Effect of Novel Engineered Aspergillus oryzae on Soybean Meal Fermentation and Optimization of Cultivation Conditions

Corresponding Author: Xiaochun Yu College of Food Science and Engineering, Tonghua Normal University, 950 Yucai Road, Tonghua, Jilin 134002, P.R. China Tel:+86 13766144076 Email: yuxc@thnu.edu.cn Abstract: The filamentous fungus Aspergillus oryzae is an important strain in fermentation industry. In this study, to test the possibilities of improving polypeptides production, the soybean meal fermentation products were used to investigate the efficiencies of four different novel recombinant engineered A. oryzae strains A. oryzae subC, A. oryzae asp, A. oryzae Ap and A. oryzae pep. Gel filtration was employed to separate the products produced by the four engineered strains and wild-type A. oryzae and the antioxidant activities of each component of the products were compared. The results showed that the proportion of small-molecule peptides produced by A. oryzae pep fermentation was higher, mainly in the range of 1000-2000 Da, when compared with that yielded by wild-type A. oryzae fermentation. The antioxidant activities of the different components of the fermentation products produced by various engineered strains, especially component I of A. oryzae pep, showed varying degrees of improvement on DPPH and ·OH free radical scavenging activities, reaching 73.1%±0.9% and 81.4%±1.2%, respectively. To further enhance the fermentation performance of A. oryzae pep, response surface methodology was used to optimize the fermentation conditions. The results indicated that the polypeptide conversion rate reached 40.4% following fermentation with 8% inoculum for 108 h at 31°C. These findings are significant for the development of A. oryzae strains and high-value utilization of soybean meal resources.


Introduction
As a filamentous fungus belonging to the genus Aspergillus, Aspergillus oryzae has been traditionally used for more than 1000 years for brewing foods such as soybean paste and soy sauce (Abe et al., 2006). Owing to its easy cultivation and absence of aflatoxin secretion, A. oryzae is widely used in food processing and industrial enzymes production (Barbesgaard et al., 1992;Brown et al., 2013). A. oryzae mainly secretes neutral protease, with relatively low alkaline protease and acid protease activities, thus, presenting low degradation and utilization of raw protein (Zhao et al., 2014). To meet the needs of the fermentation industry, researchers are committed to develop new A. oryzae strains with high protease activity to improve raw protein utilization and composition of fermentation products (Ortiz et al., 2016;Xu et al., 2011;Fang et al., 2009). Genetic engineering has been reported to significantly enhance the protease activity of A. oryzae (te Biesebeke et al., 2005;Hunter et al., 2013;Olempska-Beer et al., 2006).
As the byproduct of oil processing industry, soybean meal has high protein content and is an excellent raw protein material (Zheng et al., 2017;Rumsey et al., 1994;Mukherjee et al., 2016). However, soybean meal is currently mainly used as feed, with only a small part of it employed in fermentation and the majority of the protein material not being effectively utilized, resulting in wastage of resources (Lawrence et al., 2003;Ortega et al., 2016;Nath et al., 2017). Thus, the use of soybean meal as a raw material for further processing and comprehensive utilization has significant prospects Liu et al., 2017;Yu et al., 2018). Response Surface Methodology (RSM), which can accurately express the relationship between factors and response values, is an effective method to optimize reaction conditions. Because of its reasonable design and excellent results, it has been widely used in chemical industry, biological engineering and food engineering (Dai et al., 2018). In recent years, Box-Behnken design, an RSM with less factors and experiments, has gained more attention in the food industry, especially in the process of fermentation and optimization of operation conditions (Gunst, 2008).
In this study, engineered A. oryzae strains expressing different proteases were utilized for the fermentation of soybean meal. The molecular weight distribution and antioxidant activity of the fermentation products produced by different A. oryzae strains were compared and the fermentation conditions were optimized by RSM. The results of this study could help to develop high-yielding strains with desired properties, improve the yield of products and expand the use of soybean meal.

Strains and chemicals
Wild-type A. oryzae (huniang 3.042) preserved in our laboratory and four strains of engineered A. oryzae subC, A. oryzae asp, A. oryzae Ap and A. oryzae pep, which had been constructed before and preserved in our laboratory, were used in this study (Yu et al., 2017). The four genetically engineered A. oryzae strains expressed different proteases, respectively. Sephadex G-25 was obtained from Pharmacies Technology Co. Trypsin inhibitor (Mr 6511), vitamin B12 (Mr 1355), glutathione (Mr 307), DPPH and 3,5-dinitrosalicylic acid were purchased from Sigma Technology Co. Soybean meal was purchased from Yihai cereal oil.

Activation and Cultivation of Strains
The wild-type and four engineered A oryzae strains were activated on Potato Dextrose Agar slant medium (PDA) and cultured at 30C for 3 days. The PDA medium contained the following: 20.0% potato, 2.0% agar and 2.0% glucose (sterilized at 121C for 20 min). Then, the spores were collected, washed with 10 mL of deionized water (sterilized). The spore solution was prepared and suspended at a concentration of 10 8 spores/mL  and inoculated into solid fermentation medium (100 g) at 2% inoculum volume and 30C for 4 days. The solid-state fermentation medium comprised the following: 0.1% KH2PO4, 0.04% MgSO4•7H2O, 1.0% soluble starch and 1:1 soybean: Water. After the fermentation process, the fermented products were dried to constant weight using oven at 60C for further analysis.

Determination of Molecular Weight Distribution
Samples of fermentation of different strains were extracted with deionized water in a water bath at 50C for 6 h. To inactivate the enzyme activity, the extracts were heated in an oven at 121 C for 10 min. The supernatant was obtained by centrifugation (Sorvall Stratos, Thermo, USA) at 3000 ×g for 10 min and filtered by ultrafiltration (14-A, Ruili Shanghai, China) using an ultrafiltration membrane (cut-off molecular weight, 10 KDa) at P = 0.12 MPa (Ma et al., 2001). Subsequently, the filtrate with molecular weight less than 10 kD was collected, precooled for 2 h at −40 C in a vacuum freeze dryer (Scientz-30ND, Xinzhi Ningbo, China), freeze-dried for 10 h at a vacuum degree of less than 5 Pa and the samples were collected.
The molecular weight distribution of the peptides of the fermented products was determined by gel filtration (Xie et al., 2012). The column conditions comprised a flow rate of 1 mL/min and detection wavelength of 280 nm. As the elution buffer solution, 0.1 mol/L Tris-HCl (pH 7.0) was employed. Subsequently, 5 g of Sephadex G-25 (Pharmacies) were added to 100 mL of deionized water, placed at room temperature for 3 h for swelling, packed in a column with chromatographic system (MA99, Huxi Shanghai, China) and eluted with buffer solution for 1 h. The ultrafiltration samples of different strains with concentration of 5% (W/V) were prepared. Then, 2 mL of the samples were used for analysis under the above-mentioned conditions and the ultraviolet absorption of the effluent was recorded. The outflow volume (mL) was recorded with a measuring cylinder and the analysis was repeated again under the same conditions. Trypsin inhibitor (Mr 6511), vitamin B12 (Mr 1355) and glutathione (Mr 307) were used as standard molecular weight references.

Determination of Antioxidant Activity
In this study, the antioxidant activities such as reduction potential, DPPH radical scavenging rate and hydroxyl radical scavenging rate of each component of the fermented products produced by different strains were tested. FRAP and KCN were used to determine the reduction potential and the components of the fermented products produced by different A. oryzae strains were compared by evaluating the absorbance of the reaction solution (Yen and Duh, 1993). The absorbance value was measured at 700 nm and the reduction potential of the sample was characterized by the absorbance value. The scavenging effects of different samples on DPPH free radicals were calculated by the method developed by Parejo et al. (2000), by detecting DPPH·scavenging at a wavelength of 517 nm. Hydroxyl (OH) radicals were determined by salicylic acid method (Li et al., 2008) and their concentration was calculated by adding salicylic acid into the system and detecting absorbance at 510 nm.

Optimization of the Fermentation Conditions
RSM was applied to optimize the soybean meal fermentation conditions of engineered A. oryzae pep, which showed high content of small peptide and good antioxidant activity (Roldan-Carrillo et al., 2011). According to the design principle of the Box-Benhnken Center Combination Test, three factors that have significant influence on the polypeptides conversion rate were selected as follows: Fermentation time (X1), fermentation temperature (X2) and inoculum volume (X3). Each factor was set at three levels: Low, medium and high. The factors were coded as -1, 0 and 1 and the polypeptides conversion rate (Y/%) was considered as the measures. Design Expert 8.0 software was used for test design. The factors and levels were set as shown in Table 1. The polypeptides conversion rates achieved by fermentation with different A. oryzae strains are shown in Table 2.
The polypeptides content was determined using the biuret method (Cotton et al., 1997). In brief, 5 g of the fermentation product were mixed with 10 mL of deionized water and placed in a water bath at 50C for 6 h. The extracts were heated in an oven at 121 C for 10 min to inactivate the enzyme activity. Then, equal volume of trichloroacetic acid solution (15%) was added and the mixture was allowed to settle for 30 min to precipitate macromolecular proteins. The mixture was centrifuged at 5000 ×g for 10 min at 4 C (Sorvall Stratos, Thermo, USA) and the supernatant was stored in a 25-mL tube at constant volume. Subsequently, 1 mL of the sample solution with 5-fold dilution was mixed with 4 mL of biuret reagent and stored at room temperature for 30 min and the absorbance of the mixture was measured at 540 nm with spectrophotometer (7200, Unico, China). The polypeptides content was determined by comparing with the standard curve obtained using casein (Sigma, USA) as the standard protein and the polypeptides conversion rate was calculated (Liu et al., 2011) where soybean meal protein content is calculated as M×98.9%×42.6%, with M denoting the amount of sample, 98.9% indicating the dry matter content in soybean meal and 42.6% representing the nitrogen content in soybean meal determined by Kjeldahl method (Deacon, 1972). The equation of the standard curve is: y = 0.0854x + 0.0145.

Molecular Weight Distribution of Polypeptides in the Fermented Products
The results of gel filtration of the fermented products produced by different A. oryzae strains are shown in Fig. 1-5. The polypeptides in the fermented products produced by different strains were of various sizes and molecular weights and were continuously distributed. When compared with the fermented products produced by wild-type A. oryzae, the proportion of small molecular weight peptides in the fermented products produced by the engineered strains was higher, with A. oryzae pep exhibiting the largest peak area, indicating the higher percentage of small molecular peptides in the fermented products. When separated by chromatographic column, the peaks of the wild-type A. oryzae fermented products were at about 24 min, while those of A. oryzae Ap and A. oryzae pep fermented products were at about 42 and 48 min, respectively, indicating that the molecular weight of peptides in the fermented products of A. oryzae Ap and A. oryzae pep was lower than that of peptides in the fermented products of wild-type A. oryzae. The molecular weight distribution range of different components of the fermented products of each strain was determined from the molecular weight of standard protein. The molecular weight of the fermented product of A. oryzae pep was mainly in the range of 1000-2000 Da. The highest peak separation liquid of different strains fermented samples which the number of collection pipe is about 10-20 were collected and frozen dried respectively. These samples were recorded as gel filtration separation component I and used in subsequent tests.
Soybean polypeptide is a mixture of various peptides with different molecular weights. The molecular weight distribution of soybean polypeptide is closely related to its function. As peptide segments with different functional and physiological activities are often distributed in different molecular weight regions, it is crucial to understand the molecular weight composition of polypeptide (Suetsuna, 2000). It has been reported that during solid-state fermentation of soybean meal, microbial strains and their protease are involved in the degradation of soybean protein, increasing free amino acids, producing structural changes in soybean protein and improving the functional properties of soybean protein (Yang et al., 2013). Therefore, the microbial strains used in fermentation and their protease activity can affect the properties of the fermentation products. The results of the present study showed that the peptide composition of soybean meal fermented by different engineered strains of A. oryzae varied.

Antioxidant Capacity of the Fermented Products
Antioxidant capacity of fermented products is an important index, which can be evaluated by measuring the reduction ability of the products with potassium ferricyanide reduction system. The results of total reduction ability of each component of the fermented products produced by different A. oryzae strains are shown in Fig. 6. It can be observed from the figure that the total reduction ability of the fermented product After ultrafiltration and gel filtration, the total reducibility of the components of the fermented products presented 1.9fold increase, when compared with that of the unfiltered fermented product. It has been indicated that soybean meal proteins may exhibit antioxidant activity and that the activity of the polypeptide is closely related to its molecular weight (Xu et al., 2015;Suetsuna, 2000). When compared with the absorbance value of wild-type A. oryzae component I, the total reduction capacity of A. oryzae pep component I was 1.3-fold higher. The DPPH free radical scavenging activities of different components of the fermented products produced by different A. oryzae strains are shown in Fig. 7. It can be seen from the figure that the scavenging capacity of the fermented product extracts of wild-type A. oryzae, A. oryzae subC and A. oryzae asp showed only slight difference. The scavenging rates of A. oryzae Ap and A. oryzae pep extracts were slightly increased (24.3%±1.0% and 25.1%±1.0%, respectively), while that of component I and fermented product extract of A. oryzae pep presented the maximum value, reaching 73.1%±0.9%. The DPPH removal rate of A. oryzae pep ultrafiltrate (42%±0.9%) was 1.7-fold higher than that of the corresponding fermented extract (25.1%±1.0%). When compared with the fermented extract and ultrafiltrate, component I presented 2.9-and 1.7-fold higher clearance rate, respectively. These results showed that the DPPH scavenging activity of the fermented products improved with the decrease in the molecular weight of the polypeptides as well as increase in the proportion of small molecular peptides in the fermented products.
Currently, OH is the most active oxygen-containing free radical known and is also the most toxic free radical to living organisms (Chen et al., 2004). The scavenging effects of each component from different engineered A. oryzae strains on ·OH are shown in Fig.  8. It can be observed from the figure that the ·OH scavenging capacity of the fermented products produced by A. oryzae asp, A. oryzae Ap and A. oryzae pep was about 34%±0.9%, which was 1.5-fold higher than that produced by the wild-type A. oryzae. In particular, the scavenging capacity of the fermented products produced by A. oryzae subC only slightly differed from that produced by wild-type A oryzae, whereas the scavenging capacities of the fermented products and ultrafiltrate of A oryzae asp was higher than those of wild-type A. oryzae. Moreover, the ·OH clearance rate of the fermented product produced by A. oryzae pep reached 81.4%±1.2%.

