Final Report Summary - SYSINBIO (Systems Biology as a Driver for Industrial Biotechnology)
Executive Summary:
Metabolic engineering is an applied science focusing on developing new cell factories or improving existing ones. Metabolic engineering distinguishes itself from applied genetic engineering by the use of advanced analytical tools for identification of appropriate targets for genetic modifications and the use of mathematical models to perform in silico (computer based) design of optimized cell factories. In recent years, there has been increasing focus on using mathematical models for design and SYSINBIO has been coordinating European activities in the field of model driven metabolic engineering and activities on other technologies required for state of the art metabolic engineering, e.g. metabolomics (measurements of internal metabolites) and fluxomics (measurements of metabolic fluxes). The coordination of activities has resulted in the establishment of a database containing metabolic models for different industrially important microorganisms (the BioMet ToolBox available at www.sysbio.se). The database also contains different simulation tools required for use of these models to identify metabolic engineering targets and use of these models for analysis of omics data (high throughput measurements of biologically relevant entities such as proteins or messenger RNAs). SYSINBIO has also worked on coordinating further development of techniques required for metabolic engineering, such as metabolomics, fluxomics and identification of mutations in evolved strains through whole genome sequencing using second generation sequencing approaches. Several publications, in particular reviews, summarizing the state of the art in the field have been published in connection with SYSINBIO.
A key part of SYSINBIO was to coordinate activities on developing new and better mathematical models of metabolism that can be used for metabolic engineering. Activities on both stoichiometric models and kinetic models have been coordinated. The use of stoichiometric models has been further advanced and we have defined methods that allows for rapid setting up this kind of models. A fundamental problem with stoichiometric models is that it is necessary to perform a manual annotation and curation of the many different reactions, which is time consuming and laborious. We have therefore formulated methods and standards that will allow for faster development of this kind of models, and we have also developed a database of existing models that can be used as scaffolds for building new models. In the field of kinetic models we have performed a detailed review of the state-of-the art on the use of these models, and based on this we have drafted recommendations, in particular for use in focussed industrial metabolic engineering projects.
Project Context and Objectives:
Metabolic engineering is an applied science focusing on developing new cell factories or improving existing ones. There are several definitions, but most of these are consistent with: the use of genetic engineering to perform directed genetic modifications of cell factories with the objective to improve their properties for industrial application. In this definition the word improve is to be interpreted in its broadest sense, i.e. it also encompasses the insertion of completely new pathways with the objective to produce a heterologous product in a given host cell factory. Metabolic engineering is an enabling science, and distinguishes itself from applied genetic engineering by the use of advanced analytical tools for identification of appropriate targets for genetic modifications and the use of mathematical models to perform in silico design of optimized cell factories. Metabolic engineering is therefore often seen as a cyclic process, where the cell factory is analyzed and based on this an appropriate target is identified (the design phase). This target is then experimentally implemented and the resulting strain is analyzed again. Thus, metabolic engineering involves a continuous iteration between design and experimental work. In recent years, there has been increasing focus on using mathematical models for design. Hereby, it is expected that metabolic engineering will become faster and more efficient through the development of robust and reliable mathematical models describing the function of cell factories.
In recent years metabolic engineering has adapted tools from functional genomics and systems biology. There are many definitions of systems biology, but most of these contain elements such as mathematical modelling, global analysis (or "ome" analysis), mapping of interactions between cellular components, and quantification of dynamic responses in living cells. In most cases the objective of systems biology is to obtain a quantitative description of the biological system under study, and this quantitative description may be in the form of a mathematical model1,2. In some cases, the model may be the final result of the study, i.e. the model captures key features of the biological system and can hence be used to predict the behaviour of the system at conditions different from those used to derive the model. In other cases, mathematical modelling rather serves as a tool to extract information of the biological system, i.e. to enrich the information content in the data. There is not necessarily a conflict between the two, and generally, mathematical modelling goes hand in hand with experimental work. This partnership exemplifies the essence of systems biology, obtaining new insight into the molecular mechanisms occurring in living cells or sub-systems of living cells, through the combination of mathematical modelling and experimental biology. This does not say anything about the use of high-throughput data, e.g. transcriptome or proteome data, and clearly there are many systems biology studies that do not rely on such data. Mathematical models have, however, shown to be particularly useful for analysis of high-throughput data, as the complexity and integrative nature of biological systems makes it difficult to extract information on molecular processes from such data without the use of models as either scaffolds for the analysis or for hypothesis driven analysis of the data.
In the field of industrial biotechnology there is much focus on how systems biology can impact the development of efficient cell factories, and in particularly speed up the development process, and hereby ensure that new products can be brought to the market faster or there can be a faster improvement of existing bioprocesses.
The objectives of SYSINBIO was to coordinate research activities in the field of model driven metabolic engineering in Europe and according to the concept described above coordinate how systems biology can be used to improve the performance of metabolic models and further how mathematical models can be used for improved design of cell factories used in industrial biotechnology. The overall goal was divided into 11 specific objectives that were dealt with in 11 different workpackages that formed the basis of the project:
- Research efforts in the field of model guided metabolic engineering within Europe were coordinated through organizing yearly meetings of leading European research groups working in this field.
- Advanced education in the field of metabolic engineering in Europe was coordinated. This has involved setting up and running dedicated courses and workshops on mathematical modelling and metabolic engineering, but it also involved advanced research training through exchange of researchers between the involved laboratories.
- SYSINBIO organized an international conference with focus on the use of model guided metabolic engineering for development of efficient cell factories. The conference ensured dissemination of new results generated by the participating research groups and further pushed forward the research field by bridging between academia and industry.
- SYSINBIO coordinated the development of metabolic models and has provided a database for existing models. Concepts and methods for setting up genome-scale metabolic models were exchanged and developed resulting in implementation of best practices and efficient work flows in the partner laboratories.
- SYSINBIO also coordinated the development of dynamic models for design of metabolic engineering strategies. Concepts and methods for setting up and analyzing dynamic models was exchanged and developed. Efficient workflows for setting up dynamic models of metabolic reaction networks including parameter estimation, model validation and discrimination, parameter estimation, and simulation were defined for different types of dynamic models.
- SYSINBIO coordinated the development of novel simulation tools that allow the use of metabolic models for design of metabolic engineering strategies. Different algorithms for identification of metabolic engineering targets through the use of genome-scale metabolic models were compared and evaluated, and based on the evaluation, general guidelines for the use of metabolic models for metabolic engineering was specified.
- SYSINBIO coordinated the integration of omics data with metabolic models in order to improve their predictive strength. Different concepts and methods for integration of "omics" data into metabolic models was compared and evaluated leading to guidelines on how such data can be used for the reconstruction and evaluation of genome-scale metabolic models.
- SYSINBIO coordinating the development of novel tools for 13C-based fluxomics (both experimental and computational tools). Different tools for fluxomics were compared and evaluated, and guidelines for quantification of metabolic fluxes in microorganisms, during industrial fermentations, have been proposed.
