| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
 |
ABSTRACT |
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Anopheles gambiae and its sister species Anopheles arabiensis, the two principal vectors of malaria throughout much of Africa, display distribution patterns characterized by different levels of aridity.6,7 Anopheles arabiensis generally occurs in more arid habitats than An. gambiae. The latter species is comprised of several chromosomal forms that carry different strain-specific combinations of inversions. Various field-based studies have demonstrated that the forms also show distinct patterns of seasonal and geographic distributions.1,4,8–10 These studies provide evidence that supports the role of some chromosomal inversions (or the genetic characters statistically associated with them) in conferring tolerance to arid conditions.11 Ecologic, biogeographic, and molecular phylogenetic evidence suggests that these inversions have been transferred to An. gambiae from An. arabiensis.12
No physiologic measurements have been made to date quantifying desiccation resistance or relevant parameters related to water balance in adult anopheline mosquitoes. Investigators have suggested that the seasonal distribution of the various mosquito populations might be related to larval habitat availability, larval habitat type, and migrations between suitable habitats.3,13,14 Anopheline eggs have been shown to survive on moist soil for up to two weeks.15 No evidence of estivation has been found in these populations. However, the most frequently proposed hypothesis for the differential distribution of An. gambiae and An. arabiensis, both geographically and seasonally, is that of differential desiccation tolerance in the adults.9,11,12
In this study, we compared the desiccation resistance of female mosquitoes from two stable laboratory colonies, one of An. arabiensis and one of An. gambiae S molecular form. Males were not included in the present study. The contribution of males to the persistence of the population occurs early in their adult life.16 Conversely, females with enhanced resistance to aridity would have an increased ability to host seek, blood feed, and lay eggs during the dry season. Therefore, their potential as a vector and their reproductive success are linked to surviving longer in this stressful environment.
We predicted that a species from a stress-prone environment is more resistant to that stress. In this scenario, An. arabiensis should be more desiccation resistant than An. gambiae. From a physiologic viewpoint, desiccation resistance in insects is determined by initial water content, rate of water loss, and water content at death.17,18 We examined each of these parameters with regard to their role in differential desiccation resistance. Also, reduced resting metabolic rate has been proposed as an evolved mechanism of desiccation resistance in insects and other animals.19 We measured the resting metabolic rate of both species to determine whether this trait might be involved in adaptation to aridity of An. gambiae and An. arabiensis.
|
MATERIALS AND METHODS |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Both populations were reared in the same conditions at 28°C, a relative humidity of 70–80%, and a 12:12 light-dark cycle. Mosquito colonies were maintained by blood feeding on non-anesthetized mice, as authorized by an animal use protocol approved by the Institutional Animal Care and Use Committee of the University of California, Irvine. Larvae were reared in shallow plastic trays at a density of 150 per liter of tap water and fed a mixture of Tetramin® (Tetra Werke, Melle, Germany) fish food and yeast. Pupae were placed in pint-sized paper cups with a mesh cover and left to emerge. Adults had access to a cotton ball saturated with a 10% sucrose solution. Mosquitoes tested on the day of emergence were not provided sucrose.
Desiccation resistance and water content at death. The desiccation resistance and water content at death of female mosquitoes were tested at emergence (n = 20), four days (n = 40), and eight days (n = 20) post-emergence. For this experiment, each mosquito was placed in an individual 8-dram glass vial. A foam stopper was inserted one-third down the length of the vial. Drierite (W.A. Hammond Drierite Co., Ltd., Xenia, OH) was poured on the stopper and Parafilm® (American Can Company, Greenwich, CT) wrapped around the vial to seal it. The drierite served to absorb the water from the air inside the vial, creating an environment where the relative humidity was less than 10%. Viability was assessed visually at hourly intervals. We considered an individual dead when it could not fly or right itself. Death in this case refers to a state in which normal locomotory and righting functions are disrupted, although some capacity for response might remain. Upon determination of death, each mosquito was immediately weighed, placed in a 70°C oven overnight, and reweighed. The difference between wet body mass at death and dry mass represents water content at death. The time to death represents the desiccation resistance of the individual.
