This paper reports on a new species of the Baikal endemic sponge (fam. Lubomirskiidae) Swartschewskia khanaevi sp. nov. The description of this species is based on morphological and molecular data (ITS and mitochondrial IGRs). Morphologically, S. khanaevi sp. nov. differs from S. papyracea by loose tracts arranged in an irregular network as well as the presence on strongyles of compound spines looking like tubercles densely ornamented with simple spines. Moreover, specimens of S. khanaevi sp. nov. show a peculiar structure of the aquiferous system at the body surface that may be an adaptive trait for environmental conditions. Phylogenetic analysis has revealed that S. khanaevi sp. nov. forms a well-supported (0.99) monophyletic clade with S. papyracea and is allocated as its sister taxa.

ITS, mitochondrial IGRs, morphological analysis, Swartschewskia

Baikal is the most ancient and deepest lake on the Earth with the huge water volume. The lake is considered to be 25–30 million years old (Mazepova 1995); its maximum depth is 1641 m, and the volume of water body exceeds 23 000 km³ (Mats et al. 2001). Due to these facts, the lake is characterised by minor environmental oscillations. The family Lubomirskiidae represents the most spectacular example of endemic radiation in freshwater sponges under the specific conditions of the great lake. According to molecular phylogeny the endemic family Lubomirskiidae was monophyletic and diverged from Spongillidae (Itskovich et al. 2008; Maikova et al. 2012; Maikova et al. 2017). Existence in a stable environment over a long period of time resulted in the disappearance of gemmulation from the life cycle of Lubomirskiidae (Efremova and Goureeva 1989; Manconi and Pronzato 2002).

At present, 14 species are allocated to the family Lubomirskiidae (Efremova 2001, 2004; Manconi and Pronzato 2019). The actual number of species is most likely underestimated. Comprehensive morphological study of the Baikal sponges revealed at least five forms that showed a constant set of morphological characteristics but could not be related to any known species (Khanaev et al. 2018). In this regard, these forms were suggested to be new species. Additionally, some specimens with uncommon morphology have been described (Weinberg 2005). We also observed several sponge specimens having unusual or transitional traits that interfered with precise species identification in our previous study (Khanaev et al. 2018).

The gaps in our knowledge of Lubomirskiidae morphology and taxonomy concern some aspects in the biology of the Baikal sponges. The absence of gemmules, gemmuloscleres, and parenchymal microscleres, which often contribute to taxonomy, complicates species identification (Manconi and Pronzato 2002). Moreover, the majority of the Lubomirskiidae species were described in the late 19^th – early 20^th century. The descriptions were limited to the classical taxonomy based on diagnostic morphological characters and were often very brief.

The genus Swartschewskia Makuschok, 1927 is clearly segregated from other Lubomirskiidae genera (Dybowski 1880, Swartschewsky 1901, Makushok 1927a). Only Swartschewskia is characterised by cortex: an ectosomal skeleton, tangential arrangement of primary tracts and stout bent strongyles as megascleres. The genus includes two species: S. papyracea (Dybowski, 1880) and S. irregularis (Swartschewsky, 1902). Swartschewskia papyracea is widely distributed in the depth range of 1–80 m in Baikal. Swartschewskia papyracea morphology was reported in a number of works (Swartschewsky 1902, Makushok 1927a, Rezvoy 1936, Manconi and Pronzato 2002, 2015, 2019; Weinberg 2005, Masuda 2009). On the contrary S. irregularis is extremely rare species inhabiting sublittoral zone of Baikal (70–150 m). The species was described based on the single specimen that was not preserved (Swartschewsky 1902). During the next 120 years only one sponge specimen with a similar morphology was found but no data on its morphology were published (Efremova 2001).

Molecular approach is also limited for phylogenetic studies of the Baikal sponges due to low variability of markers (COI, ITS, rRNA-genes) usually applied for this purpose throughout the world (Itskovich et al. 2008, Harcet et al. 2010). Recently (Maikova et al. 2015, 2016), the protein coding sequences of mitochondrial DNA were used to study the phylogenetic relationship of Baikal sponges only at the genera taxonomic level. This study showed that the nucleotide substitution rate of intergenic regions (IGRs) of the Baikal sponge mtDNA is significantly higher than coding sequences, which makes them very promising for phylogenetic reconstructions of closely related species (Lavrov 2010, Maikova et al. 2012). However, only concatenated nuclear (ITS-regions) and mitochondrial (IGRs) data allowed us to separate closely related species of the family Lubomirskiidae (Maikova et al. 2017). Therefore, in this study we use ITS and mitochondrial IGRs sequences to investigate the phylogenetic position of a new species within the family Lubomirskiidae.

