Xenobennettella coralliensis a new monoraphid diatom genus characterized by the alveolate sternum valve with cavum, observed from coral reef habitats

Fieldwork

In the Western Indian Ocean, fieldwork was carried out on the 26th–29th of April 2009 in Juan de Nova coral island (Fig. 1). Coral samples were retrieved by scuba diving and preserved in alcohol (GPS data 16°59.34′S, 42°47.37′E) collected on April 28th 2009 from a water depth of 20 m.

In the South Pacific, samples were collected from Nukutavake, a small atoll located in the eastern part of the Tuamotu archipelago (Fig. 1). Nukutavake (GPS data 19°16.83′S 138°47.11′W) is 5 km in length, 13.5 km² in surface area with a sediment-filled lagoon. Samples were obtained by a scuba diver on the 22nd October to 3th November 2016, on the reef slope (<5 m deep) of the atoll. Small amounts of detritus coral sands or epiphytes on dead corals were collected in small vials (50 ml) and preserved with alcohol.

Light microscopy (LM)

For light microscopy, the samples were washed with distilled water to remove salts, treated with 30% H₂O₂ (hydrogen peroxide) for 2 h at 70 °C to remove organic matter, followed by the addition of ca. 10ml of 10% HCl to remove calcium carbonate, and rinsed several times in distilled water, alcohol-desiccated and mounted on glass slides using Naphrax® (Brunel Microscopes Ltd, Wiltshire, U.K.). Slides were examined with a Zeiss Axio Scope 100 A1 (Carl Zeiss, Jena, Germany) and Zeiss Axio Imager 2 (Carl Zeiss, Jena, Germany) light microscopes (LM), with differential interference contrast (DIC) optics at University of Perpignan and University of Szczecin, respectively. For the examination of raw or cleaned material by scanning electron microscopy (SEM), the samples were filtered through 1 µm Nuclepore® filters and rinsed twice with deionized (milliQ) water to remove salts. Filters were air-dried and mounted onto aluminum stubs before coating with gold-palladium.

Electron microscopy (EM)

Ultrastructural analysis was made with scanning and transmission electron microscopy. For the SEM examination, a drop of the cleaned sample was filtered onto Whatman Nuclepore polycarbonate membranes (Fisher Scientific, Schwerte, Germany). The filters were air-dried overnight, mounted onto aluminum stubs, and coated with gold-palladium alloy (EMSCOP SC 500 sputter coater) and examined with a Hitachi S-4500 (Hitachi, Tokyo, Japan) field emission SEM operated at 5 kV, calibrated with a TGX01silicon grating (C2M) at University of Perpignan. Additional SEM observations were made at the Goethe University in Frankfurt am Main using a Hitachi S-4500 (Hitachi, Tokyo, Japan). For TEM observations, a drop of cleaned material was left to evaporate on a copper grid at room temperature. TEM observations were done at the Warsaw University of Technology, Faculty of Materials Science and Engineering, using a Hitachi SEM/STEM S-5500 (Hitachi, Tokyo, Japan), in which the specimens were simultaneously observed in scanning and transmission mode.

Focus Ion Beam (FIB)

The FIB sectioning of the sternum valve was performed by means of a FIB/SEM Hitachi NB5000 integrated system. This system consists of an ultra-high performance focused Ga+ ion beam gun (40 kV) and high-resolution field emission gun scanning electron microscope (30 kV). This dual beam system enabled high-throughput specimen preparation, high resolution imaging and analysis, as well as precision nano-milling. During the FIB milling, particular parameters of ion beam conditions were selected to minimize the damage to our samples. Prior to cutting, a Tungsten (W) layer was applied for protection. The valve cut by FIB was selected from a sample taken from Juan de Nova from which the holotype of Xenobennettella coralliensis sp. nov. originated.

Terminology and abbreviations

The terminology used follows Anonymous (1975), Ross et al. (1979) and Round, Crawford & Mann (1990). As previously proposed by Riaux-Gobin & Romero (2013) abbreviations are as follows: for the valves with a raphe, designated the raphe valve, is RV, for valves without a raphe, designated the sternum valve is SV; for the sternum valve valvocopula is SVVC, for the RV valvocopula is RVVC.

