Synthetic Enzymatic Pathway Conversion of Cellulosic Biomass to Hydrogen Joseph Anthony ...

139 Pages·2013·13.73 MB·English

Synthetic Enzymatic Pathway Conversion of Cellulosic Biomass to Hydrogen

Joseph Anthony Rollin

Dissertation submitted to the faculty of the Virginia Polytechnic Institute and State University in
partial fulfillment of the requirements for the degree of

Doctor of Philosophy
In
Biological Systems Engineering

Yi Heng Percival Zhang, Co-Chair
Justin R. Barone, Co-Chair
Aaron S. Goldstein
Ryan S. Senger
Brenda S. J. Winkel

23 August 2013
Blacksburg, VA

Keywords: synthetic enzymatic pathway (SyPaB), biohydrogen, cellulosic biofuel, cellulose
solvent- and organic solvent-based lignocellulose fractionation (COSLIF)
Synthetic Enzymatic Pathway Conversion of Cellulosic Biomass to Hydrogen

Joseph A. Rollin

ABSTRACT

In order to meet the energy needs of a growing world in a sustainable manner, new high
efficiency, carbon-neutral fuels and chemical feedstocks are required. An emerging
approach that shows promise for high efficiency production of renewable fuels and
chemicals is the use of purified enzymes combined in one pot to catalyze complex
conversions—synthetic pathway biotransformations (SyPaB). An exemplary technology
in this burgeoning field is the production of hydrogen from biomass sugars.
Lignocellulosic biomass, which includes agricultural residues, energy crops, and
industrial waste streams, is an ideal substrate for SyPaB conversion, as it is abundant and
cheap—second only to untaxed coal on a $/energy content basis. But the structure of
biomass is highly recalcitrant, making high-yield biological conversion difficult to
achieve. In order to increase susceptibility to enzymatic digestion, thermochemical
pretreatments are applied, with the goals of removing of lignin, the simplest example
being soaking in aqueous ammonia (SAA); hemicellulose removal, most often using
dilute acid (DA); and increasing cellulose accessibility by cellulose solvent-based
pretreatments, such as cellulose solvent- and organic solvent-based lignocellulose
fractionation (COSLIF). In a comparison of the lignin removal (SAA) and accessibility
increase (COSLIF) approaches, we found that certain levels of lignin removal (~15%)
were important, but further lignin removal was less effective at achieving digestibility
gains than increasing cellulose accessibility. Pretreated biomass forms an excellent
substrate for SyPaB hydrogen generation, due to low cost and high sugar content.
Following experiments demonstrating the high yield conversion of sucrose to hydrogen
(97%) and SyPaB generation of hydrogen at a rate commensurate with the best biological
rates achieved, 157 mmol/L/h. SyPaB was combined with enzymatic hydrolysis to enable
the direct conversion of cellulosic biomass, including untreated, DA, and COSLIF corn
stover. In addition, an updated kinetic model of the system was used to rationally increase
the maximum hydrogen production rate by 70% while minimizing total enzyme loading
and without increasing substrate concentration. Together, these results demonstrate the
high level of engineering control in cell-free systems, which can enable conversion of a
variety of substrates to hydrogen at the highest possible yield and rates as high as any
biohydrogen production method.
To my Father and Mother
iii
ACKNOWLEDGEMENTS

The author wishes to thank Percival Zhang for his guidance and suggestions over
the course of this project. It was a privilege to work on the development of this important
project.
The author would also like to thank Ryan Senger for valuable discussions and
collaboration during the modeling phase of the project and thought-provoking metabolic
engineering course, Brenda Winkel for an helpful discussions and an inspiring molecular
biology laboratory course, Aaron Goldstein for his valuable suggestions, and Justin
Barone for his very helpful support, as well as all the professors who ignited a newfound
passion for biology and engineering, especially S. Ted Oyama, Ann Stevens, David
Bevan, and Kevin Edgar.
The author wishes to thank his lab mates in the Virginia Tech Biofuels lab:
Zhiguang Zhu, Noppadon Sathitsuksanoh, Suwan Myung, Hehuan Liao, Xinhao Ye,
Yiran Wang, Wenjin Liu, Niel Templeton, Xiaozhou Zhang, Fangfang Sun, Chun You,
Peng Qi, Xing Zhang, Hui Ma, Ward Tam, Allison Bakovic, and Roberto Castro, with
special thanks to Julia Martin del Campo and all of the previously mentioned labmates
who assisted in vector design and construction, protein production and purification, and
day-to-day assistance, without which this project would not have been impossible.
Additional thanks to Mary Leigh Wolfe, Denton Yoder, Barbara Wills, Susan
Rosebrough, and the rest of the BSE department staff for their help during the course of
this work.
Finally, the author would like to recognize the tremendous support of the
Department of Defense through the National Defense Science and Engineering Graduate
research (NDSEG) fellowship. This  program’s  support  enabled  significantly  greater  
research focus and effectiveness, and networking during annual conferences fostered
valuable intellectual ferment.
iv
ATTRIBUTION

Several colleagues aided in the writing and research presented in chapters 2-6 of this
dissertation. A brief description of their contributions is included here.