Components of different engineered strain fermented products
Components of different engineered strain fermented products It has been reported that polypeptides derived from casein protein using pepsin and presenting molecular weights ranging from 913 to 2951 Da have better superoxide scavenging activity (Ibrahim et al., 2018). The antioxidant activity of polypeptides is the result of a combination of factors such as amino acid composition, amino acid sequence, molecular weight and spatial structure (Han et al., 2013). In the present study, the molecular weight and amino acid composition of polypeptides in the fermented products produced by A. oryzae pep were different from those of the polypeptides in the fermented products of wildtype A. oryzae, thus, resulting in varied antioxidant activities. Nevertheless, further studies on the amino acid composition and characteristics of polypeptides in the fermentation products produced by different engineered strains are needed.

Optimization of Fermentation Conditions
By using Design Expert 8.0 analysis software, the experimental data were fitted and analyzed and the quadratic polynomial regression equation was obtained as follows: Y = 42.16+1.05X1+0.73X2+1.79X3-0.25X1X2-0.4X1X3-0.99X2X3-7.8X1 2 -5.17X2 2 -0.89X3 2 . Table 3 shows the variance analysis of the regression model. It can be observed from the table that the P value of the model is significant, but the lack of fit is not significant, indicating the success of the experimental model. The regression coefficient R 2 of response surface was 0.9317, while the corrected Adj R 2 was 0.8088, revealing that the model had a good fit and small error and could be used to analyze and predict the fermentation conditions of soybean meal by A. oryzae. In the previous single-factor experiment, the effects of inoculum volume, fermentation time, fermentation temperature and moisture content in medium on polypeptide conversion rate were studied. The results showed that the influence of moisture content in medium on polypeptide conversion rate was not significant (P > 0.05). Therefore, fermentation time, fermentation temperature and inoculum volume were selected in optimization test of RSM. The F value of variance analysis indicated that the degree of influence of each factor on polypeptides conversion rate presented the following trend: Inoculum size > fermentation time > fermentation temperature. Figure 9 shows the response surface optimization of the conditions for fermentation with A. oryzae pep. It can be seen from the figure that the three charts of the response surface are convex with open ends facing downwards. The optimum fermentation conditions and fermentation results can be predicted according to the mathematical model established by the RSM. Under the conditions of fermentation time of 110 h, fermentation temperature of 31.5C and inoculum volume of 8%, the theoretical polypeptides conversion rate obtained by RSM was 42.5%. The fermentation conditions of validation experiment were adjusted considering the actual state of the fermentation experiment and feasibility of operation as follows: Fermentation time, 108 h; fermentation temperature, 31C; and inoculum volume, 8%. Under these adjusted conditions, the polypeptide conversion rate reached 40.4%, which is in good agreement with the theoretical value, indicating that the established model is effective. The polypeptide conversion rate was 2.4 times higher than that achieved with the wild-type A. oryzae (17.0%) and is higher than that reported in the literature. In a previous study using soybean meal as raw material, the soybean polypeptides conversion rate had reached 26.4% after fermentation with A. oryzae (Gao et al., 2014). In another study on soybean meal fermentation using Bacillus subtilis, the polypeptides conversion rate reached 31.5% after optimization of the fermentation process (Liao et al., 2012). Thus, by using appropriate methods to optimize the fermentation conditions and technology, the product yield can be improved to a certain extent. In recent years, RSM has been widely used in bio-fermentation, food technology optimization and other fields (Wu and Ahn, 2018). When compared with the wild-type A. oryzae, engineered A. oryzae strains could produce polypeptides from soybean meal fermentation with high efficiency and obvious advantages. The obtained soybean polypeptides products could be further used as substrates or functional additives in food production process and other industries.

Conclusion
In this study, the fermentation efficiencies of four different novel recombinant engineered A. oryzae strains based on the molecular weight and antioxidant activities of the polypeptides produced had been investigated. The results showed that the proportion of small-molecule peptides produced by engineered A. oryzae strains was higher than that produced by wildtype A. oryzae. The antioxidant capacity of the engineered strains, especially component I of A. oryzae pep, presented varying degrees of improvement.
Under optimized fermentation conditions, the polypeptide conversion rate of A. oryzae pep reached 40.4%, which was 2.4 times higher than that of the wild-type A. oryzae. These results may help to develop efficient A. oryzae strains and effective utilization of soybean meal resources.