|
|
||||||||
| ABSTRACT |
|
|
|---|
| INTRODUCTION |
|
|
|---|
In what way Plasmodium modulates its production of total gametocytes and the proportion that are male versus female has recently been reappraised within an evolutionary framework, wherein gametocyte allocation is considered as an adaptive phenotype. In particular, the framework proposes two complementary explanations (i.e., ultimate cues) for observed gametocyte sex ratios: 1) the sex ratio is optimized to ensure successful fertilization in the short term, especially important when gametocyte densities are low,57 and 2) the sex ratio is selected over the long term according to the average genetic diversity of infections and thus the level of inbreeding, i.e., the Plasmodium population substructure results in deviations from random mating, which favors a more female-biased gametocyte investment.810 That the sex ratios of Plasmodium species adhere to evolutionary principles governing eukaryotes in general11,12 is important because it strengthens the argument for implementing the wealth of evolutionary theory to predict the consequences of an intervention strategy and for instance how will the parasite evolve following implementation of bed nets or a vaccine.13
In parallel, the examination of the proximate cues triggering sex allocation14 suggest that Plasmodium species respond to the same host cues but in a species-specific manner,15,16 which appears to depend on their blood cell preference (e.g., reticulocytes or mature erythrocytes).17 There is a growing body of evidence showing that parasite sex allocation correlates with the hematologic response of the host to infection and is manifest either through increased gametocyte production or an increased allocation of sexual stages into male versus female gametocytes. For example, in response to anemia, total gametocyte production increases in Plasmodium chabaudi,18 while an increased male investment is observed in P. vinckei and P. gallinaceum19; and in P. falciparum both variables were independently shown to increase.2023 That sex allocation in Plasmodium species responds facultatively to the same exogenous cues facilitates experimentation in controlled laboratory models and consequent extrapolation to P. falciparum.
The importance of the sex ratio for transmission of Plasmodium species in general, and of P. falciparum in particular, has been the subject of discussion.6,24 In natural infections in both humans and lizards, for example, the sex ratio can vary extensively over a few days with little clear significance for transmission success.24,25 In vitro studies of mosquito infection rates have also failed to demonstrate an effect of the gametocyte population structure, including sex ratio, on infection success.26 Given our comparative lack of knowledge of the many factors that influence transmission, establishing the significance of variability in sex ratio on transmission success requires careful examination under highly standardized conditions. Under such conditions, clear evidence has arisen from both in vitro and in vivo studies that the gametocyte sex ratio is important and that transmission success is particularly dependent on the density and proportion of male gametocytes.1,25,27,28
The study of the sex ratio phenotype in field populations has been the subject of considerable speculation, but of little empirical investigation. Furthermore most of the in vivo studies that established precise counts of P. falciparum sex ratios have been performed on populations undergoing chemotherapy.2932 Such studies are hampered by the unknown effect of antimalarials on individual gametocyte infectivity and on parasite sex determination. It is therefore of crucial importance to assess the effect of drug treatment on sex ratio to ensure that there was no treatment bias in the sex ratio data previously obtained under chemotherapy. We therefore use field P. falciparum data to assess the question of whether anti-malarial treatment has an influence on the sex ratio of gametocytes. Because no untreated controls are possible in the study of P. falciparum sex ratios, we extend the study by examining the effect of sub-curative doses of CQ on the sex ratio of a laboratory mouse model, P. vinckei petteri. The effect of chloroquine (CQ) treatment on the sex ratio of P. vinckei petteri is compared with the effect of CQ and sulfadoxine-pyrimthamine (SP) on natural infections of P. falciparum where the infection strain resistance phenotype is additionally characterized.
| MATERIALS AND METHODS |
|
|
|---|
Study design. The study was carried out in Diohine, a village approximately 110 km east of Dakar in the Niakhar area of Senegal. This area of mesoendemic malaria is characterized by seasonal transmission that culminates from August to October.
Patients were recruited from September to November 1996 at the dispensary. Children with uncomplicated P. falciparum malaria (< 5,000 parasites/µL of blood) who were not using any antimalarial drug at the time were eligible, once the verbal consent of their parents was obtained. Patients were included only if no gametocytes were observed on day 0. The study was reviewed and approved by the Senegalese Ministry of Health.
