ARTICLE

Population structure and genetic diversity in the invasivefreshwater snail Galba schirazensis (Lymnaeidae)M. Lounnas, A.C. Correa, P. Alda, P. David, M.-P. Dubois, M. Calvopiña, Y. Caron, M. Celi-Erazo,B.T. Dung, P. Jarne, E.S. Loker, O. Noya, R. Rodríguez-Hidalgo, C. Toty, N. Uribe, J.-P. Pointier,and S. Hurtrez-Boussès

Abstract: We studied the population genetic structure of the freshwater snail Galba schirazensis (Küster, 1862), a potential vectorof infectious diseases such as fascioliasis. Galba schirazensis has now a worldwide distribution but a poorly known origin becausethis species has been distinguished only recently from the morphologically similar and cosmopolitan Galba truncatula (O.F. Müller,1774). We developed specific microsatellite markers and sequenced a mitochondrial gene (cytochrome oxidase subunit I (CO1)) to studyindividuals of G. schirazensis from the Old World and the New World. We found very low genetic diversity within populations, noheterozygotes, and marked population structure — a pattern observed in other highly selfing lymnaeid species with recently enlargeddistributions as a result of biological invasions. The total lack of observed heterozygosity in the few populations of G. schirazensis thatdisplayed some allelic diversity suggests high selfing rates. We also found that the center of diversity, and by extension the origin areaof this species, should be found in the New World, whereas Old World populations should rather result from a recent introductionof a genetically uniform population. The microsatellite markers developed here will help to clarify the history of expansion ofG. schirazensis and might help to understand its role as a potential vector of infectious diseases.

Key words: lymnaeids, Lymnaea, Galba schirazensis, Galba truncatula, vector, microsatellites, selfing.

Résumé : Nous avons étudié la structure génétique des populations du gastéropode d’eau douce Galba schirazensis (Küster, 1862),un vecteur potentiel de maladies infectieuses telles que la fasciolose. Galba schirazensis présente aujourd’hui une répartitionplanétaire, mais une origine méconnue puisque cette espèce n’a que récemment été distinguée de Galba truncatula (O.F. Müller,1774), une espèce cosmopolite semblable sur le plan morphologique. Nous avons mis au point des marqueurs microsatellitesspécifiques et séquencé un gène mitochondrial (la sous-unité I de la cytochrome oxidase (CO1)) pour étudier des individus deG. schirazensis de l’Ancien Monde et du Nouveau Monde. Nous n’avons noté qu’une très faible diversité génétique au sein despopulations, aucun hétérozygote et une structure marquée des populations, des caractéristiques observées chez d’autres espècesde lymnées très autofécondes dont les aires de répartition ont beaucoup cru récemment dans la foulée d’invasions biologiques.L’absence totale d’hétérozygotie observée dans les quelques populations de G. schirazensis qui présentaient un certain degré dediversité allélique indiquerait des taux d’autofécondation élevés. Nous avons également observé que le centre de diversité et, parextension, la région d’origine de cette espèce devraient se trouver dans le Nouveau Monde, alors que les populations de l’AncienMonde seraient plutôt le résultat de l’introduction récente d’une population uniforme sur le plan génétique. Les marqueurs

Received 15 December 2016. Accepted 26 June 2017.

M. Lounnas, A.C. Correa, and C. Toty. MIVEGEC, UMR UM – CNRS 5290 – IRD 224 Maladies Infectieuses et Vecteurs : Ecologie, Génétique, Evolution etContrôle, Centre IRD, BP 64501, 34394 Montpellier CEDEX 5, France.P. Alda. MIVEGEC, UMR UM – CNRS 5290 – IRD 224 Maladies Infectieuses et Vecteurs : Ecologie, Génétique, Evolution et Contrôle, Centre IRD,BP 64501, 34394 Montpellier CEDEX 5, France; Laboratorio de Zoologı́a de Invertebrados I, Departamento de Biologı́a, Bioquı́mica y Farmacia, UniversidadNacional del Sur, San Juan No. 670, B8000ICN, Bahı́a Blanca, Buenos Aires, Argentina.P. David, M.-P. Dubois, and P. Jarne. Centre d’Ecologie Fonctionnelle et d’Evolution, UMR 5175, CNRS – Université de Montpellier – Université PaulValéry Montpellier – EPHE, 1919 route de Mende, 34293 Montpellier CEDEX 5, France.M. Calvopiña. Carrera de Medicina, Facultad de Ciencias Médicas, Universidad Central del Ecuador, Quito, Ecuador.Y. Caron and B.T. Dung. Research Unit in Parasitology and Parasitic Diseases, Fundamental and Applied Research for Animals and Health (FARAH),Faculty of Veterinary Medicine, University of Liège, Belgium.M. Celi-Erazo. CIZ, Universidad Central de Ecuador, Quito, Ecuador.E.S. Loker. Center for Evolutionary and Theoretical Immunology, Museum of Southwestern Biology, Department of Biology, University of New Mexico,Albuquerque, NM 87131, USA.O. Noya. Centro para Estudios Sobre Malaria, Instituto de Altos Estudios “Dr. Arnoldo Gabaldón” – Instituto Nacional de Higiene “Rafael Rangel” delMinisterio del Poder Popular para la Salud y Sección de Biohelmintiasis, Instituto de Medicina Tropical, Facultad de Medicina, Universidad Central deVenezuela, Caracas, Venezuela.R. Rodríguez-Hidalgo. CIZ, Universidad Central de Ecuador, Quito, Ecuador; Facultad de Medicina Veterinaria y Zootecnia, Universidad Central delEcuador, Quito, Ecuador.N. Uribe. Escuela de Bacteriología y Laboratorio Clínico, Facultad de Salud, Universidad Industrial de Santander, Bucaramanga, Colombia.J.-P. Pointier. USR 3278 CNRS–EPHE, CRIOBE Université de Perpignan, 68860 Perpignan-CEDEX, France.S. Hurtrez-Boussès. MIVEGEC, UMR UM – CNRS 5290 – IRD 224 Maladies Infectieuses et Vecteurs : Ecologie, Génétique, Evolution et Contrôle, CentreIRD, BP 64501, 34394 Montpellier CEDEX 5, France; Département de Biologie–Ecologie, Faculté des Sciences, Université Montpellier, 34095 MontpellierCEDEX 5, France.Corresponding author: Manon Lounnas (email: manon.lounnas@gmail.com).Copyright remains with the author(s) or their institution(s). Permission for reuse (free in most cases) can be obtained from RightsLink.

425

Can. J. Zool. 96: 425–435 (2018) dx.doi.org/10.1139/cjz-2016-0319 Published at www.nrcresearchpress.com/cjz on 22 November 2017.

Can

. J. Z

ool.

Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

195.

83.9

9.4

on 0

6/06

/18

For

pers

onal

use

onl

y.

microsatellites ainsi mis au point aideront à préciser l’histoire de l’expansion de G. schirazensis et pourraient aider à comprendreson rôle comme vecteur potentiel de maladies infectieuses. [Traduit par la Rédaction]

Mots-clés : lymnéidés, Lymnaea, Galba schirazensis, Galba truncatula, vecteur, microsatellites, autofécondation.

IntroductionDuring the last century, human activities and climatic change

have considerably accelerated the spread of species across theirnatural barriers and modified their areas of distribution (Kolarand Lodge 2001), threatening local biodiversity (Davis 2009). Tobetter understand these threats, it is important to more fullycharacterize the species involved and the expansion pathways.Freshwater systems are particularly sensitive to bioinvasion risks(Beisel and Lévêque 2010). The Mollusca includes a large numberof species invasive in freshwater habitats (Nunes et al. 2015). Wehere focus on the family Lymnaeidae in which several species haveshown widespread long-distance colonization (Jabbour-Zahab et al.1997; Meunier et al. 2001; Kopp et al. 2012; Lounnas et al. 2017b).Long-distance dispersal can occur as a result of human activities,such as aquarium trade (Duggan 2010). Lymnaeids display markedresistance to desiccation, which increases their survival probabil-ity (Chapuis and Ferdy 2012). If released in conducive new envi-ronments, then even a single individual can found a populationbecause lymnaeids are capable of self-fertilization, which alsohappens to be the main reproductive mode in some species (e.g.,Meunier et al. 2004b; Escobar et al. 2011; Lounnas et al. 2017a).Such colonization events may also spread food- and water-bornetrematodes carried by lymnaeids (Meunier et al. 2001; Mas-Comaet al. 2005; Correa et al. 2010), either because dispersing snails arethemselves infected or because their presence in a new locationfacilitates subsequent transmission that originates from nearbyinfected vertebrates. Importantly, the worldwide expansion of theliver fluke Fasciola hepatica Linnaeus, 1761, a parasite infecting live-stock and humans, has been facilitated by the dispersal of lym-naeids (Hurtrez-Boussès et al. 2001; Mas-Coma et al. 2005).

