Biodiversity Data Journal 11: e100525 CO) doi: 10.3897/BDJ.11.e100525 pian ees Research Article Polyphasic characterisation of Microcoleus autumnalis (Gomont, 1892) Strunecky, Komarek & J.R.Johansen, 2013 (Oscillatoriales, Cyanobacteria) using a metabolomic approach as a complementary tool Ivanka Teneva?, Detelina Belkinova?'§, Tsvetelina Paunova-Krasteval, Krum Bardarov!, Dzhemal Moten?, Rumen Mladenov', Balik Dzhambazov+ + Faculty of Biology, Plovdiv University “Paisii Hilendarski’, Plovdiv, Bulgaria § Institute of Biodiversity and Ecosystem Research, Bulgarian Academy of Sciences, Sofia, Bulgaria | The Stephan Angeloff Institute of Microbiology, Bulgarian Academy of Sciences, Sofia, Bulgaria q InoBioTech Ltd., Sofia, Bulgaria Corresponding author: lvanka Teneva (teneva@uni-plovdiv.bg) Academic editor: Christian Wurzbacher Received: 13 Jan 2023 | Accepted: 04 Apr 2023 | Published: 21 Apr 2023 Citation: Teneva |, Belkinova D, Paunova-Krasteva T, Bardarov K, Moten D, Mladenov R, Dzhambazov B (2023) Polyphasic characterisation of Microcoleus autumnalis (Gomont, 1892) Strunecky, Komarek & J.R.Johansen, 2013 (Oscillatoriales, Cyanobacteria) using a metabolomic approach as a complementary tool. Biodiversity Data Journal 11: e100525. https://doi.org/10.3897/BDJ.11.e100525 Abstract As a result of the continuous revision of cyanobacterial taxonomy, Phormidium autumnale (Agardh) Trevisan ex Gomont, 1892 has been transferred to the genus Microcoleus as Microcoleus autumnalis (Gomont, 1892) Strunecky, Komarek & J.R.Johansen, 2013. This transfer was based on a single strain and literature data. In the present study, we revise the taxonomic position of Microcoleus autumnalis by applying the classical approach of polyphasic taxonomy and additionally using metabolomics. Cyanobacterial strains identified as Phormidium autumnale and Microcoleus vaginatus (type species of the genus Microcoleus) were used for comparative analyses. In addition, the taxonomic relationship between the species Phormidium autumnale and Phormidium uncinatum was determined © Teneva | et al. This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. 2 Teneva | et al on the basis of polyphasic characteristics. Monitoring of the morphological variability of Phormidium autumnale and Microcoleus vaginatus strains showed a difference in the morphology concerning the ends of the trichomes, the shape of the apical cells, as well as the presence/absence of the calyptra and its shape. The performed TEM analysis of the thylakoid arrangement of the studied strains showed parietal arrangement of the thylakoids in the representatives of genus Phormidium and fascicular arrangement in genus Microcoleus. Molecular genetic analyses, based on 16S rDNA, revealed grouping of the investigated P autumnale strains in a separate clade. This clade is far from the subtree, which is very clearly formed by the representatives of the type species of genus Microcoleus, namely M. vaginatus. The metabolomic analysis involving P autumnale and M. vaginatus strains identified 39 compounds that could be used as potential biochemical markers to distinguish the two cyanobacterial species. Based on the data obtained, we suggest changing of the current status of Microcoleus autumnalis by restoring its previous appurtenance to the genus Phormidium under the name Phormidium autumnale (Agardh) Trevisan ex Gomont, 1892 and distinguishing this species from genus Microcoleus. Keywords Cyanobacteria, Phormidium autumnale, polyphasic, morphology, ultrastructure, TEM, 16S, phylogeny, metabolomics, biochemical markers Introduction In recent years, the taxonomy and systematics of the phylum Cyanobacteria have been actively revised and reorganised, based on new data gained mainly from different molecular genetic studies (Komarek 2006, Komarek et al. 2014, Komarek 2016, Strunecky et al. 2022). Such a process is typical for cyanobacteria, but it also affects a number of other plant, fungal and animal taxa. The problem is that sometimes as the established rules (when they exist) are interpreted in a subjective way and often giving priority to the new features, we neglect the well-functioning old, which in most cases are traditionally established and accepted (Komarek 2020). In the taxonomy of cyanobacteria, the polyphasic approach is most often applied (Anagnostidis and Komarek 1985, Hoffmann et al. 2005). This approach combines molecular-genetic, morphological, ultrastructural, biochemical and environmental data, with priority given to the molecular genetic data, while others are considered complementary (Komarek 2009, Komarek 2016). What happens in practice? Based on mainly molecular genetic data combined in most cases only with cytomorphological features, new genera are separated (often with only 2-3 representatives) and widespread species are renamed. There is no complete set of data to confirm and convincingly show the need for this change (Komarek 2020). Even assuming that the principles of polyphasic taxonomy are followed, it should be kept in mind that the 16S rDNA sequence is not a marker that allows for subgeneric identification and that the use of other genetic markers in solving taxonomic cases should not be ignored (Moten et al. 2017). Polyphasic characterisation of Microcoleus autumnalis (Gomont, 1892) Strunecky, ... 3 The presence of crypto- and morphospecies amongst representatives of the Cyanobacteria should not be overlooked. The use of complete morphological, ultrastructural, biochemical and ecological data should be a requirement in determining the taxonomic position of a certain taxon. Otherwise, taxonomic changes may occur that are contradictory and not sufficiently justified. The main targets are polyphyletic genera, such as Phormidium, Microcoleus and Leptolyngbya and taxa in which morphological criteria overlap and are not sufficiently descriptive to make definite decisions. The current study was provoked by another taxonomic change related to the species Phormidium autumnale (Agardh) Trevisan ex Gomont, 1892, which was renamed in 2013 to Microcoleus autumnalis (Gomont, 1892) Strunecky, Komarek & J.R.Johansen, 2013 (Strunecky et al. 2013). Both genera, Phormidium Kutting ex Gomont and Microcoleus Desmaziéres ex Gomont, are polyphyletic and rich in species within the order Oscillatoriales (Palinska et al. 2011, Stoyanov et al. 2014). They are amongst the earliest described genera of order Oscillatoriales (Gomont, 1892). The type species of the genus Phormidium is P. lucidum Kutzing ex Gomont (Geitler 1942) and the type species of the genus Microcoleus is M. vaginatus (Vaucher) Gomont (Geitler 1942, Drouet 1968). Although there is an available 16S rDNA sequence from the type species P. /ucidum in the GenBank, in the last updated classification of cyanobacterial orders and families, it is noted that reliable sequencing data for the type species of genus Phormidium (P._ lucidum) are missing and/or their phylogenetic placement is ambiguous (Strunecky et al. 2022). Phormidium is one of the most difficult cyanobacterial genera from a taxonomic point of view (Komarek and Anagnostidis 2005). It consists of numerous morphotypes with many transitional forms. The relatively wide range of different morphological species previously assigned to the genera Phormidium, Oscillatoria and Lyngbya were reorganised by Komarek and Anagnostidis (2005) into eight different morphological groups, which differ in the morphology of the apical ends of trichomes. The difficulty in distinguishing between the genera Phormidium and Microcoleus using the classical approach comes from the lack of sufficiently descriptive morphological criteria. According to the literature, P autumnale and M. vaginatus do not differ in cell size. The range of variation in the length and width of their cells overlap. According to Strunecky et al. (2013), the morphological difference between P autumnale and M. vaginatus is only in the form of colonies and organisation of the filaments, but the morphology of the trichomes is very similar (Strunecky et al. 2013). There is a difference in their habitats. P autumnale is a freshwater species distributed mainly in streams, rivers and waterfalls, but also amongst the overgrowth (periphyton) on underwater substrates. The main habitat of M. vaginatus is the soil. The situation is more complicated because the data obtained by molecular approaches are often inconsistent with the taxonomy based on the morphological studies (McAllister et al. 2016). Marquardt and Palinska showed that cyanobacterial strains morphologically assigned as P. autumnale are genetically different and grouped into different clusters (Marquardt and Palinska 2007). Studies of cyanobacterial blooms in New Zealand, based 4 Teneva | et al on the polyphasic approach, have reported dominance of genus Phormidium and presence of significant morphological variability between the blooming strains (P autumnale and P uncinatum Gomont ex Gomont, 1892), while other studies of the same strains, based on 16S rDNA, have identified P autumnale as the dominant species (Heath et al. 2010, Harland et al. 2014, McAllister et al. 2016). These results unequivocally show that further in-depth studies of the genus Phormidium and, in particular, P autuamnale are required. However, it is clear that this cannot happen, based only on morphology and molecular genetic criteria. The inclusion of ultrastructural characteristics (e.g. thylakoid arrangement) as well as biochemical criteria (Specific metabolites) in the process of characterisation of these taxa would help to clarify this problem. Thylakoid models are recognisable and useful in distinguishing morphologically simple single-celled and filamentous species. In addition, the various modifications in the arrangement of thylakoids are apparently related to the cryptogenera, in which their ultrastructural modification may correlate with the phylogenetic position (Komarek 2016, Mares et al. 2019). The biochemical criteria in general have always been poorly represented in the polyphasic characterisation of cyanobacteria. The reason is, on the one hand, insufficient scientific information and, on the other, the unclear taxonomic value of these criteria. Thus, in our Opinion, the demonstration of the applicability of metabolic analyses for polyphasic characterisation of cyanobacteria is very useful. Studying the metabolites of P autumnale and M. vaginatus strains, here we demonstrate the possibility of using metabolic analyses for polyphasic characterisation of cyanobacteria. In the present study, applying the classical polyphasic approach and metabolomic analysis, we showed that P autumnale and M. vaginatus belong to different genera and the Classification of Phormidium autumnale as Microcoleus autumnalis is incorrect. In addition, based on the polyphasic characterisation, we determined that the studied strains of P autumnale and P. uncinatum are different species belonging to genus Phormidium. Materials and methods Strains and culture conditions A total of 11 cyanobacterial strains from three collections were used in the present study: seven strains from the Plovdiv Algal Culture Collection (PACC), Paisii Hilendarski University of Plovdiv, Bulgaria; three strains from the Culture collection of Autotrophic Organisms (CCALA) of the Institute of Botany of the Czech Academy of Sciences, Trebon and one strain from the Culture Collection of Algae (SAG) at the University of Gottingen, Germany. Cyanobacteria were cultured for 1 month under sterile conditions (75 cm? culture flasks, TPP, Trasadingen, Switzerland) in liquid alkaline Z-nutrient medium (Staub 1961), with a photoperiod of 12 h/12 h light/dark, at a light intensity of 10 umol photons s*! m~ provided by 40 W cool-white fluorescent lamps. This cultivation was carried out in order to accumulate the cyanobacterial mass necessary for the study (morphological analysis, Polyphasic characterisation of Microcoleus autumnalis (Gomont, 1892) Strunecky, ... 5 transmission electron microscopy (TEM), preparation of extracts, DNA isolation and molecular genetic analyses. Morphological analysis Investigated cyanobacterial strains belong to three species: Microcoleus autumnalis (Gomont) Strunecky, Komarek & J.R.Johansen 2013 (previously Phormidium autumnale (Agardh) Trevisan ex Gomont 1892), Microcoleus vaginatus Gomont ex Gomont 1892 and Phormidium uncinatum Gomont ex Gomont 1892. Data for the strains originally identified as Phormidium autumnale and Phormidium uncinatum, as well as for the Microcoleus vaginatus strains, are presented in Table 1. Table 1. Origin of the investigated strains. Strain Habitat Location Isolated by Phormidium autumnale PACC S-crater Nr. 237 England, Surtsey Schwabe, 5 Aug 1968 5505 Phormidium autumnale PACC Lyophilized Germany Steubing, 30 Nov 5511 ampoule 1967 Phormidium autumnale PACC Lyophilized Germany Sprecht, 8 Dec 1967 5517 ampoule Phormidium autumnale PACC Moss cultures Germany Schwartz-Kraepelin, 5522 21 Nov 1968 Phormidium autumnale PACC Spillway Germany, Siegburg Clasen, 17 Mar 1969 5527 Phormidium autumnale PACC Meadow Germany, Schwabe, 7 May 5529 Solling mountains 1968 Microcoleus vaginatus CCALA Unknown Switzerland, Verzascatal Zehnder, 1964 145 Microcoleus vaginatus CCALA River Germany, Hamburg Marvan, 1966 152 Microcoleus vaginatus CCALA Rice field China, Hubei, Wuhan Cepak, 1991 757 Microcoleus vaginatus SAG 2211 — Soil, desert USA, New Mexico, Sevielleta Lewis, Apr 2002 LTER Phormidium uncinatum PACC Veleka river Bulgaria, Sinemorets Mladenov, 5 Oct 1987 8693 Morphological analyses were performed using a Magnum-T microscope equipped with a high-resolution 3 Mpx Si-3000 XLiCap digital camera and software (Medline Scientific Ltd., Chalgrove, UK). In the course of the work, photo documentation of the examined samples was also performed. At magnifications of 100+1000x, the variability of the following phenotypic features during the exponential growth phase of the strains was monitored: shape of the filaments; sheaths - presence and condition; trichomes - colour, shape of the 6 Teneva | et al trichome ends, mobility, presence/absence of granulations; presence of constrictions at the cross-walls, shape of the cells; apical cell of the trichome - shape, calyptra (presence/ absence, shape of the calyptra). Cell measurements: length (L) and width (W). The measurements were performed on a minimum of 50 cells. Transmission electron microscopic (TEM) analysis of thylakoid arrangement in the cells of selected strains Cultured strains were harvested by centrifugation at 3000xg for 5 min. Cyanobacterial filaments were washed with 0.1 M cacodylate buffer and fixed with 4% glutaraldehyde in 0.1 M cacodylate buffer at pH 7.2 for 4 h at 4°C. Then, the samples were washed three times with 0.1 M cacodylate buffer, after which they were fixed with 1% osmium tetroxide in 0.1 M cacodylate buffer at room temperature for 1 h. Cyanobacteria were pelleted by centrifugation and embedded in 1% agarose and cut into small cubes. Dehydration was done in an ascending alcohol series: 30%, 50%, 70%, 90%, 95% ethanol for 15 min each, 100% ethanol (2x) for 30 min and propylene oxide once for 30 min and one more time for 15 min. The dehydration was followed by impregnation with propylene oxide and resin (durcupan): propylene oxide:resin 2:1 for 30 min, propylene oxide:resin 1:1 for 30 min, propylene oxide:resin 1:2 for 30 min and pure resin overnight. The samples were polymerised at 56°C for 48 hours. Ultra-thin sections of 60-70 nm in size were cut using an ultramicrotome Reichert (Reichert-Jung Ultracut E Ultramicrotome, Optische Werke AG, Vienna, Austria). Sections were mounted on copper grids for electron microscopy and counterstained with 1% uranyl acetate in 70% methanol for 15 min, followed by Reynold's lead citrate for 20 min (Reynolds 1963). The prepared sections were examined in a_ high-resolution transmission electron microscope HR STEM JEOL JEM 2100 (JEOL Ltd., Tokyo, Japan) operating at 200 kV, equipped with a CCD camera GATAN Orius 832 SC1000 (Gatan GmbH, Munchen, Germany). DNA isolation, PCR amplification and sequencing Genomic DNA was extracted from 40 mg of fresh cyanobacterial mass using the xanthogenate-SDS (XS) extraction protocol of Tillet & Neilan (Tillett and Neilan 2001) or by the Proteinase-K extraction assay. DNA concentration and its purity were measured using a NanoDrop 2000 UV-VIS spectrophotometer (Thermo Fisher Scientific, Wilmington, DE, USA). The isolated DNA was visualised on an agarose gel by ethidium bromide and UV transillumination (MiniBIS Pro gel documentation system, DNR Bio-Imaging Systems Ltd., Jerusalem, Israel). 16S rDNA was amplified by using the primers 16S-16C-R (5- AAGGAGGTGATCCAGCCGCA-3’) and 16S-1R-F (5-AGAGTTTGATCCTGGCTCAG -3’) (Wilmotte et al. 1992, Seo and Yokota 2003 A PuReTaq™ ReadyToGo Beads kit (GE Healthcare, Buckinghamshire, UK) was used for the PCR reaction, including 1.5 U Tag DNA polymerase, 10 mM Tris-HCI pH 9, 50 mM KCI, 1.5 mM MgCl. and 200 uM dNTP. Five pmol of both primers, 100 ng genomic DNA and DEPC-water were added to the mix for each reaction to a final volume of 25 ul. Polyphasic characterisation of Microcoleus autumnalis (Gomont, 1892) Strunecky, ... 