Isopods of the superfamily Cryptoniscoidea, which belong to the suborder Epicaridea, are ectoparasites of various crustaceans [1]. Believed to be heteroxenous, they infect calanoid copepods during the pelagic juvenile stage before parasitising on its definitive host and feeds on host haemolymph and ovarian fluids [1]. Amphipods are the definitive hosts of cryptoniscoid isopods in the family Podasconidae. It consists of two genera and four described species, Podascon chevereuxi Giard & Bonnier, 1895, Po. dellavallei Giard & Bonnier, 1889, Po. haploopis Giard & Bonnier, 1895 and Parapodascon stebbingi (Giard & Bonnier, 1895) [2], typically attaching to the gills or oostegites of their amphipod hosts [3, 4]. Genus Podascon was first established for Po. dellavallei by Giard and Bonnier (1889) [5], based on parasitic specimens found from Ampelisca diadema (Ampeliscidae). The two congeneric species are also parasites of ampeliscid amphipods: Po. chevreuxi was found from A. spinimana, and Po. haploopis was from Haploops tubicola [3]. The description of the latter two species was accompanied by the establishment of the family Podasconidae [3]. On the other hand, Pa. stebbingi was originally described from Onisimus plautus of Uristidae, under the genus Podascon [3]. Shortly after, Hansen (1916) [4] established a new genus Parapodascon based on podasconid specimens which parasitised the uristid O. leucopis collected from 1618 m deep at the northeast of Iceland and north of the Faroe Islands, Denmark. Hansen [4] argued that Po. stebbingi described by Giard and Bonnier (1895) [3] exhibited a high degree of morphological similarity to the collected specimens, hence decided to assign it to the new genus. Additionally, an undescribed Parapodascon individual was discovered within the marsupium of the uristid species Anonyx nugax [4]. Barnard (1961) [6] later reported an unidentified female isopod parasite in the marsupium of Paroediceroides trepadora of Oedicerotidae, collected from 875 m of the Gulf of Panama. Lastly, Vader (1967) [7] documented isopod parasites infesting the uristid O. normani collected from Norwegian waters, with 14 out of 79 (17.7% infection rate) specimens carrying one or more podasconids, albeit no further identification was conducted for them.
Amphipods serve as hosts to a diverse array of parasites, ranging from microparasites to macroparasites, such as rotifers and crustaceans [8]. Reports of these parasites are abundant in freshwater and intertidal environments and rare in the deep sea, especially below abyssal depths (3500–6500 m). A description of a parasitic nematode from the body cavity of Hirondellea gigas of Hirondelleidae is one of the rare records of parasites of amphipods at the hadal depth [9]. Here, we report for the first time amphipods infested by parasitic isopods of Podasconidae from the abyssal and hadal (> 6500 m) depths, also documenting previously unreported amphipod hosts. Partial nucleotide sequences of the mitochondrial 16S and nuclear18S ribosomal DNAs (16S and 18S) are provided for all podasconid specimens and their phylogenetic positions are investigated, whilst those of the mitochondrial cytochrome c oxidase subunit I (COI) gene are provided for all amphipod hosts.
