|
|
||||||||
| ABSTRACT |
|
|
|---|
| INTRODUCTION |
|
|
|---|
In an attempt to study this question, we focused on one of the major components of the immune system of mosquitoes: the melanization response.7 This immune response received considerable attention as a potential resistance mechanism against Plasmodium after the artificial selection of a refractory line of the mosquito Anopheles gambiae that melanizes the parasite at the late ookinete stage.8,9 This attention was revived by the recent identification of several mosquito genes that are to some extent associated with the melanization of malaria parasites in A. gambiae.1013
However, although it is clear that melanization can lead to resistance against malaria infection at least after a selection treatment in the laboratory,8,9 it is very rare to find mosquitoes in natural populations that have melanized their parasites. For example, in one study of field-caught A. gambiae in Tanzania, < 0.5% of infected mosquitoes harbored melanized oocysts of P. falciparum.14 This is all the more surprising because, in the same study, ~90% of the mosquitoes readily melanized negatively charged Sephadex beads inoculated into the insects hemolymph.14
Thus, the lack of P. falciparum melanization by wild A. gambiae does not seem to result from a general lack of immuno-competence of the mosquitoes but rather from a mechanism allowing the parasite to bypass this immune response. One possible explanation is that the parasite is able to hide from the immune response (i.e. to evade it). For example, mosquito-derived proteins incorporated into the oocyst capsules of the avian malaria parasite Plasmodium gallinaceum might mask developing oocysts from the mosquitos immune system.15 The observation that the absence of two mosquito C-type lectins, CTL4 and CTLMA2, results in massive melanization of parasites in a susceptible line of A. gambiae supports the idea that the parasite can use some molecules produced by the mosquito to evade the immune response.10 Alternatively, the parasite might be able to suppress or at least reduce the effectiveness of the melanization response. Indeed, host immuno-suppression by pathogens is often observed, for example, inhibition of immune hemocyte function by insect polydnaviruses,16 blocking of host cell phagocytosis by Yersinia pseudotuberculosis,17,18 disruption of plant immune signalling by a fungal pathogen,19 or bacterial suppression of antibiotic peptides synthesis in Drosophila.20 In particular, both a direct and an indirect immuno-suppression of the melanization response has been shown in Aedes aegypti mosquitoes infected by P. gallinaceum.21,22 However, despite the indication that the expression of one A. gambiae immune gene, NO synthase (NOS), is repressed after P. falciparum infection,3 it is unknown whether P. falciparum can suppress the melanization immune response of Anopheles mosquitoes.23
To evaluate the possibility of suppression of the melanization response by P. falciparum in A. gambiae, we compared the melanization ability of three classes of mosquitoes: those that became infected after an infectious blood meal, those that did not become infected (or cleared the parasite) after the blood meal, and those that were fed on an uninfected blood meal. Like previous studies,21,22,24 we quantified the melanization response by inoculating negatively charged Sephadex beads into the insects hemolymph. The infection status of challenged mosquitoes was determined at 48 hours post-blood meal (pbm) by detection of Pfs25 mRNA.25,26 Bead melanization was measured between 24 and 48 hours pbm during the transition between P. falciparums ookinetes and oocysts, the developmental period of the parasite that is most susceptible to the melanization response,8 and when this response is most likely to be suppressed by P. gallinaceum.21,22
| MATERIALS AND METHODS |
|
|
|---|
Mosquitoes. The mosquitoes came from a colony of A. gambiae s.s. originating from Yaoundé that had previously been adapted to feed through Parafilm membrane.27 We recorded the mosquitoes wing lengths, measured from the tip (excluding the fringe) to the distal end of the alula with a precision of 0.04 mm, as an indication of body size.28 Where both wings could be measured, the mean of the two lengths was used.
Gametocyte carriers. Plasmodium falciparum carriers were recruited among 5- to 11-year-old children from primary schools in Mfou, a small town located 25 km east of Yaoundé, Cameroon. Thick blood smears were made from finger-prick blood samples, stained with 10% Giemsa for 20 minutes, and examined microscopically for P. falciparum gametocytes. Asymptomatic gametocyte-positive children were selected for the study after their parents or guardians had signed an informed consent form. Gametocyte carriers with malaria species other than P. falciparum were excluded from the study. All children with asexual parasitemia (> 1,000 parasites/µL) were treated with an artemisinin-amodiaquine combination on the day after the screening, according to national guidelines for the treatment of simple malaria cases. The protocol obtained approval of the National Ethics Committee of Cameroon.