7 Amplification was carried out in a TC-412 thermocycler (Techne, Cambridge Ltd., UK) using the following programme: DNA denaturation for 5 min at 94°C, followed by 30 cycles of 60 s at 95°C, 60 s at 53°C (hybridisation) and 2 min at 72°C (elongation). The reaction was completed with an elongation step of 10 min at 72°C. The obtained PCR-products were analysed by electrophoresis in a 1.5% agarose gel in Tris-Acetate-EDTA buffer (TAE). GeneRuler™ 100 bp DNA Ladder Plus (Thermo Fisher Scientific Baltics UAB, Vilnius, Lithuania) was used as a size marker. Gels were visualised with ethidium bromide and UV light. After visualisation, the correct PCR products were excised from the gel and the isolated DNA was purified using a PureLink™ PCR Purification Kit (Thermo Fisher Scientific Baltics UAB, Vilnius, Lithuania). Purified 16S rDNA products were sent for sequencing to Eurofins Genomics Germany GmbH (Ebersberg, Germany). Sequencing was conducted by using the same primers as for the PCR amplification. Obtained 16S nucleotide sequences were compared with the available 16S sequences for other cyanobacterial strains in the NCBI database using BLAST (hittos://blast.ncbi.nim.nih.gov/, accessed on 11 November 2022). All new 16S rDNA sequences from this study were deposited in the GenBank (National Center for Biotechnology Information, NCBI) under accession numbers OP626168 — OP626173 ( OP626168 for Lyngbya aerugineo-coerulea PACC 8601, currently Potamolynea aerugineo-caerulea; OP626169 for P autumnale PACC 5505; OP626170 for P. autumnale PACC 5511; OP626171 for P autumnale PACC 5517; OP626172 for P autumnale PACC 5527; OP626173 for P autumnale PACC 5529). Phylogenetic analyses For the purposes of the phylogenetic analyses, 16S rDNA sequences of identified and determined at the species level representatives of the genera Phormidium, Microcoleus, Oscillatoria, Arthrospira, Kamptonema, Trichodesmium, Dapis, Thychonema, Wilmottia, Capilliphycus, Neolyngbya and Affixifilum were retrieved from the NCBI database. Thus, the sequences of those members determined only at the generic level were not included in the analyses. The multiple alignment of the selected nucleotide sequences (106 sequences with 1531 nucleotide sites) was carried out by the MAFFT version 7 (Katoh et al. 2017) (https:// mafft.cbrc.jp/alignment/server/, accessed on 5 December 2022). Phylogenetic analyses were performed by using Maximum Likelihood (ML) and Neighbour-joining (NJ) methods with MEGA 7 (Kumar et al. 2016) and Bayesian approach with MrBayes v. 3.2.7a (Ronquist et al. 2012). The search for the best fitting models, which is a part of the phylogenetic software package MEGA 7 (Kumar et al. 2016), indicated that the Kimura 2-parameter model (K2+G+l) (Kimura 1980) is the most suitable for the analyses. This model was applied in the Maximum Likelihood (ML) and Neighbour-joining (NJ) analyses with four rate categories of the gamma distribution. The Bayesian estimation of phylogeny (Huelsenbeck and Ronquist 2001, Ronquist et al. 2012) was performed with MrBayes v. 3.2.7a on XSEDE (CIPRES, hitps:/(www.phylo.org, accessed on 5 December 2022). Two runs of eight Markov chains were calculated for ten million generations with sampling every 1000 8 Teneva | et al generations. The first 25% of the sampled trees were discarded as burn-in. Consensus phylogenetic trees were reconstructed using the MEGA 7 software. All analyses were performed with 1000 bootstrap repetitions with a total of 1531 positions in the dataset. Gloeobacter violaceus (FR798924) was used as an outgroup to root the trees. Metabolomic analysis Biomasses (500 mg) from three P. autumnale strains (PACC 5522, PACC 5527, PACC 5529) and three V. vaginatus strains (CCALA 145, CCALA 152, CCALA 757) were used for extraction of polar and non-polar metabolites. The extraction procedures and LC-MS analysis were carried out as previously described (Teneva et al. 2022). Briefly, freeze-dried cyanobacterial biomasses from the selected strains were mixed with 3 ml MeOH followed by an ultrasonic bath extraction (Branson 5510R-DTH, Wilmington, NC, USA) for 20 min and consequently, 6 ml of chloroform (for 20 min on a shaker) and 3 ml of Milli-Q water were added. After centrifugation at 4000 rpm for 20 min, the methanol/chloroform fractions were collected and filtered through 0.20 um Millex-FG hydrophobic PTFE filters (Merck KGaA, Darmstadt, Germany). Only methanol/chloroform fractions (containing non-polar compounds) were used for the LC-MS analysis. Two microlitres of each of the fractions were analysed on a Q Exactive LC-MS/MS system (Thermo Fisher Scientific, Waltham, MA, USA) composed from an Accela quaternary HPLC pump with an Accela autosampler and an HRMS Q-Exactive detector with H-ESI electrospray. The reverse phase (RP) chromatographic separation was performed on a Kinetex EVO C18 150 mm x 3 mm, 2.6 um core-shell column (Phenomenex Inc., Torrance, CA, USA). Mobile phases, mass spectral conditions and data treatment are described in detail by Teneva et al. (2022). MS/MS spectra for annotated compounds with significant fold changes (analysed by the Perseus framework of the MaxQuant proteomics software package, hitps://maxquant.net/ maxquant/, accessed on 11 November 2022) and acceptable p-value (< 0.05) between selected strain groups (P autumnale and M. vaginatus) were subjected to a FISh coverage processing, SIRIUS MS/MS processing (https://bio.informatik.uni-jena.de/software/sirius/, accessed on 11 November 2022) and MS Finder Search (http://prime.psc.riken.jo/compms/ msfinder/main.html, accessed on 11 November 2022). A limited number of compounds were validated manually by comparison with experimentally obtained or simulated MS/MS spectra from the METLIN script (Guijas et al. 2018) and MZ cloud databases, if available. Any data processing of metabolites outside Compound Discoverer was made using Xcalibur™ 2.2 (Thermo Fisher Scientific, Hemmel, UK). Statistical analysis Data (excluding metabolomics) were presented as mean + standard deviation (SD). Differences between the samples were evaluated by analysis of variance (ANOVA) and considered significant when p < 0.05. Quantitative MS data were statistically analysed and visualised by using the Perseus software package (hittps://maxquant.net/perseus/, Polyphasic characterisation of Microcoleus autumnalis (Gomont, 1892) Strunecky, ... SS) accessed on 11 November 2022). Hierarchical clustering analysis and heat map were applied to group the quantified compounds, based on their abundance after Z-score normalisation and subtraction of mean values. Two-sample t-tests, combined with permutation False Discovery Rate (FDR) to correct for multiple testing, were used. Volcano plot display was used to visualise data. Results Morphological analysis By applying the principles of the morphological approach, a description of the studied strains and measurements of their cells were performed at the beginning of the study. Morphological description of Phormidium autumnale strains Data from the performed morphological analysis are presented in Table 2. Table 2. Variability of morphological characters in Phormidium autumnale strains. S — sheath; M — motility; K — keritomy (net-like structure); L/W — mean cell length / mean cell width; (+) — presence; (+) — facultative presence. * According to Mares et al. (2019). Strain S M K L/_ Trichome Apical cells Calyptra Thylakoid W_ ends arrangement * Phormidium + + + 0.6 gradually and — elongated, rounded, weakly _ parietal thylakoids with autumnale PACC slightly rounded conical, § expressed a central fascicle 5505 narrowed slightly curved Phormidium + + + 0.6 gradually and elongated, obtuse- rounded, weakly _ parietal thylakoids autumnale PACC slightly conical, slightly expressed composed of 5511 narrowed curved peripheral fascicles Phormidium + + + 0.6 gradually and elongated, obtuse- rounded simple parietal autumnale PACC slightly conical, slightly arrangement 5517 narrowed curved Phormidium + + + 0.8 gradually and — elongated, truncated parietal thylakoids with autumnale PACC slightly rounded conical, a central fascicle; 5522 narrowed slightly curved simple parietal arrangement Phormidium + + + 1.0 gradually and _ slightly elongated, truncated or parietal thylakoids with autumnale PACC slightly curved rounded a central fascicle; 5527 narrowed simple parietal arrangement Phormidium + + + 0.8 gradually and — elongated, rounded, weakly _ parietal thylakoids with autumnale PACC slightly rounded conical, § expressed or a central fascicle 5529 narrowed slightly curved absent Thallus blue-green to dark greyish-green, forming a thin velvety membrane. Free-floating or attached to the walls of the culture flask, but also developing above the boundary of the 10 Teneva | et al membrane separating the nutrient medium from the air (aerophilic), forming creeping tufts. With ageing, the thallus detaches from the walls and floats in a common dark-green to yellowish-green mucilaginous mass on the surface of the culture flask. Filaments long, cylindrical + straight or curved and tightly interwoven heterogeneous or + parallel in places (Fig. 1A). Sheaths thin, mucilaginous, soft or clear, facultative, sometimes obscure or diffluent, colourless to amorphous, enclosing only one trichome. Figure 1. EES Photomicrographs of Phormidium autumnale strains. A, B P autumnale PACC 5517; C P autumnale PACC 5527; D P. autumnale PACC 5529. Magnification 400; Scale bar - 20 um. Trichomes bright blue-green to yellowish-green, 3.3-4.0 um wide (mean value), motile, slightly constricted at the granulated cross-walls, gradually attenuated towards ends (Fig. 1 B-D). Cells usually shorter than wide, cylindrical to + isodiametric (L/W = 0.6-1.0), with visible chromatoplasma and centroplasma or keritomised (Fig. 1D). Presence of necroidic cells. Apical cells elongated, rounded conical, curved, with rounded calyptra (Fig. 1B-D). Specific characteristics: 1) Trichomes 3.3-4.0 um wide (mean value), slightly constricted at cross-walls; cells short-cylindrical to + isodiametric (LAV = 0.6-1.0). 2) Visible chromatoplasma and centroplasma or keritomised. 3) Trichome ends gradually and slightly narrowed. 4) Apical cells elongated, with a rounded conical shape, slightly curved. 5) Calyptra weakly expressed, with a rounded shape or absent. Morphological description of Microcoleus vaginatus strains Summarised data from the performed morphological analysis are presented in Table 3. Thallus bright olive-green, dark green to black, forming fascicles at the surface and walls of Polyphasic characterisation of Microcoleus autumnalis (Gomont, 1892) Strunecky, ... 11 the culture flask. Old cultures form free-floating yellowish-green mucilaginous aerophytic or subaerophytic masses on the surface and separate yellowish aerophytic fascicles on the walls of the culture flask. Table 3. Variability of morphological characters in Microcoleus vaginatus strains. S — sheath; M — motility; K — keritomy (net-like structure); L/W — mean cell length / mean cell width; (+) — presence. * According to Mares et al. (2019). Strain S M K LI Trichome ends Apical Calyptra Thylakoid Ww cells arrangement * Microcoleus + + + 0.6 abruptly narrowed, capitate rounded to fascicular vaginatus CCALA curved to S-shaped hemispherical arrangement 145 contorted Microcoleus + + + 0.5 slightly narrowed, capitate flat to hemispherical fascicular vaginatus CCALA slightly curved arrangement 152 Microcoleus + + + 0.6 abruptly narrowed, capitate hemispherical fascicular vaginatus CCALA curved arrangement 757 Microcoleus + + + 0.6 abruptly narrowed, capitate conical, obtuse to fascicular vaginatus SAG curved to S-shaped hemispherical arrangement 2211 contorted Filaments long, cylindrical, straight or slightly curved, indiscriminately or in places + parallel arranged (Fig. 2A). Sheaths thin, mucilaginous, clear, colourless, enveloping one trichome. Sometimes diffluent, forming a shapeless yellowish mass. Trichomes bright blue-green, with keritomised contents, 4.6-5.6 um wide (mean value), motile, not constricted at the granulated cross-walls (Fig. 2C). The ends of the trichomes abruptly and strongly narrowed, curved to S-shaped contorted (Fig. 2C). The attenuation affects the last few cells, not just the apical one. Cells usually short cylindrical (LAV = 0.5-0.6), rarely + isodiametric, 2.6-3.3 ym long. Presence of necroidic cells. Apical cells capitate, with conical, obtuse to hemispherical calyptra (Fig. 2C, D). Specific characteristics: 1) Sheaths thin, mucilaginous, clear, colourless, enveloping one trichome. 2) Trichomes not constricted at cross-walls, 4.6-5.6 um wide (mean value). 3) Trichome ends abruptly and strongly narrowed (last few cells), curved to S-shaped contorted. 4) Cells short cylindrical (L/W = 0.5-0.6), rarely + isodiametric, keritomised. 5) Apical cells capitate, with conical, obtuse to hemispherical calyptra. Morphological description of Phormidium uncinatum PACC 8693 Thallus bright blue-green, forming fascicles and tufts on the surface of the nutrient medium. Old cultures black-green, tufts retain their positions in the culture flask. Filaments long, cylindrical + straight, indiscriminately or in places + parallel arranged (Fig. 3). Sheaths thin, mucilaginous, soft, obscure or diffluent, colourless to amorphous. Trichomes bright blue- green, 6.0-9.0 um wide (mean value), motile, not constricted or slightly constricted at the 12 Teneva | et al granulated cross-walls, abruptly narrowed towards the ends which are curved (Table 4). Cells short cylindrical, always distinctly shorter than wide (length % to % of the width), 1-4 um long. Apical cells capitate, with rounded conical calyptra (Fig. 3). Table 4. Variability of morphological characters in Phormidium autumnale strains. S — sheath; M — motility; K — keritomy (net-like structure); L/W — mean cell length / mean cell width; (+) — presence; (+) — facultative presence. * According to Mares et al. (2019). Strain S M K LI | Trichome ends Apical Calyptra Thylakoid Ww cells arrangement * Phormidium uncinatum + + + 0.3. abruptly narrowed, capitate rounded conical _ simple parietal PACC 8693 curved calyptra arrangement Figure 2. EESI Photomicrographs of Microcoleus vaginatus strains. A M. vaginatus CCALA 757; B M. vaginatus CCALA 145; C M. vaginatus CCALA 152; D M. vaginatus SAG 2211. Magnification 400~x; Scale bar - 10 um. Specific characteristics: 1) Trichomes 6-9 um wide, not constricted or slightly constricted at the cross-walls, abruptly narrowed towards the ends; 2) Cells always short cylindrical (length % of the width); 3) Apical cells capitate, with rounded conical calyptra. The culture strain corresponds phenotypically to P uncinatum. Polyphasic characterisation of Microcoleus autumnalis (Gomont, 1892) Strunecky, ... 13 Figure 3. EES Photomicrographs of Phormidium uncinatum PACC 8693. Magnification 400x; Scale bar - 10 um. Cell sizes According to literature data, the species Phormidium autumnale and Microcoleus vaginatus do not differ in cell size. The range of variation in the length and width of their cells overlaps (2-4 x 4-7 um and 2-5 x 3-7 um, respectively). All the strains we examined, originally designated as P autumnale and M. vaginatus, had similar cell sizes and fell within the range of variation of the two species. Results of the cellular measurements of the investigated strains are summarised in Table 5. Table 5. Cell sizes of the studied strains. RD — reference data; SD — standard deviation. Strain Length of the cells Width of the cells Mean Min Max SD Mean Min Max SD (um) (um) (um) (um) (um) (um) Phormidium autumnale (RD*) 2.0—4.0 - 5.0 —- 4.0-7.0 3.5 - - Phormidium autumnale PACC 2.5 1.5 4.0 0.7 3.7 3.0 4.0 0.5 5505 Phormidium autumnale PACC 2.4 2.0 3.0 0.5 3.8 3.0 4.0 0.4 5511 Phormidium autumnale PACC 2.3 2.0 3.0 0.5 4.0 3.0 5.0 0.2 5517 Phormidium autumnale PACC 3.1 2.0 5.0 0.6 3.9 3.0 4.5 0.4 5522 Phormidium autumnale PACC 3.2 2.0 6.0 0.9 3.3 2.0 4.0 0.6 5527 Phormidium autumnale PACC 3.0 2.0 5.0 0.6 4.0 3.0 5.0 0.4 5529 Microcoleus vaginatus (RD*) 2.0—5.0 - 6.7 - 3.0-7.0 2.5 9.0 - 14 Teneva | et al Strain Length of the cells Width of the cells Mean Min Max SD Mean Min Max SD (ym) (ym) (ym) (ym) (ym) (ym) Microcoleus vaginatus CCALA 145 2.8 2.0 4.0 0.6 46 4.0 5.0 0.5 Microcoleus vaginatus CCALA 152 2.6 1.0 4.0 0.7 56 4.0 7.0 0.7 Microcoleus vaginatus CCALA 757 3.0 2.0 4.0 0.7 4.9 4.0 5.0 0.4 Microcoleus vaginatus SAG 2211 3.3 2.0 6.0 0.9 5.2 4.0 6.0 0.5 Phormidium uncinatum (RD*) 2.0-6.0 2.0 6.0 - 5.5-9.0 4.0 9.5 - Phormidium uncinatum PACC 2.6 1.0 4.0 0.6 7.5 6.0 9.0 0.7 8693 *Komarek & Anagnostidis (2005). Transmission electron microscopy (TEM) analysis For decades, the thylakoid arrangement has been used in the classification of cyanobacteria as one of the key features for defining taxa. TEM analyses are becoming a regular part of the polyphasic characterisation of cyanobacteria, accounting for the fine structure of multiple strains. A recent comprehensive study by Mares et al. (2019) mapped the ultrastructural data of more than 200 cyanobacterial strains and classified the thylakoid arrangement. Based on visual evaluation of the TEM dataset, the types of thylakoid arrangements were divided into eight categories: 1 - thylakoids absent, 2 - parietal, 3 - radial, 4 - fascicular, 5 - parallel, 6 - irregular, 7 - Cyanothece-like, 8 - unknown or ambiguous (Mares et al. 2019). In the strains of Phormidium autumnale that we studied, the thylakoid system was organised more or less parietal (Fig. 4, Table 2). Thylakoids were usually aggregated parallel along the cell walls (Fig. 4), but often form central fascicles (Fig. 4F, G). In contrast to the parietal arrangement of thylakoids observed in the representative strains of Phormidium autumnale, in the strains of Microcoleus vaginatus, the thylakoids were characterised by a fascicular arrangement (Fig. 5). Thylakoids in Phormidium uncinatum have also parietal arrangement (Fig. 6). Phylogenetic analysis based on 16S rDNA Phylogenetic reconstructions, based on 16S rDNA (Fig. 7A), showed that investigated Phormidium autumnale strains (PACC 5505, PACC 5511, PACC 5517, PACC 5522, PACC 5527, PACC 5529, marked in bold in the phylogenetic tree) are grouped in a separate clade. This clade was supported by high bootstrap values (0.95/99/68 bootstrap support). The rest of the Phormidium autumnale strains that were used in the phylogenetic analyses formed a sister clade including also other Phormidium species. These clades are far from the subtree clearly formed by the representatives of the type species of the genus Microcoleus, namely Microcoleus vaginatus (Fig. 7B). This is further evidence supporting Polyphasic characterisation of Microcoleus autumnalis (Gomont, 1892) Strunecky, ... 15 our hypothesis, based on the morphological and TEM analyses, that Phormidium autumnale has been incorrectly transferred to the genus Microco/eus under the name Microcoleus autumnalis. \n addition, data from the metabolomic analyses also showed significant differences between the investigated Phormidium autumnale and Microcoleus vaginatus strains. Figure 4. EES] Ultrastructure of strains originally identified as Phormidium autumnale with characteristic thylakoid arrangement. A parietal thylakoids with a central fascicle in P autumnale PACC 5505: B parietal thylakoids composed of peripheral fascicles in P autumnale PACC 5511; C, D parietal thylakoids in P- autumnale PACC 5522 (varies to simple parietal); E, F parietal thylakoids with a central fascicle in P autumnale PACC 5527; G parietal thylakoids with a central fascicle in P autumnale PACC 5529; H parietal thylakoids in P autumnale PACC 5517. The type species of genus Phormidium (Phormidium lucidum Kutzing ex Gomont, 1892) was grouped together with Phormidium chlorinum (Kutzing ex Gomont 1892) Umezaki and Watanabe 1994 in a distinct clade (Fig. 7A). Taking in account that most oscillatorian genera are polyphyletic, the phylogenetic topology was congruent with the traditional genera defined by morphological features. Aside from Phormidium and Microcoleus, here we included representatives with the type species of other sister genera belonging to the family Microcoleaceae (Tychonema, Dapis, Kamptonema, Trichodesmium, Arthrospira) and family Sirenicapillariaceae (Capilliphycus, Neolyngbya, Affixifilum). In addition to genus Phormidium, from Oscillatoriaceae were included representatives of genus Oscillatoria. Most cyanobacterial strains belonging to one genus were clustered together and formed 16 Teneva | et al separate clades. Although the investigated strain Phormidium uncinatum PACC 8693 was clustered within the Phormidium clade, its position is not supported by the bootstrap values. 1 pm Figure 5. EES Ultrastructure of strains originally identified as Microcoleus vaginatus with fascicular arrangement of the thylakoids. A, B VV. vaginatus CCALA 145; C, D M. vaginatus CCALA 152; E, F MV. vaginatus CCALA 757; G, H M. vaginatus SAG 2211. A longitudinal section. B-H cross sections. A 1 soym Figure 6. EESl Ultrastructure of Phormidium uncinatum PACC 8693. A, B Parietal arrangement of the thylakoids. Polyphasic characterisation of Microcoleus autumnalis (Gomont, 1892) Strunecky, ... 17 E4588 @ Microcoleus clade 0.9/68/54 JQ712610 Microcoleus vaginatus POR 7 a i 4 artes b 0.79/59/62 — JQ712632 Microcoleus vaginatus CCALA 757 ME ane aoantd 3} = a JQ712631 Microcoleus vaginatus CCALA 145 dopalan an 0.77/52/51 JQ712605 Microcoleus vaginatus P09 nas: (Oscillatoriaceae) i KF770968 Phormidium papyraceum PACC 8600 is EF654076 Microcoleus vaginatus SEV1-KK3 ac) CAWBG523 =— al JQ712633 Microcoleus vaginatus CCALA 152 Se 2.87 0.94/58/53 | 0.878078 JQ712638 Microcoleus vaginatus $13 = CYNI06 EF654074 Microcoleus vaginatus SAG 2211 = 0.869999] pita ___0.66/59/627 ——L__ EF654064 Microcoleus vaginatus CNP3-KK2 & 1 fee -—____— EF654071 Microcoleus vaginatus NV1-KK1 5 a ape 0.75/52/54 | >-——_ MT887222 Microcoleus vaginatus Prim-3-1 BS 7 NR 112168 Oscillatoria nigro-viridis PCC 7112 0.95/93/90) 0.83/58/56 — EF654079 Microcoleus vaginatus UBI-KK2 | EF654062 Microcoleus vaginatus CJI-U2-KK1 —— JQ712618 Microcoleus vaginatus SLad22 Oscillatoriaceae se P89 */99/99 NR176515 Capilliphycus tropicalis pene [ QRG2EITS Phoradiais autemuate FAOC 5529 0. a MF190468 Capilliphycus tropicalis ALCB 114392 "999 Ope2st7l Phormidium aurunnate PACC S817 0.8/80/77 NR 176576 Capilliphycus flaviceps BLCC-M137-— | 0 a og aeeinNRIT? Sanaa tal */99/99 NR 176504 Capilliphycus salinus Capilliphycus ° */99/99 KY824052 Capilliphycus salinus ALCB114379 NR 176577 Capilliphycus guerandensis BLCC-M76 Microcoleaceac P */99/97 KY824056 Neolyngbya maris-brasilis ALCB 114380 Eggs */98/99 NR177709 Neolyngbya maris-brasilis */99/99 i MF190466 Neolyngbya arenicola ALCB 114386 Neolyngbya MF190470 Neolyngbya irregularis ALCB 114389 ‘\Nitpataad iad */98/99 MF190467 Neolyngbya tenuis ALCB 114390 ee 0.95/86/87 2 ee MT371826 Affixifilum floridanum BLCC-M62 n PEW NR 177658 Affixifilum floridanum */99/99 -———— MT371830 Affixifilum granulosum BLCC-M110_ | Affixifilum Mi 0.79/72/74 | MF190471 Affixifilum granulosum ALCB 114393 CTC AG30054. | Arthrospira 0.96/94/90 KY824053 Affixifilum granulosum ALCB 114378 B 9604 (Microcoleaceae) je (Capilliphycus, Neolyngbya, Affixifilum) 0.9: 0479B TAI2-3-DP02 Yyy1400 0.84/54/91 5 0.16 Jossitatoriacea | KC217 F798 Figure 7. (A) Maximum Likelihood (ML) phylogenetic tree, based on 106 sequences (1531 aligned positions) of the 16S rDNA gene of representatives from the genera Phormidium, Microcoleus, Oscillatoria, Arthrospira, Kamptonema, Trichodesmium, Dapis, Thychonema, Wilmottia, Capilliphycus, Neolyngbya and Affixifium. Bootstrap support values are shown as Bayesian posterior probability (BPP) / ML / NJ more than 0.50 or 50%. Asterisks indicate a BPP of 1.00. Strains used in the current study are in bold. The sequence of Gloeobacter violaceus was used to root the tree as an outgroup. (B) Exported two subtrees from the main phylogenetic tree (Figure 7A) obtained by Maximum Likelihood (ML) analysis, based on 16S rDNA. Bootstrap support values are shown as Bayesian posterior probability (BPP) / ML / NJ more than 0.50 or 50%, respectively. Asterisks indicate a BPP of 1.00. Black bold type indicate strains used in the current study. There are currently only two sequences of Phormidium papyraceum Gomont ex Gomont, 1892 in the GenBank. The BLAST search showed that one of them (OQK586776 Phormidium papyraceum ULC441) has high similarity to strains of Wi/mottia murrayi (West & G.S.West) Strunecky, Elster & Komarek 2011 and the other (KF770970 Phormidium papyraceum PACC 8693) is similar to sequences of Microcoleus vaginatus strains. In the reconstructed phylogenetic trees, they are also arranged in such a way. It was interesting that the other Microcoleus species (M. steenstrupii J.B. Petersen 1928 and M. paludosus Gomont ex Gomont 1892) were clustered together with Wé/mottia strains, but distinct from the Microcoleus vaginatus clade (Fig. 7B). This confirms the note of Komarek & Anagnostidis (Komarek and Anagnostidis 2005) that Microcoleus vaginatus should be separated from the genus Microcoleus as a special genus, which belongs to the family Oscillatoriaceae. 18 Teneva | et al Distance and similarity between 16S rDNA sequences of the strains used in the phylogenetic analyses are given in Supplementary Materials (Suppl. material 1). Metabolomic analysis To check whether strains belonging to the two genera (Phormidium and Microcoleus) cultivated under the same conditions differ in their metabolic profile, we performed a metabolomic analysis of non-polar compounds in three Phormidium strains (PACC 5522, PACC 5527, PACC 5529) and three strains of Microcoleus (CCALA 145, CCALA 152, CCALA 757) by reversed phase chromatography in positive ion mode. The positive ion mode was used as more informative to cover more compounds and provide more comprehensive compound characterisation. We chose to investigate non-polar compounds in order to identify more specific metabolites that could serve as chemo-taxonomic markers for discrimination of strains belonging to these two genera. Initial analyses showed the presence of 12,000 potential compounds. After analysis of these compounds with several software packages (including Perseus), 900 compounds were identified that differed significantly between strains of the two genera (Fig. 8). 12 "466.34552_20.59 "523,26076_19.433 - Logp , a il nin, Un Difference Figure 8. EES] Perseus volcano plot showing compounds with significantly different abundances between the Phormidium and Microcoleus strains. In the left area, red data are for compounds whose abundances were decreased in Phormidium strains and increased in Microcoleus strains. In the right side, red data are for compounds with increased abundance levels in Phormidium strains and decreased abundance levels in Microcoleus strains. Polyphasic characterisation of Microcoleus autumnalis (Gomont, 1892) Strunecky, ... 19 From them, the compounds with the greatest statistical significance were selected for further analysis and identification — a total of 39 in number, 20 with increased concentration and 19 with decreased concentration for the representatives of both genera (Fig. 9). The proposed putative identification is based on three different approaches: Compound Discoverer with FISh coverages, Sirius and MS Finder. Even if not properly annotated, these differences are statistically significant and apparent and the proposed ion features can be used to distinguish between the two cyanobacterial genera (Phormidium and Microcoleus). Therefore, these 39 compounds presented in Table 6 can be used as potential biochemical markers to distinguish between Phormidium autumnale and Microcoleus vaginatus. vu u vu 3 8 8 5 5 s fa fa fad 3 3 3 > > > on on on a a z 8 8 8 8 g & & N S 710.19707_5.798 1095.6895_29.437 258.10079_5.651 729.474_15.021 337.25919_8.397 885.83372_33.987 1052.73146_30.05 1048.69903_28.942 672.43679_13.982 786.46692_19.251 729.47352_14.11 891.52758_12.843 891.5275_12.532 745.46969_12.364 729.47355_14.595 930.49039_12.526 594.46681_16.45 568.45151_ 19.298 308.23463_19.262 790.4987_25.08 523.36489_12.368 352.29771_28.428 330.27834_11.784 324.267_27.216 442.37697_10.022 666.42967_12.978 684.43972_12.952 466.34552_20.59 592.50647_19.186 456.39218_10.789 566.41316_14.911 548.40301_ 14.987 616.50723 24.723 551.53019_17.224 747.58656_27.376 730.56071_27.369 568.50733_27.367 448.33501_20.667 580.39204_13.331 Figure 9. EES] A heatmap of the 39 significant compounds found in the investigated strains designed with the molecular weight and retention time (Y-axis). The six cyanobacterial strains (X-axis) are separated into two groups - Microcoleus strains (CCALA 145, CCALA 152, CCALA 757) and Phormidium strains (PACC 5522, PACC 5527, PACC 5529). The first 20 compounds (upper part of the Y-axis) are with increased (red) abundance levels in Phormidium strains and decreased (green) abundance levels in Microcoleus strains. The next 19 compounds (lower part of the Y-axis) are with increased abundance in Microcoleus strains and decreased abundance in Phormidium strains. The putative identities of these compounds are given in Table 6. 20 Table 6. Biochemical markers for distinguishing Phormidium autumnale and Microcoleus vaginatus. RT, retention time; (+) increased abundance; (—) decreased abundance. No RT 19 20 (min) 5.65 5.80 8.40 10.02 10.79 11.78 12.36 12.37 12.53 12.53 12.84 12.95 12.98 13.33 13.98 14.11 14.60 14.91 14.99 15.02 Teneva | et al Compound 6-Ethyl-2-methyl-4,6-dihydro-2H- [1,4]oxazino[3,2-c]quinoline-3,5-dione Unknown Unknown Ethyl N-{2-[(tert- butoxycarbonyl)amino]hexadecyl}glycinate 1,16-Hexadecanediyl bis(butylcarbamate) 2-Palmitoylglycerol 6-Hydroxy-9-[(6Z,9Z,12Z,15Z)-6,9,12,15- octadecatetraenoyloxy]-6-oxido-5, 7-dioxa-2- aza-6lambda~5~-phosphadecan-10-yl (6Z, 9Z,12Z,15Z)-6,9,12,15-octadecatetraenoate (3R)-3-{[(3alpha,5beta)-3-Hydroxy-24- oxocholan-24-yljamino}-3-phenylpropanoic acid 5-Oxo-L-prolyl-L-threonyl-L-seryl-L- phenylalanyl-L-threonyl-L-prolyl-N~5~- (diaminomethylene)-L-ornithyl-L-leucinamide Unknown Unknown (3beta,22beta)-22-[(3-Methyl-2- butenoyl)oxy]-3-{[(2E)-3-phenyl-2- propenoyljoxy}olean-12-en-28-oic acid 4-Methyl-6-oxostigmast-7-ene-3 ,22-diyl dibenzoate Phoenicoxanthin 1-Ethyl-4-(4-oxido-2,6-diphenyl-4H-1 ,4- oxaphosphinin-4-yl)piperazine Methyl N-[(3beta)-3,23-dihydroxy-28- oxolup-20(29)-en-28-yl]glycyl-L-tryptophanate Methyl! N-[(3beta)-3,23-dihydroxy-28- oxolup-20(29)-en-28-yl]glycyl-L-tryptophanate 3-Hydroxyechinenone (3'Z)-3',4'-Didehydro-beta,psi-caroten-4-one Methyl N-[(3beta)-3,23-dihydroxy-28- oxolup-20(29)-en-28-yl]glycyl-L-tryptophanate Formula C14H14N203 C32H49NgOP 283 C17H3706 CosHsoN204 CagHs2N204 C19H3gO4 C42HggNOgP C33H4gNO4 C42HegN12012 C4gH74Ns0gP CagH74N50gP C44Hgo0e6 C44Hsg05 C49H5203 C36H72N304PS C4gHe3N306 C44He3N30e6 C4gpH5402 C49Hs520 Ca4gHe3N306 Molecular P. weight 258.101 710.197 337.259 442.377 456.392 330.278 745.470 523.365 930.490 891.528 891.528 684.440 666.430 580.392 672.437 729.474 729.474 566.413 548.403 729.474 autumnale + M. vaginatus No 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 Polyphasic characterisation of Microcoleus autumnalis (Gomont, 1892) Strunecky, ... RT (min) 16.45 17.22 19.19 19.25 19.26 19.30 20.59 20.67 24.72 25.08 27.22 27.37 27.37 27.38 28.43 28.94 29.44 30.05 33.99 Compound Unknown N-heptadecanoylsphingosine 1-Palmitoyl-2-linoleoyl-sn-glycerol L-Phenylalanyl-L-leucyl-L-arginyl-L-isoleucyl- L-arginyl-L-prolyl-L-lysine Eicosapentaenoic acid methyl 9- oxooctadeca-10,12-dienoate [5-(5a,5b,8,8,11a, 13b-Hexamethyl-1 ,2,3,3a, 4,5,7a,9,10,11,11b,12,13,13a- tetradecahydrocyclopenta[a]chrysen-3-yl)-2- acetyloxyhexyl] acetate Phylloquinone oxide 2-Methyl-2-[(3E,7E,11E)-4,8,12,16- tetramethyl-3,7,11,15-heptadecatetraen-1- yl]-2H-chromen-6-ol 1-Palmitoyl-2-arachidonoyl-sn-glycerol (1R,2R,3S,4R,6S)-4,6-diamino-2-{[(2R, 15R)-16-({(1R,2R,3S,5R,6S)-3 ,5-diamino-2- [(2,6-diamino-2,6-dideoxy-alpha-D- glucopyranosyl)oxy]-6-hydroxycyclohexyl} oxy)-2,15-dihydroxy-4,13-dimethyl-7,10- dioxa-4, 13-diazahexadec-1-yl]oxy}-3- hydroxycyclohexyl 2,6-diamino-2,6-dideoxy- alpha-D-glucopyranoside Ethyl (9E)-8-oxo-9-octadecenoate [(2S)-2-hexadecanoyloxy-3-hydroxypropyl] hexadecanoate 1,2-Dipalmitoyl-3-beta-D-galactosyl-sn- glycerol (2R)-N-[(2S,3S,4R)-1-(beta-L- Allopyranosyloxy)-3,4-dihydroxy-2- undecanyl]-2-hydroxytetracosanamide 3-Octadecyloxolane-2,5-dione Unknown (2R)-N-[(2S,3R,5E)-1 ,3-Dihydroxy-5- heptadecen-2-yl]-2-hydroxyicosanamide Unknown O-[{(2R)-3-[(13Z,16Z)-13,16- Docosadienoyloxy]-2-[(4Z,7Z,10Z,13Z,16Z, 19Z)-4,7,10,13,16,19- docosahexaenoyloxy]propoxy} (hydroxy)phosphoryl]-L-serine Formula C33He3N403P C35HegNO3 C37H¢gO05 C34H73NgOgP3 C19H3203 C3gHs6Nq C31H4g03 C314H4402 C3gHegO5 C43Hg7N20gP C29H3603 C35Heg05 C41H7g019 C44HgiNO19 C22H4903 C53H19gN9O3P3S2 CggH101N202PS C47Hg9N2013 Molecular P. weight 594.467 551.530 592.506 786.467 308.235 568.452 466.346 448.335 616.507 790.499 324.267 568.507 730.561 747.587 352.298 1048.699 1095.690 1052.731 885.834 autumnale + + + + + 21 M. vaginatus 22 Teneva | et al Discussion A number of morphological and molecular genetic studies have demonstrated the polyphyleticity of the genera Phormidium and Microcoleus (Komarek et al. 2014, Stoyanov et al. 2014). The genus Phormidium presents a significant taxonomic challenge because data obtained with molecular approaches often are inconsistent with the morphological studies. For example, species morphologically assigned to Phormidium autumnale were found to be genetically distinct, grouped into different groups (Marquardt and Palinska 2007). In addition to the high biodiversity and wide distribution, like most cyanobacteria, the representatives of genus Phormidium are also characterised by a high degree of environmentally induced morphological variability (Marquardt and Palinska 2007, Heath et al. 2010). This makes them difficult for identification. Scientific reports clearly show that there is some uncertainty regarding the classification of some members of the genus, such as P autumnale and P uncinatum (McAllister et al. 2016). Based on molecular genetic analyses, as well as observations on the morphology and ultrastructure of representatives of Microcoleus vaginatus and Phormidium autumnale, Strunecky et al. (2013) transfered P autumnale to the genus Microcoleus as Microcoleus autumnalis. The authors analysed 91 Microcoleus strains and only one Phormidium autumnale strain (M. autumnalis Luznice). Although this change (based on one strain and literature data) has been accepted, the taxonomic position of P autumnale is still controversial. Proof of this is the data presented in our study, as well as the opinion of other authors who conducted research with P autumnale strains. The polyphasic approach showed that the cyanobacterial blooms observed in New Zealand were due to P autumnale and P. uncinatum strains, but molecular genetic analyses, based on 16S rDNA, identified P autumnale as the dominant species (Wood et al. 2012, Harland et al. 2014, McAllister et al. 2016). These findings strongly indicate the need for additional tools to correctly identify the cyanobacterial strains. However, it is obvious that this cannot be accomplished solely through morphology and molecular genetic studies. The inclusion of ultrastructural analysis (e.g. thylakoid arrangement), as well as metabolomic analyzes as additional tools, would, in our opinion, contribute to clarifying this issue. The polyphasic approach applied in the present study includes a detailed analysis of the morphological features of the two species Phormidium autumnale and Microcoleus vaginatus. According to Komarek and Anagnostidis (2005), the genus Phormidium contains species with trichomes 3-11 ym width, mucilaginous sheaths with only one trichome, isodiametric cells, shorter than wide, pointed or rounded apical cells with or without calyptra. Phormidium autumnale (Agardh) Trevisan ex Gomont belongs to a group that is characterised by + isodiametric cells and trichomes that are slightly narrowed at the ends forming calyptra. The morphological characteristics defining genus Microcoleus according to Strunecky et al. (2013) are narrowed ends of the trichomes, calyptra, cells shorter than wide, to more or less isodiametric and facultative presence of sheaths. Most species are Polyphasic characterisation of Microcoleus autumnalis (Gomont, 1892) Strunecky, ... 23 4—10 um in diameter. The presence of multiple trichomes in a common sheath is facultative in many, but not all species. The main cytomorphological diacritic characters for distinguishing the strains defined in the present study as P autumnale and M. vaginatus are: (1) the ends of the trichomes, (2) the shape of the apical cells in the trichome and (3) the presence/absence of a calyptra and its shape (Table 2 and Table 3, Fig. 1 and Fig. 2). The morphological difference between Microcoleus vaginatus and Phormidium autumnale according to Strunecky et al. (2013) is only in the shape of the colonies and the organisation of the filaments. According to data from the same research group, MV. vaginatus belongs to an easily recognisable clade with a specific ecology (soil biotope) and bundle-like filaments in a common sheath. Our morphological analysis did not confirm these claims. Data showed differences in the morphology of the two strains. In order to avoid the potential influence of other factors on the variability in morphology, the studied cyanobacterial strains were cultured under the same conditions. The ends of the trichomes in the studied P autumnale strains are gradually and slightly narrowed, encompassing the last few cells. In strains of MV. vaginatus, they are sharply narrowed, S-shaped, with the curve affecting the last few cells and not just the apical one. There is also a difference in the apical cells. In the P autumnale strains, they are elongated, with a rounded conical shape, slightly curved, while in the strains of /. vaginatus, the apical cells are capitate. In P. autumnale, the calyptra is absent or weakly expressed and, if it is present, it is rounded. In M. vaginatus strains, the calyptra is well developed, with a conical, obtuse to hemispherical shape. The filaments are single, which is typical for the genus Phormidium. A few trichomes in a common sheath are not observed. There is also a difference in the habitats. P autumnale is a freshwater species distributed mainly in streams, rivers and waterfalls, but also amongst the growths (periphyton) on underwater substrates. The soils are the main habitat for /. vaginatus. Regarding the ultrastructure and thylakoid arrangement, the conclusion of Strunecky et al. (2013) is that the ultrastructure of P autumnale is very similar to that of genus Microcoleus. Thylakoids usually form bundle-like aggregations arranged irregularly within the cells. In some strains, there is a radial arrangement of the thylakoids, but in the same strain, the thylakoids may form bundles of thylakoids. We observed that the arrangement of thylakoids in the two species (P autumnale and WM. vaginatus) shows significant differences. In P autumnale, the thylakoids are with parietal arrangement, sometimes with a central fascicle (Fig. 4) and in M. vaginatus strains, the thylakoids are with fascicular arrangement (Fig. 5). We agree that, due to the high degree of environmentally-induced morphological variability of cyanobacteria, the sequencing is essential for the correct taxonomic assessment of these species. Phylogenetic analyses performed by some research groups suggest that P autumnale is very close to M. vaginatus (Boyer et al. 2002, Siegesmund et al. 2008, HaSler et al. 2012, Strunecky et al. 2013). Phylogenetic analyses in the present study confirm the polyphyleticity of genus Phormidium, but clearly demonstrate the distance of the clade formed by strains of Microcoleus vaginatus from that formed by strains of Phormidium 24 Teneva | et al autumnale (Fig. 7A). This is a clear sign of the distance between the two species on a genetic basis, which excludes the identity of Phormidium autumnale with Microcoleus strains and its appurtenance to genus Microcoleus. Interestingly, the strains we studied showed genetic similarity to representatives of the genus Kampthonema separated in 2014 from the genus Phormidium (Strunecky et al. 2014). It is clear that, in certain cases, the well-developed and known morphological, ultrastructural and molecular genetic criteria are not sufficiently descriptive and do not provide a definitive answer to the question regarding the taxonomic affiliation and position of a given species. Then it is necessary to look for new characteristics to resolve such an issue. According to Komarek (2016), differences in biochemistry, for example, in the pigment content and the presence of various compounds (metabolites from the life activity of cyanobacteria), can be specific for different cyanobacterial lineages and could be considered as an additional taxonomic criterion. Some studies have reported the use of fatty acids and lipid profiles of microalgae and cyanobacteria as biomarkers to distinguish closely-related organisms at the species and generic level (Berge and Barnathan 2005, Schweder et al. 2005, Rossi et al. 2006, Lang et al. 2011). A systematic large-scale analysis of lipid profiles in microalgae was done by Lang et al. (2011), examining all available 2291 microalgal strains of the SAG culture collection. Their conclusion was that, despite the general trends in fatty acid distribution observed throughout the study reflecting the phylogenetic relationships between microalgae species and classes, the fatty acid profile alone cannot be considered as a useful marker for distinguishing between different genera and species. For this purpose, it is necessary to study and compare other metabolites, such as sterols, lipids and hydrocarbons. The conclusion is that the taxonomic value of various cell inclusions and/or the presence of biochemical compounds is not entirely clear and its evaluation and comparison with other diacritical features in the cyanobacterial taxonomy is needed. To clearly define the taxonomic position of Phormidium autumnale, we performed an additional metabolomic analysis involving three strains of Phormidium autumnale and three strains of Microcoleus vaginatus. Based on the analysis, we were able to select 39 compounds that can be used as biochemical markers to distinguish the two species. Our metabolomic analysis clearly showed a different taxonomic affiliation of Phormidium autumnale than that proposed by Strunecky et al. (2013). The limitations of applying metabolomic analysis within the polyphasic approach as a complementary tool for taxonomic identification are related to the fact that the species being compared must be cultured under the same conditions and cannot be directly applied to natural samples. Conclusions Our results conclusively demonstrate the belonging of the cyanobacterial species Phormidium autumnale to genus Phormidium and define its transfer to genus Microcoleus as incorrect. Morphological differences were found in the examined P autumnale and M. Polyphasic characterisation of Microcoleus autumnalis (Gomont, 1892) Strunecky, ... 25 vaginatus strains regarding the ends of the trichome, the shape of the apical cell and the shape of the calyptra, which are sufficiently descriptive. The ultrastructural studies also confirm the differences in the arrangement of thylakoids — parietal in P autumnale and fascicular in M. vaginatus. Molecular genetic analyses and phylogenetic reconstructions, based on 16S rDNA, strongly support our opinion that Phormidium autumnale should remain within the genus Phormidium and its transfer to the genus Microcoleus was incorrect. For the first time, based on a metabolomic analysis, 39 compounds have been selected and proposed as biochemical markers that could serve to distinguish Phormidium autumnale and Microcoleus vaginatus. Acknowledgements This research was funded by the Bulgarian National Science Fund, KP-06-N51/5. Conflicts of interest The authors have declared that no competing interests exist. References ° Anagnostidis K, Komarek J (1985) Modern approach to the classification system of the cyanophytes 1: Introduction. Algological Studies 38/39: 291-302. ° Berge J, Barnathan G (2005) Fatty acids from lipids of marine organisms: molecular biodiversity, roles as biomarkers, biologically active compounds, and economical aspects. Advances in biochemical engineering/biotechnology 96: 49-125. https://doi.org/ 10.1007/b135782 ° Boyer S, Johansen J, Flechtner V, Howard G (2002) Phylogeny and genetic variance in terrestrial Microcoleus (Cyanophyceae) species based on sequence analysis of the 16S rRNA gene and associated 16S-23S ITS region. Journal of Phycology 38 (6): 1222-1235. https://doi.org/10.1046/j.1529-8817.2002.01168.x ° Guijas C, Montenegro-Burke JR, Domingo-Almenara X, Palermo A, Warth B, Hermann G, Koellensperger G, Huan T, Uritboonthai W, Aisporna A, Wolan D, Spilker M, Benton HP, Siuzdak G (2018) METLIN: A technology platform for identifying knowns and unknowns. Analytical Chemistry 90 (5): 3156-3164. https://doi.org/10.1021/ acs.analchem.7b04424 ° Harland F, Wood S, Broady P, Gaw S, Williamson W (2014) Polyphasic studies of cyanobacterial strains isolated from benthic freshwater mats in Canterbury, New Zealand. New Zealand Journal of Botany 52 (1): 116-135. https://doi.org/ 10.1080/0028825x.2013.846266 ° Hasler P, Dvorak P, Johansen J, Kitner M, Ondrej V, Poulickova A (2012) Morphological and molecular study of epipelic filamentous genera Phormidium, Microcoleus and Geitlerinema (Oscillatoriales, Cyanophyta/Cyanobacteria). Fottea 12 (2): 341-356. https://doi.org/10.5507/fot.2012.024 26 Teneva | et al Heath M, Wood S, Ryan K (2010) Polyphasic assessment of fresh-water benthic mat- forming cyanobacteria isolated from New Zealand. FEMS Microbiology Ecology https:// doi.org/10.1111/).1574-6941.2010.00867.x Hoffmann L, KaStovsky J, Komarek J (2005) Proposal of cyanobacterial system, 2004. In Cyanoprokaryota, 2. Teil/2nd Part: Oscillatoriales, Komarek J, Anagnostidis K, Eds. Budel B, Gartner G, Krienitz L, Schagerl M, (Hrsg.). Susswasserflora von Mitteleuropa; Elsevier/Spektrum: Heidelberg, Germany, pp: 657-660. Huelsenbeck JP, Ronquist F (2001) MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics 17 (8): 754-755. https://doi.org/10.1093/bioinformatics/17.8.754 Katoh K, Rozewicki J, Yamada KD (2017) MAFFT online service: multiple sequence alignment, interactive sequence choice and visualization. Briefings in Bioinformatics 20 (4): 1160-1166. https://doi.org/10.1093/bib/bbx108 Kimura M (1980) A simple method for estimating evolutionary rates of base substitutions through comparative studies of nucleotide sequences. Journal of Molecular Evolution 16 (2): 111-120. https://doi.org/10.1007/bf01731581 Komarek J, Anagnostidis K (2005) Cyanoprokaryota. 2. Teil: Oscillatoriales. Bidel B, Gartner G, Krienitz L, Schagerl M, Eds., 19/2. Elsevier GmbH, Munchen, 759 pp. Komarek J (2006) Cyanobacterial taxonomy: Current problems and prospects for the integration of traditional and molecular approaches. Algae 21 (4): 349-375. hitps:// doi.org/10.4490/algae.2006.21.4.349 Komarek J (2009) Recent changes (2008) in cyanobacteria taxonomy based on a combination of molecular background with phenotype and ecological consequences (genus and species concept). Hydrobiologia 639 (1): 245-259. https://doi.org/10.1007/ $10750-009-0031-3 Komarek J, KaStovsky J, Mares J, Johansen JR (2014) Taxonomic classification of cyanoprokaryotes (cyanobacterial genera) 2014, using a polyphasic approach. Preslia 86: 295-335. Komarek J (2016) A polyphasic approach for the taxonomy of cyanobacteria: principles and applications. European Journal of Phycology 51 (3): 346-353. https://doi.org/ 10.1080/09670262.2016.1163738 Komarek J (2020) Quo vadis, taxonomy of cyanobacteria (2019). Fottea 20 (1): 104-110. https://doi.org/10.5507/fot.2019.020 Kumar S, Stecher G, Tamura K (2016) MEGA7: Molecular Evolutionary Genetics Analysis Version 7.0 for bigger datasets. Molecular Biology and Evolution 33 (7): 1870-1874. https://doi.org/10.1093/molbev/msw054 Lang |, Hodac L, Friedl T, Feussner | (2011) Fatty acid profiles and their distribution patterns in microalgae: a comprehensive analysis of more than 2000 strains from the SAG culture collection. BMC Plant Biology 11 (1). httos://doi.org/ 10.1186/1471-2229-11-124 Mares J, Strunecky O, Bucinska L, Wiedermannova J (2019) Evolutionary patterns of thylakoid architecture in Cyanobacteria. Frontiers in Microbiology 10 hittps://doi.org/ 10.3389/fmicb.2019.00277 Marquardt J, Palinska K (2007) Genotypic and phenotypic diversity of cyanobacteria assigned to the genus Phormidium (Oscillatoriales) from different habitats and geographical sites. Archives of Microbiology 187 (5): 397-413. https://doi.org/10.1007/ $00203-006-0204-7 Polyphasic characterisation of Microcoleus autumnalis (Gomont, 1892) Strunecky, ... 27 McAllister T, Wood S, Hawes | (2016) The rise of toxic benthic Phormidium proliferations: A review of their taxonomy, distribution, toxin content and factors regulating prevalence and increased severity. Harmful Algae 55: 282-294. https:// doi.org/10.1016/j.hal.2016.04.002 Moten D, Batsalova T, Basheva D, Mladenov R, Dzhambazov B, Teneva | (2017) Outer membrane efflux protein (OMEP) is a suitable molecular marker for resolving the phylogeny and taxonomic status of closely related cyanobacteria. Phycological Research 66 (1): 31-36. https://doi.org/10.1111/pre.12203 Palinska K, Deventer B, Hariri K, Lotocka M (2011) A taxonomic study on Phormidium- group (cyanobacteria) based on morphology, pigments, RAPD molecular markers and RFLP analysis of the 16S rRNA gene fragment. Fottea 11 (1): 41-55. httos://doi.org/ 10.5507/fot.2011.006 Reynolds E (1963) The use of lead citrate at high pH as an electron-opaque stain in electron microscopy. Journal of Cell Biology 17 (1): 208-212. https://doi.org/10.1083/jcb. 17.1.208 Ronquist F, Teslenko M, van der Mark P, Ayres D, Darling A, Hohna S, Larget B, Liu L, Suchard M, Huelsenbeck J (2012) MrBayes 3.2: Efficient Bayesian phylogenetic inference and model choice across a large model space. Systematic Biology 61 (3): 539-542. https://doi.org/10.1093/sysbio/sys029 Rossi S, Sabatés A, Latasa M, Reyes E (2006) Lipid biomarkers and trophic linkages between phytoplankton, zooplankton and anchovy (Engraulis encrasicolus) larvae in the NW Mediterranean. Journal of Plankton Research 28 (6): 551-562. https://doi.org/ 10.1093/plankt/fbi140 Schweder T, Lindequist U, Lalk M (2005) Screening for new metabolites from marine microorganisms. Advances in Biochemical Engineering/Biotechnology1-48. https:// doi.org/10.1007/b135781 Seo PS, Yokota A (2003) The phylogenetic relationships of cyanobacteria inferred from 16S rRNA, gyrB, rpoC1 and rpoD1 gene sequences. J Gen Appl Microbiol. 49 (3): 191-203. https://doi.org/10.2323/jgam.49.191 Siegesmund M, Johansen J, Karsten U, Friedl T (2008) Coleofasciculus gen. nov. (Cyanobacteria): morphological and molecular criteria for revision of the genus Microcoleus Gomont. Journal of Phycology 44 (6): 1572-1585. https://doi.org/10.1111/j. 1529-8817.2008.00604.x Staub R (1961) Ernahrungsphysiologisch-aut6kologische Untersuchungen an der planktischen Blaualge Oscillatoria rubescens DC. Schweizerische Zeitschrift fur Hydrologie 23 (1): 82-198. https://doi.org/10.1007/bf02505618 Stoyanov P, Moten D, Mladenov R, Dzhambazov B, Teneva | (2014) Phylogenetic relationships of some filamentous cyanoprokaryotic species. Evolutionary Bioinformatics 10 https://doi.org/10.4137/ebo0.s13748 Strunecky O, Komarek J, Johansen J, LukeSova A, Elster J (2013) Molecular and morphological criteria for revision of the genus Microcoleus (Oscillatoriales, Cyanobacteria). Journal of Phycology 49 (6): 1167-1180. https://doi.org/10.1111/jpy. 12128 Strunecky O, Komarek J, Smarda J (2014) Cyanobacteria, a new genus derived from the polyphyletic Phormidium on the basis of combined molecular and cytomorphological markers. Preslia 86: 193-207. 28 Teneva | et al Strunecky O, Ivanova AP, Mares J (2022) An updated classification of cyanobacterial orders and families based on phylogenomic and polyphasic analysis. Journal of Phycology https://doi.org/10.1111/jpy.13304 Teneva |, Batsalova T, Bardarov K, Moten D, Dzhambazov B (2022) A novel approach for fast screening of a complex cyanobacterial extract for immunomodulatory properties and antibacterial activity. Applied Sciences 12 (6). https://doi.org/10.3390/app12062847 Tillett D, Neilan B (2001) Xanthogenate nucleic acid isolation from cultured and environmental cyanobacteria. Journal of Phycology 36 (1): 251-258. https://doi.org/ 10.1046/j.1529-8817.2000.99079.x Wilmotte A, Turner S, Van de Peer Y, Pace NR (1992) Taxonomic study of marine Oscillatoriacean strains (Cyanobacteria) with narrow trichomes. II. Nucleotide sequence analysis of the 16S ribosomal RNA. Journal of Phycology 28 (6): 828-838. https:// doi.org/10.1111/j.0022-3646.1992.00828.x Wood S, Smith FJ, Heath M, Palfroy T, Gaw S, Young R, Ryan K (2012) Within-Mat variability in Anatoxin-a and Homoanatoxin-a production among benthic Phormidium (Cyanobacteria) strains. Toxins 4 (10): 900-912. https://doi.org/10.3390/toxins4100900 Supplementary material Suppl. material 1: Table $1 EE Authors: Ivanka Teneva, Detelina Belkinova, Tsvetelina Paunova-Krasteva, Krum Bardarov, Dzhemal Moten, Rumen Mladenov and Balik Dzhambazov Data type: Distance/Similarity Brief description: Polyphasic characterisation of Microcoleus autumnalis (Gomont, 1892) Strunecky, Komarek & J.R.Johansen, 2013 (Oscillatoriales, Cyanobacteria) using a metabolomic approach as a complementary tool. Download file (152.09 kb)