Treatment and follow-up of children. The children were allocated to two different oral treatment groups: the CQ treatment group (CQ phosphate; Société Industrielle Pharmaceutique de lOuest Africain, Dakar, Senegal), 25 mg/kg of body weight given over a three-day period: 10 mg/kg on day 0 and day 1 and 5 mg/kg on day 2, and the SP group (Fansidar®; F. Hoffmann LaRoche, Basel, Switzerland), 25 mg/kg of sulfadoxine and 1.25 mg/kg of pyrimethamine in a single dose.
The children were followed for 14 days post-treatment, and thick blood smears were prepared on days 0, 7, and 14. All children exhibiting signs of clinical failure, severe malaria, or continued parasitemia were given a second-line treatment and excluded from the study.
Determination of sex ratio. Thick blood smears were stained with Giemsa. Gametocyte and asexual parasite densities were calculated from reading 200 microscope fields (magnification x1,000). It was assumed that 0.5 µL of blood was examined in this way; thus, the sensitivity threshold was two parasites/µL. The thick blood smears prepared for days 7 and 14 were re-read to establish the sex ratios, which were not counted in the original study.33 Gametocyte sex was determined using the five classic criteria: 1) males are larger than females, 2) the ends of the cells are angular in females and round in males, 3) the nucleus is smaller in females than in males, 4) the granules of malaria pigment are centrally located in females and more widely scattered in males, and 5) the cytoplasm stains deep blue in females and pale purple in males.28,34 The gametocyte male percentage was calculated as the number of male gametocytes divided by the sum of male plus female gametocytes (the gametocytes that could not be differentiated were not included).
Inclusion criteria. Patients were included in the analysis when their gametocyte densities were 0/µL of blood on day 0 and greater than 20/µL of blood on days 7 or 14.
Statistical analysis. Due to a non-normal distribution of values, quantitative variables (patient age of treatment groups, sex ratio) were compared between treatment groups with a non parametric Mann-Whitney U test. For comparison of sex ratio at days 7 and 14 within treatment groups, data were analyzed with a non-parametric Wilcoxon Z test. The change in sex ratio over time across both treatment groups was analyzed with Pearsons correlation coefficient.
Mouse parasite. Plasmodium vinckei petteri (strain 106HW) (courtesy of I. Landau, Museum National dHistoire Naturelle, Paris, France), was used. This strain is an isolate from the rodent Thamnomys rutilans in the Central African Republic in 1988. This strain has a drug-sensitive phenotype and CQ treatment has previously been shown to have no effect on gametocyte conversion (GC) rates. Parasite development is synchronous and schizogony occurs 24 hours following the injection of frozen infected blood.35 The time between merozoite invasion and appearance of gametocytes was estimated to be 2436 hours depending on the maturation type, and the first identifiable gametocytes appear at 27 hours.36 The life span of mature gametocytes (more than 36 hours of age) is unknown but probably does not exceed 24 hours.
Infection of mice. An intraperitoneal injection of 105 parasites/mouse (810-week-old female Swiss albino mice; Charles River/IFFA-CREDO, Lyon, France) was carried out in 20 mice at day 3. Fifteen were selected after blood smears on day 0 to be included in the study; selection aimed at standardizing parasitemias and thus having similar infection parasitemias in each of the three groups. In practice, this meant elimination of two high parasitemia infections and three very low parasitemia infections. All experimental animals were maintained according to European Union guidelines.
Drug administration. Preparations of CQ diphosphate (Sigma, St. Louis, MO) were diluted in 1x phosphate-buffered saline (PBS). The 15 mice were randomly assigned in three different treatment groups on day 0. Two groups of five mice were injected intraperitoneally with CQ at sub-curative doses of 5 mg/kg of body weight (CQ5) and 10 mg/kg of body weight (CQ10), and the five remaining mice were injected with 1x PBS on day 0.
Follow-up of parasitamia and determination of sex ratio. Thin blood smears (tail blood) were prepared daily at the same time each morning, fixed with 100% methanol, and stained with 5% Giemsa in phosphate buffer (0.1 M NaPO4, pH 7.2) for five minutes. Parasitemias were calculated as percentages observed in a minimum of 100 red blood cells by reading a fixed subset of 10 different microscope fields; gametocytemias were calculated from observations on 10,000 red blood cells Mature male and female gametocytes are distinguishable after staining with Giemsa: males stain pink and females stain blue with a distinct red nucleus.37 Sex ratios based on counts of 5075 gametocytes were found to be representative.19 We calculated sex ratios from either 50,000 red blood cells or 100 gametocytes, whichever was smaller. Sex ratios are given as the proportion of males.
