Uniform design for optimizing biomass and intracellular polysaccharide production from self-flocculating Scenedesmus sp.-BH

Abstract

Four microalgae species were isolated from freshwaters in Xinjiang, one of which was identified as Scenedesmus sp.-BH. Analysis of the specimens showed not only the characterization of self-flocculation with a flocculating rate of 92.29 % in 12 h, but relatively high biomass and intracellular polysaccharide (IPS) production (250.5 and 5.55 mg/L) under original BG11 medium. Uniform design was applied to optimize the nutritional conditions of Scenedesmus sp.-BH. The maximum biomass yield of 556.7 ± 16.97 mg/L appeared at NaNO₃ 0.77 g/L, K₂HPO₄·3H₂O 0.24 g/L, MgSO₄·7H₂O 0.24 g/L, ferric ammonium citrate 0.021 g/L, and NaHCO₃ 3.85 g/L, while a maximum IPS yield of 14.3 ± 0.46 mg/L appeared at NaNO₃ 0.004 g/L, K₂HPO₄·3H₂O 0.24 g/L, MgSO₄·7H₂O 0.21 g/L, ferric ammonium citrate 0.024 g/L, and NaHCO₃ 1.42 g/L. Both biomass and IPS production were over two-fold of those cultured under the original condition, which demonstrated that uniform design could successfully improve the production of biomass and IPS.

Introduction

The Chlorophyta lives in the trunk status of the Plantae phylogenetic, but so far, studies of algal sulfated polysaccharides have usually focused on red and brown algae, while the interest in green algae has been relatively small. Green microalgae are important sources of sulfated polysaccharides. These polysaccharides containing hemi-ester sulfate groups in their sugar residues possess different therapeutic properties (Patel 2012), including strong anticoagulant (Shanmugam and Mody 2000; Mao et al. 2008), antioxidant (Costa et al. 2011; Ngo et al. 2011), and antimicrobial (Amaro et al. 2011) traits, and so on.

Several reports (Beardall et al. 2001; Giordano et al. 2002; Heraud et al. 2005; Sun and Wang 2009 Liu et al. 2010) previously revealed that microalgal cells showed dramatic reorganization of macromolecular composition in response to these changes in nutrient status. Optimizing nutritional conditions is essential to the production of microalgal polysaccharides, for example, sulfated polysaccharides for commercial mass production. In the past decade, considerable progress has been made toward understanding the effects of microalgae in response to nutritional conditions, especially carbon, nitrogen, phosphorus, and metal ion sources. Apart from molecular CO₂, bicarbonate ion was the preferred source of carbon for photosynthesis in certain marine algae (Nimer et al. 2008). Evidence by ÖSterlind (1951) reports that bicarbonate ion absorption is an active process in Scenedesmus quadricauda under certain conditions. The effects of nitrogen limitation that trigger the accumulation of polysaccharides were well documented (Sun et al. 2010; Wang et al. 2011; Yang et al. 2012) and generally, nitrate has been widely used as a form of inorganic nitrogen. The availability of bioactive trace metals such as magnesium and iron are required for biochemical processes, where they serve as cofactors for many enzymes and are crucial for catalytic activities (van Oijen et al. 2004). In addition, they affect the acquisition and assimilation of macronutrients like C, N, and P (McKay et al. 2001).

The optimization of cultivation parameters, particularly nutritional and environmental factors, is very important. In recent years, due to economic conditions and the limitations of experimental design, it has become necessary to advance the application of uniform design, an effective and powerful approach for rapidly screening key factors from a multivariable system in order to optimize culture conditions (Wang et al. 2007; Sun et al. 2010; Hu et al. 2012).

The harvest is also an essential process in the mass production of microalgae. To decrease costs involved during the biomass harvesting step,the inherent characteristics and natural behaviors of microalgae should not be underestimated. Flocculation can be induced in different ways, including the use of Zn²⁺, Fe³⁺, Al³⁺, other chemical flocculation and extreme pH, nutrient depletion, temperature changes, and so on (Salim et al. 2011). However, it should be stressed that any induced situation during culture growth to improve microalgae harvesting may also induce undesired changes in cell composition (González-Fernández and Ballesteros 2012). Thus, microalgal self-flocculation has attracted the attention of many researchers (Salim et al. 2011; Salim et al. 2012).