Despite their relevance to human and animal health, lymnaeidshave often been misidentified especially the Galba species, a phy-logenetically distinct group of small-shelled lymnaeids (Correaet al. 2010, 2011; Lounnas et al. 2017a). These species exhibit highphenotypic plasticity in shell shape and display extremely similaranatomical traits that make accurate species identification diffi-cult in the absence of molecular data (Samadi et al. 2000; Correaet al. 2011). For instance, Galba schirazensis (Küster, 1862), first de-scribed by Bargues et al (2011), has only recently been distin-guished from Galba truncatula (O.F. Müller, 1774) in Europe andAsia and from G. truncatula, Galba cubensis (L. Pfeiffer, 1839), andGalba viator (Orbigny, 1835) in the Americas (Correa et al. 2010,2011; Bargues et al. 2011). However, G. schirazensis is now describedin many regions of the world, probably favored by its wide habitatrange and its capacity to survive outside water for extended peri-ods of time (Bargues et al. 2011). Recent molecular-based studiesreported G. schirazensis in Iran, Egypt, Reunion Island, Spain, Do-minican Republic, Mexico, Colombia, Venezuela, Ecuador, andPeru (Bargues et al. 2011; Correa et al. 2010, 2011; reported asLymnaea sp. and Galba sp. by the latter authors), but went of courseunnoticed because the species was recently described. However,only highly conserved genetic markers have been used so far inspecies discrimination (Bargues et al. 2011; Correa et al. 2010, 2011).The population structure and genetic diversity of the invasivelymnaeid G. schirazensis remains therefore largely unknown.

We developed microsatellite markers in G. schirazensis and usethem to genotype 242 individuals from 18 localities in Peru, Ecua-dor, Colombia, Venezuela, USA, Spain, and Reunion Island. Wecharacterized the population structure and genetic diversity anddiscuss our result in light of the breeding system and recent his-tory of expansion of this species. We also assessed the specificity

of these markers by testing amplification in the morphologicallysimilar and closely related Galba species, i.e., G. truncatula, G. cubensis,and G. viator. We also sequenced a mitochondrial gene (cyto-chrome oxidase subunit I (CO1)) to further elucidate the phylogeo-graphic relationships among these populations and the populationsstudied by Bargues et al. (2011) and Correa et al. (2011).

Materials and methods

Development of specific microsatellite markersSnail DNA was extracted from four pooled individuals identi-

fied by Correa et al. (2011) as Galba sp. to reach a total of 2–3 �g oftotal DNA. These individuals were collected in Río Negro, Antio-quia, Colombia (06°07=21==N, 75°26=57==W). DNA was extractedfrom foot tissue using the DNeasy Blood and Tissue Kit (Qiagen)and individuals were identified to species using the nuclear genes18S (GenBank accession numbers JN614335, JN614339, JN614340,JN614342), ITS-1 (HQ283253, JN614429, JN614430, JN614432), ITS-2(HQ283263, JN614455, JN614456, JN614458), and the mtDNA geneCO1 (JN614370, JN614371, JN614373, JN614374). All these sequencesshowed 99%–100% similarity with the sequences of G. schirazensisreported by Bargues et al. (2011).

Microsatellite loci were isolated from two enriched libraries(TC10 and TG10) following protocols described in Dubois et al.(2005) and using biotin-labeled microsatellite oligoprobes andstreptavidin-coated magnetic beads. The enriched molecules werecloned into the pGEMt vector used to transform XL1-Blue Super-competent Cells. Recombinant clones were screened with TC10,TG10, and AGE1 (AAACAGCTATGACCATGATTAC) or AGE2 (TTGTAAAACGACGGCCAGTG) oligonucleotides using a modified PCRmethod (Waldbieser 1995). We screened 228 clones, 163 of whichgave a positive signal and were sequenced using an ABI Prism3100 sequencer (Applied Biosystems). Ninety-seven sequences in-cluded a repeated motif and flanking regions allowing determina-tion of PCR primers that were designed using Primer3 (Rozen andSkaletsky 2000). We retained the 25 loci that showed the largestuninterrupted stretches of repeated motifs and selected the22 ones that amplified successfully (Table 1). We tested these22 microsatellite loci in six individuals from five localities from theAmericas: Finca Jocum Bucaramanga (Colombia), Huagrahuma(Ecuador), Manto de la Novia (Ecuador), Louisiana Bedico (USA;two individuals), and Bodoque (Venezuela) (Table 2, Fig. 1). Wefinally selected the 13 loci that were polymorphic in these sixindividuals for further population genetic analyses.

To test the specificity of microsatellite markers, we also ampli-fied these loci in individuals of three species closely related toG. schirazensis: G. truncatula, G. cubensis, and G. viator. We used twoindividuals of each species that were identified using the molec-ular markers 18S, ITS-1, ITS-2, and CO1 (GenBank accession num-bers in Correa et al. 2011).

Population genetic analyses using microsatellite markersWe analyzed 238 individuals of G. schirazensis from 16 localities

from Colombia, Ecuador, Peru, USA, and Venezuela, as well asindividuals from the Old World, one from Spain and three fromReunion Island (Table 2, Fig. 1). The Spanish specimen was col-lected from within the geographic range of G. truncatula and thepopulation from Reunion Island was first ascribed to the intro-duced species G. truncatula (Griffiths and Florens 2006). The Span-ish specimen was identified by Correa et al. (2011) as G. schirazensisusing 18S, ITS-1, ITS-2, and COI (reported as Galba sp. by the authors;GenBank accession numbers JN614341, JN614436, JN614454,

426 Can. J. Zool. Vol. 96, 2018

Published by NRC Research Press

Can

. J. Z

ool.

Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

195.

83.9

9.4

on 0

6/06

/18

For

pers

onal

use

onl

y.

JN614377). We clearly identified the three individuals from Re-union Island as G. schirazensis by analyzing the CO1 gene (GenBankaccession numbers in Table 2), which presented 99%–100% ofsimilarity with the four sequences of G. schirazensis reported byBargues et al. (2011) and only 90% with the sequence of G. truncatulafrom Spain reported by the same authors (AM494011) — see alsothis study showing that all the highly specific microsatellite ofG. schirazensis successfully amplified in these four individuals. De-spite the small sample sizes, we included these individuals in ouranalyses because they constitute the first mention of G. schirazensisin these two areas. All samples were collected from small areas(<2 m2) to prevent a Wahlund effect (Meunier et al. 2004a).

Snails were killed in 70 °C water and immediately stored in 70%ethanol.

For each specimen, we removed the distal part of the foot,which was twice compressed between paper towels to removeexcess ethanol. DNA extractions were then performed using200 �L of 5% Chelex® (Chelex Bio Rad diluted in a Tris–EDTAbuffer) solution incorporating 5 �L of proteinase K (Sigma) at aconcentration of 20 mg/mL. This suspension was heated at 56 °Cfor 6 h followed by gentle vortexing and a further incubation at95 °C for 10 min. The mixture was gently vortexed and centrifugedat 10000g for 10 s. The supernatant (100 �L) was collected, diluted1:10 in deionized water, and stored at –20 °C.

Table 1. Microsatellite loci in the freshwater snail Galba schirazensis.

Locus Primer sequences (5=¡3=)Labeldye

Multiplexset

GenBankaccession No.