Statistical analysis. Statistical analyses were conducted using the statistical package Genstat 6.1 (VSN International Ltd., Oxford, United Kingdom). Because each individual mouse was included in the data set many times, we corrected for repeated measures by fitting a generalized linear mixed model that nested day within individual mouse in the random model. Sex ratio, gametocytemia, percentage of reticulocytes, and the GC rates were analyzed specifying a binomial error structure. The effect of CQ treatment groups were considered separately (control versus CQ5 versus CQ10) and combined (control versus CQ). For the overall change in sex ratio with time, the three treatment groups were combined because there was no difference in their sex ratios at any time point. Statistical significance is presented as Wald statistics, which approximate to a chisquare distribution. For the sex ratio analyses, the data were over-dispersed; they were corrected for by estimating a dispersion parameter for each analysis.
In the analyses, only the data from the day of treatment (day 0 of study) and the two days following were considered. This was because the immediate effect of CQ was considered to have an impact on the parasites up to two days post-treatment. In addition, the majority of control mice died on day 3 of the study (day 6 post-infection). The GC rates were calculated in two ways according to the estimated lifespan of mature gametocytes (> or < 12 hours). Based on the assumption that mature gametocytes take up to 36 hours to mature and subsequently live for more than 12 hours, gametocytes from the previous time period will overlap with young developing gametocytes; in this case the GC rate GCLow was calculated as the ratio of [(gametocytemia(time t) gametocytemia(time t 1))/parasitemia(time t) ]. Based on the assumption that gametocytes live less than 12 hours, the GC rate GCHigh was calculated as the ratio of [gametocytemia(time t)/parasitemia(time t)]. These rates provide upper and lower estimates of the GC rate.
| RESULTS |
|
|
|---|
No statistical difference was found between treatments and the gametocyte sex ratio was shown to be the same between the two treatments, both at days 7 and day 14 (Table 1
). When the sensitive and resistance phenotype infection groups were compared, no statistical difference was found at either days 7 and 14 for either treatment (day 7 for SP and CQ combined P = 0.50; n = 50, for CQ P = 0.44; n = 13, and for SP P = 0.92; n = 37; day 14 for SP and CQ combined P = 0.77; n = 40, for CQ P = 0.99; n = 9, and for SP P = 0.71, n = 31, by Mann-Whitney U test).
|
|
21 = 11.9, P = 0.0007).
Treatment with CQ had no effect on the gametocyte sex ratio (control versus CQ5 versus CQ10,
22 = 2.52, P = 0.28; combined CQ versus control,
21 = 2.07, P = 0.15). Neither of the CQ treatments had any significant effect on GC rates (CQ treatment on GCLow: control versus CQ5 versus CQ10,
22 = 3.44, P = 0.18; combined CQ versus control,
21. = 0.84, P = 0.36; CQ treatment on GCHigh: control versus CQ5 versus CQ10,
22 = 4.52, P = 0.11; combined CQ versus control,
.= 1.51, P.= 0.28). However, the sex ratio was found to be influenced by the density of reticulocytes independently of treatment (% reticulocytes on sex ratio:
21 = 6.44, P = 0.011), as previously found.18 A positive correlation between time and sex ratio was observed: 14.5% of the gametocytes were male at day 0 (99% confidence interval [CI] = 11.817.2%), 37.5% at day 3 (99% CI = 35.639.4%), and 47.8% at day 4 (99% CI = 43.352.3%) (
21 = 10.8, P = 0.001).
| DISCUSSION |
|
|
|---|
This contrasts with the recognized effect of drug treatment on the induction of gametocytogenesis in P. falciparum, which was indeed also noted in this study (day 7 post-treatment, Figure 1
). This induction has been previously observed following treatment with either CQ or SP in both P. falciparum and animal model malaria parasite species infections.33,3842 Indeed, treatment using other antimalarial drugs also induces gametocytogenesis,43 but the significance for gametocyte infectivity remains uncertain.44,45 Whereas SP generally reduces gametocyte infectivity,39,46 the infectiousness of CQ-treated gametocytes is variable and is probably dependent on the parasite strain, the dose of CQ administered, and additional vertebrate host immune factors.38,39,44,45,47 Drug resistance and increased gametocytogenesis have been previously found to be correlated in strains of P. chabaudi and P. vinckei petteri48 and thought to be so in P. falciparum.32,33 However, studies using the mouse model with P. yoelii nigeriensis demonstrated that drug resistance and increased gametocytogenesis are not necessarily linked.49 Indeed, our data did not show any significant differences in sex ratios between sensitive and resistant phenotypes in P. falciparum; this suggests that resistance to drugs per se is not a factor influencing sex determination. A similar conclusion cannot be drawn for P. vinckei petteri because the strain used was sensitive to CQ. Given the differential mode of action of CQ and SP, this would suggest that the drugs themselves have no direct biochemical action on the pathways involved in sexual development. It should be noted, however, that there is considerable cross-communication among signaling pathways involved in cell fate, thus enabling the same response from multiple cues.50 However, the fact that the stress induced by chemotherapy has an influence on the rate of commitment to sex in P. falciparum but not on sex ratios, strongly suggests that induction of gametocytogenesis and sex determination operate using distinct pathways.
