Isolation and Characterization of Euglena gracilis-Associated Bacteria, Enterobacter sp. CA3 and Emticicia sp. CN5, Capable of Promoting the Growth and Paramylon Production of E. gracilis under Mixotrophic Cultivation

Euglena gracilis produces paramylon, which is a feedstock for high-value functional foods and nutritional supplements. The enhancement of paramylon productivity is a critical challenge. Microalgae growth-promoting bacteria (MGPB) can improve microalgal productivity; however, the MGPB for E. gracilis remain unclear. This study isolated bacteria capable of enhancing E. gracilis growth and paramylon production under mixotrophic conditions. Enterobacter sp. CA3 and Emticicia sp. CN5 were isolated from E. gracilis grown with sewage-effluent bacteria under mixotrophic conditions at pH 4.5 or 7.5, respectively. In a 7-day E. gracilis mixotrophic culture with glucose, CA3 increased E. gracilis biomass and paramylon production 1.8-fold and 3.5-fold, respectively (at pH 4.5), or 1.9-fold and 3.5-fold, respectively (at pH 7.5). CN5 increased E. gracilis biomass and paramylon production 2.0-fold and 4.1-fold, respectively (at pH 7.5). However, the strains did not show such effects on E. gracilis under autotrophic conditions without glucose. The results suggest that CA3 and CN5 promoted both E. gracilis growth and paramylon production under mixotrophic conditions with glucose at pH 4.5 and 7.5 (CA3) or pH 7.5 (CN5). This study also provides an isolation method for E. gracilis MGPB that enables the construction of an effective E. gracilis–MGPB-association system for increasing the paramylon yield of E. gracilis.


Sewage Effluent Sample and Collection of Indigenous Bacterial Communities from the Effluent
Sewage effluent was collected from the final sedimentation tank of a conventionally activated sludge process of a sewage-treatment plant in Kofu City, Yamanashi, Japan. The water quality of the effluent was 11.8 mg L −1 of total organic carbon, 6.5 mg L −1 NH 4 -N, 0.08 mg L −1 NO 2 -N, 4.2 mg L −1 NO 3 -N, and 3.2 mg L −1 PO 4 -P. The effluent sample was first passed through a glass microfiber filter (pore size, 1.6 µm; GF/A grade; GE Healthcare UK, Ltd., Little Chalfont, UK) to remove suspended solids and organisms larger than bacteria (including microalgae) from the effluent sample. Therefore, the effluent filtrate included an indigenous bacterial community. To collect the indigenous bacteria, the effluent sample (300 mL) was filtered through a sterilized membrane filter (pore size, 0.2 µm; polytetrafluoroethylene; Merck Millipore Ltd., Cork, Ireland), which were then placed into 30 mL of C medium in a 50-mL tube, vortexed at maximum speed for 1 min, shaken at 150 rotations/min (rpm) for 60 min, and vortexed at maximum speed for 1 min to detach the bacterial cells from the filter and suspend them in C medium. The number of culturable bacteria in the bacterial suspension was quantified using R2A agar plates (0.5 g L −1 peptone, 0.5 g L −1 yeast extract, 0.5 g L −1 casamino acid, 0.5 g L −1 glucose, 0.5 g L −1 soluble starch, 0.3 g L −1 K 2 HPO 4 , 0.05 g L −1 MgSO 4 ·7H 2 O, and 0.3 g L −1 sodium pyruvate (pH 7.0), and agar 15 g L −1 ). The culturable bacterial density was 5.7 × 10 5 CFU mL −1 .

Culturing E. gracilis with a Bacterial Community Derived from Sewage Effluent and Isolating Bacteria Associated with E. gracilis
Corn steep liquor (CSL; Oji Cornstarch Co., Ltd., Tokyo, Japan) was used as an organic carbon source for the mixotrophic cultures. Ten milliliters of E. gracilis (pre-cultured in C-NH 4 medium with 0.5 g L −1 CSL at pH 4.5 or 7.5) was added to 100 mL of C-NH 4 medium with 0.5 g L −1 CSL at pH 4.5 or 7.5 in a 200-mL glass flask, into which 10 mL of the sewage-effluent bacterial suspension was inoculated. The E. gracilis sewage-effluent bacterial cultures in C-NH 4 with CSL at pH 4.5 or 7.5 were incubated in a growth chamber with shaking at 150 rpm for 10 d. Additionally, E. gracilis cultures in C-NH 4 medium with 0.5 g L −1 CSL at pH 4.5 or 7.5 but without sewage-effluent bacteria were prepared and incubated as bacteria-free control cultures. The experiments were conducted in triplicate. After 10 d of culture, 25 mL of the E. gracilis-effluent bacterial culture was transferred into a 50-mL tube and vortexed at maximum speed for 3 min to disperse the E. gracilis and bacterial cells. The sample was then filtered through a GF/A glass microfiber filter to remove E. gracilis cells. The filtrate containing bacteria was serially diluted and spread on R2A agar plates with a pH of 4.5 or 7.5, which were incubated at 28 • C for 2 weeks. In this study, bacteria obtained from cultures at pH 4.5 were defined as acidophilic bacteria, and those obtained from cultures at pH 7.5 were defined as neutrophilic bacteria. Eight acidophilic bacteria and 15 neutrophilic bacteria were isolated, and a pure culture of each strain was maintained on R2A agar at pH 4.5 or 7.5. Because CSL reportedly increases the biomass and paramylon productivity of E. gracilis [9], CSL was initially used as an organic carbon source for the mixotrophic cultures of E. gracilis; however, CSL is a mixture of organic compounds. For detailed examination, we used pure glucose as the organic carbon source for E. gracilis mixotrophic culture in subsequent experiments.

Screening of MGPB Capable of Enhancing Both E. gracilis Growth and Paramylon Production
Each isolated bacterial strain was cultured in R2A liquid medium (pH 4.5 or 7.5) at 28 • C and with shaking (150 rpm) until the late logarithmic growth phase. The cells were harvested by centrifugation (10,000× g, 24 • C, and 5 min) and washed twice with C-NH 4 medium (pH 4.5 or 7.5). Each acidophilic or neutrophilic bacterial cell culture was inoculated into E. gracilis C-NH 4 medium at pH 4.5 or 7.5 with 5 g L −1 glucose at an optical density at 600 nm (OD 600 ) of 0.05. The co-cultures of E. gracilis with each isolated bacterial strain were incubated in a growth chamber with shaking at 150 rpm for 7 d. The initial biomass of E. gracilis was approximately 40 mg dry weight L −1 . On day 7, concentrations of E. gracilis chlorophyll a + b, biomass, and paramylon were measured. Control cultures including E. gracilis cells alone (without bacterial inoculation) in C-NH 4 medium (pH 4.5 or 7.5) were also prepared and analyzed similarly for comparison. The E. gracilis growthand paramylon-production-promoting abilities of the isolated bacteria were assessed by comparing the biomass and paramylon concentrations at the end of 7-d culturing with bacterial inoculation relative to control cultures.

Identification and Characterization of the CA3 and CN5 Strains
Among the isolated bacterial strains, strains CA3 (acidophilic bacterium) and CN5 (neutrophilic bacterium) showed the highest growth-and paramylon-production-promoting abilities at pH 4.5 and 7.5, respectively. Strains CA3 and CN5 were characterized and identified using physiological and phylogenetic analyses. Physiological characterization was performed using an API 20NE Kit (BioMérieux Japan, Tokyo, Japan) according to the manufacturer's instructions. Comparative 16S rRNA gene-sequence analysis was performed, as follows: almost full-length 16S rRNA genes were amplified by PCR using the primers 8F (5 -AGAGTTTGATCCTGGCTCAG-3 ) and 1510R (5 -GGTTACCTTGTTACGACTT-3 ). Genuslevel identifications were carried out based on 16S rRNA gene-sequence similarities with those of type-strain sequences in NCBI GenBank using BLAST. The 16S rRNA sequence data [1431 base pairs (bp)] of CA3 and CN5 were submitted to the DDBJ/EMBL/GenBank databases under accession numbers LC604062 and LC604063, respectively.
Bacterial IAA production was evaluated as described previously [32], with some modifications. The bacterial colony was inoculated in 100 mL of C-NH 4 medium with 5 g L −1 glucose (pH 4.5 or 7.5) with or without 0.05% (w/v) L-tryptophan and incubated at 28 • C and 150 rpm for 1 d. The culture was collected and centrifuged (10,000× g, 4 • C, and 10 min), and 500 µL supernatant was mixed with 750 µL Salkowski reagent (98 mL 35% HClO 4 plus 2 mL of 0.5 M FeCl 3 ) and incubated at 24 • C for 30 min. The development of a pink color indicated IAA production, and the absorbance at 535 nm (A 535 ) was measured. The IAA concentration was calculated using pure IAA as a standard (Kanto Chemical Co., Inc., Tokyo, Japan).
Bacterial EPS production was evaluated as described previously [29], with specific modifications. Bacterial colonies were inoculated in 200 mL C-NH 4 medium with 5 g L −1 glucose (pH 4.5 or 7.5) and incubated at 28 • C and 150 rpm for 1 d. Each culture was collected and centrifuged (10,000× g, 4 • C, and 20 min), and the supernatant was gently mixed with three volumes of ice-cold 100% ethanol and incubated overnight at 4 • C. The precipitated EPS was collected by centrifugation (10,000× g, 4 • C, and 20 min) and dried.
This EPS was defined as free EPS. Bacterial cells were collected from the above 1-d culture by centrifugation (10,000× g, 24 • C, and 20 min) and mixed with 0.9% NaCl solution, with this mixture homogenized and shaken using a vortex for 3 min and centrifuged (10,000× g, 4 • C, and 20 min). The supernatant was gently mixed with three volumes of ice-cold 100% ethanol and incubated at 4 • C overnight. The precipitated EPS was collected by centrifugation (10,000× g, 4 • C, and 20 min), dried, and defined as cell-bound EPS. Free and cell-bound EPS were dissolved in hot distilled water, and the total sugar content of the EPS was determined in each sample by the phenol-sulfuric acid method [33]. The total protein content of the EPS was estimated in each sample using a BCA protein assay kit (Takara Bio, Shiga, Japan). Total sugar and protein contents in the EPS were calculated as mg L −1 of culture medium.

Growth of CA3 and CN5 Utilizing Glucose
Strain CA3 was pre-cultured at 28 • C in liquid R2A medium (pH 4.5) with shaking (150 rpm) until it reached the late logarithmic growth phase. Similarly, strain CN5 was pre-cultured at 28 • C in R2A medium (pH 7.5) with shaking (150 rpm) un2w1til it reached the late logarithmic growth phase. The cells of each strain were harvested by centrifugation (10,000× g, 24 • C, and 5 min) and washed twice with C-NH 4 medium (pH 4.5 or 7.5). The cells were inoculated into 200 mL C-NH 4 medium with 5 g L −1 glucose (pH 4.5 or 7.5) in a 500-mL flask until reaching an OD 600 of 0.05, after which they were incubated for 5 d at 28 • C and 150 rpm in the dark. The bacterial cell densities (OD 600 ) and glucose concentrations were monitored during the incubation period. The growth experiments were conducted in triplicate.

Co-Culturing E. gracilis with CA3 or CN5 under Acidic and Neutral pH with or without Glucose
Ten milliliters of E. gracilis pre-cultured in C-NH 4 medium with 5 g L −1 glucose (pH 4.5 or 7.5) was inoculated into 200 mL C-NH 4 (pH 4.5 or 7.5) with or without 5 g L −1 glucose. The initial biomass of E. gracilis was approximately 40 mg dry weight L −1 . CA3 or CN5 cells were pre-cultured in liquid R2A medium at 28 • C with shaking (150 rpm) at pH 4.5 or 7.5, respectively, until they reached late logarithmic growth phase. CA3 or CN5 cells were harvested by centrifugation (10,000× g, 24 • C, and 5 min) and washed twice with C-NH 4 medium (pH 4.5 or 7.5). CN3 or CN5 cells were inoculated at an OD 600 of 0.05 into E. gracilis C-NH 4 medium at pH 4.5 or 7.5 with or without 5 g L −1 glucose. Control cultures including E. gracilis cells alone (without bacterial inoculation), CA3 or CN5 cells alone (without E. gracilis), and C-NH 4 medium with 5 g L −1 glucose alone (without E. gracilis and bacterial cultures) were also prepared. All culture flasks were incubated in a growth chamber with shaking at 150 rpm for 7 d. During the experiments, chlorophyll a + b, E. gracilis biomass, paramylon, and glucose concentrations were monitored on the initial day and days 3 and 7. The experiments were conducted in triplicate. The growth-and paramylon-production-promoting effects of CA3 and CN5 were evaluated for comparison with E. gracilis axenic control culture in 7-d cultures.

Scanning Electron Microscopy (SEM) of E. gracilis Cell Surfaces
For SEM experiments, E. gracilis cells were collected from the co-culture experiment at 7 d and centrifuged at 3000× g for 5 min at 24 • C. Additionally, the collected E. gracilis cells were washed using the same biomass (dry weight)-measurement method: vortex mixing for 60 s, centrifugation (3000× g, 24 • C, and 5 min), and washing with distilled water. The E. gracilis cells (with or without the above washing process) were then fixed with 4% osmium tetroxide solution at 4 • C for 3 h, dehydrated at room temperature (~24 • C) in solutions containing progressivelyincreasing ethanol concentrations (30-100%; 15 min/incubation), and finally dried at the carbon dioxide critical point. The dried samples were coated using an osmium plasma coater (OPC80T; Filgen, Nagoya, Japan) and then examined by SEM using a JEOL SEM instrument (JSM 6320F; JEOL Ltd., Tokyo, Japan).

Analytical Procedures
The chlorophyll concentrations in the E. gracilis cultures were measured spectrophotometrically after extraction with 100% methanol for 30 min [34]. The A 665 and A 650 of each extract were measured using a UVmini-1240 spectrophotometer (Shimadzu Co. Ltd., Kyoto, Japan). Total chlorophyll (Chl; Chl a + b) concentration (µg mL −1 ) was calculated, as follows: The biomass (dry weight) of E. gracilis samples was measured, as follows: 25 mL of each culture was collected into a 50-mL centrifuge tube and vortexed for 60 s to uniformly suspend bacterial and microalgal cells. Each mixture was centrifuged (3000× g, 24 • C, and 5 min), the pellet was washed with 25 mL distilled water, and the centrifugation and wash steps were repeated one time. Subsequently, each E. gracilis pellet was suspended in 25 mL distilled water, collected on a pre-weighed GF/A glass fiber filter, dried at 90 • C for 3 h, and then weighed. We confirmed by SEM observations that the collected E. gracilis cells contained no CA3 or CN5 cells.
Paramylon was extracted from E. gracilis cells using the sodium dodecyl sulfateethylenediaminetetraacetic acid (SDS-EDTA) method. Each E. gracilis culture (25 mL) was collected into a 50-mL centrifuge tube, ultrasonicated for 1 min, vortexed for 30 s, and then centrifuged (3000× g, 24 • C, and 5 min) to remove the supernatant. The collected E. gracilis pellets were washed thrice by centrifugation (3000× g, 24 • C, and 5 min), resuspended in 25 mL distilled water, and incubated with 10 mL of ethanol for 30 min at room temperature (24 • C) before another round of centrifugation (8000× g, 24 • C, and 5 min). The collected E. gracilis cells were mixed with 10 mL SDS-EDTA reagent (1% SDS: 5% Na 2 ·EDTA), incubated in a water bath at 90 • C for 30 min, and centrifuged (8000× g, 24 • C, and 5 min), after which the supernatant was removed. Each collected E. gracilis pellet was mixed with 1 mL SDS-EDTA reagent and 9 mL distilled water, vortexed, and centrifuged (8000× g, 24 • C, and 5 min), after which each supernatant was removed from the tube. The collected E. gracilis pellets were then mixed with 20 mL distilled water, vortexed, and centrifuged (8000× g, 24 • C, and 5 min), after which each supernatant was removed from the tube. Each collected E. gracilis pellet was then mixed with 5 mL 1 mol L −1 NaOH, vortexed, and incubated at room temperature (24 • C) for 12 h. A 0.5-mL aliquot from each tube was transferred to a test tube with 0.5 mL of 5% phenol solution and 2.5 mL sulfuric acid. The test tubes were gently mixed and incubated at room temperature (24 • C) for 30 min, after which the A 480 of each solution was measured using a spectrophotometer. A standard calibration curve was prepared using commercially available 100% paramylon (β-1,3-glucan from E. gracilis; Sigma-Aldrich, St. Louis, MO, USA).
The glucose concentration in each culture was measured using a Shimadzu highperformance liquid chromatography system (Shimadzu Co. Ltd.) with a refractive index detector and a Shodex SUGAR SH1011 column (300 mm × 8.0 mm; Showa Denko K. K., Tokyo, Japan). The mobile phase was 1 mmol L −1 sulfuric acid solution, and the column was maintained at 50 • C.

Statistical Analysis
Each value presented represents the results of three replicates (n = 3) per experiment. All results are expressed as the mean ± standard deviation (SD). Statistical significance (p < 0.05) was analyzed using the paired-samples t-test with SPSS Statistics (v.22.0; IBM Corp., Armonk, NY, USA).

Isolation and Identification of Bacteria
Promoting the Growth and Paramylon Production of E. gracilis E. gracilis cells were cultured in C-NH 4 with 0.5 g L −1 CSL under mixotrophic conditions at pH 4.5 or 7.5 with or without sewage-effluent bacteria. E. gracilis clearly showed faster growth at both pH values when co-cultured with sewage-effluent bacteria than with-out ( Figure S1). Therefore, this suggested that MGPB for E. gracilis must have been present among the sewage-effluent bacteria and supported E. gracilis growth under mixotrophic conditions at pH 4.5 or 7.5.
On R2A agar, the CA3 colonies appeared convex, circular, smooth, lustrous, and white in color and were Gram-negative and rod-shaped (0.8-1.0 × 1.2-1.8 µm) ( Figure S2). CA3 tested positive for nitrate reduction, glucose fermentation, aesculin hydrolysis, and βgalactosidase activity but negative for oxidase, arginine dihydrolase, urease, and gelatinase activities. CA3 utilized D-glucose, L-arabinose, D-mannose, D-mannitol, glucosamine, D-maltose, gluconate, malate, citrate, and phenylacetate but did not utilize caprate or adipate. Almost the entire 16S rRNA gene sequence (1467 bp By contrast, colonies of strain CN5 on R2A agar were convex, circular, smooth, lustrous, and orange in color. Similar to the CA3 strain, the CN5 strain was also Gram-negative and rod-shaped (0.7-0.9 × 1.2-1.6 µm) ( Figure S2). CA5 cells were positive for oxidase, aesculin hydrolysis, and β-galactosidase activities but negative for nitrate reduction, glucose fermentation, and arginine dihydrolase, urease, and gelatinase activities. Strain CN5 utilized D-glucose, D-mannose, glucosamine, and D-maltose but did not utilize L-arabinose, Previous reports show that IAA-and EPS-producing bacteria promote E. gracilis growth and paramylon production [28,29]. Therefore, we tested the IAA-and EPSproducing activities of CA3 and CN5. CA3 and CN5 did not produce IAA in C-NH 4 + glucose medium in the absence of L-tryptophan, which is the main precursor for the IAAbiosynthesis pathways in bacteria (Table 1). Additionally, CA3 produced EPS in C-NH 4 +glucose medium at both pH 4.5 and 7.5, whereas CN5 produced EPS in C-NH 4 + glucose medium only at pH 7.5 (Table 1).  1 All values are presented as the mean ± SD (n = 3); 2 NT indicates a condition that was not tested because CN5 could not be grown at pH 4.5.

Growth of CA3 and CN5 Utilizing Glucose as a Sole Carbon Source
CA3 or CN5 were cultured in C-NH 4 medium with 5 g L −1 glucose at pH 4.5 or 7.5. CA3 rapidly utilized 5 g L −1 glucose at both pH 4.5 and 7.5, and the bacterial cell density (OD 600 ) increased in parallel with glucose uptake, reaching a stationary growth phase within 12 h ( Figure 1A). CN5 rapidly utilized 5 g L −1 glucose at pH 7.5, and bacterial growth paralleled glucose uptake, reaching stationary phase within 12 h ( Figure 1B). However, CN5 did not utilize glucose or grow at pH 4.5.

Co-Culturing E. gracilis with Strain CA3 or CN5 under Photoautotrophic or Mixotrophic
Conditions at pH 4.5 or 7.5 E. gracilis cells were co-cultured with CA3 or CN5 cells in C-NH 4 medium without glucose (photoautotrophic condition) or with 5 g L −1 glucose (mixotrophic condition) at pH 4.5 or 7.5 for 7 d. At pH 4.5 under photoautotrophic conditions without glucose, the chlorophyll, biomass, and paramylon concentrations of E. gracilis co-cultured with CA3 or CN5 and axenic control E. gracilis cultures increased at comparable rates, although slight differences were observed (Figure 2A-C). E. gracilis biomass and paramylon concentrations were slightly but significantly higher (p < 0.05) after 3 d in E. gracilis co-cultured with CA3 than when co-cultured with CN5 or in axenic control cultures ( Figure 2B,C). At pH 4.5 under mixotrophic conditions with glucose, the chlorophyll, biomass, and paramylon concentrations of E. gracilis increased rapidly and were significantly higher (p < 0.05) in E. gracilis co-cultured with strain CA3 than when co-cultured with CN5 or in axenic control cultures ( Figure 2D-F). Moreover, glucose concentration decreased more rapidly in E. gracilis co-cultured with CA3 than with CN5 or in axenic control E. gracilis cultures ( Figure 2G). After 7 d, the final biomass and paramylon concentrations of E. gracilis co-cultured with CA3 were 1.8-and 3.5-fold higher, respectively, as compared with those of control E. gracilis cultured under mixotrophic conditions with glucose ( Table 2). 1 All values are presented as the mean ± SD (n = 3); 2 The number in parentheses indicates the ratio of each value as compared with each control experimental value (co-culture with CA3/control or co-culture with CN5/control). At pH 7.5 under photoautotrophic conditions without glucose, the chlorophyll, biomass, and paramylon concentrations of E. gracilis co-cultured with CA3 or CN5 and axenic control E. gracilis cultures increased at comparable rates, although slight differences were noted ( Figure 3A-C). At pH 7.5 under mixotrophic conditions with glucose, the chlorophyll concentrations in E. gracilis co-cultured with CA3 or CN5 and control cultures increased at comparable rates ( Figure 3D).
E. gracilis biomass and paramylon concentrations increased rapidly and were significantly higher (p < 0.05) in E. gracilis co-cultured with strain CA3 or CN5 than those in axenic control cultures ( Figure 3E,F). Additionally, the glucose concentration in E. gracilis co-cultured with CA3 or CN5 decreased more rapidly than in the control E. gracilis culture ( Figure 3G). After 7 d, the final biomass and paramylon concentrations of E. gracilis cocultured with CA3 were 1.9-and 3.5-fold higher, respectively, as compared with the control E. gracilis culture. Moreover, after 7 d, the final biomass and paramylon concentrations of E. gracilis co-cultured with CN5 were 2.0-and 4.1-fold higher, respectively, as compared with the control E. gracilis culture under mixotrophic conditions (Table 2). In control cultures containing CA3 or CN5 alone (pH 4.5 or 7.5), E. gracilis biomass (collected on GF/A filters) and paramylon (extracted using the SDS-EDTA method) were not detected. CA3; open blue circles: E. gracilis co-cultured with Emticicia sp. CN5; and the closed black circles: E. gracilis axenic control cultures. Data shown represent means ± SD (n = 3). Asterisks indicate significant difference (p < 0.05) between the E. gracilis co-culture with CA3 or CN5 and the E. gracilis axenic control culture. SEM observations revealed that CA3 and CN5 attached to E. gracilis cells in co-cultures under mixotrophic conditions at pH 4.5 or 7.5, respectively. Interestingly, CA3 and CN5 attached to the flagella of E. gracilis rather than the main cell surface ( Figure 4A,B,D,E), although the reasons for this remain unclear. Moreover, the EPS matrix was observed at interfaces between the E. gracilis surface and CA3 or CN5 ( Figure 4B,E). However, after the washing process, almost all CA3 and CN5 cells had detached from the E. gracilis surfaces ( Figure 4C,F).

Discussion
In this study, Enterobacter sp. CA3 and Emticicia sp. CN5 were isolated from the surfaces of E. gracilis cells grown with sewage-effluent bacteria under mixotrophic conditions at pH 4.5 or 7.5, respectively. Both bacterial strains were capable of significantly promoting the growth and paramylon production of E. gracilis at their respective pH values during a 7-d cultivation (Figures 2 and 3). Additionally, we observed that CA3 and CN5 attached to the flagella of E. gracilis cells and presumably formed a symbiotic association with E. gracilis ( Figure 4A,B,D,E). Enterobacter sp. CA3 and Emticicia sp. CN5 represent the first isolated E. gracilis-associated bacteria capable of promoting E. gracilis growth and paramylon production under mixotrophic conditions.
Various Enterobacter spp. have been isolated from soil and water, a variety of plant species, natural animal commensals, and the human gut microbiota [35]. In the present study, Enterobacter sp. CA3 was successfully isolated from E. gracilis surfaces after growth with sewage-effluent bacteria at pH 4.5 as an MGPB. This strain is the first MGPB identified that belongs to the Enterobacter genus. Some Enterobacter spp. are acidophilic [36,37] and can grow over a wide range of pH conditions (pH 2-9) [38]. Although CA3 was isolated from E. gracilis grown at pH 4.5, it exhibited viability across a wide range of pH values and showed growth-and paramylon-promoting activities at both pH 4.5 and 7.5 (Figures 1-3). Several Emticicia spp. bacteria have also been isolated from various aquatic and soil environments [39][40][41]. Emticicia sp. EG3 was recently isolated from E. gracilis grown in sewage effluent and shown to promote its growth under autotrophic and neutral pH conditions [27]. Emticicia sp. CN5, isolated in the present study, represents the second reported E. gracilis-associated MGPB, and the first Emticicia sp. capable of promoting paramylon production.
Enterobacter sp. CA3 utilized glucose for its growth ( Figure 1) and significantly enhanced both the biomass and paramylon production of E. gracilis under mixotrophic conditions with glucose at both pH 4.5 and 7.5 but showed little or no enhancement under autotrophic conditions without glucose (Figures 2 and 3). By contrast, Emticicia sp. CN5 utilized glucose for its growth at pH 7.5 and enhanced both E. gracilis biomass and paramylon production under mixotrophic conditions with glucose at pH 7.5 but did not show such enhancement under autotrophic conditions without glucose (Figures 2 and 3). Interestingly, the growth-and paramylon-production-promoting effects on E. gracilis of the two strains were dependent on the mixotrophic condition with glucose and a viable pH condition.
V. natriegens produces IAA in E. gracilis medium with L-tryptophan and promotes the growth and paramylon production of E. gracilis [28]. We found that CA3 and CN5 produced IAA in C-NH 4 + glucose medium with L-tryptophan but not when cultured in C-NH 4 +glucose medium without L-tryptophan (Table 1). Moreover, CA3 and CN5 promoted the growth and paramylon production of E. gracilis in C-NH 4 + glucose without L-tryptophan (Figures 2 and 3). Thus, the growth-promoting factors of CA3 and CN5 must differ from that of IAA. By contrast, the EPS-producing Pseudoalteromonas sp. MEBiC 03485 promotes the growth and paramylon production of E. gracilis, and supplementation with its free EPS at 133 mg L −1 or 333 mg L −1 significantly promoted the biomass and paramylon production of E. gracilis [29]. In the present study, CA3 produced free EPS in C-NH 4 + glucose medium at concentrations of 26.7 ± 0.03 mg sugar L −1 and 16.5 ± 0.85 mg protein L -1 at pH 4.5 and 35.1 ± 0.08 mg sugar L -1 and 17.1 ± 0.11 mg protein L -1 at pH 7.5. Similarly, CN5 also produced free EPS in C-NH 4 + glucose medium at concentrations of 31.7 ± 0.02 mg sugar L −1 and 17.1 ± 0.06 mg protein L −1 at pH 7.5 (Table 1). Furthermore, CA3 and CN5 produced EPS and attached to the surface of E. gracilis cells via the EPS matrix ( Figure 4). Thus, EPS produced by CA3 and CN5 can potentially promote the growth and paramylon production of E. gracilis. However, the detected concentration of EPS produced by CA3 and CN5 was much lower than the previously reported effective EPS concentrations (133 and 333 mg L −1 ), suggesting the possibility that CA3 and CN5 might produce major factors other than EPS that are responsible for promoting growth and paramylon production.
These results clearly indicate that the growth-and paramylon-production-promoting factors of CA3 and CN5 are related to glucose metabolism or cell growth (Figures 2 and 3). For example, the promoting factors might be produced through glucose metabolism or glucose-dependent growth to a high cell density and thereby promote E. gracilis cell growth and paramylon production. By contrast, CA3 and CN5 did not promote chlorophyll synthesis (Figures 2 and 3). Chlorophylls are essential and limiting factors for photosynthesis [42] and also important in mixotrophic cultivation of E. gracilis [3]. Thus, the promoting factors of CA3 and CN5 might act more strongly in a heterotrophic mode than in an autotrophic mode under mixotrophic cultivation, which consists of autotrophic and heterotrophic growth modes. However, we did not test the effect of the bacteria under E. gracilis heterotrophic culture conditions.
In this study, we did not identify the factors or mechanisms by which CA3 and CN5 promote E. gracilis growth and paramylon production under mixotrophic conditions with glucose. MGPB can promote microalgal growth in various ways. Additionally, Zhu and Wakisaka [43] showed that the addition of ferulic acid made from rice promotes the growth and paramylon production of E. gracilis. Exogenous phytohormones, cytokinins, and abscisic acid can also promote the growth of E. gracilis [44]. Thus, it is apparent that various factors have the potential to promote the growth and paramylon production of E. gracilis, and further studies are needed to clarify the mechanisms whereby CA3 and CN5 promote these activities in E. gracilis.
After co-culturing E. gracilis cells with CA3 for 7 d under mixotrophic conditions with glucose, the biomass and paramylon productivity of E. gracilis increased by 1.8and 3.5-fold, respectively, at pH 4.5 and by 1.9-and 3.5-fold, respectively, at pH 7.5 and relative to sterile E. gracilis cultures (Table 2). Additionally, the biomass and paramylon productivity of E. gracilis increased by 2.0-and 4.1-fold, respectively, at pH 7.5 when cocultured with CN5 as compared with sterile E. gracilis culture ( Table 2). Although the cultivation conditions differed, the promoting effects of CA3 and CN5 were higher than those (a 17-23% increase in biomass and a 25-35% increase in paramylon production) reported previously for V. natriegens [28], EPS-producing Pseudoalteromonas sp. MEBiC 03485 [29], and Pseudoalteromonas sp. MEBiC 03607 [30]. Therefore, CA3 and CN5 represent promising and useful MGPBs for increasing both biomass and paramylon yields of E. gracilis under mixotrophic conditions. Furthermore, the ability to cultivate E. gracilis under acidic conditions with CA3 provides an important advantage in terms of preventing microalgal and bacterial contamination, enabling large-scale cultivation of E. gracilis. Moreover, the attachment of CA3 and CN5 to the surface of E. gracilis flagella suggests that they might form a durable symbiotic association with E. gracilis. However, for the biotechnological application of CA3 and CN5, the method of co-culturing E. gracilis with these strains needs to be optimized in further studies. These findings demonstrated that our screening and isolation methods enabled the construction of an effective and practical E. gracilis-MGPBassociation system for increasing the paramylon yield of E. gracilis.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/10 .3390/microorganisms9071496/s1. Figure S1: Images of Euglena gracilis cultures in C-NH4 medium, Figure S2: Images of Enterobacter sp. CA3 and Emticicia sp. CN5 colonies on R2A agar and their Gram staining, Table S1: List of bacterial strains isolated in this study.  Data Availability Statement: The 16S rRNA sequence data (1431 bp) of CA3 and CN5 were submitted to the DDBJ/EMBL/GenBank databases under accession numbers LC604062 and LC604063, respectively.

Conflicts of Interest:
There are no conflicts of interest to declare.