * *

x-fold 1.4 2.3 2.2 4.1 >4.0 >3.0

* sporadic measurements throughout the complete life-span Table 3. Frequency distribution of flight performances of female Ae. aegypti (recorded during single nights). Five flight ranges are arbitrarily defined. WF for water-fed, SF for sugar-fed, BF for blood-fed. TOTAL FLIGHTS FLIGHT PERFORMANCE (km/female) POOR <1 1-3 3-5 5-10 >10 * FLIERS ** 7 20 8 6 1 10 34 1 13 14 9 17 24 15 9 12

CONDITIONS

WF SF SF BF BF

teneral: 2-8 h 8 1-5 d 31 1d-8w 101 1-3d; mini *** 7 1-2d; before ovipos. 25 3-5d; after ovipos. 34

1 36 10 12

20 1 7

3 1 1

* The absolute maximal distances observed were 10.9, 13.4, 14.7, 17.2, and 18.8 km per female per night. ** shown for comparison *** small-sized females Table 4. Mobilization of teneral glycogen or lipid of females that died in the feeding experiments. The table indicates the day at which 50% of the teneral reserves (in SSCC-units) have been mobilized, and at which they fell to the minimal irreducible level, i.e. the death line. For comparison, the 50% survival times (ST) are also given. SMALL FEMALES GLYCOGEN LIPIDS 50% death 50% death 1.5 3 <1.0 <1.0 1.5 3 <1.0 <1.0 4 3 <2.0 <2.0 8 >10 5 5 *20 >20 *20 >20 *16 >10 *19 >20 *>40 >30 >40 >40 never never LARGE FEMALES GLYCOGEN LIPIDS 50% death 50% death 2.5 5 3.0 5.0 1.5 5 2.5 5.0 3.0 10 2.0 8.0 *12 >15 5.0 14.0 *10 >30 5.0 >30 *30 >30 *30 >30 *>50 *>50 never never

WF SF 0.1% SF 0.5% SF 1% SF 2% SF 5% SF 10% SF 50%

ST 2 2 5 18 22 32 35 51

ST 5.0 6.0 14 22.5 45 43 49.5 59

* These cohorts first had synthesized amounts higher than teneral values (see Table 3)

June, 2001

Journal of Vector Ecology

27

Figure 3.

Examples of temporal flight patterns of three selected individual females of Ae. aegypti of increasing age at the time when the flight trial started. The computer printouts indicate the individual flight performance (m/min) for each female, recorded during 16 h on the flight mill. At this point females were alive but did not move any more. The corresponding total flight distance (km) is at the upper right corner. Single, isolated spikes mark bursts of trivial flights, therefore black segments represent non-stop vigorous flight periods of approximately 2.5 and 9 h in these examples.

Figure 4.

Flight performance (km total distances and h duration) of female Ae. aegypti of increasing age (17 d, 3-6 weeks), with permanent access to 10% sucrose before flight, and for 1-5 d after blood meal (= 4-9 d of absolute age); oviposition (inverted triangle) between day 2 and 3. Only strong and vigorous fliers are considered. The black areas of each bar denote the sum of continuous flight episodes (mean S.E., N=8-14 for each bar).

28

Journal of Vector Ecology

June, 2001

glycogen. With respect to lipid synthesis however, large females required 5% or higher concentrations to synthesize additional reserves, whereas small females were capable of lipogenesis with 1% sucrose solutions. When the sucrose ingested was above 10%, in all females the maximal glycogen synthesis was 2-3-fold the teneral value, and the maximal lipid values were 45-fold the teneral SSCC in small females, but 3-4-fold the teneral values in large females (Table 2). Even among all dead females fed 10-50% sucrose, lipid was always above teneral SSCC-values. From this we concluded that death may be caused by glycogen depletion, but not by exhausted lipids. To test whether female Ae. aegypti were able to ingest carbohydrate from solid sources as reported by Eliason (1963), we had a population of large females set up with wicks of drinking water and with a sugar cube on the bottom of a dry cage. The 50% survival time was 58 d, i.e. about 50 d longer than with water alone. But their glycogen never exceeded teneral levels, and lipid levels fell below the teneral value, thus resembling the situation with a 1-2 % sugar diet. Flight performance of Ae. aegypti With respect to total distances flown by 294 single females, two groups could be clearly distinguished: weak fliers, all below 1000 meters per night (N=89), and strong fliers (N=205), all above 1000 meters/ female/night. To provide a reasonable estimate of the actual or maximal flight potential of Ae. aegypti, we focused on the strong fliers only, as defined by their distances and flight kinetics. In Figure 3 three typical flight patterns are presented, comparing the flight activities between newly eclosed, teneral females and females with permanent access to sucrose before the start of the flight trials. At 2-8 h only trivial flights prevailed that represent small hops, probably away from the water surface to avoid predators such as spiders. The capability for appreciable flight developed already within the first day of imaginal life. We extended the studies with sugar-fed females for up to 56 d (8 weeks) after eclosion, i.e. beyond the 50% survival-time of such a female population. The flight performances of the strong fliers are summarized in Figure 4; blood-fed females are also included. The bars represent mean flight distances (km) or duration (h), where the nonstop flight segments are in black (Figure 4). With permanent access to sucrose, maximal flight potentials persist throughout the entire life span (Figure 4). In Table 3 the numerical distributions of females are broken down to five selected flight ranges. Females deprived of sucrose and given only water died after 5 days and their flight potential remained extremely low,

