Investigation of Growth, Lipid Productivity, and Fatty Acid Profiles in Marine Bloom-Forming Dinoflagellates as Potential Feedstock for Biodiesel

: Microalgae-based biodiesel is increasingly recognized as an alternative to crop-based biodiesel. In this study, 10 local strains of dinoflagellates collected from Hong Kong waters, including a monoculture and field sample of Scrippsiella sp. isolated from an algal bloom, were evaluated against the performance of green alga Tetraselmis suecica . The specific growth rate, biomass production, lipid productivity, and fatty acid profile were investigated. The total lipid content of isolated strains ranged from 16.2% to 32.2% of the total dry biomass, whereas palmitic acid (C16:0) and docosahexaenoic acid (DHA, C22:6n3) were dominant in the fatty acid profile. Scrippsiella sp. has a high lipid productivity (47.3 mg/L/day) and fatty acid methyl esters (FAME) content (55.2–73 mg/g dry weight (dw)), which were comparable to that in green alga T. suecica . Further, monoculture and field sampled blooming Scrippsiella sp. showed no significant difference in most parameters, suggesting the possibility of harvesting a natural algal bloom population as a mitigation strategy to harmful algal bloom and to use as biodiesel feedstock. Overall, dinoflagellate species showed a slower growth rate (0.04–0.57 day –1 ) than most compared species (0.07–1.34 day – 1 ), likely due to a large genome size and low chlorophyll to carbon ratio. Notably, most investigated dinoflagellates were not ideal for mass biodiesel production due to the low growth rate and lipid productivity. However, a high level of polyunsaturated fatty acids (PUFA) in dinoflagellates are prospective for further studies in other biotechnological applications. Though effectively harvesting algal blooming biomass can be complex, it can be further explored as a strategy for algal bloom mitigation and potentially creating values at the advantage of natural bloom when applying harvested biomass for biodiesel and bioactive compounds extraction.


Introduction
Biodiesel is alternative to petroleum fuel, which can be derived from crops, such as oilseeds or other materials [1]. Microalgae are unicellular photosynthetic microorganisms living in saline, freshwater, or brackish water environments, converting sunlight, water, and carbon dioxide to algal biomass, which can be further applied for various renewable applications [2][3][4][5]. Numerous studies have suggested microalgae as a potentially rich and untapped resource for a wide range of commercial, environmental, pharmaceutical, food, and fuel applications [2,[6][7][8][9], including the production of microalgae-based biodiesel [2,10,11]. Rapid growth and high accumulation of lipids,

Algal Sampling and Cultivation Conditions
Algal samples were collected from Hong Kong coastal waters ( Figure 1) using a 10 µm mesh plankton net and transferred to the laboratory. Algal strains were isolated in samples since 1997 (Table 1). After the isolation of a single algal cell for monoculture, isolates were placed in a growth chamber (Conviron) set at 24 + 1 °C, ambient CO2 level (0.03%), 12:12 h of light:dark cycle, and with a light intensity of 120 µE m -1 s -1 supplied by cool white fluorescent tubes. Ten monocultures of dinoflagellate strains were used in the present study. To evaluate the fatty acid profile of the natural algal blooming population as potential biodiesel feedstock, a field sample of Scrippsiella sp. biomass was analyzed directly after collection and compared to the monoculture strain isolated from the same unialgal bloom. Green alga Tetraselmis suecica (CCMP904, obtained from National Center for Marine Algae and Microbiota), one of the marine microalgae readily studied for oil production [32][33][34], was analyzed in parallel with the isolated dinoflagellate strains on growth and lipid productivity. All cultures were maintained in a natural seawater-based L1 medium (30 psu, pH 8.0) [35]. Seawater was collected from Tolo Harbour and filtered with a 0.45 µm nylon membrane (Millipore, USA) prior to autoclave. Cultures were kept at an exponential growth phase by transferring to a new medium every week in 1:10 v/v, with regular monitoring under the microscope for possible contamination.

Algal Species Identification
The isolated algal cultures were identified by morphological examination under a microscope (1000X, Leica DM500, Leica Microsystems, Heerbrugg, Switzerland) or by DNA sequencing. Procedures for DNA purification, amplification, and sequencing were followed as described in [36], purified genomic DNA was extracted by a DNA extraction kit (Roche, Switzerland). The internal transcribed spacer (ITS) regions were amplified from the purified genomic DNA using ITS 1 (5'TCCGTAGGTGAACCTGCG 3') and ITS4 (5' TCCTCCGCTTATTGATATGC 3') primers [37,38]. These primers amplify ITS regions that contain 5.8S rDNA. PCR amplification was performed with C1000 Touch Thermal Cycler (Bio-Rad Laboratories, USA) at 95 °C for 5 min, followed by 35 cycles of 45 s at 94 °C, 45 s at 55 °C, and 2 min at 72 °C, and final extension at 72 °C for 10 min. Amplicons were cloned into pGEM-T easy vectors (Promega, USA) followed by DNA sequencing using the traditional dideoxy method performed by commercial facilities (1st BASE Ltd., Singapore).
Sequencing data were determined by the BLAST program and compared with published complete sequences in GenBank. Sequences were aligned by the Clustal X program [39], and the nucleotides difference was compared with the GeneDoc program [40]. The percentage of nucleotide similarity was calculated as P = (N -Nd)/N x 100%, where N and Nd are the total number of the nucleotide position compared and the total number of nucleotides differ between the sequences, respectively.

Determination of growth rate in algal strains
To unravel specific growth rates at an exponential growth phase and the maximum cell abundance, growth curves (Supplementary Figure S1) were established for each algal strain in the L1-medium. Cells were fixed with Lugol's solution and counted under a light microscope using a Sedgwick-Rafter cell counter.
The specific growth rate (µ) for the exponential growth phase was calculated using the following equation: where N0 and N1 are the cell density reading at time t0 and t1.

Determination of Algal Biomass and Total Lipid Content
Ten milliliters of culture suspension was filtered through pre-weighed GF/C glass fiber filters (Whatman GF/F, 47 mm, nominal pore size 0.7 um). During the filtration, the vacuum pressure differentials were maintained between 35 and 55 mm Hg [41]. After filtration, the filter setup was rinsed with 0.45 µm filtered natural seawater, and the filters were dried at 75 °C for 24 h until a consistent weight was attained. The weight difference of the filter before and after filtration was determined as the algal biomass in dry weight (dw).
Total lipids were extracted from freeze-dried microalgal biomass using a modified method from Bligh and Dyer [42]. The lipids were extracted with chloroform-methanol (1:1 v/v), and then separated into chloroform and aqueous methanol layers by the addition of 5% NaCl solution to a final solvent ratio of chloroform:methanol:5% NaCl solution of 2:2:1. The chloroform layer was collected and evaporated to dryness by a nitrogen flow. The total lipids were measured gravimetrically.

Determination of Fatty Acid Profile in Algal Strains
Fatty acids in microalgae underwent modified direct transesterification described by [43] to establish a fatty acid methyl esters (FAMEs) profile. Triplicates of 50 mg of samples were mixed with the transesterification reagent (i.e., 100.0 ml methanol containing 2.2 ml of sulphuric acid, 98% purity) and stirred for 2 h at 90 °C under reflux. The reaction product was then extracted with a hexane:water mixture (1:1 v/v). The FAME profile was determined by the capillary column gas chromatographic method. The FAME separation was performed on a gas chromatograph (Perkin Elmer GC 600) equipped with a flame ionization detector and a fused silica capillary column Omegawax TM 250 (30 m × 0.25 mm × 0.25 µm film thickness). The injector temperature was 250 °C, the detector temperature was 260 °C, and the injection was split at a 50:1 ratio. The thermal program was initially held at 150 °C for 16 min, then ramped at 2 °C/min to 180 °C and held for 25 min, and finally ramped at 5 °C/min to 210 °C and held for 25 min. Helium was used as carrier gas at 1.3 ml/min. Nitrogen gas was used as makeup gas (30 ml/min); the flow of hydrogen gas and synthetic air were provided at 30 and at 300 ml/min, respectively. The injections were performed with 1 µl in duplicate. FAMEs were identified by comparing the retention time with those in FAME mixed standard (18919-1AMP, Supelco) and heptadecanoic acid. Furthermore, C17:0 (Sigma-Aldrich R, USA) was used as the internal standard. Fatty acids were expressed in % of total FAME, and the FAME yield was expressed in mg/g dw.