Experimental infections. Gametocyte carriers that had been identified the previous day were brought to the laboratory. A sample of 5 mL of venous blood was collected from each gametocyte carrier in a heparinized tube. Gametocyte density per 500 leukocytes was assessed just before blood withdrawal on a blood smear (as described above) and was converted to numbers of parasites per microliter by assuming a standard leukocyte count of 8,000/µL. To standardize the blood meal, we centrifuged the blood at 37°C for 3 minutes at 2,000g and replaced the autologous serum with non-immune AB serum (the same AB serum was used for all infections), adjusting the hematocrit to 50%. A sample of venous blood from an uninfected human volunteer was treated in the same manner and used as a control blood meal. Three-day-old female mosquitoes deprived of sugar for 5 hours before blood feeding were allowed to feed on the mixture for 30 minutes through standard membrane feeders.27 In each experimental block, replicate groups of 1535 females maintained in paper cups covered with netting were simultaneously placed under membrane feeders containing either infected blood or uninfected blood. At least two different feeders were used for each infected or control blood sample, so that we could control for a potential effect of the feeder (included in the effect of replicate). After the blood meal, unfed and partially fed mosquitoes were discarded, and fully fed mosquitoes were kept in the insectary with permanent access to a 10% sucrose solution.
Melanization assays.
CM-25 Sephadex beads (Sigma-Aldrich, Steinheim, Germany) range from 40 to 120 µm in diameter; the smallest ones were selected by visual inspection for inoculation. Beads were rehydrated in saline solution containing 1.3 mmol/L NaCl, 0.5 mmol/L KCl, and 0.2 mmol/L CaCl2 (pH 6.8) and stained with 0.001% methyl green to help in visualization.24 We immobilized mosquitoes briefly on ice and injected one bead per mosquito into the thorax with < 0.1 µL of saline solution into the hemolymph, using a heat-pulled capillary needle.24 After 24 hours, mosquitoes that were able to fly were dissected in saline solution with 0.01% methyl green. Beads were recovered, and melanization was scored according to three broad categories: weak melanization (class 0), intermediate melanization (class 1), and strong melanization (class 2). Figure 1A
shows typical pictures of the three melanization classes.
|
Total RNA was extracted from each abdomen using the Tri Reagent kit (M.R.C., Ontario, Canada) according to the manufacturers instructions. Extracted RNA was treated with the Turbo DNA-free kit (Ambion) and resuspended in 10 µL of water with 1 µL of RNasin (Promega, Madison, WI). RNA was reverse transcribed, and RT products were amplified using the Access RT-PCR System kit (Promega) according to the manufacturers instructions on 1 µL of the RNA extract in a final volume of 25 µL. Reverse transcription was performed for 45 minutes at 45°C followed by 2 minutes at 94°C. Amplification conditions were five touchdown cycles (30 seconds at 94°C, 1 minute at 5550°C, 2 minutes at 68°C), followed by 20 cycles (30 seconds at 94°C, 1 minute at 50°C, 2 minutes at 68°C), and a final 7-minute elongation step (68°C). Absence of contaminating genomic DNA was systematically checked by a control without RT. A nested PCR was performed on 1 µL of a 1/100 dilution of the RT-PCR products in a final volume of 25 µL. Amplification conditions were 2 minutes at 94°C, followed by five touchdown cycles (30 seconds at 94°C, 30 seconds at 5550°C, 30 seconds at 72°C) followed by 35 cycles (30 seconds at 94°C, 30 seconds at 50°C, 30 seconds at 72°C), and a final 7-minute elongation step (72°C). The specific primers used in the two PCR series have been described elsewhere.3 Nested PCR products were analyzed on 1.2% agarose gels in Tris-borate buffer. Total RNA from cultured P. falciparum gametocytes was used as a positive control. A minimum detection threshold of 0.1 pg of total P. falciparum RNA per 100 ng of mosquito RNA was determined in a pilot experiment using dilutions of gametocyte RNA into mosquito RNA.
