Introduction

Artemia nauplii are the most common live prey utilized in cultured marine and freshwater fish species (Sorgeloos et al. 2001). Aquaculture hatcheries require live feeds similar to natural live preys larger than 1 mm, such as copepods, for further larval fish feeding. In this point of view, Artemia franciscana metanauplii are of appropriate size for crustacean (Nelson et al. 2002), fish (Lim et al. 2001), seahorse (Souza-Santos et al. 2013; Vite-Garcia et al. 2014), clownfish (Chen et al. 2020), anemones (Leal et al. 2012), soft corals (Petersen et al. 2008; Tsounis et al. 2010), and cephalopods culture (Okumura et al. 2005; Seixas et al. 2008; Guinot et al. 2013b). Therefore, juvenile or adult forms of Artemia sp. are also largely in demand in the new candidate marine cultured species. Artemia sp. is a filter-feeder fast-growing brine shrimp, which mostly consumes various small-sized particles such as bacteria (Toi et al. 2013), microalgae (Zhukova et al. 1998), and organic manure in the water column (Maldonado-Montiel et al. 2003). Among them, Artemia sp. mostly consumes various microalgae for efficient growth.

Microalgae are the first step of the marine food chain/web and directly utilized in marine fish larvae tank, which is called the green water technique (Faulk and Holt 2005; Neori 2011). They are also used as alternative feed ingredients replacing fish oil and fishmeal (Miller et al. 2007; Ganuza et al. 2008; Eryalçın et al. 2013, 2015), filter-feeder bivalve (Arney et al. 2015), rotifer (Ma and Qin 2014; Rehberg-Haas et al. 2015), Artemia sp. (Lora-Vilchis et al. 2004; Makridis et al. 2006), and copepod feeding (Rasdi et al. 2016). Different forms of microalgae biomass such as spray-dried, freeze-dried, and freshly cultured microalgae have been studied for Artemia sp. nutrition (Thinh et al. 1999; Olsen et al. 2000; Lora-Vilchis et al. 2004). Additionally, bacteria, yeast, and other suspended organic compounds and commercial feeds have been also investigated (Marques et al. 2004; Toi et al. 2014; Ahmadi et al. 2017; Planas et al. 2017; Vahdat et al. 2018). Ingestion, digestion, and chemical conversion of essential compounds of feeds are also main issues in Artemia sp. grow-out stages (Glencross et al. 2007). Based on the recent findings, the best mono-algal diet, microalgae Rhodomonas lens, was suggested for Artemia sp. (Seixas et al. 2009). However, combination of different microalgae has also been recommended for Artemia sp. nutrition in order to sustain higher growth and better nutritional quality (Pacheco-Vega et al. 2015; El-Kassas et al. 2016).

The successful culture of marine fish larvae mainly depends on high-quality live prey containing appropriate levels of essential nutrients such as highly unsaturated fatty acids (HUFA), amino acids, vitamins, and minerals. On the other hand, the bacterial load of the live feed is a concern in fish larviculture and the “content” of this load has been found to be highly correlated with the feed consumed by live feed (Hache and Plante 2011; Avila-Villa et al. 2011). The live feeds are known to have numerous pathogenic species in their microbiota and they may affect the health of the larvae by being vectors of viruses and bacteria (Skjermo and Vadstein 1993; Dhert et al. 2001; Da Silva et al. 2015). Olsen et al. (1999) found that the percentages of hemolytic bacteria and Vibrios in the intestines of halibut larvae fed with 2-day-old A. franciscana were significantly higher than those fed only short-term enriched A. franciscana nauplii.

Rearing water of marine fish larvae includes high nutrient salts and organic wastes together with feed added into the rearing water column and this leads to high proliferation of bacterial flora (Zarkasi et al. 2016). Therefore, live feeds should be also evaluated for bacterial community before given into the larvae culture tanks (Planas and Cunha 1999). Microalgae may alter the bacterial microbiota of the environment they exist in, particularly because of their ability to inhibit bacterial proliferation by secreting certain antibacterial and antiviral chemicals. Mostly, small-sized microalgae (2–6 μm) such as Nannochloropsis sp. are preferred for this purpose (Muller-Feuga 2000; Spolaore et al. 2006).

