Microbial production of cellulases and hemicellulases using distillers’ dried grains with solubles (DDGS) as the feedstock
Restricted (Penn State Only)
- Author:
- Iram, Attia
- Graduate Program:
- Biorenewable Systems
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- June 02, 2022
- Committee Members:
- Ryan Elias, Outside Unit & Field Member
Juliana Vasco-Correa, Major Field Member
Ali Demirci, Chair & Dissertation Advisor
Daniel Ciolkosz, Major Field Member
Deniz Cekmecelioglu, Special Member
Suat Irmak, Program Head/Chair - Keywords:
- Distillers’ dried grains with solubles (DDGS)
DDGS
Culture Optimization
Pretreatment
Lignocellulosic biomass
Lignocellulolytic enzymes
Cellulases
Hemicellulases
Media Optimization
RSM - Abstract:
- One of the most prominent challenges to the cost-effective production of lignocellulosic biofuels at industrial scales is the need for a sustainable, environment-friendly, and effective method of converting lignocellulosic polysaccharides into sugars. One such method is to utilize lignocellulolytic enzymes such as cellulases and hemicellulases to convert lignocellulosic biomass into simple sugars for subsequent fermentation into biofuels. Currently, such enzymes are expensive and of low quality. In addition, they need an inexpensive feedstock with high fiber and protein content. Distillers’ dried grains with solubles (DDGS), which is a byproduct of the corn ethanol industry, can be used as an economical feedstock for enzyme production due its abundance and high fiber and protein content. However, with any feedstock, the determination of ideal production logistics such as the selection and optimization of a pretreatment method, selection of best microbial strain(s), media augmentation, and optimization of culture conditions is crucial for its use at industrial scales. Therefore, present study is undertaken to evaluate DDGS’s potential as an ideal feedstock for lignocellulosic enzyme production along with various culture enhancement strategies to increase the quality of such enzymes for their final application in an integrated biorefinery where such enzymes can be produced onsite and then used in the production of lignocellulosic ethanol. This research consisted of eight phases. As the first phase, three pretreatment methods namely dilute acid hydrolysis, soaking in liquid ammonia (SAA), and semi-continuous steam explosion were selected from the literature for their further evaluation to release total reducing sugars (TRS) and inhibitory products from DDGS. Upon the treatment with the selected conditions for each treatment, it was determined that dilute acid hydrolysis releases more TRS as compared to the other two pretreatment methods. Based on these results, a central composite response surface model (R2=0.91) was developed and verified for 0.341±0.010 g TRS/ g DDGS with the elimination of conditions that produce higher than tolerable amounts of inhibitory products. The optimum conditions for dilute acid pretreatment were 13.2% (w/v) solid load, 5% (w/v) sulfuric acid concentration, and for 20 minutes for 100ml acid slurries. SAA and semi-continuous steam explosion treatment produced 0.129 and 0.055 g TRS/ g DDGS, respectively. The second phase of the study was undertaken to select the best microbial strains capable of producing high amounts of cellulases and hemicellulases using acid hydrolyzed DDGS slurry as the main feedstock. Eleven high producers of lignocellulolytic enzymes were selected from the literature to see their enzyme production levels in the DDGS media. Among them, eight (five Aspergillus niger and three Trichoderma reesei strains) were fungal strains and three were bacterial (Bacillus subtilis) strains. After 18 days of incubation, it was observed that fungal strains showed higher enzyme activities (p<0.05) as compared to the bacterial strains for both cellulase and hemicellulases. Among all the fungal strains, A. niger (NRRL 330), A. niger (NRRL 567), A. niger (NRRL 1956) and T. reesei RUTC-30 (ATTC 56765) demonstrated better stable enzyme activity production than the others. The third phase of the study was designed to evaluate the selected four fungal strains from Phase 2 for the enzyme production after the selected pretreatments in Phase 1. For this purpose, DDGS without any treatment, diluted acid-treated DDGS, and steam-treated DDGS were used as the fermentation media for the four fungal strains mentioned above, and enzyme activities for each strain were determined within 15 days of the fermentation period. Dilute acid pretreatment was significantly better (p<0.05) than untreated DDGS and steam-treated DDGS for the production of hydrolytic enzymes. In addition, conventional carbon sources such as glucose and crystallized cellulose were also evaluated along with treated and untreated DDGS to check their effectiveness as the carbon source for enzyme production. Acid hydrolyzed DDGS demonstrated better results than all other carbon sources (p<0.05). The fourth phase was undertaken to evaluate the effect of minerals (salts) and nitrogen sources on enzyme production using DDGS as the main carbon source and to enhance the enzyme production further by optimizing the effective nutrient components in the first part of Phase 4. The four fungal