ENTOMOLOGY

ENTOMOLOGY

ENTOMOLOGIE

Identification of vertebrate blood meals in tsetse flies using the Heteroduplex PCR based method

IDENTIFICATION DE L'ORIGINE DES REPAS DE SANG DE VERTEBRÉS CHEZ LES GLOSSINES

F. Njiokou^1,2, G. Simo^1,3 , A. Mbida Mbida¹, P. Truc^1,4 & S. Herder⁵

¹ Laboratoire de Recherche sur les Trypanosomoses, Organisation de Coordination pour la lutte contre les Endémies en Afrique Centrale (OCEAC), BP 288, Yaoundé, Cameroun.

² Université de Yaoundé I, Faculté des Sciences, Dept. BPA, BP 812, Yaoundé, Cameroun.

Fax : (237) 23 00 61 ; e-mail : fnjiokou@yahoo.com

³ Institut de recherche Médicale et d'étude des Plantes Médicinales (IMPM), Yaoundé

⁴ Institut de Recherche pour le Développement, UR 35 “Trypanosomoses Africaines”, Yaoundé Cameroon

⁵ Laboratoire de Recherche et de Coordination sur les Trypanosomoses (IRD - CIRAD)

TA 207/G, Campus international de Baillarguet, 34398 Montpellier cedex 5 France.

Flobert Njiokou : OCEAC, BP 288, Yaoundé, Cameroon ; Tel (237) 223 22 32 ; Fax : (237) 223 00 61 ; e-mail : fnjiokou@yahoo.com

Résumé

La faune sauvage est suspectée de jouer le rôle de réservoir de Trypanosoma brucei gambiense. Une étude des préférences trophiques des glossines a été menée pour estimer l'importance des repas de sang pris sur cette faune sauvage.

Pour ce faire, une capture des glossines a été organisée dans trois localités du sud du Cameroun (Bipindi, Campo et Nditam) et des repas de sang ont été collectés. L'origine des repas de sang a été déterminée par la technique des hétéroduplexes du gène du cytochrome B : l'ADN de référence de ces vertébrés sauvages et l'ADN extrait des repas de sang ont alors été amplifiés par les amorces du gène du cyrochrome B et hybridés avec l'ADN du rat de Gambie choisie comme "driver"; les homoduplexes et hétéroduplexes formés ont été séparés sur gel de polyacrylamide. L'origine des repas de sang a été déterminée par comparaison des profils hétéroduplexes des échantillons avec ceux des ADN de référence.

Trois mille quatre-vingts glossines ont été capturées et 96 repas de sang ont été collectés dont 77 (80%) provenaient de Glossina palpalis palpalis, 16 (17%) de G. pallicera, 2 (2%) de G. calliginea et 1 (1%) de G. nigrofusca. Quarante (42%) repas de sang ont été pris sur homme, 17 (18%) sur porc, 14 (14%) sur Guib d'eau (sitatunga) et 1 (1%) sur mouton. Dix-sept repas de sang n'ont pas pu être identifiés car ne correspondant à aucun animal domestique ni aux animaux sauvages dont nous disposions d'ADN de référence. Sept repas de sang n'ont pas pu être amplifiés.

G. p. palpalis se nourrit sur homme, animal domestique et animal sauvage et au moins 18 % de repas sont pris sur les animaux sauvages.

Summary

It has been suggested that wild and domestic animals could be reservoir hosts of Trypanosoma brucei gambiense. We studied feeding preferences of tsetse flies in order to estimate the proportion of blood meals taken by these vectors from domestic and wild animals.

Tsetse flies (3676 Glossina palpalis palpalis, 482 G. pallicera, 258 G. nigrofusca, and 77 G. calliginea) were captured, and 137 blood meals were collected in 3 localities of southern Cameroon (Bipindi, Campo and Nditam). Identification of blood meals was performed by cytochrome B heteroduplex analysis: blood meal DNAs and animal reference DNAs were amplified using cytochrome B gene primers and hybridized with Giant rat DNA as a driver. Blood meal origins were identified by comparing heteroduplex DNA profiles with reference DNAs profiles.

For all the sites, 111 blood meals were amplified. 58 blood meals were from human origin (54%), 31 from domestic animal origin (28%), and 20 were taken from wild animals (20%). 14 blood meals remained unidentified, were probably taken from wild animal.

Heteroduplex PCR based method was very useful for identification of blood meals origin but only if the corresponding reference DNA was available. G. palpalis feeds on human, domestic and wild animals, and may be a potential vector for the transmission of T. b. gambiense from animals to human.

