f 154

Polymorphic microsatellites in Simulium damnosum S.I. and their use for differentiating two savannah populations: implications for epidemiological studies

t I Valériebumas, Stéphan Herder, Kicha Bebba, Cécile Cadoux-Barnabé, Christian/Bellec, and Gé arqkuny

Abstract: In West Africa, Onchocerca volvulus, the cause of human onchocerciasis, is transmitted by sibling species of the Siinuliuin dainizosuin complex. Little is known about blackfly intraspecific variability and its consequences on.vectoria1 capacity. This study reports the use of microsatellite markers for differentiating populations of S. damiiosuin s.1. Five nlicrosatellite loci were characterized and used to analyze individuals from two savannah populations in Mali, 120 km apart. Four loci were highly polymorphic, having 8-12 alleles per locus and gene diversities ranging from 77.9 to 88.2%. A significant heterozygote deficiency was observed in the two populations. This may arise from inbreeding, population structure (the Walhund effect), or the presence of null alleles. To test this last hypothesis, new primers were designed for two loci and used to analyze homozygous individuals. After correcting for null alleles, heterozygote deficit persisted. Population subdivision in the two foci remains the most likely explanation. Our results indicate that microsatellite markers could differentiate fly populations, making them valuable tools for the study of population genetic structure.

Key words: Siinidiuin dainnosuin S.I., microsatellites, polymorphism, population structure, population genetics, null alleles.

Résumé : En Afrique de l’Ouest, Onchocerca volvulus, l’agent responsable de l’onchocercose chez l’homme, est transmis par des espèces du complexe Simulium damnosunz. La variabilité intra-spécifique et ses conséquences sur la capacité vectorielle ont été peu étudiées. Cette étude d+t l‘utilisation des marqueurs microsatellites pour différencier des populations de S. dainnosuin S.I. Cinq loci microsatkillites ont été caractérisés et utilisés pour caractériser des individus de deux populations savanicoles du Mali, distantes de 120 kms. Quatre loci se sont révélés très polymorphes avec 8-12 allèles par locus et des diversités géniques variant de 77.9 à 88.2%. Un déficit en hétérozygotes significatif a été observé dans les deux populations. Ce déficit peut être expliqué par la consanguinité, la structuration des populations (l’effet Walhund) ou par la présence d’allèles nuls. Pour tester cette demière hypothèse, de nouvelles amorces ont été désignées pour deux loci et utilisées pour analysefles individus homozygotes. Malgré cette correction des allèles nuls, le deficit en hétérozygotes persiste. La subdivision des populations dans les deux foyers reste l’explication la plus probable. Nos résultats montrent que les marqueurs microsatellites peuvent différencier des populations de simulies, ce qui en fait des outils de choix pour I’étude de la structure génétique des populations.

Mots cZés : Sirnuliurn dunziiosum SA, microsatellites, polymorphisme, structure des populations, génétique des populations, allèles nuls.

Corresponding Editor: B. Golding.

Received June 4,1997. Accepted November 15, 1997.

V. Dumas, A. Bebba, C. Cadou-Barnabé, C. BeUec, and G. Cuny.’ Laboratoire d’Epid6miologie des Maladies à Vecteurs, Institut Français pour la Recherche et le Développement en Coopération (ORSTOM), 91 1 avenue Agropolis, B.P. 5045,34032 Montpellier Cédex 1, France. S. Herder. Organisation de Coordination pour la Lutte Contre les Endemies en Afrique Centrale (OCEAC), B.P. 288, Yaoundé, Cameroon.

I Author to whom all correspondence should be addressed (e-mail: cuny@orstom.rio.net). .-

! I

l

l

I .i

I i i i

I i

i 1 i

I I !

I i l

i I !

i t

i

! ! !

I

l

I

Dumas et al.

In West Africa, blackflies (Diptera: Simuliidae) of the S. damnosum complex serve as vectors for O. volvulus. Larval polytene chromosome studies have demonstrated the existence of multiple sibling species in the S. damnosum complex (Va- jime and Dunbar 1975). Largely on the basis of fixed inversion differences, at least nine sibling species have been identified (Boakye 1993; Wilson and Post 1994), some of which are restricted to either the savannah or the forest. Simulizim squamosum, Simulium ynhense, Simulitiin sanctipauli, Simulium soubrense, Siinuliunz leonense, and Simulium konk- ourense belong to the forest group and are divided into two subcomplexes, S. sanctipauli and S. squamosum. Simulium damnosum S.S., Simulium sirbanum, and Simulium dieguer- ense are classified under the S. damnosuin subcomplex of the savannah group. This classification was made up on the basis of the different species' habitats. All these species are vectors of onchocerciasis, but with various vectorial capacities.

To better understand disease epidemiology and to develop more effective control strategies, it was necessary to examine the vectorial role of the different Simulium species according to the O. volvulus strains. Duke et al. (1966) showed that para- sites from the savannah area developed very poorly in forest flies, but well in savannah flies, and vice versa for forest para- sites, leading to the theory of vector-parasite transmission complexes.

