| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
The Anopheles gambiae s.l. complex consists of seven morphologically identical mosquito species. Anopheles gambiae sensu stricto (s.s.), with its two recognized molecular forms M and S, and A. arabiensis are the main malaria vectors in sub-Saharan Africa. Control of these mosquitoes is essential for malaria prevention, but is limited by our incomplete understanding of their population biology. Among the various natural ecologic forces controlling these vector populations, predation on An. gambiae s.l. immature stages seems to be a major factor controlling population sizes because it contributes heavily (between 13.4% and 84.5%1) to overall larval mortality (between 92.6% and 97.1%2–4). However, relatively little research has been devoted to the effects of larval predation on mosquito population structure and ecology, possibly because of difficulties in identifying and quantifying the impact of the most common predators in the field.
To facilitate studies on An. gambiae s.l. larval predation, we have developed an assay that can be used to detect predation on larvae using DNA extracts from known or potential invertebrate predators. By feeding An. gambiae s.l. larvae to known predators and analyzing them at different times after ingestion, we determined the amount of time for which An. gambiae s.l. DNA can be detected in DNA extracts from the predators.
We collected potential An. gambiae s.l. predators from larval breeding habitats in Goundri and Monomtenga (12°30'N, 1°20'W and 12°06'N, 1°17'W, respectively), two rural villages in Burkina Faso. We identified three common families of invertebrates that fed readily on An. gambiae s.l. larvae: Notonectidae (Hemiptera), Libellulidae (Odonata), and Lestidae (Odonata). To determine detection limits of the assay, we starved predators for 24 hours before feeding them a single An. gambiae s.l. larva of known identity, and killed predators in ethanol immediately after ingestion or at various times after ingestion (1, 5, 6, 12, and 24 hours in Notonectidae; 0, 1, 4, 6, and 24 hours in Lestidae; and 0, 1, 6, 8, and 24 hours in Libellulidae). The larvae were obtained from single ovipositions of field-captured female mosquitoes. The specific identity and molecular form status of adult females were determined according to Scott and others5 and Favia and others.6 Because no M/S hybrids were ever reported in these villages (Costantini C and others, unpublished data), taxonomic identification of the mother allowed accurate identification of her offspring.
To determine the best method of predator DNA extraction, we collected Libellulidae and Lestidae from the Mill River in Hamden, Connecticut (41°25'N, 7°25'W), and fed them single laboratory-raised An. gambiae s.s. Legs and heads were removed from predators before DNA extraction. We compared three different DNA extraction methods for whole predators: Qiagen DNEasy (Qiagen, Valencia, CA), Easy-DNA (Invitrogen, Carlsbad, CA), and a modified version of the protocol used by Paskewitz and Collins.7
The modified protocol of Paskewitz and Collins7 showed the most consistent results in terms of ability to detect larval mosquito DNA after ingestion. Using this protocol, we extracted DNA from the preserved predators from Burkina Faso. Each predator was rinsed in sterile water, ground in 200 µL of grinding buffer (0.5 M EDTA, 1 M Tris, pH 7.5, 5 M NaCl, 10% sodium dodecyl sulfate), placed in a 65°C water bath for 30 minutes, vortexed with 28 µL of 8 M potassium acetate, placed on ice for 30 minutes, and centrifuged at 14,000 rpm for 15 minutes. DNA was precipitated from the supernatant with 400 µL of 100% ethanol and 30 minutes on ice. After a final centrifugation (14,000 rpm for 15 minutes) and rinse with 40 µL of 70% ethanol (centrifuged at 14,000 rpm for 5 minutes), samples were dried, resuspended in 300 µL of sterile water (150 µL for Notonectidae because of their smaller mass), and incubated at 65°C for 15 minutes.
Each DNA extract from the predators was subjected to two rounds of polymerase chain reaction (PCR) amplification of a 202-basepair region of the intergenic spacer (IGS) rDNA of An. gambiae s.l., followed by Mse I restriction enzyme analysis of the PCR products. Restriction with Mse I resulted in diagnostic fragments that distinguish between the M (68 basepairs and 134 basepairs) and S (202 basepairs) forms of An. gambiae s.s., and An. arabiensis (62 basepairs and 140 basepairs). The PCR assay is designed for the same IGS rDNA region and sites as used previously,8 but using a different forward primer (783R: 5'-CGTTTCTCACATCAAGA-CAATCAAGTC-3'). This allowed the amplification of a smaller fragment (202 basepairs) than in the previous assay. Given that the prey DNA is fragmented by digestion in the predator, amplification of small DNA fragments should increase the ability to detect prey DNA over time. We used 2 µL of a 1:10 dilution of predator DNA extract in a 15-µL PCR. The PCR mixture and cycling conditions and the restriction enzyme analysis were carried out as described by Santolamazza and others.8 It should be noted that because distinguishing An. arabiensis from An. gambiae M requires separating PCR fragments that differ by only six basepairs, it is necessary to use a high-percentage agarose gel and helpful to run control DNA in each lane for comparison. We were able to differentiate among bands using a 3% agarose gel.