Introduction

The evolution of the Human African trypanosomosis (HAT) shows that currently there is a resurgence despite control activities. These observations may be explained by the presence of an active animal reservoir. Wild animals were suspected to be potential hosts of Trypanosoma brucei gambiense, the causative agent of HAT, according to previous studies (Mehlitz, 1986; Truc et al., 1997; Herder et al., 2002). Some of these wild animals were not identified as usual feeding hosts of tsetse flies (Herder et al., 2002). It was necessary to study the frequency of contacts between vertebrate hosts and tsetse flies by estimating tsetse flies feeding preferences. Immunological methods have been used for identification of bloodmeals of haematophagous insects but the process of preparation of immune sera against the potential host species were difficult (Hunter & Bayly, 1991).

A rapid method based on cellulose acetate electrophoresis of super oxide dismutase (SOD) can only differentiate human from non human bloodmeals from tsetse flies (Diallo et al., 1997). The cytochrome B gene heteroduplex PCR based method was very useful for the identification of many vertebrate feeding hosts of haematophagous insects (Boakye et al., 1999). In this study, we modified and simplified the protocol described by Boakye et al. (1999) in order to use heteroduplex analysis (in a prospective way) to investigate the ability of tsetse flies to feed on wild vertebrates in 3 foci in South Cameroon.

Materials and method

Biological material was collected in two HAT foci: Campo (2°20'N, 9°52'E) and Bipindi (3°2'N, 10°22'E), and in a non endemic locality (Nditam: 5°20'N, 11°9'E) in the Cameroonian forest.

Bloods samples were collected from 21 common wild animals including Rodents, Ungulates, Monkeys, Pangolins and Carnivores. The various domestic animals (pig, sheep, goat, cattle, dog, cat, and hen) were bled in our facilities. Blood samples were spread onto a disk (2 cm²⁾ of Whatman paper n° 1. Tsetse flies were trapped using Vavoua traps (Laveissière & Grébaut, 1990) during two weeks in every locality. They were dissected and blood meals were collected also on Whatman paper n°1, and stored into Eppendorf^® tubes.

DNAs were extracted from domestic and wild animal bloods and from blood meals using the Chelex^® 100 method (Walsh et al., 1991): 1 ml of 5% Chelex^®100 was added in each Eppendorf^® tube. These latter were heated at 56° C for one hour, and at 100° C for 30 min.. Tubes were centrifuged at 14,000 rpm for 10 min. at room temperature and supernatants were removed and used for the PCR assay. Amplification reactions were performed using 5 µl of DNA solution with vertebrate cytochrome B primers (Boakye et al., 1999). The amplified DNAs were hybridized according to Boakye et al. (1999) using Giant rat (Crycetomys gambianus) DNA as a driver. Homoduplex and Heteroduplex products were separated on a 5% polyacrylamide/urea gel without stacking gel. Heteroduplex profiles were compared with the reference animal profiles in order to identify the origin of tsetse fly blood meals.

Results

4493 tsetse flies were trapped in the three localities and dissected. Four species were identified: 3676 Glossina palpalis (81%), 482 G. pallicera (11%), 258 G. nigrofusca (6%) and 77 G. calliginea (2%). 137 blood meals were collected.

Reference animal DNAs and 125 blood meal DNAs were successfully amplified using the cytochrome B gene. Heteroduplex analysis allowed the identification of the origin of 111 blood meals by comparing their profiles with those of the reference animals (Fig. 1). 14 blood meals did not correspond to any reference profile and were not identified. A slight variation occurred between 2 Moustached Monkey and 2 Tree Pangolin from different localities. For the 20 identified blood meals from Nditam, 18 were taken from human and 2 from Sitatunga. 58 identified blood meals from Bipindi were taken from 25 humans, 30 pigs, 1 Sitatunga, 1 Golden Cat, and 1 sheep. The 33 identified blood meals from Campo were taken from 17 humans and 16 Sitatunga.

Figure 1. Heteroduplex profiles from reference animals and bloodmeal samples

M: molecular weight marker (100/20 base pair: bp); 11 to 17: reference samples (11: sitatunga; 12: dog; 13: man; 14: pig; 15: goat; 16: sheep; 17: Giant rat); 1 to 10: bloodmeal samples (2: pig; 4 and 5: man; 6, 7 and 8: sitatunga); 1, 3, 9 and 10: non identified.

