Source: VIRGINIA POLYTECHNIC INSTITUTE submitted to
METABOLIC ENGINEERING OF CLOSTRIDIUM CELLULOLYTICUM USING A GENOME-SCALE MODEL AND COMBINATORIAL APPROACHES.
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
TERMINATED
Funding Source
AFRI COMPETITIVE GRANT
Reporting Frequency
Annual
Accession No.
0220088
Grant No.
2010-65504-20346
Project No.
VA-422108
Proposal No.
2009-02209
Multistate No.
(N/A)
Program Code
95150
Project Start Date
Dec 15, 2009
Project End Date
Jun 14, 2014
Grant Year
2010
Project Director
Senger, R. S.
Recipient Organization
VIRGINIA POLYTECHNIC INSTITUTE
(N/A)
BLACKSBURG,VA 24061
Performing Department
Biological Systems Engineering
Non Technical Summary
The need for renewable fuels to replace those from oil is well-established. At this point in time, ethanol from corn represents a grand achievement in this area. But, much of the public does not care to turn their food into fuels such as ethanol. The alternative that we are exploring is to turn the non-food part of plant materials into biofuels. Thus, we hope to replace the phrase "corn to ethanol" with "corn stalks to ethanol." In order to do this, we must first engineer a bacterium that can degrade this "cellulosic" plant material effectively and convert it to biofuel. Recent studies have concluded that in order to turn this into a sustainable process economically, the same bacterium must break-down the plant material, consume it, and turn it into a biofuel. To date, no bacteria exist that can do this efficiently. One that does exist, Clostridium cellulolyticum, grows much too slowly for an industrial process. One of our objectives is to map all of the chemical reactions that occur in C. cellulolyticum metabolism and apply mathematics to determine how its metabolic rate can be improved. We draw the parallel of identifying which roads need to be widened and improved to increase traffic flow through a crowded city. At the same time, we will also explore transferring random genes from a closely related fast-growing bacterium into C. cellulolyticum in order to address this problem. This approach of screening random elements of a genome in order to improve a particular function has been termed "combinatorial metabolic engineering" and has proven widely successful.
Animal Health Component
(N/A)
Research Effort Categories
Basic
(N/A)
Applied
(N/A)
Developmental
100%
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
5114010100020%
5114010104030%
5114010202020%
5114010208030%
Knowledge Area
511 - New and Improved Non-Food Products and Processes;

Subject Of Investigation
4010 - Bacteria;

Field Of Science
2020 - Engineering; 2080 - Mathematics and computer sciences; 1000 - Biochemistry and biophysics; 1040 - Molecular biology;
Goals / Objectives
The long-term goal of this research is to develop a cellulolytic microbial strain through metabolic engineering capable of producing biofuels at sufficient rates to be used in a single stage consolidated bioprocess (CBP). We hypothesize that (i) central carbon metabolism bottlenecks in Clostridium cellulolyticum can be identified using a novel algorithm involving genome-scale flux models for both C. cellulolyticum and C. acetobutylicum and (ii) trans-genomic combinatorial metabolic engineering can be integrated with this approach to produce "super-growing" cellulolytic and solventogenic clostridia strains. The following objectives will be used to evaluate these hypotheses. (1) A genome-scale model for C. cellulolyticum will be derived from the previously developed model for C. acetobutylicum. A novel algorithm will be built that makes use of both models to identify pathway bottlenecks. (2) Rational metabolic engineering approaches will be designed based on genome-scale model simulations in effort to confer the specific growth and solvent producing rates of C. acetobutylicum to C. cellulolyticum. (3) Combinatorial metabolic engineering will be applied by over-expression of a genomic library of C. cellulolyticum and a trans-genomic library from C. acetobutylicum. These results will be built into the genome-scale models to refine bottleneck predictions. The expected outputs from this research include: (i) a functional and tested genome-scale metabolic flux model for C. cellulolyticum, (ii) a new methodology to identify metabolic bottlenecks through comparison of genome-scale models results, (iii) the production of several new cellulolytic microbial strains with improved growth rates compared to wild-type C. cellulolyticum H10, (iv) further understanding of the microbe-cellulosome synergy.
Project Methods
The genome-scale model for C. cellulolyticum H10 will be derived from mass balances given all known or predicted intracellular metabolic and membrane transport reactions. The resulting model will be solved using linear programming and fermentation data to properly constrain metabolite transport reactions. This data will be obtained from anaerobic chemostat cultures. The medium used in this study will be a modified CM3 medium. Microcrystalline cellulose will serve as the main carbon source in all experiments. Initially, it will be added at 0.37% (w/vol). We propose to run 12 chemostat cultures in parallel to expedite the process of collecting data. To do this, we will operate three multi-channel peristaltic pumps, to each feed four reactors with working volumes of 250 mL each. The reactors and media flasks will remain anaerobic by continuous nitrogen sparging. The following (extracellular) culture states and metabolites will be monitored: cell density, cellulose, lactate, acetate, ethanol, succinate, pyruvate, CO2, H2, and the culture pH. With chemostat data, proposed calculations will be applied over all central metabolites of all comparisons of C. cellulolyticum and C. acetobutylicum. From here, we will be able to distinguish metabolic pathway usage of C. cellulolyticum relative to C. acetobutylicum over a wide range of growth conditions. From these data, we will construct a multidimensional matrix of ratio values. How bottlenecks will be identified from this matrix is as follows: (i) a flux ratio is always < 1 or (ii) a flux ratio is <<1 under certain metabolic conditions. This approach will also allow us flexibility in determining which ratios of consumption fractions are most meaningful. With knowledge of potential bottlenecks, metabolic engineering experiments will be performed. Escherichia coli chemically competent Top10 cells (Invitrogen) will be used in plasmid construction. Methylated plasmids will be electrotransformed into C. cellulolyticum. To ensure effective cloning and transformation, plasmids will be purified and sequenced by the Virginia Bioinformatics Institute CORE Laboratory Services Sequencing Facility. In these studies, a plasmid control strain will be generated using the pMTL500F plasmid. Combinatorial metabolic engineering will be performed according to published protocols. Genomic DNA fragmentation through sonication will be optimized by observing resulting fragment sizes using agarose gel electrophoresis. The target fragment size is 3 kb. The vector will then be transferred into Top10 competent E. coli to generate 10,000 transformant colonies to ensure better than 90% genomic coverage. Colonies will be combined and grown overnight in liquid LB medium. Clostridial cultures will be grown in (at least) triplicate in liquid modified CM3 medium containing erythromycin. Upon reaching the mid-logarithmic growth stage, the culture will be sub-cultured into fresh media. Following enrichment, the insertion fragment will be PCR amplified and sequenced.

