First Molecular Evidence for Underestimated Biodiversity of Rhachotropis (Crustacea, Amphipoda), with Description of a New Species

The crustacean genus Rhachotropis has a worldwide distribution and amongst the largest bathymetric range known from any amphipod genus. DNA barcoding of new material from around New Zealand and the Ross Sea indicated depth-related biogeographic patterns. New Zealand Rhachotropis do not form a monophyletic clade. Species from bathyal depths on the Chatham Rise, east of New Zealand, show lower sequence divergence to bathyal species from California and the Arctic than to abyssal New Zealand species. Species sampled in the Kermadec Trench, north of New Zealand below 5000 m, seem to be more closely related to Ross Sea abyssal species than to the New Zealand shelf species. The worldwide geographic and bathymetric distribution for all Rhachotropis species is presented here. Depth may have a greater influence on phylogeny than geographic distance. Molecular and morphological investigations of Rhachotropis specimens from the Chatham Rise, New Zealand revealed a species new to science which is described in detail, including scanning electron microscopy. This increases the number of described species of Rhachotropis to 60 worldwide.


Introduction
The amphipod genus Rhachotropis (Eusiridae) contains 59 known species with a worldwide distribution (Fig. 1), [1]. Rhachotropis species appear to have a patchy distribution although some species are locally very abundant [1,2], especially in benthic slope communities [3]. Species in general have a relatively high swimming capacity, indicative of a partial pelagic lifestyle [3]. Phylogenetic analyses based on morphological characters have been unsatisfying or not possible. The numerical analysis of 20 morphological characters and corresponding character states by Bousefield & Hendrycks [4] focused on gross external morphology rather than mouthparts and reproductive features that may actually prove more significant phylogenetically as the authors suggested. Even though Rhachotropis show an impressive horizontal and vertical distribution, the genus has not been studied with molecular phylogenetic tools. This is a first preliminary analysis of the mitochondrial cytochrome oxidase c subunit 1 (COI) sequences of Rhachotropis specimens collected from bathyal and abyssal depths around New Zealand and in the Ross Sea.
This paper describes one new species collected on the Chatham Rise, east of New Zealand, and increases the number of known and described Rhachotropis species to 60, and the New Zealand Rhachotropis to four species. At least two further species from New Zealand waters appear to be new, but in too poor condition as to be formally described.

COI
Relationships for nine Rhachotropis specimens from New Zealand and the Ross Sea are shown in Fig. 2 and represent the topology inferred by both analyses. The trees were rooted with the Antarctic outgroup Eusirus sp., and include additional close matches for northern hemisphere Rhachotropis COI sequences held in Genbank: R. inflata, R. aculeata, R. inflata, R. helleri, and a putative new species from California R. sp 28 ( Table 1). The DNA barcodes revealed six well supported clades of Rhachotropis specimens from New Zealand and the Ross Sea with a further four clades for the northern hemisphere species (Fig. 2). Three specimens from the Chatham Rise, New Zealand, had identical COI sequences and were described as R. chathamensis Lörz, 2010. Two specimens from the Ross Sea had identical COI sequences and belong to R. abyssalis Lörz, 2010. A further three specimens from New Zealand had three unique COI sequences; one specimen which is described in this paper as R. novazealandica, Lörz, 2012 (Fig. 2), while the other two specimens remain undescribed: R. sp. A and R. sp. B (Fig. 2). A fourth single specimen from the Ross Sea with a unique sequence was described as R. rossi Lörz, 2010 (Fig. 2).
Sequence divergence was zero within the R. chathamensis and R. abyssalis clades, and low within the 24 specimens of R. aculeata (0.0089), 9 specimens of R. helleri (0.0003), and 13 specimen of R. inflata (0.037). A single specimen identified as Rhachotropis inflata (Cornwallis Island) is separated distinctly from the remaining clade (separated by 3% sequence divergence). And one tentative species, R. sp 28 from California, is represented by one sequence retrieved from GenBank. Inter-clade sequence divergences ranged from 0.143-0.370 with an overall average divergence 0.284. The lowest divergence (0.143, Table 2) was between R. novazealandica spec. nov. from New Zealand and R. sp. 28 from California, while the greatest divergence was between the two putative species R. sp. A and R. sp. B (0.370, Table 2) from New Zealand.
Morphological investigation revealed a species new to science which is described herein. Even though only a single damaged specimen exists, the COI sequence and detailed morphological descriptions will allow corroboration by future collections. Etymology. Rhachotropis novazealandica spec. nov. is named after New Zealand where the species was collected.
Description. Antenna 1 second article of peduncle with several plumose setae, article 2 slightly shorter than article 1, twice as long as article 3; flagellum broken after 10th-articulate. Antenna 2 peduncle article 3 longer than article 4, several plumose setae on third article; flagellum broken after 6 th article.
Mandible with smooth incisor process well developed, lacinia mobilis denticulate, molar process conical. Left and right molars have several pores in the middle. The tip of the left mandibular palp   Hypopharynx setose, outer lobes with broad gap. Gnathopod 1 coxa 1 produced, reaching to end of head, coxa 2, 3 and 4 subquadrate. Gnathopods similar in shape, subchelate. Gnathopod 1 slightly smaller than gnathopod 2, basis bearing several small spines at anterior side; merus with long setae at posteroventral corner; carpus lobe extending width of propodus, spines at terminal end of lobe; propodus widened, oval; dactylus slender, reaching end of palm.