Change in body composition, resting CO2 release rate, and respiratory pattern characteristics with age. We measured water content, dry mass, lipid content, rate of CO2 release (VCO2) at rest and respiratory pattern of individual female mosquitoes at emergence and 2, 4, 6, 8, and 10 days post-emergence. Fourteen females of each species were used per age group. First, resting VCO2 was determined via flow-through respirometry. Mosquitoes were placed in small 3-mL chambers and subjected to a constant flow of dry air at a rate of 100 mL/minute. Following an acclimation period of 90 minutes, measurements of CO2 release were made at one-second intervals for five minutes per chamber, using a CO2-H2O gas analyzer (Li-Cor model 6262, Li-Cor Biosciences, Lincoln, NE). Release of CO2 in mosquitoes at rest occurs cyclically. Variability in the release of CO2 for each female was measured by determining the standard error of the mean over a five-minute time period. Following the respiratory measurements, each mosquito was weighed, dried in a 70°C oven overnight, and weighed again to obtain dry mass and water content (also referred to as initial water content, without prior desiccation). Lipid content was determined by measuring dry weight, extracting each carcass overnight in petroleum ether, and reweighing the dry mass following extraction. The difference in weight was attributed to lipid.
VCO2 and water loss rate during desiccation. We followed the VCO2 and water loss rate of individual females until death by desiccation, using flow-through respirometry and a CO2-H2O gas analyzer (Li-Cor 6262, Lincoln, NV). Death in this circumstance refers to a VCO2 of zero. Twelve individuals of each species were measured at emergence and four days post-emergence. Mosquitoes were placed in small 3-mL chambers and subjected to a constant flow of dry air at a rate of 50 mL/minute. For each chamber, values were recorded at one-second intervals for 15 minutes every 2 hours.
Statistical analyses. We compared desiccation resistance and water content at death of both species at each age group by a t-test and between age groups by analysis of variance. For body composition and resting VCO2 we performed one-way analysis of variance. We compared both species on each day and pooled all days together to compare both species overall. For VCO2 and water loss rate during desiccation, we performed analysis of covariance for each variable and each age group using time as a covariate.
|
RESULTS |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and Figure 1). The desiccation resistance of both species was higher just after emergence than it was later in adult life (P < 0.001).
|
|
). In both species, relative water content at death significantly decreased with age (P < 0.0001). Females of both species died when their relative water content reached approximately 2 mg/mg of dry mass at emergence, 1.7 mg/mg of dry mass at 4 days post-emergence, and 1.4 mg/mg of dry mass at 8 days post-emergence.
Changes in body composition and resting VCO2 with age.
We measured VCO2 at rest, variability of the resting CO2 release pattern (represented by the SE of the mean VCO2), dry mass, water content, and lipid content of female mosquitoes at several ages between emergence and 10 days post-emergence (Figure 2). When all days were pooled, we found no difference in resting VCO2 between the two species (P = 0.070), even when the values were corrected for differences in dry mass (P = 0.738). The difference in resting VCO2 seen on day 6 is marginally significant (P = 0.046). Overall, An. arabiensis had a higher dry mass and water content (P = 0.044 and P < 0.0001, respectively) than An. gambiae. The difference in dry mass was visible at most ages, although not large enough to be significant in any one age group. Water content was significantly higher in An. arabiensis than in An. gambiae (P < 0.010) at emergence, 6 days post-emergence, and 10 days post-emergence. The difference in dry mass could be a confounding factor responsible in part for the difference in water content. To determine whether An. arabiensis contain more body water relative to their dry mass, we calculated dry mass-specific water content. We found that An. arabiensis still have more water per unit dry mass than An. gambiae (P = 0.003). Lipid contents were similar between the species (P = 0.440).
|
). Sample respiratory patterns of females at emergence and four days post-emergence are shown in Figure 3. We found no significant difference in the SE of the mean between both groups (P = 0.962). For both species, values were lower at emergence than later in adult life (P < 0.001).
|
). The water loss rates measured do not distinguish between cuticular and respiratory water loss. We found that mosquitoes at emergence had similar water loss rates during desiccation (P = 0.052). Furthermore, the water loss rates of both species seemed constant and equal throughout most of the stress and increased, to a different extent in each species, shortly before death. Mosquitoes at emergence also had similar VCO2 during desiccation (P = 0.139). Four-day-old mosquitoes had a more variable water loss rate during desiccation, but overall there was no significant difference between species (P = 0.839). At four days post-emergence, An. gambiae had a significantly lower VCO2 during desiccation than did An. arabiensis (P = 0.008). Excretory events were observed during these experiments but did not appear to differ in frequency or amplitude between the experimental groups.
|
|
DISCUSSION |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Desiccation resistance in small insects can be quantitatively partitioned into three important parameters: initial body water content, rate of water loss, and water content at death. Initial body water represents water stores in the hemolymph and in the tissues. Water loss occurs mainly through excretion, respiration, and diffusion across the cuticle. Water content at death indicates the critical body water content below which survival is no longer possible. We measured all three parameters in An. gambiae and An. arabiensis and found that these species differ only in initial body water content.