Repeatmotif Na RAS

Gsch_1 F: TTTTGGGTCAACATTAGGTTAGG PET A KT324711 (TC)13 3 213–249R: GAACATTTGACACAGTAGCGTTG

Gsch_2 F: ACGTGCACACTTCTCCCTCT PET KT324709 (CT)23 1 185R: GCCTTGGTGCAGTTTTGTATT

Gsch_3 F: TGTAGGCAAAGGCACAAAAA PET B KT324702 (CT)13 4 152–178R: AGGGTGTAAGGGCTGAATTG

Gsch_4 F: TTCATTGTCTGCTCCTGCTG VIC B KT324708 (TC)20 2 157–160R: CGCCACTGGTCGATAGACTT (CT)13

Gsch_5 F: GGTCGTTTAGCAGCCTAGCA NED A KT324710 (GT)10 2 179–185R: CGACCGTTCAAACGTTACTG (AG)13

Gsch_7 F: AAACGACCGTTTCTAAGGTGA VIC KT324707 (AG)23 1 244R: CCCTGAACCAACAGAGCATT

Gsch_9 F: GGCGGAAACGAAGAGAGTAA NED B KT324705 (TC)11 4 207–227R: TACGTGCACACACTCAACCA

Gsch_10 F: AGGAGAGTCCCATTGAGCTG PET KT324706 (GT)11 1 218R: CGAAACTGTTGAAGGCATTG

Gsch_11 F: AACACATTTCCACCCACACA FAM B KT324712 (CT)18 2 216–235R: CTTTCTTTCGCGTGGGGTAT

Gsch_12 F: CTGGGGCTAACCCAAGAATC FAM KT324713 (TG)10 1 151R: TTTGGGGTTGAGGGCTTTAT

Gsch_13 F: TCACGTTCTCGTGGTTCTCA VIC C KT324714 (CA)10 5 166–173R: GTCGATGGGGCTATGTGTCT

Gsch_14 F: CGATGGCGCCAACTTATTTA VIC A KT324715 (CA)10 2 201–222R: TCATATCGCTACGGACATTCA

Gsch_15 F: TGAGACGGGCAAGATTTCTC NED KT324716 (CT)15 1 154R: AGGGTTCGATTCCCATCTCT

Gsch_17 F: CTTCATCCACGCAGCAAGTA PET KT324717 (AC)13 1 205R: GGGGGCCGATATTAATTTTT

Gsch_18 F: GACGGACCGTGAATTAGAGC FAM KT324718 (AG)11 1 210R: ATTGTTGCGGCGGATAACTA

Gsch_19 F: CGGTATTAGGGTGTCATGTGC FAM KT324719 (CA)12 1 171R: CAGGGGGAACCATAAAGTTG

Gsch_20 F: CAGTGAGACAGAGCCACGAA FAM KT324720 (AG)31 1 192R: ATTGCCGACACGTTTGTTGT

Gsch_21 F: CGCTAGACCTTTCTTTCTGTCC NED C KT324721 (TC)14 3 239–279R: TTACAAGCTCTTGGGAACGA

Gsch_22 F: TGTGTGTGTTTGTGGAGAGAGA PET C KT324722 (AG)11 3 249–322R: GTGTTACGCATGGTGAACCT

Gsch_23 F: AATGACCCAGTGGGGAAG NED C KT324723 (AC)16 2 227–232R: TGGGGAAGGTTCAATTGTTT

Gsch_24 F: GCGTGCGTGTATGTGAAAGA PET A KT324704 (AG)10 3 167–171R: GGGGCTCTTCAAGTGTGTGT

Gsch_25 F: AGCCAGACAAAGGGGGATAG VIC C KT324703 (GA)19 2 188–190R: GGGCAGGTTCATTACTCTGTTC

Note: Description of loci and genetic variation in six individuals from five localities (polymorphic loci are set in boldface type).Annealing temperature is 55 °C for all loci. Na, number of alleles detected; RAS, range of allele size (in base pairs).

Lounnas et al. 427

Published by NRC Research Press

Can

. J. Z

ool.

Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

195.

83.9

9.4

on 0

6/06

/18

For

pers

onal

use

onl

y.

We amplified microsatellite loci in an Eppendorf Thermal Cy-cler, in a total volume of 10 �L containing 5 �L of Taq PCR MasterMix Kit (Qiagen), 1 �L of the primer mix, the forward being labeledwith a fluorochrome (for the labels of each locus see Table 1), and1 �L of DNA. PCR conditions were as follow: 15 min activation at95 °C, 35 cycles including 30 s of initial denaturation at 94 °C, 90 sof annealing at 55 °C, and 60 s of extension at 72 °C, followed by30 min of final extension at 60 °C. For genotyping, we pooled 3 �Lof diluted (1:100) PCR products with 15 �L of Hi-Di Formamide and0.15 �L of GeneScan-500 LIZ Size Standard and analyzed it on anABI Prism 3100 Genetic Analyzer (Applied Biosystems). We per-formed multiplexed locus amplification by grouping loci as men-tioned in Table 1. Multiplexing locus amplification was possible byusing PCR products characterized by different-sized loci (no over-lapping zones) and labelled with different fluorochromes. Allelesizes were read using GeneMapper® version 4.0 software (AppliedBiosystems). Raw fragment sizes and corresponding peaks weremanually checked one by one to discard error in this allele’s call-ing step. Given the marked homozygosity, we evaluated our ca-pacity to detect heterozygotes: we amplified loci combining DNAfrom two individuals exhibiting different genotypes. We bothtested situations in which alleles exhibit a small size difference(1 base pair) or a large one (40 base pairs). In all situations, we wereable to detect heterozygotes.

We used the individual genotypes of the 238 American individ-uals to estimate allelic richness, observed and expected heterozy-gosities, the F statistic FIS, and the selfing rate (s) estimated ass = 2FIS/(1 + FIS) (Hartl and Clark 1997). The global and pairwisedifferentiations among populations were estimated using bothFST and RST. We could not use software that detect null allelesbecause lymnaeids are hermaphroditic. Standard Bonferroni cor-rections were applied in the case of multiple tests (Rice 1989). Weanalyzed data using Genodive version 2.0b23 (Meirmans andVan Tienderen 2004) and FSTAT version 2.9.3.2 (Goudet 1995) soft-wares. Individuals from Spain and Reunion Island were not in-cluded in the population genetic analysis because the smallsample sizes (N = 1 and N = 3, respectively) did not allow properestimation of population genetics parameters. However, we com-pared their genotypes to those of American individuals. Multilo-cus genotypes were clustered using a discriminant analysis ofprincipal components (DAPC) using the adegenet package of R(Jombart et al. 2010). One of the main advantages of this analysis isthat it does not rely on a particular population genetic model andis thus free of assumptions about Hardy–Weinberg or linkageequilibrium. Six principal components were retained based onthe � score. This score gives the number of principal componentsoptimized to capture the best discrimination without overfitting.

Amplification, sequencing of CO1, and phylogeographicanalysis

We amplified and sequenced the CO1 gene in 21 snails from 11populations (Fig. 3). Because of amplification failure due to DNAconservation issues, we did not sequence it in individuals fromAndaracas (Ecuador), Bodoque, Zea el Amparo, La Azulita, Baila-dores, and San Eusebio (Venezuela). The individual from El Rocío(Spain) was previously sequenced by Correa et al (2011) and weretrieved the sequence from GenBank (JN614377). The PCR prim-ers were LCO1490 (5=-GGTCAACAAATCATAAAGATATTGG-3=) andHCO2198 (5=-TAAACTTCAGGGTGACCAAAAAATCA-3=) (Folmeret al. 1994). We used 2.5 �L of DNA, 12.5 �L of Taq PCR Master MixKit (Qiagen), 2.5 �L of each primer (2 �mol/L), and 5 �L of distilledwater. PCR conditions were as follows: 15 min activation at 95 °C,35 cycles including 30 s of initial denaturation at 95 °C, 1 min ofannealing at 50 °C, and 10 min of extension at 72 °C, followed by10 min of final extension at 72 °C. Unidirectional DNA sequencingwas performed by Eurofins MWG Operon (Germany) using thecycle sequencing technology (dideoxy chain termination – cyclesequencing) on ABI 3730XL sequencing machines (Applied Bio-T

able

2.Po

pu

lati

ons

ofth

efr

esh

wat

ersn

ailG

alba

schi

raze

nsis

and

thei

rge

net

icva

riab

ilit

y.

Cou

ntr

yLo

cati

onA

cron

ymC

oord

inat

esSa

mp

lin

gye

arG

enB

ank

acce

ssio

nN

os.

NN

aH

eF I

S

Esti

mat

edse

lfin

gra

te

Peru

LaJo

yade

Are

quip

aLJ

16°2

8=56==S

,71°

49=0

7==W

2012

KY

1982

50,K

Y19

8260

141.

