ARTICLE
A new species of Gemmula, Weinkauff 1875; Evidence for two clades of Philippine species in the genus Gemmula

Baldomero M. Olivera*, David R. Hillyard and Maren Watkins
Departments of Biology and Pathology, University of Utah, Salt Lake City, Utah, 84112 USA
*Corresponding author


[Reprints of this manuscript have been sent to 12 Natural History Museums in six different countries. Reprints of this article may be obtained by sending a reprint request to the Editor of this journal.]
PDF VERSION

 

 

ABSTRACT
A new species of Philippine turrid, Gemmula ambara (the amber gem turrid), is described. The type locality is Balicasag Island, Bohol, Phlippines; the holotype and several paratypes were collected at the type locality using tangle nets at depths of 70-120 fathoms. The species occurs offshore across a number of central Philippines localities; it has also been collected off Panglao Island using tangle nets and off Aliguay Island using small trawls. Gemmula ambara had been mistakenly regarded by collectors and dealers as a small variety of Gemmula diomedea, but a more careful examination of the shell morphology easily separates the two species. A preliminary analysis of the phylogeny of Gemmula ambara using standard molecular markers has revealed that in fact, this species is not closely related to Gemmula diomedea but instead, is more closely related to Gemmula rosario and Gemmula lisajoni. It appears that the species of Gemmula from the Philippines can be divided into two distinct branches, with Gemmula ambara, Gemmula rosario and Gemmula lisajoni comprising one branch, and Gemmula diomedea, Gemmula speciosa, Gemmula kieneri, and Gemmula sogodensis comprising a second branch.


INTRODUCTION
Venomous marine snails belonging to the superfamily Conoidea are a significant component of the marine biodiversity of the Philippines and their venoms are proving to be a promising resource for novel biologically active compounds. The conoidean snails have traditionally been divided into three groups: cone snails (family Conidae), auger snails (family Terebridae) and turrids (family Turridae) — for reviews of these groups see Rockel et al. 1995; Bratcher and Cernohorsky 1987; Terryn 2007; Powell 1964, 1966. In numbers of species, the turrids are the largest group of conoideans, but also the most poorly characterized. Recently, it has become clear that the family Turridae is not monophyletic (Taylor et al. 1993; Puillandre et al. 2008); however, a generally accepted taxonomic treatment for the diverse groups formerly in the family Turridae has not yet been firmly established.
‚ ‚ ‚ ‚ ‚ The type species of the family Turridae is Turris babylonia (Linnaeus, 1758); this taxon also defines the subfamily Turrinae. Included in the subfamily Turrinae are several genera, e.g., Turris, Roding 1798, Gemmula, Weinkauff, 1875, Lophiotoma, Casey, 1904, Polystira, Woodring, 1928, and Turridrupa, Hedley, 1922. The taxonomy of the Turridae requires considerable taxonomic attention, with many of the recently collected Philippine turrid species remaining undescribed. This is primarily a deep-water group, with most species not accessible to the classical shallow-water collecting techniques that provided most marine gastropods described since the work of Linnaeus. In addition, the morphological differences between many closely related species are subtle, making it easy to confuse closely related forms.
‚ ‚ ‚ ‚ ‚ In this work we describe a new species in the Turrinae, in the genus Gemmula, Weinkauff, 1875 (the “gem turrids”). We define the species based on its distinctive morphological characteristics, but also establish its phylogenetic affinities through the use of standard molecular markers. The new species, the amber gem turrid, Gemmula ambara, is only the first of many undescribed Philippine turrids that we hope will be characterized in the near future using a combination of both a shell morphology based taxonomic description, as well as molecular phylogeny.


METHODS
Specimen collection. The holotype and paratypes from Balicasag Island were collected by fishermen using tangle nets offshore. Most of the other specimens examined were purchased through commercial sources in Mactan, Cebu.
Preparation of genomic DNA. A specimen dissected by Yuri Kantor on July 1, 2004, (a male, 34 mm in length) in Balicasag Island, Philippines was used for the molecular analysis. Genomic DNA was prepared from 5 mg Gemmula ambara proboscis tissue using the Gentra Puregene DNA Isolation Kit (Gentra Systems, Minneapolis, MN) according to the manufacturer's standard protocol.
Cloning and sequencing of COI mtDNA. 10 ng of Gemmula ambara genomic DNA was used as a template for polymerase chain reaction (PCR) with oligonucleotides corresponding to COI forward primer (5’ GGT CAA CAA ATC ATA AAG AYA TGY G 3’) and COI reverse primer (5’ TAA ACT TCA GGG TGA CCA AAR AAY CA 3’). The pcr cycling profiles are as follows: Initial denaturation (950C, 60s); followed by 40 cycles of denaturation (950C, 20s); annealing (55oC, 20s) and extension (72oC, 30s). The resulting pcr product was purified by gel electropheresis, recovered from agarose using High Pure PCR Product Purification Kit (Roche Diagnostics, Indianapolis, IN). The eluted DNA fragment was annealed to pNEB206A vector using the USER Friendly Cloning Kit (New England BioLabs, Inc., Beverly, MA) following manufacturer's suggested protocol and the resulting product transformed into DH5-alpha competent cells.
‚ ‚ ‚ ‚ ‚ The nucleic acid sequences of the resulting COI-encoding clones were determined according to the standard protocol for Automated Results.
Molecular phylogeny using 12S mitochondrial ribosomal RNA. The nucleotide sequences were obtained using PCR and the phylogenetic tree constructed as described in Heralde et al. (2007).


RESULTS
Morphological characterization of a new Gemmula species
Range. Collected in the central Philippines, primarily from Balicasag Island, Bohol, using tangle nets, 70-150 fathoms, and from Aliguay Island, using small trawls, 30-100 fathoms.



Figure 1. Gemmula ambara, new species. Shown are three views of the holotype, which is 39.6mm in length.

Remarks. This medium-sized Gemmula (the holotype is shown in Fig. 1) has the same general color pattern as Gemmula diomedea, but can be separated from that species by its distinctively different subsutural region (see Fig. 2). The molecular phylogeny (see next section) suggests that this species is most closely related to Gemmula rosario and Gemmula lisajoni.



Figure 2. Gemmula ambara, new species, compared to Gemmula diomedea, Powell, 1964. The three bottom specimens are Gemmula ambara, and the top specimen is Gemmula diomedea. Note the difference in the subsutural region.


Description. Shell moderately broad fusiform, length 30-55mm. Overall color of the shell is white, flushed with a distinct violet or purplish tone in some specimens (particularly when freshly collected), except for the subsutural area, which is always light amber to dark amber brown in color. The protoconch has 3-4 translucent yellowish-brown to purplish brown whorls, axially costate over the last two protoconch whorls. Post-nuclear whorls 9 - 10.
‚ ‚ ‚ ‚ ‚ The spire whorls are sculptured with a prominent subsutural fold comprising two strong, closely-spaced spiral ribs, immediately adjacent to the suture, with the region in between moderately elevated, such that the two ribs comprise a prominent amber keel that borders the suture. In earlier spire whorls, there are two additional ribs in the anterior subsutural region and as the shell increases in size, the subsutural region can have as many as 6 to 7 spiral ribs, with the two posterior ones comprising a subsutural keel always the strongest. There are axial columns between the spiral ribs, quite strong in some specimens. A diagnostic feature of this species is that the entire subsutural region up to the peripheral sinus rib is amber in color, and the rest of the shell (except the colored protoconch) is white.
‚ ‚ ‚ ‚ ‚ In the first few post-nuclear spire whorls, the most prominent features are the amber brown subsutural region and the strongly gemmulate sinus rib. By the fourth post-nuclear whorl, an additional spiral rib anterior to the gemmulate peripheral sinus rib becomes emergent and in mature specimens, two strong additional spiral whorls are found anterior to the sinus rib in the larger spire whorls.
‚ ‚ ‚ ‚ ‚ In the body whorl, there are six principal spiral ribs and at the base, numerous weaker spiral ribs surrounding one major, somewhat crenulate rib. There are 16 spiral ribs with a varying number of weak ribs between the principal raised spirals along the canal
Comparisons. Gemmula ambara looks superficially similar to smaller specimens of Gemmula diomedea; it can readily be distinguished from members of the G. diomedea complex from the Philippines because the entire subsutural region until the sinus rib at the periphery is amber brown. In all of the G. diomedea-like specimens (there are likely to be multiple species in this complex — compare the specimens tentatively assigned to G. diomedea in Figs. 2 and 3), the one or two raised subsutural ribs are brown but the remainder of the subsutural region has the same white background color as the rest of the shell (including the region between the two brown subsutural spiral ribs — white in G. diomedea, amber in G. ambara). When preserved, the protoconch is also diagnostic (see Fig 3): brown in G. ambara, generally white and much larger in G. diomedea, (although some varieties of G. diomedea, which may be different species, have a protoconch that is light yellowish). In addition, most Philippine specimens of Gemmula diomedea have a higher and narrower spire, and a proportionately shorter canal than Gemmula ambara. Most specimens of G. diomedea have diameter:length ratio = 0.30-0.32 while for G. ambara it is 0.33-0.35.
As noted in the molecular phylogeny section that follows, Gemmula ambara is most closely related to G. rosario and G. lisajoni. G. rosario is a pure brown shell, uniformly straw brown rather than amber and white, while G. lisajoni has a distinctive deeply purplish-violet canal, with a brown body (see Fig. 3). In contrast to G. ambara, there is no white coloration in either of these species. Other Philippine forms of Gemmula such as G. speciosa, G. kieneri and G. sogodensis are both morphologically and genetically even more distant and would not be mistaken for Gemmula ambara.



Figure 3. A comparison between Gemmula ambara (A) with Gemmula lisajoni (B), Gemmula rosario (C) and Gemmula diomedea variety (D). On the bottom are two of the protoconchs; note the differences in size and color between the protoconch of Gemmula ambara and Gemmula diomedea. The specimen of Gemmula ambara shown (A) is Paratype 19; although it is a dead-collected shell with a bore-hole, it had the best-preserved protoconch of all the specimens examined. The specimen in D is the form from Aliguay Island generally assigned to G. diomedea, but similar in shape to G. congener. The G. diomedea/congener complex needs to be more carefully analyzed, but all specimens in this group can be reliably separated from G. ambara.


Etymology. The amber colored subsutural region, characteristic of this species, is the origin of the name ambara.
Types. A summary of type specimens is given in Table 1. The holotype will be deposited in the Philippine National Museum. Paratypes will be deposited in the Museum National d'Histoire Naturelle in Paris; the Academy of Natural Sciences in Philadelphia, Pennsylvania; the Harvard Museum of Comparative Zoology, in Cambridge, Massachusetts, the Natural History Museum, Geneva, Switzerland. All of these museums will also be provided with a printed version of this manuscript.


Table 1: Gemmula ambara, Types examined.



Molecular phylogeny of Gemmula ambara
A molecular phylogeny of Gemmula ambara was carried out; the results are shown in Figure 4 and Table 2. Phylogenetic trees based on 12S ribosomal mitochondrial DNA sequences were constructed using several standard methods (see Methods); the data reveal that of the species of Gemmula analyzed, Gemmula ambara is more closely related to Gemmula rosario and Gemmula lisajoni than to the other Gemmula species for which sequences were available (Gemmula sogodensis, Gemmula speciosa, Gemmula diomedea and Gemmula kieneri). This result was rather surprising, since shells of Gemmula ambara have generally been confused with smaller specimens of Gemmula diomedea, as indicated above, since these forms share a white and brown color pattern.
      ‚The COI barcode sequence for Gemmula ambara was obtained and compared to other COI Gemmula sequences previously determined. The results for the barcode of all Philippine Gemmula species analyzed so far are in Table 2. As would be expected for distinct species, the barcode sequences diverge significantly from each other within the homologous COI interval.


Table 2. Barcode sequences (CO1)



Figure 4. Phylogenetic analysis using 12S mitochondrial DNA sequences. Phylogenetic tree based on a small number of 12S rRNA sequences for the genus Gemmula showing strong support for placement of Gemmula ambara in a clade with G. lisajoni and G.rosario to the exclusion of other Gemmula species and the outgroup Turris spectabilis. Identical trees with similar support values were generated using Bayesian methods with posterior probabilities (indicated by the numbers) and maximum likelihood analysis with bootstrap analysis.

DISCUSSION
In this paper, a new turrine species from the Philippines, the amber gem turrid, Gemmula ambara, is characterized. Most specimens examined have come from the Balicasag/Aliguay region of the Central Philippines where specimens have either been caught by tangle nets or using a small trawl at depths >30 fathoms. At these localities, it appears that the species is not uncommon, but has apparently been confused with similarly colored forms in the genus Gemmula that belong to the Gemmula diomedea complex. As is described above, there are a number of distinctive morphological features that reliably differentiate the shells of Gemmula ambara from the smaller Gemmula diomedea specimens of the same size as G. ambara.
‚ ‚ ‚ ‚ ‚ The molecular phylogeny reveals that contrary to expectations from the similarities in color and shell morphology of Gemmula ambara to G. diomedea, it is most closely related to Gemmula lisajoni and Gemmula rosario, although the color pattern of G. ambara is distinct and easily separable from those species. With a more careful examination of morphological characters of these species, certain more subtle shell morphological characteristics appear to be better correlated with the true relationships than is color pattern. For example, the sculpture of the subsutural region and the size of the protoconch may prove to be a better indicator of genetic affinities, with color pattern potentially more sensitive to environmental factors.
‚ ‚ ‚ ‚ ‚ Of the species for which molecular data are available in the Philippines that are conventionally assigned to the genus Gemmula, there now appear to be two very distinctive subgroups. One includes Gemmula speciosa, Gemmula sogodensis, Gemmula kieneri and Gemmula diomedea. The other includes Gemmula rosario, Gemmula lisajoni and the species we characterized in this work, Gemmula ambara. Other forms in Gemmula need to be analyzed and added to the phylogenetic tree. Recent evidence to be described elsewhere (M. Astilla and G. Concepcion, unpublished), suggests that some Gemmula species prey on terebellid polychaetes. Whether this prey preference is shared by G. ambara remains to be established.
‚ ‚ ‚ ‚ ‚ Clearly, the first step to obtain a comprehensive framework for understanding the gem turrids is to elucidate the taxonomy and phylogenetic relationships within the group. A number of new species of Gemmula have been described from the Philippines (Olivera 1999; Olivera 2004). Recently, a molecular characterization of venom components of some of the Turrinae has been initiated (Heralde et al. 2008; Watkins et al. 2006) and therefore, it is now possible to systematically investigate pharmacologically-active compounds in this group of species, including Gemmula ambara. However, the taxonomic and phylogenetic definition of the undescribed turrine species from the Philippines clearly needs to be addressed in a timely manner, to keep pace with the molecular characterization of their venom components.

Acknowledgements. This work was supported in part by a program project grant GM48677 from the U.S. National Institute of General Medical Sciences. We thank Patricia Showers Corneli for constructing the phylogenetic trees, Yuri Kantor for dissecting the specimens from which the DNA samples were obtained, Tuong Huynh for preparing the figures and Terry Merritt for her patience in typing the innumerable drafts of this manuscript.
References
Bratcher T, Cernohorsky WO. Living Terebras of the World. New York, NY: American Malacologists, Inc. 1987.

Heralde FM III, Watkins M, Ownby JP, Bandyopadhyay PK, Santos AD, Concepcion GP, Olivera BM. Molecular phylogeny of some Indo-Pacific genera in the subfamily Turrinae H. Adams and A. Adams, 1853 (1838) (Gastropoda: Neogastropoda). Nautilus 2007; 121(3): 131-138.

Heralde FM III, Imperial J, Bandyopadhyay PK, Olivera BM, Concepcion GP, Santos AD. A rapidly diverging superfamily of peptide toxins in venomous Gemmula species. Toxicon 2008; 51(5): 890-897.

Olivera BM. The subfamily Turrinae in the Philippines: the genus Turris (Röding, 1798). Phil J Sci 1999; 128: 295-318.

Olivera BM. Evaluation of Philippine Gemmula: I. Forms Related to G. speciosa and G. kieneri. Sci Diliman 2004; 17(1): 1-14.

Powell AWB. The family Turridae in the Indo-Pacific. Part 1. The subfamily Turrinae. Indo-Pacific Mollusca 1964; 1: 227-346.

Powell AWB. The molluscan families Speightiidae and Turridae. Bull Auckland Inst Mus 1966; 5: 1-184. + 123 plates.

Puillandre N, Samadi S, Boisselier MC, Sysoev AV, Kantor YI, Cruaud C, et al. Starting to unravel the toxoglossan knot: molecular phylogeny of the "turrids" (Neogastropoda Conoidea). Mol Phylogen Evol 2008; In press.

Röckel D, Korn W, Kohn AJ. Manual of the Living Conidae (Vol. I: Indo-Pacific Region). Wiesbaden, Germany: Verlag Christa Hemmen 1995.

Taylor JD, Kantor YI, Sysoev AV. Foregut anatomy, feeding mechanisms, relationships and classification of the Conoidea (=Toxoglossa) (Gastropoda). Bull Nat Hist Mus Lond (Zool) 1993; 59: 125-170.

Terryn Y. A Collectors Guide to the Recent Terebridae (Mollusca: Neogastropoda): ConchBooks 2007.

Watkins M, Hillyard DR, Olivera BM. Genes expressed in a turrid venom duct: divergence and similarity to conotoxins. J Mol Evol 2006; 62(3): 247-256.

BACK TO TOP