Chapter 2: New Biotechnology Paradigm: Cell-Free Biosystems for Biomanufacturing
Chapter 2 was published in Green Chemistry.
Tsz Kin Tam, PhD (Cell-Free Bioinnovations, Inc.) is currently Chief Technology
Officer of Cell-Free Bioinnovations Inc., a Blacksburg-based start-up company. Dr. Tam
was a co-author of this paper. He wrote the initial draft of section 3 of this review, and
provided input during editing.
Y. H. Percival Zhang, PhD (Zhang Lab, Biological Systems Engineering Department) is
currently an Associate Professor at Virginia Tech. Dr. Zhang was the corresponding
author for this paper. He aided in the design of topics to be covered in this review,
constructed the initial outline, and provided editing input.
Chapter 3: Increasing Cellulose Accessibility Is More Important Than Removing
Lignin: A Comparison of Cellulose Solvent-Based Lignocellulose Fractionation and
Soaking in Aqueous Ammonia
Chapter 3 was published in Biotechnology and Bioengineering.
Zhiguang Zhu, PhD (Cell-Free Bioinnovations, Inc.) is currently a principal investigator
for Cell-Free Bioinnovations Inc., a Blacksburg-based start-up company. Dr. Zhu was a
co-author of this paper. He assisted with experimental determination of the cellulase
accessibility assay for all biomass samples.
Noppadon Sathitsuksanoh, PhD (Joint BioEnergy Institute) is currently a research
scientist at JBEI. Dr. Sathitsuksanoh was a co-author of this paper. He conducted SEM
images.
Marcus Foston, PhD (Foston Lab, Department of Energy, Environmental & Chemical
Engineering) is currently an Assistant Professor at Washington University. While
working as a post-doctoral researcher in the Ragauskas Lab at Georgia Tech, Dr. Foston
conducted XRD diffraction analysis of all samples.
Y. H. Percival Zhang, PhD (Zhang Lab, Biological Systems Engineering Department) is
currently an Associate Professor at Virginia Tech. Dr. Zhang was the corresponding
author for this paper. He aided in the experimental design and provided editing input.
Chapter 4: Novel Hydrogen Detection Apparatus: Improved Calibration and
Demonstrated Synthetic Pathway Biotransformation of Glucose to Biohydrogen
Chapter 4 is in preparation for submission to the International Journal of Hydrogen
Energy.
v
Julia Martin del Campo, PhD (Sobrado Lab, Department of Biochemistry) is currently a
post-doctoral researcher at Virginia Tech. She assisted with apparatus calibration and
system trouble-shooting, as well as enzyme preparation.
Xinhao Ye, PhD (Pfizer) is currently a research scientist at Pfizer. Dr. Ye ordered all
custom glassware for the hydrogen apparatus, and set up the initial lab-scale system.
Michael W.W. Adams, PhD (Adams Lab, Department of Biochemistry and Molecular
Biology) is currently Professor at the University of Georgia. Dr. Adams is a co-author of
this paper. He provided hydrogenase enzyme required for the production of hydrogen
from glucose.
Y. H. Percival Zhang, PhD (Zhang Lab, Biological Systems Engineering Department) is
currently an Associate Professor at Virginia Tech. Dr. Zhang is the corresponding author
for this paper. He provided oversight during experimentation and provided editing input.
Chapter 5: Catabolic Water Splitting for High-Speed Hydrogen Production
Powered by Sucrose through Synthetic Enzyme Cascade
Chapter 5 was submitted to Biocatalysis.
Suwan Myung, PhD (Lotte Chemical Corporation) is currently a research scientist at the
Lotte Chemical Corporation in South Korea. Dr. Myung was an equally-contributing first
author of this paper, while a graduate student in the Biological Systems Engineering
Department at Virginia Tech. He conducted enzyme production and purification and ran
experiments demonstrating the conversion of sucrose to hydrogen. Dr. Myung wrote the
initial draft of this paper.
Chun You, PhD (Zhang Lab, Biological Systems Engineering Department) is currently a
post-doctoral researcher at Virginia Tech. Dr. You was a co-author of this paper. He
assisted with enzyme production.
Fangfang Sun (Cell-Free Bioinnovations Inc.) is currently a fermentation technician for
Cell-Free Bioinnovations Inc. Ms. Sun was a co-author of this paper. She produced
several of the enzymes used in this study.
Sanjeev Chandrayan (Adams Lab, Department of Biochemistry and Molecular Biology)
is currently a graduate student at the University of Georgia. Mr. Chandrayan is a co-
author of this paper. He produced and purified hydrogenase enzyme.
Michael W.W. Adams, PhD (Adams Lab, Department of Biochemistry and Molecular
Biology) is currently Professor at the University of Georgia. Dr. Adams is a co-author of
this paper. He provided hydrogenase enzyme required for the production of hydrogen
from sucrose and glucose-6-phosphate.
Y. H. Percival Zhang, PhD (Zhang Lab, Biological Systems Engineering Department) is
currently an Associate Professor at Virginia Tech. Dr. Zhang is the corresponding author
vi
for this paper. He designed the experiments conducted in this study and provided editing
input.
Chapter 6: Metabolic Engineering of a Synthetic Enzymatic Pathway for Rate
Improvement and Utilization of Lignocellulosic Biomass
Chapter 6 is in preparation for submission to Nature Biotechnology.
Julia Martin del Campo, PhD (Sobrado Lab, Department of Biochemistry) is currently a
post-doctoral researcher at Virginia Tech. Dr. Martin del Campo is a co-author of this
paper. She assisted with enzyme preparation.
Suwan Myung, PhD (Lotte Chemical Corporation) is currently a research scientist at the
Lotte Chemical Corporation in South Korea. Dr. Myung was a co-author of this paper,
while a graduate student in the Biological Systems Engineering Department at Virginia
Tech. He conducted enzyme production and purification.
Fangfang Sun (Cell-Free Bioinnovations Inc.) is currently a fermentation technician for
Cell-Free Bioinnovations Inc. Ms. Sun was a co-author of this paper. She produced
several of the enzymes used in this study.
Allison Bakovic (Department of Physics and Chemistry) is currently an undergraduate at
the Milwaukee School of Engineering. Ms. Bakovic is a co-author of this paper. During a
summer NSF-funded fellowship, she conducted many of the experiments converting g6p
to hydrogen, and assisted with the modeling work described in this study.
Roberto Castro (Department of Chemical and Natural Gas Engineering) is currently an
undergraduate at Texas A&M University. Mr. Castro is a co-author of this paper. During
a summer NSF-funded fellowship, he conducted biomass characterization assays and
assisted with the production of hydrogen from pretreated biomass.
Sanjeev Chandrayan (Adams Lab, Department of Biochemistry and Molecular Biology)
is currently a graduate student at the University of Georgia. Mr. Chandrayan is a co-
author of this paper. He produced and purified hydrogenase enzyme.
Michael W.W. Adams, PhD (Adams Lab, Department of Biochemistry and Molecular
Biology) is currently Professor at the University of Georgia. Dr. Adams is a co-author of
this paper. He provided hydrogenase enzyme required for the production of hydrogen
from pretreated biomass and glucose-6-phosphate.
Ryan Senger, PhD (Senger Lab, Biological Systems Engineering Department) is
currently an Assistant Professor at Virginia Tech. Dr. Senger is a co-author of this paper.
He provided modeling help and global sensitivity analyses, as well as editing input.
Y. H. Percival Zhang, PhD (Zhang Lab, Biological Systems Engineering Department) is
currently an Associate Professor at Virginia Tech. Dr. Zhang is the corresponding author
for this paper. He conceived of the experiments conducted in this study and provided
editing input.
vii
TABLE OF CONTENTS

