January 1, 2007 (Vol. 27, No. 1)
Jose Manuel Otero
Lisbeth Olsson Ph.D.
Jens Nielsen Ph.D.
Petrochemical Industry Finding Success by Turning to Biotechnology
Had this article been published three to five years ago, the target audience would likely have been medical, bioprocess, or pharmaceutical biotechnologists focused on applying systems biology to areas such as regulatory responses in cell signaling pathways, protein expression, or the rational design of novel therapeutic agents. Today, however, the target audience for systems biology has expanded to include biotech and bioprocess development groups. The tools of this field now impact the design of cell factories used by major bulk, intermediate, and specialty chemical manufacturing companies, often including petroleum manufacturers and refiners.
What is the motivation behind industrial biotechnology’s recent penetration into the chemical manufacturing world? The cost of petroleum increased nearly 150% between January 2001 and 2006 from $22/bbl to $55/bbl. Global energy consumption is projected to increase by 57% to 681 trillion MJ between 2002 and 2025. The rapidly emerging economies of China, India, and Russia (2005 GDP growth rates of 9.9%, 7.6%, and 6.4%, respectively, compared to the average world GDP growth rate of 4.7%) are consuming energy sources at a record pace. These trends are driving traditional chemical manufacturing companies to fight against strong competition for their primary feedstock, petroleum.
Petroleum products refined from crude oil are generally classified into three categories: transportation fuels, finished nonfuel products, and feedstock for the chemical industry. In 2005 more than 75% of all petroleum was converted and sold as fuel, while less than 5% served as feedstock.
As most process development and manufacturing groups will agree, the startup of a new process often involves significant capital and operating expenses that over time, with improvements in technology, gains in operating efficiencies, and release of second-generation processes, will decrease.
As the process matures the largest cost fraction will be the raw materials. The industry is faced with a significant challenge in identifying sustainable raw materials that can be used in cost-effective, robust, and high end-product yield, titer, productivity, and quality processes. Industrial biotechnology coupled with recent developments in the fields of systems biology and metabolic engineering is offering such processes.
The Beginnings of Industrial Biotech
Industrial biotechnology, often referred to as white biotechnology in Europe, is the conversion of biomass via biocatalysis using microbial fermentation or enzyme catalysis to produce chemicals, materials, and/or energy. Here, we define biomass as an organic-based polymer resulting from photosynthetic carbon fixation (typically CO2), which in monomer form may include glucose, xylose, galactose, mannose, or similar monosaccharide.
Industrial biotechnology is by no means a new field. Fermentation processes for antibiotics, vitamins, organic acids, and amino acids are well established.
In each of these examples, host organisms well suited for production of the target compound were naturally isolated. Furthermore, under controlled environments, random mutagenesis followed by screening, selection, and traditional bioprocess development were used to enhance production yields, titers, productivities, and robustness. This method while providing little to no mechanistic understanding of which specific genetic perturbations lead to improved strains so that they could be further exploited, has proven to be commercially successful.
In the late 1980s and early 1990s, with recombinant DNA technology emerging from medical biotechnology, we witnessed expression of compounds previously produced via synthetic routes now being attempted in production organisms. This was made possible by the introduction of genetic sequences encoding for enzymes that were likely to catalyze desired reactions or the deletion of genes that would down-regulate undesired reactions and pathways. These approaches were largely hypothesis driven, resource intensive, and low-throughput. Such methods thus minimized the probability of successfully identifying a genotype that would elicit a significantly improved phenotype.
The real advantage of random mutagenesis, screening, and selection was the relatively large experimental space that could be covered, even if mechanistic understanding was sacrificed. The other advantage was its track record—it worked. Fast-forward approximately 10 years, and what has changed?
Although techniques that permitted manipulation of recombinant DNA existed, the annotated genome sequences of industrially relevant production hosts were not available. As of December 2006, there is a total of 470 published genome sequences, and 991 bacterial, 631 eukaryotic, and 56 archaeal sequence projects on-going. This genomic revolution is evident from the genomes sequenced between 1995 and 1999.
During this period, 24 genome sequences were made available, yet only three could be considered to have broad applicability to the industrial biotechnology sector—Saccharomyces cerevisiae, Escherichia coli, and Bacillus subtilis—while the rest were driven by the medical community. Even these three have strong medical relevance as they are important model organisms. If we move beyond 1999 many more industrially important cell factories have been sequenced.
With the substantial reduction in sequencing costs, even genome sequencing has become a tool to analyze cell factories with different phenotypes. The presence of complete genome sequences has clearly allowed better targeting of genetic modifications, and information about the complete parts lists of a given cell factory is extremely valuable.
Even though genome sequencing has clearly facilitated the use of targeted genetic modifications for construction of cell factories with desirable phenotypes, the major step forward has been through the introduction of metabolic engineering, which is an enabling science for cell factory design and construction. Metabolic engineering involves the identification of specific and targeted genetic modifications followed by implementation of these modifications via molecular biology tools that lead to re-direction of fluxes to enhance production or robustness of a given product or organism, respectively.
A key technology in the successful application of metabolic engineering is the availability of a well annotated genome, including the quantitative tools that permit careful inspection and manipulation of the genome. Among those tools has been the recent development of genome-scale metabolic models (GSMMs). To develop a model of cellular metabolism that enables the prediction of concentration profiles as functions of time, the stoichiometry and kinetic reaction rates for each biochemical reaction in a cell at physiological conditions would be required. At present, this information is not available, neither via estimation or experimental measurement.
Through careful annotation based on existing biochemical knowledge, literature review, and experimentation, however, it is possible to associate known genes with known biochemical reactions and their associated stoichiometry. The result is a biochemical model describing the formation and depletion of each metabolite. This provides mass-balance boundary conditions making possible constraint-based simulations of how the metabolic network operates at different conditions. In simpler terms, using basic stoichiometry, these models can be used to predict the relationships between genes with function in the metabolic network operating in a cell.
GSMMs have been developed for several model production organisms and was a major step in formalizing the field of systems biology.
Systems biology is the quantitative characterization of genetic, transcription, protein, metabolic, signaling, and other informational pathway responses to a clearly defined perturbation of a biological system. The perturbation may be in terms of a genetic, chemical, or environmental stimulus. At the core of systems biology is the transformation of quantitative, typically large-scale data sets, into in silico models that provide both interpretation and prediction. GSMMs provide a framework of how x-ome data may be organized and over-laid on the metabolic network.
As technologies have become more accessible for transcriptome (DNA oligonucleotide and cDNA microarrays), proteome (2-D gel electrophoresis coupled to MS or direct MS analysis), fluxome (isotopically labeled substrates coupled to detection by GC-MS), and metabolome (numerous analytical methods, including LC-MS and GC-MS) measurements, enormous data sets have been generated that require bioinformatics and quantitative models to be developed for data analysis, interpretation, and prediction.
Industrial biotechnology is beginning to exploit the benefits of these tools realizing that metabolic engineering strategies for improved process development may first be screened in silico, producing a reduced, specific, and high-probability of success list of genetic perturbations that should be experimentally validated. The process is highly iterative, with strain construction and characterization providing new x-ome data that can be used to improve the models (i.e., experimental quality control of in silico models) and metabolic engineering strategies.
Therefore, we here define industrial systems biology, acknowledging that tools established in the rapidly growing systems biology research community are graduating to commercial successes. Industrial systems biology is prevalent in two forms—either existing companies are reshaping or forming new process development groups with industrial systems biology capabilities and expertise or they are out-sourcing process development to small, recently formed entities that specialize in industrial systems biology.
It will be of no surprise that the largest industrial biotechnology product in the world, recently garnering unprecedented corporate, social, and government support, is bioethanol. In 2005, total world production was 46 billion liters, with the production volume and the total number of refineries built between 2005 and 2006 in the U.S. increasing by 2.6 billion liters and 14, respectively. The producers of bioethanol are using a variety of fermentation platforms, however, S. cerevisiae is among the more popular serving as a credit to its robustness for large-scale (>300,000 L) fermentation processes.
Critical to bioethanol economics is the purchase and pretreatment processing costs of the raw material feedstock. Although corn has been the dominant feedstock in the U.S., most experts acknowledge that other feedstocks, especially those rich in lignocellulose, will be required to ensure the sustainability and self-sufficiency of bioethanol production. For example, metabolic engineering has played a critical role in developing production strains that can process the most abundant pentose sugar in hemicelluloses, hardwoods, and crop residues—xylose—a monosaccharide second only to glucose in abundance.
Native S. cerevisiae does not metabolize xylose. Metabolic engineering using systems biology tools revealed, however, that introduction of key enzymes (xylose reductase, xylitol dehydrogenase, xylulokinase, or xylose isomerase) made possible xylose utilization. Subsequent metabolic engineering efforts have improved xylose consumption efficiency, with small decreases in growth rate and ethanol yield as compared to glucose.