two thirds flying less than 1000 m. The flight potential clearly developed only when sucrose was available. The long-lasting flights are composed of continuous flight segments of an average of 2.2 0.6 hr (N=17). The pauses usually varied between fractions of a minute and fractions of an hour, possibly representing periods of metabolite mobilization. There were similar distributions of the frequencies of flight performances whether previously fed sugar or blood (Table 3). It is noteworthy that the small females even with access to sucrose ad lib. flew as poorly as water-fed or teneral sisters of larger size. The average speed of flight for all females was 0.64 km/hr; for poor fliers it was only 0.47 km/hr, and for the vigorous, continuous flyers average speed was 1.00 km/hr with individual maxima of 1.2 km/hr. The previous data suggest that females require feeding on carbohydrates shortly after eclosion. We investigated the feeding history and its effects on the flight dynamics depending on caloric conditions. The caloric content of carbohydrates and lipids before and after flight were measured and compared with sisters that were kept motionless in a reacting tube for the duration of the flight trials. The glucose data of preflight females largely represent the crop content. Based on the glucose determinations, large females had ingested volumes in the range of 0.6-1.1 l whereas their small-sized sisters contained only half of those volumes, 0.3-0.5 l. But when their glucose content was normalized for body size, the ranges were similar: 0.009-0.015 cal/mm3 in smaller and 0.008-0.012 cal/ mm3 in larger females. In sugar-fed females the glucose and glycogen contents steadily rose during the first 35 days post eclosion to absolute levels of 0.3-0.5 cal each per female (N=34). Measurements of females flown to exhaustion, and subtraction of their mean values from unflown sisters, revealed the actual utilization of reserves for the flight activity recorded on the mill. Among 129 females with access to sugar before flight, reduction of glucose was 3319% (N=11) of pre-flight content; for glycogen these values were 4422% (N=11), and their pre-flight lipid was reduced during flight by 95% (N=11) or by 44%(N=5) in 83 blood-fed females. These figures are means with considerable variability, despite the same body sizes and identical treatments of all females before the flight trials. The five females with maximal flights over 10 km showed the following distributions. When sugar-fed before flight (1 day to 4 weeks) they consumed more of their reserves: 16-33% of glucose, up to 86% of their glycogen, and 13% of their lipid for flights between 11 and 15 km, although in an inconsistent pattern. Among the 83 blood-fed females

