P omiscuous Specia ion wi h Gene Flow in Sil e side
Fish Genus
Odon es hes
(A he ini o mes,
A he inopsidae) om Sou h Wes e n A lan ic Ocean
Basins
G aciela Ga cı
´a
1
*,Ne
´s o Rı
´os
1
, Ve o
´nica Gu ie
´ ez
1
, Jo ge Gue a Va ela
2
, Ca men Bouza Fe na
´ndez
2
,
Bele
´nGo
´mez Pa do
2
, Paulino Ma ı
´nez Po ela
2
1Seccio
´n Gene
´ ica E olu i a, Facul ad de Ciencias, UdelaR, Mon e ideo, U uguay, 2Depa amen o de Gene
´ ica, Facul ad de Ve e ina ia, Campus de Lugo, Uni e sidad de
San iago de Compos ela, Lugo, Spain
Abs ac
The p esen pape in eg a es phylogene ic and popula ion gene ics analyses based on mi ochond ial and nuclea molecula
ma ke s in sil e sides, genus Odon es hes, om a non-sampled a ea in he SW A lan ic Ocean o add ess species
disc imina ion and o de ine Managemen s Uni s o sus ainable conse a ion. All phylogene ic analyses based on he COI
mi ochond ial gene we e consis en o suppo he monophyly o he genus Odon es hes and o include O. a gen inensis,O.
pe ugiae-humensis and some O. bona iensis haplo ypes in a basal poly omy con o ming a majo de i a i e clade.
Mic osa elli es da a e ealed somewha highe gene ic a iabili y alues in he O. a gen inensis-pe ugia popula ions han in
O. bona iensis and O. pe ugia-humensis axa. Con as ing popula ion gene ics s uc u ing eme ged om mi ochond ial and
mic osa elli es analyses in hese axa. Whe eas mi ochond ial da a suppo ed wo majo g oups (O. a gen inensis-pe ugia-
humensis s. O. bona iensis-pe ugiae-humensis popula ions), mic osa elli e da a de ec ed h ee majo gene ic en i ies
ep esen ed by O. bona iensis,O. pe ugiae-humensis and an admix u e o popula ions belonging o O. a gen inensis-pe ugiae
espec i ely. The e o e, he s a COI poly omy in he ee opology in ol ing hese axa could be in e p e ed by se e al
hypo he ic scena ios such as he exis ence o sha ed ances al polymo phisms, incomple e lineage so ing in a adia ing
specia ion p ocess and/o e icula ion e en s. P esen indings suppo ha p omiscuous and ecen con ac be ween
incipien species sha ing asymme ic gene low exchanges, blu s axa bounda ies yielding complica ed axonomy and
Managemen Uni s delimi a ion in sil e side genus Odon es hes om SW A lan ic Ocean basins.
Ci a ion: Ga cı
´aG,Rı
´os N, Gu ie
´ ez V, Va ela JG, Bouza Fe na
´ndez C, e al. (2014) P omiscuous Specia ion wi h Gene Flow in Sil e side Fish Genus Odon es hes
(A he ini o mes, A he inopsidae) om Sou h Wes e n A lan ic Ocean Basins. PLoS ONE 9(8): e104659. doi:10.1371/jou nal.pone.0104659
Edi o : Vale io Ke maie , Ins i u e o Biochemis y and Biology, Ge many
Recei ed May 29, 2014; Accep ed July 11, 2014; Published Augus 15, 2014
Copy igh : ß2014 Ga cia e al. This is an open-access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion License, which pe mi s
un es ic ed use, dis ibu ion, and ep oduc ion in any medium, p o ided he o iginal au ho and sou ce a e c edi ed.
Da a A ailabili y: The au ho s con i m ha all da a unde lying he indings a e ully a ailable wi hou es ic ion. All ele an da a a e wi hin he pape and i s
Suppo ing In o ma ion iles.
Funding: This esea ch ecei ed inancial suppo om he p ojec Fondo Ma ı
´aVin
˜as_2009_1_2793 (FMV_2009_1_2793_P ojec ) g an ed by he Agencia
Nacional de In es igacio
´n e Ino acio
´n (ANII) o U uguay. The unde s had no ole in s udy design, da a collec ion and analysis, decision o publish, o p epa a ion
o he manusc ip .
Compe ing In e es s: The au ho s ha e decla ed ha no compe ing in e es s exis .
* Email: [email p o ec ed]
In oduc ion
The New Wo ld p esen s mul iple examples o a he inid species
locks o adap i e adia ions a ising om habi a ansi ions
[1,2,3]. Sil e side ish om Sou h Ame ica cons i u e a exci ing
model o unde s and he scena io o ish specia ion d i en by
di e gen na u al selec ion [1,2].
The sil e side genus Odon es hes includes 20 nominal species [4]
dis ibu ed in ma ine, es ua ine and eshwa e en i onmen s o
opical and empe a e egions in Sou h Ame ica [5]. Mos
Odon es hes species co-occu in he same habi a s and hey a e
cha ac e ized by a g ea mo phological homogenei y [6]. The low
mo phological di e gence be ween species and he high me is ic
plas ici y wi hin species oge he wi h he endency o local
popula ions o o m mic o-geog aphic habi a associa ions had led
o complica ed axonomy among sil e side axa [7].
Among eshwa e ep esen a i e species, wo o hem O.
bona iensis and O. ha che i a e endemic o i e s and lakes loca ed
eas o he Andes in sub opical and empe a e a eas [8]. The
dis ibu ion o hese species was o iginally allopa ic: O. ha che i
occu ing in he Sou h (Pa agonia), whe eas O. bona iensis
occupying Cen al and No he n A gen ina, Sou h B azil and
Pa aguay. The occu ence o he spon aneous hyb idiza ion
be ween bo h species in a communal labo a o y ank has been
epo ed [9].
