Thoridae (Crustacea: Decapoda) can penetrate the Abyss: a new species of Lebbeus from the Sea of Okhotsk, representing the deepest record of the family

Lebbeus sokhobio sp. nov. is described from abyssal depths (3303−3366 m) in the Kuril Basin of the Sea of Okhotsk. The related congeners are deep-water dwellers with a very distant distribution and very similar morphology. The new species is separated by minor morphological features, such as the armature of the rostrum and telson, meral spinulation of ambulatory pereiopods and the shape of the pleonal pleurae. This species is the deepest dwelling representative of the genus Lebbeus and the family Thoridae. A list of records of caridean shrimps recorded from abyssal depths below 3000 m is given.


Introduction
The fauna of benthic caridean shrimps (Crustacea: Decapoda: Caridea) living at depths of more than 3000 m is poorly known due to the technical diffi culties of sampling. There are many records of caridean shrimps from the abyssal depths, but it is still expected that deeper sampling in different regions of the world oceans will provide new records and interesting scientifi c data. The deepest known records of

Material and methods
The material was collected during the megafaunal sampling of the joint Russian-German SokhoBio (Sea of Okhotsk Biodiversity Study) 2015 Expedition by the R/V "Akademik M.A. Lavrentyev" in the Kuril Basin, the deepest part of the Sea of Okhotsk (see Fig. 1). Collection was made using an Agassiz Trawl (AGT) or an epibenthic sled (EBS) (see Malyutina et al. 2018). Station data for all AGT deployments are presented in Fig. 1 and also by Blagodatski et al. (2017) and Malyutina et al. (2018). Start and, sometimes, the end coordinates refer to the positions ʻon groundʼ and ʻoff groundʼ, respectively. The AGT used in the SokhoBio 2015 Expedition was of a standard design with frame dimensions (width × height) of 350 × 70 cm and a mesh size of 10 mm. In general, the AGT was deployed twice at each station; however, at some stations only one AGT was deployed. The AGT was pulled between 4 and  Blagodatski et al. 2017). Numbers 1-11 indicate the stations where benthic samples were collected; stars indicate the stations that yielded specimens of Lebbeus sokhobio sp. nov. Heterogenys microphthalma (Smith, 1885) 3197-5060 m Indian Ocean; NE Atlantic Alcock 1901;Domansky 1986;Crosnier 1987 H. monnioti Crosnier, 1987Crosnier, 2663 Indian and Atlantic Oceans
Drawings of preserved specimens were made with the help of a camera lucida attached to an Olympus binocular microscope. Postorbital carapace length (pcl., in mm), i.e., the length from the orbits to the posterodorsal margin of the carapace, and total body length (tl., in mm), i.e., the dorsal length from the tip of the rostrum to the distal margin of the telson, are used as standard measurements. The material is deposited in the Zoological Museum of Moscow State University, Moscow (ZMMU, holotype), the Zoological Museum of the National Scientifi c Center of Marine Biology FEB RAS, Vladivostok (MIMB) and Naturmuseum Senckenberg, Frankfurt am Main, Germany (SMF).