Identification of Isolated Algal Species
In this study, the potential as biodiesel feedstock in terms of growth, lipid productivity, FAME content, and fatty acid profile were investigated in 11 strains of microalgal cultures, including 10 strains of HAB-causing dinoflagellates and the green alga T. suecica. Microscopic observation of the cultures revealed its colonial existence and purity ( Figure 2). By morphological examination and DNA sequencing, seven algal species were documented in the 10 isolated dinoflagellate strains: Prorocentrum minimum (strain a, b), Prorocentrum triestinum (strain a, b), Scrippsiella sp. (strain a, b), Scrippsiella rotunda, Alexandrium affine, Polykrikos geminatum, and Akashiwo sanguinea. In pairs of species in which two isolates were identified: P. minimum, P. triestinum, and Scrippsiella sp., they shared 92%-100% of similarity on the ITS sequence analysis ( Table 2).

Growth Rate and Biomass of Selected Algal Strains
Growth rate has been one of the most studied parameters in screening microalgae for a largescale cultivation in biofuel production [44]. An initial screening indicated that some strains were promising biodiesel feedstock with a high growth rate and biomass production. The net growth rates were diverse among the examined species (Table 3). Among all strains, T. suecica had the highest growth rate at 0.72 day -1 (one division every 1.38 days) during the exponential phase of growth and reached the maximum cell abundance of 1.3 x 10 6 cell/ml at day 10. Among the isolated dinoflagellate strains, there was high variability in growth rates, ranging between 0.04 and 0.57 day -1 . Notable differences were found between isolates of P. triestinum (0.05 day -1 and 0.57 day -1 ) and Scrippsiella sp. (0.19 day -1 and 0.28 day -1 ). However, there was no significant difference between strains of P. minimum (0.38 day -1 and 0.39 day -1 ). A strain-specific growth rate and diverse responses to stressors and environmental changes in microalgae have been documented, possibly attributed by intraspecific genetic diversity accounting strain variability within algal population [45][46][47]. Metabolic variation may also partly account for the differential responses as they were collected at different times and locations of different environmental conditions. In the present study, Scrippsiella sp. (strain a) was collected in the transitional water zone in Hong Kong while Scrippsiella sp. (strain b) was collected in the eastern water zone during an algal bloom. They may belong to different genotypes where metabolic activities and responses are likely to be different.
In view of maximum cell density, the green alga T. suecica achieved high cell abundance in a relatively shorter growth period (1.3 x 10 6 cell/ml in 10 days), and in general two-fold higher in maximum cell number than the examined dinoflagellate strains. For Scrippsiella sp. (strain b) and P. minimum (strain a), maximum cell abundance between 10 and 14 days at 1.7 x 10 4 and 2.8 x 10 5 cell/ml, respectively, was reached. For other dinoflagellate strains, maximum abundance at 5.7 x 10 3 to 3.8 x 10 5 cell/ml was reached at 16-24 days. The highest cell abundance in dinoflagellate isolates was P. triestinum (strain b) with 3.8 x 10 5 cell/ml (at day 18), whereas A. sanguinea had the lowest cell number at 5.7 x 10 3 cell/ml (at day 16). Biomass is related to bio-volume and abundance in which T. suecica showed the highest cell abundance and also corresponded to the highest biomass concentration (1.82 g/L) among all tested strains. For the selected dinoflagellate strains, the biomass concentration harvested in the stationary phase varied between 1.05 and 1.48 g/L (Table 3), with over 1.3 g/L biomass in P. minimum (strain a), Scrippsiella sp. (strain b), and the field sample Scrippsiella sp. As all monocultural strains were grown at the same laboratory condition, it may not be the optimum for all the analyzed strains, and performance in growth and biomass production may vary if other growth conditions were adopted.
For initial strain screening, rapid growth and high biomass concentration are two desirable characteristics for biodiesel production. Seemingly, the green alga T. suecica was a more prospective candidate and its remarkable performance was aligned to what has been reported in the literatures [2][3][4][5]. Among the dinoflagellates, P. minimum (strain a), Scrippsiella sp. (strain b), and the field sample Scrippsiella sp. were also relatively favorable to be further evaluated for biodiesel application. Algal growth is directly affected by the availability of nutrients, light, the stability of pH, temperature, and the density of initial inoculum [44]. Refining the culture conditions may alter the growth for a more efficient production of microalgae-based biodiesel. Though a lower growth rate and/or small cell size contribute to lower biomass productivity, it may not necessarily imply a low lipid content [48], the consideration on a total lipid and fatty acid profile in different algal strains was also very important. Table 3. Growth rate, biomass, lipid content, and lipid productivity of the selected algal strains in the present study compared to reported values of freshwater and marine species. Data expressed as mean ± SD (n = 3) for the present study; mean values were adopted from literatures. ^ Field sample collected from an unialgal bloom.

Lipid Content and Productivity of Selected Algal Strains
The total lipid content is one of the key criteria for screening algal strain containing a high intracellular lipid content for biodiesel feedstock production [44]. The total lipid content and lipid productivity of algal strains are listed in Table 3. Overall, the locally isolated dinoflagellate strains accumulated lipid at 16.2%-32.2% dw, and approximately half of the isolated marine dinoflagellate strains contained over 20% of total lipid as biomass. The values were comparable to those reported in Bacillariophyceae, Chlorophyceae, and Cyanophyceae, ranging from 5% to 57%, 1.9% to 62%, and 4% to 11% dw of lipids, respectively [52]. The total lipid content of P. triestinum (strain b), T. suecica, Scrippsiella sp. (strain b), and the field sample Scrippsiella sp. exceeded 25% of dw, whereas S. rotunda, P. minimum (strain b), Scrippsiella sp. (strain a), and P. germinatum had a total lipid content of over 20% of dw. The highest lipid content was noted in Scrippsiella sp. (strain b, 32.2% of dw) which is a pure monoalgal culture isolated from algal bloom. It should be noticed that the lipid content of the Scrippsiella sp. field sample was not significantly different (p > 0.05, data not shown) from the laboratory cultured Scrippsiella sp. Slight strain differences in the total lipid content were found in between strains of P. minimum (18.2%-22% dw) and P. triestinum (16.2%-25.1% dw). Moreover, there was a notable difference of approximately 10% dw of lipid between Scrippsiella sp. isolated from the blooming population (strain b and field sample) and non-blooming population (strain a).
Accounting with growth rates and biomass production, the lipid productivity of isolated dinoflagellate strains were between 11.1 and 47.3 mg L -1 day -1 . All dinoflagellate strains have lipid productivity below 20 mg/L/day, except Scrippsiella sp. (strain b). Owing to a relatively high lipid content (32.2% dw) and short period to reach maximum cell capacity, Scrippsiella sp. (strain b) has comparable lipid productivity as the green alga T. suecica (both at 47.3 mg L -1 day -1 ).

Comparison of Growth and Lipid Productivity with Other Freshwater and Marine Species
Growth rate, biomass concentration, lipid content, and lipid productivity of dinoflagellate strains in the present study were compared to reported values of several commonly assessed algal species for biodiesel feedstock, including marine species Thalassionema sp., D. salina, Nannochloropsis sp., and Isochrysis sp., as well as freshwater species C. vulgaris and Scenedesmus sp. (Table 3). Overall, dinoflagellate species showed a slower growth rate (0.04-0.57 day -1 ) than most compared species (0.07-1.34 day -1 ), likely due to the large genome size and low chlorophyll to carbon ratio [53].
Reported biomass values showed great variation where total lipid content in dinoflagellates were comparable to other species, generally below 30% of dw, except for selected Scrippsiella sp. (30.9%-32.2%) in the present study and previously reported in D. salina (34%-44%) and selected Nannochloropsis sp. (39.5%-44.8%). For lipid productivity, most dinoflagellate species have low lipid productivity, below 20 mg/L/day, whereas most other reported species ranged at 20%-46% mg lipid/L/day. Only Scrippsiella sp. (strain b, 47.3 mg/L/day) was comparable to reported values in marine species D. salina, Nannochloropsis sp., and Isochrysis sp. (36-46 mg/L/day). Slow growth in dinoflagellates was a major factor leading to much lower lipid productivity than other algal classes, which suggests that dinoflagellates were in general not prospective for mass cultivation for harvesting lipids, or careful selection for fast-growing strains is required.
Ignition quality can be estimated by the cetane number; higher ignition quality correlates to a higher cetane number, longer chain length, and lower unsaturation [56], i.e., a cetane number of 85.9 for saturated ester (C16:0) and 22.7 for polyunsaturated ester (C18:3n3) [1,57]. C16:0 was commonly present in algal strains for combustion as biodiesel, which was of higher proportion of FAME in T. suecica (25.2%), P. triestinum (strain b, 18.6%), and P. minimum (strain b, 17.8%). Oxidative stability and low temperature properties have been suggested as common technical difficulties faced by biodiesel [1]. In terms of oxidative stability, polyunsaturated fatty acids (PUFA) such as linoleic (C18:2) and linolenic (C18:3) acids are more susceptible to oxidation, creating biodiesel instability issues. The linolenic acid methyl ester content is also specified not to exceed 12% in The Biodiesel Fuel Testing Europe (EN-14214) [58], suggesting FAME from P. triestinum (strain b), P. minimum (strain a and b), Scrippsiella sp. (strain a), and S. rotunda, with 15.4%-23.5% of linolenic acid content were not suitable for use as biodiesel. For low temperature applications, increasing unsaturation such as enriching with oleic acid (C18:1n9) or palmitoleic acid (C16:1n7) could lower the melting point and cold-filtering plugging point that improve the performance at a low temperature [1,55]. FAME derived from algal extracts ranging from 21.8 to 73 mg/g dw (Table 5). Field sample Scrippsiella sp. showed the highest FAME content (73 mg/g dw), followed by cultured Scrippsiella sp. (strain b) and T. suecica, having over 50 mg/g dw, and P. triestinum (strain b), S. rotunda, and Scrippsiella sp. (strain a) had over 30 mg/g dw. The identified fatty acids were categorized and showed great species variation in saturated fatty acids (SFA, 11.3%-44.9% of FAME), monounsaturated fatty acids (MUFA, 2.3%-55.2% of FAME), and PUFA (1.3%-87.5% of FAME). The isolated dinoflagellate strains accumulated mostly PUFA (58%-87.5% of FAME), especially DHA (47.5%-84% of FAME). To improve the quality of lipid in microalgae for biodiesel, reducing the nutrient supply (e.g., nitrogen depletion) and refining cultivation conditions may influence the unsaturated fatty acid profile for a better performance as biodiesel [15].
A high proportion of SFA and a low proportion of PUFA are considered optimal for biodiesel [1,59,60], with MUFA components that may increase oxidative stability for prolonged storage and the use in cold regions [1]. Aligned with the above selection criteria, green alga T. suecica showed the most potential in biodiesel application with a high FAME content, as well as a relatively high proportion of SFA and MUFA (over 30% of FAME). Both field and cultured Scrippsiella sp. also had high FAME yield with a considerable proportion of SFA (16.1%-22.8%) and MUFA (9.9%-15.1%), while the high PUFA content can be extracted for other bio-products. Despite both Scrippsiella sp. samples sharing similar fatty acid profiles, significant differences (p < 0.05) were noted in the percentage of the fatty acid content in each. Thus, the percentage of saturated, monounsaturated, and polyunsaturated fatty acids varied between the two samples. In addition, the field sample of Scrippsiella sp. had a higher FAME content (73 mg/g dw) than that of the laboratory-cultured isolate (55.2 mg/g dw). These variations may be caused by the differences in growth conditions experienced by the field sample in the real coastal environment and the cultivation conditions of the monoclonal culture in the laboratory. The field sample Scrippsiella sp., which was analyzed directly after collection, represents the native algal blooming population of Scrippsiella sp. and the associated microbial communities. Whereas after isolation and culturing in laboratory light and nutrient conditions, the Scrippsiella sp. were likely growing and responding differently from the blooming population in the field. Composition of microalgae-associated bacterial communities and the presence of bacterial volatile compounds can also affect growth, biomass yield, behavior, physiology, and possibly lipid productivity of microalgae [61][62][63][64]. Varying in the abundance and dominance of microbial assemblages associated with Scrippsiella sp. in the field sample and laboratory culture may also partly contribute to the divergence in growth response, and thus their chemical constituents. As the field sample Scrippsiella sp. has not been processed prior to the analysis, the presence of environmental bacteria, though believed to be of very minor proportion, may partly contribute to the fatty acids and FAME reported.
Pre-extraction of valuable constituents, such as PUFA, carotenoids in microalgae, and the use of lipid-extracted algal residue may also provide additional value and contribute to efficient bioenergy recovery [65,66]. Further, lipid-extracted algal residue are rich in amino acids, sugars, vitamins, and minerals [67], which can be supplied for animal feeds and raw material in the bioethanol or biomethane production [65,68]. Though a critical issue of inefficient production of algal biomass, complex harvesting and high operation costs are still major limitations in commercialization [69][70][71], diverse utilization of algal biomass can increase the overall marketability of large-scale microalgae cultivation for commercial purposes. Table 4. Fatty acid composition (% of total fatty acid methyl esters (FAME)) in selected algal strains. Data expressed as mean ± SD (n = 3).

The Potential of Harvesting Natural Algal Bloom Biomass for Biotechnological Purposes
A similar lipid content and fatty acid profile in both laboratory-cultured and field-sampled Scrippsiella sp. collected from an algal bloom suggest the possibility of harvesting a natural algal bloom population as a mitigation strategy to harmful algal bloom and create value as biodiesel feedstock. Algal blooming biomass are unspontaneous events that cannot be regarded as a stable source for biodiesel. There could be values in biodiesel production and bioactive extraction at the advantage of natural bloom, though effectively harvesting natural algal blooming biomass can be complex and require further exploration.
Natural algal bloom biomass may be harvested with skimmer or a patented capillary extraction belt [72] or vacuum belt [73]. Sedimentation, centrifugation, filtration, flocculation, or flotation can be adopted to concentrate biomass after harvested by skimmer. Sedimentation and flocculation require lower energy input than other methods, but sedimentation is time-consuming. Flocculation can be performed biologically with addition of bacteria; chemically, with addition of organic or inorganic flocculants; electrically, by concentrating negatively-charged cells around electrodes by electrophoresis; or induce auto-flocculation by increasing the pH [23,74]. Flocculation of dinoflagellate has been successfully performed with calcium oxide, calcium hydroxide, aluminum sulfate, and chitosan [75]. However, the addition of calcium oxide, calcium hydroxide, and aluminum sulfate can cause a shift in pH, where chitosan remains neutral (pH = 7). The use of 0.05 g/L chitosan which can flocculate over 90% of Gymnodinium catenatum (1000-3000 cell/mL) may be a desirable method of harvesting dinoflagellates, especially for pH-sensitive downstream products. The combination of chitosan, Moringa oleifera coagulant, and sand is also effective in rapidly flocculating Amphidinium carterae [76]. The Centrifugation method is reliable but of higher cost than chemical flocculation [77]. Though the use of a low efficient, high flow rate centrifugation may greatly lower costs with the selection of high lipid and a density-producing algal species [78], it is likely not applicable in harvesting slow-growing dinoflagellates.

Other Biotechnological Applications of Fatty Acids Extracted from Microalgae
The selected dinoflagellate strains produced a high level of PUFA in the form of DHA (47.5%-84% of total FAME). These omega-3 fatty acids can be applied in food additive, nutraceuticals, animal feed, and aquaculture feed. Fatty acids (DHA, eicosapentaenoic acid EPA, γ-linolenic acid GLA, and arachidonic acid AA), vitamins, trace elements, and amino acids in microalgae can be employed as food additive and nutraceuticals [24]. DHA is one of the key components in cell membrane and a precursor of lipid mediators. It plays a crucial role in fetal development as well as cardiovascular diseases, is involved in antioxidation and anti-inflammation, and relates to Alzheimer's disease [79]. DHA extracted from Crypthecodinium sp. and Schizochytrium sp. has been utilized in infant formula and their DHA oils are commercially available, respectively produced at 240 and 10 tons annually [80]. Microalgal-derived omega-3 can be a promising vegetarian source of DHA. It showed similar properties as those in fish oil, such as reducing the risk factor for cardiovascular diseases [81]. Crypthecodinium sp. and Schizochytrium sp. produce DHA at 30%-50% of total oil [80,82]. The analyzed dinoflagellates in the present study produced a high level of DHA (47.5%-84% of total FAME), which can be a potential source for DHA production. In addition to human nutrition, microalgal-derived DHA can be added in feed for poultry and aquaculture. Replacing parts of the soybean and corn meal for poultry, DHA has been observed to increase in meats, eggs, and milk under dietary intake of microalgal biomass or oil [83,84]. In aquaculture, DHA is generally enriched from fish oils [15]. Microalgae rich in long-chain PUFA have been widely applied in culturing molluscs, shrimps, and fish larvae [85]. Under a declining availability of pelagic fisheries, microalgal oil can be an alternative feed resource rich in valuable fatty acids with regard to a species-specific nutritional value and digestibility [86]. In addition, some algal species may be also applied in aquaculture with antimicrobial and immune-stimulation properties [87].
Some dinoflagellate species are known to produce toxins that can cause fish kill and pose risks to seafood safety [88]. While dinoflagellate biomass are harvested for biotechnological applications, risks associated to human exposure, environmental threats upon leaks to natural systems, and the disposal of biomass residue have to be considered [89]. The use of non-toxin-producing dinoflagellates, such as those species considered in the present study, strains of A. affine can be regarded as non-toxic with the absence of the saxitonxin gene [90] or weakly toxic [91,92], is recommended. Rat feeding safety assessment, target animal safety, and toxicity studies are also crucial prior to application in food and animal feeds. For applications with toxic dinoflagellate species, cultivation has to be carried in controlled indoor systems to reduce the risk of contamination to nearby environments. Prior toxin extraction, genetic modification, and toxicological examination on product quality could also be strategies reducing human exposure risks to the associated algal toxins.

Conclusions
The algal blooming causative agent, dinoflagellates, was evaluated to reflect the potential as renewable biodiesel feedstock. Palmitic acid (C16:0) and docosahexaenoic acid (DHA, C22:6n3) were commonly dominant in all selected algal strains. Among all assessed dinoflagellates, Scrippsiella sp. (strain b and the field sample) has relatively high biomass, total lipid content, and lipid productivity. These isolates yield a high lipid (47.3 mg/L/day) and FAME content (55.2-73 mg/g dw), which were comparable to that in green alga T. suecica. Most investigated dinoflagellates were not ideal for biodiesel production due to the low growth rate and lipid productivity. However, biotechnological applications of the high level of PUFA and other bioactive constituents in dinoflagellates are prospective for further studies. Due to high strain variations, strain-level differences should be considered when screening suitable dinoflagellates for mass cultivation. Further, this study also showed a similar lipid content and fatty acid profile in both laboratory-cultured and field-sampled Scrippsiella sp. collected from an algal bloom, suggesting the possibility of harvesting natural algal bloom population as a mitigation strategy to harmful algal bloom and create value as biodiesel feedstock. Algal blooms are unspontaneous events that cannot be regarded as a stable source for biodiesel. Though effectively harvesting algal blooming biomass can be complex and requires further evaluation, there could be potential values in biodiesel production and bioactive extraction at the advantage of natural algal bloom.