Microalgae are seen as nature’s treasure chest as they have high potential application in diverse fields ranging from industrial areas such as biofuels, chemicals and materials to functional foods, environmental remediation, and nutraceuticals [1, 2]. Although different species of land plants are considered to be applicable in these fields as well, microalgae greatly outpace land plants in terms of growth rates, land usage, and the contents of the high-value products [3, 4]. Hence, research efforts to develop algal-based industries have been ongoing worldwide, from strain development to the optimization of cultivation, harvesting and extraction methods.

While there is a great diversity of algal species in terms of their phenotypes and evolutionary origins, the green algae Chlorella spp. are considered to be one of the best candidates for industrial uses due to its rapid growth rate and high lipid content. Numerous strains of Chlorella have been present on earth for 2.5 billion years and are found worldwide in all types of habitats, from deserts to arctic regions [5-7]. As such, different Chlorella species possess a highly diverse range of phenotypes, which can also vary widely among different strains within the same species. Recently, the U.S Department of Energy (DOE) has selected Chlorella sorokiniana DOE1412 and Chlorella vulgaris CCAP 211/21A to be the best-performing strains for biomass and lipid accumulation from libraries of more than 2,000 and 600 algal strains [8]. There have also been suggestions that Chlorella sp. with various phenotypes can have possible uses in biogas upgrade processes, wastewater treatment, and nutraceutical production [9-11]. Furthermore, Chlorella sp. have an advantage of having a degree of familiarity among the general public due to their general uses as nutritional supplements, which can help facilitate its use in other industrial areas as well.

Over the past years, there have been a number of improvements in genetic engineering techniques involving microalgae. Many researchers have utilized these techniques to augment the growth and lipid productivities in a wide variety of high-performing species such as Chlorella and Nannochloropsis [12-15]. However, the use of genetically modified organisms (GMOs) remains highly restricted in most countries, and cultivation is only allowed in perfectly closed systems such as photobioreactors (PBRs). Considering the high operating costs and scalability issues associated with PBRs compared to open raceway ponds (ORPs), ORPs are a more economically attractive platform for mass cultivation [16, 17]. It has been reported that for every 1% increase in biomass production, net cash income can be increased by 0.21% and 0.10% for ORPs and PBRs, respectively [18]. In addition, as there are substantial negative social perceptions about GMOs, industrial uses of the biomass may face further barriers particularly if some of its components are to be used for food and nutraceutical production. Hence, screening and identification of outstanding wild-type strains are just as important as the development of genetically modified strains.

In this study, two novel cold-tolerant microalgae species were isolated during the winter season from a South Korean freshwater lake. These were designated as C. vulgaris ABC-002 and ABC-008 after phylogenetic analysis. The comparison with C. vulgaris UTEX265, a widely used reference strain in the algal research field [19, 20], revealed that the newly isolated strains showed outstanding growth rates in both cold- and room-temperature conditions. Further analysis of lipid content and productivities suggested the potential of using C. vulgaris strains ABC-002 and ABC-008 for biofuel production.

Isolation of Chlorella vulgaris ABC-002 and ABC-008

Algal samples were collected from a lake in Gyeonggi-do, South Korea (37°06’48.8’’N 126°58’17.0’’E), during the winter season. For the isolation of single algal cells, the samples were serially diluted and plated onto TAP (tris-acetate-phosphate) agar plates containing 100 µg/ml of ampicillin. The plates were stored at 10°C with continuous light (120 µE) illumination throughout the process to allow only for the cold-resistant strains to grow. From more than 200 colonies that appeared on the selection plate, 60 colonies (based on the size) were incubated in 25-ml cell culture flasks (SPL, Korea) with 10 ml of TAP media at 10°C to screen for strains with rapid growth rates and oleaginous traits. The screening was performed based on the culture’s optical densities and esterifiable lipid contents with these factors measured using a UV-spectrophotometer (Shimadzu, Japan) and a gas chromatograph (GC) (HP 6890, Agilent, USA), respectively. The strains with the highest growth rate and greatest lipid content were identified after a phylogenetic analysis and were designated as Chlorella vulgaris ABC-002 and ABC-008, respectively. The obtained strains were deposited into the Korean Collection for Type Cultures Center with deposit numbers KCTC18580P (ABC-002) and KCTC18581P (ABC-008).

DNA Sequencing and Phylogenetic Analyses

The cells were cultivated in 25-ml cell culture flasks (SPL) under TAP media for five days before harvesting. The harvested cells were concentrated up to 1 × 10⁸ cells/ml density and then underwent DNA extraction with InstaGene Matrix (Bio-Rad, USA) according to the manufacturer’s instructions. The extracted DNA was then subjected to PCR amplification with 18S rDNA targeting primer sets (18SF: 5’-CCTGGTTGATCCTGCCAG-3’, 18SR: 5’-TTGATCCTTCTGCAGGTTCA-3’) [21]. The amplified products were run on 1.5% agarose gel for confirmation, followed by gel extraction with the QIAquick Gel Extraction Kit (Qiagen, Germany). The products were then sequenced by Solgent Co., Ltd. (Korea) with the same primer sets used for the PCR amplification step. After comparison of the 18S rDNA with the previously identified sequences from GenBank, they were submitted to GenBank with accession numbers MF686452 and MF686487.

For the phylogenetic analyses, 18S rDNA sequences of the isolated strains were aligned with those from various algal species obtained from the NCBI database. Based upon the alignment outcome, a phylogenetic tree was constructed using the maximum likelihood (ML) method with the starting tree created by unweighted pair group method with the arithmetic mean (UPGMA) approach using the CLC workbench program (version 7.7.2). For each method, 1,000 bootstrap replicates were performed.

Batch Cultivation at Different Temperatures

The cultivation conditions of Chlorella vulgaris ABC-002 and ABC-008 were in accordance with the reference strain of Chlorella vulgaris UTEX 265 from the UTEX Culture Collection of Algae. Regarding the broth condition, cells were cultivated in TAP media [2.42 g/l Tris, 0.375 g/l NH₄Cl, 0.1 g/l MgSO₄ 7H₂O, 0.05 g/l CaCl₂ 2H₂O, 0.0108 g/l K₂HPO₄, 0.0054 g/l KH₂PO₄, 1 ml/l glacial acetic acid, and 1 ml/l Hutner’s trace elements (50 g/l Na₂EDTA 2H₂O, 22 g/l ZnSO₄ 7H₂O, 11.4 g/l H₃BO₃, 5.06 g/l MnCl2 4H₂O, 1.61 g/l C°l₂ 6H₂O, 1.57 g/l CuSO₄ 5H₂O, 1.10 g/l (NH₄)₆Mo₇O₂₄ 7H₂O, and 4.99 g/l FeSO₄ 7H₂O)] in 500-ml Erlenmeyer flasks with working volumes of 200 ml. The light and agitation conditions were held constant at 120 µE and 200 rpm, respectively, while the Fo = Minimum fluorescence yield measured at the lowest light frequency Fm = Maximum fluorescence yield measured at the saturation o o point of the sample temperature was set to either 10°C or 25°C in an artificial incubator. The cultivation was conducted in duplicates for each experiment.

Growth Analysis

Cell growth was determined according to the cell numbers and dry cell weights (DCW) throughout the cultivation. The cell numbers were counted on a daily basis using an automated cell counter (Cellometer Auto X4, USA), and the DCW was only measured at the end of the cultivation. To measure the DCW, the cells were filtered through previously weighed GF/C filter papers (USA), followed by washing with distilled water and drying overnight at 60°C.

Fatty Acid Methyl Ester (FAME) Analysis

The esterifiable lipid content of each strain was measured through the conversion of the total cellular lipid into FAME using a modified version of the Folsch process, followed by GC analysis. To prepare the samples, cells were freeze-dried for five days and ground into a fine powder. Approximately 10 mg of the prepared samples were weighed and mixed with 2 ml of a chloroform-methanol mixture (2:1, v/v) by vortexing for 20 min. After adding 1 ml of an internal standard containing (100 mg heptadecanoic acid/200 ml chloroform), transesterification was carried out through a reaction with 1 ml of methanol and 300 µl of sulfuric acid at 100°C for 20 min. After cooling the reaction to room temperature, 1 ml of distilled water was added and vortexed to wash out the residual biomass and methanol, and the chloroform phase was separated by centrifugation at 4,000 ×g for 5 min. The organic phase was recovered after filtration with a 0.20-µm RC-membrane syringe filter (Sartorius Stedim Biotech, Germany). The FAME analysis was carried out using a GC system (HP 6890, Agilent) with a flame ionization detector and an HP-INNOWax polyethylene glycol column (HP 19091 N-213, Agilent).

