Genetic Structure and Gene Flow Along an Altitudinal Gradient Among ...

1 downloads 0 Views 312KB Size Report
May 5, 2007 - KEY WORDS genetic structure, microsatellite markers, altitudinal gradient, stable flies, ..... are few migrants, the genetic diversity may stay low.
POPULATION BIOLOGY/GENETICS

Genetic Structure and Gene Flow Along an Altitudinal Gradient Among Two Stomoxyine Species (Diptera: Muscidae) on La Re´union Island JE´RE´MIE GILLES,1,2,3 ISABELLE LITRICO,4 EMMANUEL TILLARD,2

AND

GERARD DUVALLET5

J. Med. Entomol. 44(3): 433Ð439 (2007)

ABSTRACT Seasonal variations of insect population sizes are often dramatic, particularly in temperate regions and at altitudes where the climatic conditions are unfavorable to insect development during the winter. Decline of population size (or bottlenecks) and founder events may reduce the genetic variability and may create genetic differentiation between populations by drift and founder effects, but this reduction of genetic diversity is strongly inßuenced by gene ßow between populations. In this study, we determined the population genetic structure for two stomoxyine species (Diptera: Muscidae), Stomoxys calcitrans (L.) and Stomoxys niger niger Macquart, which co-occur in dairy barns along an altitudinal gradient on La Re´ union island. Using microsatellite markers, we quantiÞed the genetic variation within and among populations for different altitudes. This study displays that, contrary to expectations, genetic diversity is not correlated with altitude and that genetic differentiation is not larger among high-altitude populations than among low-altitude populations. These results attest to the small drift and founder effects in high-altitude populations despite drastic decreases in population size during the winter. Furthermore, at the island scale, the populations of S. calcitrans were slightly differentiated, but those of S. niger niger were not. Together, the results revealed large levels of gene ßow on La Re´ union Island despite the dramatic geographic barriers, and they emphasize the importance of considering agricultural practices to restrict the dispersal of stomoxyines. KEY WORDS genetic structure, microsatellite markers, altitudinal gradient, stable ßies, La Re´ union Island

Colonization and extinction cycles (Barrett and Husband 1989, Whitlock and McCauley 1990, Husband and Barrett 1996) and variations in population density (Brown 1994, Lynch et al. 1995) strongly inßuence genetic variability in space and time. Many theoretical studies (Slatkin 1977, 1993; Wade and McCauley 1988; Withlock and McCauley 1990; Austerlitz et al. 1997, 2000; Le Corre and Kremer 1998) have shown the genetic impact of population founder after extinction or reduction of population size (bottleneck). Under drift and founder effects, genetic diversity may decrease and the differentiation between populations 1 Institut fu ¨ r Vergleichende Tropenmedizin und ParasitologieLeopoldstra␤e 5, D-80802 Mu¨ nchen, Germany. 2 CIRAD-EMVT, Programme Productions Animales, Pole de Protection des Plantes, 7 chemin de l⬘IRAT, 97410 St Pierre de La Re´ union, France. 3 Corresponding author: Current address: Institut fu ¨ r Vergleichende Tropenmedizin und Parasitologie-Department of Comparative Tropical Medicine and Parasitology, Leopoldstra§e 5, D-80802 Mu¨ nchen, Germany (e-mail: [email protected]). 4 INRA, Unite ´ de Ge´ ne´ tique et dÕAme´ lioration des Plantes Fourrage` res, 86600 Lusignan, France. 5 De ´ partement Ecologie des ArthropodesÐUMR 5175 CEFE, Universite´ Paul-Vale´ ry, Route de Mende, 34199 Montpellier cedex 5, France.

may be high. When populations are founded, genetic diversity and differentiation are inßuenced by the composition of founders (Wade and McCauley 1988, Whitlock and McCauley 1990, McCauley 1991). If populations are founded by a few individuals, diversity may be low and differentiation high. But it is rare for populations to be completely isolated, and the gene ßow homogenize allelic frequencies (Reichow and Smith 2001) when we consider neutral markers. So, when bottlenecks or extinctions reoccur, a driftÐmigration equilibrium may be reached, each inßuenced by the ecology of the species and by human activities (Colson 2002). For insects, seasonal variations of population size are often dramatic (Gilles 2005, Goulson et al. 2005, Henning et al. 2005), in particular, during the winter in temperate regions or at higher altitudes where climatic conditions are unfavorable to insect development. The seasonal decline of insect population size and founder events may reduce the genetic variability and may create genetic differentiation in space and time. On La Re´union Island, two Stomoxys species (Diptera: Muscidae), the cosmopolitan Stomoxys calcitrans (L.) and the tropical Stomoxys niger niger Macquart, are

0022-2585/07/0433Ð0439$04.00/0 䉷 2007 Entomological Society of America

PPA BJP GIG GRD PPV DRV PJR LAS GPA 3BC 3BT PAL PPL PPE 1 2 3 4 5 6 7 8 9 10 11 12 13 14

*, P ⬍ 0.05; **, P ⬍ 0.01; and ***, P ⬍ 0.001.

