GCEP Research Symposium September 28. BioHydrogen Generation

Transcription

1 GCEP Research Symposium September 28 BioHydrogen Generation Jim Swartz, James H. Clark Professor Departments of Chemical Engineering and Bioengineering Stanford University H 2

2 Solar Energy Can this be One Component of our Sustainable Energy Future? Engineered Photosynthetic Microorganisms A Clean Sustainable Cycle H 2 O O 2 H 2 Fertilizer, Cement, Chemicals Fuel Enrichment, Electricity Transportation

3 What is the Potential for Solar Biohydrogen? Current U.S. Consumption Of Liquid Fuels plus Nat. Gas 60 x Btu / yr Assume: Solar Incidence of 7 kwh/m 2 -day and 15% Energy Efficiency (Overall) Need 17,000 Square Miles = 2.5% of Current Cropland (But Do Not Need to Convert Current Crop Land)

4 Lacking Effective Storage and Transport, What are the Initial Markets? Current World Market for Hydrogen is about 50 million metric tons (about $135 Billion/yr) In US, we use about 9 million metric tons/yr (95% from CH 4 ) Releasing about 60 million metric tons of CO 2 per year Equivalent to 34 million cars () Most is used for fertilizer and refineries Current Markets: 1. NH 3 Fertilizer Uses about half of the hydrogen 2. Refineries Increase fuel quality 3. Chemical Synthesis Potential Markets: 1. Portland cement about 100 million tons/yr in US ( 10x10 6 tons CO 2 /yr) 2. Short term solar energy storage for electricity

5 Direct BioConversion of Sunlight to H 2 Engineered Organisms Estimated Economics Allow About $20-30/ft 2 Temp. Control Fluid Out One Candidate Collector/Reactor Cross Section Incident Sunlight Large Surface Area Collector/Reactor Rejuvenate Culture Low Pressure* Organism Suspension Cooling Fluid Low Partial Pressure H 2 & O 2 Harvest Temp. Control Fluid In Transparent Cover Transparent Gas Permeable Membrane *or Purge Gas (N 2 )

6 Direct Photobiological Hydrogen Production: Building New Electron and Proton Pathways Solar Energy Photolysis Center 4e - O 2 4H + 2H 2 2H 2 O Reduced Ferredoxin 4e - Hydrogenase Oxidized Ferredoxin (Clostridial) Goal: Engineered Synechocystis Bacterium

7 BUT Oxygen is a Side Product and it Inactivates the Hydrogenase We Need 2H 2 O 4e - O 2 4H + Reduced Ferredoxin Oxidized Ferredoxin 4e - Hydrogenase (Clostridial) 2H 2 Initial and Primary Challenge: EVOLVE AN OXYGEN- TOLERANT HYDROGENASE

8 [Fe-Fe] Hydrogenases are the Fastest H 2 Producers But They Are VERY Complex Hydrogenase CpI from Clostridium pasteurianum 1 2H + + 2Fd red H 2 + 2Fd oxid Active Iron- Sulfur Site Ferredoxin Binding Site Electron transfer metal centers Active Site H-Cluster Has 6 Irons, 6 Sulfurs, 3 CO, 2 CN and a Bridging Dithiol 1 Peters JW, Lanzilotta WN, Lemon BJ, Seefeldt LC Science 282(5395):

9 A Structural Dynamic Model Suggests Two Dominant Gas Channels for Oxygen Entry PROTEIN EVOLUTION HYPOTHESIS Before o 2 o 2 After 2 Cohen J, Kim K, King P, Seibert M, Schulten K Structure 13(9):1321.

10 This is Our Evolution (Screening) Protocol Generate Diversity PCR Error Prone PCR DNA Shuffling Rational Design Mutant Library Isolate Mutants by Dilution Amplify Mutants PCR Gene Library Protein Library Cell-Free Protein Synthesis: Anaerobic Chamber, 2% H 2 Oxidized Methyl Viologen Hydrogen Consumption Assay 2MV ox+ H 2 2MV red + 2H + (Blue) Expose to Oxygen by Addition of Air - Saturated Buffer Jim Stapleton, Jon Kuchenreuther, Phil Smith, Alyssa Bingham

11 We Could Choose from Two [Fe-Fe] H 2 ases and Began with the Simpler One C. reinhardtii HydA1 C. pasteurainum HydA (CpI)

12 We Found a Faster One (But not with Ferredoxin) None of the Mutants was more Oxygen Tolerant Specific Activity of Hydrogen Production ( mole H 2 / Hydrogen production specific activity [umol H2/min/mg min-mg HydA1] Enzyme) Methyl Methyl Viologen viologen as e - donor Ferredoxin Ferredoxin as e - donor you get what you screen for Jim Stapleton Jon Kuchenreuther 0 Wild WT G116D N211S G116D E7 Type + N211S

13 We Then Switched to The More Complex Enzyme And Had to Reoptimize Expression C. reinhardtii HydA1 C. pasteurainum HydA (CpI)

14 Percent Activity Remaining Alyssa Bingham Found a More Oxygen- Tolerant Mutant!! With Three Mutations 25 Mutants were expressed in vivo in E.coli and purified Added 5 ul of air saturated buffer to 95 ul of enzyme solution Incubated 20 minutes (uncovered in anaerobic glove box) Percent Activity of CpI After Oxygen Exposure WT I 197 V I 197 V I 197 V I 197 V I 197 V I 197 S CpI I197V CpI I197V N 160 D A A280V 280 V N N289D 289 D N 160 D N 160 D N 289 D WT CpI CpI I197V CpI I197V N160D CpI I197V N160D N N289D 289 D CpI I197S N160D N289D Alyssa Bingham, Kunal Mehta, Rachel Bent

15 Two of the Mutations were in An Unexpected Location -- Near an Electron Conducting Center N 289 D N 160 D I 197 V Most Influential We Then Needed a Method to Test for Oxygen Tolerance During Hydrogen Production

16 % H2 in headspace We Wanted a Continuous Electron Supply from Ferredoxin The Natural Electron Source We Assembled This Pathway Using Purified Proteins glucose-p 1 H 2 O NADPH Glucose-P Dehydrogenase NADP + H + FNR* 2 Fd Red H + 2 Fd Ox H 2 ase *FNR = Ferredoxin NADPH Reductase H 2 2 H + Headspace H 2 Accumulation In Spite of Thermodynamic Concerns, It Worked 1.20% 1.00% 0.80% 0.60% 0.40% 0.20% 0.00% Time (min) Phil Smith, Alyssa Bingham

17 Unfortunately, The Oxygen-Tolerant Mutant Was NOT Tolerant When Making Hydrogen Simple gas channel hypothesis is not the whole story We will now conduct protein evolution by screening for mutants that are more oxygen tolerant when making hydrogen using ferredoxin as the electron source the natural way We will use tungsten oxide/palladium coated glass plates to sense hydrogen production Hydrogen Sensor 96-well assay plate

18 BUT We also realized that our enzyme pathway might be useful to make H 2 from Glucose To be commercially viable the process must deliver: 1. High conversion efficiencies, and 2. High Productivities It must also have an inexpensive source of enzymes. (We can t afford to separately produce and purify 15 enzymes!) Could We Conduct Cell-free Metabolic Conversions Using Crude Cell Extracts?

19 We can use the Pentose Phosphate Pathway to Get High Conversion Yields Basic Pathway glucose-p 6 H 2 O 12 NADPH 6CO 2+ 12H + Pentose 24 Fd Red Phosphate Pathway FNR 12 NADP + 12H + 24 Fd Ox H 2 ase 12 H 2 24H + 6 H 2 O Complete Pathway (Can also use xylose; i.e. cellulose hydrolysates) Glucose Glucose-6-P 6 CO 2 (C 6 H 12 O 6 ) Pentose P i Pathway Oxidative Phosphorylation ATP ADP H 2 O ½O 2 P i 1 NADPH + 1 H + 12 NADPH + 12 H + 1 NADP + 11 NADPH + 11 H + This Predicts Very High Conversion Efficiencies (92%) FNR 11 NADP + 22 Fd ox 11 H 2 22 Fd red 22 H + H 2 ase

20 The process flow diagram would look like this; We will need to express high levels of the H 2 ase Fermentor A: Make FNR (2X Volume) Niacin Iodoacetamide RNase Glucose Gas Out Semipermeable Membrane H 2 Off SDS Page Gel of UNPURIFIED Cell Lysates Homogenizer H 2 Reactor (7X Volume) Semipermeable Membrane Fermentor B: Make H 2 ase and Fd (5X Volume) O 2 Slow Feed N 2 CO 2 Recovery A New Cell Line and New Process Produces High Level of the H2ases Jon Kuchenreuther

21 What Productivities Might Be Possible? -- Use Ethanol Production as a Benchmark Vessel Fermentor A Fermentor B H 2 Reactor Ethanol: [FNR] [Fd] (µm) (µm) g EtOH L [H 2 ase] (µm) hr Fermentor A: Niacin Make FNR Iodoacetamide (2X Volume) RNase Homogenizer Fermentor B: Make H 2 ase and Fd (5X Volume) MJ g EtOH = 0.04 MJ/L-hr Glucose Gas Out H 2 Reactor (7X Volume) N 2 O 2 Slow Feed H 2 Off Semipermeable Membrane Semipermeable Membrane CO 2 Recovery Proposed Process: 86 µmole H 2 ase L 5 µmoles H 2 mole H 2 ase sec 3600 sec hr mole 10 6 µmole 2 g H 2 1 mole H MJ g H 2 = 0.40 MJ/L-hr

22 We Are Making Good Progress in Increasing Fuel Value Productivities kj L -1 Hr glucose-p 1 H 2 O Woodward et al. (2000) Nature Fuel Value Production Rates NADPH Glucose-P Dehydrogenase NADP + Zhang et al. (2007) PloS ONE H + FNR* 2 Fd Red H + H 2 ase 2 Fd Ox H 2 2 H + E. coli FNR E. coli FNR Spinach FNR Woodward et al. (2000) Nature Zhang et al. (2007) PloS ONE E. coli FNR E. coli FNR Spinach FNR Fuel value productivity (kj L -1 hr -1 ) Molar Productivity (µmole L -1 hr -1 ) H'ase Turnover number (sec -1 ) Temperature ( C) [FNR] (µm) a 50 a 30 b [Fd] (µm) c 50 c 80 c [Hydrogenase] (µm) 0.47 d 39.4 d 1 e 10 e 2 e [NADPH] (mm) a E. coli FNR; b Spinach FNR; c Synechocystis [2Fe2S] ferredoxin; d Pyrococcus furiosus hydrogenase; e Clostridium pasteurianum hydrogenase I Phillip Smith

23 Conclusions and Future Work Conclusions: Cell-free H 2 ase Production Enables Effective Protein Evolution Screen Needs to EXACTLY Replicate Desired Enzyme Use Cell-free Metabolism Offers Great Promise for Hydrogen Production from Glucose and from Cellulose Hydrolysates Future Work: Evolve Oxygen Tolerant Hydrogenase Optimize Hydrogen Production from Sunlight in Cyanobacteria Find/Evolve More Active FNR Enzyme Implement Cell-free Hydrogen Production from Biomass Develop High Cell Density Protein Production Processes

24 Questions?