Utilization of molecular tools has allowed accurate identi- fication of both parasite strains and vector species. Meredith et al. (1991) developed a method based on a combination of PCR and DNA probes that can discriminate between the two O. volvulus strains directly in vector blackflies, and we devel- oped microsatellite markers to identify parasite populations (S. Herder, V. Dumas, J.-P. Aussel, C. Bellec, and G. Cuny, submitted for publication)? Concerning vector identification, two main approaches have been used: (i) cytological analysis. of polytene chromosomes, which can only be performed in larvae, and (ii) an isoenzyme approach investigated by

/ Meredith and Townson (1981), who examined 44 enzyme sys- tems of the S. damnosum complex. However, this method al- lowed only two species of the complex to be distiguished.

A wide range of molecular techniques has been tested (Post and Flook 1992; Agatsuma et al. 1993; Brockhouse et al. 1993; Wilson and Post 19941, though none enabled differen- tiation in every sibling species in adult flies. More recently, a method developed by Tang et al. (1995), based on direct het- eroduplex analysis (DHDA) of mitochondrial DNA, allowed differentiation in every sibling species except S. sqiianzosum and S. yahense. Using this technique in combination with worm-specific probes, Toé et al. (1997) demonstrated that no preferential transmission of the two strains of O. volvulus by the different sibling species occurred in the transition zone between savannah and forest, where intermediate disease pat- tem has been detected. If transmission of the different strains of O. volvulus might occur whatever the Simuliunz species, it becomes important to identify fly populations more accurately according to their vectorial capacities and to find polymorphic markers that can discriminate between them.

To this end, we have chosen microsatellite markers, which

* S. Herder, V. Dumas, J.-P. Aussel, C. Bellec, and G. Cuny. Genetic variation of Onchocerca volvulus using microsatellite DNA sequences. Submitted for publication.

155

are powerful tools for measuring genetic variation within populations and gene flow among populations (Bruford and Wayne 1993; Jarne and Lagoda 1996). Microsatellites consist of tandem repeats of short (1-6 bp) nucleotide motifs widely dispersed throughout eucaryotic genomes. Most microsatel- lites are highly polymorphic, as a result of variation in the number of repeat units. Length polymorphism is likely to be due to slippage occurring during DNA replication (Schlötterer and Tautz 1992). Different alleles are characterized by the length in base pairs of DNA amplified by PCR using primers derived from sequences flanking the microsatellite. Because of their codominance, neutrality, and Mendelian inheritance, microsatellites can be very useful for population genetic stud- ies. They have been used recently in social insects, such as wasps (Hughes and Queller 1993; Thorén et al. 1993, ants (Gertsch et al. 1995), and bees (Estoup et al. 1993, 1995a, 1995b), and also in Anopheles gambiae, the principal vector of malaria (Lanzar0 et al. 1995), and Glossina palpalis (So- lano et al. 1997). These population genetics surveys showed the high efficiency of microsatellites for differentiating insect populations and studying their genetic structure.

Thus, microsatellites can play a role in vector control in the epidemiology of onchocerciasis, as they can be a powerful tool for (i) analyzing reinvasion phenomena in treated areas, since the migration of flies from untreated regions could lead to a recrudescence of O. volvulus infection, and (ii) identifying populations resistant to insecticides.

We report here the isolation and characterization of five microsatellite loci in S. damnosum s.l., focused on the genetic variability in two populations of S. damnoszinz s.1. from Mali.

Materials and methods

Collection of blackflies Larvae used in the construction of a partial genomic library were collected from a S. sqiiamosrim population from the Sanaga River in Cameroon.

Adult flies (provided by the Onchocerciasis Control Program (OCP)) used in this study were collected at human bait on the same day (10/14/94) from two savannah foci in Mali: N'Zana (a reinvasion focus), situated on the Baoule River (1 1"49'N, 7"1 IV), and Tienfala (a permanent focus), on the Niger River (12"44'N, 7O45'W). These two foci were under OCP control at this period. Flies were preserved in absolute ethanol and stored at 4°C.

Banding patterns of giant polytene chromosomes found in salivary glands were used for cytotaxonomic identification of larval samples collected in the rivers close to the capture points (Vajime and Dunbar 1975). At the Tienfala site, 100% of the larvae were S. sirbanurn. At the N'Zana focus, larval identification indicated the presence of two species, S. sirbunilm and S. damnosnm, in the dry season. After insec- ticide treatment, this focus became unproductive until the reinvasion process. Cytotaxonomic identifications done after reinvasion re- vealed that 95% of the samples were S. sirbanum.

We used adult samples for this study and, as is generally assumed, we considered that larvae would be an accurate reflection of adult species distribution.

DNA preparation Simulium sqziamosum DNA was extracted from a pooled larvae. Fifty larvae were frozen and thawed three times and homogenized with a mortar and pestle. The homogenate was then transferred to a tube in NET buffer (50 mM NaCl, 10 mM EDTA, plus 50 mM Tris-HC1, pH 8.0); proteinase K (Boehringer) and Triton X-100 (Sigma) were

O 1998 NRC Canada

I

. . , . . .- . . . .. -... . *_*__. <.. '- '

156

added to final concentrations of 100 pg1m.L and 1%, respectively. The mixture was then incubated at 60°C for 1 h and at 37°C overnight. RNase A (Boehringer) was added to a final concentration of 100 pg/mL and the solution was incubated for a further 30 min at 37°C. DNA was extracted twice with phenol-chloroform and precipi- tated with ethanol.

