Coral Reefs (2007) 26:399–410

DOI 10.1007/s00338-007-0210-5

REPORT

Diversity and evolution in the zoanthid genus Palythoa (Cnidaria: Hexacorallia) based on nuclear ITS-rDNA

J. D. Reimer · K. Takishita · S. Ono · T. Maruyama

Received: 22 September 2006 / Accepted: 16 February 2007 / Published online: 12 April 2007© Springer-Verlag 2007

Abstract Previous phylogenetic studies based on mito-chondrial DNA markers have suggested that the zoanthidgenus Palythoa may consist of both Palythoa species(Palythoa tuberculosa) and species formerly assigned tothe genus Protopalythoa (Palythoa mutuki, Palythoa helio-discus). In the present study various Palythoa spp. samplescollected primarily from southern Japan with additionalsamples from the Indo-PaciWc and Caribbean Sea wereexamined. The nuclear internal transcribed spacer of ribo-somal DNA (ITS-rDNA) was sequenced and aligned forphylogenetic analyses to further investigate the relationshipbetween P. tuberculosa, P. mutuki, and P. heliodiscus. ITS-rDNA analyses showed species groups forming monophy-lies with similar topology but with much higher resolutionthan seen for mitochondrial phylogenetic analyses. Theresults also conWrmed the very close relationship ofP. tuberculosa and P. mutuki. Some specimens appeared tobe a potentially undescribed Palythoa species (designatedPalythoa sp. sakurajimensis). Additionally, ITS-rDNA

sequences of P. mutuki and P. tuberculosa showed additivepolymorphic site, demonstrating for the Wrst time a poten-tial history of reticulate evolution in Palythoa.

Keywords Anthozoa · Reticulate evolution · ITS-rDNA · Zoanthid · Palythoa

Introduction

The zoanthid genus Palythoa (Cnidaria: Hexacorallia)is common in shallow subtropical and tropical watersthroughout the world. Although there are 193 species men-tioned in the literature (Fautin 2006) it is likely that manyof these nominal species have been re-described and thatthe true diversity of species in Palythoa is lower than thisnumber (see Reimer et al. 2006c).

Traditional classiWcation of Palythoa spp. has beenlargely based on morphological characteristics such aspolyp shape, oral disk size and diameter, and tentacle num-ber (see Ryland and Lancaster 2003). Additionally, Palyt-hoa spp. were deWned as having embedded “immersae”polyps and the related proposed genus Protopalythoaconsisted of similar but “liberae”-polyped (non-embeddedpolyps) species (Pax 1910). However, using molecularmarkers (mitochondrial cytochrome oxidase subunit I[COI] and mitochondrial 16S ribosomal DNA sequences[mtDNA 16S rDNA]), it has recently been shown thatPalythoa and Protopalythoa have a very close relationshipup to the level of congeners (Reimer et al. 2006c), and thatthese two taxa should be combined into a single genus,Palythoa. In addition, data from the closely related genusZoanthus have shown that polyp shape and other morpho-logical characters are not necessarily good indicators ofrelatedness (Reimer et al. 2006b).

Communicated by Biology Editor M. van Oppen.

J. D. Reimer (&) · K. Takishita · T. MaruyamaResearch Program for Marine Biology and Ecology, Extremobiosphere Research Center, Japan Agency for Marine-Earth Science and Technology (JAMSTEC), 2-15 Natsushima, Yokosuka, Kanagawa 237-0061, Japane-mail: jreimer@jamstec.go.jp

Present Address:J. D. ReimerBiological Institute on Kuroshio, 560 Nishidomari, Otsuki, Kochi 788-0333, Japan

S. OnoMiyakonojo Higashi High School, Mimata, Miyazaki 889-1996, Japan

123

400 Coral Reefs (2007) 26:399–410

Up until now, molecular phylogenetic analyses of Palyt-hoa have relied exclusively upon mitochondrial DNAmarkers (Sinniger et al. 2005; Reimer et al. 2006c), whichhave a slow rate of evolution in anthozoans compared toother animals (Shearer et al. 2002). For example, Palythoatuberculosa and Palythoa mutuki only diVer by 1–2 sitesover 870 base pairs (bp) in mtDNA 16S rDNA (Reimeret al. 2006c). On the other hand, nuclear internal tran-scribed spacer of ribosomal DNA (ITS-rDNA) has beensuccessfully used in a variety of other hexacorallian taxato delineate boundaries between many species (e.g., seeHunter et al. 1997). ITS-rDNA, despite the potential pres-ence of multiple copies (Marquez et al. 2003), has been par-ticularly useful in exploring evolutionary patterns in closelyrelated groups due to its extremely non-conservative nature(e.g., Hunter et al. 1997; Odorico and Miller 1997; Medinaet al. 1999; van Oppen et al. 2000, 2002; Diekmann et al.2001; Marquez et al. 2003; Fukami et al. 2004).

It has been proposed that the apparently unclear speciesboundaries in many genera of corals may at least partiallybe due to interspeciWc hybridization and subsequent reticu-late evolution (Veron 1995). Many diVerent species andgenera of Hexacorallia have also been shown to reproducein synchronous phase with the moon in mass spawningevents (see Levitan et al. 2004; Penland et al. 2004; Onoet al. 2005), suggesting that hybridization, reticulate evolu-tion and/or introgression (hybrids backcrossing with par-ents) in Anthozoa may be more widespread than it iscurrently known.

In this study, a nuclear marker (ITS-rDNA) was appliedto: (1) obtain a phylogeny of this genus with improved res-olution to conWrm or refute the previous hypothesis thatPalythoa and Protopalythoa are congeneric and (2) toexamine the possibility of reticulate evolution in Palythoaspp.

Materials and methods

Sampling

Samples of Palythoa spp. were collected from several sitesin Japan (Fig. 1) as well as from the Indian and Atlanticoceans between January 2004 and May 2006 (Table 1), andstored in 80–100% ethanol at ¡20°C. Samples (Fig. 2)included specimens both of the immersae ‘Palythoa’ mor-phology (nominal P. tuberculosa, as well as a sample ofPalythoa cf. caribaeorum) and the liberae ‘Protopalythoa’morphology (nominal P. mutuki, Palythoa heliodiscus, andunidentiWed specimens) (see Reimer et al. 2006c for spe-cies’ details). As samples were collected, in situ photo-graphs were taken to assist in identiWcation and forcollection of morphological data (oral disk/polyp diameter,color, tentacle count, polyp form). Sample nomenclature isexplained in the notes in Table 1.

DNA extraction, PCR ampliWcation, cloning, and sequencing

DNA was extracted from samples weighing 5–20 mg usinga spin-column DNeasy Animal Extraction protocol(Qiagen, Santa Clarita, CA, USA). PCR ampliWcation usingthe genomic DNA as a template was performed using Hot-StarTaq DNA polymerase (Qiagen, Tokyo, Japan) accord-ing to the manufacturer’s instructions. Mitochondrial 16SrDNA was ampliWed following procedures outlined inSinniger et al. (2005). The ITS-rDNA region (the 3� end of18S rDNA, ITS-1, 5.8S rDNA, ITS-2, and the 5� end of28S rDNA) was ampliWed following procedures outlined inReimer et al. (2007). The ampliWed products were visual-ized by 1.5% agarose gel electrophoresis. Some PCR-ampliWed DNA fragments were cloned and analyzed

Fig. 1 Map showing sampling locations of Palythoa spp. in this study JAPAN

Miyakejima

Wakayama (Shirahama, Kushimoto)

Yakushima

Amami (Bashayama, Tomori)

Erabu (Wanjo, Sumiyoshi, Kaito)