Progress 12/15/09 to 06/14/14

Outputs
Target Audience: The target audiences for this research include those in the biotechnology industry who are trying to convert cellulose into renewable fuels and chemicals. In addition, this project has developed new technologies and techniques for researchers in metabolic engineering and for those working with non-pathogen clostridia. The end result of this project was remarkable and enabled us to amplify very small amounts of DNA found in the environment so that it could be screened for utility in ethanol production. The new technology developed here has wide-ranging applications and could be used in areas such as medicine, genome sequencing, and forensics. Changes/Problems: As noted previously, there were problems encountered in this research, and in each case, a new technology was developed to resolve the problem and lead to an improved strain of C. cellulolyticum capable of converting cellulose to ethanol. Below is a list of the major problems encountered and what was developed for the project to move forward. 1. The predicted flux profiles of C. acetobutylicum and C. cellulolyticum were very different and did not allow us to identify metabolic bottlenecks as anticipated. In response, we identified this was due to inaccuracies of the &ldquo;biomass equations&rdquo; of both models that describe how new cells are formed. We have produced a new technology that uses Raman spectroscopy to probe cell chemical composition cheaply and in near-real-time. 2. Even with better model predictions, comparative analysis of C. acetobutylicum and C. cellulolyticum flux profiles did not lead to clear metabolic engineering strategies for improving the growth rate and alcohol production by C. cellulolyticum. In response, we developed the &ldquo;flux balance analysis with flux ratios (FBrAtio)&rdquo; algorithm to derive metabolic engineering strategies from modeling predictions. We have applied this to several cases available in the literature for validation of the method. This will be applied to C. cellulolyticum as part of follow-up research to produce even further improved strains. 3. The original methodology cited to construct and clone DNA libraries into C. cellulolyticum simply did not work. We spent significant time and resources trying to get published protocols to work in C. cellulolyticum, but ultimately this was unsuccessful. After investigating further, we learned that cloning DNA libraries by traditional methods is highly reproducible only in a few specialized laboratories. So, we sought to develop a technology to clone our DNA libraries that could be used by nearly any laboratory with some molecular biology experience. We built this new technology using degenerate oligonucleotide polymerase chain reaction (DOP-PCR). This technique allows for random amplification of DNA in a way that can be cloned easily. Our novel contribution came with the use of thermodynamics to design PCR primers that would amplify this DNA without bias. With this new technology, we successfully cloned and enriched all DNA libraries proposed and were able to prepare and clone a library produced from DNA extracted from an environmental site. This study gave us the high-ethanol and pyruvic acid producing C. cellulolyticum strain. What opportunities for training and professional development has the project provided? This research project has directly resulted in two Ph.D. degrees and one M.S. degree in the Department of Biological Systems Engineering at Virginia Tech. The students who worked on this project were all successful in completing their higher education degrees, publishing journal articles, and giving conference presentations. One Ph.D. student has moved on to postdoctoral research at Penn State, the M.S. student has moved on to Ph.D. studies in Chemical Engineering at Penn State, and the most recent Ph.D. graduate is currently deciding between postdoctoral research or industrial opportunities. This project also provided research training to 18 undergraduate researchers in the Department of Biological Systems Engineering at Virginia Tech. Many of these students have gone on to graduate school as a direct result of their experiences in the laboratory. How have the results been disseminated to communities of interest? Results of this research have been disseminated primarily through journal publications, conference presentations and abstracts, and seminar given by Dr. Ryan Senger at various universities throughout the US. In particular, this research project will be acknowledged in over 15 journal publications and in over 40 conference presentations. Efforts will be made to disseminate the important findings and new technologies of this research through Dr. Senger&rsquo;s lab website once all manuscripts are accepted for publication. What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? At the end of this project, we engineered a strain of Clostridium cellulolyticum capable of producing 250% more ethanol from cellulose than the native strain. In addition, this strain also produced and secreted 250% more pyruvic acid (a potentially valuable commodity chemical) than the native strain. These gains came at the expense of producing lactic acid (a lower value commodity chemical that is better produced by other optimized cultures). This project provided a unique opportunity to apply (i) model-guided and (ii) randomized metabolic engineering strategies. In objective 1, we were successful in producing a genome-scale metabolic flux model of C. cellulolyticum. The model is called iCCE558, and it contains 815 reactions, 794 metabolites, and accounts for 557 genes. When comparing this model to our iCAC490 model of C. acetobutylicum to identify metabolic bottlenecks, we found this analysis was highly dependent on the &ldquo;biomass equation&rdquo; used to describe cell growth of each model. We then developed a novel technology based on Raman spectroscopy to help resolve accurate portrayals of cell chemical composition. This new tool can be applied in near-real-time and is cheap compared to traditional methods of analysis. It can be applied broadly and is applicable for all cell types. In objective 2, problems were encountered between analyzing metabolic flux predictions of C. cellulolyticum and C. acetobutylicum due to the high degree of variability between predictions and the extreme influence of the biomass equations. So, we developed a new method for deriving metabolic engineering strategies from genome-scale metabolic flux models called &ldquo;flux balance analysis with flux ratios (FBrAtio).&rdquo; The approach was successful in simulating previously published strategies in different organisms, and we are now applying this to C. cellulolyticum to further increase ethanol production and growth rate. We will combine the success from objective 3 (below) with this strategy to produce further improved strains in the future. In objective 3, we aimed to clone and enrich a genomic DNA library to isolate genes or DNA fragments that would lead to improved culture growth. This turned out to be a very difficult task in C. cellulolyticum, and eventually we invented a new technology to accomplish this objective. The new technology is based on degenerate oligonucleotide primed polymerase chain reaction (DOP-PCR). We used thermodynamics to design PCR primers capable of amplifying nanogram quantities of DNA evenly (with minimal bias) to contain end-fragments capable of easily and reliably cloning these fragments into expression vectors at high efficiency. This new technology relieved a molecular cloning bottleneck that has existed for years. Now undergraduate researchers can perform this technique in a matter of days; whereas, highly specialized research labs used to require months to complete this task. Not only were we able to clone DNA libraries from C. acetobutylicum and C. cellulolyticum, but we were able to clone DNA harvested from environmental sites. Ultimately, we found a ~200 base pair DNA fragment that reordered metabolism of C. cellulolyticum to favor ethanol and pyruvic acid production. This led to the 250% increases that we have reported. While significant problems arose throughout the course of this research, we developed new technologies that will have wide applicability, and they enabled us to produce a highly productive strain of C. cellulolyticum that converts plant-based cellulose to ethanol.