For PCR analysis of microsatellites, DNA was first extracted as above from 49 adult flies of each population. Thereafter, to reduce the number of steps in sample preparation, we improved the method of extracting DNA from flies by using Chelex 100 (Bio-Rad) chelating resin (Walsh et al. 1991). Five percent Chelex 100 (100 pL) was added to one leg of a blackfly. After incubation at 56°C for 1 h, DNA was denatured at 95°C for 30 min. During boiling, Chelex would prevent the degradation of DNA by sequestering metal ions. A 20-pL aliquot of the supernatant was added to the PCR mixture and the remainder of the sample was stored frozen.

Microsatellite cloning and sequencing DNA was digested with the restriction enzyme HueIII and size- fractionated on a 1.5% low melting point agarose gel. Restriction fragments ranging from 300 to 600 bp were extracted from the gel and ligated into the EcoRV dephosphorylated site of M13BM20 vector (Boehringer Mannheim). The ligation mixture was transformed into Escherichia coli XL1 cells, using an electroporator II (Invitrogen).

After replica plating, the genomic M13 library was screened by transferring plaques onto Hybond N+ nylon membranes (Amersham). The filters were successively soaked in denaturing (0.5 M NaOH - 1.5 M NaCl) and neutralizing (0.5 M Tris (pH 8.0) - 1.5 M NaCl) buffers and air-dried. Equal mixtures (100 ng) of poly(CA),/ poly(GT), and poly(GA),/poly(CT), used as probes were labelled by random priming with 100 pCi (1 Ci=37 GBq) of [CX~~-P]~CTP, using a Megaprime DNA labelling kit (Amersham). Prehybridization was carried out in a hybridization oven for 1 h at 65°C in a buffer prepared from hybridization buffer tablets (Amersham). The klenatured probe was then added directly to the filters (in the same buffer) and hybrid- ized for 4 h at 65°C. Hybridization was followed by two washes of 15 min each in 2~ SSPE (IX SSPE: 0.18 M NaCl, 10 mM NaPO,, plus 1 mM EDTA, pH 7.7) - 0.1% SDS at room temperature, and two washes of 15 min in IX SSPE - 0.1% SDS at 65°C. Dried filters were then autoradiographecYon Hyperfilm (Amersham) at -70°C for 2-16 h. Single-stranded template DNA was prepared from all plaques showing strong hybridization signals and dotted onto a filter. After hybridization with the same probes, clones giving the strongest sig- nals were selected for sequencing.

DNA sequencing was performed using the dideoxy chain termina- tion method (Sanger et al. 1977) with the Taq Dye Primer Cycle Sequencing Kit (Applied Biosystems), and the reaction products were analyzed using an Applied Biosystems 373A DNA sequencer.

Detection of length polymorphism in microsatellite loci Specific primers were chosen from the region immediately flanking the microsatellite sequence to give PCR products in the range of 150-200 bp. These primers, 20-22 bases in length, were selected with computer assistance (OLIGO software, version 3.2, National Bio- sciences, Inc.), to minimize self-annealing, and synthesized (Euro- gentec). Standard PCRs were carried out using a DNA thermal cycler (Perkin-ElmerKetus) in 50 pJl, final volumes, containing 20 ng of purified DNA or 20 pL of DNA prepared by Chelex extraction, 20 pmol of each primer, 200 pM dNTPs, 1.5 mM MgCl,, lx reaction buffer ( 1 O m M Tris-HC1 pH 9.0,50mM KC1,1.5 mM MgCl,), and 1 U Taq DNA polymerase (Perkin-ElmedCetus). After an initial denaturation at 92°C for 5 min, samples were processed through 30 or 40 cycles (according to the locus) consisting of a denaturation step at 92°C for 30 s, an annealing step at 50°C for 30 s, and an extension step at 72°C for 1 min. The last elongation step was lengthened to 10 min. Ampli- fications were checked on 2% agarose gels visualized by ethidium bromide staining under UV light. PCR bands were then resolved on

Genome, Vol. 41, 1998

nondenaturing acrylamide gels (12.5%) and revealed by a rapid sil- ver-staining procedure (Sanguinelti 1994).

The precise size of alleles was determined by PCR in the presence of [~t~~-sldATP, followed by electrophoresis on a 6% denaturing polyacrylamide gel containing 8 M urea. The PCR conditions were those described above, except that the dNTP concentrations (75 pM dGTP, dCTP, and dTTP; 6 pM dATP; and 3 pCi [c~~~-S]~ATP) and the number of cycles (25 cycles) were lowered. The sequencing reac- tion of the M13 phage obtained using the T7 Sequencing Kit (Phar- macia) was used as a precise size marker.

