VITAMIN K2 (MENAQUINONE-7) PRODUCTION BY BACILLUS SUBTILIS NATTO IN A BIOFILM REACTOR
Open Access
- Author:
- Mahdinia, Ehsan
- Graduate Program:
- Agricultural and Biological Engineering
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- December 04, 2017
- Committee Members:
- Ali Demirci, Dissertation Advisor/Co-Advisor
Ali Demirci, Committee Chair/Co-Chair
Virendra M. Puri, Committee Member
Jeffrey M. Catchmark, Committee Member
Ryan J. Elias, Outside Member
Aydin Berenjian, Special Member - Keywords:
- MK-7
Menaquinone-7
Vitamin K
Biofilm reactor
Bacillus subtilis natto
Biofilm Reactor - Abstract:
- Menaquinone-7 (MK-7), a subtype of vitamin K, has received significant attention due to its effect on improving bone and cardiovascular health. Current fermentation strategies, which involve static fermentation without aeration or agitation, are associated with low productivity and scale-up issues and hardly justify the commercial production needs of this vitamin. Previous studies indicate that static fermentation is associated with pellicle and biofilm formations, which are critical for MK-7 secretion while posing significant operational issues. Therefore, the present study is undertaken to evaluate the possibility of using biofilm reactors as a new strategy for MK-7 fermentation. In the first phase of the study, 15 Bacillus species namely Bacillus subtilis natto, Bacillus licheniformis, and Bacillus amyloliquifaciens as well as 4 plastic composite supports (PCS) were investigated in terms of MK-7 production and biofilm formation. Results show the possibility of using a biofilm reactor for MK-7 biosynthesis. Bacillus subtilis natto strain NF1 and soybean flour yeast extract PCS (SFY) in glucose medium were found as the most efficient combination for production of MK-7. In the following phase, biofilm reactors were constructed using the selected Plastic Composite Support (PCS) and B. subtilis natto strain NF1 for MK-7 production. Using response surface methodology (RSM), optimum growth parameters including temperature, pH, and agitation were determined in a glycerol-based medium. Results were presented in a statistical model (R2=0.90), leading to optimum growth conditions of temperature (35°C), agitation (200 rpm) and pH (6.58). Model predicted MK-7 concentration was validated and MK-7 concentration of 12.1±1.2 mg/L was produced in the biofilm reactor. The obtained concentration was 58% higher as compared to the suspended-cell culture (7.7±1.5 mg/L). Then, in order to optimize the same fermentation growth parameters for MK-7 production in the glucose-based medium, Central Composite Design (CCD) was carried out along with supplementary runs to determine the optimum conditions. The biofilm reactors were able to produce a maximum concentration of 18.4±0.8 mg/L of MK-7, which was 237% higher than the suspended-cell fermentation. The next phase was undertaken to utilize biofilm reactor by optimizing the components in the glucose-based medium. Response Surface Methodology (RSM) was used to determine optimum concentrations of three major medium components (glucose, yeast extract, and casein). Maximum MK-7 concentration in biofilm reactors was achieved as 20.46±0.51 mg/L, which was 344% higher compared to the suspended-cell reactors containing the same optimum media composition. Later, optimization phases were undertaken to utilize biofilm reactors in investigating and optimizing different media components in the glycerol-based medium. By using Response Surface Methodology (RSM), the effects of glycerol, yeast extract, and soytone were studied in the fermentation medium on MK-7 production in biofilm reactor. With a composition of 45 g/L of glycerol, 5 g/L of yeast extracts, 10 g/L of soytone and 0.06 g/L of K2HPO4, MK-7 concentrations could reach 14.7±1.4 mg/L in biofilm reactors, which was 57% higher compared to the MK-7 concentration achieved in suspended-cell reactor under similar conditions. While glycerol was depleted by the end of the fifth day in the biofilm reactor, it was never depleted in the suspended-cell reactor. Evidently, biofilm reactors present a reliable strategy likely to mitigate the operational issues with MK-7 biosynthesis at the industrial scale. Then, fed-batch strategies were investigated for glucose and glycerol-based media, as carbon source addition seemed crucial in batch fermentations. Results indicated that fed-batch