June, 2001

Journal of Vector Ecology

29

two maximal fliers with 17 and 19 km/night respectively, consumed an average of 24% of their pre-flight glucose, and 11% of their glycogen, while their lipid consumption appeared very erratic (1-30%). DISCUSSION In the absence of blood meals, survival of mosquitoes strongly depends upon feeding on carbohydrates. The teneral reserves per se, a product of larval biosynthesis (Timmermann and Briegel 1999) guarantee a limited flight potential and survival time of only a few days, maintained by mobilization of most glycogen and lipid reserves present. Survival times are a very sensitive indicator for the effects of almost any physiological parameter. With access to high sugar concentrations maximal survival times were 69-90 d for large and 55-75 d for small females. The body size at eclosion and its reserve status form the baseline for further comparisons. There is a limit of caloric contents of lipid, glycogen and protein, the minimal irreducible amounts (m.i.a.), that form the border line between life and death. This level is dependent on body size too, which explains the 4 d survival of small versus 9-10 days of large females. Consequently, in mosquitoes the trait of frequent and constant seeking for and feeding on carbohydrates has evolved, as shown many times by field investigations or as routinely experienced in laboratory colonizations (Foster 1995). Because survival times are linearly correlated with the logarithm of the dietary sugar concentrations, Figure 1 allows the estimation of the 50%, 90%, or 10% survival times of small and large females Ae. aegypti for any concentration. We have to point out however, that different principles may prevail in females offered blood meals, as suggested by Scott et al. (1997) and Costero et al. (1998). By offering females increasing concentrations of sucrose, two questions were addressed: how much of a caloric input is required to prevent exhaustion of the reserves, and how efficient is the synthesis of additional reserves? A threshold level of 0.5% sucrose was found for this species. When more than 1% sucrose solutions were available, females of all body sizes increased their teneral glycogen 2-3 times. However, lipogenesis occurred differently. Small females could increase their teneral lipid level up to 5 times already with 1% sucrose, while in large females over 5% was required to produce 3-4 times the teneral lipid. Van Handel (1965) coined the term the obese mosquito for the salt marsh mosquito Ae. sollicitans. When fed 13 cal of sugar, both teneral lipid and glycogen increased by a factor of roughly 2.5. Therefore, obesity is even more

pronounced in Ae. aegypti than in Ae. sollicitans. With sucrose diets higher than 10% however, in all females the lipid content never fell below their teneral value, they even died when considerable amounts of lipids were still present. Obviously, as long as carbohydrates are available in unlimited amounts, death is not a matter of exhausted lipid reserves, but it was consistently accompanied by glycogen depletion. Lipid depletion to its minimal irreducible amount is only possible with complete starvation, i.e. with access to water only (Briegel 1990a). High sugar concentrations could also be a disadvantage; despite high caloric values, the viscosity could cause artefactual mortality in laboratory populations through glued extremities. Of course, physiological parameters other than glycogen contents definitely affect longevity, as revealed by glycogen contents far below teneral values in females still alive. Our experiment with solid sucrose cubes demonstrates that in nature any carbohydrate source is accessible as a nutritive substrate, independent of its concentration, as long as drinking water is also available. This result confirms the observations of Eliason (1963) and the stipulation by Foster (1995) that extra-floral nectars or honey-dew can be approached and utilized by mosquitoes but may not be detected in the field. This might explain why entomologists have observed so few mosquitoes on flowers (McCrae et al. 1969; McCrae 1989, and personal communication). Honeydew most likely is a major source of sugar, a feeding substrate that can easily be visited at night without an entomologists recognition. Our results on flight performance are in line with conclusions reached by Rowley and Graham (1968). The maximal distances flown by single females are similar, except that in our studies the effect of age was less pronounced; even 4 weeks old females flew as far and as long as 2 week old females. Obviously, flightmill studies are a reliable and potential means to assess maximal flight abilities of mosquitoes. The diminished flight potential, as observed by several investigators (Rowley and Graham, 1968; Johnson and Rowley 1972; Nayar and Van Handel 1971) relates to the diminution of glycogen synthesis from sugar feeding. As lipids do not contribute to flight energy (Clements 1955; Nayar and Van Handel 1971), the extent and the synthetic potential of glycogen reserves (Table 2) is of crucial significance for extended flight performances of Ae. aegypti. It should be emphasized that this approach did not claim to show flight performances encountered in the field; we rather wanted to demonstrate the absolute flight potential of these females. For that purpose forced flight conditions are appropriate. The question of the significance of sugar feeding,

30

Journal of Vector Ecology

June, 2001

raised by Foster (1995) in his broad review, has been convincingly answered. We have established that feeding on carbohydrates in whatever form or concentration can be a relevant behavior to maintain a reserve status and finally the flight potential. Our findings do not rule out the usual situations with Ae. aegypti living in extremely domestic environments, as reported by Scott et al. (1997) and Costero et al. (1998), but the blood meal story was not the central issue of this study. In conclusion, all these parameters undoubtedly increase its vector potential, of course, in combination with its haematophagy that has been thoroughly quantified previously (Briegel 1990a). Acknowledgments We thank Mr. H.J. Baumann for the skillful development of the software program for our flight mill and its computer. Mrs. I. Flckiger was responsible for the colonization of the mosquitoes, and Mrs. R. Haigis contributed with her reliable biochemical analyses. The financial support by the Swiss National Science Foundation to HB is greatly acknowledged. We also appreciate substantial suggestions by one of the reviewers. REFERENCES CITED Briegel, H. 1990a. Metabolic relationship between female body size, reserves, and fecundity in Aedes aegypti. J. Insect Physiol. 36: 165-172. Briegel, H. 1990b. Fecundity, metabolism, and body size in Anopheles (Diptera: Culicidae), vectors of malaria. J. Med. Entomol. 27: 839-850. Briegel, H. and C. Kaiser. 1973. Lifespan of mosquitoes (Culicidae, Diptera) under laboratory conditions. Gerontologia 19: 240-249. Clements, A. N. 1955. The sources of energy for flight in mosquitoes. J. Exp. Biol. 32: 547-554. Costero, A., G.M. Attardo, T.W. Scott, and J.D. Edman. 1998. An experimental study on the detection of fructose in Aedes aegypti. J. Am. Mosq. Cont. Assoc. 14: 234-242. Eliason, D.A. 1963. Feeding adult mosquitoes on solid sugars. Nature 200: 289. Foster, W.A. 1995. Mosquito sugar feeding and reproductive energetics. Annu. Rev. Entomol. 40: 443-474. Haeger, J.S. 1960. Behavior preceding migration in the salt-marsh mosquito, Aedes taeniorhynchus (Wiedemann). Mosq. News 20: 136-147. Johnson, B.G. and W.A. Rowley. 1972. Age-related ultrastructural changes in the flight muscle of the mosquito, Culex tarsalis. J. Insect Physiol. 18:

2375-2389. Klowden, M.J. 1990. The endogenous regulation of mosquito reproductive behavior. Experientia 46: 660-670. McCrae, A.W.R. 1989. Differences in sugar feeding activity between tropical and temperate mosquitoes: field observations and their implications. Vect. Ecol. Newsletter 20: 16. McCrae, A.W.R., Y. Ssenkubuge, P. Manuma, C. Mawejje, and A. Kitama. 1969. Mosquito and tabanid activity at plant sugar sources. E. Afr. Virus Res. Inst. Rep. 1968: 96-102. Nayar, J.N. and D.M. Sauerman Jr. 1971. Physiological effects of carbohydrates on survival, metabolism, and flight potential of female Aedes taeniorhynchus. J. Insect Physiol. 17: 2221-2233. Nayar, J.N. and D.M. Sauerman Jr. 1973. A comparative study of flight performance and fuel utilization as a function of age in females of Florida mosquitoes. J. Insect Physiol. 19: 1977-1988. Nayar, J.N. and D.M. Sauerman Jr. 1974. Long-term regulation of sucrose intake by the female mosquito, Aedes taeniorhynchus . J. Insect Physiol. 20: 1203-1208. Nayar, J.N. and E. Van Handel E. 1971. The fuel for sustained mosquito flight. J. Insect Physiol. 17: 471-481. Rowley, W.A. and C.L. Graham. 1968. The effect of age on the flight performance of female Aedes aegypti mosquitoes. J. Insect Physiol. 14: 719728. Rowley, W.A., C.L. Graham, and R.E. Williams. 1968. A flight mill system for the laboratory study of mosquito flight. Ann. Entomol. Soc. Am. 61: 15071514. Scott, T.W., Naksathit, A., Day, J.F., Kittayapong, P. and J.D. Edman. 1997. A fitness advantage for Aedes aegypti and the viruses it transmits when females feed only on human blood. Am. J. Trop. Med. Hyg. 57: 235-239. Timmermann, S.E. and H. Briegel. 1993. Water depth and larval density affect development and accumulation of reserves in laboratory populations of mosquitoes. Bull. Soc. Vect. Ecol. 18: 174-187. Timmermann, S.E. and H. Briegel. 1999. Larval growth and biosynthesis of reserves in mosquitoes. J. Insect Physiol. 45: 461-470. Van Handel, E. 1965. The obese mosquito. J. Physiol. 181: 478-486. Van Handel, E. 1984. Metabolism of nutrients in the adult mosquito. Mosq. News 44: 573-579. Van Handel, E. and J.F. Day. 1988. Assay of lipids, glycogen and sugars in individual mosquitoes:

June, 2001

Journal of Vector Ecology

31

correlations with wing length in field-collected Aedes vexans. J. Am. Mosq. Contr. Assoc. 4: 549550.

Van Handel, E., J.S. Haeger, and C.W. Hansen. 1972. The sugars of some Florida nectars. Am. J. Bot. 59: 1030-1032.