- SYSINBIO coordinated the development of novel tools for metabolomics. Different methods for sampling and analysis of key intracellular metabolites was compared and evaluated, and guidelines for use of different analytical tools for the analysis of intracellular metabolites, in different industrially important microorganisms, was set up.
- SYSINBIO coordinated the use of statistical methods and metabolic models for upgrading the information content in "omics" data. Different methods and concepts were evaluated for their ability to upgrade the information content in "omics" data, and in particularly it was evaluated how this kind of data can be used to identify novel metabolic engineering targets.
- Finally, SYSINBIO coordinated research activities on the use of DNA arrays, tiling DNA arrays and DNA sequencing for rapid identification of mutations that lead to improved phenotypic properties, e.g. arising in connection with adaptive evolution. The use of these different methods for mapping of mutations in strain lineages was evaluated.
Project Results:
The consortium behind SYSINBIO involved all the leading groups on metabolic engineering in Europe, and it had particularly strong competence on metabolic engineering of the yeast Saccharomyces cerevisiaae and the bacterium Escherichia coli. For these two microorganisms a number of different technologies had already been developed and implemented before the start of the project, and during the project there was further advancement in technologies. However, the strength of SYSINBIO was that there was also much experience on working with other industrially important microorganisms. The key microorganisms studied within SYSINBIO were:
- Saccharomyces cerevisiae, which is used for the production of bioethanol and a range of pharmaceutical proteins. S. cerevisiae is also considered as a general production platform for a range of different chemicals, e.g. isoprenoids (flavours, fragrances, pharmaceuticals), bulk chemicals (succinate), nutraceuticals (resveratrol and poli-unsaturated fatty acids known as PUFAs), in some of the partner laboratories. S. cerevisiae also serves as an important model organism for studying eukaryotic cells.1,3
- Escherichia coli, which is used for production of many different recombinant proteins, e.g. human growth hormone and various interleukins, and a wide range of different chemicals like amino acids, e.g. phenylalanine (used in the production of Aspartame®) and methionine (used as a feed). E. coli is also considered as a general production platform for chemical production, and it has traditionally served as the most important model organism for studying bacterial physiology and genetics.
- Aspergillus niger, which is used for the production of organic acids (citric acid and gluconic acid) and a wide range of industrial enzymes.
- Aspergillus oryzae, which is used for the production of a wide range of industrial enzymes.
- Trichoderma reesei, which is used for the production of celluloses used for processing of fibers and for biomass egradation in connection with production of fuels and chemicals from biomass.
- Corynebacterium glutamicum, which is used for the production of lysine and other amino acids used in the feed industry.
- Lactococcus lactis, which is used in starter cultures and also serves as an important model organism for lactic acid bacteria.
- Bacillus subtilis, which is used for the production of industrial enzymes, different chemicals, e.g. riboflavin, and different antibiotics, e.g. bacitracin.
- Streptomyces species, which are used for the production of antibiotics like nystatin, industrial enzymes, and recombinant proteins.
Besides broad competence on different industrially important microorganisms the partners of SYSINBIO covered competence on all key technologies that are required for state of the art metabolic engineering, namely:
- Well controlled fermentation experiments: The ability to perform well controlled fermentation experiments, e.g. using chemostat cultures or computer controlled fed-batch experiments, is essential for metabolic engineering, as it allows for detailed quantitative analysis of the cell factory. The consortium covered extensive experience in this field for a wide range of microorganisms.
- Transcriptome analysis: The partners have performed genome-wide transcription analysis of many different industrially important microorganisms, and in particularly demonstrated how this omics technique can be used for gaining insight into the operation of metabolic networks.
- Metabolome analysis: Several of the partners are among the World leaders in metabolomics of industrially important microorganisms, and has been involved in defining standards and protocols in this field.
- Fluxome analysis: The consortium includes most of the World leading research groups in the field of fluxome analysis. Partners of the consortium have pioneered the use of GC-MS analysis for measurement of isotopomers and they have recently brought the field further forward by using isotopomer measurements of key intermediates in the central carbon metabolism.
- Genome-scale metabolic models: The consortium has played an active role in setting up genome-scale metabolic models for a range of industrially important microorganisms.
- Thermodynamic analysis: The consortium has demonstrated the use of thermodynamics for analysis of large metabolic networks and illustrated how thermodynamics can be used to identify flux control coefficients in key biosynthetic pathways.
- Dynamic models: Based on its strong competences the consortium has evaluated the use of dynamic models for analysing the operation of metabolic networks and has developed several different modelling concepts for simulation of dynamic operation of metabolism.
- Model simulation and bioinformatics: The partners in the consortium have developed several new algorithms for analysis of omics data in the context of metabolic models as well as they have developed different simulation platforms that allow simulation of both stoichiometric and dynamic mathematical models. SYSINBIO ensured further advancement in this field.
Based on combination of these strong competences SYSINBIO was able to significantly advance the field and bridge stronger between academia and industry through the coordination activities described further in the following.
Training and Education
SYSINBIO was involved in organizing several advanced courses and student workshops:
- The International Yeast Course on Systems Biology that was held June 1-18, 2009 in Gothenburg. The course was sponsored by FEBS. 4 days of this course were focusing on metabolic engineering technologies, such as metabolic flux analysis, metabolic modelling, and metabolic control analysis. The course had 24 full time students, but parts of the course were offered to local students, of which there were more than 15. The 24 full time students came from 19 different countries.
- The International Course on Systems Biology of Metabolism that ran in the period from May 24 to June 11, 2010 at Chalmers University of Technology. The course had 22 participants from all over Europe. The course consisted of five blocks of teaching: 1) An introduction to systems biology; 2) Setting up and applying genome scale metabolic models; 3) Analysis of omics data and their integration with genome-scale models; 4) Dynamic mathematical models of metabolism and signalling pathays; 5) Examples in biotechnology and medicine. More information about the course can be found at: http://www.sysbio.se/ICSBM2010/.
- A student workshop that was held on December 8-9 in Istanbul . The workshop was chaired by Prof. Vassily Hatzimanikatis, EPFL. At the workshop the 20 participants (PhD students and post docs) each gave a 30 min presentation about their research work.
- A student workshop that was held on January 18-19 at the DECHEMA-Haus in Frankfurt. The workshop was chaired by Prof. Ralf Takors, University of Stuttgart, and Dr. Sergio Velasco, Chalmers. At the workshop the 20 participants (PhD students and post docs) each gave a 30 min presentation about their research work. The presentations were followed by extensive discussions involving both the participants and the organizers. In connection with the workshop there were two special lectures:
- Probabilistic Boolean framework to decrypt gene regulatory networks case studies with primary human hepatocytes and E. coli, by Prem Kumar
- Metabolome-based 13C-metabolic flux analysis; A tool for measuring in vivo metabolic activities, by Katharina Nöh
Besides the scientific presentations the workshop also allowed the participant to network through a social event organized over dinner.
