Fast-Tracking Isolation, Identiﬁcation and Characterization of New Microalgae for Nutraceutical and Feed Applications

: Microalgae contain high-value biochemical compounds including fatty acids (FA), protein and carotenoids, and are promising bioresources to enhance nutrition of food and animal feed. Important requirements for commercial strains are rapid growth and high productivities of desirable compounds. As these traits are believed to be found in aquatic environments with ﬂuctuating conditions, we collected microalgae from marine and freshwater environments that are subjected to eutrophication and/or tidal ﬂuctuations. Using this directed approach, 40 monoalgal cultures were isolated and 25 identiﬁed through 18S rDNA sequencing and morphological characterization. Based on their high growth rates (0.28–0.60 day − 1 ) and biomass productivities (0.25–0.44 g L − 1 day − 1 ) in commercial fertilizer under standardized conditions, six new strains were selected. Scenedesmus sp. GW63 produced quality FA-rich biomass with high omega-3 polyunsaturated FA (28.5% of total FA (TFA)) contents, especially α -linolenic acid (ALA; 20.0% of TFA) with a very low n-6/n-3 ratio (0.4), and high FA productivity (32.6 mg L − 1 day − 1 ). A high protein productivity (34.5 mg L − 1 day − 1 ) made Desmodesmus sp. UQL1_26 (33.4% of dry weight (DW)) attractive as potential protein-rich feed and nutrition supplement. Monoraphidium convolutum GW5 displayed valuable carotenoid production (0.8% DW) with high carotenoid accumulation capability (0.8 mg L − 1 day − 1 ). This research provides a pathway for fast-tracking the selection of high-performing local microalgae from different environments for nutraceuticals, functional foods and animal feed applications.


Introduction
Microalgae are promising bioresources that contain a wide array of valuable bioproducts to enhance the nutritional value of food and animal feed. Some of these unicellular and multicellular photosynthetic microorganisms are opportunistic and can grow rapidly under favorable conditions but also survive under extreme environmental conditions. Microalgae are highly abundant and have a rich biodiversity, with more than 72,500 species [1]. The numerous advantages of cultivating microalgae over terrestrial plants for food and feed applications are widely recognized: higher growth rates [2] and biomass productivities per land area [3]; cultivation on non-arable land [2][3][4][5]; high efficiencies in reducing greenhouse gas emission and removing CO 2 from industrial flue gases by algae bio-fixation [6,7].
Based on their potentially high biomass productivity and rich biochemical composition, there has been much interest to use microalgae for a diverse range of commercial applications. This includes the use of microalgae as renewable energy sources such as biodiesel or biocrude oils [4,5,8,9], wastewater bioremediation [10], photosynthetic gas exchangers for space travel [11], cosmetics [12], nutraceuticals [13,14], functional foods [13], and feed additives [15,16]. While there is potential for microalgae-based biofuels, the high cost of microalgae production makes such low-value commodities unfeasible under current market conditions [4,13,17]. Most promising at the current state-of-the-art is therefore to an ecological risk as invasive species [5]. However, this is contrary to most commercial microalgae operations who typically only use less than a dozen strains. Despite the fact that algae have numerous potentials and benefits to be utilized in human and animal nutrition and their high biodiversity, only few species are currently commercialized in the market.
Much effort needs to be made to investigate 'wild' local microalgal strains from unique and diverse aquatic environments with capacities of more efficient biocompound production [18]. Significant opportunities exist to identify, assess and select promising local strains, based on rapid growth in agricultural fertilizers, high biomass productivity, easy harvesting capability, high protein, carotenoid and oil content and quality, all of which are key properties for successful commercialization. As the algae industry is expanding, there is a need to fast-track effective isolation and characterization of high-performing local strains. An important foundation step is to rapidly identify, assess growth/harvesting properties, and profile and quantify the biochemical composition of local microalgae to target high-value natural compounds for nutraceuticals, food and animal feeds. The current study provides a pathway for the rapid selection of local high-performing strains by sampling from fluctuating and extreme environments, where opportunistic microalgae with high compound inducibility are likely to preferentially reside. Following isolation and identification, standardized protocols for cultivation, harvesting and compound profiling were used.

Microalgae Sample Collection and Isolation
In an effort to streamline the sampling of robust and fast-growing microalgae, water samples of microalgae were collected from a great variety of local fluctuating aquatic environments, such as freshwater lakes and ponds (subject to sporadic eutrophication), as well as environments with tidal fluctuating conditions such as a brackish river, an estuary and a coastal rocky beach of subtropical regions in Queensland, Australia (for exact location see Results section). All samples were filtered through a 50-micron-filter mesh and initially cultivated with F/2 medium (Algaboost F/2 (2000×) silicate free, AusAqua, Wallaroo, Australia) [31] for 2 weeks. Pure cultures were then isolated by micromanipulation and agar plate streaking, and subsequently kept as stock cultures following the rapid isolation protocol by Duong et al. [32,33]. In addition, one promising strain from the Northern Territory, Scenedesmus dimorphus NT8C, from the Queensland Microalgae Collection at the University of Queensland, Brisbane, Australia [34] was included in this study as a benchmark strain. All microalgal stock cultures were enriched with agricultural-grade fertilizer nutrients (Table 1) and were maintained in 50 mL Erlenmeyer flasks with constant orbital shaking (100 rpm) at 25 ± 2 • C, under 24 h white fluorescent light (120 µmol photons m −2 s −1 ). As axenic cultures often produce low biomass productivity [25], xenic, monoalgal cultures (with associated bacteria still present) were maintained and sub-cultured every three weeks. All microalgal cultures were routinely examined under a compound microscope (Olympus CX21LED, Notting Hill, Australia) for other microalgal contamination or visible bacterial presence during exponential algal growth.

Standardized Cultivation and Growth Experiments
Based on pre-screening criteria of fast growth, ease of harvesting (settling properties), and easy cultivation in fertilizer media, seven microalgal strains were pre-selected for the growth experiment (see Results section). In order to promote health of cultures and standardize cell densities, recent stock cultures were used as inoculums and initially cultured for 4 days till they reached late logarithmic (late log) phase in fertilizer medium (Table 1). Nitrate concentrations were measured with the Nitrate (NO 3− ) Test Kit (API Aquarium Pharmaceuticals) on day 4 to ensure cultures contained adequate nutrients, prior to the growth experiments. A total of 50 mL of inoculated algal cultures was used to start the standardized growth experiment with 450 mL fertilizer medium. All cultures were maintained in 500 mL Erlenmeyer flasks with aeration (without external CO 2 supply) at a temperature and light controlled laboratory (25 ± 2 • C, 16 h:8 h light:dark cycle, 150 µmol photons m −2 s −1 light from white LED lights). Three replicates of each strain were cultured, harvested and analyzed separately. Growth experiments were carried out for 3-5 days, then 250 mL of algal cultures in late log phase were harvested through centrifugation at 1700× g for 5 min. Cell culture densities (haemocytometer, Blaubrand, Merck, Bayswater, Australia) and nutrient concentrations were measured daily. As lipid production increased in some microalgal species under nutrient deprivation [2,3,14,35], the remaining nutrient deprived cultures were further cultivated for 2 days to induce lipid production. A total of 100 mL of remaining biomass was harvested and centrifuged at 1700× g for 5 min. After nutrient deprivation, 2 mL of microalgal cultures were stained with 0.5 mg mL −1 Nile red (Sigma-Aldrich, dissolved in dimethyl sulfoxide) for 15 min and photographed under a compound microscope (Zeiss AX10, AxioCam MRc camera, Oberkochen, Germany) with epifluorescent light (excitation: 510-550 nm, emission: 590 nm). All harvested biomass was frozen at −80 • C and lyophilized for 3 days. Biomass used for protein analysis was stored at −20 • C and the remaining biomass was stored at −80 • C till further analyses. The average growth rate [36] was calculated using the equation: where Nx was the number of cells at the start (tx) of the growth period (5-7 days) and Ny was the number of cells at the end (ty) of the growth period. The average doubling time [2] was calculated using the equation: T = t y − t x / log 2 N x /N y Biomass productivities (g L −1 day −1 ) were calculated as total DW over the duration of the growth experiment (late log phase (3-5 days, LL), average of the total growth experiment period (5-7 days, AVE), and the nutrient depleted period (2 days, ND)).

DNA Isolation, Sequencing and Identification
Genomic DNA was extracted from 15 mL fresh algal stock cultures of each strain using a DNeasy Plant Mini Kit (Qiagen) following the manufacturer's instructions. DNA amplification for the partial sequence of 18S rRNA gene was performed by polymerase chain reaction (PCR) with the following primers: Forward: 5 -GCGGTAATTCCAGCTCCAATAGC-3 and Reverse: 5 -GACCATACTCCCCCCGGAACC-3 . PCR was performed in a 25 µL mixture containing 1.25 U/µL TAQ DNA Polymerase (Scientifix, Clayton, Australia), 1× Reaction Buffer with 2 mM MgCl 2 (Scientifix, Clayton, Australia), 200 µM dNTP mix (Scientifix, Clayton, Australia), 0.25 µM of each primer, 18.75 µL of UltraPure Distilled Water (Thermo Fisher Scientific, Waltham, MA, USA) and 1 µL of genomic DNA (10-25'ng), and amplified on a Bio-Rad DNA Engine PCR machine with the following cycling profile: initialization at 94 • C for 5 min, denaturation at 94 • C for 30 s, annealing at 55 • C for 30 s, extension at 72 • C for 1 min, final extension at 72 • C for 10 min. PCR products were purified using a Wizard SV Gel PCR Clean-Up System (Promega, Madison, WI, USA). A volume of 5 µL of purified DNA (25 ng) was mixed with 0.83 µM of each primer separately, topped up to 12 µL in a final volume with UltraPure Distilled Water and sequenced in both directions by the Australian Genome Research Facility in Brisbane, Australia.
Sequences were manually checked, edited and submitted to GenBank (Table 1). DNA sequences were searched and compared using nucleotide BLAST in GenBank (http:// blast.ncbi.nlm.nih.gov/Blast.cgi, accessed on 27 October 2015). Based on BLAST results, closely related sequences in preference to published peer reviewed journal articles were downloaded and included for constructing the phylogenetic tree. Species names were assigned to sequences with 100% identity match, whereas genus names were adopted with 99% identity match. When sequence identity was less than a 99% match or was matched with multiple genera, morphological taxonomy and phylogenetic trees were used to identify the genus of strains. A total of 38 sequences (13 from GenBank and 25 from this study) were aligned and trimmed into standard size in MEGA 6.06. The maximum parsimony phylogenetic tree was constructed in MEGA 6.06 using the Subtree-Pruning-Regrafting (SPR) algorithm and was drawn to scale, with branch lengths calculated using the average pathway method [37,38]. The alignment consisted of 397 positions. Gaps and missing data from all positions were eliminated. The bootstrap test (1000 replicates) was used to estimate the reliability of the phylogenetic tree [39].

Fatty Acid Methyl Ester (FAME) Analyses
Lyophilized algal biomass (5 mg) was hydrolyzed and methyl-esterified in 300 µL of a 2% (v/v) H 2 SO 4 methanol solution for 2 h at 80 • C with 50 µg C21:0 (heneicosanoic acid, Sigma-Aldrich, St. Louis, MO, USA) as an internal recovery standard. A total of 300 µL of hexane and 300 µL of 0.9% (w/v) NaCl were added to the mixture. The mixture was vortexed for 20 s and centrifuged at 16,000× g for 3 min to facilitate phase separation. A total of 150 µL of the hexane layer was injected into an Agilent 6890 gas chromatograph connected to a 5975 MSD mass spectrometer. The running conditions of Agilent's RTL DBWax method (Application note: 5988-5871EN) was set up as described previously [2]. Fatty acids were identified by comparing retention times with pre-run external standards (37 FAME Mix, Supelco, identified through the NIST library). A linear calibration curve based on the internal standard C22:4(n-6) (adrenic acid) and the internal recovery standard C21:0 were used for fatty acid quantification. TFA content was determined as the sum of all fatty acids. TFA productivity (mg L −1 day −1 ) was calculated as TFA multiplied by the average biomass productivity over the entire growth period.

Total Protein Analyses
The protein content was measured by the Lowry method ( [22,40]  where C was the protein concentration (mg L −1 ) obtained from the calibration curve, V was the volume (L) of the lysis buffer, D was the dilution factor, and m was the amount of Phycology 2022, 2 91 biomass (mg) [22]. Protein productivity (mg L −1 day −1 ) was calculated as the total protein content multiplied by the biomass productivity during the late log phase.

Total Carotenoid Analyses
The total carotenoid extraction was carried out as previously described [36,41,42]. Briefly, 10 mL of 90% acetone (Merck KGaA, HPLC grade, Bayswater, Australia) was added to 10 mg of crushed lyophilized algal sample and mixed by vortexing. The mixture was then kept in the dark for 24 h at 4 • C. A spectrophotometer (Hitachi U-2800, Tokyo, Japan) was used to obtain the absorbance of the solvent at wavelengths 470, 646.6 and 663.6 nm. The carotenoid content of the biomass was calculated using the following equations: where V was the volume (mL) of the acetone and m was the amount of biomass (g). Carotenoid productivity (mg L −1 day −1 ) was calculated as the total carotenoid content multiplied by the biomass productivity during the late logarithmic phase.

Statistical Analyses
All results were obtained from three separately-grown replicate cultures (six replicates for reference strain S. dimorphus NT8C) and are shown as the mean values ± SE. Growth rate, doubling time and biomass productivity were analyzed by ANOVA followed by Tukey's test.