000

0.00

0—

—

Ecu

ador

Hu

agra

hu

ma

HU

02°4

7=32==S

,79°

16=3

1==W

2012

KT7

8130

2,K

T781

304

131.

000

0.00

0—

—M

anto

dela

Nov

iaM

N01

°24=

03==S

,78°

17=4

9==W

2011

KT7

8130

5,K

T781

315

151.

000

0.00

0—

—H

acie

nda

Cie

nag

aH

C00

°46=

18==S

,78°

37=1

0==W

2011

KT7

8130

112

1.00

00.

000

——

An

dara

cas

AN

00°2

6=10==S

,78°

32=2

2==W

2014

—14

1.07

70.

013

1(<

0.00

01)

1N

ono

NO

00°0

3=25==S

,78°

34=1

5==W

2014

KY

1982

55,K

Y19

8256

161.

000

0.00

0—

—

Col

ombi

aFi

nca

Jocu

mB

uca

ram

anga

FJ07

°06=

25==N

,73°

04=6

0==W

2012

KY

1982

53,K

Y19

8254

141.

000

0.00

0—

—

Ven

ezu

ela

Bod

oqu

eB

O08

°16=

15==N

,71°

48=5

1==W

2005

—18

1.07

70.

033

1(<

0.00

01)

1Lo

sN

evad

osLN

08°2

7=41==N

,71°

04=2

8==W

2013

KT7

8132

018

1.00

00.

000

——

Zea

elA

mp

aro

ZA08

°21=

40==N

,71°

46=0

1==W

2005

—9

1.00

00.

000

——

LaA

zuli

taLA

08°4

4=06==N

,71°

26=4

9==W

2005

—11

1.00

00.

000

——

Bai

lado

res

BA

08°1

4=05==N

,71°

50=2

6==W

2005

—18

1.00

00.

000

——

San

Euse

bio

SE08

°38=

39==N

,71°

23=4

2==W

2005

—18

1.00

00.

000

——

Saba

na

Alt

oSA

08°3

6=11==N

,71°

27=4

5==W

2013

KT7

8132

2,K

T781

323,

KT7

8132

47

1.00

00.

000

——

LaTr

amp

aLT

08°3

3=31==N

,71°

27=1

3==W

2013

KY

1982

51,K

Y19

8252

201.

154

0.04

81

(<0.

0001

)1

USA

Lou

isia

na

Bed

ico

LB30

°26=

11==N

,90°

15=0

1==W

NA

KT7

8133

221

1.07

70.

015

1(<

0.00

01)

1

Spai

nEl

Roc

íoER

37°0

8=02==N

,06°

28=1

8==W

2010

JN61

4377

1—

——

—

Reu

nio

nIs

lan

d(F

ran

ce)

Rav

ine

duG

olR

G21

°14=

26==S

,55°

25=0

7==E

2009

KY

1982

57,K

Y19

8258

,KY

1982

593

——

——

No

te:G

enet

icp

aram

eter

sw

ere

not

calc

ula

ted

inth

esa

mp

les

from

Spai

nan

dR

eun

ion

Isla

nd

beca

use

ofto

olo

wsa

mp

lesi

ze.N

,sam

ple

size

;Na,m

ean

nu

mbe

rof

alle

les;

He,e

xpec

ted

het

eroz

ygos

ity;

F IS,i

nbr

eedi

ng

coef

fici

ent

and

Pva

lue;

NA

,not

avai

labl

e.B

onfe

rron

icor

rect

ion

sw

ere

app

lied

(P=

0.01

25).

Obs

erve

dh

eter

ozyg

osit

yw

asn

ull

inal

lpop

ula

tion

s.

428 Can. J. Zool. Vol. 96, 2018

Published by NRC Research Press

Can

. J. Z

ool.

Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

195.

83.9

9.4

on 0

6/06

/18

For

pers

onal

use

onl

y.

systems). Sequencing quality was assessed based on a quality scoreQ per base (values provided by Eurofins Genomics).

In addition, we downloaded from GenBank 12 CO1 sequences ofG. schirazensis from Iran, Reunion Island, Peru, Ecuador, Colombia,Venezuela, and Mexico (JF272607–JF272610: Bargues et al. 2011;JN614370–JN614378: Correa et al. 2011), as well as one CO1 se-quence of G. truncatula from France (JN614386: Correa et al. 2011),of G. viator from Argentina (JN872451: Standley et al. 2013), and ofG. cubensis from Uruguay (JN614396: Correa et al. 2011).

These sequences were used for phylogeographic reconstruc-tion, first aligning sequences using webPRANK (Löytynoja andGoldman 2010). The alignment was manually inspected using theelectrophoregram of each sequence viewed on 4Peaks (Griekspoor andGroothuis 2005) and the quality report to discard single nucleotidepolymorphisms (SNPs) for which we had sequencing uncertainties(Q < 20). Then, we performed tree reconstruction using both max-imum likelihood (ML) and Bayesian inference (BI). ML analyseswere conducted using the best-fitting model of sequence evolu-tion. Model selection was based on Bayesian information criterionusing jModelTest (Posada 2008). We ran 5000 bootstrap replica-tions to test node robustness. BI analyses were performed withMrBayes version 3.2 (Ronquist and Huelsenbeck 2003). The treespace was explored using Markov chain Monte Carlo analyseswith random starting trees and four simultaneous, sequentiallyheated independent chains sampled every 500 trees during fivemillion generations. Suboptimal trees were discarded once the“burn-in” phase was identified and a majority-rule consensus tree,with posterior probability support of nodes, was constructedwith the remaining trees. The tree was visualized using FigTree

version 1.4.2 (http://tree.bio.ed.ac.uk/software/figtree/). Both treeswere edited with Adobe Illustrator CC version 19.0.

The number of haplotypes, the number of polymorphic sites,and the nucleotide diversity were calculated using DnaSp version 5(Librado and Rozas 2009). Genetic distances were estimated based onthe number of base-pair differences per site (p distance) among hap-lotypes using Mega7 (Kumar et al. 2016). A minimum-spanning net-work was constructed using the software SplitsTree4 (Huson andBryant 2006) to represent the distances between haplotypes.

Results

Microsatellite markersThe sequences and primers of the 22 microsatellite loci that we

identified in G. schirazensis are available in GenBank (accessionnumbers in Table 1). No amplification was observed in the closelyrelated species G. truncatula, G. cubensis, and G. viator, suggestingthat these 22 loci are specific to G. schirazensis.

We characterized the variation at the 13 polymorphic loci in allAmerican individuals. The mean (±SD) number of alleles per locus was2.846 ± 0.274, the observed heterozygosity was 0 at all loci, and themean (±SD) expected heterozygosity was 0.007 ± 0.005 (Table 2). Formost loci and populations, the lack of diversity prevented us fromestimating FIS (Table 2). Among the four loci and populations thatshowed allelic diversity at some loci (Supplementary Table S1),1 theobserved heterozygosity was always zero (FIS = 1, s = 1; Table 2).

We found high genetic differentiation among populations(global: FST = 0.979, P < 0.001; global: RST = 0.979, P < 0.001). Allpopulation pairs coming from different countries (except for

1Supplementary tables and figure are available with the article through the journal Web site at http://nrcresearchpress.com/doi/suppl/10.1139/cjz-2016-0319.

Fig. 1. Geographic location of multilocus microsatellite genotypes of the freshwater snail Galba schirazensis. Each multilocus genotype isindicated by a color and a letter. Each pie chart represents a population. For population acronyms see Table 2.

Lounnas et al. 429

Published by NRC Research Press

Can

. J. Z

ool.

Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

195.

83.9

9.4

on 0

6/06

/18

For

pers

onal

use

onl

y.