Maximum Quantum Yield Measurement

To evaluate the physiological properties of the Chlorella species at different temperatures, the maximum quantum yield (F_v/F_m) of the photosystem II (PS II) was measured during the exponential phase on day 3 after the inoculation. After dark adaptation at 25°C in a thermomixer (Eppendorf, Germany) for 20 min, cells were placed in a multi-color-PAM (pulse amplitude modulation) device (Heinz-Walz Germany). The Fv/Fm ratio was determined by an SP analysis, where the photochemically active radiation (PAR) was increased from 0 to 2,539 µmol photons•m^-2•s^-1 to determine the light curve of each strain. The Fv/Fm ratio was calculated according to the following equation:.

Microscopy Imaging and Size Distribution

The morphology of C. vulgaris strains was investigated using a Cellometer Auto X4 device on days 4 and 14 after cultivation at 10°C and 25°C. The software was set to count cells in the size range of 1 ~ 50 µm, and the size distribution was also measured with the same device.

Morphology

Two Chlorella species were isolated from a lake in South Korea during the winter season when the water temperature was close to 7°C. After cultivating the isolated strains (as described in Materials and Methods) for two months on ampicillin-containing TAP agar plates at 10°C, we verified the cold-resistant phenotypes of the cells as well as their axenic condition. These two novel Chlorella species were designated as C. vulgaris ABC-002 and ABC-008. At 25°C, the cells were 3 ~ 8 um in diameter with an ovoid shape (Fig. 1). The large chloroplast occupying almost half of the cell volumes, the visible presence of vacuoles, and lack of flagella were all consistent with the typical morphology of the genus Chlorella [5].

Figure 1. The morphologies of novel Chlorella vulgaris strains ABC–002 and ABC–008 on days 4 and 14. On day 4, the cells grown at 10°C exhibited larger cell sizes compared to the cells grown at 25°C. On day 14, no differences in the cell size were observed among the cells cultivated at different temperatures.

Noticeable changes in the cell size and morphology were observed when the cells were cultivated under low temperature. Cultivation at 10°C resulted in up to 100% increase in the cell sizes in both ABC-002 and ABC-008 during the growth phase, and the cells changed shapes into more spherical forms. However, these changes were no longer apparent when the cells entered the stationary phase, as they reverted to their respective normal features. Given these similar morphological characteristics of the two novel species, it was assumed that they were identical or very closely-related species prior to conducting the phylogenetic analyses.

Phylogenetic Analyses

The 18S rDNA of two Chlorella sp., ABC-002 and ABC-008, were sequenced and submitted to NCBI. Sequences were aligned to the 18S rDNA of other algal species from Chlorella, Ettlia, Chlamydomonas, Scenedesmus, Tetradesmus and Nannochloropsis (Fig. 2A). As the morphologies of the two novel strains were closest to the genus Chlorella, alignments were mainly done against various species of Chlorella. The phylogenetic tree revealed a clear divergence of Chlorella sp. into C. vulgaris and C. sorokiniana, and both Chlorella sp. ABC-002 and ABC-008 were found to belong to the sub-branch of C. vulgaris. A comparison of the 18s rDNA nucleotides among C. vulgaris ABC-002, ABC-008 and other representative species showed C. vulgaris SAG 211-11b-1 to be the closest species to Chlorella sp. ABC-002 and ABC-008, with only three and one nucleotide difference(s) from these species, respectively (Fig. 2B). Through these analyses, it was concluded that both of the isolated strains were of sub species of C. vulgaris, and as such, they were named Chlorella vulgaris ABC-002 and Chlorella vulgaris ABC-008, respectively.

Figure 2. (A) Phylogenetic tree of the C. vulgaris strains ABC-002 and ABC-008 and other algal species based on their 18S rDNA gene sequences. The bootstrap values are from 1,000 replicates of the sequence data. Nannochloropsis oculata (a seawater species) was used as an outgroup. (B) 18S rDNA nucleotide differences between C.vulgaris ABC-002, ABC-008 and other algal species.

Growth under Different Temperature Conditions

The growth of C. vulgaris ABC-002 and ABC-008 was characterized under both room (25°C) and low (10°C) temperatures. These two strains were benchmarked against C. vulgaris UTEX265, which is one of the most widely used strains of Chlorella vulgaris from the UTEX culture collection [19, 22-24]. The overall growth rates of the two novel strains were similar, and they both had noticeably superior growth performance in comparison to UTEX265 strain. Under cultivation at 10°C, ABC-002 and ABC-008 proliferated up to cell concentrations of 3.05 × 10⁸ cells/ml and 3.08 × 10⁸ cells/ml, respectively, owing much to their cold-resistant properties. In contrast, C. vulgaris UTEX 265 was unable to facilitate much growth (4.3 × 10⁷ cells/ml) under cold condition (Fig. 3A). When the cells were cultivated under 25°C, all three species showed better performance compared to cultivation at 10°C. Especially the UTEX265 strain, which reached a cell concentration of 2.2 × 10⁸ cells/ml, which is a five-fold increase from that at a low temperature (Fig. 3B). The two novel strains ABC-002 and ABC-008 also showed improved growth, with final cell concentrations of 3.9 × 10⁸ cells/ml and 4.2 × 10⁸ cells/ml on day 12, respectively. These changes in the growth rate were found to correspond very closely with the maximum quantum yields (Fv/Fm) as measured by multi-color PAM (Fig. 3C). At 10°C, the F_v/F_m ratios of UTEX265, ABC-002 and ABC-008 were 0.5594, 0.6747, and 0.7150 respectively, which increased up to 0.6478, 0.7134, and 0.7227 at 25°C.

Figure 3. The cultivation of Chlorella species in TAP media at (A) 10°C and (B) 25°C . (C) The maximum quantum yields of the cells were calculated under both cultivation conditions using the multi-PAM method after dark adaptation for 20 min. (D) The final dry cell weights of the strains cultivated under 10°C and 25°C were measured on day 12. Each data instance represents the mean ± SD of two replicates. The significant differences are calculated by Student’s t-test and are designated by asterisks (*p < 0.05, **p < 0.01, ***p < 0.001).

The dry cell weights (DCW) also showed relatively similar results to the growth data (Fig. 3D). Among the three strains, UTEX265 strain exhibited the lowest biomass yield of 1.5 g/l and 2.75 g/l at 10°C and 25°C, respectively. As for the two novel strains, ABC-002 (2.45 g/l, 3.4 g/l at 10°C, 25°C respectively) was found to yield higher DCWs than those of ABC-008 (2.25 g/l, 2.85 g/l at 10°C, 25°C respectively), even though ABC-008 had higher cell concentration.

Lipid Content and Composition

The total esterifiable lipid content of each strain was determined by GC after conversion to FAME, as described in Materials and Methods. At 10°C, C. vulgaris UTEX265, ABC-002, and ABC-008 accumulated lipid content per biomass of 9.33%, 16.49%, and 26.63%, respectively. The differences in lipid content among the three strains were smaller when they were cultivated at 25°C, as they achieved higher lipid content of 20.23%, 27.95% and 32.34% lipids, respectively. These findings indicate ABC-002 and ABC-008 had superior oleaginous traits in comparison to that of UTEX265. As for the fatty acid composition of the three strains under both conditions, there were no noticeable differences except for the ratio of oleic acid (C18:1) and linoleic acid (C18:3). A higher proportion of oleic acid was observed in ABC-002 and ABC-008 strains, whereas linoleic acid was more abundant in UTEX 265 (Table 1). The ratios between C18:1 and C18:3 were 0.4, 0.84, and 1.42 for these strains at 10°C and 1.12, 1.59 and 1.68 at 25°C. Accordingly, higher proportions of MUFA were found in ABC-002 and ABC-008 than in UTEX265, while PUFA was more abundant in UTEX265, especially when cultivated at 10°C.

Table 1 . The fatty acid profile of C. vulgaris ABC-002, ABC-008 and UTEX265 cultivated at 10° and 25° in TAP media..

	Cultivation temperature	FAMEs (%)	Fatty acid composition (%)												CNa	DUb	LCSFc	CFPPe
			C14:0	C16:0	C16:1	C16:2	C18:0	C18:1	C18:2	C18:3	etc.	SFA	MUFA	PUFA
C. vulgaris UTEX265	10°C	9.27	1.47	16.78	0.00	3.21	0.00	17.80	11.51	45.05	4.19	22.44	17.8	59.76	42.18	137.32	1.68	-11.20
C. vulgaris ABC-002		16.49	1.27	15.79	0.00	2.17	1.06	30.81	7.66	36.67	4.57	22.69	30.81	46.51	45.66	123.82	2.11	-9.85
C. vulgaris ABC-008		26.88	1.20	16.05	0.57	2.08	1.30	40.31	7.68	27.77	3.05	21.59	40.88	37.53	48.89	115.94	2.25	-9.40
C. vulgaris UTEX265	25°C	20.47	1.71	23.17	0.35	3.78	2.16	29.45	10.04	26.06	3.27	30.31	29.81	39.88	50.29	109.57	3.40	-5.80
C. vulgaris ABC-002		28.19	1.15	19.43	1.18	4.62	2.23	35.02	11.35	22.19	2.83	25.64	36.20	38.19	50.85	112.52	3.06	-6.86
C. vulgaris ABC-008		32.62	1.39	20.23	1.18	2.91	2.27	37.33	10.50	21.57	2.62	26.51	38.51	34.98	51.33	108.47	3.16	-6.56

The cetane number (CN), degree of unsaturation (DU), long-chain saturated factor (LCSF), and cold filter plugging point (CFPP) were estimated to evaluate the potential for use as a biodiesel..

^aThe cetane number (CN) was calculated as follows: CN = 61.1 + 0.088X₂ + 0.133X₃ + 0.152 X₄ – 0.101X₅ – 0.039X₆ – 0.243X₇ – 0.395X₈, where the variables X₂ to X₈ indicate.

the weight percentages of methyl esters, as follows: C14:0, C16:0, C18:0, C18:1, C18:2, and C18:3, respectively [37]..

^bThe degree of unsaturation (DU) was calculated as follows: DU = 1 (monounsaturated Cn:1, wt.%) + 2 (polyunsaturated Cn: 2, wt.%) [38]..

^cThe long-chain saturated factor (LCSF) was calculated as follows: LCSF = 0.1 C16 (wt.%) + 0.5 C18 (wt.%) + 1 C20 (wt.%) + 1.5 C22 (wt.%) + 2 C24 (wt.%) [38]..

^dThe cold filter plugging point (CFPP) was calculated as follows: CFPP = 3.1417 (LSCF) – 16.477 [38]..

The lipid productivity of each strain was calculated from the lipid content and DCW data shown in Fig. 3D and Fig. 4A (Fig. 4B). When cultivated at 10°C, ABC-008 showed the highest lipid productivity (49.93 mg/l/day), followed by ABC-002 (32.98 mg/l/day) and then UTEX265 (11.72 mg/l/day). Meanwhile, ABC-002 showed the highest lipid productivity (79.20 mg/l/day) at 25°C, followed by ABC-008 (76.70 mg/l/day) and C. vulgaris UTEX265 (46.42 mg/l/day).

Figure 4. The esterifiable lipid content and productivities of C. vulgaris strains under the culture conditions of 10°C and 25°C in TAP media. The (A) FAME content and (B) FAME productivities were determined on day 12. The data points represent the means of duplicate samples and the error bars are the standard deviations. The significant differences are calculated by Student’s t-test and are designated by asterisks (*p < 0.05, **p < 0.01, ***p < 0.001).

Chlorella is one of the most noteworthy algal genera, having been studied for more than a century since its discovery in 1890 [25]. Among more than 40,000 microalgal species that have been already identified or analyzed [26], Chlorella has attracted much interest due to the combination of high growth rate and relatively high lipid content [5]. Although the growth of most Chlorella species is adversely affected under lower temperatures, there have been a number of reports of cold-tolerant species from the Antarctic and Arctic regions [6, 27, 28]. In the present study, algal species ABC-002 and ABC-008 with cold-resistant phenotype were isolated from South Korea during the winter season, and they were identified as C. vulgaris upon further analysis including 18S rDNA sequencing.