Data analysis Genetic polymorphism in each population was measured as the mean number of alleles per locus, observed heterozygosity, and gene diver- sity (Nei 1987). Deviations from Hardy-Weinberg expectations, genotypic linkage disequilibria, and differentiation among popula- tions were tested using the GENEPOP software, version 1.2 (Raymond and Rousset 1995). The test of Hardy-Weinberg proportions for less than five alleles is an exact test. For five alleles or more, a Markov chain is used to obtain an unbiaised estimate of the exact probability. The genotypic linkage disequilibrium between pairs of loci was esti- mated using the common correlation coefficient (Weir 1990), and tested using Fisher's exact test on a contingency table. To test for genotypic linkage disequilibrium and genetic differentiation among populations, GENEPOP also uses a Markov chain method. In all cases, the Markov chain was set to 100 O00 steps and 1000 steps of de- memorization. Standard errors were always below 0.005. The overall significance of multiple tests was estimated by Fisher's combined probability test (Fisher 1970). FIS rates the possible decrease in hetero- zygosity of individuals in their subpopulation. A heterozygote defi- ciency is indicated by FIS > O, whereas for a subpopulation at Hardy-Weinburg equilibrium, FIS = O; FIS (inbreeding coefficient) values were computed according to Weir and Cockerham (1984), using the GENEPOP software.

Results

Characteristics of microsatellite loci The plaques of the partial M13 library were replicated onto 22 nylon discs and screened with the radiolabelled (CNGT), and (GNCT),, probes. These probes were chosen because of the high abundance of these types of microsatellite sequences in insect genomes (Thorén et al. 1995). Of approximately 4000 recombinant clones, 57 gave a positive signal. These clones were dot blotted and rescreened. to ensure specificity. The 29 clones giving the strongest signals were sequenced; 18 were false positives and microsatellite sequences were successfully obtained for 11, of which 7 were unique. The presence of false positives can be explained by an imperfect homology of se- quence between the clones and the microsatellite probes, ow- ing to the low stringency of the hybridization washes. Primer pairs could be designed for 5 of them only, as the remaining two had microsatellite repeat located too close to the cloning site to allow primer selection.

Simple DNA repeat sequences may be considered as microsatellites when the number of repeats is greater than six for dinucleotide,,motifs and greater than four for trinucleotide motifs (Stallings et al. 1991). Following these criteria, one microsatellite was excluded, because it included four repeats only: (CAA),. Four microsatellite loci were then selected for analysis by PCR amplification. A fifth microsatellite locus was identified from a published sequence of the spacer region be-

L

O 1998 NRC Canada

- Dumas et al.

Table 1. Microsatellite loci, forward and reverse primer sequences, and length of allele cloned.

Primers Locus Repeat sequence Designation 60.1 (GT)AT(GT)AT(GT),o ss 1

s s2 64.2 (GT)GC(GT),, ss3

ss4 SS3P s s 4

7.1 (ACG),(ACA)(ACG)(ACC), . SS5 (ACG)(ACA)(ACG) SS6

7.4 (GT),,IT(GT) ss7 SS8

H 3-4 (CAG),(CAA),o(CAG) ss9 SSlO SS9P SSlOP

Sequence (5'+3') CCCATTTGCCAGTTGAGGTGA CCCGTCAACATTGTGGCTACG GACGCATACCGAGTCCTTGT TACGCACACATTTTTCTATTIC ATCATGACGAGGACGCACTC TACGCACACATMTTCTATTTC AAGGAAGCCCCATGGTCGTC CTTCCAACTTACGCAGAGCC CGCTAACGCTGTGCAATATTG TGACGAAC'MTGGGACGACA

CGAAAACAACATACGAAGGG AGCAGTTI'GTTTGGTACGAC A AATTTTAAATACGTACGAGGG

CGACAA~GTGTCTCGACAAA

157

Product length (bp) 193

159

180

161

195

170

209

Fig. 1. An example of microsatellite polymorphism'at locus 64.2 in the N'Zana population. Lanes: 1-13, PCR products corresponding to 13 fly individuals; L, 20-bp ladder.

L 1 2 3 4 . 5 6 7 8 9 10 11 12 13 L

tween the two histone genes H3 and H4 (Wilson and Post 1994). These five repeat sequences are shown in Table 1.

Using the classification of Weber (1990)' loci 60.1, 64.2, and 7.4 can be classified as imperfect, with point mutations within the repeat motif, and loci 7.1 and H3.4 as compound.

Microsatellite polymorphism The five pairs of primers (Table 1) developed for the S. dam- nosum complex were first tested on larval DNA purified from a few individuals from each of the following sibling species: S. sqnamosum, S. soubreme, and S. sirbanum. All loci could be amplified in every tested species and presented a high level of polymorphism within and among species (data not shown).

To investigate whether these microsatellite loci showed an intrapopulational polymorphism in the S. damnosum complex, PCR analysis was performed on Chelex-purified DNA of 49 individuals from the N'Zana population. Amplifications of DNA from each individual of the population were successful with all primer sets. An example of the polymorphism of microsatellites resolved on nondenaturing polyacrylamide gel is shown in Fig. 1. All loci, except locus 7.1, were polymorphic.

.

180 bp

160 bp

140 bp

120 bp

The precise size of alleles was scored on a denaturing polyacrylamide gel, using an M13 sequence ladder. Alleles differed in length by multiples of 2 bp for loci 60.1'64.2, and 7.4 and of 3 bp for locus H3-4. This suggests that length vari- ation results from differences in the number of repeat units in the microsatellite region. The number of alleles observed in the N'Zana population is given in Table 2. All loci exhibited at least eight alleles, and each microsatellite was found to be highly polymorphic, with gene diversity ranging from 77.4 to 86.7% (Table 2).