In the current study, one green microalga was identified among the isolates from Xinjiang freshwaters as Scenedesmus sp.-BH, which not only had the capacity for self-flocculation, but also higher biomass and IPS production under the original BG11 medium. Therefore, Scenedesmus sp.-BH with self-flocculation was chosen to optimize the nutritional conditions for maximizing biomass and IPS production with uniform design.

Materials and methods

Microorganisms and culture conditions

Microalgae used in this experiment were isolated from freshwaters in Xinjiang, China, using the spread plate method. They are preserved in the Key Laboratory for Green Processing of Chemical Engineering of Xinjiang Bingtuan. The basal culture medium BG11 consisted of the following components: 1.5 g/L of NaNO₃, 0.0524 g/L of K₂HPO₄·3H₂O, 0.075 g/L of MgSO₄·7H₂O, 0.036 g/L of CaCl₂·H₂O, 0.006 g/L of citric acid, 0.006 g/L of ferric ammonium citrate, 0.001 g/L of EDTANa₂, 0.02 g/L of Na₂CO₃, 1.0 g/L NaHCO₃, and 1 mL/dL of trace elements solution. The trace elements solution contained the following components: 2.86 g/L of H₃BO₃, 1.86 g/L of MnCl₄·H₂O, 0.22 g/L of ZnSO₄·7H₂O, 0.39 g/L of NaMoO₄·2H₂O, 0.08 g/L of CuSO₄·5H₂O, and 0.05 g/L of Co (NO₃)₂·6H₂O. The pH was adjusted to neutral by adding 1 mol/L HCl or NaOH. Cultures were incubated in 2-L Erlenmeyer flasks containing 1.5 L of culture medium at 24 ± 2 °C under 4,000 lux continuous illumination. Algal stock culture with the absorbance value of 0.19 ± 0.01 was inoculated to the medium to give a 10 % (v/v) concentration. Cultures were shaken by hand two or three times daily to avoid sticking.

Identification of microalga

Nucleic acid extraction of algal cells was performed using the NuClean PlantGen DNA kit (Beijing ComWin Biotech Co., Ltd., China) according to the manufacturer’s instructions. The 18S rDNA was amplified by polymerase chain reaction (PCR) using the universal Primer A (5′-GTCAGAGGTGAAATTCTTGGATTTA-3′) and Primer B (5′-AGGGCAGGGACGTAATCAACG-3′) (Rasoul-Amini et al. 2009), and the ampliconic sequence was sent for analysis. The sequence obtained for the 18S rDNA gene was aligned to published sequences obtained from GenBank using Clustal X 1.83. A neighbor-joining tree was constructed from these data by the bootstrap method (1,000 replicates) using Mega 5.0 software.

Flocculating activity test

Samples of microalgal suspensions cultured in the basal BG11 medium were placed in an Erlenmeyer flask and the suspension was left to settle. During the settling period, an aliquot of the culture was withdrawn and then the optical density (OD) was measured at 680 nm at the same height in the Erlenmeyer flask to determine the flocculating rate. Similar to conventional tests, the flocculating rate was measured in the top part of the 2-L Erlenmeyer flask, where individual cells and formed flocs independently sink (Salim et al. 2011). The flocculating activity was expressed in the form of the flocculation rate, as calculated according to Eq. (1):

$$ \mathrm{Flocculating}\;\mathrm{rate}\left(\%\right)=\frac{{\mathrm{OD}}_{680}(t0)\hbox{-} {\mathrm{OD}}_{680}(t)}{{\mathrm{OD}}_{680}(t0)}\times 100, $$

(1)

where OD₆₈₀ (t ₀ ) is the turbidity of sample taken at time zero and OD₆₈₀ (t) is the turbidity of the sample taken at time t.

Growth analysis

Cultures were harvested by centrifugation at 5,000 rpm for 10 min and microalgal pellets were washed twice with distilled water and freeze-dried. The dry weight of microalgal biomass was determined gravimetrically and growth was determined regularly by measuring optical density at wavelength 680 nm (denoted as OD_680nm).

