Eur. J. Phycol. (2015), 1–14

A new definition of Adenoides eludens, an unusual marine sand-dwelling dinoflagellate without cingulum, and Pseudadenoides kofoidii gen. & comb. nov. for the species formerly known as Adenoides eludens

FERNANDO GÓMEZ1,3, RYO ONUMA2, LUIS F. ARTIGAS3 AND TAKEO HORIGUCHI4

Downloaded by [Sistema Integrado de Bibliotecas USP] at 05:09 23 February 2015

1

Laboratory of Plankton Systems, sala 100, Oceanographic Institute, University of São Paulo, Praça do Oceanográfico 191, Cidade Universitária, Butantã, São Paulo 05508-900, Brazil 2 Department of Natural History Sciences, Graduate School of Science, Hokkaido University, North 10, West 8, 060-0810 Sapporo, Japan 3 Laboratoire d’Océanologie et Géosciences, CNRS UMR 8187, Université du Littoral Côte d’Opale, MREN ULCO, 32 av. Foch, 62930 Wimereux, France 4 Department of Natural History Sciences, Faculty of Science, Hokkaido University, North 10, West 8, 060-0810 Sapporo, Japan (Received 9 Jul 2014; revised 24 Sep 2014; accepted 8 Nov 2014) The species Amphidinium eludens, as described by Herdman (1922; Proc. Trans. Liverpool Biol. Soc. 36: 18) based on her drawing in fig. 1, has been investigated for the first time by scanning electron microscopy and phylogenetic analysis. The morphological and molecular data reveal that this species is distantly related to other known dinoflagellates. Balech [1956, Rev. Algol., n. ser. 2(1–2): 30] cited Amphidinium eludens Herdman (1922, fig. 1) as the basionym of the type of Adenoides, while he described and illustrated Amphidinium kofoidii Herdman (1922, fig. 2) as Adenoides eludens. As the nomenclatural rules do not allow the change of basionym, we have re-defined the genus Adenoides based on the characteristics of Amphidinium eludens Herdman (1922, fig. 1). The thecal plate formula of Adenoides eludens is Po, 5′, 6″, 0c, 3+s, 5′″, 3p, 1″″. This species lacks a cingulum. The apical pore complex resembles that of peridinioid dinoflagellates, while the absence of a cingular groove is reminiscent of desmokont prorocentroids. We also propose Pseudadenoides kofoidii gen. & comb. nov. based on Herdman’s 1922 fig. 2 of Amphidinium kofoidii which was described by Balech in 1956 and re-named Adenoides eludens. Key words: Adenoides kofoidii, Dinophyta, microphytobenthos, molecular phylogeny, new genus, psammophilic Dinophyceae

INTRODUCTION Dinoflagellates are unicellular organisms with two dimorphic flagella. Based on the arrangement of the flagella, the dinokonts (= two flagella in grooves) are dinoflagellate cells in which two flagella are inserted ventrally; one flagellum is transverse, wrapped around the cell, and housed in a groove called the cingulum or girdle (equatorial or transverse) and the other one is a trailing longitudinal flagellum and housed in a sulcus (longitudinal). The other type, desmokonts (= two anterior flagella) are cells in which two dissimilar flagella emerge from the anterior part of the cell. They have two leading flagella inserted apically, rather than ventrally. One flagellum extends forward and the other circles its base, and there are no flagellar grooves (cingulum or sulcus). The best known Correspondence to: Fernando Gómez. (e‑mail: fernando.gomez@fitoplancton.com)

example of desmokonts is the genus Prorocentrum C.G. Ehrenberg. However, not all dinoflagellates fit into this division. The descriptions of new benthic species have largely increased in recent years (Gómez, 2012). The presence of an incomplete cingulum is reported in some sand-dwelling dinokonts (Amphidiniopsis J. Woloszynska, Cabra Sh. Murray & D.J. Patterson, Herdmania J.D. Dodge, Rhinodinium Sh. Murray, Hoppenrath, S. Yoshimatsu, S. Toriumi & J. Larsen) (Murray & Patterson, 2004; Yamaguchi et al., 2011). Planktonic dinoflagellates such as Podolampas F. Stein lack a groove (cingulum) to harbour a transversal flagellum that encircles the cell (Gómez et al., 2010). However, the absence of a cingulum has never been reported in benthic dinokonts. In one of the earlier studies of sand-dwelling dinoflagellates, Herdman (1922) described the species Amphidinium eludens and A. kofoidii. In her fig. 1 she described Amphidinium eludens as oval to ovoid

ISSN 0967-0262 (print)/ISSN 1469-4433 (online)/15/000001-14 © 2015 British Phycological Society http://dx.doi.org/10.1080/09670262.2015.1009174