Publications

  • Type: Journal Articles Status: Published Year Published: 2014 Citation: Fisher, A.K., Freedman, B.G., Bevan, D.R., Senger, R.S. (2014) A review of metabolic and enzymatic engineering strategies for designing and optimizing performance of microbial cell factories. Computational and Structural Biotechnology Journal. 11, 91-9.
  • Type: Journal Articles Status: Published Year Published: 2014 Citation: Senger, R.S., Yen, J.Y., Fong, S.S. (2014) A review of genome-scale metabolic flux modeling of anaerobiosis in biotechnology. Current Opinion in Chemical Engineering. 6, 33-42.
  • Type: Journal Articles Status: Published Year Published: 2014 Citation: Shin, J.H., Wakeman, C.A., Rodionov, D.A., Goodson, J.R., Freedman, B.G., Senger, R.S., Winkler, W.C. (2014) Transport of magnesium by a bacterial NRAMP. PLoS Genetics. 10, e1004429.
  • Type: Book Chapters Status: Published Year Published: 2013 Citation: Senger, R.S., Nazem-Bokaee, H. (2013) Resolving cell composition through simple measurements, genome-scale modeling, and a genetic algorithm. Methods in Molecular Biology. 985, 85-101.
  • Type: Journal Articles Status: Published Year Published: 2013 Citation: Yen*, J.Y., Nazem-Bokaee, H., Freedman, B.G., Athamneh, A.I., Senger, R.S. (2013) Deriving metabolic engineering strategies from genome-scale modeling with flux ratio constraints. Biotechnology Journal. 8, 581-594.
  • Type: Journal Articles Status: Awaiting Publication Year Published: 2014 Citation: Yen*, J.Y., Tanniche, I., Fisher, A.K., Bevan, D.R., Gillaspy, G.E., Senger, R.S. (2014) A review of computational methods used to construct de novo biosynthetic pathways. Advances in Genomics and Genetics. In press.
  • Type: Journal Articles Status: Under Review Year Published: 2014 Citation: Nazem-Bokaee, H., Senger, R.S. (2014) Improving microbial phenotype predictions with new genome-scale metabolic flux modeling tools. Under review.
  • Type: Journal Articles Status: Under Review Year Published: 2014 Citation: Nazem-Bokaee, H., Yen, J.Y., Athamneh, A.I.M., Apte, A.A., McAnulty, M.J., Senger, R.S. (2014) SyM-GEM: A synthetic metabolic pathway builder and genome-scale model database to aid systems metabolic engineering in biotechnology. Under review.
  • Type: Journal Articles Status: Under Review Year Published: 2014 Citation: Freedman, B.G., Senger, R.S. (2014) Improving the growth rate of Clostridium cellulolyticum on cellulosic substrates using a genomic library. Under review.
  • Type: Journal Articles Status: Under Review Year Published: 2014 Citation: Freedman, B.G., Senger, R.S. (2014) Analyzing genomic coverage of DNA libraries produced by degenerate oligonucleotide primed PCR. Under review.
  • Type: Journal Articles Status: Under Review Year Published: 2014 Citation: Freedman, B.G., Zu, T.N.K., Senger, R.S. (2014) Understanding butanol tolerant phenotypes with Raman spectroscopy. Under review.
  • Type: Journal Articles Status: Other Year Published: 2014 Citation: Nazem-Bokaee, H., Nielsen, D., Trinh, C., Senger, R.S. (2014) ToMI-FBA: A genome-scale metabolic flux based algorithm to select optimum hosts and media formulations for expressing pathways of interest. In preparation.
  • Type: Journal Articles Status: Other Year Published: 2014 Citation: Nazem-Bokaee, H., Collakova, E., Senger, R.S. (2014) Integrating Raman spectroscopy with genome-scale metabolic flux modeling to determine the phenotypic dynamics of Clostridium cellulolyticum grown on cellobiose. In preparation.
  • Type: Journal Articles Status: Other Year Published: 2014 Citation: Freedman, B.G., Senger, R.S. (2014) Review of DNA library preparation methods and their impacts in metabolic engineering. In preparation.