Determination of IPS production

After harvest, freeze-dried cells were analyzed for the determination of IPS production. The microalgal dried power was extracted with 0.1 mol/L sodium hydroxide solution at 90 °C for 2 h. After centrifugation, the supernatant was added at three-fold the volume of 95 % ethanol, and the mixture was kept for 16 h at 4 °C. The precipitate was dissolved in distilled water, then the solution was subjected to the Anthrone-sulfuric acid test with glucose as the standard (Yemm and Willis 1954). The IPS production was calculated using the following Eq. (2):

$$ \mathrm{IPS}\;\mathrm{production}\;\left(\mathrm{mg}/\mathrm{L}\right)=\left(M\times C\times V\times 100\right)/W, $$

(2)

where M is the biomass concentration of microalgae, mg/L; C is the concentration of crude polysaccharides calculated from the calibrated regression equation based on the absorbance at 620 nm, mg/mL; V is the diluted volume of the extraction solution, mL; and W is the mass of cultured microalgae, mg.

Uniform design

The experiment for the optimization of biomass and IPS production of Scenedesmus sp.-BH was arranged with five factors, each with twelve levels. The uniform design (UD) table U ₁₂ ^*(12¹⁰) was applied to arrange the experiments (Table 1), and each trial was performed in triplicate. The evaluated responses Y₁ and Y₂ were the content of biomass and IPS, respectively.

Table 1 Factors and levels of the uniform design experiment

Statistical analysis

All trials were carried out in triplicate, and the average of the biomass and IPS production was taken as the response value. Statistical and stepwise regression analyses of the data were done using data processing system (DPS) software (Version 7.55 by Hangzhou Refine Information Tech. Co., Ltd, China).

Results and discussion

Screening of self-flocculating microalga

Characteristics of four microalgae

Four microalgae were isolated and purified from freshwaters in Northern Xinjiang, China. According to morphological characteristics, including cell size, shape, and chloroplast core protein number, they were named as Scenedesmus sp.-BH, Schroederia sp.-90, Chlorella sp.-YL and Microcystis sp.-W, respectively. Micrographs of the four microalgae are shown in Fig. 1. All cultures were unialgal and conducted in the basal BG11 medium.

Fig. 1

Micrographs of four microalgae isolated from the freshwaters in Northern Xinjiang. (a) Scenedesmus sp.-BH, (b) Schroederia sp.-90, (c) Chlorella sp.-YL, (d) Microcystis sp.-W. Photomicrographs were obtained with a Primo Star microscope (*400), Carl Zeiss, Germany

To choose a microalga having relatively high biomass and IPS production, the biomass and IPS production of the four microalgae were investigated. Growth curves of the four microalgae monitored by the absorbance of microalgae suspension are shown in Fig. 2(a). We were only able to reveal the growth cycle in order to determine the harvest time of microalgae. As shown in Fig. 2(b), the maximum biomass and IPS production of Scenedesmus sp.-BH were approximately 250.5 and 5.55 mg/L more than the other three microalgae. Moreover, almost all the cells of this microalga formed small aggregates and settled at the bottom of the flask over time in the microalgal cultivation process. Therefore, tests of the flocculating rate from Scenedesmus sp.-BH were designed.

Fig. 2

Growth curves, maximum biomass (mg/L), and IPS production (mg/L) of the four green microalgae, Chlorella sp.-YL, Scenedesmus sp.-BH, Schroederia sp.-90, and Microcystis sp.-W. The culture medium was basal BG11 at 25 ± 2 °C under 4,000 lux continuous illuminations. (a) represents the growth curves of four microalgae, (b) represents the maximum biomass (mg/L) and IPS production (mg/L) of four microalgae

Identification of microalga

Using PCR amplification and subsequent DNA sequencing, we determined almost the entire length of the 18S rDNA of the microalga. The length of the 18S rDNA sequence of BH in this study was 1,591 base pairs. The 18S rDNA sequence of BH showed 99 % sequence homology with Scenedesmus obliquus (FR865738.1). It was thus named Scenedesmus sp.-BH. As shown in Fig. 3, related sequences were downloaded and used to build a phylogenetic tree with Mega 5.0 software.

Fig. 3

Phylogenic analysis of the isolated Scenedesmus sp.-BH species based on the 18S rDNA sequence

Flocculating rate of Scenedesmus sp.-BH

The sedimentation of microalgal suspensions from Scenedesmus sp.-BH was monitored for 12 h and the absorbance (A_680nm) and flocculating rate of the microalgal cells were determined over time. During the settling period, dispersed microalgal cells aggregated together, formed larger flocs with a higher sedimentation rate, and settled at the bottom of the flask, resulting in an evident separation of cells from the culture broth. The A_680nm and flocculating rate of Scenedesmus sp.-BH over time are presented in Fig. 4. The value of A_680nm dramatically decreased in the initial two hours, and then continued to decrease more slowly. In contrast, the flocculating rate dramatically increased first, and then slowly increased. The flocculating rate and efficiency (per hour) could reach up to 92.29 % and 7.69 % after 12 h standing, respectively. Having found this microalga to have a self-flocculating property, we then studied the potential to improve its biomass and IPS production.

Fig. 4

Variation tendency of A_680nm and flocculating rate (%) over time for the self-flocculating Scenedesmus sp.-BH

The self-flocculating phenomenon of green microalga Scenedesmus sp. has been reported in some previous studies. A new self-flocculating microalga S. obliquus AS-6-1 was characterized by Guo et al. (2013), who then used it as an extracting flocculating agent, which resulted in the fast flocculation of freely suspended cells of S. obliquus FSP-3 and Chlorella vulgaris CNW-11. As mentioned by Salim et al. (2011), although the flocculating microalgae Ankistrodesmus falcatus, Scenedesmus obliquus, and Tetraselmis suecica have the capacity to harvest the non-flocculating Chlorella vulgaris, Chlorella vulgaris, and Neochloris oleoabunda, respectively, the recovery efficiency is different. Salim et al. (2012) found that when adding the self-flocculating microalgae Ettlia texensis, Ankistrodesmus falcatus, and Scenedesmus obliquus to Chlorella vulgaris at a ratio of 0.25, the recovery of C. vulgaris increases from 25 % to, respectively, 40 %, 36 %, and 31 %. However, the flocculation rate of Scenedesmus sp. used in these studies was not calculated separately, and therefore, the results could not be compared with those in the current study.

Optimization of nutrients conditions for Scenedesmus sp.-BH with uniform design

Different microalgal species and strains vary greatly in terms of nutrient requirements (González-Fernández and Ballesteros 2012). There are many nutritional factors controlling the cell growth and IPS content, such as carbon, nitrogen source, and other essential macro- and micronutrients like magnesium and iron. When factors and levels influencing experimental results were greater in number, uniform design—which needs a smaller number of trials compared with orthogonal design—was a better choice. Therefore, uniform design was used for investigating the impact of the dosage of NaNO_3, K₂HPO₄·3H₂O, MgSO₄·7H₂O, ferric ammonium citrate, and NaHCO₃ in basal BG11 medium on the biomass and IPS production from Scenedesmus sp.-BH.

Optimization of biomass production

Values of responses (biomass and IPS production) at different experimental combinations are given in Table 2. There were a total of 12 runs for optimizing five individual parameters in the UD table U₁₂ ^*(12¹⁰), which was applied to the biomass and IPS production. As shown in Fig. 5, the growth tendency and the maximum A_680nm of Scenedesmus sp.-BH are significantly different in different combinations.

Table 2 Application of UD U₁₂ ^*(12¹⁰) for the optimization of biomass (mg/L) and IPS production (mg/L)

Fig. 5

Growth curves of Scenedesmus sp.-BH under different treatments with U ₁₂ ^*(12¹⁰) uniform design

The maximum value of biomass (415.8 mg/L) was recorded under the experimental condition of NaNO₃ 0.60 g/L, K₂HPO₄·3H₂O 0.18 g/L, MgSO₄·7H₂O 0.24 g/L, ferric ammonium citrate 0.002 g/L, and NaHCO₃ 3.15 g/L. By applying quadratic polynomial stepwise regression to the experimental data, Eq. (3) was established as the mathematic model for the optimization of five factors for maximum biomass production.

$$ Ybiomass=102.00-283.55{X}_2{X}_5+{X}_3{X}_5. $$

(3)

The model's adequacy was checked by an F-test and with the determination coefficient R. The analysis of variance on biomass production showed that the model was highly significant (p < 0.001) with an F-value of 27.1. The determination coefficient (R = 0.9261) was given by an analysis of variance (ANOVA) of the quadratic regression model. The value of the adjusted determination coefficient (R = 0.9089) also confirmed that the model was highly significant. Therefore, the model was found to be adequate for prediction within the range of experimental variables.

The results of the regression analysis for biomass listed in Table 3 indicated that the interaction between K₂HPO₄·3H₂O, NaHCO₃, MgSO₄·7H₂O, and NaHCO₃ had a significant influence on biomass (p < 0.01). The model predicted the maximum biomass yield of 579.8 mg/L that appeared at NaNO₃ 0.77 g/L_, K₂HPO₄·3H₂O 0.24 g/L, MgSO₄·7H₂O 0.24 g/L, ferric ammonium citrate 0.021 g/L, and NaHCO₃ 3.85 g/L.

Table 3 Results of the stepwise regression analysis for the optimization of biomass

Optimization of IPS production

The maximum value of IPS production (11.4 mg/L) was recorded under the experimental condition of NaNO₃ 0.90 g/L, K₂HPO₄·3H₂O 0.24 g/L, MgSO₄·7H₂O 0.060 g/L, ferric ammonium citrate 0.020 g/L, and NaHCO₃ 2.8 g/L. By applying quadratic polynomial stepwise regression to the experimental data, Eq. (4) was established as the mathematic model for the optimization of five factors for maximum IPS production.

$$ YIPS\;\mathrm{prosuction}=9.60-2.23{X}_1-4105.97{X}_4^2-5.69{X}_1{X}_2+1277.25{X}_2{X}_4. $$

(4)

The goodness of fit of the model was also examined using the determination coefficient (R = 0.9688), which implied that the sample variation of more than 96 % was attributed to the variables. The test statistic F value for the overall regression was significant at the upper 5 % level, which further supported that the second-order model was very adequate in approximating the response surface of the experimental design. The regression analysis for IPS production presented in Table 4 showed that the X₁ linear coefficient, which constituted the most significant term, i.e., NaNO₃, had a very significant effect on IPS production (p < 0.01); K₂HPO₄·3H₂O, and ferric ammonium citrate had a significant influence on IPS production (p < 0.05). However, the interaction between NaNO₃, K₂HPO₄·3H₂O, and quadratic ferric ammonium citrate had an influence on IPS production, but did not show a significant effect.

Table 4 Results of the stepwise regression analysis for the optimization of IPS production

The model predicted that a maximum IPS yield of 14.6 mg/L appeared at NaNO₃ 0.0039 g/L, K₂HPO₄·3H₂O 0.24 g/L, MgSO₄·7H₂O 0.21 g/L, ferric ammonium citrate 0.024 g/L, and NaHCO₃ 1.42 g/L. This predicted value was higher than any from the twelve experiments.

Verification of the models

In order to validate the optimal values, three dependent replicates were performed from the optimum nutritional conditions obtained by DPS software, listed in Table 5. Colors of the cultivation broth were significantly different, with the original BG11 medium appearing green, while the broths cultured in optimum nutritional conditions for biomass and IPS production were dark green and yellow, respectively, due to the different chlorophyll concentrations that each growth mode had. As shown in Fig. 6, growth curves of the Scenedesmus sp.-BH cultured in three types of culture medium were dramatically different, indicating that the five nutritional factors considered in this experiment could markedly impact the growth of Scenedesmus sp.-BH.

Table 5 The anticipated and experimental indices in optimum medium components

Fig. 6

Growth curves of the Scenedesmus sp.-BH cultured in different culture conditions. 1 presents the Scenedesmus sp.-BH cultured in basal BG11 medium as the control; 2 presents the Scenedesmus sp.-BH cultured in optimum medium components for biomass; 3 presents the Scenedesmus sp.-BH cultured in optimum medium components for IPS production

The final experimental results were in good agreement with predicted maximum indices (Table 5), which also proved the credibility of the mathematic models. The biomass and IPS production of Scenedesmus sp.-BH cultured in the optimum medium was over two-fold of the initial BG11 medium, which is commonly used for the cultivation of blue and green freshwater microalgae.

The experiments in our study present a comprehensive study on the optimization of critical medium components for the production of biomass and IPS from Scenedesmus sp.-BH. The mathematical models obtained from uniform design software reveal the effects of single factors and their interaction with the biomass and IPS production of Scenedesmus sp.-BH. From quadratic equations, we found that the properties and significant differences between individual variables vary with the target product. For instance, the biomass correlates with the interaction of K₂HPO₄·3H₂O and NaHCO₃(X₂X₅), MgSO₄·7H₂O and NaHCO₃(X₃X₅) concentration, but the IPS production is the function of NaNO₃ concentration (X₁), the interaction of K₂HPO₄·3H₂O, and ferric ammonium citrate (X₂X₄) concentration. The maximum output of both biomass and IPS production is observed at high K₂HPO₄·3H₂O, MgSO₄·7H₂O, and ferric ammonium citrate concentration. The optimal concentration of NaNO₃ for the maximum production of IPS is 0.004 g/L, which is lower than that for maximal biomass (0.77 g/L) (Table 5).

Considering different algal strains, cultivation methods, and cultivation scales, the required optimum culture conditions are quite different, while the effects of culture conditions on biomass and intracellular polysaccharides of algae are basically similar. The activities of some key enzymes of algal cells, such as high ATP-citrate lyase, can promote microalgae to shift from sugar storage sugar to lipids (Bellou and Aggelis 2013) and sequentially decrease the content of polysaccharides. Phosphorus plays a significant role in most cellular processes, especially those involved in the formation and transformation process of energy metabolism. Liang et al. (2013) found that different phosphorus levels could significantly influence the lipid, carbohydrate, and protein content of freshwater microalga Chlorella sp. Yingying and Changhai found that both the growth and biochemical compositions (such as chlorophyll, protein, and polysaccharide) in Isochrysis galbana can be affected by phosphorus. The interrelationships between phosphorus availability and the basic physiological reactions such as growth, reproduction, and biochemical composition have also been reported in other microalgae (Fabregas et al. 2000; Yim et al. 2003). In our experiments, the range of K₂HPO₄·3H₂O concentration from 0.065 g/L to 0.24 g/L was beneficial for cell growth, and intracellular polysaccharide biosynthesis. The bicarbonate ion was another main source of carbon for the growth and metabolism of microalgae. The results obtained in our study showed that the interaction of K₂HPO₄·3H₂O and NaHCO₃, MgSO₄·7H₂O and NaHCO₃ concentration had an important effect on the growth of Scenedesmus sp.-BH, with a biomass increase from 250.5 mg/L to 556.7 mg/L.

The nitrogen source also influences the intracellular polysaccharide biosynthesis of microalgae. Wang et al.(2011) reported that the bound polysaccharide and total polysaccharide content of Microcystis aeruginosa cultured in 5 % N of standard BG-11 media were significantly higher than those of M. aeruginosa cultured in all other nitrogen concentration media (10 %, 25 % and 100 %). Yang et al. (2012) also found that bound polysaccharides of M. aeruginosa cultured in 0.26 mg/L nitrogen concentration were obviously higher than those cultured in 2.55 and 25.47 mg/L at all experimental temperatures and light intensities. In our experiments, when NaNO₃ concentration decreased from 1.5 g/L to 0.004 g/L, it significantly influenced the production of intracellular polysaccharides of Scenedesmus sp.-BH (p < 0.001).

Iron, one of the structural elements of organic components, plays an essential role in photosynthesis and biomacromolecule synthesis of microalgae. As van Oijen et al. (2004) had already reported, iron stress appears primarily to affect photosynthesis in the Antarctic diatom Chaetoceros brevis, causing a reduction in the diurnal production of water-extractable carbohydrates. Liu et al. (2008) found that adding FeCl₃ into the media not only prolonged the period of the exponential growth phase and increased the final cell density of Chlorella vulgaris, but also greatly accumulated the lipid only when the iron concentration in the initial medium reached a certain value. In our study, the NaNO₃ concentration and the interaction of K₂HPO₄·3H₂O and ferric ammonium citrate concentration showed a significant effect on the IPS production of Scenedesmus sp.-BH, with an overall content increase from 5.55 mg/L to 14.3 mg/L.

Conclusions

A new self-flocculating microalga, identified as Scenedesmus sp.-BH, was isolated from freshwaters in Xinjiang, China. We found that this microalga contained a higher biomass and rate of IPS production compared with the other three microalgae in our lab. The biomass and IPS production of Scenedesmus sp.-BH cultured in optimized medium with uniform design could reach up to 556.7 ± 16.97 and 14.3 ± 0.46 mg/L, respectively, which were more than two-fold greater than those obtained from the original culture medium. The results obtained in this study showed that uniform design combining stepwise regression analysis was a powerful method for the optimization of culture conditions from microalgae.