On he o he hand, ma ine sil e sides gene ally ha e simila li e
his o y s a egies, occu ing in la ge numbe s in semi-isola ed
popula ions in es ua ies and coas al lagoons [10,11,12,1]. Ten
species o Odon es hes a e endemic o a chain o small shallow lakes
sp ead along he Sou h Wes e n A lan ic Ocean coas al plain
[13,14]. Among hem, in Pa os Lagoon es ua y and i s adjacen
ma ine coas al a ea occu s O. a gen inensis and O. incisa, whe eas
PLOS ONE | www.plosone.o g 1 Augus 2014 | Volume 9 | Issue 8 | e104659
in he eshwa e habi a s o Pa os-Mi im lagoon sys em can be
ound O. bona iensis, O. humensis,O. e opinnis and O. a .
pe ugiae.
Mos Odon es hes species ep esen economically impo an
esou ces o a isanal and ec ea ional ishe ies in Sou h Ame ica
and pa icula ly O. bona iensis shows a g ea po en ial o
aquacul u e de elopmen [15].
The iden i ica ion o incipien ecological species ep esen s an
oppo uni y o in es iga e he cu en e olu iona y p ocess whe e
adap i e di e gence and ep oduc i e isola ion a e associa ed [2].
Behe ega ay and Sunnucks [2] ound ha niche di e gence due o
es ua ine coloniza ion by ma ine sil e side ish led o isola ion by
adap a ion and specia ion in he p esence o high gene low, one o
he mos con incing epo s o pa apa ic specia ion in aqua ic
o ganisms om he Sou he n Hemisphe e. Behe ega ay e al. [16]
explo ed he ole o adap i e di e si ica ion and ecen sea-le el
changes as e olu iona y d i e s in he O. pe ugiae species complex
which comp ises se e al allopa ic and sympa ic mo pho ypes
ound in he lakes and i e s o sou he n B azil, U uguay and
no he n A gen ina [17]. Mos mo pho ypes ha e unce ain
axonomic s a us and a e endemic o he as sys em o lakes o
he Coas al Plain o Rio G ande do Sul S a e (CPRS), sou he n
B azil [17]. Behe ega ay e al. [16] pe o med a phylogeog aphic
econs uc ion o adia ions in he Sou h Ame ican coas al
eshwa e O. pe ugiae species complex, and also epo ed some
o he mos apid specia ion a es o a e eb a e g oup.
Gene low among hyb idizing species wi h incomple e ep o-
duc i e ba ie s blu s species bounda ies, while selec ion unde
he e ogeneous local ecological condi ions o along s ong g adien s
may coun e ac his endency [18]. Thus, phylogeog aphic
app oach p o ides a aluable amewo k o iden i y signa u es o
di e gen na u al selec ion associa ed wi h ecological di e gence
and he possible occu ence o e icula ion e en s among
incomple e ep oduc i ely isola ed axa.
In his s udy we implemen a phylogeog aphic analysis based on
m DNA coding sequences (cy och ome oxidase subuni I, COI)
and en mic osa elli e loci o access in he species bounda ies and
o es possible e icula ion and in og ession e en s among
Odon es hes axa om he SW A lan ic Ocean, he Rı
´odela
Pla a es ua y and in he U uguay Ri e basins. A he same ime
his in o ma ion will con ibu e o a long- e m success o
Managemen Uni s o sus ainable conse a ion o hese axa in
ishe ies and aquacul u e.
Ma e ials and Me hods
Sample collec ion and DNA ex ac ion
All sampling p o ocols o his scien i ic s udy we e app o ed by
CNEA (Comisio´n Nacional de Expe imen acio´n Animal) om
U uguay.
A o al o 163 indi iduals o Odon es hes om 20 sampling si es
h ough h ee majo egions, he Rı
´o de la Pla a (RP) es ua y
(N = 45), Lowe U uguay and Neg o i e (UNR) basins (N = 23)
Figu e 1. Dis ibu ion map o 20 sampling si es h ough h ee majo a eas, lowe U uguay and Neg o i e (UNR) basins, he Rı
´ode
la Pla a (RP) es ua y, and associa ed coas al lagoons and si es om SWA lan ic Ocean (AC) in
Odon es hes
as ollows: UNR- Las
Can
˜as beach (CA
B
), Yagua e e s eam (YA
S
), Pa o
´n s eam (PV
S
), Baygo ia dam (BA
D
), Rinco
´n del Bone e dam (RB
D
), Ansina own
(AN
T
); RP- Buceo Po (B
P
), Ha che y and Ca asco lake (CA
L
), Pando s eam (PA
S
), Pina beach (PN
B
), Solis Chico s eam (SC
S
); Solis
G ande s eam (SG
S
); Pi iapolis beach (PR
B
), Sauce Lagoon (S
L
), Chascomus Lagoon (CH
L
), A gen ina; AC-, Ga zo
´n Lagoon (G
L
),
Rocha Lagoon (R
L
), Cas illos Lagoon (C
L
), Valizas s eam (VA
S
), Na ional Ins i u e o Fishe ies Resea ch and De elopmen (INIDEP),
A gen ina.
doi:10.1371/jou nal.pone.0104659.g001
P omiscuous Specia ion Sil e side Fish Odon es hes SW A lan ic Ocean
PLOS ONE | www.plosone.o g 2 Augus 2014 | Volume 9 | Issue 8 | e104659
and A lan ic coas (AC) si es o SWA Ocean (N = 95) we e
included in he p esen s udy. All hese samples we e p ima ily
asc ibed o O. bona iensis (Ob), O. pe ugiae species complex (Op),
O. humensis (Oh), O. a gen inensis (Oa) and only wo specimens
om O. incisa (Oi) acco ding o Dye [19] mo phological
diagnosis. Tissue samples we e ob ained om a isanal gillne s
ishe ies ope a ing in hese a eas du ing 2006–2012. The sampled
a eas a e shown in Figu e 1 and Appendix S1. Sample codes a e
as ollows: collec ing si e name and he co esponding en i on-
men s in lowe case (i.e.: S = s eam, P = po , B = beach, L =
lagoon o lake, D = dam, T = own). Tissues o he ouche
specimens we e deposi ed in he collec ion o he E olu iona y
Gene ics Sec ion in he Facul y o Sciences, Uni e si y o he
Republic, Mon e ideo, U uguay.
Genomic DNA o sac i iced specimens was isola ed om muscle
issue ( ixed in e hanol 95%) using sodium chlo ide p o ein
p ecipi a ion, ollowed by e hanol p ecipi a ion modi ied om
Med ano e al. [20].
PCR ampli ica ions and sequencing o he mi ochond ial
COI gene
A agmen o 650 bp om he COI gene was ampli ied using
FishF2 and FishR1 p ime s [21]. Reac ion olume was 10 mL
con aining 10X supplied bu e , 0.25 mM MgCl
2
, 0.2 mM o each
dNTP (10 mM), 0.25 mM o each p ime (10 mM), 0.1 uni s o Taq
DNA polyme ase (In i ogen) and app oxima ely 100 ng/ul o
empla e DNA. Cycling condi ions consis ed o one ini ial
dena u a ion a 94uC o 5 min ollowed by 35 cycles o 94uC
o 30 s, 52uC o 30 s, 72uC o 1 min and a inal ex ension o
72uC o 10 min.
Ampli ied COI p oduc s we e sequenced di ec ly on bo h
s ands in a Pe kin-Elme ABI P ism 377 Au oma ed Sequence
(MACROGEN, Seoul, Ko ea). Sequence alignmen s we e pe -
o med using Clus al X 1.8 [22].
S a is ical analyses o sequences om COI da a se
Co ec ed es ima es o pai wise sequence di e gence we e
ob ained using Kimu a’s [23] wo-pa ame e algo i hm (K2P)
implemen ed in MEGA 5.0 [24]. Wi hin a popula ion, DNA
polymo phism was measu ed by calcula ing he p opo ion o
seg ega ing si es (S), he haplo ype di e si y (h) [25], and he
nucleo ide di e si y (p) [25] wi h ARLEQUIN 3.11 [26] and
DnaSP e sion 4.50 [27] p og ams. Tajima’s [28] and Fu’s [29]
es s implemen ed in DnaSP 4.50 [27] we e pe o med o check
he mu a ion/d i equilib ium and any depa u e om neu ali y.
Signi icance o Fu’s Fs [29] and Tajima’s D[28] alues was
e alua ed using he coalescen algo i hm compa ing he obse ed
alue wi h a null dis ibu ion gene a ed by 10,000 eplica es, and
gi ing an empi ical popula ion sample size and he obse ed
numbe o seg ega ing si es.
Phylogene ic analysis and di e gence ime es ima es o
he mi ochond ial gene
The phylogeog aphic ela ionships among mi ochond ial COI
haplo ypes in Odon es hes popula ions om he sampled a ea we e
assessed by using wo di e en me hodologies. A non-model based
me hod (MP, maximum-pa simony) was implemen ed in PAUP*
4.0b10 [30] ollowing an equally weigh ed MP analysis using
heu is ic sea ch (MULPARS op ion, s epwise addi ion, ee-
bisec ion- econnec ion [TBR] b anch swapping, 100 eplica es).
A s ic consensus be ween i al ees was compu ed o econcile
equally pa simonious opologies. The deg ee o con idence
assigned o nodes in he ees was assessed by boo s apping wi h
500 eplica es.
On he o he hand, wo model based app oaches we e also
used, i.e., maximum-likelihood (ML) and Bayesian in e ence (BI),
implemen ed in PAUP* 4.0b10 [30] and BEAST .1.5.4 [31],
espec i ely.
In ML and BI analyses, he bes - i ed nucleo ide subs i u ion
model o each da a se was de e mined in Model es .3.7 [32]
based on he Akaike in o ma ion c i e ion [33], which simul a-
neously compa es mul iple nes ed o non-nes ed models. In he
COI da a se among he 56 models o nucleo ide subs i u ion, he
bes i was he HKY+Cmodel [34] wi h gamma dis ibu ion (C).
The gamma dis ibu ion shape pa ame e alue was 0.18. The
likelihood sco es es ima ed o hese models we e used as he p io
se ings o he ML analysis in he da a se (2lnL = 21602.50).
Heu is ic sea ch (again wi h 100 eplica es o s epwise addi ion
and TBR b anch swapping) in ML analyses was implemen ed in
PAUP* 4.0b10 [30]. The obus ness o he nodes was de e mined
a e 1,000 boo s apping eplica es as implemen ed in PhyML 3.0
(h p://a gc.li mm. /phyml), acco ding o he algo i hm de el-
oped by Guindon e al. [35]. In his case, he NNI (a as nea es
neighbou edge in e change sea ch) swapping algo i hm op ion
was implemen ed. Nonpa ame ic boo s ap alues abo e 75%
we e conside ed o be obus suppo o clades [36].
All ees we e oo ed by means o an ou g oup c i e ion using
sequences o O. egia,O. incisa,O. smi i,O. ha che y and O.
pla ensis and a mo e dis an ly axon A he ina hepse us e ie ed
om he GenBank.