Na

5.50 6.00 4.83 5.16 5.33 5.16 5.66 5.33 5.66 5.66 5.33 4.83 4.50 5.33 0.00 0.09* 0.12** 0.00 0.11* 0.20*** 0.08* 0.14** 0.00 0.12* 0.00 0.18** 0.00 0.16*

FIS Ho

0.6528 (0.1702) 0.6447 (0.1498) 0.5639 (0.2605) 0.6534 (0.1909) 0.5830 (0.2029) 0.5488 (0.1670) 0.6000 (0.1520) 0.5494 (0.1900) 0.6850 (0.1242) 0.5907 (0.1574) 0.6619 (0.1055) 0.5495 (0.1400) 0.6139 (0.1386) 0.5403 (0.1128) 0.6802 (0.1058) 0.6971 (0.1041) 0.6268 (0.2052) 0.6574 (0.1155) 0.6468 (0.1303) 0.6730 (0.0931) 0.6434 (0.1380) 0.6284 (0.1183) 0.7027 (0.0.0975) 0.6567 (0.1226) 0.6463 (0.1166) 0.6453 (0.1385) 0.5643 (0.1361) 0.6213 (0.1260) 30 30 30 30 28 30 30 30 30 30 30 24 22 30 6.56 6.33 5.44 6.00 5.67 6.00 5.78 6.00 6.22 5.22 6.00 6.00 5.33 0.08** 0.10*** 0.07* 0.06* 0.00 0.08** 0.20*** 0.00 0.00 0.12** 0.08** 0.00 0.10** 0.5815 (0.2129) 0.5807 (0.1886) 0.5659 (0.2099) 0.584 (0.188) 0.5963 (0.1792) 0.5803 (0.151) 0.497 (0.2065) 0.6259 (0.208) 0.6481 (0.2495) 0.5471 (0.2014) 0.5934 (0.1623) 0.6384 (0.1657) 0.5578 (0.2321) 0.6207 (0.1867) 0.6375 (0.1660) 0.5975 (0.1986) 0.6153 (0.1739) 0.6229 (0.1764) 0.6236 (0.1729) 0.6116 (0.1975) 0.6303 (0.2124) 0.6419 (0.2027) 0.6043 (0.1821) 0.64 (0.1515) 0.6481 (0.1581) 0.6097 (0.1811) 30 30 30 30 30 30 30 30 30 22 30 30 27 100 900 900 1,200 1,200 1,600 1,600 80 900 1,300 1,300 300 1,100 1,100 21⬚ 40⬘ S, 55⬚ 49⬘ E 21⬚ 24⬘ S, 55⬚ 55⬘ E 21⬚ 24⬘ S, 55⬚ 55⬘ E 21⬚ 21⬘ S, 55⬚ 56⬘ E 21⬚ 22⬘ S, 55⬚ 55⬘ E 21⬚ 19⬘ S, 55⬚ 57⬘ E 21⬚ 20⬘ S, 55⬚ 59⬘ E 21⬚ 20⬘ S, 55⬚ 35⬘ E 21⬚ 16⬘ S, 55⬚ 36⬘ E 21⬚ 03⬘ S, 55⬚ 21⬘ E 21⬚ 05⬘ S, 55⬚ 21⬘ E 20⬚ 55⬘ S, 55⬚ 31⬘ E 21⬚ 06⬘ S, 55⬚ 39⬘ E 21⬚ 05⬘ S, 55⬚ 41⬘ E

