بهینه‏ سازی محیط کشت برای افزایش تولید گاما- لینولنیک اسید توسط موکور همیلیس

Introduction

Recently, microbial production of lipids containing polyunsaturated fatty acids (PUFA) such as omega-3 and omega-6 compounds have attracted much attention. γ-Linolenic acid (GLA) (6, 9, 12, cis, cis, cis-octadecatrienoic acid), an omega-6 PUFA, is an essential fatty acid in human nutrition, which has been suggested for pharmacological and dietetic purposes. The numerous advantages mentioned for GLA can be summarized as suppressing human T-cell proliferation and activation, decreasing cardiovascular risk factors, impressing against rheumatoid arthritis, normalizing nerve conduction velocity and sciatic endoneurial blood flow (1- 9).

γ-Linolenic acid conventionally produces from plant sources such as seeds of evening primrose, borage and blackcurrant. However, to meet the increasing demand, extensive research is being carried out for its production from microbial sources as an alternative to the plant based production (10- 11).

The cells of some algae and fungi species have been identified to contain GLA. For example, producing substantial quantities of GLA by fungi the order Mucorales especially Mucor, Mortierella, and Cunninghamella genus has been reported. The first trials of PUFA production by Mucor fungi with the target being γ-Linolenic acid were carried out in the UK and Japan. Thereafter, microbial production of various PUFAs has been investigated with the aim of understanding effective production (12- 14).

Various studies have been conducted to optimize the production of fatty acids by fungi. In this way, the traditional one-at-a-time optimization strategy is relatively simple, and the individual effects of medium factors can be graphically depicted without statistical analysis. It has a low efficiency in experimental runs and does not consider the interactions. Moreover, it frequently fails to determine the region of optimum response. Consequently, factorial based designs have been suggested as efficient approaches to optimization. In fact, factorial is a useful experimental design to investigate the effect of several variables influencing the responses by varying them simultaneously and carrying out a limited number of experiments (15).

The main objective of this study was to screen the significant variables by the fractional factorial design, and further to optimize the screened variables exactly for GLA production by employing Mucor hiemalis PTCC 5292 in submerged fermentation using full factorial design.

Materials and methods

Microorganism, inoculum preparation and culture conditions: Mucor hiemalis PTCC 5292 was purchased from Iranian Research Organization for Science and Technology (IROST). Mucor hiemalis was kept on potato dextrose agar (PDA) slants by incubating at 28˚C for four days. Then, 2 ml of sterile distilled water was added to each sporulated slant and the spores were harvested by scrapping them from the PDA surface. 1ml of spore suspension (around1×10⁷ spores) was used to inoculate 250 ml Erlenmeyer flasks containing 50 ml of basal fermentation medium according to the experimental design method (Table 1) and incubated in a rotary shaker-incubator at 180 rpm and 28˚C for 3 days (16).

The basal fermentation medium included the following salt components (g/l): NH₄NO₃ 0.286, KH₂PO₄ 0.75, MgSO₄·7H₂O 0.4 and CaCl₂ 0.4.

Cell dry mass determination and lipid extraction: The mycelia were harvested from medium through Whatman No. 1 filter paper and thoroughly washed with distilled water, then dried at 95 + 2˚C for 24 hours and the dried biomass was weighed (17).

The lipid production from mycelia was assayed by extraction according to the modified procedure of Bligh and Dyer (18) using chloroform/methanol (1:1). Chloroform phase was recovered. The same process was repeated thrice. The whole solvent was evaporated and dried.

Methyl ester preparation and analysis of GLA: Free fatty acids and o-acyl lipids were esterified by heating with a large excess of anhydrous methanol in the presence of an acidic catalyst. Fatty acids were methyl esterified according to the procedure of Christie (19). The sample of lipid (50 mg) was dissolved with 1ml toluene in a test tube fitted with a condenser. Then 2ml of 1% sulfuric acid in methanol was added to it and maintained at 50°C for 24 h. The methylated fatty acids were separated from the aqueous layer by adding 5ml of 5% NaCl and then dissolved in 2×5ml n-hexane. The separated n-hexane layer using Pasteur pipettes washed by 4 ml of water containing 2% potassium bicarbonate and dried over anhydrous sodium sulfate. The solution is filtered and the solvent removed under reduced pressure in a rotary film evaporator or in a stream of nitrogen.

Gas chromatographic conditions: GC^[1] was performed on Agilent 19091J-413 Series Gas Chromatograph equipped with a FID^[2] and the capillary column HP5 (30 m ´ 0.25 mm i. d., 0.25 µm film thickness; USA). Injector and detector temperatures were maintained at 260°C and 300°C, respectively. The oven was programmed for 2 min at 100°C, then was increased to 160°C at 3 min, was maintained for 2 min at 215°C, was increased further to 217°C at 2 min, then was maintained for 2 min at 218°C and finally was increased to 260°C at 2 min. The carrier gas, nitrogen, was used at a flow rate of 1.5 mL/ min. The injection volume was 1 µL, with a split ratio of 50:1.