Fo he da a se , di e gence ime o nodes and he age o he
mos ecen common ances o ( MRCA) we e es ima ed wi h he
BEAST .1.5.4 so wa e [31]. This p og am pe o ms Bayesian
s a is ical in e ences o pa ame e s by using MCMC (Mon e Ca lo
Ma ko chain) as a amewo k. Inpu iles we e gene a ed wi h
Beau i .1.5.4 [31] assuming unco ela ed logno mal ees and a
Yule specia ion p ocess as p io in o ma ion. The nucleo ide
subs i u ion model and i s pa ame e alues we e selec ed
acco ding o he Model es .3.7 [32] esul s. An unco ela ed
logno mal elaxed molecula clock, which allows a e a ia ion
among lineages, was implemen ed using an es ima ed a e o
mi ochond ial genome o 0.023 [2]. We ca ied ou wo
independen uns o 10 million gene a ions. T ees and pa ame e s
we e sampled e e y 1,000 i e a ions, wi h a bu n in o 10%.
Resul s o each un we e isualized in he T ace .1.5 p og am
[37] o ensu e ha s a iona i y has been achie ed and ha
con e gence has been eached. Each analysis was epea ed many
imes o op imize he ope a o s o pa ame e s un il no sugges ion
message appea ed in he log ile. The iming o clade di e gence
and he MRCA we e es ima ed in million yea s ago (Mya) wi h a
mean and a 95% HPD (lowe and uppe ). Pos e io p obabili ies
and he maximum c edibili y ee we e calcula ed using he
T eeAnno a o .1.5.4 so wa e [31].
AMOVA, isola ion by dis ance and his o ical demog aphy
To de e mine he gene ic s uc u e o Odon es hes samples he
a iance componen s among hie a chical pa i ions in he da ase
we e assessed by Analysis o Molecula Va iance (AMOVA) [38].
The Euclidean me ic o Exco ie e al. [38] was used o cons uc
he pai wise dis ances ma ix. The gene ic a ia ion was
pa i ioned in o h ee componen s, i.e., among g oups (F
CT
),
among popula ions wi hin g oups (F
SC
), and among indi iduals
wi hin popula ions (F
ST
), a e dis ega ding ei he hei o iginal
popula ions o hei g oups. Fo bo h molecula ma ke s,
popula ions we e asc ibed o h ee majo sampling a eas such as
A lan ic coas (AC), Rio de la Pla a (RP) and U uguay and Neg o
P omiscuous Specia ion Sil e side Fish Odon es hes SW A lan ic Ocean
PLOS ONE | www.plosone.o g 3 Augus 2014 | Volume 9 | Issue 8 | e104659
i e basins (UNR), and di e en g ouping hypo heses o he
popula ions we e es ed. The signi icance o he obse ed F-
s a is ics was es ed using he null dis ibu ion gene a ed om
3,000 non-pa ame ic andom pe mu a ions o he da a ma ix
a iables and P- alues we e adjus ed wi h sequen ial Bon e oni
co ec ions o mul iple compa isons [39].
Rela ionships and geog aphical dis ibu ion o he haplo ypes
we e analysed in he haplo ype ne wo k cons uc ed wi h
NETWORK . 4.6.0.0 (h p://www. luxus-enginee ing.com/
sha ene .h m), which implemen s he median-joining me hod, in
he absence o ecombina ion [40]. The ne wo k was op imized
using maximum pa simony c i e ion.
Popula ion subdi ision and he le el o gene ic isola ion among
sampling si es we e measu ed assuming an in ini e si es model
[41]. Pai wise es ima es F-s a is ics we e calcula ed in ARLE-
QUIN 3.11 [26].
To de e mine o wha ex en he geog aphic dis ance could
explain he gene ic di e en ia ion among loca ions, a es o
isola ion by dis ance was pe o med using he Man el es [42]. In
his case, his es de e mines i he e is a signi ican co ela ion
be ween he geog aphic dis ance ma ix ( ep esen ed by he
minimum coas line o i e con ou dis ance in kilome e s) and he
pai wise Fs ma ix be ween collec ing si es. The signi icance o he
Z alue (Man el coe icien ) was calcula ed using andom
pe mu a ion p ocedu es implemen ed in he Man el Non-
pa ame ic Tes Calcula o 2.0 [43]. S a is ical signi icance was
accessed h ough 1,000 pe mu a ions.
To assess o he his o ical demog aphy o Odon es hes we
compa ed he obse ed equency dis ibu ion o pai wise
nucleo ide di e ences among haplo ypes (i.e., misma ch dis ibu-
ion) in ela ion o he expec ed unde a sudden popula ion
expansion model [44] implemen ed in ARLEQUIN 3.11 [26]
and DnaSP e sion 4.50 [27] p og ams. The signi icance o he
assumed model was es ed using he sum o squa es de ia ions
(SSD) be ween he obse ed and expec ed da a by means a
pa ame ic boo s apping app oach (1,000 pe mu a ions) and
conside ing he Ha pending’s aggedness index [45]. The
misma ch dis ibu ion will be mul imodal in s able popula ions
and unimodal in expanding ones. The ime o a possible
popula ion expansion ( ) can be calcula ed as =2u [44], whe e
is he mode o he misma ch dis ibu ion and uis he mu a ion
a e o he sequence (such ha u=mm
T
, whe e mis he mu a ion
a e/si e/gene a ion and m
T
is he numbe o nucleo ide base
pai s). I he sudden expansion model was no ejec ed, hen was
con e ed o ime since expansion ( ) in yea s be o e p esen as
ollows: [YBP ( = /2u)]. Fo Odon es hes sil e sides, he m DNA
subs i u ion a e was es ima ed in 0.023 [2]. Because ime ( )is
measu ed in gene a ions and he age a sexual ma u i y o
Odon es hes was calcula ed as minimum popula ion doubling ime
1.4–4.4 yea s (h p://www. ishbase.o g), o con e o ime since
expansion in yea s, we ha e mul iply by he gene a ion ime o a
mean 2.9 yea s.
