Spartina versicolor Fabre: Another case of Spartina trans-Atlantic introduction?

HAL Id: hal-01355664 h ps://hal.science/hal-01355664 Submi ed on 25 Ap 2018 HAL is a mul i-disciplina y open access a chi e o he deposi and dissemina ion o sci- en i ic esea ch documen s, whe he hey a e pub- lished o no . The documen s may come om eaching and esea ch ins i u ions in F ance o ab oad, o om public o p i a e esea ch cen e s. L’a chi e ou e e plu idisciplinai e HAL, es des inée au dépô e à la di usion de documen s scien i iques de ni eau eche che, publiés ou non, émanan des é ablissemen s d’enseignemen e de eche che ançais ou é ange s, des labo a oi es publics ou p i és. Spa ina e sicolo Fab e: Ano he case o Spa ina ans-A lan ic in oduc ion? Alex Baumel, Ma hieu Rousseau-Gueu in, C. Sapienza-Bianchi, Agnès Ga eil, N. Duong, Hélène Rousseau, Oli ie Co i on, Rachid Ami ouche, S. Sciand ello, B. Dua e, e al. To ci e his e sion: Alex Baumel, Ma hieu Rousseau-Gueu in, C. Sapienza-Bianchi, Agnès Ga eil, N. Duong, e al.. Spa ina e sicolo Fab e: Ano he case o Spa ina ans-A lan ic in oduc ion?. Biological In a- sions, 2016, 18 (8), pp.2123-2135. �10.1007/s10530-016-1128-z�. �hal-01355664� INVASIVE SPARTINA Spa ina e sicolo Fab e: Ano he case o Spa ina ans-A lan ic in oduc ion? A. Baumel .M. Rousseau-Gueu in .C. Sapienza-Bianchi .A. Ga eil .N. Duong . H. Rousseau .O. Co i on .R. Ami ouche .S. Sciand ello .B. Dua e .I. Cac¸ado . J. M. Cas illo .M. Ainouche Recei ed: 23 Ma ch 2015 / Accep ed: 20 Ma ch 2016 ÓSp inge In e na ional Publishing Swi ze land 2016 Abs ac In e con inen al in oduc ions a e wide- sp ead in he genus Spa ina, wi h impo an ecolog- ical and e olu iona y consequences. The na i e o in oduced s a us o Spa ina species is hen c i ical wi h ega d o biodi e si y assessmen , especially o ulne able Medi e anean coas line ecosys ems. Spa - ina e sicolo was i s eco ded in sou he n F ance in 1849, hen successi ely in a ious places on he Eu opean and No h-A ican Medi e anean and A lan ic coas s. This species is conside ed o be ei he a Eu opean na i e o an in asi e species in oduced om No h Ame ica which has a high mo phological simila i y o he A lan ic Ame ican species Spa ina pa ens. We pe o med ex ensi e sampling o S. e sicolo in Eu ope and No h A ica ( om na u al popula ions and he ba ium collec ions) and compa ed hese samples o o he Eu opean and Ame ican Spa ina species (including S. pa ens). Ch omosome coun s we e epo ed o he i s ime and e ealed ha S. e sicolo is e aploid (2n=4x=40). Phy- logene ic analyses based on chlo oplas and nuclea ibosomal DNA sequences did no e eal any molec- ula a ia ion wi hin S. e sicolo . In his species, a single haplo ype, ha is iden ical o one haplo ype o S. pa ens, was ound in he ou chlo oplas and he nuclea ibosomal ITS egions in es iga ed. In addi- ion, simple sequence epea ma ke s we e used and e ealed a low le el o gene ic di e si y wi hin S. e sicolo , sugges ing ha he in oduc ion o S. Gues edi o s: Alan G ay and Malika Ainouche/In asi e Spa ina. Elec onic supplemen a y ma e ial The online e sion o his a icle (doi:10.1007/s10530-016-1128-z) con ains supple- men a y ma e ial, which is a ailable o au ho ized use s. A. Baumel N. Duong Aix Ma seille Uni e si e ´, Ins i u Me ´di e ane ´en de Biodi e si e ´e d’Ecologie (IMBE, UMR CNRS, IRD, A ignon Uni e si e ´), Technopo ˆle de l’En i onnemen A bois-Me ´di e ane ´e, BP 80, 13545 Aix-en-P o ence Cedex 04, F ance M. Rousseau-Gueu in C. Sapienza-Bianchi  A. Ga eil H. Rousseau M. Ainouche (&) UMR CNRS 6553 Ecobio, OSUR (Obse a oi e des Sciences de l’Uni e s de Rennes), Uni e si e ´de Rennes 1/Uni e si e ´Eu ope ´enne de B e agne, 35042 Rennes, F ance e-mail: [email p o ec ed] M. Rousseau-Gueu in INRA, UMR 1349, Ins i u de Ge ´ne ´ ique, En i onnemen e P o ec ion des Plan es, 35653 Le Rheu Cedex, F ance O. Co i on Pla e-Fo me de Cy oge ´ne ´ ique Mole ´culai e, INRA, UMR 1349, Ins i u de Ge ´ne ´ ique, En i onnemen e P o ec ion des Plan es, 35653 Le Rheu Cedex, F ance R. Ami ouche Uni e si e ´des Sciences e de la Technologie Houa i Boumediene, Labo a oi e de Biologie e Physiologie des O ganismes, BP 32 El-Alia, 16111 Bab-Ezzoua , Alge , Alge ia 123 Biol In asions (2016) 18:2123–2135 