S. niger niger

HE n Na FIS Ho

S. calcitrans

HE n Altitude (m) Coordinates Region Pop code

Study Sites and Sample Collection. La Re´ union is a volcanic island (2,507 km2) ⬇800 km east of Madagascar (21⬚ 20⬘ S, 55⬚ 15⬘ E). The climate is humid tropical, with heavy rains in summer. At the end of summer (April 2003), 379 S. calcitrans adults were allocated into 13 populations and 404 S. niger niger adults were allocated into 14 populations (Table 1; Fig. 1) after capture in Vavoua traps (Laveissie` re and Gre´ baut 1990) in dairy barns. Seven populations (population 1Ð7; Fig. 1) were located along an altitudinal gradient (100 Ð1,600-m elevation), and the abundance of the two species was recorded weekly during a 90-wk period with Vavoua traps. Trapped ßies were identiÞed in the laboratory according to Zumpt (1973). DNA Extraction and Microsatellite Protocol. The individuals were crushed with tungsten microbals, and their DNA was extracted following the DNeasy tissue kit (QIAGEN, Hilden, Germany) protocol. The S. calcitrans were genotyped with nine microsatellite loci: seven loci (ScA7, ScD10, ScD7, ScA6, ScA3, ScA11, and ScF7) identiÞed by Gilles et al. (2004) on this species and two (SnB10 and SnF1) identiÞed on S. niger niger (Gilles et al. 2005). The S. niger niger were genotyped using six microsatellite loci: four loci (SnE8, SnF2, SnB10, and SnF1) identiÞed by Gilles et al. (2005) on this species and two (ScA7 and ScD10) identiÞed on S. calcitrans (Gilles et al. 2004). The other loci identiÞed for S. calcitrans and S. niger niger (Gilles et al. 2004, 2005) were excluded from the analysis,

Pop no.

Materials and Methods

Values (mean ⴞ SE) of Na, HE, Ho, and FIS in populations of S. calcitrans and S. niger niger

found in dairy barns. These blood-sucking insects associated with cattle (Zumpt 1973) have become pests and cause important economic losses. Their bites are painful and they are potential mechanical vectors of pathogens (Gilles et al. 2007). Gilles (2005) observed large 1) spatial variation along an altitudinal gradient and 2) seasonal variation in the abundance of these two species in the south of La Re´ union island. These variations were related to climatic factors, mainly temperature (positive relationship) and relative humidity (negative relationship). In unfavorable season (winter), Stomoxys populations decrease; the higher the altitude, the higher the decrease. When environmental conditions (e.g., temperature and humidity) become favorable, at the beginning of summer, Stomoxys populations increase. Our aim is to better understand the impact of population size ßuctuations on the genetic structure of two species of Stomoxys, S. calcitrans and S. niger niger, on La Re´ union Island and to determine their colonization dynamics. We addressed the following questions: 1) Is the diversity of high-altitude populations, which show greater size reduction, lower than that of low-altitude populations? 2) Is the differentiation among high-altitude populations higher than among low-altitude populations? and 3) How are Stomoxys populations differentiated at island scale? Answers to each of these questions may help with the development of control strategies on La Re´ union Island.

Vol. 44, no. 3

R1 R1 R1 R1 R1 R1 R1 R2 R2 R3 R3 R4 R5 R5

JOURNAL OF MEDICAL ENTOMOLOGY

Table 1.

434

May 2007

GILLES ET AL.: GENETIC STRUCTURE AND GENE FLOW

435

Fig. 1. (a) Location of La Re´ union Island. (b) Sampled breeders (ßy populations) for genotyping individuals of S. calcitrans and S. niger niger. The Þve regions (r1, r2, r3, r4, and r5) used in grouped genetic analyses are delimited by plain font black lines.

because of the probable occurrence of null alleles. Primer sequences, ampliÞcation conditions, and allele scoring were as reported previously (Gilles et al. 2004, 2005). Polymerase chain reaction (PCR) products were analyzed on an ABI 3100 automated sequencer (Applied Biosystems, Foster City, CA), and subsequent analysis was carried out using GeneScan Analysis 3.7 (Applied Biosystems). Fourteen individuals were replicated to check repeatability. Genetic Diversity within Populations. For each species, we calculated the mean number of alleles per locus (Na), NeiÕs unbiased expected heterozygosity (HE) (Nei 1978), observed heterozygosity (Ho), and WrightÕs inbreeding coefÞcient (FIS) according to Weir and Cockerham (1984) by using Genetix version 4.04 (Belkhir et al. 2001). The test for departure from HardyÐWeinberg equilibrium was conducted using 1,000 permutations in each population by Genetix version 4.04 (Belkhir et al. 2001). For the seven populations located along the altitudinal gradient (1Ð7), Spearman rank correlation coefÞcients (SAS Institute 2001) were calculated to test for a relationship between HE and population size at the sampling time (MarchÐApril 2003). Genetic Differentiation among Populations. To consider the differentiation between populations, the FST were calculated according to the Weir and Cockerham (1984) procedure and tested using 1,000 permutations of individuals among populations or regions