The observed heterozygosities at the four loci ranged be- tween 55.1 and 81.6%. Comparisons between observed hetero- zygosity and gene diversity resulted in significant hetero- zygote deficiency at loci 64.2 and H3-4. The Hardy-Weinberg test was performed for each locus. The hypothesis of the Hardy-Weinberg equilibrium (HWE) was rejected (p < 0.05) for two of the four loci (64.2 and H3-4) (Table 2). Note that the level of significance of departure from HWE is around 6% for locus 7.4. When multiple tests were combined (Fisher's method) for overall significance, HWE was rejected ( p < IO4). FIS represents the extent of overall deviation from

O 1998 NRC Canada

158 Genome, Vol. 41, 1998

Table 2. Number of alleles, observed heterozygosity, gene diversity, exact probability for departure from Hardy-Weinberg proportions, and FIS values for each locus in the two populations.

No. of Observed Gene FIS Locus Population alleles heterozygosity (%) diversity (%) HW test

60.1 N’Zana . 10 81.6 Tienfala 11 . 74.0

64.2 N’Zana 11 70.8 (55.1)

Tien fala 12 63.3 7.4 N’Zana 8 71.4

Tienfala 9 71.7 H3-4 N’Zana 8 64.6

(57.1)

86.1 86.7 87.1

(84.3) 88.2 79.9 83.4 77.9

(77.4)

0.925 0.06 0.034 0.001

0.0004 0.061. 0.001 0.007

(404)

( 4 0 3

0.14 0.19

(0.35) 0.28 0.11 0.14 0.17

(0.26) Tienfala 10 74.0 79.6 0.12 0.07

Note: Null alleles were taken into account in all analyses. Values in parentheses are those obtained with the original primers.

HWE. FIS values were positive at all loci (Table 2), indicating extensive heterozygote deficiencies, with a combined value over all loci of 0.195. This apparent heterozygote deficiency could be due to many factors: inbreeding, population subdivi- sion, or presence of null alleles.

Null alleles Mutations or deletions within the DNA sequences comple- mentary to PCR primers may inhibit their binding, leading to a reduced or complete loss of allele amplification. Such non- detectable or null alleles produce heterozygote deficiency, since some heterozygous individuals are mistyped as homozy- gotes. Null alleles can be revealed through lo@@ring the prim- ing stringency or redesigning primers (Callen et al. 1993). New primers flanking the original primers were designed to generate larger products*for the two loci showing heterozygote deficiency (64.2 and H3-4), and all the originally homozygous individuals were reawlyzed. The sequences of these primers and the new expected sizes of the amplified DNA are given in Table 1. Only the forward primer was redesigned for locus 64.2, as the reverse strand was too short to allow new primer selection. Of 22 homozygotes at locus 64.2 and 21 homozy- gotes at locus H3-4,7 and 2 individuals, respectively, became heterozygous with the new primers. The molecular basis of null alleles was investigated by sequencing the seven PCR products of locus 64.2. Sequence data revealed: (i) the null allele was caused by five point mutations with respect to the original forward primer, which prevented correct annealing in PCR experiments; (ii) the seven sequenced alleles exhibited the same mutations at the same positions; and (iii) these muta- tions are associated with two null alleles that have different numbers of (GT) repeats, i.e., alleles (GT)13 and (GT)14.

The genotypic analysis was performed again, taking into account null alleles. Heterozygosity rose from 55.1 to 70.8% at locus 64.2 and from 57.1 to 64.6% at locus H3-4 (Table 2). However, heterozygote deficiencies were still observed. The results of the Hardy-Weinberg tests are given in Table 2. Loci 64.2 and H3-4 still showed significant deviations from HWE.

Genetic variability among populations To assess the potential of the microsatellite approach for popu- lation genetics of S. ~ i i ~ t u l i ~ m , we decided to perform a similar study in another geographic area, the Tienfala focus, 120 km

away. The results are shown in Table 2. All loci were highly polymorphic, except locus 7.1, with 9 alleles at the least vari- able locus (7.4) and 12 alleles at the most variable locus (64.2). Gene diversities were high, varying between 79.6 and 88.2%. Of the four polymorphic loci, three (60.1, 64.2, and 7.4) showed significant heterozygote deficiencies and significant deviation from HWE. Combined probability tests over all loci revealed a significant departure from HWE for the two popu- lations (p = 0.0003 for N’Zana andp < lo4 for Tienfala), and combined FE values revealed important heterozygote defi- ciencies (FIS = 0.13 for N’Zana and FI, = 0.16 for Tienfala).

The distributions of allele frequencies at the four polymor- phic loci are given in Fig. 2 for the two populations studied. Comparison of these distributions indicates some differences, except at locus 7.4. Some alleles occurred in a single popula- tion only, such as alleles 201 and 203 at locus 60.1 in the N’Zana population and allele 185 at the same locus in the Tienfala population. However these particular alleles are pre- sent at a low frequency in each population, Given the relatively small number of individuals examined, we cannot conclude for certain as to the presence of fixed alleles.

The exact test for population differentiation was computed for all loci and among the two populations. The mean p value (p < lo4) indicates that there is a significant difference be- tween the two populations.