Downloaded by [Sistema Integrado de Bibliotecas USP] at 05:09 23 February 2015

F. Gómez et al. in shape, has a barely visible episome and a hump in the sulcal area (Herdman, 1922, p. 22, fig. 1). In her fig. 2 she described Amphidinium kofoidii as round to squarish in shape, with a more visible button-like episome (Herdman, 1922, p. 26, fig. 2). Balech (1956) proposed the new thecate genus Adenoides based on Herdman’s fig. 2 of Amphidinium kofoidii. However, Balech (1956) proposed as basionym Amphidinium eludens Herdman (1922, fig. 1). Later, Dodge (1982) proposed Adenoides kofoidii for Amphidinium kofoidii Herdman 1922, fig. 2. The taxonomic and nomenclatural history of Adenoides eludens is reported in Hoppenrath et al. (2003). Adenoides eludens based on Balech’s description and as described by Herdman’s fig. 2 of Amphidinium kofoidii was investigated again by Dodge & Lewis (1986) and Hoppenrath et al. (2003), and molecular data have been available since Saldarriaga et al. (2001). To date, there are 45 sequences of different molecular markers of Adenoides eludens and another 39 sequences labelled as two uncultured Adenoides. The sequences of Adenoides eludens available in GenBank are from the cultures CCCM 683 and CCMP 1891 isolated from the Pacific coasts of Canada or field material collected from the same location. Another culture CCMP 2081 was isolated from Europe. However, there are no molecular data for Adenoides eludens from the European Atlantic coast where the species was described. While Adenoides eludens (= Amphidinium kofoidii) has been subjected to morphological and molecular studies, little is known about the species described as Amphidinium eludens Herdman (1922, fig. 1). This organism is one of the first described sand-dwelling dinoflagellates and according to Herdman (1922) is responsible for discolourations in the sands of the European Atlantic coasts. However, the detailed morphology of this species remains unknown. In this study, we provide light microscopy pictures, the first scanning electron microscopy pictures and the first molecular data for Amphidinium eludens Herdman (1922, fig. 1), and the first molecular data for Amphidinium kofoidii Herdman (1922, fig. 2) from the French coasts of the English Channel, where Balech (1956) described the genus Adenoides. The morphological and molecular data support the conclusion that the two species described in figs 1 and 2 by Herdman (1922) belong to independent genera. Amphidinium eludens is an unusual dinoflagellate, characterized by the lack of a cingular groove similar to desmokont dinoflagellates.

MATERIALS AND METHODS Source, isolation and microscopy observations This study was undertaken in the soft sandy sediments of the shore of Wimereux, France (50º46′12″N, 1º36′42″E)

2 collected during low tide in June, 2012. Two sites on the beach in front of the LOG laboratory (MREN ULCO and Marine Station of Wimereux UL1; Gómez & Artigas, 2014) were sampled: the moist sands around the border of a large pool (~50 m diameter, ~1 m depth), and several smaller pools and moist sands showing a faint brownish discolouration. The upper centimetre of sand was collected with a spoon and deposited into a bottle containing seawater collected at the same location. Then, the sand with seawater was stirred vigorously and the suspension settled in a composite settling chamber. The settled material was examined with an inverted microscope (Nikon Eclipse TE2000-S, Tokyo) and photographed with a Nikon Digital Sight DS-2M camera. The cell size, described as length (apical to antapical axis) and depth, i.e. the length along the lateral sides (ventral to dorsal distance), was measured in 25 specimens. The width (transdiameter) was measured in five specimens. For scanning electron microscopy, the sand samples with seawater were stirred vigorously, and the suspension was fixed with glutaraldehyde (5%) and filtered onto a 0.8 µm pore size Whatman Nuclepore® membrane filter, washed with distilled water, fixed with osmium tetroxide, dehydrated with a graded series of ethanol and critical point dried with CO2. Filters were mounted on stubs, sputter-coated with gold and viewed using a Hitachi S4800 scanning electron microscope. Images were presented on a black background using Adobe Photoshop CS3.

PCR amplification and DNA sequencing For molecular analysis, each specimen of Amphidinium eludens and A. kofoidii was micropipetted individually with a fine capillary into a clean chamber and washed several times in serial drops of 0.2 μm filtered and sterilized seawater. Finally, 1–5 specimens of each species were deposited in a 0.2 ml Eppendorf tube filled with several drops of absolute ethanol. The sample was kept at room temperature and in darkness until the molecular analysis could be performed. Prior to DNA extraction, the 0.2 ml Eppendorf tubes were centrifuged for 10 min at 14462 × g in a TOMY MX-201 centrifuge (Tokyo, Japan). Ethanol was then evaporated in a vacuum desiccator. Cells were resuspended in 10 μl of QuickExtract DNA extraction solution (Epicenter, Madison) and incubated at 56ºC for 1 h and 98ºC for 2 min in a thermal cycler (GeneAmp PCR System 9700, Applied Biosystems, Foster City). The product was used as DNA template for the following polymerase chain reaction (PCR). In the first round of PCR, to obtain almost complete SSU rDNA and partial LSU rDNA sequences, two sets of primers (SR1-SR12b for SSU rDNA and D1RF1 28-1483R for LSU rDNA, respectively) were simultaneously applied. In the second round of PCR, 0.5 µl of the first PCR products was used as DNA template, and three pairs of primers (SR1SR5, SR4-SR9p and SR8-SR12b; see Yamaguchi et al., 2006) were used for SSU rDNA amplification and two pairs of primers (D1RF-25R1 and D3A-28-1483R; see Daugbjerg et al., 2000) were applied for LSU rDNA amplification. PCR conditions for both rounds of amplification consisted of one initial cycle of denaturation at 94ºC for 5 min, followed by 35 cycles of denaturation at 94ºC for 30 s, annealing at 55°C for 30 s, and extension at 72ºC for 30 s. The PCR process was completed by a final extension cycle

Pseudadenoides gen. & comb. nov. and Adenoides at 72ºC for 7 min. The PCR products were directly sequenced using the ABI PRISM BigDye Terminator Cycle Sequencing Kit (Applied Biosystems) and a DNA autosequencer ABI PRISM3100 Genetic Analyzer (Applied Biosystems). Both forward and reverse strands were sequenced.