- A PhD course entitled "Industrial Biotechnology for lignocellulose based processes" that was held at Chalmers, Gothenburg, in the period 16th to 21st October 2011 and organized by Professor Lisbeth Olsson. The course was relevant for PhD students and researchers working in the field of lignocellulose based processes, i.e. the production of fuels and other chemicals using plant cell wall as raw material. The idea of the course was to encompass all parts of the subject from raw material composition and sources, to pretreatment and hydrolysis, enzymes that act on plant cell wall material, microorganisms and their improvement for production of targeted biofuels and chemicals, the fermentation process and last but not least analytics. The course program included an evening poster session, where the participants had the chance to present their research and visit Professor Olsson laboratories. The course program was closed with a day of seminars on "Hot topics in industrial biotechnology". The course was attended by 25 students of 17 different nationalities, coming from institutions in the following countries: Sweden, Denmark, Norway, Finland, United Kingdom, Belgium, Germany, Italy, Spain and Turkey. SYSINBIO sponsored 7 students from the SYSINBIO consortium. 15 lecturers from Sweden, Denmark, Finland, Norway and Italy was involved in the course. Several of them were partners of the SYSINBIO.
International Conference
SYSINBIO organized a 2 day conference "Industrial Systems Biology: Sustainable production of fuels and chemicals". The conference was held at Chalmers University of Technology, Gothenburg, Sweden August 18-19, 2010 (the conference was organized as one of the conferences in the Gothenburg Life Science Conference series). In connection with the conference there was also organized a dedicated student workshop on August 17 (3 hours) where one speaker from academia (Professor Gregory Stephanopoulos, MIT, USA) and one speaker from industry (Dr. Alan Berry, Novozymes, USA/Denmark) discussed job prospects and industrial demand for future students.
At the conference there were 20 oral presentations and 40 poster presentations. There were more than 170 participants at the conference. Among the speakers were: Jens Nielsen, Chalmers, Gothenburg Sweden; Vassily Hatzimanikatis, EPFL, Lausanne, Switzerland; Kiran Patil, DTU, Lyngby, Denmark; Matthias Heinemann, ETH Zürich, Switzerland; Elmar Heinzle, Saarland University, Germany; Sang Yup Lee, KAIST, Daejon, Korea; Christoph Wittmann, TU Braunschweig, Braunschweig, Germany; Andrew C. Eliot, DuPont, Wilmington, USA; Jack Pronk, TU Delft, The Netherlands; Alan Berry, Novozymes, Davis, USA; Verena Siewers, Chalmers, Gothenburg, Sweden; Jochen Förster, Fluxome, Copenhagen, Denmark; Eric Appelman, Perstorp, Sweden. A detailed report from the conference has been published in Biotechnology Journal4. Furthermore, based on oral presentations several papers were published in a special issue of Biotechnology journal (issue 3, 2011) entitled "Systems biology for industrial applications", and the editorial provides perspectives based on conclusions of the conference5.
Mathematical Models and Their Use in Industrial Biotechnology
A key task of SYSINBIO was to coordinate activities on developing new and better mathematical models of metabolism that can be used for metabolic engineering. Activities on both stoichiometric models and on kinetic models have been coordinated. Stoichiometric models, often also referred to as genome-scale metabolic models (GEMs), have the advantage that they are directly linked to genomic information6,7, and methods for set up and use of GEMs has been further advanced. In particular SYSINBIO was active in defining methods that allows for faster setting up this kind of models8. A fundamental problem with stoichiometric models is that it is necessary to perform a manual annotation and curation of the many different reactions, and this is time consuming and laborious. We have therefore formulated methods and standards that will allow for faster development of this kind of models, and we have also developed a database of existing models that can be used as scaffolds for building new models9. We have also worked on different user-friendly graphical interfaces that allows for visualization of these complex models. For kinetic models we have performed an extensive survey of existing models in the partner laboratories (and of published models), and we have used this to discuss how these models can be used in metabolic engineering8,10. A fundamental problem with kinetic models is the requirement for parameter estimation, but partners within SYSINBIO are using different modelling concepts that partly circumvent this problem, e.g. Monte Carlo simulations and use of linearized kinetics in the logarithmic domain11,12. Furthermore, workflows for setting up kinetic models are in the process of being defined and these will be used to define the requirements for experimental data for validation and parameter estimations (unpublished).
The coordination of the development of more reliable and standardized microbial models has allowed for the improvement of model predictions and therefore made it possible to apply simulation and optimization algorithms that can identify gene modifications that can increase the production yields of desired compounds. We have coordinated the development of reliable and effective computational and mathematical methods for the design of rational metabolic engineering strategies13,14,15,16. Through activities we have ensured that the simulation and optimization tools developed obey the following requisites: 1) robustness regarding the different models used (stoichiometric versus dynamic models) and metabolic engineering objectives; 2) compatibility with existing model storage standards (SBML and other formats); and 3) straightforward incorporation in user-friendly software tools to facilitate its use by non-expert users. Another important part of the activities has been the coordination of the development of user friendly software tools, something that is allowing for a wider use of mathematical modelling in the field of metabolic engineering16. Finally SYSINBIO has been working on improving the predictive strength of GEMs through incorporation of regulatory information17.
Integration of Omics Data
A key challenge in metabolic engineering is to be able to integrate omics data, e.g. transcriptomics, proteomics and metabolomics, with mathematical models with the objective to identify novel metabolic engineering targets. In SYSINBIO there were several ways to advance this. Flux data obtained from 13C-labelled experiments have been shown to allow constraining the fluxes in GEMs and hereby obtain better constraints on the operation of these large metabolic networks18. We have further used a combination of thermodynamics and experimental fluxes to identify flux constraining reactions19. This has substantially improved the predictive strength of these models for finding metabolic engineering targets. SYSINBIO also demonstrated how a combination of random sampling of GEMs combined with transcriptome data can be used to identify whether flux control is at the transcriptional or metabolic level20.
Another key challenge in this field is visualization of the rather complex data, but through the use of GEMs as scaffold it has been possible to easily visualize how omics data link to specific enzymatic reactions21, and this has been provide in a user-friendly software algorithm.
Due to the high connectivity of the different metabolic reactions within the metabolic network, there has been much interest in exploiting tools from functional genomics for mapping of global regulatory structures or even using high-throughput experimental techniques provided by the various omics for dissecting how the fluxes through the different branches of the metabolic network are controlled. In particular it is interesting to map the level of control of fluxes, i.e. is it at the metabolism level or is there control of flux at the transcriptional, translational or post-translational level. This kind of information can only be obtained through a combination of mathematical models and high-quality experimental data on metabolites and fluxes, and possible other omes. The partners of SYSINBIO have in several studies demonstrated how integrative analysis of omics data through the use of GEMs and other network structures can lead to new biological insight, in particular on how metabolic fluxes are controlled. This has involved both the identification of key transcription factors controlling parts of metabolism in yeast22,23 and how the key protein kinase Snf1 control yeast metabolism24. These concepts were also demonstrated to be useful to gain insight into how different yeast strains control their energy metabolism25. However, these concepts have not only been used for studies of yeast, as the concepts have also been used to gain new insight into enzyme production by Aspergillus oryzae26. As part of these studies there was also performed a detailed characterization of different transcriptome platforms for yeast, and even though these results were specific for yeast they are likely to have translational character and hence some general guidelines for transcriptome analysis in industrial microorganisms has been proposed25.