As stated in Materials and methods, a.cryptic species might account for 5% of the samples in ,the N’Zana focus. Alleles found only in this population could be attributed to individuals of the cryptic species, and we eliminated 3 samples with alleles 201 and (or) 203 at locus 60.1 from new calculations. Changes in heterozygosity and FIS values were low, at most 1.5%, and a high heterozygote deficiency persisted. The exact test for population differentiation remained highly significant, with the mean p value < lo4.

No significant genotypic linkage disequilibrium between loci was detected in any population (p > 0.05 for each pairwise comparison), suggesting that the loci give independent esti- mates of population genetic parameters.

Discussion Our study suggests that microsatellites may be a useful tool for population genetic studies in the S. darnnosum complex. We

O 1998 NRC Canada

Dumas et al. 159

Fig. 2. Distribution of allele frequencies at four microsatellite loci (60.1,64.2,7.4, and H3-4) for the two S. simzdizrm populations (N’Zana and Tienfala).

T 64.2 T 0.3 t 0.3 4

0.2

0.1

O

7.4 T 0.3 +

H N’&a

O Tienfala

T r l-llll 0.1 O

m - u m m m - 7 -

isolated and characterized five loci. Primer sequence conser- vation was examined across some species of the complex, in- cluding S. squamosum (Sanaga, Cameroon), S. soubrense (Soubre, Ivory Coast), and S. sirbanzim (Zandji, Benin). At all loci, the primer pairs were found to successfully amplify the target sequence across every species tested. These results sug- ,.y gest an interspecific conservation of the flanking regions of microsatellites within the S. damnosum complex, making these markers useful in population genetics, whatever the spe- cies of the complex.

Of the five microsatellite loci used in this study, four were

I

I

I

highly polymorphic in the two populations tested, having 8-12 alleles per locus and gene diversities ranging from 77.9 to 88.2%. Significant heterozygote deficiencies were observed in the two populations. Such deficiencies may be generated by numerous factors, including inbreeding, population structure (the Walhund effect), and null alleles. By redesigning primers at two loci, we demonstrated the existence of null alleles in some samples. Sequence analyses suggested that the failure of PCR amplifications was due to point mutations in the primer regions. This could be explained by the fact that the genomic library was done with S. sqziamosum DNA (a forest species), and that our populations were captured in geographic areas where only savannah flies occur. Moreover, population studies of flies captured in the Sanaga region (Cameroon), where S. squamosum is predominant, indicate no such heterozygote deficiencies when using the original primer pairs (P. Bar- bazan, personal communication). However, even after correct- ing for null alleles, heterozygote deficiencies persisted in the

I !

population. Thus, the deficit in heterozygotes at two loci could not be explained by the presence of null alleles alone. Inbreed- ing could also be the cause of deficiency, but it seems unlikely, since such effects would be expected to be evident for all loci. Population subdivision is a further cause of heterozygote de- ficiency: the two sites studied are within the area of OCP con- trol, where populations are recurrently eradicated through insecticide treatments. If recolonization occurs and is very re- cent, a Walhund effect may ensue. Note that this would pro- duce heterozygote deficiencies of the order observed here only if colonizing populations were markedly differentiated and outcrossed reproduction had not occurred. Another explana- tion could be the presence of two species in our populations. Estrada-Franco et al. (1992) demonstrated that an excess of homozygotes in Anopheles populations was due to mixed spe- cies of the Anopheles quadrimaculatus complex, leading to the description of a new species. The hypothesis of mixed species can be ruled out for the Tienfala site, since only S. sirbanuin was found. At the N’Zana site, S. damnosum S.S. and S. sir- banum were found at the same rate during the dry season, but after the reinvasion process, cytotaxonomical data indicated a large majority of S. sirbanum (95% of samples). The very low percentage of samples belonging to a different species does not significantly affect heterozygote deficiency. Therefore, population subdivision in the two foci remains the most likely explanation.

Microsatellite markers allow a clear distinction to be made between the N’Zana and Tienfala populations. The exact test of population differentiation clearly demonstrates that locus

O 1998 NRC Canada

160

64.2 can discriminate between the two populations. These re- sults are in good agreement with entomological data, which indicate that the Tienfala focus is a permanent one, where reinvasion by migratory flies has never been shown by the OCP. On the other hand, the N’Zana focus is essentially con- stituted by invading flies, probably arising from genetically differentiated foci.

Locus 7.1 appears monoallelic in S. sirbaiiuiv populations; . interestingly, preliminary results show high polymorphism in S. squamosuriz populations from Sanaga (Cameroon). By con- trast, locus H3-4 appears monomorphic for these latter samples (P. Barbazan, personal communication). Although these mo- lecular markers were not developped for species diagnosis, their allelic distribution and polymorphism might then be use- ful for fly identification.

The recent work of Toé et al. (1997) demonstrated that there was no preferential transmission of the two strains of O. volvulus by the different species of S. daiizaosum s.1. in the transition area between the savannah and the forest where an intermediate disease pattern occurs. Since we have developed identical population genetic tools, i.e., microsatellite markers, for O. voZvulus (S. Herder, V. Dumas, J.-P. Aussel, C. Bellec, J.-P. Chippaux, M. Boussinesq, and G. Cuny, in preparati~n)~ and S. damnoslm s.l., they can be used to determine: (i) if some fly populations possess a better vectorial capacity ac- cording to the different parasite strains; and (ìi) if some adap- tative phenomenon allows transmission of savannah-strain parasites by forest-dwelling black fly populations in transition areas.

In conclusion, contributions using microsatellites may shed some light on the complexity of S. damnosuh s.1. and the vector-parasite relationships, the migration behavior of adult flies in relation to their breeding site, and the recolonization process, which can be source of reactivation of foci, and assist in the identification of insecticide-resistant populations. Population genetic studies seem to be a promising approach to understanding vector transmission and the epidemiology and control of onchocerciasis.

Acknowledgements We thank Jean-Marc Hougard, chief of The Vector Control Unit, Onchocerciasis Control Program (OCP), for providing the adult flies from Mali; Yiriba Bissan, Coordinator of Ento- mological Research, OCP, and Laurent Toé, Responsible for Insecticide Labratory, OCP, for data on species identification and foci information; and Michel Raymond and François Rousset for providing the GENEPOP software. We are grateful to Philippe Barbazan, Pierre Guillet, and Philippe Jarne for help- ful discussions and constructive comments on the manuscript.

References Agatsuma, T., Hirai, H., Ochoa, J.O., and Tada, I. 1993. Typing of

Simulium using mitochondrial DNA PCRproducts with restriction endonuclease digestion. Ann. Trop. Med. Parasitol. 87: 307-309.

Boakye, D.A. 1993. A pictoral guide to the chromosomal identifica- tion of members of the SiinuZiurn damizosum Theobald complex in

3 S. Herder, V. Dumas, J.-P. Aussel, C. Bellec, J.-P. Chippaux, M. Boussinesq, and G. Cuny. Genetic variation of Onckocerca volvulus using microsatellite DNA sequences. In preparation.

Genome, Vol. 41, 1998

West Africa with particular reference to the Onchocerciasis Con- trol Programme area. Trop. Med. Parasitol. 44: 223-243.

Brockhouse, C.L., Vajime, C.G., Martin, R., and Tanguay, R.M. 1993. Molecular identification of onchocerciasis vector sibling species in blackflies (Diptera: Simuliidae). Biochem. Biophys. Res. Commun. 194: 628-634.

Brudford, M.W., and Wayne, R.K. 1993. Microsatellites and their application to population genetic studies. Curr. Opin. Genet. Dev. 3: 939-943.

Callen, D.F., Thompson, A.D., Shen, Y., Phillips, H.A., Richards, R.I., Mulley, J.C., and Sutherland, G.R. 1993. Inci- dence and origin of ”null” alleles in the (AC), microsatellite mark- ers. Am. J. Hum. Genet. 52: 922-927.

Duke, B.O.L. 1981. Geographical aspects of onchocerciasis. Ann. Soc. Belge Med. Trop. 61: 179-186.

Duke, B.O.L. 1990, Human onchocerciasis - an overview of the disease. Acta Leiden. 59: 9-24.

Duke, B.O.L., Lewis, D.J., and Moore, P.J. 1966. Oizclzo- cercaSiinulium complexes. I. Transmission of forest and Sudan

. savanna strains of Onchocerca volvulus from Cameroon, by Sinzuliuiiz darnnosum from various West African bioclimatic zones. Ann. Trop. Med. Parasitol. GO: 318-336.

Erttmann, K.D., Unnasch, T.R., Greene, B.M., Albiez, E.J., Boateng, J., Denke, A.M., Ferraroni, J.J., Karam, M., Schulz-Key, H., and Williams, P.N. 1987. A DNA sequence for forest form Oiicho- cerca volvulus. Nature (London), 327: 415-417.

Estoup, A., Solignac, M., Harry, M., and Cornuet, J.-M. 1993. Char- acterization of (GT), and (CT), microsatellites in two insect spe- cies: Apis rnellifera and Boinbus terrestris. Nucleic Acids Res. 6:

Estoup, A., Scholl, A., Pouvreau, A., and Solignac, M. 1995a. Monoandry and polyandry in bumble bees (Hymenoptera; Bom- binae) as evidenced by highly variable microsatellites. Mol. Ecol.

Estoup, A., Solignac, M., Cornuet, J.-M., Goudet, J., and Scholl, A. 1995b. Genetic differentiation of continental and island popula- tions of Bonzbus terrestris (Hymenoptera: Apidae) in Europe.

Estrada-Franco, J.G., Ma, M., Lanzaro, G.C., Gwadz, R., Galvan- Sanchez, C., Cespedes, J.L., Vargas-Sagarnaga, R., and Ro- driguez, R. 1992. Genetic evidence of a species complex in Anopheles pseudopunctipeiznis. Boletin de la Oficina Sanitaria Panamericana, 69: 219-236.

Fisher, R.A. 1970. Statistical methods for research workers. 14th ed. Olivier and Boyd, Edinburgh.

Gertsch, P., Pamilo, P., and Varvio, S.-L. 1995. Microsatellites reveal high genetic diversity within colonies of Cainponotus ants. Mol. Ecol. 4: 257-260.

Hughes, C.R., and Queller, D.C. 1993. Detection of highly polymor- phic microsatellite loci in a species with little allozyme polymor- phism. Mol. Ecol. 2: 131-137.

Jarne, P., and Lagoda, P.J.L. 1996. Microsatellites, from molecules to populations and back. Trends Ecol. Evol. 11: 424-429.

Lanzaro, G.C., Zheng, L., Toure, Y.T., Traore, S.F., Kafatos, F.C., and Vernick, K.D. 1995. Microsatellite DNA and isozyme vari- ability in a West African population of Anopheles gambiae. Insect Mol. Biol. 4: 105-112.

Meredith, S.E.O., and Townson, H. 1981. Enzymes for species iden- tification in the Simulium damnosuin complex from West Africa. Tropenmed. Parasitol. 32: 123-129.

Meredith, S.E.O., Lando, G., Gbakima, A.A., Zimmer”, P.A., and Unnasch, T.R. ,1991. Onchocerca volvulus: application of the po- lymerase chain reaction to identification and strain differentiation of the parasite. Exp. Parasitol. 73: 335-344.

Nei, M. 1987. Molecular evolutionary genetics. Colombia University Press, New York.

Post, R.J., and Flook, P. 1992. DNA probes for identification of mem-

1427-1 43 1.

4: 89-93.

Mol. Ecol. 4: 252,l-13.

O 1998 NRC Canada

Dumas et ai.

bers of the Simulium damnosum complex (Diptera: Simulidae). Med. Vet. Entomol. 6 379-384.

Prost, A., Rougemont, A., and Omar, M.S. 1980. Caractères épidémiologiques, cliniques et biologiques des onchocercoses de savane et de forêt en Afrique Occidentale. Ann. Parasitol. Hum. Comp. 55: 347-355.

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

Sanger, F., Nicklen, S., and Coulson, A.R. 1977. DNA sequencing with chain-terminating inhibitors. Proc. Natl. Acad. Sci. U.S.A.

Sanguinetti, C.J., Dias Neto, E., and Simpson A.J.G. 1994. Rapid silver staining and recovery of PCR products separated on polyacrylamide gels. BioTechniques, 17: 914-921.

Schlötterer, C., and Tautz, D. 1992. Slippage synthesis of simple sequence DNA. Nucleic Acids Res. 20: 21 1-215.

Solano, P., Duvallet, G., Dumas, V., Cuissance, D., and Cuny, G. 1997. Microsatellite markers for genetic population studies in Glossina palpalis (Diptera: Glossinidae). Acta Trap. 65: 175-180.

Stallings, R.L., Ford, A.F., Nelson, D., Tomey, D.C., Hildebrand, C.E., andMoyzis, R.K. 1991. Genomics, 10: 807-815.

Tang, J., Toé, L., Back, C., Zimmerman, P.A., Pruess, K., and Un- nasch, T.R. 1995. The Simulium damnosum species complex: phylogenetic analysis and molecular identification based upon mi- tochondrially encoded gene sequences. Insect Mol. Biol. 4: 79-88.

Thorin, P.A., Paxton, R.J., and Estoup, A. 1995. Unusually high fre- quency of (CT), and (GT), microsatellite loci in yellowjacket

74: 5463-5467.

!

I

161

wasp, Vespula rufa (L.) (Hymenoptera: Vespidae). Insect Mol. Biol. 4: 141-148.

Toi, L., Tang, J., Back, C., Katholi, C.R., andunnasch, T.R. 1997. Vector-parasite transmission complexes for onchocerciasis in West Africa. Lancet, 349: 163-166.

Vajime, C.G., and Dunbar, R.W. 1975. Chromosomal identification of eight species of the subgenus Edwardsellum near and including Siinzilirim (Edwardsellum) damnosum Theobald (Diptera: Simuli- idae), Trop. Med. Parasitol. 25: 95-102.

Walsh, P.S., Metzger, D.A., and Higuchi, R. 1991. Chelex 100 as a medium for simple extraction of DNA for PCR-based typing from forensic material. BioTechniques, 10: 506-5 13.

Weber, J.L. 1990. Informativeness of human (dC-dA),-(dG-dT), polymorphisms. Genomics, 7: 524-530.

Weir, B.S. 1990. Genetic data analysis. Sinauer Associates, Sunder- land, Mass.

Weir, B.S., and Cockerham, C.C. 1984. Estimating F-statistics for the analysis of population structure. Evolution, 38: 1358-1370.

Wilson, M.D., and Post, R.J. 1994. Integratiorf of morphometric, cytogenetic and molecular techniques: a case study of Simiiliirm damnosum. I n The identification and characterization of pest or- ganisms. Edited by D.L. Hawksworth. Commonwealth Agricul- tural Bureaux International, Wallingford, U.K.

Zimmer", P.A., Dadzie, K.Y., De Sole, G., Remme, J., Soumbey AIley, E., and Unnasch, T.R. 1992. Onchocerca volvzdz(s DNA probe classification correlates with epidemiologic pattems of blin- dess. J. Infect Dis. 165: 964-968.

O 1998 NRC Canada

National Research Conseil national Council Canada de recherches Canada .

Aac-clwc ~ Genome Génome

Volume 41, Number 2, April 1998 Volume 41, numéro 2, avril 1998