New record of Nausithoe werneri (Scyphozoa, Coronatae, Nausithoidae) from the Brazilian coast and a new synonymy for Nausithoe maculata

Abstract The order Coronatae (Scyphozoa) includes six families, of which Nausithoidae Haeckel, 1880 is the most diverse with 26 species. Along the Brazilian coast, three species of the genus Nausithoe Kölliker, 1853 have been recorded: Nausithoe atlantica Broch, 1914, Nausithoe punctata Kölliker, 1853, and Nausithoe aurea Silveira & Morandini, 1997. Living polyps (n = 9) of an unidentified nausithoid were collected in September 2002 off Arraial do Cabo (Rio de Janeiro, southeastern Brazil) at a depth of 227 m, and have been kept in culture since then. We compared these specimens with three species cultured in our laboratory: Nausithoe aurea (from Ilhabela, São Paulo, Brazil), Nausithoe maculata Jarms, 1990 (from Cuba and Puerto Rico), and Nausithoe werneri Jarms, 1990 (from the Atlantic Ocean off Morocco and from the Mediterranean Sea). The criteria used for comparison were: main aspects of the morphology, life cycle, and DNA sequences (18S, 28S, and COI). The results indicate that the unidentified polyps belong to N. werneri. Furthermore, N. aurea is considered a junior synonym of N. maculata.

The high morphological similarity among polyps in this order and the difficulty in relating the polyp to the medusa (as they are usually not collected together) have led to two classification systems. Polyps were historically identified only as "Coronatae polyps" or were classified as belonging to the genus Stephanoscyphus Allman, 1874(later changed to Stephanoscyphistoma Jarms, 1990, to accommodate these specimens).
been described or recorded: the medusae of Nausithoe atlantica Broch, 1913 andNausithoe punctata Kölliker, 1853, and the polyp of Nausithoe aurea Silveira & Morandini, 1997(Oliveira et al. 2016. The purpose of this study was to identify coronate polyps from the Brazilian continental slope off Arraial do Cabo (southeastern Brazil, western South Atlantic) and to compare them with the other known species from the Atlantic Ocean and the Brazilian coast, and thus, to improve our knowledge of coronate biodiversity and distribution.

Materials and methods
Deep-sea specimens from Brazil were collected during a cruise of the Navio Oceanográfico Prof. Wladimir Besnard off the coast of Arraial do Cabo (Rio de Janeiro state, southeastern Brazil, 23°45.80'S, 41°44.40'W) on 15 September 2002, at a depth of 227 m. Calcareous substrates were collected using a box-corer, and the polyps were found on these. The collected polyps were provisionally named Nausithoe sp. and numbered as follows: AC01, AC02, AC08, AC10, AC17, AC18, and AC20 (Table 2). They were kept alive in small dishes for a few days aboard ship (under room temperature 20-23 °C) and brought to the laboratory on 18 September 2002. Since their establishment in culture, three polyps (AC01, AC02, and AC10) reproduced asexually by forming tissue balls from the ephyrae (Fig. 1A, B), consistent with the observations of Silveira et al. (2003).
The study was conducted at the Laboratório de Cultivo e Estudos de Cnidaria, in the Zoology Department of the Biosciences Institute, University of São Paulo (IB-USP). For comparison, we used living polyps of several species: nine Nausithoe sp., nine Nausithoe werneri Jarms, 1990, 14 Nausithoe maculata Jarms, 1990, and three Nausithoe aurea (Table 1). They were selected due to their availability in culture at our laboratory and for their distribution in the Atlantic Ocean, which we considered important for comparison. Table 1. Data from studied species of Nausithose (Nausithoe sp., N. werneri, N. maculata, and N. maculata (= N. aurea Jarms (1990) and Jarms et al. (2002). Specimens of Nausithoe sp. were measured twice, in 2002 and 2018 (except for tissue balls, measured only in 2018). Da = diameter at aperture, Dbd = diameter of basal disc, Db = diameter just above basal disc, D 2 mm = diameter at 2 mm height, D 5 mm = diameter at 5 mm height, L tot = total length, Nwt = number of whorls of internal cusps, Nw = number of cusps per whorl, -= not measured. Voucher specimens of Nausithoe sp. were deposited in the Museu de Zoologia da Universidade de São Paulo as: MZUSP8502 (one medusa), MZUSP8503 (two polyps), MZUSP8504 (50+ ephyrae).

Cultivation
All polyps were fed with 1-2-day-old nauplii of Artemia sp. once a week. Strobilation usually produced hundreds of ephyrae (Fig. 1E), and excessive feeding to increase the number of jellyfish was not necessary. Polyps and ephyrae were kept separately in 200-mL acrylic containers (maximum of 20 ephyrae per dish) in incubation chambers at the appropriate temperature for each species (N. werneri and Nausithoe sp. at 15 °C; N. maculata and N. aurea at 20 °C). Once ephyrae were released, they were fed daily with macerated mussel gonads (Perna perna). As soon as they grew large enough to catch Artemia, we began to vary their diet, adding 1-day-old nauplii. Food was provided in abundance. The water in the polyp dishes was changed 1 day after feeding, and in the medusae and ephyrae containers, it was changed 1 hour after feeding (as these stages are more responsive to variations in water quality). Measurements and photographs of all stages were taken for comparison using a Nikon SMZ800 stereomicroscope and a Nikon Eclipse 80i microscope.

Morphological analyses
Measurements followed the protocols established by Jarms (1990Jarms ( , 1991 and Jarms et al. (2002). The main measurements were: i) total tube length, ii) tube diameter at 2 mm above the base, iii) tube diameter at 5 mm above the base, iv) diameter of the basal disc, v) diameter just above the basal disc, and vi) aperture diameter. The type of external ornamentation (number of transverse rings per 4-mm length) and whether these rings were more or less prominent with respect to an imaginary line tangential to the tube contour was noted for possible comparison (according to Morandini and Jarms 2005). For the more translucent tubes, we were able to observe the number of whorls of the cusps and the number of cusps per whorl. We also took scanning electron microscope (SEM) photographs of the internal cusps of some polyps (Nausithoe sp. and N. werneri) in order to observe their shape and ornamentation (with special attention to whether they had additional cusps on the margin and surface). Characters and measurements of the ephyrae and medusae are shown in Fig. 2. The SEM observations were conducted at the Laboratório de Biologia Celular e Microscopia of the IB-USP, and the method of preparing the specimens followed the protocols of Morandini and Jarms (2005, 2010.

Cnidome analyses
Capsule types and sizes of nematocysts in the different life-cycle stages (polyp, ephyra, medusa) were measured (Mariscal 1974;Östman 2000) using a Nikon Eclipse 80i light microscope. A total of 60 measurements were performed on each type per lifecycle stage of Nausithoe sp. and on the medusa tentacles of N. werneri and N. maculata.

DNA analyses
Because hundreds of ephyrae were produced, most of the molecular tissue for DNA extractions came from them (and some from mature medusae). We extracted DNA from specimens representing all the putative species in this study: Nausithoe sp., Nausithoe aurea, Nausithoe maculata, and Nausithoe werneri. DNA was extracted using an ammonium acetate-based protocol adapted from Fetzner (1999). Preliminary tests established that the minimum number of ephyrae for a satisfactory extraction was around 30, considering their size. Three partial genes were amplified by polymerase chain reaction (PCR): mitochondrial protein coding cytochrome oxidase subunit I (COI), and nuclear ribosomal markers 18S and 28S, using published primers and PCR conditions (Table 3). PCR products were purified with the Agencourt AMPure kit (#A63881). The BigDye reaction was conducted using the same primers and annealing temperature in each case. These steps were performed in the Laboratório de Evolução Molecular of the Zoology Department (IB-USP). Finally, the precipitated DNA were sequenced at the Laboratório de Sinalização de Redes Regulatórias de Plantas at the Botanical Department, IB-USP, with a Hitachi 3730xl DNA Analyzer. Chromatograms (.abi files) and consensus sequences were created and checked to identify potential sequencing errors and/or contamination, using Geneious software 9.1 (Kearse et al. 2012; all COI sequences were translated to check potential mitochondrial pseudogenes (NUMTs) or indel artifacts -genetic code: Invertebrate Mitochondrial). Sequence alignments were performed with the Geneious 9.1 MAFFT module for 18S and 28S markers (auto-mode; Katoh and Standley 2013), and Translation Align module for COI marker (genetic code: Invertebrate Mitochondrial; MAFFT alignment: E-INS-i); this way, we avoid inserting any spurious gap in the COI alignment. To avoid any difference (artifact) on COI distances (the most relevant dataset to discuss species delimitation), we edited ends of COI alignment. Doing so, all sequences present the same basic information (sequence length). Corrected distance values from each molecular marker alignment were obtained using MEGA X -Kimura 2 Parameters (Kumar et al. 2018).

Results
All the morphological features observed on the available specimens are summarized in Tables 2, 4. The time periods until medusae developed distinct morphological characters are listed in Table 5. Metagenetic species. Solitary polyp with typical periderm tube, dark to light brown (Fig. 1C, D), conical, with transverse rings on the surface with longitudinal striations (5 rings every 0.4 mm at 2 mm above base). Polyps 5.05-20.2 mm long; basal disk 0.63-0.66 mm in diameter; diameter just above the basal disk 0.14-0.4 mm; tube diameter at 2-mm height 0.09-0.45 mm, and at 5-mm height 0.35-0.45 mm; tube aperture diameter 1.28 mm. Tubes with 3-10 whorls of internal cusps (Fig. 1C); closer to base, number of internal cusps per whorl is 16: 4 large (perradius) with additional cusps on the surface, 4 intermediate (interradius), and 8 small (adradius) (Fig. 3A).

Systematics section
Upper whorls with only 8 cusps: 4 large (perradius) with no additional cusps, and 4 intermediate (interradius) (Fig. 3B, C). Polyps with 26-37 filiform tentacles (Fig. 1D). Polydisc strobilation, with more than 100 ephyrae at a time. Medusa (Figs 2A, B, 4) entirely translucent, with slightly flattened smooth umbrella; 16 slightly elongated lappets with rounded margins; 8 rhopalia with statocyst and red ocelli. Live specimens measuring up to: 9.5 mm total diameter; 7.74 mm diameter between opposite rhopalia; 2.83 mm coronal groove diameter; gastric cirri approximately 0.9 mm in length; and tentacle length up to half the total diameter of the medusa. Stomach with 4 gastric septa, each with 2 gastric filaments (8 in total) (Figs 2A, 4D, E). Gonads, 8, but not fully developed ( Fig. 4A-C); in fact, only a single individual had them and they were malformed (Fig. 4B, C), so complete data about their morphology are lacking. Cnidome composed of only two nematocyst types: holotrichous isorhiza and heterotrichous microbasic eurytele ( Fig. 5; Table 6).   A more-basal series, with 16 cusps and additional cusps over 4 larger perradial cusps B and C two moredistal series (C being the highest along the tube), each with 8 cusps.

Metagenetic species.
Solitary polyp with typical periderm tube, dark to light brown, conical, with transverse rings on surface with longitudinal striations (4-5 rings every 0.4 mm at 2 mm above the base). Polyps 2.56-31.46 mm long. Basal disk 0.24-0.53 mm in diameter. Diameter of aperture 0.33-1.29 mm. Diameter just above basal disc 0.07-0.23 mm. Diameter at 2-mm height 0.07-0.12 mm, and at 5-mm height 0.07-0.17 mm. Six to 14 whorls of 8 internal cusps: 4 large (perradius) and 4 intermediate (interradius), with additional cusps. Polyp with up to 40 filiform tentacles. Polydisc strobilation, with more than 100 ephyrae at a time. Medusa with smooth umbrella (Fig. 7) with high central dome; 16 slightly elongated lappets with rounded margins, and 8 rhopalia with statocyst and red ocelli. Live specimens measuring up to: 6.8 mm total diameter; 5 mm between opposite rhopalia; 2.5 mm coronal groove diameter; gastric cirri approximately 1.9 mm in length; and tentacle length up to 3 mm. Stomach with 4 gastric septa, each with 1 gastric filament (4 in total). Sequences from all species studied are available on GenBank (Table 7). DNA comparisons of the sequenced markers of the species are summarized in Tables 8 and 9. We were not able to amplify the three proposed markers for all polyps. Therefore, Nausithoe sp. specimens AC01 (no 28S data), AC17 (no COI data), and AC18 (no COI data) were not included in the molecular analyses. Polyps with all three genes sequenced were compared with each other and with N. maculata (= N. aurea from Brazil), N. maculata (Cuba), and N. werneri. Using only the COI marker, Nausithoe sp. (specimens AC02, AC08, AC10, AC20) and N. werneri had less than 6% genetic difference from each other and almost 20% genetic difference from N. maculata (Cuba) and N. maculata (= N. aurea from Brazil). Nausithoe maculata from Cuba had less than 7% difference from N. maculata (= N. aurea) from Brazil. The remaining DNA extractions from cultures are also considered as vouchers (Table 7).