Venezuela and Colombia) showed significant FST after Bonferroniadjustment (P = 0.0004; Supplementary Table S2).1 In contrast,populations from the same country or from adjacent countriesusually showed no significant differentiation (Ecuador: FST < 0.019,NS (not significant); Venezuela and Colombia: FST < 0.235, NS). Anexception is the Venezuelan population from La Trampa that differssignificantly from most other Venezuelan and Colombian popula-tions with FST values between 0.3 and 0.45 (Supplementary Table S2).1

Nine multilocus genotypes were found (Fig. 1). Some loci (1–8)could not be genotyped in some individuals (N = 65). Thus, we didnot ascribe these individuals with incomplete genotypes to anymultilocus genotype. However, the loci that could be genotypedin these individuals were exactly the same as in the individualswhere all the loci were genotyped (Supplementary Table S1).1 Toanalyze the genetic structure of populations using DAPC, we usedall the individuals except for 12 individual genotypes that hadmore than three loci that did not amplify (Supplementary Table S1).1

Four clusters were detected by DAPC, which grouped together

individuals from (1) Colombia, Venezuela, Spain, and ReunionIsland; (2) Peru; (3) USA; and (4) Ecuador (Fig. 2; SupplementaryTable S11). Each cluster consisted of a single main multilocusgenotype with small within-cluster variation (SupplementaryTable S1).1 Differences between the main genotypes and variantsdid not exceed one repeat (2 base pairs) and was of 1 base pair insome cases. The exception is the Spanish genotype that differedfrom the main type at two loci (Supplementary Table S1).1 Themain genotype from each cluster differed from those of otherclusters at five loci or more (Supplementary Table S1).1 The largestdifference was observed between clusters from Ecuador and USA(all 13 loci fixed for different alleles).

CO1 sequencesWe obtained 21 CO1 sequences of G. schirazensis (Table 2). The

analysis of these sequences and sequences of G. schirazensis re-trieved from GenBank revealed five haplotypes (number of se-quences = 37, mean (±SD) nucleotide diversity = 0.005 ± 0.001,

Fig. 2. Genetic population structure of the freshwater snail Galba schirazensis. Scatterplot of 230 individuals on the first two DAPC axes(first axis: 61% of total variance; second axis: 22% of total variance). Individuals group in four clusters. Color version online.

430 Can. J. Zool. Vol. 96, 2018

Published by NRC Research Press

Can

. J. Z

ool.

Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

195.

83.9

9.4

on 0

6/06

/18

For

pers

onal

use

onl

y.

mean (±SD) haplotype diversity = 0.499 ± 0.083) determined by10 polymorphic sites without indels (Supplementary Table S3).1 TheCO1 sequences diverged from 0.2% (USA and Mexico; USA andEcuador) to 1.6% (USA and Peru: from Bargues et al. 2011) (Supple-mentary Table S4).1

The best model describing the evolution of the CO1 sequenceswas HKY+I. The ML tree showed that these sequences pooled in

three clades: (1) Ecuador (cluster 4 in DAPC), (2) the single se-quence from USA (cluster 3), and (3) a worldwide clade includingPeru (cluster 2), Venezuela, Colombia, Reunion Island, and Spain(cluster 1). GenBank sequences from Iran, Venezuela, Colombia,and Reunion Island fell in the worldwide clade, those from Ecua-dor in the Ecuador clade, and those from Mexico were close to theUSA sequence (Fig. 3). The BI tree resulted in the same grouping

Fig. 3. Maximum-likelihood phylogenetic tree of the freshwater snail Galba schirazensis based on mtDNA CO1 sequences. Bootstrap values areindicated at each node. Sequences are their GenBank accession numbers. Colored sequences are those obtained in this study, whereas theothers were retrieved from GenBank. Galba truncatula JN614386 sequence (Correa et al. 2011), Galba viator JN872451 sequence (Standley et al.2013), and Galba cubensis JN614396 sequence (Correa et al. 2011) were used as the outgroups. For each sequence obtained here, the color coderefers to the cluster number derived from DAPC. Color version online.

Lounnas et al. 431

Published by NRC Research Press

Can

. J. Z

ool.

Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

195.

83.9

9.4

on 0

6/06

/18

For

pers

onal

use

onl

y.

(Supplementary Fig. S1).1 However, the USA sequence groupedwith the Ecuador clade in the ML tree, whereas it grouped with theworldwide clade in the BI tree. In both trees, these nodes were notwell supported by posterior probabilities and ML bootstrap (52%–72%).

The topologies of the ML tree and the BI tree were consistentwith the results from DAPC, grouping those individuals from thesame cluster in the same clade (Figs. 2 and 3; SupplementaryFig. S11). These analyses showed congruence between nuclear andmitochondrial genetic structures. Each mitochondrial haplotypeincluded one to five multilocus microsatellite genotypes, and nomultilocus microsatellite genotype was found associated with dif-ferent mitochondrial haplotypes. However, the distance betweenDAPC clusters was in some cases lower than the phylogeneticdistances. For instance, clusters 1 (worldwide) and 2 (Peru) weregenetically distant from each other in the DAPC, but they gath-ered in the same worldwide clade (Figs. 2 and 3). Also, the clus-ters 3 (USA) and 4 (Ecuador) were genetically distant from each

other in the DAPC (Fig. 2), but they were genetically close in thetree and the CO1 sequences diverged a little from each other (0.2%;Supplementary Tables S3 and S4;1 Figs. 3 and 4). In the minimum-spanning network, North American haplotypes (3 and 4) appearedto be intermediates between worldwide and Ecuadorian haplo-types (1 and 5; Fig. 4). This inconsistency probably reflects the factthat microsatellite loci are evolving much faster than mtDNAsequences.

DiscussionWe developed 13 polymorphic genetic markers to study the

population structure and genetic diversity of G. schirazensis, aninvasive freshwater snail inhabiting Asia, Europe, and the NewWorld. These microsatellite loci appear to be specific to G. schirazensisbecause we did not find any amplification in the morphologicallysimilar and most closely related species G. truncatula, G. cubensis, andG. viator.

Fig. 4. Minimum-spanning network of CO1 haplotypes from the freshwater snail Galba schirazensis. Colors and haplotypes are those referred inSupplementary Table S4.1 Each circle represents a haplotype, its size being proportional to the frequency of occurrence of a certain haplotype.Numbers of haplotype occurrences are indicated in each circle. Color version online.

432 Can. J. Zool. Vol. 96, 2018

Published by NRC Research Press

Can

. J. Z

ool.

Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

195.

83.9

9.4

on 0

6/06

/18

For

pers

onal

use

onl

y.

Analyzing New World populations, we observed very low ge-netic diversity within populations, a total lack of heterozygotes,and marked population structure. The lack of observed heterozy-gosity in those few populations that showed allelic diversity sug-gests that G. schirazensis mainly reproduces by self-fertilization.This is consistent with the fact that under laboratory conditions,individuals isolated from birth are able to self-fertilize (Bargueset al. 2011). Our population genetic evidence for self-fertilizationin natural population only relies on five locus-by-population com-binations, all involving relatively rare alleles segregated in thepopulations (Table 2; Supplementary S11). In the absence of self-fertilization, rare alleles are expected to be mainly in theheterozygous state, so it would be very improbable for them toappear only as homozygotes in these five cases. Therefore, a highselfing rate is the most likely explanation for the lack of heterozy-gosity. In addition, related lymnaeid species exhibiting selfingrates in the 80%–100% range include G. cubensis (Lounnas et al.2017a), G. truncatula (Trouvé et al. 2000; Meunier et al. 2001, 2004a),Omphiscola glabra (O.F. Müller, 1774) (Hurtrez-Boussès et al. 2005),and Pseudosuccinea columella (Say 1817) (Nicot et al. 2008), which alsoshow very low or no within-population allelic diversity, high FIS,and large FST. In contrast, most populations of mainly outcrossinglymnaeids such as Lymnaea stagnalis (Linnaeus, 1758) (Puurtinenet al. 2007; Kopp et al. 2012) and Radix sp. (Pfenninger et al. 2011)have high polymorphism, low FIS, and moderate FST. Therefore,G. schirazensis shares the common characteristics of highly selfinglymnaeid species studied so far. It also shares a similar ecologywith G. cubensis, G. truncatula, and O. glabra. All of them live mainlyin temporary habitats that undergo frequent flooding anddroughts that create strong bottlenecks (Meunier et al. 2004b). Inthis context, self-fertilization allows a single individual to colo-nize vacant habitats and to produce a new population (Meunieret al. 2004a). Finally, the recently analyzed transcriptomes of bothG. truncatula and G. schirazensis bear molecular signatures consis-tent with ancient self-fertilization, dating back to their commonancestor (Burgarella et al. 2015).