To study molecular genetic barcodes, fragments of the mitochondrial gene coding cytochrome c oxidase subunit I (COI mtDNA), mitochondrial 16S ribosomal RNA (16S rRNA) and nuclear 28S ribosomal RNA (28S rRNA) gene markers were amplifi ed and sequenced. Total genomic DNA was extracted from muscle tissue using the innuPREP DNA Micro Kit (AnalitikJena, Germany) following the manufacturer's protocol. The COI mtDNA gene marker was amplifi ed with the help of the universal primers LCO1490 (5'-ggtcaacaaatcataaagatattgg-3'), HC02198 (5'-taaacttcagggtgaccaaaaaatca-3') (Folmer et al. 1994), for 16S rRNA (16SAR-cgcctgtttatcaaaaacat, 16SBR-ccggtctgaactcagatcacgt) after Palumbi et al. (2002), for 28S rRNA (28SA-gacccgtcttgaaacacgga, 28SB-tcggaaggaaccagctacta) primers after Whiting et al. (1997). PCR was performed with the T100 amplifi cator (Bio-Rad, USA) under the following conditions: initial denaturation at 96°C for 1.5 minutes followed by 42 cycles of 95°C for 2 minutes, 49°C for 35 seconds, and 72°C for 1.5 minutes, followed by chain extension at 72°C for 7 minutes. A volume of 10μl of the reaction mixture contained 1μl of total DNA, 2μl of 5×PCR mix (Dialat, Russia) and 1μl of each primer. The amplifi cation products were separated by using gel electrophoresis of nucleic acids on a 1.5% agarose gel in 1×TBE, and then stained and visualized with 0.003% EtBr using imaging UV software. DNA nucleotide sequences were determined using Genetic  Iwasaki & Nemoto 1987;Tiefenbacher 1991Tiefenbacher , 1994Gorny 1999 Sergestes arachnipodus (Cocco, 1832) 3300-4000 m Atlantic: Mediterranean Sea Company et al. 2004 Sergia robusta (Smith, 1882)  The aligned sequences of the COI mtDNA gene markers, 635 base pairs in length, were analyzed for pairwise sequence divergence (p-distances) and used to construct the phylogenetic relationships, whereas data on 16S and 28S are only presented in the ʻGenBank accession numbersʼ section, as there are no sequences to compare them to in GenBank (NCBI) or any other genetic database. The dataset of COI mtDNA gene marker alignments used in this study is presented in Appendix 1. The best evolutionary substitution model was determined using MEGA ver. 7.0. and jModeltest ver. 2.1.141 via the CIPRES Science Gateway ver. 3.3 (http://www.phylo.org/). Phylogenetic analysis was performed using MrBayes ver. 32.6 for the Bayesian analysis (BA) with an NKY + I + G evolutionary model and using RA×ML ver. 8.0.0 with a GTR + I + G evolutionary model for the Maximum-Likelihood analysis (ML). Bayesian analysis was carried out by sampling one tree every 1000 generations over 1 000 000 generations. Values of confi dence (bootstrap support) >50% are presented for BA/ML analyses; the divergences of p-distances are calculated using the Kimura-2-parameter (K2P) model in MEGA. The phylogenetic tree obtained based on COI mtDNA is presented in Fig. 7; there are no data on other species of Lebbeus based on the other gene markers (16S rRNA and 28S r RNA) for any valuable analysis, so they are just presented in this paper for future research.
Existing records of caridean shrimps below depths of 3000 m are presented in Table 1. It is based on the available literature found on the Internet using keywords as well as on a special search for scientifi c information in libraries using reference journals and electronic catalogs to search in publications not found on the Internet. It is possible that some pelagic species (e.g., members of Oplophoridae, Pasiphaeidae and Sergestidae) can be caught by trawls when they were hauled up from abyssal depths. The table includes all published records of caridean shrimps when the sampling depth was indicated as being deeper than 3000 m, without any assessment. Moreover, pelagic shrimps are often eurybathic and recorded by video vehicles from abyssal depths (e.g., Lörz et al. 2012;Jamieson et al. 2009;Wicksten et al. 2017;pers. obs.). Some remarks, for example for Galatheacaris abyssalis Vereshchaka, 1997, are presented in the text (see Table 1).

Etymology
This new species is named after the SokhoBio (Sea of Okhotsk Biodiversity Study) Expedition 2015, which allowed the collection of numerous deep-sea species such as this one.
Description CARAPACE (Figs 2, 3A-B). Smooth, without setae; dorsal surface slightly convex in males and gibbous in females, with well-marked postrostral median ridge armed with 2 postrostral teeth located at about anterior 0.2 of carapace length (Fig. 5); antennal tooth situated slightly below suborbital angle (Figs 3A-B, 5); supraorbital tooth large, directed forward, with deep notch below base, situated anterior rostral base; suborbital lobe prominent, triangular; anterolateral margin between antennal and pterygostomial teeth strongly sinuous, with deep concavity below antennal tooth; pterygostomial tooth acute, smaller and more slender than antennal and supraorbital teeth, overreaching anterior margin of carapace ( Fig. 3A-B).
ORBITS. Well developed, orbital margin with slight convexity posteriorly, base of eyestalk located between this convexity and suborbital lobe.