Biochemical composition of Artemia sp. is directly affected by consumed microalgae diets (Fabregas et al. 2001), and microalgae species are known to significantly differ in their specific biocompounds such as different essential fatty acids (EFA). Most importantly, crustaceans and other invertebrates are not capable of synthesizing n-3 and n-6 polyunsaturated fatty acids (PUFA) efficiently. For that reason, enrichment process or feeding with high-quality feeds is desired. The four microalgae species A. viridis, D. salina, C. vulgaris, and C. reinhardtii have not been evaluated for Artemia franciscana nauplii feeding for long-term grow-out culture so far. The aim of this study was to determine the effect of feeding with these two marine and two freshwater microalgae alone or in combination on survival, growth, and fatty acid composition of A. franciscana metanauplii and on predominant bacterial species of the rearing water.

Materials and methods

Microalgae culture

Two marine microalgae A. viridis (strain number: CCAP1001/1) and D. salina (strain number: CCAP19/12) and two freshwater microalgae C. vulgaris (strain number: CCAP211/12) and C. reinhardtii (strain number: CCAP 11/32A) were obtained from CCAP (Culture Collection Algae and Protozoa). For marine microalgae, artificial seawater was prepared as 32 ppt in a polyethylene conic bottom tank with the following inside dimensions: 50 cm × 100 cm × 80 cm by addition of artificial sea salt (Instant Ocean®) with tap water. Prepared seawater was chemically sterilized. For that purpose, 3 mL L−1 sodium hypochlorite was added to the total tank volume and aerated for 24 h. After 24 h, neutralization was carried out by adding 150 mg L−1 sodium thiosulfate (Merck, Germany) and then aerated for an additional 24 h to the desired seawater pH. Marine microalgae A. viridis and D. salina were cultured in medium f/2+Si, f/2 (Guillard and Ryther 1962), and freshwater microalgae C. vulgaris and C. reinhardtii were cultured in 3N-BBM+V medium. All sub-cultures were maintained at 27 °C and salinity of 32 ppt under an initially 12L:12D photoperiod. Microalgae culture volumes were increased from 50-mL test tubes to 250-mL Erlenmeyer flasks. The contents of 250-mL flasks were transferred to 1-L, 5-L, and then 30-L volume culture vessels. The population growth of microalgae was determined by using Neubauer hemocytometer counter daily. Inoculation was performed by reaching population to late log-phase into the larger volume.

Experimental design and diets

A. franciscana nauplii were obtained from the hatching of Great Salt Lake A. franciscana cysts (INVE Aquaculture Nutrition, Dendermonde, Belgium). Cysts were decapsulated as described by Bruggeman et al. (1980) and then incubated for 24 h in a 5-L cylindrical conical glass tank containing seawater (38 g L−1 salinity) at 28°C. Gentle aeration was placed in the bottom of the glass tank with continuous illumination. After hatching, instar I nauplii were moved into cylindrical containers at 23 °C and rinsed with UV-treated seawater for 20 min. Initial A. franciscana stock density was 60 nauplii mL−1 in all experimental feeding culture tanks. A. franciscana nauplii groups were fed every day selected microalgae during 15 days of the experiment. Microalgae diets were (a) A. viridis (AV) (b) D. salina (DS), (c) C. vulgaris (CV), (d) C. reinhardtii (CR), and (e) mix (MX) (1:1:1:1) due to the high nutritional values. Concentration of microalgal density was 107 cell/mL−1 for 106 A. franciscana constant during the culture period. Each group was tested in triplicate. Water exchange renewal rate in all vessels was adjusted as 10% of the total volume per day.

Growth performance and sampling

Dry weight was determined by sampling 5 mL of microalgal culture filtered on previously weighed pre-combusted Whatman GF/C (Maidstone, England; diameter 47 mm) fiberglass filters. Microalgal biomass was washed three times with 5 mL of 0.5-M ammonium formate in order to remove salts (Zhu and Lee 1997). Filters were dried overnight at 80 °C and dry weight was determined gravimetrically. Microalgal biomass (25–50 mL) were sampled by centrifugation and immediately frozen at −80°C. One hundred fifty milliliters of A. franciscana culture was filtered on a sieve (45-μm mesh size), rinsed with distilled water, and frozen at −80 °C for proximate and fatty acid composition at the end of the experiment. During the experiment, ten A. franciscana from each culture vessel were sampled in 5% formalin solution every 3 days. A. franciscana growth was determined by measuring the total length of 10 A. franciscana at 3, 6, 9, 12, and 15 days after hatching using an Olympus BX-51 microscope. At the end of the experiment, survival rate was determined by counting all live A. franciscana in each culture vessel.

Proximate composition and fatty acid analysis

Moisture, protein, and crude lipid contents of microalgae and A. franciscana were analyzed according to Folch et al. (1957) and AOAC (1990) (Table 1, Table 3). Fatty acid methyl esters in microalgae, diets, and experimental fish were obtained by transmethylation with 1% sulfuric acid in methanol (Christie 1989). Fatty acid methyl esters were separated by GC (GC-14A; Shimadzu, Tokyo, Japan) in a Supercolvax-10 fused silica capillary column (constant pressure with 100KPa, length: 30 m; internal diameter: 0.32 mm; 0.25 i.d (Ref.: 24080-U) Supelco, Bellefonte, PA, USA) using helium as a carrier gas. Column temperature was 180 °C for the first 10 min, increasing to 220 °C at a rate of 2 °C min−1 and then held at 220 °C for 15 min. Fatty acid methyl esters were quantified by FID following the conditions described to Izquierdo et al. (1990) and identified by comparison with external standards and well-characterized fish oils (EPA 28, Nippai, Ltd Tokyo, Japan).

Table 1 Proximate composition of experimental microalgae (% dw)
Full size table

Determination of the heterotrophic plate count and identification of the predominant bacterial species in the rearing water

The rearing water of A. franciscana was sampled three times during the experiment (day 0, 2, and 15), and these samples were inoculated onto Marine Agar 2216 and TCBS agar to determine the heterotrophic plate counts (HPC) and total Vibrio sp. counts (Austin 1988; Makridis et al. 2006). The morphological and physiological characteristics of predominant bacterial colonies from each plate were determined by using conventional biochemical and physiological tests and rapid identification kits (API 20E) according to Buller (2004) and Austin and Austin (2007). For molecular identification, the isolates displaying the same morphological and biochemical characteristics were selected and inoculated into Marine Broth 2216 (Difco) and incubated overnight at 22 °C in a shaking incubator. Genomic DNA was then extracted by the PureLink Genomic DNA Mini Kit (Invitrogen) according to the manufacturer’s instructions and used as a template for PCR. An approximately 540-bp-long fragment of the 16S rRNA gene was amplified using the universal bacteria primer set; primer S-20 (5’ AGA GTT TGA TCC TGG CTC AG 3’) and primer A-18 (5’ GWA TTA CCG CGG CKG CTG 3’) (Suau et al. 1999). Amplification was performed using a thermal cycler (Biometra, TPersonal) with the following parameters: initial denaturation at 95 °C for 3 min, followed by 30 cycles of amplification (denaturation at 95 °C for 30 s, annealing at 56 °C for 1 min, extension at 72 °C for 1 min) and a final extension step of 72 °C for 4 min. PCR products were purified and sequenced bidirectionally and sequence editing and analysis were performed in Bioedit v7.0.0 (Hall 1999) using the ClustalX 2.1 (Larkin et al. 2007) and BLASTN 2.2.20 algorithm (Zhang et al. 2000).

Statistical analysis

All data were tested for normality with the Shapiro-Wilk test and homogeneity of variances by Levene’s test and treated using one-way ANOVA. Means were compared by the two-tailed Student’s t test (P < 0.05) and Duncan’s multiple range test (DMRT) using SPSS software (SPSS for Windows 11.5; SPSS Inc., Chicago, IL, USA). Duncan’s multiple range test (DMRT) is a post-hoc test to measure differences between pairs of means (which are shown as a, b, c, d, etc. in relevant figures below). In this test, whilst the same letters mean no significant difference, different letters indicate significant differences among pairs of means. Values presented in the figures and tables are means ± standard deviations SD. All values presented as percentages were arc cosine transformed before performing any statistical test.

Results

Growth performance of Artemia franciscana

In our study, A. franciscana fed CR (1760±63μm), MX (1725±25μm), and AV (1650±36μm) diet showed significantly higher total body length than A. franciscana fed DS (1196±38μm) and CV (1310±32μm) at the 12th day of the experiment (P < 0.05) (Fig. 1). However, feeding A. franciscana nauplii with a CR diet provided significantly higher body length (2520±45μm) among other treatments at the end of the experiment (P < 0.05).

Fig. 1
figure 1

Total length of A. franciscana groups fed experimental microalgae (CV, C. vulgaris; CR, C. reinhardtii; DS, D. salina; AV, A. viridis; MX, mix. Lower case letters (a, b, c) denote significant variable differences between groups using Duncan’s multiple range test, P < 0.05)

Full size image

Survival of Artemia franciscana

Final survival was found similar in CV, AV, and MX groups and significantly higher than DS and CR groups (P < 0.05). A. franciscana fed DS diet showed a moderate level of survival, whereas the lowest survival rate was observed in A. franciscana fed CR diet (Fig. 2).

Fig. 2
figure 2

Survival of A. franciscana groups fed experimental microalgae diet at the end of the experiment (CV, C. vulgaris; CR, C. reinhardtii; DS, D. salina; AV, A. viridis; MX, mix. Lower case letters (a, b, c) denote significant variable differences between groups using Duncan’s multiple range test, P < 0.05)

Full size image

Proximate composition and fatty acid profile of Artemia franciscana metanauplii

CR and MX microalgae diets contained higher crude lipid levels than other diets. CR diet contained higher crude protein level among experimental microalgae (Table 1). Fatty acid profiles of microalgae are shown in Table 2

Table 2 The fatty acid compositions of cultured microalgae (% total fatty acids)
Full size table

At the end of the study, crude protein content of A. franciscana metanauplii was found similar in CR and MX groups and these two groups were significantly higher than other groups (P < 0.05). Similarly, lipid and ash contents of the A. franciscana fed MX diet showed significantly higher values than those of other groups (P < 0.05) (Table 3).

Table 3 Proximate composition of Artemia franciscana metanauplii at the end of the experiment (% dw)
Full size table

A. franciscana fed CR diet presented significantly (P < 0.05) higher oleic acid (OA; 18:1n-9) content than other A. franciscana groups. Linoleic acid (LA; 18:2n-6) was found also higher in A. franciscana fed CR diet. α-linoleic acid (ALA; 18:3n-3) level was increased by utilization of MX diet (P < 0.05). Arachidonic acid (ARA; 20:4n-6) level was found significantly higher in both DS and AV groups (P < 0.05). Additionally, Artemia fed AV diet showed a significantly higher level of docosahexaenoic acid (DHA; 22:6n-3) (P < 0.05). Eicosapentaenoic acid (EPA; 20:5n-3) level was enhanced by utilization of DS diet. Total n-3 content was higher in the MX group, whereas total n-3 HUFA levels were found significantly (P < 0.05) higher in A. franciscana fed DS and AV diet (Table 4).

Table 4 The fatty acid compositions of Artemia after 15 days of culture (% total fatty acids)
Full size table

HPC, total Vibrio count, and predominant bacterial species

Bacterial load of the rearing water of A. franciscana changed significantly during the experiment (Fig. 3). HPC and Vibrio sp. counts decreased gradually in all rearing water of A. franciscana during the experiment regardless of the algal species used. However, at the end of the experiment, we found that rearing water containing D. salina, C. reinhardtii, and A. viridis and MX groups had the highest number of HPC (Fig. 3A). The rearing water containing MX diet showed the highest Vibrio sp. counts at the beginning of the experiment, whereas C. vulgaris had the lowest number of both counts among the algal species on day 2 and at the end of the experiment (Fig. 3B). A total of 40 bacterial isolates were obtained from the rearing water samples. According to both conventional identification and 16S rRNA gene sequencing results, four different bacterial species were found predominantly in the rearing water of A. franciscana including Carnobacterium maltaromaticum (IUET-ME1), Pseudomonas stutzeri (IUET-ME2), Pseudoalteromonas sp. (IUET-ME3), and Vibrio sp. (IUET-ME4) (accession numbers: MF503245-48) (Table 5; Fig. 4).

Fig. 3
figure 3

A Total aerobic mesophilic bacteria of A. franciscana culture water and B total Vibrio sp. count of A. franciscana culture water (CV, C. vulgaris; CR, C. reinhardtii; DS, D. salina; AV, A. viridis; MX: mix. Lower case letters (a, b) denote significant variable differences between groups using Duncan’s multiple range test, P < 0.05)

Full size image
Table 5 Morphological and phenotypic characteristics of the strains isolates
Full size table
Fig. 4
figure 4

Agarose gel electrophoresis image of the PCR products. Lane 1: (L) GoldBio 100 bp PLUS™ DNA Ladder. Lanes 2–6: Partially amplified 16S rRNA gene of the isolates found predominantly in the rearing water

Full size image

Discussion

Microalgae is the first step of the marine and freshwater food chain/web (Spolaore et al. 2006; Hemaiswarya et al. 2011). Several microalgae diets have been evaluated for rotifer (Rehberg-Haas et al. 2015), Artemia sp. (Pacheco-Vega et al. 2015), Daphnia sp. (Taipale et al. 2011), and copepods (Rasdi et al. 2016) in order to improve growth performance and nutritional value of cultured species. In this study, two marine and two freshwater microalgae alone or in combination were used for long-term feeding of A. franciscana. Nutritional composition of microalgae can directly reflect to live prey. Vismara et al. (2003) showed Artemia sp. fed Euglena gracilis contained high α-tocopherol levels which is a very important antioxidant source for larval fish (Betancor et al. 2011). In another study, Sánchez-Saavedra and Paniagua-Chávez (2017) used refrigerated cyanobacteria Synechococcus elongatus and compared it to fresh microalgae Chaetoceros muelleri for A. franciscana culture and refrigerated S. elongatus performed similarly as fresh microalgae maintaining growth and proximate composition. This result shows that frozen microalgae can also be efficiently utilized for Artemia sp. feeding and enhancing growth performance in advanced.

Different microalgae such as Isochrysis sp. and Chaetoceros muelleri have been tested for A. franciscana, and Chaetoceros muelleri has been selected as a better microalgae diet due to the high EFA conversion rate and EPA content (Lora-Vilchis et al. 2004). More recently, endemic microalgae species Grammatophora sp., Rhandonema sp., Navicula sp., and Schizochytrium sp. have been selected as potential food for A. franciscana and performed better than commonly used microalgae Chaetoceros muelleri (Pacheco-Vega et al. 2015). This result was also similar to other combination of different microalgae for Artemia sp. grow-out experiments (El-Kassas et al. 2016).

Concentration, ingestion, and assimilation rates of microalgae in crustacean feeding have vital importance (Evjemo and Olsen 1999). Artemia sp. growth is related to the size of the ingested amount of feeds and these diets should be at a certain size (Chaoruangrit et al. 2018). Suitable size of microalgae is reported between 7 and 28 μm, and optimum size is described as 16 μm for Artemia sp. feeding. A. viridis is a diatom with a diameter of 14.7 μm, and its beneficial effects have been reported in the Haliotis asinina metamorphosis (Jing Ding et al. 2017). In another trial, fairy shrimp (B. thailandensis) fed Chlorella vulgaris was recorded as 19.13 mm in total length for 15 days of culture period and lower concentration of microalgae lead to reduce growth (Chaoruangrit et al. 2018). Moreover, not only size but also nutritional properties of microalgae play an important role in growth (Fabregas et al. 1996). In our study, the highest total length of A. franciscana was achieved by utilization of C. reinhardtii (CR) which resulted in a 25.20-mm total length of A. franciscana. This result could be related to the CR diet containing high protein level (see Table 1). Additionally, these psychrophilic microalgae can contain a higher amount of polyunsaturated fatty acids (Giroud et al. 1988; Siaut et al. 2011; Kim et al. 2016). Nutritional quality of Artemia sp. is also related to the application time of the feeds or enrichers. For instance, short-time enrichment with Chlorella marina supported the growth and nutritional value of Artemia sp. However, it has been reported that combination of different Chlorella species could be better for long-time enrichment (El-Kassas et al. 2016).

Protein level of A. franciscana (five days old) fed Tetraselmis suecica was recorded as 63.6% and A. franciscana fed Rhodomonas lens was 67.7% of the dry weight of 5-day-old A. franciscana (Seixas et al. 2009). Our protein values were lower than those values; one of the reasons could be the age of the A. franciscana culture. Lipid levels of A. franciscana fed Tetraselmis suecica and Rhodomonas lens were recorded as 15.2% and 12.3%, respectively (Seixas et al. 2009). However, in comparison to the lipid levels in our study, A. franciscana fed MX diet showed the highest lipid value (18.71%) than that obtained by Seixas et al. (2009). In the same study, protein and lipid levels were found similar to that obtained by Artemia fed Chaetoceros muelleri comparing to A. franciscana fed 8 weeks of refrigerated cyanobacteria Synechococcus elongatus (Sánchez-Saavedra and Paniagua-Chávez 2017). In our study, protein levels of A. franciscana fed experimental diets ranged between 53.3 and 58.45% and these values were found higher than those of Artemia sp. fed refrigerated Synechococcus elongatus and Chaetoceros muelleri by Sánchez-Saavedra and Paniagua-Chávez (2017). Enhancement of protein value can be attributed to the combined utilization of microalgae. Chlorella species and C. reinhardtii have a thick cell wall, and this feature results in low digestibility of these microalgae (Dhont and Lavens 1996). On the other hand, these two species contain highly active biocompounds such as lutein, α-carotene, ß-carotene, ascorbic acid, and α-tocopherol which are valuable biocompounds for lipid preservation and utilization (Hashimoto et al. 1982; Shibata et al. 2001, 2003). Artemia is strongly recommended to be enriched with astaxanthine and lutein (Hemaiswarya et al. 2011). In this study, one of the highest survivals was found in the Artemia group fed C. vulgaris. This could be related to the antioxidant effect of these microalgae (see Fig. 2). Additionally, recent studies showed that C. vulgaris extracts can prevent lipid peroxidation induced by some toxic effects (Vijayavel et al. 2007). Similar survival level was maintained by Chlorococcum humicola in fairy shrimp (B. thailandensis) which resulted in 70% rate of survival.

