|
|
||||||||
| ABSTRACT |
|
|
|---|
| INTRODUCTION |
|
|
|---|
Meiotic drive, an example of a common gene drive system, subverts meiosis in favor of a particular gamete, and has been reported in a wide variety of organisms including, plants, fungi, insects, and mammals.16,17 During meiosis, the meiotic driver product blocks maturation of gametes bearing a sensitive responder locus. The driver gene and responder locus are located on opposite members of homologous chromosomes, respectively. Because driver-carrying chromosomes selectively destroy their homolog, they have the potential for increasing in frequency in natural populations despite possible deleterious effects on fitness. The best characterized meiotic drive system is Segregation Distorter (SD), discovered in 1956 in Drosophila melanogaster.18 The major components of SD are the driver gene (Sd), an enhancer (E(SD)), and the responder (rsp). These reflect strong linkage disequilibrium due to suppression of recombination across the peri-centromeric region of chromosome 2, near the eye color genes bw and cn.19 During spermatogenesis, the Sd gene causes a failure in chromatin condensation of sensitive responder-bearing chromosomes. Thus, the phenotype of the driver can be detected by biased eye color ratios among the progeny.
The existence of an endogenous meiotic drive system has been reported in two mosquito species, Aedes aegypti and Culex pipiens.20,21 Meiotic drive in these mosquitoes is tightly linked to the sex determining locus on chromosome 1 and, therefore, it is a sex linked meiotic drive system rather than an autosomal one as in D. melanogaster. The driver gene (D) of Ae. aegypti is tightly linked to the male determining allele (M), and the responder locus is tightly linked to the female determining allele (m) on the homologous chromosome. Neither the driver gene linked to a female determining allele nor the responder linked to the male determining allele function properly to produce the drive phenotype.22 In males heterozygous for the driver (MD) and a sensitive responder (ms), the products of the MD gene cause fragmentation of ms-bearing gametes during spermatogenesis. Therefore, the observed phenotype of the meiotic drive system in Ae. aegypti is a highly male-biased sex ratio.
Introducing strong driver males into a sensitive mosquito population has been proposed as a mechanism to reduce population size. However, the use of meiotic drivers to suppress Ae. aegypti populations has been investigated with laboratory cage studies with limited success. Hickey and Craig observed quick reversion of the sex ratio after introducing driving males into sensitive female cage populations.23 After reversion of distorted sex ratio, driver alleles were still present and active in the population. They suggested accumulation of a suppressor reducing the sensitivity of the responder, which was identified as the tolerance gene on chromosome 1 by Wood and Ouda.24 Therefore, the meiotic drive gene may impart strong selective pressure for driver suppressors, thereby desensitizing females to meiotic drive. For a successful application of the MD system, two alternative ways of using the mosquito meiotic drive system were suggested and tested successfully. Wood and others tested population replacement potential in the MD system using an eye color marker gene.25 Curtis and others26 and Suguna and others27 combined the meiotic drive system with the double translocation homozygote technique and tested for population suppression potential. Although both cases were successful in cage experiments, there were no field trials. These studies showed that MD has a strong potential for population replacement and suppression in spite of its strong negative selection potential against drive sensitivity.
Release of an Ae. aegypti strain carrying a strong meiotic driver and an insensitive responder (MDmi) into a sensitive population should have strong selection potential for both the driving chromosome and its insensitive homolog. The potential use of the MD system for population replacement could include strategies that couple a desired transgene with either a drive insensitive mi allele or with the MD gene itself.28 In this study, we investigated the population replacement potential of the MD system in laboratory cage populations. We used the Ae. aegypti T37 strain that carries a strong meiotic driver and insensitive responder, and the drive-sensitive RED strain to maintain three population types.23,28,29 The first population type consisted of 100% driving males and 100% sensitive females, and the second and third population types were initiated with 10% and 1% driver males, respectively. With the latter two populations, we subsequently added 10% and 1% driver males to them every other generation. We also selected one strain diagnostic restriction fragment length polymorphism (RFLP) marker on each chromosome, including one marker on chromosome 1 that is tightly linked to the MD system as a proxy for examining replacement potential. Our results suggest that the strong negative selection potential of the MD gene can be used as a strong population replacement strategy.
