3.3 Determination of influencing factors for succinic acid production from xylan using microbial consortium containing strain M5 and 130Z
The influence of single factors, such as mixed time (inoculation time ofA. succinogenes to T. thermosaccharolyticum M5), initial xylan concentration, initial yeast extract concentration, pH, and MgCO3 concentration on succinic acid production was explored. During these variables, mixed time, initial xylan concentration and MgCO3 concentration were found to have greater impacts on succinic acid production. As shown in Fig 4A, the optimal inoculation time of strain A. succinogenes 130Z to T. thermosaccharolyticum M5 was 72 h. The possible reason could be that when the inoculation time was too early, xylanase and β-xylosidase activities secreted by strain M5 was relatively low (Fig 4B), which was not conducive to the subsequent xylan degradation. However, if the inoculation time was too late, strain M5 was in the decline period, and xylanase and β-xylosidase secretion efficiency would be reduced. When strain 130Z was inoculated at 72 h, strain M5 was in the exponential growth period, and a large amount of xylanase and β-xylosidase was secreted. In this microbial co-cultivation system, xylanase activity was relatively stable with 0.41 U/mL when temperature was shifted to 37℃, indicating that xylan could still be continuously degraded.
Generally, higher xylan concentration would lead to higher xylose production in the microbial co-cultivation system. However, when xylan concentration was increased to a certain extent, the growth of strain M5 would be inhibited, namely the substrate inhibition occurred. Moreover, a product feedback inhibition of xylanase and β-xylosidase caused by xylose also existed, which has been mentioned above (Fig 4C). Hence, when xylan concentration was at 80 g/L, the highest succinic acid concentration of 22.07 g/L was obtained (Fig 4D). Further increase of xylan would reduce succinic acid production, as only 21.81 g/L of succinic aicd was produced from 100 g/L of xylan.
Organic acids produced in the fermentation process could significantly reduce the pH value. MgCO3 was usually added as an acid neutralizer to maintain the pH values in the fermentation medium. During the reaction, MgCO3 can also release CO2 to maintain the CO2 environment needed for succinic acid fermentation (Hyohak et al., 2010). In addition, MgCO3 can provide magnesium ions for phosphoenolpyruvate carboxykinase, which is a key enzyme in succinic acid production (Zou, Zhu, Li, & Tang, 2011). As reflected in Fig 4E, when MgCO3 concentration was 60 g/L, the highest succinic acid titer was up to 28.97 g/L. When MgCO3 concentration was increased from 40 g/L to 60 g/L, succinic acid concentration was also increased simultaneously. However, further increase in MgCO3 concentration reduced succinic acid production. For instance, when MgCO3 concentration was increased to 80 g/L, succinic acid titer was just 17.30 g/L.
3.4 Optimization of fermentation conditions for enhancedsuccinic acid production from xylan using microbial co-cultivation system containing strain M5 and 130Z
Based on the one-factor-at-a-time (OFAT) experiments, the optimal conditions for three important factors (mixed time, initial xylan concentration, and MgCO3 concentration) were determined using response surface methodology (RSM). For the response surface analysis, 17 experiments with triplicates were conducted according to RSM design as shown in Table 1, where X1 denotes initial xylan concentration (g/L), X2 denotes mixed time, X3 denotes MgCO3 concentration, and Y is succinic acid production (g/L). According to multiple regression analysis of experimental data, second-order polynomial equation of empirical relationship between response and screening variables was obtained as follows:
(2)
The adequacy of this model was further verified using analysis of variance (ANOVA), which was tested using Fisher’s statistical analysis (Table 2) (Isar, Agarwal, Saran, Kaushik, & Saxena, 2007). Value of ”Prob > F” was less than 0.0500, indicating that model terms were significant (Bezerra, Santelli, Oliveira, Villar, & Escaleira, 2008). The lack of fit F-value of 1.71 implied that the lack of fit was not significant. The non-significant lack of fit suggested that obtained experimental responses sufficiently fit with the model (Ganjali Dashti, Abdeshahian, Wan Yusoff, Kalil, & Abdul Hamid, 2014). The response surface is a three-dimensional spatial surface map, which can visually reflect the interaction between various factors and response values, and show the relationship between the observed response value and test parameter level (Cui et al., 2009). The above multiple regression equation was used to make the response surface curve. As shown in Fig 5A-C, there was interaction between each pair of variables, and significant interaction between the three selected variables. The optimal succinic acid production value predicted from the response surface model was 32.39 g/L under the optimal conditions, in which xylan concentrations was 84.05 g/L, mixed time was 67 h and MgCO3 concentration was 62.35 g/L, respectively. To verify the accuracy of model prediction, three sets of parallel experiments were carried out under the optimal conditions. As seen in Fig 5D, succinic acid production could reach 32.50 g/L with yield of 0.41 g/g, which was basically consistent with the model prediction value. In this microbial co-cultivation system, strain M5 was able to constantly hydrolyze xylan to xylose during first 67 h, and approximately 12.23 g/L of xylose occurred in the fermentation broth before strain 130Z was inoculated. After dosage of strain 130Z, xylose could be immediately consumed, and succinic acid was accumulated. Owning to the constant hydrolysis of xylanase and β-xylosidase, xylose concentration in the fermentation broth was maintained at 6-7 g/L. In addition, the activities of xylanase and β-xylosidase were always maintained at 0.38-0.43 U/mL and 6.02-7.34 U/mL, respectively during the co-cultivation process. Before 96 h, succinic acid production increased exponentially, and the highest succinic acid titer of 32.50 g/L was obtained at 240 h. Taken together, this microbial co-cultivation system could not only reduce feedback inhibition of enzymes, but also make it possible to produce succinic acid directly from xylan through CBP.