Analysis o mic osa elli e ma ke s
A o al o 120 indi iduals om 13 popula ions we e analyzed
using hese nuclea ma ke s (Appendix S1). Ten polymo phic
mic osa elli e loci de eloped o Odon es hes we e ampli ied:
Odon02, Odon09, Odon27, Odon38, Odon39 [2]; and Obo01,
Obo26; Obo46; Obo54 and Obo77 [46]. The o wa d p ime o
each pai was luo escen ly labeled as ollows: Odon02, Odon25,
Odon39, Obo01, Obo54 wi h 59-FAM; Odon27, Odon38,
Obo26 and Obo77 wi h 59-HEX; and inally Odon09 wi h 59-
NED.
PCR ampli ica ions we e ca ied ou in a eac ion olume o
10 ml ( inal concen a ions in pa en hesis) each con aining DNA
ex ac (400 ng/ul); dNTPs (0.1 mM each); p ime s (10 mM each);
MgCl2 In i ogen (0.8–2.5 mM); Taq DNA Polyme ase In i o-
gen (0.04 U/ml); and In i ogen bu e (1X). Ampli ica ion
condi ions we e hose p oposed by [2] and [46] espec i ely.
The PCR eac ions we e ca ied ou in a Ve i y 96-Well The mal
Cycle (Applied Biosys ems) and he PCR p oduc s sepa a ed on
an ABI 377 au oma ed sequence . The ampli ied agmen s we e
geno yped using an ABI 3730 DNA Sequence (Applied Biosys-
ems) and isualiza ion o he esul s was pe o med using he
p og am GeneMappe 3.7 so wa e (Applied Biosys ems). Alleles
we e sco ed using a GeneScan 500 LIZ Size S anda d and
Geno ype so wa e (Applied Biosys ems, Inc.).
S a is ical and popula ion s uc u e analyses based on
nuclea ma ke s
Among all popula ions, only 13 we e analyzed wi h mic osa -
elli es. To implemen he analysis o he Odon es hes da a se ,
based on biogeog aphic c i e ia and o a oid s a is ical bias due o
he low numbe o samples in some collec ing si es, he popula ions
we e i s asc ibed o di e en axa as ollows: Oa including
popula ion om G
L
collec ing si e; Ob belonging o popula ions
om CA
L
,S
L
and C
L
;Oph popula ions om CA
B
,BA
D
,RB
D
;
inally Oap emb acing popula ions om B
P
,PN
B
,PA
S
,SC
S
,SG
S
,
PR
B
and R
L
. The numbe o alleles, he allelic ichness, he
expec ed he e ozygosi y co ec ed o sampling bias, he obse ed
he e ozygosi y, he polymo phic in o ma ion con en and he
es ima ed null allele equency we e calcula ed o each locus in
he whole popula ion pe axon using CERVUS e sion 3.0.3
[47]. GENEPOP 4.0.10 [48] was used o pe o m he exac es o
Ha dy-Weinbe g (HW) equilib ium by mic osa elli e loci ( es
mul i-popula ion) and by popula ion ( es mul i-locus) using he
Ma ko chain me hod wi h 1,000 i e a ions. Linkage disequilib-
ium be ween loci and de ia ions om Ha dy-Weinbe g equilib-
ium o each locus we e es ed by a Ma ko chain me hod
ollowing he algo i hm o Guo and Thompson [49] and using he
Bon e oni [50] co ec ion o mul iple compa isons (a= 0.05). All
he analyses ou lined abo e we e implemen ed in GENEPOP
4.0.10 [48]. W igh ’s F-s a is ics (Fis, Fs , and Fi [51]) o e
popula ions and loci we e calcula ed by FSTAT e sion 2.9.3.2
[52]. To de ec he p esence o sco ing e o s o he possible
p esence o null alleles, we analyzed he geno ypic ma ices
ob ained wi h he Mic o-Checke so wa e [53].
Neighbo Joining ee based on D
A
dis ance [54] was
cons uc ed using Popula ions, 1.2.30 so wa e package [55].
An analysis o popula ion subdi ision and clus e ing o
indi idual geno ypes was implemen ed wi h STRUCTURE .
2.2 [56] by a MCMC me hod. We conside ed 1 o 13 di e en
popula ions (K = 1 o K = 13). Ten independen uns employing
an admix u e model we e implemen ed wi h a bu n-in pe iod
leng h o 50,000 i e a ions, ollowed by 100,000 MCMC
eplica es. The a e age o hese independen uns was calcula ed
and he ue alue o ‘‘K’’ was accessed ollowing he app oach
de ailed in he manual o STRUCTURE . 2.2 (h p://p i ch.bsd.
uchicago.edu/s uc u e.h ml).
Di e en g oups o hypo heses and popula ions as sou ces o
a ia ion we e assessed in he AMOVA conside ing all en loci
using ARLEQUIN 3.1 so wa e package [26]. Fu he mo e, F
ST
alues o pai wise compa isons o he 13 Odon es hes popula ions
and hei signi icance le el o gene ic di e en ia ion (P = 0.05)
and Rs we e es ed addi ionally wi h FSTAT [52].
P omiscuous Specia ion Sil e side Fish Odon es hes SW A lan ic Ocean
PLOS ONE | www.plosone.o g 4 Augus 2014 | Volume 9 | Issue 8 | e104659
Popula ion di e gence and mig a ion a es om bo h
molecula ma ke s
To disc imina e be ween he ela i e e ec s o di e gence and
gene low on he specia ion p ocess, we analyzed ou da a se
unde he Isola ion wi h Mig a ion model [57]. The ‘‘isola ion
wi h mig a ion’’ model in IMa does no assume gene low and
gene ic d i a e in equilib ium, making i he mos app op ia e o
ecen ly di e ged popula ions ha sha e haplo ypes and alleles due
o bo h gene low and ances al polymo phism. The model
assumes ha an ances al popula ion spli s in o wo descendan
popula ions ha may con inue o exchange genes a e sepa a ion.
Following [1] we conside Ob as a eshwa e sis e axon o Oa
and Oph, sha ing a common eshwa e ances o wi h hese axa.
The me hod es ima es pos e io p obabili y dis ibu ions o
bo h ances al and ac ual popula ion sizes, di ec ional mig a ion
a es be ween he wo popula ions, and he ime elapsed since
popula ion spli ing. An MCMC app oach is used o d aw a
sample om he pos e io dis ibu ion o genealogies and o
es ima e h ee ypes o popula ion pa ame e s: popula ion size
(h=4Nu), spli ing ime ( =Tu, whe e Tis he ime in gene a ions
since he common ances y, and i is o he same o de o 4N) and
mig a ion a es (2NM =4Nu6m/2). The p io s we e inally se as
ollows: he uppe bound o popula ion sizes q= 10, spli ing imes
= 4 and mig a ion a es m= 2, espec i ely. We un he MCMC
simula ions wi h 100,000 bu n-in s eps and 10,000.000 sampled
genealogies. The pos e io dis ibu ions o mig a ion a es and
popula ion sizes a e de i ed analy ically om he sampled
genealogies.
Resul s
Gene ic a ia ion in he mi ochond ial COI gene in
Odon es hes species om SWA Ocean basins
This s udy includes a da a se o 655 bp o mi ochond ial COI
gene om 156 indi iduals belonging o popula ions o O.
a gen inensis, O. pe ugia, O. humensis and O. bona iensis
(GenBank accession numbe s: KJ854753–KJ854894, see Appen-
dix S1). Mo eo e , o he sequences om Odon es hes species and
one mo e dis an ly ela ed gene a (A he ina hepse us) we e
e ie ed om he GenBank and included o bo h he pai wise
dis ance compa isons and he phylogene ic analyses.
Among 36 COI haplo ypes ini ially assigned o O. a gen inensis,
30% o hem we e sha ed wi h O. pe ugiae and 25% wi h O.
humensis espec i elly. Among 7 COI haplo ypes ini ially g ouping
O. bona iensis sequences, 58% o hem we e sha ed wi h O. pe ugiae
and 8% wi h O. humensis. The e o e we pa i ioned he s a is ical
analysis in wo di e en da a se s: O. a gen inensis-pe ugiae-humensis
(Oaph)andO. bona iensis-pe ugiae-humensis (Obph).
The Oaph popula ions showed highe haplo ype di e si y (h) and
nucleo ide di e si y (p) hanObph (Table 1). Thi y six haplo ypes
we e ound in Oaph popula ions whe eas only se en in Obph axa.
Excep o he h ee mos common haplo ypes (H_1 and H_2 in
Obph and H_6 in Oaph), mos haplo ypes ep esen ed a e a ian s
ha explained he obse ed haplo ype di e si y in each axa
(Appendix S1). A signi ican excess o low- equency haplo ypes and
he eby nega i e and signi ican alues o bo h Tajima’s and Fu’s
neu ali y es s we e obse ed in Oaph indica ing a depa u e om
neu ali y, whe eas Obph p esen ed only nega i e and signi ican
alues in Tajima’s D es (Table 1). These alues would be
consis en wi h popula ions ha expe ienced demog aphic expan-
sion scena ios o al e na i ely selec i e sweeps.
In Oaph popula ions he a e age o he co ec ed pai wise K2P
sequence di e gence be ween COI haplo ypes was highe han in
Obph (Table 1). The a e age pai wise dis ances be ween haplo-
Table 1. Es ima es o DNA polymo phism in COI gene o Odon es hes popula ions om SW A lan ic Coas , Rı
´o de la Pla a es ua y and U uguay-Neg o Ri e basins.
Base pai s Va iable Si es S Numbe o Haplo ypes Haplo ype Di e si y p
Kimu a 2P Dis ance
(T
+
Ts) D Fs
Oaph 684 40 36 36 0.843 (0.034) 0.003 (0.013) 0.007 (0.001) 22.397 (P,0.01) 254.046 (P,0.00)
Obph 684 20 10 7 0.696 (0.058) 0.002 (0.006) 0.004 (0.002) 22.797 (P,0.001) 20.715 (P.0.10)
S = A e age o polymo phic seg ega ing si es; Haplo ype di e si y (h = gene) (Nei, 1987); p= Nucleo ide di e si y (Nei, 1987). Co ec ed Kimu a 2P dis ances (1980). D = Neu ali y es (Tajima, 989). Fs = Neu ali y es (Fu,
1997). S anda d de ia ion in b acke s (SD).
doi:10.1371/jou nal.pone.0104659. 001
P omiscuous Specia ion Sil e side Fish Odon es hes SW A lan ic Ocean
PLOS ONE | www.plosone.o g 5 Augus 2014 | Volume 9 | Issue 8 | e104659

ypes o O. a gen inensis-pe ugiae-bona iensis-humensis axa and
o he Odon es hes species included in p esen s udy (O. egia,O.
pla ensis,O. smi i and O. ha che y) was 0.04760.011 (mean 6
SE), whe eas he di e gence be ween he o me and O. incisa was
0.07560.018. The a e age di e gence be ween he ing oup and
he ou g oup A. hepse us was 0.61160.153.
Phylogene ic analyses
P esen phylogene ic analyses included 43 haplo ypes om
Oaph and Obph popula ions. All pe o med phylogene ic analyses
(ML and BEAST) conduc ed using he HKY+Gmodel o sequence
e olu ion, clea ly iden i ied a majo monophyle ic and ecen ly
de i a e clade wi h a high pos e io p obabili y o occu ence,
including mino monophyle ic clades which collapsed in a basal
poly omy joining mos o he 41 haplo ypes o Oaph and a well
suppo ed clade o Obph (Fig. 2). O he mino clade in eg a ed by
wo O. bona iensis haplo ypes collapsed basal o he majo clade.
O he species om he genus Odon es hes (O. egia,O. smi i,O.
ha che y and O. pla ensis) we e ecip ocally monophyle ic in
ela ion o he mino clade o O. bona iensis and he majo
de i a e clade, whe eas O. incisa was he mos basal axon o he
genus Odon es hes. Simul aneously we used a e e ence calib a ion
ime o nodes by assuming a subs i u ion a e o con en ional a e
o mi ochond ial genome o 0.023 mu a ions/si e/pe million
Figu e 2. T ee opology gene a ed using he HKY
+
Cmodel o molecula e olu ion based on 43 COI gene haplo ypes (H) o
Odon es hes
om lowe U uguay and Neg o i e basins, he Rı
´o de la Pla a es ua y, SWA Ocean basins. Bayesian phylogeog aphic
in e ence amewo k implemen ed in BEAST 1.5.4 and he es ima ed di e gence da es. Numbe s abo e b anches e e o he Bayesian pos e io
p obabili y o occu ence o clades while boo s ap suppo alues om ML boo s ap a e shown below b anches. The bo om ba summa izes he
ime-scale di e gence da es in Mya.
doi:10.1371/jou nal.pone.0104659.g002
P omiscuous Specia ion Sil e side Fish Odon es hes SW A lan ic Ocean
PLOS ONE | www.plosone.o g 6 Augus 2014 | Volume 9 | Issue 8 | e104659
Table 2. Pai wise F
ST
alues based on COI da a se o Odon es hes popula ions om SW A lan ic Coas , Rı
´o de la Pla a es ua y and U uguay-Neg o Ri e basins.
YA
S
CA
B
PV
S
BA
D
RB
D
AN
T
R
L
SC
S
SG
S
PA
S
PN
S
G
L
B
P
C
L
VA
S
S
L
CA
L
PR
B
CH
L
INIDEP
YA
S
_CA
B
_PV
S
(Op)0
BA
D
_RB
D
_AN
T
(Oph) 0.036 0
R
L
(Oap)0.229 0.085 0
SC
S
(Oap)0.261 0.092 0.023 0
SG
S
(Oap)0.315 0.103 20.031 0.013 0
PA
S
(Oap) 0.191 0.076 20.04 20.043 20.074 0
PN
S
(Oap) 0.147 0.040 0.024 20.042 0.049 20.013 0
G
L
(Oa)0.249 0.088 20.026 0.022 20.038 20.035 0.036 0
B
P
(Oap)0.262 0.086 20.020 20.056 20.065 20.057 0.009 20.026 0
C
L
_VA
S
(Obp) 0.206 0.207 0.405 0.582 0.616 0.551 0.483 0.503 0.539 0
S
L
(Ob) 0.133 0.086 0.259 0.305 0.340 0.301 0.215 0.288 0.302 0.196 0
CA
L
(Ob)0.480 0.313 0.312 0.400 0.450 0.387 0.283 0.355 0.377 0.709 0.220 0
PR
B
(Oap)0.333 0.195 0.211 0.253 0.313 0.250 0.026 0.243 0.255 0.667 0.332 0.499 0
CH
L
(Ob) 0.067 0.021 0.238 0.273 0.318 0.270 0.173 0.260 0.272 0.147 0.087 0.450 0.333 0
INIDEP (Oa)0.195 0.112 0.122 0.092 0.189 0.103 20.051 0.141 0.141 0.503 0.246 0.321 0.100 0.214 0
F
ST
signi ican alues a e in bold (P= 0.05). (See Fig. 1 and Appendix S1).
doi:10.1371/jou nal.pone.0104659. 002
P omiscuous Specia ion Sil e side Fish Odon es hes SW A lan ic Ocean
PLOS ONE | www.plosone.o g 7 Augus 2014 | Volume 9 | Issue 8 | e104659
yea s [16] o cap u e a plausible ime in e al (lowe and uppe
es ima e) o clade di e gence. All Odon es hes clades showed a
high p obabili y o ha e di e ged be ween 0.1 and 2.5 Mya
(Qua e na y). The di e gence be ween he genus A he ina and
Odon es hes occu ed in he Miocene.
Popula ion gene ic s uc u e, isola ion by dis ance and
his o ical demog aphy
Table 2 shows he pai wise F
ST
alues o he COI da a se
among he 15 collec ing si es analyzed wi h his ma ke . Low
popula ion gene ic s uc u e was de ec ed among locali ies
asc ibed o Oap om RP es ua y and AC a eas espec i ely.
Ne e heless, hese locali ies appea ed di e gen o hose asc ibed
o Obph om some RP es ua y si es, and UNR basins.
Rema kably he CA
L
collec ing si e seems o be he mos di e gen
om all he emaining ones. The mos plausible popula ion
s uc u ing based on COI da a se among es ed hypo heses in he
AMOVA was add essed ollowing wo di e en g ouping c i e ia:
(a) assigning all popula ions o wo-g oup o samples; (b) o ming
h ee g oups o popula ions (Table 3). The wo-g oup hypo hesis
(a) poin ed ou ha mos gene ic a ia ion was dis ibu ed among
g oups (W
CT
), sugges ing a ema kably highe le el o gene ic
s uc u e when samples om ma ine and es ua ine mo phs
Table 3. Analysis o molecula a iance (AMOVA) based on COI gene o Odon es hes popula ions om SW A lan ic Coas , Rı
´odela
Pla a es ua y and U uguay-Neg o Ri e basins.
Hypo hesis Sou ce o a ia ion
d
Sum o
squa es
Va iance
componen s
Pe cen age o
a ia ion Ws a is ics
a Among g oups 1 8.864 0.10574 Va 21.21 W
CT
=0.21209
Among popula ion wi hin g oups 13 8.690 0.03416 Vb 6.85 W
SC
= 0.08695
Wi hin popula ions 139 49.855 0.35867 Vc 71.94 W
ST
= 0.28060
b Among g oups 2 9.588 0.09463 Va 19.44 W
CT
=0.19441
Among popula ion wi hin g oups 11 7.419 0.03316Vb 6.81 W
SC
= 0.08456
Wi hin popula ions 138 49.539 0.35897 Vc 73.75 W
ST
= 0.26253
Two g ouping hypo heses among all es ed: a) con o ming wo g oups o popula ions as ollows: 1-popula ions om S
L
,C
L
,CA
L
and CH
L
s. 2-popula ions om VA
S
,R
L
,
SC
S
,SG
S
,PA
S
,PN
B
,G
L
,B
P
,PR
B
, INIDEP, CA
B
,YA
S
,BA
D
,RB
D
and PV
S
;b) sepa a ing h ee g oups o samples as ollows:1-popula ions om CA
B
,PV
S
,YA
S
,BA
D
and RB
D
;2-
popula ions om VA
S
,R
L
,SC
S
,SG
S
,PA
S
,PN
B
,G
L
,B
P
,PR
B
, INIDEP and 3- popula ions om S
L
,C
L
,CA
L
and CH
L
. (See Fig. 1 and Appendix S1).
doi:10.1371/jou nal.pone.0104659. 003
Figu e 3. Haplo ype ne wo k (cons uc ed wi h NETWORK . 4.6.0.0 so wa e) o
Oaph
and
Obph
axa. Black do s ep esen missing
haplo ypes and ci cle size is p opo ional o haplo ype equency. Di e en colou s in each ci cle indica e he collec ing si es as desc ibed in he
Figu e 1.
doi:10.1371/jou nal.pone.0104659.g003
P omiscuous Specia ion Sil e side Fish Odon es hes SW A lan ic Ocean
PLOS ONE | www.plosone.o g 8 Augus 2014 | Volume 9 | Issue 8 | e104659
asc ibed o Oaph popula ions was conside ed as a sepa a e g oup
om he o he including eshwa e samples om Obph axa.
The haplo ype ne wo k based on COI gene (Fig. 3) showed a
s ikingly s a -shaped opology including he wo mos equen
haplo ypes wi h a high p opo ion o single ons, ypical o
popula ions ha ha e su e ed a ecen demog aphic expansion.
One o he mos equen and cen al haplo ypes (H_6) including
samples o Oaph is p esen in 11 sampling si es and is sho ly
in e connec ed by one o h ee s ep-mu a ions o mos haplo ypes
o he ne wo k, belonging o Oaph popula ions. A single s ep
mu a ion sepa a ed H_6 om he o he mos equen haplo ype
(H_2), which included samples belonging o Obph axa om 6
sampling si es. Rema kably, he ne wo k opology showed some
loops in ol ing he cen al H_6 and H_2 haplo ypes and hei
espec i e de i a e ones. The e o e hese al e na i e links may be
ep esen ing equally good connec ions due o homoplasy o
pe haps he exis ence o pe iphe al e icula ion e en s among
hem.
Taking in o accoun all collec ing si es, nega i e alues in
Man el es we e obse ed ( = 20.126, p = 0.050), showing a
nega i e co ela ion be ween gene ic and geog aphic dis ances and
excluding he isola ion by dis ance model o popula ion di e en-
ia ion in Odon es hes.
Figu e 4 shows an unimodal misma ch dis ibu ion pa e n in
he COI da a se which adjus ed o he dis ibu ion p edic ed by
he g ow h–decline popula ion model [44] in he Oaph popula-
ions. The sum o squa es de ia ions was SSD = 0.148 (P.0.06)
and Ha pending’s Raggedness index was 0.486. The es ima e
pa ame e unde he model was = 3.218. The ime o expansion-
decline in Oaph based on a subs i u ion o his ma ke was
es ima ed o ha e s a ed a ound 227,000 YBP. In he Obph
popula ions da a se he sum o squa es de ia ion alue was
SSD = 0.180 (P.0.05) and Ha pending’s Raggedness index was
0.492, he e o e popula ion g ow h-decline hypo hesis was accep -
ed o his axon. The es ima e pa ame e unde he model was
= 7.843 and he ime o expansion-decline in Obph popula ions
we e es ima ed o ha e s a ed a ound 497,000 YBP.
Gene ic a iabili y in mul i-locus nuclea da a
The measu es o mic osa elli e gene ic a ia ion including o al
numbe o alleles, allelic ichness, he e ozygosi y and HWE
de ia ion o each locus and he co esponding a e age ac oss
all loci pe axon a e showed in Table 4. Signi ican depa u es
om Ha dy-Weinbe g equilib ium we e ound a some loci o he
ollowing popula ions: Oap (Odon27 and Obo54); Oph (Obo26,
Odon27, Odon39 and Odon02); Ob (Obo39, Obo54 and
Odon02). This could be due o a Wahlund e ec , wi h a educ ion
Figu e 4. Misma ch dis ibu ion in
Odon es hes
species unde he g ow h–decline popula ion model using m DNA COI da a se . (a).
Oaph and (b). Obph popula ions.
doi:10.1371/jou nal.pone.0104659.g004
P omiscuous Specia ion Sil e side Fish Odon es hes SW A lan ic Ocean
PLOS ONE | www.plosone.o g 9 Augus 2014 | Volume 9 | Issue 8 | e104659