Downloaded by [Sistema Integrado de Bibliotecas USP] at 05:09 23 February 2015

Phylogenetic analyses Both SSU rDNA and LSU rDNA sequences were aligned based on the secondary structure of the rDNA molecule (database no longer available), and the alignments were refined manually. We used sequences from GenBank of Amoebophrya spp. as outgroups for SSU rDNA and a ciliate Tetrahymena pyriformis (C.G. Ehrenberg) Lwoff and an apicomplexa Eimeria tenella (Raillet & Lucet) Fantham were used for outgroups of LSU rDNA analyses, respectively. The accession numbers of sequences included in the alignment are indicated in each tree. The aligned sequences were examined using maximum likelihood (ML) analyses with PAUP version 4.0b10 (Swofford, 2003), and Bayesian analysis with MrBayes 3.2.1 (Huelsenbeck & Ronquist, 2001). The program Modeltest version 3.04 (Posada & Crandall, 1998), which employs the hierarchical likelihood ratio test (hLRT) was used to explore the best ML sequence evolution model for the dataset. The hRLT model selected for ML analysis of the dataset was TrN+I+G. In the ML analysis, a heuristic search was performed with a TBR branch-swapping algorithm, and the starting tree was obtained by the neighborjoining (NJ) method. The parameters in this analysis were: assumed nucleotide frequencies A = 0.2756, C = 0.1898, G = 0.2392 and T = 0.2954; substitution rate matrix with A<->C = 1.0000, A<->G = 3.5126, A<->T = 1.0000, C<->G = 1.0000, C<->T = 7.7026 and G<->T = 1.0000; proportion of sites assumed to follow a gamma distribution with shape parameter = 0.5633; and number of rate categories = 4. Bootstrap analysis for ML was calculated for 100 pseudoreplicates. For LSU rDNA analysis, the hLRT model selected for ML analysis of the dataset was TrN+I+G. In the ML analysis, a heuristic search was performed with a TBR branch-swapping algorithm, and the starting tree was obtained by the NJ method. The parameters in these analyses were: assumed nucleotide frequencies A = 0.2865, C = 0.1806, G = 0.2701 and T = 0.2628; substitution rate matrix with A<->C = 1.0000, A<->G = 2.8299, A<->T = 1.0000, C<->G = 1.0000, C<->T = 6.6145 and G<->T = 1.0000; proportion of sites assumed to follow a gamma distribution with shape parameter = 0.6675; and number of rate categories = 4. Bootstrap analysis for ML was calculated for 100 pseudo-replicates. For Bayesian analysis, GTR+I+G model was selected by MrModeltest 2.2 (Nylander et al., 2004) as a suitable evolutionary model. Markov chain Monte Carlo iterations were carried out until 3 000 000 generations were attained for SSU rDNA phylogeny, while 5 500 000 generations were required for LSU rDNA phylogeny, when the average standard deviations of split frequencies fell below 0.01, indicating a convergence of the iterations. Our sequences were deposited in DDBJ/ EMBL/GenBank under accession numbers LC002839LC002848.

3 RESULTS Based on the morphological and molecular data (see below) the species described as Amphidinium eludens Herdman (1922, fig. 1) and Amphidinium kofoidii Herdman (1922, fig. 2) belong to separate genera. If we had proposed a new genus name based on Herdman (1922, fig. 1), it would have been illegitimate as a superfluous later homotypic synonym of Adenoides [International Code of Nomenclature for algae, fungi, and plants (I.C.N.), art. 52.1, McNeill et al., 2012]. As defined by Balech (1956), the basionym of the type of Adenoides, Adenoides eludens, is Amphidinium eludens Herdman (1922, fig. 1). According to article 7.3 of I.C.N.: ‘A new combination or a name at new rank (Art. 6.10) is typified by the type of the basionym even though it may have been applied erroneously to a taxon now considered not to include the type (but see Art. 48.1)’. The characteristics of the genus Adenoides are defined by Amphidinium eludens Herdman (1922, fig. 1), independently of the fact that Balech (1956) provided for Adenoides eludens the description and illustrations of the species described as Amphidinium kofoidii by Herdman (1922, fig. 2). Hereafter, we redefined the genus Adenoides based on the morphology of Amphidinium eludens Herdman (1922, fig. 1). Taxonomic descriptions Adenoides Balech emended F. Gómez, R. Onuma, Artigas & Horiguchi DIAGNOSIS: Armoured cell laterally compressed, lacking the cingulum and flagella inserted ventrally. Thecal plate formula Po, 5′, 6″, 0c, 3+s, 5′″, 3p, 1″″. An alternative interpretation is Po, cp, 4′, 7″, 1c, 1+s, 5′″, x, 2p, 1″″. TYPE SPECIES: Adenoides eludens (Herdman) Balech. BASIONYM: Amphidinium eludens Herdman 1922, p. 22, fig. 1. SYNONYM: Adenoides kofoidii sensu Dodge 1982. Adenoides eludens F. Gómez, R. Onuma, Artigas & Horiguchi epitype Given the lack of type material, we have designated fig. 1 in Herdman (1922) as the lectotype for this species and the cell shown in Fig. 23 of this study as the epitype under Article 9.7 of the I.C.N. (McNeill et al., 2012). A detailed description of the epitype is presented below. EPITYPE: Fig. 23. A SEM stub was deposited in the herbarium of the Faculty of Science, Hokkaido University as SAP114711.

