An integrative redescription of Echiniscus quadrispinosus quadrispinosus Richters, 1902 (Heterotardigrada, Echiniscidae) from the terra typica in Taunus Mountain Range (Europe; Germany)

In the present study, we used an integrative taxonomy approach to redescribe a population of Echiniscus quadrispinosus quadrispinosus Richters, 1902 from the neotype locality in the Taunus Mountain Range (Germany). We found clear differences in the chaetotaxy formula between the life stages of E. q. quadrispinosus. The body appendages B are, in general, absent in juveniles. Moreover, in larvae all body lateral appendages, except for E, are absent. We also obtained DNA sequences of 18S rRNA, 28S rRNA, ITS-2, and COI of E. q. quadrispinosus from the neotype locality and three Norwegian populations. Comparison with the sequences available in GenBank showed low genetic differences between the neotypic population and specimens from other localities. Therefore, we decided to establish our specimens from Taunus Mountain Range as neotype and paraneotypes of E. q. quadrispinosus. We also discussed and amended the taxonomic status of three subspecies E. q. brachyspinosus Bartoš, 1934, E. q. cribrosus Murray, 1907 and E. q. fissispinosus Murray, 1907 and established them as junior KACZMAREK Ł. et al., Integrative redescription of Echiniscus q. quadrispinosus 103 synonyms of E. q. quadrispinosus. Finally, we also confirmed E. lichenorum Maucci, 1983 as a valid species, clearly different from E. q. quadrispinosus.


Introduction
Tardigrades inhabit terrestrial and aquatic (freshwater and marine) environments from the highest mountains to the deepest oceans, from the polar regions to the tropics. They are found in mosses, lichens, soil, leaf litter, sediments and on aquatic plants (Nelson et al. 2020). Up to now, ca 1400 species of tardigrades have been described across the world (Guidetti & Bertolani 2005;Degma & Guidetti 2007;Degma et al. 2009Degma et al. -2021.
The genus Echiniscus C.A.S. Schultze, 1840 (amended by Gąsiorek et al. 2017) is characterized mainly by red eyes, a rigid buccal tube without stylet supports or with fine, fibrillar stylet supports, two pairs of paired plates, two or three median plates (sometimes transversally subdivided), notches on the terminal plate, pseudosegmental plates absent and ventral plates sometimes present. To date, 120 species and subspecies have been attributed to this genus, which makes it one of the most species-rich tardigrade genera (Degma et al. 2009(Degma et al. -2021. Echiniscus q. quadrispinosus Richters, 1902 was described from the Taunus Mountain Range (Germany) without either detailed diagnosis or specified type locality. Not long after, two varieties, i.e., E. q. cribrosus Murray, 1907 andE. q. fissispinosus Murray, 1907, were described from the Shetland Islands and Scotland, respectively (Murray 1907). Finally, Bartoš (1934) described a third variety E. q. brachyspinosus Bartoš, 1934 from the Carpathians. Ramazzotti & Maucci (1983) elevated E. q. cribrosus and E. q. fissispinosus to the subspecies level and suggested E. q. brachyspinosus to be a different form only. All these taxa differ from typical E. q. quadrispinosus by rather doubtful morphological characters. Gąsiorek et al. (2019a) included E. q. quadrispinosus into the E. quadrispinosus subgroup characterized by the presence of "plates with dominant circular pores, trunk appendages in the form of filiform cirri and additional spines". Two other species included in this group are E. lentiferus Claxton &Dastych, 2017 andE. lichenorum Maucci, 1983 nom. inq. (suggested as conspecific with E. q. quadrispinosus) (Gąsiorek et al. 2019a).

Morphometrics and morphological nomenclature
All measurements are given in micrometres (μm). Structures were measured only if their orientation was suitable. Body length was measured from the anterior extremity to the end of the body, excluding the hind legs. Terminology used for the description of dorsal plates follows Kristensen (1987). Lengths of the claws were measured from the base of the claw (in its middle point) to its top. The sp index is the ratio of the length of a given structure to the length of the scapular plate expressed as a percentage (length of structure × 100 ⁄ length scapular plate) (Dastych 1999, and later independently proposed as the psc index by Fontoura & Morais 2011). Configuration and arrangement of body appendages (chaetotaxy) is given according to Gąsiorek et al. (2017).

Species identification
Specimens were identified using the taxonomic key and morphological description in Ramazzotti & Maucci (1983) and compared with the original description (Richters 1902).

Comparative molecular analysis
The obtained sequences were checked for quality and manually aligned in BioEdit ver. 7.2.5 (Hall 1999). The COI haplotypes were retrieved using the DnaSP ver. 5.10.01 software (Librado & Rozas 2009). All the COI haplotypes were translated into amino acid sequences using the EMBOSS-TRANSEQ application (Rice et al. 2000;Goujon et al. 2010) to check for internal stop codons and indels. The translation was successfully carried out with the invertebrate mitochondrial codon table and the -2 th reading frame.
Basic Local Alignment Search Tool (BLAST, Altschul et al. 1990) searches with sequences deposited in the NCBI database were performed to verify the identity and homology of the amplified nuclear and mitochondrial barcode sequences. For molecular comparisons, single sequences of species belonging to the genus Echiniscus C.A.S. Schultze, 1840 were downloaded and aligned using the ClustalW Multiple Alignment tool (Thompson et al. 1994) implemented in BioEdit ver. 7.2.5. Only the available GenBank sequences that coincided with nrDNA and mtDNA fragments obtained in our study have been applied. Alignment sequences were trimmed to 685, 672, 276, and 510 bp for 18S rRNA (23 species), 28S rRNA (20 species), ITS2 (11 species) and COI (21 species) molecular markers, respectively. Calculation for the uncorrected p-distances was performed using the software MEGA X (Kumar et al. 2018). Uncorrected pairwise distances are provided as supplementary materials (Supp. file 1).

Remarks
Animals were mounted on microscope slides in Hoyer's medium, 45 animals prepared for SEM, and 15 prepared for molecular analyses. However, DNA sequences were obtained from two females and one male only (exoskeleton, No GR5/S, GR8/S and GR10/S) which was later mounted on microscope slide in Hoyer's medium.

Description of the neotypic population
Females (measurements and statistics in Table 1, Supp. file 2 and Figs 1-4) Body orange in live specimens and transparent/light yellow after preparation. Eyes red. Apart from the head appendages which include internal and external cirri and cylindrical cephalic papillae (secondary clava), only appendage A with clava (primary clava) near its base is present (Fig. 1D). Dorsal and lateral appendages in the shape of short, and long filaments and/or spines are present at positions A-B-C-C d -D-D d -E ( Fig. 1A-C). However, appendage B may be absent, at least on one side of the body, for more details see Morphological variability below.
Dorsal plates well developed. Head and scapular plates not faceted. Under PCM, lateral portions of the scapular plate appear to be detached from the dorsal plate, forming small additional plates (one on each side of the body) (Fig. 1E, arrowheads) divided from the lateral margin of the scapular plate by a thin bright stripe. This division is formed by the narrow stripe without pores, as it is clearly visible in SEM along with a small additional plate (Fig. 1C, arrowhead). Paired plates I and II are divided into two parts -narrow anterior part and a wider posterior part -by smooth stripe without sculpture ( Fig. 2A, arrows). Anterior parts most often divided longitudinally into two parts ( Fig. 2A-B, indented arrowheads). Median plate 1 and 2 divided into anterior part ( Fig. 2A, asterisks) and posterior part ( Fig. 2A, filled arrowhead; which is especially visible in lateral position), m3 undivided ( Fig. 2A, empty arrow). The terminal plate with two notches ( Fig. 2A, empty arrowhead).
All dorsal plates, covered with double sculpture under PCM ( Fig. 2A-E), i.e., regular polygonal or roundish black 'granules', i.e., endocuticular pillars (0.6-1.0 µm in diameter on scapular plate and similar in size also on other plates) and white roundish or oval pores which especially in larger specimens may be merged (seen as white spots, 1.0-4.2 µm in diameter on scapular plate and similar in size also on other plates); but see below for more details. Pores absent on anterior parts of m1 and m2 ( Fig. 2A, asterisks), with typical double sculpture present on the posterior parts ( Fig. 2A, filled arrowheads). Central part of terminal and scapular plates with cross-like pattern, and on terminal plate a transverse line of the cross is an extension of notches ( Fig. 2C-E). Sometimes, on the scapular plate, the transverse line is absent and then only the longitudinal line visible in the middle of this plate or an additional transverse line is present and the plate is poorly divided into six rectangles ( Fig. 2C-D). Under SEM the plates with regularly distributed pores (Fig. 2F), which means that where the white pores visible under PCM are absent (i.e., neck plate, lateral portions of the scapular plate and anterior parts of paired plates) the plates appear to be smooth or with poorly visible granulation (Fig. 2F).
Two poorly marked ventral rectangular plates, arranged transversally, are present below the head ( Fig. 3A, C, arrowheads). Another two rounded square plates are present on lateral sides of the gonopore ( Fig. 3B-C, arrows). Ventral cuticle possesses tiny and regular granulation (due to the presence of dense endocuticular pillars). Granulation is a little larger on the plates around the gonophore (0.3-0.4 µm diameter) than in other parts of the ventral cuticle, 0.1-0.2 µm diameter) (Fig. 3D).
Outer cuticle of legs I-III with clearly visible stripes of tiny and regular granulation (0.1-0.4 µm in diameter): a thin frontal stripe on the upper part of the leg (Fig. 4A, empty arrow), a wide stripe in the central part of the leg covering frontal and lateral side of the leg (Fig. 4A, empty arrowhead) and the most distal, thin stripe above claws on the ventral side of the leg (Fig. 4A, filled indented arrowhead); white pores absent. On legs IV only frontal stripe on the upper part of the leg (just above the plate with dentate collar) (Fig. 4C, E, empty arrows) and thin stripe above claws on the ventral side of the leg are present. Triangular spine on leg I ( Fig. 4B-C, filled indented arrowhead) and finger-like papilla on leg IV, present ( Fig. 4D-E, filled arrow). Legs IV with dentate collar with seven to ten sharp, triangular teeth and the plate with the same sculpture as dorsal plates ( Fig. 4D-E). External claws of all legs I-IV smooth, internal with spurs directed downwards ( Fig. 4C-F)). The gonopore with the typical six-petal rosette ( Fig. 3C, asterisk).  Table 2, Supp. file 2 and Fig. 5) Males are, in general, similar to females in the morphology of plates and chaetotaxy (Fig. 5A, C-D). However, there are differences in the lengths of some morphological structures (especially slightly shorter head appendages, i.e., cirri internal and external and body appendages) (compare values in Tables 1-2). On the dorsal side, regular polygonal or roundish black 'granules' 0.3-0.9 µm in diameter (on scapular plate and similar in size also on other plates) and white roundish or oval pores 1.0-3.1 µm   in diameter (on scapular plate and similar in size also on other plates). Gonopore round and without the six-petal rosette (Fig. 5B). Table 3, Supp. file 2 and Fig. 6) In general, juveniles were similar to adults of both sexes in the morphology of plates (Fig. 6A). However, there are differences in chaetotaxy (Fig. 6A-D), in general, lack of filaments B. Moreover, appendages are shorter in juveniles than in adult females (compare values in Tables 1-3) (Fig. 6A-D). On the dorsal side, regular polygonal or roundish black 'granules' 0.4-0.9 µm in diameter (on scapular plate and similar in size also on other plates) and white roundish or oval pores 1.0-2.7 µm in diameter (on scapular plate and similar in size also on other plates). Gonopore absent. Table 4, Supp. file 2 and Fig. 7G-H) All head and body appendages much shorter than in adults and juveniles (compare values in Tables 1-4) (Fig. 7G-H). Moreover, morphology and sculpture of plates are also different. Anterior parts of the paired plates not divided. Although, sculpture on dorsal plates composed of regular polygonal or roundish black 'granules' and white roundish or oval pores as in adults and juveniles. However, on head plate, anterior parts of paired plates and on entire median plate m3, pores completely absent and black 'granules' poorly marked. On scapular plate pores are distributed mainly on the margins of the plate and almost absent in the centre. On the dorsal side, regular polygonal or roundish black 'granules' 0.4-0.9 µm in diameter (on scapular plate) and white roundish or oval pores 0.7-1.4 µm in diameter (on    (Fig. 7G). Gonopore absent (Fig. 7H).

Eggs
Smooth, light orange and deposited in the exuviae up to 6 in one exuvium.

DNA sequences
We obtained good quality sequences for the applied molecular markers:

Morphological variability
A strict chaetotaxy was analysed in 36 females, 23 males, 35 juveniles, 3 larvae. In all adult specimens, both females and males typical chaetotaxy, i.e., A-B-C-C d -D-D d -E was observed (Fig. 1A-B), whereas, in juveniles appendages B are most often absent (chaetotaxy: A-C-C d -D-D d -E). The other dorsal and lateral appendages were in general shorter in juveniles than in adults. In all studied larvae, chaetotaxy was always A-C d -D d -E and all appendages were much shorter than in juveniles and adults (compare values in Tables 1-4). Moreover, some modifications in chaetotaxy were observed in juveniles and adults. Richters, 1902, ♀. A*. Leg I outer cuticle with clearly visible stripes of tiny and regular granulation: a thin frontal stripe on the upper part of the leg (empty arrow), a wide stripe in the central part of the leg covering frontal and lateral side of the leg (empty arrowhead) and the most distal, thin stripe above claws on the ventral side of the leg (filled indented arrowhead) (PCM). B. Spine on leg I (arrowhead) (PCM). C. Spine on leg I (arrowhead) and thin frontal stripe on the upper part of the leg (empty arrow) (SEM). D*. Claws IV with dentate collar and finger-like papilla (filled arrow) (PCM). E. Vlaws IV with dentate collar and finger-like papilla (filled arrow); empty arrow indicates thin frontal stripe on the upper part of the leg (SEM). F. Claws of the II leg (PCM). * = manually assembled deep-focus image. Scale bars in micrometres (μm).

Fig. 4. Echiniscus quadrispinosus
In two juveniles appendages B were present on both sides of the body and in five juveniles appendages B were present only on one side of the body (chaetotaxy: A-B-C-C d -D-D d -E) (Fig. 6A-B). Moreover, in one juvenile appendages C d and D d were present only on one side of the body, and in another one D d was present on both sides and C d only on one side (chaetotaxy: A-C-C d -D-D d -E) (Fig. 6C-D).
In two females appendages B were present only on one side of the body (chaetotaxy: A-B-C-C d -D-D d -E) (Fig. 7A). Two females had additional small spines near the base of normally developed appendages B (chaetotaxy: A-B-C-C d -D-D d -E) (Fig. 7B-D). In other two females only appendage C d on one side of the body was present and appendages D d were completely absent (chaetotaxy: A-B-C-C d -D-E) (Fig. 7E-F

Based on these assumptions chaetotaxy formula for adults and juveniles of this species is in general A-(B)-C-C d -D-D d -E and for larvae A-C d -D d -E.
The other observed aberrations in chaetotaxy are only accidental.

Genetic variability
The obtained eight COI sequences (GenBank accession numbers: MZ798397-MZ798404) of E. q. quadrispinosus consisted of four COI haplotypes. Haplotype 1 was found in the Norwegian (169/7 sequence, population code: 169) and German (GR8 and GR10 sequences, population code: GR) populations whereas haplotypes 2, 3 and 4 were identified in different Norwegian populations (haplotype 2 -169/8 sequence, population code: 169; haplotype 3 -184/3 and 184/8 sequences, population code: 184; haplotype 4 -187/4 and 187/7 sequences, population code: 187). The value of uncorrected genetic p-distances between obtained COI haplotypes ranged from 0.2% to 0.8%. In turn, the analysis of the p-distances between E. q. quadrispinosus and compared 20 taxa of the genus Echiniscus ranged  from the most similar 1.2% for E. quadrispinosus (GenBank accession number: JX683821, Vincente et al. 2013) to the least similar 21.8% for E. tantulus Gąsiorek, Bochnak, Vončina & Kristensen, 2020(GenBank accession number: MT107427, Bochnak et al. 2020, with an average p-distance of 14.4%. In the conservative 18S rRNA gene fragment we observed no differences between our eight sequences from the German and Norwegian populations (GenBank accession numbers: MZ798389-MZ798396) and sequences of E. quadrispinosus deposited in NCBI (GenBank accession number: MK529684). In turn, the uncorrected genetic p-distances between the other 21 taxa of the genus Echiniscus showed that the least similar was E. belloporus Gąsiorek & Kristensen, 2018(GenBank accession number: MK529674, Gąsiorek et al. 2019a) with a genetic distance value of 3.1% and an average p-distance was 1.4%.
The analysis of the p-distances between our eight sequences of 28S rRNA from the German and Norwegian populations (GenBank accession numbers: MZ816972-MZ816979; two groups of sequences, i.e., the first consisted of GR8, GR10,169/8,169/9,184/3,184/8,187/1 sequences and the secondone 187/8 sequence) indicated that the genetic distance was 1%. Comparison with other 19 taxa of the genus Echiniscus, for which GenBank sequences are available, are as follows: the most similar was E. quadrispinosus (GenBank accession number: MK529714, Gąsiorek et al. 2019a) with 1% value of the p-distance and the least similar was E. belloporus Gąsiorek & Kristensen, 2018(GenBank accession numbers: MK529702, Gąsiorek et al. 2019a) -5.4%, with an average p-distance of 2.5%.
No genetic differences were observed between our eight ITS2 sequences from the German and Norwegian populations (GenBank accession numbers: MZ816980-MZ816987). The ranges of uncorrected genetic p-distances between our sequences and the other 10 species of the genus Echiniscus indicated that the most similar was E. virginicus Riggin, 1962 (GenBank accession number: MN545756, Gąsiorek et al. 2019b) -0.42% and the least similar was E. blumi Richters, 1903(GenBank accession number: EF620383, Jørgensen et al. 2007) -34.5%, with an average p-distance of 21.8%. There were no available ITS2 sequences of E. quadrispinosus in the GenBank database.

Establishing of the neotype and paraneotypes of E. q. quadrispinosus
The search for the type material of E. q. quadrispinosus in various collections did not bring positive results. We can probably assume that the type material of E. q. quadrispinosus no longer exists. Taking into consideration that accurate diagnoses of the species were poorly provided in the past, it is necessary to establish a neotype series of this species. For this reason, we designated the neotype and 108 paraneotypes of E. q. quadrispinosus which agree with the original description and were collected in the terra typica in the Taunus Mountain Range (Germany). The neotype series was deposited at the Department of Animal Taxonomy and Ecology, Adam Mickiewicz University in Poznań and Institute of Systematics and Evolution of Animals, Polish Academy of Sciences, Poland. All the above-mentioned statements are in accordance with the International Commission on Zoological Nomenclature (ICZN) acts dedicated to the establishment of neotype series.

Discussion
In our study, we used integrative taxonomy to describe E. q. quadrispinosus specimens from its terra typica -Taunus Mountain Range (Germany). Our analysis has shown that our specimens are morphologically compatible with specimens described by Richters (1902). However, in the studied population we found morphological differences between adults and juveniles in chaetotaxy.
Intraspecific and interspecific variability of Echiniscidae has been studied for many years and is still problematic. However, despite the well-documented intraspecific variability in genera such as Echiniscus, Mopsechiniscus du Bois-Reymond Marcus, 1944, or Pseudechiniscus Thulin, 1911, even in more recently published species descriptions concerning the variability, in relation to life stage or sex, was ignored (for a review see, e.g., Bartylak et al. 2019).
We analysed separately females, males, juveniles and larvae of E. q. quadrispinosus to check possible differences in chaetotaxy between different life stages and sex of the animals. We found a clear pattern related to the presence or absence of some filaments in different life stages, but not between sexes. This is similar to the E. tristis Gąsiorek & Kristensen, 2018, for which such differences in chaetotaxy were found to be connected to the life stages or sex of the animals (Bartylak et al. 2019). Such results underline how important it is to analyse a large number of specimens from any given species to avoid incorrect species identification or wrong/incomplete description of a new species, subspecies or forms as was done for E. q. quadrispinosus.
Apart from the nominal E. q. quadrispinosus, three other subspecies are known: E. q. brachyspinosus, E. q. cribrosus and E. q. fissispinosus. Echiniscus q. brachyspinosus was proposed as a variety only by Ramazzotti & Maucci (1983) who suggested that the described differences (body appendages in the form of short, wide spines in E. q. brachyspinosus instead of filaments in nominal species) should be rather considered as species variability and the difference between young and adult specimens. This is additionally confirmed by the fact that such spines were found only in small specimens which were part of the population of the typical E. q. quadrispinosus. Moreover, also Cuénot (1932) observed short spines C and D in small specimens of E. q. quadrispinosus. The same was observed by us in the present study and, what is more, such variability seems to be frequent in species of Echiniscus (e.g., Guil 2008; Bartylak et al. 2019). We think that in such a situation this form should be considered as young specimens of E. q. quadrispinosus. Echiniscus q. cribrosus and E. q. fissispinosus differ from the nominal species by the absence of 'supplementary plates', which are in fact separated parts of paired plates I and II in E. q. quadrispinosus. Such pseudoplates are often present in other species of Echiniscus (see e.g., Claxton 1996;Kaczmarek & Michalczyk 2010;Claxton & Dastych 2017;Gąsiorek & Kristensen 2018;Bartylak et al. 2019;Gąsiorek & Vončina 2019;Gąsiorek & Michalczyk 2020b;Gąsiorek et al. 2020;Kiosya et al. 2021). Additionally, some differences were also reported in chaetotaxy (e.g., lacking of appendages B, reduced or absent lateral or dorsal appendages). Moreover, in E. q. fissispinosus lateral appendages D are doubled on one or both sides of the body. Since the visibility of 'supplementary plates' may depend on slide preparation or size of the specimens, differences in chaetotaxy (number of appendages, their shape and duplication) are typical in species of Echiniscus (for a review see, e.g., Guil 2008; Bartylak et al. 2019), we suggest that E. q. cribrosus should be considered a typical E. q. quadrispinosus.
Moreover, Gąsiorek et al. (2019a) suggested E. lichenorum as conspecific with E. q. quadrispinosus and proposed for this species a taxonomic status of nomen inquirendum. However, although we agree that dorsal plate sculpture in E. lichenorum is very similar or even identical as in E. q. quadrispinosus, both species are easy to differentiate because of the presence of appendages E in E. q. quadrispinosus. Based on this assumption, although we did not examine the type material of E. lichenorum, we think that this species should be considered as valid, even if some morphological details are not listed in the original description.
Echiniscus q. quadrispinosus is reported from many localities throughout the world, most of them in the Holarctic (McInnes 1994;Meyer 2013;Kaczmarek et al. 2016;McInnes et al. 2017). Due to its single reports from New Zealand, Central and South America, McInnes (1994) considered this species as "probably cosmopolitan". Although this is possible (see, e.g., the distribution of E. testudo by Gąsiorek et al. 2019c), in our opinion reports outside Holarctic need a confirmation.
Summarizing, we propose E. lichenorum as a valid species and the subspecies E. q. brachyspinosus, E. q. cribrosus and E. q. fissispinosus as part of the nominal E. q. quadrispinosus, meaning that the nomenclature of the nominal species needs to be changed to E. quadrispinosus.