(Genetix version 4.04; Belkhir et al. 2001) (Fig. 1; Table 2). Regions were deÞned according to distance between populations and to geographical barriers (e.g., mountains and valleys), which may be important obstacles to ßy migrations. To test the isolation by distance, correlations between genetic distances (measured as FST) and spatial distances between pairs of populations were calculated and tested following the Mantel permutation procedure (Genetix version Table 2. region

Regions defined on Fig. 1 and sample size (n) for each

Region

n of S. calcitrans

n of S. niger niger

R1

210

208

R2

60

60

R3

52

60

R4 R5

30 27

24 52

Pop/region 1 2 3 4 5 6 7 8 9 10 11 12 13 14 (only for S. niger niger)

Pop altitude (m) 100 900 900 1,200 1,200 1,600 1,600 80 900 1,300 1,300 300 1,100 1,100

436

JOURNAL OF MEDICAL ENTOMOLOGY

Table 3. Organization by altitude and altitude code of the seven breeders for the altitudinal transect, and sample size (n) for each altitude Altitude (code)

Size of S. calcitrans pop

Size of S. niger niger pop

A1 A2

30 60

30 60

A3

60

58

A4

60

60

Pop

Pop altitude (m)

1 2 3 4 5 6 7

100 900 900 1,200 1,200 1,600 1,600

4.04; Belkhir et al. 2001). In addition, the genetic differentiation was tested by grouping 1) populations per region (Fig. 1; Table 2) and 2) populations (1Ð7) per altitude (Table 3). Results Genetic Diversity within Populations. The mean number of alleles per locus was 5.88 and 5.30 for S. calcitrans and S. niger niger, respectively. The mean number of alleles per locus and per population varied from 5.22 to 6.55 for S. calcitrans and from 4.50 to 6.00 for S. niger niger (Table 1). For NeiÕs HE, values ranged from 0.60 to 0.65 for S. calcitrans and from 0.56 to 0.70 for S. niger niger (Table 1). For Ho, values ranged from 0.50 to 0.65 for S. calcitrans and from 0.54 to 0.68 for S. niger niger (Table 1). For both species, no correlation

Vol. 44, no. 3

was observed between NeiÕs HE values and altitude (P ⬎ 0.5) and between NeiÕs HE values and the population size (P ⬎ 0.5). However, for S. calcitrans, the mean number of alleles per locus was signiÞcantly correlated with the size of population in the sampled period (Rs ⫽ 0.54; P ⬍ 0.05), but there was no correlation (P ⬎ 0.1) for S. niger niger. Regarding FIS, the global value for S. calcitrans was 0.078 (P ⬍ 0.001), and the FIS values per population were relatively low: from 0 to 0.12 (Table 1) except for the population 7 (FIS ⫽ 0.20; P ⬍ 0.001). In the same way, the global FIS value for S. niger niger was 0.094 (P ⬍ 0.001), and the FIS per population ranged from 0 to 0.20 (Table 1). FIS values varied between loci (Fig. 2), probably because the null alleles are present for some loci. Thus, this variation between loci does not support the notion that inbreeding explains the Fis values in our study. Genetic Differentiation among Populations. On the island scale, the FST value for S. calcitrans was 0.02 (P ⬍ 0.001), whereas that for S. niger niger was not significantly different (P ⬎ 0.1). Several pairwise differences among populations occurred, however, as population 13 of S. niger niger differed from all others (Table 4). Furthermore, for both species, when we grouped populations by region (Fig. 1; Table 2) or by altitude for the seven populations located along the transect (Table 3), the differences were low but signiÞcant: FST ⫽ 0.01, P ⬍ 0.001 and FST ⫽ 0.023. P ⬍ 0.001 for S. calcitrans and FST ⫽ 0, P ⬎ 0.1 and FST ⫽ 0, P ⬎ 0.5 for S. niger niger. But when we grouped populations by region or altitude, genetic differences

Fig. 2. Mean FIS value per locus for S. calcitrans (a) and S. niger niger (b).

May 2007 Table 4.

