**1. Introduction**

The genus *Micractinium* with its type species, *M. pusillum*, was described by Fresenius [1] for a coccoid green alga, which formed colonies of 2–4 cells and produced bristles. Since the first description, several species of *Micractinium* were established based on cell shape, number of bristles, and arrangemen<sup>t</sup> of cells into colonies [2]. All species occurred in all kinds of freshwater habitats, such as lakes and small ponds, and were typical planktonic species. Phylogenetic analyses of *Micractinium* surprisingly showed that *M. pusillum* is closely related to the genus *Chlorella*, a unicellular green alga without any cell appendices. Luo et al. [3,4] have demonstrated that the colony and bristle formation was a response on grazing through the rotifer *Brachionus calyciflorus*. The SSU and ITS rDNA sequences revealed that *M. pusillum* represented a cryptic species complex [4,5]. In addition, Pröschold et al. [6] transferred the genus *Diacanthos* with its type species *D. belenophorus* to the genus *Micractinium*. Apart from these free-living species of *Micractinium*, Pröschold et al. [7] indicated that a green algal endosymbiont of the ciliate *Paramecium bursaria* also belonged to *Micractinium*. Brandt [8] was the first who discovered that "chlorophyll-bearing bodies" in *Paramecium bursaria* and

*Stentor polymorphus* were independent organisms and not plastids. Since then, endosymbiotic algae in ciliates, heliozoa, amoeba, or other invertebrates have been of special interests in phycology as well as in zoology, microbiology, and virology. Within ciliates, green algal endosymbionts are widely distributed. Around 40 species of ciliates and other protists live in symbiosis with green algae [9]. For most of these endosymbionts, the origin and phylogenetic position are unknown. The majority of the investigated green algae belong to the *Chlorella* clade of the Trebouxiophyceae ([7] and references therein). Interestingly, the endosymbionts do not form a single lineage within the *Chlorella* clade, but are closely related to free-living species of *Chlorella*, *Micractinium* [7], and *Meyerella* [10], and sometimes formed an own genus like *Carolibrandtia* [11,12].

Symbiotic interactions between green algae and ciliates are known to be of different nature. Some mixotrophic ciliates always bear zoochlorellae in their cells and rarely occur algae-free. Such obligate endosymbiosis is found for example in *Paramecium bursaria*, one of best investigated ciliate species [13]. In contrast, several ciliates live only facultatively in symbiosis with green algae. One of these ciliates is the recently described *Tetrahymena utriculariae*, which lives in symbiosis with the alga *Micractinium* [14]. *T. utriculariae* lives inside bladder traps of *Utricularia reflexa*, a carnivorous aquatic plant. The ciliate survives the typically anoxic and nutrient-rich milieu inside traps, most likely because of its green algal endosymbionts. Cultivated outside the traps under oxygenic conditions, the ciliates lose their endosymbionts and switch to a heterotrophic way of life. This clearly indicated that the green algal endosymbiont *Micractinium* has a special function by providing oxygen to its hosts [15].

The aim of this study was to clarify the phylogenetic position and the taxonomic status within *Micractinium*. We isolated the strain from its host *Tetrahymena utriculariae* and deposited it under the number SAG 2587 in the Culture Collection of Algae at the University of Göttingen. We used an integrative approach (morphology and phenotypic plasticity, SSU, and ITS rDNA sequences including their secondary structures) for comparing this strain with existing described species of *Micractinium*.

#### **2. Material and Methods**

The strain SAG 2587 was isolated from the host as described in Pitsch et al. [14] and cultivated on agarized basal medium with peptone (ESP; medium 1b in [16]). For morphological investigations, we cultivated the strain at 18 ◦C, with 50 μmol photons/m2s1 provided by daylight fluorescent tubes (Osram L36W/954 Lumilux de lux daylight, Munich, Germany), and light:dark cycle of 16:8 hrs for two to three weeks. The light microscopic investigations were conducted using an Olympus BX-60 microscope (Olympus, Tokyo, Japan) and the micrographs were taken with a ProgRes C14plus camera using the ProgRes CapturePro imaging system (version 2.9.0.1, both from Jenoptik, Jena, Germany).

The genomic DNA of the strain was extracted using the DNeasy Plant Mini Kit (Qiagen, Hilden, Germany) following the instructions provided by the manufacturer. The SSU and ITS rDNA was amplified in PCR reactions using the Taq PCR MasterMix Kit (Qiagen, Hilden, Germany) with the primers EAF3 and ITS055R [17]. The PCR product was purified and sequenced as described by Darienko et al. [18]. The SSU and ITS rDNA sequence is available in the EMBL, GenBank, and DDBJ sequence databases under the accession number MT359915. This sequence was aligned and included into a data set of a total of 40 sequences (2602 bp) of representatives of the Chlorellaceae (Trebouxiophyceae). The data set was aligned according to the secondary structures. The secondary structures were folded using the software mfold [19], which uses the thermodynamic model (minimal energy) for RNA folding. GenBank accession numbers of all sequences used are given in the figure. For the phylogenetic analyses, the dataset with unambiguously aligned base positions was used. To test which evolutionary model fitted best for the data set, we calculated the log-likelihood values of 56 models using the automated model selection tool implemented in PAUP, version 4.0b167 [20], and the best model according to the Akaike criterion by PAUP was chosen for the analyses. The setting of the best model is given in the figure legend. The following methods were used for the phylogenetic

analyses: distance, maximum parsimony, maximum likelihood, and Bayesian inference. Programs used included PAUP version 4.0b167 [20], RAxML version 8.2.12 [21], MrBayes version 3.2.7a [22], and the PHASE package 2.0 [23–27]. For the Bayesian calculations, the secondary structure models of SSU and ITS (doublet in MrBayes and RNA7D in PHASE) were also taken into account.
