INTRODUCTION
Microalgae are unicellular, eukaryotic organisms, and an evolutionarily diverse group containing approximately 200,000 species (Norton et al. 1996, Singh and Saxena 2015). Microalgae have been used as diverse biological resources and are also treated as one of the most promising sources for new high-value products and applications (Pulz and Gross 2004, Jang et al. 2017). Thus, several species are currently exploited for a variety of biotechnological purposes, including cosmeceuticals, animal feed, functional foods, fatty acids, other biologically active compounds, wastewater treatment, and biofuel (Fabregas and Herrero 1985, Becker 2004, Pulz and Gross 2004, Spolaore et al. 2006, Matos et al. 2017). Microalgae have some important advantages over conventional land plants because they can be produced all year around and have greater biomass productivity (Singh and Gu 2010, Buono et al. 2014). The microalgal biomass in the world has been estimated at a value of US $1.25 × 109 per year and produces 5,000 tons y−1 of dry matter (Pulz and Gross 2004). However, the biochemical compositions of microalgal are very diverse and depend on the species (Andersen 2013). Therefore, successful algal biotechnology mainly depends on choosing the right algal species with relevant properties for specific purposes (Pulz and Gross 2004). Thus, the selection, isolation, and study of new candidate microalgal species that may possess unique functional ingredients are needed.
Several microalgal species have been utilized in aquaculture as important food sources or food additives for aquaculture animals such as mollusks, shrimp, rotifers, and fish (Borowitzka 1997, Bleakley and Hayes 2017). To produce 3.8 × 106 tons of shrimps globally in 2010, 612 tons of microalgae (in dry weight) were estimated to be required (Muller-Feuga 2013). Over 20 years, several microalgal genera such as the chlorophyte Chlorella, prasinophyte Tetraselmis, prymnesiophytes Isochrysis, Pavlova, bacillariophytes Phaeodactylum, Chaetoceros, Skeletonema, Thalassiosira, and eustigmatophyte Nannochloropsis have been used in aquaculture either as individual diets or components of mixed diets (Brown 2002, Spolaore et al. 2006, Hemaiswarya et al. 2011, Roy and Pal 2015). However, to increase production of aquaculture animals, more effort is required to discover new microalgal species with higher nutritional quality. In particular, to use a microalgal species as a source of aquaculture feed, it should be easy to culture, does not produce toxic materials, but has highly nutritious proteins (Hemaiswarya et al. 2011).
Shrimps cannot synthesize all amino acids and thus they must obtain some amino acids from their diet. The amino acids that an organism cannot synthesize on its own (or are incapable of synthesizing sufficient amounts) are referred to as essential amino acids (National Research Council 2011). Arginine, histidine, isoleucine, leucine, lysine, methionine, phenylalanine, threonine, tryptophan, and valine are essential amino acids that are required in diets for shrimps (National Research Council 2011). Moreover, taurine may be conditionally essential for some juvenile marine fish and shrimp (National Research Council 2011). Furthermore, some salmonoids, shrimp, and lobsters need to consume carotenoid pigments such as astaxanthin and canthaxanthin which causes the pink color in their flesh (Gouveia et al. 2008, Bleakley and Hayes 2017). Therefore, the microalgal species which satisfy such requirements can be used for animal feeds.
Dinoflagellates live in freshwater and marine waters (Taylor et al. 2008). They are a major component of diverse marine ecosystems and play diverse roles as primary producer, predator, prey, and symbiotic partners in marine ecosystems (Jeong et al. 2010, 2015, Holmes et al. 2014, LaJeunesse et al. 2018). They also contain diverse useful materials such as omega-3, macrolides, mycosporine-like amino acids, and carotenoids (Shimizu 1996, Mendes et al. 2009, Rosic and Dove 2011, Onodera et al. 2014, Jang et al. 2017). However, only few dinoflagellate species have been studied for their amino acid compositions (Okaichi 1974, Hayashi et al. 1986, Flynn and Flynn 1992, Markell and Trench 1993), and their potential as a food source in aquaculture has been rarely studied (Sánchez 1986). To the best our knowledge, among dinoflagellates, only Gymnodinium sp. and Crypthecodinium cohnii have been used as a food material for aquatic animals (Harel and Clayton 2004, Pulz and Gross 2004, Gouveia et al. 2008, Muller-Feuga 2013). Therefore, more dinoflagellate species should be studied to develop new effective species for the aquaculture industry.
In the present study, the crude protein (CP) contents and compositions of amino acids of six dinoflagellate species, Heterocapsa rotundata, Ansanella granifera, Alexandrium andersonii, Takayama tasmanica, Takayama helix, and Gymnodinium smaydae, were analyzed. These dinoflagellates belong to diverse families such as Heterocapsaceae, Suessiaceae, Ostreopsidaceae, Brachidiniaceae, and Gymnodiniaceae (Guiry and Guiry 2018). Thus, it is worthwhile to test whether they have the CP contents and compositions of amino acids similar to each other or not, although their taxonomical distances are large. Moreover, the levels of essential amino acid of the dinoflagellates were compared to the requirement levels of essential amino acid for juvenile shrimps. Based on the average requirement of essential amino acids, essential amino acid indexes of the dinoflagellates were determined. Furthermore, nutritional values of dinoflagellates were compared to those of other microalgae, which are frequently used for aquaculture. The results of this study provide the biochemical profiles of dinoflagellates belonging to diverse families and allow the possibility to use them as a protein source for aquaculture.
MATERIALS AND METHODS
Preparation of experimental organisms
The dinoflagellates H. rotundata (HRSH1201) and A. granifera (AGSW10) were isolated from the surface water or sediment samples of Shiwha Bay, Korea in January 2012 and September 2010, respectively (Table 1). The dinoflagellate A. andersonii (AAJH1505) was isolated from Jinhae Bay, Korea in May 2015. These clonal cultures were established following two serial single-cell isolations (Jeong et al. 2014, Lee et al. 2016). Furthermore, the mixotrophic dinoflagellate G. smaydae (GSSW10) was isolated from Shiwha Bay, Korea in May 2010 (Table 1). A clonal culture of G. smaydae was established by providing H. rotundata as prey. Moreover, T. helix (CCMP3082) was obtained from the National Center for Marine Algae and Microbiota, USA and T. tasmanica (CAWD115) was obtained from the Cawthron Institute’s Culture Collection of Microalgae, New Zealand.
Cell culturing and harvesting
To obtain large volumes of H. rotundata, A. granifera, A. andersonii, T. tasmanica, and T. helix, a clonal culture of each species was maintained in enriched f/2-Si seawater media (Guillard and Ryther 1962) in 2-L polycarbonate bottles at 20°C under an illumination of 50 μmol m−2 s−1 provided by cool white fluorescent light on a 14 : 10 h light : dark cycle. However, the culture of G. smaydae was mixotrophically cultured by continuously providing H. rotundata as a prey species (ca. 30,000–50,000 cells mL−1) every two days. The bottles of G. smaydae culture were also maintained under the temperature and light conditions described above.
To determine the growth rate of each dinoflagellate species, a 5-mL aliquot was removed from each bottle every two days, or before and after providing prey cells. The samples were fixed with 5% Lugol’s solution, and all or >200 cells in three 1-mL Sedgwick-Rafter chamber were enumerated. When the growth reached the early-stationary phase, the alga was collected by continuous centrifugation at 4,000 rpm and stored at −70°C in the deep freezer until analyzed. When the cultures were harvested, the final density and concentrated volume of each culture was determined to calculate the total number of the cells.
CP and amino acid analysis
The frozen concentrated samples were dried at −70°C. Triplicated samples were analyzed except A. granifera, which were analyzed in duplicates. The CP was calculated as total nitrogen in each sample multiplied by a factor of 4.78, which was developed for marine microalgae (Lourenço et al. 2004). Total nitrogen was determined by multiplying the nitrogen content of a cell with the total number of cells in each sample. The nitrogen content of a cell was calculated from the Menden-Deuer and Lessard equation used to convert cell volume to nitrogen content for dinoflagellates (Menden-Deuer and Lessard 2000). To calculate the cell volume, cell size was measured based on >20 randomly selected cells using an inverted light microscope (Carl Zeiss Axiovert 200M; Carl Zeiss Microscopy GmbH, Jena, Germany). The volume of H. rotundata, T. tasmanica, and T. helix cells in cultures was estimated with the equation for a rotational ellipsoid, while that of A. andersonii cells was estimated with the equation for a sphere. Moreover, the volume of A. granifera and G. smaydae was obtained from our previous studies (Lee et al. 2014a, 2014b).
The 20–40 mg of each sample was hydrolyzed in 6 N HCl at 130°C for 24 h for analyzing the amino acids. Amino acid profiling included analysis of the following essential and non-essential amino acids: histidine, isoleucine, leucine, lysine, methionine, phenylalanine, threonine, valine, serine, aspartic acid, glutamic acid, arginine, glycine, alanine, proline, tyrosine, taurine, and gamma-aminobutyric acid. However, tryptophan and cysteine were not obtained because they are completely lost with acid hydrolysis, and methionine was not fully obtained because it could be destroyed to varying degrees with acid hydrolysis (Lourenço et al. 2004). Thus, in the present study, tryptophan and cysteine were omitted and the value for methionine could be underestimated. The amino acids were analyzed using Ultimate 3000 High Performance Liquid Chromatography (HPLC/Fluorescence model; Thermo Fisher Scientific, Waltham, MA, USA).
The amount of each amino acid was expressed as a percentage in CP or a percentage in the analyzed total amino acids in the present study. The essential amino acid index (EAAI) is a common index for estimating the quality of a given protein (Oser 1959, Becker 2004).
The EAAI of a dinoflagellate was determined from the formula:
where aa1, …, aan are the amino acid contents of a dinoflagellate and AA1, …, AAn are the average required levels of these amino acids in juvenile shrimps.
The geometric mean, EAAI should be the ratio of all essential amino acids of a dinoflagellate to those of whole egg (Oser 1959). However, EAAI has been modified and used with various standard proteins in the previous studies (Brown 1991, Brown and Jeffrey 1992, Becker 2004), and thus the essential amino acid requirements for juvenile shrimp served as a standard for evaluating the quality of protein in the dinoflagellate in the present study. The juveniles of the prawn Penaeus japonicus and Penaeus monodon were chosen as the standard animals representing each requirement level of essential amino acids (EAAs) because they are the major groups of shrimp used in the aquaculture industry (Madrigal et al. 2017). Because methionine could be partially destroyed during the acid hydrolysis, eight essential amino acids, histidine, threonine, arginine, valine, phenylalanine, isoleucine, leucine, and lysine were used to determine EAAI. Moreover, in the evaluation, a EAAI score of >0.95 defined a ‘high’ quality of protein, while a score of 0.86–0.95 signified a ‘good’ quality protein, a score of 0.75–0.86 signified a ‘useful’ protein, and a score of ≤0.75 indicated an ‘inadequate’ protein (Zhang et al. 2009).
The amino acid showing the greatest difference in percentage from its required level is the limiting amino acid.
Artemia test for toxicity
A bioassay for potential toxicity of each of the six dinoflagellates tested was conducted using the brine shrimp Artemia salina. Encysted eggs of A. salina were hatched in 500 mL of filtrated natural seawater under artificial light at 20°C for 48 h. Six-well plate chambers were prepared with triplicate wells of several densities of each dinoflagellate and 10 A. salina nauplii were placed in each well of the plate chambers. The experimental densities were 100, 1,000, 5,000, 10,000, 100,000, and 200,000 cells mL−1 for H. rotundata and A. granifera, 10, 100, 1,000, 2,000, and 5,000 cells mL−1 for T. tasmanica, 10, 100, 1,000, 2,000, and 4,000 cells mL−1 for T. helix, and 100, 1,000, 5,000, 10,000, and 20,000 cells mL−1 for G. smaydae. In addition, triplicate wells containing a mixture of 10 nauplii and each dinoflagellate-filtrate as well as triplicate wells containing 10 nauplii only were established. The filtrate was obtained through filtration (Whatman grade GF/C filter; GE Healthcare, Buckinghamshire, UK) of each dense culture. As a positive control, 7,000 cells mL−1 of Alexandrium minutum was used in an identical experimental setup (Lim et al. 2015). The plate chambers were incubated at 20°C under a 14 : 10 h light : dark cycle of cool white fluorescent lights at 20 μmol m−2 s−1. At the beginning of the incubation and 6, 12, 24, and 48 h later, the plate chambers were placed under a dissecting microscope, and living and dead nauplii were counted at a magnification of ×7–40.
RESULTS
CP and amino acid composition
The CP of six dinoflagellates tested in the present study ranged from 26 to 65% of dry mass (Table 2, Fig. 1A). The percentage of the amount of CP relative to dry weight of T. tasmanica was the highest (65%), that of H. rotundata, A. granifera, T. helix, and G. smaydae the second highest (53–63%), and A. andersonii the lowest (26%).
