UNDERSTANDING METHANOTROPHIC POLYHYDROXYBUTYRATE (PHB) PRODUCTION ACROSS SCALE: LIFE CYCLE ASSESSMENT, PURE CULTURE EXPERIMENTATION, AND PATHWAY/GENOME DATABASE DEVELOPMENT
A DISSERTATION SUBMITTED TO THE DEPARTMENT OF CIVIL AND ENVIRONMENTAL ENGINEERING AND THE COMMITTEE ON GRADUATE STUDIES OF STANFORD UNIVERSITY IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
Katherine Helen Rostkowski June 2012
© 2012 by Katherine Helen Rostkowski. All Rights Reserved. Re-distributed by Stanford University under license with the author.
This work is licensed under a Creative Commons Attribution- Noncommercial 3.0 United States License. http://creativecommons.org/licenses/by-nc/3.0/us/
This dissertation is online at: http://purl.stanford.edu/mc120yq3299
ii I certify that I have read this dissertation and that, in my opinion, it is fully adequate in scope and quality as a dissertation for the degree of Doctor of Philosophy.
Craig Criddle, Primary Adviser
I certify that I have read this dissertation and that, in my opinion, it is fully adequate in scope and quality as a dissertation for the degree of Doctor of Philosophy.
Michael Lepech
I certify that I have read this dissertation and that, in my opinion, it is fully adequate in scope and quality as a dissertation for the degree of Doctor of Philosophy.
Perry McCarty
I certify that I have read this dissertation and that, in my opinion, it is fully adequate in scope and quality as a dissertation for the degree of Doctor of Philosophy.
Peter Karp
Approved for the Stanford University Committee on Graduate Studies. Patricia J. Gumport, Vice Provost Graduate Education
This signature page was generated electronically upon submission of this dissertation in electronic format. An original signed hard copy of the signature page is on file in University Archives.
iii ABSTRACT
While the 140 million tons of plastics produced each year may contribute to quality of life, they also come at a significant cost. Their production requires large quantities of nonrenewable resources, contributing to climate change; they accumulate in landfills and natural environments; and they often contain harmful additives. One way to address the multiplicity of problems that arise from the widespread use of synthetic plastics–without compromising convenience and disposability—would be to replace them with functionally equivalent materials that are biobased, biodegradable, and biocompatible, such as polyhydroxyalkanoates—a class of bioplastics. Bacteria known as “methanotrophs” consume methane as feedstock, and some produce the PHA polymer poly-ß-hydroxybutyrate (PHB). PHB production from methane could take advantage of the abundant biogas methane that is currently flared or allowed to escape to the atmosphere by the waste sector (landfills and wastewater treatment plants) to produce a valuable product that biodegrades to methane at end-of-life, creating a closed-loop cycle. This research evaluates methanotrophic growth and PHB production across scale. (1) Life Cycle Assessment (LCA) is used to anticipate the environmental impacts of PHB production from waste biogas by extrapolation from laboratory scale studies. LCA is used as an early-stage design tool to identify opportunities for pollution prevention, reduce resource consumption, guide environmental performance improvements, and identify research needs. The LCA also enables comparison with published LCAs for PHB produced from other feedstocks. (2) Stoichiometry and kinetics are evaluated and modeled for PHB-producing methanotrophs to describe the effects of oxygen and nitrogen on growth and PHB production by two PHB-producing methanotrophs. Significant differences were observed, with major implications for the use of these species in biotechnology applications. Such analyses can better inform bioreactor design, scale-up models, and life cycle assessments (LCAs). (3) A pathway genome database is developed for Methylosinus trichosporium OB3b as a model
iv organism using pathway reconstruction to predict the metabolic composition. The database provides a platform for the visualization of experimental data from omics experiments, facilitates comparative studies of pathways across species, and provides a resource for biotechnology applications of methanotrophs, such as through flux balance analysis.
v ACKNOWLEDGMENTS
I would like to thank my adviser, Craig Criddle, for his unending enthusiasm. I would also like to thank my committee members: Peter Karp for his patience and approachability, Perry McCarty for his thoughtful suggestions, and Michael Lepech for his attentiveness. I must also acknowledge my funding agencies; the research presented here was supported by the National Science Foundation Graduate Research Fellowship (NSF GRFP), the Stanford Graduate Fellowship (SGF), and by the California Environmental Protection Agency (CalEPA), Cal Recycle and the Department of Toxic Substances Control.
I would like to express my gratitude to the other students in the Criddle group, but especially Andy Pfluger, Allison Pieja, Kurt Rhoads, and Eric Sundstrom who have been research collaborators, mentors, and friends.
I also want to thank my family and friends for their support throughout my Ph.D. Most importantly, I’d like to thank my boyfriend, John, for providing love, encouragement, stability, and the occasional weather forecast.
vi TABLE OF CONTENTS
Abstract ...... iv Acknowledgments...... vi List of Tables ...... ix List of Figures ...... x Introduction ...... 1 Problems caused by Petroleum-based Plastics ...... 1 Polyhydroxyalkanoates (PHAs): A Biodegradable and Biocompatible Plastic Alternative...... 2 Type I versus Type II Methanotrophs ...... 4 Industrial Ecology Principles in Bioplastic (PHB) Production...... 4 The Methane Opportunity: Waste Valorization and Industrial Symbiosis ...... 5 Ecobiotechnology: Natural Selection for Plastic Production...... 6 Research Objectives ...... 8 Chapter 1: Cradle-to-Gate Life Cycle Assessment for a Cradle-to-Cradle Cycle: Biogas-to-Bioplastic (and Back) ...... 10 Abstract ...... 10 Introduction ...... 10 Methodology ...... 16 Goal and Scope Definition ...... 17 Inventory Analysis ...... 17 Results ...... 23 Impact Assessment...... 23 Discussion ...... 25 Chapter 2: Stoichiometry and Kinetics of the PHB-producing Type II Methanotrophs Methylosinus trichosporium OB3b and Methylocystis parvus OBBP ...... 28 Abstract ...... 28 Introduction ...... 28
vii Requirements for Oxygen and Reducing Equivalents in Methanotrophic Proteobacteria ...... 30 Materials and Methods ...... 31 Cultures ...... 31 Culture Growth Conditions ...... 31 Growth Monitoring and PHB production ...... 33 PHB Measurement ...... 34 Modeling of Stoichiometry ...... 35 Modeling of Kinetics ...... 37 Results ...... 39 Discussion ...... 43 Chapter 3: MethanoCyc: A Database for Methylosinus trichosporium OB3b ...... 46 Abstract ...... 46 Introduction ...... 46 Pathway/Genome DatabasE Construction & Metabolism ...... 48 Visualization of Experimental Data ...... 51 Species Comparison ...... 52 Biotechnology Applications: Flux Balance Analysis ...... 54 Conclusion ...... 57 Conclusions ...... 59 Life Cycle Assessment ...... 59 Stoichiometry and Kinetics ...... 60 Pathway/Genome Database ...... 60 References ...... 61
viii LIST OF TABLES
Number Page
Table 1 Life Cycle Stages for Petrochemical Plastics and PHB...... 16 Table 2 Parameterization of PHB production from methane...... 19 Table 3 Life Cycle Inventory of LCA System...... 22 Table 4 Impact Assessment for 1 kg of PHB Production from Waste Biogas ...... 24 Table 5 Headspace composition for each experimental setup...... 33 Table 6 Stoichiometric equations used to describe methanotrophic growth and PHB production...... 36
Table 7 Substrate partitioning parameters (fe, fs), cellular yield (YX), and % PHB production (by cell dry weight) of each strain by oxygen partial pressure and nitrogen source...... 40 Table 8 Microbial kinetic parameters, maximum specific growth rate (µmax), and maximum specific rate of substrate utilization (qmax)...... 42 Table 9 Substrate partitioning parameters (fe, fs,) yield during PHB production after growth under optimal conditions: for strain OB3b, nitrate as N source and 0.3 atm oxygen; for strain OBBP, ammonium as N source and 0.3 atm oxygen...... 43 Table 10 Kinetic values reported for methanotrophic growth...... 44 Table 11 Parameterization of PHB production from methane. Target: Production of 1.00 g PHB...... 45 Table 12 MethanoCyc Database Summary Statistics...... 49 Table 13 Flux Balance Analysis Nutrients, Secretions, and Biomass Metabolites...... 55
ix LIST OF FIGURES
Number Page
Figure 1 Left: The chemical formula for PHAs. PHAs consist of 100-30,000 repeated
monomer units (n). For PHB, R is a methyl group (-CH3). Right: Transmission electron microscopy illustrating PHB granules in methanotrophic bacteria (Photo credit: Pieja & Sundstrom, 2009)...... 3 Figure 2 Using selective pressures to enrich for high PHB-producing methanotrophs...... 7 Figure 3 Cradle-to-cradle Feedstock Cycle for PHB and biogas methane...... 13 Figure 4 Process Flow Diagram (PFD) of LCA System...... 18 Figure 5 Methane-PHB Cycle...... 20 Figure 6 Relative Impact Assessment of several PHB recovery methods...... 25 Figure 7 Methane consumption, oxygen consumption, carbon dioxide production, and biomass accumulation (growth) of (left) Methylosinus trichosporium OB3b with nitrate at 0.3 atm O2 and (right) Methylocystis parvus OBBP with ammonium at 0.3 atm O2...... 39 Figure 8 Methane Assimilation by Methanotrophs...... 50 Figure 9 Serine Cycle: formaldehyde assimilation in Type II methanotrophs...... 51 Figure 10 Omics viewer image of ratio of metabolomics data from Methylosinus trichosporium OB3b growth and polyhydroxybutyrate production experiment...... 52 Figure 11 Methylosinus trichosporium OB3b cellular overview with reactions highlighted that are shared with Methylomonas methanica MC09...... 53 Figure 12 Visualization of Flux Balance Analysis of Methylosinus trichosporium OB3b (nitrate as nitrogen source)...... 57
x INTRODUCTION
PROBLEMS CAUSED BY PETROLEUM-BASED PLASTICS
Globally, more than 140 million tons of plastics are produced each year, including polyethylene, polystyrene, polyvinyl chloride, and polyurethanes (DiGregorio 2009). Many of these materials have contributed to an improved quality of life, with applications ranging from life-saving medical devices to improved packaging. Other products–such as disposable beverage containers—have provided convenience. But these benefits have come at a significant cost, namely, the widespread accumulation of plastic, with all the attendant problems. Conventional plastics are produced from petroleum. Plastics production accounts for nearly 10% of all the oil and gas that the United States produces and imports and the market is expected to grow at a rate of 15% per year (DiGregorio 2009). Plastic production facilities consume approximately 270 million tons of oil and gas annually worldwide to supply power and raw materials (Gerngross and Slater 2000), resulting in high emissions of greenhouse gases. Franklin Associates (2007) quantified the total global warming potential for several plastic resins, with values ranging from 1.477 kg
CO2 eq/kg resin for the production of low-density polyethylene (LDPE) to 3.149 kg CO2 eq/kg resin for the production of acrylonitrile-butadiene-styrene (ABS) (FranklinAssociates 2007). Superficially, plastic products appear to be well suited for on-the-go societies, but they do not decompose in the environment and accumulate in landfills (Volova 2004) and the marine environment (Keshavarz and Roy 2010; Law et al. 2010; Moret-Ferguson et al. 2010). In 2005, only 5.7% of the 4.4 million tons of petrochemical plastics discarded in the United States was recovered and recycled (DiGregorio 2009). Meanwhile, the estimated accumulation rate of plastics in 1996 was 25 million tons per year (Lee 1996b). Researchers have estimated that plastics occupy about 20% of landfill volume (Braunegg et al. 1998) and can persist for upwards of 2,000 years (DiGregorio 2009).
1 Additional consequences of plastics use and disposal include in-use leaching of potentially harmful additives and release of unwanted residues during incineration (Keshavarz and Roy 2010). Some of the fundamental chemicals used in plastics, such as bisphenol A (BPA) and phthalates, are known endocrine disrupters and can lead to developmental problems (Walsh 2010). Upon incineration, plastics release carbon dioxide as well as other chemicals of concern (e.g. dioxins, sulfur oxides, hydrogen chloride, cadmium, lead, zinc, and arsenic) (Harding et al. 2007).
POLYHYDROXYALKANOATES (PHAS): A BIODEGRADABLE AND BIOCOMPATIBLE PLASTIC ALTERNATIVE
One way to address the multiplicity of problems that arise from the widespread use of synthetic plastics–without compromising convenience and disposability—would be to replace them with functionally equivalent materials that are biodegradable and biocompatible. Among the alternatives are polylactic acid (PLA) and polyhydroxyalkanoates (PHAs)—a class of bioplastics that has received increasing attention since the 1980s (Byrom 1987). Many microorganisms produce PHAs. Industrial production is a two-step fermentation in which a period of balanced growth is followed by a period of unbalanced growth. In the balanced growth period, the carbon feedstock and nutrients required for cell replication are supplied; in a subsequent period of unbalanced growth, the carbon feedstock is provided but one or more of the nutrients needed for replication (e.g., N, P, S, Fe, Na, K, Mg, Mn) becomes growth limiting. Under these conditions, many bacteria produce intracellular storage granules made of one or more PHA polymers (Anderson and Dawes 1990; Lee 1996b). Over 100 PHA molecules have been identified (Steinbuchel and Valentin 1995). All contain long polyester chains, with 100-30,000 repeated monomer units (n), configured as shown in Figure 1. This structure confers thermoplastic properties needed for molding and extrusion, and makes them suitable replacement options for synthetic plastics (e.g., polypropylene and polyethylene) in many applications (Byrom 1987; Steinbuchel and Fuchtenbusch 1998). Details of the properties of specific PHAs are extensively described elsewhere in the literature (Anderson and 2 Dawes 1990; Lee 1996b; Braunegg et al. 1998; Reddy et al. 2003). PHB, for example, is highly crystalline with a melting temperature around 180°C and a glass transition temperature around 15°C. Its properties are similar to isotactic polypropylene (Doi 1990). Different microorganisms can produce different types of PHA, and the nature of the carbon feedstock affects the type of PHA produced (Steinbuchel and Valentin 1995; Steinbuchel and Fuchtenbusch 1998; Lee 1996a; Braunegg et al. 1998). Bacteria known as “methanotrophs” consume methane as feedstock, and some produce the PHA polymer poly-ß-hydroxybutyrate (PHB). For PHB, the R group in Figure 1 is a methyl group (-
CH3). This PHB is stored within the cell as cytoplasmic granules 0.3-1.0 µm in diameter (Doi 1990) (Figure 1) and is oxidized, as needed, to meet intracellular demands for reducing power (Pieja et al. 2011a).
Figure 1 Left: The chemical formula for PHAs. PHAs consist of 100-30,000 repeated monomer units (n). For PHB, R is a methyl group (-CH3). Right: Transmission electron microscopy illustrating PHB granules in methanotrophic bacteria (Photo credit: Pieja & Sundstrom, 2009).
PHAs degrade in a variety of environments, including soil, sludge, and seawater (Akmal et al. 2003). They are first enzymatically depolymerized to hydroxyalkanoate monomers, then further metabolized --- aerobically to carbon dioxide-- and, anaerobically, to carbon dioxide and methane (Doi 1990). Under optimal conditions of adequate water and nutrients, biodegradation is rapid (period of weeks). The methanogenic communities in anaerobic digester sludge rapidly degrade PHAs and PHA- containing biocomposites to biogas methane (Morse 2009).
3 TYPE I VERSUS TYPE II METHANOTROPHS
There are two types of methanotrophic bacteria, the type I (gamma proteobacteria) and the type II (alpha proteobacteria) (Hanson and Hanson 1996). Type I methanotrophs use the ribulose monophosphate pathway for carbon, while the type II methanotrophs use the serine cycle (Anthony 1982). In addition, the two types produce different methane monooxygenase (MMO) enzymes. There are two forms of MMO in methanotrophs (Hanson and Hanson 1996; Dumont and Murrell 2005). All known methanotrophs can form particulate or membrane-bound MMO (pMMO), which is an integral membrane metalloenzyme (Lieberman and Rosenzweig 2005). Type I methanotrophs use the pMMO, while type II methanotrophs produce a cytoplasmic enzyme, soluble MMO (sMMO) in addition to the pMMO (Hanson and Hanson 1996). Type II methanotrophs have been shown to be PHB-producing (Pieja et al. 2011b; Pieja et al. 2011a; Pieja et al. 2011c). The two types appear to have different survival strategies. Type I methanotrophs are faster growing but have a high rate of die-off under stress conditions, while Type II methanotrophs are slower growing but survive under stress. Type II methanotrophs out-compete type I methanotrophs under oxygen- and nitrogen-limiting conditions (Hanson and Hanson 1996; Graham et al. 1993).
INDUSTRIAL ECOLOGY PRINCIPLES IN BIOPLASTIC (PHB) PRODUCTION
Currently, PHAs have only limited market application. This is largely due to production cost (Braunegg et al. 1998; Lee 1996b). Application of the principles of industrial ecology (Graedel and Allenby 2003) , i.e., use of waste carbon as feedstock, ecobiotechnology, and networking of waste and production systems, may make it possible to decrease PHA production costs. As described by Ryan, “[a]cross industry and government, the predominant response to climate change is to talk of the need for more eco-design and rapid ‘innovation for sustainability’ or ‘eco-innovation,’ but there are few clear examples of how to move from the redesign of existing products and services to systems innovation” (Ryan 2008). As shown in Figure 3 of Chapter 1, synthesis of PHB from methane 4 provides a clear path whereby carbon from methane, a potent greenhouse gas, can be sequestered in a valuable product. Design for environment principles (Graedel et al. 1996) can be incorporated into each step of this cycle. As proposed by Frosch and Gallopoulos, “[w]aste from one industrial process can serve as the raw materials for another, thereby reducing the impact of industry on the environment” (Frosch and Gallopoulos 1989). Use of methane for PHB production is thus an example of waste valorization, i.e., “the treatment of waste for beneficial use as raw material or as an energy carrier, with emphasis on processes and practices that reduce emissions and related environmental impacts” (Nzihou and Lifset 2010), and results in environmental benefits from the greater goal of loop closing (Nzihou 2010). Through industrial symbiosis—including “physical exchanges of materials, energy, water, and by-products among diversified clusters of firms” (Chertow 2007)—a landfill or wastewater treatment plant (or both, since they are often located in close proximity) can be co-located with a PHB production facility, allowing use of the biogas methane as a high value carbon feedstock .