| MATERIALS AND METHODS |
|
|
|---|
The effect of the MD gene on A. aegypti population dynamics was evaluated across three experimental population types designed to examine the effects of the meiotic driver on sex ratio distortion and allele replacement potential, and maintained until the F15 generation. We established three experimental population types that were initiated with 100% (500), 10% (50), and 1% (5) male mosquitoes carrying a strong meiotic driver (MDmi: T37 strain) and 500 sensitive females (msms: RED strain). Thus, the 10% cages and 1% cages were initiated with 10% and 1% driving males and 90% (450) and 99% (495) non-driving males (Mdms: RED strain), respectively. At every even numbered generation (F2, F4, F6...F14), we added 10% (50) and 1% (5) T37 males to the 10% and 1% cage populations respectively, which simulates repeated introductions of a meiotic driver into natural populations. Each population type was replicated three times. At each generation, 1,0001,500 eggs per cage were hatched to determine sex ratios and initiate the next generation. Observed sex ratios were tested for a 1:1 sex ratio using chi-square analysis.
Genotyping and allele frequencies.
One strain diagnostic RFLP marker was identified on each chromosome to track the change of allele frequencies associated with the meiotic drive system using our standard Southern blotting procedures.30,31 An RFLP marker closely linked to the sex determining locus on chromosome 1 was identified as a proxy to track the MD and mi alleles, respectively. Bulked genomic DNA mixtures from 30 mosquitoes from each strain were digested with EcoRI at 37°C for three hours, separated by electrophoresis on a 0.9% agarose gel, and transferred onto Hybond N+ filters (Amersham Biosciences, Piscataway, NJ). Radiolabeled probe was synthesized using
-P32 dCTP and polymerase chain reaction amplicons of previously mapped RFLP markers as templates.30 The RFLP markers, which showed no common alleles between the two strains, were selected and used for genotyping the cage populations over multiple generations.
We extracted genomic DNA from 72 males and 72 females per cage for selected generations. Genomic DNA extractions were digested with EcoRI at 37°C for two hours, separated by electrophoresis on a 0.9% agarose gel, and transferred onto Hybond N+ filters. We genotyped and calculated allele frequencies using strain diagnostic RFLP markers as radio labeled probes up to the F15 generation using our standard protocol.31 To test for the effect and frequency of the tolerance gene, which is known to be linked to the red-eye locus, we also determined the eye color of 200 F15 males from the 100% cage populations.24
| RESULTS |
|
|
|---|
|
|
10% cage populations. The 10% T37 male cage populations showed a consistent increase in male frequencies until peaking at 75.13 ± 4.9% at the F6 generation. Male frequencies remained at fairly constant levels from the to F7 F13 generations, then decreased to less than 70% males after the F13 generation.
1% cage populations. The 1% cage populations remained relatively constant at approximately 54% males until the F5 generation. However, beginning about the F6 generation, male frequencies slowly but constantly increased, and exceeded 65% by the F15 generation.
Effect of MD and mi on RFLP allele frequencies.
Selected strain-diagnostic RFLP markers were LF284 (chromosome 1), LF355 (chromosome 2), also known as RpL17A, and LF323 (chromosome 3) (Figure 2
). LF284 is tightly linked with the sex determining locus, and the other two markers are autosomal. Therefore, the frequency of the T37-LF284 allele in the male population is highly correlated with frequency of MD and mi, respectively, while in the female population, the T37-LF284 allele represents the mi allele frequency. The RED-LF284 allele in the male population is highly correlated with frequency of Md and ms, and the RED-LF284 allele in the female population represents the ms allele frequency, respectively. The observed changes in allele frequencies for each marker in each cage are shown in Figure 3
.
|
|
10% cage populations.