Fluxomics
Metabolic fluxes can either be estimated through the use of flux balance analysis (FBA) using GEMs or through the use of 13C-labelled substrate feeding followed by analysis of the labelling patterns in intracellular metabolites. Due to their abundance and stability, 13C-based methods have conventionally used proteinogenic amino acids to detect labelling patterns. Recently, however, methods for direct analysis in the free pool of metabolites have been developed. Both methods require a tight integration of experiments and model simulations for estimation of the fluxes, and in particular the mathematical complexity involved in flux estimation has been a barrier for a wider use of fluxomics. Besides providing general information on how the metabolic network is operating at different growth conditions, metabolic flux analysis is very well suited for analysis of the effects of growth on different media, specific mutations and screening of different mutants. Thus, flux analysis today represents a standard technique for rapid phenotypic characterization of metabolically engineered strains, and this tool is likely to gain even wider use in the future. Within SYSINBIO there has been coordination of developments in this field, and this has resulted in several key reviews and papers specifying guidelines for how to perform metabolic flux analysis27,28,29,30.
Metabolomics
Another important analytical tool in metabolic engineering is metabolome analysis, as this allows for direct quantification of the levels of intracellular metabolites. As these levels determine the kinetics of the different reactions, the metabolite levels together with the fluxes represents key information on how metabolism can be engineered. SYSINBIO has coordinated activities on metabolome analysis in the different partner laboratories. Many of the partner labs were involved in a thorough inter-laboratory comparison of metabolome analysis of the yeast Saccharomyces cerevisiae and this allowed to provide an overview of the advantage and disadvantage of different methods for metabolome analysis25. Based on this we have provided an overview of best practice in the field and also provide guidelines for performing metabolome analysis in other industrially important microoganisms. This inter-laboratory study has been supplemented with more specific studies on extraction of metabolites from yeast by SYSINBIO partners31,32. Whereas there has been good progress on metabolomics of yeast it has been found to be far more difficult to obtain truly quantitative data on intracellular metabolite levels in bacteria, as illustrated in a study on E. coli33.
Adaptive Evolution
Random mutagenesis has traditionally been used to optimize the performance of microbial strains used in industry, and it is still being used as a very powerful tool. Also in metabolic engineering it is often necessary to evolve strains further in order to adapt them to certain environmental conditions. Often directed genetic modifications that results in a desirable phenotype with respect to production of a certain metabolite lead to a reduction of the specific growth rate of the cells, and this may often result in low productivities of the strain. However, there have been several examples illustrating that following such directed genetic modifications resulting in engineering of the metabolic network, it is possible to evolve the recombinant strains towards faster growth using the concept of adaptive evolution. Hereby the productivity is increased and a strain with both high yield and productivity is obtained. In cases where random mutagenesis or evolution is used to identify microbial strains with desirable phenotypes there is much interest to identify the underlying mutations, as this information can be used to further improve our models of microbial metabolism and hereby this information can be used actively to further metabolic engineer the production strain. With the development of next generation sequencing technologies it has become possible to sequence microbial strains and hereby identify mutations acquired during adaptive evolution. In connection with the SYSINBIO different strategies for use of sequencing have been discussed and this has resulted in the several studies evaluated the use of sequencing for SNP identification has been evaluated. This has led to the development of a specific genome-browser for a widely applied S. cerevisiae strain (CEN.PK) and this genome-browser can be used as a database for queering specific questions34. The use of genome-sequencing has furthermore been demonstrated for identification of governing mutations acquired during adaptive evolution of yeast growth on galactose35.
Potential Impact:
The SYSINBIO project has ensured coordination of research efforts in the field of model guided metabolic engineering within Europe, and in particular ensured improved dialogue between academia and industry. This has led to a substantially improved transfer of ideas and technology from academia to industry and hence contributed to improve the competiveness of the European biotech industry. The project has also been instrumental in coordinating advanced education in the field of metabolic engineering in Europe by stimulating the exchange of students and by sponsoring workshops that have the graduate student as their main focus. It is further expected that this will have increased the awareness among all the partners of existing courses in the different partner countries, and this will have sustained impact following the project end.
In the last period of the project an international conference was organized that ensured the dissemination of new results generated by the participating research groups and further pushed forward the research field by bridging between academia and industry. The project has also resulted in coordination of the development of metabolic models and has led to the establishment of a database for existing models. This has led to exchange of concepts and methods for setting up genome-scale metabolic models. In the last part of the project this database has been further populated and it will be maintained by the partner groups following the project, which represents a major outcome of the project with a wide dissemination potential.
We have coordinated the development of dynamic models for design of metabolic engineering strategies. The concepts and methods for setting up and analyzing dynamic models have been summarized in a review paper. This involved defining efficient workflows for setting up dynamic models of metabolic reaction networks including parameter estimation, model validation and discrimination etc. We have in the project discussed the application of different simulation tools that allow the use of metabolic models for design of metabolic engineering strategies. This has helped to define some general guidelines for the use of metabolic models for metabolic engineering and move towards a consensus between the different research groups working on this field in Europe. We have evaluated different concepts for integration of "omics" data into metabolic models and we will in the last part of the project provide guidelines on how such data can be used for the reconstruction and evaluation of genome-scale metabolic models. These guidelines will allow for improved and faster model reconstructions in the future.
Very solid reports on novel tools for 13C-based fluxomics (both experimental and computational tools) have been drafted. These reports have already been disseminated to several companies. We have coordinated research activities on the use of DNA arrays, tiling DNA arrays and DNA sequencing for rapid identification of mutations that lead to improved phenotypic properties. In the last part of the project we will further evaluate the use of genome-sequencing for analysis of evolved mutants. This will lead to development of guidelines for which methods are preferential to use for analysis of evolved mutants in connection with strain design and optimization.
SYSINBIO has also attempted to raise the public awareness of industrial biotechnology. Several news and cover stories have been featured in Swedish newspapers, including Dagens Nyheter and Ny Teknik. In each case this was handled by providing press releases through assistance from Chalmers central administration.
As a global conclusion we can say that SYSINBIO has been instrumental to advance the industrial applications of systems biology in Europe both by coordinating research among different groups in academia and between industry and academia and by facilitating transfer of ideas and technology from academy to industry. SYSINBIO has also played a very relevant role in education of graduate students by facilitating exchanges and by giving graduate students in the partner academic groups a broad overview of the activities that are being carried out by different actors in the field.
