Identification of 13 Spirogyra species (Zygnemataceae) by traits of sexual reproduction induced under laboratory culture conditions

The genus Spirogyra is abundant in freshwater habitats worldwide, and comprises approximately 380 species. Species assignment is often difficult because identification is based on the characteristics of sexual reproduction in wild-collected samples and spores produced in the field or laboratory culture. We developed an identification procedure based on an improved methodology for inducing sexual conjugation in laboratory-cultivated filaments. We tested the modified procedure on 52 newly established and genetically different strains collected from diverse localities in Japan. We induced conjugation or aplanospore formation under controlled laboratory conditions in 15 of the 52 strains, which allowed us to identify 13 species. Two of the thirteen species were assignable to a related but taxonomically uncertain genus, Temnogyra, based on the unique characteristics of sexual reproduction. Our phylogenetic analysis demonstrated that the two Temnogyra species are included in a large clade comprising many species of Spirogyra. Thus, separation of Temnogyra from Spirogyra may be untenable, much as the separation of Sirogonium from Spirogyra is not supported by molecular analyses.

www.nature.com/scientificreports www.nature.com/scientificreports/ necessary for correct assignment to species and genera 4,6,7 . Previous taxonomic studies have used wild-collected filaments containing mature zygospores 9,11 . However, the seasonality of mature zygospore formation often prevents the collection of sexually mature wild filaments 12 . Less than 10% of the species of Spirogyra-like algae have been subjected to reliable taxonomic and phylogenetic analyses 9,11 . Laboratory cultivation to induce sexual reproduction is a promising approach for the development of a modern taxonomic system for Spirogyra and its relatives.
Allen 13 induced conjugation of three vegetative types of Spirogyra (Groups I-III) in culture using agar plates. She identified only a single species, S. pratensis, within Group I, based on vegetative and sexual characteristics. Hoshaw et al. 14 used a similar agar plate method to induce conjugation of Spirogyra vegetative filaments in culture and identified three species, S. singularis, S. communis and S. fragilis. Zwirn et al. 15 investigated the induction of conjugation in cultures of 95 Spirogyra strains under diverse experimental conditions including cultures on agar plates. Although they identified eight strains as six species of Spirogyra, morphological traits that are essential to correctly classify these Spirogyra species 4,6 were not shown. Recently, zygospores were formed with high efficiency by incubating cultured vegetative cells on agar plates of newly modified medium in two species of Spirogyra 16 .
The procedure has not been tested on other species in the Zygnemataceae. In this study, we established clonal cultures of Spirogyra-like algae from many samples collected in diverse localities around Japan. Zygospores or aplanospores were induced in 13 species using the newly-modified method, which was based on the procedures of Allen 13 and Ikegaya et al. 16 .

Methods
Sample collection and isolation. Samples containing filaments of Spirogyra-like algae were collected from ponds or paddy fields in Japan (Supplementary Table S1). Clones were established from fragmented filaments using a pipette-washing procedure 17 . The clones were grown in 100 mL of Closterium (C) medium 18 Table S1).

Induction of conjugation.
The 122 isolated strains were classifiable into 52 rbcL-types by differences in rbcL sequences; thus, we selected 52 strains with different rbcL sequences for further study (JPS001-JPS052; Supplementary Table S1). We attempted to induce conjugation in these strains.
In order to induce conjugation of Spirogyra, Allen 13 and Ikegaya et al. 16 incubated vegetative filaments on agar plates under 70-90 μmol photons m −2 s −1 illumination (500 ft-c 13 = ca. 70 μmol photons m −2 s −1 ) on a 16:8 or 12:12 hour light/dark cycle, respectively. In this study, we used the agar plates of the medium of Ikegaya et al. 16 and increased the light intensity; light intensity is reportedly important for the induction of conjugation in Spirogyra 13,19 .
Following the procedures of Ikegaya et al. 16 , we suspended agar powder (Wako, Osaka, Japan) in Artificial Pond Water medium (APW: 0.1 mM KCl, 0.1 mM CaCl 2 , 1 mM NaCl with 1 mM HEPES-Na buffer; pH 7.0) to make up 1% (w/v) agar plates in Petri dishes (90 mm × 15 mm). Actively growing filaments (ca. 100) in each culture were rinsed with liquid APW for 5 min and transferred onto agar plates, which were then sealed with surgical tape. The plates were incubated for ca. 2 weeks at 20 °C under 120 μmol photons m −2 s −1 illumination (cool-white fluorescent lamps, FL40SW; NEC Lighting, Tokyo, Japan) on a 14:10 hour light/dark cycle. Conjugation was observable after this time period when sexual reproduction had been successfully induced.

Light microscopy.
Microscopic observations were made with a BX53 microscope (Olympus, Tokyo, Japan) equipped with Nomarski differential interference optics. DNA sequencing. DNA extraction was performed following previously described procedures 20,21 . Cells were shaken with ceramic beads in chloroform and cetrimonium bromide using a ball mill (Mixer Mill MM 300; Retsch, Haan, Germany). DNA was extracted with an Illustra ™ blood genomic Prep Mini Spin Kit (GE Healthcare UK, Little Chalfont, UK). RuBisCO Large subunit (rbcL) genes and ATP synthase beta subunit (atpB) were amplified by polymerase chain reaction (PCR) using previously designed primers 10,22,23 (Supplementary   Tables S2, S3). PCR products were purified with an Illustra ™ GFX PCR DNA and Gel Band Purification Kit (GE Healthcare UK). Purified PCR products were sequenced directly using an ABI PRISM 3100 s Genetic Analyzer (Applied Biosystems, Foster City, CA, USA) with a BigDye Terminator Cycle Sequencing Ready Reaction Kit (ver. 3.1; Applied Biosystems).

Phylogenetic analyses.
We analysed 138 ingroup (Zygnematophyceae) and five outgroup (Coleochaete) operational taxonomic units (OTUs) of different rbcL sequences (Supplementary Tables S1, S4). The 138 ingroup OTUs included our 52 rbcL-types from Japan (Supplementary Tables S1, S4). The 1,332 base pairs of the rbcL gene sequences from the 143 OTUs were aligned using Clustal X software 24 . In addition, 1206 base pairs of the atpB gene sequences from the same 143 OTUs were aligned by Clustal X. The combined data set from rbcL and atpB genes [available from TreeBASE (https://www.treebase.org/treebase-web/home.html); study ID S24057] was subjected to maximum likelihood (ML) and Bayesian inference (BI) analyses. Then, ML analysis with 1,000 bootstrap replications 25 was performed with RAxML v. 8.0.0 software 26 using the GTR + CAT + I model and partitioning the dataset into first, second and third codon. BI analysis was performed with MrBayes v. 3.2.6 software 27 , as previously described 28 (1,000,000 generations of Markov chain Monte Carlo iterations; the first 25% of the iterations were discarded as burn-in). The substitution models for each partition of BI were GTR + I + G (first and third codon positions) and K2 + I + G (second codon position), which were selected with MEGA 7.0.21 software 29 .

Results
Sexual reproduction. Among the 52 strains with different rbcL sequences (Supplementary Table S1), we induced formation of zygospores or aplanospores in 15 (Figs 1-4) that were assigned to 13 species (Table 1).
www.nature.com/scientificreports www.nature.com/scientificreports/ Twelve species formed zygospores and one species formed aplanospores. Nine of the twelve species that formed zygospores developed conjugation tubes from both male and female gametangia. The other three of the twelve that formed zygospores produced conjugation tubes from only the male gametangia; two of these three underwent unequal division of the mother gametangial cell to form one small and one large daughter cell, which developed into sterile and gametangial cells, respectively. We classified the observed modes of zygospore and aplanospore formation into five categories: Types C1-C5 ( Supplementary Fig. S2). Type C1: formation of aplanospores.
Only Spirogyra mirabilis formed aplanospores ( Fig. 3f) and its aplanosporangia were inflated. Although Transeau 4 reported that S. mirabilis originating from North America formed both aplanospores and ladder-like conjugation tubes (Type C3), we did not observe conjugation in cultured material from Japan. Type C2: lateral conjugation (conjugation between adjoining cells in a single filament).
Nine species belonged to this category (S. dentireticulata Type C4: ladder-like conjugation; conjugation tubes often formed by a papilla from the male gametangium, lacking sterile cells adjacent to gametangia. S. hopeiensis (Fig. 1f) and S. chenii (Fig. 2j) belonged to this category. Type C5 (Temnogyra-type): ladder-like conjugation; conjugation tubes often formed by a papilla from the male gametangium, sterile cells adjacent to gametangia.
Mature zygospores and aplanospores. We recognised three types of zygospores or aplanospores (Types Z1-Z3; Supplementary Fig. S3). In the first type, the zygospores and aplanospores were ovoid in shape (resembling a watermelon) 4 , and the mesospores were single-or double-layered (Type Z1). In the second type, the zygospore was ellipsoidal (resembling an American football) 4 and the mesospores were single-layered (Type Z2). In the third type, the zygospore was lenticular (a compressed spheroid) and the mesospores were single-or double-layered (Type Z3). www.nature.com/scientificreports www.nature.com/scientificreports/ Type Z1: ovoid zygospore/aplanospore.
Vegetative morphology. Vegetative cells of the 52 strains were classified into three categories (Types V1-V3; Supplementary Fig. S4 and Table S5), based on differences in the transverse cell walls and numbers of chloroplasts. Type V1: plane transverse wal l and single chloroplast.

Discussion
The 15 strains in which zygospores or aplanospores were induced were clearly assignable to 13 species (Table 1). Two of the thirteen species (S. corrugata and S. punctata) were assignable to Temnogyra according to their sexual reproduction characteristics (Fig. 4). These two species formed a robust clade separated from Sirogonium species within a large monophyletic group (SST group) that included all of the Spirogyra OTUs examined in our phylogenetic analysis (Fig. 5a). Therefore, the two "Temnogyra" species should be re-assigned to the genus Spirogyra. However, information on the type species of Temnogyra, T. collinsii I. F. Lewis 31 , was not available during the present study. Further studies that include T. collinsii will be needed to clarify the taxonomic status of the genus Temnogyra.
UTEX 1745 is labelled "Spirogyra liana" 32 in the Texas culture collection; it has been the subject of previous phylogenetic analyses, and we included it in our study (Fig. 5b) 11 . S. liana has been recognised as a member of "Temnogyra" based on its conjugation characteristics 7 . However, there is no information on the sexual reproduction traits of UTEX 1745. This strain was phylogenetically separated from the two species assigned to Temnogyra (S. corrugata and S. punctata) based on their peculiar sexual reproduction characteristics (Fig. 4).
We obtained molecular information on eight Spirogyra species for the first time. Among these, there were three rare species: the Japanese endemic S. minuticrassoidea, and two species that had not previously been recorded in Japan, S. pseudomaxima and S. dentireticulata. These species probably seldom reproduce sexually in www.nature.com/scientificreports www.nature.com/scientificreports/ the wild. Hence, our cultivation procedure for inducing conjugation in vegetative cells proved useful for the identification of Spirogyra species that seldom, or rarely, sexually reproduced in nature. It is likely that more cryptic species will be revealed by inducing sexual reproduction in a range of established culture strains.  Table S1). Note that OTUs of I-VII correspond to those of Clades I-VII of Stancheva et al. 11 and "S. sp. " may belong to Temnogyra or Sirogonium because of lack of sexual reproduction characteristics. www.nature.com/scientificreports www.nature.com/scientificreports/ Among 52 genetically different strains, we were able to identify 15 that were assignable to 13 species. We were not able to obtain adequate information on conjugation in the remaining 37 strains, which were consequently unidentifiable at the species level. The difficulty of inducting sexual reproduction may have been a product of the procedure that we used. We attempted to induce conjugation using single clonal cultures to identify homothallic sexuality. Although heterothallic sexuality has not been demonstrated previously in Spirogyra or Sirogonium 11,33,34 , it cannot be ruled out for the culture strains that we were unable to assign to species. The procedures for induction should therefore be modified and improved so that the taxonomic status of strains such as these can be resolved.
Spirogyra-like algae are diverse with approximately 380 species widely distributed all over the world [4][5][6]9,11 . However, only 13 species of these algae were correctly identified in at species level using cultured materials originating only from Japan (Table 1). Thus, further taxonomic studies are needed based on more culture strains of Spirogyra-like algae from various localities of the world in order to construct more reliable and through taxonomic systems of Spirogyra and its related genera. However, some problems are recognized in species with wide distribution. S. varians UTEX479 collected in England and S. varians chi0102 (JPS015) from Japan are closely related (Fig. 5b). In contrast, S. longata RSS031 from California and S. longata chiA305 (JPS005) and kit0201 (JPS006) collected in Japan (Supplementary Table S1) are phylogenetically separated (Fig. 6). Similarly, S. majuscula RSS006 from California and S. majuscula chi0202 (JPS007) (Supplementary Table S1) from Japan are separated, as well as S. pratensis UTEX928 from USA and S. pratensis UTEX1746 from India as discussed in the previous studies (Figs 5b and 6) 8,11 . Therefore, morphological species concept based on only light microscopic characteristics of vegetative and reproductive phases may not delineate natural or monophyletic species as previously suggested 35 . Although it is necessary to induce and examine sexual reproduction characteristics of more worldwide culture strains to progress taxonomic studies of Spirogyra-like algae, more detailed observations using scanning and transmission electron microscopes combined with molecular phylogeny are need to recognize actual species of these algae.

Data Availability
All the other data generated and analysed during this study are included in this published article and its Supplementary Information. Figure 6. Details of part of the ML tree based on the combined data set from rbcL and atpB genes (Fig. 5a) showing a section corresponding to Clades I, II and VII of Stancheva et al. 11 . For details of the explanation of the tree, see Fig. 5.