After a single PCR amplification followed by digestion with Mse I (Figure 1
), we detected An. gambiae s.l. DNA immediately after ingestion in Libellulidae at a success rate of 81% (21 samples) and in Lestidae at a success rate of 100% (16 samples). We detected larval DNA from second, third, and fourth instars in both Lestidae and Libellulidae. We did not feed any of the predators first-instar larvae. We obtained a positive result one hour after ingestion in Lestidae (1 sample) but did not detect mosquito DNA in Notonectidae (14 samples) with a single PCR. This may be due to Notonectidae feeding behavior; they inject digestive enzymes into their prey, ingest the liquefied material, and discard the rest of the body. This reduces the amount and quality of larval DNA in Notonectidae compared with Libellulidae and Lestidae, which ingest the entire larva.
|
). These increased detection times in Lestidae and Libellulidae and the ability to detect Anopheles DNA in Notonectidae demonstrate the increased sensitivity obtained by performing two sequential PCRs.
|
This PCR-based assay is more sensitive and less expensive and time-consuming than the precipitin test,1,3 which is unable to distinguish between species and molecular forms within the An. gambiae complex and requires antibody production. This new PCR-based test also represents an improvement over a similar PCR method of Morales and others,9 which reported positive prey detection in a single predator (Libellulidae) one hour after ingestion. In contrast, our assay detects larval DNA up to six hours after ingestion by Libellulidae predators. The use of a single PCR amplification rather than two and the targeting of a larger IGS rDNA region than in our study (390 basepairs versus 202 basepairs) could be responsible for the reduced sensitivity of their method compared with ours.
This assay successfully detects larval DNA immediately after ingestion in all predator DNA extracts but prey DNA detection varies over time and for different predators, ranging from 4 hours to 24 hours after ingestion. This variation in DNA prey detection appears to be related to the predator digestion type (external versus internal), which in turn affects the amount and quality of prey DNA. This PCR-based assay will be useful for future studies of An. gambiae larval ecology, particularly in identifying larval predators in natural breeding sites, and determining whether predation pressure differs among M and S An. gambiae s.s. forms and An. arabiensis. An assay similar to the one described here could also be developed to detect predation on adult mosquitoes.
Further refinements of this method should include optimization of DNA extractions and storage conditions. Another modification of this general protocol that could facilitate field predation studies would include an optimization of prey DNA detection at intervals tailored to the predators’ feeding rates. Detection of prey DNA over an extended period of time can actually bias estimates of the feeding rates. Because this assay can only detect the presence or absence of Anopheles larval DNA, multiple feedings cannot be discerned when ingestion intervals of the predator are shorter than the prey DNA detection time. Although we detected larval DNA in the second through the fourth instars immediately after digestion, we did not test the effect of larval instar on detection after digestion. This would be a useful next step in the development of this assay. In addition, we did not address whether larval DNA can be detected in a predator that consumes another predator that has fed on An. gambiae.
Received September 19, 2006. Accepted for publication May 8, 2007.
Acknowledgments: We thank Antoine Sanou and the staff at the Centre National de Recherche et de Formation sur le Paludisme for assistance with field collections and laboratory work in Burkina Faso, and Dr. L. Zheng (Yale University) for providing larval G3 An. gambiae.
Financial support: This study was supported by a Yale Institute for Biosphere Studies Field Ecology Pilot Grant and the Department of Ecology and Evolutionary Biology Chair’s Discretionary Fund to Erika Schielke, and National Institutes of Health grant R01 AI46018 to Jeffrey Powell.
* Address correspondence to Jeffrey Powell, Department of Ecology and Evolutionary Biology, Yale University, PO Box 208106, New Haven, CT 06520-8106. E-mail: jeffrey.powell{at}yale.edu 
Authors’ addresses: Erika Schielke, Jeffrey Powell, and Adalgisa Caccone, Department of Ecology and Evolutionary Biology, Yale University, PO Box 208106, New Haven, CT, 06520-8106, E-mails: erika.schielke{at}yale.edu, jeffrey.powell{at}yale.edu, and adalgisa.caccone{at}yale.edu. Carlo Costantini, Institut de Recherche pour le Développement, Antenne Institut de Recherche pour le Développement de Bobo-Dioulasso, 01 BP 171, Bobo-Dioulasso, Burkina Faso, E-mail: carlo.costantini{at}ird.bf. Gianmaria Carchini, Department of Biology, Tor Vergata University, Via della Ricerca Scientifica snc, I 00133, Rome, Italy, E-mail: carchini{at}uniroma2.it. N’Falé Sagnon, Centre National de Recherche et de Formation sur le Paludisme, 01 BP 2208 Ouagadougou, Burkina Faso, E-mail: n.fale.cnlp{at}fasonet.bf.
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