Discussion

Reference animals and blood meal DNAs were successfully amplified using vertebrate cytochrome B primers. This confirmed that Chelex^®100 was a suitable technique for extraction of low quantities of DNA (Walsh et al., 1991). DNA extraction from animal tissue, which was usually done by a phenol/chloroform method, was not necessary anymore for this DNA amplification. Heteroduplex profiles were highly specific of each reference DNA sample when using Giant rat DNA as driver. These profiles could be used as a reference for identification of others blood meals. However, a slight variation has been observed between different Moustached Monkey and Tree Pangolin. They may probably belong to different subspecies. 125 blood meals (91%) were successfully amplified, and 111 blood meals (89%) were identified. 54 % of identified blood meals were from human origin while 28% and 18% respectively, were taken from domestic animals (pig and sheep) and wild animals (Sitatunga : Tragelaphus spekei, and Golden cat : Profelis aurata). These results confirmed previous study by Grébaut et al. (unpublished data) at Bipindi, where 80% of blood meals were from human and pig origins, and by Laveissière et al., (1985) results in West Africa where 46% of blood meals were taken from human and 46% from another antelope species Tragelaphus scriptus. The results also confirmed the feeding preferences of tsetse flies for these animals. Sitatunga blood meals were identified in all studied localities. Fourteen amplified blood meals (11%) could not be identified since their heteroduplex profiles did not correspond to any reference DNAs. These blood meals showed three different profiles and, therefore, suggested that they were originated from three animal species. They were probably from wild animals origin since their profiles were different from all domestic animals species tested.

The simplified Heteroduplex method was very useful for identification of blood meals origin but only if the corresponding reference DNA is available. At least 41% of tsetse fly blood meals were taken from wild and domestic animals. Therefore, this suggested a frequent contact between some animals and tsetse flies. While human is known as the main feeding host of G. p. palpalis, tsetse fly appears to feed also on animals. Thus a potential transmission of HAT may occur from wild and domestic animals to human in Cameroon.

Reference

Boakye, D.A., Tang, J., Truc, P., Merriweather, A. & Unnasch, T. R.; 1999. Identification of bloodmeal in haematophagous Diptera by cytochrome B heteroduplex analysis. Med. Vet. Entomol. 13, 282-287.

Diallo, P.B., Truc, P. & Laveissière, C.; 1997. A new method for identifying blood meals of human origin in tse tse flies. Acta Trop. 23, 61 - 64.

Herder, S., Simo, G., Nkinin, S. & Njiokou, F.; 2002. Identification of trypanosomes in wild animals from Southern Cameroon using the polymerase chain reaction (PCR). Parasite. 9, 345-349.

Hunter, F.F. & Bayly, R. 1991.; ELISA for identification of blood meal source in black flies (Diptera: Simulidae). J. Med. Entomol. 28, 527-532.

Laveissière, C., Couret, D., Staak, C. & Hervouet, J.P.; 1985. Glossina palpalis et ses hôtes en secteur forestier de Côte d'Ivoire. Relation avec l'épidémiologie de la trypanosomiase humaine. Cah. ORSTOM, Sér. Ent. Méd. Parasitol. 23, 297-303.

Laveissière, C. & Grébaut, P.; 1990. Recherches sur les pièges à glossine (Diptera : Glossinidae) Mise au point d’un modèle économique : le piège Vavoua. Trop. Med. Parasitol.; 41, 185-192.

Mehlitz, D.; 1986. Le réservoir animal de la maladie du sommeil à Trypanosoma brucei gambiense. Etudes et Synthèses de L'I.M.V.T. 156 p.

Truc, P., Formenty, P., Diallo, P.D., Konoin-Oka, C. & Lauginie, F.; 1997. Confirmation of two distinc classes of zymodemes infecting man and wild animals in Côte d'Ivoire : suspected differences in pathogenicity. An. Trop. Med. Parasitol. 91 (8), 951 - 956.

Walsh Sean, P., Metzger, D. A. et Higuchi, R.; 1991.; Chelex ^® 100 as a medium for simple extraction of DNA for PCR-based typing from forensic material. Bio Technique. 10, 506 - 513.

Genetic differentiation in natural populations of Glossina palpalis s.l. using microsatellite DNA polymorphism : artifacts, demes, or cryptic species?

DIFFERENCIATION GENETIQUE DES POPULATIONS NATURELLES DE GLOSSINA PALPALIS S.L. GRACE AUX MICROSATELLITES : ARTEFACTS, DEMES OU ESPECES JUMELLES?

P. Solano¹, S. Ravel², T. de Meeus³, S. de La Rocque⁴, B. Sane¹, D. Zeze², L. Ndri¹, V. Jamonneau² & G. Cuny²

¹IRD UR 035, Centre Pierre Richet Bouaké, Côte d’Ivoire.

²IRD UR 035, LRCT IRD/CIRAD, Montpellier, France

³UMR CNRS/IRD 062, CEPM, Montpellier, France

⁴CIRAD-EMVT, Montpellier, France

P. SOLANO, Centre Pierre Richet Bouaké, Côte d’Ivoire.

Address : Institut de Recherche pour le Développement, 15 BP 917, ABIDJAN 15, Côte d’Ivoire.