Experimental design and data analysis: In order to select the significant variables for GLA production, the independent variables such as carbon sources (glucose), mineral nitrogen sources (ammonium nitrate), organic nitrogen sources (peptone) and pH were considered and screened via 1/2 fractional factorial design. A total of four variables were included for screening and each independent variable was tested at two levels, high level (+1) and low level (-1) (Table 1). For a 1/2 fractional factorial with the 4 variables, 11 experimental runs with three replications of the center points are required (Table 2). The variables and their experimental matrix are listed in Tables 2 and 3 respectively. The main effect of each variable on GLA production was estimated as the difference between the two averages of measurements made at the higher and lower levels. The analysis of variance (ANOVA^[3]) was used to determine the significance of each variable (Table 4). The p-value, as the most common means of assessing significance, was also determined for each variable. The p-value is the probability that the magnitude of a parameter estimate is due to random process variability. The present study was accepted only confidence levels above 95%. Therefore, it is possible to gradate the variables with respect to their effects on the GLA concentration.

Table 1- Experimental variables at different levels used for the fractional and full factorial designs.

Table 2- Lipid production, experimental design and results of the 1/ 2 fractional factorial method for GLA concentration by Mucor hiemalis 5292.

Strain PTCC 5292 was cultivated at 28°C for 72h. Representation of biomass (X, g/L), biomass yield on the consumed glucose (Y_X/S, g/g), cellular lipid in biomass (Y_L/X, %) and lipid yield on the consumed glucose (Y_L/S, g/g) when the maximum quantity of cellular lipids (L_max, g/L) was achieved. Data are presented as mean values from duplicate experiments conducted by using different inocula. a For the determination of the initial C/N ratio in the medium, it was assumed that peptone contained 45.95% wt/wt, of carbon and 15.5% wt/wt, of nitrogen.

Table 3- Lipid production, experimental design and results of the full factorial method for GLA concentration by Mucor hiemalis 5292.

Strain PTCC 5292 was cultivated at 28°C for 72h. Representation of biomass (X, g/L), biomass yield on the consumed glucose (Y_X/S, g/g), cellular lipid in biomass (Y_L/X, %) and lipid yield on the consumed glucose (Y_L/S, g/g) when the maximum quantity of cellular lipids (L_max, g/L) was achieved. Data are presented as mean values from duplicate experiments conducted by using different inocula. _a For the determination of the initial C/N ratio in the medium, it was assumed that peptone contained 45.95% wt/wt, of carbon and 15.5% wt/wt, of nitrogen.

Table 4- ANOVA showing the effect of factors and their interactions for GLA concentration in the 1/2 fractional factorial design.

DF refers to degrees of freedom, SS refers to sum of squares, MS refers to mean square, F and p refers to F and p values respectively.

The next step was to determine the optimum levels of the significant independent variables for GLA production. Therefore, the full factorial design was used for the completion of previous experiments (1/2 fractional factorial design) and was effective to improvement the time and cost of experiments. The levels of variables were similar to 1/2 fractional factorial design, as is shown in Table 1. The experimental design and the results of them are listed in Table 3.

All of the experiments were repeated. The sum of experiments was carried out in the 35 runs. The statistical software Minitab 16 was used to design and analysis of all experiments and statistical software Design-Expert 8.0.7.1 was applied to depict a three-dimensional response surface plot for investigating the effect of variables on the GLA production.

Results

Screening of the significant variables using the fractional factorial design: The results of the 1/2 fractional factorial design for GLA concentration in the screening of the significant variables step has been presented in Table 2.

The variables showing the p-values less than 0.05 were determined to have a significant effect on the response. Therefore, the variables of glucose, peptone, ammonium nitrate and pH were selected for further optimization through a full factorial design (Table 3). Table 4 indicates the analysis of variance of the variables on the response in the screening step.

Optimization of the significant variables using full factorial design: The design and the matching outcomes of full factorial experiments to evaluate the effects of four independent variables are presented in Table 3.

Variables and responses were evaluated for the GLA production consequently by Minitab version 16 to create a model based on design, which comprised main effects and interaction effects. The model planed model based on the analysis related GLA production with variables of the experiments. The subsequent equation was acquired from the coded data:

Where, Y is GLA concentration; A, glucose concentration; B, peptone concentration; C, ammonium nitrate concentration and D, pH. The analysis of variance of this fitting is showed in Table 5.

(1)

Y=123.35 + 67.62A - 87.03B + 18.58C - 36.32D - 58.91AB + 2.20AC - 27.31AD - 14.48BC + 44.63BD - 48.20CD - 10.49ABC + 29.99ABD - 38.99ACD + 45.02BCD + 45.40ABCD

Table 5- ANOVA showing the effect of factors and their interactions for GLA concentration in the full factorial design.

DF refers to degrees of freedom, SS refers to sum of squares, MS refers to mean square, F and p refers to F and p values respectively.