The morphologies of the two Chlorella species ABC-002 and ABC-008 were similar to those of typical Chlorella species and were not much different from C. vulgaris UTEX265 strain, which was used as a comparison (Fig. 1). However, the growth rates of ABC-002 and ABC-008 were superior to that of the UTEX265 strain at both 10°C and 25°C, as the final cell density was nearly seven and two times higher, respectively (Figs. 3A and 3B). These results indicate that the ABC-002 and ABC-008 strains possess not only cold-resistant phenotypes, but are also capable of achieving higher growth rate and cell densities under normal room temperature. The higher maximum quantum yields (F_v/F_m) in ABC-002 and ABC-008 compared to that of the UTEX265 strain also support the superior photosynthetic performance of the two novel strains (Fig. 3C) [29]. Furthermore, a decline of the F_v/F_m ratio is often indicative of cells entering stress condition. Thus it can be inferred that the photosynthetic performance of the ABC-002 and ABC-008 strains was not hindered substantially under low temperatures, as their F_v/F_m showed only slight decreases at 10°C of 5.43% and of 1.07%, respectively. In contrast, the F_v/F_m ratio of UTEX265 showed a greater decline of 15.8%, indicating that the cells were in a highly stressed condition at low temperature.

The results of this study also demonstrated that the two novel strains not only showed superior biomass yield than the UTEX265 strain, but also possess 1.4 to 2.9-fold higher lipid content depending on the condition. Oftentimes the growth rate and lipid content are inversely correlated, which hinders the attempts at increasing the lipid productivity via genetic engineering or by optimizing the cultivation conditions [30, 31]. Hence, the high performance in terms of both lipid content and growth rate in ABC-002 and ABC-008 imply that these strains innately possess more efficient metabolism or photosynthetic activities compared to those of C. vulgaris UTEX265 or other strains isolated from the same lake.

UTEX265 strain is known to possess favorable fatty acid composition, which makes it suitable for the production of biodiesel [32]. Therefore, UTEX265 was used as a benchmark against the ABC-002 and ABC-008 strains to judge the latters’ potential for industrial uses (Table 1). Generally, the quality of biodiesels can often be estimated by parameters such as CN (cetane number), DU (degree of unsaturation), LCSF (long-chain saturated factor), and CFPP (cold filter plugging point), which are determined by the FAME composition [33, 34]. Among these parameters, CN is directly related to the combustion quality and ignition delay, where a high CN value is beneficial for cold starts and reduces the amounts of white smoke emission. The FAMEs for all strains had adequate CN values for use as biodiesel according to the American standards (ASTM D6751, >47) when they were cultivated at 25 C, while the CN values deteriorated when the algae were cultivated at 10°C. Only C. vulgaris ABC-002 strain was able to meet the standard limits under cold temperature. This is related to the higher portion of poly-unsaturated fatty acids in the lower cultivation conditions [35, 36], particularly in UTEX265 strain compared to the ABC-002 and ABC-008 strains. Accordingly, the degree of unsaturated fatty acids (DU) was higher in the UTEX265 strain, which is less favorable due to its association with ignition delay. The CFPPs of all strains cultivated at both temperatures showed stable values below –5°C. This is low enough to be usable in most fuel systems without much concern, as additives are capable of lowering the CFPP of biodiesels. Overall, these results demonstrate that the ABC-002 and ABC-008 possess favorable lipid composition that can be used for the production of biodiesel.

Taken together, the novel microalgal species C. vulgaris ABC-002 and ABC-008 are expected to have high potential for biofuel production and are particularly suitable for outdoor cultivation in countries with distinct seasonal variation in climate, such as Korea.

This research was supported by the Advanced Biomass R&D Center (ABC) of the Global Frontier Project, funded by the Ministry of Science and ICT (ABC-2010-0029728, 2011-0031343 and 2011-0031350).

The authors have no financial conflicts of interest to declare.