In the 10% cage populations, the T37-LF284 allele frequencies increased in both male and female populations (Figure 3C and D
). However, the male population showed a quicker change and attained much higher frequencies than female populations. The T37-LF355 and T37-LF323 allele frequencies gradually increased in the early generations, and fluctuated between 0.2 and 0.3 after the F6 generation in both male and female populations.
1% cage populations.
The 1% cage populations showed similar patterns of changes in T37-LF284 allele frequencies to the 10% cage populations, although at reduced levels comparatively (Figure 3E and F
). The T37-LF284 allele frequencies increased from 0.01 at the F1 to 0.34 at the F15 generations in the male populations. The T37-LF284 allele frequencies increased in female populations but at lower levels than observed in male populations. The T37-LF355 and T37-LF323 allele frequencies increased slightly, but remained at frequencies less than 0.1 in both male and female populations across all generations.
Impact of MD and mi in cage populations.
The impact of a meiotic driver on the observed sex ratio at each generation is determined by the frequency of driving males (MDms) in the previous generation. Furthermore, the MDms frequency at each generation is determined by the allele frequency of the sensitive responder (ms) in female populations of the previous generation. Therefore, the extent of a male biased sex ratio at each generation is highly correlated with the allele frequency of sensitive responders in the population of its grandmother generation (Figure 4
). This indicates that increased mi allele frequencies in the 100% cages are responsible for the reduction in male frequencies at later generations.
|
|
| DISCUSSION |
|
|
|---|
In our 100% cage populations, after mating T37 males with RED females, all F1 males and females were heterozygotes with genotypes MDms and mims, respectively. The products of the MD allele in the F1 males acted on the sensitive responder, and therefore, the F2 generation showed maximum maleness. After peaking at the F2 generation (84.8 ± 0.99%), male frequencies decreased gradually to 61.2 ± 0.72% at the F15 generation. Both males and females in the 100% cage populations showed quick increases in T37-LF284 allele frequencies, which represents the increase of mi allele frequencies. The decrease in male frequencies is highly proportional to the increase of mi allele frequencies. That is, mi replaced ms at the later generations, and the male-biased sex ratios gradually reverted back to approximately 1:1.
Previously, Wood and Ouda reported quick selection of tolerance genes originating from the RED strain under driving conditions.24 The tolerance gene is linked to the red-eye locus on chromosome 1 and suppresses responder sensitivity to a driver. Although the maximum distortion of the male frequency at the F2 generation (84.8 ± 0.99%) was reduced to 61.2 ± 0.72% at the F15 generation, strong selection of tolerance genes linked to the red-eye locus was not observed in the 100% cages. This implies that the insensitive responder originating from T37 males may have higher fitness than the tolerance gene originating from the sensitive RED strain under driving conditions in 100% cage populations or that the tolerance gene is no longer segregating in our RED strain. Increased frequencies of the insensitive responder, which was absent in the cage populations of Wood and Ouda, at the later generations in the female populations of the 100% cages probably served as the primary reason for reversion of the male biased sex ratio back to 1:1 under driving conditions. However, the T37-LF284 frequencies in female populations decreased after the F12 generation in the 100% cages. The observed decrease may be due to random sampling variance or recombination between the responder locus and our diagnostic RFLP marker. Although unlikely, it may be due to selection for driver suppressors as Wood and Ouda suggested.24
Repeated introduction of 50 MDmi males at every even numbered generation greatly influenced the increase of T37-LF284 allele frequency in the 10% cage populations. Although our 10% cage populations were initiated with only 10% MDmi males, the allele frequency of T37-LF284 quickly increased in both male and female populations (Figure 3C and D
), which implies quick increases of both MD and mi in the populations. However, the increase of T37-LF284 allele frequencies were greater in males than in female populations. The MD alleles quickly spread in sensitive populations due to the higher male-biased segregation ratios than observed with normal male determining alleles. As the frequencies of the MD allele increases in the 10% cage populations, mi alleles have strong selection potential. Conversely, mi alleles do not have strong selection potential under low driver frequencies. In addition, after the F6 generation obvious increases in the T37-LF284 allele frequency are evident in the 1% cage populations, which were supplemented with only five MDmi males at every even numbered generation (Figure 3E and F
). Of note are the obvious increases in the T37-LF284 alleles among females.