Data analysis. Because the use of Sephadex beads has the disadvantage that some mosquitoes die of the side effects of inoculation, we first controlled for a potential bias caused by post-inoculation mortality. We analyzed the arsine-transformed mortality rates among replicates as a function of the experimental treatment (control blood versus challenge by P. falciparum), including the experimental block as a potential random confounder.
We analyzed the data in two steps. First, we determined whether the melanization response of challenged mosquitoes (i.e., those that fed on infected blood) was associated with infection success at 48 hours pbm. The level of bead melanization (scored as three categories) was analyzed with an ordinal logistic analysis as a function of the mosquitos infection status (infected or uninfected). Wing length (an indication of body size), the experimental block, the parasite isolate, and the replicate were included as potential confounders. As different isolates were used in each block, isolate was nested within block. Replicate was nested within block and isolate; block, isolate, and replicate were considered as random factors. Second, we compared the melanization response between mosquitoes challenged with P. falciparum and control mosquitoes fed on an uninfected blood meal. Because the first step of the analysis had revealed differences in the melanization level of infected and uninfected mosquitoes, we compared these two groups to control mosquitoes separately. For each group (infected or uninfected), the level of bead melanization (scored into three categories) was analyzed with an ordinal logistic analysis as a function of the blood-meal type (infectious or control). The wing length, the experimental block, and the replicate were included as potential confounders. Replicate was nested within block; block and replicate were considered as random factors. We did not include the isolate as a confounder because the first analysis showed that there was no difference between isolates.
All analyses were performed with the software JMP version 5.0 (http://www.jmpdiscovery.com).
| RESULTS |
|
|
|---|
On average, melanization was stronger in the second experimental block (Figure 1
), which may be because of the difference in body size of the two corresponding batches of mosquitoes. Indeed, the mean wing length was 3.09 ± 0.020 (SE) mm in the first block and 3.39 ± 0.015 mm in the second block, and melanization was overall positively associated to the wing length in a one-way logistic analysis (
2 = 6.04, P = 0.014).
Among females fed on infectious blood, the mosquitoes with live parasites at 48 hours pbm had a significantly stronger melanization response than those where no parasite was detected (Table 1
). Although 40% and 93.8%, respectively, of the infected mosquitoes strongly melanized the bead in the first and the second block, only 30.8% and 70.6% did so in the uninfected mosquitoes (Figure 1
).
|
|
| DISCUSSION |
|
|
|---|
That the melanization response was slightly stronger in infected than in uninfected mosquitoes could result from a trade-off between bead melanization and parasite resistance, whereby mosquitoes able to clear early stages of the parasite would have a weaker melanization response against beads and vice-versa. Such a trade-off could be caused by a negative genetic correlation between different components of the immune system.31 This idea is supported by the observation that alleles responsible for P. falciparum melanization are associated to high parasite numbers in wild pedigrees of A. gambiae,32 but contrasts a previous study showing that the genetic correlation between two arms of the immune systemantibacterial defense and melanizationis positive rather than negative in a natural population of A. gambiae.33 However, the existence of a trade-off between parasite resistance and bead melanization would imply in our study that the mosquitoes that cleared the parasite would have, on average, a weaker melanization response than mosquitoes fed on un-infected blood. This was not the case; rather, the melanization response was similar in the two categories of mosquitoes (Table 2
). It is thus more likely that the melanization response had been triggered after midgut invasion by the early stages of malaria parasites. This could simply result from the cell damage caused by ookinetes during their migration through the midgut epithelium,34 because melanization is also intimately tied to wound healing,7 or from an immunologic priming35 following the previous up regulation of immune components caused by the detection of invading malaria parasites.3
The lack of immuno-suppressive ability of the melanization response in a natural population of P. falciparum suggests that the parasite survives this response by evading it. Note that this might not hold for other immune responses that can affect Plasmodium development. For example, the expression of the NOS gene of A. gambiae is repressed after P. falciparum infection,3 suggesting that the parasite might reduce the production of NO, a molecule known to limit Plasmodium development in the mosquito.36 Immune evasion mechanisms by parasitic protozoa in their vertebrate hosts include antigenic variation, shedding of surface proteins, antigenic mimicry, and intracellular hiding,37 but such mechanisms in the vector remain unknown. Candidate molecules enabling the evasion of the melanization response by the parasite include mosquito immune proteins acting as protective agonists. For instance, two C-type lectins, CTL4 and CTLMA2, prevent the melanization of P. berghei by a susceptible line of A. gambiae.10 Moreover, gene silencing of CTL4 or CTLMA2 does not affect the melanization of Sephadex beads in A. gambiae, supporting the idea that the parasite might specifically use these molecules to evade the immune response.38
In conclusion, this study showed that experimental infection of A. gambiae by P. falciparum is not associated with an immuno-suppression but rather an enhanced melanization response. This contrasts earlier laboratory-based studies with P. gallinaceum, where the parasite suppressed the melanization response of Ae. aegypti.21,22 Thus, together with another study,3 our finding confirms the need to validate the observations made in laboratory models of mosquitomalaria interactions in field transmission systems.23,39 It also stresses the importance of addressing the strategies used by malaria parasites to overcome mosquito defense mechanisms.