strains that were selected in Phase 2 and used in Phase 3 were also used in this phase. The results showed that salts do not have a significant positive effect on enzyme production for most of the fungal strains. Later, three nitrogen sources namely yeast extract, peptone, and ammonium sulfate were further optimized by Box-Behnken RSM design for A. niger (NRRL 330) and A. niger (NRRL 567), because these two strains demonstrated higher enzyme activity levels than other strains in the study. A statistical model with a composite desirability value of 0.97 was recommended consisting of 16.5 g/L yeast extract, 5 g/L peptone, and 1.9 g/L of ammonium sulfate for A. niger (NRRL 330). Maximum experimentally verified enzyme activity was 0.54±0.02 IU/ml for cellulases and 48.71±2.05 for xylanase. These activity levels were increased from 0.28 and 27.55 IU/ml respectively thus showing a 93% increase in cellulase activity and 76% increase in the xylanase (hemicellulase) activity. Therefore, A. niger (NRRL 330) was selected for further evaluation and culture optimization in the later stages of this research. In Phase 5, the effect of dilution factor, agitation, and aeration in 2 L bench-top bioreactors was evaluated with A. niger (NRRL 330) and dilute acid-treated DDGS slurries. At this point, the DDGS were not filtered out of the acid slurry and the dilution was accomplished by adding DI water into the DDGS media. The enzyme production levels gradually increased from day 1 to day 9 of the fermentation. A sharp increase in the enzyme activities was observed with higher agitation and aeration rates. The highest cellulase activity (0.76 IU/ml) and hemicellulase activity (29.04 IU/ml) were observed at the lowest dilution rate with the highest aeration (1 vvm) and agitation (500 rpm) rates at later days of fermentation. The sixth phase of this study was undertaken to optimize the fermentation parameters for enzyme production using statistical optimization strategies in 2 L benchtop bioreactors. Based on the results of the previous phase, aeration (0.5-2 vvm) and agitation rates (100-500rpm) were selected for optimization along with inoculum size (1-10%). The optimization increased the enzyme production to 0.82 IU/ml for cellulase and 52.76 IU/ml for xylanase. The composite desirability was 0.76 for this model. The optimized conditions were determined as to be 6.5% inoculum size, 310 rpm agitation rate, and 1.4 vvm aeration. The results showed a further 52% increase in cellulase activity and a 9% increase in xylanase activity from the shake flask study. One problem with using a fungal strain for and especially with A. niger is that it produces high levels of biomass and mycelial clumps, which can hinder enzyme production in submerged fermentation. Therefore, the application of microparticles such as aluminum oxide and magnesium silicate (talc) was then evaluated in Phase 7 of the study. The aluminum oxide and talc concentrations of 0, 5, 10, 15, 20, 25, and 30 g/L with optimized nitrogen sources of Phase 4 added to the DDGS media. Samples were incubated with A. niger (NRRL 330) for nine days in shake flask fermentations. The addition of 5 g/L aluminum oxide increased cellulase production from 0.52±0.02 to 1.15±0.04 IU/ml on the ninth day. On the other hand, the addition of 10 g/L of ammonium oxide increased xylanase production from 11.65±0.1 to 21.73±1.98 IU/ml on the third day of fermentation. These results also show that aluminum oxide was better microparticle in terms of enzyme production as compared to the talc microparticles. In addition, the addition of 10 g/L of aluminum oxide microparticles in the 2 L bioreactor vessels increased the cellulase activity to 0.93±0.01 IU/ml on day 9. As the final phase (Phase 8), technoeconomical analysis (TEA) was performed for the production of DDGS in a commercial scale corn ethanol biorefinery, which is used for the onsite production of cellulase and hemicellulases and using those enzymes to make more ethanol from corn stover in an integrated first generation and second-generation (1G/2G) biorefinery. A TEA model was developed using SuperPro Designer software with a feedstock capacity of 128 ton/h of corn grain and 110 tons/h of corn stover. All the equipment, utilities, and raw material prices were estimated from previous TEA models developed by the U.S. Department of Agriculture, Agricultural Research Service (USDA-ARS), and National Renewable Energy Laboratory (NREL) with the implementation of proper inflation rates. The integrated production will result in the unit production cost of $0.96/kg ($0.75/l or 2.84/gal) of ethanol with $1.05/kg ($0.82/l or $3.10/gal) of lignocellulosic ethanol. The results show the economic feasibility of producing enzymes from DDGS and using those enzymes for the conversion of lignocellulosic biomass into ethanol. Overall, this research shows the potential of DDGS as the microbial feedstock for enzyme production. DDGS produced more cellulases and hemicellulases compared to conventional carbon sources such as glucose, because DDGS has cellulose and hemicellulose fibers which after pretreatment can induce enzyme production. In addition, DDGS also has protein and other nutritional components which can further help in enzyme production.