Oceanological and Hydrobiological Studies Identification of Diatom

Oceanological and Hydrobiological Studies International Journal of Oceanography and Hydrobiology Vol. XXXIX, No.3 Institute of Oceanography (3-20) University of Gdańsk ISSN 1730-413X 2010 eISSN 1897-3191

DOI 10.2478/v10009-010-0031-7 Received: September 22, 2009 Original research paper Accepted: August 10, 2010

Identification of diatom isolates from the Gulf of Gdańsk: testing of species identifications using morphology, 18S rDNA sequencing and DNA barcodes of strains from the Culture Collection of Baltic Algae (CCBA)

Filip F. Pniewski1, Thomas Friedl2, Adam Latała1

1Department of Marine Ecosystem Functioning Institute of Oceanography, University of Gdańsk Al. Marszałka Piłsudskiego 46, 81-378 Gdynia, Poland 2Albrecht-von-Haller-Institut für Pflanzenwissenschaften Abteilung Experimentelle Phykologie und Sammlung von Algenkulturen Universität Göttingen, Untere Karspüle 2, 37073 Göttingen, Germany

Key words: 18S rDNA, Baltic Sea, centrics, diatoms, molecular phylogeny, pennates

Abstract

Eighteen strains of diatoms isolated from different phytoplankton and microphytobenthic assemblages in the Gulf of Gdańsk (Baltic Sea, Poland) and maintained at the Culture Collection of Baltic Algae (CCBA) were investigated for their species identifications. The latter were tested by phylogenetic analyses of nearly full 18S rDNA sequences as well as through sequence comparisons of 5.8S+ITS2 rDNA fragments for use as DNA barcode. The planktonic species were readily identified as a member of the Cyclotella meneghiniana species complex and Skeletonema marinoi, respectively, by both 18S phylogenies and DNA barcoding. In contrast, for the pennate diatom strains the suitability of DNA sequence comparisons for species identification

4 F.F. Pniewski, T. Friedl, A. Latała appeared to be still rather limited. Only one strain could be identified to the species level (Navicula gregaria) using DNA barcodes, and no closest relative barcode sequences were available for the other strains. The 18S rDNA phylogenies supported species identification for only one strain (Bacillaria cf. paxillifera). In all other pennate strains identification was only possible to the genus level. Because for most species identified by morphology no closest neighbouring 18S rDNA sequences were available, the CCBA strains may serve as references to represent certain species which have been well characterized by morphology, 18S rDNA sequence analyses and DNA barcodes. Relatively high 18S rDNA differences between pairs of strains of Fistulifera saprophila and Navicula perminuta indicated that each species may represent a complex of several cryptic species. In contrast, the five isolates of Nitzschia microcephala had identical sequences as well as the two isolates of Nitzschia cf. fonticola. With the addition of new species to the 18S rDNA phylogeny of pennate raphid diatoms, the monophyly of the genera Fistulifera, Haslea and Navicula s.str. were confirmed, while Nitzschia appeared paraphyletic. The traditional family Naviculaceae was paraphyletic, in agreement with previous phylogenetic studies.

INTRODUCTION

Algal culture collections are specific repositories for living organisms and play a crucial role in phycology and microbiology as they are a source of high quality research material. Moreover, culture collections also serve purposes in education, ex situ conservation and industry, e.g. production of biofuels or dietary supplements (Latała et al. 2006). Growing interest in the practical usage of algal strains calls for their full description, including: morphology, autecology, toxicity, production of different biomolecules etc., as their specific and unique features are essential for their possible applications. Moreover, it is important to mention that the effective use of an algal strain relies on its correct identification. Recently, a few papers strongly emphasized the need for accurate taxonomic strain identification, showing the consequences of previous mistakes (e.g. Bortolus 2008). Diatoms are ecologically and economically one of the most important protists (Moniz&Kaczmarska 2009). However, despite well described taxonomically significant microstructures of their silica frustules, diatoms are often difficult to identify, especially for the untrained observer. Nowadays, DNA sequence analyses have become a standard approach in diatom research, opening a new window into their systematics and evolution. The proven discriminatory power of the molecular approach ensures its increasing, future role in the delimitation of diatom species (Alverson 2008). Up to now mainly the nuclear-encoded small ribosomal subunit (18S) and large ribosomal subunit (28S), rRNA genes, the plastid-encoded large subunit of RUBISCO (rbcL) and the mitochondrial cytochrome c oxidase 1 (coxI) have been used as genetic markers for investigating the diversity, systematics and evolution of diatoms. Among them 18S rRNA genes have been most frequently sequenced resulting in the largest reference sequence dataset among the

Molecular phylogeny of same diatoms... 5 different markers used for diatoms (Moniz & Kaczmarska 2009). 18S rDNA, as a functionally stable evolutionary marker which evolved independently of morphology (Medlin et al. 2000 and references therein), was discovered to be especially useful for inferring phylogenetic relationships among diatoms at all taxonomic levels (e.g Beszteri et al. 2001, Medlin & Kaczmarska 2004, Sorhannus 2004, Sorhannus 2007, Theriot et al. 2009). Most importantly, similar tree topologies have also been recovered from analyses of LSU rDNA and rbcL (Sorhannus et al. 1995, Lundholm et al. 2002, Kaczmarska et al. 2005, Choi et al. 2008). Taking into consideration the importance of a correct, fast and inexpensive identification of organisms the concept of “barcoding” has been developed. Recently it has also been introduced into diatom taxonomy. The idea of barcoding is based on the assumption that differences within a short DNA fragment reflect biological separation of species and therefore can be used as a genetic tag allowing species identification (Moniz&Kaczmarska 2009 and references therein). Recently, Moniz & Kaczmarska (2010) have identified species-specific levels of divergence within the DNA region which starts at the beginning of the 5.8S and ends in the highly conserved region of helix III of the ITS2 transcript. This fragment was proposed as a DNA-barcode for diatoms. The Culture Collection of Baltic Algae (CCBA) was established in the 1980s in order to isolate, maintain and supply microalgal strains for research, education and commercial use (Latała et al. 2006). The main focus of the collection is on the Baltic Sea which is the largest brackish sea in the world. Its modern history goes back merely 13,000 years and the biological diversity of the sea is relatively poor since its short history did not allow for the creation of a large number of species. In spite of this, flora and fauna of the Baltic Sea are unique with many species of microalgae being well adapted to the low salinity. At present the collection houses more than 100 strains, all from Baltic habitats, and covers diatoms, green algae, diatoms, several microflagellate groups and cyanobacteria (http://www.ocean.univ.gda.pl/~ccba/). Diatoms constitute a large number of the CCBA strains. Most of the diatom strains of the CCBA collection have already been used in various ecological, ecophysiological and ecotoxicological analyses (e.g. Latała 2003; Latała et al. 2006, 2009a,b,c, 2010; Jodłowska & Latała 2010). Therefore, the correct identification of the strains is crucial with respect to their possible applications. The main objective of this study was to test the morphology-based species identification of a variety of common diatoms isolated from the phytoplankton and microphytobenthos of the Gulf of Gdańsk and maintained in the CCBA collection. 18S rRNA gene sequences and the recently proposed “diatom barcode” (5.8S+ITS2 rDNA fragment) were sequenced, analysed and compared with sequences already available from databases.

www.oandhs.org

6 F.F. Pniewski, T. Friedl, A. Latała

MATERIALS AND METHODS

Algal cultures Eighteen strains of diatoms were obtained from natural samples of Baltic water collected at six coastal stations of the Gulf of Gdańsk (Table 1). Unialgal cultures were established using the dilution method, washing with microcapillary pipettes or streaking on agar (Hoshaw&Rosowski 1979). The uni-algal nature of cultures was checked by microscopic observation. Non- axenic diatom strains were grown in liquid f/2 medium (Guillard 1975) prepared from Baltic water. The following culture conditions were applied: salinity 6.9 PSU, temperature 18ºC, PAR intensity 20 µmol photons m-2 s-1 and 16:8 L:D photoperiod.

Morphology and Identification of Baltic diatom strains The culture strains were studied using a microscope with 40× objective lens. Diatom cells were cleaned with hydrogen peroxide (33%) at 30ºC, 60ºC and 90ºC for 3h at each step and mounted in Naphrax (Batterbee 1986). Permanent slides were analysed with the same microscope under ×100 oil immersion objective lens. For scanning electron microscopy (SEM) cleaned frustules were coated with gold and/or carbon and studied using a Hitachi S- 4700 scanning electron microscope. For diatom species identification the following literature was used: Krammer&Lange-Bertalot 1988, Krammer&Lange-Bertalot 2008, Lange-Bertalot 2001, Sarno et al. 2005, Snoeijs&Vilbast 1994, and Witkowski et al. 2000. DNA extraction, PCR and sequencing. Cells of investigated diatom species were centrifuged, frozen in liquid nitrogen and mechanically broken using a micropestell in a plastic reaction tube (Eppendorf 1.5 ml). DNA was then extracted using the Invisorb Plant Spin Kit (Invitek, Berlin, Germany) and Genomic Mini Kit (A&A Biotechnology, Gdynia, Poland). In both cases DNA extractions were carried out with extracting buffer as recommended by the manufacturers. To amplify SSU rDNA, diatom-specific primers (Brinkmann et al., unpubl.) were used, while for the ITS1+5.8S+ITS2 region it was the pair NS7 and LR1850 (White et al. 1990, Bhattacharya 1996). In both cases, PCR was performed in a 50 μl reaction volume containing a reaction mix of 2 mM MgCl2, 4% DMSO, 0.2 mM of each dNTP, 0.4 μM of each PCR primer, 5 μl template DNA and 1 U of Taq DNA polymerase (Bioline, Luckenwalde, Germany). The PCR protocol was also the same; after the initial denaturing step at 95°C for 5 min, 33 cycles of denaturing at 94°C for 1 min, annealing at 51°C 30 s and elongation at 72°C for 2 min followed by final extension at 72°C for

Table 1

Strains and species of Baltic diatoms used in this study, information on their collection, morphology and 18S rDNA and DNA- barcode sequence accession numbers..

Strain Collection information Literature for Species designation Accession number (isolation site and date) morphology comparison in the CCBA

Amphora coffeaeformis (C.A. Agardh) Kützing (1844) BA16 Gulf of Gdańsk; costal station Gdynia, Mar. 2003 Garduño et al.1996 HM805019a Sala et al. 1998 HM805021b Witkowski et al. 2000 Witkowski et al. 2000 HM805020a Bacillaria cf. paxillifer (O.F. Müller) Hendey (1964) BA14c Gulf of Gdańsk, May 1988 Schmid 2007 Beszteri et al. 2005 HM805030a Cyclotella meneghiniana F.T. Kützing (1844) BA10 Gulf of Gdańsk; costal station Gdynia, May 1986 Krammer & Lange-Bertalot 2008 HM805022b HM805031a Entomoneis punctulata (Grunow) Osada & Kobayasi (1990) BA83 Gulf of Gdańsk; costal station Swarzewo, Aug. 2003 Witkowski et al. 2000 HM805023b Fistulifera saprophila (H. Lange-Bertalot & K. Bonik) H. Lange-Bertalot (1997) BA55 Gulf of Gdańsk; costal station Sopot, Apr. 2003 Lange-Bertalot 2001 HM805032a HM805024b BA56 Gulf of Gdańsk; costal station Sopot, Apr. 2003 HM805033a Haslea spicula (W.J. Hickie) H. Lange-Bertalot (1997) BA28 Gulf of Gdańsk, costal station Władysławowo, Jul. 2003 Lange-Bertalot 2001 HM805034a Snoeijs & Vilbaste 1994 HM805037a Navicula gregaria Donkin (1861) BA102 Gulf of Gdańsk; costal station Kuźnica, Oct. 2005 Lange-Bertalot 2001 HM805026b Navicula perminuta Grunow in Van Heurck (1880) BA32 Gulf of Gdańsk; costal station Kuźnica, Oct. 2003 Lange-Bertalot 2001 HM805043a BA30 Gulf of Gdańsk; costal station Sopot, Sep. 2003 Lange-Bertalot et al. 2003 HM805044a HM805028b Nitzschia cf. fonticola Grunow in Cleve & Möller (1878) BA34 Gulf of Gdańsk; costal station Gdynia, Aug. 2003 Krammer & Lange-Bertalot 1988 HM805035a BA31 Gulf of Gdańsk; costal station Puck, Oct. 2003 HM805036a HM805025b Nitzschia microcephala Grunow in Cleve & Möller (1878) BA33 Gulf of Gdańsk; costal station Jurata, May 2003 Krammer & Lange-Bertalot 1988 HM805038a HM805027b BA85 Gulf of Gdańsk; costal station Jurata, Jul. 2004 Snoeijs & Vilbaste 1994 HM805039a BA100 Gulf of Gdańsk; costal station Gdynia, Aug. 2005 Witkowski et al. 2000 HM805040a BA101 Gulf of Gdańsk; costal station Puck, Oct. 2005 HM805041a BA29 Gulf of Gdańsk; costal station Puck, Oct. 2003 HM805042a Skeletonema marinoi Sarno et Zingone (2005) BA98 Gulf of Gdańsk; costal station Gdynia, Mar. 2005 Sarno et al. 2005 HM805045a HM805029b a – 18S rDNA sequence accession number, b – DNA-barcode accession number

8 F.F. Pniewski, T. Friedl, A. Latała

1 min. PCR products were purified with the Nucleo® Spin Extraction Kit (Macherey-Nagel, Düren, Germany) and quantified on 1% agarose gel. Purified PCR products were sequenced using BigDye® Terminator v 3.1 Cycle Sequencing Kit (Applied Biosystems, Darmstadt, Germany). For the sequencing the following standard primers were used: 34F, 569F, 891F, 1122F, 370R, 585R, 1122R, 1263R for 18S rDNA fragments, and NS7, ITS4, 5.8SbR, 5.8SbF, 1617F for the ITS1+5.8S+ITS2 region. Sequencing reaction was carried out on an ABI Prism 3100 (Applied Systems) sequencer. Overlapping sequence fragments were assembled into complete sequences using SeqAssem (Hepperle 2004b).

Alignment and phylogenetic analyses All sequences were first compared to sequences in the GenBank nucleotide database using BLAST search. The 18S rDNA sequences which were obtained were aligned with previously published diatom SSU rDNA sequences (e.g. Medlin et al. 1996, Besztari et al. 2001, Sinninghe Damsté et al. 2004, Sorhannus 2007) using BioEdit (Hall 1999). The final data set contained 82 taxa, including 80 diatoms and two species of Bolidomonas (B. mediterranea AF123596 and B. pacifica AF123595) which were employed as an outgroup. Subsequently, the alignment was manually corrected using the Align Manual Sequence Alignment Editor (Hepperle 2004a). Two independent types of analyses, i.e. maximum parsimony (MP) and Bayesian interference (BI), were used to construct the 18S rDNA phylogenies. Maximum parsimony analysis was implemented with the PAUP* V4.0b10 (Swofford 2001). For this alignment positional homology was assumed for 1614 positions out of 1848 possible and 553 of them were parsimony informative. Gaps were treated as missing data. Maximum parsimony trees were obtained using the tree-bisection- reconnection (TBR) branch-swapping algorithm and heuristic search with random addition of sequences. Stability of the tree topology was evaluated by bootstrap (BS) tests using 1000 replicates. Bayesian analysis was carried out using MrBayes 3.1 software (Ronquist&Huelsenbeck 2003). The analysis was run with GTR+G+I model; nucmodel=4by4, nst=6, rates=invgamma for 2 200 000 Markov chain Monte Carlo (MCMC) generations, saving every 10000th tree. To measure tree strength, bipartition posterior probabilities (BPP) were calculated. The 50% majority rule consensus trees were constructed and displayed using TreeView (Page 1996).

Distance rates estimations for “diatom barcodes” For the description of diatom barcode (Moniz&Kaczmarska 2009, 2010), DNA fragments of c. 1110-1300bp, including partial 18S rDNA + ITS1 + 5.8S

Molecular phylogeny of same diatoms... 9

+ ITS2 + partial 28S rDNA, were analysed. All fragments were first aligned using BioEdit (Hall 1999). The partial 18S and 28S as well as full 5.8S rDNA sequences were easily aligned clearly marking hypervariable ITS regions. The beginning of the barcode coincides with the beginning of 5.8S, thus it was easy to identify. The end of the barcoding sequence was found by searching short 5- letter sequence YGGTA in the highly conserved motif on helix III. Obtained barcodes were compared with other, recently published, sequences downloaded from the GenBank database (Moniz&Kaczmarska 2009, 2010). Next, genetic variability between sequences was estimated using uncorrected pairwise (p) distances in MEGA 4 (Tamura et al. 2007).

RESULTS AND DISCUSSION

Morphological strain identification The isolated diatom strains were identified to species level using morphological characters as visible by light and scanning electron microscopy (Fig. 1, Table 1). Characteristic diagnostic features are given in the legend to Fig. 1. Among the 18 isolated strains were planktonic and benthic species as well. Two strains represented planktonic centric diatom species of the order Thalassiosirales, i.e. C. meneghiniana, which is present in the waters of the Gulf of Gdańsk throughout the whole year, and a Skeletonema costatum-like strain which is known to form blooms in the Baltic in early spring. Of the 16 benthic pennate diatoms strains, 8 strains represented 3 species of nitzschioid diatoms (members of the order Bacillariales), i.e. Bacillaria cf. paxillifer (1), Nitzschia cf. fonticola (2) and N. microcephala (5). All other benthic species were naviculoid diatoms, identified as members of the orders Thalassiophysales, Amphora coffeaeformis, Surirellales, Entomoneis punctulata, and Naviculales, Fistulifera saprophila (2), Haslea spicula, Navicula gregaria and N. perminuta (2) (Fig. 1, Table 1).

18S rDNA phylogenetic analyses Nearly full 18S rDNA sequences were determined for 18 Baltic diatom strains here. The five strains of Nitzschia microcephala shared an identical sequence with each other. Also the sequences of the two strains of N. cf. fonticola were identical. Therefore, just a single sequence for each species was used in the phylogenetic analyses, i.e. BA29 and BA34, respectively. Both types of phylogenetic analyses, maximum parsimony (MP) and Bayesian inference (BI, Fig. 2), revealed rather similar topologies with the same overall relationships among the main diatom lineages. They were also in agreement with other current diatom phylogenies (e.g. Kooistra&Medlin 1996,

www.oandhs.org

10 F.F. Pniewski, T. Friedl, A. Latała

Fig. 1. Scanning electron (SEM) micrographs of Baltic diatoms from the CCBA collection: (A), (B) Fistulifera saprophila (BA-56) and (C), (D) Fistulifera saprophila (BA-55), valve outside showing slit-like opening of the “fistula” at one side of the central nodule (arrow) and valve inside showing distinct circular pore of the “fistula” at one side of the central nodule (arrowhead). System of straight striae forming an angle of 75-80º with the raphe. (E) Navicula perminuta (BA-30) and (F) Navicula perminuta (BA-32). (G) and (H) Amphora coffeaeformis (BA- 16). (I) Nitzschia microcephala (BA-29). (J) Cyclotella meneghiniana (BA-10), valve with tangential undulation of central area and with one central fultoportula (arrow) and marginal fultuportulae (arrowhead). (K) Skeletonema marinoi (BA-98), valves with intercalary fultoportula processes (IFPPs) connected with 1:1 plain joints and the intercalary rimoportula process IRPP (arrow). (L) and (M) Navicula gregaria (BA-102). (N) and (O) Entomoneis punctulata (BA-83). (P) Bacillaria cf. paxillifer (BA-14c). (Q) Nitzschia cf. fonticola (BA-34). (R) Haslea spicula (BA-28). A, C, E, G, I, J, M, N, O, Q and R - external view of the valve; B, D, F, H, L and P – internal view of the valve.

Molecular phylogeny of same diatoms... 11

Fig. 2. 18S rDNA phylogeny of 13 strains of Baltic diatoms from the CCBA culture collection and other members of centric and pennate diatoms (fifty percent majority rule Bayesian tree). Bootstrap values and bipartition posterior probabilities above 50 are shown on the nodes that were recovered in both of the molecular analyses (MP/BI); MP, Maximum Parsimony; BI, Bayesian Inference. Two strains of Bolidophyceae were used as outgroup taxa but pruned away from the tree.

www.oandhs.org

12 F.F. Pniewski, T. Friedl, A. Latała

Medlin et al. 1996, 2000, Medlin&Kaczmarska 2004, Sorhannus 2004, Theriot et al. 2009). The centric diatoms formed an array of several successive lineages at the base and received moderate to high bootstrap support. The Radial Centrics (represented by members of Paraliales, Coscinodiscales and Rhizosoleniales) were most basal followed by the “bi(multi)polar” orders Chaetocetorales and Lithodesmiales, and the Thalassiosirales clade. The two CCBA strains of planktonic centric diatoms formed monophyletic clades with strains representing the genera Cyclotella and Skeletonema, as was expected from their morphological identification. Baltic strain BA10 which was identified as C. meneghiniana by morphology (Fig. 1) was most closely related to strain G8W4 of Beszteri et al. (2005b) (sequence AY496212); both strains differed in only two sequence positions. C. meneghiniana is a common freshwater species (Finlay et al. 2002), although it has been considered to be a typical component of Baltic phytoplankton (Pliński 1995, Stoń et al. 2004) as well. C. meneghiniana has been found to be genetically diverse corroborating the great variability observed in their silica frustules (Beszteri et al. 2005a, b). Together with the finding that the species is well adapted to an extremely wide variety of habitats (Håkansson 2002), the assumption has arisen that C. meneghiniana may actually contain several cryptic species (Beszteri et al. 2005b, Kaczmarska et al. 2005). Phylogenetic analyses as well as comparisons of barcode sequences (described below) indicated that Baltic strain BA10 must be classified as C. meneghiniana. Another Baltic strain of centric diatoms, BA98, previously was assigned to Skeletonema costatum based on morphology (Fig. 1). Early investigations at the protein level (Gallagher 1982) have revealed great genetic diversity within S. costatum, which has led to the conclusion that it could be a species complex. Finally, results of recent thorough investigations allowed the distinction of several new Skeletonema species (Medlin et al. 1991, Sarno et al. 2005, Zingone et al. 2005). A BLAST search revealed that the Baltic Skeletonema-like strain BA98 is most similar (100%) to S. marinoi which is one of the newly described species. Our phylogenetic analyses showed strain BA98 nested within a clade representing the genus Skeletonema and, as expected, formed a clade together with S. marinoi (AJ632216). Our analyses also confirmed S. dohrnii as the closest relative of S. marinoi as suggested by the SSU+LSU rDNA phylogenies of Sarno et al. (2005) and the LSU rDNA phylogeny of Godhe et al. (2006). Pennate diatoms were supported as a monophyletic clade that was divided into “araphids” and “raphids” (Fig. 2). The Araphids (class Fragilariophyceae sensu Round et al. 1990) were found to be paraphyletic, while raphids (class Bacillariophyceae sensu Round et al. 1990) remained a well-supported monophyletic group, in agreement with previous studies ( Medlin&Kaczmarska 2004, Sims et al. 2006, Sorhannus 2007, Theriot et al. 2009).

Molecular phylogeny of same diatoms... 13

The three Baltic pennate diatoms of the order Bacillariales (“nitzschoid diatoms”), B. cf. paxillifer, N. cf. fonticola and N. microcephala, were divided into two sister subclades by both MP and BI analyses. The five N. microcephala strains (represented by strain BA29 in Fig. 2, Table 1), together with Fragilariopsis cylindrum, Nitzschia inconspicua and three Pseudo-nitzschia species, formed a well-supported subclade whose sister-group relationship with the genus Cylindrotheca was very significant (Fig. 2). However, the phylogenetic position of the species could not be unambiguously resolved. In the BI tree the N. microcephala strains were basal within the subclade (Fig. 2), while the MP analysis grouped both Nitzschia species together, albeit with rather weak BS support. The other two Baltic members of Bacillariales were positioned within another distinct subclade (Fig. 2). Strain B. cf. paxillifer BA14c grouped together with another strain representing B. paxillifer (M87325). The two Baltic strains of Nitzschia cf. fonticola (represented by BA34 in Fig. 2) and N. apiculata were grouped together in a common lineage which, however, was not supported in the significance tests. In the Bayesian analyses the two Bacillaria strains were sister to the Achnanthes brevipes/Psammodictyon panduriforme lineage, but this was also not supported in the significance tests (Fig. 2). The common origin of the other representatives of the genus Nitzschia received no BS support and was rather weakly supported by BI (Fig. 2). MP analyses did not resolve relationships within this subclade, i.e. there was a polytomy with strains BA14c and BA34 forming two independent lineages (not shown). The species of Nitzschia were intermixed with species of Bacillaria, Cylindrotheca, Fragilariopsis and Pseudo-nitzschia (Fig. 2), which confirmed that Nitzschia in its present circumscription is an artificial taxon. Previous analyses with a larger sample of nitzschioid species using 18S rDNA (Sinninghe Damsté et al. 2004, Sorhannus 2007) or large subunit (LSU) rDNA sequences (Lundholm & Moestrup 2002, Lundholm et al. 2002) have already suggested paraphyly of Nitzschia as currently circumscribed. These studies also confirmed monophyly of the genera Cylindrotheca, Bacillaria and Pseudo-nitzschia (which previously have been separated from Nitzschia, Lundholm et al. 2002), while the branching order within the nitzschioid diatoms differed among these studies. Baltic benthic pennate diatom strains of the order Naviculales were distributed on two independent lineages rendering the Naviculales in its present circumscription paraphyletic. The Baltic strains identified as Fistulifera saprophila by morphology formed a well supported clade together with F. pelliculosa (AY485454). The strains BA55 and BA56 differed by six positions in their 18S rDNA sequences. Strain BA55 had six sequence positions different with the F. saprophila AJ867025 sequence, but 4, 91 and 91 sequence differences as compared to the three F. pelliculosa sequences AY485454, AJ544657 and

www.oandhs.org

14 F.F. Pniewski, T. Friedl, A. Latała

EU260468. The corresponding strain BA56 sequence differed by 4, 93 and 93 positions from the latter three sequences, but had only a single difference between it and the F. saprophila AJ867025 sequence. Kaczmarska et al. (2005) suggested that five or more base substitutions may provide evidence for a species complex. Thus, both Fistulifera species may actually comprise several cryptic species that may be distinguished only at the molecular level, but this will need further morphological and molecular studies. The addition of two more sequences in the present study supported the monophyly of Fistulifera (Fig. 2). Moreover, the genus Fistulifera appeared distant from Navicula which supports its separation from the latter genus (Fig. 2). The three Baltic strains of Navicula, N. gregaria BA102, N. perminuta BA30 and BA32, formed a well supported clade with 5 other species of Navicula that were used as references to represent the genus (Fig. 2). Monophyly of N. gregaria BA102 with N. sclesciscensis was well supported in the Bayesian, but not in the MP analysis (Fig. 2). Both strains of N. perminuta, which were morphologically identical, had 18 positions different in their 18S rDNA sequences, but their monophyletic origin was well supported (Fig. 2). Therefore, following Kaczmarska et al. (2005), N. perminuta may also be regarded as a complex of several cryptic species. The Baltic strain Haslea spicula BA28 was with other species of the genus in a common clade, but monophyly of the genus received only moderate support (Fig. 2). The common origin of both genera, Haslea and Navicula, was, however, well supported. Our phylogenetic analyses clearly resolved the monophyletic origin of the Baltic strain A. coffeaeformis BA16 with A. montana within in a well-supported monophyletic Amphora clade (Thalassiophysales, Fig. 2). It was sister to a clade representing the order Surirellales. The Baltic strain E. punctulata BA83 was within the Surirellales clade, but its exact position remained ambiguous (Fig. 2). Neither the monophyletic origin of the order Naviculales sensu Round et al. (1990) nor monophyly of the traditional family Naviculaceae were resolved in our phylogenetic analyses. One part of the traditional Naviculaceae, the clade uniting Navicula and Haslea (naviculoids sensu stricto of Medlin & Kaczmarska 2004), was sister group to the family Pleurosigmataceae. The genera Fistulifera and Eolimna whose species have formerly been included in Navicula sensu Round et al. 1991 were confirmed here as entities clearly separated from Navicula, in congruence with their formal separation from the latter as based on morphology by Lange-Bertalot (1997). Interestingly, Eolimna was found paraphyletic as it has already been revealed in previous studies (e.g. Beszteri et al. 2001, Medlin & Kaczmarska 2004). While E. subminuscula was well resolved as a close relative to Fistulifera spp., E. minima shared a well supported single origin with Sellaphora pupula (Fig. 2). Fistulifera and Eolimna formed together with Phaeodactylum tricornutum, Sellaphora pupula, the orders Surirellales and Thalassiophysales, another

Molecular phylogeny of same diatoms... 15 deeply diverging clade of pennate raphid diatoms (Fig. 2). An increasing number of ecological studies in the Baltic Sea have revealed the occurrence of numerous unusually small-celled taxa belonging to Navicula s. str. (Lange- Bertalot et al. 2003), which can hardly be identified at the species level using morphological characters. The taxonomy of naviculoid diatoms is far from being resolved yet and phylogenetic analyses using rDNA sequences will certainly reinforce a systematic study of the group and reveal their actual diversity.

DNA barcodes and genetic distance estimation DNA barcode sequences were determined for 9 out of the 13 Baltic strains used in the 18S rDNA phylogeny (Fig. 2). For three strains, B. cf. paxillifer BA14c, F. saprophila BA56 and N. perminuta BA30, PCR amplification of the ITS2 region was unsuccessful; the strain Haslea spicula BA28 got extinct. Intra- and interspecific genetic (p) distance were calculated with the DNA barcodes according to Moniz&Kaczmarska (2009). The alignments for distance analyses were prepared with a gap penalty of 10 and a gap extent penalty of 1.2 as in Moniz&Kaczmarska (2010). Comparisons of the barcodes from strains Cyclotella meneghiniana BA10 and Skeletonema marinoi BA98 to the NCBI sequence database using BLAST (Altschul et al. 1990) confirmed the identities of both strains. The comparison of the strain BA10 barcode sequence with corresponding sequences of 30 other C. meneghiniana strains available in the NCBI database showed uncorrected genetic distances (p) varying from 0.01 to 0.07 diff./site (differences per site) with a mean of 0.02 diff./site. These values fitted the range of intraspecific distance changes (maximum distance between two sequences within the same morpho-species; 0.00-0.08 diff./site) as calculated by Moniz & Kaczmarska (2010) for species belonging to the Thalassiosiraceae. With the comparison of the strain S. marinoi BA98 barcode with corresponding available sequences, the range of intraspecific distance variations was even lower, i.e. 0.00-0.01 diff./site with an average value of 0.004 diff./site. However, also the average interspecific distances (minimum distance between two sequences from the same morpho-genus) were very low, comparable with intraspecific variations observed in other species. For example, the average p value calculated from comparing the S. marinoi BA98 barcode to 7 corresponding sequences of S. dohrnii was just 0.03 diff./site. Variations of interspecific distances calculated for species from the Skeletonemataceae family by Moniz & Kaczmarska (2010) were the smallest (0.02-0.08 diff./site) among all analysed families which suggests a very close relationship among the species of this family.

www.oandhs.org

16 F.F. Pniewski, T. Friedl, A. Latała

For six out of the seven barcode sequences obtained for Baltic pennate diatoms the BLAST searches did not find any significant match. Only for the strain N. gregaria BA102 was its taxonomic identity confirmed using DNA barcode sequences. Intraspecific distances among strain BA102 and three available N. gregaria sequences was 0.04 diff./site. Pairwise comparisons of N. gregaria BA102 and another Baltic strain of the same genus, N. perminuta BA32, showed an interspecific distance of 0.15 diff./site which is within the range of the interspecific variability within the family Naviculaceae as calculated by Moniz & Kaczmarska (2010), i.e. p=0.13 to 0.34 diff./site. By comparing the barcodes of the two Baltic Navicula strains with corresponding available sequences of five other species of the genus (N. arenaria, N. salinicola, N. incerta, N. cf. phyllepta and Navicula sp. CCMP 2122) it was found that p distances varied within the range of 0.09-0.27 diff./site which fitted well within the same range of interspecific sequence variation as calculated by Moniz&Kaczmarska (2010). However, genetic distances of the Baltic strain F. saprophila BA55 (formerly included in Navicula) with the other mentioned Navicula species were much higher, i.e. they ranged from 0.28 to 0.35 diff./site with an average value of 0.31 diff./site. It clearly confirmed that strain BA56 is genetically distant from other species of Navicula. The A. coffeaeformis BA16 barcode sequence did not match those of the same species available in the NCBI database (GQ330306, GQ330307, GQ330309). Uncorrected genetic distance estimated for those sequences were within the range 0.27-0.28 diff./site which is rather high and suggests that strain BA16 represents a different species. For Baltic strain E. punctulata BA83 no barcode sequences representing the same genus were available. Its lowest p distance was with Baltic strain Amphora coffeaeformis BA16 (0.23 diff./site) which is in agreement that both strains formed closely related sister-groups in the 18S rDNA phylogeny (Fig. 2). The pairwise distances between both Baltic strains of Nitzschia (N. cf. fonticola BA34 and N. microcephala BA29) were with 0.18 diff./site within the range of the interspecific distance variations (0-0.43 diff./site) estimated for the Bacillariaceae by Moniz & Kaczmarska (2010). The comparison of the barcodes from the two Baltic Nitzschia strains with those of 17 other members of Bacillariaceae (Cylindrotheca, GQ330326, FJ864277, FJ864278; Nitzschia, AY574377, AY574378, AY574379; Pseudo-nitzschia, DQ813827, GQ330424, GQ330422, GQ330384, GQ330382, GQ330380, EU599141, DQ813835, DQ813834, DQ330430, DQ813836) revealed a range of 0.13 to 0.19 diff./site, which again fitted well into the range estimated for the Bacillariaceae by Moniz&Kaczmarska (2010).

Molecular phylogeny of same diatoms... 17

ACKNOWLEDGEMENTS

We would like to thank Prof. Bożena Bogaczewicz-Adamczak and Dr Aleksandra Zgrundo for identifying diatom species, Dr Małgorzata Witak and MSc. Dorota Jankowska for fruitful discussion on the diatoms morphology and Anna Łatkiewicz for help producing the SEM images. Sincere thanks to Svenja Bruns and Dr Nicole Brinkmann for providing assistance to F.P. with the sequence analyses. Parts of this work were supported by the German Science Foundation (DFG) by a grant extended to T.F. (Fr 905/13-2) and by the Polish State Committee for Scientific Research (N304 219035).

REFERENCES

Altschul S.F., Gish W., Miller W., Myers E.W., Lipman D.J., 1990, Basic local alignment search tool, Journal of Molecular Biology, 215: 403-410 Alverson A.J., 2008, Molecular Systematics And The Diatom Species, Protist, 159: 339-353 Batterbee R.W., 1986, Handbook of Holocene Palaeoecology and Palaeohydrology, John Wiley & Sons Ltd. Beszteri B., Ács É., Makk J., Kovács G., Márialigeti K., Kiss, K.T., 2001, Phylogeny of six naviculoid diatoms based on 18S rDNA sequences, Inter. J. Sys. and Evolut. Microbiol., 51: 1581-1586. Beszteri B., Ács É., Medlin L.K., 2005a, Conventional and geometric morphometric studies of valve ultrastructural variation in the two closely related Cyclotella species (Bacillariophyta). Eur. J. Phycol., 40: 89-103. Beszteri B., Ács É., Medlin L.K., 2005b, Ribosomal DNA sequence variation among sympatric strains of Cyclotella meneghiniana complex (Bacillariophyceae) reveals cryptic diversity. Protist, 156: 317-33 Bhattacharya D., 1996, Analysis of bootstrap tree lengths using the maximum parsimony method. Mol. Phylogenet. Evol. 6: 339-350 Bortolus A., 2008, Error cascades in the biological sciences: the unwanted consequences of using bad taxonomy in ecology. Ambio, 37: 114-118. Choi H-G., Joo H.M., Jung W., Hong S.S., Kang J.-S., Kang S.-H., 2008, Morphology and phylogenetic relationships of some psychrophilic polar diatoms (Bacillariophyta). Nov. Hedwig. Beih., 133: 7-30 Finlay B.J., Monaghan E.B., Maberly S.C., 2002, Hypothesis: the rate and scale of dispersal of freshwater diatom species is a function of their global abundance. Protist, 153: 261-73 Gallagher J.C., 1982, Physiological variation and electrophoretic banding patterns of genetically different seasonal populations of Skeletonema costatum (Bacillariophyceae). J. Phycol. , 18: 148-162 Garduño R.A., Hall B.D., Brown L., Robinson M.G., 1996, Two distinct colonial morphotypes of Amphora coffeaeformis (Bacillariophyceae) cultured on solid media. J. Phycol., 32: 469-478 Godhe A., Mcquoid M.R., Karunasagar I., Karunasagar I., Rehnstam-Holm A.-S., 2006, Comparison of three common molecular tools for distinguishing among geographically separated clones of the diatom Skeletonema marinoi Sarno et Zingone (Bacillariophyceae). J. Phycol., 42: 280-291

www.oandhs.org

18 F.F. Pniewski, T. Friedl, A. Latała

Guillard R.L., 1975, Culture of phytoplankton for feeding marine invertebrates. In Culture of marine invertebrate animals. (Smith, W.L. & Chanle, M.N. editors), 29-60. Plenum Press, New York Håkansson H., 2002, A compilation and evaluation of species in the genera Stephanodiscus, Cyclostephanos and Cyclotella with new genus in the family Stephanodiscaceae. Diatom Res., 17: 1-139 Hall T.A., 1999, BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symp. Ser., 41: 95-8 Hepperle D., 2004a, Align Ver.07/04©. Multisequence alignment-editor and preparation/manipulation of phylogenetic datasets. Win32-Version. Distributed by the author via: http://www.sequentix.de Hepperle D., 2004b, SeqAssem©. A sequence analysis tool, counting assembler and trace data visualization tool for molecular sequences. Win32-Version. Distributed by the author via: http://www.sequentix.de Hoshaw R.W., Rosowski J.R., 1979, Methods of microscopic algae. In Handbook of phycological methods. Cultures methods & growth measurements. (Stein, J. R. editor), 53-67. Cambridge University Press, Cambridge Jodłowska S., Latała A., 2010, Photoacclimation strategies in the toxic cyanobacterium Nodularia spumigena (Nostocales, Cyanobacteria). Phycologia, 49: 203-211 Kaczmarska I., Beaton M., Benoit A.C., Medlin L.K., 2005, Molecular phylogeny of selected members of the order Thalassiosirales (Bacillariophyta) and evolution of the fultoportula. J. Phycol., 42: 121-138 Kooistra W.H.C.F., Medlin L.K., 1996, Evolution of the diatoms (Bacillariophyta). IV. A reconstruction of their age from small subunit rRNA coding regions and the fossil record. Mol. Phylogenet. Evol., 6: 391-407 Krammer K., Lange-Bertalot H., 1988, Bacillariophyceae 2. Epithemiaceae. Bacillariaceae, Surirellaceae. In Süsswasserflora von Mitteleuropa 2 (Ettl, H., Gerloff, J., Heynig, H. & Mollenhauer, D., editors), t. 2. Fischer. Stuttgart. Krammer K., Lange-Bertalot H., 2008, Bacillariophyceae 2. Centrales. Fragilariaceae, Eutoniaceae. In Süsswasserflora von Mitteleuropa 2 (Ettl, H., Gerloff, J., Heynig, H. & Mollenhauer, D., editors), t. 2. Fischer. Stuttgart. Lange-Bertalot H., 1997, Frankophila, Mayamaea und Fistulifera: drei neue Gattungen der Klasse Bacillariophyceae. Archiv für Protisten Kunde. 148: 65-76 Lange-Bertalot H., 2001, Navicula sensu stricto. 10 genera separated from Navicula sensu lato. Frustulia. In Diatoms of Europe. Diatoms of the European inland waters and comparable habitats. (Lange-Bertalot, H. editor), Vol. 2. A.R.G. Gantner Verlag K.G. Lange-Bertalot H., Witkowski A., Bogaczewicz-Adamczak B., Zgrundo A., 2003, Rare and new small-celled taxa of Navicula s. str. in the Gulf of Gdansk and in its freshwater affluents. Limnologica, 33: 258-270 Latała A., 2003, Autecological characteristic of some algal strains from Culture Collection of Baltic Algae (CCBA). In Biological resource centers and the use of microbes (Lima, N. & Smith, D., editors), 323-345. Braga, Portugal Latała A., Jodłowska S., Pniewski F., 2006, Culture Collection of Baltic Algae (CCBA) and characteristic of some strains by factorial experiment approach. Algological Studies, 122: 137-154 Latała A., Nędzi M., Stepnowski P., 2009a, Toxicity of imidazolium and pyridinium based ionic liquids towards algae. Chlorella vulgaris, Oocystis submarina (green algae) and Cyclotella meneghiniana, Skeletonema marinoi (diatoms). Green Chemistry, 11: 580-588

Molecular phylogeny of same diatoms... 19

Latała A., Nędzi M., Stepnowski P., 2009b, Toxicity of imidazolium and pyridinium based ionic liquids towards algae. Bacillaria paxillifer (a microphytobenthic diatom) and Geitlerinema amphibium (a microphytobenthic blue green alga). Green Chemistry, 11: 1371-1376 Latała A., Nędzi M., Stepnowski P., 2009c, Acute toxicity assessment of perfluorinated carboxylic acids towards the Baltic microalgae. Environmental Toxicology and Pharmacology, 28: 167-171 Lundholm N., Moestrup Ø., 2002, The marine diatom Pseudo-nitzschia galaxiae sp. nov. (Bacillariophyceae): morphology and phylogenetic relationships. Phycologia, 41: 594-605 Lundholm, Daugbjerg N.N., Moestrup Ø., 2002, Phylogeny of the Bacillariaceae with emphasis on the genus Pseudo-nitzschia (Bacillariophyceae) based on partial LSU rDNA. Eur. J., Phycol., 37:115-134 Medlin L.K., Elwood H.J., Strickel S., Sogin M.L., 1991, Morphological and genetic variation within the diatom Skeletonema costatum (Bacillariophyta): evidence for a new species, Skeletonema pseudocostatum. J. Phycol., 27: 514-524 Medlin L.K., Kaczmarska I., 2004, Evolution of diatoms: Morphological and cytological support for the major clades and a taxonomic revision. Phycologia, 43: 245-270 Medlin L.K., Kooistra W.H.C.F., Gersonde R., Wellbrock U., 1996, Evolution of the diatoms (Bacillariophyta). II. Nuclear-encoded small-subunit rRNA sequences comparisons confirm paraphyletic origin for centric diatoms. Mol. Biol. Evol., 13: 67-75 Medlin L.K., Kooistra W.C.H.F., Schmid A-M.M., 2000, A review of the evolution of the diatoms – a total approach using molecules, morphology and geology. In The origin and early evolution of diatoms: fossil, molecular and biogeographical approaches (Witkowski, A. & Sieminska, J. editors), 13-35. W. Szafer Institute of Botany, Polish Acadamy of Sciences, Cracow Moniz M.B.J., Kaczmarska I., 2009, Barcoding diatoms: Is there a good marker?, Molecular Ecology Resources, 9: 65-74 Moniz M.B.J., Kaczmarska I., 2010, Barcoding of diatoms: Nuclear encoded ITS revisited. Protist, 161: 7-34 Page R.D.M., 1996, Treeview: An application to display phylogenetic trees on personal computers. Comput. Appl. Biosci., 12: 357-8 Pliński M., 1995, Fitoplankton in Zatoka Pucka (Korzeniewski, K., editor), 395-415. Wydawnictwo Uniwersytetu Gdańskiego, Gdańsk. (in Polish) Ronquist F., Huelsenbeck J.P, 2003, Mrbayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics, 19:1572-1574 Round F.E., Crawford R.M., Mann D.G., 1990, The Diatoms. Biology and Morphology of the Genera. Cambridge University Press, Cambridge Sala S.E., Sar E.A., Ferrario M.E., 1998, Review of materials reported as containing Amphora coffeaeformis (Agardh) Kützing in Argentina. Diatom Res., 13: 323-336 Sarno D., Kooistra W.H.C.F., Medlin L.K., Precopo I., Zingone A., 2005, Diversity in the genus Skeletonema (Bacillariophyceae). II. An assessment of the taxonomy of S. costatum-like species with the description of four new species. J. Phycol., 41: 151-176. Schmid A-M., 2007, The “paradox” diatom Bacillaria paxillifer (Bacillariophyta) revisited. J. Phycol., 43: 139-155 Sims P.A., Mann D.G., Medlin L.K., 2006, Evolution of diatoms: insights from fossil, biological and molecular data. Phycologia, 45: 361-402 Sinninghe Damsté S., Muyzer G., Abbas B., Rapen S.W., Masse G. et al., 2004, The rise of the rhizosolenoid diatoms. Science, 304: 584-7 Snoeijs P., Vilbast S., 1994, Intercalibration and Distribution of Diatom Species in The Baltic Sea. Vol. 2. Opulus Press. Uppsala

www.oandhs.org

20 F.F. Pniewski, T. Friedl, A. Latała

Sorhannus U., 2004, Diatom phylogenetics inferred based on direct optimization of nuclear- encoded SSU rRNA sequences. Cladistics. 20: 487-497 Sorhannus U., 2007, A nuclear-encoded small-subunit ribosomal RNA timescale for diatom evolution. Marine Micropaleontology, 65: 1-12. Sorhannus U., Gasse F., Perasso R., Baroin Tourancheau A.,1995, A preliminary phylogeny of diatoms based on 28S ribosomal RNA sequence data. Phycologia, 34: 65-73 Stoń J., Kosakowska A., Łotocka M., 2002, Pigment composition in relation to phytoplankton community structure and nutrient content in The Baltic Sea. Oceanologia, 44: 419-37 Swofford D.L., 2001, PAUP*. Phylogenetic Analysis Using Parsimony (* And Other Methods). Version 4.0b8. Sinauer, Sunderland, MA Tamura K., Dudlmy J., Nei M., Kumur S., 2007, MEGA4: Molecular Evolutionary Genetics Analysis (MEGA) software version 4.0. Molecular Biology and Evolution, 24:1596-1599 Theriot E.C., Cannone J.J., Gutell R.R., Alverson A.J., 2009, The limits of nuclear-encoded SSU rDNA for resolving the diatom phylogeny. Eur. J., Phycol., 44: 277-290 White T.J., Bruns T.D., Lee S.B., Taylor J.W., 1990, Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. [In:] PCR protocols: a guide to methods and applications. Academic Press, London, 315-322 Witkowski A., Lange-Bertalot H., Metzeltin, D., 2000, Diatom Flora of Marine Coast. I. In Iconographia Diatomologica 7 (Lange-Bertalot, H. Gantner A.R.G., editors), Verlag K.G Zingone A., Percopo I., Sims P.A., Sarno D., 2005, Diversity in the genus Skeletonema (Bacillariophyceae). I. A reexamination of the type material of S. costatum with the description of S. grevillei sp. nov. J. Pycol., 41: 140-150