Discussion
This is the first molecular study of Rhachotropis and has revealed a high level of diversity among specimens from the northern and southern hemispheres. The historic Rhachotropis collections, including the type material for most the species, were preserved in formalin or other DNA degrading media and are therefore not suitable for routine molecular investigations. Some fragile Rhachotropis specimens collected on recent expeditions were damaged and unsuitable for detailed morphological descriptions, but were fixed in ethanol to enable molecular studies. This study continues the integrative approach of DNA barcoding and classic taxonomy.
Most barcode projects aim to develop open-access libraries derived from referenced (vouchered) specimens that will improve understanding of biodiversity, highlight cryptic species, and provide rapid tools for identification of a wide range of species [5,6,7]. While barcoding has its limitations, especially the discrimination of recently diverged species that underwent introgressive hybridisation, the COI barcode region has been shown to be appropriate for discrimination between closely related species across diverse animal phyla [7][8][9][10]. Barcoding can highlight potentially cryptic species that appear in discrete clades with high sequence divergences as in the Rhachotropis case here. High intra-specific divergences indicate that additional data are required to distinguish potential new species from known species. The barcode databases, once established can be applied to the DNA identification of specimens where traditional morphological methods are inappropriate such as stomach contents in fishes [11,12], fish fillets [13,14] and environmental barcoding for biomonitoring [15].
Although there are ongoing discussions about the level of intraand inter-specific divergences in amphipods and the concept of species to be used [16,17] molecular species recognition is mostly based on the barcode ''gap'' between intra-and interspecific variations, with high inter-and low intra-clade sequence divergences indicative of cryptic species. Based on the barcode gap and consistent morphological differences, Lörz et al. [18] described and redescribed species of Antarctic Amphipoda and suggested that benthic species of Amphipoda do not occur circum Antarctic.
The inter-and intra-specific divergences of the Rhachotropis clades are in the same order of magnitude as for other deep sea Amphipoda (e.g. [10,18,19]). Interspecific uncorrected COI sequence distances in the Antarctic Iphimediidae varied from 7.9% (Echiniphimedia scotti to E. hodgsoni) to 29.5% (Iphimediella cyclogena to I. georgei) [18]. The deepwater Antarctic Rhachotropis species from the Admiralty seamount and Scott Island, to the north of the Ross Sea were in the same range, 28%.
Within the Epimeriidae sequence divergence varied from 8.5% (E. schiaparelli to E. macrodonta) to 26.15% (E. horsti to E. annabellae) [18]. Sequences of species from New Zealand's seamounts, Epimeria horsti and E. bruuni were more similar to each other than to any of the remaining Antarctic Epimeria species, but the distance between them was high with nearly 20%. The Antarctic Epimera species formed a monophyletic clade [18] while this study found the New Zealand Rhachotrois not to be monophyletic with the largest genetic distance of 37% between species..
Rhachotropis specimens are found in all major oceans of the world: Arctic, Atlantic Ocean, Mediterranean Sea, Carribean Sea, Indian Ocean, Pacific Ocean and the Southern Ocean (see Fig. 1 map). Rhachotropis specimens have been collected in all water depths (see Fig. 8a,b), from the shelf (e.g. [20]) to abyssal and hadal sampling sites (R. rossi, R. abyssalis    (e.g. [21]). Specimens used in this study are from three oceans, the Arctic, Southern and Pacific Oceans. Generally more species are currently known from the shelf and upper slope area, however, the observed depth pattern is heavily sample/collection biased and areas with more stations show more species. For example, detailed sampling at one deepwater location (2700 m Iceland Basin) shows four species. Similar results are found for Southern Ocean species in general [22] and in specific groups, such as isopods and gastropods [23].
The worldwide and broad depth distribution makes Rhachotropis an ideal model group to test the relationship between shelf and trench faunas or biogeographic ''processes'' such as sub-or emergence events. Submergence describes the downwards movement/shift of taxa from the shelf/shallow water depth to deeper waters (continental slopes and abyss) while emergence represents the upward movement of taxa from deeper to shallower depth [24] Currently there is insufficient specimen or habitat coverage to provide such comparisons and present a phylogeny of the genus, but this snapshot of Rhachotropis' molecular biodiversity provides an indication of what could be found with integrative methods and extensive sampling.
Our preliminary study suggests that the New Zealand Rhachotropis fauna is not monophyletic (Fig. 2), with the highest sequence divergence among all Rhachotropis specimens found between two species from New Zealand waters, R. sp. A from the Kermadec Trench (.5000 m), and R. sp. B, sampled from the Chatham Rise, east of New Zealand (418 m). Their position in the tree remains to be inconclusive with no node support. This suggests the use of additional molecular markers in subsequent studies. Based on COI New Zealand bathyal species seem to be closer related to Californian and Arctic shelf species than to New Zealand abyssal species. The New Zealand trench specimen shows a divergence of 30% to the Antarctic abyssal species, sampled below 3000 m at the Admiralty seamount and Scott Island. We therefore hypothesise that depth has a greater influence on the phylogeny of Rhachotropis than geography. The Kermadec Trench is one of the coldest trenches in the world [25]. The Deep Western Boundary Current purges Antarctic Bottom Water from the southern entrance into the Kermadec Trench [26], and it appears likely that the New Zealand trench species derive from Antarctica. However, further studies with additional molecular markers are needed to better resolve the tree and to support this hypothesis.
Further specimens from a comprehensive species set, from the shelf to abyssal and hadal depths, and additional genetic markers are required to test sub-or emergence theories. Our preliminary analyses testing DNA divergence against geography (Fig. 1) and depth (Fig. 8a, b), indicate that Rhachotropis could be a deep-sea taxon that has undergone several speciation events establishing it at bathyal depths (Emergence) in oceans around the world.