During the 2016 expeditions, unusual sponges were sampled in Olkhonskiye Vorota Strait. These sponges were identified as a separate species based on their morphological and molecular phylogenetic data. The paper describes a new species of Swartschewskia and we present additional data on the morphology of Swartschewskia papyracea (Dybowski, 1880) and provide diagnostic keys for the species belonging to the genus Swartschewskia.

Two species, S. papyracea and S. irregularis, were included in the genus Swartschewskia before the present study. We used the following sources for comparative analysis of diagnostic morphotraits. The original description of S. papyracea was made by W. Dybowski (1880) on several exemplars. Afterwards the type material was lost (Efremova 2001). Due to the impossibility of comparing our data with type material we were guided by the generally accepted recent descriptions of Manconi and Pronzato (2002, 2019). Additionally, we studied the morphology of S. papyracea specimens from the type locality of S. khanaevi sp. nov. Data on S. irregularis morphology are extremely poor. For more than 200 years of studies of the Baikal sponges, only two specimens of S. irregularis were collected. Both specimens are no longer available; therefore, we relied on the original description of the species (Swartschewsky 1902).

Based on molecular data, the new species belongs to the genus Swartschewskia (fam. Lubomirskiidae). The limitations of the molecular approach were previously shown for phylogenetic studies of the Baikal sponges due to low variability of markers (COI, ITS, rRNA-genes) usually applied for this purpose throughout the world (Itskovich et al. 2008, Harcet et al. 2010). The protein coding sequences of mtDNA allowed phylogenetic relationships within Lubomirskiidae to be resolved only at the genera taxonomic level (Maikova et al. 2015, 2016). The mtDNA intergenic regions, as we suggested, could be suitable for the separation of closely related species of Baikal sponges (Maikova et al. 2012) due to their increased rate of substitution accumulation (Lavrov 2010). But on the phylogenetic tree based on mtDNA intergenic regions, the species of Baikal sponges did not form separate clades at the species level, with some exceptions (deep-sea sponges, for example) (Maikova et al. 2012). The concatenated nuclear (ITS-regions) and mitochondrial (IGRs) data were most suitable for studying phylogenetic relationships within the family Lubomirskiidae at the moment (Maikova et al. 2017). The limitation of the molecular approach is apparently related to the low evolutionary rate of both nuclear and mitochondrial DNA of Baikal sponges (Lavrov 2010), which is enhanced by the relatively recent divergence of many species. An exception is the S. papyracea, which shows the acceleration of the accumulation of nucleotide substitutions in mtDNA to be twice relative to other species of the family Lubomirskiidae (Maikova et al. 2015). We hypothesise that this species is one of the most ancient of the existing species. On the phylogenetic tree based on protein-coding mtDNA genes, S. papyracea is closer to a common ancestor than all other species of the Baikal sponges (Maikova et al. 2016). In this study, based on concatenated nuclear (ITS) and mitochondrial (IGRs) data, the maximum interspecific genetic distances were between S. papyracea and other Lubomirskiidae species. Within the genus Swartschewskia the intraspecific and interspecific genetic distances do not overlap. This shows the genetic subdivision of the species within the genus and the genetic isolation of the genus Swartschewskia within the family Lubomirskiidae. The division of the new species into two groups inside the Clade B is not reflected in their morphology.

Swartschewskia khanaevi sp. nov. has skeleton structure and spicules typical for the genus (Manconi and Pronzato 2002, 2019). Swartschewskia khanaevi sp. nov. differs from S. papyracea by the clustering of pores in pore fields, less ordered structure of the ectosomal skeleton and unusual secondarily microspined tuberculated spines on strongyles. In S. papyracea distribution of dermal pores is uniform, ectosomal skeleton is highly ordered alveolar network with polygonal meshes, spicules bear spines grouped in rosettes (Manconi and Pronzato 2002, 2019). Generally, oscula are similar in both species, but S. papyracea has an alternative rather rare kind of osculum, which is sieve-like. It consists of a number of exhalant apertures inside rounded vallum (Manconi and Pronzato 2002). Pore fields of S. khanaevi sp. nov. could hardly be misinterpreted as sieve-like oscula. The total shape of the latter is always roundish; apertures are packed very closely and are noticeably larger than dermal pores (mean size 281 × 219 μm vs 42 × 54 μm). Ectosomal skeleton of S. irregularis even less ordered than of S. khanaevi sp. nov. It lacks polygonal network and looks like randomly arranged spicules. Strongyles of S. irregularis are smooth.