ANTENNULA. Antennular peduncle (Fig. 3A, D-E) well developed; basal segment about twice as long as wide, with dorsodistal margin armed with 3 slender spines; stylocerite well developed, acute, nearly reaching distal margin of basal segment, mesial margin sinuous; intermediate segment (article 2) stout, about 1.5-2.0 times as long as wide, with slightly convex mesial margin bearing long plumose setae and slender distolateral tooth; distal segment (article 3) short, about as long as wide, about half the length of intermediate segment, with acute dorsolateral subdistal tooth, with long plumose setae along mesial margin; upper antennular fl agellum with aesthetasc-bearing portion consisting of 10-12 articles. No sexual dimorphism detected. ANTENNA (Fig. 3F). Normal, well developed; basicerite armed with small tooth ventrolaterally; carpocerite overreaching midlength of scaphocerite; fl agellum well developed; scaphocerite wide, greatly overreaching antennular peduncle, about 3 times as long as maximal width, with well-developed distolateral tooth reaching distal margin of blade.
MOUTHPARTS. Typical for genus, without distinctive features. Mandible with 2-segmented palp; incisor process well-marked, terminating in sharp tip, bearing 4 distinct teeth and several additional denticles; molar process terminating distally. Maxilla I consisting of well-developed and partly fused endites, armed with spiniform setae and unsegmented bilobed palp. Maxilla II with simple, slender blunt palp; upper endite bilobed, fringed with setae; lower endite reduced; scaphognathite well developed, with rounded posterior lobe. Maxilliped I with partly fused endites, bearing short stout setae along distal margin as well as some elongated setose setae along distodorsal angle; exopod well developed, with   well-marked caridean lobe with many setae; palp 2-segmented; epipod ear-shaped, bilobed distally. Maxilliped II with well-developed exopod, fringed with setae distally; ischium stout, with long setae along lateral margin; propodus short, length equal to that of dactylus, with convex dorsal margin furnished with long simple setae, ventral margin unarmed; dactylus convex, armed with numerous stout, long, simple setae along distal margin; exopod fl agellate; epipod well-marked, distally bilobed, with podobranch. Maxilliped III (Fig. 4B-C) moderately long and stout, slightly overreaching scaphocerite and antennular peduncle; epipod well developed; exopod absent; antepenultimate article about 6 times as long as wide, slightly tapering distally, with longitudinal row of long spiniform setae along lateral surface and 3 long spiniform setae on distal margin; penultimate article about twice as long as wide, smooth; terminal segment about 7 times as long as wide, with distal margin oblique, armed with row of spines along distomesial margin.
PEREIOPOD I (Fig. 4D). Moderately robust; coxa with epipod and setobranch; basis stout, unarmed; ischium stout, about twice as long as wide, with long, simple setae along ventral margin; merus slender, about 4 times as long as wide, with row of spiniform setae proximally (Fig. 4F); carpus robust, about half the length of merus and slightly shorter than propodus, about twice as long as wide, slightly fl aring distally; distal margin slightly overlapping carpo-propodal articulation; mesial surface with grooming apparatus consisting of shallow concavity and complex of short, stiff setae; propodus (palm) about 3 times as long as wide, subcylindrical, smooth; fi ngers ( Fig. 4E) stout, about half the length of palm, subspatulate, about as long as wide; cutting edges straight, with strong distal teeth.
PEREIOPOD II (Fig. 4G). Relatively slender, unarmed; coxa with setobranch and epipod; basis small, about as long as wide; ischium about 4 times as long as wide, smooth; merus about 7 times as long as wide; carpus subdivided into 7 sub-articles with ratio of about 1:1:4:2:1:1:2; propodus (palm) subcylindrical, slightly shorter than distal carpal segment, about twice as long as wide and twice as long as fi ngers, with straight smooth margins; fi ngers slender, about 1.5 times as long as wide, with straight cutting edges.