DOI 10.1007/s10530-016-1128-z e sicolo occu ed om a na ow gene ic pool o S. pa ens om No h Ame ica. Keywo ds Co dg ass Gene ic di e si y Species s a us Medi e anean Mic osa elli es Phylogeny In oduc ion We land habi a s a e among he mos h ea ened in he Medi e anean as a consequence o in ense u baniza- ion, an h opogenic dis u bance and inc eased numbe o in asi e axa (Me ´dail and Ve laque 1997). In F ance, Medi e anean we lands a e among he habi- a s ha a e he mos colonized by in asi e species (Ve laque e al. 2002). Because su eys o biodi e si y a e gene ally poo ly coo dina ed in he Medi e anean biodi e si y ho spo (Ma ignani e al. 2014), in e ence o he na i e o in oduced plan species s a us is no i ial. This s a us is an essen ial pa ame e o biodi e si y managemen and conse a ion biology. I also ep esen s c i ical in o ma ion wi h ega d o popula ion and species e olu iona y his o y. Es ab- lishing na i e s a us o a species in a gi en egion is no an easy ask and equi es a combina ion o di e en app oaches o elucida e he o igin, mode o o ma ion and biogeog aphy o he conside ed axon. The inc eased oppo uni ies o long-dis ance human- media ed species dispe sal make hese esea ches e en mo e complex (Kowa ik 2003). In his con ex , molecula ma ke s and e olu iona y gene ics p o ide impo an insigh o ace back popula ion, species o igin and mig a ion his o y (Mansion e al. 2008; Ha dion e al. 2014). In Spa ina (co dg asses), ecu en in e con inen al in oduc ion e en s and biological in asions a e pa ic- ula ly common and well-documen ed (Daehle and S ong 1996a,b;SanLe ´on e al. 1999;Baumele al. 2001;Sanche ´z-Gullon 2001; Ay es e al. 2004;Ane al. 2007;Ainouchee al.2009; Campos e al. 2004; Lona d e al. 2010; Saa ela 2012; S ong and Ay es 2013). This g ass genus (Poaceae, Chlo idoideae) ep esen s a well- suppo ed monophyle ic lineage (Baumel e al. 2002; Fo une e al. 2007) closely ela ed o some membe s o he pa aphyle ic Spo obolus genus and Calamo il a (Pe e son e al. 2014). I is composed o abou 15 pe ennial species ha ha e di e si ied mos ly in he New Wo ld (Mobbe ley 1956). In oduc ion o species ou side hei na i e ange o e he pas 150 yea s has accele a ed di e si ica ion by acili a ing hyb idiza ion wi h na i e species, in og ession o specia ion, esul - ing in se e al supe imposed di e gen genomes ha coexis in he species cu en ly ound in he wild (Ainouche e al. 2012). The basic (haploid) ch omosome numbe in Spa ina is conside ed o be x=10 (Ma chan 1968), and all species eco ded o da e a e polyploid, anging om e aploids o dodecaploids. Molecula phylogenies om nuclea and chlo oplas DNA sequences ha e indica ed ha genus Spa ina has e ol ed h ough wo main lineages including e aploid and hexaploid species espec i ely (Baumel e al. 2002). The e aploid lineage is composed o species na i e o he New Wo ld, colonising coas al o inland sal ma shes om ei he No he n (Spa ina pa ens,Spa - ina bake i, Spa ina g acilis,Spa ina cynuso oides, Spa ina pec ina a)o Sou he n(Spa ina cilia a, Spa ina a undinacea) hemisphe es. The e aploid S. a gen inensis (syn. S. spa inae), which has a disjunc dis ibu ion in No h-Cen al Ame ica and in Sou h- Ame ica, is sis e o he hexaploid lineage. This la e clade is composed o Spa ina ma i ima,Spa ina al e ni lo a,andSpa ina oliosa, all colonizing low ma sh zones. Spa ina ma i ima, na i e o he Wes e n Eu ope and A ican A lan ic coas s, isone o he ew Old Wo ld na i e species wi h ecen axa o hyb id o igin and he con o e sial S. e sicolo (see below). Acci- den al o delibe a e in oduc ions lead o a ious hyb idiza ion e en s wi hin o be ween he e aploid and hexaploid lineages ( e iewed in Ainouche e al. 2012; S ong and Ay es 2013). In Eu ope, in oduc ions o he hexaploid S. al e ni lo a, na i e o he A lan ic Ame ican coas s and i s subsequen hyb idiza ion wi h hexaploid S. S. Sciand ello Depa men o Biological, Geological and En i onmen al Sciences, Uni e si y o Ca ania, ia Alongo 18, 95125 Ca ania, I aly B. Dua e I. Cac¸ado MARE – Ma ine and En i onmen al Sciences Cen e, Facul y o Sciences o he Uni e si y o Lisbon, Campo G ande, 1749-016 Lisbon, Po ugal J. M. Cas illo Depa amen o de Biologı ´a Vege al y Ecologı ´a, Facul ad de Biologia, Uni e