Our results suggest that under the influence of a strong meiotic driver, the alleles linked with the mi allele on chromosome 1 are effectively replacing those linked with the ms allele. Although introducing MDmi males into sensitive populations has proven unsuccessful as a tool for population suppression due to the strong selection potential of MD for mi alleles, genes linked to MD and mi would have strong population replacement potential. The distribution of the meiotic driver in Ae. aegypti populations does not appear to be worldwide,32 which may imply a short evolutionary history. This suggests that release of MDmi males into sensitive populations can be an effective tool for population replacement. However, Wood reported between family variation in sex ratios within the Trinidad-derived Ae. aegypti T-30 strain and suggested that there were at least six different alleles at the responder locus with varying sensitivities to the driver.33 Clearly, a better understanding of the complexities of interactions among the drive system components is needed. Our present study provides important empiric data that will facilitate additional studies, including experimental population studies and simulations by mathematical modeling.
Mori and others suggested two strategies for using the MD system for introducing novel genes into natural populations.28 One strategy would be to insert a desired transgene adjacent to or within an mi allele on chromosome 1, and then use selection by a strong driver to increase the transgene frequency in naive, drive-sensitive populations. The second strategy would be to couple the transgene with a strong driver gene, and then insert this construct into an autosome. Release of males carrying the driver/transgene construct and homozygous for mi alleles would result in selection for both, thereby effecting population replacement with the transgene. Therefore, meiotic drive seems particularly advantageous because it represents a relatively stable system that has the potential to be linked with effector genes that influence vector competence by transgenic insect techniques without significant risk for horizontal transfer in natural populations.17
Aedes aegypti is the primary vector for the yellow fever and dengue fever viruses. Efforts to test the utility of meiotic drive as a novel vector control strategy will require the molecular characterization of the MD system. The segregation distorter (SD) system in D. melanogaster is the best characterized meiotic driver.34 The SD gene has been determined to be the result of a 3'-truncated tandem duplication of the RanGAP (Ran GTPase activation protein) gene.35 Kusano and others reported that the mislocalization of RanGAP was responsible for the meiotic drive phenotype.34 However, additional studies are needed to test whether expression of truncated Ran-GAP will result in meiotic drive in Ae. aegypti. Alternatively, other genes involved in the Ran signaling pathway could be considered as candidates for MD.36
Received June 3, 2005. Accepted for publication August 25, 2005.
Financial support: This work was supported by grant PO1-AI45123 from the National Institute of Allergy and Infectious Diseases, National Institutes of Health.
* Address correspondence to David W. Severson, Center for Tropical Disease Research and Training, Department of Biological Sciences, University of Notre Dame, Notre Dame, IN 46556. E-mail: severson.1{at}nd.edu ![]()
Authors addresses: Sung-Jae Cha, Akio Mori, and David W. Severson, Center for Tropical Disease Research and Training, Department of Biological Sciences, University of Notre Dame, Notre Dame, IN 46556, Telephone: 574-631-3826, Fax: 574-631-7413, E-mail: Severson.1{at}nd.edu. Dave D. Chadee, Department of Life Sciences, University of the West Indies, St. Augustine, Trinidad and Tobago, West Indies, E-mail: chadee{at}tstt.net.tt.
Reprint requests: David W. Severson, Center for Tropical Disease Research and Training, Department of Biological Sciences, University of Notre Dame, Notre Dame, IN 46556.
| REFERENCES |
|
|
|---|
This article has been cited by other articles:
![]() |
S.-J. CHA, D. D. CHADEE, and D. W. SEVERSON Population dynamics of an endogenous meiotic drive system in aedes aegypti in trinidad. Am J Trop Med Hyg, July 1, 2006; 75(1): 70 - 77. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |