Isolation and characterization of a motility-defective mutant of Euglena gracilis

Strains, culture, and media

The wild-type Z strain, which is identical to E. gracilis SAG 1224-5/25, was provided by the culture collection of the Institute of Applied Microbiology (IAM), University of Tokyo, Tokyo, JPN. The ${\mathrm{M}}_3^{-}$ZFeL mutant strain produced in this study is deposited in the microbial culture collection at the National Institute for Environmental Studies (NIES, Tsukuba, JPN) with registration number NIES-4440. The cells were cultured using Cramer–Myers (CM) (Cramer & Myers, 1952) and Koren–Hunter (KH) media (Koren & Hutner, 1967) for autotrophic and heterotrophic growth, respectively, prepared at pH 3.5. The CM medium does not include a carbon source, while the KH medium includes 12 g L⁻¹ of glucose and various organic acids and amino acids as carbon sources. The seed culture was maintained in KH medium with each month of subculture.

Fe-ion irradiation and mutant screening

The mutant strain was obtained through Fe-ion irradiation using the same conditions as previously reported (Yamada et al., 2016a; Yamada et al., 2016b), and mutant screening was based on the algal phenotypes. Two milliliters of an E. gracilis cell suspension (4 × 10⁵ cells mL⁻¹) was placed in hybridization bags that were segmented into 5 × 7 cm compartments. They were then irradiated by Fe ions (linear energy transfer [LET]: 650 keV µm⁻¹) at a dose of 50 Gy in the RIKEN RI-beam factory (Wako, Saitama, JPN). After one week of recovery by culture inoculation in KH culture medium, two rounds of mutant screening were conducted.

About 1 × 10⁵ mutagenized cells with independent genetic backgrounds were placed at one end of a culture dish filled with 10 mL of liquid KH medium with 0.5% methyl cellulose, which was added to solidify and prevent unexpected stirring of the culture, and were illuminated from the opposite end with 50 µmol photons m⁻² s⁻¹ of fluorescent light. After one week of culture incubation following Fe-ion beam irradiation, cells proliferated more than 100 times, resulting in 1 × 10⁷ mutagenized cells, and most of them were found at the illuminated side through phototaxis. About 1,000 of the cells remained at the initial inoculation point; these were collected as putative non-motile mutant cells with a micropipette.

The collected population was supposed to include a high proportion of motile cells, which resided at the initial inoculation point due to random dispersal; therefore, they were subjected to a second round of screening. The cells were placed at the opening end of a 15 mL conical tube filled with 5 mL of liquid KH medium with 0.5% methyl cellulose, illuminated with 50 µmol photons m⁻² s⁻¹ of fluorescent light, and shaded with aluminum foil, except at the closing end. After 2 weeks of static horizontal incubation in the tubes, the cells proliferated more than 1,000 times. Most cells were gathered at the unshaded end to seek light. We retrieved about 1,000 of the cells that showed no movement from the initial inoculation position and cultured them for an additional two weeks.

The proliferated cells were randomly isolated using fluorescence-activated cell sorting (MoFlo XDP; Beckman Coulter, Brea, CA, USA) to establish clonal lines in individual wells of 96-well plates (Tissue Culture Test Plate; TPP, Trasadingen, CHE) filled with 200 μL of KH medium. Populations without cell motility were selected by microscopic observations at 26 °C and identified as the motility-defective strains. The above-mentioned screening was performed for two independently mutagenized populations. During the screening process, cells proliferated more than 10⁶ times; therefore, each population included many genetically identical cells. Based on this possibility, one mutant strain was established from each population, i.e., two strains were established from the two independent populations.

Morphological characterization

The cells were observed under an upright light microscope (DM2500B; Leica, Wetzlar, DEU) equipped with a differential interference contrast module. Scanning electron microscopy (SEM) was conducted using a field emission scanning electron microscope (SU8200; Hitachi, Tokyo, JPN) after following a general sample preparation. Briefly, 1 mL of culture in KH medium was centrifuged at 2,000 × g for 1 min to collect the cells. The precipitated cells were pre-fixed by adding a solution containing 2.5% glutaraldehyde, 2% paraformaldehyde, and 0.1 M sodium cacodylate, and fixed again using a solution containing 1% osmium tetroxide and 0.1 M sodium cacodylate. The fixed samples were completely dried using a critical point dryer, sputter-coated with osmium, and then subjected to SEM photographing.

For quantifying the proportion of cells with flagella, the cells were cultured in CM or KH medium and used in their logarithmic growth phase for analysis. For the CM medium, the cells were cultured in 50 mL of medium using a 100 mL volume test tube aerated with 50 mL min⁻¹ of air containing 5% CO₂. For the KH medium, each strain was cultured in 50 mL of medium using a 100 mL volume conical flask with rotary shaking at 100 rpm. Each culture was conducted at 26 °C with 100 µmol m⁻² s⁻¹ of constant illumination and sub-cultured every week to maintain a stable proliferation. The cells were then fixed with glutaraldehyde (0.025%) and observed for the presence of flagella under an inverted microscope (CKX41; Olympus, Tokyo, JPN) equipped with a phase contrast module. To exclude subjective judgment, short and intact flagella were not differentiated and were counted as cells with flagella.

Growth tests

The algae growth rate was evaluated in 100 mL test tubes containing 50 mL of medium. The cells were inoculated with an initial optical density (OD) of 0.1 and precultured for three days in 100 mL volume conical flasks containing KH medium with continuous shaking (100 rpm at 26 °C, and 100 m⁻² s⁻¹ of constant illumination). The cultured cells were then collected by centrifuging at 2,000 × g for 5 min and washed twice with either KH or CM medium, which was subsequently used for the culture test. Next, the cell suspension was inoculated in culture tubes with an initial OD of 0.1 and incubated at 26 °C with 100 µmol m⁻² s⁻¹ of constant illumination. KH and CM cultures were aerated with 50 mL min⁻¹ of air containing 5% CO₂. ODs from culture growth were determined over time by spectrophotometer (UVmini-1240; Shimadzu, Kyoto, JPN) at λ 680 nm, immediately after placing the suspension in the cuvette.

Quantitative analysis of motility

We used microscopic observations with video image capture to quantitatively evaluate algae motility. Observation and image processing techniques were used as previously reported (Ozasa et al., 2011). We confined the cells in suspension (0.7 µL, containing 300–500 cells) in a closed circular polydimethylsiloxane (PDMS) micro-chamber (2.49 mm diameter and 140 µm depth), and observed red-light, bright-field transmission images with a 5 × objective lens at 26 °C. Approximately 300–500 cells were measured in the micro-chamber. Cell movements were visualized by differentiating, thresholding, and superimposing the video images sequentially (trace image), and evaluated by counting the spatial sum of the trace pixels in the trace image (Ozasa et al., 2013; Ozasa et al., 2014). The number of trace pixels was named the “trace momentum” (TM) (Ozasa et al., 2013; Ozasa et al., 2014). The measurement rate of the TM values was 0.67 Hz (one frame per 1.6 s). The TM value is a fair measure of all locomotive activity observed within the chamber. The typical swimming traces were individually superimposed onto the cell distribution image and prepared as a binary image to demonstrate the cell’s swimming activity (Ozasa et al., 2016).

Sedimentation analysis

Cellular sedimentation rates in KH and CM media at 26 °C were evaluated in 1.5 mL microtubes. The culture was prepared as indicated above for the morphology analysis. In brief, when CM medium was used, cells were cultured in 100 mL volume test tubes with aeration, whereas with KH medium, they were cultured in a 100 mL volume conical flask with rotary shaking at 100 rpm. Each culture was conducted with constant light illumination (100 µmol m⁻² s⁻¹). Culture ODs average at λ 680 nm were 23.4 and 22.0 by KH culture, and 4.1 and 4.4 by CM culture of wild-type and ${\mathrm{M}}_3^{-}$ZFeL strains, respectively. Images of the cell suspension in each microtubule were taken at 1–5 min intervals with a general compact digital camera. The transparent supernatant area was detected and quantified from the images using the Image J software, with the threshold values set as 0 to 50 after converting to greyscale. Identical rectangular regions were selected inside the images of microtube images, and the transparent area inside each region was quantified.

To evaluate the sedimentation speed at 20 cm depth, the E. gracilis wild type and the ${\mathrm{M}}_3^{-}$ZFeL mutant strains were cultured in cuboid acryl beakers (10 × 10 × 30 cm) using 2 L of CM medium at 29 °C and 1,300 µmol m⁻² s⁻¹ of overhead illumination (light–dark ratio of 12:12 h). Beakers were filled with approximately the same concentrations (g L⁻¹) of cells, 0.37 and 0.32 g L⁻¹ for the wild-type and M⁻3ZFeL strains, respectively. After stirring the culture, sedimentation was periodically observed for 3.5 h, and supernatant was carefully removed to obtain a concentrated culture. Sediment concentration was then quantified by weighing the dried cells after filtration with a glass fiber filter (GA-55; ADVANTEC, Tokyo, JPN).

Quantification of carbohydrates and lipids

The harvested algal cells were freeze-dried (FDV-1200; EYELA, Tokyo, JPN), and the deproteinized paramylon and lipid components were extracted as previously reported (Inui et al., 1982; Suzuki et al., 2015). The dried cells (10 mg) were sonicated twice in 10 mL of acetone for 90 s (UD-201; TOMY, Tokyo, JPN) with dial setting at 4 in 50 mL volume of plastic centrifuge tubes. The paramylon was subsequently separated from the residual components using centrifugation, boiled for 30 min in 10 mL of 1% sodium dodecyl sulfate aqueous solution, and washed twice with 10 mL of water. The extracted paramylon was then quantified using the phenol–sulfuric acid method (Montgomery, 1957), which can quantify total carbohydrates. Similarly, neutral lipids were extracted from 100 mg of dried cells using n-hexane as a solvent; 10 mL of n-hexane was added to the dried cells in 50 mL glass centrifuge tubes. The suspension was then homogenized for 90 s using a sonicator (UD-201; TOMY, Tokyo, JPN) with dial setting at 4, and then filtered with a piece of glass fiber filter paper (GF/C; Whatman, Little Chalfont, Buckinghamshire, UK), followed by an additional step of residue extraction. After evaporating the collected organic solvent dissolving lipids, the weight of the residue left in the flask was quantified as that of the extracted total neutral lipid.

Statistical analyses

The results with error bars, except that of motility quantification, are represented as mean  ± standard error from three independent experiments. The results of motility quantification are represented as mean  ± standard deviation (SD) for the whole measured time points. Statistical significance was analyzed using Student’s t-test. For multiple comparisons, Bonferroni’s corrections were applied. p < 0.05 was considered significant.

Characterization of ${\mathrm{M}}_3^{-}$ ZFeL

Through our screening for non-motile mutants of E. gracilis, we established two mutants with independent genetic backgrounds. The mutant strains were named ${\mathrm{M}}_3^{-}$ZFeL and M⁻₄ZFeL according to the typical nomenclature (Schiff, Lyman & Russell, 1980), with the phenotypic designation “M” for motility and the mutagen designation “Fe” for Fe-ion irradiation (Yamada et al., 2016a). The strain ${\mathrm{M}}_3^{-}$ZFeL showed a stable behavioral phenotype, and was thus used in this study.

The E. gracilis ${\mathrm{M}}_3^{-}$ZFeL strain showed non-motile phenotype with few defects in proliferation. This phenotype was identified by the formation of colonies from single-cell inoculation in liquid culture. The ${\mathrm{M}}_3^{-}$ZFeL strain showed colony formation in liquid KH culture medium after a week, whereas wild-type cells were dispersed in the same culture medium (Figs. 1A and 1B). As judged by both light and electron microscopy, the M⁻₃ZFeL cells were more rounded than were the wild-type cells. In addition, the ${\mathrm{M}}_3^{-}$ZFeL flagella were shorter than those of the wild type (Figs. 1C–1F). Moreover, a significantly higher proportion of ${\mathrm{M}}_3^{-}$ZFeL cells lacked flagella both in the CM (p = 5.6 × 10⁻⁵, t(4) = 15.1) and KH (p = 0.04, t(4) = 2.29) media (Figs. 1G and 1H).

The autotrophic and heterotrophic growth rates of the wild-type and M⁻₃ZFeL strains were evaluated using KH and CM culture media (Fig. 2). The mutant strain showed slightly slower and faster growth than did the wild type in the KH and CM culture, respectively. The ${\mathrm{M}}_3^{-}$ZFeL culture ODs at the third (p = 0.020, t(4) = −5.81) and fourth (p = 1.5 × 10⁻⁴, t(4) = −20.48) days of KH culture, which included high amounts of glucose, were significantly lower than those of the wild type (Fig. 2A). On the other hand, the ODs on the second (p = 0.020, t(4) = −6.13), fourth (p = 0.042, t(4) = −4.98), and eleventh (p = 0.036, t(4) = −5.18) days of CM culture were significantly higher than those of the wild type (Fig. 2B).

Our quantification of motility showed that the ${\mathrm{M}}_3^{-}$ZFeL TM values were less than one-thousandth that of the wild type. Following previous studies, cells cultured in a liquid KH medium were adequately stirred and subjected to TM measurements. Figure 3A shows the cell distribution and swimming traces over approximately 8 s. The absence of swimming traces for the mutant strain revealed that cells could not move using their flagella. The total TM value ± SD for the wild-type strain was 11,161 ± 269 for 368 cells, whereas that for ${\mathrm{M}}_3^{-}$ZFeL was only 9 ± 6 for 467 cells, as shown in Fig. 3B. The TM value for the mutant strain cannot be distinguished from the measurement noise, indicating that the ${\mathrm{M}}_3^{-}$ZFeL strain was almost completely non-motile. Although some of the ${\mathrm{M}}_3^{-}$ZFeL cells possessed a short flagellum (Figs. 1C–1H), our results indicated that the short flagellum did not function well.

Potential of ${\mathrm{M}}_3^{-}$ ZFeL for industrial application

${\mathrm{M}}_3^{-}$ZFeL demonstrated faster sedimentation than did the wild type in 1.5 mL microtubes. As shown in Fig. 4, the ${\mathrm{M}}_3^{-}$ZFeL cell sedimentation rate was higher than that of the wild-type cells under both heterotrophic (Fig. 4A) and autotrophic (Fig. 4B) growth conditions using KH and CM media, respectively. The values at the final time point (40 min) were significantly different under both heterotrophic (p = 0.010, t(4) = −3.69) and autotrophic (p = 0.011, t(4) = −3.60) conditions (Figs. 4A and 4B).

By analyzing sedimentation in a larger-scale culture, ${\mathrm{M}}_3^{-}$ZFeL was found to be advantageous for industrial harvesting. To assess the effect of fast sedimentation on the harvest of cells cultured using photosynthesis, wild-type and ${\mathrm{M}}_3^{-}$ZFeL strains were cultured using CM medium and their sedimentation rates were measured in 20 cm-deep beakers. ${\mathrm{M}}_3^{-}$ZFeL cells showed a significantly faster sedimentation rate than did the wild-type cells (Fig. 5). The ${\mathrm{M}}_3^{-}$ZFeL cell density at the bottom of the beaker (34.1 g L⁻¹), was more than 10 times that of the wild type (3.0 g L⁻¹).

The paramylon storage in ${\mathrm{M}}_3^{-}$ZFeL cells was equal to or greater than that in the wild-type cells, whereas lipid accumulation in ${\mathrm{M}}_3^{-}$ZFeL cells was less than (autotrophic) or equal to (heterotrophic) that that in the wild-type cells. As shown in Figs. 6A and 6B, ${\mathrm{M}}_3^{-}$ZFeL cells produced a higher amount of paramylon (1.6 times; p = 0.033, t(4) = −2.51) than did the wild-type cells under autotrophic conditions, whereas there was no significant difference under heterotrophic conditions. In contrast, the ${\mathrm{M}}_3^{-}$ZFeL strain showed a lower lipid content than did the wild type, especially under autotrophic conditions (p = 0.034, t(4) = −2.47). (Figs. 6C and 6D).