GILLES ET AL.: GENETIC STRUCTURE AND GENE FLOW

437

Pairwise genetic differentiation between populations (FST) of S. calcitrans (top matrix) and S. niger niger (bottom matrix)

Pop

1

2

3

4

5

6

7

8

9

10

11

12

13

1 2 3 4 5 6 7 8 9 10 11 12 13 14

Ñ ⫺0.0004 ⫺0.0041 ⫺0.0004 0.0006 ⫺0.0060 0.0053 ⫺0.0017 0.0025 ⫺0.0062 ⫺0.0058 0.0032 0.0360 0.0127

0.0004 Ñ 0.0076 0.0091 ⫺0.0037 0.0032 ⫺0.0011 ⫺0.0014 0.0001 ⫺0.0069 ⫺0.0032 0.0010 0.0405 0.0183

0.0400 0.0393 Ñ ⫺0.0009 0.0048 0.0041 0.0073 0.0004 0.0110 ⫺0.0111 ⫺0.0009 ⫺0.0057 0.0440 0.0118

0.0147 ⫺0.0005 0.0473 Ñ ⫺0.0012 0.0006 0.0012 0.0017 0.0005 ⫺0.0073 ⫺0.0110 ⫺0.0097 0.0270 0.0058

0.0294 0.0230 0.0157 0.0230 Ñ 0.0016 ⫺0.0018 ⫺0.0052 0.0048 ⫺0.0111 ⫺0.0057 0.0066 0.0391 0.0049

0.0011 0.0034 0.0268 0.0231 0.0258 Ñ 0.0049 ⫺0.0043 ⫺0.0037 ⫺0.0052 ⫺0.0012 0.0016 0.0081 0.0050

0.0163 ⫺0.0001 0.0634 0.0025 0.0429 0.0193 Ñ 0.0020 ⫺0.0007 ⫺0.0083 ⫺0.0104 ⫺0.0007 0.0270 0.0074

0.0196 0.0072 0.0233 0.0052 0.0072 0.0246 0.0134 Ñ 0.0043 ⫺0.0086 ⫺0.0074 ⫺0.0014 0.0298 0.0138

0.0221 0.0116 0.0107 0.0144 0.0033 0.0132 0.0210 ⫺0.0011 Ñ ⫺0.0020 0.0000 ⫺0.0121 0.0106 0.0100

0.0040 ⫺0.0636 0.0451 0.0071 0.0253 0.0046 ⫺0.0022 0.0030 0.0043 Ñ ⫺0.0132 ⫺0.0069 0.0249 0.0005

0.0052 ⫺0.0044 0.0291 0.0064 0.0192 ⫺0.0039 0.0125 0.0140 0.0131 0.0047 Ñ ⫺0.0079 0.0202 0.0014

0.0299 0.0120 0.0354 0.0073 0.0145 0.0323 0.0167 ⫺0.0028 0.0039 0.0083 0.0217 Ñ 0.0019 ⫺0.0058

0.0396 0.0457 0.0082 0.0543 0.0088 0.0338 0.0684 0.0307 0.0174 0.0447 0.0366 0.0387 Ñ ⫺0.0148

SigniÞcance values for pairwise test of population differentiation. Plain font, P ⬍ 0.05.

were not correlated with geographic distances (Mantel permutation procedure not signiÞcant). For populations on the altitudinal transect (1Ð7), high-altitude populations (⬎1,200 m) were not more differentiated than low-altitude populations (S. calcitrans: FST ⫽ 0.02, P ⬍ 0.001 versus FST ⫽ 0.02, P ⬍ 0. 001 and S. niger niger: FST ⫽ 0, P ⬎ 0.1 versus FST ⫽ 0, P ⬎ 0.1). Discussion Key results are, Þrst, that genetic diversity was not correlated with the altitude of populations and that genetic differentiation was not greater for high-altitude populations than low-altitude populations. Despite the large reduction of size during the winter in high-altitude populations, our results attest to low drift and founder effects. Second, unlike S. calcitrans whose populations are slightly differentiated, we observed no genetic differentiation between the populations of S. niger niger at the island scale and no distance isolation for both species. The latter result indicates that gene ßow is important at the small and large scales. Variable Degrees of Genetic Differentiation between Species and Populations. At the island scale, genetic differentiation among populations varied according to species. Indeed, there was no genetic differentiation for S. niger niger, but populations of S. calcitrans were slightly differentiated. Why do these close species differ? The high ßight capacity of these ßies may explain the important gene ßow at the island scale, but passive transport might play an important part in the gene ßow on La Re´ union Island. Cultural practices and manure or sugarcane (Saccharum L.) transports between different parts of the island may allow large gene ßow despite dramatic geographic barriers. Some eggs, larvae, and pupae of Stomoxys are found in sugarcane leaves and in manures, these substrates constituting the laying site (personal observation; Kunz and Monty 1976, Barre´ 1981). The transport of these substrates seems to strongly contribute to the passive transport of the Stomoxys spp. Moreover, whereas manures are laying sites for S. calcitrans, sugarcane leaves constitute the preferential laying sites for S. niger niger (Kunz and Monty 1976, Barre´ 1981),