Previous data on the morphology of Swartschewskia species do not contain records of strongyles ornamented with tuberculated spines or pore fields (Dybowski 1880; Swartschewsky 1901, 1902; Makushok 1927a; Rezvoy 1936; Manconi and Pronzato 2002, 2019). Weinberg (2005) mentioned an unusual specimen of S. papyracea as a thin encrusting sponge with numerous oscula and bearing: (a) spicules as stout and bent strongyles thinner than usual in the species and ornamented with massive complex spines; (b) skeleton of clearly divided ectosomal and choanosomal parts, but intensive study of the skeleton was not carried out. The sponge was collected from the Maloye More Strait (the precise locality was unknown). Based on these facts, we suppose that Weinberg met a specimen of S. khanaevi sp. nov. Taking into account sites, where S. khanaevi sp. nov. was collected, the species is most likely a local endemic of the Olkhonskiye Vorota Strait or Maloye More Strait as a whole.

Fossil spicules similar to S. khanaevi sp. nov. were found in the Late Pliocene sediments (interval of 3.2−2.8 Ma) of Lake Baikal (Veynberg 2009). They were described as spicules of extinct species Palaeoswartschewskia sp. 1, they were some thinner and longer than in S. khanaevi sp. nov. and had complex spines. These spines, with a smaller number of simple spines than tuberculated spines and less expressed tubercles (see Veynberg 2009), represent intermediate variant between rosette spines of S. papyracea and tuberculated spines of S. khanaevi sp. nov. In this regard, we cannot ascertain the unambiguous identity of Palaeoswartschewskia sp. 1 and S. khanaevi sp. nov., but these two species are doubtlessly morphologically close.

Non-uniform localisation of pores in S. khanaevi sp. nov. is uncommon amongst the Baikal sponges. Normally, in lubomirskiids pores are evenly distributed throughout the sponge surface. The bottom at the study site consists of stones (a substrate for sponges) and sandy areas located nearby. The latter saturate the water with suspended grains of sand. The number of suspended particles combined with hydrodynamic activity can lead to clogging of the aquiferous system (Bell et al. 2015). The concentration of inhalant pores at restricted areas of the body surface was previously described as an adaptive trait of some sponge species living under the conditions of high sedimentation (Rützler 1974; Werding and Sanchez 1991; Pronzato et al. 1998). A larger size of inhalant apertures of S. khanaevi sp. nov. in comparison with S. papyracea can also prevent clogging.

The presence of sessile ciliates and dense aggregation of diatom algae on the sponge surface is not common for Lubomirskiidae. Isolated diatom algae can be observed sometimes on the lubomirskiids surface. There are no descriptions of mass diatom accumulations on the surface of a number of specimens. Any attached ciliates on sponges in Baikal also have never been mentioned. However, ectosymbiotic sessile ciliates of the Lagenophrys genus were described on Baikal endemic amphipods cuticle (Khalzov et al. 2018). Probably the emergence of a unique epibiotic community on S. khanaevi sp. nov. is possible due to unusual structure of aquiferous system. Permanent exhalant and inhalant water currents are normally presented at the sponge surface and prevent its colonisation. In S. khanaevi sp. nov., exhalant and inhalant apertures are concentrated in restricted areas. Therefore, up to 80 % of the body surface has no currents, which is a favourable substrate for epibiotic organisms.

A new species Swartschewskia khanaevi sp. nov. was described based on morphological traits and sequences of nuclear (ITS1 and ITS2) and mitochondrial (IGRs) markers. In the molecular phylogeny the specimens of S. khanaevi sp. nov. are clustered within a well-defined group containing S. papyracea as the most closely related species. Indeed, the specimens’ morphological traits clearly indicate their belonging to Swartschewskia: well-developed ectosomal skeleton of tangential spicular fibres and sparsely developed choanosomal skeleton, stout bent strongyles as megascleres. The major morphological traits that distinguish S. khanaevi sp. nov. from other congeners are the structure of ectosomal skeleton and compound spines on strongyles. Swartschewskia khanaevi sp. nov. was sampled only from the Olkhonskiye Vorota Strait, and we assumed it to be a local endemic of this strait. We suggest the non-uniform localisation of pores on the sponge surface may be an adaptation to biotope conditions.

This study was performed within the framework of the State Tasks Nos. 0345-2019-0002 and 0345-2019-0009 and supported by the RFBR grant No. 19-04-00787A. The authors are grateful to Dr Igor V. Khanaev and Valery I. Chernykh for samples collection. Authors thank Dr Tatyana Ya. Sitnikova (Limnological Institute SB RAS, Russia), who organised the expedition in 2003 and identified ciliates on the sponge surfaces, and Dr Nina A. Bondarenko (Limnological Institute SB RAS, Russia) for identification of diatom algae on sponges. SEM analyses were carried out in the Center ‘Ultramicroanalysis’ (Limnological Institute SB RAS).