It is very difficult to know a priori where G. schirazensis waspresent before 2010, hence whether each particular population isnative or invasive. Galba schirazensis has probably been misidenti-fied in previous reports of small lymnaeid species from all overthe world, either as G. truncatula (everywhere) or as G. cubensis andG. viator (New World) (Correa et al. 2010, 2011; Bargues et al. 2011),because they all have very similar shell morphology and internalanatomy. Thus, it is impossible to trace invasion routes usingreports from the literature, except for places known from histor-ical records to be devoid of native small-shelled lymnaeids, suchas Reunion Island where the G. schirazensis populations necessarilyresult from introduction(s). The geographic origin of G. schirazensisremains unknown. Bargues et al. (2011) assumed that this speciesis native to the Middle East because it was first described in Iran(Küster 1862). From this postulate Bargues et al. (2011) hypothesizetwo spreading waves: first an Old World spread (from Iran intoAfrica, Europe, and Asia) and second a Trans-Atlantic spread intothe Americas through the livestock transported by the Spanishconquerors during the early colonization period. However, as it istypical of species descriptions of that time, only a brief descrip-tion of the shell of G. schirazensis was provided, which could in prac-tice fit with that of many small-shelled lymnaeids, including thevery widespread G. truncatula. Therefore, the snail lineage studied inthe present paper, re-described based on molecular sequences asG. schirazensis by Bargues et al. (2011), as Lymnaea sp. by Correa et al.(2010), and as Galba sp. by Correa et al. (2011), may not necessarilybe (i) the same snail as that named G. schirazensis by Küster (1862)and (ii) native from the Middle East. Its current occurrence in Iran(Bargues et al. 2011) could well result from an introduction.

In this confused taxonomical situation, it is important to clearlylist the available evidence on the possible geographical origin ofG. schirazensis. Any mention of this species can only be ascertained

based on molecular evidence to avoid misidentification. This elim-inates all mentions prior to 2010, including the first description byKüster (1862). Consequently, we must rely only on genetic diver-sity and phylogeny to identify likely origin areas. Bargues et al.(2011) showed that all the individuals they sampled in the OldWorld (Iran, Egypt, and Spain) have exactly the same haplotypesat the nuclear genes 18S, ITS-2, and ITS-1, and at the mitochondrialgenes 16S and CO1. However, they observed two to four differenthaplotypes for each gene in the American populations (DominicanRepublic, Mexico, Peru, Ecuador, and Venezuela), including theOld World haplotype. Our microsatellite study corroborates thispattern, as each of the four different regions sampled in the NewWorld forms well-defined and differentiated clusters and the fewindividuals from two very distant places in the Old World (Re-union Island and Spain), including one from a known recent in-troduction (Reunion Island), are genetically related to only oneof these clusters (Fig. 1). The phylogeographic tree is consistentwith this pattern, as the American CO1 sequences are variable,whereas the sequences from the Old World showed no variationand all belong to the large worldwide clade (Fig. 3; SupplementaryTable S21). All this suggests that the center of diversity, and byextension the origin area of this species, should be found in theNew World, whereas Old World populations should rather resultfrom a recent introduction of a genetically uniform population. Inother snail species with the ability to self-fertilize, genetic unifor-mity over large distances and many populations is indeed thesignature of recent introductions; for example, in L. stagnalisintroduced in New Zealand (Kopp et al. 2012), G. truncatula intro-duced on the Bolivian Altiplano (Meunier et al. 2001), or P. columellaintroduced in many parts of the world (Lounnas et al. 2017b). Fi-nally, an American origin for G. schirazensis is also corroborated bythe phylogeny of Correa et al. (2010), who concluded that thewhole Galba clade likely has an American origin, because most ofits species are native there (Galba cousini (Jousseaume, 1887),G. viator, Galba neotropica (Bargues, Artigas, Mera y Sierra, Pointierand Mas-Coma, 2007), G. cubensis, Galba humilis (Say, 1822)). Thehypothesis of a post-Columbian introduction of G. schirazensisfrom Europe to the New World (Bargues et al. 2011) is not consis-tent with our findings: it would have been very unlikely to accu-mulate fixed differences at 13 distinct loci (as we observedbetween USA and Ecuador) in such a short time, while at the sametime Venezuelan, Colombian, and Old World populations wouldremain nearly identical and devoid of diversity over huge geo-graphical distances.

Although our data together along with the report of Bargueset al. (2011) seem sufficient to reject the hypothesis of a singlerecent introduction of G. schirazensis in the New World, more ex-tensive sampling is needed (i) to test whether all the populationsfrom the Old World have a single recent introduced origin, anhypothesis so far consistent with the results of conservative(Bargues et al. 2011; Correa et al. 2011; this study) and polymorphic(this study) markers and (ii) to more fully understand the geneticvariation observed among South and North America isolates. Theapplication of the genetic markers developed here to a more ex-tensive sample that covers a worldwide range will therefore helpto clarify the geographic origin and invasion routes of G. schirazensis.

Combining markers with contrasted mutation rates, we ob-served discordances in genetic structuration between phyloge-netic and DAPC analyses. Populations from Peru and populationsdistributed worldwide appeared to be significantly differentiatedat a microsatellite level, whereas they grouped in the same cladeat a mitochondrial level (haplotype 1). Microsatellite loci havehigh mutation rates (between 1 × 10−2 and 1 × 10−4 per generation;Jarne and Lagoda 1996; Ellegren 2002) and tens of generations cangenerate mutations at least at one locus, while several thousandsof years are needed to generate one base substitution at the CO1gene (around 1.35 × 10−8/site per year according to Wilke et al.2009). This result suggests that Peruvian population and popula-

Lounnas et al. 433

Published by NRC Research Press

Can

. J. Z

ool.

Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

195.

83.9

9.4

on 0

6/06

/18

For

pers

onal

use

onl

y.

tions that belong to the worldwide genotype probably divergedrecently. Similarly, we found that populations from Ecuador andUSA were strongly differentiated at the nuclear level, while theywere genetically close at the mitochondrial level (0.2% of divergence),suggesting a recent divergence event. Moreover, sequences fromUSA and Mexico branched differentially between the BI tree andthe ML tree with moderate posterior probabilities and ML boot-strap. North American haplotypes appeared to be intermediatebetween haplotype 1 (worldwide) and haplotype 5 (Ecuador),which were strongly differentiated at both mitochondrial andmicrosatellite markers (1.2%; Fig. 4). Finally, we cannot excludethe possibility of mtDNA introgression, which is often responsibleof mito-nuclear discordance (Toews and Brelsford 2012). This lackof clear structuration of North and South American populationsat a mitochondrial level shows that our sample size is too limited;more genetic markers would have been required to understandthe fine-scale population structure and to bridge gaps in our un-derstanding of both ancient population splits and more recentexpansion events of G. schirazensis.

Understanding the population structure and the genetic diver-sity of G. schirazensis can also help elucidating its role in spreadinginfectious diseases. Caron (2015) reported high prevalences ofF. hepatica in G. schirazensis sampled in Ecuador. However, Bargueset al. (2011), who experimentally infected snails from one Egyptianlocality and one Spanish locality, observed that the parasite didnot complete its life cycle. Dreyfuss et al. (2014), in experimentalinfections of successive generations of Colombian G. schirazensis,detected prevalences ranging from 0% to 15.3%, but free cercariawere only observed in 2 of the 400 studied snails. It is worthmentioning that, in the two studies cited above, only allopatriccombinations have been tested and no experiment has been con-ducted so far with sympatric parasites. Therefore, it would beprudent to consider G. schirazensis as a putative, perhaps occa-sional, intermediate host of F. hepatica. However, our capacity toidentify which parasites are transmitted by G. schirazensis has hith-erto been hindered by the morphological similarity among Galbaspecies (Correa et al. 2011). We need to identify by molecularmeans morphologically cryptic Galba species and to identify therole of G. schirazensis as an intermediate host of food- and water-borne trematodes such as F. hepatica.

In conclusion, our analysis provides strong support of the ideasthat this species is predominantly self-fertilizing and has an Amer-ican origin with recent colonization of the Old World by a genet-ically uniform strain related to populations from Venezuela andColombia — a hypothesis that awaits confirmation by more ex-tensive sampling.