and sugarcane leaves are widely transported in the island, particularly during the harvest. Thus, the lack of genetic structure of S. niger niger populations could be the result of massive transport of sugarcane residues at the island scale. At a lower scale, regarding differentiation along the altitudinal gradient, S. calcitrans and S. niger niger populations showed no isolation by distance, but some populations, populations 3 and 13, with different farming methods are more differentiated than the others (apart from the geographic location). The differentiation of these two populations may result from drift effect in relation to small numbers. This assumption is supported by diversity values among S. calcitrans in these two populations and S. niger niger in population 13, because the mean number of alleles per locus and heterozygosity values were low. Harper et al. (2003) showed that among Lepidoptera the genetic diversity is all the higher as the population size is high. The diversity and differentiation of these populations may depend on intrinsic (or local) factors of the breeders, factors that may inßuence the population size. Gilles (2005) suggested that the quantity of larval resources could limit the density of stable ßy species. These larval resources are directly associated with agricultural practices. No Drift Effect at High Altitude Despite the Reduction of Population Size. Although many studies (Wright 1931, Whitlock and McCauley 1990, Brown 1994, Husband and Barrett 1996, Brookes et al. 1997) showed population size ßuctuations govern genetic diversity, our results display no impact of recurrent reduction of population size on the diversity. Indeed, the genetic diversity and differentiation values were not correlated with the altitude of populations, although high-altitude populations have extreme reductions of size compared with low-altitude populations. These results concur with two possible scenarios of population dynamics for S. calcitrans and S. niger niger. 1) Variations in relative abundance of the two stomoxyine species during winter did not result from a real decline of the population size. This situation has been reported for other species. For example, in Australia, Chapman et al. (1999) showed that the popu-

438

JOURNAL OF MEDICAL ENTOMOLOGY

lations of Aedes vigilax (Skuse) almost disappear when environmental conditions are unfavorable, but stocks of immature forms (eggs) persist several months and allow the population to keep its initial size and then to limit the drift effect. 2) The population bottleneck may really exist but, at the end of winter, new populations may be created from a lot of individuals from other populations at lower altitudes (source populations). If there are many migrants, the probability for postbottleneck populations to reach the genetic diversity of source populations may be high, and the population differentiation will be small. But, if there are few migrants, the genetic diversity may stay low and the population differentiation high (Slatkin 1977). In addition, Slatkin (1995) showed that a few migrant individuals of second generation could be enough to prevent drift effects. So, although these two scenarios of population colonization are possible and are not incompatible, it is unlikely that the functioning of high-altitude populations is independent on La Re´ union Island. Indeed, regarding the degree of differentiation, the populations of S. calcitrans differed slightly on the island scale and those of S. niger niger did not differ. In addition, genetic differentiation was independent of geographic distance. These results point out large gene ßow between populations, yet this is not surprising given the high ßight capacity of stable ßies. Some studies (Bailey et al. 1979, Hogsette et al. 1987) showed that S. calcitrans is able to ßy far (several hundred kilometers), and Chevillon et al. (1995) showed no isolation by distance at the scale of Sardinia among Culex pipiens L. with individuals migrating long distances and thus limiting the genetic differentiation. In our study, it is difÞcult to determine the part of migrant individuals and the part of immature stages wintering in manure or in vegetal substrates in decomposition, yet we think immature stocks were probably very large (personal observations). Thus, it would be of interest 1) to perform a captureÐmarkÐrelease protocol to know effective migrations of the two Stomoxys species and 2) to determine the genetic structure at different periods of the year. In conclusion, we emphasize the importance of considering all stages of the life cycle of insects to understand the impact of periodic variations of the population size on the genetic structure. And particular focus should be placed on immature stages of these two stomoxyine species, which are able to develop and then spend the winter “protected” in speciÞc substrates. Genetic analyses should be carried out at different periods of the year to establish the population dynamics of stomoxyine species, in particular the role of migrants and immature stocks in the colonization by high-altitude populations on La Re´ union. In addition, agricultural practices seem to be an important factor to take into account to understand population dynamics of S. calcitrans and S. niger niger in farms of the Re´ union Island. This calls for better knowledge of the breeding site. Otherwise, as we observed only small or null genetic differentiation between populations at the island

Vol. 44, no. 3

scale, our results may have implications in the control of the two Stomoxys species. Indeed, large gene ßow between different parts of the island may in part be explained by the transport of larval substrates such as manure or sugarcane leaves and not only by the dispersal of adults. These Þndings should allow us to orientate ßy control programs. Acknowledgments This work was supported by funds from CIRAD-Poˆ le de Protection des Plantes, Conseil Re´ gional from La Re´ union Island and UMR 5175 CEFE. We thank the Service des Marqueurs Ge´ ne´ tiques at Centre dÕEcologie Fonctionnelle et Evolutive-Centre National de la Recherche ScientiÞque (Montpellier), particularly Chantal Debain and the BM team at CIRAD-3P (Emmanuelle Chapier and Emmanuel Jouen).

References Cited Austerlitz, F., B. Jung-Muller, B. Godelle, and P. H. Gouyon. 1997. Evolution of coalescence times, genetic diversity and structure during colonization. Theor. Popul. Biol. 51: 148 Ð164. Austerlitz, F., S. Mariette, N. Machon, P. H. Gouyon, and B. Godelle. 2000. Effects of colonization processes on genetic diversity: difference between annual plants and tree species. Genetics 154: 1309 Ð1321. Bailey, D. L., T. L. Whitfield, and B. J. Smittle. 1979. Flight and dispersal of the stable ßy. J. Econ. Entomol. 66: 410 Ð 411. Barre´, N. 1981. Les stomoxes ou mouches boeuf a` La Re´ union. Pouvoir pathoge` ne, e´ cologie, moyen de lutte. Maison Alfort (France)ÑGroupement dÕEtudes et de Recherches pour le Developpement de lÕAgronomie Tropicale-IEMVT. Barrett, S.C.H., and B. C. Husband. 1989. The genetics of plant migration and colonization, pp. 254Ð277. In A.H.D. Brown, M. T. Clegg, A. L. Kahler, and B. S. Weir [eds.], Plant population genetics, breeding, and genetic resources. Sinauer, Sunderland, MA. Belkhir, K., P. Borsa, L. Chikhi, N. Raufaste, and F. Bonhomme. 2001. Genetix 4.02, logiciel sous Windows pour la ge´ ne´ tique des populations. Laboratoire Ge´ nome, Populations, Interactions, Centre National de la Recherche ScientiÞque Unite´ Mixte de Recherche 5000, Universite´ de Montpellier II, Montpellier, France. Brookes, M. I., Y. A. Graneau, P. King, O. C. Rose, C. D. Thomas, and J.L.B. Mallet. 1997. Genetic analysis of founder bottlenecks in the rare british butterßy Plebejus argus. Conserv. Biol. 11: 648 Ð 661. Brown, D. S. 1994. Freshwater snails of Africa and their medical importance. Taylor & Francis, London, United Kingdom. Chapman, H. F., J. M. Hughes, C. Jennings, B. H. Kay, and S. A. Ritchie. 1999. Population structure and dispersal of the salt marsh mosquito Aedes vigilax in Queensland Australia. Med. Vet. Entomol. 13: 423Ð 430. Chevillon, C., G. Addis, R. Raymond, and A. Marchi. 1995. Population structure in Mediterranean islands and risk of genetic invasion in Culex pipiens L. (Diptera: Culicidae). Biol. J. Linn. Soc. 55: 329 Ð343. Colson, I. 2002. Selection and gene ßow between microenvironment: the case of Drosophila at lower Natal Oren, Mount Carmel, Israel. Mol. Ecol. 11: 1311Ð1316. Gilles, J., I. Litrico, P. Sourrouille, and G. Duvallet. 2004. Microsatellite DNA markers for the stable ßy: Stomoxys

May 2007

GILLES ET AL.: GENETIC STRUCTURE AND GENE FLOW

calcitrans (Diptera: Muscidae). Mol. Ecol. Notes 4: 635Ð 637. Gilles, J. 2005. Dynamique et ge´ ne´ tiques des populations d⬘insectes vecteurs. Les stomoxes, Stomoxys calcitrans et Stomoxys niger niger dans les e´ levages bovins re´ unionnais. Ph.D. dissertation, Universite´ de La Re´ union. Gilles, J., I. Litrico, and G. Duvallet. 2005. Microsatellite loci in the stable ßy, Stomoxys niger niger (Diptera: Muscidae) on La Re´ union Island. Mol. Ecol. Notes. 5: 93Ð95. Gilles, J., J.-F. David, G. Duvallet, S. de La Rocque, and E. Tillard. 2007. . EfÞciency of traps for Stomoxys calcitrans and Stomoxys niger on Reunion Island. Med. Vet. Entomol. (in press). Goulson, D., L. Derwent, M. E. Hanley, D. W. Dunn, and S. R. Abolins. 2005. Predicting calyptrate ßy populations from the weather, and probable consequences of climate change. J. Appl. Ecol. 42: 795Ð 804. Harper, G. L., N. Maclean, and D. Goulson. 2003. Microsatellite markers to assess the inßuence of population size, isolation and demographic change on the genetic structure of the UK butterßy Polyommatus bellargus. Mol. Ecol. 12: 3349 Ð3357. Henning, J., F.-R. Schnitzler, D. U. Pfeiffer, and P. Davies. 2005. Inßuence of weather conditions on ßy abundance and its implications for transmission of rabbit haemorrhagic disease virus in the North Island of New Zealand. Med. Vet. Entomol. 19: 251Ð262. Hogsette, J. A., J. P. Ruff, and C. J. Jones. 1987. Stable ßy biology and control in the northwest Florida. J. Agric. Entomol. 4: 1Ð11. Husband, B. C., and S.C.H. Barrett. 1996. A metapopulation perspective in plant population biology. J. Ecol. 84: 461Ð 469. Kunz, S. E., and J. Monty. 1976. Biology and ecology of Stomoxys nigra Macquart and Stomoxys calcitrans (L.) (Diptera, Muscidae) in Mauritius. Bull. Entomol. Res. 66: 745Ð755. Laveissie`re, C., and P. Gre´baut. 1990. Recherche sur les pie` ges a` glossines (Diptera: Glossinidae). Mise au point dÕun mode` le e´ conomique: le pie` ge “Vavoua.” Trop. Med. Parasitol. 41: 185Ð192.

439

Le Corre, V., and A. Kremer. 1998. Cumulative effects of founding events during colonization on genetic diversity and differentiation in an island stepping stone model. J. Evol. Biol. 11: 495Ð512. Lynch, M., J. Conery, and R. Burger. 1995. Mutation accumulation and the extinction of small populations. Am. Nat. 146: 489 Ð518. McCauley, D. E. 1991. Genetic consequences of local population extinction and recolonisation. Trends Ecol. Evol. 6: 5Ð 8. Nei, M. 1978. Estimation F-statistics for the analysis of population structure. Evolution 38: 1358 Ð1370. Reichow, D., and J. M. Smith. 2001. Microsatellites reveal high levels of gene ßow among populations of the California squid Loligo opalescens. Mol. Ecol. 10: 1101Ð1109. SAS Institute. 2001. SAS/STAT software release 8.2. SAS Institute, Cary, NC. Slatkin, M. 1977. Gene ßow and genetic drift in a species subject to frequent local extinction. Theor. Popul. Biol. 12: 253Ð262. Slatkin, M. 1993. Isolation by distance in equilibrium and nonequilibrium populations. Evolution 47: 264 Ð279. Slatkin, M. 1995. A measure of population subdivision based on microsatellite allele frequencies. Genetics 139: 457Ð 462. Wade, M. J., and D. E. McCauley. 1988. Extinction and recolonization: their effects on the genetic differentiation of local populations. Evolution 42: 995Ð1005. Weir, B. S., and C. C. Cockerham. 1984. Estimating F-statistics for the analysis of population structure. Evolution 38: 1358 Ð1370. Whitlock, M. C., and D. E. McCauley. 1990. Same population genetic consequences of colony formation and extinction: genetic correlation within founder groups. Evolution 44: 1717Ð1724. Wright, S. 1931. Evolution in Medelian populations. Genetics 10: 97Ð159. Zumpt, F. 1973. The stomoxyine biting ßies of the world. Taxonomy, biology, economic importance and control measures. Gustav Fischer Verlag, Stuttgart, Germany. Received 28 September 2006; accepted 6 February 2007.