Investigating the variability of the morphological characters classically used to describe

species of Opheodesoma (Echinodermata: Holothuroidea: Synaptidae)

Laura J. Kenyon

UFID: Redacted

Senior Honors Thesis

Advisor: Dr. Gustav Paulay

Florida Museum of Natural History, University of Florida, Gainesville, FL 32611-7800 USA

ABSTRACT:

Opheodesoma is one of the genera belonging to the sea cucumber family Synaptidae, found in shallow tropical waters. There are 11 species names currently used. The species described were differentiated based on morphological characters exclusively. However,

Opheodesoma specimens exhibit a lot of variation in their morphology and species names have been revised throughout the past century. Due to the extent of the variation observed, and because the species were described from a limited number of individuals, it is probable that some of the currently described species represent intra-specific variation. We hypothesize that morphological characters previously used are too variable to distinguish between species. A phylogenetic tree based on a portion of the mitochondrial marker CO1 sequenced from 40 specimens reveals three genetically distinct groups. We tested several morphological characters to see if they co-varied with the mtDNA data. We did not find diagnostic morphology to distinguish between the two closest related groups, supporting our hypothesis. If further investigations are unable to distinguish the two groups, they must represent a single species.

However, the third genetically distinct group did have morphological characters that co-varied with the mtDNA sequence data and appears to be an un-described species.

KEYWORDS: phylogenetics, COI, sea cucumber, Synaptinae, Micrournae

2 INTRODUCTION:

Opheodesoma belongs to Apodida, a unique lineage of sea cucumbers. Apodida is sister to all other sea cucumbers and is unified by the lack of tube feet, respiratory trees, and radial water canals (Kerr 2001). Apodida contains three families: Chiridotidae, Myriotrochidae, and

Synaptidae. Of these, synaptids are the most diverse in shallow water habitats and are only found on reefs or hard bottoms. This monophyletic group is distinguished from the others by the presence of anchor and anchor plate ossicles (calcium deposits). Synaptidae contains 11 genera belonging to two subfamilies and includes Opheodesoma (Clark 1907).

Opheodesoma is part of Synaptinae, the subfamily of Synaptidae found in tropical waters

(Clark 1907). Opheodesoma belongs in the subgroup Micrournae which also includes Synapta,

Polyplectana, Synaptula, and Euapta (Heding 1928). These genera are united by their ossicles.

The anchor ossicles have smooth arms and minute knobs at the vertex and the anchor plates contain a bridge around the base. There are also miliary granules dispersed around the body wall

(Heding 1928). Micrournae have a distinct form of ciliated funnel (Heding 1928). Ciliated funnels are small (70 to 150 µ) funnel-shaped bodies made of connective tissue that are ciliated and reside in the mesenteries (Clark 1907).

Micrournae is found in shallow warm waters. It includes the longest sea cucumber in existence, Synapta maculata, which can reach over 3 meters, yet also contains species that only reach a few millimeters. They generally contain 12-15 tentacles like other synaptids, but

Polyplectana is characterized by having around 25 tentacles (Clark 1924). Micrournae reproductive physiology is also variable, ranging from hermaphrodism to sexual dimorphism and from free-spawning to viviparity (Clark 1924). They are deposit-feeders and are known to

3 occasionally form large (up to 1,000 animals) aggregations where an abundance of food is available (Berrill 1965). Micrournae is an important and dominant group on coral reefs and are often encountered by divers.

Opheodesoma is one of the 5 monophyletic clades in Micrournae and is sister to Euapta.

It is differentiated from Euapta by the presence of numerous Polian vesicles, which are part of the water-vascular system (WVS) in sea cucumbers, and the presence of numerous madreporites, which also function in the WVS. Opheodesoma is further differentiated from Euapta by the notable constriction of the base of Opheodesoma anchor plates.

The current species names of Opheodesoma are still poorly understood. The genus was first described by Fisher in 1907 and contains 11 currently used names based strictly on morphological characters. There are over 14 available names. Currently used names are often disagreed upon. For instance, O. mauritiae and O. africana were used until they became junior synonyms to O. grisea (Cherbonnier 1988). However, O. mauritiae was used in a 2003 publication (Samyn 2003). O. ramispicula and O. australiensis were both described by Heding in

1931, but O. ramispicula was later synonymized with O. australiensis (Clark 1946).

The presently accepted species of Opheodesoma are characterized by body wall coloration, webbing between the digits of the tentacles, thickness of the cartilaginous ring, the presence of rods (ossicles) in the oral disc, the presence of rods in the tentacles, and the color of the calcareous ring (Table 1). Other characters, less consistently used, include the thickness of the body wall, the shape of the tentacles, the distribution of miliary granules, the size of the anchors, and the size of the anchor plates. The lack of clarity in the species descriptions and the tendency for species names to be revised throughout history is due to the high amount of intra-

4 and inter-specific variability in morphology. Previous authors have recognized this variability but lacked other ways to differentiate organisms (Clark 1907). Also, many of these characters are subjective and difficult to interpret. In addition, previous species descriptions have often been based on only one or a few specimens. Thus, the amount of variation possible within a single species or within a single specimen has not been explored. As a result, we are unclear of the valid species present.

We must expand this investigation past just morphological data. With modern genetic technology burgeoning, we propose to incorporate molecular data with morphology. The use of genetic data will provide insight to the current valid species present. Since the mitochondrial maker Cytochrome Oxidase I (COI) has been widely used in Echinoderms, including many genera of sea cucumber, we propose the use of this maker in Opheodesoma (Ward 2008). As a mitochondrial marker, COI is expected to evolve at a faster rate than most nuclear markers, and can provide resolution among closely related organisms. A combination of COI data with morphology will reveal the extent of the variability seen in classically used characters.

The purpose of our investigation is to first see how many genetically distinct groups are present in the Florida Museum of Natural History collection of over 40 specimens, and then to assess the variability of morphology within these genetic groups. Given the extent of the variation recorded in historical descriptions, we hypothesize that the amount of variation seen within genetically distinct groups will parallel or exceed the amount of variation seen between groups. This would suggest that the characters classically used are too variable to describe species.

5 METHODS:

We studied 40 specimens from the Florida Museum of Natural History collections. We examined many of these specimens for the characters historically used in classification. We also included a new morphological trait to examine. We tested to see if the morphology covaried with

COI data.

Methods for the morphological investigation:

Cartilaginous ring: We did not mark the thickness of the cartilaginous ring. The anterior portion of the animal must be dissected, with tissue removed, in order to see the cartilaginous ring. It is invasive and can destroy the preserved specimen. Heding marked the ring as either

“voluminous” or “not voluminous or wanting” with no specific measurements, making the character state very subjective (Heding 1928). The trait is known to be highly variable in a single species and can grow with age (Clark 1924). Thus it would not be useful in this study.

Anchors and Anchor Plates: We decided not to mark the size of the anchors or anchor plates. The sizes of both of these ossicle types are variable and overlap among specimens (Clark

1924). Also, anchors and anchor plates grow throughout development, making size designation an unreliable character (Clark 1924).

Body wall coloration: Striped/Not Striped. We only examined live photos since preservation can alter coloration (Cherbonnier 1951). We were only able to examine 15 out of the 40 specimens, due to the lack of live photos available. We determined if the body wall had longitudinal striping present or absent. If the animal was blotched with different colors, or had small horizontal bands, it was marked as having no striping.

6 Webbing between digits: Present/Absent. We marked 23 specimens. Webbing exists between the digits (of the tentacles) as a thin membrane. This is hard to see if the animal is contracted or not-well preserved. Specimens that were too contracted or did not have well preserved digits were excluded from the examination. Examination requires extensive microscopy due to the minute size of the digits. This trait is continuous, thus we only marked specimens where the presence or absence was obvious. To maintain a consistent marking strategy, one person alone went through each of the specimens to mark the trait; thus, ambiguity associated with different definitions of webbing would be avoided.

Rods in the oral disc: Present/Absent. We examined a piece of oral disc tissue under a light-microscope. If the tissue was not thin enough to detect ossicles, the tissue was dissolved in bleach, washed three times with de-ionized water, dried, and mounted onto a viewing slide with

Euperal. The slides were then analyzed for ossicles. We marked 12 specimens.

Rods in the tentacles: Present/Absent. We marked 29 organisms. We isolated a piece of tissue from the tentacle and examined it with a microscope. Like with the oral disc, if the tissue was not transparent enough to see through, we prepared a permanent slide.

Color of the calcareous ring: White/Tinted Green. We only marked 2 specimens since this investigation is invasive. We must dissect the anterior portion of the animal and potentially remove surrounding body-parts. We attempted to mark the character in an adequate number of specimens to obtain usable results, while also trying to minimize the number of specimens dissected.

7 Thickness of the body wall: Thick/Thin. We determined if each specimen had a thick or thin body wall. A thick body wall was one that we could not see internal organs through. A thin body wall was one that was transparent. We marked 25 specimens for this trait. Results can be skewed by the relative contraction of the animals. If an animal is contracted upon preservation, the body wall may appear thicker than it is in a relaxed form. However, there is not an accurate way to account for this contraction, so we marked specimens as they were.

Shape of the tentacles: Thick, short, rounded/Thin, long, tapered. We marked 25 specimens. Long, thin, and tapered tentacles also had long digits compared to the short and thick tentacles. This trait is highly subjective, so one person marked each of the specimens based on a preconceived notion of what would be considered long compared to short. The length of the tentacles could be measured; however, because the length of the animal can vary dramatically with relative contraction, we would not have a good basis of comparison for our measurements.

Thus we decided to mark the character based on the overall perceived shape.

Miliary granules in the oral disc: Present/Absent. We examined 8 specimens. Using a microscope we examined thin tissue samples. If the tissue was too thick it was dissolved and examined as a permanent slide.

Miliary granules in the tentacles: Present/Absent. We examined 28 specimens. We used the same technique as we did with miliary granules in the oral disc.

Coloration of the oral disc: Same as tentacles/Different than tentacles. This is not a character used in previous literature, but it seemed to be variable among specimens. The oral disc either had a different color from the surrounding tentacles, or the same color. A microscope was

8 only used on very small specimens. We excluded those specimens that were too contracted to clearly view the oral disc. We marked 24 specimens.

Methods for COI analysis:

We sequenced a 650 base-pair portion of the mitochondrial marker Cytochrome Oxidase

I (COI). To do this, we extracted the DNA using DNAzol, a ready-to-use reagent, in order to isolate the DNA to later be used for PCR. We cleaned the DNA with QUIAGEN PCR cleaning kit, which removed all enzymes, salts, oligomers, etc. from the DNA. To isolate the desired portion of the COI marker we used the primers COIeF and COIeR (5’-

ATAATGATAGGAGGRTTTGG-3’ and 5’-ATGGAYGTAGAYACACGAGC-3’ respectively).

We performed a polymerase chain reaction (PCR), and sequencing was done at the

Interdisciplinary Center for Biotechnology Research (ICBR), a University of Florida world-class research center.

The sequences were aligned with MUSCLE, which is a public domain multiple alignment software that can be used with nucleotide sequencing data. We used Bayesian analysis to obtain the phylogenetic tree.

RESULTS:

The 40 specimens fall into three distinct phylogenetic groups based on COI sequencing

(Figure 1). One of the groups only contains one specimen, but it has several characters unique to it. Since we have morphological characteristics that agree with the phylogenetic data, we can confidently call this group an “Evolutionary Significant Unit” and we have specifically called it

“ESU 3”. The other two genetically distinct groups will loosely be referred to as “ESU 1” and

9 “ESU 2”, although at the present time we can not technically consider them to be ESUs. An ESU is defined as a group that is genetically distinct and has at least two characters that co-vary with the genetic data. Thus our results do not support ESU 1 and ESU 2 as true ESUs.

We found that none of the characters investigated differentiated ESU 1 and ESU 2 (Table

2). For every character, the traits were either the same in each ESU, or each ESU contained both of the variations. After considering collection locations, geography could not distinguish the groups either. We found that all specimens examined (28) had miliary granules in the tentacles.

We also found that all specimens examined (29) did not have rods in the tentacles.

ESU 3 did have morphological characters that differentiated it from ESU 1 and 2. The bases of the anchor plates have numerous perforations rather than the constant 7 found in all other Opheodesoma specimens (Figure 2). Also, ESU 3 was a solid burgundy color, with neither blotching nor striping. This coloration has not been documented in previous literature. Like other specimens, there were miliary granules present in the tentacles. All other specimens had either rods or miliary granules in the oral disc, but ESU 3 did not have either. Instead there were unknown salt clumps that have not been seen before.

DISCUSSION:

The purpose of this investigation was to test the variability of characters classically used to define species of Opheodesoma. When combining these characters with COI data, we found that they are too variable within genetic groups to differentiate between groups, which supported our hypothesis. We did not find any characters that distinguished ESU 1 from ESU 2. These two groups are not isolated geographically either. Therefore the term “ESU” does not strictly apply to

10 either of these two groups. Instead we suspect that these two groups represent a single species.

However, ESU 3 did have morphological characters that co-varied with COI data, suggesting that it is a different species.

Astoundingly, the webbing did not differentiate ESU 1 and 2. This is surprising since it has been emphasized in previous literature as an important character for defining species

(Heding 1928). While there is some ambiguity with discerning this character, we only included examples in this study where the character state was obvious. These results suggest that webbing is phenotypically plastic in Opheodesoma, and may result as a response to age, diet, feeding method, or habitat.

Also interesting was the absence of rods in the tentacles. Clark in 1924 suggested that the lack of rods in the tentacles was a synapomorphy for Opheodesoma. However, later work suggested that some specimens did contain rods in the tentacles, and this was a useful character for defining species (Heding 1928). This investigation suggests that Clark’s observation in 1924 might be correct. However, more specimens will need to be examined in order to make definitive conclusions.

Our investigation of the color of the calcareous ring lacked an adequate sampling size.

Due to the invasive nature of this investigation, we only examined one specimen from ESU 1 and one from ESU 2. However, we found bright white calcareous rings in both of these specimens. While more samples are needed, this tentatively suggests that the color of the ring does not vary between groups. It can be speculated that the color may depend on the animals’ diet or microhabitat. For instance, the tinted green rings may be found in animals that eat certain algae.

11 The thickness of the body wall may not be a reliable trait. It did not differentiate between groups and it is greatly influenced by relative contraction of the specimen. Since we had no other option, we marked only preserved specimens. There is no way to measure the relative contraction of the animals once they are preserved. Therefore we cannot determine if body wall thickness is just a by-product of contraction (i.e. if an animal is well contracted it appears to have a thick body wall). This might also be a product of specimen coloration and ossicle density. If the animal is very dark in color or contains a high concentration of ossicles, it may appear to have a thick body wall. Since body wall tissue is dissolved before ossicles can be examined, it is difficult to determine ossicle concentration or distribution. Future investigations using a SEM

(scanning electron microscope) where we can see the surface patterns of the body wall might yield new insight to this problem.

The shape of the tentacles could vary greatly with contraction of the animal and did not differentiate ESU 1 and 2. Like with the body wall, we are unable to determine the extent of tentacle contraction. The tentacles can contract with the rest of the body, but it is unknown if the digits can as well. Thus the length of the digits may not change with contraction, making this character more reliable. Further investigations that examine the ability of the digits to contract will be useful.

We used the terms ESU 1 and ESU 2 only for discussion purposes since it appears that these groups might represent a single species. We were surprised by the lack of differentiating characters due to the results from the COI data. This data indicates that there is a ~2% divergence between ESU 1 and 2. This is comparable to the 2% divergence of COI in two species of Euapta, the sister to Opheodesoma. E. lappa and E. godeffroyi are known to have

12 diverged with the closure of the Isthmus of Panama 3.5 million years ago. They have distinguishing morphological characters. This suggests that a divergence of 2% in COI can be a result of 3.5 million years of isolation. This gives us adequate reasoning to pursue ESU 1 and 2 to see if we eventually find distinguishing morphology.

One possibility that could explain the COI divergence but lack of morphological divergence is that these were once diverging groups (separated by some boundary) that did not reach complete reproductive isolation. This boundary was later destroyed and the groups began to interbreed more frequently which maintained the group as a single species. These two groups are found in similar locations, but perhaps this was not always the case. Only speculations can be made about what caused them to diverge in the past.

Another scenario is that these groups represent a single species that has been diverging over time, yet has not reached reproductive isolation. This requires that there is something currently driving their divergence. However, this seems unlikely. They are found in superficially similar habitats and have undistinguishable morphology. It is hard to guess what would be driving the divergence today.

Another possibility is that these two groups are two species that are reproductively isolated, and morphology has not caught up with COI divergence. Perhaps they are isolated by unique chemical coatings on their gametes, or unique chemical recognition signals that we are unable to detect. Their behaviors might be isolating them as well. Some avoid sunlight and hide under rocks or remain inactive during the day (Berrill 1965). This could be due to their photoreceptors being highly light sensitive (Clark 1924). It is possible that one group evolved the ability to withstand more light than the other group and overtime they evolved different

13 affiliations for daylight. One group may only be active (and breeding) at night, and the other during the day. This would explain that lack of distinguishing morphology, since morphology

(excluding the photoreceptors) would not be directly selected for. These are speculations and further investigations are needed.

Unlike ESU 1 and 2, ESU 3 shows unique morphology and is probably its own species.

The extra perforations in the anchor plate bases are extremely unique and have never been documented in previous literature. This specimen is a color that has never been documented. It did not contain any ossicles in the oral disc and instead had abnormal clumps of salt crystals.

This suggests that it is a species distinct from ESU 1 and 2, and an un-described species. More sequence data supporting this is needed, but this evidence is quite suggestive.

Future investigations must include more genetic and morphological data. Sequencing nuclear markers will be helpful in determining if ESU 1 and 2 are reproductively isolated. The calcareous ring is made of radial and inter-radial pieces, and some literature has suggested that these pieces show variability between organisms (Heding 1928). This could be a useful thing to investigate. We only looked for ossicles in the body wall, tentacles, and oral disc. However, ossicles can be present wherever there is connective tissue (Clark 1924). Therefore, other body parts could be explored for unique ossicles.

Behavioral studies may also reveal differences between these groups. This is extremely difficult but might be necessary. Synaptids do not survive for long periods of time in captivity, so investigations would need to be done in the wild. This would mean that the researcher must know which genetic group the animals belong to in order to make comparisons. Clearly this is a circular problem: the animal must be collected in order to obtain genetic data, but the organism

14 must be alive to analyze behavior. Thus the researcher could watch the behavior of an organism for a period of time and then collect it to use in genetic analysis. This type of work could provide valuable and unique information.

Although more research is needed, this investigation suggests that morphology alone is unable to differentiate species in Opheodesoma. This brings into question all of the currently used species names and urges a major revision. While the 40 specimens analyzed may not include all species that exist, this does represent the most commonly found species across the

Indo-Pacific. Thus it is likely that Opheodesoma contains fewer species than currently recognized.

This study reveals the amount of insight that we can gain from incorporating genetic data into phylogenetic studies. It urges the investigation of other clades that have been distinguished purely by morphology. In order to protect animals that are potentially at risk of extinction, we must have a firm understanding of the biodiversity that is present by doing investigations of this nature.

15 ACKNOWLEDGEMENTS:

I owe a lot of graditude to Dr. Gustav Paulay for his advice, motivation, and constant enthusiasm. I would also like to thank François Michonneau for always giving me valuable advice and for being an excellent mentor and role model. Lastly, the rest of the “Cuke Team” deserve thanks: Julie Zill, J.D. Paulsen, and the sea cucumbers in the FLMNH collections.

16 BIBLIOGRAPHY:

Berrill, M. 1965. The Ethology of the Synaptid Holothurian, Opheodesoma spectabilis.

Canadian Journal of Zoology. Volume 44: 457 -482.

Cherbonnier, G. 1951. Les Holothuries de Lesson. Bulletin au Museum. 2. XXIIL. Pp. 295-301

Cherbonnier, G. 1988. Échinodermes: Holothurides. Faune de Madagascar. ORSTOM 70 292

Clark, H.L. 1907. The Apodus Holothurians: A monograph of the Synaptidae and Molpadiidae.

City of Washington: Smithsonian Institution.

Clark, H.L. 1924. The Holothurians of the Museum of Comparative Zoology: The Synaptidae.

Bulletin of the Museum of Comparative Zoology LXV, no. 13.

Clark, H.L. (1946). The echinoderm fauna of Australia: its composition and its origin. Carnegie

Institution of Washington Publication, 566. Carnegie Institution of Washington:

Washington DC, USA. 1-567

Fisher, W . K. 1907. The holothurians of the Hawaiian Islands. Proc. U. S. N at. Mus. 32:

637- 744.

17 Heding SG. 1928. Papers from Dr. Th. Mortensen's Pacific expedition 1914–16. XLVI.

Synaptidae. Saertryk af Videnskabelige Meddelelser fra Dansk naturhistorisk Forening

85:105–323, pi. II and III.

Heding SG. 1931. Uber die Synaptiden des Zoologischen Museums zu Hamburg. Zoologische

Jarbaucher (Systematik) 61:637–696, pi. 11.

Kerr, Alexander M. 2001. Phylogeny of the Apodan Holothurians ( Echinodermata ) inferred

from morphology. Zoological Journal of the Linnean Society 53: 53-62.

Kerr, Alexander M. and Kim, J. 2001. Phylogeny of Holothuroidea (Echinodermata) inferred

from morphology. Zoological Journal of the Linnean Society 133: 63-81.

Samyn, Y. 2003. Towards an understanding of the shallow-water Holothuroid fauna

(Echinodermata: Holothuroidea) of the western Indian Ocean. PhD Thesis. Vrije

Universiteit Brussel: Brussel, Belgium. III, 384 + 1 cd-rom.pp.

Ward, Robert D., Bronwyn H. Holmes, and Tim D. O’Hara. 2008. DNA barcoding discriminates

echinoderm species. Molecular Ecology Resources 8, no. 6 (November): 1202-1211.

18 TABLES AND FIGURES:

Table 1: Descriptions of currently named species. Character states are labelled as follows: Body Color: 0= no striping, 1=striping; Webbing: 0=no webbing; 1=webbing; Rods in OD: 0=no rods; 1=rods; Rods in Tentacles: 0=no rods; 1=rods; Cart. Ring: 0=voluminous; 1=not voluminous; Color of Calc: 0=white; 1=green. If the character was not described, the space is left blank. The description of O. kamaranensis is only available in a foreign language and an English translation is not currently available.

Color: Species Name Body Webbing Rods: Oral Rods: Cartilagenous Calcareous Color Disc Tentacles Ring Ring O. spectabilis 0 1 1 0 0 1 O. serpentina 0 0 1 1 1 O. sinevirga 1 O. kamaranensis Most recent description unavailable. O. clarki 0 1 1 1 1 O. lineata 1 O. variabilis 0 0 0 0 1 O. australiensis 0 1 1 1 O. radiosa 1 0 1 0 0 O. glabra 0 1 1 0 1 O. grisea 0 0 1 0 0

19 Table 2: Number of specimens examined for each character. The number of specimens found to exhibit each character state are indicated. If no specimens were found to exhibit a particular character state, the space was left blank.

Character Character ESU 1 ESU 2 ESU 3 State Body Color Striped 2 6 Not Striped 4 2 1 Webbing Present 5 7 1 Absent 3 6 Rods in Oral Present 5 4 Disc Absent 1 1 1 Rods in Present Tentacles Absent 11 17 1 Color of White 1 1 Calcareous Tinted Green Ring Thickness of Thick 6 10 Body Wall Thin 2 7 Thick, short, 4 12 1 Shape of rounded Tentacles Long, thin, 4 4 tapered Miliary Present 10 17 1 Granules in Absent Tentacles Miliary Present 2 3 Granules in Absent 2 1 Oral Disc Same as 5 12 1 Color of Oral tentacles Disc Different than 2 4 tentacles

20

Figure 1: Phylogenetic tree based on COI morphological data and Bayesian analysis. Each genetically distinct group is labelled and color coded. UF identification numbers are reported and locations can be found in the Florida Museum of Natural History specimen database.

21

Figure 2: Anchor plates of ESU 1 (A), ESU 2 (B and C), and ESU 3 (D). Notice the 7 perforations in the base in A,B, and C. Notice the numerous perforations in the base in D. Note that the actual sizes of these anchor plates are between 150-250 µ in length and 140-185 µ in width.

22