sidad de Se illa, Apa ado 1095, 41080 Se ille, Spain 2124 A. Baumel e al. 123 ma i ima led o he o ma ion o wo s e ile F1 hyb ids in Sou he n England (S. x ownsendii) and in Sou h- wes F ance (S. x ney au ii). Genome duplica ion in he B i ish hyb id esul ed in he igo ous and e ile allododecaploid S. anglica (Hubba d 1968; Gue ´ne ´gou e al. 1988; G ay e al. 1990; G ay e al. 1991). This species apidly expanded in ange and sp ead na u ally o wes e n Eu opean sal ma shes. I is now in oduced in a ious con inen s, leading o a ious a emp s o con ol o e adica e he species (e.g. Hacke e al. 2001; Co e e al. 2007). In oduced S. al e ni lo a is p og essing along he wes e n A lan ic coas s o F ance and Spain (Baumel e al. 2003; Campos e al. 2004). Ano he in oduced Spa ina species in Eu ope is he na i e Sou h-Ame ican hep aploid species S. densi lo a (Fo une e al. 2008) ha is in ading Medi e anean sal ma shes o he Ibe ian peninsula (Bo olus 2006; Cas illo e al. 2008), whe e i hyb idized wi h he hexaploid S. ma i ima (Cas illo e al. 2010). In he wes e n Medi e anean, damp dep essions in dune habi a s a e colonized by Spa ina e sicolo Fab e (Fig. 1) ha g ows also in b ackish ma shes in he A lan ic Coas o he Sou hwes Ibe ian Peninsula. This species, also named Spa ina juncea o Spa ina du ieui (Che alie 1923; Sain -Y es 1932), had a con o e sial axonomic s a us. I was ini ially eco ded almos simul aneously in se e al Medi e - anean places: i s in Sou he n F ance nea Agde (Fab e 1849), hen in I aly (Pa la o e 1848–1850), Alge ia (Cosson and Maisonneu e 1867), and Po u- gal (Da eau 1897). In 1901, Ney au de ec ed his axon on he Sou hwes F ench A lan ic coas , nea A cachon (Cos e 1906). Since hen, S. e sicolo has es ablished all along he wes e n Medi e anean coas s: in Co sica (Jeanmonod and Bu de 1989) as well as on he A lan ic and Medi e anean coas s o he Ibe ian Peninsula (Sanche ´z-Gullo ´n2001). S. e sicolo was conside ed as ei he a na i e Medi e anean plan (e.g. Sanche ´z-Gullo ´n2001; Giuliano and S anisci 2010; Tison e al. 2014a,b), o an in asi e species in oduced om Ame ica (Sanz Elo za e al. 2004; Tison and de Foucaul 2014). Based on mo phological simila i ies, Mobbe ley (1956) conside ed S. e si- colo as synonymous o S. pa ens, assuming ha he Medi e anean popula ions we e in oduced om he A lan ic No h Ame ican coas whe e S. pa ens is abundan in high ma sh and dunes. Recen s udies on sal ma shes along he Spanish A lan ic coas ha e unde lined he p esence o S. pa ens in Eu ope (San Leon 1999; Page e al. 2010) and enewed in e es in he s a us o S. e sicolo . P ie o e al. (2011) examined h ee S. e sicolo indi iduals om no h- e n Spain (As u ias) using In e nal ansc ibed Space (ITS) sequences o nuclea ibosomal DNA genes, and compa ed hese sequences o hose ini ially published in genus Spa ina by Baumel e al. (2002) and Fe is e al. (unpublished). These indi iduals exhibi ed simila ITS sequence o S. pa ens, which led hese au ho s o sugges ha S. e sicolo should be conside ed as S. pa ens. Bu i canno be excluded ha a na i e Medi e anean Spa ina species exis s besides he in oduc ion o S. pa ens in Spain. Cases o c yp ic in asion ha e al eady been documen ed in he ecen his o y o Spa ina (e.g. Bo olus e al. 2015) and can be di icul o de ec (Val uena e al. 2011). Fig. 1 Spa ina e sicolo a igo ous popula ion om Vieux Salins (Hye ` es, F ance) bmi o ic ch omosomes coun e s ained wi h DAPI (2n =40), ba ep esen s 5 lm Spa ina e sicolo Fab e: Ano he case o Spa ina ans-A lan ic in oduc ion? 2125 123 In his s udy, we aim a answe ing he ollowing ques ions: Is S. e sicolo Fab e om Eu ope con- speci ic wi h S. pa ens om No h Ame ica and i so is Eu opean S. pa ens ano he case o ans-A lan ic in oduc ion in o Eu ope? As highligh ed abo e, wo ldwide in asions o Spa ina a e common phe- nomena and he Medi e anean coas may be one o he a ious places whe e Ame ican Spa ina ha e se led a e dispe sal by ships du ing he 18 h o 19 h cen u ies. Gaining insigh in o he na i e o in oduced s a us o S. e sicolo , o i s ela ionship wi h o he Spa ina species will be o c i ical impo ance in o de o be e unde s and he biogeog aphy and di e si i- ca ion o Spa ina species, as well as o de e mine he conse a ion p io i y le el and managemen policy o S. e sicolo in he Old Wo ld. To answe hese ques ions, popula ions o S. e sicolo sampled om a ious Medi e anean and A lan ic si es in Eu ope and No h-A ica (including e e ence ypes om he ba ia) a e analyzed using cy ogene ic and molec- ula (mic osa elli e, nuclea and chlo oplas DNA sequences) da a, and compa ed o No h-Ame ican Spa ina species. Ma e ials and me hods Plan ma e ial Fi y-se en indi iduals o S. e sicolo we e sampled om nume ous Medi e anean si es (n =47) and om he ba ium collec ions (n =10) (Table S1). The analyzed samples include specimens om a ious popula ions in F ance (including Co sica), I aly, Po ugal, Spain and Alge ia as well as he i s S. e sicolo plan s om Agde (F ance) disco e ed by Fab e (1849). Eigh Spa ina pa ens samples we e ob ained om he A lan ic No h Ame ican coas (Table S1). Rep esen a i es om eigh o he Spa ina species (S. a gen inensis,S. al e ni lo a,S. a undi- nacea,S. bake i,S. densi lo a,S. oliosa,S. ma i ima and S. pec ina a) and om Spo obolus c yp and us we e addi ionally in oduced in phylogene ic analyses. Ch omosome coun s The ch omosome numbe o S. e sicolo was de e - mined on mi o ic ch omosomes ob ained om wo plan s collec ed in F ance (Vieux Salins and Sain - Louis du Rhone). Mi o ic ch omosomes we e obse ed on me aphasic cells isola ed om oo ips. The oo s ips o 0.5–1.5 cm leng h we e ea ed wi h 0.04 % 8-hyd oxiquinoline o 2 h a 4 °C in he da k ollowed by 2 h a oom empe a u e o accumula e me aphases, hen ixed in e hanol-ace ic acid (3:1, / ) o 12 h a 4 °C and s o ed in e hanol 70 % a -20 °C. A e washing in 0.01 M enzyme bu e (ci ic acid- sodium ci a e pH 4.5) o 15 min, he oo s we e diges ed in a solu ion o 5 % Onozuka R-10 cellulase (Sigma) and 1 % Y23 pec olyase (Sigma) a 37 °C o 30 min. The oo ips we e hen washed wi h dis illed wa e o 30 min. Roo ips ans e ed on a slide we e squashed in a d op o 3:1 e hanol-ace ic acid ixa ion solu ion. A e ai -d ying, slides we e s ained wi h 4,6- diamidino-2-phenylindole (DAPI). Fluo escence images we e cap u ed using a CoolSnap HQ came a (Pho ome ics, Tucson, A iz) on an Axioplan 2 mic oscope (Zeiss, Obe kochen, Ge many) and anal- ysed using Me aVue TM (Uni e sal Imaging Co po a- ion, Downing on, PA). DNA isola ion, PCR ampli ica ion and DNA sequencing Genomic DNA was isola ed om 100 mg o esh (o 30 mg o he ba ium) lea es om each indi idual using he NucleoSpin Ò Plan II Ki (Mache ey– Nagel), ollowing ins uc ions p o ided by he man- u ac u e . DNA concen a ions we e es ima ed using he Nanod op Spec opho ome e ND 1000 (The mo Fische Scien i ic). Fou chlo oplas and en nuclea egions we e ampli ied. Chlo oplas sequences we e chosen among he mos a iable in e genic egions iden i ied in Poaceae (Rousseau-Gueu in e al. 2015) o wi hin Spa ina (Blum e al. 2007; Kim e al. 2013): i included he ndhC- nV, pe A-psbJ (p ime s designed om Rousseau-Gueu in e al. 2015), and he nL- nF and nT- nL in e genic egions (Tabe le e al. 1991). Nuclea egions included In e nal ansc ibed Space s (ITS) o nuclea ibosomal genes ( DNA) (Whi e e al. 1990) and nine mic osa elli e ma ke s (SSR 6, 40, 44, 72, 109, 122, 161, 172, 188) iden i ied om S. ma i ima Bac e ial A i icial Ch omosome end sequences (Fe ei a de Ca alho e al. 2013). The p ime sequences used in his s udy a e indica ed in Table S2. 2126 A. Baumel e al. 123 Ampli ica ions o chlo oplas and ITS egions we e ca ied ou using he high ideli y KOD polyme ase (Toyobo, No agen) in a o al olume o 50 ll. The eac ion mix included 1X o KOD bu e , 1.5 mM MgSO4, 0.2 mM dNTP, 0.3 lM o each p ime , 0.02 U o KOD polyme ase, and 20 ng empla e DNA. Cycling condi ions we e 94 °C o 2 min, ollowed by 32 ounds o 94 °C o 20 s, 59.5 °C o 10 s and an ex ension a 70 °C o 15 s. Chlo oplas and nuclea (ITS) PCR p oduc s we e pu i ied using he PCR Clean-up Gel ex ac ion ki (Mache ey–Nagel) and he pu i ied p oduc s we e sen o Mac ogen Eu ope (Ams e dam, Ne he lands) o di ec sequencing. Long PCR p oduc s ( om he nT- nL, nL- nF and ndhC- nV egions) we e sequenced om bo h sides. Sequences we e cleaned and e i ied isually on he ch oma og ams (no double peaks obse ed). Simple Sequence Repea (SSR) de ec ion was pe o med o 60 samples (52 Old Wo d S. e sicolo and 8 New Wo ld S. pa ens) using 9 mic osa elli e loci. Ampli ica ions o all mic osa elli es we e pe - o med in 20 lL ha con ained 10 ng DNA, 4 lLo 5X bu e , 4 mM MgCl2, 0.2 mM dNTPs, 0.4 lMo each p ime , and 0.4 uni s o Taq polyme ase (Q- Biogen) in a PTC-200 G adien The mal Cycle (MJ Resea ch), ollowing ouchdown PCR p o ocols (Miglio e e al. 2013). The 50ends o he o wa d p ime s we e labelled wi h PET o NED. The luo es- cen ly labelled PCR p oduc s we e dilu ed (1/40) and we e sepa a ed by capilla y elec opho esis, wi h a 500 bp size s anda d (LIZ500), using an ABI P ism Ò 3730xl (Applied Biosys ems) au oma ic sequence . Alleles we e sized using PEAK SCANNER 1.0 so wa e (Applied Biosys ems). Geno yping (PCR and elec opho esis) was epea ed o 8 samples o e i y he ep oducibili y o he peak pa e ns. Phylogene ic analyses The da a ma ices gene a ed o indi idual o con- ca ena ed chlo oplas egions as well as he nuclea ITS ma ix we e ob ained a e aligning all sequences using Geneious (D ummond e al. 2010) and adjus ing hem manually. These ma ices we e i s subjec ed o phylogene ic analyses using maximum pa simony. Sequence da a we e analyzed using PAUP* 4.0b10 (Swo o d 2001) wi h heu is ic sea ch and he de aul sea ch op ions. The phylogene ic analyses we e pe - o med using sequences om Spo obolus c yp and us, a closely ela ed species o Spa ina (Pe e son e al. 2014) as ou g oup. Boo s ap analyses we e pe o med wi h 1000 eplica es (Felsens ein 1985). In addi ion, hese da a ma ices we e subjec ed o Maximum Likelihood phylogene ic analyses. The bes - i ed model o sequence e olu ion o each egion (indi idual o conca ena ed) was de e mined by using JModel es (Posada 2008) implemen ed in MEGA 5.0 (Tamu a e al. 2011). Maximum likelihood analyses we e hen pe o med o each ma ix using PhyML (Guindon and Gascuel 2003), wi h 1000 eplica es o boo s ap. Mic osa elli e analyses Ou o he nine SSR loci, six we e conside ed o be eliable and we e subsequen ly analyzed. A e com- pa isons o eplica es, all dubious peaks we e emo ed. Since mos geno ypes ha e a leas h ee alleles in his polyploid species (see below), SSR ma ke s we e analyzed as bina y da a and he ma ix o allele size was con e ed in o p esence/absence da a ma ix. Genalex 6.51 so wa e (Peakall and Smouse 2006) was used o sea ch o ma ching geno ypes, bea ing e idence o iden ical clones. The sea ch o ma ching geno ype was epea ed accoun ing o one, wo o h ee allelic e o s. Allelic accumula ion cu es we e pe o med (specaccum unc ion, egan R pack- age, Oksanen e al. 2013) o ma ch allelic ichness o S. e sicolo and S. pa ens. S uc u e o gene ic di e si y was analyzed using mul i a ia e analyses o ade4 and adegene packages o R (D ay and Du ou 2007; Jomba 2008). P incipal coo dina e analysis (PcoA, dudi.pco unc ion, ade4 R package) was based on Jacca d dis ances (Jacca d 1901) compu ed on SSR p esence/absence (dis .bina y unc ion, ade4 R package). Fo ma ching samples ha ing iden ical geno ypes (clones), only one geno- ype was kep in analyses based on indi idual geno- ypes and hese we e iden i ied as ‘‘clones’’ on he PcoA plo . A disc iminan analysis was conduc ed on he co ela ion o allele p esence/absence o dis in- guish gene ic g oups acco ding o he clus e ing p ocedu e designed by Jomba e al. (2010) (DAPC analysis, adegene R package). Finally, allele equen- cies wi hin hese g oups we e used o compu e Nei dis ances (Nei 1972) and o build a Neighbo Joining ne wo k o explo e ela ionships be ween DAPC Spa ina e sicolo Fab e: Ano he case o Spa ina ans-A lan ic in oduc ion? 2127 123 gene ic g oups (nj unc ion, ape R package, Pa adis e al. 2004). Resul s Ch omosome coun s e ealed ha in bo h analyzed popula ions, S. e sicolo indi iduals ha e 2n =40 ch omosomes (Fig. 1b), indica ing ha his axon is a e aploid species. No sequence he e ogenei y was obse ed in ei he chlo oplas o nuclea (ITS) sequences. In a-genomic polymo phism migh be expec ed in polyploid nuclea genomes, bu he ITS egions (belonging o he DNA gene amily) seem o ha e unde gone conce ed e olu ion as p e iously ound in o he Spa ina species (Baumel e al. 2002; Bou e e al. 2015). Maximum Pa simony (MP) analyses we e pe o med using he ndhC- nV (601 bp), pe