The source of the observed difference in water content can involve several mechanisms that are not mutually exclusive. First, An. arabiensis might have a higher hemolymph volume than An. gambiae. This has been found in Drosophila selected for enhanced desiccation resistance.22 Such a physiologic adaptation requires an increase in the amount of hemolymph solutes to maintain homeostasis. Female mosquitoes are faced with a large flux of fluid through the hemolymph at each blood meal. Anopheles ingest more than their weight in blood and although part of the blood meal is excreted via the anus during feeding, the larger portion of the fluid ingested is transferred to the Malpighian tubules via the hemocoel.23 Since female mosquitoes are accustomed to managing large fluid volumes, adaptation to desiccating conditions by increasing hemolymph volume is not improbable. Hemolymph volume is technically difficult to measure in such small insects, but has been done in Drosophila gravimetrically and by several other methods, such as injection of radioactive markers.17,22
Another possible avenue of enhanced water storage is water bound to glycogen and stored intracellularly. This energy storage compound can bind up to five times its weight in water, and it has been suggested that insects can store more water by favoring energy storage in the form of glycogen rather than lipid.24,25 Anopheles arabiensis contain an average 0.18 mg more water than An. gambiae. Theoretically, the amount of glycogen necessary to store this additional water is 0.036 mg.24 This would require practically all the dry mass difference observed between species to be due to glycogen alone (An. arabiensis have on average 0.042 mg more dry mass than An. gambiae). If the difference in dry mass between both species is due to glycogen, then that amount of glycogen is sufficient to explain the difference in water content. Therefore, glycogen cannot be ruled out as a reasonable candidate for permitting the storage of intracellular water.
Finally, An. arabiensis could simply replenish their crop content more often than An. gambiae. When mosquitoes feed on a source of sugar, the contents of this meal enters the crop and is slowly passed to the midgut to be absorbed little by little. Replenishing this fluid more often might enhance chances of survival in a dry environment. Khan and Maibach found that thirsty mosquitoes feed more often, suggesting that the crop provides a reservoir for water as well as sugar.26 It would be interesting to compare the feeding frequency of both species to determine whether An. arabiensis have evolved a modified feeding behavior in response to a more desiccating environment.
We found no difference in water content at death between An. arabiensis and An. gambiae. Various studies of Drosophila have demonstrated identical water contents at death in 20 separate populations of Drosophila that had undergone different selection regimens, including strong selection for enhanced desiccation resistance for more than 150 generations.18,22,25 Those results clearly indicate that Drosophila can show evolved changes in multiple parameters strongly affecting desiccation resistance, but water content at death is not one of them. It is possible that the terrestrial environment has pushed tolerance of low water content to its limit early on in the evolution of these small dipterans, resulting in no genetic variability in this trait.
We followed the water loss rate and VCO2 of individual females during desiccation. The VCO2 is a measure of the metabolic expenditure of an animal and a value higher than resting VCO2 may reflect an increase in activity, or alternatively, an increase in non-locomotory energy expenditure during the desiccation stress. Increases in respiratory water loss rate parallel the rate of gas exchange associated with activity. Thus, activity levels during a desiccation stress might significantly influence desiccation resistance.27,28 However, interspecific comparisons have shown that respiratory water loss contributes more to total water loss in xeric species than in mesic species. The reason for this is that xeric species have a less permeable cuticle, with the consequence that water loss is proportional to metabolic rate in xeric but not in mesic species.29 Cross-species comparisons in Drosophila have shown water loss rate to be negatively correlated with desiccation resistance.25,30,31 Gibbs and others showed that Drosophila from mesic and xeric habitats differed in water loss rate, but not in metabolic rate, body water content, or water content at death.30 Although they could not demonstrate differences in the characteristics of cuticular lipids in the species studied, it is nevertheless possible that more xeric species had a less permeable cuticle than mesic ones. We found in this study no difference in water loss rate or VCO2 between species during desiccation at emergence. At four days, An. gambiae had a lower VCO2 than An. arabiensis but the same water loss rate. Within each species, we found no correlation between VCO2 and water loss rate. Our results suggest no difference in respiratory or cuticular water loss between these two closely related mosquito species.
