Phylogenetic Hypotheses of Octocoral Species Using Predicted RNA Secondary Structures of the Internal Transcribed Spacer 2 (ITS2)

Phylogenetic hypotheses of octocoral species using predicted RNA secondary structures of the internal transcribed spacer 2 (ITS2)

Catalina Aguilar Hurtado

UNIVERSIDAD DE LOS ANDES FACULTAD DE CIENCIAS DEPARTAMENTO DE CIENCIAS BIOLOGICAS BOGOTA D.C. 2006 Phylogenetic hypotheses of octocoral species using predicted RNA secondary structures of the internal transcribed spacer 2 (ITS2)

Catalina Aguilar Hurtado

A thesis submitted in partial fulfillment of the requirements for the degree of

BIOLOGIST

Director JUAN ARMANDO SANCHEZ Marine Biologist Ph.D

Codirector SILVIA RESTREPO Biologist Ph.D

UNIVERSIDAD DE LOS ANDES FACULTAD DE CIENCIAS DEPARTAMENTO DE CIENCIAS BIOLOGICAS BOGOTA D.C. 2006

2 ABS TRACT

Octocoral systematics is a relatively new field of study. Morphological aspects together with molecular identification are tools that have been recently applied to solve systematic questions in octocorals due to their intricate morphology. For this group, as for many others, the use of a molecular marker that solves systematic problems at different taxonomic levels would be extremely useful for species identification and classification. Here we use the internal transcribed spacer 2 (ITS2) predicted RNA secondary structure for reconstructing phylogenetic hypotheses at different taxonomic levels. These structures are very important for the ribosome assembly since they are involved in mRNA folding and processing. For this reason, mutations that do not enable the proper folding of the sequence have to be compensated or suppressed. Hence, ITS2 are said to evolve in concert, which leads to a homogenization of all the copies of this gene throughout the genome. ITS2 secondary structures are useful for higher taxonomy systematics because a common core of the structures has being found. The secondary structures presented here, in the majority of octocoral species, have the six helicoidal ring model presented in Acropora spp. and other eukaryotes such as yeast, dinoflagellates, arthropods and mammals among others, they have the typical hallmarks that are standard for ITS2 . With these molecular and structural guides, a molecular morphometrics approach was made for solving phylogenies.

In this study there were three taxonomic levels: (1) interfamily, between Plexauridae and Gorgoniidae, (2) intrafamily, within Gorgoniidae species, and (3) intrageneric, within the Eunicea genus. The tree parts of this thesis are presented as standing alone chapters, which correspond to articles in press or recently submitted to peer review at indexed journals. For all of these analyses, saturation plots were made. In the first approach, saturation was complete since transversions were considerably more frequent than transitions; in the last two approaches, saturation was also found but in a lesser degree. This was made for reassuring the importance of using secondary structure and not the primary sequence alignment for phylogenetic reconstruction at this level. Moreover, direct

3 comparisons between primary and secondary alignment (e.g., molecular morphometrics) phylogenies were made in order to find differences between these two approaches. M olecular morphometrics interfamily analyses gave high support to the recently described species Alaskagorgia aleutiana as a member of the Plexauridae family, in consequence with previous mtDNA results. This analysis also retained clades from the two families of octocorals that were studied.

At the intrafamily level the phylogenetic hypotheses using molecular morphometrics gave high support to the placement of Africagorgia schoutedeni and A. africana within a Gorgoniidae clade. Whereas, the same species were placed as a basal clade in the non-morphometrics phylogram using the primary sequence alignment. Muriceopsis spp. had certain secondary structure characteristics that drove this taxon as a basal clade in respect with other gorgoniids in the molecular morphometrics analysis. Moreover, long-branch attraction was found for Plexaurella nutans and Pterogorgia citrina species, which mean that they had to be excluded from the analysis. For this reason, the systematic positioning of Plexaurella spp., Pterogorgia citrina and Muriceopsis spp. is still unresolved.

For the lowest taxonomical level analyzed, molecular morphometrics was combined with secondary structure alignment using Bayesian inference and partitions for each helix producing a well-supported phylogram. Eunicea flexuosa, a divergent species previously classified within the outgroup Plexaura, was given high support as part of Eunicea, which corroborated what has been found with the primary alignment and morphological characters. M olecular morphometrics results differed from the primary alignment at species from the Euniceopsis subgenus that were placed grouping the outgroups Pseudoplexaura and Plexaura whereas E. mammosa was placed inside the Eunicea subgenus. As morphological characters have been found to be highly homoplasic in octocorals, molecular data are important to reconstruct octocoral systematics. For this reason, the addition of

4 ITS2 secondary structures brings more characters and hence more credibility to Eunicea systematics.

ITS2 molecular morphometrics is presented here as a new and promising phylogenetic tool. This sequence has proven to be important for the organism ribosome’s proper function and it turned out to be a useful marker for studying octocoral systematics. Even though more octocorals representatives are needed to resolve most of this subclass systematics, this study relying on ITS2 secondary structures for reconstructing phylogenies at different taxonomic levels demonstrated the great potential of ITS2 and molecular morphometrics for diverse systematic applications.

5 Hipótesis filogenética de especies de octocorales utilizando estructuras secundarias del espaciador interno transcrito 2 (ITS2)

RESUMEN

Debido a su compleja morfología la sistemática de octocorales es una nueva área de estudio; recientemente han sido aplicadas herramientas como identificación molecular, junto con aspectos morfológicos, para resolver preguntas sobre la sistemática de los octocorales. Para este grupo, como para muchos otros, se podría utilizar marcadores moleculares para resolver problemas sistemáticos a diferentes niveles taxonómicos, estos serían de gran ayuda para la identificación y clasificación de especies. En este estudio se utilizó el modelo de estructura secundaria del espaciador interno transcrito 2 (ITS2) del ARN para construir hipótesis filogenéticas a diferentes niveles taxonómicos. Estas estructuras son muy importantes para el ensamblaje de ribosomas, ya que estan involucradas en el pliegue y procesamiento correcto del ARNm. Por esta razón, mutaciones que no permitan la adecuada formación de estas estructuras serán compensadas o eliminadas. Como consecuencia de lo anterior se ha propuesto que el ITS2 evoluciona en concierto, lo que lleva a una homogenización de todas las copias de este gen en el genoma. Las estructuras secundarias del ITS son útiles para la identificación taxonómica, ya que se a encontrado una forma común en todas las estructuras construidas. Según los resultados en la mayoría de especies de octocorales las estructuras secundarias del ITS2 mostraron las típicas marcas presentes en ciertos corales y otros eucariotas como levaduras, dinoflagelados, artrópodos, y mamíferos entre otros, estos presentan un modelo de seis hélices con un anillo central. Estas guías moleculares y estructurales se utilizaron para resolver filogenias por medio de morfometría molecular.

En este estudio se analizaron tres niveles taxonómicos: (1) interfamiliar, entre la familia Plexauridae y Gorgoniidae, (2) intrafamiliar, entre las especies de Gorgoniidae, e

6 (3) intragenérico, dentro del género Eunicea. Las tres partes de esta tesis son presentadas por capítulos separados, que corresponden a artículos publicados o recientemente sometidos a revisión en revistas indexadas. Para estos análisis se realizaron gráficas de saturación. En el mayor nivel (interfamiliar) la saturación del alineamiento primario fue completa, ya que las transversiones fueron considerablemente más frecuentes que las transiciones, en los dos niveles restantes también se presentó saturación pero en menor grado. Esto se realizó con el fin de comprobar la importancia de utilizar estructuras secundarias y no el alineamiento primario para la reconstrucción filogenética en estos niveles. Adicionalmente, se realizaron comparaciones directas entre las filogenias del alineamiento primario y secundario (e.g., morfometría molecular) para encontrar diferencias entre las dos metodologías. El análisis de morfometría molecular interfamiliar mostró altos soportes a la especie recientemente descrita Alaskagorgia aleutiana que hace parte de la familia Plexauridae, lo cual es consecuente con resultados obtenidos anteriormente con ADN mitocondrial. Este análisis también retiene clados de las dos familias de octocorales estudiadas.

Las hipótesis filogenéticas utilizando morfometría molecular a nivel intrafamiliar mostraron un alto soporte de las especies Africagorgia schoutedeni y A. africana dentro del clado de Gorgoniidae. Por otro lado estas mismas especies se encontraron básales en filogramas que no eran de morfometría sino de alineamiento primario de las secuencias. Las especies del género Muriceopsis presentaron ciertas características en sus estructuras secundarias que las llevaron a posicionarse basalmente con respecto a las demás gorgonias en el análisis de morfometría molecular. Así mismo, se excluyeron del análisis las especies Plexaurella nutans y Pterogorgia citrina debido a que se encontró atracción de ramas largas en los filogramas con la presencia de estas. Por esta razón la sistemática de Plexaurella spp., Pterogorgia citrina y Muriceopsis spp. sigue sin resolver.

7 Para el menor nivel taxonómico analizado, se encontró un filograma con buenos soportes pues se combinó la morfometría molecular con el alineamiento secundario utilizando inferencia Bayesiana y particiones para cada hélice. La especie divergente Eunicea flexuosa que era clasificada anteriormente dentro del grupo externo Plexaura, se encontró con un alto soporte dentro del género Eunicea, lo que corrobora lo encontrado con el alineamiento primario del ITS2 y caracteres morfológicos. Los resultados de morfometría molecular se diferenciaron del alineamiento primario con las especies del subgénero Euniceopsis que se encontraban con los grupos externos Pseudoplexaura y Plexaura mientras E. mammosa se encontró dentro del subgénero Eunicea. Como se ha visto que los caracteres morfológicos son altamente homoplásicos en octocorales, los datos moleculares son importantes para construir la filogenia de éstos. Por esta razón, el uso de estructuras secundarias del ITS2 aumenta credibilidad a la sistemática de Eunicea debido a un mayor aporte de caracteres.

La morfometría molecular del ITS2 se presenta como una nueva herramienta filogenética. Esta secuencia ha demostrado ser importante para el correcto funcionamiento de los ribosomas en los organismos y resulto ser un marcador útil para estudiar la sistemática de los octocorales. Aunque se necesita una mayor representación de especies para resolver la mayoría de la sistemática de esta subclase, este estudio que utiliza las estructuras secundarias del ITS2 para la reconstrucción de filogenias a diferentes niveles taxonómicos, demostró el gran potencial del ITS2 y la morfometría molecular para diversas aplicaciones sistemáticas.

8 ACKNO WLEDG EM ENTS

I would like to specially thank Dr. Juan Armando Sánchez for his support and collaboration throughout this project. Dr. Silvia Restrepo for her corrections. To the students of BIOMMAR at Universidad de los Andes for their support in this last two semesters. To the Universidad de los Andes for providing resources to perform this paper analysis. To Dr. Meg Daly at the University of Ohio (and the Cnidaria Tree-of-life project) for providing some of the sequences needed for this project. I would like to thank my parents, brother, and grandmother for their unconditional support over the years.

9 TAB LE O F CO NTENTS

LIST OF TABLES………………………………………………………………………..12 LIST OF FIGURES………………………………………………………………………12 LIS T OF APPENDIX…………………………………………………………………….15

CHAPT ER 1: MOLECULAR MORPHOMET RICS : CONTRIBUT ION OF ITS 2 SEQUENCES AND PREDICTED RNA SECONDARY STRUCTURES TO OCTOCORAL SYSTEMATICS

Abstract…………………………………………………………………..16 Introduction………………………………………………………………17 Materials and Methods…………………………………………………...20 Results……………………………………………………………………22 Discussion………………………………………………………………..28 Acknowledgments………………………………………………………..29 References

CHAPTER 2: PHYLOGENETIC HYPOTHESES OF GORGONIID OCTOCORALS ACCORDING TO ITS 2 AND THEIR PREDICT ED RNA S ECONDARY S TRUCTURES

Abstract…………………………………………………………………..33 Introduction………………………………………………………………34 Materials and Methods…………………………………………………...36 Results……………………………………………………………………39 Discussion………………………………………………………………..46 Acknowledgments………………………………………………………..49 References

CHAPT ER 3: MOLEC ULAR S YS TEMAT ICS OF THE GEN US EUNICEA (PLEXAURIDAE: OCTOCORALLIA: CNIDARIA) USING THE S ECONDARY S TRUCTURE OF THE INTERNAL T RANS CRIBED SPACER 2

10 Abstract…………………………………………………………………..56 Introduction………………………………………………………………57 Materials and Methods…………………………………………………...59 Results……………………………………………………………………62 Discussion………………………………………………………………..66 Acknowledgments………………………………………………………..68 References

APPENDIX………………………………………………………………………………..72

11 LIST OF TABLES

Table 1.- Species / Characters matrix from molecular morphometrics. Character numbering refers to RNA predicted secondary structure as in Fig. 1.3………………….25

Table 2.1 - Species name and GenBank Accession numbers from the reanalyzed Internal Transcript Spacer, partial sequence………………………………………….36

Table 2.2- Species / Characters matrix from molecular morphometrics. Character numbering refers to RNA predicted secondary structure as in Fig. 2.1……...39

Table 3.- Species / Characters matrix from molecular morphometrics. Character numbering refers to RNA predicted secondary structure as in Fig. 3.1………………….62

LIST OF FIGURES

Figure 1.1.-Phylogenetic hypothesis of 14 octocorals species using ITS2 sequences alignment. A. Bayesian phylogram, numbers in the nodes are Bayesian posterior probabilities or clade credibility values (generations=10000, burn- in=100); model (SYM+I+G) selected by AIC; rates invariant among sites with gamma. B. Maximum Likelihood phylogram with the assumed model (TVM ef+I+G) selected by AIC, tree searches had six substitution rates (A-C, 1.4269; A-G, 3.3394; A-T, 1.3979; C-G, 0.5815; C-T, 3.3394; and G-T, 1.0000), an assumed proportion of invariant sites (0.0644), and a gamma- shape parameter (2.0199). C. Maximum Parsimony phylogram with INDELS codified in GapCoder. D. MP phylogram with INDELS codified as fifth base. All values above the nodes correspond to bootstrap values from 1000 replicates. Letters in parentheses indicate family names; Pri: Primnoidae; Sub: Subsselifloriidae; Al: Alcyoniidae; P: Plexauridae; G: Gorgoniidae…………………………………………………………………..23

Figure 1.2.- Saturation plots of 14 octocorals species ITS2 sequences. Tamura-Nei distance obtained with assumed proportion of invariable sites (0.0644), and a gamma-shape parameter (2.0199)……………………..………………………………………………..24

12 Figure 1.3.- ITS2 predicted RNA secondary structures for 4 species of octocorals with their designated character as used in molecular morphometrics (Table 1), and their enthalpy values of structure formation according to MFOLD. (A) Alaskagorgia aleutiana,∆G= -59.26 kcal/mole; (B) Muriceopsis flavida, ∆G= -80.63 kcal/mole; (C) Muricea muricata, ∆G= -39.87 kcal/mole; (D) Eunicea laciniata, ∆G= -104.17 kcal/mole. ………………………………………….25

Figure 1.4.- ITS2 predicted RNA secondary structures for 10 species of octocorals and their enthalpy values of structure formation according to MFOLD: (A) Pacifigorgia stenobrochis, ∆G= -69.53 kcal/mole; (B) Gorgonia ventalina, ∆G= -69.63 kcal/mole; (C) Leptogorgia virgulata, ∆G= -67.66 kcal/mole; (D) Pterogorgia citrina, ∆G= -68.97 kcal/mole; (E) Plexaurella nutans, ∆G= - 73.34 kcal/mole; (F) Plexaura kuna,∆G= -26.74 kcal/mole; (G) Calyptrophora japonic, ∆G= -97.36 kcal/mole; (H) Alcyonium digitatum, ∆G= -53.71 kcal/mole; (I) Sclereobelemon theseus, ∆G= -84.23 kcal/mole; (J) Lobophytum sp., ∆G= -77.86 kcal/mole. ……………………………………26

Figure 1.5.- Phylogenetic hypotheses of molecular morphometrics data. A. 50% Majority- rule consensus with its support above the node of 15 equally parsimonious trees. B. Adams consensus from the same trees. Values below the nodes correspond to bootstrap values from 1000 replicates. Letters in parentheses indicate family names; Pri: Primnoidae; Sub: Subsselifloriidae; Al: Alcyoniidae; P: Plexauridae; G: Gorgoniidae. ………………………………27

Figure 2.1.- ITS2 predicted RNA secondary structures for 9 species of octocorals and their enthalpy values of structure formation according to MFOLD: (A) Pseudopterogorgia bipinnata CBC (Carrie Bow Cay, Belize), numeration represents characters used for molecular morphometry as in Table 1, ∆G= - 58.96 kcal/mole; (B) Eunicea laciniata, ∆G= -103.02 kcal/mole; (C) Tobagogorgia hardyi, ∆G= -67.47 kcal/mole; (D) Africagorgia schoutedeni, ∆G= -80.92 kcal/mole; (E) Lophogorgia miniata, ∆G= -62.67 kcal/mole; (F) Muriceopsis bayeriana, ∆G= -80.27 kcal/mole; (G) Gorgonia ventalina, ∆G= -69.67 kcal/mole; (H) Leptogorgia setacea, ∆G= -67.66 kcal/mole; (I) Pacifigorgia stenobrochis,∆G= -69.53 kcal/mole. ………………………….40

13 sampling every 100 simulation, burn-in 10000) node support from the >50%- majority rule consensus from the 1001 sampled trees. B, maximum parsimony phylogram from corrected sequence alignment using branch and bound algorithm, Length=638, Consistency index=0.726, Retention index= 0.697, support nodes bootstrap 50% consensus, 1000 replicates. C, molecular morphometrics phylogram of 36 characters, Length=204, Consistency index=0.554, Retention index= 0.643, (G-test from 100 trees g1= -0.54 g2= 0.29, showing support nodes from maximum parsimony non-parametric 100 bootstrap replicates / Bayesian 50% majority rule consensus of 1001 trees analysis using partitions by helix (simulations as above). CBC= Carrie Bow Cay, Belize; B= San Salvador, Bahamas. …………………………………...41

Figure 2.3.- Saturation plots of ITS2 sequences of 22 octocorals species. Tamura-Nei distance obtained with assumed proportion of invariable sites (0.1699), and a gamma-shape parameter (2.2647)……………………………………………42

Figure 2.4.- Phylogenetic hypotheses (phylograms) including or excluding Plexaurella nutans and Pterogorgia citrina. A. Maximum likelihood tree with sequence alignment. B. Maximum likelihood tree with sequence alignment including all the species (with Pl.nutans and Pt.citrina). C-F. Molecular morphometrics maximum parsimony. C. Molecular morphometrics maximum parsimony tree including all the species. D. Excluding Muriceopsis-Tobagorgia species. E. Excluding Pterogorgia citrina F. Excluding. Plexaurella nutans. Overall methods and program settings as in Fig. 2.2…………………………………45

Figure 3.1.- ITS2 predicted ARN secondary structures for 9 species of the genus Eunicea and their enthalpy values of structure formation according to M FOLD: (A) Eunicea flexuosa, numeration represents characters used for molecular morphometry as in Table 1, ∆G= -58.96 kcal/mole; (B) Eunicea laciniata, ∆ G= -104.17 kcal/mole; (C) E. colombiana, ∆G= -104.17 kcal/mole; (D) E.tayrona, ∆G= -71.45 kcal/mole; (E) E. laxispica, ∆G= -62.67 kcal/mole; (F) E. fusca, ∆G= -65.96 kcal/mole; (G) E tourneforti, ∆G= -69.67 kcal/mole; (H) E. mammosa, ∆G= -67.66 kcal/mole; (I) E. pallida,∆G= -69.53 kcal/mole……………………………………………………………………..63

14 a gamma-shape parameter (1.4130). Bayesian probabilities and bootstrap values of 1,000 replicates are separated by a slash. C. Consensus tree of the Bayesian analysis with 10, 000,000 generations of the combined data of predicted secondary structure and secondary sequence alignment of the ITS 2 region. ……………………………………………………………………….65 APPENDIXES

Appendix 1. DCSE Alignment showing secondary structures of fourteen species of octocorals (Helix numbering as in Fig. 1; Character coding in Table 1.)……72

Appendix 2. DCSE Alignment showing secondary structures of twenty-four species of octocorales (Helix numbering as in Fig. 1; Character coding for molecular morphometrics in Table 2.2)…………………………………………………….75

Appendix 3. DCSE Alignment showing secondary structures of twelve species of octocorals (Helix numbering as in Fig. 1; Character coding in Table 3.)……78

15 CHAPT ER 1 MOLECULAR MORPHOMETRICS: CONTRIBUTION OF ITS2 SEQUENCES AND PREDICT ED RN A S ECONDARY S TRUCTURES TO OCTOCORAL SYSTEMATICS1

1.1. ABSTRACT Octocorals are among the largest and most diverse invertebrates on seamounts and in deep water but most of their systematics remains misunderstood. Molecular studies have produced new insights at higher levels. Unfortunately, most DNA sequences from both mitochondrial and nuclear genes have exhibited a great deal of conservation preventing their use for closely-related species. The internal transcribed spacers (ITSs) from the nuclear ribosomal-DNA have shown considerable variation among octocorals. The ITS2 sequence has turned out to be a promising region. Here we provide new sequences and predicted RNA secondary structures for the ITS2 of fourteen octocorals. The sequences exhibited the highly conserved six-helicoidal ring-model structure found in yeast, insects, and vertebrates. A molecular morphometrics approach of 14 octocoral species produced 49 characters and 15 equally parsimonious trees. Consensus trees retained most relationships found with conserved mtDNA sequences. For instance, the node grouping Alaskagorgia aleutiana with Muricea muricata was highly supported, which comprises independent support for the placement of this recently described deep-water species with the Plexauridae, in spite of having poor affinities according to morphology. Molecular morphometrics skips the issue of dealing with multiple INDELS and saturation in the primary information from sequence alignments. Nonetheless, the reliability and phylogenetic signal of ITS2 is better for intrageneric studies. Key words: M olecular morphometrics, Octocorallia, octocoral systematics, ITS2, RNA secondary structure

1 Aguilar, C., Sánchez, J.A. (In press). Molecular morphometrics: contribution of ITS2 sequences and predicted RNA second ary structures to octocoral systematics. Bulletin of Marine Science.

1.2. INTRODUCTION

Notwithstanding that invertebrate systematics is entering into the “DNAbarcoding era” (e.g., Hajibabaei et al., 2006), octocoral systematics and taxonomy remain in their infancy. Morphological characters in octocorals have a great deal of individuality, e.g., each calcite sclerite is unique compared to each other, preventing the use of landmarks and related tools to study morphology, consequently taxonomy has to be based on multiple characteristics and a range of variation for each structure (e.g., Sánchez, 2001). Each structure, for example a surface cortex sclerite, will range in shape according to the positioning with respect to the octocoral axis, the polyp aperture and even locality or habitat (e.g., Sánchez, 2004). Modern taxonomy focuses on finding a gap between intraspecific and interspecific variation (Meyer& Pauley, 2005). Unfortunately in the case of octocorals there seems to be a high overlap in the morphological continuum among species. On the other hand, octocoral molecular systematics has not been entirely the solution. Molecular studies have brought along new insights regarding octocoral systematics mostly on higher taxa relationships (e.g., Sánchez et al., 2003a). Even though mitochondrial and nuclear DNA sequences have shown good phylogenetic resolution for comparing families and genera (e.g., Bernston et al., 2001; Sánchez et al., 2003a, 2003b), the same genomic regions have proven to be nearly invariable at the close related species level (France & Hoover, 2002). The search for DNA regions and molecular approaches with resolution at the species level is a priority in octocoral molecular systematics. The ITS sequences (Internal Transcribed Spacers), found within the tandem arrays of the nuclear ribosomal RNA genes in eukaryotes, have not been considered useful in molecular systematics in some invertebrates, mostly due to saturation, excessive INDELS, and/or intragenomic variation (e.g., Frabry et al., 1999; Harris & Crandall, 2000). Nonetheless, they seem to be a promising nuclear region for lower eukaryotes, particularly at the intra-familial level and

17 down to closely-related species, where other known regions are nearly invariable (e.g., LaJeunesse 2001). In scleractinian corals, excepting the species of Acropora, ITS sequences have reliable phylogenetic information especially if the RNA predicted secondary structure is compared (Chen et al., 2004). In the case of octocorals, ITS sequences have been very useful to compare species of alcyoniids octocorals as well as to point out hybridization among species (McFadden & Hutchinson 2004). These sequences across many octocoral taxa have shown considerable variation within genera (unpublished) and perhaps are among the most variable genomic regions in these organisms in which mitochondrial DNA and other sequences are very conserved (Berntson et al., 2001; Shearer et al., 2002). Nonetheless, ITS alignment always produced multiple and variable INDELS (insertions-deletions) depending on the different gap opening and/or extension penalties (e.g., ClustalW), which makes somewhat unreliable the phylogenetic inference. An alternative to using ITS2 sequences is to process them into RNA predicted secondary structures for correcting alignments and/or molecular morphometrics. RNA is said to catalyze reactions, direct the site-specific modification of RNA nucleotides, modulate protein expression, and serve in protein localization (Mathews, 2005). Furthermore, RNA is used in molecular systematics because it has more options of developing secondary structures due to the presence of uracil, which can bond to both adenine and guanine. RNA secondary structure formation tends to minimize energy whereas maximizing stable G-C bonds, which often appear in the framework of helixes’ stems. This is why, based on sequence pairing, secondary structure can be reconstructed independently from the tertiary ones. For cladistic analysis, secondary structures are particularly useful because they include “morphologically” informative characters not found in the primary sequence (Billoud et al., 2003; Swain & Taylor, 2003). In addition, RNA secondary structures are of importance in phylogenetic analysis, because they are used to correct alignments between species that sometimes are not accurate by simple base alignments (e.g., Sánchez et al., 2003a). The most reliable method for reconstructing secondary structures is by crystallography. However, this method is expensive and not often performed for systematic

18 studies. Alternatively, the RNA secondary structure can be determined by an alignment with sequences of already known structures to depict common base-pairing patterns in the absence of crystallized structures. Additionally, these structures have highly conservative zones that ease reconstructing unknown RNA secondary structures. As a result, most RNA secondary structures known today have been constructed in that fashion. In the case of octocorals, we know the secondary structures from 21 species comprising a part of the region LSU-rRNA (16S) for a phylogenetic analysis (Sanchez et al., 2003a). The ITS2 secondary structure is a helicoidal ring with 5 to 6 helixes. Compelling evidence proposed by Joseph et al (1999) in yeast, shows the fundamental role of the helicoidal ring of the ITS2 transcript during in the pre-rRNA processing, which lacking otherwise blocks the production of the mature 25S rRNA when specific structures of the ring model can not be formed (Cote et al., 2002). Consequently, secondary structure of the ITS2 region seems to be essential when it comes to ribosome assembly and the likely reason why ring model structure has been reconstructed in a wide range of eukaryotes ranging from dinoflagellates and corals to insects and invertebrates (Joseph et al., 1999; Chen et al., 2004; Gottschling & Plötner. 2004). In this study, we explored the phylogenetic resolution of ITS2 predicted RNA secondary structure to compare a variety of octocoral taxa; moreover, we compared this results with the ITS2 primary sequence alignment. Particularly, we generated new ITS2 sequences and RNA secondary structures for reconstructing phylogenetic relationships of the deep-water octocoral Alaskagorgia aleutiana Sánchez & Cairns (2004), a species whose morphological characters overlap between the families Plexauridae and Gorgoniidae and mtDNA places it together with plexaurid species (Sánchez & Cairns, 2004). It is worth noting that mtDNA sequences in anthozoans in general are extremely conserved (see discussion in Shearer et al., 2002). Would comparisons among RNA predicted secondary structures from one of the most variable sequences in nature (ITS2) produce similar results to those comparisons made with one of the less variable sequences (cnidarians mtDNA)?

19 Besides answering this question, it is a goal of this paper to discuss the systematic value of ITS2 secondary structures in octocorals.

1.3. MATERIALS AND METHODS

1.3.1. ITS2 Sequences Sequences were obtained using primers designed by Sánchez (submitted), which targets the region of the 5’ end of the 5.8s ribosomal genes and the 3’ end of the 28S ribosomal gene containing the complete ITS2 (5.8S-436: 5’- AGCATGTCTGTCTGAGTGTTGG-3’ and 28S-663: 5’- GGGTAATCTTGCCTGATCTGAG-3’, numbers respect to the sequence of Alcyonium digitatum, Genbank Acc. No. AF262347: McFadden et al., 2001). DNA extractions were made according to Sánchez et al. (2003b) from alcohol preserved material (ethanol 95%) and the DNeasy kit (Qiagen) applied to the extractions with high concentration of dark pigments only that usually did not amplify properly. Template DNA for sequencing was obtained from the combination of two PCR reaction tubes containing 56 µl total (volume completed with ddH2O) with 1 µl of DNA template (1:50 dilutions of genomic DNA extractions), 2 units Taq polymerase (Promega), 3 µl of 10X Buffer (Promega), MgCl2, 0.15 µM dNTPs mix, 0.16 µM of each primer. PCR conditions had one initial period of 2.0 min at 94°C followed by 30 cycles of 30 sec at 94°C, 45 sec at 56°C, one 45 sec at 72°C, and a final extension step for 5.0 min at 72°C. Gene purification was made using the Edge Biosystems kit and sequencing usingBigDye 3.1 (AB 3100, capillary electrophoresis automated sequencer). Consensus sequences were obtained automatically by assembling the two complementary DNA chromatograms (Sequencher software).

1.3.2. ITS2 Sequence Analysis

20 Fourteen octocorals (Calyptrophora japonica [Primnoidae], Muricea muricata, Alaskagorgia aleutiana, Plexaura kuna, Eunicea laciniata, Plexaurella nutans, Muriceopsis flavida [Plexauridae], Pterogorgia citrina, Pacifigorgia stenobrochis, Gorgonia ventalina, Leptogorgia virgulata [Gorgoniidae], Lobophytum sp., Alcyonium digitatum [Alcyoniidae], Sclerobelemon theseus [Subsselifloriidae]) sequences were aligned in BioEdit (Hall 1999) using Clustal W multiple alignment with the default gap and extension penalties used by this program; this for performing different sequence analysis. First, Bayesian inference of phylogeny was done using MrBayes (Huelsenbeck and Ronquist, 2001). Maximum likelihood (ML) phylogeny was carried out with the best-fit model from ModelTest (Posada and Crandall, 1998) obtained by the Akaike Information Criterion (AIC). Maximum parsimony (MP) phylogeny was carried out using the exhaustive branch and bound algorithm in PAUP* (Swofford, 2002) with two different INDEL codifications: INDELS codified in GapCoder (Young and Healy 2003), and INDELS codified as fifth base. To analyze sequences saturation Tamura-Nei distance, transitions values, transversions values, and their radio (Ti/Tv) was obtained in PAUP* (Swofford, 2002).

1.3.3. ITS2 Predicted RNA Secondary Structures ITS2 alignment for used for predicting secondary structures by homologous structures already published, like Acropora spp. (Odorico & M iller, 1997). Moreover, manual alignment was performed by visual homology for the construction of a Dedicated Comparative Sequence Editor (DCSE) format in order to perform different phylogenetic analysis. DCSE format uses square brackets ([ or ]) to delimitate each helix, braces ({ or }) for bulges and loops, and hyphens for gaps in the alignment, terminal loops appear separated by inverse brackets (] and [). The acquired structures with restrictions and constrains were submitted in MFOLD (http://www.bioinfo.rpi.edu/applications/mfold/old/rna/). RNA was folded at a fixed temperature of 37°C, and the structure chosen from different output files was the one with

21 the highest negative free energy. The obtained secondary structures were used to construct a matrix for a cladistic analysis by means of molecular morphometrics based on geometric features and base number (Billoud et al., 2000). This means, secondary structures helixes were numbered and treated like a character that will vary depending on their base number. Bulges and internal loops were given the number of the helix followed by a letter, ending helixes loops and separation segments where designated with the “i” letter. Loops complement, helix complement, and separation segments were indicated with an apostrophe (e.g. 4a loop with its complement 4a’). Nucleotides were count for each character and arranged in discrete character states ranging from 0 to 5 (being 0 the character with less nucleotide number) as seen in table 1. Phylogenetic analyses using Fitch maximum parsimony were carried out in PAUP* (Swofford, 2002)

1.4. RESULTS

The ITS2 region varied in length form 309 bp in Lobophyton sp. to the shortest in P. citrina with 197 bp, and the remaining ranged between these two numbers (see appendix 1). Four different phylogenetic hypotheses were obtained from the primary ITS2 alignment. There were 266 total characters, from which 208 were parsimony-informative. All of them presented consistence and retention index of 50%. Bayesian analysis retained two equally long trees (L= 858; CI= 0.614; RI= 0.504) (Fig. 1.1A); ML retained one tree (L= 855; CI = 0.616; RI= 0.509) (Fig. 1.1B); MP retained one tree with both INDELS analysis: fifth base (L= 854 CI= 0.617; RI = 0.510) (Fig. 1.1D), and GapCoder (L= 935; CI = 0.640; RI=0.531) (Fig. 1.1C) which contained more characters (337 characters from which 237 were parsimony-informative). Bootstrap 50% majority-rule consensus trees (replicates=1000) were obtained for all the analysis expect for Bayesian obtained topology. The saturation plots obtained with the Tamura-Nei distance, showed clearly that the ITS2 primary sequences were saturated (Fig. 1.2), which could have leaded to misleading results.

Fi gure 1.1.-Phylogenetic hypothesis of 14 octocorals species using ITS2 sequences alignment. A. Bayesian phylogram, numbers in the nodes are Bayesian posterior probabilities or clade credibility values (generations=10000, burn-in=100); model (SYM+I+G) selected by AIC; rates invariant among sites with gamma. B. Maximum Likelihood phylogram with the assumed model (TVMef+I+G) selected by AIC, tree searches had six substitution rates (A-C, 1.4269; A-G, 3.3394; A-T, 1.3979; C-G, 0.5815; C-T, 3.3394; and G-T, 1.0000), an assumed proportion of invariant sites (0.0644), and a gamma-shape parameter (2.0199). C. Maximum Parsimony phylogram with INDELS codified in GapCoder. D. MP phylogram with INDELS codified as fifth base. All values above the nodes correspond to boot strap values from 1000 replicates. Letters in parentheses indicate family names; Pri: Primnoidae; Sub: Subsselifloriidae; Al: Alcyoniidae; P: Plexauridae; G: Gorgoniidae

Fi gure 1.2.- Saturation plots of 14 octocorals species IT S2 sequences. Tamura-Nei distance obtained with assumed proportion of invariable sites (0.0644), and a gamma-shape parameter (2.0199).

Fourteen octocorals secondary structures were reconstructed, including the recently described Alaskagorgia aleutiana (Figs. 1.3, 1.4). The structures present the six-helicoidal ring-model in all except in an outgroup (Sclerobelemnon theseus that has only five helixes). The molecular morphometrics approach produced 49 characters (48 parsimony informative). A maximum parsimony search found 15 equally long (L=262) trees. Consistency and retention indexes were about 50% (CI=0.565; RI=0.457), indicating a great deal of homoplasy. As an explicit measure of phylogenetic content, the skewness of the tree-length distribution of 1000 randomly generated trees in PAUP* was assessed with G-tests (Hillis and Huelsenbeck 1992). The lengths of random trees had a mean of 319.8 (SD = 8.68) and the tree-length frequency distribution was significantly rightskewed (g1= - 0.56, P<0.01), suggesting that the primary content was not from homoplasy. In addition, to examine the noise (e.g., excessive homoplasy) in the data set the Permutation Tail Probability (PTP) was also determined using PAUP* (e.g., Fu and M urphy 1999). The PTP test with 1000 replicates clearly showed that all replicates had more steps (>282) than the unpermuted tree length 262 (P<0.001), which also supports a good content of phylogenetic signal.

Ta bl e 1. Species / Characters matrix from molecular morphometrics. Character numbering refers to RNA predicted secondary structure as in Fig. 1.3.

Spp . / Characte rs 1 1a 1i 2 2i 2' 2a' 2i' 3 3a 3b 3c 3d 3i 3' 3'a 3b' 3c' 3d' 3i' 4 4a 4b 4c 4i 4' 4a' 4b' 4c' 4i 5 5a 5b 5c 5d 5e 5i 5' 5'a 5'b 5'c 5'd 5e' 5i 6 6i 6' 6i' 1' Calyptrophora japonica 210 2320 100 000 0000 002 4230 033 203 3412 105 302 1000 434 42 Muricea muricata 003 1010 341 121 2421 110 5311 051 325 0400 002 020 0000 000 00 Alcyonium digitatum 003 1010 122 100 0220 000 3330 232 300 4122 111 413 1222 232 00 Eunicea laciniata 210 2120 220 000 0200 002 1200 210 003 4133 232 423 3211 444 32 Plexaura kuna 003 0100 331 120 3323 100 0120 301 101 1121 005 111 4112 212 20 Plexaurella nutans 101 3130 010 000 1110 001 3210 230 200 3133 211 301 0002 151 11 Sclerob elemon t heseus 321 1210 010 000 2100 001 2200 021 102 4113 313 414 1131 000 33 Lobophytum sp. 000 3031 021 000 2221 003 5331 052 322 1231 334 121 4220 424 20 Muriceopsis flavida 100 4040 111 000 0110 001 4031 040 313 3022 201 322 3210 363 11 Pacifigorgia stenobrochis 102 2120 221 000 0210 001 3320 330 204 2133 200 213 2204 202 11 Gorgonia ventalina 102 2120 102 000 3000 002 1300 111 004 2133 200 213 2204 242 11 Leptogorgia virgulata 102 2120 210 000 1100 001 4310 140 204 2133 200 213 3202 262 11 Pterogorgia citrina 100 4140 110 000 0110 001 3000 330 005 1203 201 121 2202 252 11 Alaskagorgia aleutiana 003 2320 342 121 1412 113 1030 212 112 0000 003 000 0000 232 30

25 Fi gure 1.3.- ITS2 predicted RNA secondary structures for 4 species of octocorals with their designated character as used in molecular morphometrics (Table 1), and their enthalpy values of structure formation according to MFOLD. (A) Alaskagorgia aleutiana,∆G= - 59.26 kcal/mole; (B) Muriceopsis flavida, ∆G= -80.63 kcal/mole; (C) Muricea muricata, ∆G= -39.87 kcal/mole; (D) Eunicea laciniata, ∆G= -104.17 kcal/mole.

26 kcal/mole; (E) Plexaurella nutans, ∆G= -73.34 kcal/mole; (F) Plexaura kuna,∆G= -26.74 kcal/mole; (G) Calyptrophora japonic, ∆G= -97.36 kcal/mole; (H) Alcyonium digitatum, ∆G= -53.71 kcal/mole; (I) Sclereobelem on theseus, ∆G= -84.23 kcal/mole; (J) Lobophytum sp., ∆G= -77.86 kcal/mole.

Trees found retained clades of the families Plexauridae and Gorgoniidae (Majority rule consensus: Fig. 1.5A). However, M. flavida and P. nutans, species of the family Plexauridae, were found in defined nodes with Gorgoniidae (Adams consensus: Fig. 1.5B). M. flavida (Plexauridae) was found in the same node with P. citrina (Gorgoniidae). E. laciniata (Plexauridae), with a predominantly long helix V (Fig. 1.4D), did not group with any other taxa and remained in the basal polytomy together with the outgroups C. japonica (Primnoidae) and S. theseus (Sessilifloridae) (Fig. 1.4G, 1.4I). The other members of Plexauridae included in the analysis, M. muricata and P. kuna, group with A. aleutiana, despite the secondary structure morphology of P. kuna being different (helix V is longer). The two soft corals (Alcyoniidae) were basal with respect to that group (Fig. 1.1B). Helix III length is the main characteristic that closely relates A. aleutiana with M. muricata, due to its vast number of bases and loops in comparison to other species (e.g., Fig. 1.3).

27 Figure 1.5.- Phylogenetic hypotheses of molecular morphometrics data. A. 50% Majority- rule consensus with its support above the node of 15 equally parsimonious trees. B. Adams consensus from the same trees. Values below the nodes correspond to bootstrap values from 1000 replicates. Letters in parentheses indicate family names; Pri: Primnoidae; Sub: Subsselifloriidae; Al: Alcyoniidae; P: Plexauridae; G: Gorgoniidae.

1.5. DISCUSSION

The ITS2 sequences among the studied octocorals had between 4 and 20% (without INDELS) total divergence. The sequences used span highly divergent taxa according to what is known about mitochondrial and nuclear DNA sequences, such as pennatulaceans, primnoids (Calcaxonia), and several soft and gorgonian corals. That level of divergence produced misleading phylogenetic results using ITS2 primary sequence due to saturation, since Ti/Tv radio was below one also obvious when transversions were more frequent that transitions (Fig. 1.2). Consequently primary ITS2 sequence analysis could be alatorious, which leads to branch attraction and moreover inconsistent results even with maximum likelihood methods. The phylogenetic trees obtained by means of primary ITS2 sequence analysis were not consistent with the mitochondrial DNA analysis done by Sánchez and Cairns (2004) due to the placement of A. aleutiana (Fig. 1.1) in different no supported nodes. In spite of that, the molecular morphometrics produced a phylogenetic hypothesis that retains some of the results found previously. The ITS2 predicted RNA secondary structure morphometrics supported some branches of the phylogram obtained by Sanchez and Cairns (2004) with much conserved mitochondrial DNA sequences for most of the species used in this study. For instance, the node that clusters A. aleutiana with M. muricata, also found with mtDNA, was highly supported with the molecular morphometrics results (Fig. 1.5, bootstrap value of 92), which gives independent support for the placement of A.aleutiana with the Plexauridae. Likewise, obvious affinities according to morphology and found also with mtDNA such as the sister relationship of L. virgulata with P. stenobrochis, were also observed here.

28 Some other phylogenetic relationships found with mtDNA and considered unexpected due to morphological affinities, such as the sister relationship of P. citrina with M. flavida (Sánchez et al., 2003; Sanchez and Cairns, 2004; Wishing et al., 2005), were also recovered in this study. Aspects above drive us to support the idea of the ITS2 region and its predicted RNA secondary structure as a good phylogenetic tool for octocorals. It is important to note that there were some inconsistencies with our results. The genus Eunicea has clear affinities with Plexaura spp and Muricea spp. with both molecular and morphological analyses (Sánchez et al., 2003; Sánchez and Cairns, 2004; Sánchez, unpublished). However, in this study Eunicea was found in different trees together with C. japonica, S. theseus (Figs. 1.4G, 1.4I) or A. digitatum (Fig. 1.4H). E. laciniata had a helix V in ring model predominantly larger than the others. Consequently, it was evident that comparing too distant species (e.g., different families and orders) could lead to misleading results even with the molecular morphometrics approach. In conclusion, the ITS2 RNA predicted secondary structures can be used with reliability to reconstruct phylogenetic relationships among octocorals. Using the molecular morphometrics approach skips the issue of dealing with multiple INDELS from the alignment of the primary sequences and could have phylogenetic signal even if the primary sequences are saturated. The reliability and phylogenetic signal of ITS2 will be better for intrafamilial and intrageneric studies (Sánchez & Aguilar, in prep.).

1.6. ACKNOWLEDGEMENTS The George Institute for Biodiversity Sustainability, and the proceeding editors Bob George and Stephen Cairns, provided support and encouragement to J.A. Sánchez to attend and present these results at the 3rd International Symposium on Deep-Sea Corals. J.A. Sánchez acknowledges a postdoctoral fellowship from the Smithsonian Institution-National Museum of Natural History (2002-2003), which made possible getting most of the DNA in this study, and Facultad de Ciencias, Department of Biological Sciences, and BIOMMAR students at Universidad de los Andes for providing partial

29 resources to do the analysis and writing the paper. The Systematics Association (UK) grant to J.A. Sánchez supported partially the analyses of this paper. Assistance and advice from the staff of the laboratories of analytical biology at Smithsonian (molecular systematics [Liz Zimmer, Ken Schallop and Lee Weight] and SEM [Scott Witaker and Susan Branden]) is greatly appreciated.

REFERENCES

Berntson, E. A., F. M. Bayer, A. G. McArthur, and S. G. France. 2001. Phylogenetic relationships within the Octocorallia (Cnidaria : Anthozoa) based on nuclear 18S rRNA sequences. Mar. Biol. 138: 235-246. Billoud, B., M. A. Guerrucci., M. Masselot., and J. S. Deutsch. 2000. Cirripede phylogeny using a novel approach: molecular morphometrics. Mol. Biol. Evol. 17:1435–1445. Chen, C. A., C. C. Chang, N. V. Wei, C. H. Chen, Y. T. Lein, H. E. Lin, C. F. Dai, and C. C. Wallace. 2004. Secondary structure and phylogenetic utility of the ribosomal internal transcribed spacer 2 (ITS2) in Scleractinian Corals. Zool. Stud. 43: 759-771. Cote, C. A., C. L. Greer, and B. A. Peculis. 2002. Dynamic conformational model for the role of ITS2 in pre-rRNA processing in yeast. RNA 8: 786-797. Fabry S., A. Kohler, and A. W. Coleman. 1999. Intraspecies analysis: Comparison of ITS sequence data and gene intron sequence data with breeding data for a worldwide collection of Gonium pectorale. J. Mol. Evol. 48: 94-101. France, S. C., and L. L. Hoover. 2002. DNA sequences of the mitochondrial COI gene have low levels of divergence among deep-sea octocorals (Cnidaria : Anthozoa). Hydrobiologia. 471: 149-155. Fu, J., and R. W. Murphy. 1999. Discriminating and locating covariance: an application of permutation tail probability (PTP) analyses. Syst. Biol. 48: 380-395

30 Gottschling M. and J. Plötner. 2004. Secondary structure models of the nuclear internal transcribed spacer regions and 5.8S rRNA in Calciodinelloideae (Peridiniaceae) and other dinoagellates. Nucl. Acids Res. 32: 307-315 Hajibabaei, M., D. H. Janzen, J. M. Burns, W. Hallwachs, and P. D. Hebert. 2006. DNA barcodes distinguish species of tropical Lepidoptera. Proc. Natl. Acad. Sci. U. S. A. 103: 968-971. Harris, D. J, and K. A. Crandall. 2000. Intragenomic variation within ITS1 and ITS2 of freshwater crayfishes (Decapoda : Cambaridae): Implications for phylogenetic and microsatellite studies. Mol. Biol. Evol. 17: 284-291. Hillis, D. M., and J. P. Huelsenbeck. 1992. Signal, noise, and reliability in molecular phylogenetic analysis. J. Hered. 83: 189-195. Huelsenbeck, J. P., and F. Ronquist. 2001. MrBayes: Bayesian inferrence of phylogenetic trees. Bioinformatics. 17: 754-755. Joseph, N., E. Krauskopf, M. I. Vera, and B. Michot. 1999. Ribosomal internal transcribed spacer 2 (ITS2) exhibits a common core of secondary structure in vertebrates and yeast. Nucl. Acids Res. 27: 4533-4540.

LaJeunesse, T. C. 2001. Investigating the biodiversity, ecology, and phylogeny of endosymbiotic dinoflagellates in the genus Symbiodinium using the its region: in search of a “species” level marker. J. Phycol. 37: 866–880.

Mathews, D. H. 2005. Predicting a set of minimal free energy RNA secondary structures common to two sequences. Bioinformatics. 21: 2246–2253.

McFadden, C. S., R. Donahue, B. K. Hadland, and R. Weston. 2001. A molecular phylogenetic analysis of reproductive trait evolution in the soft coral genus Alcyonium. Evolution. 55: 54-67. Meyer, C. P. and G. Paulay. 2005. DNA barcoding: Error rates based on comprehensive sampling. Plos Biology 3: 2229-2238. Posada, D., and K. A. Crandall. 1998. Model Test: Testing the model of DNA substitution. Bioinformatics. 14:817–818.

31 Sánchez, J.A. (submitted) Morphological and M olecular Systematics of the Candelabrum Gorgonian Corals (Eunicea Lamouroux, Plexauridae; Octocorallia; Cnidaria) with description of new species and aspects of their natural history. Zootaxa Sánchez J.A., and S.D. Cairns. 2004. An unusual new gorgonian coral from the Aleutian Islands, North Pacific, Alaska. Zool. Med. 78: 265-274. Sánchez, J. A, H. R. Lasker, and D. J. Taylor. 2003a. Phylogenetic analyses among octocorals (Cnidaria): mitochondrial and nuclear DNA sequences (lsu-rRNA, 16S, and ssu-rRNA, 18S) support two convergent CLADES of branching gorgonians. Mol. Phylogenet. Evol. 29: 31-42. Sánchez, J.A., C. S. Mcfadden, S. C. France, and H. R. Lasker. 2003b. Molecular phylogenetic analyses of shallow-water Caribbean octocorals. Mar. Biol. 142: 975- 987. Shearer T. L., M. J. H. van Oppen, S. L. Romano, and G. Wörheide. 2002. Slow mitochondrial DNA sequence evolution in the Anthozoa (Cnidaria). Mol. Ecol. 11: 2475-(12), 2475-2487. Swain, T. D. and D. J. Taylor. 2003. Structural rRNA characters support monophyly of raptorial limbs and paraphyly of limb specialization in water fleas. Proc. Roy. Soc. London B 270: 887-896. Swofford, D.L. 2002. PAUP*. Phylogenetic Analysis Using Parsimony (* and other methods). Version 4.0b10. Sinauer Associates, Sunderland, MA.

Young, N. D. and J. Healy. 2003. GapCoder automates the use of indel characters in phylogenetic analysis. BMC Bioinformatics 4:6. Wirshing, H.H., C.G. Messing, C.J. Douady, J. Reed, M.J. Stanhope, and M.S. Shivji. 2005. Molecular evidence for multiple lineages in the gorgonian family Plexauridae (Anthozoa: Octocorallia). M arine Biology 147(2): 497-508.

32 CHAPT ER 2 PHYLOGENETIC HYPOTHESES OF GORGONIID OCTOCORALS ACCORDING TO ITS 2 AND THEIR PREDICT ED RNA S ECONDARY S TRUCTURES 2 2.1. ABSTRACT

Octocoral systematics rely on several traits such as colonial form, skeleton, polyp, and sclerites but due to the reduced number of homologous characteristics among species it’s systematics is one of the most difficult among invertebrates. Gorgoniid octocorals taxonomy (Cnidaria; Octocorallia; Gorgoniidae) includes diagnostic characters not well defined at the generic level, and based on the family diagnosis some species could be classified in either Gorgoniidae or Plexauridae. In this study, we used sequences from the Internal Transcribed Spacer 2 (ITS2) and their predicted RNA secondary structure to both correct the alignment and reconstruct phylogenies using molecular morphometrics for 24 octocorals from the Atlantic and Pacific. ITS2 exhibited the six-helicoidal ring-model structure found in eukaryotes, and provided 38 parsimony-informative characters. The proposed phylogenies, though differing between sequence- and structure-base results, provided consistent support for recently described taxa and some internal clades. ITS2 offered a large number of characters validating neglected monophyletic Gorgoniidae genera.

Keywords: Gorgoniidae, gorgonians, octocorals, phylogeny, molecular systematics, ITS2, RNA secondary structure, molecular morphometrics.

2 Aguilar, C., Sánchez, J. A. (submitted). Phylogenetic hypotheses of gorgoniid octocorals according to ITS2 and their predicted RNA secondary structures. Molecular Phylogenetics & Evolution.

33 2.2. INTRODUCTION

The Internal Transcribed Spacers (ITSs), found within the nuclear ribosomal RNA tandem arrays in eukaryotes, have not been considered appropriate for molecular phylogenetics in some invertebrates, mostly due to excessive INDELS (insertions- deletions), saturation, and/or intragenomic variation (e.g., Frabry et al., 1999; Harris and Crandall, 2000; Marquez et al., 2003; Volmer and Palumbi, 2004). Even so, this is a promissory nuclear region for some lower eukaryotes, particularly at the intra-familial level and down to closely-related species, where other known regions are nearly invariable (e.g., LaJeunesse, 2001). In scleractinian corals, excepting some species of Acropora (Marquez et al., 2003), ITS sequences have reliable phylogenetic information especially if the RNA predicted secondary structure is compared (Chen et al., 2004). In the case of octocorals, the ITS sequences have been very useful to compare species of alcyoniid octocorals as well as to point out hybridization among species (e.g., McFadden et al., 2001). These sequences across many octocoral taxa have shown considerable variation within genera (see Chapter I) with description of new species and aspects of their natural history) and perhaps they are among the most variable genomic regions in these organisms where mitochondrial DNA and other sequences are much conserved (Berntson et al., 2001; Shearer et al., 2002). Nonetheless, ITS alignment always produced multiple and variable INDELS depending on the different gap opening and/or extension penalties (e.g., multiple alignment), which makes somewhat unreliable the phylogenetic inference. Consequently, an approach to study comparatively these sequences is necessary in order to improve the homology among sequences. One such approach is to rely on the use of predicted RNA secondary structure. RNA secondary structure can be determined by an alignment against sequences with already known structures to depict common base-pairing patterns in the absence of crystallized structures. Secondary structures are particularly useful in systematics because they include “morphological” informative characters not found in the primary sequence (e.g., molecular morphometrics: Caetano-Anollés, 2002; Billoud et al., 2000; Swain and Taylor, 2003). In addition, these structures are of importance in phylogenetic analysis,

34 because they are used to correct alignments between species that sometimes are not accurate by multiple alignments (e.g., Sánchez et al., 2003b). These structures have conserved zones that ease reconstructing unknown RNA secondary structures. As a result, most RNA secondary structures known today have been constructed using this approach. For octocorals, the secondary structures from 21 species (Octocorallia) comprising a part of the region Lsu-rRNA (16S) have been determined (Sanchez et al., 2003b). The ITS2 secondary structure is a helicoidal ring with 5 to 6 helixes. Compelling evidence proposed by Joseph et al. (1999) in yeast shows the fundamental role of the helicoidal ring of the ITS2 transcript during the pre-rRNA processing, whose lack otherwise blocks the production of the mature 25S rRNA when specific structures of the ring model can not be formed (see also Cote et al., 2002; Chen et al., 2004; Gottschling and Plötner, 2004). Octocoral systematics is one of the most puzzling among invertebrates due to the reduced number of homologous characteristics, which are oftentimes also highly plastic and variable within and among species (Bayer, 1961). In addition, there are no landmarks in any of octocoral morphological characters (Sánchez and Lasker, 2003) and comparisons against molecular phylogenies have shown a great deal of convergence and homoplasy (Sánchez et al., 2003a; 2003b; Sánchez 2004). Therefore, classification rearrangements relying on the continuity of morphological forms are frequent for octocoral taxonomy (e.g., Grasshoff, 1988). In brief, the problem in octocoral systematics has been the absence of an independent source of characters to critically test taxonomic observations. Opportunely, DNA sequence characters have been able to turn these observations into testable hypotheses creating a rebirth for taxonomy (e.g., Knowlton, 2000; Mallet and Willmott, 2003). Using ITS2 DNA sequences and their predicted RNA secondary structure the goals of the paper were to review the Gorgoniidae relationships with emphasis on species with asymmetrically spiny sclerites, including Muriceopsis spp., and new gorgoniid taxa recently described by Sánchez (submitted).

35 2.3. MATERIALS AND METHODS 2.3.1. ITS2 sequences

Representative species from the Gorgoniidae with and without asymmetrical spiny sclerites in the surface coenenchyme, as well as a few Plexauridae as outgroups, were examined for molecular analyses. Table 2.1- Species name and GenBank Accession numbers from the reanalyzed Internal Transcript Spacer, partial sequence.

Species Accecion no. Africagorgia schoutedeni AY587533 Eunicea laciniata AF052926. Filigorgia africana AY587532 Gorgonia flabellum AY587521 Gorgonia mariae AY587523 Gorgonia ventalina AY587522 Leptogorgia setacea AY587525 Leptogorgia virgulata AY587526 Lophogorgia euryale AY587530 Lophogorgia hebes AY587529 Lophogorgia miniata AY587528 Lophogorgia stheno Mureceopsis flavida AY587537 Muriceopsis bayeri AY587538 Pacifigorgia stenobrochis AY587531 Pinnigorgia flava AY587535 Pinnigorgia platysoma AY587536 Plexaurella nutans Pseudopterogorgia bipinnata AY587524 Pterogorgia citrina Tobagorgia hardyi AY587534

DNA extractions were made according to Sánchez et al. (2003) from alcohol preserved material (Ethanol 95%) and using the DNeasy kit (Qiagen) for the extractions with high concentration of dark pigments that usually did not amplify properly. Gorgoniidae and Plexauridae were sequenced according to the methods and protocols presented in Sánchez et al. (2003b) consisting in amplifying the region between the 5’ end

36 of the 5.8s ribosomal gene and the 3’ end of the 28S ribosomal gene containing the complete Inter-transcribed Spacer-2 or ITS2 (5.8S-436: 5’- AGCATGTCTGTCTGAGTGTTGG-3’ and 28S-663: 5’- GGGTAATCTTGCCTGATCTGAG-3’, numbers respect to the sequence of Alcyonium digitatum, Genbank Acc. No. AF262347: McFadden et al., 2001). PCR conditions had one initial step of 2.0 min at 94°C followed by 30 cycles of 30 sec at 94°C, 45 sec at 56°C, 45 sec at 72°C, and a final extension step for 5.0 min at 72°C. DNA for sequencing was acquired from the mixture of two PCR reactions containing 56 µl total (volume completed with ddH2O) each with 1 µl of DNA template (1:50 dilutions of genomic DNA extractions), 2 units Taq polymerase (Promega), 3 µl of 10X Buffer (Promega), MgCl2, 0.15 µM dNTPs mix, 0.16 µM of each primer. Gene purification was made using the Edge Biosystems kit and sequencing using BigDye 3.1 (AB 3100, capillary electrophoresis automated sequencer, Molecular Systematics, Laboratories of Analytical Biology, Museum Support Center-NHMH, Suitland, Maryland, USA). The raw sequences were deposited in GENBANK. Additional, information on voucher specimens is available in Sánchez (submitted-a).

2.3.2. ITS2 predicted RNA secondary structures. Secondary structures for 24 octocoral species were reconstructed by aligning their sequences (using Bioedit Hall, 1999) with homologous structures already published, like Acropora spp. (Odorico and M iller, 1997). Moreover, manual alignment was performed by visual homology for the construction of a Dedicated Comparative Sequence Editor (DCSE) format in order to perform different phylogenetic analysis (e.g., Sánchez et al., 2003b) . DCSE format uses square brackets ([ or ]) to delimitate each helix, braces ({ or }) for bulges and loops, and hyphens for gaps in the alignment, terminal loops appear separated by inverse brackets (] and [). The acquired structures with restrictions and constrains were submitted in MFOLD (http://www.bioinfo.rpi.edu/applications/mfold/old/rna/). RNA was folded at a fixed temperature of 37°C, and the structure chosen from different output files

37 was the one with the highest negative free energy. The obtained secondary structures were used to construct a matrix for a cladistic analysis by the means of molecular morphometrics based on geometrical features and base number (Billoud et al., 2000; Swain and Taylor, 2003). This means, secondary structures helixes were numbered and treated like a character that will vary depending on their base number. Bulges and internal loops were given the number of the helix followed by a letter, ending helixes loops and separation segments where designated with the “i” letter. Loops complement, helix complement, and separation segments were indicated with an apostrophe (e.g. 4a loop with its complement 4a’). Nucleotides were count for each character and arranged in discrete character states ranging from 0 to 5 (being 0 the character with less nucleotide number) as seen in appendix 2.

2.3.3. ITS2 Sequence Analysis For both molecular morphometrics and sequences alignment, were accomplished through different phylogenetic programs. First, a molecular morphometry hypothesis was obtained by maximum parsimony using the exhaustive branch and bound algorithm in PAUP* (Swofford, 2002). Also in PAUP*, both maximum parsimony and maximum likelihood analyses were made for sequence alignment corrected by the secondary structure (gaps as missing), and for the primary alignment made by ClustalW. For maximum likelihood, settings were made according to the best model suggested by the Akaike Information Criterion (AIC) in Modeltest (Posada and Buckely, 2004). Bayesian inference of phylogeny was done using MrBayes (Huelsenbeck and Ronquist, 2001) for both matrices (secondary sequence alignment and molecular morphometrics) using partitions for each helix (see figure legends for tree details). Different phylogenetic hypothesis were constructed excluding and including in multiple combinations Plexaurella nutans, Pterogorgia citrina and Muriceopsis-Tobagorgia. To analyze sequences saturation Tamura-Nei distance, transitions values, transversions values, and their radio (Ti/Tv) was obtained in PAUP* (Swofford, 2002) for the corrected alignment.

2.4. RESULTS New sequences and predicted RNA secondary structures for the ITS2 of 24 octocorals were obtained (Fig. 2.1; Appendix 2). ITS2 sequences, aligned for the secondary structure base pairing, had 465 positions including DCSE symbols for RNA secondary structures (Appendix 2). All sequences were unique and a total of 56 constant characters were useful for keeping the similarity among sequences (Fig. 2.1). The unpermuted most- parsimonious tree was significantly shorter than 1000 permutation replicates (>803, PTP: P<0.001) whereas the length frequency distribution of the random trees (mean = 1193.5: SD = 54.88) was significantly skewed towards less optimum trees (g1= -0.89, P<0.05 g1- test Hillis and Huelsenbeck; 1992) with respect to the most parsimonious tree (L= 569: Fig. 2.2). These two analyses indicate predominance of phylogenetic signal over homoplasy and noise. Only one species, Muriceopsis. bayeriana, showed tandem repetitive sequences, but this did not affect the informative characters, secondary structure reconstruction, or the overall alignment. The saturation plots obtained with Tamura-Nei distance, showed clearly that the ITS2 secondary sequences alignment were saturated (Fig. 2.3), which could have leaded to misleading results.

Ta bl e 2 .2 Species / Characters matrix from molecular morphometrics. Character numbering refers t o RNA predict ed secondary struct ure as in Fig. 2.1.

Spp./Characters 1i 2 2i 2' 2i' 3 3a 3i 3' 3'a 3i' 4 4a 4b 4i 4' 4'c 4'b 4'a 4i' 5 5a 5b 5c 5d 5i' 5' 5'd 5'c 5'b 5'a 5i' 6 6i 6' 6i' G_ventalina 2131112 41 0 112 0 21 0 1 0 221 3 1 1 1 2 021170401 G_mariae 2131102 40 0 122 0 02 0 1 0 221 3 1 1 1 2 021170401 G_flabellum 2131112 41 0 112 0 21 0 1 0 221 3 1 1 1 2 021170401 Pi_flava 2121351 15 0 120 2 33 3 0 0 200 0 1 2 0 0 220230401 Pi_Platysoma 2121330 05 1 143 1 33 3 0 0 202 3 1 2 0 0 220330201 T_hardyi 1232030 33 1 001 0 30 0 1 0 100 1 1 1 3 0 023241211 A_schoutedeni 1303230 03 0 121 0 12 0 1 0 302 1 1 2 3 0 222330500 F_africana 1303210 02 0 122 0 12 0 1 0 311 4 1 0 2 3 220130500 P_stenobrochis 2131241 04 1 031 2 42 1 1 0 131 3 1 1 1 2 011160001 L_Stheno 2131230 23 0 042 1 24 1 1 0 111 4 1 2 1 1 022151112 L_euryale 2131230 23 0 042 1 24 1 1 0 111 4 1 2 1 1 022151112 L_hebes 0333020 12 0 042 1 24 1 1 0 111 4 1 2 1 1 022151112 Lo_miniata 2131220 12 0 042 1 24 1 1 0 111 4 1 2 1 1 022151112 L_violacea 2131240 14 0 042 1 24 1 1 0 111 4 1 2 1 1 022160302 L_virgulata 2131230 33 0 042 1 24 1 1 0 111 4 1 2 1 1 022120401 Le_setacea 2131230 33 0 042 1 24 1 1 0 111 4 1 2 1 1 022120401 M_flavida 0414121 23 1 040 2 15 2 0 1 030 1 0 1 3 4 113101611 M_bayeriana 0414152 25 1 043 1 05 3 0 1 130 0 1 1 3 4 023210602 E_laciniata 2131250 15 0 111 0 01 0 0 0 030 4 1 3 4 4 223013233 Pl_nutans 1 2 32030 33 1 031 1 32 0 0 1 031 1 1 3 3 4 001020501 Pt_citrina 0333120 22 1 030 0 32 0 0 0 202 2 0 0 3 0 012120501 P_bipinnata_CBC 2030210 21 0 020 1 02 2 1 0 222 1 0 1 3 2 001103023 P_bipinnata_B 2030210 01 1 102 1 21 2 1 0 221 2 1 1 2 2 021103033 P_elisabethae_CBC 2030200 40 0 102 1 01 1 1 0 231 2 1 1 2 3 02114232139

Fi gure 2.1.- ITS2 predicted RNA secondary structures for 9 species of octocorals and their enthalpy values of structure formation according to MFOLD: (A) Pseudopterogorgia bipinnata CBC (Carrie Bow Cay, Belize), numeration represents characters used for molecular morphometry as in Table 1, ∆G= -58.96 kcal/mole; (B) Eunicea laciniata, ∆G= - 103.02 kcal/mole; (C) Tobagogorgia hardyi, ∆G= -67.47 kcal/mole; (D) Africagorgia schoutedeni, ∆G= -80.92 kcal/mole; (E) Lophogorgia miniata, ∆G= -62.67 kcal/mole; (F) Muriceopsis bayeriana, ∆G= -80.27 kcal/mole; (G) Gorgonia ventalina, ∆G= -69.67 kcal/mole; (H) Leptogorgia setacea, ∆G= -67.66 kcal/mole; (I) Pacifigorgia stenobrochis,∆G= -69.53 kcal/mole.

Figure 2.2.- Molecular phylogenetic hypothesis of 22 species of Gorgoniidae and Plexauridae (Muriceopsis spp. and Eunicea laciniata). A, maximum likelihood phylogram from corrected sequence alignment, showing support nodes from maximum likelihood TVM+I+G model, invariant sites and gamma distribution / Bayesian-estimated likelihood (settings according to MrModeltest) tree using partitions by helix, 10 million Monte Carlo markovian chain generations (Bayesian-Monte Carlo simulation by MrBayes sampling every 100 simulation, burn-in 10000) node support from the >50%-majority rule consensus from the 1001 sampled trees. B, maximum parsimony phylogram from corrected sequence alignment using branch and bound algorithm, Length=638, Consistency index=0.726, Retention index= 0.697, support nodes bootstrap 50% consensus, 1000 replicates. C,

41 molecular morphometrics phylogram of 36 characters, Length=204, Consistency index=0.554, Retention index= 0.643, (G-test from 100 trees g1= -0.54 g2= 0.29, showing support nodes from maximum parsimony non-parametric 100 boot strap replicat es / Bayesian 50% majority rule consensus of 1001 trees analysis using partitions by helix (simulations as above). CBC= Carrie Bow Cay, Belize; B= San Salvador, Bahamas

Fi gure 2.3.- Saturation plots of ITS2 sequences of 22 octocorals species. Tamura-Nei distance obtained with assumed proportion of invariable sites (0.1699), and a gamma-shape parameter (2.2647).

Excluding Plexaurella nutans and Pterogorgia citrina from the analyses (see below), there were four different phylogenetic hypotheses corresponding to maximum likelihood and maximum parsimony, using the secondary sequence alignment (Appendix 1), one with primary sequence alignment, and maximum parsimony using the molecular morphometrics matrix (Table 2). Main differences among the four hypotheses consisted in the placement of Muriceopsis spp. and Africagorgia-Filigorgia (Fig. 2.2). Maximum parsimony, maximum likelihood, using secondary sequence alignment, and maximum likelihood using primary alignment had Africagorgia-Filigorgia basal respect other gorgoniids and Muriceopsis together with Tobagogorgia (Fig. 2.2A-B, 2.4A ). The ITS2 sequences exhibited the highly conserved six-helicoidal ring-model structure found in yeast, insects, and vertebrates and provided 38 parsimony-informative characters using molecular morphometrics (g1= -0.54, P<0.01). Maximum parsimony,

42 using the molecular morphometrics matrix (Table 2), placed Muriceopsis as basal of the rest of gorgoniids and Africagorgia-Filigorgia with Tobagogorgia and Pinnigorgia (Fig. 2.2C). Overall, conflicting results among the four phylogenetic methods are rooted in a few characters for the basal nodes evident with the low support from bootstrap or Bayesian clade credibility values (Fig. 2.2, 2.3A), resulting in competing hypothesis in optimal trees. According to the classification proposed by Sánchez (submitted-a) based on electron microscopy of sclerites “Gorgoniids with capstan derived sclerites” (e.g., Pacifigorgia, Lophogorgia, and Leptogorgia) were characterized by a long helix IV with one internal loop and a helix V with four internal loops (Fig. 2.1E, H, I); these characteristics support this clade together in all hypotheses. “Gorgoniids with C-shaped sclerites” had a predominantly long helix V if compared to helix IV, but genera such as Gorgonia spp. and Pseudopterogorgia spp. presented observable differences among them (Fig 2.1A, G). “Gorgoniids with asymmetric spiny sclerites” (Muriceopsis, Tobagogorgia, and Pinnigorgia) exhibited one or two lateral bulges in the V helix (e.g., Fig. 2.1C, F), though this clade was only monophyletic in the maximum likelihood tree using the sequence alignment (Fig. 2.4A). M. flavida and M. bayeriana were characterized by a big loop in helix VI and an internal one in helix III (Fig. 2.1F; M. flavida structure not shown) which probably droved them apart to a basal clade in the morphometry analysis (Fig. 2.2C). “Gorgoniids with spindles” (Africagorgia and Filigorgia) had a short helix IV and a big central ring (Fig. 2.1D), which was observed in the outgroup Eunicea laciniata (Fig. 2.1B). This last result brought along support for the three sequence alignment hypotheses, in which Africagorgia-Filigorgia are placed basal to the rest of gorgoniids and closer to the outgroup (Fig. 2.2A,B; Fig. 2.4A). Plexaurella nutans and Pterogorgia citrina were initially excluded from the analyses because the comprised very long branches resulted in inconsistent results. P. nutans and P. citrina grouped together in a clade with Muriceopsis spp. and Tobagogorgia hardyi (Fig. 2.4 B, C). Muriceopsis-Tobagorgia was a likely source of long-branch attraction; however, in the absence of Plexaurella (Fig. 2.4F), Pterogorgia grouped with

43 Pinnigorgia (still an inconsistent result compared to morphology, see Sánchez submitted- b), in absence of Pterogorgia (Fig. 2.4E), Plexaurella still group with Muriceopsis- Tobagorgia and using sequence data it was placed basal of all studied gorgoniids (data not shown). Excluding Muriceopsis-Tobagorgia from the analyses, Plexaurella and Pterogorgia grouped together as a sister branch of Pinnigorgia spp. (Fig. 2.4D). In summary, it was evident that long-branch attraction was resulting in phylogenetic inconsistencies for the position of Plexaurella nutans and Pterogorgia citrina.

45 Muriceopsis-Tobagorgia species. E. Excluding Pterogorgia citrina F. Excluding. Plexaurella nutans. Overall methods and program settings as in Fig. 2.2.

2.5. DISCUSSION DNA sequences from the nuclear ITS2 region, including information from predicted RNA secondary structure, despite their reduced length, provided a great number of characters and phylogenetic information among Gorgoniidae genera and species. In contrast, mitochondrial DNA sequences provided limited support to resolve phylogenetic relationships within Gorgoniidae (e.g., Sánchez et al., 2003b). It was interesting that the potential inconsistent results with ITS2 such as the placement of Pterogorgia spp., Plexaurella spp. and Muriceopsis spp. grouping separate and basal to most gorgoniids (Fig. 2.4B,C) were also encountered with mtDNA (Sánchez et al., 2003b), which suggests it is a problem beyond a particular genome (e.g., nuclear or mitochondrial) or DNA sequence. ITS2 sequences also recovered other phylogenetic hypotheses for the new and enigmatic deep-water octocoral Alaskagorgia aleutiana as a member of Plexauridae (see Chapter I), as observed with mtDNA (Sánchez and Cairns, 2004). Even though ITS2 information presented numerous INDELS, the information presented here did not used them (e.g., gaps as missing) and tests disclosed that saturation or noise was present due to the great number of transversions above transitions (fig. 2.3). Fortunately, there is no ITS2 intra-genomic variation in octocorals (Sánchez, unpublished; Gutierrez-Rodriguez, personal communication) as it has been observed in other corals (e.g., Marquez et al., 2003 but see Chen et al., 2004). ITS2 sequences have also been useful in resolving phylogenetic relationships within the plexaurid genus Eunicea (Sánchez, submitted-b) with description of new species and aspects of their natural history; Grajales et al., unpublished) and ITS2 is certainly a better marker for intrageneric comparison. Consequently, ITS2 offers a great tool, with moderate effort (e.g., fast and straightforward PCR and sequencing), for octocorals within a certain range of species affinity (e.g., subfamilies and genera), and because ITS2 is short (<300 bp), even partially degraded DNA from museum specimens (e.g., Africagorgia spp., Filigorgia spp. and Pinnigorgia spp. presented here: see details in

46 Sánchez submitted-b) can be readily amplified and sequenced. It is important to remark that there are still many genes to survey in the octocoral genome but it is a good start to have already a species marker available. Mitochondrial DNA is nearly invariable in octocorals, and other anthozoans (Shearer et al., 2002) therefore DNA sequences from other genomes should be used for DNA barcoding in octocorals, and perhaps ITS2 is one such region. In addition, ITS2 has the extra information in its predicted RNA secondary structures, which can ease the bioinformatics analyses towards a barcode system in octocorals. ITS2 Secondary structures found in these octocorals presented common hallmarks found in other corals (e.g. Acantherastra spp., Chen et al. 2004) and eukaryotes (Schultz et al. 2005), which gave reliability to the phylogenies; since, it was made based in conserved characteristics of the whole eukarya kingdom. This for demonstrating that ITS2 secondary structure can work for mega-systematic analysis down to population level (data not shown). The phylogenetic hypotheses presented here provided valuable information and support for describing new taxa (Fig. 2.2) (e.g., Tobagogorgia hardyi and Africagorgia), validating neglected taxa (e.g., Filigorgia) and a preliminary internal division of Gorgoniidae (Sánchez, submitted-a). Usually, Gorgoniidae species include diagnostic characters that are not well defined at the generic level, and based on the family diagnosis some species could be classified in either Gorgoniidae or Plexauridae, the two most common gorgonian groups in shallow waters. For instance, gorgoniids have been characterized by having reduced sclerites that are regularly sculpted or ornate in thin branches (Grasshoff and Alderslade, 1997) but some genera descriptions include traits such as tubercles arranged in whorls, though irregular in some species (Bayer, 1961; Grasshoff and Alderslade, 1997), whose irregularity could correspond to Plexauridae. The sclerite size and coenenchyme thickness are also confusing among the two families. The genus Plexaurella has reduced (<0.5 mm), regularly ornate sclerites whereas having the thickest coenenchyme of Plexauridae. Muriceopsis, classified in the same family, has also sclerites of intermediate size among the two families and thin coenenchyme as in Gorgoniidae (Sánchez, 2001). According to morphology Pterogorgia should be with “Gorgoniids with C-shaped sclerites” (Sánchez,

47 submitted-b) and Plexaurella should be a plexaurid (Bayer, 1961, but see Sánchez et al., 2003a). Both Muriceopsis spp. and Plexaurella spp. have sclerite types different from those of other plexaurid species and previous molecular phylogenies of Plexauridae and Gorgoniidae, according to three mtDNA coding genes show how Muriceopsis and Plexaurella group with Gorgoniidae (Sánchez et al., 2003b; Wirshing et al., 2005). This affinity with gorgoniids was also evident from ITS2. Unfortunately, the long-branch attraction effect confuses any straightforward conclusion on their natural classification (Fig. 2.4). Species from Leptogorgia and Lophogorgia were not closely related with Africagorgia-Filigorgia, which were considered previously as part of the same genus (Grasshoff, 1988). Particularly the assertion that the gorgoniids with asymmetrical spiny sclerites are part of a continuum within Gorgoniidae is not supported by recently published Scanning Electron Microscopy (SEM) (Sánchez, submitted-b) and the phylogenic hypotheses of ITS2. Supporting and additional evidence comes from mitochondrial DNA sequences from the mut-S homolog gene (MSH1). Using MSH1, Lepard (2003) found that supposed Leptogorgia “outgroups” such as Pseudopterogorgia, Pacifigorgia, and Eugorgia were intermingled with various clades of Leptogorgia providing a glimpse of the poor current systematic treatment of both Lophogorgia and Leptogorgia. Results by LePard (2003) clearly show that some of the African fauna from Leptogorgia sensu lato are not monophyletically related as it was found here as well. Finally it is important to mention that the phylogeny of Gorgoniidae is not by any means complete, the inclusion of other genera unfortunately not available for DNA analyses such as the enigmatic Pacific Pseudopterogorgia, which morphology does not resemble their Caribbean counterparts (e.g., Bayer, 1961; Williams and Lindo, 1997), should be included to better understand the phylogeny of this important group of octocorals.

48 2.6. ACKNOWLEDGEMENTS Special thanks to the Facultad de Ciencias, Department of Biological Sciences, and BIOMMAR at Universidad de los Andes for providing partial resources to perform the analysis and to develop this paper. The Systematics Association (UK) grant supported partially the analyses of this paper. J.A. Sánchez acknowledges a postdoctoral fellowship from the Smithsonian Institution-NMNH (2002-2003). We are also indebted to Andrea Lepard, who provided unpublished phylogenetic results for preliminary comparisons and helpful insight on her valuable molecular results on gorgoniids. This work is part of an on- going project of biodiversity of Tobago, which is being conducted by Jerry D. Hardy, Jr. for the Tobago House of Assembly. Travel funds for our work in Tobago provided by J.D. Hardy. Patricia Turpin, Man-o-War Bay Cottages, Charlotteville, Tobago, provided housing for the Tobago expedition. Stephen D. Cairns, Frederic Bayer (NMNH Smithsonian Institution) Scott France, and Odalisca Breedy provided helpful discussions. M. Grasshoff (Senkenberg museum) kindly sent fragments of African specimens for molecular and SEM analyses. Phil Alderslade (Northern Territory museum, Darwin) and Catherine McFadden (Harvey Mudd college) kindly sorted and provided Indopacific specimens. The Smithsonian Institution, Invertebrate Workshop at Bocas del Toro, Panama (Rachel Collin, director) provided facilities to obtain Leptogorgia spp. and Muriceopsis spp. Assistance and advice from the staff of the laboratories of analytical biology (Liz Zimmer, Ken Schallop, and Lee Weight) are significantly appreciated.

49 REFERENCES

Aurivillius, M ., 1931. The gorgonarians from Dr. Sixten Bock’s expedition to Japan and Bonin Islands 1914. Kungliga Svenska Vetenskapsakademiens Handlingar. 9, 1-337. Bayer, F.M., 1953. Zoogeography and evolution in the octocorallian family Gorgoniidae. Bull. Mar. Sci. Gulf Caribbean. 3,100-119. Bayer, F.M., 1961, The shallow water Octocorallia of the West Indian region. Stud. Fauna Curaçao. 12, 1-373. Bayer, F.M., Macintyre, I.G., 2001. The mineral component of the axis and holdfast of some gorgonacean octocorals (Coelenterata: Anthozoa), with special reference to the family Gorgoniidae. Proc. Biol. Soc. Wash. 114, 309-345. Berntson, E. A., F. M. Bayer, A. G. McArthur, and S. G. France. 2001. Phylogenetic relationships within the Octocorallia (Cnidaria : Anthozoa) based on nuclear 18S rRNA sequences. Mar. Biol. 138: 235-246. Billoud, B., Guerrucci, M. A., M. Masselot., Deutsch, J. S., 2000. Cirripede phylogeny using a novel approach: molecular morphometrics. Mol. Biol. Evol. 17, 1435–1445. Breedy, O., Guzman, H.M., 2003. Octocorals from Costa Rica: The genus Pacifigorgia (Coelenterata: Octocorallia: Gorgoniidae). Zootaxa 281, 1-60. Caetano-Anollés G., 2002. Tracing the evolution of RNA structure in ribosomes. Nucleic Acids Res. 30, 2575 - 2587. Chen, C. A., C. C. Chang, N. V. Wei, C. H. Chen, Y. T. Lein, H. E. Lin, C. F. Dai, and C. C. Wallace. 2004. Secondary structure and phylogenetic utility of the ribosomal internal transcribed spacer 2 (ITS2) in Scleractinian Corals. Zool. Stud. 43: 759-771. Cote, C. A., C. L. Greer, and B. A. Peculis. 2002. Dynamic conformational model for the role of ITS2 in pre-rRNA processing in yeast. RNA 8: 786-797. Deichmann, E., Bayer, F.M., 1959. The lemon-colored plexaurids from the West Indies and Brazil. Breviora 115, 1-17.

50 Fabricius, K., Alderslade, P., 2001. Soft corals and sea fans: a comprehensive guide to the tropical shallow water genera of the central-west Pacific Ocean and the Indian Ocean and Red Sea. Melbourne: New Litho. Fabry S., A. Kohler, and A. W. Coleman. 1999. Intraspecies analysis: Comparison of ITS sequence data and gene intron sequence data with breeding data for a worldwide collection of Gonium pectorale. J. Mol. Evol. 48: 94-101. Filatov, D., 2001. Processor of sequences manual. University of Birmingham. Available at http://www.biosciences.bham.ac.uk/labs/filatov/proseq.html Fu, J., Murphy, R.W., 1999. Discriminating and locating covariance: an application of permutation tail probability (PTP) analyses. Syst. Biol. 48, 380-395. Fukami, H., Budd, A.F., Paulay, G., Sole´-Cava, A., Chen, C.A., Iwao, K., Knowlton, N, 2004. Conventional taxonomy obscures deep divergence between Pacific and Atlantic corals. Nature 427, 832-835. Goldberg, W.M., 2001. The sclerites and geographic distribution of the gorgonian Swiftia exserta (Coelenterata: Octocorallia: Holaxonia). Bull. Biol. Soc. Wash. 10, 100- 109. Gottschling M. and J. Plötner. 2004. Secondary structure models of the nuclear internal transcribed spacer regions and 5.8S rRNA in Calciodinelloideae (Peridiniaceae) and other dinoagellates. Nucl. Acids Res. 32: 307-315 Grasshoff, M., 1977. Die Gorgonarien des östlichen Nordatlantik und des Mittelmeeres. III. Die Familie Paramuriceidae (Cnidaria, Anthozoa)."M eteor" Forsch.-Ergebnisse 27, 5-76. Grasshoff, M., 1988. The genus Leptogorgia (Octocorallia: Gorgoniidae) in West Africa. Atlantide Report 14, 91-147. Grasshoff, M., 1992. Die Flachwasser-Gorgonarien von Europa und Westafrika (Cnidaria, Anthozoa). Forschungsinstitut Senckenberg 149, 1-135.

51 Grasshoff, M. and P. Alderslade. 1997. Gorgoniidae of Indo-Pacific reefs with descriptions of two new genera (Coelenterata: Octocorallia). Senckenbergiana biologica 77(1): 23-25. Guzmán H. M., Cortés, J., 1985. Organismos de los arrecifes coralinos de Costa Rica. IV. Descripción y distribución geográfica de octocoralarios (Cnidaria: Anthozoa) de la costa Caribe. Brenesia 24, 125-173. Hall, T.A., 1999. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symp. Ser. 41, 95-98. Harris, D. J, and K. A. Crandall. 2000. Intragenomic variation within ITS1 and ITS2 of freshwater crayfishes (Decapoda : Cambaridae): Implications for phylogenetic and microsatellite studies. Mol. Biol. Evol. 17: 284-291. Higgins, D.G., Thompson, J.D., Gibson, T.J., 1996. Using CLUSTAL for multiple sequence alignments. Methods Enzymol. 266, 383-402 Hillis, D.M ., Bull, J.J., 1993. An empirical test of bootstrapping as a method for assessing confidence in phylogenetic analysis. Syst. Biol. 142, 182-192. Hillis, D.M ., Huelsenbeck, J.P., 1992. Signal, noise, and reliability in molecular phylogenetic analysis. Journal of Heredity 83, 189-195. Huelsenbeck, J.P., Ronquist, F., 2001. MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics 17, 754–755. Joseph, N., E. Krauskopf, M. I. Vera, and B. Michot. 1999. Ribosomal internal transcribed spacer 2 (ITS2) exhibits a common core of secondary structure in vertebrates and yeast. Nucl. Acids Res. 27: 4533-4540.

Knowlton, N., 2000. Molecular genetic analyses of species boundaries in the sea. Hydrobiologia 420, 73-90. Kukenthal, W. 1919. Gorgonaria. Wiss. Ergebn. Dt. Tiefsee-Exped. “Valdivia” 13, 1-946. LaJeunesse, T. C. 2001. Investigating the biodiversity, ecology, and phylogeny of endosymbiotic dinoflagellates in the genus Symbiodinium using the its region: in search of a “species” level marker. J. Phycol. 37: 866–880.

52 LePard, A., 2003. Analysis of variation in the mitochondrial encoded MSH1 in the genus Leptogorgia (Cnidaria: Octocorallia) and implications for population and systematics studies. M.S. thesis. University/College Charleston, Charleston, USA. Lewis, J.C., Von Wallis, E.,1991. The function of surface sclerites in gorgonians (Coelenterata Octocorallia). Biol. Bull. 181, 275-288. McFadden, C.S., Donahue R., Hadland B.K. Weston R., 2001. A molecular phylogenetic analysis of reproductive trait evolution in the soft coral genus Alcyonium. Evolution 55, 54-67. M allet, J., Willmott., K., 2003. Taxonomy: renaissance or tower of babel? Trends Ecol. Evol. 18, 57-59. M arquez, L., Miller, D., M acKenzie, J., Van Oppen, M.J.H., 2003. Pseudogenes contribute to the extreme diversity of nuclear ribosomal DNA in the hard coral Acropora. Mol. Biol. Evol. 207, 1077-1086. Odorico D. M ., Miller D. J, 1997. Variation in the ribosomal internal transcribed spacers and 5.8S rDNA among five species of Acropora (Cnidaria; Scleractinia): Patterns of variation consistent with reticulate evolution. Mol. Biol Evol. 14 (5): 465-473. Posada, D., Buckley, T.R., 2004. Model selection and model averaging in phylogenetics: advantages of the AIC and Bayesian approaches over likelihood ratio tests. Syst. Biol. 53, 793-808 Schultz, J., Maisel, S., Gerlach, D., Müller, T., Wolf, M. (2005). A common core of secondary structure of the internal transcribed spacer 2 (ITS2) throughout the Eukaryota. RNA 11: 361-364. Sánchez, J.A., 2001. Systematics of the Southwestern Caribbean Muriceopsis Aurivillius (Cnidaria: Octocorallia) with description of a new species. Bull. Biol. Soc. Washington 10, 160-180. Sánchez, J.A., 2004. Evolution and dynamics of branching colonial form in marine modular Cnidarians: Gorgonian Octocorals. Hydrobiologia 530, 283-290.

53 Sánchez, J.A. (submitted-a) New taxa of Atlantic gorgoniid octocorals (Gorgoniidae: Octocorallia). Journal of Natural History. Sánchez, J.A. (submitted-b) Morphological and molecular systematics of the Caribbean candelabrum octocorals (Eunicea Lamouroux, 1816; Cnidaria; Octocorallia; Plexauridae) with description of new species and aspects of their natural history. Invertebrate Systematics. Sánchez J.A., Cairns, S.D., 2004. An unusual new gorgonian coral from the Aleutian Islands, North Pacific, Alaska. Zool. Med. 78, 265-274. Sánchez J.A., Lasker, H.R., 2003. Patterns of Morphologic Integration in Branching Colonies of Marine Modular Organisms: supra-module organization in Gorgonian Corals. Proc. Roy. Soc. London B 270, 2039-2044. (DOI: 10.1098/rspb.2003.2471) Sánchez J.A., Lasker, H.R., Taylor, D.J., 2003a. Phylogenetic analyses among octocorals (Cnidaria) according to mitochondrial and nuclear DNA sequences (lsu-rRNA 16S, and ssu-rRNA 18S) support two convergent clades of branching gorgonians. Mol. Phylogenet. Evol. 29, 31-42. (DOI:10.1016/S1055-7903(03)00090-3) Sánchez, J.A., Mcfadden C.S., France, S.C., Lasker, H.R., 2003b. Molecular Phylogenetic analyses of shallow-water Caribbean octocorals. Marine Biology. 142, 975-987. (DOI10.1007/s00227-003-1018-7) Shearer T. L., M. J. H. van Oppen, S. L. Romano, and G. Wörheide. 2002. Slow mitochondrial DNA sequence evolution in the Anthozoa (Cnidaria). Mol. Ecol. 11: 2475-(12), 2475-2487. Stiasny, G., 1939. Gorgonaria von Kap Blanco, Senegal und Rio d’Ouro (Aus dem Zoologischen Museum, Amsterdam). Revue Zool. Bot. Afr. 32, 285-322. Stiasny, G., 1937. Gorgonaria von Cap Blanco, gesammelt durch Dr. Theodore Monod. Temmincka 11, 297-315. Strimmer, K., von Haeseler, A., 1996. Quartet puzzling: a quartet maximum likelihood method for reconstructing tree topologies. Mol. Biol. Evol. 13, 964-969.

54 Swain, T. D. and D. J. Taylor. 2003. Structural rRNA characters support monophyly of raptorial limbs and paraphyly of limb specialization in water fleas. Proc. Roy. Soc. London B 270: 887-896. Swofford, D.L., 2002. PAUP*. Phylogenetic Analysis Using Parsimony (*and other methods). Version 4.0b10. Sunderland: Sinauer Associates. Tixier-Durivault, A., d'Hondt, M.J., 1974. Les Octocoralliaires de la campagne Biacores. Bull. Mus. Nat. Hist. Nat. Zool. 174, 1361-1433. Verrill, A.E., 1868. Notes on Radiata in the Museum of Yale College, Number 6: Review of the corals and polyps of the West Coast of America. Trans. Connecticut Acad. Arts Sci. 1, 377-422. Vollmer S. V., S. R. Palumbi., 2004. Testing the utility of internally transcribed spacer sequences in coral phylogenetics. Mol. Ecol. 13, 2763–2772. Williams, G.C., Lindo, K.G., 1997. A review of the Octocorallian genus Leptogorgia (Anthozoa: Gorgonidae) in the Indian Ocean and subantarctic, with description of a new species and comparisons with related taxa. Proc. California Acad. Sci. 49, 499- 521. Williams, G.C., Vennam, J.S., 2001. A revision of the Indo-West Pacific taxa of the gorgonian genus Pseudopterogorgia (Octocorallia: Gorgoniidae), with the description of a new species from western India. Bull. Biol. Soc. Washington 10, 71-95. Wirshing, H.H., Messing, C.G., Douady, C.J., Reed, J., Stanhope, M.J., Shivji, M.S., 2005. M olecular evidence for multiple lineages in the gorgonian family Plexauridae (Anthozoa: Octocorallia). M ar. Biol. 147, 497-508.

55 CHAPT ER 3 MOLECULAR SYSTEMATICS OF THE GORGONIAN GENUS EUNICEA (PLEXAURIDAE: OCTOCORALLIA: CNIDARIA) USING THE PREDICTED RNA S ECONDARY S TRUCTURE OF THE INTERNAL T RANS CRIBED S PACER 2 (ITS2)3

3.1. ABSTRACT In spite of the amount of new molecular studies on octocorals, there is no consensus for a marker with enough phylogenetic resolution to solve intrageneric or closely related species relationships. Moreover, intrageneric morphological information by itself are not always accurate for producing phylogenies since intra-species comparisons can present greater differences than intra-generic ones. For this reason the use of independent characters is necessary. Here we present three separate topologies of phylogenetic relationship of nine species of the Caribbean gorgonian genus Eunicea with two sister taxa as outgroups (genus Plexaura and Pseudoplexaura), based on molecular morphometrics of RNA predicted secondary structures (ITS2) , primary sequence analyses., and a Bayesian combined data analysis. Twelve secondary structures with the 6 helicoidal ring model, also present in other groups of corals and eukaryotes, were reconstructed and provided the necessary information for phylogenetic reconstruction. This study presents Plexaura flexuosa as a member of Eunicea complementing previous analysis with mitochondrial data, external morphology and sclerite structures. We suggest the new combination Eunicea flexuosa, based on this and previous studies, which was best supported by the combined data analysis. Eunicea mammosa and Eunicea laxispica are presented in the same clade with the subgenus Euniceopsis, different from what is found with primary alignment results. ITS2 is a reliable marker for intrageneric studies in gorgonian octocorals due to the amount of changes that produce phylogenetic signal. In addition ITS2 secondary

3 Grajales, A., C. Aguilar, J. A. Sánchez. (submitted). Molecular systematics of the gorgonian genus Eunicea (Plexauridae: Octocorallia: Cnidaria) using the predicted RNA secondary structure of the Internal T ranscribed Spacer 2 (IT S2). Hydrobiologia

56 structure molecular morphometrics does not rely on INDELS whereas providing more informative characters. Keywords: Eunicea, gorgonian coral, Octocorallia, ITS2, RNA secondary structure, molecular morphometrics.

3.2. INTRODUCTION

In spite of the progress to know the Octocorallia phylogeny using mitochondrial gene sequences (e.g., mtDNA: McFadden et al., in press), there is no consensus on a sequence region with enough resolution for closely-related species level relationships. The most variable regions know so far for octocorals are the ITSs, which have provided good resolution among species as well as to point out hybridization (McFadden, 2001), that would otherwise be very difficult to distinguish at morphological level due to excessive homoplasy of morphological characters (Sánchez et al., 2003a; 2003b). However, this sequence shows like in other groups, saturation and in some cases high number of INDELS, which can produce misleading results (see Chapter I). The analysis of the predicted secondary structure of some genes such as 16S (Sanchez et al., 2003b) and ITS2 (Chen et al., 2004) have been taken as an alternative to test phylogenetic hypotheses. In this study, we approached predicted RNA secondary structures of ITS2 to resolve closely- related relationships witing the gorgonian genus Eunicea (candelabrum corals). The ITS2 (Internal Transcribed Spacers2), found in the tandem arrays of the nuclear ribosomal RNA have not been considered useful for molecular systematics in some invertebrate groups mostly due to excessive INDELS and/or intragenomic variation (e.g., Frabry et al., 1999; Harris & Crandall, 2000). Nonetheless, they seem to be a promissory nuclear region for lower eukaryotes, particularly at the intra-familial level and down to closely related species, where other known regions are nearly invariable (e.g., Knowlton, 2000; Berntson et al., 2001; LaJeunesse 2001, Shearer et al., 2002). In scleractinian corals, excepting the species of Acropora, ITS sequences have reliable phylogenetic information especially if the RNA predicted secondary structure is compared (Chen et al., 2004). These

57 sequences have shown considerable variation within octocoral genera providing information for classifying a recently described species Alaskagorgia aleutina and placed it within the Plexauridae; even thought with its morphological characteristics it was uncertain and the results were also consistent with mitochondrial DNA analysis (Sánchez and Cairns, 2004; see Chapter I). In addition variation within the Gorgoniidae family was also found, in which secondary structures of the ITS2 retained and classified different gorgonian genera (see Chapter II). Perhaps they are among the most variable genomic regions in these organisms where mitochondrial DNA and other sequences are very conserved (e.g., Berntson et al., 2001; Shearer et al., 2002; France and Hoover, 2002). For this reason, ITS aligning always produce multiple and variable INDELS (insertions-deletions) depending on the different gap opening and/or extension penalties (e.g., ClustalW), which make somewhat unreliable the phylogenetic reconstruction. The known ITS2 RNA predicted secondary structure is a model that has a common core in many eukaryotes (Schultz et al. 2005). It is a large-scale marker that is not limited to a specific taxonomic level and has even been used in mega-systematics analysis (Schultz et al. 2005). Other advantages of this approach includes the possibility of obtaining parsimony informative characters the are not found in the primary sequence (Billoud et al., 2003; Swain & Taylor, 2003); and also that it is possible to correct primary alignments based on the information of the secondary structure (e.g., Sánchez et al., 2003a). For the Eunicea genus no phylogeny based on secondary structure has been done, this could be important to support the subgenus Euniceopsis that have been classified as being part of Eunicea with sclerites (Verill 1907) and other morphological characteristics (Bayer, 1961). One recent study includes a phylogeny based on combined morphological characters with ITS2 primary sequences alignment; though this study was made with less sequences (Sánchez, submitted). In this study, we present a phylogenetic reconstruction of the Eunicea genus, based on the ITS2 sequence information. In order to contrast the possible emerging differences, we present a phylogeny based on the predicted secondary structure of the ITS2 sequences (molecular morphometrics), as well as a primary alignment

58 phylogeny of the, and a Bayesian dataset analysis which a combined both morphological and molecular data in order to make partitions for each helix.

3.3. MATERIALS AND METHODS

3.3.1. ITS2 Sequences Sequences were obtained using primers designed by Sánchez (submitted), which targets the region of the 5’ end of the 5.8s ribosomal genes ribosomal gene and the 3’ end of the 28S ribosomal gene containing the complete ITS2 (5.8S-436: 5’- AGCATGTCTGTCTGAGTGTTGG-3’ and 28S-663: 5’- GGGTAATCTTGCCTGATCTGAG-3’, numbers respect to the sequence of Alcyonium digitatum, Genbank Acc. No. AF262347: McFadden et al., 2001). DNA extractions were made according to Sánchez et al. (2003b) from alcohol preserved material (Ethanol 95%) and the DNeasy kit (Qiagen) applied to the extractions with high concentration of dark pigments that usually did not amplify properly. Template DNA for sequencing was obtained from the combination of two PCR reaction tubes containing 56 µl total (volume completed with ddH2O) with 1 µl of DNA template (1:50 dilutions of genomic DNA extractions), 2 units Taq polymerase (Promega), 3 µl of 10X Buffer (Promega), MgCl2, 0.15 µM dNTPs mix, 0.16 µM of each primer. PCR conditions had one initial period of 2.0 min at 94°C followed by 30 cycles of 30 sec at 94°C, 45 sec at 56°C, 1.0 45 sec at 72°C, and a final extension step for 5.0 min at 72°C. Gene purification was made using the Edge Biosystems kit and sequencing using BigDye 3.1 (AB 3100, capillary electrophoresis automated sequencer). Consensus sequences were obtained automatically by assembling the two complementary DNA chromatograms (Sequencher software).

3.3.2. ITS2 Predicted RNA Secondary Structures ITS2 alignment for used for predicting secondary structures by homologous structures already published, like Acropora spp. (Odorico & M iller, 1997). Moreover,

59 manual alignment was performed by visual homology for the construction of a Dedicated Comparative Sequence Editor (DCSE) format in order to perform different phylogenetic analysis. DCSE format uses square brackets ([ or ]) to delimitate each helix, braces ({ or }) for bulges and loops, and hyphens for gaps in the alignment, terminal loops appear separated by inverse brackets (] and [). The acquired structures with restrictions and constrains were submitted in MFOLD (http://www.bioinfo.rpi.edu/applications/mfold/old/rna/). RNA was folded at a fixed temperature of 37°C, and the structure chosen from different output files was the one with the highest negative free energy.

3.3.3. ITS2 Sequence Analysis Sixteen sequences of the ITS 2 region of eight Eunicea species (Eunicea colombiana , E.fusca, E. mammosa, E. tourneforti, E. laxispica, E. laciniata, E.pallida, .E tayrona and E.flexuosa) and three corresponding to Pseudoplexaura crucis, Plexaura homomalla and Plexaura kuna were aligned in BioEdit (Hall 1999) using Clustal W multiple alignment (Higgins et al. 1996) with the default gap and extension penalties used by this program; this for performing different sequence analysis.

3.3.4. Phylogenetic Analysis M aximum parsimony The obtained secondary structures were used to construct a matrix for a cladistic analysis by the means of molecular morphometrics based on geometrical features and base number (Billoud et al., 2000). This means, secondary structures helixes were numbered and treated like a character that will vary depending on their base number. Bulges and internal loops were given the number of the helix followed by a letter, ending helixes loops and separation segments where designated with the “i” letter. Loops complement, helix complement, and separation segments were indicated with an apostrophe (e.g. 4a loop with its complement 4a’). Nucleotides were count for each character and arranged in discrete

60 character states ranging from 0 to 5 (being 0 the character with less nucleotide number) as seen in table 1. A semi exhaustive search using maximum parsimony and the branch and bound algorithm were carried out in PAUP* (Swofford, 2002). Branch support was given using a 50% majority-rule consensus in PAUP* (Hillis and Bull, 1993).

Maximum likelihood and Bayesian inference Primary ClustalW sequences alignment was submitted to the best-fit model from ModelTest (Posada and Crandall, 1998) obtained by the Akaike Information Criterion (AIC), this for performing a search with the branch and bound algorithm in PAUP* (Swofford, 2002). INDELS were treated as missing data. Bayesian inference of phylogeny was done using MrBayes (Huelsenbeck and Ronquist, 2001), Bayesian-estimated likelihood (settings according to M rModeltest), 10 million M onte Carlo markovian chain generations (Bayesian-Monte Carlo simulation by MrBayes sampling every 100 simulation, burn-in 10000).

Bayesian combined dataset analysis A third phylogenetic analysis was carried using the Bayesian approach with combined datasets (Nylander et al., 2004) with the software Mrbayes 3.1 (Huelsenbeck and Ronquist, 2001). In this approach, each data partition is allowed to have different evolution rates (using the option prset ratepr=variable). The molecular morphometrics data was treated as DATATYPE=standard mode and no evolutionary model was selected, but a different rate of change for each helix was settled (γ) . The model for the sequence dataset was searched using the software Mrmodeltest. Two separate chains were run with 10 million generations each, to test for the convergence of the parameter estimation form two different depart points of the search. Likelihood comparisons between the different phylogenetic hypotheses were done in PAUP * (Swofford, 2002) using the Shimodaira-Hasegawa test with 1000 replicates, and the Kishino-Hasegawa test to calculate two tail distribution.

61 3.4. RESULTS

ITS2 nuclear region varied from 252bp in Eunicea laciniata to the shortest of 185bp in Plexaura kuna, with this region twelve new predicted RNA secondary structures were constructed, all following the 6 helicoidal ring model excepting E. pallida. There were two mayor common features present in these structures: (1) 5’ G-C, R-R, G-C, G-C, bulge on helix IV and (2) 5’ GUGC, bulge, CAAGG with its complementary pair base on helix V (Fig. 3.1)

Ta bl e 3 . Species / Characters matrix from molecular morphometrics. Character numbering refers to RNA predicted secondary structure as in Fig. 3.1.

Spp. / Characters 1 1a 1i 2 2i 2´ 2i` 3 3a 3i 3´ 3á 3i´ 4 4a 4b 4i 4´ 4´b 4á 4i´ 5 5a 5b 5c 5d 5e 5f 5g 5i 5´ 5f' 5e' 5´d 5ć 5´b 5á 5i´ 6 6i 6´ 6i´ 1´ 1á Pseudoplexaura crucis 11 0333461371 21205220320 0001004110 0000122201 2 Plexaura homomalla 20 1313530540 43303211410 0000305130 0000200001 1 Plexaura kuna 20 1313430540 41313221420 0000005130 0000200001 1 Eunicea tayrona 20 1414140451 04201210162 1102101422 2220221201 1 Eunicea pallida 20 1122022420 15213422240 0211314312 2110200001 1 Eunicea colombiana 20 1414010210 34201400072 2312412422 1331122202 1 Eunicea mammosa 20 2121141651 46122501531 2220002221 2130311111 1 Eunicea laxispica 20 2112230441 24204202152 0022101312 0410113112 1 Eunicea laciniata 20 1414320132 35202400271 2012302422 0421133322 1 Eunicea tourneforti 20 1414320230 54201300182 2302215322 1431422201 1 Eunicea fusca 10 1415220530 22203102343 0202203322 0400421201 1 Eunicea flexuosa 20 1122122521 06202402140 1302101311 2000112111 1

Fi gure 3.1.- ITS2 predicted ARN secondary structures for 9 species of the genus Eunicea and their enthalpy values of structure formation according to MFOLD: (A) Eunicea flexuosa, numeration represents characters used for molecular morphometry as in Table 1, ∆G= -58.96 kcal/mole; (B) Eunicea laciniata, ∆ G= -104.17 kcal/mole; (C) E. colombiana, ∆G= -104.17 kcal/mole; (D) E.tayrona, ∆G= -71.45 kcal/mole; (E) E. laxispica, ∆G= -62.67 kcal/mole; (F) E. fusca, ∆G= -65.96 kcal/mole; (G) E tourneforti, ∆G= -69.67 kcal/mole; (H) E. mammosa, ∆G= -67.66 kcal/mole; (I) E. pallida,∆G= -69.53 kcal/mole.

63 Three different phylogenetic analyses were obtained form the ITS2 sequence alignment. The molecular morphometrics analysis produced 44 characters and 2 equally parsimonious trees (L=205). Consistency and retention indexes were about 60% (CI= 0.634; RI=0.662), and homoplasy index was of 0.366. The resolution of the topology based only on the predicted RNA secondary structure of the ITS2 region resolved most relationships among the studied species (Fig. 3.2A). Maximum likelihood analysis of the primary sequence alignment retained one tree (L= 301) with. the assumed model (TVMef+G) selected by AIC, tree searches had six substitution rates (A-C, 1.2431; A-G, 3.0064; A-T, 1.0809; C-G, 0.3376; C-T, 3.0064; and G-T, 1.0000), no assumed proportion of invariant sites, and a gamma-shape parameter (1.413). Supports from 1000 bootstrap replicates were obtained (Fig. 3.2B).Bayesian analysis of the same alignment retained the same topology and gave support to the same branches. The best resolved phylogenetic hypothesis with secondary structure data was the one done with Bayesian inference in which helixes were to evolve independently at different rates in a mixed matrix (fig.3.2C). The most parsimonious molecular morphometrics cladogram (fig. 3.2A) showed similar topology but did not had as high support values for each node as the Bayesian approach. Moreover, the group containing E. pallida, E. mammosa, E. laxispica and E. flexuosa (new combination from Plexaura flexuosa) was highly supported by Bayesian probabilities which gave support to the monophyly of the group.

Fi gu re 3.2.- A. Phylogenetic hypotheses of molecular morphometrics data (table 1), 50% Majority rule consensus of 2 equally parsimonious trees of 9 species of the genus Eunicea. Numbers represent bootstrap values of 10,000 replicates. B. Maximum Likelihood phylogram of the ITS2 primary alignment with the assumed model (TVMef+G) selected by AIC, tree searches had six subst it ution rat es (A-C, 1.2431; A-G, 3.0064; A-T, 1.0809; C-G, 0.3376; C-T , 3.0064; and G-T , 1.0000), an assumed proport ion of invariant sit es (= 0), and a gamma-shape parameter (1.4130). Bayesian probabilities and bootstrap values of 1,000 replicates are separated by a slash. C. Consensus tree of the Bayesian analysis with 10, 000,000 generations of the combined data of predicted secondary structure and secondary sequence alignment of the ITS 2 region.

An important finding about the different phylogenetic hypotheses obtained was that E. mammosa, a member from the Eunicea subgenus, was placed in a different position grouping. The analysis of the combined dataset as well as the molecular morphometrics

65 placed E. mammosa within the Euniceopsis subgenus (fig. 3.2 A.C), whereas the primary alignment placed it as a sister species of Plexaura kuna (Fig. 3.2B). E. flexuosa that was as a sister taxa of E. mammosa in molecular morphometrics (fig. 3.2A) ocurred within the Eunicea genus in the three topologies (Fig. 3.2). E. tayrona was presented as an unresolved branch from the node with E. laciniata and E. fusca (Fig. 3.2C). The ITS2 predicted secondary structure models showed that E. mammosa, E. pallida, E. laxispica and E. flexuosa exhibited a long helix IV (Fig. 3.1A, E, H, I) that could be the feature that drove them in the same clade. Different from the outgroup E. homomalla that lacked one helix (figure not shown, see appendix). Moreover, the test for comparing the different phylogenetic hypotheses gave significant differences between pair comparisons (all the p values were less than 0.05).

3.5. DISCUSSION

Eunicea species morphological identification is based mostly on microscopic characters (Sanchez, submitted). For this reason, molecular identification has become an important alternative for this genus and octocorals in general. ITS2 has proven to be a good molecular marker that gave resolution to octocorals species at different taxonomic levels (e.g., see Chapter I). Particularly, for the Eunicea genus the ITS2 marker gave a new look at its phylogeny. ITS2 secondary structures presented similar models that grouped conserved species features in the clade (long helix IV for one group), though it is important to have in mind that this molecular structures can also produce homoplasies due to selective pressures acting on particular helix features, which is evident in the conservation of the 6-helicoidal ring core present in most eukaryotes. ITS2 structures are important for the proper pre- rRNA maturation; any change in their helix conformation can affect the process of 25S production (Cote et al. 2001). In this study, helix partition was done for each helix as if

66 they had evolved under different rates and substitution bias because not all of them have the same role in this dynamic conformation (Cote et al. 2002). This approach resulted in a well-supported Bayesian inference phylogram (Fig. 3.2C). For the phylogenetic hypotheses obtained in this study the congruence test between morphological and molecular data showed significant differences (P < 0,05), which corresponded to different species placement in each tree. The major difference between the molecular morphometrics and sequence alignment phylogenies was the placement of E. mammosa in and out the Euniceopsis subgenus, which did not give any precise information about this species status. The different species combination suggested there was not clear differentiation between these two subgenera using ITS2, differing to what has been described with morphologic characters (Bayer, 1961). The combined molecular-morphological Bayesian approach data matrix has not been widely used, mainly because the methods of combining datasets were until recently only parsimony-based analysis, which has been proved to be insufficient when the combined data present different evolution rates (Pollock and Bruno, 2000). This type of analysis combining molecular data with “typical” morphological characters has presented some failures to asses a proper character evolution model (Lewis , 2001). Here we combined the corrected primary alignment and the molecular morphometrics data. Traditionally, some problems arise from the assumptions of character state change by assuming that each character has exactly the same state in a particular time in opposition of punctuated equilibria. In the case of molecular morphometrics the possibility of change is based in the number of nucleotides present in some helix, which can be readily identified. The other issue, corresponding to the comparison between character states were resolved using the correction only from the parsimonious informative sites (Lewis, 2001) Combined data phylogram resolved at some level the incongruence between both phylogenies (e.g., Fig 3.2A-B), and gave high support to all nodes when compared to any of the other topologies. The position of E. mammosa in fig. 3.2C confirmed the monophyly of the group whereas the sister relationships of E. mammosa and the exact relations for the

67 two subgenera is still unclear. For this, it is so say that a more informative matrix of ITS2 secondary structure analysis is needed for a better resolved and supported Eunicea phylogeny. Nevertheless, due to the polytomies found in some branches (Fig. 3.2C), and for the inconsistencies in the placement of a few species, this study needs to be completed with more Eunicea representatives for a better understanding of this group systematics. The molecular study of this genus, the most specious among shallow-water Caribbean gorgonian corals, is just beginning. Consequently, ITS2 secondary structure analysis could be a great tool for telling apart new species and to complete Eunicea systematics. This, taking into account that morphologic phylogenetic reconstruction has difficulties (e.g., Sánchez, submitted) and ITS2 secondary structure contained more information than the usual primary sequences alignment.

3.6. ACKNOWLEGMENTS pecial thanks to Universidad de los Andes (Faculty of Sciences and Department of Biological Sciences) for resources, Howard Lasker (SUNY, University at Buffalo) for field collections at the Bahamas, Klaus Ruetzler (CCRE, Smithsonian) for filed collections at Carrie Bow Cay, Belize, and a postdoctoral fellowship (J.A. Sánchez, 2002-2003) from the Smithsonian Institution-NMNH (Liz Zimmer, Ken Schallop, Lee Weight and Stephen D. Cairns) for initial molecular analyses. Colleagues and students from BIOMMAR provided helpful discussions for the paper.

68 REFERENCES

Bayer, F. M. 1961. The shallow water Octocorallia of the West Indian region. Stud. Fauna Curaçao. 12, 1-373. Berntson, E. A., F. M. Bayer, A. G. McArthur, and S. G. France. 2001. Phylogenetic relationships within the Octocorallia (Cnidaria : Anthozoa) based on nuclear 18S rRNA sequences.Mar. Biol. 138: 235-246. Billoud, B., Guerrucci, M. A., M. Masselot., Deutsch, J. S., 2000. Cirripede phylogeny using a novel approach: molecular morphometrics. Mol. Biol. Evol. 17, 1435–1445. Chen, C. A., C. C. Chang, N. V. Wei, C. H. Chen, Y. T. Lein, H. E. Lin, C. F. Dai, and C. C. Wallace. 2004. Secondary structure and phylogenetic utility of the ribosomal internal transcribed spacer 2 (ITS2) in Scleractinian Corals. Zool. Stud. 43: 759-771. Cote, C. A., C. L. Greer, and B. A. Peculis. 2002. Dynamic conformational model for the role of ITS2 in pre-rRNA processing in yeast. RNA 8: 786-797. Cote, C. A., B. A. Peculis. 2001. Role of the ITS2-proximal ítem and evidence for indirect recognition of processing sites in pre-rRNA processing in yeast. Nucleic Acids Research.29: 2106-2116. Fabry S., A. Kohler, and A. W. Coleman. 1999. Intraspecies analysis: Comparison of ITS sequence data and gene intron sequence data with breeding data for a worldwide collection of Gonium pectorale. J. Mol. Evol. 48: 94-101. France, S. C., and L. L. Hoover. 2002. DNA sequences of the mitochondrial COI gene have low levels of divergence among deep-sea octocorals (Cnidaria : Anthozoa). Hydrobiologia. 471: 149-155. Hall, T.A. 1999. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucl. Acids. Symp. Ser. 41:95-98. Harris, D. J, and K. A. Crandall. 2000. Intragenomic variation within ITS1 and ITS2 of freshwater crayfishes (Decapoda : Cambaridae): Implications for phylogenetic and microsatellite studies.Mol. Biol. Evol. 17: 284-291.

69 Knowlton, N., 2000. Molecular genetic analyses of species boundaries in the sea. Hydrobiologia 420, 73-90. LaJeunesse, T. C. 2001. Investigating the biodiversity, ecology, and phylogeny of endosymbiotic dinoflagellates in the genus Symbiodinium using the its region: in search of a “species” level marker. J. Phycol. 37: 866–880.

Lewis, P.O. (2001). A likelihood approach to estimating phylogeny from discrete morphological character data. Syst. Biol. 50:. 913–925.

McFadden, C. S., R. Donahue, B. K. Hadland, and R. Weston. 2001. A molecular phylogenetic analysis of reproductive trait evolution in the soft coral genus Alcyonium. Evolution. 55: 54-67. Nylander, Johan A. A.,1 Ronquist Fredrik, Huelsenbeck Joh P, Nieves-Aldrey Jose Luis 2004 Bayesian Phylogenetic Analysis of Combined Data Syst. Biol. 53(1):47–67 Odorico, D.M ., Miller, D.J., 1997. Internal and external relationships of the Cnidaria: implications of primary and predicted secondary structure of the 50-end of the 23S-like rDNA. Proc. R. Soc. Lond. B 264, 77–82. Pollock, D.D., and Bruno, W.J. (2000). Assessing an unknown evolutionary process: Effect of increasing site-specific knowledge through taxon addition. Mol. Biol. Evol. 17: 1854–1858. Sánchez, J.A. (submitted) Morphological and molecular systematics of the Caribbean candelabrum octocorals (Eunicea Lamouroux, 1816; Cnidaria; Octocorallia; Plexauridae) with description of new species and aspects of their natural history. Invertebrate Systematics. Sánchez J.A., Cairns, S.D., 2004. An unusual new gorgonian coral from the Aleutian Islands, North Pacific, Alaska. Zool. Med. 78, 265-274. Sánchez J.A., Lasker, H.R., Taylor, D.J., 2003a. Phylogenetic analyses among octocorals (Cnidaria) according to mitochondrial and nuclear DNA sequences (lsu-rRNA 16S, and ssu-rRNA 18S) support two convergent clades of branching gorgonians. Mol. Phylogenet. Evol. 29, 31-42. (DOI:10.1016/S1055-7903(03)00090-3)

70 Sánchez, J.A., Mcfadden C.S., France, S.C., Lasker, H.R., 2003b. Molecular Phylogenetic analyses of shallow-water Caribbean octocorals. Marine Biology. 142, 975-987. (DOI10.1007/s00227-003-1018-7) Schultz, J., Maisel, S., Gerlach, D., Müller, T., Wolf, M. (2005). A common core of secondary structure of the internal transcribed spacer 2 (ITS2) throughout the Eukaryota. RNA 11: 361-364. Shearer T. L., M. J. H. van Oppen, S. L. Romano, and G. Wörheide. 2002. Slow mitochondrial DNA sequence evolution in the Anthozoa (Cnidaria). Mol. Ecol. 11: 2475-(12), 2475-2487. Swain, T. D. and D. J. Taylor. 2003. Structural rRNA characters support monophyly of raptorial limbs and paraphyly of limb specialization in water fleas. Proc. Roy. Soc. London B 270: 887-896. Swofford, D.L. 2002. PAUP*. Phylogenetic Analysis Using Parsimony (* and other methods). Version 4.0b10. Sinauer Associates, Sunderland, MA. Verril, A.E. (1907) ‘The Bermuda Islands. Part 5. Characteristics life of Bermuda corals reefs.’ Transactions of the Connecticut Academy of Arts and Science. 12, 204-348, 413-418.

71 APPENDIX

1. DCSE Alignment showing secondary structures of fourteen species of octocorals (Helix numbering as in Fig. 1.1; Character coding in Table 1.1).

....|....| ....|....| ....|....| ....|....| ....|....| ....|....| 5 15 25 35 45 55 Helix numbering -----1------1a------1i---2------2i---2 '--2a'---- Calyptrophora japonica [UGUCUGUCU G{A}GUG]------[UUU G]------GU UCUUCUA[CA GA]------Sclerobelemon theseus [UGUCUGUCU {GA}GUGUUU G]---G[UUC ]------UCGAAC[GA A]------Alcyonium digitatum [UGUCUGUCU ]------GAGUG[UUA ]------GGU[UA A]------Lobophytum sp. [UGUCUGUCU ]------[GA{ GUG}UUA]------GGU[UA A{AAGA}UC] Gorgonia ventalina [UGUCUGUCU GAG]------UG[UUU G]------GUUUA[CG AA]------Pacifigorgia stenobrochis [UGUCUGUCU GAG]------UG[UUU G]------GUUAA[CA AA]------Leptogorgia virgulata [UGUCUGUCU GAG]------UG[UUU G]------GUUUA[CA AA]------Pterogorgia citrina [UGUCUGUCU GAG]------[UGU UUG]------GUUAU[CA AACA]----- Muricea muricata [UGUCUGUCU ]------GAGUG[UUA ]------GGU[UA A]------Eunicea laciniata [UGUCUG-UC UGAG{U}GU] ------[UUG G]------UCAAA[CC GA]------Plexaura kuna [UGUCUGUCU ]------GAGUG[UU] ------AGGUC[AA ]------Plexaurella nutans [UGUCUGUCU GAG]------U[GUU UG]------GUUAA[CA AAC]------Muriceopsis flavida [UGUCUGUCU GAG]------[UGU UUGG]------UUA[CC AAACA]---- ALASKAGORGIA ALEUTIANA [UGUCUGUCU ]------GAGU[GUU A]------G GUUAAGA[UG AU]------

....|....| ....|....| ....|....| ....|....| ....|....| ....|....| 65 75 85 95 105 115 Helix numbering ----2i'--- 3--3a------3b------3 c-----3d------Calyptrophora japonica -----AA[CG AGCG]------Sclerobelemon theseus ------A[UC GCGUCGC]------UU Alcyonium digitatum ----GA-[UG UU{AA}UGUU UGU{AAG}CG ]------Lobophytum sp. ------[GA ACGGAAC{A} AC]------Gorgonia ventalina -----AC[GA {GC}UCGCG] ------CGUGCG Pacifigorgia stenobrochis ---AUAU[GA UCG{U}GUGC G]------Leptogorgia virgulata ---ACAU[GA UUGUGUG]------Pterogorgia citrina -----AU[GA ACG-GCA]------Muricea muricata GAACAUG[UU G{U}UG{UU} GUGUUA{GA} GGAUGCA{U} G{UG}C{AA} GGUCA]-AAG Eunicea laciniata -CAAA--[UG CGCGGACGC] ------Plexaura kuna --GAGAA[GC {A}GG{GA}G CGUGCG{AA} CAAGG]------ACCAG Plexaurella nutans ------U[GA ACGCACG]------Muriceopsis flavida -----GU[GA ACGG{C}GGC ]------ALASKAGORGIA ALEUTIANA --GAGAA[AC G{ACA}ACGG AUGC{U}GCG CU{U}GC{AA }GGGCAA{A} GUUU]-----

....|....| ....|....| ....|....| ....|....| ....|....| ....|....| 125 135 145 155 165 175 Helix numbering --3i------3'--3d'-- -3c'-----3 b'------3a'------Calyptrophora japonica ----UGCG[C GCUCG]------Sclerobelemon theseus UUGCGAAA[G CGGCGCGG]------Alcyonium digitatum ----UAGA[C G{AC}GCGAA CGAACA]------Lobophytum sp. AUAUCCAC[G U{AC}GUUC{ A}CGUUU]------CA[GGC]U Gorgonia ventalina UUUGCGCU[C GCGAUC]------Pacifigorgia stenobrochis ------AC[U GUGC{U}CGA UC]------Leptogorgia virgulata --UAUCAU[C ACGCGAUC]------Pterogorgia citrina ----CACG[U GC{U}CGUUC ]------Muricea muricata GAUACUUU[U GGCCG{UG}C {C}UGCAUCC {G}UGACAU{ U}UA{UAU}C AA]------Eunicea laciniata -----CGU[G CGUUCGUGCG ]------Plexaura kuna CUUUACCG[C CUUGCGUGC{ AC}GU{CGG} CC{A}GC]------Plexaurella nutans ---UUUUU[C G{A}UGCGUU C]------Muriceopsis flavida ----GUAU[G CU{C}UCGUU C]------ALASKAGORGIA ALEUTIANA ---UAUUU[A GGC{C}UUGU UC{AC}GCAG CGUGCGUU{G AA}CGU{GC} UGU]------

....|....| ....|....| ....|....| ....|....| ....|....| ....|....| 185 195 205 215 225 235

72 Helix numbering ------3i'---4 -----4a------4b------4c------Calyptrophora japonica ------GCG[UUG GGA{CGU}CA CGGAU{CGC} GCG]------Sclerobelemon theseus ------CG[UUG GGA{CGU}CG UGGCG]------Alcyonium digitatum ------A[GCG AUG{CUAU}C AAGG{ACAC} AAA]------Lobophytum sp. GCCGUC[GCC ]---AA[UGU C{AUCUAC}G CUGU{GCC}U GAGG{U}CAA GGU{GUU}AA Gorgonia ventalina ------GAC[UGA GG{CUGU}CG CA----CGC] ------Pacifigorgia stenobrochis ------GA[UUG AGG{CUGU}C GC{GU}CGUU G]------Leptogorgia virgulata ------GA[UUG AGG{CUGU}C GCGU{C}GUU G-UUU]------Pterogorgia citrina ------GA[UUG GGGG{U}CGC GUAG]------Muricea muricata ------U[UUG GAGGG{AUGA }AC{G}AACG {AUACCG}UG {C}GGUG]-- Eunicea laciniata ------ACU[GGG GC{GUC}GCG CCGC]------Plexaura kuna ------A[UG{ GA}AU{UC}G AG]------Plexaurella nutans ------GA[UUG AGG{CGU}CG CGU-{U}CGC ]------Muriceopsis flavida ------GA[UUG GGGG{U}CG{ G}GUAG{CGU }UGCG{C}GU ]------ALASKAGORGIA ALEUTIANA ------ACUUG[GAG GG{A}CGCG{ AAA}GAA]------

....|....| ....|....| ....|....| ....|....| ....|....| ....|....| 245 255 265 275 285 295 Helix numbering ------4i----- 4'---4c'------4b'------4a'----- Calyptrophora japonica ------UUUU[UG C{AAAA}AUC CGUG{UC}UC CCGA]------Sclerobelemon theseus ------CGUA[CG CCGCG{UC}U CC{G}UAA]------Alcyonium digitatum ------UGUUAC[UU U{AGG}CCUU G{UUCACA}C GUUGU]------Lobophytum sp. AUGA]------UAUC[UC GUUU{UAA}G CCUUGCUUCG {UACA}ACAG U{UUC}GACA Gorgonia ventalina ------CCCGC[GU GUGCG{UC}C CUCG]------Pacifigorgia stenobrochis -----UCUCC GUCGGAG[CA ACG{U}GCG{ UC}CCUCGA] ------Leptogorgia virgulata ------CCAGC[AA A{C}UAAC-G UGCG{UC}CC UCGA]------

Pterogorgia citrina ------CGU UGCAUUA[CU ACGCG{U}CU CUCGA]------Muricea muricata ------CAA[CG CC{U}CA{AA }CGUUGU{GU UG}CCU{A}U UU{UUA}AA] Eunicea laciniata ------CGCCGCC[GC GGCGCGUCCC ]------Plexaura kuna -----GGACG CGUAUAA[CU U{AU}AU{A} UA]------Plexaurella nutans ------A ACAUAAC[GU G{C}ACGCG{ UC}CCUCGA] ------Muriceopsis flavida ------GUAC[GC {A}CGCA{UU UA}CUAC{G} CG{U}CUCUC GA]------ALASKAGORGIA ALEUTIANA ------UGUUGU[UU C{AAA}CGCG {A}C{U}CCU C]------

....|....| ....|....| ....|....| ....|....| ....|....| ....|....| 305 315 325 335 345 355 Helix numbering ------4i'- --5---5a------5b------5 c------5d------Calyptrophora japonica ------AGUU C[AGCG{AGA AU}CGUCGAG {U}GUC{AA} UAGCG{A}UU CGCGAAG]-- Sclerobelemon theseus ------AUG C[AGAGGG{A U}CG{U}GCG AGCGUCGCCG UC{UAC}CCG {UACG}CGGU Alcyonium digitatum ------C[GAACGUGC GUC{GA}ACG {CAC}UGC{A C}UUCGAAGG {A}CGCG{A} Lobophytum sp. ]------AC A[GC{GAA}C GC{GACA}GA {GCUAUCGUA }CGUCGC{U} GU{AUUUC}G Gorgonia ventalina --AAGUGCAG C[ACG{UU}C GAC{GCCC}G UGUCGUGC{G AA}CAAGG{C C}CAAAG]-- Pacifigorgia stenobrochis ---AGUGCAG C[ACG{UU}C GAC{GCUC}G UGUCGUGC{G AA}CAAGG{C C}CAAAG]-- Leptogorgia virgulata ---AGUGCAG C[ACG{UU}C GAC{GCUAU} UGUCGUGC{G AA}CAAGG{C C}CAAAG]-- Pterogorgia citrina ----AAUGCA GUAC[GUUCC GCU{UUC}UG -----CAC{G AA}CAAGGC{ CU}AA]---- Muricea muricata -CACGCACUC A[GG{CAGAC }AUGC]------Eunicea laciniata ----UCGAAG U[GC{CA}CC GUG{A}[GC] ACAC[GC]CG CCGUU{CAUC A}GUGC{C}G Plexaura kuna ------G U[GCG{AC}G CCU{CAA}UC GUCGA{C}GA CGAAAU]------Plexaurella nutans ------A[GUAC{A}G CAC{GUUC}G ACG{CUU}GU GCGC{GA}AC AAGG{CA}UA Muriceopsis flavida ------AUUG C[AGAACGUU CCGCUCU{CG }GCAC---{G A}GCAAGGC{ CU}AA]---- ALASKAGORGIA ALEUTIANA ------AAU C[GAAC-GCG CG]------

....|....| ....|....| ....|....| ....|....| ....|....| ....|....| 365 375 385 395 405 415 Helix numbering ------5e------5i------5'--5e '------5d Calyptrophora japonica ------CCAA AAAUCAUCAC GA[CUUCGCG AACGCUG{CC Sclerobelemon theseus AG{A}GGGCG A]------CCUA UA[UCGC{AA AUCACGAAGC Alcyonium digitatum CG{A}GC]------GA AC[GCCGCG{ A}CGCCU{CA Lobophytum sp. AGAGA{CGC} GA]------UUUCUU UU[UC{CU}U CUCUCAU{C} Gorgonia ventalina ------AA[CUUUG{U }CCUUG{AGA Pacifigorgia stenobrochis ------AA[CUUUG{U }CCUUG{CGA Leptogorgia virgulata ------AA[CUUUG{C }CCUUG{CGA Pterogorgia citrina ------AG UG[UU{UU}G {C}CCUUG{C Muricea muricata ------UCU CG[GCAU{AA }CC]------

73 Eunicea laciniata CCGUGC{GAA }CAAGG{CCU A}AAG]------UGU UG[CUU{CGA }CCUUG{CCA Plexaura kuna ------UC AAUUUCAUCA UC[A{U}UUU CGUU{C}UUG Plexaurella nutans AA]------GA CA[----UUU A---{C}CCU Muriceopsis flavida ------AG UU[UU{UU}G -CCCUUGU{C ALASKAGORGIA ALEUTIANA ------CCGA AA[CGCGCGU UC]------

....|....| ....|....| ....|....| ....|....| ....|....| ....|....| 425 435 445 455 465 475 Helix numbering '------5c'------5b'------5a'------Calyptrophora japonica }GACCUCGAC GC{C}GCU]------Sclerobelemon theseus CAAAAU}UC{ A}UUGCUG{A CUU}CGG{U} GAUGG{U}CG {CGUU}AUGC UCGCCGCCCU Alcyonium digitatum A}---UCGAA {C}GCG{CUU C}CGU{CGA} GAC{GAGA}G CGCGUUC]------Lobophytum sp. GCGACG{CCC CAUUCAA}UC {GA}GCG{GC A}GC]------Gorgonia ventalina }GCGCGACGU {UC}GUCG{U C}CGU]------Pacifigorgia stenobrochis }GCACGACGU {CU}GUCG{U U}CGU]------Leptogorgia virgulata }GCACGACG{ UCU}GUCG-{ UU}CGU]------Pterogorgia citrina CA}GUGCG{U U}AGUGGAGC ]------Muricea muricata ------Eunicea laciniata }GCACGGCGC GC{CGCC}GA CGGCG{CA}C GCGG{A}GC] ------Plexaura kuna AC{CAACACU CA}GA{C}-- AGGC--{A}U GC]------Plexaurella nutans UGU{AA}GUG CA-UCGUCGU GCGU---AC] ------Muriceopsis flavida A}GUGC{AGU AGU}AGAGC{ UU}G{C}GAG CGU{C}UCU] ------ALASKAGORGIA ALEUTIANA ------

....|....| ....|....| ....|....| ....|....| ....|....| ....|....| 485 495 505 515 525 535 Helix numbering ------5 i'---6------6i----6'- Calyptrophora japonica ------C[CGUGU ACACAC]------UAAGA[GUGU Sclerobelemon theseus UU]------Alcyonium digitatum ------A CGUU[UUG]------UUUAC[CAA] Lobophytum sp. ------[GGGCG CAUG{CAC}G CC]------GCUC[GGC{ Gorgonia ventalina --CGAUGCCA ACAACCCCUA CAGG[UUA]------C AUCAU[UAA] Pacifigorgia stenobrochis ------CCCUUU UCUA[UUA]------U[UAA] Leptogorgia virgulata ------CCCU[UUA]------UCA UCUAU[UAA] Pterogorgia citrina ------UCUC[UUA]------CC AUUCU[UAA] Muricea muricata ------Eunicea laciniata ------UUC[GAAAU CG]------U UUCAU[CGAU Plexaura kuna ------UCCC[GGU]------AUA[ACC] Plexaurella nutans ------CUUU[UU]------CA UUUUU[AA]- Muriceopsis flavida ------U[UUAA] ------CCAUUUU CAUUC[UUAA ALASKAGORGIA ALEUTIANA ------C[GUU]------UUGUU[AAC]

....|....| ....|....| ....|....| ....|....| ....|....| .. 545 555 565 575 585 Helix numbering ------6 i'------1'------Calyptrophora japonica GUACACG]AU U[UU]GCAC[ AA]UCCUCUC A[CAC{A}UA GA{U}CAGGC A] Sclerobelemon theseus ------UU U[CAAACAUA GA{U}CAGGC A] Alcyonium digitatum ------[A GA{U}CAGGC A] Lobophytum sp. UCAAUU}CGU GUG{U}ACC] ------AA[A GA{U}CAGGC A] Gorgonia ventalina ------C[CUCA GA{U}CAGGC A] Pacifigorgia stenobrochis ------C[CUCA GA{U}CAGGC A] Leptogorgia virgulata ------C[CUCA GA{U}CAGGC A] Pterogorgia citrina ------C[CUCA GA{U}CAGGC A] Muricea muricata ------GA[A GA{U}CAGGC A] Eunicea laciniata {CG}UUC]------UUA[ACCUCA GA{U}CAGGC A] Plexaura kuna ------AA[A GA{U}CAGGC A] Plexaurella nutans ------C[CUCA GA{U}CAGGC A] Muriceopsis flavida ]------C[CUCA GA{U}CAGGC A] ALASKAGORGIA ALEUTIANA ------CAA[A GA{U}CAGGC A]

74 2. DCSE Alignment showing secondary structures of twenty-four species of octocorales (Helix numbering as in Fig. 1; Character coding for molecular morphometrics in Table 2.1).

....|....| ....|....| ....|....| ....|....| ....|....| ....|....| 5 15 25 35 45 55 Characters # ------1------1i---2---- 2i------2' ------2i '----3--3a E_laciniata [UGUCUGUCU GAG{U}GU]- --[UUGG]-- UCAAA[CCGA ]------C AAA[UGC-GC G_ventalina [UGUCUGUCU GAG]-----U G[UUUG]--G UUUA--[CGA A]------AC ----[GA{GC G_mariae [UGUCUGUCU GAG]-----U G[UUUG]--G UUUA--[CGA A]------AC ----[GA{GC G_flabellum [UGUCUGUCU GAG]-----U G[UUUG]--G UUUA--[CGA A]------AC ----[GA{GC Pi_flava [UGUCUGUCU GAG]-----U G[UUUG]--G UUA---[CAA A]------A AGAU[GAACG Pi_platysoma [UGUCUGUCU GAG]-----U G[UUUG]--G UUA---[CAA A]------A AGAU[GAACG A_schoutedeni [UGUCUGUCU GAG]-----U -[GUUU]-GG ----UC-[AA AC]---CGAC ----[GAAUG A_africana [UGUCUGUCU GAG]-----U -[GUUU]-GG ----UC-[AA AC]---CGAC ----[GAAUG P_stenobrochis [UGUCUGUCU GAG]-----U G[UUUG]--G UUAA--[CAA A]------AUAU[GAUC- L_stheno [UGUCUGUCU GAG]-----U G[UUUG]--G UUUA--[CAA A]------ACAU[GAUU- L_euryale [UGUCUGUCU GAG]-----U G[UUUG]--G UUUA--[CAA A]------ACAU[GAUU- L_hebes [UGUCUGUCU GAG]-----[ UGUUUG]--G UUUA--[CA{ U}AACA]------U[GAUU- Lo_miniata [UGUCUGUCU GAG]-----[ UGUUUG]--G AUUA--[CAA AACA]------U[GAUU- L_violacea [UGUCUGUCU GAG]-----U G[UUUG]--G UUUA--[CAA A]------ACGU[GAUU- L_virgulata [UGUCUGUCU GAG]-----U G[UUUG]--G UUUA--[CAA A]------ACAU[GAUU- Le_setacea [UGUCUGUCU GAG]-----U G[UUUG]--G UUUA--[CAA A]------ACAU[GAUU- M_flavida [UGUCUGUCU GAG]-----[ UGUUUGG]-- UUA--[CCAA ACA]------GU[GAACG M_bayeriana [UGUCUGUCU GAG]-----U [GUUUGG]-- UUA--[CCAA AC-]------CGU[GAACG T_hardyi [UGUCUGUCU GAG]-----U [GUUUG]--G UUAA--[CAA AC-]------U[GAACG Pl_nutans [UGUCUGUCU GAG]-----U [GUUUG]--G UUAA--[CAA AC]------U[GAACG Pt_citrina [UGUCUGUCU GAG]-----[ UGUUUG]--G UUAU--[CAA ACA]------AU[GAACG P_elisabethae_CBC [UGUCUGUCU GAG]-----U G[UU-G]--G UUAA--[CGA ]A------ACCG[GAUCG P_bipinnata_CBC [UGUCUGUCU GAG]-----U G[UU-G]--G UUAA--[CGA ]A------ACC-[UGGU- P_bipinnata_B [UGUCUGUCU GAG]-----U G[UU-G]--G UUAA--[CGA ]A------ACCU[GAUCG

....|....| ....|....| ....|....| ....|....| ....|....| ....|....| 65 75 85 95 105 115 Characters # ------3i------3' --3'a----- E_laciniata -----GGACG C]------C-GU------[GCG UUCGUGCG]- G_ventalina }------UCG CG]------CGUGCGUUUG C-GCU------[ CGCGAUC]-G G_mariae }------UCG C]------AG CGUGCGUUUG CCGCCGCC------[GCGAUC]-C G_flabellum }------UCG CG]------CGUGCGUUUG C-GCU------[ CGCGAUC]-G Pi_flava -AACG~{C}G C]------AUC------[GU-C-G UUCGUUC]-G Pi_platysoma -AAC-~~~~A CG]C------AU------[CGU{C}G UUCGUUC]-G A_schoutedeni C------G CG]------GA------[CG CGCAUUC]-G A_africana ------CG]------[ CGCAUUC]-G P_stenobrochis ---G{U}GUG CG]------AC------[UGUGC---{ U}CGAUC]-G L_stheno ------GUG UG]------UAUC------[CG CACGAUC]-G L_euryale ------GUG UG]------UAUC------[CG CACGAUC]-G L_hebes ------GUG U]------AUU------[A CGCAAUC]-G Lo_miniata ------GUG U]------AUU------[A CGCAAUC]-G L_violacea ------GUG UAU]------UAC------[GUA CACGAUC]-G L_virgulata ------GUG UG]------UAUCAU ------[CA CGCGAUC]-G Le_setacea ------GUG UG]------UAUCAU ------[CG CACGAUC]-G M_flavida G{C}------GGC]------GUAU------[GCU{C} U-CGUUC]-G M_bayeriana G{CUUUU}-- -GGCAUAU]------AUAU------[ AUAUGCU{CU }CCGUUC]-G T_hardyi CACG]------UUUUUG ------[CG{A} UGUGUUC]-G Pl_nutans CACG]------UUUUU------[CG{A} UGCGUUC]-G Pt_citrina GCA]------C ACG------[UGC---{ U}CGUUC]-G P_elisabethae_CBC ]------CGCCUG CCUGCAUUCU ------[CGAUC]-G P_bipinnata_CBC ------CG C]------GUAC------[GCGAUCG]- P_bipinnata_B CG]------AA------[ CGCGAUC]-G

....|....| ....|....| ....|....| ....|....| ....|....| ....|....| 125 135 145 155 165 175 Characters # 3i'---4--- -4a------4b------

75 E_laciniata ACU[-GGGG- C{GUC}GCGC CGC]CGCC------G_ventalina AC[-UGAGG{ CUGU}CGCAC GC]------G_mariae AC[-UGAGG{ CCGU}CGCAC GCG]------G_flabellum AC[-UGAGG{ CUGU}CGCAC GC]------Pi_flava AA[-UGGGG- -G{U}CGCGU -AG{CGC}GC AA]------CUUAC Pi_platysoma AA[-UGGGG- -G{U}CGCGU -AG]------CGCGCAACUU ACUUACUUGC A_schoutedeni AC[-UGGGG{ C-GU}CGCGC CGC]------C------A_africana AC[-UGGGG{ C-GU}CGCGC CGC]------C------P_stenobrochis A-[UUGAGG{ CUGU}CGC{G U}CGUUG]------UCUCCGU------L_stheno A-[UUGAGG{ CUGU}CGCAU {C}GUUG-UU U]------L_euryale A-[UUGAGG{ CUGU}CGCAU {C}GUUG-UU U]------L_hebes A-[UUGAGG{ CUGU}CGUGU {C}GUUG-UU U]------Lo_miniata A-[UUGAGG{ CUGU}CGCGU {C}GUUG-UU U]------L_violacea A-[UUGAGG{ CUGU}CGCGU {C}GUUG-UU U]------L_virgulata A-[UUGAGG{ CUGU}CGCGU {C}GUUG-UU U]------Le_setacea A-[UUGAGG{ CUGU}CGCGU {C}GUUG-UU U]------M_flavida A-[UUGGGG- -G{U}CG{G} ---GUAG{CG U}UGCG{C}G U]------M_bayeriana A-[UUGGGG- -G{U}CG--- {C}GUAG{CA U}UGCAU]------T_hardyi A-[UUGGGG- C-GU]------CGCGU-GCGC AACAU-AACA Pl_nutans A-[UUGAGG{ CGU}CGCGU{ U}CGC]------AACA Pt_citrina A-[UUGGGGG -{U}CGCGUA G]------CGUUGCA P_elisabethae_CBC AU[-UGAGG{ UUGU}CGC{U }UGC]------P_bipinnata_CBC AU[-UGAGG{ U}CG{U}CGC {U}CGC]------P_bipinnata_B AU[-UGAGG{ U-CG-U}CGC {U}GGC]------

....|....| ....|....| ....|....| ....|....| ....|....| ....|....| 185 195 205 215 225 235 Characters # -4i------4'--4'c--- -4'b------4'a------4 i'------E_laciniata GCC------[G CGGCGCGUCC C]-UCGAAGU GCCACCGUGA G_ventalina CCCGC------[G ------UGUG CG{UC}CCUC G]-AAGUGC------AGC G_mariae -CCG------[C G-----CGUG CG{UC}CCUC G]-AAGUGC------AGC G_flabellum CCCGC------[G ------UGUG CG{UC}CCUC G]-AAGUGC------AGC Pi_flava UUAC------[UUGC{U ACUA}CUACG CG{U}CCCUC G]-AAGUGC------AGC Pi_platysoma UACUA------[CUACG CG{U}CCCUC G]-AAGUGC------AGC A_schoutedeni GCC------[GCGGCG CG{UC}CCUC G]-AAGUGCC A------A_africana GCC------[GCGGCG CG{UC}CCUC G]-AAGUGCC A------P_stenobrochis CGGAG------[CAACG{U}G CG{UC}CCUC GA]-AGUGC------AGC L_stheno -CCAGC------[AAA {C}UAACGUG CG{UC}CCUC GA]-AGAGC------AGC L_euryale -CCAGC------[AAA {C}UAACGUG CG{UC}CCUC GA]-AGAGC------AGC L_hebes -CCAGC------[AAG {C}UAACGUG CG{UC}CCUC GA]-AGUGC------AGC Lo_miniata -CCAGC------[AAG {C}UAACGUG CG{UC}CCUC GA]-AGUGC------AGC L_violacea -CCAGC------[AAA {C}UAACGUG CG{UC}CCUC GA]-AGUGC------AGC L_virgulata -CCAGC------[AAA {C}UAACGUG CG{UC}CCUC GA]-AGUGC------AGC Le_setacea -CCAGC------[AAA {C}UAACGUG CG{UC}CCUC GA]-AGUGC------AGC M_flavida -GUAC-[GC{ A}CGCA{UUU A}CUA-C{G} CG{U}CUCUC GA]-AUUGC------M_bayeriana -GUU--[GUG CA--{CAUUU A}CUA-C{G} CG{U}CUCUC GA]-AUUGCA GC------T_hardyi AGCAU------[G CG{UC}CCUC GA]-AGUGC------AGC Pl_nutans UAAC------[GUG{C}ACG CG{UC}CCUC GA]------A Pt_citrina UUA------[CUACG CG{U}CUCUC GA]AA-UGC------AGUAC P_elisabethae_CBC --CUA------[G CG{U}----G CG{UC}CCUC G]-AAGUGCA GC------P_bipinnata_CBC --CUA------[G UG{U-C}GCG CG{UC}CCUC G]-AAGUACA GC------P_bipinnata_B --CUAAC------[G U---C{GC}G CG{UC}CCUC G]-AAGUACA GC------

....|....| ....|....| ....|....| ....|....| ....|....| ....|....| 245 255 265 275 285 295 Characters # ------5 ------5a------5b------5c------5d---- E_laciniata [GC{ACA}CG CCGCCGUU{C AUCA}GUGC{ C}GCCGUGC{ GAA}----CA AGG{CCUA}A G_ventalina [------ACG--{UU}C GAC-{GCCC} GUGUCGUGC{ GAA}----CA AGG{CC}CAA G_mariae ------[ACG{UU}C GAC-{GCCC} GUGUCGUGC{ GAA}----CA AGG{CC}CAA G_flabellum ------[ACG{UU}C GAC-{GCCC} GUGUCGUGC{ GAA}----CA AGG{CC}CAA Pi_flava ------[ACG-UU-{ CAGCUCUA}- ----CAU--{ GAA}----CA AGG{CCU}AA Pi_platysoma ------[ACG-UU-C {AGCU}CU-- -{A}CAU--{ GAA}----CA AGG{CCU}AA A_schoutedeni ------[C CGAAG-U{GA }GC{GAGC}C G{C}CGUGC{ GAA}----CA AGG{CCUAA} A_africana ------[C CGAAG-U{GA }GC{AAGC}C G{C}CGUGC{ GAA}----CA AGG{CCUAA} P_stenobrochis ------[ --ACG{UU}C GAC-{GCUC} GUGUCGUGC{ GAA}----CA AGG{CC}CAA

76 L_stheno ------[ --ACG{UU}C GAC-{GCUGU }UGUCGUGC{ GAA}----CA AGG{CC}CAA L_euryale ------[ --ACG{UU}C GAC-{GCUGU }UGUCGUGC{ GAA}----CA AGG{CC}CAA L_hebes ------[ --ACG{UU}C GAC-{GCUCU }UGUCGUGC{ GAA}----CA AGG{CC}CAA Lo_miniata ------[ --ACG{UU}C GAC-{GCUCU }UGUCGUGC{ GAA}----CA AGG{CC}CAA L_violacea ------[ --GCG{UU}C GAC-{GCUCU }UGUCGUGC{ GAA}----CA AGG{CC}CAA L_virgulata ------[ --ACG{UU}C GAC-{GCUAU }UGUCGUGC{ GAA}----CA AGG{CC}CAA Le_setacea ------[ --ACG{UU}C GAC-{GCUAU }UGUCGUGC{ GAA}----CA AGG{CC}CAA M_flavida ------[A GAACG-UUCC GCUCU{CG}------GCAC{ GA}G----CA AGGC{CU}AA M_bayeriana ------[ACGUUCC GCUCU-CCGC AC------{ GAA}----CA AGG{CC}UAA T_hardyi ------[ACGUU{C GA}CGC{UC} ---GUGCGC{ GAA}----CA AGG{CA}UAA Pl_nutans --[GUAC{A} GCAC{GUUC} GACG-{CUU} ---GUGCGC{ GA}----ACA AGG{CA}UAA Pt_citrina ------[----GUUCC GCU{UUC}UG ------CAC{ GAA}----CA AGGC{CU}AA P_elisabethae_CBC ------[ --ACG{UU}C GACG{CUC}- GUGUCGUGC{ GAA}----CA AGG{CC}CAA P_bipinnata_CBC ------[ ACG{CGUU}C GGCG{CU}CG U{G}UCGC-{ G}CG{AA}CA AGG{CC}CAA P_bipinnata_B ------[ACG{UU}C GGCG{CUC}G UGUCGCGC-{ GAA}----CA AGG{CC}CAA

....|....| ....|....| ....|....| ....|....| ....|....| ....|....| 305 315 325 335 345 355 Characters # ------5i'--5' ---5'd------5'c------5'b-----5 E_laciniata AG]------UGUUG[--C UU{CGA}CCU UG{CCA}GCA CGGCGCGC{C GCC}GACGGC G_ventalina AG]------AA-[--C UUUG{U}CCU UG{AGA}GCG CGACGU---{ UC}GUCG-{U G_mariae AG]------AA-[--C UUUG{U}CCU UG{AGA}GCG CGGCGU---{ UC}GUCG-{U G_flabellum AG]------AA-[--C UUUG{U}CCU UG{AGA}GCG CGACGU---{ UC}GUCG-{U Pi_flava AG]------U--[-UU UU{CGU}CCU UG{CUA}GUG ------{AAGCGCU Pi_platysoma AG]------U--[-UU UU{CGU}CCU UG{CUA}GUG --{A}AG------{CGCU A_schoutedeni AG{U}GCG]U GCGU--[CGU UU{CGA}CCU UG{CCA}GCA CG{A}CG--{ AAG}ACGC-U A_africana AG{U}GCG]U GCGU--[CGU CU{CGA}CCU UG{CCA}GCA CG{A}CG--{ AAG}GCGC-U P_stenobrochis AG]------AA-[CUU UG--{U}CCU UG{CGA}GCA CGACG----U {CU}GUCG{U L_stheno AG]------AA-[CUU UG--{U}CCU UG{CUA}GCA CGACG----{ UCU}GUCG{U L_euryale AG]------AA-[CUU UG{C}--CCU UG{CUA}GCA CGACG----{ UCU}GUCG{U L_hebes AG]------AA-[UUU UG--{U}CCU UG{CCA}GCA CGACG----{ UCU}GUCG{U Lo_miniata AG]------AA-[UUU UG--{U}CCU UG{CCA}GCA CGACG----{ UCU}GUCG{U L_violacea AG]------AA-[UUU UG--{U}CCU UG{CCA}GCG CGACG----{ UCU}GUCG{U L_virgulata AG]------AA-[CUU UG{C}--CCU UG{CGA}GCA CGACG----{ UCU}GUCG{U Le_setacea AG]------AA-[CUU UG{C}--CCU UG{CGA}GCA CGACG----{ UCU}GUCG{U M_flavida ]------AGUU-[UU{ UU}-G-CCCU UGU{CA}GUG C{AGUAGU}- ---AGAGC{U M_bayeriana AG]------UUGC[CUU UG{C}--CCU UG{CAA}GUG C{AUUA}GG{ U}AGAGC{CU T_hardyi A]------GUUA[UUU A-{C}--CCU UG{CAA}GCG CAC{CGUC}- ----GCG{CA Pl_nutans A]------GACA[-UU UA{C}--CCU UGU{AA}GUG CAU-CGUC------Pt_citrina ]------AGUG-[UU{ UU}G{C}CCU UG{CCA}GUG CG------{U P_elisabethae_CBC AG]------AUU-[--U UUUG{U}CCU UG{CAA}GCG UGGCG------{U P_bipinnata_CBC A]------GAAC[--- UUUG{U}CCU UGCG{A}GCG -GGCG------{U P_bipinnata_B A]------GAAC[--- UUUG{U}CCU UG{CGA}GCG CGGCG------{U

....|....| ....|....| ....|....| ....|....| ....|....| ....|....| 365 375 385 395 405 415 Characters # 'a------5i'------6------E_laciniata G{CAC}GCG{ GA}GC]--UU C------[GAAA UCG]----UU G_ventalina C}------CGU]------CGAUGCCAAC AACCCCUACA GG----[UUA ]------G_mariae C}------CGU]------CAAUUCCAAC AACCACUCUA GG----[UUA ]------G_flabellum C}------CGU]------CAAUGCCAAC AACCCCUACA GG----[UUA ]------Pi_flava U}GAGAAC-- -GU]-----U CUUU------[UUA ]------CUU Pi_platysoma UGA}GAAC-- -GU]-----U CUUU------[UUA ]------CUU A_schoutedeni U------CGG]------AA AACCUCUCGU CGACC[GUU] ------CUU A_africana U------CGG]------AA AACCUCUCAU CGACC[GUU] ------CUU P_stenobrochis U}------CGU]------CCCUUUUC UA----[UUA ]------L_stheno U}------CGU]------CCCUUUA------[UAAU ]------L_euryale U}------CGU]------CCCUUUA------[UAAU ]------L_hebes U}------CGU]------CCCCUUA------[UAAU ]------Lo_miniata U}------CGU]------UCCUUUA------[UAAU ]------L_violacea U}------CGU]------CCCUUUAU UA---[UAA] ------L_virgulata U}------CGU]------CCCU------[UUA] -----UCAU- Le_setacea U}------CGU]------CCCU------[UUA] -----UCAU- M_flavida U}--G{C}GA GCGU{C}UCU ]------U------[UUAA ]----CCAUU M_bayeriana U}--G{C}GA ACGU]------UUC------[UAA] -----CCAUU

77 T_hardyi U}--AAC{A} --GU]------ACC-UUU------[UUAA ]------UU Pl_nutans GUG------CGUAC]------UUU------[UU]------CAUU Pt_citrina U}AGUGGAGC ]------UCUC------[UUA] -----CCAUU P_elisabethae_CBC U}CGUCG{GC }CGU{U}CAC ]------A AACGC------[UAAU G]----CAUA P_bipinnata_CBC U}CGCCG{UC }CGU]------C-[AAU UAU]--ACU- P_bipinnata_B U}CGUCG{UC }CGU]------C-[AAU UAU]--ACU-

....|....| ....|....| ....|....| ....|....| ....| 425 435 445 455 465 Characters # --6i------6'------6i'------1'------. E_laciniata UCAU---[CG AU{CG}UUC] ---UUA[ACC UCAGA{U}CA GGCA] G_ventalina -CAUCAU[UA A]------C[C UCAGA{U}CA GGCA] G_mariae -CAUCAU[UA A]------C[C UCAGA{U}CA GGCA] G_flabellum -CAUCAU[UA A]------C[C UCAGA{U}CA GGCA] Pi_flava UCU----[UA A]------C[C UCAGA{U}CA GGCA] Pi_platysoma UCU----[UA A]------C[C UCAGA{U}CA GGCA] A_schoutedeni AAUU---[AA C]------[C UCAGA{U}CA GGCA] A_africana AAUU---[AA C]------[C UCAGA{U}CA GGCA] P_stenobrochis ---U---[UA A]------C[C UCAGA{U}CA GGCA] L_stheno --CU---[AU UA]------AC[C UCAGA{U}CA GGCA] L_euryale --CU---[AU UA]------AC[C UCAGA{U}CA GGCA] L_hebes --CU---[AU UA]------AC[C UCAGA{U}CA GGCA] Lo_miniata --CU---[AU UA]------AC[C UCAGA{U}CA GGCA] L_violacea -UCUC--[UU A]------AC[C UCAGA{U}CA GGCA] L_virgulata --CUAU-[UA A]------C[C UCAGA{U}CA GGCA] Le_setacea --CUAU-[UA A]------C[C UCAGA{U}CA GGCA] M_flavida UUCAUUC[UU AA]------C[C UCAGA{U}CA GGCA] M_bayeriana UUCAUUC[UU A]------AC[C UCAGA{U}CA GGCA] T_hardyi U------[UU AA]------C[C UCAGA{U}CA GGCA] Pl_nutans UUU------[ AA]------C[C UCAGA{U}CA GGCA] Pt_citrina CU------[U AA]------C[C UCAGA{U}CA GGCA] P_elisabethae_CBC UA-----[CA UUA]------UAUUAAC[C UCAGA{U}CA GGCA] P_bipinnata_CBC ------[AU A{CAUC}AUU ]----AAC[C UCAGA{U}CA GGCA] P_bipinnata_B ------[AU A{CAUC}AUU ]----AAC[C UCAGA{U}CA GGCA]

3. DCSE Alignment showing secondary structures of twelve species of octocorals (Helix numbering as in Fig. 1; Character coding in Table 3.1.).

....|....| ....|....| ....|....| ....|....| ....|....| ....|....| 5 15 25 35 45 55 helix numbering ---1--1a------1i- --2------2 i------2 '-----2i'------3 Pseudo. crucis [UGUC{UG}U UUUGAG]-[U GUU]----GG U-CAAG-[AA CG]---AAAA ------[UG{ Plexaura homomalla [UGUCUGUUC UGAG]--U-[ GUUG]GU]-- --CAA-[GAA C]---GAAAA UGAUGA[UGA Plexaura kuna [UGUCUGU-C UGAG]--U-[ GUUG]GU--- --CAA-[GAA C]---GAAAA ------[UGA Eunicea tayrona [UGUCUGUUC UGAG]--U-[ GUUGGU]--- --CAA-[ACC GAC]------A--[AG Eunicea pallida [UGUCUGU-C UGAG]--U-[ GUU]----GG UU-----[AA C]------[AAA Eunicea colombiana [UGUCUGU-U UGAG]--U-[ GUUGGU]--- --CAA-[ACC GAC]------[GAA Eunicea mammosa [UGUCUGU-C UGAG]--UG- [UU]----GG U-C----[AA ]------G-----[AA Eunicea laxispica [UGUCUGUUU UGAG]--UG- [UUG]----G UU--AA-[CA A]AC------[UGA Eunicea_laciniata [UGUCUGU-C UGAGU]--GU [UUGG]---- -UCAAA-[CC GA]--CAAA------[UG- Eunicea tourneforti [UGUCUGU-C UGAG]--U-[ GUUGGU]--- --CAA-[ACC GAC]------Eunicea fusca [UGUCUGU-C UGA]---U[U GUUGGU]--- --CAA-[ACC GACA]-GAU------Eunicea flexuosa [UGUCUGU-C UGAG]--U-[ GUU]----GG UU-----[AA C]------A[AAC

....|....| ....|....| ....|....| ....|....| ....|....| ....|....| 65 75 85 95 105 115 helix numbering ----3a------3i------3'--3'a------3i '---4---4a Pseudo. crucis A}UGAUGAUG GAAU]----- CGG-[AUUUC GUCGUCG{A} CG]------A CU[GAGG{CG

78 Plexaura homomalla UGA-UGGAA] ------UUAGA[UUUC GUCGUCG]------ACNA UU[GAGG{CG Plexaura kuna UGA-UGAAA] ------UUGAA[UUUC GUCGUCG]------ACGA UU[GAGG{CG Eunicea tayrona UGCGCGGACG ]------CGUG--[C GUUCGUGC{G }ACU]------[GGGG{UG Eunicea pallida AC{UGA}AUG C]------NCAC-[GC GU{GUGUGU} GUUUU]---A [UUGAGG{UG Eunicea colombiana UGCGCGGACG ]------CGUG--[C GUCCGCGCAU UC]-----GA CU[GGGGC{G

Eunicea mammosa C{G}AAAAU{ GA}UAAU]GA UGAUGGA[AU UG{G}AUUUU GUU]-GUCGA C-[GAUUG{A Eunicea laxispica ACGCACGU]------UUUU--[GC G{A}UGUGUU CG]-AUU--- --[GGGG{CG Eunicea_laciniata -CGC-GGACG C]------CGU-[GC GUUCGUGCG] ----ACU--- -[-GGGGC{G Eunicea tourneforti [GA{AU}GCG CGGACG]------CACG-[C GUCCGCGC{A U}UC]---GA CU[GGGGC{G Eunicea fusca [GCGCGGACG C]------UGUGC-[GC GUCCGCGC]------GA C[UGGGG{CG Eunicea flexuosa --{UGAAC}G CACG]------UUUUU--[C G{A}UGCGUU ]------[CGA{UU}G

....|....| ....|....| ....|....| ....|....| ....|....| ....|....| 125 135 145 155 165 175 helix numbering -----4b------4 i-----4'-4 'b---4'a------Pseudo. crucis U}CGCG]------CUACAUCUA AGCAAU---[ GCGCG{UC}C CUCG]----- Plexaura homomalla UC}GCGCUA] ------CAUCUA AGUAA-[UA{ C}GCGU{C}C CUC]------Plexaura kuna UC}GC{AC}U A]------UAUAUA AGUUA-[UA{ CGC}GU{C}C CUC]------Eunicea tayrona U}CGCGCCG] ------CCG------[CGGCGC{G }UCCC]UC-- Eunicea pallida U}CG{UG}CA UU]------AAAAUUA AAA----[AA UG{UG}CG{U C}CCUCGA]- Eunicea colombiana UC}GCGCCGC ]------CGCC------[GCGGCGC GUCCC]---- Eunicea mammosa }GGU{CUC}G C{ACU}GUAU ]------AUA AGUA----[A UACGCGUU{G }C]------G Eunicea laxispica U}CGCGUGC] ------GCAACAUAA CAA------[GCAUGCG {UC}CCUC]- Eunicea_laciniata UC}GCGCCGC ]------CGCCGCC------[GCGGCGC GUCCC]---- Eunicea tourneforti UC}GCGCCGC ]------CGCC------[GCGGCGC GUCCC]---- Eunicea fusca U}CGCG]------CCGCUCGA AGU------[CGCG{UC}C CUCG]----- Eunicea flexuosa GGGCGUCGCG UU]------CG CAACAU[AAC GUG{CAC}GC GUCCCUCG]-

....|....| ....|....| ....|....| ....|....| ....|....| ....|....| 185 195 205 215 225 235 helix numbering ------4i' ------5------5a------5b -----5c--- Pseudo. crucis --AAUUCCAU GCUCGCC------Plexaura homomalla -GAAUUCCAU GCUCGCCGAC ------Plexaura kuna -GAAUUCCAU GCUGGCCGAC ------Eunicea tayrona ------[ GAAG-UG{CC AC}CG{C}GA GC{A}----- Eunicea pallida --AGUGCAGC ------[AAG{UU C}GA------Eunicea colombiana ------[UC GAAG{UGCCA }CCGAAGU{G A}GC{AAGC} Eunicea mammosa UAUUAAUUCC AACUCAAAUU CCA------[UUC UCGCC{AA}- Eunicea laxispica ------G------[AAGUGC-{A GCAC}-GUUC GACG------Eunicea_laciniata ----UCGAAG UGCCACCGUG A------[G C{ACA}CGCC GCCGUU{CAU CA}GUGC{C} Eunicea tourneforti ------U------[C GAAG{UGCCA }CCGAAGUG{ AG}CG{AGCC Eunicea fusca ------AAG UGCCACCGUG AGCA------[CGCCG {CCGUCCGA} UCGGC-{GC} Eunicea flexuosa AA------[GUAC{A}G CA{AGUU}CG

....|....| ....|....| ....|....| ....|....| ....|....| ....|....| 245 255 265 275 285 295 helix numbering --5d------5e------5f------5e------5i'------Pseudo. crucis ------[GU- -CGUGC{A}A GCAAGACU]------UAAA GCU------Plexaura homomalla ------[GUG {C}-AA-GCA AG{ACAU}AA ]------AGCUGG UCCUUG---- Plexaura kuna ------[GUG {C}-A-CGCA AG-GNGGUAA ]------AGCUGG UCCUUG---- Eunicea tayrona CGCCGCCGUG C--{GAA}CA AGG{CC}UAA ]------AG U-GUUG---- Eunicea pallida -{U}GUUGUG {C}GA{A}CA AGG{CCUA}A AU{AAAA}UU UGC]------CCUU----- Eunicea colombiana C-G{C}CGUG C--{GAA}CA AGG{CCUAA} AG{U}GCG]------UGC Eunicea mammosa ------CGUG U{AC}--GCA AG{UCG}UAG ]------UAAAG------Eunicea laxispica --{CUC}GUG CGC{GAA}CA AGG{CA}UAA A]------GUUA----- Eunicea_laciniata --G-CC-GUG C--{GAA}CA AGG{CCUA}A AG]------UGUUG----- Eunicea tourneforti GC}---CGUG C--{GAA}CA AGG{CCU}AA A{GU}GC]------G UGC------Eunicea fusca C---GCCGUG C--{GAA}CA AGG{CCU}AA A]------G UGUUG----- Eunicea flexuosa AC-GCUU{GU GCGCGA}ACA AGG{CA}UAA A]------GACA ------

....|....| ....|....| ....|....| ....|....| ....|....| ....|....| 305 315 325 335 345 355 helix numbering ------5 '--5f'------5e'------5'd----5' c------5'b----- Pseudo. crucis ------[GGUCCUU GUU{C}GCAC GC]------

79 Plexaura homomalla ---[UU{CGC ACGC}UCCCU GCUUCUC]------Plexaura kuna ---[UU{CGC ACGCUC}CCU GCUUCUC]------Eunicea tayrona ------[U UA{CGA}CCU UG{CCG}GCA CGGCG{C}GC G{GA}GCUC{ CU}CG{AGAC Eunicea pallida [GCAAG{C}A UU{GCA}UUU UG{AACUCA} UC{A}CACAA C{C}UC{CAU }CUU]----- Eunicea colombiana GU----[CGU UU{CGA}CCU UG{CCA}GCA CG{A}CG{AA G}GCGCU-CG G{AAAACCU} Eunicea mammosa ------[CUG --{NNC}CUU GU{UC}ACAC G{CUG}GGCG {C}AG{C}GA ]------Eunicea laxispica ------[U UUA-{C}CCU UG{CAA}GCG CACCGUCG{C GCAU}AAC{A }GUA{C}CUU Eunicea_laciniata ----[--C-- UU{CGA}CCU UG{CCA}GCA CGGCGCGC{C GCC}--GACG GCG{CAC}GC Eunicea tourneforti --[GU{CG}U UU{CGA}CCU UG{CCA}GCA CG{A}CG{AA GA}CGCUUCG G{AAAAC}CU Eunicea fusca ------[U UU{CGA}CCU UG{CCA}GCA CGGCG{CGCC }GCCGA-CGG CG]------Eunicea flexuosa ------[U UUA-{C}CCU UGUAAGUG{C A}UCG{UCG} UGCGUAC]------

....|....| ....|....| ....|....| ....|....| ....|....| ....|....| 365 375 385 395 405 415 helix numbering ---5'a------5i'------6 ------6 i------6' Pseudo. crucis ------UCC------[ UUG]------CUUCUCU[UA Plexaura homomalla ------UUAA C------Plexaura kuna ------UUGA C------Eunicea tayrona }CGUUUC]------ACCG AUC------[G UU]------CUU----[AA Eunicea pallida ------UAA C------Eunicea colombiana CU{CA}UCGA ]------CCG -UC------[U AA]-----UA C------[UU Eunicea mammosa ------CCUCCCUGC AGC------[ UU]------CUCUU--[AA Eunicea laxispica ]------U------[ UU]-----AA UUUUU--[AA Eunicea_laciniata G{GA}GC]------UUC------[G AAAUCG]--- -UUUCAU[CG Eunicea tourneforti {C}UCG]------UCGACC------[G UU]------CUUAAUU[AA Eunicea fusca ------CCG CGCGCAGCCU CGAACCAC[G UU]------CUU----[AA Eunicea flexuosa ------CU [UU]-----U -CAUUUUU[A

....|....| ....|....| ....|....| ... 425 435 445 helix numbering ------6 i'------1' --1'a------. Pseudo. crucis A]------C--[CUCA GA{UCA}AGG CA] Plexaura homomalla ------[CUCA GA{U}CAGGC A]. Plexaura kuna ------[CUCA GA{U}CAGGC A]. Eunicea tayrona C]------[CUCA GA{U}CAGGC A]. Eunicea pallida ------[CUCA GA{U}CAGGC A]. Eunicea colombiana A]------CC--[CUCA AA{U}CAGGC A]. Eunicea mammosa ]------C--[CUCA GA{U}CAGGC A]. Eunicea laxispica ]------C--[CUCA GA{U}CAGGC A]. Eunicea_laciniata AU{CG}UUC] UUA[ACCUCA GA{U}CAGGC A]. Eunicea tourneforti C]------[CUCA GA{U}CAGGC A]. Eunicea fusca C]------[CUCA GA{A}CAGGC A]. Eunicea flexuosa A]------C--[CUC AGA{U}CAGG CA]