Received September 13, 2006. Accepted for publication November 18, 2006.
Acknowledgments: We thank the inhabitants of Mfou for cooperation and Engelbert Manga, Constance Efemba, and Emmanuel Bozewan for kind support during the fieldwork in Mfou. The authors thank Isaac Tchikangwa for rearing mosquitoes, Rose Nyambam and Sylvie Kemleu-Zebaze for assistance in the screening of gametocyte carriers, Kristin Michel for help during the fieldwork and for the images of beads, Isabelle Thiéry for providing cultured P. falciparum gametocytes, and an anonymous reviewer for helpful comments on an earlier version of the manuscript.
Financial support: This study was supported by the VIHPAL Program of the French Ministry of Research.
* Address correspondence to Louis Lambrechts, Department of Entomology, University of California, One Shields Ave., Davis, CA 95616. E-mail: llambrechts{at}ucdavis.edu ![]()
Authors addresses: Louis Lambrechts and Jacob C. Koella, Laboratoire de Parasitologie Evolutive, Unité Mixte de Recherche 7103, Centre National de la Recherche Scientifique-Université Pierre et Marie Curie, Case Courrier 237, 7 quai St Bernard, 75252 Paris Cedex 05, France, Telephone: 33144275204, Fax: 33144273516. Present address of Louis Lambrechts, Department of Entomology, University of California, One Shields Ave., Davis, CA 95616, Telephone: 5307520565, Fax: 5307521537. Present address of Jacob C. Koella, Division of Biology, Imperial College London, Silwood Park Campus, Ascot SL5 7PY, UK, Telephone: 442075942254. Louis Lambrechts, Jean-Claude Jacques, and Catherine Bourgouin, Centre de Production et dInfection des Anophèles, Institut Pasteur, 28 rue du Dr Roux, 75724 Paris Cedex 15, France, Telephone: 33144389577, Fax: 33140613089. Isabelle Morlais, Parfait H. Awono-Ambene, and Frédéric Simard, Laboratoire de Recherche sur le Paludisme, Institut de Recherche pour le Développement-Organisation de Coordination pour la Lutte contre les Endémies en Afrique Centrale, BP 288, Yaoundé, Cameroon, Telephone: 2372232232/9932111, Fax: 2372230061/2201854. Isabelle Morlais, Anna Cohuet, and Frédéric Simard, Unité de Caractérisation et Contrôle des Populations de Vecteurs, Institut de Recherche pour le Développement, 911 Avenue Agropolis, BP 64501, 34394 Montpellier Cedex 5, France, Telephone: 33467041924, Fax: 33467542044. Catherine Bourgouin, Unité de Biologie et Génétique du Paludisme, Institut Pasteur, 28 rue du Dr Roux, 75724 Paris Cedex 15, France, Telephone: 33145688224, Fax: 33140613089.
Reprints requests: Louis Lambrechts, Department of Entomology, University of California, One Shields Ave., Davis, CA 95616. E-mail: llambrechts{at}ucdavis.edu.
| REFERENCES |
|
|
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |