**1. Introduction**

Symbiosis of green algae with protists and invertebrates has been studied for more than 100 years ([1] and references therein). Living in a mutualistic relationship has advantages for both ciliate and algae. The ciliate can survive starvation under nutrient limitation and prevent against damages induced by solar irradiation. The algae are protected against infection by chloroviruses, which lyse the endosymbionts outside of their hosts [2–4]. Such mixotrophic ciliates are widely distributed in many freshwater habitats and belong to di fferent phylogenetic lineages [5–7]. The endosymbiotic green algae are commonly assigned to *Chlorella*-like organisms or simply named as zoochlorellae [1,8,9]. So far, *Paramecium bursaria* is the most studied model ciliate in endosymbiosis research because of its easiness of cultivating and cloning and it can be identified rather easily from its morphology [10,11]. *P. bursaria* contains up to 500 green algal endosymbionts, which are located in perialgal vacuoles [12]. In contrast, identification of the endosymbionts of *P. bursaria* based solely on morphology is almost impossible

(Figure 1), and the isolation and cultivation of these zoochlorellae is quite difficult and time-consuming, which was already recognized by Pringsheim [13] and Loefer [14]. The few algal strains available in public culture collections were characterized by Pröschold et al. [1] using an integrative approach based on their morphology and phylogenetic analyses. Pröschold et al. [1] identified four different green algal species isolated from different *Paramecium bursaria* strains: *Chlorella variabilis*, *C. vulgaris*, *Micractinium conductrix*, and an unidentified species of *Scenedesmus*. However, most strains of *Paramecium bursaria* bore either *Chlorella variabilis* or *Micractinium conductrix* as endosymbionts and they were assigned to an American (or Southern) and a European (or Northern) group by Gaponova et al. [15], Hoshina et al. [16–18], and Hoshina and Imamura [19,20], respectively. A clear identification of the endosymbionts is of special interests since the discovery of highly specific chloroviruses, which infected these green algae when they were released from their hosts ([21] and references therein).

**Figure 1.** Morphology of *Chlorella variabilis* and *Micractinium conductrix* in their host *Paramecium bursaria* strains and cultivated as axenic strains. (**A**,**B**) *Paramecium bursaria* strains CIL-16 (**A**) and SAG 27.96 (**B**), respectively, (**C**) CCAP 211/84 = NC64A; (**D**) SAG 241.80.

The aim of this study was to develop (i) a protocol for the isolation of green algal endosymbionts and (ii) a quick and precise identification method without previously isolating them from their hosts. The isolation method focuses on the separation of green algal endosymbionts from other free-living organisms and the special nutrient requirement for the growth of these zoochlorellae. The easy diagnostic PCR approach uses species-specific primers, which focused on the internal transcribed spacer region 2 (ITS-2) of the nuclear ribosomal operon, often used for species delimitation among the Chlorellaceae before ([1,22] and references therein). Moreover, the ITS-2 region was selected on the basis of the exact delineation at the species level using the compensatory base change (CBC) concept introduced by Coleman [23]. This concept uses the CBCs in the conserved region of the ITS-2 secondary structure. Coleman [23–25] found that if two specimens differed in at least one CBC in the conserved region of ITS-2 (helices II and III), both were not able to mate and therefore represented

two di fferent biological species. Pröschold et al. [1] demonstrated that the endosymbiotic species had several CBCs in their ITS-2 secondary structures and can be clearly distinguished. In addition, Pröschold et al. [1] raised the question if the occurrence of a green algal taxon was correlated with the geographical origin of its host or/and with the a ffiliation to a certain ciliate syngen (= biological species). Bomford [26] investigated the mating behavior among several isolates of *P. bursaria* and discovered six syngens by conjugation experiments. Greczek-Stachura et al. [9] confirmed the syngen pattern by sequencing of the nuclear ITS rDNA, the mitochondrial cytochrome *c* oxidase subunit I (COI), and the histone H4 gene. However, they focused only on the ciliate phylogeny and therefore, which green algal endosymbionts were present in the di fferent syngens remains unknown.

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

#### *2.1. Cultivation and Molecular Characterization of Paramecium bursaria*

The investigated strains were collected from around the world and cultivated in modified Bold Basal Medium (3N-BBM+V; medium 26a in [27]) with the addition of 30 mL of soil extract per liter final medium (called S/BBM). The soil extract was prepared as described in Schlösser [28]. Origin and details about the investigated *Paramecium bursaria* strains are listed in Table 1. All cultures were maintained at 15–21 ◦C under a light:dark cycle of 12:12 h (photon flux rate up 50 μmol m<sup>−</sup><sup>2</sup> s<sup>−</sup>1). Genomic DNA of the green algae was extracted using the DNeasy Plant Mini Kit (Qiagen GmbH, Hilden, Germany). The SSU and ITS rDNA were amplified using the Taq PCR Mastermix Kit (Qiagen GmbH, Hilden, Germany) with the primers EAF3 and ITS055R [29]. The SSU and ITS rDNA sequences of the *Paramecium bursaria* strains were aligned according to the secondary structures, resulting in a dataset of 19 sequences (2197 bp). GenBank accession numbers of all newly deposited sequences were given in the phylogenetic tree and Table 1. For the phylogenetic analyses, we calculated the log-likelihood values of 56 models using the automated selection tool implemented in PAUP version 4.0b167 [30] to test which evolutionary model fitted best for the dataset. The best model according to the Akaike criterion by PAUP was chosen for the analyses. The settings of the best model are given in the figure legend. The following methods were used for the phylogenetic analyses: Distance, maximum parsimony, and maximum likelihood, all included in PAUP version 4.0b167 [30].

The secondary structures were folded using the software mfold [31], which uses the thermodynamic model (minimal energy) for RNA folding. The visualization of the structures was done using the program PseudoViewer 3 [32].



#### *2.2. Isolation of the Green Algal Endosymbionts*

The endosymbiotic green algae are often not easy to distinguish from free-living green algae in the surrounding media. Therefore, a special isolation method needed to be developed to avoid that free-living algae, or contaminations grow during the isolation process. With the following procedure as demonstrated in the work scheme (Figure 2), we obtained several axenic clonal strains in culture from different mixotrophic ciliates, such as *Paramecium bursaria*.

**Figure 2.** Working scheme for isolating the green algal endosymbionts of *Paramecium bursaria*.

For the isolation of their green algal endosymbionts, single ciliate cells were washed several times and transferred into fresh S/BBM medium (step 1). After starvation and digestion of any food, after approximately 24 h, cells were washed again (step 2) and the ciliates were transferred onto agar plates containing basal medium with beef extract (ESFl; medium 1a according to Schlösser, [28]). Before placement of the ciliates onto agar plates, 50 μL of an antibiotic mix (mixture of 1% penicillin G, 0.25% streptomycin, and 0.25% chloramphenicol) were added to prevent bacterial growth. The agar plates were kept under the same conditions as described. After growth (~40 days), the algal colonies were transferred onto agar slopes (1.5%) containing ESFl medium and were kept under the described culture conditions (step 3).

#### *2.3. Diagnostic PCR Amplification*

The isolation of the green algal endosymbionts is time-consuming and not always successful, especially if *Chlorella variabilis* is the endosymbiont. To investigate which green algal endosymbiont was present in a set of *P. bursaria* strains without isolating them, we developed a diagnostic PCR method using species-specific primers. Pröschold et al. [1] demonstrated that the SSU and ITS rDNA of *Chlorella variabilis* and *Micractinium conductrix* contain three introns and one intron, respectively, which makes an easy PCR amplification not possible. Therefore, we amplified the ITS-2 sequences for both *C. variabilis* and *M. conductrix*, a sequence which is diagnostic. For designing species primers, the internal transcribed spacer 1 (ITS-1) rDNA sequences of all representatives (58 species) belonging to the Chlorellaceae were compared to find characteristic positions. The following diagnostic primers for both species were designed based on the SSU and ITS rDNA sequences (*C. variabilis*, NC64A = CCAP 211/84; FN298923, and *M. conductrix*, SAG 241.80, FM205851; see Figure 3): CvarF: CTCGTGCTGTCTCACTTCGGTG and MconF: GTAGGCGCAGCCTCGTGTTGTGAC.

**Figure 3.** ITS-1 secondary structure of *Chlorella variabilis* (blue) and *Micractinium conductrix* (yellow). The positions of the diagnostic primers CvarF and MconF are highlighted in white boxes in the structures.

These primers were tested in combination with the reverse primer ITS055R [29], both on the isolated axenic algal cultures and their hosts. For identification of the endosymbionts, both primer combinations (CvarF/ITS055R and MconF/ITS055R) were tested on 18 investigated *Paramecium bursaria* strains. All PCR amplifications were done on a thermocycler with the following program: 5 min initial denaturation at 95 ◦C, followed by 30 cycles (1 min at 95 ◦C, 2 min at 55 ◦C and 3 min at 68 ◦C), and final synthesis for 10 min at 68 ◦C.