ACKNOWLEDGEMENTS……………………………………………….........…......................iv
ATTRIBUTION..............................................................................................................................v
LIST OF FIGURES……………………………………………………………...........................ix
LIST OF TABLES……………………………………………………………….......................xiii
ABBREVIATIONS.....................................................................................................................xiv
CHAPTERS
1. Introduction…………………………………………………………….............................1
2. New Biotechnology Paradigm: Cell-Free Biosystems for Biomanufacturing....................9
3. Increasing Cellulose Accessibility Is More Important Than Removing Lignin: A
Comparison of Cellulose Solvent-Based Lignocellulose Fractionation and Soaking in
Aqueous Ammonia..................................................................……………………..........22
4. Novel Hydrogen Detection Apparatus: Improved Calibration and Demonstrated
Synthetic Pathway Biotransformation of Glucose to Biohydrogen..................................32
5. Catabolic Water Splitting for High-Speed Hydrogen Production Powered by Sucrose
through Synthetic Enzyme Cascade.…………….............................................................44
6. Metabolic Engineering of a Synthetic Enzymatic Pathway for Rate Improvement and
Utilization of Lignocellulosic Biomass…………….........................................................48
7. Conclusions.....…………………………………………………………..........................77
APPENDICES
A. Supplementary Materials for Catabolic Water Splitting for High-Speed Hydrogen
Production Powered by Sucrose through Synthetic Enzyme Cascade..............................83
B. Supplementary Materials for Metabolic Engineering of a Synthetic Enzymatic
Pathway for Rate Improvement and Utilization of Lignocellulosic Biomass...................93
C. Example MATLAB Programs for Metabolic Engineering of a Synthetic Enzymatic
Pathway for Rate Improvement and Utilization of Lignocellulosic Biomass...................98
D. Metabolomics Assay for the Determination of Cell-Free Biotransformation
Intermediates...................................................................................................................121