Downloaded by [Sistema Integrado de Bibliotecas USP] at 05:09 23 February 2015

F. Gómez et al. The slightly laterally flattened cells are ellipsoidal in lateral view. The sulcal area is depressed and lies on the anterior third of the cell, neither extending onto the epitheca nor reaching the antapex. In lateral view, the ventral contour of the cell showed a hump (Figs 2, 6). The cells were 30 ± 2 (27–37) µm long, 23 ± 2 (18–27) µm deep, and 13–17 µm wide (Figs 1–15). There is an intergradation of the size with large and small specimens co-existing in the same sample (Fig. 6). The epitheca is directly connected to the hypotheca, without any cingulum. We presume that the vestigial cingular groove was placed in the suture between the preand postcingular plates that is at the level of the flagellar insertion. The height of the epitheca is about one-third the length of the cell. Some specimens showed a large pusule in the anterior dorsal part of the cell (Figs 1–4). The transversal flagellum encircled the cell (Fig. 5). The cell possesses yellow-brown plastids. The species Amphidinium kofoidii sensu Herdman is more pigmented and brownish when compared with Adenoides eludens (Fig. 7). Some specimens of Adenoides were found devoid of pigments and with numerous granules (Figs 8–10). The cell showed two pyrenoids with a starch sheath, each pyrenoid placed in the sides of the mid anterior hypotheca (Figs 6, 11). The nucleus is oval and situated in the posterior region, although hardly visible because it is hidden by plastids. Some specimens experienced ecdysis or exuviation, releasing empty thecae (Figs 11–12). The empty theca was usually divided into two parts, which are considered the epitheca and hypotheca (Figs 13–15). One plate that is interpreted as a sulcal anterior plate is attached to the epitheca (Fig. 13). The tabulation is illustrated in detail in Figs 16–46. The thecal plate pattern is an apical pore plate (Po), five apical plates (5′), six precingular plates (6″), no cingular plate (0c), at least three sulcal plates (3+s), five postcingular plates (5′″), three posterior intercalary plates (3p) and one antapical plate (1″″). The cell surface is smooth with round pores (approximately 0.15 µm). The pores are evenly distributed in the plates, with a trend to form rows near the sutures (Fig. 19). There is an accumulation of pores in the posterior end of the first and second posterior intercalary plates (1p, 2p) and in the dorsal end of the antapical plate (Figs 16, 40–41). The pores are absent in the sulcal plates (Figs 22, 27–30, 34). The boundary of the plates shows a thick (~1 µm wide) smooth and depressed margin. These intercalary bands were more conspicuous in the hypotheca (Figs 19, 21). The apical pore plate showed a pentagonal to round shape being bordered by five apical plates (Figs 24, 37). The junction of the apical pore plate and the apical series formed a ridge, except in the first apical plate suture which is shorter than in the other apical plates (Figs 24, 25, 37). The apical pore plate contains a row of marginal pores (10–13 pores) and a

4 round central cover plate (Figs 26, 37). From the central cover plate emerged a rim that protruded and extended towards the anterior part of the first apical plate (Figs 24, 26, 37). There are five apical plates (Figs 20, 23, 37). Plate 3′ is the largest and is located in the dorsal side of the epitheca. Plates 2′ and 5′ are intermediate in size. Plates 4′ and 1′ are the smallest of the apical series. The apical plates 2′ to 5′ showed a curved ridge in the suture with the apical pore plate. Plate 1′ did not show a ridge in the junction with the apical pore plate. Plate 1′ is a six-sided irregular smooth-surfaced polygon, with very few pores when compared with the other apical plates (Figs 24–26). An alternative interpretation of the apical tabulation is to consider plate 1′ as a wide canal plate (cp). Plates 2′ and 4′ are pentagonal, especially plate 4′ which is a quasi-regular pentagon. Plates 3′ and 5′ are six-sided. Plate 5′ shows a distinctive undulating flange in the sutures, except in the anterior end in the suture with plates 5″ and 6″ (Figs 25–26, 37). At the level of the precingular plate series, there are six precingular plates and a large smooth-surfaced and concave plate. This plate is joined to the precingular plates when the epitheca is separated from the hypotheca (Fig. 13). Based on the proximity to the flagellar pores, the differences in the shape (concave) and the absence of ornamentation (smooth-surfaced lacking pores and intercalary bands) as in the other sulcal plates, we have considered it as the right anterior sulcal plate (S.d.a.) (Figs 21–22, 34). An alternative interpretation is to consider this large plate as plate 1″. Following the interpretation as an anterior sulcal plate, the first precingular plate (1″) is displaced to the left side in the ventral view. Plates 1″ and 5″ are the biggest of the epitheca, plates 2″, 3″ and 4″ are intermediate in size and plate 6″ is the smallest of the precingular series (Figs 16–20, 35–36). Plate 1″ is polygonal and curved in the suture with the anterior sulcal plate (Figs 21–23). Plates 2″, 4″ and 5″ are six-sided and plates 3″ and 6″ are quadrangular (Figs 16–20). Plate 3″ is nearly square-shaped (Fig. 20) and plate 6″ is an elongated rectangle oriented along the antero-posterior axis (Figs 28, 34–35). The cell does not have a cingulum. However, when the theca is empty, it tended to separate along the suture between the precingular and postcingular plate series (Figs 13–15). This suggests that this thick suture could be occupying the space of the cingulum of a hypothetical ancestor. The observations of live specimens revealed that the transversal flagellum encircles this area (Fig. 5). The cell shows a depression in the sulcal area (Figs 21–22, 32–35). The left anterior sulcal (S.s.a.) plate is smooth and located between the right anterior sulcal plate (S.d.a.) and a large plate 1′″. This S.s.a. plate is elongated and concave and it could be alternatively interpreted as a first cingular plate (Figs 21–23, 34).

5

Downloaded by [Sistema Integrado de Bibliotecas USP] at 05:09 23 February 2015

Pseudadenoides gen. & comb. nov. and Adenoides

Figs 1–15. Light microscopy pictures of live specimens and empty thecae of the redefined Adenoides eludens (Fig. 7 also includes Amphidinium kofoidii). Figs 1–4. Different views of the same specimen. Fig. 5. Note the transversal flagellum that encircles the cell. Fig. 6. Note the different sizes. Fig. 7. Note the different pigmentation between Adenoides eludens (yellowish, on the left) and Amphidinium kofoidii (brownish, on the right). Figs 8–10. Specimens lacking pigmentation. Figs 11–12. Ecdysis or exuviation from the theca. Figs 13–15. Empty theca with separated epitheca and hypotheca. Fig. 13. Note that the sulcal right (= dexter) anterior plate (S.d.a.) is attached to the epitheca. LF = longitudinal flagellum. TF = transversal flagellum. Scale bar = 10 µm.

There are two flagellar pores in the depressed area of the sulcus. The longitudinal flagellum is inserted in a pore located in the right side of the sulcus, adjacent to the suture of plates 6″ and 5′″ (Figs 22, 27–30, 34). The right half of this pore is surrounded by a prominent ridge (Figs 28–29). The pore of the transversal flagellum is located in the middle of the sulcus. It is bordered by a rectangular structure

(Figs 28–30). Below the pores, a large sulcal posterior plate with a rounded posterior end is subdivided into three plates connected by wedge-shaped contours (Figs 29–30). The right anterior sulcal (S.d.a.) plate is in contact with the flagellar pores (Fig. 22). This plate shows an anterior-posterior oriented line, which suggests the presence of a subdivision of the plate (Fig. 30).

6

Downloaded by [Sistema Integrado de Bibliotecas USP] at 05:09 23 February 2015

F. Gómez et al.

Figs 16–30. Scanning electron micrographs of the re-defined Adenoides eludens. Figs 16–17. Right lateral view. Fig. 18. Ventral view. Fig. 19. Left lateral view. Fig. 20. Dorsal view. Figs 21–23. Ventral lateral view. Figs 24–25. Detail of the apical plates. Fig. 26. Detail of the apical pore plate. Figs 27–30. Detail of the sulcal plates. Fig. 30. The arrowheads point out tentative sutures in the sulcal plates. S.d.a. = Sulcal dexter (= right) anterior plate. S.s.a. = Sulcal sinister (= left) anterior plate. S.p. = Sulcal posterior plate. LFP = longitudinal flagellar pore. TFP = transversal flagellar pore. Scale bar = 10 µm, except Figs 22, 24–30 where scale bar = 1 µm.

There are five postcingular plates (Figs 16, 19). Plate 1′″ is the longest plate of the cell and extended for 2/3 the height of the hypotheca. The anterior part

of this five-sided plate is curved and overhangs at the anterior end (Figs 18, 22, 34). Plates 2′″ and 4′″ are large and broad, plate 3′″ is intermediate in size

Downloaded by [Sistema Integrado de Bibliotecas USP] at 05:09 23 February 2015

Pseudadenoides gen. & comb. nov. and Adenoides

7

Figs 31–41. Scanning electron micrographs of a single specimen of Adenoides eludens. Fig. 31. Ventral view. Figs 32–33. Right lateral view. Fig. 34. Detail of the sulcal area. Fig. 35. Latero-ventral view. Figs 36–37. Detail of the epitheca. Figs 38–39. Dorsal view. Figs 40–41. Antapical view. Fig. 40. The inset shows the pores of the plates 2p (up) and 1″″ (down). S.d.a. = Sulcal dexter (= right) anterior plate. S.s.a. = Sulcal sinister (= left) anterior plate. S.p. = Sulcal posterior plate. Scale bar = 20 µm, except Figs 34 and 37 where scale bar = 1 µm.

and plate 5′″ is the smallest of the postcingular series. Plate 2′″ is four sided, plate 3′″ is a regular pentagon and plate 4′″ is five-sided and elongated posteriorly. Plate 5′″ is smaller and more anteriorly placed than the other plates of the postcingular series (Figs 34–35). An alternative interpretation is that plate 5′″ is equivalent to plate x reported in some sand-dwelling dinoflagellates with incomplete cingulum. Four large plates that do not belong to the sulcal and postcingular series are considered posterior intercalary and antapical plates. Three plates are considered

posterior intercalary plates and the plate situated at the bottom of the cell is considered the antapical plate 1″″ (Figs 16, 19). The antapical plate is bordered by plate 1′″ and three posterior intercalary plates (1p, 2p, 3p). The first and second posterior intercalary plates (1p, 2p) are large and with the shape of an oblique irregular pentagon (Figs 16, 19). Plates 1p and 2p have an accumulation of pores in the posterior end, at the contact with the antapical plate (Figs 39–41). Plate 3p is an elongated pentagon oriented in the longitudinal axis (Figs 32–33). The antapical plate is hexagonal and it showed, at the dorsal end, the special

Downloaded by [Sistema Integrado de Bibliotecas USP] at 05:09 23 February 2015

F. Gómez et al.

8

Figs 42–46. Line drawings of the re-defined Adenoides eludens with the Kofoid system of tabulation. Fig. 42. Right lateral. Fig. 43. Left lateral. Fig. 44. Apical. Fig. 45. Antapical. Fig. 46. Ventral.

Figs 47–56. Light microscopy pictures of live specimens (47–50) and empty thecae (51–56) of Pseudadenoides kofoidii gen. & comb. nov. Fig. 47. The arrows point the transversal flagella encircling the cingulum. Fig. 50. Megacytic cell. Figs 51–56. Empty thecae. Scale bar = 10 µm.

pore field found in plates 1p and 2p (Fig. 41). An alternative interpretation of the tabulation of the hypotheca may appear when plate 5′″ is considered to be an x plate (Fig. 48), then plate 3p is considered to be 5′″ plate. A new genus name for Amphidinium kofoidii Herdman (1922, fig. 2)

We had to redefine the genus Adenoides due to the designation of Amphidinium eludens Herdman (1922, fig. 1) as basionym of the type of Adenoides. Morphological and molecular data reveal that the species Amphidinium kofoidii Herdman (1922, fig. 2) does not belong to the new definition of the genus Adenoides or to the other known dinoflagellate genus. Therefore, a new genus name is proposed here. The

Pseudadenoides gen. & comb. nov. and Adenoides

9

description of the genus and its type species is here simplified because the morphological characteristics have been already reported from the English Channel and North Sea (Balech, 1956; Dodge & Lewis, 1986; Hoppenrath et al., 2003). For that reason, we have summarized the description which is similar to that found, with more detail, in Hoppenrath et al. (2003).

central cover plate. From the central cover plate emerged a narrow rim that protruded and extended ventrally towards the anterior part of the first apical plate (Fig. 62). The apical plates 1′ and 4′ are in contact with the sulcus (Fig. 62). Plate 2′ is very narrow and 4′ is relatively large, covering nearly the entire right half of the epitheca. There are no precingular plates. The shallow cingulum consists of six plates and there are four small sulcal plates surrounding the flagellar pore (s.a., s.s., s.p., s.d.) (Figs 59–60). The hypotheca consists of 11 plates, comprised of five postcingular, five posterior intercalary plates and one antapical plate. Plates 1′″ and 2′″ lie at the left lateral cell side in the upper fourth of the hypotheca. The small, pentagonal plate 3′″ lies dorsally, and the pentagonal, posteriorly pointed plate 4′″ is relatively large, lying at the right lateral side. The six-sided plate 5′″ is in contact with the sulcus. All five posterior intercalary plates are large and cover most of the hypotheca (Figs 57–59). The plates 1p and 5p are in contact with the sulcus and bordered its posterior margin. Both are in contact with each other and form a ‘ventral suture’ (Figs 59–60).

Pseudadenoides F. Gómez, R. Onuma, Artigas & Horiguchi gen. nov.

Downloaded by [Sistema Integrado de Bibliotecas USP] at 05:09 23 February 2015

DIAGNOSIS: Armoured cell laterally compressed, with a minute epitheca. The plate formula is Po, 4′, 0″, 6c, 4s, 5′″, 5p, 1″″ (or alternatively it can be interpreted as Po 4′, 0″, 6c, 5s, 5′″, 3p, 2″″). SYNONYM: Adenoides sensu Balech 1956. ETYMOLOGY: Pseudo-, pseud- (before vowels), Ancient Greek ψευδής (pseudes): false, not genuine, fake. The type species has been confused with Adenoides eludens. The gender is feminine (art. 62.4 of I.C.N.). TYPE SPECIES: Pseudadenoides kofoidii (Herdman) F. Gómez, R. Onuma, Artigas & Horiguchi comb. nov. BASIONYM: Amphidinium kofoidii Herdman (1922, p. 26, fig. 2) SYNONYM: Adenoides eludens sensu Balech 1956 EPITYPE: Fig. 60 Pseudadenoides kofoidii (Herdman) F. Gómez, R. Onuma, Artigas & Horiguchi comb. nov. Cells are asymmetrical, round to squarish, slightly flattened laterally, 28–37 µm long and 21–29 µm deep. The minute epitheca is cup-shaped, depressed and scarcely visible (Figs 47–50). The cingulum lies almost at the anterior end of the cell, completely encircling the epitheca and meeting without displacement. The sulcal area lies on the anterior third of the cell, neither extending onto the epitheca nor reaching the antapex, and is slightly depressed. The transverse flagellum completely encircles the cell at the cingulum level (Fig. 47). Most of the cells show one large pusule in the anterior hypotheca. The cell is full of brown plastids. There are two conspicuous pyrenoids. The nucleus is oval and situated in the posterior region, although hardly visible because it is hidden by plastids (Fig. 47–50). The thecal plate pattern is an apical pore plate (Po), four apical plates (4′), without precingular plates (0″), six cingular plates (6c), four sulcal plates (4s), five postcingular plates (5′″), five posterior intercalary plates (5p) and one antapical plate (1″″) (Figs 51–62). The apical pore plate is an angular square-shape bordered by four apical plates which form a ridge running around it (Fig. 60). The apical pore plate contains few marginal pores (~7 pores) and a round

Molecular phylogeny We have sequenced specimens of the species Amphidinium eludens Herdman (1922, p. 22, fig. 1, now Adenoides eludens emended), and the species Amphidinium kofoidii Herdman (1922, p. 26, fig. 2, now Pseudadenoides kofoidii gen. & comb. nov.). The SSU- and LSU rDNA sequences of Amphidinium kofoidii Herdman from specimens of the Pacific Ocean are available in GenBank (retrieved as Adenoides eludens). We have also provided the first SSU- and LSU rDNA sequences from specimens of the European Atlantic coasts, where Amphidinium kofoidii was described. The aligned length for SSUrDNA was 1725 bp and 954 bp for LSU rDNA (D1– D3 region excluding the variable region). The sequences of the Atlantic and Pacific specimens of Amphidinium kofoidii were identical (Figs 63, 64). We provided the first SSU- and LSU rDNA sequences of Amphidinium eludens. All the sequences obtained from different samples were identical (Figs 63, 64). The species Amphidinium eludens and Amphidinium kofoidii described in Herdman (1922) are distantly related in the SSU rDNA (Fig. 63) and LSU rDNA trees (Fig. 64). Our data strongly support the splitting of these two species into two distinct genera based on the considerable evolutionary distance of their respective SSU- and LSU rDNA sequences. Because of the low resolution of basal positions of the phylogenetic trees, it is also difficult to establish the accurate relationships between these genera with other dinoflagellates (Figs 63, 64).

Downloaded by [Sistema Integrado de Bibliotecas USP] at 05:09 23 February 2015

F. Gómez et al.

10

Figs 57–62. Scanning electron micrographs of Pseudadenoides kofoidii gen. & comb. nov. Figs 57–58. Right lateral view. Fig. 59. Ventral view. Fig. 60. Apical view. Fig. 61. Dorsal view. Fig. 62. Detail of the pore and apical plates. Scale bar = 20 µm, except Fig. 62 where scale bar = 1 µm.

DISCUSSION In pioneering studies of sand-dwelling dinoflagellates, Herdman (1922) described in her fig. 1 Amphidinium eludens as forming discolourations in a beach of the British Isles. Despite the tradition of taxonomic studies in the European Atlantic coasts and the relative abundance of A. eludens, to date that species has remained under investigation. Confusion has been present since the original description. Herdman (1922) illustrated Amphidinium eludens as slightly larger than A. kofoidii in line drawings lacking any reference to the size and reported a length of 70 µm and 50–80 µm for A. eludens and A. kofoidii, respectively. However, in a note at the end of a subsequent publication (Herdman, 1923, p. 63) she explained that the size measurements were overestimated and should be 30 µm and 25–40 µm for Amphidinium eludens and A. kofoidii, respectively. Consequently, the cells of Amphidinium kofoidii tend to be larger than those of A. eludens. This

contrasts with the relative sizes of Herdman’s original figures. In our observations, we found an average length of 30 µm for Amphidinium eludens, which is usually slightly less than for A. kofoidii. The error in the relative size of the original description could have contributed to further confusion in the identification of both species. When Balech (1956) proposed Adenoides eludens, he considered that Herdman’s figure 1 and 2 corresponded to the same species. Dodge & Lewis (1986, p. 224) maintained this view when they reported ‘the cell contents of the two species is very similar it may be that they are only variants’. As a consequence, the species Amphidinium eludens has received little attention, and has been cited as Adenoides eludens sensu Balech (Paulmier, 1992) a situation that has deprived us from knowing the unusual morphology of Amphidinium eludens until now. Our specimens studied from a sandy beach in the English Channel correspond to Herdman’s fig. 1

11

Downloaded by [Sistema Integrado de Bibliotecas USP] at 05:09 23 February 2015

Pseudadenoides gen. & comb. nov. and Adenoides

Fig. 63. Maximum likelihood phylogenetic tree of Adenoides eludens and Pseudadenoides kofoidii with other dinoflagellates inferred from SSU rDNA sequences. The species newly sequenced in this study are bold. The numbers at each node represent bootstrap value (maximum likelihood, ML) and posterior probability (PP). Only values of > 70% (bootstrap) and > 0.9 (PP) are indicated. Closed circle denotes 100%/1.0 = bootstrap/PP.

12

Downloaded by [Sistema Integrado de Bibliotecas USP] at 05:09 23 February 2015

F. Gómez et al.

Fig. 64. Maximum likelihood phylogenetic tree of Adenoides eludens and Pseudadenoides kofoidii with other dinoflagellates inferred from partial LSU rDNA sequences. The species newly sequenced in this study are bold. The numbers at each node represent bootstrap value (maximum likelihood, ML) and posterior probability (PP). Only values of > 70% (bootstrap) and > 0.9 (PP) are indicated. Closed circle denotes 100%/1.0 = bootstrap/PP.

Downloaded by [Sistema Integrado de Bibliotecas USP] at 05:09 23 February 2015

Pseudadenoides gen. & comb. nov. and Adenoides reported as Amphidinium eludens. The shared characteristics with the original description including (1) general cell contour and size; (2) lack of apparent raised epitheca viewed from the lateral side; (3) the presence of a hump in the ventral contour of the cell where flagella arise; and (4) the position and shape of the nucleus and pyrenoids (Figs 1–6). The main characteristic of the redefined Adenoides eludens is the absence of the cingular groove. With the exception of Podolampas and allied genera, the absence of a cingular groove is a rare feature in planktonic species with a globular shape. Although there is no groove, the transversal flagellum nonetheless encircles the cell. As a general trend, the cell flattening of the benthic species makes it difficult to interpret the tabulation. This is even more difficult in some sand-dwelling dinoflagellates that have an incomplete cingulum. Adenoides is an extreme case, where the cingulum is absent or the sulcal left anterior plate could be interpreted as a vestigial first cingular plate. We can speculate an evolution towards the loss of the cingular groove as an adaptation to benthic habitats. In fact, dinoflagellates that lack the cingular groove such as Prorocentrum are widespread and diverse in benthic habitats. Then, we can speculate that the loss of the cingular groove may be an advantage in benthic habitats, and tentatively propose that the rotational movement of the transversal flagellum is less useful in interstitial space between the sand grains than in pelagic habitats. With the molecular phylogeny we can find a relationship between the planktonic and benthic dinoflagellates that is useful to interpret their tabulation. For example, as suggested by Saldarriaga et al. (2003), the sand-dwelling genus Roscoffia Balech is related to Podolampas (Gómez et al., 2010). The genera Amphidiniopsis or Herdmania are related to planktonic Protoperidinium-like cells such as Archaeperidinium minutum (Kofoid) E. Jørgensen (Gómez et al., 2011; Yamaguchi et al., 2011). The morphology of the apical pore of Adenoides is reminiscent of peridinioid genera such as Heterocapsa F. Stein or Azadinium Elbrächter & Tillmann (Tillmann et al., 2009). The tabulation of the epitheca of Adenoides is 5′ and 6″, but it can also be interpreted as 4′ and 6″ or 5′ and 7″. These are common tabulation features of dinoflagellates from different phylogenetic origins. Taxonomic sampling in the phylogenetic trees remains incomplete, and for the moment there are no dinoflagellate sequences close to Adenoides or Pseudadenoides. Numerous sand-dwelling dinoflagellates also remain as monotypic genera, lacking any known relative with which they can be classified, at least at the family level. The lack of the cingular groove in Adenoides or the absence of precingular plates in Pseudadenoides makes these genera particularly interesting for the evolution of dinoflagellates.

13 ACKNOWLEDGEMENTS We thank Pierre Compère and Gerry Moore for their useful advice on nomenclature.

FUNDING F.G. was supported by a UL1 post-doctoral grant and a CNRS convention of research, and an invited lecturer grant from Université Littoral-Côte Opale. F.G. is currently supported by the Brazilian Conselho Nacional de Desenvolvimento Científico e Tecnológico (grant number BJT 370646/2013-14).

AUTHOR CONTRIBUTIONS F. Gómez: collection, isolation, light and electron microscopy, drafting and editing manuscript; R. Onuma: molecular analysis; L.F. Artigas: collection and editing manuscript; T. Horiguchi: phylogenetic analysis and editing manuscript.

REFERENCES BALECH, E. (1956). Étude des dinoflagellés du sable de Roscoff. Revue Algologique, Nouvelle Serie, 2: 29–52. DAUGBJERG, N., HANSEN, G., LARSEN, J. & MOESTRUP, Ø. (2000). Phylogeny of some of the major genera of dinoflagellates based on ultrastructure and partial LSU rDNA sequence data, including the erection of three new genera of unarmoured dinoflagellates. Phycologia, 39: 302–317. DODGE, J.D. (1982). Marine Dinoflagellates of the British Isles. Her Majesty’s Stationery Office, London. DODGE, J.D. & LEWIS, J. (1986). A further SEM study of armoured sand-dwelling marine dinoflagellates. Protistologica, 22: 221–230. GÓMEZ, F. (2012). A quantitative review of the lifestyle, habitat and trophic diversity of dinoflagellates (Dinoflagellata, Alveolata). Systematics and Biodiversity, 10: 267–275. GÓMEZ, F. & ARTIGAS, L.F. (2014). High diversity of dinoflagellates in the intertidal sandy sediments of Wimereux (north-east English Channel, France). Journal of the Marine Biological Association of the United Kingdom, 94: 443–457. GÓMEZ, F., MOREIRA, D. & LÓPEZ-GARCÍA, P. (2010). Molecular phylogeny of the dinoflagellates Podolampas and Blepharocysta (Peridiniales, Dinophyceae). Phycologia, 49: 212–220. GÓMEZ, F., LÓPEZ-GARCÍA, P. & MOREIRA, D. (2011). Molecular phylogeny of the sand-dwelling dinoflagellates Amphidiniopsis hirsuta and A. swedmarkii (Peridiniales, Dinophyceae). Acta Protozoologica, 50: 255–262. HERDMAN, E.C. (1922). Notes on dinoflagellates and other organisms causing discolouration of sand at Port Erin. II. Proceedings and Transactions of the Liverpool Biological Society, 36: 15–30. HERDMAN, E.C. (1923). Notes on dinoflagellates and other organisms causing discolouration of the sand at Port Erin. III. Proceedings and Transactions of the Liverpool Biological Society, 38: 58–63. HOPPENRATH, M., SCHWEIKERT, M. & ELBRÄCHTER, M. (2003). Morphological reinvestigation and characterization of the marine, sand-dwelling dinoflagellate Adenoides eludens (Dinophyceae). European Journal of Phycology, 38: 385–394. HUELSENBECK, J.P. & RONQUIST, F. (2001). MrBayes: Bayesian inference of phylogenetic trees. Bioinformatics, 17: 754–755. MCNEILL, J., BARRIE, F.R., BUCK, W.R., DEMOULIN, V., GREUTER, W., HAWKSWORTH, D.L., HERENDEEN, P.S., KNAPP, S., MARHOLD, K., PRADO, J., PRUD’HOMME VAN REINE, W.F., SMITH, G.F., WIERSEMA, J.H. & TURLAND, N.J. (2012). International Code of Nomenclature

F. Gómez et al.

Downloaded by [Sistema Integrado de Bibliotecas USP] at 05:09 23 February 2015

for Algae, Fungi, and Plants (Melbourne Code). Regnum Vegetabile 154. Koeltz, Königstein. MURRAY, S. & PATTERSON, D.J. (2004). Cabra matta, gen. nov., sp. nov., a new benthic, heterotrophic dinoflagellate. European Journal of Phycology, 39: 229–234. NYLANDER, J.A.A., RONQUIST, F., HUELSENBECK, J.P. & NIEVESALDREY, J.L. (2004). Bayesian phylogenetic analysis of combined data. Systematic Biology, 53: 47–67. PAULMIER, G. (1992). Catalogue illustré des microphytes planctoniques et benthiques des côtes Normandes (Rapport interne RV-92.007RH). IFREMER, Issy-Les-Moulineaux. POSADA, M.A. & CRANDALL, K.A. (1998). Modeltest: testing the model of DNA substitution. Bioinformatics, 14: 817–818. SALDARRIAGA, J.F., TAYLOR, F.J.R., KEELING, P.J. & CAVALIER-SMITH, T. (2001). Dinoflagellate nuclear SSU rRNA phylogeny suggests multiple plastid losses and replacements. Journal of Molecular Evolution, 53: 204–213. SALDARRIAGA, J.F., LEANDER, B.S., TAYLOR, F.J.R. & KEELING, P.J. (2003). Lessardia elongata gen. et sp. nov. (Dinoflagellata,

14 Peridinales), Podolampaceae and the taxonomic position of the genus Roscoffia. Journal of Phycology, 39: 368–378. SWOFFORD, D.L. (2003). PAUP*. Phylogenetic Analysis using Parsimony (*and Other Methods). Version 4. Sinauer Associates, Sunderland, MA. TILLMANN, U., ELBRÄCHTER, M., KROCK, B., JOHN, U. & CEMBELLA, A. (2009). Azadinium spinosum gen. et sp. nov. (Dinophyceae) identified as a primary producer of azaspiracid toxins. European Journal of Phycology, 44: 63–79. YAMAGUCHI, A., KAWAMURA, H. & HORIGUCHI, T. (2006). A further phylogenetic study of the heterotrophic dinoflagellate genus Protoperidinium (Dinophyceae) based on small and large subunit ribosomal RNA gene sequences. Phycological Research, 54: 317–329. YAMAGUCHI, A., HOPPENRATH, M., POSPELOVA, V., HORIGUCHI, T. & LEANDER, B.S. (2011). Molecular phylogeny of the marine sanddwelling dinoflagellate Herdmania litoralis and an emended description of the closely related planktonic genus Archaeperidinium Jörgensen. European Journal of Phycology, 46: 98–112.