During the expedition KH-23-5 by R/V Hakuho-maru in September 2023, amphipods were sampled from abyssal (3500–6500 m) and hadal depths (>6500 m) of the Japan and the Kuril–Kamchatka Trenches (Fig. 1), using a beam trawl and an epibenthic sledge. All amphipods were briefly sorted and identified onboard and were fixed with 80% ethanol and kept in 99% ethanol. The samples were stored at 4°C until transferred to the Atmosphere and Ocean Research Institute, The University of Tokyo (AORI) for more detailed sorting and searching for amphipods infested by podasconids. The identification of amphipods followed Barnard (1961), Barnard and Karaman. (1991a; 1991b), and Shimomura and Tomikawa (2016) [5, 10–12]
Among the 2134 amphipods collected, with over 50 amphipod genera, four individuals (0.0019% infection rate) harboured a total of seven male podasconid parasites. The parasite familial assignment followed Williams and Boyko (2012) and Boyko et al. (2013) [1, 13]. Present male specimens showed no contradiction in the morphological characteristics with available descriptions [3, 4], for the familial assignment. The gross morphology of these parasites allowed us to assign them to three morphospecies. Five out of seven podasconid specimens (Podasconidae sp. A) were obtained from two females of Aristias sp. (Aristiidae), and the remaining podasconids, Podasconidae sp. B and sp. C were obtained from Byblisoides arcillis (Ampeliscidae, male) and Epimeria abyssalis (Epimeriidae, male). These host species were collected from stations F5 (4556–4565 m, 39º32.516'N; 143º51.305'E), F9 (6539–6523 m, 39º32.540'N; 144º09.524'E) and F14 (5804–5802 m, 39º27.983'N; 144º48.982'E), respectively (Fig. 1; Fig. 2). Two amphipod hosts are exclusively benthic species; B. arcillis possesses a silk gland for tube building and E. abyssalis is referred to a “reptantic” form, adapted for walking on the seafloor [14–15]; Aristias sp. is also considered a benthic species, as it is often found associated with other sessile invertebrates [16]. Aristias sp. obtained in this study was possibly dislodged from biotic substrates during a collection. Notably, one specimen of Aristias sp. was infested by four podasconids and the others were infested by one (Fig. 2). All podasconids were found attaching to the host tissue probably using the oral cone, antennule and first few pereopods (Fig. 2). They were attached to the gills of the host amphipods, except Podasconidae sp. B to the base of the pleopod rami of B. arcillis (Fig. 2).
These morphospecies were distinguished from each other by their body size and morphology of the uropod and article 1 of antennule. Podasconidae sp. B exhibited straight sympods compared to the slender sympods of uropods, which were narrower distally, of Podasconidae sp. A and sp. C (Fig. 3). Podasconidae sp. A exhibited serration on article 1 of antennule, bearing seven teeth, however, sp. B and sp. C exhibited no teeth. Although Podasconidae sp. A and sp. C were morphologically similar to each other, but sp. C is 31% and 62% longer in dorsal length to sp. A and sp. B, respectively (sp. A: 2.09 mm; sp. B: 1.04 mm; sp. C: 2.74 mm) (Fig. 3). These morphological discrepancies would warrant further investigation, to definitively determine if they represent different species. However, the identification of isopod parasites was left at the family level, due to the limited numbers of the specimens and the difficulties in distinguishing Podascon and Parapodascon, particularly from a male specimen [4]. In this study, we prioritized getting genetic data and detailed morphological analysis is required in future.
Total DNA was extracted, using a DNeasy Blood and Tissue Kit (QIAGEN, Germany), from whole podasconid specimens, except Podasconidae sp. C, in which a few appendages (pereopods and pleopods) were removed to extract DNA. The exoskeleton was mounted for a future morphological study after dissolving its content as post-extraction specimens with the whole body. The preparation was done using Hoyer’s medium. Total DNA was also extracted from four amphipod hosts, by removing two pleopods from the right side, except for E. abyssalis, where a single pleopod was adequate.
Partial nucleotide sequences of the mitochondrial 16S and nuclear 18S were determined for all podasconids. Those of the mitochondrial COI gene were determined for host amphipods. Primers used in this study are listed in Table 1. Amplifications of 16S were performed by using a DNA thermal cycler with reaction conditions following Lörz et al. (2018) [17] and amplifications of COI were by reaction conditions following Hou et al. (2007) [18], while these reactions were performed using TaKaRa ExTaq HS (TaKaRa Bio, Japan). For 18S, amplifications were performed with KOD FX Neo (TOYOBO Life Science, Japan) and reaction conditions were 94°C for 2 min; 45 cycles of 98°C for 10 s, 52°C for 30 s, and 68°C for 120 s; and 68°C for 3 min. All PCR products were diluted with double-distilled water by 10 folds and purified using Exo-SAP Express (Thermo Fisher Scientific, USA). Purified samples were used for sequencing reaction using a BigDye Terminator Cycle Sequence Kit v3.1 (Thermo Fisher Scientific, USA). For 18S, eight additional internal primers were used (Table 1). Products of sequencing reactions were further purified using a BigDye Xterminator Purification Kit (Thermo Fisher Scientific, USA) and the nucleotide sequences were determined using ABI 3500 XL DNA sequencer at the AORI. Obtained sequences were deposited in the DDBJ/GenBank/EMBL databases with accession numbers: Amphipods COI, LC823198–LC823201; Podasconidae 18S, LC823202–LC823208; Podasconidae 16S, LC823209–LC823215.