In the composition of amino acids, the top 5 amino acids were the same in H. rotundata, A. granifera, A. andersonii, and T. tasmanica; glutamic acid, aspartic acid, arginine, alanine, and leucine (Fig. 1B). Moreover, the top four amino acids, glutamic acid, aspartic acid, alanine, and leucine, were the same in T. helix and G. smaydae (Fig. 1B). In all six dinoflagellates, glutamic acid was the most dominant amino acid in CP, ranging from 7.4 to 15.8% (Table 3). The second main amino acid was aspartic acid for H. rotundata, A. granifera, T. helix, and G. smaydae, whereas arginine and leucine were the second main amino acids for A. andersonii and T. tasmanica, respectively (Table 3, Fig. 1B). Moreover, taurine and gamma-aminobutyric acid were detected in H. rotundata, A. granifera, A. andersonii, and G. smaydae although its percentages in CP were as low as <1%. However, these two amino acids were not detected in T. tasmanica and T. helix (Table 3).
Regarding the EAAs, the percentage of histidine in CP was the highest in A. granifera, whereas the percentages of threonine, arginine, valine, phenylalanine, isoleucine, leucine, and lysine in CP were the highest in A. andersonii (Fig. 2). The percentage of total detected amino acids (TDAAs) in CP varied from 50.1 to 98.2%; the lowest percentage was found in G. smaydae, while the highest percentage was in A. andersonii (Table 3). The nonessential amino acids of H. rotundata, A. andersonii, T. helix, and G. smaydae were greater than their EAAs, whereas those of A. granifera and T. tasmanica were slightly lower than their EAAs (Table 3).
Evaluation of amino acids as animal feed in aquaculture
The percentage of each EAA of each of the six dinoflagellate species was greater or lower than that required for the juvenile shrimps (Table 3, Fig. 3); the percentages of all EAA except lysine of A. andersonii were greater than those required, whereas the percentages of all EAA of G. smaydae were lower than those required. In H. rotundata, the percentages of histidine and arginine were greater than those required, whereas those of isoleucine and lysine were lower than those required (Fig. 3A). In A. granifera, the percentage of histidine was much greater than those required, whereas those of isoleucine and lysine were lower than those required (Fig. 3B). In T. tasmanica, the percentages of threonine and leucine were greater than those required, whereas those of arginine and lysine were lower than those required (Fig. 3D). In T. helix, the percentages of threonine and leucine were greater than those required, whereas those of histidine and lysine were lower than those required (Fig. 3E). The first limiting amino acid in H. rotundata, A. granifera, A. andersonii, T. tasmanica, and T. helix cells was lysine, while that in G. smaydae cells was arginine (Fig. 3). The second limiting amino acid in H. rotundata, A. granifera, and G. smaydae cells was isoleucine, while those of T. tasmanica and T. helix cells were arginine and histidine, respectively (Fig. 3).
The EAAI of six dinoflagellate species ranged from 0.79 to 1.59 (Table 3); the EAAI of A. andersonii was the highest, whereas that of G. smaydae was the lowest.
Artemia test for toxicity
None of the Artemia salina nauplii were dead in wells incubated with high concentrations of H. rotundata, A. granifera, T. tasmanica, T. helix, and G. smaydae for 48 h, but 1–2 nauplii were dead when incubated with high concentrations of Alexandrium minutum, which is known to be toxic to Artemia and thus provided a positive control. During the incubation period, abnormal behaviors of the nauplii were not observed.
DISCUSSION
The percentages of the amount of the CP relative to dry weight (PPD) of the six dinoflagellates, Heterocapsa rotundata, Ansanella granifera, Alexandrium andersonii, Takayama tasmanica, Takayama helix, and Gymnodinium smaydae tested in the present study fell within the range of other reported dinoflagellates (6.0–71.9%); 71.9% for the phototrophic dinoflagellate Prorocentrum triestinum but 6.0–21.5% for the phototrophic dinoflagellates Amphidinium carterae, Prorocentrum rhathymum, and Symbiodinium sp. (Okaichi 1974, Shah et al. 2016) (Fig. 4). Thus, the PPD of the dinoflagellates tested in this study were greater than that of the phototrophic dinoflagellates except for P. triestinum in the literature. Moreover, the PPD of the dinoflagellates except for A. andersonii tested in this study are greater than those of the other microalgal groups Chlorella sp., Tetraselmis suecica, Tetraselmis chui, Isochrysis galbana, Pavlova lutheri, Phaeodactylum tricornutum, Chaetoceros calcitrans, Nannochloropsis oculata, Skeletonema costatum, and Thalassiosira pseudonana that are most frequently used for aquaculture feeds (19–35%) (Fig. 4) (Brown 1991, Brown and Jeffrey 1992, Spolaore et al. 2006, Hemaiswarya et al. 2011). In particular, the PPD of T. tasmanica is greater than that of any microalgae except for P. triestinum.
Interestingly, taurine and gamma-aminobutyric acid were not detected in either T. tasmanica or T. helix (family Brachidiniaceae), whereas they were detected in H. rotundata (Heterocapsaceae), A. granifera (Suessiaceae), A. andersonii (Ostreopsidaceae), and G. smaydae (Gymnodiniaceae). Therefore, the absence of taurine and gamma-aminobutyric acid may be a unique character of the species in the family Brachidiniaceae. Taurine is considered to be conditionally essential to shrimps (National Research Council 2011). Therefore, all the dinoflagellates except T. tasmanica and T. helix may be useful diets for shrimps.
Aspartic acid and glutamic acid, two dominant amino acids, were abundant in H. rotundata, A. granifera, T. helix, and G. smaydae as in most microalgae reported (Brown 1991, Brown and Jeffrey 1992). The percentages of arginine (9.6–9.7%), the second or third highest amino acid of H. rotundata and A. andersonii, are greater than that of Chlorella sp., I. galbana, Pavlova salina, P. lutheri, P. tricornutum, Chaetoceros gracilis, C. calcitrans, N. oculata, S. costatum, and T. pseudonana (6.3–8.4%) but lower than that of T. chui and T. suecica (15.0% and 13.2%, respectively) (Fig. 5A). Therefore, H. rotundata and A. andersonii can be used for the formulation of arginine-rich mixed algal diets. The percentage of leucine (10.0%), the second highest in TDAAs of T. tasmanica, is greater than that of the other dinoflagellates (7.5–9.1%) and the marine microalgae (7.1–9.0%) (Fig. 5B). Therefore, T. tasmanica may be useful for the formulation of leucine-rich mixed algal diets. Among the reported microalgae, A. granifera has the highest percentage of histidine (7.5%), and T. helix has the highest percentage of threonine (8.3%). T. tasmanica has the highest percentage of valine (6.8%), and G. smaydae has the highest percentage of lysine (7.3%) (Table 4). Thus, A. granifera, T. helix, T. tasmanica, and G. smaydae can be used for histidine, threonine, valine, and lysine sources, respectively.
The EAAIs of A. andersonii, H. rotundata, and A. granifera (1.03–1.59) can be scored to be ‘high quality protein’ for animal feeds, that of T. tasmanica ‘good quality protein’, but those of T. helix and G. smaydae ‘useful quality protein’. Therefore, the nutritional values of the protein possessed by all the dinoflagellates tested in the present study are high enough to be used as a food source for the shrimps represented here. The EAAIs of A. andersonii, H. rotundata, and A. granifera are greater than that of Chlorella spp. and T. chui (0.84–0.91) (Brown and Jeffrey 1992) which have been used for animal feeds in the aquaculture industry.
The presence or absence of toxic materials can affect the use of target microalga as an animal feed. The strains of A. andersonii and the other five dinoflagellates tested in the present study were revealed not to be toxic to brine shrimp Artemia salina (Kim et al. 2017). Therefore, these dinoflagellate strains in the present study can be used as a food source of shrimps.
To use dinoflagellates commercially as animal feeds, they should be cultivated at a large scale. The methods of cultivating each of A. andersonii, A. granifera, G. smaydae, H. rotundata, T. helix, and T. tasmanica have been developed; the mixotrophic growth rates of A. andersonii, A. granifera, G. smaydae, and T. helix on optimal prey species are considerably greater than their autotrophic growth without prey (Lee et al. 2014b, 2016, Jeong et al. 2016, Lim et al. 2018). Using suitable prey, these mixotrophic dinoflagellates can be easily cultivated. The growth rates of these dinoflagellates may be elevated if they are cultivated in optimal light, temperature, salinity, and nutrient conditions. Thus, it is worthwhile to explore the optimal conditions for each dinoflagellate species.
The dinoflagellates tested in the present study have useful materials other than amino acids; A. granifera has beta-carotene (Jeong et al. 2014), T. tasmanica and T. helix have violaxanthin and / or fucoxanthin (de Salas et al. 2003), and G. smaydae has zeaxanthin (Kang et al. 2014). Astaxanthin, a metabolite of zeaxanthin, is incorporated into the diets of some fish, shrimp, and lobster for muscle color (Gouveia et al. 2008). Therefore, the dinoflagellates tested in the present study could be used for providing useful amino acids with useful pigments to animals in the aquaculture industry.