  • Type: Conference Papers and Presentations Status: Published Year Published: 2014 Citation: Freedman, B.G., Senger, R.S. (November 2014) Combinatorial metabolic engineering with genomic libraries constructed from degenerate oligonucleotide-primed PCR with thermodynamically designed primers. American Institute of Chemical Engineers (AIChE) Annual Meeting. Atlanta, GA (oral presentation)
  • Type: Conference Papers and Presentations Status: Published Year Published: 2014 Citation: Nazem-Bokaee, H., Yen, J.Y., Senger, R.S. (November 2014) Novel tools in genome-scale metabolic flux modeling to identify metabolic engineering targets and predict microbial phenotypes. American Institute of Chemical Engineers (AIChE) Annual Meeting. Atlanta, GA (oral presentation)
  • Type: Conference Papers and Presentations Status: Published Year Published: 2014 Citation: Senger, R.S. (May 2014) Flux Balance Analysis with Flux Ratios (FBrAtio): A genome-scale modeling methodology to design metabolic engineering strategies. 3rd Annual Constraint Based Reconstruction and Analysis Conference. Wintergreen, VA (oral presentation)
  • Type: Conference Papers and Presentations Status: Published Year Published: 2014 Citation: Nazem-Bokaee, H., Zu, T.N.K., Collakova, E., Senger, R.S. (May 2014) Reconciling observed and predicted cellular metabolic fluxes using genome-scale modeling, a genetic algorithm, and Raman spectroscopy. 3rd Annual Constraint Based Reconstruction and Analysis Conference. Wintergreen, VA (poster presentation)
  • Type: Conference Papers and Presentations Status: Published Year Published: 2013 Citation: Freedman, B.G., Senger, R.S. (November 2013) Combinatorial metabolic engineering of Clostridium cellulolyticum using clonable genomic library fragments generated from degenerate oligonucleotide primed PCR. American Institute of Chemical Engineers (AIChE) Annual Meeting San Francisco, CA (oral presentation)
  • Type: Conference Papers and Presentations Status: Published Year Published: 2013 Citation: Nazem-Bokaee, H., Senger, R.S. (November 2013) Optimizing the biomass equation of a genome-scale model using simple measurements obtained by Raman spectroscopy. American Institute of Chemical Engineers (AIChE) Annual Meeting San Francisco, CA (oral presentation)
  • Type: Conference Papers and Presentations Status: Published Year Published: 2013 Citation: Yen, J.Y., Nazem-Bokaee, H., Freedman, B.G., Athamneh, A.I., Senger, R.S. (March 2013) Deriving metabolic engineering strategies with flux ratios genome-scale modeling. Institute for Biological Engineers (IBE) International Conference. Durham, NC (oral presentation)
  • Type: Conference Papers and Presentations Status: Published Year Published: 2013 Citation: Nazem-Bokaee, H., Senger, R.S. (March 2013) Host selection for synthetic pathways using a computational systems biology approach to explore biodiversity. Institute for Biological Engineers (IBE) International Conference. Durham, NC (oral presentation)
  • Type: Theses/Dissertations Status: Published Year Published: 2014 Citation: Nazem-Bokaee, H. Systems metabolic engineering through application of genome-scale metabolic flux modeling. PhD Dissertation. Department of Biological Systems Engineering at Virginia Tech. 2014.
  • Type: Theses/Dissertations Status: Published Year Published: 2014 Citation: Freedman, B.G. Degenerate oligonucleotide primed amplification of genomic DNA for combinatorial screening libraries and strain enrichment. PhD Dissertation. Department of Biological Systems Engineering at Virginia Tech. 2014.


Progress 12/15/11 to 12/14/12

Outputs
OUTPUTS: During this reporting year, we have finalized the genome-scale model of the cellulose consumer, Clostridium cellulolyticum H10. We have developed a "dynamic" biomass equation for this model. This means that we have developed a comprehensive mathematical model for C. cellulolyticum that describes how complex substrate (e.g., cellulose) are utilized by the cell and transformed into biofuels (e.g., ethanol) and macromolecules (e.g., protein, lipids, cell wall, DNA/RNA, etc.) required of cell growth. Novel during this reporting period is that we have realized that the chemical composition of C. cellulolyticum cells changes dramatically over the course of fermentation and that this change must be incorporated into the model to achieve accurate modeling results. We have developed a method to incorporate this change using simple measurements obtained from Raman spectroscopy. This approach is novel and will revolutionize genome-scale modeling of microbial growth and metabolism. We are currently preparing a journal publication on these methods. We have published our comprehensive genome-scale model database to our research website and have made it publicly available. Users can download a working genome-scale model for organisms relevant to biotechnology (there are currently 9 models to choose from). Users can also add synthetic metabolic pathways to these models using a graphical interface available on the website. We have two journal publications ready to submit on this topic. We have finished development of the new approach using "flux ratios" to derive metabolic engineering strategies using genome-scale modeling. We published one journal article on this concept, and it is to be a "featured article" in Biotechnology Journal. We also delivered 11 research presentations at scientific conferences or invited seminars during this reporting period. PARTICIPANTS: The following individuals worked on this project during the reporting period: Ryan S. Senger, Assistant Professor, Biological Systems Engineering Department (BSE), Virginia Tech; Hadi Nazem-Bokaee, PhD student, BSE, Virginia Tech; Benjamin Freedman, PhD student, BSE, Virginia Tech; Michelle Halsted, Undergraduate, BSE, Virginia Tech. TARGET AUDIENCES: The target audience for this research is the biofuels industry and researchers in academia or industry who focus on metabolic engineering of microbes. This research will play a significant role in the creation of industry and technologies related to the use of lignocellulose for the production of biofuels and bio-commodity chemicals. PROJECT MODIFICATIONS: The only project modification to report during this period is the adoption/development of the DOP methods for generating and cloning genomic libraries as discussed earlier. This method has proven effective and will allow us to accomplish the project goals. In addition, we able to expand upon the original project goals in that we can now (i) induce errors in DNA library fragments and (ii) obtain DNA from relevant environmental sites (e.g., decaying grasses and agricultural wastes). We also have incorporated Raman spectroscopy as a method for determining how cell composition of C. cellulolyticum changes over the course of batch fermentation. This was found necessary for achieving accurate modeling of metabolic activity. These improvements will better enable us to engineer a strain of C. cellulolyticum capable of effectively converting plant materials (e.g., cellulose) into biofuels and chemicals in a consolidated bioprocess.

Impacts
Here, the following are described related to this research project: (i) changes in knowledge, (ii) changes in action, and (iii) changes in conditions. For the changes in knowledge, our progress with cloning a "genomic library" in clostridial vectors has finally been successful. To do this, we have developed a method of "degenerate oligonucleotide priming (DOP)" PCR to generate high-quality DNA to comprise the library. This was a major hindrance of our previous approaches. We have also adopted new methods of TA cloning using linearized T-tailed vectors and the Gateway technology invented by Invitrogen to put our DNA library into a clostridial vector. Results have shown very high efficiency, and we have now successfully cloned the necessary DNA libraries to complete the goals of this project. The new approach using the DOP techniques has opened many doors for new developments. These will be explored as changes in action. In particular, for changes in action, we will now explore the use of "error-prone" PCR in generating DNA libraries. This is an exciting development as slight changes can be engineered in cloned DNA fragments to test for improved functionality. Methods involving mutated sequences have proven very successful in previous research, and we have now developed a method to genomic library screening. In addition, the new DOP methods will allow us to explore DNA beyond the genomic DNA of C. cellulolyticum or C. acetobutylicum as initially proposed. We will now go to environmental sites (e.g., decaying grass or agricultural residues) and to extract DNA and determine if it can enhance cellulose degradation and improve growth rate of C. cellulolyticum on several complex substrates, including recycled paper. We do not have any changes in conditions to report for this reporting period.

Publications

  • Yen, J.Y., Nazem-Bokaee, H., Freedman, B.G., Athamneh, A.I., Senger, R.S. (2013) Deriving metabolic engineering strategies from genome-scale modeling with flux ratio constraints. Accepted to Biotechnol J.
  • Senger, R.S., Nazem-Bokaee, H. (2013) Resolving cell composition through simple measurements, genome-scale modeling and a genetic algorithm, in Methods in molecular biology (H. Alper, Ed.) ISSN: 1064-3745. Springer.
  • Senger, R.S. (April 2012) Engineering a BioEconomy with synthetic biology. Virginia Commonwealth University Chemical and Life Sciences Engineering Department Seminar. Richmond, VA (presentation abstract)
  • Yen, J.Y., Nazem-Bokaee, H., Freedman, B.G., Athamneh, A.I., Senger, R.S. (October 2012) Metabolic engineering in silico enabled by manipulating metabolic pathway flux ratios. American Institute of Chemical Engineers (AIChE) Annual Meeting. Pittsburgh, PA (presentation abstract)
  • Freedman, B.G., Senger, R.S. (October 2012) Metabolic production and cloning of error-prone genomic libraries. American Institute of Chemical Engineers (AIChE) Annual Meeting. Pittsburgh, PA (presentation abstract)
  • McAnulty, M.J., Yen, J., Freedman, B.G., Senger, R.S. (July 2012) A new approach to modeling gene regulation using flux ratios and the implications for butanol production from clostridia. American Society for Agricultural and Biological Engineers (ASABE) Annual International Meeting. Dallas, TX (presentation abstract).
  • McAnulty, M.J., Yen, J., Freedman, B.G., *Senger, R.S. (June 2012) Metabolic engineering in silico enabled by genome-scale models with flux ratio constraints. Metabolic Engineering IX Conference. Biarritz, France (presentation abstract)
  • Nazem-Bokaee, H., Senger, R.S. (May 2012) Host selection using a novel database of standardized genome-scale models. Biobased Materials Research Center Graduate Research Symposium. Blacksburg, VA (presentation abstract)
  • Nazem-Bokaee, H., Senger, R.S. (March 2012) Mathematics of cell metabolism. American Society for Agricultural and Biological Engineers (ASABE) Southeastern Regional Rally. Blacksburg, VA (presentation abstract)
  • Senger, R.S. (March 2012) Flux ratio constraints to enable metabolic engineering in silico of clostridial metabolism. Institute of Biological Engineers (IBE) International Conference. Indianapolis, IN (presentation abstract)


Progress 12/15/10 to 12/14/11

Outputs
OUTPUTS: During this reporting year, we have created the following tangible outputs as a result of this funded project. We have reconciled our previously constructed genome-scale metabolic reconstruction of Clostridium cellulolyticum H10 and are preparing to share this updated model freely through our research website. We have also developed a new database of existing genome-scale models related to biotechnology and biofuels production. We have compiled these into a single format that can be downloaded and used immediately by any user. This research has also led to a new concept called "flux ratios" modeling that has led to the submission of a manuscript for publication. This new technique will lead to much more effective metabolic engineering strategies designed by computer models. Our application also allows users to easily add synthetic pathways to any of the available genome-scale models. This is a critical tool that is currently missing from the systems biology research community. This resource will be made public through our research website once the journal publication is accepted. We have also published two more journal articles and one book chapter during this reporting period. Our research group also contributed 11 research presentations at scientific conferences or invited seminars during this reporting period. PARTICIPANTS: The following individuals worked on this project during the reporting period: Ryan S. Senger, Assistant Professor, Biological Systems Engineering Department (BSE), Virginia Tech Hadi Nazem-Bokaee, PhD student, BSE, Virginia Tech Benjamin Freedman, PhD student, BSE, Virginia Tech Mike McAnulty, MS student, BSE, Virginia Tech Parker Lee, Undergraduate, BSE, Virginia Tech Ashley Umberger, Undergraduate, BSE, Virginia Tech TARGET AUDIENCES: The target audience for this research is the biofuels industry and researchers in academia or industry who do metabolic engineering using bacteria. This research will play a significant role in the creation of industry and technologies related to the use of lignocellulose for the production of biofuels and bio-commodity chemicals. PROJECT MODIFICATIONS: The only project modification made during this reporting period is the addition of a comprehensive genome-scale model database. This was enabled through the extremely efficient work of a graduate student funded by this project who figured out a method for compiling the genome-scale model of C. cellulolyticum H10 quickly and efficiently. This database has enabled easy comparison of the C. cellulolyticum H10 model prepared by us with the models detailed in the Project Modifications of the last reporting period. This development will also enable comparisons between metabolic networks of several cellulolytic organisms and solventogenic clostridia to uncover modifications that could aid the construction of an organism capable of consolidated bioprocessing lignocellulose to biofuels.

