Introduction

Marine microalgae are valuable sources for various purposes such as fuels, animal feeds, vitamins, pharmaceuticals, etc. [1]. In recent years, there has been a growing interest in researching microalga polysaccharide because they are easily to be cultivated in large quantity and stable quality. The unicellular marine microalga Porphyridium cruentum (P. cruentum) can synthesize and secrete sulfated polysaccharides to the cultural medium when the cells grow to its steady period. The extracellular polysaccharides from P. cruentum are highly sulfated acidic heteropolymers composed of xylose, galactose and glucose. The molecular weight is in the range of 2–7 × 106 Da [2]. Different from the polysaccharides extracted from plants or macroalgae, microalgae polysaccharides have some special properties for industrial uses such as in cosmetics, health food, anti-herpes, antiviral and anti-irritant medication [3–5].

Previous studies have shown that nutrient and light availability affected the production of cell biomass and extracellular polysaccharides [6–8]. Nitrogen determines the production of polysaccharides and influences the secretion of soluble polysaccharides [9]. At present, most researches on microalga cultivation are carried out in batch culture system. Nevertheless, the productivity of microalga biomass or polysaccharide is usually considerably higher in continuous or semi-continuous culture system [10, 11]. Due to the operational simplicity of semi-continuous culture, several parameters such as renewal rate and nutrient concentration were generally studied to control the productivity and biochemical composition [12, 13]. Although several data are available on the biochemical variability of P. cruentum controlled by changes of culture parameters like renewal rate, nutrient concentration in semi-continuous systems [10, 14], the previous research mostly focused on the single factor such as renewal rate or nutrients. The integrative factors and their interaction affecting the output rate and yield of exopolysaccharide still need to be studied.

In this work, we introduced the uniform design method to investigate the interaction between nutrition and renewal regime on extracellular polysaccharides production of P. cruentum. Culture parameters such as nitrogen concentration, renewal rate and renewal period were obtained by regression analysis for the maximum recovery yield and output rate of polysaccharides. The optimized method, culture technology, and the technical conditions achieved in this work could be an important reference for harvesting extracellular polysaccharide of P. cruentum in large-scale.

Materials and methods

Microorganism and the cultural conditions

Porphyridium cruentum was provided by the Institute of Oceanology, Chinese Academy of Sciences. The high-yield strain P.C-03 was isolated, purified and cultivated in the optimum medium OMII [15]. The P. cruentum cells were grown in a newly designed flat-plate organic glass photobioreactor (FPPBR-50) (Fig. 1). The size was 50 × 5 × 60 cm, and the actual cultural volume was 15 L. The temperature and pH were monitored by thermometer and pH meter installed on the top of the bioreactor. The temperature of culture system was controlled at 25 °C (±1 °C) by air-conditioning in culture room. Light source was supplied by 15 W Philips cool-white lamps (F15T12/CW), and the light intensity was adjusted at 80 μE m−2 s−1. The culture system was agitated by charging 1% CO2 in compressed air at 10 L/min through an aeration tube extended to the bottom of the photobioreactor. The initial inoculation density was about 60 × 104 cells/mL, and the cultures were kept in this semi-continuous regime for 23 days.

Fig. 1
figure 1

Schematic drawing of flat-plate photobioreactor system employed by this study. A air compressor, B air reservoir, C CO2 tank, D press meter, E flow meter, F oil and water seperator, G gas analyzer, H aeration tube (diameter 5 mm), R airlift flat-plate type photobioreactor, I Periplastic pump, K inlet reservoir, L sampling port, T thermometer, P pH probe, O DO probe

Full size image

Analytical determination

Cell numbers were counted six times under a microscope (OLYMPUS CH-2) using a bright-light hemacytometer. The value was mean ± SD. The biomass dry weight (DW) W and growth rate K were calculated by Eqs. 1 and 2, in which N 2 and N 1 were cell densities (cells/mL) at two different time points t 2 and t 1 (days).

$$ W({\text{g}}/{\text{L}}) = 1.4054 \times {\text{OD}}_{ 6 0 4} - 0.0727\quad (r^{2} = 0.9923) $$
(1)
$$ K = \log_{2} (N_{2} /N_{1} )/(t_{2} - t_{1} ) $$
(2)

To determine the concentration of extracellular polysaccharide (EPS), the recovery suspension was centrifuged at 4,500 rpm for 10 min. Then, the top layer solution was subjected to the phenol-sulfuric acid test with glucose as the standard [16]. Each test was performed in triplicate. Value was the mean ± SD.