Discussion
The objective of this study was to identify the polyps of Nausithoe sp. from deep waters off southeastern Brazil by comparing them with previous records along the Brazilian coast. We used two approaches: morphology combined with life-cycle observations, and molecular data. Also, we present previously unpublished data on nematocysts for N. werneri and N. maculata from Cuba. So far, only three species of Nausithoe have been recorded from the Brazilian coast: N. aurea, N. atlantica, and N. punctata (Oliveira et al. 2016). Nausithoe aurea is endemic Figure 5. Photomicrographs of the two nematocyst types found in Nausithoe sp. A three heterotrichous microbasic euryteles discharged B holotrichous isorhiza capsules of two different sizes (large, grey arrow; small, black arrow) C a small discharged holotrichous isorhiza. Figure 6. Aboral view of Nausithoe maculata medusae from Cuba's polyp culture. A four-month-old medusa with mature male gonads (g), long tentacles (t), and malformations in the lappets and central disc. B three-month-old medusa with mature female gonads (g), showing the rhopalium (r), coronal groove (cg), marginal lappets (ml) with pigment spot (ps), gastric cirri (gc), and tentacles (t).

Figure 7.
Nausithoe werneri male medusae from Mediterranean's polyp culture (A-C 5 months old D 3 months old) A aboral view of an adult medusa with mature gonads (g), contracted lappets, and extended tentacles (t) B detail of gastrovascular cavity, focusing on the gastric cirri (gc) C lateral view, focusing on the lips (l) and the extension of the manubrium D beginning of gonad (g) development (aboral view).  to the Brazilian coast and can be found in shallow waters (Silveira and Morandini 1997;Morandini and Silveira 2001). The polyps are solitary with 16 smooth cusps per series; and the medusae have a characteristic single yellow spot in the center of each marginal lappet (Silveira and Morandini 1997). These features show that this species is not identical to Nausithoe sp. Nausithoe punctata is a cosmopolitan species that lives in shallow waters (Jarms and Morandini 2019); in Brazil it was reported by Goy (1979) and Neumann-Leitão et al. (2008). The polyp stage is colonial and inhabits sponges (Werner 1979;Uriz et al. 1992). Medusae are transparent with a pale-pink disc and yellowish lappets; the umbrella is flattened and reaches up to 15 mm in diameter (Kölliker 1853;Kramp 1961;Werner 1970). With these features, we can also discard this species as identical to Nausithoe sp. As pointed out by Jarms and Morandini (2019), there is extensive confusion in the taxonomy and identification of N. punctata, and many records around the world might be erroneous.
Nausithoe atlantica is known only from the medusa stage (Jarms and Morandini 2019). Although records from the Pacific Ocean exist (Oliveira et al. 2016), the only confirmed identifications are from the North Atlantic, near the type locality (Broch 1913;Russell 1956). The adult medusa is dark yellowish-brown, up to 35 mm in diameter, and has rhopalia without ocelli, oblong gonads, and more than 160 gastric filaments (Russell 1970). The Brazilian record provided by Oliveira et al. (2016) derives from a thesis in which the specimens were collected off Santa Catarina state. The description provided is identical to Russell (1970), and no voucher material is available for comparison. Considering the divergent features and doubtful record, we discard this species as identical to Nausithoe sp.
The two types of nematocysts found in Nausithoe sp. (heterotrichous microbasic euryteles and holotrichous isorhizas) are the same as in N. aurea (Silveira and Morandini 1997) and in N. planulophora (Werner 1971). These are the most common types in scyphozoans (Östman 2000), which indicate that this information might not be useful in differentiating species of the group. There is an overlap of the measurements obtained from Nausithoe sp. and N. werneri, and polyp nematocysts are slightly larger than ephyrae and medusae ones. For now, much more work on the scyphozoan cnidome is needed to improve understanding of the types of cnidae in the group's evolutionary history.
As do most of the solitary Nausithoidae polyps, those of Nausithoe sp. and N. werneri resemble each other. As stated by several authors, the most useful features to distinguish Nausithoe polyps are the number and shape of the internal cusps of the tube (e.g., Jarms 1991; Morandini and Jarms 2012). Although these features are used widely in the systematics of the group, the variation among specimens has not been thoroughly studied, in part because of the relatively few samples that can be used for scanning electron microscopy. It is also unknown how laboratory conditions might affect the growth of the animals and the shape of the cusps. SEM observations of the cusps of Nausithoe sp. showed 16 internal cusps per whorl close to the base and eight cusps in the upper whorls of the tube (Fig. 3), whereas in N. werneri, only eight cusps are found per whorl, both in the literature (Jarms 1990) and in our observations.
Both the ephyrae and medusae of the Brazilian deep-sea Nausithoe sp. are morphologically similar to N. werneri: translucent body, rhopalium with statocyst and red ocelli, lappets slightly elongated with rounded margins, and total diameter ( Table 4). The only differences we noted were the shape and size of the manubrium, the shape  of the central disc, and the number of gastric cirri. In Nausithoe sp., the manubrium is wider (Fig. 4F-H) than in N. werneri (Fig. 6D). The elevated central dome of N. werneri is also not present in Nausithoe sp. Although Jarms (1990) described N. werneri with two or three gastric cirri per quadrant, our cultivated specimens (derived from Jarms's culture) had only one gastric cirrus per quadrant (Fig. 6B); Nausithoe sp. has eight filaments in the stomach. These variations in morphological features might be related to plasticity in the response of the species under different environmental conditions and food supply. Certainly, the food provided in our laboratory differs from that used by Jarms (1990). Although our specimens were kept in the same conditions, there might be some individual variation. To ascertain the utility of these morphological features it would be necessary to examine many specimens from a broad population. An interesting feature is the difference in the time taken to develop gonads between Nausithoe sp. and our specimens of N. werneri (Table 5). Nausithoe sp. took more than 20 weeks to begin the differentiation and development of the reproductive organs, while N. werneri required only nine weeks. These times are comparable because all species were kept in the same kind of container and temperature, with similar population densities, and fed equally.
Genetic divergence related to species delimitation is difficult to discern, especially for clades with limited molecular markers and specimens. For Discomedusae, the sister clade of Coronamedusae, Gómez Daglio and Dawson (2017) proposed that the mean intraspecific pairwise genetic distance (x ± s.d.) is 0.006 ± 0.005 and the mean interspecific distance between congeneric species (x ± s.d.) is 0.12 ± 0.04. Our molecular data showed that specimens of Nausithoe sp. and N. werneri had less than 6% of genetic difference for the COI gene (Table 8) and each of them differed by ~20% from N. maculata (from Cuba) and N. maculata (= N. aurea from Brazil). Together with the morphological similarities, we state they are the same species, according to the criteria of Gómez Daglio and Dawson (2017). Following the same rationale, we found less than 7% of genetic difference in the COI gene between N. maculata (from Cuba) and N. maculata (= N. aurea from Brazil). In these two species, the morphological resemblance is obvious, and the only difference is the formation of planuloids inside the periderm tube of N. aurea (Silveira and Morandini 1997), which was considered one of the distinguishing features of this species, not observed in N. maculata. Considering that Table 9. Genetic similarity (percent) between each Nausithoe sp. polyp (AC02, AC08, AC10, AC20), N. maculata (= N. aurea from Brazil), N. maculata (from Cuba), and N. werneri. COI in italic; all three markers combined in white. Boldface indicates higher similarity that we are considering to be the same species. coronamedusae species were mostly defined by a few specimens, we propose that this feature represents possible genetic plasticity not previously recorded for N. maculata.
Comparing the life cycle and morphology of scyphomedusae is extremely important to help in identifying and describing species. However, the simple structure of these animals, as evidenced by the traditional use of certain uninformative characters in the description of specimens (e.g., Kramp 1961;Russell 1970), can yield insufficient information for a precise analysis (Dawson 2005). Future approaches, including additional genetic data from other species of Coronatae, will add detail to the systematics of this clade. A broader sampling of molecular markers, individuals, populations, species, and clades in general will allow for novel insights to be applied to Coronamedusae, the proposal of a more refined "genetic gap" for the delimitation of species, and specific genetic diagnoses (DeSalle and Goldstein 2019). Nevertheless, the use of genetic data must be combined with information on morphology and life cycles to confirm or reject the validity of species and to identify new taxa.
To conclude, based on both the morphological and molecular data obtained, we identify the deep-sea Nausithoe sp. specimens from off the Brazilian coast as Nausithoe werneri, thus expanding the distribution of this species to the western South Atlantic. Additionally and also based on molecular and morphological data, we consider the species Nausithoe aurea as a junior synonym of Nausithoe maculata.