Description

EM (Figs. 3–7)

Sternum valve (SV; Figs. 3, 4 and 7): valve external surface generally domed with narrow, plain mantle (Fig. 3B). Abrupt transition from valve surface to mantle. Sternum relatively broad, lanceolate in shape and concave, expanding towards valve centre into relatively broad unilateral central area (Figs. 3A–3F). Alveolate transapical striae parallel in the middle, becoming radiate towards apices, 24–32 in 10 µm with coarse transapical ribs (virgae) between stria (Fig. 3F). The transapical ribs connected with short and robust apically oriented viminae (Fig. 7C). Each alveola opens internally on margin by/via a round foramina (Figs. 7C and 7E). Externally, striae composed of biseriate rows of small, densely packed areolae (55–65 in 10 µm) that became triseriate rows towards valve margin, forming a quincunx pattern over whole SV surface. Areolae occluded by hymenes bearing radiate slits; hymenes positioned somewhat below valve surface (Figs. 6C–6D). Internally, alveolae covered with a structureless siliceous membrane and open into valve interior near valve mantle (Figs. 4 and 7C). The alveolae openings (or foraminae) small and oblong to elliptic in shape (Figs. 4 and 6A–6B). In the mid-valve interior, sealed to unilateral central area, distinctive relatively large cavum with a thin solid siliceous wall. Cavum interior flat and embedded with the same thin siliceous membrane that encloses the alveola in whole SV interior (Figs. 4, 6A, 7A and 7D). Girdle narrow composed of plain bands with narrow SVVC with short triangular fimbriae (Figs. 3D, 3E and 6B).

Raphe valve (RV; Fig. 5): valve external surface flat with somewhat elevated margins and raphe system. Sternum very narrow, separated from valve face by a row of small, densely packed areolae, up to 60–65 in 10 µm (Figs. 5B and 5D). Raphe filiform with external proximal raphe endings slightly expanded and approximate to each other (Fig. 5A). Apical raphe endings terminate on valve mantle (in a few cases observed below valve apex) and bent into the same direction (Figs. 5A–5B). Transapical striae close to valve margin composed of pyramidal groups of small areolae, grouped in triseriate rows and decrease towards valve center to biseriate or dispersed, solitary areolae (Figs. 5A–5B). On valve surface, each stria delineated by a slightly elevated rib that continues up to raphe sternum. Transapical striae at valve center parallel, becoming radiate towards apices, 24–26 in 10 µm. Areolae small, circular, internally occluded by hymenes bearing radiate slits, the same as observed in SV (Fig. 5E). Valve margin internally with structures resembling chambers, but open to cell lumen (Figs. 6E–6F). Moreover, internal valve margin strongly bent towards valve interior, with rather robust ribs between stria (Fig. 6A). Unfortunately, despite the numerous analyzed valves, we did not clearly observe the RV valvocopula (RVVC). The raphe system distinctly elevated, raphe slit opens laterally. Internal proximal raphe endings bent in opposite directions, whereas distal raphe ends terminate in small, simple helictoglossa at apices (Figs. 5C–5D).

Taxonomic position of Xenobennettella

Since the publication of Round, Crawford & Mann (1990), monoraphid genera primarily included in two major and heterogenous Achnanthes sensu lato and Cocconeis sensu lato genera have been split into a number of genera based on the re-evaluation of distinctive and shared morphological characteristics. The major common feature of the latter taxa was their heterovalvy, with both a RV and a SV, while other characteristics e.g., the girdle including the valvocopulae, orientation and shape of areola, valve outline or central area structure—showed a great deal of morphological variation. Taxonomic revisions of Achnanthes Bory and Cocconeis date back to the end of 19th century (e.g., Cleve & Grunow, 1880), but continued into the 20th century (Holmes, 1985; Hustedt, 1933; Hustedt in Schmidt, 1937; Lange-Bertalot & Ruppel, 1980). Round, Crawford & Mann (1990) resurrected the “forgotten” genera, e.g., Achnanthidium Kützing and Eucocconeis Cleve. However, the former genus was defined too broadly and further taxonomic work was required to conform the taxa included in this genus to the generitype Achnanthidium microcephalum Kützing. In the following papers, the diagnosis of Achnanthidium was refined further and several new genera established, including Planothidium Round & Bukhtiyarova, Psammothidium Bukhtiyarova & Round, Karayevia Round & Bukhtiyarova ex Round and Kolbesia Round & Bukhtiyarova ex Round (Round & Bukhtiyarova, 1996; Round, 1998). The process of transferring taxa from Achnanthes and Cocconeis sensu lato into morphologically appropriate, newly-established genera has continued over the last three decades and additional new genera have been established: Lemnicola Round & Basson, Pogoneis Round & Basson (Round & Basson, 1997), Astartiella Witkowski, Lange-Bertalot & Metzeltin, Amphicocconeis (De Stefano & Marino, 2003), Scalariella (Riaux-Gobin, Witkowski & Ruppel, 2012), Gliwiczia (Kulikovskiy, Lange-Bertalot & Witkowski, 2013) and Madinithidium Witkowski & Desrosiers (Desrosiers et al., 2014). Despite these revisions, small groups of achnanthoid and cocconeid taxa remain without detailed generic accommodation. An example of such a group is described here as a new genus Xenobennettella.

