G. Charles Dismukes
Department of Chemistry & Chemical Biology
Waksman Institute. Room1014A
Piscataway, NJ 08854-0759
Microbial metabolism and renewable energy production
Food vs. Biomass Energy Feedstocks: Biomass energy feedstocks currently in commercial use or projected for the near future nearly all come from terrestrial sources that compete with agricultural products (corn, seed plants, switchgrass, etc.). In the US there has been a redirection of our agricultural subsidy program from food to energy which arises from an historical (entrenched) government program –the Farm Bill, currently $150 Billion/year - which was originally created to protect consumers and farmers. Agricultural food products and biomass energy products are not the same and efforts to force acceptance of food biomass as energy biomass are ill advised and economically more costly. There are sound scientific and technological reasons for developing optimal biomass energy crops, as noted below. Equally important for the American consumer, it makes no economic sense for the government to make the critical choices about how to manage our emerging biomass energy crops using the old ideas for managing our terrestrial food crops. These choices are often incompatible.
To mitigate many of the potentially deleterious environmental and agricultural consequences associated with current landbased-biofuel feedstocks, we advocate the use of biofuels derived from aquatic microbial oxygenic photoautotrophs (AMOPs), more commonly known as algae, cyanobacteria and diatoms. AMOPs are buoyant phototrophs that consume CO2 and make proteins, lipids and energy storage molecules, but are devoid of structural biopolymers found in higher plants that are so recalcitrant to digestion. Compared to terrestrial crops, AMOPs are inherently more efficient solar collectors, use less or no land, can be converted to liquid fuels using simpler technologies than cellulose, and offer secondary uses that fossil fuels do not provide. AMOPs pose a different set of technological challenges if they are to contribute as biofuel feedstocks. Our research on AMOPs addresses: 1) their demonstrated productivity in mass culturing and future potential as biomass energy crops, 2) use as cell factories for production of gaseous fuels (H2, CH4), 3) fundamental photosynthetic physiology and mechanisms, 4) genetic transformants for understanding mechanisms and improving fuel production.
• Biofuel Production via Microbial Cell Factories
Algae and cyanobacteria have flexible genomes that allow expression of different gene products capable of controlling the product distribution of carbon byproducts that form during catabolism of energy storage molecules. This flexibility is controlled at the genome level through genus/strain specific metabolic networks and externally by physical (environmental) stresses and chemical (nutrient) selection. Our research seeks to identify and characterize native strains taken from the field or culture collections that are optimally suited for biofuel production. This research involves the dissection of the metabolic networks through the application of systems biological approaches. We quantify the metabolite pools excreted from and within cells, using both homebuilt tools and commercial state-of-the-art tools. Key collaborations with other principal investigators on campus support this effort, including: Joshua Rabinowitz (Genomics), Herschel Rabitz (Chemistry). The following projects rely on this same general approach.
“Renewable Biosolar Hydrogen Production from Robust Oxygenic Phototrophs” which is highlighted on the project site.
“Quantitative Tools for Dissection of Hydrogen-Producing Metabolic Networks"
With an eye towards eventually enabling rational optimization of microbial H2 production, here we aim to develop novel tools for quantitative dissection of H2-producing metabolic networks. These tools aim to bring innovations both in high-throughput experimental metabolomics and in algorithms for predictive modeling/analysis of experiments.
1: Develop quantitative experimental tools for simultaneous measurements of multiple metabolite concentrations and fluxes;
2: Map the organisms’ dynamic metabolic responses to changing environmental conditions;
3: Develop predictive models of H2-metabolism. Organisms: Clostridium acetobutylicum (possessing the fastest and highest yielding hexose fermentation pathway to H2 of any microbe yet reported), and a thermophilic cyanobacterium utilizing a nitrogenase-dependent pathway to H2.
• Cofermentation for Enhanced Biomass Conversion Efficiency
The current best autofermentation conversion yields to H2 produced from intracellular energy reserves (mainly carbohydrates) by AMOPs are limited to about 4 H2 out of a potential 12 maximum per glucose equivalent. The conversion of AMOPs via a two-stage fermentation using nonphotosynthetic anaerobic microorganisms is well suited to utilize this feedstock for the co-production of H2 and CH4. This process involves a two-stage fermentation using acidogenic bacteria and metanogenic archea. Energy Conversion Efficiencies equal to 60% of the total enthalpy content of the fermented biomass should be achievable based on composition. This ECE exceeds current commercial conversion yields for cellulosic ethanol and could provide a more practical route for commercial applications.
In a related approach we are investigating methods for the cofermentation of AMOP feedstocks in which the organic acids produced during fermentative hydrogen generation are selectively utilized by a methanogen for the anaerobic production of methane. In this method, an increase in the yield and rate of hydrogen formation during cofermentation occurs by the principle of mass action which prevents the hydrogen uptake reaction. This approach requires the cocultivation of hydrogen-evolving bacteria and hydrogen-tolerant methanogen. By recovering the energy of organic acid fermentative products through co-production of hydrogen and methane, this overall process is predicted to have a high ECE.