Tel : (225) 21 24 37 79 ; fax : (225) 21 75 47 26 ; e-mail : solano@mpl.ird.fr or solano@ird.ci

Résumé

Une des méthodes de lutte anti-tsé-tsé est la technique du « mâle stérile » (SIT), qui suppose une connaissance précise de la génétique des populations des espèces cibles. Nous présentons les travaux menés sur les deux sous-espèces de Glossina palpalis à des échelles géographiques variées, grâce à l’ADN microsatellite.

Chez G.p. gambiensis, deux loci montrent des différences importantes entre les glossines du Sénégal et du Burkina Faso. Les glossines du Sénégal sont aussi plus petites que celles du Burkina Faso (taille des ailes).

A l’échelle d’une zone agro-pastorale du sud-ouest du Burkina Faso, une structuration de population est trouvée au sein de glossines vivant en sympatrie (Fst=0,07, p<0,001). Cette structuration pourrait refléter le mélange en ce point de populations originaires de bassins versants différents. Des différences dans le taux d’infection et l’identité des trypanosomes sont trouvées chez ces deux groupes de tsé-tsé.

Dans le foyer de maladie du sommeil de Bonon (Côte d’Ivoire), 5 loci microsatellites ont montré de très forts déficits d’hétérozygotes (Fis moyen = 0,366, p<0,001). Des allèles nuls sont identifiés à deux loci, mais ici aussi la coexistence au sein des mêmes pièges, de deux ou plusieurs groupes génétiquement distincts est suspectée.

Ces résultats montrent que des barrières aux échanges génétiques existent de manière répétée dans les populations de Glossina palpalis, et que les conséquences en terme d’épidémiologie et de lutte méritent d’être approfondies.

Mots-clés : Glossina palpalis- Afrique de l’Ouest- microsatellite- Trypanosomose- Déficits d’hétérozygotes

Summary

One of the main methods to fight the tsetse could be the sterile insect technique (SIT), which implies a thorough knowledge of population genetics of targeted species. The work presented here was conducted at various spatial scales in West Africa using microsatellite DNA polymorphism on both subspecies of Glossina palpalis, G. p. gambiensis and G. p. palpalis.

Two microsatellite loci showed significant differences between West African populations. Differences were also found in the size of the wings, those belonging to tsetse from Senegal appearing significantly smaller than those from Burkina Faso.

In an agropastoral area in South West Burkina Faso, population structuring appeared between tsetse living in sympatry (Fst=0.07, p<0.001). This might reflect the junction, at this point, of two different populations (species?) of G. palpalis gambiensis originating from different river basins. Different infection rates and different trypanosome species were found in these tsetse.

In the sleeping sickness focus of Bonon (Côte d’Ivoire), five microsatellite loci showed huge heterozygote deficiencies in the whole sample (Fis over loci=0.366, p<0.001). Null alleles were identified at two loci but again a Wahlund effect is suspected.

The impact of these repetitive departures from random mating in Glossina palpalis might be of importance on the epidemiology and control of human and animal trypanosomoses. The design and efficiency of tsetse control by genetic methods could benefit from more extensive studies.

Key words: Glossina palpalis- West Africa- microsatellite- trypanosomosis- heterozygote deficiency

Introduction

Tsetse flies (Diptera: Glossinidae) are the main vector of trypanosomes (Kinetoplastida: Trypanosomatidae), causal agents of human and animal trypanosomoses in intertropical Africa. These diseases have a considerable impact on public health and on the economic development in this part of the world. In some ecological conditions, tsetse eradication can be one of the ways to solve the problem (Politzar & Cuisance, 1984; Vreysen et al., 2000), and it is the aim of African countries through the PATTEC initiative (Kabayo, 2002; Rogers & Randolph, 2002). The use of this method in the past has proved that it has to be effective and therefore precisely targeted, and with a scenario, first reducing the tsetse populations by various classical methods, then using the SIT (Sterile Insect Technique) to achieve eradication. This implies a thorough knowledge of the epidemiology of the disease, including the biology, ecology and population genetics of the targeted tsetse species and subspecies.

However, our knowledge on the flies still suffer many gaps, especially because of the lack of tools to study the genetic variability of these insects. The main objective of this work was to investigate the genetic variability of the populations of a tsetse species at the intraspecific level, and its possible consequences on the epidemiology and control of African trypanosomoses. Here, we review the work which has been undertaken at various spatial scales on both subspecies of Glossina palpalis sensu lato: G. p. gambiensis, one of the main vector of animal trypanosomoses in the soudano- Guinean belt of West Africa, and G. p. palpalis the main vector of Human African trypanosomosis (sleeping sickness) in West African forest zones.