SIZE. Largest female (holotype) has pcl. 19.0 mm and tl. 62 mm. Largest male has pcl. 17.0 mm and tl. At the same time, the phylogenetic signifi cance of grouping based on marked morphological features is rather doubtful ) and has not yet been proven due to the lack of suffi cient genetic data (see below). However, based on the presence of epipods on the basis of pereiopods I-III, the shape and armament of telson, with 4 pairs of dorsal spines, and the relatively slender dactyli of the ambulatory pereiopods, armed with numerous small accessory spinules, the new species may be close to L. africanus from Mauritania, at a depth of 1500 m (Fransen 1997), L. antarcticus from the South Ocean, at depths of 450-1775 m (Nye et al. 2013b), L. bidentatus known from off Chile, at a depth of 1680 m (Zarenkov 1976;Fransen 1997), L. carinatus collected off Peru, at depths of 1680-1860 m (Zarenkov 1976;Fransen 1997) (Rathbun 1902(Rathbun , 1904Wicksten & Mendez 1982) and L. virentova from the Mid-Cayman Spreading Center, Caribbean Sea, at a depth of 2294-2375 m (Nye et al. 2013a). The morphological features shared among these species include: short styliform rostrum, not reaching distal margin of second antennular segment, armed with 4 or more dorsal teeth, including postrostral teeth and with more than 1 ventral tooth; distinct U-or V-shaped notch inferior to base of supraorbital tooth; sinuous anterolateral margin of carapace between antennal and pterygostomial teeth with deep excavation below antennal tooth; pleomere II with distinct anterior transverse groove on tergum; basal antennular segment with 2 or 3 dorsodistal teeth; dactyli of ambulatory pereiopods armed with accessory spiniform spinules over entire length of fl exor margin (after Nye et al. 2012, with some modifi cations).
Some morphological features allow the separation of L. sokhobio sp. nov. from some of the other species in this group mentioned above. The new species is distinguishable from L. java by the different rostral formula and the presence of 4 (2+2) dorsal rostral teeth (vs 3 (2+1) in L. java); the presence of 4 or 5 pairs of dorsal spines on the telson (vs 3 in L. java); inner distal spines on the telson much shorter than those in L. java : fi g. 2e); shorter stylocerite of the basal antennular segment not reaching the distal margin of the segment (vs reaching the distal margin in L. java; Komai et al. 2019: fi g. 2b); distal part of penultimate article of maxilliped III (Fig. 4A) with fewer but more slender spines than in L. java : fi gs 2g, 3a); merus of pereiopod III armed with 6-9 movable teeth at the distal angle (vs a maximum of 5 in L. java); and the different number of lateral spines on the meri of pereiopods III-V (see Fig. 4F-G, I-K vs Komai et al. 2019: fi gs 3e-f).
Another very geographically distant species, L. laurentae, although rather poorly described, can be separated from L. sokhobio sp. nov. by the more slender distal part of the rostrum and its feebly marked dorsal and ventral armature (Wicksten 2010;Komai et al. 2012) in contrast to the rostrum of the new species, which has well-marked dorsal teeth and some extension in the distal part, with well-developed ventral teeth (Figs 3B, 5).
Lebbeus sokhobio sp. nov. can be separated from L. antarcticus by the more slender distal part of the rostrum (Nye et al. 2013b: fi g. 8b) (vs rostrum with some extension in the distal part, with well-developed ventral teeth in the new species; Figs 3B, 5) and the presence of 3 postrostral teeth (see Nye et al. 2013b: fi g. 8b) (vs 2 in the new species).
Lebbeus cristatus and L. formosus differ from the new species in having a more slender and short rostrum (Ahyong 2010: fi g. 1a-c;Chang et al. 2010: fi g. 4a-b), a different armature of the distal margin of the basal antennular segment (Ahyong 2010: fi g. 1d;Chang et al. 2010: fi g. 4a-b) and of the posterior margin of the telson (Ahyong 2010: fi g. 1g;Chang et al. 2010: fi g. 4e), and a smaller number of lateral spines on the meri of pereiopods III-V (Ahyong 2010: fi g. 1d-g;Chang et al. 2010: fi g. 5e, g-h).