viii
LIST OF FIGURES
Chapter 2
Figure 1:  Classification  of  biotransformation  based  on  biocatalysts………………….....11
Figure 2: Major targets for cell-free technologies, grouped by type. Proteins, high-value
compounds, and commodities each present different production challenges. The log-log
plot allows a single graph to show very high-value, low volume products, as well as low-
value, high volume biocommodities.................................................………………….....11
Figure 3: SEPB conversion pathways from various starting carbohydrates to
hydrogen…………………………………………………………………………………14
Figure 4: Pathway for complete oxidation of glucose mediated by PQQ-dependent
dehydrogenases. Electrons harnessed from these reactions provide the electrode potential
for an enzymatic fuel cell (adapted from Xu and Minteer55). Here PQQ is used as a
cofactor to transfer the charge to the electrode..............………………………………...14
Figure 5: Pathway for the production of isobutanol from glucose (adapted from Guterl et
al.62). In this pathway, first pyruvate and reducing equivalents are generated, after which
these were converted into isobutanol. An alternative pathway allows the production of
ethanol from pyruvate in the second half of the pathway (not shown).............................15
Figure 6: Two pathways for the production of fructose from starch. Conversion using the
traditional pathway (A) is limited by the reaction equilibrium of glucose isomerization,
whereas the more recent pathway (B) proposed by Moradian and Brenner67 uses
phosphorylation to enable complete conversion…………………...................................16
Figure 7: Schematic representation of synthetic metabolons. Different mechanisms may
be used to specifically bind several different enzymes on the same scaffold in sequence,
including cohesin-dockerin pairs and DNA-zinc finger protein pairs. (A) Three-enzyme
cascade, linear design. (B) Three-enzyme cascade, mirrored design (adapted from
Conrado et al.71)…………………………........................................................................16
Figure 8: Structures and costs of natural cofactors: NADP (A) and NAD (B),
nicotinamide mononucleotide (NMN) (C), nicotinamide riboside (NR) (D), biomimetic
cofactors (E, 1-benzyl-3-carbamoyl-pyridinium; F, 1-buta-3-carbamoyl-pyridinium and
G, nicotinamide glucoside).................................................................................………...17
Figure 9: Redox enzyme engineering strategies................................................................18

ix
Chapter 3
Figure 1: Conceptual images of the effects of SAA and COSLIF pretreatments. Areas of
the relative components correspond to their percentage of the microfibril or pretreated
material. Cellulose surfaces susceptible to enzymatic attack (110 face) are highlighted in
red. SAA lignin is greatly decreased, but remains widely distributed. The quantity of
COSLIF lignin is high, but it may form clusters as depicted here. Cellulose accessibility
increases greatly following COSLIF pretreatment............................................................24
Figure 2: COSLIF-pretreated (A) and SAA-pretreated (B) switchgrass at standard and
low enzyme loadings, with and without BSA blocking. In these graphs circles represent a
standard enzyme loading (15 FPU/g glucan), and triangles represent a low enzyme
loading  (3  FPU/g  glucan).  All  hydrolysis  runs  were  supplemented  with  10  U  β-
glucosidase. Solid data points represent hydrolysis conducted without BSA, and open
data points represent hydrolysis conducted after BSA blocking. Error bars represent one
standard  deviation………………………………………..................................................27
Figure 3: SEM micrographs of untreated (A and B), SAA-treated (C and D) and
COSLIF-treated (E and F) switchgrass. Magnification of samples A, C, and E is
approximately 350X, while magnification was increased to approximately 3,000X for B,
D, and F. In the untreated material (A), vascular bundles and parenchyma are highlighted
with the upper and lower arrows, respectively..................................................................27
Figure 4: Thioredoxin–GFP–CBM fusion protein schematic, with TGC proteins
represented in green and BSA proteins in blue. The TGC protein is similar in size to T.
reesei EG1 (A). To determine total substrate accessibility to cellulase (TSAC), TGC
equilibration is conducted without BSA (B). When BSA blocking is used prior to TGC
equilibration, cellulose accessibility to cellulase (CAC) may be determined (C). Cellulose
surfaces (110 face) susceptible to cellulase binding are highlighted  in  red…...................28
Figure 5: XRD diffraction spectra for untreated, COSLIF, and SAA switchgrass samples.
The major peak seen for all three samples is the (0 0 2) peak, used for peak height-based
CrI determination. (1 0 1) and (0 4 0) peaks are also apparent for untreated and SAA
samples...............................................................................................................................28
Figure 6: Digestibility as a function of delignification and CAC. Data points were
obtained from this study and previous studies in the literature (Hong et al., 2007; Kumar
and Wyman, 2009; Kumar et al., 2009; Sathitsuksanoh et al., 2009, 2010; Wyman et al.,
2009; Zhu et al., 2009a). Here a correlation between delignification, CAC, and 72 h
enzymatic digestibility is suggested, for a broad range of feedstocks and
pretreatments………………………………………………….........................................29

x