THE METHANE OPPORTUNITY: WASTE VALORIZATION AND INDUSTRIAL SYMBIOSIS
Use of biogas methane offers important advantages for PHB synthesis. The choice of substrate has a significant impact on cost (Byrom 1987; Keshavarz and Roy 2010). Ideally, the feedstock should be available in large quantities and its availability should not be dependent upon political influence on price or supply (Byrom 1992). Biogas methane is less sensitive to such influences than conventional sugar feedstocks derived from cultivated biomass, such as corn and sugar cane. Use of a feedstock that is already generated and collected at landfills and at anaerobic digesters avoids the costs associated with growing, harvesting, and transporting cultivated biomass; the need for feedstock hydrolysis; fluctuating prices due to the vagaries of supply and demand; and land use for cultivated biomass with all of the associated environmental and social impacts (fertilizer demand, adverse changes in soil chemistry; groundwater and surface water contamination, habitat loss, and negative effects upon food supplies (Runge and Senauer May - Jun., 2007; Tenenbaum 2008). One analysis of the unintended
5 consequences of cultivated biomass use for biofuel production states that the large quantity of corn used by the ethanol industry is “sending shock waves through the food system” (Runge and Senauer May - Jun., 2007). Use of corn as a biofuel feedstock has resulted in higher corn prices. In March of 2007, prices rose to over $4.38 a bushel, the highest level in ten years (Runge and Senauer May - Jun., 2007). PHA production from corn and sugar cane feedstock is subject to the same market forces. According to one report, the cost of cultivated PHA feedstock accounts for 40-50% of total production costs (Shen et al. 2009). One study shows that corn-based PHA production requires 22% more steam, 19 times more electricity, and 7 times more water than polystyrene production (Gerngross 1999). Another study concluded that PHA production in genetically modified plants would require four times the energy compared to production of the same quantities of petroleum-based plastics (Dove 2000). Currently, much biogas is either not collected or simply flared, contributing to greenhouse gas emissions. Its conversion into a higher value product creates economic incentives for its more efficient capture. While energy production is currently a competing market for biogas, the quality of the gas can be problematic: biogas often contains trace contaminants including hydrogen sulfide, halides, and siloxanes that may damage combustion engines and result in expensive repairs and service interruptions (Schweigkofler and Niessner 2001). This can result in significant clean-up costs for the energy production from landfill and digester biogas. Consequently, biogas is often simply flared so as to release carbon dioxide rather than methane (the more potent greenhouse gas) with no effort at energy recovery.
ECOBIOTECHNOLOGY: NATURAL SELECTION FOR PLASTIC PRODUCTION
Production costs may also be decreased through ecobiotechnology, also known as “selective pressure for product formation,” (Kleerebezem and van Loosdrecht 2007). Ecobiotechnology takes advantage of natural selection and competition to enrich methanotrophic communities capable of high levels of PHB production. By beginning with a complex inoculum containing a diverse microbial community, such as activated
6 sludge at wastewater treatment plants, and applying strong selective pressures, only those microorganisms that grow optimally under the applied conditions survive and grow. When methane is provided as the sole carbon source, methanotrophs are favored. Some methanotrophs produce PHB and some do not. Research is ongoing to identify factors that favor growth of the PHB-producing methanotrophs while selecting against methanotrophs that do not produce PHB (Pieja et al. 2011b). As shown in Figure 2, the goal is to create conditions that select for a stable a community of PHB-producing methanotrophs.
Figure 2 Using selective pressures to enrich for high PHB-producing methanotrophs.
In conventional PHB production facilities, pure cultures of PHB-producing strains (i.e., cultures containing a single microorganism) are grown in bioreactors, and the PHB is harvested after a period of unbalanced growth (Johnson et al. 2009b). Such processes have high substrate costs, high capital costs, and consume large quantities of energy 7 (Johnson et al. 2009b) and are thus less favorable for scale-up production. Ecobiotechnology makes use of microbial communities, as opposed to pure cultures, often genetically engineered. Ecobiotechnology takes advantage of use the diversity and adaptive capacity of microbial communities, avoids the need for repeated sterilization of growth media and equipment, facilitates use of mixed and variable substrates, and may enable long-term stable operation (Kleerebezem and van Loosdrecht 2007). The diversity of the bioreactor community may also confer resistance to specialized predators, such as phage. The use of selection pressures to manage microbial community composition and product formation is thus an attractive alternative to traditional pure culture technology (Chen 2009). To date, however, the use of methanotrophic communities has not yet enabled the high percentages of PHA obtained in other systems. Research that the author has contributed to but is not presented elsewhere in this dissertation describes selection strategies that may implement the concept of eco- biotechnology. Specifically, in order to identify conditions favorable selection, we began with a diverse activated sludge inoculum and used medium typically recommended for methanotroph enrichment. Conditions that selected for enrichments dominated by PHB- producing Type II methanotrophs were (1) use of nitrogen gas as the sole nitrogen source in the absence of copper, (2) use of a dilute mineral salts media in the absence of copper, and (3) use of media prepared at pH values of 4-5 (Pieja et al. 2011b). Hypothesis- formulation, laboratory evaluation, and analysis of condition 1 was this author’s work.
RESEARCH OBJECTIVES
The objective of this research is to evaluate Type II methanotrophic growth and polyhydroxybutyrate (PHB) production across scale. The specific goals of this research are to: 1. Conduct a predictive life cycle assessment (LCA) to anticipate the environmental impacts of PHB production from waste biogas by extrapolation from laboratory scale studies. In Chapter 1, LCA is used as an early-stage design tool to identify opportunities for pollution prevention, reduce resource consumption, guide
8 environmental performance improvements, and identify research needs. The LCA also enables comparison with published LCAs for PHB produced from other feedstocks. 2. Evaluate the stoichiometry and kinetics of Type II methanotrophic growth and PHB production. In Chapter 2, the effects of oxygen and nitrogen source on stoichiometry and kinetics of growth and PHB production in the Type II methanotrophs Methylosinus trichosporium OB3b and Methylocystis parvus OBBP are described. Significant differences were observed, with major implications for the use of these species in biotechnology applications. Such analyses can better inform bioreactor design, scale-up models, and life cycle assessments (LCAs). 3. Develop a pathway genome database for a model organism. In Chapter 3, a pathway genome database is developed (i) using pathway reconstruction to predict the metabolic composition of Methylosinus trichosporium OB3b as a representative organism for methanotrophs; (ii) to provide a platform for the visualization of experimental data from omics experiments; (iii) to facilitate comparative studies of pathways across species; and (iv) to provide a resource for biotechnology applications of methanotrophs, such as through flux balance analysis.
9 CHAPTER 1: CRADLE-TO-GATE LIFE CYCLE ASSESSMENT FOR A CRADLE- TO-CRADLE CYCLE: BIOGAS-TO-BIOPLASTIC (AND BACK)
This work is in review for publication in Environmental Science and Technology, 2012.
Katherine H. Rostkowski, Craig S. Criddle, and Michael D. Lepech
ABSTRACT
At present, most synthetic organic materials are produced from fossil carbon feedstock that is regenerated over time scales of millions of years. Bio-based alternatives can be rapidly renewed in cradle-to-cradle cycles (1-10 years). Such materials extend landfill life and decrease undesirable impacts due to material persistence. This work develops a LCA for synthesis of polyhydroxybutyrate (PHB) from methane with subsequent biodegradation of PHB back to biogas (40-70% methane, 30-60% carbon dioxide). The parameters for this cradle-to-cradle cycle for PHB production are developed and used as the basis for a cradle-to-gate LCA. PHB production from biogas methane is shown to be preferable to its production from cultivated feedstock due to the energy and land required for the feedstock cultivation and fermentation. For the PHB- methane cycle, the major challenges are PHB recovery and demands for energy. Some or all of the energy requirements can be satisfied using renewable energy, such as a portion of the collected biogas methane. Oxidation of 18-26% of the methane in a biogas stream can meet the energy demands for aeration and agitation, and recovery of PHB synthesized from the remaining 74-82%. Effective coupling of waste-to-energy technologies could thus conceivably enable PHB production without imported carbon and energy.
INTRODUCTION
Globally, more than 140 million tons of plastics are produced each year (DiGregorio 2009). The benefits of these materials have come at a significant cost. 10 Including both material and energy inputs for plastic production, conventional plastics account for nearly 10% of the oil and gas produced and imported in the United States (DiGregorio 2009). This market is expected to grow at a rate of 15% per year (DiGregorio 2009). Plastic production facilities consume approximately 270 million tons of oil and gas annually worldwide to supply power and raw materials (Gerngross and Slater 2000), resulting in high greenhouse gases emissions (IPCC 2007). Franklin Associates (2007) quantified the global warming potential for several plastic resins, with
values ranging from 1.477 kg CO2 eq/kg resin for the production of low-density
polyethylene (LDPE) to 3.149 kg CO2 eq/kg resin for the production of acrylonitrile- butadiene-styrene (ABS) (FranklinAssociates 2007). Most plastic products are recalcitrant and accumulate in landfills (Volova 2004) and the marine environment (Keshavarz and Roy 2010; Law et al. 2010; Moret-Ferguson et al. 2010). In 2010, only 8% of the 31 million tons of the plastic discarded in the United States was recovered and recycled (EPA 2012). Meanwhile, the estimated accumulation rate of plastics in 1996 was 25 million tons per year (Lee 1996b). Municipal Solid Waste (MSW) contains plastics ranging from 13.2-15.8% by wet mass (Staley and Barlaz 2009). Plastics occupy about 20% of landfill volume (Braunegg et al. 1998) and can persist for over 2,000 years (DiGregorio 2009). Other concerns include in-use leaching of potentially harmful additives (Keshavarz and Roy 2010), such as bisphenol A (BPA) and phthalates (Walsh 2010) and release of unwanted residues during incineration (e.g. dioxins, sulfur oxides, hydrogen chloride, cadmium, lead, zinc, and arsenic) (Harding et al. 2007; Keshavarz and Roy 2010). One way to address the challenges that arise from the widespread use of synthetic plastics–without compromising convenience and disposability—would be to replace them with functionally equivalent materials that are biodegradable and biocompatible, such as polyhydroxyalkanoates (PHAs). Such materials have received increasing attention since the 1980s (Byrom 1987). Under growth limiting conditions, many microorganisms produce PHAs as intracellular storage granules made of one or more PHA polymers (Anderson and Dawes 1990; Lee 1996b). Over 100 PHA molecules have been identified (Steinbuchel and Valentin 1995), containing long polyester chains, with 100-30,000
11 repeated monomer units (n). This structure confers thermoplastic properties needed for molding and extrusion, making them suitable replacement options for synthetic plastics in many applications (Byrom 1987; Steinbuchel and Fuchtenbusch 1998). Details of the properties of specific PHAs are extensively described elsewhere (Anderson and Dawes 1990; Lee 1996b; Braunegg et al. 1998; Reddy et al. 2003). Different microorganisms can produce different types of PHAs, and the nature of the carbon feedstock affects the type of PHA produced (Steinbuchel and Valentin 1995; Steinbuchel and Fuchtenbusch 1998; Lee 1996a; Braunegg et al. 1998). The PHA is stored within the cell as cytoplasmic granules 0.3-1.0 µm in diameter (Doi 1990) (Figure 1). When the stored PHA is needed to meet carbon or energy requirements, it is degraded to acetyl-CoA, a key metabolic intermediate (Uchino et al. 2007). In soil, sludge, and seawater, PHA resins degrade rapidly (Akmal et al. 2003), with aerobic mineralization to carbon dioxide and anaerobic biodegradation to biogas (Doi 1990). Both PHAs and biocomposites containing PHAs rapidly degrade to biogas in methanogenic bioreactors (Morse 2009). The market for PHAs has expanded from initial applications in packaging (Griffin 1994) to industrial and agricultural applications (Philip et al. 2007) to medical applications, where they are now marketed at low-volume and high-price as premium biocompatible bioplastics (Keshavarz and Roy 2010). Further expansion of the market is limited by production cost (Braunegg et al. 1998; Lee 1996b). Major factors affecting production cost are the use of cultivated feedstock, such as corn and sugar cane, with the associated water, chemicals, and energy required for growth, harvesting, transport, and feedstock processing, and the energy and chemical costs of PHA synthesis, extraction, and purification (Keshavarz and Roy 2010) (Runge and Senauer May - Jun., 2007; Tenenbaum 2008). According to one report, the cost of cultivated PHA feedstock accounts for 40-50% of total production costs (Shen et al. 2009). The volatile and increasing price of crude oil (Greene et al. 2006; Bentley 2010) and increasing awareness of the adverse environmental impacts of petrochemical-based plastics have revitalized research in PHA production (Keshavarz and Roy 2010) and the potential use of organic waste as a feedstock opens the door to new methods of production that do not rely upon cultivated feedstock and imported energy.
12 As shown in the feedstock life cycle of Figure 3, synthesis of PHB from methane provides a clear path whereby carbon from methane, a potent greenhouse gas, can be sequestered in a valuable product using design for environment principles (Graedel et al. 1996).
Step 1 Methanotrophs consume methane and store it as PHB PHB Methane granules (Carbon in Source) bacteria
Step 4 Step 2 Plastic products are PHB is extracted collected and landfilled from bacteria methane PHB
Plastic PHB Products Powder
Step 3 PHB is manufactured into plastic products
Figure 3 Cradle-to-cradle Feedstock Cycle for PHB and biogas methane.
The cradle-to-cradle feedstock cycle for PHB and biogas methane (Figure 3) takes advantage of the abundant biogas (typically consisting of 40-70% methane and 30-60% carbon dioxide) that is often flared or allowed to escape to the atmosphere by the waste sector, the third largest contributor to global emissions of non-carbon dioxide greenhouse
gases (accounting for 15% of these emissions) (USEPA 2006). Methane (CH4) is a greenhouse gas with a global warming potential (GWP) of 25 over a 100-year period (Solomon et al. 2007). The two largest sources within the waste sector—solid waste landfills and wastewater treatment plants (USEPA 2006). The US EPA estimated global methane emissions for 2005 at ~36 Tg from landfills and ~26 Tg from wastewater 13 treatment facilities (USEPA 2006), and these emissions are expected to double by 2030 (Matthews and Themelis 2007). Demand for biogas as a feedstock could provide market incentives for more efficient capture of such emissions. While energy production is currently a market for biogas, biogas often contains trace contaminants including hydrogen sulfide, halides, and siloxanes that can damage combustion engines and result in expensive repairs and service interruptions (Schweigkofler and Niessner 2001). Current biogas production facilities could theoretically sustain PHB production levels of approximately 27-41 Tg/yr, assuming theoretical yields of 0.45-0.67 g PHB/g methane (Yamane 1993) and a 0.5% fugitive loss (Jimenez-Gonzalez et al. 2000). These values suggest that PHB from existing landfills and anaerobic digesters theoretically replace 20- 30% of the total plastics annual market (DiGregorio 2009). In Step 1 of Figure 3, methanotrophic bacteria are used to produce PHB. Methanotrophs are the major terrestrial sink for methane, obtaining both energy and carbon from it (Mancinelli 1995; Hanson and Hanson 1996), and are a subset of the methylotrophs, bacteria that metabolize one-carbon compounds (Hanson and Hanson 1996; Murrell 2010; Lidstrom 2006). Large and active methanotrophic populations naturally assemble when methane and oxygen are simultaneously present (Dworkin and Falkow 2006; Murrell 2010). As early as 1970, researchers discovered that some methanotrophs could make PHB under nutrient-limiting conditions (Whittenb.R et al. 1970). The subset that can produce PHB are known as the “Type II methanotroph” (Wendlandt et al. 2001; Helm et al. 2008; Choi and Lee 1999; Wendlandt et al. 2005; Pieja et al. 2011a; Pieja et al. 2011b). When diverse Type II methanotrophs were screened for PHB production, levels of PHB produced ranged from 9 to 44% by dry mass (Pieja et al. 2011b). Others have reported levels of 51% (Wendlandt et al. 1998, 2001) and 52% (Wendlandt et al. 2010), under optimized conditions. After PHB is extracted from cells and purified (Step 2), the resulting bioplastic resin can be used to make a wide range of products (Step 3). At end of life, these products are ideally either recycled directly or returned to a controlled anaerobic environment, such as a landfill with efficient biogas capture or an anaerobic digestion facility (Step 4). In such environments, waste products containing PHB are broken down
14 and the resulting biogas becomes feedstock for PHB production (Step 1). The PHB can be of high molecular weight (>1 million Da), enhancing its value compared to PHB from other sources (Wendlandt et al. 1998). Use of biogas methane for PHB production is thus an example of waste valorization (Nzihou and Lifset 2010), and results in environmental benefits from the greater goal of “loop closing” (Nzihou 2010; Lifset 2002). Opportunities for industrial symbiosis become apparent (Chertow 2007). For example, a landfill or wastewater treatment plant (or both, since they are often located in close proximity) could be co-located with a PHB production facility, allowing use of a continuous and stable supply of biogas methane, or treated wastewater effluent for cooling. Several studies have evaluated cradle-to-gate processes for the production of PHB(Gerngross 1999; Akiyama et al. 2003; Nonato et al. 2001; Harding et al. 2007; Kim and Dale 2005, 2008), but most only consider energy requirements and global warming impacts rather than undertaking a more thorough environmental impact assessment. In addition, the results show high variability (Kim and Dale 2008; Gerngross 1999; Gerngross and Slater 2000; Akiyama et al. 2003; Harding et al. 2007; Kurdikar et al. 2000; Tabone et al. 2010). Literature evaluating the use of corn as a feedstock ranges in energy requirements from 2.5 (Kim and Dale 2008) to 81.0 MJ/kg PHB (Gerngross and Slater 2000). Often, data is not available or is provided by industry and may be inaccurate or biased. There have been no studies evaluating PHB beyond the resin phase and no studies considering the use of waste methane as a feedstock for PHB production. Because no full life cycle assessment exists, the full environmental benefits of PHBs are unknown (Harding et al. 2007). There are several ways in which the PHB production method from waste biogas differs from the production of commercially available plastics and bioplastics. Table 1 summarizes these stages -- Raw Material Acquisition, Material Processing, Retirement & Recovery, and Treatment & Disposal-- for polypropylene produced from fossil fuels, PHA produced from cultivated feedstock, and PHB produced from biogas methane.