Additional areas of bioethanol process development include identification of thermostable cellulases, development of ethanologenic bacteria, and further metabolic engineering of S. cerevisiae to decrease formation of by-products, such as glycerol and organic acids. Bioethanol is both a catalyst and advocate for future applications of industrial biotechnology, particularly bulk chemical manufacturing.
An example of a product being launched, previously produced via petrochemical conversion and now made possible by industrial biotechnology, is produced by DuPont’s new technology platform, DuPont Bio-Based Materials, and is called Bio-PDO™, commonly known as 1,3-propanediol (PDO). PDO is a critical intermediate in the production of polymers commonly used for apparel, fiber, and carpet industries, and serves as an intermediate for DuPont’s new polymer platform, Sorona™. Bio-PDO is produced in E.coli via the conversion of D-glucose through the central carbon metabolism to dihydroxyacetone phosphate, further to glycerol, and finally to 1,3-propanediol.
The development of the cell factory was done in close collaboration with Genencor (www.genencor.com), whereas the process development was done in collaboration with Tate & Lyle. The manufacturing facility was completed in 2006, and using corn as the principle feedstock, it will produce 45 million kg/year at full capacity, resulting in 40% less energy use compared with the petrochemical equivalent. Bio-PDO is among the first success stories for metabolic engineering and industrial biotechnology in the bulk chemical industry. Yet it should be realized that greater then 10 years of development and significant resources were invested to reach this milestone.
The final example is a compound produced via petrochemical conversion of maleic anhydride, for which an industrial biotechnology process is highly desired. Succinic acid is a four carbon organic acid identified by the United States Department of Energy in 2004 as a top 10 building block molecule for production via biotechnology. Succinic acid is the starting material for synthesis of many chemicals, including butanediol, tetrahydrofuran, gamma-butyrolactone, and poly-amides, used in solvents and polymers with an annual world market size exceeding $1.3 billion.
Numerous groups are exploring succinic acid production with significant metabolic engineering using Anaerobiospirillum succiniciproducens, Actinobacillus succinogenes, E. coli, and Mannheimia succiniciproducens.
Of particular interest is M. succiniciproducens isolated in 2002 by researchers at the Korea Advanced Institute of Science & Technology. In less than four years they have sequenced and annotated the genome, developed a genome-scale metabolic model, collected transcriptome, proteome, and fluxome data, and experimentally demonstrated significant improvements in succinic acid productivity predicted by in silico simulations.
The Center for Microbial Biotechnology at the Technical University of Denmark is using similar approaches to metabolically engineer S. cerevisiae and Aspergillus niger for overproduction of succinic acid, exploiting the x-ome data and systems biology tools already available in these industrially proven production organisms.
Although significant effort is still required to reach commercialization, the progress and speed of development is enhanced due to the systems biology toolbox now in practice.
Succinic acid and other compounds, such as aspartic acid, promise to be realistically achievable targets for industrial biotechnology. Industrial systems biology is being successfully applied and will continue to be in demand, not only by specialty biotechology companies, but by the large chemical manufacturing companies that traditionally relied on synthetic chemistry.
Further evidence is the emergence of start-up companies focused on providing industrial systems biology expertise to process development groups. Although small, these companies have significant collaborations with many of the major chemical manufacturing, nutraceutical, pharmaceutical, and petrochemical companies.
Looking forward, the goal will be to integrate industrial biotechnology processes that utilize common raw materials, production hosts, and platform technologies into a biorefinery. The term biorefinery was first coined in 1999, when it was proposed that lignocellulosic raw materials could be converted to numerous biocommodities via integrated unit processes, where the output of the biorefinery would adjust to market demands similar to present day petrochemical refineries.
For the biorefinery platform to be commercialized there are two essential components. First, the economic and socio-political landscape must support and warrant the significant financial investment, favorable legislative policy, and consumer driven demand that will be required.
Second, the advances and tools developed within systems biology must graduate to industrial systems biology, enabling metabolic engineering in a rapid, dynamic, and commercial environment. The transition to industrial systems biology, while requiring significant investment, promises to offer a return worth waiting for.
Jose Manuel Otero (jomo@biocentrum. dtu.dk) is a Merck Doctoral Fellow, Lisbeth Olsson, Ph.D., is professor, and Jens Nielsen, Ph.D., is professor and center director at the Center for Microbial Biotechnology, BioCentrum-Technical University of Denmark and CSO of Fluxome Sciences.