Sample Collection, Isolation and Identification
A total of 40 microalgal strains were successively isolated and cultured from marine, brackish and freshwater environments that are subjected to frequent fluctuating conditions (eutrophication and/or tidal influences). Based on initial observations under a compound microscope, freshwater strains were dominated by Chlorophyceae, whereas marine strains were dominated by Bacillariophyceae. More than half of the marine algae were unhealthy (displaying lysed cells or fractured cell walls) and ultimately died upon culturing in agricultural fertilizer media, leaving 25 cultures. BLAST results of the 18S rRNA gene sequences were able to assign identity to 20 out of 25 isolated new strains, and those with less than 99% sequence identity matches were confirmed by morphological taxonomy and the maximum parsimony phylogenetic tree (Table 2, Figure 1).
UQL1_19 and UQL1_26 were identified as Desmodesmus sp. as they formed a wellsupported clade (99% bootstrap support) with Desmodesmus armatus (Genbank accession: KP281291 and KP281290), Desmodesmus communis (Genbank accession: KF864475), Desmodesmus armatus var. subalternans (Genbank accession: KF673362), and Desmodesmus opoliensis (Genbank accession: AB917107). The morphological description matched with this finding, where the formation of spines was the distinct difference between Desmodesmus and Scenedesmus [43]. Non-spiny strains were identified as Scenedesmus, while spiny strains were assigned into the genus Desmodesmus. The sequence of Scenedesmus sp. GW63 was 99% identical to Scenedesmus parvus (Genbank accession: FR865718). UQL1_13 formed a distinct clade (99% bootstrap support) with Ankistrodesmus gracilis (Genbank accession: Y16937) and Ankistrodesmus sp. (Genbank accession: KM676973), and therefore was assigned to Ankistrodesmus species. GW5 sequence was identical to the Genbank sample Monoraphidium convolutum (AY846377). However, this strain clustered with Ankistrodesmus sp. UQL1_13 rather than other Monoraphidium species. UQL1_20 had a sequence identity of 100% with Monoraphidium sp. (Genbank accession: AY846386 and AY846387) and Monoraphidium pusillum (Genbank accession: AY846383) and was ultimately identified as a Monoraphidium species. Clear monophyletic clades of Scenedesmaceae, Selenastraceae, Chlorophyceae, and Bacillariophyceae were formed in the 18S rRNA phylogeny in this study ( Figure 1). Among the 25 identified new monoalgal strains (Table 2), six displayed initial fast growth, easy cultivation with agricultural fertilizer-grade nutrients and settling as an indicator of good harvesting potential ( Figure 2). These were subjected to standardized growth and lipid induction experiments in triplicate cultures to allow a direct comparison.  mately identified as a Monoraphidium species. Clear monophyletic clades of Scenedesmaceae, Selenastraceae, Chlorophyceae, and Bacillariophyceae were formed in the 18S rRNA phylogeny in this study (Figure 1). Among the 25 identified new monoalgal strains (Table 2), six displayed initial fast growth, easy cultivation with agricultural fertilizer-grade nutrients and settling as an indicator of good harvesting potential ( Figure 2). These were subjected to standardized growth and lipid induction experiments in triplicate cultures to allow a direct comparison.

Cultivation and Growth
After inoculation for the standardized assay, most cultures maintained exponential growth with the exception of Scenedesmus sp. GW63, which did not grow during day 1 and resumed growth on day 2. A lag phase was observed in Monoraphidium sp. UQL1_20

Cultivation and Growth
After inoculation for the standardized assay, most cultures maintained exponential growth with the exception of Scenedesmus sp. GW63, which did not grow during day 1 and resumed growth on day 2. A lag phase was observed in Monoraphidium sp. UQL1_20 on day 3, and Desmodesmus sp. UQL1_26 on day 2 and day 4. All cultures attained exponential growth in relatively homogeneous patterns till the end of the experiment (Figure 3). S. dimorphus NT8C maintained relatively the same growth rate (on average) before and after nutrient depletion. The average growth rate was slightly reduced in Ankistrodesmus sp. ULQ1_13, Desmodesmus sp. UQL1_19, Monoraphidium sp. UQL1_20 upon nutrient deprivation, and slightly increased in M. convolutum GW5, Scenedesmus sp. GW63 and Desmodesmus sp. UQL1_26 (Figure 3). Nutrient draw down was the fastest in S. dimorphus NT8C, with no residual nitrate detected in the medium on day 3. Nitrate in culture media was consumed by all other strains by day 5 (Figure 4).
M. convolutum GW5 reached the maximum cell density (cell density Max ) of 176.50 × 10 6 cells mL −1 , while Desmodesmus sp. UQL1_26 reached the minimum cell density of 14.96 × 10 6 cells mL −1 among the seven studied strains (Table 3). S. dimorphus NT8C was the highest biomass producer when grown till late log phase (0.17 g L −1 day −1 ) and a high biomass producer during the whole period of the growth experiment (0.24 g L −1 day −1 ) and nutrient deprivation phase (0.37 g L −1 day −1 ). However, Scenedesmus sp. GW63 became the most productive strain with its high average biomass productivity over the duration of the growth experiment (0.27 g L −1 day −1 ) and its biomass productivity over the nutrient depleted period (0.37 g L −1 day −1 ). The fast-growing strain Ankistrodesmus sp. UQL1_13 showed considerable strength in producing biomass during nutrient depletion (0.37 g L −1 day −1 ). Notably, algal biomass productivity achieved during the nutrient depleted period was higher overall compared to other growth phases; however, patterns were highly variable between species.  Table 3). The lowest growth rate was observed in Desmodesmus sp. UQL1_26 (0.28 day −1 ). Average growth rates were highly variable within the same genus apart from two Scenedesmus spp. According to growth rates, cell doubling time was the shortest in Ankistrodesmus sp. UQL1_13 (1.15 days) and the longest in Desmodesmus sp. UQL1_26 (2.45 days) ( Table 3). Biomass productivityLL-biomass productivity at the late log phase, Biomass productivityAVE-biomass productivity on average of the total growth experiment period, Biomass productivityND-biomass productivity at the nutrient depletion period. Different small letters indicate significant (p < 0.05) differences based on ANOVA followed by Tukey's test. * Value was underestimated due to biomass loss in centrifugation during harvesting.
M. convolutum GW5 reached the maximum cell density (cell densityMax) of 176.50 × 10 6 cells mL −1 , while Desmodesmus sp. UQL1_26 reached the minimum cell density of 14.96 × 10 6 cells mL −1 among the seven studied strains (Table 3). S. dimorphus NT8C was the highest biomass producer when grown till late log phase (0.17 g L −1 day −1 ) and a high   Biomass productivity LL -biomass productivity at the late log phase, Biomass productivity AVE -biomass productivity on average of the total growth experiment period, Biomass productivity ND -biomass productivity at the nutrient depletion period. Different small letters indicate significant (p < 0.05) differences based on ANOVA followed by Tukey's test. * Value was underestimated due to biomass loss in centrifugation during harvesting.