Impacts
Here, the following are described related to this research project: (i) changes in knowledge, (ii) changes in action, and (iii) changes in conditions. For the changes in knowledge, we have continued to discover more effective methods for cloning genomic libraries in very "difficult to clone" clostridial vectors. We have found that repeatedly re-culturing or enriching these plasmids initially led to degradation of the plasmid or truncation of the antibiotic resistance genes. This was surprising, but it led to necessary changes in plasmid construction and allowed us to re-examine and optimize our methods of culture enrichment. Our new methods of cloning will be published when the genomic library enrichments are complete. Due to the problems associated with plasmid degradation, this research took longer than expected but is now in-line with the project timeline. For the changes in action for this reporting period, we first report that the project remains on schedule with the proposed timeline. Since the metabolic network reconstruction for C. cellulolyticum H10 was much easier than anticipated to complete, time resources were available to build our new model and closely related genome-scale models into a comprehensive database. This allows these models to be downloaded and used immediately with the freely available COBRA toolbox in MATLAB. This is enabling direct modeling comparisons between C. cellulolyticum and the closely related bio-butanol producers C. acetobutylicum and C. beijerinckii. Research related to parallel chemostat construction is on-time and progressing. We have grown E. coli cultures in this system and are currently dealing with challenges of maintaining an anaerobic environment to grow clostridial cultures. We do not have any changes in conditions to report for this reporting period.

Publications

  • Milne, C.B., Eddy, J.A., Raju, R., Ardekani, S., Kim, P.J., Senger, R.S., Jin, Y.S., Blaschek, H.P., Price, N.D. (2011) Metabolic network reconstruction and genome-scale model of butanol-producing strain Clostridium beijerinckii NCIMB 8052. BMC Syst Biol 5:130.
  • Reed, J.L., Senger, R.S., Antoniewicz, M.R., Young, J.D. (2011) Computational approaches in metabolic engineering. J Biomed Biotechnol 2010:207414.
  • Freedman, B.G., Athamneh, A.I.M., McAnulty, M.J., Zhou, R., Yue, Y., Senger, R.S. (2011) Metabolic engineering the clostridia for biobutanol production. Submitted Book Chapter.
  • Freedman, B.G., *Senger, R.S. (October 2011) Improving the growth rate of cellulolytic clostridia through genomic library enrichment. American Institute of Chemical Engineering (AIChE) Annual Meeting. Minneapolis, MN (presentation abstract)
  • McAnulty, M.J., *Senger, R.S. (October 2011) Rigorous proton balancing and other new tools to reduce the phenotypic solution space of a genome-scale model. American Institute of Chemical Engineering (AIChE) Annual Meeting. Minneapolis, MN (presentation abstract)
  • *Nazem-Bokaee, H., Yen, J., Athamneh, A, Apte, A.A., McAnulty, M.J., Senger, R.S. (October 2011) Host selection for synthetic pathways using a genome-scale model database. American Institute of Chemical Engineering (AIChE) Annual Meeting. Minneapolis, MN (presentation abstract)
  • *Nazem-Bokaee, H., Senger, R.S. (October 2011) Genetic regulation of the cellulosome in Clostridium acetobutylicum. American Institute of Chemical Engineering (AIChE) Annual Meeting. Minneapolis, MN (presentation abstract)
  • *Senger, R.S. (October 2011) Research in the metabolic engineering and systems biology laboratory at Virginia Tech. Virginia Tech Biological Systems Engineering Department Seminar. Blacksburg, VA (presentation abstract)
  • *Senger, R.S. (August 2011) Genome-scale modeling for the production of value-added bioproducts from microbes. American Society for Agricultural and Biological Engineering (ASABE) International Conference. Louisville, KY (presentation abstract)
  • *Lee, P.W., Senger, R.S. (August 2011) Combinatorial metabolic engineering of Clostridium cellulolyticum to create a host capable of converting cellulose to advanced liquid biofuels. Virginia Tech Summer Undergraduate Research Fellowship (SURF) Symposium. Blacksburg, VA (presentation abstract)
  • *Senger, R.S. (June 2011) Engineering phenotypes using genome-scale modeling and Raman spectroscopy. 1st Annual Constraint Based Reconstruction and Analysis Conference. Reykjavik, Iceland (presentation abstract)
  • *Senger, R.S. (May 2011) Novel biofuels through synthetic biology. Oak Ridge National Laboratory Biofuels Symposium. Oak Ridge, TN (presentation abstract)
  • *McAnulty, M.J., Senger, R.S. (April 2011) Total proton flux and balancing in genome-scale models: The case for the updated model of Clostridium acetobutylicum ATCC 824. Virginia Tech ICTAS Bio-Based Materials Center Symposium. Blacksburg, VA (presentation abstract)
  • *Senger, R.S. (March 2011) Narrowing the solutions of genome-scale modeling through proton efflux, rigorous proton balancing, and optimizing the biomass equation. Institute of Biological Engineers (IBE) International Conference. Atlanta, GA (presentation abstract)


Progress 12/15/09 to 12/14/10

Outputs
OUTPUTS: We have created the following tangible outputs as a direct result of this funded project. First, an initial draft of the Clostridium cellulolyticum H10 genome-scale metabolic reconstruction has been created. We are now in the process of simulating metabolism of this organism on the computer in order to identify targets for metabolic engineering. Our metabolic reconstruction will serve as a platform for all future metabolic engineering in this cellulose-consuming organism. This genome-scale metabolic network reconstruction will be disseminated through our website and through a journal publication that we will submit in the next six months. We have also published a journal article investigating the impact of the biomass equation in genome-scale models. In summary, we have found that the equations describing the allocation of cellular resources for growth are extremely significant for genome-scale modeling, especially for organisms producing biofuels. These findings will lead to much more accurate models in this area and will have profound impacts on the computational aspects of metabolic engineering. We have constructed a continuous microreactor system capable of running 18 chemostat reactors in parallel. This will streamline data acquisition needed for accurate modeling and testing metabolic models. Finally, we have created and cloned genomic libraries of C. cellulolyticum and C. acetobutylicum into clostridial expression plasmids. We are in the process of making these libraries available to other researchers through our website. PARTICIPANTS: Individuals: Ryan S. Senger (PD) Graduate Students Trained: 2 Undergraduate Students Trained: 2 TARGET AUDIENCES: Nothing significant to report during this reporting period. PROJECT MODIFICATIONS: In response to the change in condition mentioned previously, we have made the following project modification. We are preparing the genome-scale metabolic reconstruction, as proposed. However, we now will also prepare a version using published automated building methods. We will then compare the performance of (i) our model, (ii) the previously published model, and (iii) the automated build model. From this, we can identify the pros/cons of each and create a composite model. This will be a very good study to assess the suitability of automated network building and will be of significant importance to the field. Furthermore, we will use this opportunity to produce a superior model of C. cellulolyticum metabolism.

Impacts
Here, the following are described related to this research project: (i) changes in knowledge, (ii) changes in action, and (iii) changes in conditions. For the changes in knowledge, we have developed protocols for the effective extraction of genomic DNA in C. cellulolyticum and its effective cleavage into 1-5kb fragments (through nebulization) that are suitable for cloning. We have identified several changes to existing clostridial protocols and have achieved superior results with library generation. We are also the first to methylate plasmids for transformation into C. cellulolyticum using in vivo methods (as opposed to methylase treatment in vitro). We have achieved superior results using TA cloning for library fragments, and we have demonstrated how the preparation of "home-grown" competent E. coli K12 DH5a cells can be used to propagate genomic libraries. This results in a dramatic cost savings for producing a genomic library, and these results will be well-received by the community. We have also developed knowledge related to the genome-scale metabolic network of C. cellulolyticum and have compiled this to form a novel metabolic model. For the changes in action, we first report that this project is on-time and in-line with the proposed timeline. The metabolic network reconstruction for C. cellulolyticum was proposed for Year 1. This task is complete. We also proposed to be half-way completed with construction and operation of a continuous reactor system. This system has been built and tested. Its operation with clostridial cultures is to begin shortly. Finally, we proposed to be half-way finished with library generation and enrichment by the end of Year 1. This work is ahead of schedule. We have now created libraries for C. cellulolyticum and C. acetobutylicum and transformed into C. cellulolyticum. We are now ready to begin the enrichment process. The only change in action to report is that we will now prepare multiple C. cellulolyticum metabolic network reconstructions and then determine which is best from chemostat data. This is in response to an important change in conditions, which is described next. During Year 1 of this project, a genome-scale model for C. cellulolyticum was published by a competing group. We have inspected this model and believe that higher-quality models can be built. Furthermore, an algorithm for automating the network reconstruction process has also been published in the past year. This has led to project modifications that are discussed in a later section.

Publications

  • Senger, R.S. (2010) Improving biofuels production with genome-scale models: The role of cell composition. Biotechnol J 5, 671-685.
  • Senger, R.S. (2010) The genome-scale model biomass constituting equation plays a large role in predicting biofuels productions. AIChE Annual Meeting. Salt Lake City, UT. (abstract)
  • Freedman, B.; Senger, R.S. (2010) A combinatorial approach to genetic library enrichment of Clostridium cellulolyticum on lignocellulose. AIChE Annual Meeting. Salt Lake City, UT. (abstract)
  • Senger, R.S. (2010) Improving the growth rate of Clostridium cellulolyticum H10 with genomic libraries and genome-scale modeling. Clostridium XI Conference. San Diego, CA. (abstract)
  • Senger, R.S. (2010) Increased microbial biofuels production through systems biology. ASABE International Meeting. Pittsburgh, PA. (abstract)
  • Freedman, B.; Senger, R.S. (2010) A combinatorial approach to genetic library enrichment of Clostridium cellulolyticum on cellulose. ASABE International Meeting. Pittsburgh, PA. (abstract)
  • Senger, R.S. (2010) Identification fo cellulosome repressors in solventogenic clostridia. Metabolic Engineering VIII Conference. Jeju, S. Korea. (abstract)