Our original sampling in the Scattered Islands involved only a collection of material preserved in 70% alcohol, which excludes any chance of the isolation of single specimens and establishing a clonal culture and thus the use of molecular markers. Its systematic position is based on several characteristics crucial for the monoraphid diatoms. Cox (2015) included Achnanthaceae in the Order Mastogloiales, whereas the remaining monoraphid clades were placed into the Order of Cocconeidales and two Families Cocconeidaceae and Achnanthidiaceae (cf. Round, Crawford & Mann, 1990). Taking into consideration the characteristics shared by Xenobennettella with other monoraphid genera, i.e., (1) presence of sternum and cavum on one SV valve, (2) the chambered internal RV margin and alveolate SV and (3) internal proximal raphe ends bent in opposite directions on RV, plus areola occlusions with short radiate slits, we place the new genus in a clade with Planothidium, which belongs to Achnanthidiaceae (Cox, 2015).

Xenobennettella key characteristics for identification

Presence of cavum (1). Xenobennettella is the third monoraphid genus bearing this structure and is the first purely marine organism with a cavum. In Planothidium, a cavum or rimmed depression in the SV valve occurs facultatively and has never been observed in marine forms (Riaux-Gobin et al., 2011; Riaux-Gobin, Compère & Al-Handal, 2011; Witkowski, Lange-Bertalot & Metzeltin, 2000). Gliwiczia includes purely freshwater forms but differs from the two other cavum bearing genera possessing it on both SV and RV (Kulikovskiy, Lange-Bertalot & Witkowski, 2013; Potapova, 2016). In all of the above listed genera, the presence of cavum is manifested on valve surface as a fascia. Among the particular genera, cavum shows 3-D structural differences. In Xenobennettella, the cavum expands from valve margin towards sternum and occupies valve inner surface corresponding to an external unilateral fascia. In Planothidium, cavum is more or less similar to Xenobennettella as shown in numerous examples illustrated by Jahn et al. (2017) and Stancheva et al. (2020). In Planothidium taxa, when present, it expands from valve margin towards sternum and its presence is marked by unilateral fascia on valve exterior. However, cavum is significantly different in Gliwiczia for which here we refer to as Achnanthes calcar Cleve—the name used by Montgomery (1978) when he observed the first ever SV of Xenobennettella in Florida waters. Achnanthes calcar has been transferred in Gliwiczia (Cleve) by Kulikovsky, Lange-Bertalot & Witkowski. As raised in Kulikovskiy, Lange-Bertalot & Witkowski (2013), but cf. Potapova (2016), no RV of Achnanthes calcar was illustrated in Cleve (1891). Fortunately, the original material studied by Cleve (1891) from Finland has been sent in the past to the National Academy of Sciences in Philadelphia and in slides made from this material RV of Achnanthes (Gliwiczia) calcar was illustrated (cf. Kulikovskiy, Lange-Bertalot & Witkowski, 2013; Potapova, 2016). The cavum in Gliwiczia and in G. calcar, in particular, is different from Xenobennettella and Planothidium, firstly occurring on both RV and SV, secondly it is rather small and made of a thick silica layer. It extends from the valve margin to ca. $\frac{1}{2}$ distance to mid sternum. An internally terminating cavum prolongs into a distinct and elevated fascia that shows a decrease towards the opposite valve margin. Like in Xenobennettella and Planothidium, cavum in Gliwiczia is expressed externally, but in the latter genus on valve face as a fully developed fascia with a somewhat diminishing size towards the opposite valve margin. This is completely different when compared to Planothidium and Xenobennettella that both have unilateral fascia. With the unique major characteristics presented above for Xenobennettella, as typified by X. coralliensis, this is an interesting discovery. This new diatom genus was found in the sublittoral zone of coral reefs of the Indo-Pacific, habitats that are fertile grounds for undiscovered biodiversity and diatom morphologies (e.g., Al-Handal, Compère & Riaux-Gobin, 2016; Kryk et al., 2021; Lobban et al., 2012; Riaux-Gobin, Compère & Al-Handal, 2011; Riaux-Gobin, Compère & Al-Handal, 2011; Risjani et al., 2021). The diatom genus we studied bears a set of characteristics previously seen in monoraphid genera from very different ecologies: the freshwater Planothidium and Gliwiczia and the epizoic Bennettella and Epipellis (Denys & De Smet, 2010; Ferrario et al., 2019). Many monoraphid diatom species are characterized by the presence of a horseshoe (based on a German term Hufeisen), but renamed recently as hood or cavum (https://diatoms.org/genera/planothidium/guide). SEM observations show that this structure is a rimmed depression resembling a cave. The first cavum structures were observed in Achnanthes taxa, which are now transferred to Planothidium Round & Bukhtiyarova and Gliwiczia Kulikovskiy, Lange-Bertalot & Witkowski. Whereas Planothidium possesses a cavum only on the SV, Gliwiczia species possess it on both valves (Krammer & Lange-Bertalot, 1988; Kulikovskiy, Lange-Bertalot & Witkowski, 2013; Round, Crawford & Mann, 1990). In terms of habitat, all known taxa bearing a cavum have been recorded in freshwater (most Gliwiczia occur in Lake Baikal) or slightly brackish-waters (Planothidium), making Xenobennettella the first exclusively marine taxon to bear this structure with a specific coral reef habitat from the Indo-Pacific and Florida Keys. The cavum structure has been observed thus far only in Planothidium (in some species) and Gliwiczia (in all species; Table 1).

