Application of uniform design to evaluate the different conditions on the growth of algae Prymnesium parvum

Overview of the study area

The experimental algae P. parvum was collected from the fishponds in Dawukou, Ningxia, China. Algal water samples were filtered by medium size filter paper and centrifuged, and cultured with F/2 culture medium in the following environmental conditions for 5 days; light intensity of 5000 lx, light/dark ratio of 12 h: 12 h, water temperature of 18.5 ± 0.5 ℃, pH of 8.5 ± 0.1 and salinity of 1.2 ± 0.1 mg/l^20,21. The plate separation method was used to separate and purify the cultured algae²². After microscopy, the colony of pure algal cells were transferred to different volumes of triangular glass bottles which contains sterilized F/2 medium for expansion culture. Algae P. parvum propagates vegetatively by cell division, the cell density of algae increases exponentially during the process of propagation thus requires more space. To accommodate this increasing space requirement, different sizes of the triangular glass bottles were used as 50 ml, 250 ml and 10 l. The expansion cultures were maintained in the environmental conditions similar to the initial culture. The algal cells were used for the experiment when they reach the logarithmic growth stage (the logarithmic growth stage was reached in 10 days).

Data collection and experimentation

The water sample from the 10 L expansion culture of P. parvum was collected. The initial nutrient concentrations and environmental factors were determined using appropriate methods and equipment in the laboratory. The initial nutrient concentrations and environmental conditions of the algae culture used in this experiment is presented in Table 1.

Experimental factors and their levels for each nutrient concentrations and environmental factors were designed based on the above reference as shown in Table 2. We have designed eight levels for environmental factors (i.e., water temperature, pH and salinity) and ten levels for nutrient concentrations (i.e., nitrogen, phosphorous, silicon and iron).

1. 1.

   Evaluation of the effects of environmental factors on the growth of P. parvum

To study the effects of environmental factors on the growth of P. parvum, water temperature, pH and salinity were used as the experimental factors by adopting the uniform design^23,24,25 of three factors and eight levels as shown in Table 3.

A 250 ml triangular glass bottle was used to implement each level of the above experiment with three replicates for each level (total of 24 bottles). The algae culture was allowed to grow in F/2 culture medium in the nutrient solution of 100 ml with an inoculation ratio of 1:10 (V/V). These bottles were kept in the light intensity of 5000 lx with light/dark ratio of 12 h: 12 h, while maintaining all other growth conditions to meet the experimental design requirements. The nutrient concentrations of N, P, Si and Fe were maintained at the level of initial concentrations (Table 1). Inoculated algae were cultured in a shaker for 10 days until it reaches its logarithmic growth stage and the growth rate was quantified.

1. 2.

   Evaluation of the effects of nutrient concentrations on the growth of P. parvum

To study the effects of nutrient concentrations on the growth of P. parvum, nitrogen, phosphorus, silicon and iron were used as experimental factors by adopting the uniform design^5,26 of four factors and ten levels as shown in Table 4. The culture medium was prepared with sodium nitrate (NaNO₃) as the nitrogen source, monosodium phosphate (NaH₂PO₄) as the phosphorus source, sodium metasilicate (Na₂SiO₃) as the silicon source, and ferric citrate (FeC₆H₅O₇) as a source iron to obtain the appropriate concentrations of nitrogen, phosphorous, silicon and iron as designed for this experiment (Table 2).

A 250 ml triangular glass bottle was used to implement each level of the above experiment with three replicates for each level (total of 30 bottles). The algae culture was allowed to grow in F/2 culture medium with a volume of 100 ml and an inoculation ratio of 1:10 (V/V). These bottles were kept in the light intensity of 5000 lx, light/dark ratio of 12 h: 12 h, water temperature of 18.5 ± 0.5 ℃, pH of 8.5 ± 0.1 and salinity of 1.2 ± 0.1 mg/l. Inoculated algae were cultured in a shaker for 10 days until it reaches its logarithmic growth stage and the biomass density was quantified.

1. 3.

   Determination of the growth rate of P. parvum

The algal cell density of the culture of each experimental level was measured using a 0.1 ml count plate under an optical microscope (Leica biological microscope DM1000, Leica Corporation, Oskar-Barnack-Straße, Germany) both at the beginning of the experiment and following 10 days of incubation period as the growth of the algae can reach its logarithmic growth stage at 10 days. Based on the algal cell density measurement, biomass density was calculated using the following formula (Eq. 1) described by Wei and Zhang;

where t is the duration of the experiment in days, N₀ is the initial cell density (cell/ml) at the beginning of the experiment, and Nt is the cell density (cell/ml) at the end of day t of the experiment.

Data analysis and results

1. 1.

   Establishment of the regression model between environmental factors and the growth rate

The growth rate of P. parvum under different levels of environmental factors are shown in Table 5, and the growth curve with time is shown in Fig. 1.

In multiple quadratic stepwise regression analysis, water temperature (X₁), pH (X₂) and salinity (X₃) were taken as independent variables, and the growth rate (Y) was taken as the dependent variable. From this analysis a quadratic polynomial regression equation (Eq. 2) was developed as follows:

Correlation coefficient (R) of the above equation was 0.9994 and probability (P) of the regression equation was 0.025 (p < 0.05) as tested by the F-test, which indicates the significant relationship between the growth rate of P. parvum and the environmental factors. Therefore, the above regression model could robustly represent the relationship between the selected environmental factors and the growth rate of P. parvum. The standardized regression coefficients of each environmental factor of this model in the stepwise regression with the growth rate are shown in Table 6.

Accordingly, the magnitude of the effect of each environmental factor on the growth rate was in the order of X₂ > X₃ > X₁. Thus, the contribution of pH > salinity > water temperature on the growth rate of P. parvum.

1. 2.

   Evaluation of the effect of environmental factors on the growth rate of P. parvum

The environmental conditions that would result in the maximum growth rate of P. parvum were determined by optimizing the regression equation (Eq. 2). The following simple regression models (Eqs. 3–5) of multiple quadratic stepwise regression analyses reveal the relationships between individual environmental factors and the growth rate. These models were obtained by dimensionality reduction analysis in which the other factors were maintained at optimal levels.

The influence curves of each environmental factor on growth rate of P. parvum are shown in Fig. 2. The behavior of the curves is similar where the growth rate increases initially, then reaches a theoretical maximum and finally declines with increasing level of each environmental factor. Accordingly, P. parvum reaches its theoretical maximum growth rate (0.789) when the water temperature, pH and salinity is 18.11 ℃, 8.39 and 1.23‰, respectively. Therefore, Fig. 2 can be considered as the growth model of P. parvum as affected each of the respective environmental factors.

1. 3.

   Establishment of regression model between nutrient concentrations and the growth rate

The growth rates of P. parvum under the different levels of nutrient concentrations are shown in Table 7, and the growth curve with time is shown in Fig. 3.

A quadratic polynomial regression equation (Eq. 6) was generated using N (X_i), P (X_ii), Si (X_iii) and Fe (X_iv) as independent variables and the growth rate (Y′) as the dependent variable by using multiple quadratic stepwise regression analysis as follows:

Correlation coefficient (R) of the above equation was 0.9994 and probability (P) of the regression equation was 0.035 (< 0.05) as tested by F-test, which indicates that the relationship between the growth rate of P. parvum and nutrient concentration is significant. Hence, the above regression model could robustly represent the relationship between the concentration of N, P, Si, Fe and the growth rate of P. parvum. The standardized regression coefficients of each nutrient in the main model (Eq. 6) of the stepwise regression with the growth rate are shown in Table 8.

Accordingly, the magnitude of the impact of each nutrient on the growth rate of P. parvum was in the order of X_i > X_ii > X_iv > X_iii. Therefore, the contribution of nitrogen > phosphorous > iron > silicon for the growth of P. parvum.

1. 4.

   Evaluation of the effect of nutrient concentrations on the growth rate of P. parvum

Multifactor square stepwise regression model was used to analyze the influence of individual nutrient concentration following the dimensionality reduction. To evaluate the influence of individual nutrient concentration on the growth rate, following sub-models (Eqs. 7–10) were developed by fixing other factors at the optimal level.

The influence curves of each nutrients on growth rate of P. parvum are shown in Fig. 4. The behavior of the curves shows an initial increase of the growth rate, then the growth rate reaches a theoretical maximum and finally declines with increasing level of concentrations of each nutrient. Accordingly, P. parvum reaches its theoretical maximum growth rate (0.896) when the concentration of nitrogen, phosphorous, silicon and iron is 3.41, 1.05, 0.69, 0.53 mgl⁻¹, respectively. Therefore, Fig. 4 may be considered as the growth model of P. parvum as affected each of the respective nutrients.

Source: Ecology - nature.com