Material and Methods

Contrary to other better known organisms, no DNA sequences from tsetse flies were available in data banks. It was thus necessary to isolate microsatellite sequences from a partial genomic bank using classical cloning methods. Chronologically, G.p. gambiensis was the first subspecies under study. Three polymorphic microsatellite loci were obtained from a genomic bank of G.p.gambiensis (Solano et al., 1997). The two first loci (Gpg55.3 and Gpg19.62) were located on the X chromosome (one band in the males, XY; one or two bands in the females, XX). The third locus (Gpg69.22) was autosomal. Parental transmission of these microsatellite alleles was checked with tsetse individuals reared in an insectarium. It was possible to confirm the Mendelian inheritance of these alleles and to observe an absence of null alleles on this insectarium, two important points before undertaking genetics studies on the natural populations.

These microsatellite loci were also found to be polymorphic in other tsetse species of medical and veterinary importance (G. fuscipes fuscipes and G. tachinoides).

For G. p. palpalis, 12 polymorphic microsatellite loci (two of them derived from G. p. gambiensis) were tested, as described by Luna et al. (2001). We tried several of these loci on some G. p. palpalis individuals and selected those which were most polymorphic and easily scorable. The study was extended on five polymorphic loci, which are the following: Gpg55.3 (Solano et al., 1997), pGp1, pGp11, pGp 13, pGp24 (Luna et al., 2001).

Data Analyses

A microsatellite sequence is constituted by short tandem repeats from 2 to 6 nucleotides. After identification of these microsatellites sequences (each one constituting a locus), primers flanking both sides of the microsatellite were designed for PCR amplification. The variation in the number of repetition of this microsatellite determines the length of the amplified fragments.

From each single tsetse captured in the field, two to three legs per individual were simply removed and put into an eppendorf tube. These legs, incubated in an aqueous Chelex® solution (chelating resin), provided the DNA subjected to amplification at microsatellite loci (see Solano et al., 1999 for details). According to the size of the observed bands (which represent the alleles) after the PCR with each microsatellite, one can infer the genotype of the individuals. Three software developed for population genetics were used in a complementary way to analyse the data: GENEPOP (Raymond & Rousset, 1995), FSTAT (Goudet, 1995), and GENETIX (Belkhir et al., 1999). The main measured parameters were: the departure from Hardy-Weinberg equilibrium, the linkage disequilibrium between loci and the genetic differentiation between populations. In each population, Wright’s Fis (within sample heterozygote deficiency) and Fst (measure of population differentiation) were estimated using Weir & Cockerham’s (1984) unbiased estimators (f for Fis, q for Fst). The measure of Fis and its significance were conducted only on the females, since males were haploid on X-linked loci (i.e. they have only one X-chromosome). For random mating (within samples) or random distribution of individuals (between samples), F values are expected to be zero. Factorial Correspondence Analyses (FCA) were conducted on multilocus genotypes when necessary.

Other epidemiological information recorded

During the field works conducted at the local scale in G. p. gambiensis and G. p. palpalis area, various information were also recorded for each tsetse individual: sex, trap location recorded by GPS, apparent density in the trap, ovarian age (for the females), presence/absence of trypanosomes by microscopical examination, PCR identification of trypanosomes when present, presence/absence of blood meal, and identification of blood meal source when present (see also de La Rocque et al., 1998; Lefrançois et al., 1998).

Results

Geographic location of all the samples of the two subspecies is shown in Figure 1, where the limit between the two subspecies (adapted from Challier et al., 1983) is drawn: the northern distribution limit of the species, and the different ecological zones in West Africa.

Figure 1: Glosina palpalis in West Afric: Sampling sites for population genetic studies

G.p. gambiensis at its distribution scale in West Africa

Two micro-satellite loci of G. p. gambiensis showed marked differences between the flies originating from Senegal and Burkina Faso. Morph metric analysis of theses flies, conducted on the size of the wings with semi automatic software developed for this purpose (Fly Picture Measurements, Borne et al., 1999), strengthened these genetic differences, flies from Senegal being significantly smaller than those from Burkina Faso.

G.p. gambiensis at the local scale of various agropastoral areas in SouthWest Burkina Faso

At the regional level, two agro-pastoral areas were studied in South Western Burkina Faso. In Samorogouan, captures were carried out during dry and rainy seasons (250 individuals). The results obtained in this area suggest that flies behave like a "panmictic unit" in which individuals seem to reproduce randomly. The more frequent movements occurring in the rainy season were likely to homogenize gene flow between groups of individuals which might be separated during the dry season.

In the Sideradougou area, two campaigns were carried out in 1997 and 1998. A total of 360 G. p. gambiensis were analyzed, originated from two villages fifteen kilometers one from the other: the village of Yéguéré in the eastern part and the village of Nyarafo in the western part of the area. At this scale, a first population structuring appears, which persists over each year of capture. The average Fst parameter (measure of genetic differentiation between populations) reached a highly significative value of 0.07 (p<0,0001), indicating a significant genetic differentiation between these populations which belong to the same hydrographic network. The epidemiological results obtained in this area also distinguish different transmission models, with contrasted infection rates and different trypanosomes species identified by PCR (de La Rocque et al., 1998; Lefrançois et al., 1998).