Table 1 List of primers used in this study
Primers
|
|
|
|
Sequence (5’-3’)
|
|
References
|
18S
|
18S-a1F
|
|
Forward
|
|
GGYGAAACCGYGAAWGGYTC
|
|
Kakui et al. (2011) [34]
|
|
18S-b3F
|
|
Forward
|
|
CCTGAGAAACGGCTACCACAT
|
|
Kakui and Shimada (2017) [35]
|
|
18S-b4F
|
|
Forward
|
|
TGCGGTTAAAAAGCTCGTAGTTG
|
|
Kakui et al. (2011) [34]
|
|
18S-b4R
|
|
Reverse
|
|
TCCAACTACGAGCTTTTTAACC
|
|
Kakui et al. (2011) [34]
|
|
18S-b5F
|
|
Forward
|
|
GATCGAAGGCGATYAGATACC
|
|
Kakui et al. (2011) [34]
|
|
18S-b6F
|
|
Forward
|
|
CCTGCGGCTTAATTTGACTC
|
|
Kakui et al. (2011) [34]
|
|
18S-a6R
|
|
Reverse
|
|
AACGGCCATGCACCAC
|
|
Kakui et al. (2011) [34]
|
|
18S-b8R
|
|
Reverse
|
|
TCTAAGGGCATCACAGACCTG
|
|
Kakui et al. (2011) [34]
|
|
18S-b8F
|
|
Forward
|
|
GGTCTGTGATGCCCTTAGATG
|
|
Kakui et al. (2011) [34]
|
|
18S-a9R
|
|
Reverse
|
|
CCTTGTTACGACTTTTAGTTCC
|
|
Kakui et al. (2011) [34]
|
16S
|
16SFt_amp
|
|
Forward
|
|
GCRGTATIYTRACYGTGCTAAGG
|
|
Lörz et al. (2018) [17]
|
|
16SRt_amp2
|
|
Reverse
|
|
CTGGCTTAAACCGRTYTGAACTC
|
|
Lörz et al. (2018) [17]
|
COI
|
LCO1490-JJ
|
|
Forward
|
|
CHACWAAYCATAAAGATATYGG
|
|
Astrin and Stüben, (2008) [36]
|
|
HCO2198-JJ
|
|
Reverse
|
|
AWACTTCVGGRTGVCCAAARAATCA
|
|
Astrin and Stüben, (2008) [36]
|
Seven 18S sequences of the present Podasconidae were aligned with those of 15 Bopyroidea, 4 other Cryptoniscoidea and 2 outgroup taxa, which were retrieved from the GenBank, using MAFFT ver. 7 [19] with the “Q-INS-I” strategy [20]. Alignment ambiguous sites were removed with Gblocks 0.91b [21] with a “relaxed” parameter [22], resulting in the final matrix of 28 taxa and 1483 sites. Phylogenetic analyses were performed using the Bayesian Inference approach using MrBayes 3.2.7 [23] under the GTR+G+I substitution model for 5000000 generations of four Markov chains with a sampling frequency of 100, and the first 25% as burn-in fractions. Alignment, masking of alignment ambiguous sites and the Bayesian phylogenetic analysis were done on NGPhylogeny.fr [24]. Additionally, the Maximum-Likelihood (ML) tree was built using the GTR+G+I model, determined by the ModelTest-NG [25] integrated in raxmlGUI 2.0 [26], estimating nodal support from bootstrap proportion values (BS) in 1000 pseudoreplicates.
