Speciation Versus Phenotypic Plasti

Speciation Versus Phenotypic Plasticity in Coral Inhabiting Barnacles:
Darwin’s Observations in an Ecological Context
O. Mokady,
1
Y. Loya,
2
Y. Achituv,
3
E. Geffen,
1
D. Graur,
2
S. Rozenblatt,
4
I. Brickner
2,3
1
Institute for Nature Conservation Research, The George S. Wise Faculty of Life Sciences, Tel Aviv University, Tel Aviv 69978, Israel
2
Department of Zoology, The George S. Wise Faculty of Life Sciences, Tel Aviv University, Tel Aviv 69978, Israel
3
Faculty of Life Sciences, Bar Ilan University, Ramat Gan 52900, Israel
4
Department of Molecular Microbiology and Biotechnology, The George S. Wise Faculty of Life Sciences, Tel Aviv University,
Tel Aviv 69978, Israel
Received: 18 January 1999 / Accepted: 9 May 1999
Abstract. Speciation and phenotypic plasticity are two
extreme strategic modes enabling a given taxon to populate a broad ecological niche. One of the organismal
models which stimulated Darwin’s ideas on speciation
was the Cirripedia (barnacles), to which he dedicated a
large monograph. In several cases, including the coralinhabiting barnacle genera Savignium and Cantellius
(formerly PyrgomaandCreusia,respectively), Darwin
assigned barnacle specimens to morphological “varieties” (as opposed to species) within a genus. Despite having been the subject of taxonomic investigations and revisions ever since, the significance of these varieties has
never been examined with respect to host-associated speciation processes. Here we provide evidence from molecular (12S mt rDNA sequences) and micromorphological (SEM) studies, suggesting that these closely related
barnacle genera utilize opposite strategies for populating
a suite of live-coral substrates.Cantelliusdemonstrates a
relatively low genetic variability, despite inhabiting a
wide range of corals. The speciesC. pallidusalone was
found on three coral families, belonging to distinct
higher-order classification units. In contrast,Savignium
barnacles exhibit large between- and within-species
variations with respect to both micromorphology and
DNA sequences, withS. dentatum“varieties” clustering
phylogenetically according to their coral host species (all
of which are members of a single family). Thus, whereas
Savignium seems to have undergone intense hostassociated speciation over a relatively narrow taxonomic
range of hosts, Cantelliusshows phenotypic plasticity
over a much larger range. This dichotomy correlates with
differences in life-history parameters between these barnacle taxa, including host-infestation characteristics, reproductive strategies, and larval trophic type.
Key words: Phylogenetic reconstruction — Speciation — Phenotypic plasticity — Pyrgomatine barnacles
—Cantellius — Savignium— 12S mt rDNA
Introduction
The high level of diversity among coral reef invertebrates and its potential for improving our understanding
of the evolutionary mechanisms governing speciation
processes have attracted considerable attention in recent
years (e.g., Romano and Palumbi 1996). Advanced
methodologies, including high-resolution morphological
techniques [e.g., scanning electron microscopy (SEM)
studies] and molecular tools, enable identification of
complexes of sibling species (Knowlton et al. 1992; Van
Veghel and Bak 1993; Knowlton 1993) for readdressing
ecological and evolutionary questions (Knowlton and
Jackson 1994). Correspondence to:O. Mokady:e-mail:[e-mail]
J Mol Evol (1999) 49:367–375
©Springer-Verlag New York Inc. 1999
Here we examine coral-inhabiting barnacles (subfamily Pyrgomatinae) belonging to two genera (Savignium
andCantellius). Obligate coral symbionts in this group
are said to have become markedly specialized for living
within a continuously growing substratum, such as living
coral colonies, in both morphological and growth characteristics (Ross and Newman 1973; Young and Christoffersen 1984).
In barnacles, which reproduce by internal fertilization,
adaptations for epizoic life that promote larval host
specificity will form a reproductive barrier. Reproductive
isolation, in turn, will lead to speciation (Templeton
1989). Alternatively, adaptations which do not promote
host specificity will probably lead to some degree of
phenotypic plasticity, in response to variations among
the coral hosts. Savignium,reported to be specific to
coral suborders (Ross and Newman 1973) and genera
(Ogawa and Matsuzaki 1992), is considered to demonstrate a higher degree of host specificity thanCantellius.
These two pyrgomatine barnacles were therefore chosen
as model systems for contrasting speciation and phenotypic plasticity.
Coral-inhabiting barnacles were reported to exhibit
substantial phenotypic plasticity by Darwin (1854) and
numerous times since. Darwin’s assignment of the three
Savignium dentatum“varieties” is embedded in the current taxonomy of barnacles, which is considered a wellfounded one. A number of studies have subsequently
assigned barnacles to these “varieties,” with no attempt
to interpret the observed differences ecologically (e.g.,
Hiro 1935; Foster 1980; Soong and Chang 1983).
Barnacle taxonomy, fundamentally established over a
century ago by Darwin (1854), has only recently been
reevaluated with the aid of molecular tools (e.g., Van
Syoc 1995). In the current study we use molecular data
(12S mt rDNA sequences), in conjunction with SEM
observations, to test for indications of speciation within
acknowledged species (sibling species). We use the data
to reexamine some of the ecological conclusions that
have traditionally been drawn upon the currently accepted taxonomy and correlate the findings with several
known life-history characteristics of these barnacles.
Materials and Methods
Animal Collection and SEM Observations
Fragments of coral colonies inhabited by barnacles were observedin
situand collected at a depth of 1–30 m near the northern tip of the Gulf
of Eilat, Red Sea, Israel. Twelve species of scleractinian corals and one
hydrocoral (Millepora dichotoma), hosting six currently recognized
species of barnacles (fourSavigniumand twoCantelliusspecies) were
sampled (see Table 1 for a complete list of hosts and symbionts).
Barnacle shells were carefully removed from the coral and the soft
parts and calcareous parts (i.e., shell plates and valves) were separated
for identification and description (Brickner 1994). Shells and valves of
coral-inhabiting barnacles were dehydrated in an alcohol series and
coated with gold–palladium. Specimens thus prepared were examined
by SEM (JEOL-840).
In addition, we collected two species of rock-inhabiting barnacles,
Tetraclita squamosaand Balanus amphitrite,from nearby intertidal
rocks, to serve as outgroups for the molecular analysis.
DNA Preparation, Amplification, and Sequencing
To extract total cellular DNA, the whole soft tissue of individual barnacles was homogenized in lysis buffer (10 mMTris–HCl pH 8.0, 100
mMNaCl, 20 mMEDTA, 0.5% lauryl sarcosine). The lysate was
digested for1hbyproteinase K (25–50mg/ml) at 55°C and extracted
with phenol:chloroform (1:1). Nucleic acids were precipitated overnight with 0.1 vol of 3Msodium acetate and 2 vol of 100% ethanol, at
−20°C. The pelleted nucleic acids were washed in 100% ethanol, dried,
and resuspended in 100mlH2
O.
The polymerase chain reaction (PCR) was employed to amplify a
fragment of the 12S subunit of the mitochondrial rDNA using the
Table 1. Red Sea coral-inhabiting and free-living barnacles examined in this study and their typical substrates
Barnacle species Substrate
Savignium dentatum Cyphastrea chalcidicum
Favites abdita
Favia favus
Platygyra lamellina
Savignium elongatum Echinopora gemmacea
Savignium crenatum Acanthastrea echinata
Platygyra lamellina
Savignium milleporum Millepora dichotoma
Cantellius pallidus Cyphastrea chalcidicum
Montipora erythraea
Pavona cactus
Cantellius arcuatus Porites lobata
Pocillopora damicornis
Tetraclita squamosa Intertidal rock (free living)
Balanus amphitrite Intertidal rock (free living)
>
Fig. 1. Partial sequence of the 12S mt rDNA from coral-inhabiting
barnacles and free-living barnacles (EMBL accession numbers
X78234–X78254). The substrate inhabited by each barnacle species is
indicated in parentheses. Each sequence represents an individual barnacle, collected from a separate coral colony or intertidal rock, except
for the sequences of S. milleporumandS. dentatum(Fav) I, each of
which represents two identical sequences obtained from two individuals from separate colonies. A dotin a sequence indicates that the
nucleotide in this position is the same as in theC. pallidus (Cyp)
sequence. Stem coding regions are indicatedabovethe sequences, and
the numbers (32–48) correspond to the numbering used by Hickson et
al. (1996). Complementary sequences assumed to form a stem are
marked by the same number (e.g., 32 and 328).Asterisksdenote positions for which full complementarity is observed between stemforming sequences. The “A” at position 31 and the “T” at position 327
(underlined)correspond to positions 1160 and 1468 in the human sequence (Anderson et al. 1981), respectively.Aca, Acanthastrea echinata; Cyp, Cyphastrea chalcidicum; Ech, Echinopora gemmacea; Fav,
Favia favus; Fat, Favites abdita; Mil, Millepora dichotoma; Mon,
Montipora erythraea; Pav, Pavona varians; Pla, Platygyra lamellina;
Poc, Pocillopora damicornis; Por, Porites lobata.
368
369
primer set of Kocher et al. (1989) as modified by Mokady et al. (1994):
58-GAAACCAGGATTAGATACCC and 58-TTTCCCGCGAGCGACGGGCG. The reaction buffer consisted of 10 mMTris–HCl (pH
9.0), 50 mMKCl, 0.1% Triton X-100, and 3.5 mMMgCl2
. Fifty-five
picomoles of each primer was added for each reaction, along with 2.5
U of Taq DNA polymerase (Promega, Madison, WI), a 300mMconcentration of each dNTP, and 1ml of template DNA solution in a total
volume of 100ml. The PCR cycle consisted of 2 min of denaturation
at 92°C, 2 min of annealing at 54°C, and 3 min of elongation at 72°C.
This cycle was repeated 29 times, with a final cycle in whi

Hasil (Bahasa Indonesia) 1:

Spesiasi Versus fenotipik plastisitas di Barnacles karang menghuni:Pengamatan Darwin dalam konteks ekologiO. Mokady,1Y. Loya,2Y. Achituv,3E. Geffen,1D. Graur,2S. Rozenblatt,4I. Brickner2,31Institut Riset pelestarian alam, George S. Wise Fakultas 69978 Life Sciences, Universitas Tel Aviv, Tel Aviv, Israel2Departemen Zoologi, George S. Wise Fakultas 69978 Life Sciences, Universitas Tel Aviv, Tel Aviv, Israel3Fakultas Ilmu kehidupan, Universitas Bar Ilan, Ramat Gan 52900, Israel4Departemen Mikrobiologi molekuler dan bioteknologi, Fakultas George S. Wise Life Sciences, Tel Aviv University,Tel Aviv 69978, IsraelMenerima: 18 Januari 1999 / diterima: 9 Mei 1999Abstrak. Spesiasi dan plastisitas fenotipik adalah duaekstrim mode strategis memungkinkan takson yang diberikan untuk mengisi ceruk ekologi yang luas. Salah satu organismemodel gagasan Darwin yang dirangsang spesiasiadalah Cirripedia (teritip), yang ia berdedikasibesar monografi. Dalam beberapa kasus, termasuk coralinhabiting barnacle genera Savignium dan Cantellius(sebelumnya PyrgomaandCreusia, masing-masing), Darwinditugaskan barnacle spesimen morfologi "varietas" (sebagai lawan dari spesies) dalam genus. Meskipun telah subjek penyelidikan taksonomi dan revisi sejak, pentingnya varietas ini memilikinever been examined with respect to host-associated speciation processes. Here we provide evidence from molecular (12S mt rDNA sequences) and micromorphological (SEM) studies, suggesting that these closely relatedbarnacle genera utilize opposite strategies for populatinga suite of live-coral substrates.Cantelliusdemonstrates arelatively low genetic variability, despite inhabiting awide range of corals. The speciesC. pallidusalone wasfound on three coral families, belonging to distincthigher-order classification units. In contrast,Savigniumbarnacles exhibit large between- and within-speciesvariations with respect to both micromorphology andDNA sequences, withS. dentatum“varieties” clusteringphylogenetically according to their coral host species (allof which are members of a single family). Thus, whereasSavignium seems to have undergone intense hostassociated speciation over a relatively narrow taxonomicrange of hosts, Cantelliusshows phenotypic plasticityover a much larger range. This dichotomy correlates withdifferences in life-history parameters between these barnacle taxa, including host-infestation characteristics, reproductive strategies, and larval trophic type.Key words: Phylogenetic reconstruction — Speciation — Phenotypic plasticity — Pyrgomatine barnacles—Cantellius — Savignium— 12S mt rDNAIntroductionThe high level of diversity among coral reef invertebrates and its potential for improving our understandingof the evolutionary mechanisms governing speciationprocesses have attracted considerable attention in recentyears (e.g., Romano and Palumbi 1996). Advancedmethodologies, including high-resolution morphologicaltechniques [e.g., scanning electron microscopy (SEM)studies] and molecular tools, enable identification ofcomplexes of sibling species (Knowlton et al. 1992; VanVeghel and Bak 1993; Knowlton 1993) for readdressingecological and evolutionary questions (Knowlton andJackson 1994). Correspondence to:O. Mokady:e-mail:[e-mail] Mol Evol (1999) 49:367–375©Springer-Verlag New York Inc. 1999Here we examine coral-inhabiting barnacles (subfamily Pyrgomatinae) belonging to two genera (SavigniumandCantellius). Obligate coral symbionts in this groupare said to have become markedly specialized for livingwithin a continuously growing substratum, such as livingcoral colonies, in both morphological and growth characteristics (Ross and Newman 1973; Young and Christoffersen 1984).In barnacles, which reproduce by internal fertilization,adaptations for epizoic life that promote larval hostspecificity will form a reproductive barrier. Reproductiveisolation, in turn, will lead to speciation (Templeton1989). Alternatively, adaptations which do not promotehost specificity will probably lead to some degree ofphenotypic plasticity, in response to variations amongthe coral hosts. Savignium,reported to be specific tocoral suborders (Ross and Newman 1973) and genera(Ogawa and Matsuzaki 1992), is considered to demonstrate a higher degree of host specificity thanCantellius.These two pyrgomatine barnacles were therefore chosenas model systems for contrasting speciation and phenotypic plasticity.Coral-inhabiting barnacles were reported to exhibitsubstantial phenotypic plasticity by Darwin (1854) andnumerous times since. Darwin’s assignment of the threeSavignium dentatum“varieties” is embedded in the current taxonomy of barnacles, which is considered a wellfounded one. A number of studies have subsequentlyassigned barnacles to these “varieties,” with no attemptto interpret the observed differences ecologically (e.g.,Hiro 1935; Foster 1980; Soong and Chang 1983).Barnacle taxonomy, fundamentally established over acentury ago by Darwin (1854), has only recently beenreevaluated with the aid of molecular tools (e.g., VanSyoc 1995). In the current study we use molecular data(12S mt rDNA sequences), in conjunction with SEMobservations, to test for indications of speciation withinacknowledged species (sibling species). We use the datato reexamine some of the ecological conclusions thathave traditionally been drawn upon the currently accepted taxonomy and correlate the findings with severalknown life-history characteristics of these barnacles.Materials and MethodsAnimal Collection and SEM ObservationsFragments of coral colonies inhabited by barnacles were observedinsituand collected at a depth of 1–30 m near the northern tip of the Gulfof Eilat, Red Sea, Israel. Twelve species of scleractinian corals and onehydrocoral (Millepora dichotoma), hosting six currently recognizedspecies of barnacles (fourSavigniumand twoCantelliusspecies) weresampled (see Table 1 for a complete list of hosts and symbionts).Barnacle shells were carefully removed from the coral and the softparts and calcareous parts (i.e., shell plates and valves) were separatedfor identification and description (Brickner 1994). Shells and valves ofcoral-inhabiting barnacles were dehydrated in an alcohol series andcoated with gold–palladium. Specimens thus prepared were examinedby SEM (JEOL-840).In addition, we collected two species of rock-inhabiting barnacles,Tetraclita squamosaand Balanus amphitrite,from nearby intertidalrocks, to serve as outgroups for the molecular analysis.DNA Preparation, Amplification, and SequencingTo extract total cellular DNA, the whole soft tissue of individual barnacles was homogenized in lysis buffer (10 mMTris–HCl pH 8.0, 100mMNaCl, 20 mMEDTA, 0.5% lauryl sarcosine). The lysate wasdigested for1hbyproteinase K (25–50mg/ml) at 55°C and extractedwith phenol:chloroform (1:1). Nucleic acids were precipitated overnight with 0.1 vol of 3Msodium acetate and 2 vol of 100% ethanol, at−20°C. The pelleted nucleic acids were washed in 100% ethanol, dried,and resuspended in 100mlH2O.The polymerase chain reaction (PCR) was employed to amplify afragment of the 12S subunit of the mitochondrial rDNA using theTable 1. Red Sea coral-inhabiting and free-living barnacles examined in this study and their typical substratesBarnacle species SubstrateSavignium dentatum Cyphastrea chalcidicumFavites abditaFavia favusPlatygyra lamellinaSavignium elongatum Echinopora gemmaceaSavignium crenatum Acanthastrea echinataPlatygyra lamellinaSavignium milleporum Millepora dichotomaCantellius pallidus Cyphastrea chalcidicumMontipora erythraeaPavona cactusCantellius arcuatus Porites lobataPocillopora damicornisTetraclita squamosa Intertidal rock (free living)Balanus amphitrite Intertidal rock (free living)>Fig. 1. Partial sequence of the 12S mt rDNA from coral-inhabitingbarnacles and free-living barnacles (EMBL accession numbersX78234–X78254). The substrate inhabited by each barnacle species isindicated in parentheses. Each sequence represents an individual barnacle, collected from a separate coral colony or intertidal rock, exceptfor the sequences of S. milleporumandS. dentatum(Fav) I, each ofwhich represents two identical sequences obtained from two individuals from separate colonies. A dotin a sequence indicates that thenucleotide in this position is the same as in theC. pallidus (Cyp)sequence. Stem coding regions are indicatedabovethe sequences, andthe numbers (32–48) correspond to the numbering used by Hickson etal. (1996). Complementary sequences assumed to form a stem aremarked by the same number (e.g., 32 and 328).Asterisksdenote positions for which full complementarity is observed between stemforming sequences. The “A” at position 31 and the “T” at position 327(underlined)correspond to positions 1160 and 1468 in the human sequence (Anderson et al. 1981), respectively.Aca, Acanthastrea echinata; Cyp, Cyphastrea chalcidicum; Ech, Echinopora gemmacea; Fav,Favia favus; Fat, Favites abdita; Mil, Millepora dichotoma; Mon,Montipora erythraea; Pav, Pavona varians; Pla, Platygyra lamellina;Poc, Pocillopora damicornis; Por, Porites lobata.368369primer set of Kocher et al. (1989) as modified by Mokady et al. (1994):58-GAAACCAGGATTAGATACCC and 58-TTTCCCGCGAGCGACGGGCG. The reaction buffer consisted of 10 mMTris–HCl (pH9.0), 50 mMKCl, 0.1% Triton X-100, and 3.5 mMMgCl2. Fifty-fivepicomoles of each primer was added for each reaction, along with 2.5U of Taq DNA polymerase (Promega, Madison, WI), a 300mMconcentration of each dNTP, and 1ml of template DNA solution in a totalvolume of 100ml. The PCR cycle consisted of 2 min of denaturationat 92°C, 2 min of annealing at 54°C, and 3 min of elongation at 72°C.This cycle was repeated 29 times, with a final cycle in whi