In addition, high Fis values (within population heterozygote deficiency) were observed inside the western part of the zone, with the flies caught in a sacred wood close to the village of Nyarafo. After having tested several biological and technical assumptions, a Factorial Correspondence Analysis was carried out on the genotypes of these tsetse flies. This analysis revealed two groups of individuals (Figure 2), each of them not showing any more heterozygote deficiency, but exhibiting highly significant genetic differentiation between-groups. These results reflect the existence of a Wahlund effect with, two sympatric genetically different groups of individuals artificially gathered (Solano et al., 2000).

Fig 2. Factorial correspondence Analysis on G. p gambiensis within sideradougou area Burkina Faso

G.p.palpalis at the scale of a Human African Trypanosomosis (HAT) focus in Center West Côte d’Ivoire

The study was undertaken in the focus of Bonon (6°55’N, 6°W), in the preforest zone of central-western Côte d’Ivoire. During a medical survey conducted in April 2000, a total of 96 patients were diagnosed positive for HAT (sleeping sickness) (Solano et al., 2003). Entomological surveys were conducted in November 2000 (end of rainy season) and January 2001 (cold dry season), with the settling of 320 Vavoua traps (Laveissière & Grébaut, 1990) in the living places of the patients (home, water supply sites, fields).

Genetic analyses were first conducted with five microsatellite loci on the tsetse population of January 2001 (101 individuals). Weir & Cockerham parameters indicated huge heterozygote deficiencies inside the population (Fis over loci=0.366, p<0.001, with values of Fis by locus ranging from 0.12 to 0.56). All the five loci deviated from Hardy Weinberg expectations, showing a much lower number of heterozygotes than expected. For the five loci 55.3, 1, 11 , 13 and 24, the number of distinct alleles recorded was respectively 17, 16, 11, 13 and 13.

At one locus (Gpg55.3) showing high Fis values, the presence of null alleles could be ruled out, because this locus is situated on the X chromosome, and expected “null” males in the case of null alleles were not observed.

Primers were redesigned on the two loci with the highest Fis values (pGp24, Fis=0.56 and pGp13, Fis=0.48), then amplifications were conducted again. The Fis decreased to 0.42 and 0.39 respectively but still were very significant. It is likely that null alleles exist at these loci, that they could account for part of the deficiency but other reasons also have to be investigated because null alleles do not explain all the deficiencies. Analyses are still being conducted to elucidate these within-group heterozygote deficiencies, in particular taking into account a spatial compound and an age-dependent phenomenon.

Discussion

We should first remind that the two subspecies of flies live in different ecozones, as illustrated by their distribution limits on Figure 1. Challier et al., 1983, define G. p. gambiensis as a subspecies living in humid and dry savannas, whereas G. p. palpalis lives in forests and humid savannas.

G. p. gambiensis

Thanks to the high polymorphism of microsatellite sequences, a structuration of the populations of G. p. gambiensis has been detected on distinct geographic scales. It is obvious that a higher number of loci would have improved the analysis, but preliminary conclusions can already be drawn. To our knowledge it is the first time that such a subdivision of tsetse populations is observed at the intraspecific level, in a limited spatial scale of a few kilometers. In the Sidéradougou area, this subdivision persisted during time. An epidemiological “structuration” is superimposed to the genetic one: various populations of vectors belonging to the same subspecies seem to be able to play different roles in the efficiency of the transmission of animal trypanosomes. In a sympatric situation near Nyarafo, mixed populations were detected. Among several assumptions, this mixture might reflect the presence of two different populations of G. p. gambiensis coming from different river basins. In this place, indeed, a contact point is established between the Basin of Comoé River, which runs South towards the Côte d’Ivoire, and the basin of the Mouhoun (Volta Noire) which runs North and to which belongs the Koba river (Figure 3).

Figure 3: Hypothesis of the River Basin to explain genetic differentiation on G. palpalis gambiensis in Sideradougou

At the scale of the distribution range, the genetic differences underlined do not appear simply related to geographic distance. It is likely that other ecological factors, specific to each deme, play a part in the structuration of the populations of this vector.

Comparing to the few other studies which addressed to genetic studies on natural populations of tsetse, it can be noted that deviations from random mating were also observed in the 3 subspecies of G. morsitans: G.m. submorsitans, G.m. morsitans, and G.m. centralis, where Krafsur & Endsley (2002) found very restricted gene flow between populations of this savannah species. In G. pallidipes, also a savannah species, genetic distances were correlated with geographical distance between Kenya and South Africa (Krafsur, 2002).

G.p. palpalis

In the focus of Bonon, in the Center-West of Côte d’Ivoire, G. p. palpalis is the vector of Human Sleeping Sickness. It has been the only tsetse found in the traps, with apparent densities of 2.6 flies/trap/day in January 2001 (cold dry season), ranging from 0 to more than 100 in some of the traps. Highly significant values of Fis, the within-population heterozygote deficiency index, were observed on the 5 loci.

The presence of null alleles was investigated and was found at two of the 5 loci. When redesigning new primers at these loci, the Fis decreased but remained highly significant.

Analyses are still being conducted, and it already appears that we face a situation which looks very much like the situation of G. p. gambiensis: a geographical structuration, together with a Wahlund effect. In G. p. gambiensis we might have expected a population structure based on river basin origin, because the linear distribution of the tsetse in savannah areas (Cuisance et al., 1985) is known. But the result was not expected in G. p. palpalis because the pattern is different in forest areas, where the spatial distribution of this subspecies is more homogeneous on a macrogeographic scale (the presence of water itself does not become anymore a key parameter). When “zooming” progressively at microgeographic scales, the distribution then depends on local parameters: presence of water hole, presence of pigs in a village, age of the coffee or cocoa plantation, etc (Gouteux, 1982; Laveissière et al., 1986).

The analyses are still in progress and questions remain, since the cause of this Wahlund effect has to be elucidated. For the moment, spatial and age-dependent (Sané et al., 1999) factors are studied, and the existence of cryptic species (as already evoked for G. p. gambiensis) might be an explanation but remains to be investigated further. The results are very difficult to compare with others, because to our knowledge very few genetic studies were conducted on forest tsetse species.

Conclusion

Heterozygote deficiencies seem to be the rule within and among populations of both subspecies of G. palpalis. In G. p. gambiensis in Burkina Faso, genetically differentiated groups appeared to be able to have different vectorial capacities. If this phenomenon was also observed in G. p. palpalis, the consequences on the transmission of Human Sleeping Sickness might be of importance. The causes of these repetitive lacks of heterozygotes in tsetse populations in different studies, and over different tsetse species, and also on vectors of other diseases (e.g. Dumas et al., 1998) have to be explained, and show how little is known on vector population genetics and biology.

The information expected from such approaches might be helpful for designing control strategies, which could be of different efficacy according to the degree of genetic isolation/differentiation of some demes or subpopulations. For instance, if the aim is the eradication of the vectors by SIT (Politzar & Cuisance, 1984; Vreysen et al., 2000), these studies will allow identification if the target population is genetically compatible with the released individuals, and to specify the probability of possible reinvasions of treated zones.

It is recommended that more genetic studies be done on these vectors of Human and Animal diseases with the view of assessing the design and probability of success of control strategies, as African countries have taken the initiative of fighting against these diseases of public health and economic importance.

Acknowledgements

This work is the result of close collaboration between scientists from IRD (previously ORSTOM), CIRAD-EMVT, Institut Pierre Richet (Côte d’Ivoire), and CIRDES (Burkina Faso). The work benefited funds from IRD, CIRAD-EMVT, FAO/IAEA, CNRS, AUF, french cooperation.

References

Kabayo J.P. (2002). Aiming to eliminate tsetse from Africa. Trends Parasitol., 18, 473-475.

Rogers D.J. & Randolph S. E.(2002). A response to the aim of eradicating tsetse from Africa. Trends Parasitol., 18, 534-536.

Politzar H. & Cuisance D. (1984). An integrated campaign against riverine tsetse Glossina palpalis gambiensis and G. tachinoides by trapping and the release of sterile males. Insect Sci. Appli., 5, 439-442.

Vreysen M.J.B., Saleh K.M., Ali M.Y., Abdulla A., Zhu Z.R., Juma K.G., Dyck V.A., Msangi A.R., Mkonyi P.M. & Feldmann U (2000). The use of the Sterile Insect Technique (SIT) for the eradication of the tsetse fly Glossina austeni on the island of Unguja (Zanzibar). J. Econ. Entomol., 93, 123-135.

Solano P., Duvallet G., Dumas V., Cuisance D. & Cuny G. (1997). Microsatellite markers for genetic population studies in Glossina palpalis. Acta Tropica, 65, 175-180.

Luna C., Bonizzoni M., Cgeng Q., Aksoy S. & Zheng L. (2001). Microsatellite polymorphism in the tsetse flies (Diptera: Glossinidae: Nemorhina). J. Med. Entomol., 376-381.

Solano P., de La Rocque S., Cuisance D., Geoffroy B., de Meeüs T., Cuny G. & Duvallet G. (1999). Intraspecific variability in natural populations of Glossina palpalis gambiensis from West Africa, revealed by genetic and morphometric analyses. Med. Vet. Entomol., 13, 401-407.