Yoron (Shin's Reef, Chabana)

Kerama (Kuroshima-kita, Paradise-Iso)

Ishigaki (Onsen, Kataguwa)

PACIFIC OCEANEAST CHINA SEA

30 N

130 E 140 E

0 300km

Hachijojima

Iriomote (Hoshizuna, Haemida)

Otsuki (Nishidomari, Futabai)

Chibishi

Sakurajima

Other sample locations:Pacific (Saipan)Indian (Madagascar, Israel)Atlantic (Honduras)

KOREA

123

Coral Reefs (2007) 26:399–410 401

Tab

le1

Pal

ytho

a sa

mpl

es u

sed

in th

is s

tudy

Sam

ple

nam

eaL

ocat

ionb

Dep

thSa

mpl

ing

date

Col

lect

ed b

ycM

orph

olog

ical

iden

tiWca

tion

mtD

NA

16S

rD

NA

A

cces

sion

num

ber

ITS-

rDN

A

Acc

essi

on n

umbe

rPh

ylog

enet

ic

conc

lusi

on

PhE

rabu

Kai

to1

Kai

to, E

rabu

¡19

.0M

ay 2

005

JDR

P. h

elio

disc

usA

B21

9224

dD

Q99

7882

P. h

elio

disc

us

PhIs

higa

ki K

ata2

Kat

aguw

a, I

shig

aki

¡9.

5Fe

b 20

05JD

RP

. hel

iodi

scus

DQ

9978

59D

Q99

7885

P. h

elio

disc

us

PhIs

higa

ki K

ata3

Kat

aguw

a, I

shig

aki

¡9.

5Fe

b 20

05JD

RP

. hel

iodi

scus

NA

DQ

9978

84P

. hel

iodi

scus

PhIs

higa

ki K

ata5

Kat

aguw

a, I

shig

aki

¡10

.0Fe

b 20

05JD

RP

. hel

iodi

scus

DQ

9978

43N

AP

. hel

iodi

scus

PhIs

higa

ki K

ata1

1K

atag

uwa,

Ish

igak

i¡

12.0

Dec

200

5JD

RP

. hel

iodi

scus

DQ

9978

61D

Q99

7880

P. h

elio

disc

us

PhSa

ipan

Lau

Lau

1L

au L

au, S

aipa

n¡

3.0

Dec

200

4JD

RP

. hel

iodi

scus

AB

2192

23d

DQ

9978

83P

. hel

iodi

scus

PhSa

ipan

Lau

Lau

2L

au L

au, S

aipa

n¡

2.0

Dec

200

4JD

RP

. hel

iodi

scus

DQ

9978

44D

Q99

7881

P. h

elio

disc

us

PWak

ayam

a Sh

ira1

Shi

raha

ma,

Wak

ayam

a+

0.5

Apr

200

6JD

R a

nd H

Fun

know

n P

alyt

hoa

sp.

DQ

9978

63D

Q99

7887

P. s

p. s

akur

ajim

ensi

s

PSak

ura

Hak

ama1

Hak

amag

oshi

, Sak

uraj

ima

¡4.

0Fe

b 20

06JD

R, S

O,

JT, a

nd A

Iun

know

n P

alyt

hoa

sp.

DQ

9978

42D

Q99

7886

P. s

p. s

akur

ajim

ensi

s

PEW

anjo

N1

Wan

jo-n

orth

, Era

bu+

0.5

May

200

6JD

Run

know

n P

alyt

hoa

sp.

DQ

9978

62N

AP

. sp.

sak

uraj

imen

sis

PmM

iyak

eIzu

I1Iz

ushi

ta, M

iyak

ejim

a0.

0Ju

ne 2

005

JDR

P. m

utuk

i 1A

B21

9225

dD

Q99

7889

P. m

utuk

i 1

PmY

akuS

ango

1S

ango

ham

a, Y

akus

him

a0.

0Ju

ne 2

003

JDR

P. m

utuk

i 1A

B21

9222

dD

Q99

7890

P. m

utuk

i 1

PmY

aku

Sang

o2S

ango

ham

a, Y

akus

him

a0.

0Ju

ly 2

004

JDR

P. m

utuk

i 1D

Q99

7875

DQ

9978

92P

. mut

uki 1

PmA

mam

i Tom

ori1

Tom

ori,

Am

ami

+0.

5A

ug 2

004

JDR

P. m

utuk

i 2A

B21

9220

dD

Q99

7891

P. m

utuk

i 2

PmA

mam

i Tom

ori2

Tom

ori,

Am

ami

+1.

0A

ug 2

004

JDR

P. m

utuk

i 2A

B21

9221

dN

AP

. mut

uki 2

PmE

rabu

Sum

iyos

hi1

Sum

iyos

hi, E

rabu

+0.

5M

ay 2

006

JDR

P. m

utuk

i 1D

Q99

7847

DQ

9978

94P

. mut

uki 1

PmIr

iom

ote

Hos

hi1

Hos

hizu

na, I

riom

ote

0.0

Feb

2006

JDR

P. m

utuk

i 2D

Q99

7841

DQ

9978

88P

. mut

uki 2

PmIr

iom

ote

Hae

mid

a3H

aem

ida,

Iri

omot

e0.

0Fe

b 20

06JD

RP

. mut

uki 2

DQ

9978

40D

Q99

7893

P. m

utuk

i 2

PtM

iyak

eIzu

1Iz

ushi

ta, M

iyak

ejim

a¡

2.0

June

200

5JD

RP

. tub

ercu

losa

AB

2192

18d

NA

*P

. tub

ercu

losa

PtH

achi

joji

ma

Bor

a2B

orah

ama,

Hac

hijo

jim

a¡

1.5

Jan

2006

JDR

P. t

uber

culo

saD

Q99

7868

DQ

9978

98P

. tub

ercu

losa

PtW

akay

ama

Kus

hi1

Kus

him

oto,

Wak

ayam

a¡

1.0

Aug

200

4JD

RP

. tub

ercu

losa

DQ

9978

74D

Q99

7899

P. t

uber

culo

sa

PtO

tsuk

i Nis

hido

mar

i1N

ishi

dom

ari,

Ots

uki

¡6.

0Ja

n 20

06JD

RP

. tub

ercu

losa

DQ

9978

57N

AP

. tub

ercu

losa

PtO

tsuk

i Nis

hido

mar

i2N

ishi

dom

ari,

Ots

uki

¡4.

0Ja

n 20

06JD

RP

. tub

ercu

losa

DQ

9978

67N

AP

. tub

ercu

losa

PtO

tsuk

i Nis

hido

mar

i3N

ishi

dom

ari,

Ots

uki

¡3.

0Ja

n 20

06JD

RP

. tub

ercu

losa

DQ

9978

53D

Q99

7939

P. t

uber

culo

sa

PtO

tsuk

i Fut

abae

1F

utab

ai, O

tsuk

i¡

12.0

Jan

2006

JDR

P. t

uber

culo

saD

Q99

7865

DQ

9979

18,

924-

926,

944

, 945

fP

. tub

ercu

losa

PtY

akuS

ango

3S

ango

ham

a, Y

akus

him

a¡

12.0

July

200

4JD

RP

. tub

ercu

losa

DQ

9978

58N

AP

. tub

ercu

losa

PtY

akuS

ango

4S

ango

ham

a, Y

akus

him

a¡

1.5

July

200

4JD

RP

. tub

ercu

losa

DQ

9978

64D

Q99

7903

, 92

8, 9

35-9

38f

P. t

uber

culo

sa

PtA

mam

i Tom

ori2

Tom

ori,

Am

ami

¡1.

5A

ug 2

004

JDR

P. t

uber

culo

saD

Q99

7839

DQ

9978

97P

. tub

ercu

losa

PtA

mam

i Bas

haya

ma1

Bas

haya

ma,

Am

ami

¡2.

0O

ct 2

005

JDR

P. t

uber

culo

saD

Q99

7845

DQ

9979

23P

. tub

ercu

losa

PtA

mam

i Bas

haya

ma2

Bas

haya

ma,

Am

ami

¡2.

0O

ct 2

005

JDR

P. t

uber

culo

saD

Q99

7851

DQ

9979

00P

. tub

ercu

losa

123

402 Coral Reefs (2007) 26:399–410

Tab

le1

cont

inue

d

* N

Ada

ta n

ot a

cqui

red

aSa

mpl

e na

mes

fol

low

the

conv

enti

on o

f G

enus

-spe

cies

-Loc

atio

n-L

ocat

ion-

sam

ple

num

ber,

exc

ept f

or s

peci

es f

rom

Isr

ael a

nd H

ondu

ras,

whe

re lo

cati

on is

sim

ply

coun

try

nam

e. A

dditi

onal

ly,

whe

n on

ly g

enus

nam

e w

as k

now

n, n

o sp

ecie

s de

sign

atio

n w

as a

dded

to th

e sa

mpl

e na

me

bL

ocat

ions

for

sam

ples

fro

m o

cean

s ot

her

than

the

PaciW

c in

dica

ted

in p

aren

thes

es. A

ll lo

cati

ons

in J

apan

unl

ess

othe

rwis

e no

ted

cN

ame

abbr

evia

tions

: JD

R=

J. R

eim

er, H

F=

H. F

ukam

i, S

O=

S. O

no, J

T=

J. T

suka

hara

, AI

=A

. Iw

ama,

FS

=F.

Sin

nige

r, O

P=

O. P

olak

dFr

om R

eim

er e

tal.

(200

6c)

eFr

om R

eim

er e

tal.

(200

6a)

fS

ampl

es w

ith

mor

e th

an o

ne I

TS

-rD

NA

seq

uenc

e A

cces

sion

Num

ber

wer

e cl

oned

. Clo

nes

are

indi

cate

d in

Fig

.4 b

y se

quen

ces

with

hyp

hena

ted

sam

ple

num

bers

Sam

ple

nam

eaL

ocat

ionb

Dep

thS

ampl

ing

date

Col

lect

ed b

ycM

orph

olog

ical

id

entiW

catio

nm

tDN

A 1

6S r

DN

A

Acc

essi

on n

umbe

rIT

S-r

DN

A

Acc

essi

on n

umbe

rPh

ylog

enet

ic

conc

lusi

on

PtE

rabu

Wan

jo2

Wan

jo, E

rabu

¡2.

0M

ay 2

006

JDR

P. t

uber

culo

saD

Q99

7850

NA

P. t

uber

culo

sa

PtE

rabu

Wan

jo3

Wan

jo, E

rabu

¡2.

0M

ay 2

006

JDR

P. t

uber

culo

saD

Q99

7846

DQ

9979

02P

. tub

ercu

losa

PtE

rabu

Sum

iyos

hi2

Sum

iyos

hi, E

rabu

¡9.

0M

ay 2

006

JDR

P. t

uber

culo

saD

Q99

7855

NA

P. t

uber

culo

sa

PtY

oron

Shin

1Sh

in’s

Ree

f, Y

oron

¡1.

0M

ay 2

005

JDR

P. t

uber

culo

saA

B21

9219

dD

Q99

7921

P. t

uber

culo

sa

PtY

oron

Shin

2Sh

in’s

Ree

f, Y

oron

¡2.

0M

ay 2

005

JDR

P. t

uber

culo

saD

Q99

7877

NA

P. t

uber

culo

sa

PtY

oron

Cha

bana

2C

haba

na, Y

oron

¡2.

0M

ay 2

005

JDR

P. t

uber

culo

saD

Q99

7879

DQ

9979

22P

. tub

ercu

losa

PtC

hibi

shi N

aga1

Nag

annu

-kit

a, C

hibi

shi

¡3.

0Ju

ne 2

004

JDR

P. t

uber

culo

saD

Q99

7860

DQ

9978

96,

927,

930

, 932

-934

fP

. tub

ercu

losa

PtC

hibi

shi N

aga2

Nag

annu

-kit

a, C

hibi

shi

¡4.

0Ju

ne 2

004

JDR

P. t

uber

culo

saD

Q99

7869

NA

P. t

uber

culo

sa

PtK

eram

a K

uro3

Kur

oshi

ma-

kita

, Ker

ama

¡2.

0Ju

ne 2

004

JDR

P. t

uber

culo

saD

Q99

7856

NA

P. t

uber

culo

sa

PtK

eram

a Pa

radi

se1

Para

dise

-Iso

, Ker

ama

¡7.

0Ju

ne 2

004

JDR

P. t

uber

culo

saD

Q99

7854

N

AP

. tub

ercu

losa

PtIs

higa

ki O

nsen

1O

nsen

, Ish

igak

i¡

8.5

Feb

200

5JD

RP

. tub

ercu

losa

NA

DQ

9979

19, 9

29f

P. t

uber

culo

sa

PtIs

higa

ki K

ata1

Kat

aguw

a, I

shig

aki

¡10

.0F

eb 2

005

JDR

P. t

uber

culo

saD

Q99

7866

DQ

9979

20P

. tub

ercu

losa

PtIs

higa

ki K

ata4

Kat

aguw

a, I

shig

aki

¡10

.0F

eb 2

005

JDR

P. t

uber

culo

saD

Q99

7873

NA

P. t

uber

culo

sa

PtIs

higa

ki K

ata6

Kat

aguw

a, I

shig

aki

¡12

.0D

ec 2

005

JDR

P. t

uber

culo

saD

Q99

7852

NA

P. t

uber

culo

sa

PtIr

iom

ote

Hos

hi1

Hos

hizu

na, I

riom

ote

0.0

Feb

200

6JD

RP

. tub

ercu

losa

DQ

9978

48D

Q99

7904

-917

fP

. tub

ercu

losa

PtSa

ipan

Lau

Lau

1L

au L

au, S

aipa

n¡

3.0

Dec

200

4JD

RP

. tub

ercu

losa

DQ

9978

72D

Q99

7895

P. t

uber

culo

sa

PMad

agas

car

289

Pisc

ine-

Saka

tia,

M

adag

asca

r (I

ndia

n)¡

10.0

July

200

5FS

Pal

ytho

a sp

.D

Q99

7878

DQ

9979

01P

. tub

ercu

losa

PtIs

rael

1E

ilat

, Isr

ael (

Indi

an)

¡3.

0M

ay 2

006

OP

P. t

uber

culo

saD

Q99

7849

DQ

9979

31, 9

40, 9

41f

P. t

uber

culo

sa

PtIs

rael

2E

ilat

, Isr

ael (

Indi

an)

¡1.

0M

ay 2

006

OP

P. t

uber

culo

saD

Q99

7876

NA

P. t

uber

culo

sa

PcH

ond1

Util

a, H

ondu

ras

(Atl

antic

)¡

8.0

Feb

200

4FS

P. c

arib

aeor

umN

AD

Q99

7942

, 943

, 946

fP

. tub

ercu

losa

gro

up

ZsS

H23

Hak

amag

oshi

, Sak

uraj

ima

¡9.

0Ju

ly 2

004

JDR

Z. s

ansi

bari

cus

AB

2191

87e

NA

Z. s

ansi

bari

cus

ZsE

S1Su

miy

oshi

, Era

bu0.

0M

ay 2

006

JDR

Z. s

ansi

bari

cus

DQ

9978

71N

AZ

. san

siba

ricu

s

ZgY

S1Sa

ngoh

ama,

Yak

ushi

ma

¡1.

5Ju

ly 2

004

JDR

Z. g

igan

tus

AB

2191

92e

NA

Z. g

igan

tus

ZkY

S1Sa

ngoh

ama,

Yak

ushi

ma

¡1.

5Ju

ly 2

004

JDR

Z. k

uros

hio

AB

2191

91e

NA

Z. k

uros

hio

ZS

H50

Hak

amag

oshi

, Sak

uraj

ima

¡5.

0F

eb 2

005

JDR

Zoa

nthu

s sp

.D

Q99

7870

NA

Z. k

uros

hio

grou

p

123

Coral Reefs (2007) 26:399–410 403

following procedures outlined in Reimer et al. (2007).Between 2 (sample PtIsO1) and 14 (sample PtIrHo1) cloneswere sequenced (see Table 1).

Phylogenetic analyses

New sequences obtained in the present study were depos-ited in DDBJ and GenBank (accession numbersDQ997839–DQ997946). By using CLUSTAL X version1.8 (Thompson et al. 1997), the nucleotide sequences of mt16S rDNA from samples were aligned with previously pub-lished mtDNA 16S rDNA sequences (Table 1) from Palyt-hoa (AB219218-AB219225), and Zoanthus spp. sequences(AB219187, AB219191, AB219192) were used as the out-group. Zoanthus ITS-rDNA sequences (particularly ITS-1and ITS-2 spacers) were highly divergent from Palythoasequences (see Reimer et al. 2007) and thus an ITS-rDNAalignment consisting only of Palythoa sequences wasgenerated, using Palythoa heliodiscus sequences as theoutgroup for the subsequent phylogenetic analyses. Thealignments were inspected by eye and manually edited. Allambiguous sites were removed from the dataset before phy-logenetic analyses. Consequently, two alignment datasetswere generated: (1) 759 alignment positions of 52

sequences (mtDNA 16S rDNA); and (2) 686 alignmentpositions of 67 sequences (ITS-rDNA). The alignment dataare available on request from the corresponding author andalso from the European Molecular Biology Lab (EMBL)(alignment numbers: mtDNA 16S rDNA = ALIGN_001072,ITS-rDNA = ALIGN_001073).

The alignments of mtDNA 16S rDNA and ITS-rDNAwere tested for optimal Wt of various nucleotide substitutionmodels using MODELTEST version 3.06 (Posada andCrandall 1998). The base frequencies, proportion of invari-able sites (and a gamma distribution) were estimated fromthe datasets. For the mtDNA 16S rDNA and ITS-rDNAdatasets, the TN model (Tamura and Nei 1993) incorporat-ing invariable sites (TN + I) and the Hasegawa, Kishinoand Yano (HKY) model (Hasegawa et al. 1985) incorporat-ing invariable sites and a discrete gamma distribution (fourcategories) (HKY + I + �) were selected by MODEL-TEST, respectively. The maximum-likelihood (ML) analy-ses with PhyML (Guindon and Gascuel 2003) of thesedatasets were independently performed using an input treegenerated by BIONJ (Gascuel 1997) with the modelsselected by MODELTEST. PhyML bootstrap trees (500replicates) were constructed using the same parameters asthe individual ML trees.

Fig. 2 a Palythoa tuberculosa at Hoshizuna, Iriomote, Okinawa, depth 0.5 m. Note the two colonies with diVerent coloration. b Palythoa mutuki 2 at Haemida, Iriomote, Okinawa, depth 0.0 m, c P. mutuki 1 at Izushita, Miyakejima, Tokyo, depth 0.0 m. Note the diVerences in oral disk coloration and the slight diVerence in polyp thickness between P. mutuki 1 and P. mutuki 2. d Palythoa sp. sakurajimensis at Hakamagoshi, Sakurajima, Kagoshima, depth 4.0 m, and e Palythoa heliodiscus at Sumiyoshi, Erabu, Kagoshima, depth 5.0 m. Scale bars 1 cm

123

404 Coral Reefs (2007) 26:399–410

ML distances of the two datasets were calculated underthe optimal models described above with PAUP* Version4.0 (SwoVord 1998). Distance trees were constructed usingthe neighbor-joining (NJ) method (Saitou and Nei 1987).The ML distance bootstrap analyses with 1,000 replicateswere also performed.

Bayesian trees were reconstructed by using the programMrBayes 3.1.2 (Ronquist and Huelsenbeck 2003) under thegeneral time reversible (GTR) model (Rodriguez et al. 1990)of nucleotide substitution incorporating invariable sites(GTR + I) for the mtDNA 16S rDNA dataset, and underHKY + I + � for the ITS-rDNA dataset [both modelsselected by MrModeltest (Nylander 2004. MrModeltest v2.Program distributed by the author. Evolutionary BiologyCentre, Uppsala University)]. One cold and three heatedMarkov chain Monte Carlo (MCMC) chains with default-chain temperatures were run for 1,000,000 generations, sam-pling log-likelihoods (InLs), and trees at 100-generationintervals (10,000 InLs and trees were saved during MCMC).The likelihood plot for mtDNA 16S rDNA and ITS-rDNAdatasets suggested that MCMC reached the stationary phaseafter the Wrst 30,000 and 300,000 generations, respectively.Thus, the remaining 9,700 and 7,000 trees of mtDNA 16SrDNA and ITS-rDNA were used to obtain posterior proba-bilities and branch-length estimates, respectively.

To examine the levels of sequence variation within eachITS-rDNA species group, all obtained sequences of theentire ITS-rDNA region (ITS-1 + 5.8S rDNA + ITS-2, 18SrDNA and 28S rDNA portions not included) for each spe-cies group were aligned and edited, and the total number ofvariable sites was estimated.

Results

The mtDNA 16S rDNA tree (Fig. 3) showed P. heliodiscusforming one highly supported clade (ML bootstrapsupport = 99%, NJ bootstrap support = 96%, but posteriorprobability of Bayesian inference [PP] = 0.69), and theP. mutuki species group derived from P. tuberculosa.P. mutuki was divided into two separate clades (P. mutuki1 and 2), with P. mutuki 1 diVering from P. tuberculosa by1 bp, and P. mutuki 2 only by 2 bp. Both P. mutuki groupsshowed generally low bootstrap support values, althoughthe P. mutuki 2 clade has a posterior probability of 0.95 inthe Bayesian analysis. Included in the P. tuberculosaspecies group were distant samples (from Japan) from thePaciWc and Indian Ocean (PMadagascar289, PtIsrael1,PtIsrael2), as well as P. cf. caribaeorum from Honduras(PcHond1 = sample PcH1 in Reimer et al. 2006c). Whilebootstrap probability was relatively low for the mono-phyly of P. tuberculosa (ML = 72%, NJ = 61%), posteriorprobability was moderately high (0.90). Additionally,

three unidentiWed Palythoa samples (PErabuWanjoN1,PSakuraHakama1, PWakayamaShira1) formed a separateclade with relatively low support (ML = 82%, NJ = 78%,PP = 0.80), designated Palythoa sp. sakurajimensis in thisstudy. Variation between mtDNA 16S rDNA sequenceswithin species groups was very low, i.e., <0.5% (Table 2).

The ITS-rDNA tree (Fig. 4) had four main Palythoagroups, three of which corresponded to known species(P. tuberculosa, P. mutuki, and P. heliodiscus) and theP. sp. sakurajimensis clade comprising two of the‘unknown Palythoa sp.’ samples (PSakuraHakama1 andPWakayamaShira1). P. heliodiscus formed a well-sup-ported clade (ML, NJ = 100%, PP = 1.00), as did P. sp. sak-urajimensis with moderate to high support (ML = 99%,NJ = 74%, PP = 0.97). A large clade containing both theseparate P. tuberculosa and P. mutuki clades was also mod-erately to highly supported (ML = 98%, NJ = 61%,PP = 0.96), as were both the P. mutuki (ML = 96%,NJ = 71%, PP = 0.95) and the P. tuberculosa groups(ML = 96%, NJ = 92%, PP = 0.99) within this larger clade.ITS-rDNA sequences from samples distant from Japan(PMadagascar289, PtIsrael1, PtIsrael2, PcHond1) wereclearly within the P. tuberculosa clade.

However, ITS-rDNA showed larger diVerences betweensequences within each species group. P. mutuki formed fourseparate subclades (two each of P. mutuki 1 and P. mutuki 2)with moderate to very high support, but the two subcladeseach of P. mutuki 1 and P. mutuki 2 mtDNA 16S rDNA didnot separately form a monphyly with the other subclade oftheir putative species group in the ITS-rDNA tree (Fig. 4).However, putative (based on mtDNA 16S rDNA) P. mutuki1 and 2 sequences were not found within the same subclade(Fig. 4). P. tuberculosa also showed intraspeciWc variation,with samples PtYoronShin1 and PtYoronChabana2 forminga highly supported basal group (ML = 98%, NJ = 99%, butPP = 0.54). In some cases, clones from the same P. tuber-culosa specimens clustered together within the same subc-lade (PtYakuSango4 and PcHond1). In other cases clonedsequences from individual specimens grouped in more thanone subclade (PtIriomoteHoshi1, PtOtsukiFutabae1) (Fig. 4).The P. mutuki and P. tuberculosa species groups seemed tobe very closely related in comparison to the other Palythoaspecies groups.

IntraspeciWc ITS-rDNA sequence variation rangedbetween 0.4% (P. sp. sakurajimensis) and 23.7% (P. tuber-culosa) (Table 2). P. heliodiscus (n = 5) and P. sp. sakura-jimensis (n = 2) groups showed little variation in ITS-rDNAsequences within their own species groups for both lengthand sequence identity (Table 2), although sample sizeswere small. On the other hand, P. mutuki and P. tubercul-osa groups showed much higher levels of ITS-rDNA lengthand sequence variation within their respective speciesgroups (Table 2). P. tuberculosa ITS-rDNA sequences,

123

Coral Reefs (2007) 26:399–410 405

while monophyletic and distinct from other Palythoaspecies groups, not only showed variability in length andsequences both between samples (Tables 2, 3) but alsobetween cloned sequences within individuals (maximumvariation 11.9% [74/621 bp] between clones [n = 14] fromsample PtIriomoteHoshi1). For all four species groups,ITS-1 was found to be the most variable region, followedby ITS-2 and then 5.8S rDNA, respectively (Table 2).

Discussion

The utility of ITS-rDNA in species delineation of the genus Palythoa

ITS-rDNA has diVerent levels of divergence within diVer-ent genera of hard corals [i.e., only 2% variation in theMontastraea annularis complex (Medina et al. 1999), 4.9%

in Madracis spp. (Diekmann et al. 2001), 11% in Poritesspp. (Hunter et al. 1997), and almost 60% reported in ITS-2rDNA of Acropora spp. (Marquez et al. 2003)]. ITS-rDNAin Palythoa spp., while suYcient for separating speciesgroups from one another, does not have the huge variabilityseen previously between Zoanthus spp., where diVerencesbetween species groups of up to approximately 70% of basepairs in the ITS-1 region made alignment impossible (seeReimer et al. 2007). As ITS-rDNA variation levels widelyvary among hexacorallian genera, it is somewhat diYcult toconWdently assign the level of taxonomic relationship(s)seen between P. tuberculosa, P. mutuki, P. heliodiscus, andP. sp. sakurajimensis based solely on ITS-rDNA sequences.However, with an examination of both ITS-rDNA andmitochondrial marker data (mtDNA 16S rDNA here andCOI—see Reimer et al. 2006c) from Palythoa, it wouldappear that these four groups have a congeneric-level rela-tionship. DiVerences in interspeciWc variation of ITS-rDNA

Fig. 3 Maximum likelihood tree of mitochondrial 16S ribosomal DNA (mtDNA 16S rDNA) sequences. Values at branches represent ML and NJ bootstrap probabilities, respec-tively (>50%). Bayesian poster-ior probabilities of >95% are represented by thick branches. For sample name abbreviations see Table 1

69/58

54/56

72/61

60/--

83/--

82/78

99/96

98/95

98/98

97/74

100/100

ZsSH23 (AB219187)ZsES1 (Z. sansibaricus)

ZgYS1(Z. gigantus) (AB219192)ZkYS1 (AB219191)

ZSH50 (Z. kuroshio group)Z

oanthus

PhIshigakiKata2

PhIshigakiKata5PhSaipanLauLau2

PhSaipanLauLau1PhErabuKaito1 (AB219224)

PhIshigakiKata11

P. heliodiscus

PErabuWanjoN1

PSakuraHakama1PWakayamaShira1

P. sp. sakurajimensis

PtAmamiTomori2

PtChibishiNaganu1

PtYoronShin1 (AB219219)

PtKeramaParadise1

PtKeramaKuro3

PtAmamiBashayama2

PtMiyakeIzu1 (AB219218)PtYoronChabana2

PtYoronShin2PtIshigakiKata6

PtErabuSumiyoshi2

PtErabuWanjo2

PtIriomoteHoshi1

PtOtsukiNishidomari1

PtIsrael1

PtYakuSango3

PtOtsukiNishidomari3

PtErabuWanjo3

PtIsrael2

PtIsshigakiKata1PtOtsukiNishidomari2

PtOtsukiFutabae1

PtHachijojimaBora2

PtChibishiNaga2

PtWakayamaKushi1

PtSaipanLauLau1

PtIshigakiKata4

PtYakuSango4

PtAmamiBashayama1

PMadagascar289

P. tuberculosa group

PmErabuSumiyoshi1PmMiyakeIzu1 (AB219225)

PmYakuSango1 (AB219224)PmYakuSango2

PmIriomoteHoshi1PmIriomoteHaemida3

PmAmamiTomori1 (AB219220)

PmAmamiTomori2 (AB219221)

P. mutuki group

P. mutuki 2

P. mutuki 1

P. tuberculosa group

0.002 substitutions/site ML/NJ

123

406 Coral Reefs (2007) 26:399–410

within a genus (relatively lower in Palythoa and higher inZoanthus) may be due to the examined Palythoa spp. beingmore recently diverged than Zoanthus spp., or due to muchfaster ITS-rDNA evolution in Zoanthus. How useful ITS-rDNA is in examining congeners from other zoanthid gen-era besides Palythoa and Zoanthus remains to be seen,although preliminary data indicate it may be useful in dis-tinguishing between Parazoanthus spp. and other genera inthe former suborder Brachycnemina (F. Sinniger, personalcommunication).

5.8S rDNA sequences in the ITS-rDNA region, whichclearly resolved the Zoanthus congeners Z. sansibaricus,Z. kuroshio, and Z. gigantus (Reimer et al. 2007), showedno variation between P. tuberculosa and P. mutuki. Simi-larly, mtDNA COI sequence diVerences between P. tuber-culosa and P. mutuki were not observed, and while mtDNA16S rDNA does distinguish between P. tuberculosa and P.mutuki, this is only by 1–2 bp (Reimer et al. 2006c). Forthese reasons it is concluded that the best means of speciWcPalythoa identiWcation is the analyses of ITS-rDNA (inparticular ITS-1 and/or ITS-2) sequences with conWrmatorymtDNA 16S rDNA sequences.

Palythoa phylogeny, and comparison with ecological data

Phylogenetic data based on mitochondrial markers (COIand 16S rDNA) suggest Protopalythoa and Palythoa areone genus [=Palythoa (Reimer et al. 2006c)], and the ITS-rDNA analyses in this study further conWrm this hypothe-sis. ITS-rDNA was observed to be more variable thanmtDNA 16S rDNA (Table 2). In the ITS-rDNA phyloge-netic analyses, P. tuberculosa is sister to P. heliodiscus andP. sp. sakurajimensis but more closely related to P. mutuki(Fig. 4) than to P. heliodiscus or P. sp. sakurajimensis. Inthe mtDNA 16S rDNA tree, P. tuberculosa is derived fromP. sp. sakurajimensis, and P. mutuki derived from P. tuber-culosa (Fig. 3). Despite the ITS-rDNA tree having thesediVerences in topology from the mtDNA 16S rDNA tree,ITS-rDNA sequences again clearly show the close relation-ship between P. tuberculosa and P. mutuki (Fig. 4). Fromboth the mtDNA 16S rDNA and ITS-rDNA results it isconcluded that P. mutuki (liberae ‘Protopalythoa’ morphol-ogy) is more closely related to P. tuberculosa (immersae‘Palythoa’ morphology) than to other species with Protop-alythoa morphology. These results also support the hypoth-esis posed by Reimer et al. (2006b) that in zoanthids grossmorphology (i.e., polyp shape and colony features) oftendoes not reXect relatedness among congeners.

The close relationship inferred between P. tuberculosaand P. mutuki based on DNA sequence data reXects theirsimilar ecological and distribution data, as they often occursympatrically in shallow areas of high sunlight, water Xowand wave activity. P. heliodiscus appears to be somewhatT

able

2In

tras

peciW

c se

quen

ce v

aria

tion

leve

ls o

f IT

S-rD

NA

and

mt 1

6S r

DN

A f

or P

alyt

hoa

tube

rcul

osa,

P. m

utuk

i, P

. sp.

sak

uraj

imen

sis

and

P. h

elio

disc

us

aN

ote

that

IT

S-1

and

ITS

-2 r

DN

A s

eque

nce

leng

ths

are

diV

eren

t tha

n in

Tab

le3

as s

eque

nces

her

e ar

e al

igne

d w

hile

leng

ths

in T

able

3 ar

e fr

om u

nalig

ned

sequ

ence

s

DN

A r

egio

nIT

S-1a

ITS

-2a

5.8S

rD

NA

ITS-

1+

5.8S

+IT

S-2

mtD

NA

16S

rD

NA

Spe

cies

/gro

upn

DiV

eren

ces/

tota

l ba

se p

airs

%

DiV

eren

ceD

iVer

ence

s/to

tal

base

pai

rs%

D

iVer

ence

DiV

eren

ces/

tota

l ba

se p

airs

%

DiV

eren

ceD

iVer

ence

s/to

tal

base

pai

rs%

D

iVer

ence

DiV

eren

ces/

tota

l ba

se p

airs

%

DiV

eren

ce

P. t

uber

culo

sa50

96/2

9932

.151

/190

26.8

6/15

63.

815

3/64

623

.72/

760

0.3

P. m

utuk

i7

82/3

2125

.526

/174

14.9

0/15

60.

010

8/65

116

.62/

760

0.3

P. s

p. s

akur

ajim

ensi

s2

3/38

90.

80/

170

0.0

0/15

60.

03/

715

0.4

2/76

00.

3

P. h

elio

disc

us6

20/3

515.

78/

197

4.1

1/15

60.

129

/704

4.1

2/76

00.

3

123

Coral Reefs (2007) 26:399–410 407

more distantly related to the other three Palythoa groupsand ecologically this species is also somewhat diVerentfrom the others, as it is often found in deeper areas withmuch lower sunlight, and thus appears to be more mixo-trophic, obtaining energy from both its symbiotic zooxan-

thellae and perhaps directly from particles in the watercolumn.

Despite extensive sampling throughout the course of thisstudy, only three samples of P. sp. sakurajimensis werefound, suggesting that this species may be less common ormore cryptic than other Palythoa species in the sub-tropicaland tropical waters of Japan. More samples are neededbefore the ecology of this potentially new species group canbe discussed, but it is noteworthy that all three sampleswere found in habitats where other Palythoa species werenot found [e.g., at the relatively cold locations (winterminimums approximately 15°C) of Sakurajima and Shira-hama, and relatively high in the tidal zone at Wanjo North,Okinoerabu Island]. This species may be more adapted to“marginal” or variable conditions than other Palythoa spe-cies, similar to Zoanthus sansibaricus (found at the colderSakurajima site) compared to other Zoanthus species inJapan (Reimer et al. 2006a).

Fig. 4 Maximum likelihood tree of internal transcribed spac-er ribosomal DNA (ITS-rDNA) sequences. Values at branches represent ML and NJ bootstrap probabilities, respectively (>50%). Bayesian posterior probabilities of >95% are repre-sented by thick branches. For sample name abbreviations see Table 1. Sample names with Accession Numbers are from previous studies (see Table 1). Sample names with a hyphen-ated number or letter ending rep-resent clones. Samples with Wlled in target symbols are also shown in the alignment in Fig. 5

PtIriomoteHoshi1-9PtIriomoteHoshi1-14PtIriomoteHoshi1-8PtIriomoteHoshi1-1PtIriomoteHoshi1-7PtIriomoteHoshi1-4PtIriomoteHoshi1-2

80/69

74/61

PtOtsukiNishidomari3PtErabuWanjo3

PtAmamiTomori2PtYakuSango4

PtIriomoteHoshi1-10PtWakayamaKushi1

72/82

79/80 PtAmamiBasahyama2PtHachijojimaBora282/92PtOtsukiFutabae1-6PtOtsukiFutabae1-8PtOtsukiFutabae1-5

100/6666/97

PtIriomoteHoshi1-6100/99PtIriomoteHoshi1-11PtIriomoteHoshi1-5PtIriomoteHoshi1-3PtIriomoteHoshi1-13PtIriomoteHoshi1-12

96/7772/--

59/--

PtSaipanLauLau1PtAmamiBashayama1PtIshigakiKata1

93/96

86/--

95/97

PtYoronChabana2PtYoronShin1

92/72

98/99

59/--

PmErabuSumiyoshi173/79

99/10071/77

96/92

96/71

PWakayamaShira1

98/61

99/74

88/--66/5353/70

100/100

--/86

0.1 substitutions/site ML/NJ

PhSaipanLauLau1PhIshigakiKata3

PhErabuKaito1PhSaipanLauLau2PhIshigakiKata11PhIshigakiKata2

P. heliodiscus

PSakuraHakama1 P. sp. sakurajimensis100/100

PmIriomoteHoshi1PmIriomoteHaemida3

P. mutuki 2

PmMiyakeIzu1 P. mutuki 1

PmAmamiTomori1 P. mutuki 2PmYakuSango1PmYakuSango2 P. mutuki 1

PcHond1-3PcHond1-8PcHond1-4

PtIsrael1-5PtIsrael1-2

PtIsrael1-3

PMadagascar289

PtIshigakiOnsen1-3PtIshigakiOnsen1-1

PtYakuSango4-7PtYakuSango4-4PtYakuSango4-1PtYakuSango4-8PtYakuSango4-2

PtOtsukiFutabae1-4PtOtsukiFutabae1-7

PtOtsukiFutabae1-1

PtChibishiNaga1-4PtChibishiNaga1-5

PtChibishiNaga1-3PtChibishiNaga1-1PtChibishiNaga1-7

PtChibishiNaga1

P. tuberculosa group

Table 3 Length variation of diVerent regions of obtained ITS-rDNAsequences for diVerent Palythoa spp

a Note that ITS-1 and ITS-2 rDNA sequence lengths are diVerent thanin Table 2; sequence lengths here are from unaligned sequences, whileTable 2 shows aligned sequence lengths

Species/group n ITS-1a 5.8S ITS-2a

P. tuberculosa 50 243–271 156 171–192

P. mutuki 7 283–312 156 168–174

P. sp. Sakurajimensis 2 389 156 170

P. heliodiscus 6 339 156 203

123

408 Coral Reefs (2007) 26:399–410

Implications of ITS-rDNA variation within P. tuberculosa and P. mutuki

Veron (1995) described a theory of reticulate evolution inhard corals. Under this scenario, species groups undergo

repeated sexual reproductive isolation, diVerentiation, andsecondary contact based on continuously changing patternsof distribution, perhaps due to changing ocean currents and/or ocean levels. Many hard corals and other Anthozoa(including at least some zooxanthellate Zoantharia—see

Fig. 5 Alignment of the ITS-rDNA region for various Palythoa mutu-ki and Palythoa tuberculosa species groups’ sequences. Areas in openboxes represent shared common areas of sequence variation betweenspecies groups, and shaded gray areas represent “minor” areas of se-quence variation within a species group (diVerent from the majority ofsequences within a species group). Thus, other areas of variation in theP. tuberculosa species group not marked with an open box or a gray

area represent diVerences between all P. mutuki and P. tuberculosa se-quences shown. Sample names are abbreviated due to limited space inthe alignment: PmMiI1 PmMiyakeIzu1, PmES1 PmErabuSumiyoshi1,PmIrHo1 PmIriomoteHoshi1, PmAT1 PmAmamiTomori1, PtYoS1PtYoronShin1, PtYoC2 PtYoronChabana2, PtSaiLL1 PtSaipanLau-Lau1, PtIrHo1-6 and 1–10 PtIriomoteHoshi1 clones 6 and 10, PtYS4PtYakuSango4

123

Coral Reefs (2007) 26:399–410 409

Ono et al. 2005) undergo mass spawning events (Levitanet al. 2004; Penland et al. 2004), providing an opportunityfor hybridization and reticulate evolution. Many studieshave detailed molecular and/or reproductive evidence ofpotential reticulate patterns in zooxanthellate hard corals(e.g., Odorico and Miller 1997; Hatta et al. 1999; Medinaet al. 1999; van Oppen et al. 2000, 2002; Diekmann et al.2001).

In this study there were moderate amounts of intraspe-ciWc variation in ITS-rDNA sequences (see Table 3) fromP. mutuki and both intraspeciWc and intragenomic variationfrom P. tuberculosa. An examination of the ITS-rDNAsequence alignment reveals potential “reticulate evolution”patterns, such as additive polymorphic sites (APS).

Additive polymorphic sites can potentially indicatehybridization and reticulate evolution, particularly in ITS-rDNA (Sang et al. 1995; Campbell et al. 1997; Whittallet al. 2000; Aguilar and Feliner 2003; Feliner et al. 2004).One kind of APS that hints at hybridization is when somesequences share portions of their sequences with sequencesnormally found only within another clade (Aguilar andFeliner 2003). In the case of P. tuberculosa, samples PtYo-ronShin1 and PtYoronChabana2 show such a pattern. BothITS-rDNA sequences are clearly in the P. tuberculosaclade, yet both have two regions that are found inP. mutuki, and not in any other P. tuberculosa sequences(Fig. 5, marked regions 2 and 6). Figure 5 shows 13 mainpotential ITS-rDNA regions shared between P. mutuki andP. tuberculosa that have this type of APS (numbered areas).The question of whether PtYoronShin1 and PtYoronCha-bana2 are P. tuberculosa–P. mutuki hybrids or not remainsto be investigated, as although PtYoronShin1 displayed“intermediae” polyps more clear and free of the coenenc-hyme than most P. tuberculosa specimens and reminiscentof P. mutuki, PtYoronChabana2 had “immersae” polypscommon to P. tuberculosa.

Another kind of APS that hints at a reticulate history iswhen observed intragenomic variation patterns (such as anindel, etc.) are also found in other individuals (Sang et al.1995; Campbell et al. 1997; Whittall et al. 2000; Aguilarand Feliner 2003). This APS type can also be seen in theITS-rDNA alignment (Fig. 5, grey regions). For example,P. tuberculosa PtIriomoteHoshi1 clones (PtIriomoteHo-shi1-6, PtIriomoteHoshi1-10, i.e., distinct intragenomicsequences) have many variable regions diVerent from eachother that are also found in P. tuberculosa sequences fromother samples (Fig. 5). It is possible that the results are dueto slow concerted evolution with the presence of multipleITS-rDNA copies (ancestral polymorphism). However,variable ITS-rDNA regions of PtIriomoteHoshi1 matchwith regions found in other individuals (not only with justother P. tuberculosa specimens but even with P. mutukisequences—see region 4—Fig. 5), supporting the hybrid-

ization hypothesis. The data suggest that P. tuberculosa andP. mutuki potentially have a history of reticulate evolution.

The data further show that morphology between evenclosely related zoanthid congeners can vary drastically (asbetween P. tuberculosa and P. mutuki) and highlight theneed for additional genetic investigation of zoanthid diver-sity. To investigate more fully the possibility of a potentialreticulate evolutionary history in Palythoa, studies of sex-ual reproduction timing and patterns (as conducted in Zoan-thus, Ono et al. 2005) need to be conducted. The moleculartechniques used in this study to distinguish between cong-eners and identify a potential reticulate evolutionary historymay provide a reliable method for analyses of other Zoan-tharia groups.

Acknowledgments The authors wish to thank the following peoplefor their support and help throughout the course of this study. Dr. JunzoTsukahara (Kagoshima University) for his continuing support of zo-anthid research; John Byron, Ryan Cox, Dr. Yasuo Furushima (JAMS-TEC), Shin Hisashi, Yuichi Nagata, Hideo Noda, Denny Probizanski,Mika Reimer, Kei Sasahara, and Yuki Wada for their help in collectingsamples in Japan; Dr. Hironobu Fukami (Kyoto University) for his ad-vice on the nature of the ITS-rDNA region; Omer Polak (Interuniver-sity for Marine Science, Eilat, Israel) for kindly providing samplesfrom Eilat; Frederic Sinniger (University of Geneva) for productivediscussions, comments on this article, and samples from Honduras andMadagascar; and Masoru Kawato and Hirokazu Kuwahara for helpwith sequencing at JAMSTEC. JDR was supported at JAMSTEC by aJapan Society for the Promotion of Science Post-doctoral fellowship(#P04868). Dr. Dhugal Lindsay (JAMSTEC) helped with English edit-ing. Fumihito Iwase (BIK) gave us invaluable advice on scientiWcnomenclature. Three anonymous reviewers greatly improved the qual-ity of this manuscript.

References

Aguilar JF, Feliner GN (2003) Additive polymorphisms and reticula-tion in an ITS phylogeny of thrifts (Armeria, Plumbaginaceae).Mol Phylogenet Evol 28:430–447

Campbell CS, Wojciechowski MF, Baldwin BG, Alice LA, DonoghueMJ (1997) Persistent nuclear ribosomal DNA sequence polymor-phism in the Amelanchier agamic complex (Rosaceae). Mol BiolEvol 14:81–90

Diekmann OE, Bak RPM, Stam WT, Olsen JL (2001) Moleculargenetic evidence for probably reticulate speciation on the coralgenus Madracis from a Caribbean fringing reef slope. Mar Biol139:221–233

Fautin DG (2006) Hexacorallians of the World. http://www.geoportal.kgs.ku.edu/hexacoral/anemone2/index.cfm

Feliner GN, Larena BG, Aguilar JF (2004) Fine-scale geographicalstructure, intra-individual polymorphism and recombination innuclear ribosomal internal transcribed spacers in Armeria(Plumbaginaceae). Ann Bot 93:189–200

Fukami H, Budd AF, Levitan DR, Jara J, Kersanach R, Knowlton N(2004) Geographic diVerences in species boundaries amongmembers of the Montastraea annularis complex based on molec-ular and morphological markers. Evolution 58:324–327

Gascuel O (1997) BIONJ: an improved version of the NJ algorithmbased on a simple model of sequence data. Mol Biol Evol 14:685–695

123

410 Coral Reefs (2007) 26:399–410

Guindon S, Gascuel O (2003) A simple, fast, and accurate algorithm toestimate large phylogenies by maximum likelihood. Syst Biol52:696–704

Hasegawa M, Kishino H, Yano T (1985) Dating of the human-apesplitting by a molecular clock of mitochondrial DNA. J Mol Evol22:160–174

Hatta M, Fukami H, Wang W, Omori M, Shimoike K, Hayashibara T,Ina Y, Sugiyama T (1999) Reproductive and genetic evidence fora reticulate evolutionary history of mass-spawning corals. MolBiol Evol 16:1607–1613

Hunter CL, Morden CW, Smith CM (1997) The utility of ITS sequencesassessing relationships among zooxanthellae and corals. Proc 8thInt Coral Reef Symp 2:1559–1602

Levitan DR, Fukami H, Jara J, Kline D, McGovern TM, McGhee KE,Swanson CA, Knowlton N (2004) Mechanisms of reproductiveisolation among sympatric broadcast-spawning corals of theMontastraea annularis species complex. Evolution 58:308–323

Marquez LM, Miller DJ, MacKenzie JB, van Oppen MJH (2003)Pseudogenes contribute to the extreme diversity of nuclear ribo-somal DNA in the hard coral Acropora. Mol Biol Evol 20:1077–1086

Medina M, Weil E, Szmant AM (1999) Examination of the Montast-raea annularis species complex (Cnidaria: Scleractinia) usingITS and COI sequences. Mar Biotechnol 1:89–97

Odorico DM, Miller DJ (1997) Variation among the ribosomal InternalTranscribed Spacers and 5.8S rDNA among Wve species of Acro-pora (Cnidaria; Scleractinia): Patterns of variation consistent withreticulate evolution. Mol Biol Evol 14:465–473

Ono S, Reimer JD, Tsukahara J (2005) Reproduction of Zoanthussansibaricus in the infra-littoral zone at Taisho Lava Field, Sak-urajima, Kagoshima, Japan. Zool Sci 22:247–255

Pax F (1910) Studien an westindischen Actinien. In Spengel JW (ed)Ergebnisse einer Zoologischen Forschungreise nach Westindienvon Prof. W. Kukenthal und Dr. R. Hartmeyer im Jahre, 1907.Zoologische Jahrbucher Supplement 11, pp 157–330

Penland L, Kloulechad J, Idip D, van Woesik R (2004) Coral spawningin the western PaciWc Ocean is related to solar insolation: evi-dence of multiple spawning events in Palau. Coral Reefs 23:133–140

Posada D, Crandall KA (1998) MODELTEST: testing the model ofDNA substitution. Bioinformatics (Oxf) 14:817–818

Reimer JD, Ono S, Iwama A, Takishita K, Tsukahara J, Maruyama T(2006a) Morphological and molecular revision of Zoanthus (An-thozoa: Hexacorallia) from southwestern Japan with descriptionof two new species. Zool Sci 23:261–275

Reimer JD, Ono S, Iwama A, Tsukahara J, Maruyama T (2006b) Highlevels of morphological variation despite close genetic related-ness between Zoanthus aV. vietnamensis and Zoanthus kuroshio(Anthozoa: Hexacorallia). Zool Sci 23:755–761

Reimer JD, Ono S, Takishita K, Tsukahara J, Maruyama T (2006c)Molecular evidence suggesting species in the zoanthid genera

Palythoa and Protopalythoa (Anthozoa: Hexacorallia) are conge-neric. Zool Sci 23:87–94

Reimer JD, Takishita K, Ono S, Tsukahara J, Maruyama T (2007)Molecular evidence suggesting intraspeciWc hybridization inZoanthus (Anthozoa: Hexacorallia). Zool Sci 24 (in press)

Rodriguez F, Oliver JL, Marin A, Medina JR (1990) The general sto-chiatic model of nucleotide substitution. J Theor Biol 142:485–501

Ronquist F, Huelsenbeck JP (2003) Bayesian phylogenetic inferenceunder mixed models. Bioinformatics (Oxf) 19:1572–1574

Ryland JS, Lancaster JE (2003) Revision of methods of separating spe-cies of Protopalythoa (Hexacorallia: Zoanthidea) in the tropicalWest PaciWc. Invert Systemat 17:407–428

Saitou N, Nei M (1987) The neighbor-joining method: a new methodfor reconstructing phylogenetic trees. Mol Biol Evol 4:406–425

Sang T, Crawford DJ, Stuessy TF (1995) Documentation of reticulateevolution in peonies (Paeonia) using internal transcribed spacersequences of nuclear ribosomal DNA: implications for biogeog-raphy and concerted evolution. Proc Natl Acad Sci USA92:6813–6817

Shearer TL, van Oppen MJH, Romano SL, Worheide G (2002) Slowmitochondrial DNA sequence evolution in the Anthozoa (Cnida-ria). Mol Ecol 11:2475–2487

Sinniger F, Montoya-Burgess JI, Chevaldonne P, Pawlowski J (2005)Phylogeny of the order Zoantharia (Anthozoa, Hexacorallia)based on mitochondrial ribosomal genes. Mar Biol 147:1121–1128

SwoVord DL (1998) PAUP*. V Phylogenetic analysis using parsi-mony (*and other methods), version 4. Sinauer Associates, Sun-derland

Tamura K, Nei M (1993) Estimation of the number of nucleotide sub-stitutions in the control region of mitochondrial DNA in humansand chimpanzees. Mol Biol Evol 10:512–526

Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG(1997) The ClustalX windows interface: Xexible strategies formultiple sequence alignment aided by quality analysis tools.Nucleic Acids Res 25:4876–4882

van Oppen MJH, Willis BL, Van Vugt HWJA, Miller DJ (2000)Examination of species boundaries in the Acropora cervicornisgroup (Scleractinia, Cnidaria) using nuclear DNA sequence anal-yses. Mol Ecol 9:1363–1373

van Oppen MJH, Willis BL, Rheede TV, Miller DJ (2002) Spawningtimes, reproductive compatibilities and genetic structuring in theAcropora aspera group: evidence for natural hybridization andsemi-permeable species boundaries in corals. Mol Ecol 11:1363–1376

Veron JEN (1995) Corals of the world: the biogeography and evolutionof the Scleractinia. University of New South Wales Press, Sydney

Whittall J, Liston A, Gisler S, Meinke RJ (2000) Detecting nucleotideadditivity from direct sequences is a SNAP: an example fromSidalcea (Malvaceae). Plant Biol 2:1–7

123