Biosolar Fuels Production:
(AFOSR, DOE-GTL). Our key achievements thus far include:
1) development of the strongest fermentative H2 producing cyanobacterium and conditions for "milking" to enhance
the yield and rate of H2 production,
2) construction of genetic knock-outs of metabolic pathways with improved fuel production,
3) metabolomics (LC-tandem-MS and cryoprobe-assisted NMR methods) for metabolic pathway elucidation from cellular metabolites,
4) acceleration of fermentation via environmental stresses,
5) optimization of yield via photoautotrophic growth/fermentation conditioning,
6) strain selection & bioprospecting.
All oxygenic phototrophs extract electrons and protons from water and use them to reduce NADP+ and plastoquinone for use as energy sources for metabolism such as CO2 fixation via the Calvin cycle and other pathways. However, some microbial oxygenic phototrophs (cyanobacteria and microalgae) can transiently produce H2 gas under anaerobic conditions via proton reduction catalyzed by a hydrogenase in competition with other intracellular processes. They do so by redirecting the electrons and protons obtained from reoxidation of carbohydrate storage intermediates at the level of ferredoxin/NADPH into hydrogenase. The carbohydrate intermediate is ultimately produced from CO2, water and sunlight during photosynthesis. Other phototrophs such as green algae and diatoms store a greater fraction of their carbon fixation intermediates as neutral lipids and thus are better candidates for hydrocarbon or biodiesel production.
Biohydrogen. We are engaged in metabolic engineering to control the flux of electrons and protons into H2 and lipid production in genetically tractable strains of cyanobacteria, microalgae and diatoms. Our research has identified the metabolic pathways for intracellular reductants and protons to produce H2 within cyanobacteria. Cyanobacteria use a pathway that is distinct from that used by the microalgae, thus affording new opportunities for enhanced solar H2 production from water. This distinction enables temporal separation of H2 from photosynthetic O2 production, a requirement for applications in energy generation. Large increases in H2 production have been demonstrated using selective environmental stresses (osmotic pressure, ion exchange, selective nutrient deprivation -nitrate, carbonate and phosphate) and biosynthesis optimization (Ni2+ loading) that have practical utility (Ananyev, et al 2008); Carrieri et al. 2008.
Metabolic Engineering. Experimental observations of intracellular reductant accumulation (fluorescence of NADH) have shown that the availability of reducing equivalents (NADH) produced during fermentation of carbohydrate storage molecules (glycogen and osmolytes) limits H2 production capacity in some cyanobacteria (Ananyev et al., 2008). We therefore initiated studies of fermentative metabolite production in aquatic phototrophs (Carrieri et al, 2008; 2009). To test our hypotheses we have created genetic knock-outs of pyruvate metabolism by antibiotic resistance cartridge mutagenesis; specifically the genes for lactate dehydrogenase and pyruvate-ferredoxin oxidoreductase in Synechococcus 7002 (McNeely, et al. 2009) These mutant strains have substantially increased H2 production (up to 5 fold). This genetic approach enables the design of tailored made microbes suitable for improved H2 production and can be applied to other fuel precursors. Extension of this genetic approach to engineering of other cyanobacteria with more robust H2 production capacity is underway.
GeneticTransformation. My lab is working to transform the filamentous strain Arthrospira (Spirulina) maxima,the most vigorous H2 producing cyanobacterium. The whole genome was recently sequenced at DOE-JGI based on a grant proposal I submitted with our collaborator Donald Bryant. The strategy that my lab is exploring in collaboration with Oliver Lenz (Humboldt University) is to inactivate the natural restriction enzyme systems in this organism which serve to digest foreign DNA taken up from exogenous sources. The Arthrospira genome contains 11 different restriction systems, including Spal, SpdII, SpaIII and SpaIV, which are isoschizomers of Tth111I, PvuI, PvuII and HindIII (Tagut et al., J.Appl Phycol. 1995, 7, 561-4; Zhao Physiol Genomics 24:181-190, 2006). To protect native DNA from self-digestion the cell uses methylation of selective sites to block the restriction enzymes. The CpG methyltransferase, M.SssI, methylates all cytosine residues (C5) within the double-stranded dinucleotide recognition sequence 5'...CG...3'. This enzyme is available as a commercial kit and is being used to methylate four plasmids that we have constructed for modification of the hox H subunit of hydrogenase. Transformation of the genome using these protected plasmids is underway.
Lipds/Biodiesel. Recently we began working on a new project to examine green microalgae and diatoms as more efficient sources for accumulating neutral lipids as primary energy storage molecules. As part of our bioprospecting efforts, we have discovered several strong lipid producing strains of diatoms from acidic thermal sources which are promising as biodiesel precursors (Figure).
Figure. False-colored image of diatom cell isolated from Norris Hot Spring (YNP): left: bright field microscopy, length 35 mm; right: (red) Chlorophyll fluorescence in the range 670-690 nm, and (green) Nile Red fluorescence in the range 570-620 nm Green light (520-550 nm) used for fluorescence excitation of Chl and Nile Red.
This bioprospecting effort is complemented by collaborative studies with the Falkowski lab (Rutgers Univ.) of lipid metabolism using two marine diatoms strains with sequenced genomes that are genetically transformable: Thalassiosira pseudonana and Phaeodactylum tricornutum. Our goal is to elucidate the pathways for lipid production using the special tools we have developed coupled with knock-out mutagenesis to test hypotheses for metabolic pathways involved. For example, using our ultrasensitive H2 electrode we recently learned that both diatoms exhibit dark anaerobic H2 production (autofermentation) and photo-induced H2. Tp is predicted to have the hydA gene coding for an FeFe-hydrogenase homologous to the Thermatoga enzyme, but no hyd maturase genes. Thus, this appears to be a unique construct capable of assembly even without maturases. Pt is predicted to have no hydrogenase genes at all (no hyd structural genes nor hyd maturases!, but does have a NARF like sequence encoding a type of ironsulfur cluster. Neither possesses nitrogenase genes. The latter discovery suggests that we have identified a new H2 evolving system not previously known in biology.
Bioprospecting for Water-Splitting Enzymes: The Search for Photosynthetic "Weirdophiles": (NIH, Dreyfus, AFOSR). We are searching for novel oxygenic phototrophs that split water using unconventional mechanisms distinct from that typified by terrestrial plants. We call these "weirdophiles" to distinguish them from most extremophiles which merely use tuned versions of conventional enzymes and standard mechanisms. Our first "weirdophile" is an alkalophillic cyanobacterium isolated from alkaline soda lakes of E. Africa (pH >10 carbonate ~0.4 M). Using homebuilt tools for in vivo measurements of PSII activity we have discovered that cells of Arthrospira m. have a 5x faster turnover rate of the O2-evolving complex (OEC) compared to all other phototrophs examined thus far (Ananyev and Dismukes, 2005). This organism uses carbonate directly in the OEC for water splitting, possibly as a proton acceptor (Carrieri, et al., 2007). Other unusual aquatic phototrophs are under study which may have the potential to replace carbonate by borate.
Instrumentation Development for Renewable Energy Science (NSF-IDBR, NASA-NAI, AFOSR) Important discoveries are often made by those with the best tools. We have designed and built several powerful instruments that offer major advantages for detection of dissolved O2 and H2 gases, intracellular fluorescence detection of pigments and pyridine nucleotides, and magnetic resonance. These tools are being applied to projects in photobiology, photochemistry and geomicrobiology.
Detection of Dissolved Hydrogen. In another approach we are developing ultrasensitive tools for screening for microbial hydrogen production activity from diverse natural habitats. Strains isolated from these screens have been shown to possess better metabolic properties more suited for large scale H2 production. This strategy has identified novel strains from volcanic soda lakes, thermophillic sources and hypersaline aquifers. Data from one of the three powerful tools that we have built is illustrated in the figure, showing detection of dissolved H2 (at 10-8 M sensitivity) produced by induction of hydrogenase activity in the green microalga, Chlamydomonas reinhardtii. The figure illustrates how trains of light pulses of variable pulse duration give rise to different kinetics of H2 production. At higher light duration, more O2 is produced which poisons the hydrogenase enzyme thereby suppressing H2 production. This behavior differs among individual strains and between different species.
Detection of Intracellular NADH+NADPH by Fluorescence. These cofactors are believed to serve as the sole reductants for the hox-class of bidirectional NiFe-hydrogenases. This claim has been inferred almost entirely from in vitro assays. However, there is no direct evidence identifying the cognate reductants for hydrogenase in vivo. Moreover, there is ambiguity in the literature on whether NADH or NADPH or both can serve as reductant for the same hydrogenase in some species. To examine these issues with greater precision, we have constructed an instrument for detection of intracellular fluorescence emission produced by reduced pyridine nucleotide cofactors (total NADH+NADPH) (see Fig.). This instrument includes a second channel for simultaneous electrochemical detection of dissolved H2 using our homebuilt design (2 nanomolar sensitivity). Completed in May 2007, this unique instrument is providing a wealth of new information about intracellular redox regulation via pyridine nucleotides.
Detection of Charge Separation by Fluorescence. Once of these instruments is a fast-repetition rate laser-based fluorometer that has the capability of measuring the speed and error frequency of photochemical turnover of the PSII O2-evolving complex (OEC) (Ananyev and Dismukes, 2005). The instrument has increased by 100 fold the range of flash excitation rates previously examined using oximetry and allows measurements on intact cells and leaves, samples previously inaccessible without perturbation. With this tool we have been able to characterize the efficiency of PSII-OEC turnover over its full range of turnover frequencies using intact cells and leaves without the need to isolate the PSII enzymes. Fitting of the data to kinetic models for PSII turnover have revealed new control features of the OEC.