List of Websites:
Further information about SYSINBIO can be obtained by contacting the coordinator:
Professor, dr.techn. PhD Jens Nielsen
Department of Chemical and Biological Engineering
Chalmers University of Technology
Kemivägen 10
SE412 96 Gothenburg
Sweden
Tel: 0046 31 702 3804
E-mail: nielsenj@chalmers.se
Or via the SYSINBIO web-site: www.sysbio.se/SYSINBIO
Metabolic engineering is an applied science focusing on developing new cell factories or improving existing ones. Metabolic engineering distinguishes itself from applied genetic engineering by the use of advanced analytical tools for identification of appropriate targets for genetic modifications and the use of mathematical models to perform in silico (computer based) design of optimized cell factories. In recent years, there has been increasing focus on using mathematical models for design and SYSINBIO has been coordinating European activities in the field of model driven metabolic engineering and activities on other technologies required for state of the art metabolic engineering, e.g. metabolomics (measurements of internal metabolites) and fluxomics (measurements of metabolic fluxes). The coordination of activities has resulted in the establishment of a database containing metabolic models for different industrially important microorganisms (the BioMet ToolBox available at www.sysbio.se). The database also contains different simulation tools required for use of these models to identify metabolic engineering targets and use of these models for analysis of omics data (high throughput measurements of biologically relevant entities such as proteins or messenger RNAs). SYSINBIO has also worked on coordinating further development of techniques required for metabolic engineering, such as metabolomics, fluxomics and identification of mutations in evolved strains through whole genome sequencing using second generation sequencing approaches. Several publications, in particular reviews, summarizing the state of the art in the field have been published in connection with SYSINBIO.
A key part of SYSINBIO was to coordinate activities on developing new and better mathematical models of metabolism that can be used for metabolic engineering. Activities on both stoichiometric models and kinetic models have been coordinated. The use of stoichiometric models has been further advanced and we have defined methods that allows for rapid setting up this kind of models. A fundamental problem with stoichiometric models is that it is necessary to perform a manual annotation and curation of the many different reactions, which is time consuming and laborious. We have therefore formulated methods and standards that will allow for faster development of this kind of models, and we have also developed a database of existing models that can be used as scaffolds for building new models. In the field of kinetic models we have performed a detailed review of the state-of-the art on the use of these models, and based on this we have drafted recommendations, in particular for use in focussed industrial metabolic engineering projects.
Project Context and Objectives:
Metabolic engineering is an applied science focusing on developing new cell factories or improving existing ones. There are several definitions, but most of these are consistent with: the use of genetic engineering to perform directed genetic modifications of cell factories with the objective to improve their properties for industrial application. In this definition the word improve is to be interpreted in its broadest sense, i.e. it also encompasses the insertion of completely new pathways with the objective to produce a heterologous product in a given host cell factory. Metabolic engineering is an enabling science, and distinguishes itself from applied genetic engineering by the use of advanced analytical tools for identification of appropriate targets for genetic modifications and the use of mathematical models to perform in silico design of optimized cell factories. Metabolic engineering is therefore often seen as a cyclic process, where the cell factory is analyzed and based on this an appropriate target is identified (the design phase). This target is then experimentally implemented and the resulting strain is analyzed again. Thus, metabolic engineering involves a continuous iteration between design and experimental work. In recent years, there has been increasing focus on using mathematical models for design. Hereby, it is expected that metabolic engineering will become faster and more efficient through the development of robust and reliable mathematical models describing the function of cell factories.
In recent years metabolic engineering has adapted tools from functional genomics and systems biology. There are many definitions of systems biology, but most of these contain elements such as mathematical modelling, global analysis (or "ome" analysis), mapping of interactions between cellular components, and quantification of dynamic responses in living cells. In most cases the objective of systems biology is to obtain a quantitative description of the biological system under study, and this quantitative description may be in the form of a mathematical model1,2. In some cases, the model may be the final result of the study, i.e. the model captures key features of the biological system and can hence be used to predict the behaviour of the system at conditions different from those used to derive the model. In other cases, mathematical modelling rather serves as a tool to extract information of the biological system, i.e. to enrich the information content in the data. There is not necessarily a conflict between the two, and generally, mathematical modelling goes hand in hand with experimental work. This partnership exemplifies the essence of systems biology, obtaining new insight into the molecular mechanisms occurring in living cells or sub-systems of living cells, through the combination of mathematical modelling and experimental biology. This does not say anything about the use of high-throughput data, e.g. transcriptome or proteome data, and clearly there are many systems biology studies that do not rely on such data. Mathematical models have, however, shown to be particularly useful for analysis of high-throughput data, as the complexity and integrative nature of biological systems makes it difficult to extract information on molecular processes from such data without the use of models as either scaffolds for the analysis or for hypothesis driven analysis of the data.
In the field of industrial biotechnology there is much focus on how systems biology can impact the development of efficient cell factories, and in particularly speed up the development process, and hereby ensure that new products can be brought to the market faster or there can be a faster improvement of existing bioprocesses.
The objectives of SYSINBIO was to coordinate research activities in the field of model driven metabolic engineering in Europe and according to the concept described above coordinate how systems biology can be used to improve the performance of metabolic models and further how mathematical models can be used for improved design of cell factories used in industrial biotechnology. The overall goal was divided into 11 specific objectives that were dealt with in 11 different workpackages that formed the basis of the project:
- Research efforts in the field of model guided metabolic engineering within Europe were coordinated through organizing yearly meetings of leading European research groups working in this field.
- Advanced education in the field of metabolic engineering in Europe was coordinated. This has involved setting up and running dedicated courses and workshops on mathematical modelling and metabolic engineering, but it also involved advanced research training through exchange of researchers between the involved laboratories.
- SYSINBIO organized an international conference with focus on the use of model guided metabolic engineering for development of efficient cell factories. The conference ensured dissemination of new results generated by the participating research groups and further pushed forward the research field by bridging between academia and industry.
- SYSINBIO coordinated the development of metabolic models and has provided a database for existing models. Concepts and methods for setting up genome-scale metabolic models were exchanged and developed resulting in implementation of best practices and efficient work flows in the partner laboratories.
- SYSINBIO also coordinated the development of dynamic models for design of metabolic engineering strategies. Concepts and methods for setting up and analyzing dynamic models was exchanged and developed. Efficient workflows for setting up dynamic models of metabolic reaction networks including parameter estimation, model validation and discrimination, parameter estimation, and simulation were defined for different types of dynamic models.
- SYSINBIO coordinated the development of novel simulation tools that allow the use of metabolic models for design of metabolic engineering strategies. Different algorithms for identification of metabolic engineering targets through the use of genome-scale metabolic models were compared and evaluated, and based on the evaluation, general guidelines for the use of metabolic models for metabolic engineering was specified.
- SYSINBIO coordinated the integration of omics data with metabolic models in order to improve their predictive strength. Different concepts and methods for integration of "omics" data into metabolic models was compared and evaluated leading to guidelines on how such data can be used for the reconstruction and evaluation of genome-scale metabolic models.
- SYSINBIO coordinating the development of novel tools for 13C-based fluxomics (both experimental and computational tools). Different tools for fluxomics were compared and evaluated, and guidelines for quantification of metabolic fluxes in microorganisms, during industrial fermentations, have been proposed.
- SYSINBIO coordinated the development of novel tools for metabolomics. Different methods for sampling and analysis of key intracellular metabolites was compared and evaluated, and guidelines for use of different analytical tools for the analysis of intracellular metabolites, in different industrially important microorganisms, was set up.
- SYSINBIO coordinated the use of statistical methods and metabolic models for upgrading the information content in "omics" data. Different methods and concepts were evaluated for their ability to upgrade the information content in "omics" data, and in particularly it was evaluated how this kind of data can be used to identify novel metabolic engineering targets.
- Finally, SYSINBIO coordinated research activities on the use of DNA arrays, tiling DNA arrays and DNA sequencing for rapid identification of mutations that lead to improved phenotypic properties, e.g. arising in connection with adaptive evolution. The use of these different methods for mapping of mutations in strain lineages was evaluated.
Project Results:
The consortium behind SYSINBIO involved all the leading groups on metabolic engineering in Europe, and it had particularly strong competence on metabolic engineering of the yeast Saccharomyces cerevisiaae and the bacterium Escherichia coli. For these two microorganisms a number of different technologies had already been developed and implemented before the start of the project, and during the project there was further advancement in technologies. However, the strength of SYSINBIO was that there was also much experience on working with other industrially important microorganisms. The key microorganisms studied within SYSINBIO were:
- Saccharomyces cerevisiae, which is used for the production of bioethanol and a range of pharmaceutical proteins. S. cerevisiae is also considered as a general production platform for a range of different chemicals, e.g. isoprenoids (flavours, fragrances, pharmaceuticals), bulk chemicals (succinate), nutraceuticals (resveratrol and poli-unsaturated fatty acids known as PUFAs), in some of the partner laboratories. S. cerevisiae also serves as an important model organism for studying eukaryotic cells.1,3
- Escherichia coli, which is used for production of many different recombinant proteins, e.g. human growth hormone and various interleukins, and a wide range of different chemicals like amino acids, e.g. phenylalanine (used in the production of Aspartame®) and methionine (used as a feed). E. coli is also considered as a general production platform for chemical production, and it has traditionally served as the most important model organism for studying bacterial physiology and genetics.
- Aspergillus niger, which is used for the production of organic acids (citric acid and gluconic acid) and a wide range of industrial enzymes.
- Aspergillus oryzae, which is used for the production of a wide range of industrial enzymes.
- Trichoderma reesei, which is used for the production of celluloses used for processing of fibers and for biomass egradation in connection with production of fuels and chemicals from biomass.
- Corynebacterium glutamicum, which is used for the production of lysine and other amino acids used in the feed industry.
- Lactococcus lactis, which is used in starter cultures and also serves as an important model organism for lactic acid bacteria.
- Bacillus subtilis, which is used for the production of industrial enzymes, different chemicals, e.g. riboflavin, and different antibiotics, e.g. bacitracin.
- Streptomyces species, which are used for the production of antibiotics like nystatin, industrial enzymes, and recombinant proteins.
Besides broad competence on different industrially important microorganisms the partners of SYSINBIO covered competence on all key technologies that are required for state of the art metabolic engineering, namely:
- Well controlled fermentation experiments: The ability to perform well controlled fermentation experiments, e.g. using chemostat cultures or computer controlled fed-batch experiments, is essential for metabolic engineering, as it allows for detailed quantitative analysis of the cell factory. The consortium covered extensive experience in this field for a wide range of microorganisms.
- Transcriptome analysis: The partners have performed genome-wide transcription analysis of many different industrially important microorganisms, and in particularly demonstrated how this omics technique can be used for gaining insight into the operation of metabolic networks.
- Metabolome analysis: Several of the partners are among the World leaders in metabolomics of industrially important microorganisms, and has been involved in defining standards and protocols in this field.
- Fluxome analysis: The consortium includes most of the World leading research groups in the field of fluxome analysis. Partners of the consortium have pioneered the use of GC-MS analysis for measurement of isotopomers and they have recently brought the field further forward by using isotopomer measurements of key intermediates in the central carbon metabolism.
- Genome-scale metabolic models: The consortium has played an active role in setting up genome-scale metabolic models for a range of industrially important microorganisms.
- Thermodynamic analysis: The consortium has demonstrated the use of thermodynamics for analysis of large metabolic networks and illustrated how thermodynamics can be used to identify flux control coefficients in key biosynthetic pathways.
- Dynamic models: Based on its strong competences the consortium has evaluated the use of dynamic models for analysing the operation of metabolic networks and has developed several different modelling concepts for simulation of dynamic operation of metabolism.
- Model simulation and bioinformatics: The partners in the consortium have developed several new algorithms for analysis of omics data in the context of metabolic models as well as they have developed different simulation platforms that allow simulation of both stoichiometric and dynamic mathematical models. SYSINBIO ensured further advancement in this field.
Based on combination of these strong competences SYSINBIO was able to significantly advance the field and bridge stronger between academia and industry through the coordination activities described further in the following.
Training and Education
SYSINBIO was involved in organizing several advanced courses and student workshops:
- The International Yeast Course on Systems Biology that was held June 1-18, 2009 in Gothenburg. The course was sponsored by FEBS. 4 days of this course were focusing on metabolic engineering technologies, such as metabolic flux analysis, metabolic modelling, and metabolic control analysis. The course had 24 full time students, but parts of the course were offered to local students, of which there were more than 15. The 24 full time students came from 19 different countries.
- The International Course on Systems Biology of Metabolism that ran in the period from May 24 to June 11, 2010 at Chalmers University of Technology. The course had 22 participants from all over Europe. The course consisted of five blocks of teaching: 1) An introduction to systems biology; 2) Setting up and applying genome scale metabolic models; 3) Analysis of omics data and their integration with genome-scale models; 4) Dynamic mathematical models of metabolism and signalling pathays; 5) Examples in biotechnology and medicine. More information about the course can be found at: http://www.sysbio.se/ICSBM2010/.
- A student workshop that was held on December 8-9 in Istanbul . The workshop was chaired by Prof. Vassily Hatzimanikatis, EPFL. At the workshop the 20 participants (PhD students and post docs) each gave a 30 min presentation about their research work.
- A student workshop that was held on January 18-19 at the DECHEMA-Haus in Frankfurt. The workshop was chaired by Prof. Ralf Takors, University of Stuttgart, and Dr. Sergio Velasco, Chalmers. At the workshop the 20 participants (PhD students and post docs) each gave a 30 min presentation about their research work. The presentations were followed by extensive discussions involving both the participants and the organizers. In connection with the workshop there were two special lectures:
- Probabilistic Boolean framework to decrypt gene regulatory networks case studies with primary human hepatocytes and E. coli, by Prem Kumar
- Metabolome-based 13C-metabolic flux analysis; A tool for measuring in vivo metabolic activities, by Katharina Nöh
Besides the scientific presentations the workshop also allowed the participant to network through a social event organized over dinner.
- A PhD course entitled "Industrial Biotechnology for lignocellulose based processes" that was held at Chalmers, Gothenburg, in the period 16th to 21st October 2011 and organized by Professor Lisbeth Olsson. The course was relevant for PhD students and researchers working in the field of lignocellulose based processes, i.e. the production of fuels and other chemicals using plant cell wall as raw material. The idea of the course was to encompass all parts of the subject from raw material composition and sources, to pretreatment and hydrolysis, enzymes that act on plant cell wall material, microorganisms and their improvement for production of targeted biofuels and chemicals, the fermentation process and last but not least analytics. The course program included an evening poster session, where the participants had the chance to present their research and visit Professor Olsson laboratories. The course program was closed with a day of seminars on "Hot topics in industrial biotechnology". The course was attended by 25 students of 17 different nationalities, coming from institutions in the following countries: Sweden, Denmark, Norway, Finland, United Kingdom, Belgium, Germany, Italy, Spain and Turkey. SYSINBIO sponsored 7 students from the SYSINBIO consortium. 15 lecturers from Sweden, Denmark, Finland, Norway and Italy was involved in the course. Several of them were partners of the SYSINBIO.
International Conference
SYSINBIO organized a 2 day conference "Industrial Systems Biology: Sustainable production of fuels and chemicals". The conference was held at Chalmers University of Technology, Gothenburg, Sweden August 18-19, 2010 (the conference was organized as one of the conferences in the Gothenburg Life Science Conference series). In connection with the conference there was also organized a dedicated student workshop on August 17 (3 hours) where one speaker from academia (Professor Gregory Stephanopoulos, MIT, USA) and one speaker from industry (Dr. Alan Berry, Novozymes, USA/Denmark) discussed job prospects and industrial demand for future students.
At the conference there were 20 oral presentations and 40 poster presentations. There were more than 170 participants at the conference. Among the speakers were: Jens Nielsen, Chalmers, Gothenburg Sweden; Vassily Hatzimanikatis, EPFL, Lausanne, Switzerland; Kiran Patil, DTU, Lyngby, Denmark; Matthias Heinemann, ETH Zürich, Switzerland; Elmar Heinzle, Saarland University, Germany; Sang Yup Lee, KAIST, Daejon, Korea; Christoph Wittmann, TU Braunschweig, Braunschweig, Germany; Andrew C. Eliot, DuPont, Wilmington, USA; Jack Pronk, TU Delft, The Netherlands; Alan Berry, Novozymes, Davis, USA; Verena Siewers, Chalmers, Gothenburg, Sweden; Jochen Förster, Fluxome, Copenhagen, Denmark; Eric Appelman, Perstorp, Sweden. A detailed report from the conference has been published in Biotechnology Journal4. Furthermore, based on oral presentations several papers were published in a special issue of Biotechnology journal (issue 3, 2011) entitled "Systems biology for industrial applications", and the editorial provides perspectives based on conclusions of the conference5.
Mathematical Models and Their Use in Industrial Biotechnology
A key task of SYSINBIO was to coordinate activities on developing new and better mathematical models of metabolism that can be used for metabolic engineering. Activities on both stoichiometric models and on kinetic models have been coordinated. Stoichiometric models, often also referred to as genome-scale metabolic models (GEMs), have the advantage that they are directly linked to genomic information6,7, and methods for set up and use of GEMs has been further advanced. In particular SYSINBIO was active in defining methods that allows for faster setting up this kind of models8. A fundamental problem with stoichiometric models is that it is necessary to perform a manual annotation and curation of the many different reactions, and this is time consuming and laborious. We have therefore formulated methods and standards that will allow for faster development of this kind of models, and we have also developed a database of existing models that can be used as scaffolds for building new models9. We have also worked on different user-friendly graphical interfaces that allows for visualization of these complex models. For kinetic models we have performed an extensive survey of existing models in the partner laboratories (and of published models), and we have used this to discuss how these models can be used in metabolic engineering8,10. A fundamental problem with kinetic models is the requirement for parameter estimation, but partners within SYSINBIO are using different modelling concepts that partly circumvent this problem, e.g. Monte Carlo simulations and use of linearized kinetics in the logarithmic domain11,12. Furthermore, workflows for setting up kinetic models are in the process of being defined and these will be used to define the requirements for experimental data for validation and parameter estimations (unpublished).
The coordination of the development of more reliable and standardized microbial models has allowed for the improvement of model predictions and therefore made it possible to apply simulation and optimization algorithms that can identify gene modifications that can increase the production yields of desired compounds. We have coordinated the development of reliable and effective computational and mathematical methods for the design of rational metabolic engineering strategies13,14,15,16. Through activities we have ensured that the simulation and optimization tools developed obey the following requisites: 1) robustness regarding the different models used (stoichiometric versus dynamic models) and metabolic engineering objectives; 2) compatibility with existing model storage standards (SBML and other formats); and 3) straightforward incorporation in user-friendly software tools to facilitate its use by non-expert users. Another important part of the activities has been the coordination of the development of user friendly software tools, something that is allowing for a wider use of mathematical modelling in the field of metabolic engineering16. Finally SYSINBIO has been working on improving the predictive strength of GEMs through incorporation of regulatory information17.
Integration of Omics Data
A key challenge in metabolic engineering is to be able to integrate omics data, e.g. transcriptomics, proteomics and metabolomics, with mathematical models with the objective to identify novel metabolic engineering targets. In SYSINBIO there were several ways to advance this. Flux data obtained from 13C-labelled experiments have been shown to allow constraining the fluxes in GEMs and hereby obtain better constraints on the operation of these large metabolic networks18. We have further used a combination of thermodynamics and experimental fluxes to identify flux constraining reactions19. This has substantially improved the predictive strength of these models for finding metabolic engineering targets. SYSINBIO also demonstrated how a combination of random sampling of GEMs combined with transcriptome data can be used to identify whether flux control is at the transcriptional or metabolic level20.
Another key challenge in this field is visualization of the rather complex data, but through the use of GEMs as scaffold it has been possible to easily visualize how omics data link to specific enzymatic reactions21, and this has been provide in a user-friendly software algorithm.
Due to the high connectivity of the different metabolic reactions within the metabolic network, there has been much interest in exploiting tools from functional genomics for mapping of global regulatory structures or even using high-throughput experimental techniques provided by the various omics for dissecting how the fluxes through the different branches of the metabolic network are controlled. In particular it is interesting to map the level of control of fluxes, i.e. is it at the metabolism level or is there control of flux at the transcriptional, translational or post-translational level. This kind of information can only be obtained through a combination of mathematical models and high-quality experimental data on metabolites and fluxes, and possible other omes. The partners of SYSINBIO have in several studies demonstrated how integrative analysis of omics data through the use of GEMs and other network structures can lead to new biological insight, in particular on how metabolic fluxes are controlled. This has involved both the identification of key transcription factors controlling parts of metabolism in yeast22,23 and how the key protein kinase Snf1 control yeast metabolism24. These concepts were also demonstrated to be useful to gain insight into how different yeast strains control their energy metabolism25. However, these concepts have not only been used for studies of yeast, as the concepts have also been used to gain new insight into enzyme production by Aspergillus oryzae26. As part of these studies there was also performed a detailed characterization of different transcriptome platforms for yeast, and even though these results were specific for yeast they are likely to have translational character and hence some general guidelines for transcriptome analysis in industrial microorganisms has been proposed25.
Fluxomics
Metabolic fluxes can either be estimated through the use of flux balance analysis (FBA) using GEMs or through the use of 13C-labelled substrate feeding followed by analysis of the labelling patterns in intracellular metabolites. Due to their abundance and stability, 13C-based methods have conventionally used proteinogenic amino acids to detect labelling patterns. Recently, however, methods for direct analysis in the free pool of metabolites have been developed. Both methods require a tight integration of experiments and model simulations for estimation of the fluxes, and in particular the mathematical complexity involved in flux estimation has been a barrier for a wider use of fluxomics. Besides providing general information on how the metabolic network is operating at different growth conditions, metabolic flux analysis is very well suited for analysis of the effects of growth on different media, specific mutations and screening of different mutants. Thus, flux analysis today represents a standard technique for rapid phenotypic characterization of metabolically engineered strains, and this tool is likely to gain even wider use in the future. Within SYSINBIO there has been coordination of developments in this field, and this has resulted in several key reviews and papers specifying guidelines for how to perform metabolic flux analysis27,28,29,30.
Metabolomics
Another important analytical tool in metabolic engineering is metabolome analysis, as this allows for direct quantification of the levels of intracellular metabolites. As these levels determine the kinetics of the different reactions, the metabolite levels together with the fluxes represents key information on how metabolism can be engineered. SYSINBIO has coordinated activities on metabolome analysis in the different partner laboratories. Many of the partner labs were involved in a thorough inter-laboratory comparison of metabolome analysis of the yeast Saccharomyces cerevisiae and this allowed to provide an overview of the advantage and disadvantage of different methods for metabolome analysis25. Based on this we have provided an overview of best practice in the field and also provide guidelines for performing metabolome analysis in other industrially important microoganisms. This inter-laboratory study has been supplemented with more specific studies on extraction of metabolites from yeast by SYSINBIO partners31,32. Whereas there has been good progress on metabolomics of yeast it has been found to be far more difficult to obtain truly quantitative data on intracellular metabolite levels in bacteria, as illustrated in a study on E. coli33.
Adaptive Evolution
Random mutagenesis has traditionally been used to optimize the performance of microbial strains used in industry, and it is still being used as a very powerful tool. Also in metabolic engineering it is often necessary to evolve strains further in order to adapt them to certain environmental conditions. Often directed genetic modifications that results in a desirable phenotype with respect to production of a certain metabolite lead to a reduction of the specific growth rate of the cells, and this may often result in low productivities of the strain. However, there have been several examples illustrating that following such directed genetic modifications resulting in engineering of the metabolic network, it is possible to evolve the recombinant strains towards faster growth using the concept of adaptive evolution. Hereby the productivity is increased and a strain with both high yield and productivity is obtained. In cases where random mutagenesis or evolution is used to identify microbial strains with desirable phenotypes there is much interest to identify the underlying mutations, as this information can be used to further improve our models of microbial metabolism and hereby this information can be used actively to further metabolic engineer the production strain. With the development of next generation sequencing technologies it has become possible to sequence microbial strains and hereby identify mutations acquired during adaptive evolution. In connection with the SYSINBIO different strategies for use of sequencing have been discussed and this has resulted in the several studies evaluated the use of sequencing for SNP identification has been evaluated. This has led to the development of a specific genome-browser for a widely applied S. cerevisiae strain (CEN.PK) and this genome-browser can be used as a database for queering specific questions34. The use of genome-sequencing has furthermore been demonstrated for identification of governing mutations acquired during adaptive evolution of yeast growth on galactose35.
Potential Impact:
The SYSINBIO project has ensured coordination of research efforts in the field of model guided metabolic engineering within Europe, and in particular ensured improved dialogue between academia and industry. This has led to a substantially improved transfer of ideas and technology from academia to industry and hence contributed to improve the competiveness of the European biotech industry. The project has also been instrumental in coordinating advanced education in the field of metabolic engineering in Europe by stimulating the exchange of students and by sponsoring workshops that have the graduate student as their main focus. It is further expected that this will have increased the awareness among all the partners of existing courses in the different partner countries, and this will have sustained impact following the project end.
In the last period of the project an international conference was organized that ensured the dissemination of new results generated by the participating research groups and further pushed forward the research field by bridging between academia and industry. The project has also resulted in coordination of the development of metabolic models and has led to the establishment of a database for existing models. This has led to exchange of concepts and methods for setting up genome-scale metabolic models. In the last part of the project this database has been further populated and it will be maintained by the partner groups following the project, which represents a major outcome of the project with a wide dissemination potential.
We have coordinated the development of dynamic models for design of metabolic engineering strategies. The concepts and methods for setting up and analyzing dynamic models have been summarized in a review paper. This involved defining efficient workflows for setting up dynamic models of metabolic reaction networks including parameter estimation, model validation and discrimination etc. We have in the project discussed the application of different simulation tools that allow the use of metabolic models for design of metabolic engineering strategies. This has helped to define some general guidelines for the use of metabolic models for metabolic engineering and move towards a consensus between the different research groups working on this field in Europe. We have evaluated different concepts for integration of "omics" data into metabolic models and we will in the last part of the project provide guidelines on how such data can be used for the reconstruction and evaluation of genome-scale metabolic models. These guidelines will allow for improved and faster model reconstructions in the future.
Very solid reports on novel tools for 13C-based fluxomics (both experimental and computational tools) have been drafted. These reports have already been disseminated to several companies. We have coordinated research activities on the use of DNA arrays, tiling DNA arrays and DNA sequencing for rapid identification of mutations that lead to improved phenotypic properties. In the last part of the project we will further evaluate the use of genome-sequencing for analysis of evolved mutants. This will lead to development of guidelines for which methods are preferential to use for analysis of evolved mutants in connection with strain design and optimization.
SYSINBIO has also attempted to raise the public awareness of industrial biotechnology. Several news and cover stories have been featured in Swedish newspapers, including Dagens Nyheter and Ny Teknik. In each case this was handled by providing press releases through assistance from Chalmers central administration.
As a global conclusion we can say that SYSINBIO has been instrumental to advance the industrial applications of systems biology in Europe both by coordinating research among different groups in academia and between industry and academia and by facilitating transfer of ideas and technology from academy to industry. SYSINBIO has also played a very relevant role in education of graduate students by facilitating exchanges and by giving graduate students in the partner academic groups a broad overview of the activities that are being carried out by different actors in the field.
List of Websites:
Further information about SYSINBIO can be obtained by contacting the coordinator:
Professor, dr.techn. PhD Jens Nielsen
Department of Chemical and Biological Engineering
Chalmers University of Technology
Kemivägen 10
SE412 96 Gothenburg
Sweden
Tel: 0046 31 702 3804
E-mail: nielsenj@chalmers.se
Or via the SYSINBIO web-site: www.sysbio.se/SYSINBIO