Materials and Methods
All necessary permits were obtained for the described field studies. Studies in the Ross Sea were undertaken under permit number AMLR07/005/ Tangaroa The amphipod specimens were identified to species level by the first author using identification keys based on morphological characters.
The amphipod specimens including the type material have been registered and curated at the National Institute for Water & Atmospheric Research (NIWA) Invertebrate Collection (NIC) in Wellington, New Zealand.

Morphological description
The specimen of the new species was dissected under a Leica MZ12 stereomicroscope and drawn using a camera lucida. All illustrations were digitally 'inked' following Coleman [35,36]. Inking was done with the software Adobe Illustrator 14.0 and an A3 drawing table (Wacom Intuos 9612).
Parts of selected specimens (mouthparts, antennae, coxal plates) were dried, coated with gold-paladium and investigated via a Scanning electron microscope LEO1525.
Nomenclatural Acts. The electronic version of this document does not represent a published work according to the International Code of Zoological Nomenclature (ICZN), and hence the nomenclatural acts contained in the electronic version are not available under that Code from the electronic edition. Therefore, a separate edition of this document was produced by a method that assures numerous identical and durable copies, and those copies were simultaneously obtainable (from the publication date noted on the first page of this article) for the purpose of providing a public and permanent scientific record, in accordance with Article 8.1 of the Code. The separate print-only edition is available on request from PLoS by sending a request to PLoS ONE, 1160 Battery Street, Suite 100, San Francisco, CA 94111, USA along with a check for $10 (to cover printing and postage) payable to ''Public Library of Science''.
In addition, this published work and the nomenclatural acts it contains have been registered in ZooBank, the proposed online registration system for the ICZN. The ZooBank LSIDs (Life Science Identifiers) can be resolved and the associated information viewed through any standard web browser by appending the LSID to the prefix ''http://zoobank.org/''. The LSID for this publication is: urn:lsid:zoobank.org:pub:B21B0DED-2543-40F0-BB02-3883D-F06A245 The LSID for Rhachotropis novazealandica spec. nov. is: urn:lsid:zoobank.org:act:F270B26E-A63D-42A2-B9F0-62A502-E2EFB4 Figure 8. a) Depth distribution of Rhachotropis species showing that the genus spans from the shallow to the abyss; b) the depth range of the 59 named plus the 3 spp from this study (A, B and sp. 28 California) Rhachotropis species. doi:10.1371/journal.pone.0032365.g008