Raymond M. & Rousset F. (1995). GENEPOP (version 1.2): Population genetics software for exact tests and ecumenicism. J. Hered., 86, 248-249.

Goudet J. (1995). FSTAT (version 1.2): A computer program to calculate F-statistics. J. Hered., 86 (6), 485-486.

Belkhir K, Borsa P, Goudet J, Chikhi L, Bonhomme F, 1996 - 1998. GENETIX, logiciel sous WindowsTM pour la génétique des populations. Laboratoire Génome et Populations, CNRS UPR 9060, Université de Montpellier II, Montpellier (France).

De La Rocque, S., Lefrançois, T., Reifenberg, J.M., Solano, P., Bengaly, Z., Kaboré, I., Cuisance D. (1998). PCR analysis and spatial repartition of trypanosomes infecting tsetse flies in Sideradougou area (Burkina Faso). Ann. N.Y. Acad. Sci. 849, 32-38.

Lefrançois T., Solano P., de La Rocque S., Bengaly Z., Reifenberg J.M., Kaboré I. & Cuisance D. (1998). New epidemiological features on animal trypanosomosis by molecular analysis in the pastoral zone of Sidéradougou, Burkina Faso. Mol. Ecol., 7, 897-904.

Challier A., Gouteux J.P. & Coosemans M. (1983). La limite géographique entre les sous-espèces Glossina palpalis palpalis (Rob.-Desv.) et Glossina palpalis gambiensis Vanderplanck en Afrique occidentale. Cah. O.R.S.T.O.M., sér. Ent. Méd. Parasitol., XXI (4), 207-220.

Borne F., Petiteau L., Geoffroy B., de La Rocque S. & Cuisance D. (1999). Fly Picture Measurement, un nouvel outil informatique pour l’étude des glossines. Revue Elev. Méd. vét pays Trop., 52, 19-21.

Solano P., de La Rocque S., de Meeüs T., Cuny G., Duvallet G. & Cuisance D. (2000). Microsatellite DNA markers reveal genetic differentiation among populations of Glossina palpalis gambiensis collected in the agro-pastoral zone of Sideradougou, Burkina Faso. Insect Mol. Biol., 9, 433-439.

Solano P., A. Koné, A. Garcia, B. Sané, V. Michel, J.F. Michel, B. Coulibaly, V. Jamonneau, D. Kaba, S. Dupont & F. Fournet (2003). Rôle des déplacements des malades dans l’épidémiologie de la Trypanosomose Humaine Africaine dans le foyer de Bonon, Côte d’Ivoire. Médecine Tropicale, in press.

Laveissière C. & Grébaut P. (1990). Recherches sur les pièges à glossines. Mise au point d’un modèle économique : le piège « Vavoua ». Trop. Med. Parasitol., 41, 185-192.

Krafsur E.S. & Endsley M.A. (2002). Microsatellite diversities and gene flow in the tsetse fly, Glossina morsitans s.l. Med. Vet. Entomol., 16, 292-300.

Krafsur E.S. (2002). Population structure of the tsetse fly Glossina pallidipes estimated by allozyme, microsatellite, and mitochondrial gene diversities. Insect Mol. Biol., 11, 37-45.

Cuisance D., Fevrier J., Dejardin J., Filledier J. (1985b). Dispersion linéaire de Glossina palpalis gambiensis et G. tachinoides dans une galerie forestière en zone soudano-guinéenne. Rev. Elev. Méd. Vét. Pays trop, 38, 2, 153-172.

Gouteux J.P. (1982). Observations sur les glossines d’un foyer forestier de trypanosomiase humaine en Côte d’Ivoire. 4. Peuplement de trois plantations de caféiers d’âge différent. Cah. O.R.S.T.O.M., sér. Ent. Méd. Parasitol., XX (1), 29-39.

Laveissière C., Hervouet J.P. & Couret D. (1986). Localisation et fréquence du contact homme/glossine en secteur forestier de Côte d’Ivoire. 1. Recherche des points épidémiologiquement dangereux dans l’environnement végétal. Cah. O.R.S.T.O.M., Sér. Ent. Méd. Parasitol., 24, 21-35.

Sané B., Garcia A., Fournet F. & Laveissière C. (1999). Répartition des groupes d’âge de Glossina palpalis palpalis femelle dans les plantations et les talwegs en zone forestière de Côte d’Ivoire. Bull. Soc. Pathol. Exot., 92, 210-212.

Dumas, V., Herder, S., Bebba, A., Cadoux-Barnabe, C., Bellec, C. & Cuny, G. (1998) Polymorphic microsatellites in Simulium damnosum s.l. and their use for differentiating two savannah populations. Implications for epidemiological studies. Genome, 41, 154-161.