15 Table 1 Life Cycle Stages for Petrochemical Plastics and PHB.
Life Cycle Life Cycle Stage Description Stage Petrochemical Plastics PHAs from Cultivated PHB from (e.g. polypropylene) Feedstock Waste Methane Raw Fossil fuel feedstock Feedstock cultivation Renewable Material (oil and natural gas) (e.g. corn, sugar cane) feedstock Acquisition extraction and fermentation (e.g. acquired from corn to glucose) waste (e.g. landfill waste methane) Material Cracking Microbial synthesis of PHB granules Processing Polymerization Recovery and purification of granules
Manufacture Extrusion, Injection molding, or Stretch blow molding & Assembly Use & Filling (if applicable), Retail, and Use Service Retirement & Reuse, Collection for Reuse, Collection for Reuse, Recovery Remanufacture, Landfilling/Incineration Collection for Closed-loop Recycling, reconversion Open-loop recycling, or into bioplastic Landfilling/Incineration PHB Treatment & Landfilling/Incineration Biodegradation in a Landfill, Anaerobic Disposal (Persistent) Bioreactor, or Composting Facility
METHODOLOGY
This life cycle assessment (LCA) evaluates the production of PHB by methanotrophs from waste biogas. The analysis consists of four components: goal and scope definition, inventory analysis, impact assessment, and interpretation (Jimenez- Gonzalez et al. 2000). We have used the findings to identify research areas that will be critical for industrial scale production of PHB generally and, more specifically, for the use of waste biogas as a feedstock.
16 GOAL AND SCOPE DEFINITION
The goal in this study was to anticipate the environmental impacts of PHB production from waste biogas by extrapolation from laboratory scale studies (Pfluger et al. 2011; Pieja et al. 2011b; Pieja et al. 2011a; Wendlandt et al. 2005; Wendlandt et al. 1998, 2001; Wendlandt et al. 2010). LCA is used as an early-stage design tool(Rebitzer et al. 2004) to identify opportunities for pollution prevention, reduce resource consumption (Rebitzer et al. 2004), guide environmental performance improvements (Vink et al. 2003), and identify research needs. The LCA also enables comparison with published LCAs for PHB produced from other feedstocks. This study considers 9 environmental impact categories using the Tool for the Reduction and Assessment of Chemical and other environmental Impacts (TRACI) 2.0 V 3.01 impact assessment method (Bare 2011) developed by the U.S. Environmental Protection Agency. This tool was designed specifically for US using input parameters consistent with US locations. It is a midpoint oriented LCIA method including the following impact categories: Global Warming, Acidification, Carcinogenics, Noncarcinogenics, Respiratory effects, Eutrophication, Ozone Depletion, Ecotoxicity, and Smog. Using SimaPro software, the study considers Cradle-to-resin production of PHB from waste biogas. Cradle-to-resin production is used as a boundary to facilitate comparison of this study with others that have evaluated plastic production. In addition, the Manufacture & Assembly stage and the Use & Service stage are omitted because PHAs can be processed with equipment already in use for traditional plastics(Steinbuchel and Fuchtenbusch 1998) and are functionally equivalent to existing petrochemical plastics during use. Results are presented on a per mass basis (functional unit: 1 kg of PHB produced) for consistent comparison with other datasets.
INVENTORY ANALYSIS
The Process Flow Diagram (PFD) in Figure 4 below is a schematic depicting the boundary of the study, highlighting the processes and flow of materials for PHB production from methane (also shown in figures 3 and 5).
17 Process Inputs Process Inputs Common processes for petrochemical plastics and polyhydroxybutyrate
Cells PHB Containing Powder/ Plastic Product Platic Product Anaerobic PHB Resin Manufacture Use Collection for Biodegradation PHB Production PHB Recovery (e.g. (e.g. Filling, Waste Treatment (e.g. Landfill, Injection,Stretch Retail, & Biodigester) Excess Blow Molding) Consumer Use) Methane Cell Process Outputs Process Outputs Material
Process Inputs Sewage Sludge Excess Cell Digestion and Material Use Biogas Recovery
LCA System Boundary Process Outputs
Figure 4 Process Flow Diagram (PFD) of LCA System.
Figure 5 illustrates the methane-PHB cycle for the production of 1.0 g of PHB. Table 2 defines the parameterization of PHB production from methane, the values that were used, and the associated references. In order to produce 1.0 g PHB, approximately 5.2 g of methane (this value is conservative and would be <3 for optimized systems), 20.9 g of oxygen, and 0.12 g of nitrogen are required for cell growth and PHB production. Because 1.0 g of PHB biodegrades to 0.4 g of methane, the remainder of the methane requirement is met by anaerobic digestion of waste cell residue left after PHB extraction and recovery (0.4 g methane) and from biodegradation of additional waste organic matter in a landfill or anaerobic bioreactor (4.5 g methane from about 16.9 g of organic waste).
18
Table 2 Parameterization of PHB production from methane.
Parameter Value Reference % PHB achieved 50 % (0.5g PHB/g total Measured value mass) (Listewnik et al. 2007) Yield of PHB on 0.55 g PHB/g methane Lumped average of Methane measured and theoretical values (Asenjo and Suk 1986; Wendlandt et al. 2001; Yamane 1993, 1992)] Growth Yield 0.345 g biomass/g Average value (Leak and methane Dalton 1986) Oxygen Requirement 4 g oxygen/g methane Thermodynamic estimate Nitrogen Requirement 0.12 g nitrogen/g biomass Typical 12% N in microbial biomass. PHB Recovery by 90 % Reported value: (Byrom Extraction 1987; Jacquel et al. 2008) Methane Recovery from Biomass waste 0.32 g methane/g biomass Calculated from the Carbohydrate organic 0.27 g methane/g organic empirical formula for waste waste each waste type PHB 0.38 g methane/g PHB (Rittmann and McCarty 2001)
19 Methane Methane Production Consumption (Landfill, (Methanotrophic digester, etc) PHB Bioreactor)
Disposal
0.4 g methane 20.9 g oxygen 0.1 g nitrogen 1.1 g biomass 3.2 g methane
16.9 g Organic Waste 4.5 g methane 5.2 g methane 1.1 g PHB 2.0 g methane
PHB Extraction & Recovery 0.4 g methane Use and Disposal 1.0 g PHB
Figure 5 Methane-PHB Cycle.
Table 3 summarizes the inventories for all of the modeled processes in the LCA system (inputs, outputs, flows to next process) shown in Figure 4. The modeled PHB production strategy assumes use of Type II methanotrophic communities that self- assemble under appropriate selection conditions, such as those identified by Pieja et al. (Pieja et al. 2011b) and Pfluger et al. (Pfluger et al. 2011). Inputs and outputs to the life cycle inventory of PHB production by methanotrophs are based on experimental data and system parameterization using literature values and stoichiometric calculations (see Table 2 and Figure 5). The system is credited with use of methane as an input but includes environmental burdens associated with accessing waste biogas (i.e. modeled as “Biogas from Sewage Sludge, at storage” from Ecoinvent database, accounting for raw sludge digestion and gas storage). Supply chains of biochemical products can be quite complex with limited information about the chemical inputs due to legal and intellectual property concerns(Jimenez-Gonzalez et al. 2000); when data was unavailable, best estimates were used. During PHB production, methanotrophs also use process inputs for the production of cellular biomass (i.e. cell growth). We refer to this as “excess cell material,” a potentially valuable high-protein byproduct. While allocation in joint production is still unresolved for LCA methodologies (Baumann and Tillman 2004), in this study we
20 allocate all environmental burdens associated with cell growth and PHB production to PHB production. Several PHB recovery methods were modeled (solvent-based extraction (Jacquel et al. 2008), selective dissolution(Yu and Chen 2006), surfactant digestion (Jacquel et al. 2008), based on literature data and best-available scale-up information (Patel et al. 2006). The excess cell material is treated as a beneficial byproduct and the model considers two uses for it: (i) combustion, or (ii) anaerobic digestion. Environmental burdens associated with these processes involved in these uses are allocated to the use of excess cell material. To obtain realistic full-scale power requirements (Jimenez-Gonzalez et al. 2000), energy requirements for industrial scale processes (i.e. Agitation and Aeration during PHB Production; Centrifugation, Heating, Drying, Filtration, and Pumping during PHB Recovery) were calculated using the BREW generic approach (Patel et al. 2006). Energy was initially modeled as production mix of the Western Electricity Coordinating Council (WECC) according to eGrid 2005. Because energy requirement had a large contribution to environmental impacts, energy was also modeled assuming that a portion of the biogas methane could be allocated to meet energy demands, accounting for digestion and biogas capture, impacts associated with purification to methane, combustion in a gas turbine, and release of carbon dioxide. For all processes involving liquids, a 1% fugitive loss was assumed. For gases, a 0.5% fugitive loss was assumed (Jimenez-Gonzalez et al. 2000).
21
Table 3 Life Cycle Inventory of LCA System.
Inputs Flow to Next Process Outputs PHB Production Methane 5.26 kg Cell Culture 222.22 L Methane 0.03 kg Containing Losses PHB Oxygen 21.04 kg Cell material 1.11 kg Oxygen 0.11 kg Losses Water 224.44 L PHB 1.11 kg Water Losses 2.24 kg Chemical 0.43 kg Energy 57.60 MJ PHB Recovery: All Methods Cell Culture 222.22 L PHB 1.00 kg Wastewater 219.44 L Containing PHB Excess Cell 1.22 kg Material Solvent Extraction New Solvent 4.89 L Waste 4.89 L Solvent Recovered 40.00 L Waste 444.44 L Solvent Ethanol Ethanol 448.89 L Solvent 0.45 L Losses Energy 6.13 MJ Ethanol 4.49 L Losses PHB Recovery: Surfactant Digestion Cell Culture 1.00 L Containing PHB Surfactant 0.62 kg Waste 0.61 kg Surfactant Hypochlorite 25.92 kg Waste 25.66 kg Hypochlorite Energy 1.45 MJ Surfactant 0.01 kg Losses Hypochlorite 0.26 kg Losses PHB Recovery: Selective Dissolution Cell Culture 0.00 L Containing PHB Acid 0.25 kg Waste Acid 0.25 kg Base 0.21 kg Waste Base 0.21 kg Hypochlorite 2.96 kg Waste 2.96 kg
22 Hypochlorite Water 30.46 kg Hypochlorite 0.00 kg Losses Acid Losses 0.00 kg Energy 1.47 MJ Base Losses 0.00 kg Hypochlorite 0.03 kg Losses Water Losses 0.30 L Excess Cell Material Use: All Options Excess Cell 1.22 Material Excess Cell Material Use: Combustion Energy 20.21 MJ Excess Cell Material Use: Anaerobic Degradation Methane 0.39 kg Carbon 0.64 kg dioxide
RESULTS
IMPACT ASSESSMENT
Table 4 below summarizes impacts resulting from the production of 1 kg of PHB,
as determined by modeling with TRACI 2.0. The most common commercially used
method for PHB recovery is solvent extraction (Jacquel et al. 2008). For steps 1-3 of
Figure 3, over 90% of the contribution of each of the impact categories is due to PHB
recovery when solvent extraction is used. For this reason, the production of 1 kg of PHB
from cradle-to-intracellular resin before recovery is specifically evaluated and also listed
in Table 4. These values are normalized, using the U.S. Environmental Protection
Agency (EPA) normalization database values (Bare et al. 2006). The normalized values represent relative contributions for the production of 1kg PHB to annual U.S. emissions by impact category. First, we note that most of the normalized values are low or negative, implying a low or net positive impact, respectively. Thus, the overall
23 production method is favorable. While most of the negative impacts have low positive values, they are all primarily attributed to energy use. In the case of global warming, for instance, energy use is the only process that detrimentally contributes at least 5% to the global warming potential impact. A substantial portion of the process energy requirement (57.60 MJ) for PHB production can be offset by energy generation from combustion of the excess cell material, producing 20.21 MJ, resulting in a net 37.39 MJ energy requirement.
Table 4 Impact Assessment for 1 kg of PHB Production from Waste Biogas
Cradle-to- Cradle-to- Normalization Cradle- intracellular- intracellular- Value Impact Indicator Unit to-resin resin resin (Bare et al. Value Normalized Value 2006) Value kg CO 9.42 x 12 -13 Global Warming 2 -1.94 6.85 x 10 -2.83 x 10 eq 102 H+ 9.25 x Acidification moles 2.62 2.08 x 1012 1.26 x 10-12 101 eq kg Carcinogenics benzene 1.02 1.02 x 10-2 7.21 x 107 1.42 x 10-10 eq kg 8.84 x Noncarcinogenics toluene 3.15 x 101 4.11 x 1011 7.66 x 10-11 102 eq kg - Respiratory 3.95 x 10 -2 10 -13 PM2.5 1.42 x 10 2.13 x 10 6.65 x 10 Effects 1 eq Eutrophication kg N eq 1.06 1.11 x 10-3 5.02 x 109 2.22 x 10-13 - kg CFC- 5.08 x 10 -7 7 -15 Ozone Depletion 4 4.32 x 10 8.69 x 10 4.97 x 10 11 eq kg 2,4-D 4.20 x 10 -10 Ecotoxicity 1 4.08 2.06 x 10 1.98 x 10 eq 10 kg NO -2 10 -13 Smog x 3.28 1.83 x 10 3.38 x 10 5.41 x 10 eq
24 Figure 6 illustrates the relative impacts of the different recovery methods on the PHB cradle-to-resin LCA. Solvent extraction was the least favorable option, and selective dissolution was the most favorable. While recovery methods other than solvents result in less harm, all of the recovery methods nonetheless incur the most negative impacts. Improved methods of PHB recovery could address these issues and would likely lead to lower costs and improved profitability, as recovery level and purity increase (Jacquel et al. 2008).
Figure 6 Relative Impact Assessment of several PHB recovery methods.
DISCUSSION
The impact assessment suggests two primary research priorities: (1) a more environmentally benign PHB recovery method that is less energy intensive and does not use harmful solvents, and (2) reduction of energy requirements and potential use of biogas-to-energy technology to off-set demands for imported energy. PHB recovery from cell material was a primary contributor to all environmental impact categories in PHB production, especially when solvent extraction was used. Others have also noted that recovery of PHB will be a major barrier in production, specifically noting that solvent-based extraction should be rejected for two reasons: (i) 25 high cost at large scale due to solvent demand and solvent recovery (Byrom 1992) with estimates showing that over half of the production cost of PHB is associated with the recovery and purification process (Ling et al. 1997); and (ii) environmental concerns due to large quantities of chlorinated solvents (Byrom 1992) that are unsuitable for mass production of bioplastic (Byrom 1987). Other LCA studies of PHB have included recovery with little process detail resulting in less significant impacts, suggesting that such studies either have access to better industrial scale information, have underestimated recovery impacts, or this study overestimates impacts (or some combination of these). The other two PHB recovery methods modeled, surfactant digestion and selective dissolution, showed some improvement over solvent extraction but were still major contributors to adverse impacts. Literature has suggested some promising methods that may further reduce environmental impacts and operating costs for PHB recovery such as genetic alterations that allow “spontaneous liberation of PHB” (Jung et al. 2005) and a “lysis system that allows the PHB granules to be released gently and efficiently” (Fidler and Dennis 1992). In any case, recovery of PHB will be a barrier to more widespread commercialization regardless of production strategy. There are several ways to improve the energy requirement. The energy impact per mass of PHB produced can be reduced by increasing cell growth rates, by increasing the PHB produced per unit of energy consumed, or by decreasing energy inputs per unit of PHB produced. Cell growth can be increased through the optimization of growth medium composition, adjusting gas ratios, or novel techniques that increase the mass transfer rates of methane and oxygen enabling enhanced cell growth and higher final cell density (Han et al. 2009). PHB production may be increased through the cell growth optimization, use of different limiting nutrients, genetic engineering, and/or long-term selection processes (Pieja et al. 2011c). Lastly, some or all of these energy requirements can be satisfied using renewable energy, such as a portion of the collected biogas methane. If selective dissolution is used to recover PHB, the total energy requirement for PHB Production and Recovery of would be 38.86-59.07 MJ/kg PHB, where the range depends upon whether the energy can be offset by combustion of excess cell material. As is the case for traditional petrochemical plastics, a single input, in this case, biogas
26 methane, can be used as both the feedstock (7.88 m3) and as the source of process energy (1.78-2.71 m3) source. Oxidation of 18-26% of the methane can provide sufficient energy to meet the energy demands for aeration and agitation, separation, and purification of the polymer. PHB is unique in that it rapidly biodegrades under anaerobic conditions, a factor that is not captured in the boundary of the LCA system studied here. At end-of-life, PHB-based products can thus be digested to recover biogas feedstock. This biogas can be reused as a feedstock for further PHB production, completing the life cycle and eliminating the typical problems associated with end-of-life of plastics such as accumulation. Since all approaches to PHB production will require PHB recovery and are likely to be paired with the best available method, Total energy requirement and Global warming potential for PHB production alone can be compared. The total energy requirement for PHB production form waste biogas is 37.4 MJ/kg PHB, while production from corn requires 41.9 MJ/kg PHB (Akiyama et al. 2003). Because of this energy savings and the high global warming potency of methane, the benefit for global warming potential is more substantial than for a feedstock that sequesters carbon dioxide during growth. In fact, using biogas methane for PHB production, results in a global warming potential of -1.94 kg CO2 eq and can be as low as -6.06 if excess cell material is combusted and biogas is used to satisfy energy requirements while PHB from corn
feedstock has a global warming potential of only -0.1 kg CO2 eq (Akiyama et al. 2003).
27 CHAPTER 2: STOICHIOMETRY AND KINETICS OF THE PHB-PRODUCING TYPE II METHANOTROPHS METHYLOSINUS TRICHOSPORIUM OB3B AND METHYLOCYSTIS PARVUS OBBP
This work in preparation for submission for publication.
Katherine H. Rostkowski, Andrew R. Pfluger, and Craig S. Criddle
ABSTRACT
In addition to being the major terrestrial sink for methane, a major greenhouse gas, methanotrophs are of biotechnological interest for a variety of purposes (e.g. single- cell protein production, polyhydroxybutyrate (PHB) production, bioremediation). Optimizing growth of Type II methanotrophs and their capacity for PHB production specifically is of commercial and environmental interest. In this study, we describe how oxygen and nitrogen source affect the stoichiometry and kinetics of growth and PHB production in the Type II methanotrophs Methylosinus trichosporium OB3b and Methylocystis parvus OBBP. Significant differences were observed, with major implications for the use of these species in biotechnology applications. Such analyses can better inform bioreactor design, scale-up models, and life cycle assessments (LCAs).
INTRODUCTION
Methanotrophs, a subset of the methylotrophs, obtain both their carbon and energy from methane (Mancinelli 1995; Hanson and Hanson 1996) and, as such, are the major terrestrial sink for methane (Hanson and Hanson 1996; Murrell 2010; Lidstrom 2006). When methane and oxygen are simultaneously present, large and active methanotrophic communities can self-assemble (Dworkin and Falkow 2006; Murrell 2010). In industrial applications, methanotrophy is of interest (Trotsenko et al. 2005) for single-cell protein production (Smith et al. 2010), production of polyhydroxybutyrate (PHB) (Wendlandt et al. 2001; Helm et al. 2008; Choi and Lee 1999; Wendlandt et al. 28 2005; Whittenbury et al. 1970a; Pieja et al. 2011b; Pieja et al. 2011a), specialty chemicals, and bioremediation through co-metabolism (Smith et al. 2010; Anderson and McCarty 1997; Dalton and Stirling 1982). Under nutrient-limiting conditions, different methanotrophs have evolved different strategies for growth and survival. Type II methanotrophs can synthesize intracellular PHB granules when methane and oxygen are available, but another growth factor, such as nitrogen, is absent (Whittenbury et al. 1970a; Pieja et al. 2011b). These granules can subsequently serve as a source of reducing equivalents, facilitating more rapid growth when reducing power is needed (Pieja et al., 2011a). In screening assays, Type II methanotrophs produced 9-44% PHB (by dry weight) (Pieja et al. 2011b). Under optimized conditions, PHB production can exceed 50% of dry weight (Wendlandt et al. 1998, 2001). PHB is of industrial interest as a biodegradable substitute for fossil carbon-based plastics with physical properties similar to polypropylene (Anderson and Dawes 1990). To date, however, the costs of polhydroxylalkanoate (PHA) production, including PHB, have limited large-scale production (Braunegg et al. 1998; Lee 1996b). One strategy for cost reduction is to use biogas—a mixture of methane and carbon dioxide—generated at wastewater treatment plants and landfills as feedstock for Type II methanotrophs. A Life cycle assessment of such a process indicates a more favorable energy outcome (37 MJ/kg PHB from biogas compared to 42 MJ/kg PHB from corn-derived sugar) and potential for enhanced carbon sequestration (~2 kg CO2 equivalents fixed/ kg methane compared to
0.1 kg CO2 equivalents fixed/kg of corn-derived sugar (Chapter 1). Optimizing growth of Type II methanotrophs and their capacity for PHB production for such an application is thus of commercial and environmental interest. In this study, we describe the effect of oxygen and nitrogen source on the stoichiometry and kinetics of growth and PHB production in the Type II methanotrophs Methylosinus trichosporium OB3b and Methylocystis parvus OBBP. Such analyses can better inform scale-up models and provide more accurate results in predictive life cycle assessments (LCAs).
29 REQUIREMENTS FOR OXYGEN AND REDUCING EQUIVALENTS IN METHANOTROPHIC
PROTEOBACTERIA
The availability of oxygen and reducing equivalents is a critical factor in methane metabolism, beginning with the monooxygenase-mediated oxidation of methane to methanol, a step that requires both oxygen and NAD(P)H. Subsequent oxidation of methanol to formaldehyde, formaldehyde to formate, and formate to carbon dioxide,
release reducing equivalents to an electron transport chain that terminates with O2 reduction to water and enables ATP synthesis. Formaldehyde also serves as the key C1 building block for anabolic reactions that require reducing equivalents and ATP (Costa et al. 2001; Trotsenko and Murrell 2008; Tonge et al. 1975; Dalton and Stirling 1982). Nitrogen metabolism also affects the demand for oxygen and reducing equivalents. Obligate methanotrophs can assimilate nitrate and ammonia, and some can fix dinitrogen (Murrell and Dalton 1983a). Type I methanotrophs assimilate ammonia by reductive amination of α-ketoglutarate or pyruvate and via glutamine synthetase (Shishkina and Trotsenko 1979). Type II methanotrophs use both glutamine synthetase and glutamate synthase in the glutamate cycle (Shishkina and Trotsenko 1979; Trotsenko and Murrell 2008; Murrell and Dalton 1983a; Trotsenko 1983). Ammonium is fortuitously oxidized to toxic hydroxylamine by methane monooxygenase (Dalton and Stirling 1982). The presence of gene clusters that encode hydroxylamine reduction to ammonium in Methylosinus trichosporium OB3b (Stein et al. 2010) suggests a hydroxylamine-ammonium “futile” cycle. Such a cycle would facilitate ammonium assimilation, but at the expense of increased demand for oxygen and reducing equivalents.
The ability to fix N2 was once believed to be limited to Type II methanotrophs (Oakley and Murrell 1988; Auman et al. 2001; Murrell and Dalton 1983b), and, in fact,
delivery of N2 as sole nitrogen source can select for Type II methanotrophs (Pieja et al. 2011b), but nifH genes coding for nitrogenase are also found in Type I species (Auman et al. 2001). In one study, pairing of N2-fixation with low O2 concentrations allowed Type II methanotrophs to become dominant within the biofilms of a fluidized bed reactor previously dominated by Type I methanotrophs that had dominated at higher levels of 30 dissolved oxygen with nitrate as the N-source (Pfluger et al. 2011). Nitrogenase activity is sensitive to oxygen, but sensitivity varies among species (Dedysh et al. 2004). Most
N2-fixing species grow optimally at low oxygen partial pressure (4-10%) and produce sMMO (Chu and Alvarez-Cohen 2000). The common PHB-producing genuses, Methylocystis, Methylosinus, and Methylocella only fix nitrogen at low oxygen concentrations (0.05-0.15 bar) (Vorob'ev and Dedysh 2008). Quantifying oxygen sensitivity is thus important for optimization of growth. In one study, for example, nitrogenase was completely inactivated at partial pressures of 28% O2 in the gas phase,
and activity was restored at lower O2 levels (Debont and Mulder 1974). Similarly, the growth rate of Methylocystis strain T-1 with N2 as sole N-source decreased at increased
O2 partial pressure. Under 10% O2 atmosphere, no growth occurred (Takeda 1988).
MATERIALS AND METHODS
CULTURES
The strains evaluated in this study were Methylosinus trichosporium OB3b and Methylocystis parvus OBBP. Both are obligate aerobic methane-oxidizing alpha proteobacteria (Stein et al. 2010). Strain OB3b is often referred to as the “work horse” organism for research on the physiology, biochemistry, and molecular biology/genetics of methanotrophy since Whittenbury’s initial isolation in 1970 (Murrell and Jetten 2009; Stein et al. 2010). Strain OBBP has been studied for its PHB-producing ability (Pieja et al. 2011a). Cultures of strains OB3b and OBBP were provided by J. Semrau (University of Michigan).
CULTURE GROWTH CONDITIONS
All cultures were grown under six different oxygen partial pressures in 158-ml serum bottles. Glassware was acid-washed with 10% HCl for ≥1 h and triple-rinsed in Milli-Q water before use to remove trace metal contamination. Each culture was grown
in 50 ml of W1 media (containing 0.8 mM MgSO4·7H2O, 0.14 mM CaCl2·2H2O, 1.2
31 mM NaHCO3, 2.35 mM KH2PO4, 3.4 mM K2 HPO4, 20.7 μM Na2MoO4·2H2O, 1 μM
CuSO4·5H2O, 10 μM FeEDTA), 1 ml trace metal solution (containing, per liter: 500 mg
FeSO4 · 7H2O, 400 mg ZnSO4 · 7H2O,20 mg MnCl2 · 7H2O , 50 mg CoCl 2 · 6H2O , 10
mg NiCl2·6H2O,15 mg H3BO3, 250 mg EDTA), and 10 ml vitamin solution (containing, per L: 2.0 mg biotin, 2.0 mg folic acid,5.0 mg thiamine·HCl, 5.0 mg calcium pantothenate, 0.1 mg vitamin B12, 5.0 mg riboflavin, and 5.0 mg nicotiamide). The medium was sterilized by autoclaving at 121°C for 40 min. All cultures were grown with
one of the following nitrogen sources: nitrate (10 mM NaNO3 in W1 media), ammonium
(10 mM as NH4Cl), or nitrogen gas(N2 in bottle headspace). For cultures grown with nitrate and ammonium, the headspace of the serum bottles was pressurized with helium (He) after alternatively vacuum-degassing and refilling with He eight times (1 min vacuum-degas and 1 min He refill). Pressure in the headspace was released using a 23G needles to obtain ambient atmospheric pressure prior to addition of oxygen (O2) and
methane (CH4) gases. For samples grown with N2 as the sole N-source, the headspace was refilled with N2 gas instead of He. Each serum bottle was inoculated with 2.5 mL
(5%) of active culture (Optical Density at 670nm wavelength (OD670) ≥ 0.4, grown under the same conditions) and pressurized by adding 30 ml of methane, and different amounts of oxygen (0.05 atm of O2: injected 5.5 ml O2; 0.10 atm of O2: injected 11 ml O2; 0.15
atm of O2: injected 16.5 ml O2; 0.20 atm of O2: injected 22 ml O2; 0.30 atm of O2:
injected 33 ml O2; and 0.40 atm of O2: injected 44 ml of O2). Table 5 summarizes headspace composition. All cultures were incubated horizontally on orbital shake tables at 150 rpm and 30°C.
32
Table 5 Headspace composition for each experimental setup.
Setup Headspace Composition by Nitrogen Source Oxygen Partial Nitrate Ammonium Nitrogen Gas - + Pressure (atm) (NO3 ) (NH4 ) (N2) 30 ml methane 0.05 5.5 ml oxygen
Balance: helium Balance: helium Balance: N2 gas 30 ml methane 0.10 11 ml oxygen
Balance: helium Balance: helium Balance: N2 gas 30 ml methane 0.15 16.5 ml oxygen
Balance: helium Balance: helium Balance: N2 gas 30 ml methane 0.20 22 ml oxygen
Balance: helium Balance: helium Balance: N2 gas 30 ml methane 0.30 33 ml oxygen
Balance: helium Balance: helium Balance: N2 gas 30 ml methane 0.40 44 ml oxygen
Balance: helium Balance: helium Balance: N2 gas
GROWTH MONITORING AND PHB PRODUCTION
For both strains, 18 incubation conditions (6 levels of O2 and 3 nitrogen sources) were evaluated for growth. Bottle headspace was sampled periodically to evaluate methane
33 consumption, oxygen consumption, nitrogen consumption, and carbon dioxide production. 0.5 ml samples were injected into a Gow-Mac GC (GOW-Mac Instrument Co., Bethlehem, PA) equipped with a thermal conductivity detector (TCD) with a CTR1 column (Alltech Associates Inc., Deerfield, IL) with helium (He) as carrier gas, and gas composition was calibrated to prepared standards at each sampling point. In the case that there was no remaining oxygen in the headspace but methane was still available, the original amount of oxygen was injected. Cell growth was measured via optical density at 670 nm using a spectrophotometer at each sampling point. Cultures that did not attain exponential growth grew slowly or not at all. Cultures that entered the exponential growth phase were subsequently assayed for PHB production after the added methane (30 ml) disappeared. PHB production was induced by incubation with 1:1 methane:oxygen in the absence of nitrogen. Cultures were centrifuged at 4,816×g (4,700 rpm) for 8 min, washed once with W1 medium, re-centrifuged, and re- suspended in the same volume of W1 medium. After 22 h of incubation, cultures were harvested, immediately frozen at −20°C, and lyophilized for PHB analysis. Both strains were grown in triplicate and induced for PHB production by transfer to serum bottles with 1:1 methane:oxygen (no nitrogen), then assayed for PHB production after 22 hours. Aqueous phase concentrations of methane and oxygen were computed from partial pressure measurements using Henry’s law constant (T =30°C, solubility in pure water); the Henry’s constant for CH4 was 0.0014 mol/kg ∙ bar with temperature dependence constant, K = 1600, and the Henry’s constant for O2 was 0.0013 mol/kg ∙ bar, bar with temperature dependence constant, K = 1500(CRC Handbook of Chemistry and Physics 1995).
PHB MEASUREMENT
Methodology described in (Pieja et al. 2011b) was used for each sample measurement: 3–6 mg of lyophilized biomass was added to a 12-mL glass vial with a PTFE-lined plastic cap (Wheaton Science Products). A modified version of the protocol described by Braunegg et al. was used for the PHB assay (Braunegg et al. 1978). The organic phase of the resulting mixture was analyzed using an Agilent 6890N gas
34 chromatograph equipped with an HP-5 column (containing (5% phenyl)- methylpolysiloxane, Agilent Technologies) and FID detector. DL-β-hydroxybutyric acid sodium salt (Sigma) was used as a standard.
MODELING OF STOICHIOMETRY
During cell synthesis, a fraction of the electrons (fe) derived from methane is used
to reduce oxygen to water for energy generation and the remaining fraction (fs) is used
for cell synthesis. The resulting electron balance is fe + fs = 1. Three half reactions are
needed to describe cell growth: one for the oxidation of the electron donor (Rd), one for
reduction of the electron acceptor reaction (Ra), and one for cell synthesis (Rc), normalized to 1 mole of electrons. The electron acceptor reaction is straightforward:
The half reaction for the electrons used for energy generation is feRa: + - fe (¼ [O2 + 4H + 4e 2H2O]) (1) In the case of monooxygenase-mediated reactions, not all the oxygen consumed is reduced to water. Some O2 is a reactant in the initial attack on methane. This step is described by the following stoichiometry: + - CH4 + O2 + 2H + 2e CH3OH +H2O (2)
The O2 that is used for the monooxygenase-mediated attack on methane (equation 2) is stoichiometricially, physiologically, and energetically distinct from the oxygen used as the terminal electron acceptor for energy production (equation 1). For the monooxygenase-mediated reaction, one atom of oxygen is incorporated into methane to produce methanol. When methane is used for energy methanol is converted linearly to carbon dioxide: + - CH3OH +H2O CO2 + 6H + 6e (3) Therefore, the electron donor reaction, is the sum of half reactions 2 and 3:
Electron donor reaction (Rd): + - 1/4 [CH4 + O2 CO2 + 4H + 4e ] (4) The half reaction for the electrons used for cell synthesis (fsRc) depends upon the nature of the nitrogen source and other elements used to synthesize biomass. To produce
35 biomass with the empirical formula C5H7O2N (Rittmann and McCarty 2001) and nitrate is the sole source of nitrogen, the synthesis half reaction is: - + - 5CO2 + NO3 + 29H + 28e C5H7O2N + 11H2O (5)
Dividing equation 5 by 28 results in a normalized cell synthesis half reaction (fsRc) that
accounts for the monooxygenase-mediated conversion of methane to methanol with O2 as a reactant: - + - fs (1/28 [5CO2 + NO3 + 29H + 28e C5H7O2N + 11H2O]) (6) The total reaction (R) for nitrate as N source incorporates equations 1, 4, and 6,
where R = Rd + feRa + fsRc: - + Rnitrate = (1/4) CH4 + (1/4 + fe/4) O2 + (fs/28) NO3 + (29/28 fs + fe -1) H (1/4 –
5fs/28) CO2 + (fe/2 + 11fs/28) H2O + (fs/28) C5H7O2N
Using the above stoichiometry, the ratio of O2 consumed to CH4 consumed = 1+ fe,
where fe + fs = 1 during growth. The mass ratio is 2(1+ fe). The biomass yield, YX, (g
VSS/gCH4) depends upon the nitrogen source: for nitrate, it is 113fs/28 : 4; for
ammonium, it is 113 fs/23 : 4; and for nitrogen gas, it is 113 fs/25 : 4. During the PHB
production phase, the “biomass” is PHB with an empirical formula of C4H6O2. No
nitrogen is required in this case, and the PHB yield, YPHB, (gPHB/gCH4) is 86 fs/18 : 4. Table 6 summarizes reaction stoichiometry for both the growth phase and the PHB production phase. Table 6 Stoichiometric equations used to describe methanotrophic growth and PHB production.
Nitrogen Total Reaction Source GROWTH PHASE - + Nitrate (1/4) CH4 + (1/4 + fe/4) O2 + (fs/28) NO3 + (29/28 fs + fe - 1) H - (NO3 ) (1/4 – 5fs/28) CO2 + (fe/2 + 11fs/28) H2O + (fs/28) C5H7O2N - + + Ammonium (1/4) CH4 + (1/4 + fe/4) O2 + (fs/23) HCO3 + (fs/23) NH4 + (20fs/23 + fe – 1) H + (NH4 ) (1/4 – 4fs/23) CO2 + (fe/2 + 9fs/23) H2O + (fs/23) C5H7O2N Nitrogen gas (1/4) CH4 + (1/4 + fe/4) O2 + (fs/50) N2 (N2) (1/4 – fs/5) CO2 + (fe/2 + 8fs/25) H2O + (fs/25) C5H7O2N PHB PRODUCTION PHASE No Nitrogen (1/4) CH4 + (1/4 + fe/4) O2 (1/4 – 4fs/18) CO2 + (fe/2 + fs/3)H2O + (fs/18) C4H6O2 36
The analysis of Table 6 can be paired with measured gas consumption and production, such as that depicted in Figure 7, to calculate fe and fs, cell yield (YX), and
PHB yield (YPHB), shown in Table 7 and 8.
MODELING OF KINETICS
The system can be described as a batch system with a controlled volume with no mass entering or exiting. The mass rate of substrate (methane) accumulation is:
= + −훥푀푠 where C is the concentration in the gas훥 phase,퐶퐺푉퐺 V훥퐶 is퐿푉 the퐿 volume of the gas, C is the G 훥푡 G L concentration of the liquid phase, and VL is the volume of the liquid. Similarly, the mass of organism accumulation can be considered based on typical microbial growth kinetics, described by the Monod equation (Rittmann and McCarty 2001): S = max K + S where µ is the specific growth rate µdue toµ synthesis, S is the concentration of the rate-
limiting substrate, µmax is the maximum specific growth rate, and K is the concentration of S that gives one-half the maximum rate. The maximum specific growth rate, µmax ,
can be used with the biomass yield, YX, to calculate the maximum specific rate of
substrate utilization, qmax:
q = Ymax max 휇 With known partial pressure in the headspace xduring sample points, we calculate the concentration in the media of both methane and oxygen at each sample point using Henry’s law constant with temperature dependence for our experimental temperature,
namely 30°C, (CRC Handbook of Chemistry and Physics 1995). Typical values of KCH4
and KDO for methanotrophic cultures are in the ppb range (Arcangeli and Arvin 1997; Dunfield and Conrad 2000), well below the levels investigated in this study. Accordingly, S >> K, and zero order kinetics apply and the mass rate of biomass accumulation is: 37 = 훥푀X where Xa is the biomass concentration, expressed푞푚푎푥푋푎 푉as퐿 mass of volatile suspended solids 훥푡 per liter (mg VSS/L) at time t. Using a dimensionless Henry’s constant, HC, based on ratio of concentrations in the gas phase to the liquid phase, the mass rate of substrate (methane) accumulation and the mass rate of biomass accumulation can be equated:
= + 푑퐶퐿 푞푚푎푥푋푎푉퐿 − Multiplying the above equation by 푑푡YX on both퐻퐶 푉sides,퐺 푉 퐿we see the effect that the gas phase (term HCVG) has on µmax, where X represents the mass of the biomass:
Y = (Y ) 퐿 푚푎푥 +푎 퐿 X 푑퐶 X 푞 푋 푉 �− 퐶 퐺 퐿� 푑푡 = 퐻 푉 푉 + 푑푋 휇푚푎푥푋푎푉퐿 − Using experimental data for dCL/dt푑푡 and Xa퐻 at퐶 푉each퐺 푉time퐿 step the nonlinear least-squares fitting (NLSF) described by Kemmer and Keller (Kemmer and Keller 2010) was used to calculate qmax for each incubation. This method minimizes the sum of the squared
differences of the experimentally calculated dCL/dt and values computationally determined . 푞푚푎푥푋푎푉퐿 퐶 퐺 퐿 The value− of퐻 푉qmax+푉 can also be determined mathematically, where A is matrix
representing values dCL/dt and B is a matrix representing values for each −푋푎푉퐿 experimental setup and solved using Matlab: 퐻퐶푉퐺+푉퐿
=
퐀 푞=푚푎푥 ∗ 퐁 ( ) 퐓 퐓 퐀퐁 ( 푞)푚푎푥 =퐁퐁 퐓 퐓 −1 Microbial kinetic parameters, maximum퐀퐁 퐁퐁 specific푞 푚푎푥growth rate (µmax), and maximum specific rate of substrate utilization (qmax) are listed in Table 8.
38 RESULTS
Figure 7 below shows the gas composition and biomass production of two representative experimental setups. The trends show rate of methane and oxygen consumption as well as rates of carbon dioxide production and biomass accumulation (growth).
Figure 7 Methane consumption, oxygen consumption, carbon dioxide production, and biomass accumulation (growth) of (left) Methylosinus trichosporium OB3b with nitrate at 0.3 atm O2 and (right) Methylocystis parvus OBBP with ammonium at 0.3 atm O2.
Table 7 lists fe and fs, cell yield (YX) based on the stoichiometric analysis of Table 6 and gas composition and biomass accumulation data of each experiment, such as the examples shown in Figure 7. Table 7 also summarizes PHB production (as percent of dry weight) for each strain after growth under different conditions followed by incubation without nitrogen.
39 Table 7 Substrate partitioning parameters (fe, fs), cellular yield (YX), and % PHB production (by cell dry weight) of each strain by oxygen partial pressure and nitrogen source.
fe fs YX % PHB after nitrogen (g VSS/g methane) limitation† (cell dry weight)
) ) ) ) ) ) ) ) - - - - + + + + (atm) ) ) ) ) Strain 3 3 3 3 4 4 4 4 2 2 2 2 ( N ( N ( N ( N Nitrate Nitrate Nitrate Nitrate Nitrate Nitrate ( NO ( NO ( NO ( NO (NH (NH (NH (NH Ammonium Ammonium Ammonium Ammonium Nitrogen Gas Nitrogen Gas Nitrogen Gas Nitrogen Gas Oxygen Partial Pressure Pressure Partial Oxygen 0.1 0.31 0.57 0.15 0.69 0.43 0.85 0.69 0.53 0.96 22 14 20 0 0.1 0.34 0.59 0.64 0.66 0.41 0.36 0.66 0.51 0.41 22 13 28
5 0.2 B3b 0.31 0.34 0.71 0.69 0.66 0.29 0.69 0.82 0.32 20 9 45
O 0
0.3 0.38 0.50 N/A 0.62 0.50 N/A 0.63 0.61 N/A 29 12 N/A 0 0.4 0.36 0.19 N/A 0.64 0.81 N/A 0.66 0.99 N/A 24 13 N/A 0 Av 0.34± 0.44 0.66 0.56 0.66 0.69 24 11 29 ± St N/A N/A N/A 0.03 ±0.17 ±0.03 ±0.17 ±0.03 ±0.21 ±4 ±4 ±11 Dv 0.1 0.41 0.39 N/A 0.59 0.61 N/A 0.59 0.75 N/A 19 50 N/A 0 0.1 0.46 0.56 N/A 0.54 0.44 N/A 0.55 0.54 N/A 11 41 N/A
5 0.2 0.48 0.44 N/A 0.52 0.56 N/A 0.53 0.69 N/A 8 37 N/A 0 OBBP 0.3 0.44 0.41 N/A 0.56 0.59 N/A 0.57 0.73 N/A 14 60 N/A 0 0.4 0.47 0.15 N/A 0.53 0.85 N/A 0.53 1.05 N/A 6 42 N/A 0 Av 0.45 0.39 0.55 0.61 0.55 0.75 14 46 ± St N/A N/A N/A N/A ±0.03 ±0.15 ±0.03 ±0.15 ±0.03 ±0.19 ±8 ±8 Dv N/A: Not applicable because the cultures did not show significant growth. † PHB production was induced by incubation with 1:1 methane:oxygen in the absence of nitrogen.
For strain OB3b, the highest value of fs (0.85) occurred when cells were grown with nitrogen gas at low O2 partial pressure (0.10 atm). Overall, however, fs was maximum when nitrate was the nitrogen source, with fs = 0.66±0.03. A lower and more variable value resulted when cells were grown with ammonium or N2 gas. For strain
OBBP, the highest fs (0.85) occurred when cells were grown with ammonium at high O2
40 partial pressure (0.40 atm). Overall, ammonium was the preferred nitrogen source, with an fs value of 0.61±0.15. The value for fs was lower and less variable with nitrate, at 0.55±0.03. High variability with oxygen was observed in both strains when ammonium was the nitrogen source. This may reflect the high level of reducing equivalents required for hydroxylamine reduction. Strain OB3b produced more PHB after growth with nitrate and nitrogen gas, while strain OBBP produced more PHB after growth with ammonium. Both cultures had low variability in PHB production after growth on nitrate. For strain OB3b, variability was much higher after growth with N2 as the N-source, 11%. Only one sample of strain
OBBP could be tested with N2 as nitrogen source due to lack of growth. Table 8 summarizes kinetic values for strains Ob3b and OBBP. These values are valuable for growth optimization and reactor design.
41
Table 8 Microbial kinetic parameters, maximum specific growth rate (µmax), and maximum specific rate of substrate utilization (qmax).
µmax qmax -1
(d ) (mg methane/ mg VSS d-1)
Strain ) ) ) )
- - + + ) ) 3 3 4 4 2 2 N N ( ( NO NO Pressure (atm) Pressure Oxygen Partial Partial Oxygen ( ( Nitrate Nitrate Nitrate Nitrate (NH (NH Ammonium Ammonium Ammonium Nitrogen Gas Nitrogen Gas 0.10 5.94 3.11 5.83 8.57 5.86 6.08 0.15 4.32 2.60 1.67 6.52 5.13 4.07 0.20 4.99 7.18 1.50 7.22 8.81 4.66 OB3b 0.30 3.36 3.92 N/A 5.35 6.39 N/A 0.40 4.96 6.06 N/A 7.67 6.11 N/A 3.93 4.57 3.00 7.07 6.46 4.94 Av±St Dev ±2.10 ±1.97 ±2.45 ±1.21 ±1.39 ±1.04
0.10 2.92 4.21 N/A 4.93 5.63 N/A 0.15 2.71 2.63 N/A 4.98 4.88 N/A 0.20 2.28 4.25 N/A 4.31 6.16 N/A 0.30 2.16 3.67 N/A 3.81 5.05 N/A
OBBP 0.40 5.55 5.94 N/A 3.07 5.67 N/A 2.63± 4.14± N/A 4.72± 5.48±0. N/A Av±St Dev 0.40 1.20 0.67 51 N/A: Not applicable because the cultures did not show significant growth.
Strains OB3b and OBBP were tolerant of oxygen when either nitrate or ammonium was the sole nitrogen source, as indicated by the low variability in µmax values in Table 8. Under nitrogen-fixing conditions, however, both strains were sensitive to oxygen. Strain OBBP was the most sensitive, with growth only occurring at O2 levels
≤ 0.05 atm (data not shown), while strain OB3B only grew at O2 levels < 0.3 atm .
Using the PHB synthesis reaction in Table 6, values of fe and fs were determined for the PHB production after growth under “optimal” conditions for each strain. Conditions chosen for more detailed evaluation of PHB production stoichiometry were identified based upon the magnitude of µmax in the growth phase, the level of volatile suspended solids achieved in the growth phase, and the %PHB achieved in the PHB 42 accumulation phase. For strain OB3b, nitrate as N source and 0.3 atm oxygen; for strain OBBP, ammonium as N source and 0.3 atm oxygen. These values are summarized in Table 9. Table 9 Substrate partitioning parameters (fe, fs,) yield during PHB production after growth under optimal conditions: for strain OB3b, nitrate as N source and 0.3 atm oxygen; for strain OBBP, ammonium as N source and 0.3 atm oxygen.
Strain fe fs YPHB (g PHB/g methane) Methylosinus trichosporium 0.05±0.02 0.95±0.02 1.13±0.02 OB3b Methylocystis parvus OBBP 0.26±0.10 0.74±0.10 0.88±0.12
Table 9 indicates that strain OB3b can achieve a higher fs, 0.95 than strain OBBP, fs, = 0.74. This allows strain OB3b to achieve a higher yield of PHB per unit of methane (1.13) than strain OBBP (0.88) during PHB production phase.
DISCUSSION
Stoichiometry and kinetics are critical to the economics of PHB production from a biogas or natural gas feedstock. High specific growth rates, high substrate utilization rates, and high levels of PHB production are desirable to minimize environmental and economic costs and to maximize benefits. Both strains were less sensitive to oxygen when nitrate or ammonium was provided as the nitrogen source. Dinitrogen allowed only slow growth and exhibited an obvious oxygen threshold, likely due to the sensitivity of nitrogenase to oxygen. Low oxygen is also undesirable for other reasons. Intracellular PHB degradation has been observed in one methanotroph under anaerobic conditions in the absence of an exogenous carbon source (Vecherskaya et al. 2009) and some methanotrophs, such as strain OB3b,
43 have been shown to sporulate when oxygen-starved (Titus et al. 1982; Whittenbury et al. 1970b). Table 10 below lists specific growth rates found in this study to those found in the literature. Although the growth kinetics of methanotrophic bacteria for the purposes of TCE degradation has been well studied, particularly in mixed cultures, data on kinetics of pure culture groth under conditions of different nitrogen sources and specific atmospheric pressures of oxygen are unavailable. It is obvious from this table that these parameters have significant variability, likely dependent on growth conditions. The values reported here are within the range found in literature. Table 10 Kinetic values reported for methanotrophic growth.
-1 Source µmax (d ) This study 1.5-7.18 (Anderson and McCarty 1994) 2.17 (Broholm et al. 1992) 0.344 (Ferenci et al. 1975) 4.4-4.6 (Heijnen and Roels 1981) 0.96-8.16 (Oldenhuis et al. 1991) 4.18 Adapted from (Arcangeli and Arvin 1997) The values given in Tables 7-9 are critical for bioreactor design and process evaluations, such as life cycle assessments (LCAs). Table 11 compares parameters identified for “optimal conditions” from this work with parameters used for life cycle modeling, using literature values and best estimates. A key parameter is the methane required to produce a unit of PHB. As shown in Table 10, a prior LCA (Chapter 1) may have vastly underestimated the benefits of PHB production for strain OBBP, indicating that process viability is highly dependent on which methanotrophic culture is used. Analyses such as this one may also serve to better select for other high yield strains, and further optimization may be possible with communities rather than pure cultures (Johnson et al. 2009a; Pfluger 2010; Pieja et al. 2011b; Pieja et al. 2011c).
44
Table 11 Parameterization of PHB production from methane. Target: Production of 1.00 g PHB.
Parameter Units Model Observed Observed Value in for Strain for Strain Chapter 1 OB3b OBBP LCA Percent PHB % 50 29 60 (g PHB/ (Table 7) (Table 7) g dry weight) Non-PHB Biomass = g biomass 1.00 2.45 0.66 1 g 1g % PHB �Methanotrophic� − Growth Yield g biomass/ 0.345 0.63 0.73 g methane (Table 7) (Table 7) Methane Requirement for Non- g methane 2.90 3.89 0.92 PHB Biomass = Non PHB Biomass
Methanotrophic Growth Yield − �Yield of PHB on Methane � g PHB/ 0.55 1.13 0.88 g methane (Table 9) (Table 9) Methane Requirement for 1.00 g g methane 1.82 0.88 1.13 PHB = 1.00 g PHB
Yield of PHB on Methane
� � Total Methane Requirement = g 4.72 4.77 2.05 (Methane Requirement for methane Non-PHB Biomass + Methane Requirement for 1.00 g PHB)
45 CHAPTER 3: METHANOCYC: A DATABASE FOR METHYLOSINUS TRICHOSPORIUM OB3B
This work is in preparation for submission for publication.
Katherine H. Rostkowski, Peter D. Karp, and Craig S. Criddle
ABSTRACT
MethanoCyc is an organism-specific pathway/genome database for Methylosinus trichosporium OB3b, an obligate aerobic methane-oxidizing alpha proetobacterium, that has been generated using the Pathway Tools Software. It can aid in the study of cellular processes in Methylosinus trichosporium OB3b and methanotrophs in general. MethnoCyc is available for public access at http://www.biocyc.org/organism- summary?object=MOB3B. Pathway reconstruction was used to predict the metabolic composition of Methylosinus trichosporium OB3b as a representative organism for methanotrophs, resulting in a pathway/genome database (PGDB) of 976 reactions. This metabolic network provides a platform for the visualization of experimental data from omics experiments, such as differences in metabolites during growth and during polyhydroxybutyrate (PHB) production. Additionally, the PGDB can be used to facilitate comparative studies of pathways across species, such as the comparison to a non-PHB- producing methanotroph shown here. Lastly, MethanoCyc provides a resource for biotechnology applications of methanotrophs, such as through flux balance analysis.
INTRODUCTION
Methanotrophs, discovered in 1970 by Whittenbury, are gram-negative aerobes utilizing only methane and methanol as combined carbon and energy sources (Whittenbury et al. 1970a). These microorganism are a subset of the methylotrophs, bacteria that metabolize one-carbon compounds (Hanson and Hanson 1996; Murrell 2010; Lidstrom 2006; Mancinelli 1995). Being widely distributed in the environment 46 (Murrell and Jetten 2009), wherever there is an exchange of methane and oxygen (Dworkin and Falkow 2006), they are the major terrestrial sink for methane (Hanson and Hanson 1996; Murrell 2010; Lidstrom 2006; Murrell and Jetten 2009). Microbes that produce and consume methane, methanogens and methanotrophs, respectively, “harbor many secrets that need to be disclosed” for a complete understanding of the biogeochemical methane cycle in order to make global predictions on the cycling of this important greenhouse gas (Murrell and Jetten 2009). Understanding methanotrophy may also be of biotechnological interest (Trotsenko et al. 2005). Beginning in the 1970s when methanotrophs were first discovered, there was interest in the inexpensive production of single-cell proteins and more recently in the production of added value protein products such as fish feed in Denmark and Norway (Smith et al. 2010). Methanotrophs are also researched for the production of the bioplastic polyhydroxybutyrate (PHB) (Wendlandt et al. 2001; Helm et al. 2008; Choi and Lee 1999; Wendlandt et al. 2005; Whittenbury et al. 1970a; Pieja et al. 2011b; Pieja et al. 2011a). In addition, the methane monooxygenase systems in methanotrophs have made them interesting for synthetic chemistry and bioremediation applications (Smith et al. 2010), most notably for the co-oxidation of trichloroethylene (TCE) and other chlorinated solvents in contaminated environments (Anderson and McCarty 1997). Much is to be gained from the recent sequencing of the methanotroph, Methylosinus tricosporium OB3b (‘oddball’ strain 3b) (Stein et al. 2010). It may be considered a representative organism for methanotrophs, often referred to as the “work horse” organism in research on the physiology, biochemistry and molecular biology/genetics of methanotrophy since Whittenbury’s initial isolation in 1970 (Murrell and Jetten 2009; Stein et al. 2010). This strain is an obligate aerobic methane-oxidizing alpha proteobacterium (Stein et al. 2010). The genome is the first reported in the Methylocystaceae family in the order Rhizobiales (Stein et al. 2010). It was sequenced, assembled, and annotated by the US Department of Energy Joint Genome Institute (JGI) (Stein et al. 2010).
47 The goals of this research were: (i) to use pathway reconstruction for predicting the metabolic composition of Methylosinus trichosporium OB3b as a representative organism for methanotrophs; and (ii) to provide a platform for the visualization of experimental data from genomics, transcriptomics, proteomics, and metabolomics. Additionally, the long-term goals are: (iii) to facilitate comparative studies of pathways across species; and (iv) to provide a resource for biotechnology applications of methanotrophs such as through flux balance analysis.
PATHWAY/GENOME DATABASE CONSTRUCTION & METABOLISM
An organism’s genome can be used to construct a representative pathway/genome database. MetaCyc is used as a reference database in conjunction with the PathoLogic component of the Pathway Tools software (Karp et al. 2002; Dale et al. 2010) to computationally predict the metabolic network of the organism from its genome and create a pathway/genome database (PGDB) (Caspi and Karp 2007; Paley and Karp 2002). The current version of MetaCyc (http://metacyc.org) contains 1747 pathways from more than 2170 different organisms (Paley and Karp 2002; Krieger et al. 2004; Caspi et al. 2008; Caspi and Karp 2007; Karp et al. 2006) with more than 90% of its pathways manually curated with literature citations and species information (Zhang et al. 2005). The Pathway Tools software has been optimized such that it outperforms expert analyses in metabolic pathway prediction (Paley and Karp 2002). The PGDB describes each gene, the metabolic network of the organism (pathways, reactions, enzymes, and metabolites), and the regulatory network of the organism (operons, transcription factors). Pathway Tools allows the user to create and update the contents of a PGDB, publish a PGDB, as well as perform complex queries, visualization, and analysis (Karp et al. 2002). Several such databases have been constructed and curated (May et al. 2009; Keseler et al. 2011; Sumner and Urbanczyk-Wochniak 2007; Mueller et al. 2003; Cherry et al. 1998). Pathway Tools has many tools for computational analysis, including comparative analysis and analysis of omics data in a pathway context, and can be useful in biochemistry, molecular biology, biotechnology, bioinformatics, metabolic engineering, and systems biology (Caspi et al. 2008). In a post-genomic era with modern high- 48 throughput technologies, model organism databases can be important for the integration of new experimental data for a holistic understanding of cellular processes (Karp et al. 2002; May et al. 2009). MethanoCyc is a web accessible PGDB created from the Methylosinus trichosporium OB3b genome, which was downloaded from JGI, and can serve as a model organism database for methanotrophs. It is available at http://www.biocyc.org/organism- summary?object=MOB3B. The metabolic reconstruction was evaluated by manually verifying and curating known methanotrophic pathways described in the literature (Whittenbury et al. 1970a; Hanson and Hanson 1996; Lidstrom 2006; Asenjo and Suk 1986; Hakemian and Rosenzweig 2007; Bowman 2006; Lieberman and Rosenzweig 2004; Mancinelli 1995; Murrell 2010; Semrau et al. 1995; Leak and Dalton 1983; Bowman et al. 1993; Smith et al. 2010; Cornish et al. 1984; Vecherskaya et al. 2001). MethanoCyc is currently the most comprehensive genome-wide metabolic database available for a methanotroph. Table 11 summarizes the MethanoCyc Database. Table 12 MethanoCyc Database Summary Statistics.
Pathways 187
Enzymatic Reactions 976
Polypeptides 4472
Enzymes 616
Compounds 727
Citations 385 Methanotrophs are unique in that they use methane monooxygenases to catalyze the oxidation of methane to methanol (Hanson and Hanson 1996; Dumont and Murrell
2005). The net reaction of methane oxidation in the presence of oxygen is: CH4 +2O2
CO2 +2H2O (Mancinelli 1995). The pathway for methane assimilation is linear. Figure 8 is a visualization from MethanoCyc, showing the metabolism of substrates by methanotrophs, the central role of formaldehyde as an intermediate, and the pathways
49 employed for the synthesis of intermediates (Hanson and Hanson 1996). Methane oxidation by aerobic methanotrophs is initiated by MMOs that use two reducing equivalents to split the O-O bonds of dioxygen. One of the oxygen atoms is reduced to
water (H2O), while the other is incorporated into methane to form methanol, CH3OH (Hanson and Hanson 1996).
Figure 8 Methane Assimilation by Methanotrophs.
There are two forms of MMO in methanotrophs (Hanson and Hanson 1996; Dumont and Murrell 2005). All known methanotrophs can form particulate or membrane-bound MMO (pMMO), which is an integral membrane metalloenzyme (Lieberman and Rosenzweig 2005). Type I methanotrophs use the pMMO, while Type II methanotrophs and Type X methanotrophs produce a cytoplasmic enzyme, soluble MMO (sMMO) in addition to the pMMO (Hanson and Hanson 1996). In all cases, methanol is then oxidized to formaldehyde by a periplasmic methanol dehydrogenase (MDH) in gram-negative methylotrophs and by an NAD-linked methanol dehydrogenase in gram- positive methylotrophs (Hanson and Hanson 1996). The assimilation of formaldehyde forms intermediates of the central metabolic routes that can be used for the biosynthesis of cell material (Hanson and Hanson 1996). Type I and Type X methanotrophs use the ribulose monophosphate pathway for carbon, 50 while the Type II methanotrophs use the serine cycle (Anthony 1982). As Methylosinus trichosporium OB3b is a Type II methanotroph, the serine cycle is shown in Figure 9.
Figure 9 Serine Cycle: formaldehyde assimilation in Type II methanotrophs.
VISUALIZATION OF EXPERIMENTAL DATA
Because PGDB construction creates a cellular overview of the metabolic network of an organism, omics data can be overlaid and visually represented. Figure 10, for example, shows the ratio of metabolomic measurements of Methylosinus trichosporium OB3b in growth phase and in polyhydroxybutyrate production phase. Metabolites in red, such as PHB intermediates, are those that are present at higher concentrations in whole broth samples during PHB production phase, while those in blue, such as protein-building blocks, are those that are present at higher concentrations in whole broth samples during
51 growth phase determined using rapid quenching (Canelas et al. 2008) and ethanol extraction (Lange et al. 2001) Other metabolites, such as decreases in succinate and malate during PHB production and increase in oxalate during PHB production are undergoing further evaluation.
Figure 10 Omics viewer image of ratio of metabolomics data from Methylosinus trichosporium OB3b growth and polyhydroxybutyrate production experiment.
SPECIES COMPARISON
The PGDB also helps facilitate comparison across species. By comparing the cellular overview for Methylosinus trichosporium OB3b, a Type II methanotroph, with that of Methylomonas methanica MC09, a representative Type I methanotroph, it is easy to identify the similarities (Figure 11) and difference in their metabolisms.
52
Figure 11 Methylosinus trichosporium OB3b cellular overview with reactions highlighted that are shared with Methylomonas methanica MC09.
It is not surprising that Type I and Type II methanotrophs would have similar metabolisms, but this tool could help identify valuable differences. Figure 11 highlights the PHB production pathway in Type II methanotrophs that is not present in the Type I methanotroph. Until recently, the literature contained conflicting evidence of as to which methanotrophs produce PHB and which do not (Pieja et al. 2011b). There had been several reports of PHB production in both Type I methanotrophs (Asenjo and Suk 1986; Bowman 2006; Bowman et al. 1993; Vecherskaya et al. 2001; Zhang et al. 2008; Wendlandt et al. 2001) and Type II methanotrophs (Asenjo and Suk 1986; Helm et al. 2006; Shah et al. 1995; Helm et al. 2008; Vecherskaya et al. 2001; Wendlandt et al. 1998, 2001; Zhang et al. 2008). Because the reports in Type I methanotrophs were based on qualitative evidence, there was a general misconception about PHB-producing ability of all methanotrophs. A more recent screening study found all type I strains tested negative for phaC (PHB producing gene) and PHB production; all Type II strains tested positive for phaC and PHB production (Pieja et al. 2011b). The species comparison tool could have visually suggested this difference much faster than misinterpreted laboratory studies suggested. This comparison platform could be used to compare this representative Type
53 II methanotroph to other species of interest such as soil bacteria, other PHB-producing bacteria, or other species of methanotrophs.
BIOTECHNOLOGY APPLICATIONS: FLUX BALANCE ANALYSIS
Because methanotrophs are of biotechnological interest for a variety of reasons, one goal of a pathway genome database is to provide a resource for biotechnology applications of methanotrophs such as through flux balance analysis. Flux balance analysis (FBA) is an approach to studying genome-scale biochemical networks and the flow of metabolites through such networks (Orth et al. 2010) that has become central for studying the systems biology of metabolism (Thiele and Palsson 2010). FBA allows us to quantitatively simulate the microbial metabolism (Kauffman et al. 2003). Typically, FBA development is time-consuming; however, MetaFlux, the software used to create the MethanoCyc FBA model, links FBA with pathway genome databases to speed the creation of FBA models. MetaFlux is a multiple gap-filling method to accelerate the development of FBA models using mixed integer linear programing (MILP) to suggest corrections to the sets of reactions, biomass metabolites, nutrients, and secretions that make up an FBA model (Latendresse et al. 2012). For MethanoCyc, we began with the known metabolites for E.coli from EcoCyc (Karp et al. 2009) and removed metabolites specific to E.coli only (e.g. spermidine, B12, 5-methyl THF). Using MetaFlux, the adjustments to the PGDB included making the reactions in aspartate production reversible as well as malyl-coA lyase and glycerone transferase (these are reversible in MetaCyc). The table below describes the necessary nutrients and secretions to produce the listed biomass for a functioning FBA model. To trace PHB polymer production, a specific piece of code had to be written to implement polymerization in the organism. Alternatively, monomer units, hydroxybutanoyl-coA could be monitored.
54
Table 13 Flux Balance Analysis Nutrients, Secretions, and Biomass Metabolites.
Nutrients (7) Methane Oxygen Nitrate or Ammonium Pi Sulfate Ferrous Iron (Fe 2+) Coenzyme A Secretions (3) Carbon dioxide Water Proton Biomass Metabolites (48) L-glutamate glycine L-alanine L-lysine L-aspartate L-arginine L-glutamine L-serine L-methionine L-tryptophan L-phenylalanine L-tyrosine L-cysteine L-leucine L-histidine L-proline L-asparagine L-valine L-threonine L-isoleucine GTP CTP UTP dATP dGTP dCTP dTTP N-acetylmuramoyl-L-alanyl-D-glutamyl-meso-2,6- 55 diaminopimelyl-D-alanyl-D-alanine- diphosphoundecaprenyl-N-acetylglucosamine NAD+ NADH NADP+ NAPH coenzyme A FAD pyridoxal 5'-phosphate S-adenosyl-L-methionine riboflavin ubiquinol-8 heme o di-trans,octa-cis-undecaprenyl diphosphate glutathione sulfate H2O ATP ADP phosphate diphosphate H+
Because the tool works with pathway genome databases directly, fluxes computed from the FBA model are easily queried and visualized on the metabolic network (Latendresse et al. 2012). With this model 258 reactions carry non-zero flux when ammonium is available as the nitrogen source and 262 reactions carry non-zero flux when nitrate is available as the nitrogen source, see Figure 12.
56
Figure 12 Visualization of Flux Balance Analysis of Methylosinus trichosporium OB3b (nitrate as nitrogen source).
The MethanoCyc FBA model can be used by future researchers for the purposes of metabolic engineering, enhancing the understanding of the metabolic network, testing experimental conditions computationally. FBA can also be used to predict the growth rate of an organism or the rate of production of a metabolite of biotechnological interest (Orth et al. 2010).
CONCLUSION
Pathway Tools Software has facilitated the generation of MethanoCyc. MethanoCyc is an organism-specific pathway/genome database for Methylosinus trichosporium OB3b, an obligate aerobic methane-oxidizing alpha proetobacterium, available at http://www.biocyc.org/organism-summary?object=MOB3B. Besides playing an important role in the global methane cycle, methanotrophs are biotechnologically relevant. The goals of this research were: (i) to use pathway reconstruction for predicting the metabolic composition of Methylosinus trichosporium OB3b as a representative organism for methanotrophs; (ii) to provide a platform for the visualization of 57 experimental data from omics experiments; (iii) to facilitate comparative studies of pathways across species; and (iv) to provide a resource for biotechnology applications of methanotrophs such as through flux balance analysis.
58 CONCLUSIONS
While the 140 million tons of plastics produced each year may contribute to quality of life, they also come at a significant cost. Their production requires large quantities of nonrenewable resources, contributing to climate change; they accumulate in landfills and natural environments; and they often contain harmful additives. One way to address the multiplicity of problems that arise from the widespread use of synthetic plastics–without compromising convenience and disposability—would be to replace them with functionally equivalent materials that are biobased, biodegradable, and biocompatible, such as polyhydroxyalkanoates—a class of bioplastics. Bacteria known as “methanotrophs” consume methane as feedstock, and some produce the PHA polymer poly-ß-hydroxybutyrate (PHB). PHB production from methane could take advantage of the abundant biogas methane that is currently flared or allowed to escape to the atmosphere by the waste sector (landfills and wastewater treatment plants) to produce a valuable product that biodegrades to methane at end-of-life, creating a closed-loop cycle. This research evaluates methanotrophic growth and PHB production across scale.
LIFE CYCLE ASSESSMENT
This work develops a LCA for synthesis of polyhydroxybutyrate (PHB) from methane with subsequent biodegradation of PHB back to biogas (40-70% methane, 30- 60% carbon dioxide). The parameters for this cradle-to-cradle cycle for PHB production are developed and used as the basis for a cradle-to-gate LCA. PHB production from biogas methane is shown to be preferable to its production from cultivated feedstock due to the energy and land required for the feedstock cultivation and fermentation. For the PHB-methane cycle, the major challenges are PHB recovery and demands for energy. Some or all of the energy requirements can be satisfied using renewable energy, such as a portion of the collected biogas methane. Oxidation of 18-26% of the methane in a biogas stream can meet the energy demands for aeration and agitation, and recovery of PHB synthesized from the remaining 74-82%. Effective coupling of waste-to-energy 59 technologies could thus conceivably enable PHB production without imported carbon and energy.
STOICHIOMETRY AND KINETICS
In addition to being the major terrestrial sink for methane, a major greenhouse gas, methanotrophs are of biotechnological interest for a variety of purposes (e.g. single- cell protein production, polyhydroxybutyrate (PHB) production, bioremediation). Optimizing growth of Type II methanotrophs and their capacity for PHB production specifically is of commercial and environmental interest. In this study, we describe how oxygen and nitrogen source affect the stoichiometry and kinetics of growth and PHB production in the Type II methanotrophs Methylosinus trichosporium OB3b and Methylocystis parvus OBBP. Significant differences were observed, with major implications for the use of these species in biotechnology applications. Such analyses can better inform bioreactor design, scale-up models, and life cycle assessments (LCAs).
PATHWAY/GENOME DATABASE
Pathway Tools Software has facilitated the development of MethanoCyc. MethanoCyc is an organism-specific pathway/genome database for Methylosinus trichosporium OB3b, an obligate aerobic methane-oxidizing alpha proetobacterium, available at http://www.biocyc.org/organism-summary?object=MOB3B. This research (i) uses pathway reconstruction for predicting the metabolic composition of Methylosinus trichosporium OB3b as a representative organism for methanotrophs; (ii) provides a platform for the visualization of experimental data from omics experiments; (iii) facilitates comparative studies of pathways across species; and (iv) provides a resource for biotechnology applications of methanotrophs, such as through flux balance analysis.
60 REFERENCES
Akiyama, M., T. Tsuge, and Y. Doi. 2003. Environmental life cycle comparison of polyhydroxyalkanoates produced from renewable carbon resources by bacterial fermentation. Polymer Degradation and Stability 80(1): 183-194.
Akmal, D., M. N. Azizan, and M. I. A. Majid. 2003. Biodegradation of microbial polyesters P(3HB) and P(3HB-co-3HV) under the tropical climate environment. Polymer Degradation and Stability 80(3): 513-518.
Anderson, A. J. and E. A. Dawes. 1990. Occurrence, Metabolism, Metabolic Role, and Industrial Uses of Bacterial Polyhydroxyalkanoates. Microbiological Reviews 54(4): 450-472.
Anderson, J. E. and P. L. McCarty. 1994. Model for Treatment of Trichloroethylene by Methanotrophic Biofilms. Journal of Environmental Engineering-Asce 120(2): 379-400.
Anderson, J. E. and P. L. McCarty. 1997. Transformation yields of chlorinated ethenes by a methanotrophic mixed culture expressing particulate methane monooxygenase. Appl Environ Microbiol 63(2): 687-693.
Anthony, C. 1982. The biochemistry of methylotrophs. London ; New York: Academic Press.
Arcangeli, J.-P. and E. Arvin. 1997. Modelling of the growth of a methanotrophic biofilm. Water Science and Technology 36(1): 199-204.
Asenjo, J. A. and J. S. Suk. 1986. Microbial Conversion of Methane into Poly-Beta- Hydroxybutyrate (Phb) - Growth and Intracellular Product Accumulation in a Type-Ii Methanotroph. Journal of Fermentation Technology 64(4): 271-278.
61 Auman, A. J., C. C. Speake, and M. E. Lidstrom. 2001. nifH sequences and nitrogen fixation in type I and type II methanotrophs. Applied and Environmental Microbiology 67(9): 4009-+.
Bare, J. 2011. TRACI 2.0: the tool for the reduction and assessment of chemical and other environmental impacts 2.0. Clean Technologies and Environmental Policy 13(5): 687-696.
Bare, J., T. Gloria, and G. Norris. 2006. Development of the method and U.S. normalization database for Life Cycle Impact Assessment and sustainability metrics. Environmental Science & Technology 40(16): 5108-5115.
Baumann, H. and A.-M. Tillman. 2004. The hitchhiker's guide to LCA : an orientation in life cycle assessment methodology and application.
Bentley, R. W. 2010. The expected dates of resource-limited maxima in the global production of oil and gas. Energy Efficiency 3(2): 115-122.
Bowman, J. 2006. The Methanotrophs — The Families Methylococcaceae and Methylocystaceae. In The Prokaryotes, edited by M. Dworkin, et al.: Springer New York.
Bowman, J. P., L. I. Sly, P. D. Nichols, and A. C. Hayward. 1993. Revised Taxonomy of the Methanotrophs - Description of Methylobacter Gen-Nov, Emendation of Methylococcus, Validation of Methylosinus and Methylocystis Species, and a Proposal That the Family Methylococcaceae Includes Only the Group-I Methanotrophs. International Journal of Systematic Bacteriology 43(4): 735-753.
Braunegg, G., B. Sonnleitner, and R. M. Lafferty. 1978. Rapid Gas-Chromatographic Method for Determination of Poly-Beta-Hydroxybutyric Acid in Microbial Biomass. European Journal of Applied Microbiology and Biotechnology 6(1): 29- 37.
62 Braunegg, G., G. Lefebvre, and K. F. Genser. 1998. Polyhydroxyalkanoates, biopolyesters from renewable resources: Physiological and engineering aspects. Journal of Biotechnology 65(2-3): 127-161.
Broholm, K., T. Christensen, and B. Jensen. 1992. Modelling TCE degradation by a mixed culture of methane-oxidizing bacteria. Water Research 26: 1177-1185.
Byrom, D. 1987. Polymer Synthesis by Microorganisms - Technology and Economics. Trends in Biotechnology 5(9): 246-250.
Byrom, D. 1992. Production of poly-B-hydroxybutyrate: poly-B-hydroxyvalerate copolymers. FEMS Microbiol Rev 103: 247-250.
Canelas, A. B., C. Ras, A. ten Pierick, J. C. van Dam, J. J. Heijnen, and W. M. Van Gulik. 2008. Leakage-free rapid quenching technique for yeast metabolomics. Metabolomics 4(3): 226-239.
Caspi, R. and P. D. Karp. 2007. Using the MetaCyc pathway database and the BioCyc database collection. Curr Protoc Bioinformatics Chapter 1: Unit1 17.
Caspi, R., H. Foerster, C. A. Fulcher, P. Kaipa, M. Krummenacker, M. Latendresse, S. Paley, S. Y. Rhee, A. G. Shearer, C. Tissier, T. C. Walk, P. Zhang, and P. D. Karp. 2008. The MetaCyc Database of metabolic pathways and enzymes and the BioCyc collection of Pathway/Genome Databases. Nucleic Acids Research 36(Database issue): D623-631.
Chen, G. Q. 2009. A microbial polyhydroxyalkanoates (PHA) based bio- and materials industry. Chemical Society Reviews 38(8): 2434-2446.
Cherry, J. M., C. Adler, C. Ball, S. A. Chervitz, S. S. Dwight, E. T. Hester, Y. K. Jia, G. Juvik, T. Roe, M. Schroeder, S. A. Weng, and D. Botstein. 1998. SGD: Saccharomyces Genome Database. Nucleic Acids Research 26(1): 73-79.
63 Chertow, M. R. 2007. "Unocvering" Industrial Symbiosis. In Journal of Industrial Ecology.
Choi, J. I. and S. Y. Lee. 1999. High-level production of poly(3-hydroxybutyrate-co-3- hydroxyvalerate) by fed-batch culture of recombinant Escherichia coli. Applied and Environmental Microbiology 65(10): 4363-4368.
Chu, K. H. and L. Alvarez-Cohen. 2000. Treatment of chlorinated solvents by nitrogen- fixing and nitrate-supplied methane oxidizers in columns packed with unsaturated porous media. Environmental Science & Technology 34(9): 1784-1793.
Cornish, A., K. M. Nicholls, D. Scott, B. K. Hunter, W. J. Aston, I. J. Higgins, and J. K. M. Sanders. 1984. In vivo 13C NMR Investigations of Methanol Oxidation by the Obligate Methanotroph Methylosinus trichosporium OB3b. Journal of General Microbiology 130(10): 2565-2575.
Costa, C., M. Vecherskaya, C. Dijkema, and A. J. Stams. 2001. The effect of oxygen on methanol oxidation by an obligate methanotrophic bacterium studied by in vivo 13C nuclear magnetic resonance spectroscopy. J Ind Microbiol Biotechnol 26(1- 2): 9-14.
CRC Handbook of Chemistry and Physics. 1995. Edited by D. R. Lide and H. P. R. Frederikse. 76th Edition ed. Boca Raton, FL: CRC Press, Inc.
Dale, J. M., L. Popescu, and P. D. Karp. 2010. Machine learning methods for metabolic pathway prediction. Bmc Bioinformatics 11.
Dalton, H. and D. I. Stirling. 1982. Co-Metabolism. Philosophical Transactions of the Royal Society of London Series B-Biological Sciences 297(1088): 481-496.
Debont, J. A. M. and E. G. Mulder. 1974. Nitrogen-Fixation and Co-Oxidation of Ethylene by a Methane-Utilizing Bacterium. Journal of General Microbiology 83(Jul): 113-121.
64 Dedysh, S. N., P. Ricke, and W. Liesack. 2004. NifH and NifD phylogenies: an evolutionary basis for understanding nitrogen fixation capabilities of methanotrophic bacteria. Microbiology 150(Pt 5): 1301-1313.
DiGregorio, B. E. 2009. Biobased performance bioplastic: Mirel. Chem Biol 16(1): 1-2.
Doi, Y. 1990. Microbial polyesters. New York, N.Y.: VCH.
Dove, A. 2000. Experts disagree over color of biomass. Nature Biotechnology 18(5): 490.
Dumont, M. G. and J. C. Murrell. 2005. Community-level analysis: key genes of aerobic methane oxidation. Methods Enzymol 397: 413-427.
Dunfield, P. F. and R. Conrad. 2000. Starvation Alters the Apparent Half-Saturation Constant for Methane in the Type II MethanotrophMethylocystis Strain LR1. Applied and Environmental Microbiology 66(9): 4136-4138.
Dworkin, M. and S. Falkow. 2006. The prokaryotes : a handbook on the biology of bacteria. 3rd ed. New York ; [London]: Springer.
EPA, U. 2012. Plastics: Just the Facts. Accessed March 11, 2012.
Ferenci, T., T. Strom, and J. R. Quayle. 1975. Oxidation of Carbon-Monoxide and Methane by Pseudomonas-Methanica. Journal of General Microbiology 91(Nov): 79-91.
Fidler, S. and D. Dennis. 1992. Polyhydroxyalkanoate production in recombinant Escherichia coli. FEMS Microbiol Rev 9(2-4): 231-235.
FranklinAssociates. 2007. Cradle-To-Gate Life Cycle Inventory of Nine Plastic Resins and Two Polyurethane Precursors.
Frosch, R. A. and N. E. Gallopoulos. 1989. Strategies for Manufacturing. Scientific American 261(3): 144-152.
65 Gerngross, T. U. 1999. Can biotechnology move us toward a sustainable society? Nature Biotechnology 17(6): 541-544.
Gerngross, T. U. and S. C. Slater. 2000. How green are green plastics? Scientific American 283(2): 36-41.
Graedel, T. E. and B. R. Allenby. 2003. Industrial ecology. 2nd ed, Prentice-Hall international series in industrial and systems engineering. Upper Saddle River, N.J.: Prentice Hall.
Graedel, T. E., B. R. Allenby, and American Telephone and Telegraph Company. 1996. Design for environment. Upper Saddle River, N.J.: Prentice Hall.
Graham, D. W., J. A. Chaudhary, R. S. Hanson, and R. G. Arnold. 1993. Factors Affecting Competition between Type-I and Type-Ii Methanotrophs in 2- Organism, Continuous-Flow Reactors. Microbial Ecology 25(1): 1-17.
Greene, D. L., J. L. Hopson, and J. Li. 2006. Have we run out of oil yet? Oil peaking analysis from an optimist's perspective. Energy Policy 34(5): 515-531.
Griffin, G. J. L. 1994. Chemistry and technology of biodegradable polymers. 1st ed. London ; New York: Blackie Academic & Professional.
Hakemian, A. S. and A. C. Rosenzweig. 2007. The Biochemistry of Methane Oxidation. Annual Review of Biochemistry 76(1): 223-241.
Han, B., T. Su, H. Wu, Z. Gou, X. H. Xing, H. Jiang, Y. Chen, X. Li, and J. C. Murrell. 2009. Paraffin oil as a "methane vector" for rapid and high cell density cultivation of Methylosinus trichosporium OB3b. Appl Microbiol Biotechnol 83(4): 669-677.
Hanson, R. S. and T. E. Hanson. 1996. Methanotrophic bacteria. Microbiol Rev 60(2): 439-471.
Harding, K. G., J. S. Dennis, H. von Blottnitz, and S. T. Harrison. 2007. Environmental analysis of plastic production processes: comparing petroleum-based 66 polypropylene and polyethylene with biologically-based poly-beta- hydroxybutyric acid using life cycle analysis. J Biotechnol 130(1): 57-66.
Heijnen, J. J. and J. A. Roels. 1981. A Macroscopic Model Describing Yield and Maintenance Relationships in Aerobic Fermentation Processes. Biotechnology and Bioengineering 23(4): 739-763.
Helm, J., K. D. Wendlandt, G. Rogge, and U. Kappelmeyer. 2006. Characterizing a stable methane-utilizing mixed culture used in the synthesis of a high-quality biopolymer in an open system. Journal of Applied Microbiology 101(2): 387-395.
Helm, J., K. D. Wendlandt, M. Jechorek, and U. Stottmeister. 2008. Potassium deficiency results in accumulation of ultra-high molecular weight poly-beta-hydroxybutyrate in a methane-utilizing mixed culture. Journal of Applied Microbiology 105(4): 1054-1061.
IPCC. 2007. Climate change 2007: the physical science basis. Contribution of working group 1 to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. New York:
Jacquel, N., C. W. Lo, Y. H. Wei, H. S. Wu, and S. S. Wang. 2008. Isolation and purification of bacterial poly (3-hydroxyalkanoates). Biochemical Engineering Journal 39(1): 15-27.
Jimenez-Gonzalez, C., S. Kim, and M. R. Overcash. 2000. Methodology for Developing Gate-to-Gate Life Cycle Inventory Information. Int J LCA 5(3): 153-159.
Johnson, K., Y. Jiang, R. Kleerebezem, G. Muyzer, and M. C. M. van Loosdrecht. 2009a. Enrichment of a Mixed Bacterial Culture with a High Polyhydroxyalkanoate Storage Capacity. Biomacromolecules 10(4): 670-676.
Johnson, K., Y. Jiang, R. Kleerebezem, G. Muyzer, and M. C. van Loosdrecht. 2009b. Enrichment of a mixed bacterial culture with a high polyhydroxyalkanoate storage capacity. Biomacromolecules 10(4): 670-676. 67 Jung, I. L., K. H. Phyo, K. C. Kim, H. K. Park, and I. G. Kim. 2005. Spontaneous liberation of intracellular polyhydroxybutyrate granules in Escherichia coli. Res Microbiol 156(8): 865-873.
Karp, P. D., S. Paley, and P. Romero. 2002. The Pathway Tools Software. Bioinformatics 18: S225-S232.
Karp, P. D., R. Caspi, H. Foerster, C. A. Fulcher, R. Hopkinson, J. Ingraham, P. Kaipa, M. Krummenacker, S. Paley, J. Pick, S. Y. Rhee, C. Tissier, and P. F. Zhang. 2006. MetaCyc: a multiorganism database of metabolic pathways and enzymes. Nucleic Acids Research 34: D511-D516.
Karp, P. D., I. M. Keseler, C. Bonavides-Martinez, J. Collado-Vides, S. Gama-Castro, R. P. Gunsalus, D. A. Johnson, M. Krummenacker, L. M. Nolan, S. Paley, I. T. Paulsen, M. Peralta-Gil, A. Santos-Zavaleta, and A. G. Shearer. 2009. EcoCyc: A comprehensive view of Escherichia coli biology. Nucleic Acids Research 37: D464-D470.
Kauffman, K. J., P. Prakash, and J. S. Edwards. 2003. Advances in flux balance analysis. Current Opinion in Biotechnology 14(5): 491-496.
Kemmer, G. and S. Keller. 2010. Nonlinear least-squares data fitting in Excel spreadsheets. Nature Protocols 5(2): 267-281.
Keseler, I. M., J. Collado-Vides, A. Santos-Zavaleta, M. Peralta-Gil, S. Gama-Castro, L. Muniz-Rascado, C. Bonavides-Martinez, S. Paley, M. Krummenacker, T. Altman, P. Kaipa, A. Spaulding, J. Pacheco, M. Latendresse, C. Fulcher, M. Sarker, A. G. Shearer, A. Mackie, I. Paulsen, R. P. Gunsalus, and P. D. Karp. 2011. EcoCyc: a comprehensive database of Escherichia coli biology. Nucleic Acids Research 39: D583-D590.
Keshavarz, T. and I. Roy. 2010. Polyhydroxyalkanoates: bioplastics with a green agenda. Curr Opin Microbiol 13(3): 321-326.
68 Kim, S. and B. E. Dale. 2005. Life cycle assessment study of biopolymers (Polyhydroxyalkanoates) derived from no-tilled corn. International Journal of Life Cycle Assessment 10(3): 200-210.
Kim, S. and B. E. Dale. 2008. Energy and greenhouse gas profiles of polyhydroxybutyrates derived from corn grain: a life cycle perspective. Environ Sci Technol 42(20): 7690-7695.
Kleerebezem, R. and M. C. van Loosdrecht. 2007. Mixed culture biotechnology for bioenergy production. Curr Opin Biotechnol 18(3): 207-212.
Krieger, C. J., P. F. Zhang, L. A. Mueller, A. Wang, S. Paley, M. Arnaud, J. Pick, S. Y. Rhee, and P. D. Karp. 2004. MetaCyc: a multiorganism database of metabolic pathways and enzymes. Nucleic Acids Research 32: D438-D442.
Kurdikar, D., L. Fournet, S. C. Slater, M. Paster, K. J. Gruys, T. U. Gerngross, and R. Coulon. 2000. Greenhouse Gas Profile of a Plastic Material Derived from a Genetically Modified Plant. Journal of Industrial Ecology 4(3): 107-122.
Lange, H. C., M. Eman, G. van Zuijlen, D. Visser, J. C. van Dam, J. Frank, M. J. T. de Mattos, and J. J. Heijnen. 2001. Improved rapid sampling for in vivo kinetics of intracellular metabolites in Saccharomyces cerevisiae. Biotechnology and Bioengineering 75(4): 406-415.
Latendresse, M., M. Krummenacker, M. Trupp, and P. D. Karp. 2012. Construction and completion of flux balance models from pathway databases. Bioinformatics 28(3): 388-396.
Law, K. L., S. Moret-Ferguson, N. A. Maximenko, G. Proskurowski, E. E. Peacock, J. Hafner, and C. M. Reddy. 2010. Plastic accumulation in the North Atlantic subtropical gyre. Science 329(5996): 1185-1188.
69 Leak, D. J. and H. Dalton. 1983. In vivo Studies of Primary Alcohols, Aldehydes and Carboxylic Acids as Electron Donors for the Methane Mono-oxygenase in a Variety of Methanotrophs. Journal of General Microbiology 129(11): 3487-3497.
Leak, D. J. and H. Dalton. 1986. Growth Yields of Methanotrophs .2. A Theoretical- Analysis. Applied Microbiology and Biotechnology 23(6): 477-481.
Lee, S. Y. 1996a. Plastic bacteria? Progress and prospects for polyhydroxyalkanoate production in bacteria. Trends in Biotechnology 14(11): 431-438.
Lee, S. Y. 1996b. Bacterial polyhydroxyalkanoates. Biotechnology and Bioengineering 49(1): 1-14.
Lidstrom, M. 2006. Aerobic Methylotrophic Prokaryotes. In The Prokaryotes, edited by M. Dworkin, et al.: Springer New York.
Lieberman, R. L. and A. C. Rosenzweig. 2004. Biological Methane Oxidation: Regulation, Biochemistry, and Active Site Structure of Particulate Methane Monooxygenase. Critical Reviews in Biochemistry and Molecular Biology 39: 147-164.
Lieberman, R. L. and A. C. Rosenzweig. 2005. Crystal structure of a membrane-bound metalloenzyme that catalyses the biological oxidation of methane. Nature 434(7030): 177-182.
Lifset, R. 2002. Closing the Loop and Honing Our Tools. Journal of Industrial Ecology 5(4): 1-2.
Ling, Y., H. H. Wong, C. J. Thomas, D. R. Williams, and A. P. Middelberg. 1997. Pilot- scale extraction of PHB from recombinant E. coli by homogenization and centrifugation. Bioseparation 7(1): 9-15.
70 Listewnik, H. F., K. D. Wendlandt, M. Jechorek, and G. Mirschel. 2007. Process Design for the Microbial Synthesis of Poly-B-hydroxybutyrate (PHB) from Natural Gas. Engineering in Life Sciences 7(3): 278-282.
Mancinelli, R. L. 1995. The Regulation of Methane Oxidation in Soil. Annual Review of Microbiology 49: 581-605.
Matthews, E. and N. Themelis. 2007. Potential for Reducing Global Methane Emissions from Landfills 2000-2030. Paper presented at International Landfill Symposium, Sardinia, Italy.
May, P., J. O. Christian, S. Kempa, and D. Walther. 2009. ChlamyCyc: an integrative systems biology database and web-portal for Chlamydomonas reinhardtii. Bmc Genomics 10.
Moret-Ferguson, S., K. L. Law, G. Proskurowski, E. K. Murphy, E. E. Peacock, and C. M. Reddy. 2010. The size, mass, and composition of plastic debris in the western North Atlantic Ocean. Mar Pollut Bull 60(10): 1873-1878.
Morse, M.-C. 2009. Anaerobic Biodegradation of Biocomposites for the Building Industry. Dissertation thesis, Civil & Environmental Engineering, Stanford University, Stanford, CA.
Mueller, L. A., P. F. Zhang, and S. Y. Rhee. 2003. AraCyc: A biochemical pathway database for Arabidopsis. Plant Physiology 132(2): 453-460.
Murrell, J. C. 2010. The Aerobic Methane Oxidizing Bacteria (Methanotrophs). In Handbook of Hydrocarbon and Lipid Microbiology, edited by K. N. Timmis: Springer Berlin Heidelberg.
Murrell, J. C. and H. Dalton. 1983a. Ammonia Assimilation in Methylococcus- Capsulatus (Bath) and Other Obligate Methanotrophs. Journal of General Microbiology 129(Apr): 1197-1206.
71 Murrell, J. C. and H. Dalton. 1983b. Nitrogen-Fixation in Obligate Methanotrophs. Journal of General Microbiology 129(Nov): 3481-3486.
Murrell, J. C. and M. S. M. Jetten. 2009. The microbial methane cycle. Environmental Microbiology Reports 1(5): 279-284.
Nonato, R. V., P. E. Mantelatto, and C. E. V. Rossell. 2001. Integrated production of biodegradable plastic, sugar and ethanol. Applied Microbiology and Biotechnology 57(1-2): 1-5.
Nzihou, A. 2010. Toward the Valorization of Waste and Biomass. Waste Biomass Valorization 1: 3-7.
Nzihou, A. and R. Lifset. 2010. Waste Valorization, Loop-Closing, and Industrial Ecology. Journal of Industrial Ecology 14(2): 196-199.
Oakley, C. J. and J. C. Murrell. 1988. Nifh Genes in the Obligate Methane Oxidizing Bacteria. Fems Microbiology Letters 49(1): 53-57.
Oldenhuis, R., J. Y. Oedzes, J. J. Vanderwaarde, and D. B. Janssen. 1991. Kinetics of Chlorinated-Hydrocarbon Degradation by Methylosinus-Trichosporium Ob3b and Toxicity of Trichloroethylene. Applied and Environmental Microbiology 57(1): 7- 14.
Orth, J. D., I. Thiele, and B. O. Palsson. 2010. What is flux balance analysis? Nature Biotechnology 28(3): 245-248.
Paley, S. M. and P. D. Karp. 2002. Evaluation of computational metabolic-pathway predictions for Helicobacter pylori. Bioinformatics 18(5): 715-724.
Patel, M., M. Crank, V. Domburg, B. Hermann, L. Roses, B. Husing, L. Overbeek, F. Terragni, and E. Recchia. 2006. Medium and Long-term Opportunities and Risks of the Biotechnological Production of Bulk Chemicals from Renewable Resources - The Potential of White Biotechnology The BREW Project. Utrecht:
72 Pfluger, A. R. 2010. Selective Growth of Type II Methanotrophic Bacteria in a Fluidized Bed Reactorthesis, Civil and Environmental Engineering, Stanford University, Stanford.
Pfluger, A. R., W. Wu, A. J. Pieja, J. Wan, K. H. Rostkowski, and C. Criddle, S. 2011. Selection of type I and type II methanotrophic proteobacteria in a fluidized bed reactor under non-sterile conditions. Bioresource Technology.
Philip, S., T. Keshavarz, and I. Roy. 2007. Polyhydroxyalkanoates: biodegradable polymers with a range of applications. Journal of Chemical Technology and Biotechnology 82: 233-247.
Pieja, A. J., E. Sundstrom, and C. S. Criddle. 2011a. Poly-3-hydroxybutyrate metabolism in the Type II methanotroph Methylocystis parvus OBBP.
Pieja, A. J., K. H. Rostkowski, and C. S. Criddle. 2011b. Distribution and Selection of Poly-3-hydroxybutyrate Production Capacity Among Methanotrophic Bacteria. Microbial Ecology.
Pieja, A. J., E. R. Sundstrom, and C. S. I. s. Criddle. 2011c. Evaluation of poly-3- hydroxybutyrate production capacity in mixed cultures of methanotrophic bacteria in cyclically stressed environments. In Submission.
Rebitzer, G., T. Ekvall, R. Frischknecht, D. Hunkeler, G. Norris, T. Rydberg, W. P. Schmidt, S. Suh, B. P. Weidema, and D. W. Pennington. 2004. Life cycle assessment Part 1: Framework, goal and scope definition, inventory analysis, and applications. Environment International 30(5): 701-720.
Reddy, C. S., R. Ghai, Rashmi, and V. C. Kalia. 2003. Polyhydroxyalkanoates: an overview. Bioresour Technol 87(2): 137-146.
Rittmann, B. E. and P. L. McCarty. 2001. Environmental biotechnology : principles and applications, McGraw-Hill series in water resources and environmental engineering. Boston: McGraw-Hill. 73 Runge, C. F. and B. Senauer. May - Jun., 2007. How Biofuels Could Starve the Poor. Foreign Affairs 86(3): 41-53.
Ryan, C. 2008. Climate change and ecodesign - Part I: The focus shifts to systems. Journal of Industrial Ecology 12(2): 140-143.
Schweigkofler, M. and R. Niessner. 2001. Removal of siloxanes in biogases. Journal of Hazardous Materials 83(3): 183-196.
Semrau, J., A. Chistoserdov, J. Lebron, A. Costello, J. Davagnino, E. Kenna, A. Holmes, R. Finch, J. Murrell, and M. Lidstrom. 1995. Particulate methane monooxygenase genes in methanotrophs. J. Bacteriol. 177(11): 3071-3079.
Shah, N. N., M. L. Hanna, K. J. Jackson, and R. T. Taylor. 1995. Batch cultivation of Methylosinus trichosporium OB3B: IV. Production of hydrogen-driven soluble or particulate methane monooxygenase activity. Biotechnology and Bioengineering 45(3): 229-238.
Shen, L., J. Haufe, and M. K. Patel. 2009. Personal Communication with Shen, L., J. Haufe, and M. K. Patel, Product overview and market projection of emerging bio- based plastics. Utrecht 2009.
Shishkina, V. N. and Y. A. Trotsenko. 1979. Pathways of Ammonia Assimilation in Obligate Methane Utilizers. Fems Microbiology Letters 5(3): 187-191.
Smith, T. J., Y. A. Trotsenko, and J. C. Murrell. 2010. Physiology and Biochemistry of the Aerobic Methane Oxidizing Bacteria. In Handbook of Hydrocarbon and Lipid Microbiology, edited by K. N. Timmis: Springer Berlin Heidelberg.
Solomon, S., Intergovernmental Panel on Climate Change., and Intergovernmental Panel on Climate Change. Working Group I. 2007. Climate change 2007 : the physical science basis : contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge ; New York: Cambridge University Press. 74 Staley, B. F. and M. A. Barlaz. 2009. Composition of Municipal Solid Waste in the United States and Implications for Carbon Sequestration and Methane Yield. Journal of Environmental Engineering-Asce 135(10): 901-909.
Stein, L. Y., S. Yoon, J. D. Semrau, A. A. Dispirito, A. Crombie, J. C. Murrell, S. Vuilleumier, M. G. Kalyuzhnaya, H. J. Op den Camp, F. Bringel, D. Bruce, J. F. Cheng, A. Copeland, L. Goodwin, S. Han, L. Hauser, M. S. Jetten, A. Lajus, M. L. Land, A. Lapidus, S. Lucas, C. Medigue, S. Pitluck, T. Woyke, A. Zeytun, and M. G. Klotz. 2010. Genome sequence of the obligate methanotroph Methylosinus trichosporium strain OB3b. J Bacteriol 192(24): 6497-6498.
Steinbuchel, A. and H. E. Valentin. 1995. Diversity of Bacterial Polyhydroxyalkanoic Acids. Fems Microbiology Letters 128(3): 219-228.
Steinbuchel, A. and B. Fuchtenbusch. 1998. Bacterial and other biological systems for polyester production. Trends Biotechnol 16(10): 419-427.
Sumner, L. W. and E. Urbanczyk-Wochniak. 2007. MedicCyc: a biochemical pathway database for Medicago truncatula. Bioinformatics 23(11): 1418-1423.
Tabone, M. D., J. J. Cregg, E. J. Beckman, and A. E. Landis. 2010. Sustainability Metrics: Life Cycle Assessment and Green Design in Polymers. Environmental Science & Technology 44(21): 8264-8269.
Takeda, K. 1988. Characteristics of a Nitrogen-Fixing Methanotroph, Methylocystis T-1. Antonie Van Leeuwenhoek Journal of Microbiology 54(6): 521-534.
Tenenbaum, D. J. 2008. Food vs. fuel: diversion of crops could cause more hunger. Environ Health Perspect 116(6): A254-257.
Thiele, I. and B. O. Palsson. 2010. A protocol for generating a high-quality genome-scale metabolic reconstruction. Nature Protocols 5(1): 93-121.
75 Titus, J. A., W. M. Reed, R. M. Pfister, and P. R. Dugan. 1982. Exospore formation in Methylosinus trichosporium. J Bacteriol 149(1): 354-360.
Tonge, G. M., C. J. Knowles, D. E. F. Harrison, and I. J. Higgins. 1975. Metabolism of one carbon compouds. Cytochromes of methane- and methanol-utilizing bacteria. FEBS Letters 44: 106-110.
Trotsenko, Y. A. 1983. Metabolic Features of Methane-Utilizing and Methanol-Utilizing Bacteria. Acta Biotechnologica 3(3): 269-277.
Trotsenko, Y. A. and J. C. Murrell. 2008. Metabolic Aspects of Aerobic Obligate Methanotrophy. In Advanced Applied Microbiology.
Trotsenko, Y. A., N. V. Doronina, and V. N. Khmelenina. 2005. Biotechnological potential of aerobic methylotrophic bacteria: A review of current state and future prospects. Applied Biochemistry and Microbiology 41(5): 433-441.
Uchino, K., T. Saito, B. Gebauer, and D. Jendrossek. 2007. Isolated poly(3- hydroxybutyrate) (PHB) granules are complex bacterial organelles catalyzing formation of PHB from acetyl coenzyme A (CoA) and degradation of PHB to acetyl-CoA. Journal of Bacteriology 189(22): 8250-8256.
USEPA. 2006. Global Anthropogenic Non-CO2 Greenhouse Gas Emissions: 1990 – 2020, edited by U. S. E. P. Agency. Washington, DC.
Vecherskaya, M., C. Dijkema, and A. J. Stams. 2001. Intracellular PHB conversion in a Type II methanotroph studied by 13C NMR. J Ind Microbiol Biotechnol 26(1-2): 15-21.
Vecherskaya, M., C. Dijkema, H. Ramirez Saad, and A. J. Stams. 2009. Microaerobic and anaerobic metabolism of a Methylocystis parvus strain isolated from a denitrifying bioreactor. Environmental Microbiology Reports 1(5): 442-449.
76 Vink, E. T. H., K. R. Rabago, D. A. Glassner, and P. R. Gruber. 2003. Applications of life cycle assessment to NatureWorks (TM) polylactide (PLA) production. Polymer Degradation and Stability 80(3): 403-419.
Volova, T. G. 2004. Polyhydroxyalkanoates--plastic materials of the 21st century : production, properties, applications. Hauppauge, N.Y.: Nova Science Publishers.
Vorob'ev, A. V. and S. N. Dedysh. 2008. [Use of enrichment cultures for assessing the structure of methanotrophic communities in peat soil: the problem of representativity of the results]. Mikrobiologiia 77(4): 566-569.
Walsh, B. 2010. The Perils of Plastic. TIME, April 12, 2010, 44-54.
Wendlandt, K. D., M. Jechorek, J. Helm, and U. Stottmeister. 1998. Production of PHB with a high molecular mass from methane. Polymer Degradation and Stability 59(1-3): 191-194.
Wendlandt, K. D., M. Jechorek, J. Helm, and U. Stottmeister. 2001. Producing poly-3- hydroxybutyrate with a high molecular mass from methane. Journal of Biotechnology 86(2): 127-133.
Wendlandt, K. D., W. Geyer, G. Mirschel, and F. A. Hemidi. 2005. Possibilities for controlling a PHB accumulation process using various analytical methods. Journal of Biotechnology 117(1): 119-129.
Wendlandt, K. D., U. Stottmeister, J. Helm, B. Soltmann, M. Jechorek, and M. Beck. 2010. The potential of methane-oxidizing bacteria for applications in environmental biotechnology. Engineering in Life Sciences 10(2): 87-102.
Whittenb.R, K. C. Phillips, and Wilkinso.Jf. 1970. Enrichment, Isolation and Some Properties of Methane-Utilizing Bacteria. Journal of General Microbiology 61: 205-&.
77 Whittenbury, R., K. C. Phillips, and Wilkinso.Jf. 1970a. Enrichment, Isolation and Some Properties of Methane-Utilizing Bacteria. Journal of General Microbiology 61: 205-&.
Whittenbury, R., S. L. Daview, and J. Davey. 1970b. Exospores and Cysts Formed by Methane-utilizing Bacteria. Journal of General Microbiology 61: 219-226.
Yamane, T. 1992. Cultivation engineering of microbial bioplastics production. FEMS Microbiology Reviews 103: 257-264.
Yamane, T. 1993. Yield of poly-D(-)-3-hydroxybutyrate from various carbon sources: a theoretical study. Biotechnology and Bioengineering 41(1): 165-170.
Yu, J. and L. X. Chen. 2006. Cost-effective recovery and purification of polyhydroxyalkanoates by selective dissolution of cell mass. Biotechnol Prog 22(2): 547-553.
Zhang, P. F., H. Foerster, C. P. Tissier, L. Mueller, S. Paley, P. D. Karp, and S. Y. Rhee. 2005. MetaCyc and AraCyc. Metabolic pathway databases for plant research. Plant Physiology 138(1): 27-37.
Zhang, Y. X., J. Y. Xin, L. L. Chen, H. Song, and C. U. Xia. 2008. Biosynthesis of poly- 3-hydroxybutyrate with a high molecular weight by methanotroph from methane and methanol. Journal of Natural Gas Chemistry 17(1): 103-109.
78