![]() |
|
![]() |
|
Production of Cellulases in Tobacco and Potato Plant Bioreactors
Objective: The expression of Acidothermus cellulolyticus endoglucanase (E1) gene and Trichoderma reesei cellobiohydrolase (CBH1) gene in transgenic tobacco (Nicotiana tabaccum) and potato (Solanum tuberosum) was examined in this study. Our primary goal was to test the feasibility of plant bioreactors for economically viable production of fungal and bacterial cellulases. Approach/Background: CBH1 and E1 produced in transgenic tobacco plants were found to be biologically active and to accumulate in leaves at levels of up to 0.05% and 2.6% of total soluble protein, respectively. The biochemical characteristics of plant-based recombinant E1 enzyme were similar to those of natural E1 purified from bacterial culture. In addition, plant-derived E1 resisted plant proteolysis at different developmental stages. Transgenic plants exhibited normal growth and developmental characteristics with photosyn-thetic rates similar to those of untransformed SR1 tobacco plants. Based on E1 activity and E1 protein accumulation in leaf extracts, E1 expression in potato is much higher than that measured in transgenic tobacco bearing the same transgene constructs. E1 expression under the control of the RbcS-3C promoter was specifically localized in leaf tissues, while E1 was expressed in both leaf and tuber tissues under the control of the constitutive Mac promoter. This suggests dual-crop applications in which potato vines serve as enzyme production bioreactors while tubers are preserved for culinary applications. Economic analysis demonstrated that the cost of large-scale E1 production, based on potato-vinebased expression, could be as low as $1.40/kg enzyme. Status/Accomplishments: FY 1998: E1 endo-glucanase expression was optimized in tobacco plants. Seventeen separate production lines were tested for production of active enzyme. The most prolific plant lines produced up to 1.3% of the total soluble protein in tobacco. FY 1999: E1 endoglucanase expression was further optimized in dual-crop potato plants, using genetic regulatory elements that limited transgene protein production to leafy vines, while leaving the tubers unchanged. In potato systems, the production level was increased to 2.6% of the total soluble protein in potato leaves. In addition, E1 endoglucanase enzyme was characterized in tobacco plants developed in FY 1998. Publications and Presentations:
Summary Date: March 2000 |
|
Improved Cellulase Enzymes at NREL
Objective: Protein engineering is being used to develop technologies able to deliver cellulase to the bioethanol process at a cost of $0.07$0.10/gal ethanol produced. Approach/Background: Reducing the cost of cellulase enzyme is critical to meet ethanol production cost targets. This conclusion was drawn from a series of industrial colloquies and a recent Cellulase Research Assessment Workshop. The review panel concluded that cellulase systems with higher specific activities than current fungal cellulases are required to meet targeted objectives for FY 2010 and beyond. The measure of success for cellulase technology in this midterm context is one capable of producing the required cellulase activity for one-tenth the cost currently estimated. The key to this strategy is to increase the specific activity of cellulases (thus deriving more activity per gram of protein produced) the feedstoci is pretreated biomass. A ten-fold increase should be both rewarding from a process perspective and technically feasible. The second issue, equally important, is the development of an expression system capable of producing large quantities of recombinant enzymes at low cost and preferably from low-value processing plant streams. Our primary research goal remains reducing the cost of cellulases acting on pretreated biomass used for bioethanol production by increasing the specific activity of the enzymes in the cellulase complex. Previously, we showed that EI functions with a high degree of synergism with fungal exoglucanases, especially Trichoderma reesei CBH I. Furthermore, we have recently shown that the artificial ternary cellulase system consisting of Acidothermus cellulolyticus EI, T. reesei CBH I, and T. fusca E3 is capable of releasing 85% as much reducing sugar from crystalline cellulase as the native T. reesei ternary system (i.e., EGI, CBH I, and CBH II). This result is encouraging for the ultimate success of engineered cellulase systems, because the artificial enzyme system described above was necessarily tested at 50oC, a temperature far below that considered optimal for one key component, EI, in order to spare the more heat-labile exoglucanases, CBH I and E3. The objective of our FY 1999 annual operating plan built upon the accomplishments of prior years by targeting further improvements using enzyme engineering in EI, as well as T. fusca E3 (by subcontract) and T. reesei CBH I. It is well recognized that the roles played by the enzymatic components in fungal cellulase systems must be understood before better artificial systems can be designed. We will use new methods in two-dimensional electrophoresis to begin mapping these cellulase component enzymes. It is hoped that correlations between inducers, cellulosic substrate, and specific activity of the resulting cellulase complex and the ratios of enzymes produced can eventually be used to improve the design of cellulase cocktails. Status/Accomplishments: Site-directed mutagenesis of Acidothermus EI: Our continuing key performance objective is to improve the kinetic efficiency of A. cellulolyticus EI on pretreated biomass. The native structure of biomass will be modified, using known principles of enzyme engineering, by targeted amino acid replacement to ensure optimal surface interaction of enzyme and cellulose (biomass). We succeeded in building polymerase chain reaction (PCR) mutation followed by DNA insert sequence confirmation and recombinant enzyme purification) and testing eleven new rEI enzymes with combined (double and triple) mutations identified last year as individually increasing performance. Unfortunately, most combined mutants displayed activity equivalent to the control (native EI and CBH I) or to a single mutation site, and, in some cases, the new combined mutants had less activity than that found for the single site mutations. In no case did the combined mutations yield a digestion value higher than that conferred by the beneficial EIY245G mutant alone. We thus conclude that for the classes of mutations tested in FY 1998 and FY 1999, i.e., those amino acids that directly interact with the cellodextrin substrate, nature may have developed a nearly optimal structure for this enzyme. Fungal expression of EI: A fungal expression system was developed for the production of the catalytic domain of Acidothermus cellulolyticus EI (EIcat). Aspergillus awamori was selected because of our recent successes in expressing T. reesei CBH I with this system and its capacity for the synthesis and secretion of foreign proteins. The A. awamori expression system provides an excellent opportunity to produce large quantities of EIcat protein for additional studies and for subcontractor evaluation. To achieve EI expression in A. awamori, mutagenic PCR was used to introduce Not1 and Xba1 restriction sites on the N-terminus and C-terminus coding sequence of the enzyme. The PCR product was then ligated into fungal shuttle vectors, followed by transformation of E. coli XL1 blue cells. Restriction digest analysis confirmed that the transformants contained the EI coding sequence and sufficient DNA was produced and purified for fungal transformation. Spheroplasted A. awamori cells were transformed. Western blot analysis indicated that we successfully transformed A. awamori with the EI gene and that the expressed protein is the same apparent molecular weight as the catalytic domain produced in E. coli. Improving T. reesei CBH I by protein engineering: The coding sequence for T. reesei CBH I has been successfully inserted and expressed in A. awamori using an E. coliAspergillus shuttle vector that contains elements required for maintenance in both hosts. This construct permits site-directed mutagenesis in E. coli and expression of new mutant proteins in A. awamori. The CBH I gene is expressed under the control of the A. awamori glucoamylase promoter and includes the glucoamylase secretion signal peptide. In order to have the signal peptide properly cleaved during secretion, the initial construction of this plasmid required the addition by PCR of a Not1 site and Xba1 site to the native coding sequence of CBH I. The Not1 site addition resulted in a change of the N-terminal-most amino acid in the protein (glutamine) to glycine. This glycine has subsequently been changed back to the native glutamine using site-directed modifications of the construct. This new construct was successfully used to transform A. awamori and express rCBH I as confirmed by Western blot analysis of culture supernatant. The recombinant CBH I expressed in A. awamori tends to be slightly overglycosylated as shown by the higher molecular weight, observed on western blot analysis (ca. 15%). It is important to note that this level of glycosylation is much less severe than that encountered for the Pichia pastoris rCBH I. The notion that this increased molecular weight is due to glycosylation has been confirmed by digestion of the recombinant protein by endoglycosidases. Following digestion with endoglycosidases H and F, the higher molecular weight form of the protein was shown to collapse to a molecular weight consistent with native CBH I. T. reesei cellulase induction and fingerprinting: Understanding the roles and relationships of component enzymes from the Trichoderma reesei cellulase system acting on complex substrates is key to the development of efficient artificial cellulase systems for the conversion of lignocellulosic biomass to sugars. Polysaccharide-degrading enzymes produced by T. reesei have been subjected to fingerprint analysis by high-resolution two-dimensional gel electrophoresis. Forty-six spots from a 11 x 25 cm Pharmacia gel loaded with a commercial cellulase preparation have been subjected to matrix-assisted laser desorption ionizationtime of flight (MALDI-TOF) mass spectrometry for direct peptide sequence analysis. Thirty-four of these species are identifiable as T. reesei enzymes and 11 are unidentifiable (they do not correlate with the current database). Proteonomics analysis of this gel indicates tremendous fragmentation and microheterogeneity throughout the gel. Most dominant glycosyl hydrolases exhibit both charge and mass heterogeneity. More troubling is that most gel spots showed multiple speciation. After ruling out sampling errors, we conclude that this level of two-dimensional separation is not sufficient to separate all protein species contained in an aged commercial fungal preparation. Understanding the roles and relationships of component enzymes from the Trichoderma reesei cellulase system acting on complex substrates is key to the development of efficient artificial cellulase systems for the conversion of lignocellulosic biomass to sugars. Publications and Presentations:
Summary Date: March 2000 |
|
Develop and Test Improved Thermomonospora fusca Cellulase Systems
Objective: This work attempts to improve three properties of E3 by altering its amino acid sequence: thermostability, resistance to feedback inhibition by cellobiose, and specific activity on pretreated biomass. Approach/Background: Several cellulases have been identified at NREL that are promising both for production by recombinant DNA technology and for their activity in the degredation of biomass cellulose. These cellulases are EI from Acidothermus cellulolyticus, CBH I from Trichoderma reesei, and E3 from Thermomonospora fusca. Even though T. fusca E3 is thermostable (100% activity after 16 h at 60oC), it is not as stable as EI. A three-dimensional structure for E3 is currently being determined at Cornell, and structures for two related enzymes, T. reesei CBH II and T. fusca E2, are already available to D. Wilson's laboratory. This information will be used to identify residues in E3 to be modified by site-directed mutagenesis. Status/Accomplishments: Site-directed muta-genesis of T. fusca exocellulase, Cel6B, began in 1998. The subcontractor produced and characterized a mutant containing a new disulfide linkage, which joined the two loops that form the active site tunnel. This did not increase stability or drastically inhibit activity, showing that opening the active site tunnel is not required for activity. A Trp residue located at the entrance to the active site tunnel was also mutated to Ala and to Tyr. The mutant enzymes show lower activity on crystalline cellulose, but the effect is not as large as was seen for T. reesei, Cel6A. A mutant of Cel6B with reduced inhibition by cellobiose was also produced. Publications and Presentations:
Summary Date: March 2000 |
|
Improve Thermomonospora fusca Cellulase Enzymes by Protein Engineering
Objective: This work will continue to improve three properties of E3 by altering its amino acid sequence: thermostability, resistance to feedback inhibition by cellobiose, and specific activity on pretreated biomass. Approach/Background: Several cellulases have been identified at NREL that are promising both for production by recombinant DNA technology and for their activity in the degredation of biomass cellulose. These are EI from Acidothermus cellulolyticus, CBH I from Trichoderma reesei, and E3 from Thermomonospora fusca. Even though T. fusca E3 is thermostable (100% activity after 16 h at 60oC), it is not as stable as EI. A three- dimensional structure for E3 is currently being determined at Cornell, and structures for two related enzymes, T. reesei CBH II and T. fusca E2, are already available to D. Wilson's laboratory. This information will be used to identify residues in E3 to be modified by site-directed mutagenesis. Status/Accomplishments: In 1999, the Sub-contractor continued site-directed mutagenesis of T. fusca exocellulase, Cel6B. In all, 18 mutants were produced and characterized. Six were in residues in the active site tunnel and twelve were in residues in the loops that enclose the tunnel. None of the mutants increased thermostability, showing that approaches that work for mesophilic proteins do not work well on a thermophilic protein. Two mutations increased activity on crystalline cellulose and a double mutant had higher activity than either single mutant on most substrates. Four mutations reduced feedback inhibition by cellobiose. The three-dimensional structure of Cel6B began to be determined in collaboration with J. Sakon at the University of Arkansas. Publications and Presentations: Zhang, S.B., K. Barr, and D.B. Wilson. 2000 (in press). Effects of non-catalytic residue mutations on substrate specificity and ligand binding of Thermomonospora fusca endocellulase E2. European Journal of Biochemistry. Summary Date: March 2000 |
|
Provide High Resolution X-Ray Structures of Selected Cellulases
Objective: Our objective is to develop high quality X-ray crystallographic structures of important cellulases identified by NREL or NREL subcontractors. Approach/Background: In the past, J. Sakon and P.A. Karplus analyzed the structure of A. cellulolyticus EI at 2.4-Å resolution using the Cornell biochemistry departments in-house rotating anode source. In 1999, unfrozen crystals were diffracted to 1.8-Å resolution using the Cornell high energy synchrotron source and a higher-resolution structure for EI was solved. Attempts were also made to improve the resolution of the liganded structure. Data from these structures was used to craft a mutation strategy for EI, and this site-directed mutagenesis work was conducted at NREL. Genetically improving an enzymatic activity can best be approached in a directed fashion if a reliable three-dimensional crystal structure is known for the target protein. X-ray crystallography provides the only known approach to the solution of this problem for a protein the size of the EI catalytic domain. Status/Accomplishments: When Tyr245 in endocellulase EI from Acidothermus cellulolyticus was changed to Gly (Y245G) by designed mutation at NREL, the enzymatic activity (measured as release of soluble sugars from biomass) was increased by 12% ± 2%. This is the first reported improvement in the specific activity of a cellulase by protein engineering. Using protein crystallography, computer modeling and enzyme kinetics, the subcontractor explained the enhanced activity on a molecular basis and concluded that the effect was due to reduced product inhibition. Refined structures for T. fusca E2 and the new endo/exocellulase E4 were also developed. Publications and Presentations:
Summary Date: March 2000 |
|
Perform Computer Modeling of Selected Cellulases and CBDs
Objective: In order to guide protein engineering studies, a series of molecular mechanics (MM) calculations of the A. cellulolyticus EI thermal tolerant endoglucanase and selected mutants designed to completely characterize binding specificity and active site dynamics will be undertaken. In a second, but related study, the interactions of T. reesei CBH I and the crystalline cellulose substrate will be modeled. Approach/Background: Molecular dynamics calculations, which can provide a precise microscopic description of molecular structuring and dynamical events in model carbohydrate-protein systems, involve the numerical integration of Newton's equations of motion for all atoms in a system as they move in response to the forces acting on them. These forces are computed directly from the derivatives of an empirical potential energy function of the atomic coordinates. In such calculations, physical observables (properties) are calculated as time averages over the various states that arise during the course of the simulation, since in principle the frequency of occurrence of each state and the time spent in each state will, for simulations of sufficient length, converge to the value determined by the Boltzmann distribution. Because atomic motions are simulated directly, entropic effects that do not appear in conformational energy minimizations are implicitly included in MD simulations, and the complete histories of the time sequence of events produced by MD calculations can provide rate information that cannot be obtained from Monte Carlo calculations. For a sufficiently realistic potential energy function, such simulations provide information of unparalleled detail about the microscopic behavior of a system, which often cannot be obtained by any currently available experimental methods. Status/Accomplishments: This project used Molecular Mechanics computer modeling calculations to probe the mechanisms of cellulase activity and substrate binding interactions, as a guide to mutagenesis experiments seeking to improve the activity of these enzymes. Molecular Dynamics simulations were used to explore the mechanism of hydrolysis of the E2 cellulase from Thermomonospora fusca and to explore the possibility of conformational changes in this protein upon substrate binding. Simulations were also used to explore the effects of point mutations in E1 from Acidothermus cellulolyticus upon substrate binding affinity. Publications and Presentations:
Summary Date: March 2000 |
|
![]() |
![]() Last updated: Wednesday, 30-Aug-2000 08:03:43 EDT |