The ML and Bayesian phylogenetic analyses for the 18S dataset have constructed trees that are almost identical in topology. Podasconidae sp. A and sp. C formed a robust clade [BS = 100%, Bayesian posterior probability (PP) = 1]. Podasconidae sp. B was sister to Cryptoniscoidea sp. which was found on an ostracod [13] with sufficient support values (76%, 0.98) (Fig. 4). The monophyly of these two clades was supported with the maximum BS and PP. The results of phylogenetic analyses therefore inferred the non-monophyletic nature of Podasconidae (Fig. 4). The pairwise p-distance for sequences of the seven present podasconids were then calculated based on both 18S (1891 sites after the removal of any gaps and missing sites) and 16S (421 sites), using MEGA11 [27]. The pairwise p-distance between individuals of Podasconidae sp. A was 0.00% for 18S, whilst it ranged between 0.24%–1.2% for 16S. Pairwise 18S p-distances among the three species were 6.50% (sp. A–sp. B), 2.64% (sp. A–sp. C) and 6.38% (sp. B–sp. C). Pairwise 18S p-distances of present podasconids and other cryptoniscoids were from 5.85% (sp. B–Cryptoniscoidea sp.) to 15.1% (sp. B–Zonophoryxus dodecapus) [13]. Pairwise 16S p-distances were 23.2%–24.2% (sp. A–sp. B), 47.8%–49.2% (sp. A–sp. C) and 48.8% (sp. B–sp. C).
Previously, the Podasconidae have been reported from shallower waters, up to the bathyal depths (200–3500 m) in the Northern Atlantic Ocean and Western Pacific Ocean. This study presents the first record of three podasconid species from three amphipod hosts (Aristias sp., B. arcillis and E. abyssalis) that have never been reported previously. Furthermore, this research significantly extends the depth record of Podasconidae to the hadal depths. Podasconids inhabiting the abyssal and hadal depths may be more speciose, as highly diverse amphipod taxa were collected from such depths and different species of podasconids parasitised different amphipods.
The observed morphological characters (Figs. 2, 3) and a close phylogenetic relationship with Cryptoniscoidea sp., which was found on an ostracod (Fig. 4), suggest Podasconidae sp. B belongs to a different genus to the sp. A or sp. C. Morphologically, sp. B showed a resemblance to male Parapodascon in the shape and length of the biramous uropod sympods, however, the attachment site and the number of teeth at the antennule basis of sp. B differed from species of Parapodascon or Podascon. Furthermore, distinguishing adult male cryptoniscoids and cryptoniscus larvae is challenging due to their similar morphology. This raises the possibility that the present podasconid specimens to be juvenile cryptoniscoid, especially sp. B. In that case, although speculative, it might be suggested that B. arcillis could serve as intermediate hosts. Additionally, cryptoniscoids were also found infesting the abdomen of cumaceans of families Bodotoridae and Leuconidae caught from abyssal depths of the present expedition (K. Okamoto, pers. comm.). Similarly, the German–Russian joint expedition KuramBio II has retrieved juveniles of cryptoniscoids attached to the Ascothoracidae (Dendrogastrida, Thestraca) and cumaceans, collected from the abyssal depth (5120–5741 m) of the Kuril–Kamchatka Trench [28]. In addition, they were juveniles, cryptoniscus larvae also were collected at a depth of 8738.9 m, which is the current deepest record of Epicaridea [28].
The penetration of parasites to the deepest of the trenches in the Northwest Pacific region appears to be a consistent pattern, as also observed in the digenean parasite [29]. The high surface productivity observed in the Japan and Kuril-Kamchatka Trenches [28, 30] might explain the successful penetration of parasites into the abyssal and hadal depths. Increased productivity supports a higher abundance and diversity of organisms, which creates a more favourable environment for parasites to encounter and infect suitable hosts. Such a convenient environment may have induced cryptoniscoids to diversify via host-switching among distantly related crustaceans or between different species of amphipods with varying lifestyles. Further sampling efforts in other productive hadal zones, like the Atacama Trench [31], could provide crucial insights into large-scale patterns of deep-sea parasitism and how these interactions influence biodiversity in extreme environments.
This study provided 16S and 18S ribosomal DNA sequences of the Podasconidae collected from abyssal and hadal oceans, confirming the diversity of these epicaridean isopods in the abyssal and hadal environments. The present study indicated podasconid parasites of amphipods are non-monophyletic and accumulation of DNA sequences and the Cryptoniscoidea from other crustacean groups are worthy of investigation to untangle the phylogenetic relationship of cryptoniscoid families. Moreover, studying the interaction of these parasitic isopods with the other Peracarida could reveal new interactions in the deep sea.