A-psbJ (470 bp), nL- nF (577 bp), nT- nL (631 bp), chlo oplas conca ena ed (2278 bp) o nuclea ibosomal ITS (457 bp) ma ices. These analyses esul ed in 23, 34, 20, 23, 3 and 19 equal mos pa simonious ees. In hese analyses, S. e sicolo , S. pa ens and S. bake i always belonged o he same clade (Fig. 2). Fo he mos esol ed ee co esponding o he ITS egions, hese h ee species belong o a well-suppo ed clade (99 %) and a e as posi ioned as a sis e clade o S. a undincacea and S. densi lo a (100 % boo s ap suppo ). The sequences ob ained om all S. e si- colo (including he sequences ob ained by P ie o e al. 2011) and S. pa ens samples we e iden ical, apa om a single subs i u ion obse ed in one accession o S. pa ens (Cheesequake s a e Pa k, Flo ida) o he ndhC- nV egion. S. e sicolo and S. pa ens a e closely ela ed o S. bake i, p esen ing only wo subs i u ions (one o he ndhC- nV and one o he ITS egions). Since simila ee opologies we e ob ained using Maximum Likelihood, only he MP ees a e p esen ed he e. Thi y-se en alleles we e eco ded o e he six SSR loci. All SSR geno ypes we e he e ozygous wi h mos ly 3 o 4 alleles pe locus. Wi hin S. e sicolo , sea ch o ma ching geno ypes e ealed 35 geno ypes among 52 samples, 31 being unique and 4 being epea ed om 4 o 11 imes. These iden ical geno ypes a e e e ed as ‘‘clones’’. Accoun ing o one, wo o h ee allele e o s we ound 41, 44 and 50 ma ching geno ypes in S. e sicolo , whe eas no ma ching geno ypes we e ound among he 8 samples o S. pa ens e en accoun ing o 3 allele e o s. Accumu- la ion cu e (Fig. 3) accoun ing o he unequal sampling be ween S. e sicolo (n =52) and S. pa ens (n =8) e ealed highe allelic di e si y in S. pa ens han in S. e sicolo : i.e. o 6 samples 32 alleles we e encoun e ed in S. pa ens agains 26 in S. e sicolo (Fig. 3). The PcoA analysis based on Jacca d dis ances compu ed be ween indi idual geno ypes (Fig. 4) e ealed ha he main s uc u e is due o di e en ia ion be ween 6 ou o 8 S. pa ens geno ypes. The o he 2 S. pa ens samples a e mo e simila o S. e sicolo geno ypes. He ba ium specimens a e sca e ed among S. e sicolo geno ypes, excep one (collec ed in Ca non, F ance, in 1880 by Jou e) ha is g ouped wi h S. pa ens (H6, Fig. 4). The he ba ium specimen collec ed by Fab e in (1849)inAgdehasoneo he geno ypes eco ded on many indi iduals (‘‘ma ching geno ypes’’ see Ma e ials and me hods) and iden i ied as clone ‘‘b’’ (Fig. 4). This clone is ound in F ance, Basque a ea, I aly and Sicily. Acco ding o he DAPC analysis (Fig. 4) and NJ ne wo k (Fig. 5), he SSR geno ypes we e op imally clus e ed in six g oups. The gene ic clus e numbe 4 ( ep esen ed in g een, Figs. 5, 6) was composed o Eu opean samples om F ance (na u al popula ions and mos he ba ium samples), I aly, Co sica, and No h Spain. The indi iduals om he sou he n Ibe ic Peninsula (sou h o Spain and Po ugal) and Alge ia a e g ouped in wo (1 and 6) closely ela ed clus e s. Th ee gene ic clus e s could be dis inguished in he Ame ican samples o S. pa ens:(1) Mexico-Flo ida-Delawa e (2) New Je sey and (3) New- Je sey Hampshi e. The he ba ium sample om F ance sampled in Ca non (He aul ) was assigned in a g oup wi h S. pa ens (Mexico-Flo ida-Delawa e g oup). c Fig. 2 Molecula phylogeny o Spa ina based on chlo oplas (andhC- nV: 601 bp; bpe A-psbJ: 470 bp, c nL- nF: 577 bp, d nT- nL: 631 bp, econca ena ed sequences: 2278 bp) o nuclea ibosomal ITS sequences (457 bp) using he maximum pa simony me hod. Fo each phylogeny, one o he equally pa simonious ees ha is opologically iden ical o he 50 % majo i y- ule consensus ee (pe A-psbJ: 34 equally pa simonious ees; conca ena ed chlo oplas sequences: h ee equally pa simo- nious ees; ITS: 19 equally pa simonious ees) o he 50 % majo i y- ule consensus ee (ndhC- nV: 23 pa simonious ees; nL- nF: 20 pa simonious ees) is p esen ed. The boo s ap pe cen ages (1000 eplica es) a e shown in bold abo e he b anches and he numbe o changes is indica ed below. The ee is oo ed using ei he Spo obolus c yp and us o So ghum bicolo 2128 A. Baumel e al. 123 5 S. bake i S. pa ens (New Je sey, USA) S. e sicolo (Ba ayo, As u ias, Spain) S. pec ina a S. a undinacea S. densi lo a S. al e ni lo a S. ma i ima Spo obolus c yp and us S. oliosa S. pa