Marine fish larvae require sufficient amount of essential fatty acids for high growth and survival (Izquierdo and Koven 2010). These fatty acids should be delivered by rotifer, Artemia sp. and/or copepods to fish larvae. Artemia sp. is known to be able to incorporate and concentrate some dietary lipids. However, they are limited to elongate and desaturate 18C to 22C fatty acids and deposition and utilization of fatty acids. Feeding of Nannochloropsis oculata and Phaedactylum tricornatum have resulted in low levels of EPA Artemia sp. fed Nannochloropsis oculata and Phaedactylum tricornatum (Zhukova et al. 1998). Recent studies showed that on-growing Artemia sp. prefers DHA being preferentially transferred from dietary phospholipids to triacylglycerols (Guinot et al. 2013a). Therefore, after feeding process with qualified diets, HUFA enrichment should be applied at the metanauplii stage. In our trial, experimental microalgae were grown under phototrophic conditions, which lead to lower long-chain fatty acid conversion and accumulation in their body cell, whereas A. viridis enhanced higher ARA, DHA, and Σn-3 HUFA in A. franciscana at the end of the experiment. Amphora sp. can contain high amount of lipid deposition under stress conditions (Chtourou et al. 2015), and AV has been reported to have the highest EFA level among other diatoms (Huang et al. 2007a; Huang et al. 2007b). This feature of AV supported the post-larval growth of Haliotis asinina (Jing Ding et al. 2017). Moreover, AV is also used as a feed supplement for Penaeus monodon post larvae production (Khatoon et al. 2009). Monospecific diets may result in nutritional deficiencies due to the inadequate nutritional content of several essential compounds. Combination of more than one microalga could meet the nutritional requirements of the live prey (Brown et al. 1989; Smith et al. 1992). Similar positive results were observed in A. franciscana fed MX diet in our study. Recent studies on copepods also showed combination of different microalgae could be the optimum diet for survival and growth (Milione et al. 2007).

Utilization of feeds affects distinctively bacterial community of rearing water and gut of live feeds. In order to grow out Artemia sp. nauplii, several commercial algae powders (Claus et al. 1979; Griffiths et al. 2016) and fresh culture of microalgae have been studied so far (Hemaiswarya et al. 2011). Especially, in algae-limited conditions, bacterial microbiota may significantly contribute as a dietary source for Artemia sp. production (Toi et al. 2013). According to this study, higher survival of A. franciscana can be enhanced by utilization of C. vulgaris, A. viridis, and combination of four microalgae. A. franciscana nutritional requirement may change in further development stages. It is possible that this may be one of the reasons why the combination of the four microalgae diets works best in terms of growth. This study suggests the combined utilization of marine and freshwater microalgae species could promote survival and total length of A. franciscana.

Since many studies have shown that there are many pathogenic species in the microbiota of live feeds, various algal species including Skeletonema sp., Chaetoceros sp., Nannochloropsis sp., Isochrysis sp., Pavlova sp., Tetraselmis sp., and Chlorella sp. have been investigated to reduce the bacterial load of Artemia sp. or to decontaminate these feeds and successful results have been reported in various degrees (Salvesen et al. 2000; Tolomei et al. 2004; Makridis et al. 2006; Interaminense et al. 2014). Among these, the closest study to our work was made by Makridis et al. (2006) by using another Chlorella species (Chlorella minutissima) and they found this species was very successful in reducing both the bacterial load and Vibrio sp. counts as we determined in our study with C. vulgaris. We think that this microalgae species caused the highest decrease in the number of bacterial counts due to having strong antimicrobial properties (Falaise et al. 2016). It was also found that the use of different algal species might play a role in altering the total number of Vibrio sp. members that are known to contain disease-causing strains in marine fish. The predominant species Vibrio sp., Pseudomonas sp., and Pseudoalteromonas sp. identified in Artemia sp. rearing water are compatible with some other studies (Hameed and Balasubramanian 2000; McIntosh et al. 2008), but we also identified Carnobacterium maltaromaticum as lactic acid bacteria (LAB). Carnobacterium species are commonly found in various environments including humans, healthy and diseased fish, seafood, meat, and poultry and dairy products both in fresh and vacuum-packed products and in probiotics (Laursen et al. 2005; Afzal et al. 2010).

Future studies should be focused on different forms of microalgae biomass such as spray-dried or lyophilized microalgae powder for Artemia sp. grow-out feeding. Additionally, the current knowledge about the microbiota of the live feeds used in aquaculture is very limited and in order to detect microorganisms introduced in the aquaculture systems by live feeds and to better understand the effects of these microorganisms on cultured species, community profiling must be done by the latest and more powerful methodologies such as next-generation sequencing in future works.