AcknowledgementsWe thank N. Bonel for useful comments on an earlier draft of

the manuscript. We thank R. Ayaqui for providing us specimensfrom Peru. M.L. was supported by a doctoral fellowship from theUniversity of Montpellier and a postdoctoral grant from LabexCeMeb; A.C.C. was supported by a doctoral fellowship from IRD;P.A. was supported by the Foundation Méditerranée Infection;S.H.-B. was supported by the University of Montpellier. This studywas financially supported by IRD, CNRS, and the University ofMontpellier.

ReferencesBargues, M.D., Artigas, P., Khoubbane, M., Flores, R., Glöer, P., Rojas-García, R.,

Ashrafi, K., Falkner, G., and Mas-Coma, S. 2011. Lymnaea schirazensis, an over-looked snail distorting fascioliasis data: genotype, phenotype, ecology,worldwide spread, susceptibility, applicability. PLoS ONE, 6: e24567. doi:10.1371/journal.pone.0024567.

Beisel, J.N., and Lévêque, C. 2010. Introductions d’espèces dans les milieux aqua-tiques : faut-il avoir peur des invasions biologiques? Éditions Quae, c/o INRA,Versailles, France.

Burgarella, C., Gayral, P., Ballenghien, M., Bernard, A., David, P., Jarne, P.,Correa, A., Hurtrez-Boussès, S., Escobar, J., Galtier, N., and Glemin, S. 2015.

Molecular evolution of freshwater snails with contrasting mating systems.Mol. Biol. Evol. 32(9): 2403–2416. doi:10.1093/molbev/msv121. PMID:25980005.

Caron, Y. 2015. Aspects malacologiques du cycle de Fasciola hepatica en Belgiqueet en Equateur. Docteur en Sciences Vétérinaires, Université de Liège, Liège,Belgique.

Chapuis, E., and Ferdy, J.B. 2012. Life history traits variation in heterogeneousenvironment: the case of a freshwater snail resistance to pond drying. Ecol.Evol. 2(1): 218–226. doi:10.1002/ece3.68. PMID:22408738.

Correa, A., Escobar, J., Durand, P., Renaud, F., David, P., Jarne, P., Pointier, J.-P.,and Hurtrez-Boussès, S. 2010. Bridging gaps in the molecular phylogeny ofthe Lymnaeidae (Gastropoda: Pulmonata), vectors of Fascioliasis. BMC Evol.Biol. 10(1): 381. doi:10.1186/1471-2148-10-381. PMID:21143890.

Correa, A.C., Escobar, J.S., Noya, O., Velásquez, L.E., González-Ramírez, C.,Hurtrez-Boussès, S., and Pointier, J.-P. 2011. Morphological and molecularcharacterization of Neotropic Lymnaeidae (Gastropoda: Lymnaeoidea), vec-tors of fasciolosis. Infect. Genet. Evol. 11(8): 1978–1988. doi:10.1016/j.meegid.2011.09.003. PMID:21968212.

Davis, M.A. 2009. Invasion biology. Oxford University Press, Oxford. doi:10.2134/jeq2004.2384.

Dreyfuss, G., Correa, A.C., Djuikwo-Teukeng, F.F., Novobilský, A., Höglund, J.,Pankrác, J., Kašný, M., Vignoles, P., Hurtrez-Boussès, S., Pointier, J.-P., andRondelaud, D. 2014. Differences in the compatibility of infection betweenthe liver flukes Fascioloides magna and Fasciola hepatica in a Colombianpopulation of the snail Galba sp. J. Helminthol. 89(6): 720–726. doi:10.1017/S0022149X14000509. PMID:25000491.

Dubois, M.-P., Jarne, P., and Jouventin, P. 2005. Ten polymorphic microsatellitemarkers in the wandering albatross Diomedea exulans. Mol. Ecol. Notes, 5(4):905–907. doi:10.1111/j.1471-8286.2005.01108.x.

Duggan, I.C. 2010. The freshwater aquarium trade as a vector for incidentalinvertebrate fauna. Biol. Invasions, 12(11): 3757–3770. doi:10.1007/s10530-010-9768-x.

Ellegren, H. 2002. Microsatellite evolution: a battle between replication slippageand point mutation. Trends Genet. 18: 70. doi:10.1016/S0168-9525(02)02631-8.

Escobar, J.S., Auld, J.R., Correa, A.C., Alonso, J.M., Bony, Y.K., Coutellec, M.A.,Koene, J.M., Pointier, J.-P., Jarne, P., and David, P. 2011. Patterns of mating-system evolution in hermaphroditic animals: correlations among selfingrate, inbreeding depression, and the timing of reproduction. Evolution (N.Y.),65(5): 1233–1253. doi:10.1111/j.1558-5646.2011.01218.x.

Folmer, O., Black, M., Hoeh, W., Lutz, R., and Vrijenhoek, R. 1994. DNA primersfor amplification of mitochondrial cytochrome c oxidase subunit I fromdiverse metazoan invertebrates. Mol. Mar. Biol. Biotechnol. 3(5): 294–299.PMID:7881515.

Goudet, J. 1995. FSTAT (Version 1.2): a computer program to calculate F-statistics.J. Hered. 86: 485–486. doi:10.1093/oxfordjournals.jhered.a111627.

Griekspoor, A., and Groothuis, T. 2005. 4Peaks: a program that helps molecularbiologists to visualize and edit their DNA sequence files. Version 1.7 [com-puter program]. Available from http://nucleobytes.com.

Griffiths, O.L., and Florens, F.B.V. 2006. A field guide to the non-marine molluscsof the Mascarene Islands (Mauritius, Rodrigues and Réunion) and the north-ern dependencies of Mauritius. Bioculture Press, Mauritius.

Hartl, D., and Clark, A. 1997. Principles of population genetics. Sinauer Associ-ates, Inc., Sunderland, Mass.

Hurtrez-Boussès, S., Meunier, C., Durand, P., and Renaud, F. 2001. Dynamics ofhost–parasite interactions: the example of population biology of the liverfluke (Fasciola hepatica). Microbes Infect. 3: 841–849. doi:10.1016/S1286-4579(01)01442-3. PMID:11580979.

Hurtrez-Boussès, S., Pendino, A., Barnabé, C., Durand, P., Rondelaud, D.,Durand, C., Meunier, C., Hurtrez, J., and Renaud, F. 2005. Comparison be-tween shell morphology and genetic diversity in two sympatric lymnaeidsnails, vectors of fasciolosis. Can. J. Zool. 83(12): 1643–1648. doi:10.1139/z05-150.

Huson, D.H., and Bryant, D. 2006. Application of phylogenetic networks in evo-lutionary studies. Mol. Biol. Evol. 23: 254–267. doi:10.1093/molbev/msj030.PMID:16221896.

Jabbour-Zahab, R., Pointier, J.-P., Jourdane, J., Jarne, P., Oviedo, J.A., Bargues, M.D.,Mas-Coma, S., Anglés, R., Perera, G., Balzan, C., Khallayoune, K., and Renaud, F.1997. Phylogeography and genetic divergence of some lymnaeid snails, inter-mediate hosts of human and animal fascioliasis with special reference tolymnaeids from the Bolivian Altiplano. Acta Trop. 64(3–4): 191–203. doi:10.1016/S0001-706X(96)00631-6. PMID:9107366.

Jarne, P., and Lagoda, P.J. 1996. Microsatellites, from molecules to populationsand back. Trends Ecol. Evol. 11: 424–429. doi:10.1016/0169-5347(96)10049-5.PMID:21237902.

Jombart, T., Devillard, S., and Balloux, F. 2010. Discriminant analysis of principalcomponents: a new method for the analysis of genetically structured popu-lations. BMC Genet. 11(1): 94. doi:10.1186/1471-2156-11-94. PMID:20950446.

Kolar, C.S., and Lodge, D.M. 2001. Progress in invasion biology: predicting invaders.Trends Ecol. Evol. 16(4): 199–204. doi:10.1016/S0169-5347(01)02101-2. PMID:11245943.

Kopp, K.C., Wolff, K., and Jokela, J. 2012. Natural range expansion and human-assisted introduction leave different genetic signatures in a hermaphroditicfreshwater snail. Evol. Ecol. 26(3): 483–498. doi:10.1007/s10682-011-9504-8.

Kumar, S., Stecher, G., and Tamura, K. 2016. MEGA7: Molecular evolutionary

434 Can. J. Zool. Vol. 96, 2018

Published by NRC Research Press

Can

. J. Z

ool.

Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

195.

83.9

9.4

on 0

6/06

/18

For

pers

onal

use

onl

y.

genetics analysis version 7.0 for bigger datasets. Mol. Biol. Evol. 33(7):msw054. doi:10.1093/molbev/msw054.

Küster, H. 1862. Die Gattungen Limnaeus, Amphipeplea, Chilina, Isidora und Phy-sopsis. In Systematisches Conchylien-Cabinet. 2nd ed. Edited by F.H.W. Martiniand J.H. Chemnitz. Bauer and Raspe, Nürnberg, Germany. p. I.17 b: issues180–182, pp. 1–48, pls. 1–11 (1862); i.

Librado, P., and Rozas, J. 2009. DnaSP v5: a software for comprehensive analysisof DNA polymorphism data. Bioinformatics, 25(11): 1451–1452. doi:10.1093/bioinformatics/btp187. PMID:19346325.

Lounnas, M., Vázquez, A.A., Alda, P., Sartori, K., Pointier, J.-P., David, P., andHurtrez-Boussès, S. 2017a. Isolation, characterization and population-geneticanalysis of microsatellite loci in the freshwater snail Galba cubensis (Lymnaeidae).J. Mollusc. Stud. 83(1): 63–68. doi:10.1093/mollus/eyw041.

Lounnas, M., Correa, A.C., Vázquez, A.A., Dia, A., Escobar, J.S., Nicot, A., Arenas, J.,Ayaqui, R., Dubois, M.-P., Gimenez, T., Gutiérrez, A., González-Ramírez, C., Noya, O.,Prepelitchi, L., Uribe, N., Wisnivesky-Colli, C., Yong, M., David, P., Loker, E.S.,Jarne, P., Pointier, J.-P., and Hurtrez-Boussès, S. 2017b. Self-fertilization, long-distance flash invasion and biogeography shape the population structure ofPseudosuccinea columella at the worldwide scale. Mol. Ecol. 26(3): 887–903.doi:10.1111/mec.13984. PMID:28026895.

Löytynoja, A., and Goldman, N. 2010. webPRANK: a phylogeny-aware multiplesequence aligner with interactive alignment browser. BMC Bioinformatics,11: 579. doi:10.1186/1471-2105-11-579. PMID:21110866.

Mas-Coma, S., Bargues, M., and Valero, M. 2005. Fascioliasis and other plant-borne trematode zoonoses. Int. J. Parasitol. 35: 1255–1278. doi:10.1016/j.ijpara.2005.07.010. PMID:16150452.

Meirmans, P.G., and Van Tienderen, P.H. 2004. GENOTYPE and GENODIVE: twoprograms for the analysis of genetic diversity of asexual organisms. Mol. Ecol.Notes, 4(4): 792–794. doi:10.1111/j.1471-8286.2004.00770.x.

Meunier, C., Tirard, C., Hurtrez-Boussès, S., Durand, P., Bargues, M.D.,Mas-Coma, S., Pointier, J., Jourdane, J., and Renaud, F. 2001. Lack of molluscanhost diversity and the transmission of an emerging parasitic disease in Bo-livia. Mol. Ecol. 10(5): 1333–1340. doi:10.1046/j.1365-294X.2001.01284.x. PMID:11380888.

Meunier, C., Hurtrez-Boussès, S., Durand, P., Rondelaud, D., and Renaud, F.2004a. Small effective population sizes in a widespread selfing species,Lymnaea truncatula (Gastropoda: Pulmonata). Mol. Ecol. 13(9): 2535–2543. doi:10.1111/j.1365-294X.2004.02242.x. PMID:15315668.

Meunier, C., Hurtrez-Boussès, S., Jabbour-Zahab, R., Durand, P., Rondelaud, D.,and Renaud, F. 2004b. Field and experimental evidence of preferential selfingin the freshwater mollusc Lymnaea truncatula (Gastropoda, Pulmonata). He-redity (Edinb.), 92(4): 316–322. doi:10.1038/sj.hdy.6800410.

Nicot, A., Dubois, M.-P., Debain, C., David, P., and Jarne, P. 2008. Characteriza-tion of 15 microsatellite loci in the pulmonate snail Pseudosuccinea columella(Mollusca, Gastropoda). Mol. Ecol. Resour. 8(6): 1281–1284. doi:10.1111/j.1755-0998.2007.02065.x. PMID:21586021.

Nunes, A.L., Tricarico, E., Panov, V.E., Cardoso, A.C., and Katsanevakis, S. 2015.Pathways and gateways of freshwater invasions in Europe. Aquat. Invasions,10(4): 359–370. doi:10.3391/ai.2015.10.4.01.

Pfenninger, M., Salinger, M., Haun, T., and Feldmeyer, B. 2011. Factors and pro-cesses shaping the population structure and distribution of genetic variationacross the species range of the freshwater snail Radix balthica (Pulmonata,Basommatophora). BMC Evol. Biol. 11(1): 135. doi:10.1186/1471-2148-11-135.PMID:21599918.

Posada, D. 2008. jModelTest: phylogenetic model averaging. Mol. Biol. Evol.25(7): 1253–1256. doi:10.1093/molbev/msn083. PMID:18397919.

Puurtinen, M., Knott, K.E., Suonpää, S., Nissinen, K., and Kaitala, V. 2007. Pre-dominance of outcrossing in Lymnaea stagnalis despite low apparent fitnesscosts of self-fertilization. J. Evol. Biol. 20(3): 901–912. doi:10.1111/j.1420-9101.2007.01312.x. PMID:17465901.

Rice, R. 1989. Analyzing tables of statistical tests. Evolution (N.Y.), 43(1): 223–225.doi:10.2307/2409177.

Ronquist, F., and Huelsenbeck, J.P. 2003. MrBayes 3: Bayesian phylogenetic in-ference under mixed models. Bioinformatics, 19(12): 1572–1574. doi:10.1093/bioinformatics/btg180. PMID:12912839.

Rozen, S., and Skaletsky, H. 2000. Primer3 on the WWW for general users and forbiologist programmers. In Bioinformatics methods and protocols. Edited byS. Misener and S.A. Krawetz. Methods in Molecular Biology™, Vol. 132. HumanaPress, Totowa, N.J. pp. 365–386.

Samadi, S., Roumegoux, A., Bargues, M.D., Mas-Coma, S., Yong, M., andPointier, J.-P. 2000. Morphological studies of lymnaeid snails from the humanfascioliasis endemic zone of Bolivia. J. Mollusc. Stud. 66: 31–44. doi:10.1093/mollus/66.1.31.

Standley, C.J., Prepelitchi, L., Pietrokovsky, S.M., Issia, L., Stothard, J.R., andWisnivesky-Colli, C. 2013. Molecular characterization of cryptic and sympat-ric lymnaeid species from the Galba/Fossaria group in Mendoza Province,Northern Patagonia, Argentina. Parasit. Vectors, 6: 304. doi:10.1186/1756-3305-6-304. PMID:24499569.

Toews, D.P., and Brelsford, A. 2012. The biogeography of mitochondrial andnuclear discordance in animals. Mol. Ecol. 21(16): 3907–3930. doi:10.1111/j.1365-294X.2012.05664.x. PMID:22738314.

Trouvé, S., Degen, L., Meunier, C., Tirard, C., Hurtrez-Boussès, S., Durand, P.,Guegan, J., Goudet, J., and Renaud, F. 2000. Microsatellites in the hermaph-roditic snail, Lymnaea truncatula, intermediate host of the liver fluke, Fasciolahepatica. Mol. Ecol. 9(10): 1662–1664. doi:10.1046/j.1365-294x.2000.01040-2.x.PMID:11050562.

Waldbieser, G. 1995. PCR-based identification of AT-rich tri- and tetranucleotiderepeat loci in an enriched plasmid library. Biotechniques, 19(5): 742–744.PMID:8588909.

Wilke, T., Schultheiß, R., and Albrecht, C. 2009. As time goes by: a simple fool’sguide to molecular clock approaches in invertebrates. Am. Malacol. Bull. 27:25–45. doi:10.4003/006.027.0203.

Lounnas et al. 435

Published by NRC Research Press

Can

. J. Z

ool.

Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

195.

83.9

9.4

on 0

6/06

/18

For

pers

onal

use

onl

y.