Using our data for desiccation resistance as well as mass of body water prior to and following desiccation, we estimated the average water loss rate of these mosquitoes in still dry air. Figure 5 shows a plot created using dry mass-specific initial water content, dry mass-specific water content at death, and desiccation resistance values for each group. The slopes obtained represent the mean rate of water loss during desiccation in still dry air. The values were nearly identical for both species at emergence (slope = –0.057 corresponding to 57 µg of water/hour) as well as at four days post-emergence (slope = –0.020). This reinforces the conclusion from the flow-through desiccation experiment that water loss rates are not different between the populations studied.
|
Our results highlight some notable differences in respiration and water balance between newly emerged and older adult mosquitoes. We found that both species have a higher desiccation resistance, lower amplitude of respiratory pattern, and lower water loss rates at emergence compared with later ages. Mosquitoes are very vulnerable in the hours following emergence. After exiting their pupal case, they must expand their appendages, diurese, and perhaps make a short flight to a resting place. In the next few hours, their cuticle hardens. In two days, maturation of several organs such as salivary glands, midgut epithelium, and flight organs occurs.37 Our emergence experiments were performed on females that were at most a few hours old. These mosquitoes are not yet ready to mate or seek hosts. They have a relatively low resting VCO2 at this time. The low amplitude of their respiratory pattern suggests that they are not controlling gas exchange as efficiently as they do at later ages, perhaps because the cuticle is not yet fully hardened. However, in response to a desiccation stress, emergence mosquitoes survive longer. As shown in Figure 5, they have more body water relative to their dry mass than do four-day-old mosquitoes. Figure 4 shows that their water loss rates are lower and more constant than those of older mosquitoes. Their quiescent behavior brought about by the maturation period and the extra water they contain relative to their mass both contribute to their enhanced survival.
In conclusion, our study indicates that the variation in desiccation resistance among these anopheline species is due to body water content. Anopheles arabiensis carry more water than An. gambiae. Our populations are laboratory colonies that were created from field populations belonging to two sister species recently diverged.12 Differential desiccation resistance has persisted in these populations despite hundreds of generations of laboratory rearing at high relative humidity. These results were obtained in mosquitoes reared under identical conditions, indicating that the physiologic differences are genetically based. Evidence accumulated over the past 20 years suggests that some chromosomal inversions, fixed in An. arabiensis but polymorphic in An. gambiae, are somehow linked to aridity tolerance. This study provides the first physiologic evidence of differential resistance to aridity in the adults of these species and an indication of the mechanisms involved. We now need to extend these physiologic studies to the different chromosomal forms of An. gambiae s.s. By doing so, we hope to characterize and quantify the physiologic specializations of these disease vectors. This work is also a necessary step towards the understanding of the genetic basis of physiologic traits that confer tolerance to aridity in An. gambiae and of the possible involvement of gene suites carried within chromosomal inversions.
Received March 8, 2005. Accepted for publication April 13, 2005.
Acknowledgments: We thank Marc Benedict (Entomology Branch, National Center for Infectious Diseases, Centers for Disease Control and Prevention, Atlanta, GA) and Anthony James (Departments of Microbiology and Molecular Genetics and Molecular Biology and Biochemistry, University of California, Irvine, CA) for providing mosquito eggs for the establishment of colonies. We also thank the anonymous reviewers for their comments.
Financial support: This work was supported by a Graduate Assistance in Areas of National Need fellowship to Emilie M. Gray.
* Address correspondence to Emilie M. Gray, Department of Ecology and Evolutionary Biology, University of California, Irvine, CA 92697-2525. E-mail: djemilie{at}gmail.com 
Authors’ address: Emilie M. Gray and Timothy J. Bradley, Department of Ecology and Evolutionary Biology, University of California, Irvine, CA 92697-2525, Telephone: 949-824-7038, Fax: 949-824-2181, E-mails: djemilie{at}gmail.com and tbradley{at}uci.edu.
|
REFERENCES |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. B. Benoit and D. L. Denlinger Suppression of water loss during adult diapause in the northern house mosquito, Culex pipiens J. Exp. Biol., January 15, 2007; 210(2): 217 - 226. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. M. Gray and T. J. Bradley Evidence from mosquitoes suggests that cyclic gas exchange and discontinuous gas exchange are two manifestations of a single respiratory pattern J. Exp. Biol., May 1, 2006; 209(9): 1603 - 1611. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