For both Plasmodium species examined here, the sex ratio increased over the course of the infection (Figure 1
). One plausible explanation is that male gametocytes have a longer half-life than females. A recent study by Reece and others51 addressed sex-specific gametocyte longevity using the rodent malaria parasite P. chabaudi and curative CQ drug treatment to remove the asexual population (and thus the supply of new gametocytes). This study concluded that male gametocytes have a higher natural longevity. This contrasts with findings from a similar study on P. falciparum infections, in which no change in sex ratio was observed following drug treatment, albeit based on a small number of individuals.30 An alternative explanation concerns the response of the parasite to anemia: previous studies have demonstrated a positive correlation between sex ratio and reticulocytosis19; this was again borne out here for P. vinckei petteri. The absence of an effect of sub-curative doses of CQ on sex ratio of would suggest that any differential sex mortality is insignificant compared with the natural shift to increased male sex allocation in response to hematologic recovery. These differential species effects of exogenous cues on sex ratio increasingly highlight the importance of an appreciation of the characteristics associated with the ecology of the species. Therefore, particular attention must be paid to parasite life history characteristics to interpret the gametocyte phenotypic response to exogenous cues or during an infection.
The parasite seems to have evolved two distinct ways in which to respond to the environment with respect to their transmission to the mosquito host: gametocyte production, and sex allocation. However, the cues that activate one do not necessarily act upon the other. If anemia influences both, it now seems probable that drug treatment only affects gametocyte production.
Received March 8, 2004. Accepted for publication June 9, 2004.
Acknowledgments: We thank El Hadj Bâ for the microscopic examinations. Arthur M. Talman, Richard E. L. Paul, Olivier Domarle, Frédéric Ariey, and Vincent Robert are members of the Grand Program Horizontal Anopheles of the Institut Pasteur. We are very grateful to the reviewers for their constructive criticism.
Financial support: Fieldwork was support in part by the Institut de Recherche pour le Développement and PAL+. Richard E. L. Paul was supported by the Institut Pasteur and the French Ministère de la Recherche (Programme de Recherches Fondamentales en Microbiologie, Maladies Infectieuses et Parasitologie).
Authors addresses: Arthur M. Talman, Groupe de Recherche sur le Paludisme, Institut Pasteur de Madagascar, BP 1274, Antananarivo 101, Madagascar and Department of Biological Sciences, Imperial College, Exhibition Road, SW7 2AZ London, United Kingdom, E-mail: atalman{at}pasteur.mg. Richard E. L. Paul, Unité de Biochimie et Biologie Moléculaire des Insectes, Institut Pasteur, 28 rue du Dr Roux, 75724 Paris Cedex 15, France, E-mail: rpaul{at}pasteur.sn. Cheikh S. Sokhna and Jean-François Trape, UR Paludisme Afro-Tropical, Institut de Recherche pour le Développement, BP 1386, Dakar, Senegal, E-mails: sokhna{at}ird.sn and trape{at}ird.sn. Olivier Domarle and Frédéric Ariey, Groupe de Recherche sur le Paludisme, Institut Pasteur de Madagascar, BP 1274, Antananarivo 101, Madagascar, E-mails: domarle{at}pasteur.mg and ariey{at}pasteur.mg. Vincent Robert, Groupe de Recherche sur le Paludisme, Institut Pasteur de Madagascar, BP 1274, Antananarivo 101, Madagascar and Unité de Recherche Paludisme Afro-Tropical, Institut de Recherche pour le Développement, BP 1386, Dakar, Senegal, Telephone: 261-20-22-412-72, Fax: 261-20-22-415-34, E-mail: robert{at}pasteur.mg.
| REFERENCES |
|
|
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |