Breaking the Biological Barriers to Cellulosic Ethanol: A

Joint Research Agenda

A Research Roadmap Resulting from the Biomass to Biofuels
Workshop Sponsored by the U.S. Department of Energy

December 7–9, 2005, Rockville, Maryland

DOE/SC-0095, Publication Date: June 2006

Office of Science, Office of Biological and Environmental Research, Genomics:GTL Program
Office of Energy Efficiency and Renewable Energy, Office of the Biomass Program

DOE Genomics:GTL
GTL Biofuels
Home Page This Document

Chapter PDFs
• Executive Summary (257 kb)
• Introduction (1524 kb)
• Technical Strategy: Development of a Viable Cellulosic Biomass
to Biofuel Industry (263 kb)
• System Biology to Overcome Barrier to Cellulosic Ethanol Current File
ƒ Lignocellulosic Biomass Characteristics (794 kb)
ƒ Feedstocks for Biofuels (834 kb)
ƒ Deconstructing Feedstocks to Sugars (632 kb)
ƒ Sugar Fermentation to Ethanol (1367 kb)
• Crosscutting 21st Century Science, Technology, and Infrastructure
for a New Generation of Biofuel Research (744 kb)
• Bioprocess Systems Engineering and Economic Analysis (66 kb)
• Appendix A. Provisions for Biofuels and Biobased Products in the
Energy Policy Act of 2005 (54 kb)
• Appendix B. Workshop Participants and Appendix C. Workshop
Participant Biosketches (529 kb)

John Houghton Sharlene Weatherwax John Ferrell

Office of Science Office of Science Office of Energy Efficiency
Office of Biological and Office of Biological and and Renewable Energy
Environmental Research Environmental Research Office of the Biomass
301.903.8288 301.903.6165 Program
John.Houghton@ Sharlene.Weatherwax@ 202.586.6745
science.doe.gov science.doe.gov John.Ferrell@
hq.doe.gov

base url: www.doegenomestolife.org

Lignocellulosic Biomass
Characteristics
Makeup, Structure, and Processability

L
ignocellulosic biomass has long been recognized as a potential low-
cost source of mixed sugars for fermentation to fuel ethanol. Plant
biomass has evolved effective mechanisms for resisting assault on its
structural sugars from the microbial and animal kingdoms. This property
underlies a natural recalcitrance, creating technical barriers to the cost-
effective transformation of lignocellulosic biomass to fermentable sugars.
Moderate yields and the resulting complex composition of sugars and
inhibitory compounds lead to high processing costs. Several technologies
have been developed over the past 80 years, often in wartime, that allow
this conversion process to occur, yet the clear objective now is to make the
process cost-competitive in today’s markets.
Cell walls in lignocellulosic biomass can be converted to mixed-sugar solu-
tions plus lignin-rich solid residues by sequential use of a range of thermo­
chemical pretreatments and enzymatic saccharification. The low rate at
which biomass is converted to sugars and the coproduction of fermenta-
tion inhibitors increase equipment size and result in high pretreatment and
enzyme costs. New approaches for designing improved energy feedstocks,
deconstructing plant cell walls, and transforming their polysaccharides
to fermentable sugars are needed. A systematic understanding of enzyme
interactions with plant cell architecture and hierarchy, as well as cellu­lose,
hemicellulose, and lignin structure during chemical and enzymatic hydrol­
ysis, will allow the prediction of plant-tissue response to hydrolytic attack
and the creation of new systems.
Significant technology development will be needed for creation of large-
scale bioenergy and biorefinery industries that can handle a billion tons
made up of a variety of biomass each year. In the DOE-USDA Billion-
Ton Study, corn stover and perennial crops such as switchgrass and hybrid
poplar make up about half the potential 1.3 billion tons of biomass that
could be available by the mid-21st Century (Perlack et al. 2005). Under-
standing the structure and function of these and other biomass resources
will be critical to enhancing their processability.
The result of analysis and research described here will be to increase the
efficiency with which the solid (substrate) interacts with large protein
macromolecules (enzymes) at its surface while the surface itself is being
eroded into soluble oligosaccharides [see sidebar, Image Analysis of Bio-
energy Plant Cell Surfaces at the OBP Biomass Surface Characterization
Lab (BSCL), p. 40]. This knowledge, combined with development of new
proteins that catalyze these transformations as well as microbial systems

Biofuels Joint Roadmap, June 2006 • Office of Science and Office of Energy Efficiency and Renewable Energy • U.S. Department of Energy 39
 LIGNOCELLULOSIC BIOMASS
Image Analysis of Bioenergy Plant Cell Surfaces at the OBP Biomass Surface
Characterization Lab (BSCL)

M
any aspects of current biomass conversion technology are becoming better understood, and a nascent
biomass processing industry is emerging for some niche markets. To reach the mid- and long-term
goals stated in the DOE Office of the Biomass Program’s Multi Year Program Plan: 2007-2012,
however, enhanced fundamental understanding of feedstocks and all biorefinery processes is critical. For exam-
ple, detailed knowledge about plant
cell-wall ultrastructure and func-
tion to formulate improved enzyme
mixtures and pretreatments will
reduce the cost of producing sug-
ars. (See images, right.)
In many cases, we know how to
describe biomass compositionally.
That is, we can conduct chemi-
cal or spectroscopic analyses and
determine the percentages of
individual sugars, protein, uronic
acids, and lignin. When we study
biomass conversion of corn sto-
ver, hardwoods, or rice straw, for
example, we are in fact working
primarily with the plant’s struc-
tural parts, most of which are cell
wall. Therefore, more knowledge is Fig. A. Collage of Scanning Electron Microscopy Images Showing a
needed about the natural organiza- Rind and Adjacent Pith Section Cut from a Field-Dried Corn-Stem
tion and structure of polymers and Cross Section. The rind shows a higher density of vascular elements made
chemicals in plant tissue that affect from thick-walled cells. The pith section (shown longitudinally) shows a
chemical pretreatment, enzymatic greater number of thin-walled parenchyma cells. Overall, most cellulose
digestibility, and the generation of needed for biomass conversion is located in the rind, although the pith rep-
compounds inhibiting fermentative resents most of the stem volume. Closeups of a cell-wall pit also are shown
microorganisms used to produce (~150,000×). These structures are thought to aid transfer of chemicals and
the final fuel or chemical. The study enzymes used in processing within the biomass bulk.
of plant cell walls at the submi-
cron or macro­molecular scale is
challenging. Imaging and image
analysis are at the cutting edge of
botany, molecular biology, bio-
chemistry, chemistry, and material
and computer sciences. Descrip-
tions of microscopies important
for ultrastructure imaging are in
the Imaging Technologies section
of the Crosscutting Technologies
chapter, p. 163, and sidebar, Some
Imaging Technologies Relevant to
Feedstock Characterization, p. 163.
Fig. B. AFM: Corn Parenchyma Cell Wall.

40 Biofuels Joint Roadmap, June 2006 • Office of Science and Office of Energy Efficiency and Renewable Energy • U.S. Department of Energy
for fermentation and consolidation, will enable the design of procedures
and hardware that dramatically speed up the process, improve yield, and
lower costs.

Structure and Assembly of Cell Walls

Plant cell walls are a complex and dynamic mixture of components that
perform many functions (see Fig. 1. Simplified Cell Wall and Fig. 2.
Conceptual Illustration of Cell-Wall Biogenesis, p. 42; and sidebar, Under-
standing Biomass, pp. 53 to 55). The cell walls are intricate assemblages of
celluloses, hemicelluloses (i.e., xyloglucans, arabinoxylans, and glucoman-
nans), pectins (i.e., homogalaturonans, rhamnogalacturonan I and II, and
xylogalacturonans), lignins, and proteoglycans (e.g., arabinogalactan-pro-
teins, extensins, and proline-rich proteins). Most mass in the plant cell
wall is in the form of polysaccharides (cellulose and hemicelluloses). The
next most abundant polymer is lignin, which is composed predominantly
of phenylpropane building blocks. Lignins perform an important role in
strengthening cell walls by cross-linking polysaccharides, thus providing
support to structural elements in the overall plant body. This also helps the
plant resist moisture and biological attack. These properties of lignin, how-
ever, interfere with enzymatic conversion of polysaccharide components.
Additionally, since lignin is not converted readily to ethanol, we must find
other uses in the process if we are to maximize energy yield from biomass.
Several thousand gene products are estimated to participate in synthesis,
deposition, and function of cell walls, but very few associated genes have
been identified and very little is known about their corresponding enzymes.
Many questions remain, for example, regarding how polysaccharides and
lignin are synthesized, how wall composition is regulated, and how com-
position relates to cell-wall biological functions. To answer these questions,
we need to discover the functions of many hundreds of enzymes, where
proteins are located within cells, whether or not they are in complexes,
where and when corresponding genes are expressed, and what factors and
genes control expression and activities of the proteins involved. Applica-
tion of new or improved biological, physical, analytical, and mathemati-
cal tools will facilitate a detailed mechanistic understanding of cell walls.
That knowledge will permit optimization of various processes involved in
producing biomass and converting it to fuels.
Productivity and conversion-process efficiencies can be increased by alter-
ing fundamental aspects of plant growth, development, and response to
biotic and abiotic stress. Altering cell-wall composition to increase the
relative amount of cellulose and decrease lignin, for example, could have
significant effects (see sidebar, Optimizing Lignin Composition for More
Efficient Bioethanol Production, p. 43). Eventually, a systems cell-wall
model incorporating biophysical aspects with structural properties and
knowledge of proteins involved in synthesis will aid in rational develop-
ment of highly productive feedstock species whose cell walls are opti-
mized for conversion.

Biofuels Joint Roadmap, June 2006 • Office of Science and Office of Energy Efficiency and Renewable Energy • U.S. Department of Energy 41
 LIGNOCELLULOSIC BIOMASS
Fig. 1. Simplified Cell
Wall. For more details, see
sidebar, Understanding
Biomass, p. 53. [Adapted
with permission from
C. Somerville et al.,
Science 306, 2206–11
(2004); © 2004 AAAS.]

Fig. 2. Conceptual Illustration of Cell-Wall Biogenesis. The Golgi apparatus participates in hemicellulose and lignin
biosynthesis. Cellulose microfibrils (CMF) are laid separately in the swollen gel of hemicelluloses. As lignin is deposited,
the cell wall becomes hydrophobic. Water removal from the swollen gel, together with peroxidase (PO) and calcium,
causes anisotropic shrinkage perpendicular to the CMFs. This shrinkage drives further oligolignol polymerization by
orienting the lignin aromatic ring parallel to the cell-wall surface. [Adapted from N. Terashima et al., “Comprehensive
Model of the Lignified Plant Cell Wall,” pp. 247–70 in Forage Cell Wall Structure and Digestibility, ed. H. G. Jung et al.,
American
Society of
Agronomy,
Crop Sci-
ence Society
of America,
and Soil Sci-
ence Society
of America
(1993).]

42 Biofuels Joint Roadmap, June 2006 • Office of Science and Office of Energy Efficiency and Renewable Energy • U.S. Department of Energy
Plants can have two types of cell walls, primary and secondary. Primary
cell walls contain cellulose, which consists of hydrogen-bonded chains of
thousands of β-1,4-linked glucose molecules, in addition to hemicelluloses
and other materials woven into a nanoscale network with the cellulose.
Cellulose in higher plants is organized into microfibrils, each measuring
about 3 to 6 nm in diameter and containing up to 36 cellulose chains. Each

Optimizing Lignin Composition for More Efficient Bioethanol Production

Plant lignin (guaiacyl and syringyl) interferes with the release and hydrolysis of cell-wall polysaccharides. Meta-
bolic engineering of the lignin biosynthetic pathway has been suggested as a method for modifying lignin con-
tent in feedstocks. Studies in Arabidopsis demonstrated that overexpression of the enzyme ferulate 5-hydroxylase
(F5H) increases lignin syringyl monomer content and abolishes the tissue specificity of its deposition (Fig. A).
To determine whether or not this enzyme has a similar regulatory role in woody plants, F5H was overexpressed
in poplar trees using a cinnamate 4-hydroxylase promoter to drive F5H expression. Transgenic trees displayed
enhanced lignin syringyl monomer content, indicating that F5H overexpression is a viable metabolic engineer-
ing strategy for modifying lignin biosynthesis. These high-syringyl lignin poplars demonstrated a significant
increase in chemical pulping efficiency. [R. Franke et al., “Modified Lignin in Tobacco and Poplar Plants Over-
Expressing the Arabidopsis Gene Encoding Ferulate 5-Hydrosylase,” Plant J. 22(3), 223–34 (2000).] Similar
metabolic engineering strategies hold promise for developing improved feedstocks for bioethanol production.
Many aspects of lignin biosynthesis remain matters of debate. Although most genes involved in the biosyn-
thetic pathway have been cloned and functions assigned, mechanisms that regulate the pathway still are largely
unknown, as is its relationship with other cell-wall biochemical pathways and plant development. Topics to
be studied include regulation of lignin deposition and tissue specificity, identity of proteins involved in mono-
lignol transport and polymerization, and ways in which lignin content and composition can be modified (see
Fig. 4. Phenylpropanoid Pathway Leading to Lignin Biosynthesis in Plants, p. 49). Also needed is a detailed
understanding of lignin-biodegradation mechanisms, including that accomplished by white rot fungi, which
break down yellow lignin and leave behind crystalline white cellulose (see sidebar, White Rot Fungus, p. 93).
Comprehensive explorations of lignin biosynthesis and degradation are required to maximize energy yield
from biomass crops.

S lignin Fig. A. Lignin Composition Controlled by Genetic Manipulation and Monitored via
S lignin G lignin Histochemical Staining for Lignin Monomer Composition in Arabidopsis Stem Cross
Sections. In the lignified cells of wild-type Arabidopsis stems, the presence of syringyl (S)
or guaiacyl (G) monomers can be visualized by histochemical staining of S lignin (red)
and G lignin (yellow).
Histochemical staining allows a diagnosis of the effects of experiments to manipulate
G lignin lignin composition. For example, eliminating one enzyme of the lignin biosynthetic path-
way in an Arabidopsis mutant (fah1-2) leads to a pure G lignin, and overexpression of the
same enzyme leads to a homogeneous deposition of S lignin (35S-F5H). The fah1 gene
encodes ferulate-5-hydroxylase, a cytochrome P450–dependent monooxygenase that
catalyzes hydroxylation of coniferaldehyde and coniferyl alcohol in the pathway leading to
S lignin syringyl lignin. [Figures published in C. Chapple et al., “Lignin Monomer Composition
is Determined by the Expression of a Cytochrome P450–Dependent Monooxygenase in
Arabidopsis,” Proc. Natl. Acad. Sci. USA 95, 6619–23 (1998); ©1998 National Academy of
Sciences, U.S.A.]

Biofuels Joint Roadmap, June 2006 • Office of Science and Office of Energy Efficiency and Renewable Energy • U.S. Department of Energy 43
 LIGNOCELLULOSIC BIOMASS

cellulose chain is a linear collection of thousands of glucose residues. Pairs

of glucose residues (cellobiose) make up the repeating unit of cellulose.
Like steel girders stabilizing a skyscraper’s structure, the mechanical
strength of the primary cell wall is due mainly to the microfibril scaffold.
A microfibril’s crystalline and paracrystalline (amorphous) cellulose core
is surrounded by hemicellulose, a branched polymer composed of pentose
(5-carbon) and, in some cases, hexose (6-carbon) sugars. In addition to
cross-linking individual microfibrils, hemicelluloses in secondary cell walls
also form covalent associations with lignin, a complex aromatic polymer
whose structure and organization within the cell wall are not completely
understood (see Fig. 3. Association of Lignin with Polysaccharides, this
page). The crystallinity of cellulose and its association with hemicellulose
and lignin are two key challenges that prevent the efficient breakdown of
cellulose into glucose molecules that can be fermented to ethanol.
Many enzymes involved in cell-wall synthesis
or modification are thought to be located in
protein complexes. Within the plasma mem-
brane are rosettes composed of the enzyme
cellulose synthase; these protein complexes
move laterally along the membrane to synthe-
size cellulose molecular chains (36 per rosette),
which crystallize into microfibrils. Movement
of the rosette molecular machine is associ-
ated with cortical microtubules that underlie
the membrane, but that linkage also is poorly
understood. The interaction of cellulose syn-
thase with the cytoskeleton has an impact on
cellulose fibril orientation and perhaps length.
Understanding the function of these com-
plexes and their interactions with metabolic
pathways that produce sugars will be impor-
tant for eventually controlling cell-wall com-
position. A number of cellulose-synthase genes
have been cloned for a variety of plants. (See
sidebar, Understanding Biomass, beginning on
p. 53, for an illustrated explanation.)

Fig. 3. Association of Lignin with Polysaccharides. The sche- Factors in Recalcitrance of

matic diagram shows possible covalent cross-links between
polysaccharides and lignin in cell walls. Lignin is bonded to the
Lignocellulose Processing to Sugars
cellulose and hemicellulose polysaccharides and serves as a stiff- Organization and interactions among polymers
ening and hydrophobic agent, complicating biomass breakdown. of the cell wall—constructed for strength and
[Source: Adapted from K. Iiyama, T. Lam, and B. Stone, “Cova- resistance to biological, physical, and chemi-
lent Cross-Links in the Cell Wall,” Plant Physiol. 104(2), 318 cal attack—constitute a barrier to access by
(1994). Reprinted with permission of American Society of Plant
depolymerizing enzymes and must be partially
deconstructed in the bioconversion pretreatment
Biologists, ©ASPB 1994.]
step before saccharification can occur. Although

44 Biofuels Joint Roadmap, June 2006 • Office of Science and Office of Energy Efficiency and Renewable Energy • U.S. Department of Energy
various pretreatments have been developed, we still do not have a detailed
understanding of fundamental physical and chemical features of lignocel-
lulosic biomass that limit its breakdown to sugars and ultimate bioconversion
efficiency. Improved methods must be developed to characterize biomass and
its interaction with various chemical treatments, as well as with deconstruc-
tion and saccharification enzymes.
Natural factors believed to contribute to lignocellulosic feedstock’s recalci-
trance to chemicals or enzymes include plant and cell-wall architecture and
molecular structure.

Plant Architecture
The organs of a plant (leaf, stem or trunk, and root) are composed of
myriads of cells with different functions in the plant’s economy. Each has
its own particular type of cell wall whose composition is related directly to
cell function [e.g., support (fibers), protection (epidermis), and transport
(xylem and phloem)]. Leaf, stem, and root tissues invariably contain cells
of more than one type. In tissues, individual cells are closely associated at
their cell-wall interfaces to give a compact tissue structure. This structure
must be disassembled by milling (comminution) to allow liquid access to
cell walls.
• The waxy barrier comprising grass cuticle and tree bark impedes pen-
etration of enzymes.
• Even milled plant stems and woody tissues limit liquid penetration by
their nature.

Cell-Wall Architecture
The nanoscale composite nature of the plant cell wall restricts penetration
of chemicals and enzymes to their substrates. The lignin-hemicellulose
coating on the cell wall’s cellulosic microfibrils affects the following:
• Conformation of cellulose and noncellulosic polysaccharides making
up the microfibril limits accessibility of hydrolytic enzymes to their
substrates.
• Lignin-carbohydrate complexes limit enzymatic hydrolysis of biomass
polysaccharides.

Molecular Structure
Cellulose crystallinity severely restricts cellulase attacks. Cellulases must
physically release individual cellulose chains from microfibril crystals for
subsequent catalytic hydrolysis to sugars. Limiting factors:
• Inherent difficulty of enzymes in acting on poorly hydrated cellulose surfaces.
• Amount and composition (including heterogeneity) of lignin.
• Chemical heterogeneity and strength of covalent interactions between
lignin and noncellulosic polysaccharides.
• Robustness of hydrogen bonding in cellulose microfibrils arising from
extended hydrogen-bond periodicity.

Biofuels Joint Roadmap, June 2006 • Office of Science and Office of Energy Efficiency and Renewable Energy • U.S. Department of Energy 45
 LIGNOCELLULOSIC BIOMASS

In addition, all native celluloses undergo physical modifications that can

inhibit saccharification as they are dehydrated in traditional methods of
isolation or storage after harvest and in pretreatment processes. Charac-
terizing the effects of conditions and storage environments would point
to modifications in harvesting and storage for biomass resources. Under-
standing physicochemical characteristics of the cell-wall polysaccharide
system would guide genomic modifications of bioenergy crops to facilitate
processing and resist deleterious physical modification as much as possible.
For example, structural elements of many lignocellulosic materials react to
pretreatment in ways that reduce enzymatic digestibility.
• High mechanical pressure such as that from plug feeders collapses the
natural vascular structure.
• Dilute-acid chemical pretreatments may permit cellulose to reanneal,
leading to “hornification” of cellulose in microfibrils.
• Ambient or elevated temperatures may accelerate denaturation (e.g., the
tendency of most (beta-1,4)-pentosans and hexosans to have inverse
water-solubility relationships with temperature).
• Some pretreatments may permit lignin to become soluble and “plate
out” on cellulose surfaces during the cool-down phase.
These “process-induced” causes of recalcitrance must be understood and
overcome through process modification or biomass design.
After cellulose, hemicellulose is
Optimizing Hemicellulose Acetylation in Cell Walls the next most abundant polysac-
charide in native biomass feed-
Hemicellulose Acetylation Degradation Products Are Toxic to Microbes stocks. Structural information

A
cetyl side groups from hemicellulose biomass polymers are released on these polymeric substrates is
during current pretreatment steps. These small acetyl molecules necessary, and mechanistic models
often are toxic and inhibit the microbial activity that converts sug- must be developed to identify
ars to ethanol. Hemicellulosic components such as xyloglucan and gluc- “bottlenecks” in hemicellulose
uronoarabinoxylan and pectic cell-wall components often are O-acetylated. bioconversion (see sidebar, Opti-
For instance, O-acetyl groups may be present on the glucan backbone of mizing Hemicellulose Acetylation
xyloglucan or on galactose or arabinose residues of side chains. The degree in Cell Walls, this page).
of sugar-residue O-acetylation of pectins varies from 0 to 90% depend- A systematic approach to under-
ing on the tissue, species, and method of preparation. The role of O-acetyl standing these factors will
substituents in vivo is not known, but in vitro experiments suggest that one promote more effective use of
function may be their involvement in hindering enzymatic polysaccharide lignocellulosic biomass in biocon-
breakdown. O-acetyl substituents also affect polysaccharide solubility and version systems. Fortunately, we
pectin’s gelation properties (Pauly and Scheller 2000). now have new biological, physi-
Plant genes exhibit weak sequence similarity to putative bacterial acetyl- cal, analytical, and mathematical
transferase genes. Genetic tools in plants such as Arabidopsis will enable tools that can help in reliably
the identification of gene products catalyzing polysaccharide acetylation identifying and quantifying the
and the determination of acetylation’s role in cell-wall structure and func- relative importance of various
tion. Such studies will provide insights into the possibility of developing potentially limiting factors. We
biomass crop varieties with significantly reduced polysaccharide acetyla- also have tools to identify and
tion and thus improving the fermentation process. optimize facilitating factors, for
example, through plant breeding.
46 Biofuels Joint Roadmap, June 2006 • Office of Science and Office of Energy Efficiency and Renewable Energy • U.S. Department of Energy
The goal is to provide a rational basis for design of practical, effective, and
economical pretreatments, including controlling the physical modification
of native celluloses and related cell-wall polysaccharides during thermal
and chemical treatments. Current thermochemical treatments ultimately
will be replaced with more benign enzymatic treatments to the degree
feasible. Necessary detailed analyses are discussed in the chapter, Decon-
structing Feedstocks to Sugars, p. 85.

Optimization of Plant Cell Walls

Optimal efficiency of biofuel production depends on maximizing fuel yield
from a unit of biomass and minimizing energy inputs. The plant cell wall, a
complex assembly that plays a primarily, but not exclusively, structural role
during plant growth and development, may be particularly amenable to the
application of engineering principles in redesigning the cell wall to meet
energy needs. To breed plants in which cell-wall composition is optimized
for conversion efficiency, understanding how cell walls are made, how
composition is regulated, and the roles of various polymers in supporting
plant growth and development will be necessary. The long-term goal is to
develop a systems-level understanding to facilitate rational improvement
of plant cell-wall composition in dedicated energy crops. Such knowledge
of plant cell walls is in a very primitive stage because of scientific and
technical challenges that have impeded scientific progress. Future research
on cell-wall synthesis and function requires interdisciplinary approaches
ranging from genomics to synthetic carbohydrate chemistry and biophys-
ics. Model organisms are important in facilitating advances in basic biology
and in bringing the most sophisticated biological tools to the problem.
Several new plant models closely related to species selected for energy
crops are advocated. A powerful first step is to obtain comprehensive DNA
sequences for these organisms.

Understanding Cell-Wall Structure and Function

Increasing the production of biofuels begins with increasing biomass pro-
ductivity, either by making more cell walls or making cell walls with more
carbon. In addition, changes in cell-wall composition could have major
effects on the efficiency with which biomass can be converted to fuels; rela-
tive amounts of certain sugars could be increased or wall polymers could be
made more amenable to enzymatic hydrolysis, thus improving the yield of
sugars delivered to the biorefinery as raw feedstock.
Important questions remain about the structures of cell-wall polymers, how
they are made, and their functions in plant growth and development. To
optimize the amount, composition, and structure of walls for biofuel produc-
tion, we must identify the genes involved in synthesis of cell-wall polymers,
the design principles for cell walls, and factors that control the amounts and
organization of various types of enzymes and resultant polymers. Prelimi-
nary evidence suggests that cell-wall biophysical properties important to
plant growth and development may be achieved in many different ways with
regard to chemical composition. Thus, cell-wall composition of energy crops

Biofuels Joint Roadmap, June 2006 • Office of Science and Office of Energy Efficiency and Renewable Energy • U.S. Department of Energy 47
 LIGNOCELLULOSIC BIOMASS

probably can be altered so they are better suited for fuel production. Desir-
able improvements include increasing the amount of such useful polysac-
charides as cellulose or certain hemicelluloses and minimizing the content of
such undesirable components as lignin or acetyl groups.
Evidence indicates that photosynthetic CO2 fixation is regulated by plants
in response to demand for fixed carbon, so understanding photosynthate
flux into cell-wall polymers relative to other pathways of primary carbon
metabolism and storage is important. Understanding mechanisms that
regulate carbon flux and synthesis of various polysaccharides may make
possible the development of plants that accumulate significantly more
polysaccharide per cell. Expected significant increases in the ratio of
carbon to nitrogen and mineral nutrients would have a beneficial effect on
agricultural inputs (e.g., planting, fertilizing, cultivating, and harvesting),
costs, and sustainability.
Progress in this area requires broad approaches to achieve a foundation
of knowledge about cell-wall structure and function that will be the basis
for a systems approach to predicting and controlling biomass composi-
tion. Before a systems approach can be implemented, a comprehensive
understanding is needed about what reactions are performed by the many
hundreds of enzymes involved in cell-wall synthesis and deposition, where
and when relevant genes are expressed, and what genes control expression
and activity of proteins involved in polysaccharide and lignin synthesis
and modification. Indeed, one “grand challenge” in systems biology may
be understanding how to engineer cell walls that meet the need of chemi-
cal biorefineries for optimized feedstocks yet still meet the plant’s need for
development, robustness, and maximal rates of growth.
GTL capabilities could provide extensive support for research on cell-wall
synthesis, structure, and function. Sequencing support for model organisms
(see below) and for identifying relevant genes in energy crops is an imme-
diate goal. The development of populations of transgenic experimental
plants with epitope-tagged proteins would greatly facilitate the determina-
tion of subcellular protein localization and the application of proteomic
techniques to identify protein complexes. DNA chips, in conjunction with
advanced genetic technologies, can be used for a systems-level understand-
ing of transcriptional control of cell-wall synthesis and modification path-
ways. Epitope tagging also may be used to facilitate mRNA purification
from single cells, facilitating insights into processes specific to cell types.
Ultimately, GTL capabilities in systems analysis will permit an integrated
systems model that can be used to support directed modification of cell
walls for specific applications.
Efforts to understand and modify cell walls need to be coordinated with
bioconversion and plant cell-wall deconstruction initiatives to optimize
feedstock composition based on pretreatment and conversion methods
and effects. These objectives also need coordination to develop analytical
and visualization methods, computational facilities, and organic-chemistry
methods for production of enzyme substrates and standards used in phe-
notyping and gene characterization.
48 Biofuels Joint Roadmap, June 2006 • Office of Science and Office of Energy Efficiency and Renewable Energy • U.S. Department of Energy
Control of Lignin Synthesis and Structure
Although lignin is not converted readily to ethanol, lignin biomass may be
amenable to chemical or thermal processing to achieve such liquid fuels as
low-grade diesel or fuel oil. One aspect of optimizing biomass composi-
tion for ethanol production is minimizing lignin content. Alternatively,
developing plants with modified lignin that can be removed easily during
biomass processing may be possible.
Lignin is a complex aromatic polymer associated with polysaccharides in
secondary cell walls (see Fig. 3, p. 44, and Fig. 4. Phenylpropanoid Pathway
Leading to Lignin Biosynthesis in Plants, this page). Lignin constitutes
a significant barrier in biomass conversion to fuels by inhibiting enzyme
access to polysaccharides and by releasing toxins during degradation that
inhibit organism growth during fermentation of cell-wall hydrolysates to
ethanol. Genetic studies have indicated that lignin reductions may cause
deleterious changes in plant growth and development. However, lignin
possibly may be reduced with or without harmful effects on plant growth
if compensating changes could be made in the amount of cell-wall poly-
saccharides. Some early experiments are under way. The degree to which

Fig. 4. Phenylpro-
panoid Pathway
Leading to Lignin
Biosynthesis in
Plants. Horizontal
reactions are ring
modifications;
vertical reactions are
side-chain modi-
fications. [Figure
source: C. Fraser
and C. Chapple,
Purdue University]

4CL, 4-(hydroxy)cinnamoyl CoA ligase; C3'H, p-coumaroyl shikimate/quinate 3'-hydroxylase; C4H, cinnamate 4-hydroxylase; CAD, .
cinnamyl alcohol dehydrogenase; CCoAOMT, caffeoyl CoA O-methyltransferase; CCR, cinnamoyl CoA reductase; COMT, caffeic acid/.
5-hydroxyferulic acid O-methyltransferase; F5H, ferulate 5-hydroxylase; HCALDH, hydroxycinnamaldehyde dehydrogenase; HCT,
hydroxycinnamoyl-CoA shikimate/quinate hydroxycinnamoyltransferase; PAL, phenylalanine ammonia-lyase.

Biofuels Joint Roadmap, June 2006 • Office of Science and Office of Energy Efficiency and Renewable Energy • U.S. Department of Energy 49
 LIGNOCELLULOSIC BIOMASS

cellulose amount can be increased with or without simultaneous changes in

hemicellulose content and composition must be ascertained.
Exploring lignin biosynthesis and its regulation in a comprehensive fash-
ion may make possible the formulation of methods for limiting and alter-
ing lignification to maximize biomass-to-energy conversion. For instance,
the gene for ferulate-5-hydroxylase has been used to increase the syringyl
monomer content of poplar lignin (see sidebar, Optimizing Lignin Compo-
sition for More Efficient Bioethanol Production, p. 43). The resulting trees
had normal growth and development, but the pulping time was reduced
by more than 60% (Huntley et al. 2003). Similarly, opportunities exist to
modify lignification cell specificity so its impact on energy conversion can
be minimized. A goal is to create a lignification toolbox to manipulate
polymer depositions genetically and analyze the impact of those manipula-
tions with advanced analytical organic chemistry. Such detailed knowledge
could create novel opportunities for fundamentally changing how biomass
is synthesized and subsequently processed for biofuels. For instance, novel
monomers might be incorporated to generate lignins with unique, useful
chemistries—readily cleavable linkages that could facilitate lignin depoly-
merization under more benign conditions (i.e., with enzymes).

Improved Methods, Tools, and Technologies

New analytical methods, tools, and technologies will accelerate the under-
standing of cell-wall synthesis, makeup, structure, and function and will
speed breeding or rational modification of energy crop varieties.
At the basic research level, new and improved methods are needed to ana-
lyze wall composition and nanoscale structure. Ideally, these methods could
be applied to analysis of a small number of cells. Molecules in cell walls
range from 2 to 5 angstroms (0.2 to 0.5 nm) in diameter (i.e., a polysac-
charide chain) and to many microns in length. Primary cell walls are from
50 to 100 nm in thickness and, in some cells, are thought to be chemically
differentiated from one side to another. New imaging modalities that take
advantage of various chemically specific imaging tags will support the long-
term vision of in situ images of living plant cell walls. Images will reveal key
molecular processes occurring in real time during the full life cycle of cell-
wall formation, maturation, transformation, dehydration, and processing
into simple feedstocks. The understanding obtained through research using
such imaging is expected to result in quantitative, predictive modeling as a
guide to formulating advanced feedstocks and their subsequent processing.
A systematic approach is required to identify plant biomarkers and specific
antibodies or other molecular tags useful in feedstock improvement.
Poorly understood now, the fine structure of intact walls must be studied to
determine how the parts fit together to comprise the whole wall’s physical
properties. Some aspects of the general problem may be resolved simply by
encouraging the application of such existing methods as very high resolution
electron and scanning probe microscopy (see sidebar, Image Analysis of Plant
Cell Surfaces at the OBP Biomass Surface Characterization Lab, p. 40).

50 Biofuels Joint Roadmap, June 2006 • Office of Science and Office of Energy Efficiency and Renewable Energy • U.S. Department of Energy
Similarly, greater use of nuclear magnetic resonance (NMR) and mag-
netic resonance imaging may allow the development of 2D and 3D maps
of cell-wall composition from important experimental and production
species such as Arabidopsis and poplar. Use of NMR may be expanded by
isotopic labeling and further development of solvents capable of dissolv-
ing cell-wall components. Complete annotation of 2D maps could
facilitate greatly the analysis of genetic variation in cell-wall composition
and the assignment of function to genes implicated in wall biosynthesis
and modification.
Other approaches meriting investment include expanded collections of
antibodies and aptamers to cell-wall components, the use of enzyme-
based polysaccharide fingerprinting, pyrolysis gas chromatography–mass
spectrometry (GC-MS), and related methods. A challenge is associated
with characterizing enzymes that synthesize cell-wall polysaccharides:
Many enzymes add sugars to preexisting polysaccharides (i.e., “accep-
tors” or “primers”) that are not readily available as standards and reagents.
Focused investments in carbohydrate chemistry will be required to
construct substrates—including labeled substrates—for measuring the
activity of many wall biosynthetic enzymes. Expertise in carbohydrate
synthetic chemistry also would be a needed complement to proteomic
and metabolomic capabilities envisioned in GTL capability suites.
Expanded capabilities in synthetic carbohydrate chemistry could open
up new high-throughput methods for characterizing carbohydrate-active
enzymes based on high-density and high-diversity “glycochips.” In this
method, the activity or binding of a target protein could be evaluated
simultaneously with hundreds or thousands of potential substrates and
very small amounts of reagent.
High-throughput methods of cell-wall analysis are needed for plant
breeding and improvement, allowing timely analysis with the most
sophisticated analytical techniques. Ultimately, infield characterization
is required to support breeding, molecular marker mapping, and studies
involving such environmental variables as fertilizers and various biotic
and abiotic stresses. Methods must be accurate and relatively inexpen-
sive for the large numbers of samples typically handled during a breed-
ing program. Additionally, they should be applicable to a wide variety
of materials, from corn stover to wood. In principle, a high-throughput
sample analysis may be enabled by detailed analysis of the relationship
between cell-wall composition and features of Fourier transform infrared
spectroscopy spectra or pyrolysis GC-MS chromatograms, combined
with computational methods.

Technical Milestones
Within 5 years
• Develop rapid, accessible tools and methods for consistent biomass
compositional analysis in bulk and fractions (see section, Characterizing
Cell Walls Using High-Throughput Methods, p. 108).

Biofuels Joint Roadmap, June 2006 • Office of Science and Office of Energy Efficiency and Renewable Energy • U.S. Department of Energy 51
 LIGNOCELLULOSIC BIOMASS

• Identify genes for enzymes that catalyze synthesis of major polysaccha-

ride backbones.
• Identify a substantial fraction of enzymes that catalyze synthesis of
polysaccharide sidechains and determine sidechain biological function
in model plant species.
• Identify enzymes that acetylate polysaccharides, and establish biological
function for such modifications.
• Identify genetic regulatory factors that control lignin synthesis and
deposition.

Within 10 years
• Clarify regulation of polysaccharide biosynthesis, including key steps
that regulate carbon flow from photosynthesis into cell-wall polymers.
• Define mechanisms that control cellulose amount and fibril length and
angle.
• Modify celluloses with altered numbers of glycan chains in secondary
walls, and produce and test them in model species.
• Make available for testing biomass crop plants with decreased lignin
and increased amounts of cellulose or other polysaccharides.
• Develop new tools and methods to help us understand cell-wall struc-
ture, including highly parallel computational simulations and high-
sensitivity 2D NMR and MS instrumentation for analysis of lignin in
small tissue samples.
• Identify all genes that catalyze synthesis of polysaccharide sidechains.

Within 15 years
• Determine regulatory genes that control amounts of major polysaccha-
rides, including cellulose.
– Develop methods for manipulating polysaccharide composition of
any particular cell type within a specific tissue.
• Make available plants with improved wall composition. These plants will
have increased yields of fermentable sugars, requiring less costly prepro-
cessing; cell-wall degradation will result in insignificant levels of inhibi-
tory compounds (in the fermentation process).
• Develop a detailed model of lignin monomer transport, polymerization
initiation, and the interactions of lignin polymers with polysaccharide
components of the plant cell.

52 Biofuels Joint Roadmap, June 2006 • Office of Science and Office of Energy Efficiency and Renewable Energy • U.S. Department of Energy
 A First Step to Optimizing Feedstocks
for Fuel Production

O
ptimizing plant biomass for more efficient processing requires a better understanding of
plant cell-wall structure and function (see next two pages). Plant cell walls contain long
chains of sugars (polysaccharides) that can be converted to transportation fuels such as
ethanol. The saccharification process involves using enzymes to break down (hydrolyze) the poly-
saccharides into their component sugars for fermentation by microbes to ethanol (see sidebar, From
Biomass to Cellulosic Ethanol, p. 26). Significant challenges for efficient conversion are presented
by both the large number of enzymes required to hydrolyze diverse sugar linkages and the physical
inaccessibility of these compounds to enzymes because other cell-wall components are present.
Plant cell walls contain four different polymer types—cellulose microfibrils, hemicelluloses,
pectins, and lignins. Microfibrils perform an important role in strengthening cell walls, thus
providing support to the overall plant body. Some properties of lignin, however, interfere with
enzymatic conversion of polysaccharide components. Additionally, since lignin is not readily
converted to ethanol, we must find other ways it can be used if we are to maximize the yield of
energy from biomass.
Several thousand genes are estimated to participate in cell-wall synthesis, deposition, and func-
tion, but very few genes have been identified and very little is known about their corresponding
enzymes. Many questions remain, for example, regarding how polysaccharides and lignin are
synthesized, how wall composition is regulated, and how composition relates to the biological
functions of cell walls. To answer these questions, we need to discover the functions of many
hundreds of enzymes, where proteins are located within cells, whether or not they are in com-
plexes, where and when the corresponding genes are expressed, and which genes control the
expression and activities of proteins involved. Application of new or improved biological, physi-
cal, analytical, and mathematical tools will facilitate a detailed mechanistic understanding of
cell walls. That knowledge will permit optimization of various processes involved in producing
biomass and converting it to fuels.
Major opportunities exist to increase productivity and conversion-process efficiencies by alter-
ing fundamental aspects of plant growth, development, and response to biotic and abiotic stress.
Altering cell-wall composition to increase the relative amount of cellulose and to decrease lignin,
for example, could have significant effects (see sidebar, Optimizing Lignin Composition for
More Efficient Bioethanol Production, p. 43). Eventual development of a comprehensive physi-
ological cell-wall model incorporating biophysical aspects with structural properties and knowl-
edge of proteins involved will aid in rational development of highly productive feedstock species
whose cell walls are optimized for conversion.

Biofuels Joint Roadmap, June 2006 • Office of Science and Office of Energy Efficiency and Renewable Energy • U.S. Department of Energy 53
54 Biofuels Joint Roadmap, June 2006 • Office of Science and Office of Energy Efficiency and Renewable Energy • U.S. Department of Energy
Biofuels Joint Roadmap, June 2006 • