1 Introduction
Crustaceans are the fourth most species-rich group of animals on the planet, following insects, chelicerates, and molluscs. Foremost among the crustaceans in number and morphological diversity are the decapods. With nearly 15,000 extant species, they include the well-known crabs (Brachyura), hermit crabs and their relatives (Anomura), shrimps (Dendrobranchiata, Caridea, and Stenopodidea), lobsters (Astacidea, Gebiidea, and Axiidea), and other lesser known groups: Achelata, Polychelida and Glypheidea [1–3].
The Glypheidae Winckler 1883 are a group of lobster-like decapods that first appeared in the Lower Triassic, flourished during the Jurassic, became less abundant in the Cretaceous, and apparently disappeared before the end of the Eocene [4]. Glypheids were therefore thought to be an extinct group until 1975, when a male specimen of the species Neoglyphea inopinata, caught off the Philippines at a depth of 185 meters in 1908 and kept in the collections of the Smithsonian Institution, was described [4]. Subsequent expeditions in 1976, 1980, and 1985, captured additional specimens at the same site, and another specimen was trawled in the Timor Sea. This discovery stimulated new research on the group, and subsequently the glypheid lobsters were recognized as a separate infraorder [5]. More recently, in October 2005, during the EBISCO cruise (MNHN-IRD), a single female specimen of another glypheid species was discovered on a seamount in the Coral Sea, near New Caledonia, at a depth of 400 meters. Nicknamed ‘Jurassic shrimp’, it received some attention from the media [6]. This specimen has been named Laurentaeglyphea neocaledonica in 2006 [7,8].
Although several markers are now available that can resolve deep and shallow relationships within the Decapod Tree of Life [9,10], only four rRNA gene fragments (12S, 16S, 18S and 28S rRNA) have been sequenced from Neoglyphea inopinata. Consequently, the exact phylogenetic position of the Glypheidea is still unresolved [9]. Here, we sequenced four additional protein-coding genes, both mitochondrial and nuclear (cytochrome oxydase I, histone H3, phosphoenolpyruvate carboxykinase and sodium-potassium ATPase α-subunit) from the two extant glypheid species in order to investigate their relationships among decapods.
2 Material and methods
We constructed an expanded decapod molecular data set, which included representatives from Dendrobranchiata, Stenopodidea, Caridea, Brachyura, Anomura, Axiidea, Astacidea, Polychelida, Achelata, and the two extant glypheid species Neoglyphea inopinata and Laurentaeglyphea neocaledonica. The genes sampled are three mitochondrial genes: cytochrome oxydase I (COI), 12S and 16S rRNA, and five nuclear genes: histone H3, 18S and 28S rRNA, phosphoenolpyruvate carboxykinase (PEPCK) and sodium-potassium ATPase α-subunit (NaK).
2.1 Data collection
Table 1 lists the taxa used in the study. All the specimens sequenced for the study are housed in the crustacean collection at the Muséum national d’histoire naturelle in Paris and preserved in 80% ethanol.
List of taxa and markers used in this study.
| COI | 12S | 16S | H3 | 18S | 28S | PEPCK | NaK | ||||
| Dendrobranchiata | Penaeoidea | Aristeus ssp. | ----- | HQ211500 | HQ241511 | HQ241556 | HQ241523 | HQ241534 | EU427212 | EU427143 | |
| Penaeus ssp. | AY781297 | EF584003 | AF192088 | ---- | AF186250 | AF124597 | EU427213 | EU427144 | |||
| Pleocyemata | Caridea | Crangonoidea | Glyphocrangon ssp. | HQ241546 | HQ241502 | HQ241513 | HQ241558 | HQ241525 | HQ241537 | EU427242 | EU427173 |
| Pandaloidea | Heterocarpus ssp. | HQ241547 | HQ241503 | HQ241514 | HQ241559 | HQ241526 | HQ241537 | EU427243 | EU427174 | ||
| Plesionika ssp. | HQ241552 | HQ241507 | HQ241519 | HQ241564 | HQ241530 | HQ241541 | EU427249 | EU427180 | |||
| Stenopodidea | Stenopus ssp. | AF125441 | ----- | DQ079734 | DQ079701 | DQ079769 | DQ079812 | EU427247 | EU427178 | ||
| Axiidea | Callianassoidea | Callianassidae | AF436025 | ----- | DQ079777 | DQ079664 | DQ079741 | DQ079777 | EU427236 | EU427167 | |
| Achelata | Palinuroidea | Panulirus ssp. | AF339470 | ----- | AF337976 | DQ079697 | DQ079765 | DQ079808 | EU427230 | EU427161 | |
| Puerulus angulatus | HQ241554 | HQ241509 | HQ241521 | HQ241566 | HQ241532 | HQ241543 | EU427233 | EU427164 | |||
| Anomura | Galatheoidea | Munida ssp. | HQ241548 | HQ241504 | HQ241515 | HQ241560 | HQ241527 | HQ241538 | EU427188 | EU427119 | |
| Lomosoidea | Lomis hirta | AY595672 | AY595547 | AF436052 | DQ079680 | AF436013 | AY596101 | EU427187 | EU427118 | ||
| Paguroidea | Coenobita ssp. | HQ241545 | HQ241501 | HQ241512 | HQ241557 | HQ241524 | HQ241535 | EU427184 | EU427115 | ||
| Astacidea | Nephropoidea | Homarus ssp. | AF370853 | DQ298427 | AF370876 | DQ079675 | AF235971 | AY859581 | EU427219 | EU427150 | |
| Nephropsis ssp. | HQ241551 | HQ241506 | HQ241518 | HQ241563 | HQ241529 | HQ241540 | EU427221 | EU427152 | |||
| Thaumastocheles ssp. | HQ241555 | HQ241510 | HQ241522 | HQ241567 | HQ241533 | HQ241544 | EU427223 | EU427154 | |||
| Brachyura | Cancridea | Cancer ssp. | AF060771 | ----- | DQ079708 | DQ079668 | DQ079743 | DQ079781 | EU427196 | EU427127 | |
| Portunoidea | Portunidae | AY616443 | AY919085 | AJ130811 | DQ079669 | DQ079744 | DQ079782 | EU427205 | EU427136 | ||
| Glypheidea | Neoglyphea inopinata | HQ241549 | DQ298431 | HQ241516 | HQ241561 | AY583968 | AY583986 | ||||
| Laurentaeglyphea neocaledonica | HQ241550 | HQ241505 | HQ241517 | HQ241562 | HQ241528 | HQ241539 | HQ241499 | HQ241568 | |||
| Polychelida | Eryonoidea | Polycheles ssp. | HQ241553 | HQ241508 | HQ241520 | HQ241565 | HQ241531 | HQ241542 | EU427234 | EU427165 | |
DNA extraction was performed using the QIAamp DNA Micro Kit commercialized by Qiagen. Amplification and sequencing were performed using the following sets of primers: L1490 and H2198 for COI [11]; 16Sar and 16Sbr for 16S rRNA [12]; H3F1 and H3R1 or H3F2 and H3R2 for Histone H3 [13]; 12S1 and 12S3 [14] or 12SF and 12SR [15] for 12S rRNA; 1F-5R, 3F-18Sbi and 18Sa2.0-9R [16,17] for 18S rDNA; 28SC’1 [18] and 28SC2 or 28SD2 [19] for 28S rDNA; PEPCK for, PEPCK for2, PEPCK rev and PEPCK rev3 [20] for PEPCK; NaK for-b and NaK rev [20] for NaK.
PCR products were sequenced using the BigDyeTerminator V3.1 kit (Applied biosystem) and the ABI3730XL sequencer. Both strands of the PCR products were sequenced. The two strands obtained for each sequence were combined using the BioEdit Sequence Alignment Editor program [21].
The 70 sequences generated for this work have been deposited in GenBank under accession numbers HQ241499-HQ241568.
Nucleotide sequences were aligned by eye. Alignment was straightforward for the protein-coding genes, COI, H3, NaK and PEPCK, resulting in datasets of 666, 330, 534, and 570 base pairs (bp), respectively. Variable regions of the 16S, 18S and 28S genes were difficult to align reliably and were removed from the alignments. The final aligned sequences consisted of 438 bp for 16S, 1869 bp for 18S, 358 bp for 28S, and 364 bp for 12S. The few remaining gaps were treated as missing data in all further analyses. Alignments can be obtained from Marie-Catherine Boisselier.
2.2 Phylogenetic analyses
We performed both separate and combined analyses. We built phylogenies using Maximum Likelihood (ML) and Bayesian methods of inference. ML analyses were performed with RAxML 7.0.4 [22,23], and Bayesian analyses were performed with MrBayes 3.1 [24]. We used eight data partitions in the combined analyses, corresponding to the eight genes here sampled.
Bayesian analyses were performed by running 5,000,000 generations in four chains, saving the current tree every 100 generations, with the model inferred by Modeltest using the AIC criterion [25] applied to each partition (GTR). The last 48,000 trees were used to construct a 50% majority rule consensus tree. For the ML analysis, we used the same eight partitions and performed 1000 bootstrap replicates.
3 Results and discussion
As all recent molecular studies found Dendrobranchiata, Stenopodidea and Caridea to be the most basal decapod infraorders, we used them as outgroups [1,9,10,20,26,27]. The remaining decapods (Reptantia) are divided into two major groups supported by ML BP and Bayesian PP values of 100%, one including Brachyura, Anomura, and Axiidea, and the other including Astacidea, Polychelida, Achelata, and Glypheidea (Fig. 1).
Bayesian tree obtained from the combined data set (COI, 12S and 16S rRNA, 18S and 28S rRNA, H3, PEPCK and NaK; 5129 bp). Nodes with values are supported by Bayesian posterior probabilities values and ML bootstrap values above 95%.
The phylogenetic position of the Glypheidea has long been debated [5,28–33] and our study is the first molecular one to strongly support the inclusion of Glypheidea within one of the two major reptantian clades.
Our study also strongly shows that the two extant genera of glypheids cluster together. It further confirms the status of Glypheidea as a separate infraorder [5,34], in disagreement with several authors who considered glypheoids as Astacidea [33,35,36].
Among the major clade identified here that includes Astacidea, Polychelida, Achelata, and Glypheidea, the position of the latter is however unresolved. In the ML analysis, Glypheidea is placed as sister-group to Achelata (BP value < 50%, a result also obtained with weak support by Bracken et al. [9]) while the Bayesian analysis places Glypheidea as the sister-group to Achelata plus Polychelida (both forming the ancient Palinura infraorder) with a PP value of 72%.
Glypheids somewhat resemble crayfish or lobsters and are hypothesized to have an astacidean-like habitus (lifestyle) as well as a growth close to that observed for Nephrops norvegicus [34]. They also appear to be close to the Astacidea through their thoracic and abdominal structures, by their mouthparts and gills, and by the presence of gonopods [7]. The presence of a diaeresis on the uropodal exopod, with a spinose and distinct margin is another common feature and a horizontal position of the first pereiopod has been inferred to be a synapomorphy of the Astacura (Glypheoidea plus Astacidea) [32]. On the basis of nodal robustness, our results do not exclude a sister relationship of glypheids with astacideans [5].
Conflict of statement
No conflict of interest for any of the authors.
Acknowledgements
We thank A. Crosnier, R. Cleva and J. Brisset for their help in collection management. This work was supported by the “Consortium national de recherche en génomique”, and the “Service de systématique moléculaire” of the Muséum national d’histoire naturelle (CNRS UMS 2700). It is part of the agreement No. 2005/67 between the Genoscope and the Muséum national d’histoire naturelle on the project “Macrophylogeny of life” directed by Guillaume Lecointre.