Valve structure (2). Whereas in LM the new genus due to the presence of a cavum may resemble Planothidium (but also with the proximal internal raphe endings bent in opposite directions and hymenate areolae occluded with slits), its SV alveolate structure eliminates the taxa from the latter genus. Xenobennettella coralliensis, as the first representative of the genus has clearly chambered internal RV margin and alveolate SV (Table 1), which rules out any resemblance between the two genera. Our newly described genus shares several features with Bennettella with regards to the alveolate SV ultrastructure. In both genera, the alveolae are internally closed with solid siliceous coverings and possess marginal foraminae. Comparing to Bennettella the key is also transapical striae of the SV and RV of Xenobennettella, which are composed of small areolae arranged in bi- to triseriate rows with each stria delineated by a transapical rib visible on the valve surface and interior. It is worth to remember, that internally, the raphe sternum is elevated over the valve surface in both genera, with the raphe slit opening laterally (see Ferrario et al., 2019). In addition, in the Bennettella RV, the fascia is complete over the whole valve (bilateral) and axially distorted, which is simple in Xenobennettella. Interestingly, Xenobennettella is also the first record of this structure associated with an alveolate valve, as Gliwiczia and Planothidium are both characterized by a simple porous valve ultrastructure.

Raphe system (3). The last major difference between other genera and our newly described genus is the RV raphe system. In other similar genera, the raphe is sinusoid and terminates below the apices whereas in Xenobennettella it is straight and its apical ends are bent to the same side at the apices with raphe branches terminating on the valve mantle (externally; Table 1).

Distribution and ecology

Xenobennettella was common in coral sand from 20 m deep at Juan the Nova, a coral island in the Mozambique Chanel, west of Madagascar, but also observed in Florida (Montgomery, 1978, Pl. 3: D). Montgomery (1978) imaged a single SV of Xenobennettella coralliensis from the coral sand of Florida Keys and identified it as Achnanthes calcar. Due to the fact that Achnanthes (Gliwiczia) calcar inhabits freshwaters, Montgomery possibly considered this taxon as redeposited from a terrestrial habitat. Thus far we have observed two species of Xenobennettella with X. coralliensis in Juan the Nova Island of the Western Indian Ocean and possibly a second species in the Tuamotu Archipelago in the South Pacific.

Our research continues to document the seemingly independent evolution of many ultrastructural characteristics across diatoms. Apparently, the cavum in Xenobennettella is possibly another example of homoplasy in diatom valve ultrastructure, along with the alveolate cross-section. With the documented paraphyly of monoraphid diatoms (Davidovich et al., 2017; Górecka et al., 2021; Riaux-Gobin et al., 2021), we are hopeful that we will be able to test in the future the hypothesis that this genus may represent a secondary gain of the monoraphid frustule close to the pinnularioid diatoms, based on similarities in the alveolate valve ultrastructure.