Genetic differences
The intraspecifi c pairwise genetic distances (p-distances) within the studied population of Lebbeus sokhobio sp. nov. (n = 7) is 0.004 ±0.002 (d ±ES), which is rather low. Also, the genetic differences between specimens from different stations and the intraspecifi c differences among specimens from one station are very similar. Genetic p-distances between known species of the genus vary from 0.014 to 0.16 substitutions per 100 nucleotide positions (see Table 2), showing that the interspecifi c genetic differences of closely related species from different, sometimes very distant, regions of the World Ocean (e.g., L. antarcticus, L. virentova and L. sokhobio sp. nov.; see Fig. 7) are only slightly different from the intraspecifi c differences within the Kuril Basin of the Sea of Okhotsk. Unfortunately, much genetic data from genetic markers other than COI mtDNA are not currently available. However, it is very interesting that the genetically closest (= phylogenetically related) species among representatives of the genus Lebbeus are distributed most distantly -L. antarcticus from the Southern Ocean and L. virentova from the Caribbean (see Fig. 7). The genetic p-distances (Table 2) between these species are lower than previously documented for caridean shrimps (Knowlton et al. 1993;Knowlton & Weigt 1998;Hebert et al. 2003;Sites & Marshall 2004;Zakšek et al. 2007;Lefébure et al. 2006aLefébure et al. , 2006bMarin 2017).
At the same time, available barcoding data show that all of the deepest dwelling species belong to the same phylogenetic clade (see Fig. 7; 'Deep-sea Lebbeus' clade) showing a low level of divergence among species, whereas their species are very widely distributed. Similar small interspecifi c distances of about 1-2% are also known from very distantly living species of other deep-sea caridean shrimps (e.g., Mirocaris Vereshchaka, 1997 (Alvinocarididae); Shank et al. 1999;Vereshchaka et al. 2015; data from GenBank), as well as other deep-sea invertebrate taxa such as bivalve mollusks (e.g., Abyssogena Krylova et al., 2010 (Vesicomyidae); Liu & Zhang 2018) and octocorals (France & Hoover 2002). Low interspecifi c genetic differences in COI mtDNA were observed exclusively in deep-sea taxa, but, for example, not in all studied deep-sea caridean shrimps (e.g., Shank et al. 1999;Vereshchaka et al. 2015;Zhang et al. 2017). As suggested by France & Hoover (2002), possible explanations for such reduced rates of divergence include a lower rate of evolution for octocoral mitochondrial genomes (also supported by Shearer et al. 2002) and the presence of a gene, mtMSH, which may code for a mitochondrial DNA mismatch-repair system (Culligan et al. 2000). The purpose of this study is not to try to answer the question of why the interspecifi c distances of the deep-sea clade within the genus Lebbeus as so low, given the small amount of genetic data available, but it can be concluded that interactions (= gene fl ow) between populations from the Sea of Okhotsk, the Caribbean and Antarctica are more diffi cult to imagine than to assume the presence of some mechanism interfering with the standard rates of evolution in the COI mtDNA gene of deep-sea species. In addition, in closely related shallow-water taxa, such as the  Fig. 7. Phylogenetic relationships within Thoridae (Hippolyte Leach, 1814 as outgroup) based on COI mtDNA gene markers. Bootstrap support (> 50%) is shown by the numbers along the branches (BA/ML analysis). Line thickness is also correlated with support values. widely distributed and abundant Thor amboinensis (De Man, 1888) (Thoridae), geographic variability is well refl ected in genetic changes (Fig.7;Titus et al. 2018 for T. amboinensis). The Canadian clade of Lebbeus polaris (Sabine, 1824) (see Fig. 7) shows a higher degree of COI mtDNA variability than deep-sea Lebbeus species from different regions of the world. Perhaps the use of other gene markers will allow deep-sea species to be divided more clearly, using molecular genetic methods, but at the moment there is an insuffi cient amount of genetic information for comparison in international depositories (e.g., GenBank (NCBI) database).

Distribution
Lebbeus sokhobio sp. nov. is so far known only from the Kuril Basin of the Sea of Okhotsk and is probably endemic for this region in accordance with current knowledge of the limited geographical ranges of species in the genus Lebbeus (e.g., Hayashi 1992;Komai et al. 2004Komai et al. , 2012Komai 2015;Anosov et al. 2018). In the same bathymetric range along the neighboring Kuril-Kamchatka Trench and the adjacent abyssal plain of the northwestern Pacifi c, no specimens of this genus were collected, neither during earlier expeditions to the area or as the result of the more recent deep-sea trawling of the KuramBio I-II (Kuril-Kamchatka Biodiversity Studies) Expeditions (Brandt & Malyutina 2012;Brandt et al. 2016;Malyutina et al. 2018;pers. obs.).