Total Carotenoid Content
Striking variation (ranging from 0.1 to 8.1 mg g −1 DW) of the total carotenoid content was found across the studied strains ( Figure 6). M. convolutum GW5 accumulated the largest amount of carotenoids (8.1 mg g −1 DW), followed by Ankistrodesmus sp. UQL1_13

Sample Collection, Isolation, Identification and Selection of Microalgal Strains
Selecting suitable local microalgal strains is a fundamental step for achieving reliable and feasible commercial production of bioproducts. It requires algae to produce the

Sample Collection, Isolation, Identification and Selection of Microalgal Strains
Selecting suitable local microalgal strains is a fundamental step for achieving reliable and feasible commercial production of bioproducts. It requires algae to produce the desired biocompounds under the local environmental conditions of which large-scale cultivation is envisaged [44]. Thus, the habitat-centered collection strategy [45] was adopted in the hope that these algae are better adapted and resilient to local conditions. In the present study, microalgae were collected from local sites with suboptimal conditions, and subsequently isolated and identified. The choice of these sites was driven by the notion that aquatic environments with fluctuating and at times extreme conditions provide the best habitats for both, fast-growing, opportunistic as well as resilient microalgae. In our study, variable conditions included fluctuating salinity, temperature, and nutrient availability. A rapid isolation protocol was combined with initial screening and standardized productivity assays that enable direct comparisons between strains as well as benchmarking to previously characterized strains (Scenedesmus dimorphus NT8C in our study). In this assay, parameters, such as light irradiation, cultivation time and temperature were kept constant. While this led to the rapid identification of local high-performing strains for various applications, it should be considered that the standardized assay is not optimized for each strain. It therefore serves as a decision tool to select microalgae for large-scale cultivation, rather than providing definite answers on which strains can be used commercially under fully-optimized conditions. Growth optimization, large-scale cultivation and stimulation of biosynthesis of various valuable compounds, such as carotenoids, will still be required. Nevertheless, the most critical issues (rapid growth in commercial fertilizer, high compound productivity, and cost-effective harvesting by settling) have been addressed by the current study.
Partial sequence of the 18S rRNA gene is a widely accepted and commonly used approach in microalgae identification [46][47][48][49]. However, the analyses of sequence, phylogeny and morphology were inconclusive to identify all microalgal strains to the species level in this study. Resolution and reconstruction of relationships at the species level is rather limited due to the high conservation of the 18S rRNA gene [43,48,50], and the fact that small subunit ribosomal (SSU) RNA often underestimates the true number of species in planktonic microalgae [48,50]. Although the majority of microalgae taxa are described based on morphological taxonomy, species identification remains challenging, due to the difficulty of observing specific morphological characteristics and the morphological variations at different life stages [51]. In addition, phenotypic plasticity caused by environmental conditions [52] and genotypic differences have been reported in many microalgae [53,54]. The studied taxa Monoraphidium, Ankistrodesmus and Kirchneriella formed a monophyletic clade within Chlorophyceae, but interestingly, two Monoraphidium spp., (M. convolutum GW5 and Monoraphidium sp. UQL1_20) did not form a distinct clade. This finding supports the view of Fawley et al. [55], and Krienitz et al. [47,56], where species of the family Selenastraceae under existing taxonomic criteria display a polyphyletic nature. Further analyses using a combination of morphological characteristics, including ultrastructure [46,47,51] and DNA sequence including the internal transcribed spacer 2 (ITS2) rDNA, which has higher substitution rates in comparing with 18S rRNA gene [43,47,48,50,56], are required for discrimination of algal species.

Cultivation and Growth
Rapid growth rate and high biomass productivity are key desirable characteristics in selecting microalgae for cost-effective commercial cultivation [44]. The standard growth experiment under unoptimized laboratory conditions [2] was used to establish and compare growth rates, biomass productivity, and biochemical contents of seven microalgal strains in this study. Overall, the average growth rate (ranged from 0.28 to 0.60 day −1 ) was comparable to, or greater than those recorded in other studies [57][58][59][60], aside from M. convolutum [61] and S. dimorphus NT8C [34]. S. dimorphus NT8C is a fast growing and easily harvested alga with a wide range of temperature tolerance. The average growth rate of Ankistrodesmus sp. UQL1_13, Desmodesmus sp. UQL1_19, and Monoraphidium sp. UQL1_20 was found to be higher than S. dimorphus NT8C. However, as shown in this study and previous literature [62][63][64], fast growth rates do not necessarily represent high biomass productivities. The largest biomass producer Scenedesmus sp. GW63 was shown to produce 18.9% more biomass than S. dimorphus NT8C. The highest biomass productivity (0.25 to 0.44 g L −1 day −1 ) was achieved during nutrient depletion where high biomass density was accumulated, despite the lack of CO 2 supply, prolonged photoperiods, and high nitrogen availability. This is in the median or upper range compared to the estimated average microalgae biomass productivity of 0.18 g L −1 day −1 found by Griffiths and Harrison [35] and Rodolfi et al. [3]; previous studies of the same species ranged from 0.01 to 0.26 g L −1 day −1 [3,14,34,59,60,65], and current commercialized freshwater species Chlorella vulgaris ranged from 0.01 to 0.73 g L −1 day −1 [3,14,28,35,64], and Spirulina maxima ranged from 0.21 to 0.25 g L −1 day −1 [14,66].
Genotypic differences and phenotypic plasticity between strains and species may be causing variation in algae growth [67]. In fact, it is important to consider that cultivation conditions such as light, nutrients, temperature, growth phase, and even culture senescence (aging) have substantial impact on growth and biochemical content of algae [68,69]. Nitrate concentration used in commercial agricultural fertilizer medium in this study was low compared to other commonly used culturing media for green algae such as BBM (Bold's Basal Medium, [70]) and BG-11 [71]. Increased initial nitrogen concentrations have been associated with increases in growth rate [2,72,73]. Light intensity is not likely to be a limiting factor for growth and biomass production at a small scale when cell densities are low [2] and individual cells receive the optimal irradiance [74]. Nevertheless, the LED lights used in this study are more effective in energy conversion of photosynthesis and therefore might have increased algae growth rate and biomass productivity [68].
Overall, the investigated strains had fatty acid profiles characterized by relative low proportions of SFA (21.0-31.8% of TFA) and MUFA (28.5-35.4% of TFA) and high proportions of PUFA (33.0-41.6% of TFA). The dominant fatty acids were C16:0, C18:1(n-9) and C18:3(n-3) which is typical for green freshwater algae [69]. The investigated strains also had moderate or relative high proportions of omega-3 PUFA (22.0-31.2% of TFA) compared to other species such as Scenedesmus quadricauda (5.4% of TFA) [26], C. vulgaris (26.1% of TFA) [76], but lower than C. muelleri (32.4% of TFA) and P. salina (61.3% of TFA) [77]. The predominant omega-3 PUFA was ALA C18:3(n-3) which is considered essential for human nutrition and has benefits for cardiovascular diseases, liver function and can redistribute body fat in animals [79]. Furthermore, the here investigated microalgae had extremely low n-6/n-3 ratios (0.3-0.6), which makes them ideal for health and nutrition, as the typical Western diet is characterized by too high n-6/n-3 ratios [80,81] and therefore has been associated with cancer, and cardiovascular, inflammatory and autoimmune diseases [82]. While ALA is a precursor for EPA and DHA [83], the synthesis efficiency in animals and humans is generally low and additional intake of EPA and DHA is recommended [84]. Overall, the high contents of PUFA, and specifically omega-3 PUFA with associated low n-6/n-3 ratios and coupled with relatively low contents of SFA found in seven microalgae in this study are ideal for human and animal nutrition. However, the fatty acid content and composition of microalgae varied between species and strains, and these can also be strongly influenced by culture conditions such as light, nutrients, temperature and growth phase [69]. Nitrogen limitation is the major stimulus of FA accumulation and composition [69]. While this may increase the total FA content by increasing the SFA and MUFA content [3], the reduction in biomass productivity [3] and the PUFA content might not be desirable for omega-3 PUFA production [2].
While the content of TFA and the FA compositions are important selection criteria for oil-producing microalgae, of particular importance is the biomass productivity and the associated FA productivity, as this determines the final FA production in a culture system. The highest FA productivity was found in Scenedesmus sp. GW63 (32.6 mg L −1 day −1 ), which not only accumulated more FA than those of Ankistrodesmus sp. UQL1_13 (32.2 mg L −1 day −1 ) and S. dimorphus NT8C (30.8 mg L −1 day −1 ), but also had a high quality of FA with a low SFA content and a high PUFA content. This FA productivity is in the upper range of other commercial microalgae such as C. vulgaris (12.6-15.6 mg L −1 day −1 ), Spirulina sp. (approximately 4 mg L −1 day −1 ) [85], Isochrysis sp. (6.4-21.1 mg L −1 day −1 , lipid productivity), and Nannochloropsis sp. (4.6-20.0 mg L −1 day −1 , lipid productivity) [62]. In addition, the high content of total omega-3 PUFA (28.5% of TFA) and the essential PUFA ALA (20.0% of TFA), and the very low n-6/n-3 ratio (0.4) make Scenedesmus sp. GW63 a good choice for producing FA-rich biomass with high nutritional and health value for nutraceuticals, food, and livestock feed applications for species with the ability to bioconvert ALA into EPA and/or DHA. Subsequent enhancement of FA content and desired composition in this strain with special focus on omega-3 PUFA content and especially ALA content should be achieved through manipulations of culturing conditions [86].

Total Protein Content
Green algae have a high quantity of protein, with typical yields between 10 and 30% DW [87]. The crude protein content of microalgae evaluated in this study is within this range. The protein value of Desmodesmus sp. UQL1_26, Desmodesmus sp. UQL1_19, Monoraphidium sp. UQL1_20, Ankistrodesmus sp. UQL1_13 was found to be over 30% DW, which was almost twice the amount of those found in S. dimorphus NT8C (17.8% DW). Moreover, these top four protein producers exceeded the 20% DW protein supplement baseline set by the Ontario Ministry of Agriculture and Food for animal feeds [76,88]. Notably, the overall protein contents obtained in the present study were lower than those reported previously for the same species [26,34,65,76], apart from Monoraphidium sp. [89]. Not surprisingly, varied previous studies on commercial species C. vulgaris and Spirulina sp. also showed comparably lower values than the ones commonly reported [29,76,85]. Key cultivational attributes of microalgae that contribute to its protein production-light intensity and photoperiod [26,90], growth phase [14], nutrient availability [18]-are crucial considerations when evaluating protein content from different studies. In addition, some of the variability in protein content was driven by different analytical methods [22,76,91]. For example, the commonly used Kjeldahl method overestimates the protein content in microalgae with the nitrogen to protein conversion factor of 6.25 [22,91].
Nevertheless, protein content is a critical factor determining the nutritional value of microalgae in human diet, animal livestock feed, and aquaculture nutrition [16,92]. The world's most prominent commercial alga Spirulina is utilized in human nutrition due to its high protein content and other nutritive values, such as the presence of vitamins. It is renowned for improving several health conditions such as hyperlipidaemia, hypertension [16], anaemia, and immune function [76]. Microalgal biomass is also proven to be suitable for animal feed supplement [15,16,21] and about 30% of global production is currently sold for this application [15]. It can be used as partial replacement for conventional protein such as soybean and fish meal in livestock feeds [16,21]. Overall, Desmodesmus sp. UQL1_26 has the highest potential to be used for protein production as feed and nutrition supplement, due to its high protein content (33.4% of DW; Figure 5), productivity, and fast settling. Furthermore, it is capable of withstanding harsh environmental conditions and forms protective eight-celled colonies to avoid predation from small-sized grazers [52]. These features suggest that Desmodesmus sp. UQL1_26 is likely to be a reliable competitive strain when cultivated in mass outdoor open systems. To further investigate the nutritional quality of protein, the content and quantity of individual amino acids, especially essential amino acids, should be analyzed and compared to FAO/WHO guidelines [21]. Mutagenesis of promising microalgal strains can also lead to further protein contents. For example, a 60% increase in protein content (to 48.7% of DW) has been reported recently for a Chlorella vulgaris mutant with reduced chlorophyll contents when grown heterotrophically [93].

Total Carotenoid Content
Microalgae are also good sources of carotenoids, with an average content of 0.1 to 0.2% DW [16,21]. The overall carotenoid content of the here investigated strains (0.1 to 0.8% DW) are similar to, or higher than this average value, aside from Scenedesmus sp. GW63 (0.01% DW). The carotenogenic strain M. convolutum GW5 was capable of accumulating 0.8% DW carotenoids without further stress induction. This value is greater than those in commonly reported species Scenedesmus obliquus (0.3% DW) and aquaculture feed species C. muelleri (0.4% DW), Isochrysis galbana (0.5% DW), Tetraselmis suecica (0.6% DW), Pavlova salina (0.5% DW) [30,77], the commercial alga C. vulgaris (0.01-0.1% DW) [23,76], and even the famous β-carotene producer D. salina in some studies (0.7% DW) [30] and the natural astaxanxin source H. pluvialis (0.4% DW) [94]. However, as expected, the carotenoid content of M. convolutum GW5 is still considerable low compared to commercial carotenogenic H. pluvialis and D. salina, which are currently dominating the lucrative carotenoid market [4,16,17]. While these high carotenoid contents of commercial producers are usually achieved under specific stress-induced production, the microalgae presented in this study did not grow under any of the following carotenogenesis conditions: high light intensity (175 µmol photons m −2 s −1 ), phosphate or sulphate starvation and salt stress (0.8% of NaCl) [95], which might explain their relative low values. However, stress-induced carotenogenesis is likely to reduce algae growth rates [95], biomass productivity [96,97] and therefore increases the production costs [4,17]. In addition, variation in carotenoid content and the extremely low value obtained from Scenedesmus sp. GW63 can be due to the algae cell disruption method used, the pigment-protein binding efficiency [94] and loss of pigments within time [98] in aqueous acetone extraction.
Carotenoid contents and properties, especially its antioxidant activity and pigmentation are critical measurements in determining microalgae's suitability in the food and feed industry. Carotenoid-rich microalgae can be applied to food products as vitamin supplements, food colorant, nutraceuticals, and functional foods to promote overall health; and as feed additives for poultry, livestock, fish, and crustaceans to enhance coloration and survival [16,21]. Monoraphidium sp. GK12, a similar strain of M. convolutum GW5, has been proposed as a potential astaxanthin producer. In addition, it has been successfully cultivated in large-scale outdoor cultures and trialed in prawn farming for improving prawn pigmentation and survival [57,89]. In summary, M. convolutum GW5 is an attractive strain for carotenoid applications based on its high content of carotenoids (8 mg g −1 DW; Figure 6) and the ability to achieve high productivity. Furthermore, it has great potential to be cultivated at a large scale for astaxanthin production as demonstrated by the promising performance of Monoraphidium sp. GK12. Further research on M. convolutum GW5 as a source of valuable carotenoid production will rest on detailed individual carotenoid profiling, quantity and quality optimization of desired pigments through the carotenogenesis process. Recent reports on Chlorella spp. demonstrate that further increases in carotenoid biosynthesis and profile variations can be achieved by mutagenesis and induction by high light irradiation, glucose, NaCl, and nitrogen deficiency [99,100].

Conclusions
Using streamlined protocols, this study isolated and identified novel microalgal strains, and assessed their growth, biomass productivity, fatty acid content and composition, protein content, and carotenoid content. The results clearly demonstrate the naturally high growth rates and productivities of multiple promising strains under standardized conditions. In addition, the essential PUFA ALA, together with the very low n-6/n-3 ratio, and the high FA productivity suggested the suitability of Scenedesmus sp. GW63 for the production of high-quality FA oils. The high protein content and productivity, and its unique physiological response to grazers suggest that Desmodesmus sp. UQL1_26 is likely to be a reliable competitive alga when cultivated in mass outdoor open systems for protein-rich foods and feeds. Exceptionally high carotenoid content and productivity of M. convolutum GW5 were identified and highlighted as desirable traits, suggesting that this alga is an ideal candidate for high-value carotenoid production. All these characteristics confirmed great potential of Scenedesmus sp. GW63, Desmodesmus sp. UQL1_26, and M. convolutum GW5 to be exploited commercially for the production of nutraceuticals, functional foods and animal feeds. Furthermore, the combination of these three microalgae in one multi-nutrient supplement would provide add-on health and nutritional advantages, including high levels of quality FA, protein and carotenoids, respectively. Further investigations should be carried out under large-scale outdoor cultivation. Subsequently, future improvements and optimizations of desired biocompounds through manipulation of cultivation conditions, and necessary toxicity tests for these algae to be consumed by humans and animals are required. This study laid the fundamental framework for targeting high-value natural compounds from promising local microalgae for food and feed applications, through careful collection, isolation, and identification of suitable microalgae, and comparisons of biochemical contents under a standard cultivation assay. The streamlined methods are applicable to most other locations to identify high-performing local microalgal strains.