additions can be significantly effective in glucose-based medium, increasing the end product concentrations to 28.7±0.3 mg/L of MK-7 which renders the biofilm reactors a potential replacement for static fermentation strategies with a maximum 32.5±0.4 mg/L of MK-7. Moreover, morphological changes of the applied B. subtilis strain was tracked during the 12 day long runs and finally, SEM investigations confirmed robust biofilm and extracellular matrices formed on the Plastic Composite Supports (PCS) in the biofilm reactors. In conclusion, biofilm reactors especially with fed-batch fermentation regimes seem to be an effective tool to enhance MK-7 productions on industrial scales. In order to reach the desired MK-7 quality, several downstream processing including extraction, drying, ultrasonication, etc. must be carried out after its biosynthesis in the broth. These processes, however, need to be carried out in such way to ensure least amount of losses and maximum recovery into the end-product. Therefore, in this phase of the study, drying, storage, and ultrasonication steps were evaluated under different conditions. Results showed that drying under forced air flow is not only fastest, but also demonstrated a better preservation of the vitamin and should replace the vacuum drying. Ultrasonication for 15 minutes seem to be harmless and sufficient for phase transition in analysis. Also, storage at refrigerated temperatures seem to preserve MK-7 at least for one week. Static liquid fermentations were conducted in McCartney bottles to explore the maximum MK-7 secretion potentials in different glycerol and glucose-based media compositions that were optimized in our previous studies. Maximum 32.5±0.4 mg/L and 14.6±0.4 mg/L concentrations were achieved in glycerol and glucose-based media respectively. Furthermore, fermentations in deeper culture tubes indicated how the MK-7 concentrations are distributed in different zones of the static liquid broth. Results in general, showed a clearer road map to ensuring better quality and preservation of the valuable end-product and enlightened more the path to further scaling up the fermentation process when compared with results obtained in optimized biofilm reactors in the previous steps. Last phase in the study was to mathematically model the findings in batch fermentations in the biofilm reactors and thus further elucidate the conditions governing fermentation. The logistic equation was modified to correlate substrate consumption with fermentation time and was utilized to model the substrate consumption in the four batch fermentations. Results indicated very accurate fits and therefore there is no need for more complex equations (R^2>0.953). Then, this successfully modified-logistic equation was inserted into the basic Luedeking-Piret equation modifying it to model MK-7 production based on substrate consumption. Furthermore, modified-Gompertz model was also used for the same purpose. Results indicated more accurate fits by the modified Luedeking-Piret equation (R^2=0.9705,0.943,0.970,and 0.959) compared to the modified-Gompertz (R^2=0.914,0.943,0.949 and 0.860). Yet, the modified Luedeking-Piret equation was a more complex model compared to the modified-Gompertz. In summary, bench-top biofilm reactors were successfully constructed using suitable PCS and B. subtilis natto combination. Growth parameters including temperature, pH and agitation along with medium components including carbon and nitrogen sources in glycerol and glucose-based media were optimized using RSM. Results indicated significant enhancement of MK-7 production in biofilm reactors compared to suspended-cell bioreactors by up to 344%. Furthermore, the fermentation profiles indicated robust carbon source consumptions leading to substrate depletions around halfway of the fermentations. Thus, fed-batch substrate additions were investigated and results indicated up to 28.7±0.3 mg/L MK-7 concentrations in the glucose-based medium with fed-batch additions. These concentrations are comparable with maximum concentrations coming from static fermentation counterparts (32.5±0.4 mg/L). Finally batch results were mathematically modeled and optimum processing procedures were investigated. The result from this study show a potential for larger pilot scale fermentations in biofilm reactors and investigating downstream and in-situ recovery techniques on pilot scales.