Total recovery yield (Y total ), productivity of cell biomass and extracellular polysaccharides produced by P. cruentum were calculated by Eqs. 3 and 4, in which C was the concentration of biomass and EPS of the renewal broth each time; n was the renewal times; V was the total culture volume (L); R was the renewal rate (%); P was the output rate; V R was the total recovery volume (L) and t g was the total culture time of P. cruentum (day).

$$ Y_{\text{total}} = \sum\limits_{i = 1}^{n - 1} {C_{i} } VR + VC_{n} $$
(3)
$$ P = \frac{{Y_{\text{total}} }}{{V_{R} t_{g} }} $$
(4)

Nitrogen concentrations were measured by ultraviolet spectrophotometry [17].

Uniform design

The experiment for optimization of renewal regime was arranged as three factors, and each was at seven levels. The UD table U7 (73) was applied to arrange the experiments (Table 1) [18]. Each trial was performed in triplicate. The semi-continuous regime was started at seventh day when the cell growth had reached the logarithmic middle-phase.

Table 1 Factors and levels in the uniform design experiments
Full size table

Statistical analysis

All experiments were performed in triplicate. The averages of total recovery yield, productivity of biomass and polysaccharides were taken as response values. The statistical and stepwise regression analysis was performed with UD software (Uniform Design Version 3.0, developed by Wang Y. F., Jilin, China). The Tukey test (p < 0.05) was used to compare different treatments (SPSS for windows Version 11.5).

Results

The change of cultural parameters in semi-continuous cultivation

The renewal culture began on the seventh day because cells reached the highest growth activities in this period. The changes of cell density in different renewal regimes were shown in Fig. 2. The final cell density of batch culture was (1,262.84 ± 23.7) × 104 cells/mL. In semi-continuous culture system, the cell density reached equilibrium within each renewal period in experiments No. 1 and No. 2. In experiments No. 4, 5, 6 and 7, the cell density in the recovery suspensions exhibited periodic increase, the maximum was (1,113.26 ± 21.7) × 104 cells/mL on the 23th day of No. 6. The lowest cell density (555.30 ± 17.6) × 104 cells/mL was observed in experiment No. 3 due to its high renewal rate and short renewal period.

Fig. 2
figure 2

Growth curve of P. cruentum under different treatments

Full size image

The renewal regimes also affected the production of extracellular polysaccharides in recovery suspensions. As shown in Fig. 3, in experiments No. 4, 5 and 6, the concentration of polysaccharides exhibited periodic increase within each renewal period. In experiment No. 4, the maximum polysaccharide yield was 1,108 ± 13.2 mg/L. The polysaccharide concentration decreased dramatically in experiment No. 3 because of high renewal rate and short renewal period. Meanwhile, the polysaccharide was only 252.92 ± 3.87 mg/L after the last renewal cycle. In batch culture, the highest polysaccharide concentration was 1.25 g/L at the 16th day. However, with the increase of polysaccharide secretion, the culture medium turned viscous, and the cells autolyzed. Therefore, the concentration of polysaccharide cannot be determined exactly.

Fig. 3
figure 3

Variety of polysaccharide concentration in recovery suspension under different treatments

Full size image

The consumption of NaNO3 in the cultural media was shown in Fig. 4. In experiment No. 2, a high average concentration of 2.0 ± 0.19 g/L was observed with a low standard deviation of 0.19, which indicated the establishment of a balance between the addition and consumption of nitrogen. The result could farther explain why cell density kept equilibrium and the highest growth rate was obtained in experiment No. 2 (Fig. 2; Table 2). In experiment No. 3, the concentration of NaNO3 was at a lower average level of 0.78 g/L with a high standard deviation of 0.62. The results indicated that renewal regime affected the consumption-addition balance of nitrogen. Therefore, there would be an optimal renewal regime to maintain nitrogen concentration in culture liquid at stable level, and made cells growth steady.

Fig. 4
figure 4

Varieties of concentration of NaNO3 in culture medium in different treatments

Full size image
Table 2 Experiment of UD U7 (73) for the optimization of recovery yield of biomass, EPS and growth rate mean ± SD
Full size table

Optimization on renewal regime for maximum recovery yield and productivity

According to Table 2, the average growth rate (K avg) between different runs had significant difference (p < 0.05). The higher average growth rates were obtained in experiments No. 2 and 3, and both of which had short renewal period and high renewal rate. The total recovery yield of biomass and polysaccharides between different runs also had significant differences (ANOVA, p < 0.01), and their maximal value were obtained in experiments No. 2 with 63.9 and 27.7 g, respectively. Besides, the recovery volume (V R) in semi-continuous culture was larger than that in batch culture, and the larger V R was observed in experiments No. 2 and 3 too (Table 2). The results indicated that the final recovery yield correlated with the K avg and V R, and recovery volume was the function of renewal rate and renewal period. Therefore, the optimum renewal regime was asked in order to obtain the maximal biomass and polysaccharide production. The output rate between different runs had slight significant difference (p < 0.05). The highest output rates of cell biomass and polysaccharides were obtained in experiments No. 6 and 4.

Equation 5 was established as the mathematic model for the optimization of three factors for maximum productivity. The quadratic equations relating to different indexes were expressed in terms of the values of all the independent variables and by neglecting the statistically insignificant terms. The mathematic expressions were established as Eqs. 6–10. The analysis of variance for the mathematical models was listed in Table 3.

Table 3 Variance analysis by stepwise regression on mathematic models
Full size table
$$ Y = b_{0} + b_{1} x_{1} + b_{2} x_{2} + b_{3} x_{3} + b_{4} x_{1}^{2} + b_{5} x_{1} x_{2} + b_{6} x_{1} x_{3} + b_{7} x_{2}^{2} + b_{8} x_{2} x_{3} + b_{9} x_{3}^{2} $$
(5)
$$ K_{\text{avg}} = 0.126 + 0.0001x_{1} + 0.991x_{2} - 0.04x_{3} + 0.004x_{1} x_{3} - 1.12x_{2}^{2} $$
(6)
$$ Y_{{c {\text{-total}}}} = 36.4 + 233x_{2} - 6.26x_{3} - 0.424x_{1}^{2} + 1.28x_{1} x_{3} - 354x_{2}^{2} $$
(7)
$$ Y_{{{\text{EPS}} {\text{-total}}}} = 15.28 + 84.32x_{2} + 1.70x_{1} x_{2} + 0.154x_{1} x_{3} - 168x_{2}^{2} - 0.091x_{3}^{2} $$
(8)
$$ P_{\text{biomass}} = 22.4 + 26.2x_{1} + 7.47x_{3} - 5.17x_{1}^{2} - 1.11x_{1} x_{3} - 86.6x_{2}^{2} $$
(9)
$$ P_{\text{EPS}} = 19.81 - 1.99x_{1} + 8.4x_{3} + 10.59x_{1} x_{2} - 53.41x_{2}^{2} - 16.14x_{2} x_{3} $$
(10)

All mathematic expressions were found to have high correlation (R 2 > 0.99) (Table 3), which suggested that the model was suitable for evaluating the semi-continuous system. Generally, the calculated F value should be greater than that of the critical value, if the model was a good prediction for experimental results. Here, all of the computed F values for the equations were greater than that of the critical values, and the regression coefficients of equations were statistically significant (Table 3). Therefore, the equations were authentic.

Equation 6 was the quadratic function of average growth rate (K avg) of microalga cells in semi-continuous culture. t test for the regression coefficient of variable revealed that renewal rate (X 2) and renewal period (X 3) had more significant effect (p < 0.05) on growth rate than the nitrogen concentration (X 1). The t test results of variable coefficients in Eq. 8 showed that all three factors (X 1, X 2, X 3) had significant effect (p < 0.05) on total recovery yield of polysaccharides (Y EPS-total). t test for Eq. 7 showed that only the renewal rate (X 2) and the renewal period (X 3) had a significant effect on total recovery yield of biomass (Y c-total), but the interaction between nitrogen concentration (X 1)and renewal period (X 3) had no significant effect (Table 3). In Eq. 9, none of three factors was found significant (p > 0.05), but square of renewal rate (X 2) had a significant effect on the productivity of biomass. Similarly, only renewal period (X 3) and interaction of X 2 and X 3 (X 2 X 3) had a significant impact on productivity of polysaccharides in Eq. 10.

Based on the model evaluation, the optimal test variables and anticipated maximum indexes were calculated by UD software with gridding attempt method (Table 4). Experimental confirmations were carried out with each set of calculated optimal renewal regimes, and the culture time was 23 days. Regularity for change of cell growth, EPS production and nitrogen consumption in the process of culture were similar to the above results showed in Figs. 2, 3, 4. The final experimental results were in good agreement with predicted maximum indexes (Table 4), which also proved the credibility of the mathematic models.

Table 4 The optimized renewal regime and concentration of NaNO3 for maximum index
Full size table

Discussion

The concentration of nutrient salts and light attenuation were the key limiting factors for cell growth and secondary metabolites synthesis in batch culture. The semi-continuous culture could change the nitrogen concentration and light utilization efficiency, accordingly overcame the restrictions and kept microalgal cells growing steadily. Renewal rate, renewal period, nitrogen concentration, and their interaction had significant effects on the cell growth and recovery yield (Table 3). In our previous study on growth dynamics of P. cruentum [19], we have validated that light attenuation was the function of incident light intensity (I 0), cell density, and light-path of photobioreactor (LP). With increase of cell density, the degree of light attenuation increased. In this study, renewal rate correlated with nitrogen concentration and cell density in the culture media, so it affected cell growth rate and polysaccharide production. Renewal period determined recovery time and consumption rate of nitrogen, and nitrogen concentration influenced the growth of cells and the production of polysaccharides [9]. Therefore, an optimal renewal regime is needed for obtaining the maximum value of different index.

If the cell growth period, namely, the special time required for the cell density updated to the initial level before renewal, was longer than the renewal period, cell density in the recovery suspension could be increased by applying a longer renewal period. It was understandable that the highest and the lowest concentration of biomass were observed in experiments No. 6 and 3, respectively (Fig. 2). As shown in Fig. 3, the concentration of polysaccharides in the recovery suspension increased with addition of renewal degrees in experiments No. 4, 5 and 6, which elucidated that low renewal rate and long renewal period would facilitate the synthesis of polysaccharides. Nitrogen concentration remained steady when supply and consumption was at equilibrium. However, if the supply did not keep up with consumption, the nitrogen concentration decreased. In experiment No. 4, the average nitrogen concentration was 1.56 ± 0.43 g/L in culture liquid. Accordingly, the concentration of polysaccharides reached the highest value (Figs. 3, 4). The result could agree well with our previous work about culture medium optimization, which showed that the highest polysaccharides production from P. cruentum was obtained at NaNO3 concentration of 1.5 g/L.

The mathematic models revealed the effects of single factors and their interaction on growth rate, total recovery yield and output rate. From the quadratic equations, it was found that the property and significant difference of individual variable varied with the target product. For instance, the total recovery yield of biomass (Y c-total) correlated with renewal rate (X 2), renewal period (X 3), and the interaction between nitrogen concentration and renewal period (X 1 X 3), but the Y EPS-total was the function of renewal rate (X 2), X 1 X 2, and X 1 X 3. The maximum average growth rate and total recovery yield of biomass were achieved under renewal condition of short renewal period and high renewal rate (Table 4), which can be explained by that short renewal period and high renewal rate could keep the nitrogen concentration of culture system in a normal level, the same time, it could reduce light attenuation degree. Thus, the growth rate and biomass were increased. However, extension of renewal period was in favor of the increase of total recovery yield of polysaccharides (Table 4). The maximum output rate of both biomass and polysaccharides were observed at low renewal rate and long renewal period. The optimal concentration of NaNO3 for maximum productivity of polysaccharides was 0.5 g/L, which was lower than that for maximal productivity of cell biomass (1.78 g/L) (Table 4). The results indicated that the deficiency of nitrogen would stimulate synthesis of polysaccharide in some microalgal cells [20, 21]. Besides, different microalga species and different culture system have dissimilar optimal renewal regime. For instance, the optimum renewal rate for Chlorella spp. cultured in column photobioreactor was 33%, whereas the optimum renewal rate for Phaeodactylum tricornutum in 250 mL Erlenmeyer flasks was 18.2% [22, 23].

According to the experimental results (Table 4), the average production of biomass and polysaccharide were calculated as 4.58 and 1.96 g/L, respectively, in semi-continuous culture system (the average production was the ratio of total yield to volume of photobioreactor 15 L). However, only biomass production of 2.22 g/L and EPS production of 1.25 g/L were achieved in batch culture system. The results confirmed that semi-continuous culture mode was more efficient for increasing the production. In this study, the maximal output rate of polysaccharide was 68.64 mg/L per day, which was 2.02 times higher than the one cultivated in 20 cm LP flat plate photobioreactor 15 L. The polysaccharide concentration of P. cruentum cultured in batch or semi-continuous mode were mostly in the ranged of 500–950 mg/L [1, 6, 14]. Although some researchers reported higher polysaccharide concentration [10, 24], they cultured the alga in 80 mL glass tubes. The difference maybe caused by the quantitative method. Generally, the total recovery yield and output rate are the function of renewal rate and renewal period, also affected by nutrients. The high polysaccharide concentration would not always imply high total recovery yield and high output rate. For example, the highest exopolysaccharide concentration was 1.10 g/L in experiment No. 4 (Fig. 3), but the maximal total recovery yield of EPS was harvested in experiment No. 2, not in No. 4 (Table 2). Similarly, the highest polysaccharide concentration was 1.25 g/L in batch culture. However, the total recovery yield was 4.5-fold and the output rate was 7.8-fold lower than that in semi-continuous culture with optimal renewal regime.

The established mathematic models were credible and efficient in the determination of optimal renewal regime for growth rate, total recovery yield and productivity. The optimal renewal condition was objective dependent and could be designed accordingly.