ens (Tamaulipas, Mexico) S. pa ens (New Hampshi e, USA) S. pa ens (Cheesequake s a e, New Je sey, USA) S. pa ens (Flo ida Deso o Pa k, Flo ida, USA) S. pa ens (Delawa e, USA) S. pa ens (Bay iew a enue, New Je sey, USA) S. e sicolo (Ped ena, Can ab e, Spain) S. e sicolo (O u , As u ias, Spain) S. e sicolo (Pan ano, Sicily, I aly) S. e sicolo (Boudigou de To eilles, Py énées o ien ales, F ance) S. e sicolo (Kouali, Tipaza, Alge ia) S. e sicolo (Les A esquie s, Hé aul , F ance) S. e sicolo (Po Sain Louis, Bouches du Rhône, F ance) S. e sicolo (E ang des Pesquie s, Va , F ance) 99 70 16 16 6 3 1 21 1 1 7 9 8 2 100 6 93 92 5 F S. e sicolo (P ie o e al. 2011; JN133292) S. e sicolo (P ie o e al. 2011; JN133290) S. e sicolo (P ie o e al. 2011; JN133291) 99 80 4 22 3 3 2 5 27 3 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 3 S. bake i S. pa ens (New Je sey, USA) S. pec ina a S. a undinacea S. densi lo a S. al e ni lo a S. ma i ima So ghum bicolo S. pa ens (Cheesequake s a e, New Je sey, USA) S. pa ens (Flo ida Deso o Pa k, Flo ida USA) S. pa ens (Delawa e, USA) S. e sicolo (Pan ano, Sicily, I aly) S. e sicolo (Boudigou de To eilles, Py énées o ien ales, F ance) S. e sicolo (Kouali, Tipaza, Alge ia) S. e sicolo (Les A esquie s, Hé aul , F ance) S. e sciolo (Po Sain Louis, Bouches du Rhône, F ance) S. e sciolo (Vieux salins, Va , F ance) S. e sicolo (Guadiana i e , Andalusia, Spain) S. e sciolo (Banda bianca, Co sica, F ance) S. e sicolo (Tagus es ua y, Po ugal) S. e sicolo (G ande mo e, Hé aul , F ance) S. e sciolo (E ang des Pesquie s, Va , F ance) S. pa ens (Chissahowi zka, Flo ida, USA) A S. bake i S. pa ens (New Je sey, USA) S. pec ina a S. a undinacea S. densi lo a S. al e ni lo a S. ma i ima Spo obolus c yp and us S. pa ens (Cheesequake s a e, New Je sey, USA) S. pa ens (Flo ida Deso o Pa k, Flo ida, USA) S. pa ens (Delawa e, USA) S. e sciolo (Bou digou de To eilles, Py énées o ien ales, F ance) S. e sicolo (Kouali, Tipaza, Alge ia) S. e sicolo (Les A esquie s, Hé aul , F ance) S. e sicolo (Po Sain Louis, Bouches du Rhône, F ance) S. e sicolo (Vieux salins, Va , F ance) S. e sicolo (Guadiana i e , Andalusia, Spain) S. e sicolo (Banda bianca, Co sica, F ance) S. e sicolo (Tagus es ua y, Po ugal) S. e sicolo (G ande mo e, Hé aul , F ance) S. e sicolo (E ang des Pesquie s, Va , F ance) S. pa ens (Chissahowi zka, Flo ida, USA) S. pa ens (Tamaulipas, Mexico) S. pa ens (New Hampshi e, USA) S. pa ens (Bay iew a enue, New Je sey, USA) S. e sicolo (Ped ena, Can ab e, Spain) S. e sciolo (O u , As u ias, Spain) S. e sicolo (Pan ano, Sicily, I aly) S. e sicolo (Ba ayo, As u ias, Spain) S. e sicolo (Eb o i e , Andalusia, Spain) S. e sicolo (Aspe illo dunes, Andalusia, Spain) S. e sicolo (Roche, Andalusia, Spain) S. e sicolo (Pan ano, Sicily, I aly) S. oliosa 1 64 1 1 1 1 1 1 5 3 65 96 B S. bake i S. pa ens (Tamaulipas, Mexico) S. pec ina a S. al e ni lo a Spo obolus c yp and us S. pa ens (New Je sey, USA) S. pa ens (New Hampshi e,USA) S. pa ens (Cheesequake s a e, New Je sey, USA) S. pa ens (Flo ida Deso o Pa k, Flo ida, USA) S. e sicolo (Aspe illo dunes, Andalusia, Spain) S. e sicolo (Roche, Andalusia, Spain) S. e sicolo (Pan ano, Sicily, I aly) S. e sicolo (Bou digou de To eilles, Py énées o ien ales, F ance) S. e sicolo (Kouali, Tipaza, Alge ia) S. e sicolo (Banda bianca, Co sica, F ance) S. oliosa S. ma i ima S. densi lo a S. g acilis S. a undinacea 67 3 1 1 2 7 4 3 2 2 3 88 C S. bake i S. pec ina a S. a undinacea S. densi lo a S. al e ni lo a S. ma i ima Spo obolus c yp and us S. pa ens (Cheesequake s a e, New Je sey, USA) S. e sciolo (Bou digou de To eilles, Py énées o ien ales, F ance) S. e sicolo (Kouali, Tipaza, Alge ia) S. e sicolo (Les A esquie s, Hé aul , F ance) S. e sicolo (Po Sain Louis, Bouches du Rhône, F ance) S. e sicolo (Vieux salins, Va , F ance) S. e sicolo (Guadiana i e , Andalusia, Spain) S. e sicolo (Tagus es ua y, Po ugal) S. e sicolo (G ande mo e, Hé aul , F ance) S. e sicolo (E ang des Pesquie s, Va , F ance) S. e sicolo (Ped ena, Can ab e, Spain) S. e sciolo (O u , As u ias, Spain) S. e sicolo (Pan ano, Sicily, I aly) S. e sicolo (Eb o i e , Andalusia, Spain) S. e sicolo (Aspe illo dunes, Andalusia, Spain) S. e sicolo (Roche, Andalusia, Spain) S. e sicolo (Pan ano, Sicily, I aly) S. oliosa 1 1 1 2 2 1 1 2 4 9 53 91 94 D S. bake i S. pa ens (Cheesequake s a e, New-Je sey USA) S. e sicolo (Bou digou de To eilles, Py énées o ien ales, F ance) S. pec ina a S. a undinacea S. densi lo a S. al e ni lo a S. ma i ima So ghum bicolo 100 85 67 66 54 68 85 12 5 6 9 3 3 3 3 2 1 1 1 0 E Spa ina e sicolo Fab e: Ano he case o Spa ina ans-A lan ic in oduc ion? 2129 123 The S. pa ens clus e being he mos simila o S. e sicolo (NJ ne wo k Fig. 6) is obse ed all along he A lan ic No h Ame ican coas om Mexico o Delawa e. The clus e 4 which has mos o he S. e sicolo samples, and he he ba ium sample collec ed by Fab e, a e he mos simila o S. pa ens acco ding o he NJ ne wo k (Fig. 6). Discussion Ou esul s e eal ha S. e sicolo is a e aploid species wi h 40 ch omosomes, as ound in S. pa ens (Ma chan 1968) and no Eu opean o No h-A ican popula ions analyzed can be di e en ia ed gene ically om No h Ame ican S. pa ens samples as hey exhibi simila DNA ITS and cpDNA sequences. In his monog aph o Spa ina, Mobbe ley (1956) s essed he mo phological simila i ies be ween hese wo axa, al hough se e al pheno ypes we e desc ibed o S. pa ens (Mobbe ley 1956). Ou esul s suppo he hypo hesis ha all Eu opean and A ican popula ions o S. e sicolo a e in ac No h Ame ican S. pa ens in oduced be o e o a he beginning o he nine- een h-cen u y. Al hough some mic osa elli e a ia ion was de ec ed be ween S. pa ens and S. e sicolo , only ew geno ype di e ences we e obse ed. Gene ic di e ences ega d- ing mic osa elli e alleles would mos likely esul om in aspeci ic gene ic di e si y in S. pa ens- e sicolo popula ions; his is suppo ed by gene ic simila i ies be ween some No h-Ame ican S. pa ens and S. e sicolo samples as can be seen on PcoA esul s (Fig. 4). The in oduc ion o igin could be in he a eas co e ed by he gene ic clus e 4 (Fig. 5), i.e. F ance, No h Spain o I aly because he co esponding geno- ype is he mos simila o S. pa ens in he NJ ne wo k (Fig. 6). This pa e n is also clea in he PcoA analysis (Fig. 4). The S. e sicolo samples om Po ugal and Sou h Spain, o Alge ia a e ei he de i ed om his in oduc ion o esul ed om a second in oduc ion. Spa ina pa ens is a highly a iable hizoma ous species, exhibi ing high ecological ampli ude along he A lan ic coas o No h Ame ica om Canada o Cen al Ame ica, colonizing high sal ma sh zones, beaches and sand dunes, wi h a iable seed se (Silande and An ono ics 1979). Allozyme s udies in na i e S. pa ens popula ions (Silande 1984) e ealed impo an polymo phism wi h a dec eased ole o ege a i e ep oduc ion om dune o ma sh habi a s. In con as mic osa elli e geno yping e ealed educed gene ic di e si y in S. e sicolo compa ed o No h Ame ican S. pa ens samples (Fig. 3), which is consis- en wi h a gene ic bo leneck ollowing in oduc ion in Eu ope oge he wi h he p edominan clonal p opaga ion o he in oduced plan s. Indeed we ound 17 ma ching SSR geno ypes wi hin S. e sicolo bu his numbe inc eased o 41 when accoun ing o one allele e o indica ing ha gene ic a ia ion wi hin S. e sicolo could be mainly o somaclonal a ia ion. P elimina y su eys in S. e sicolo popula ions e ealed s e ile pollen (R. Ami ouche, unpublished da a), which is in ag eemen wi h he obse a ion ha his axon a ely p oduces seeds (Fab e 1849; San Leon e al. 1999; Tison e al. 2014; ou pe sonal obse a ions). Fu he sampling and phylogeog aphic analyses a e needed in he na i e egion o S. pa ens o be e documen he his o y o his axon and o iden i y he p ecise popula ions ha we e in oduced in he Medi e anean. Al hough a ious s udies ha e documen ed dis ibu ion, ecology and plas ici y o S. pa ens in No h Ame ica (e.g. F asco and Good 1982; Bu dick and Mendelssohn 1987; Bu dick e al. 1989; Foo e and Reynolds 1997; Lona d e al. 2010), e y ew s udies ha e documen ed gene ic di e si y in S. pa ens (e.g. Wu 2012) and he e is a g ea need o de elop DNA-based analyses a he genome le el in he na i e ange o his species, which plays an impo an ecological ole, p e en ing coas al e osion and being used in dune es o a ion. Samples Alleles 123456 20 25 30 35 40 S. pa ens Fig. 3 Ra e ac ion cu es o allelic di e si y o he se en SSR loci 2130 A. Baumel e al. 123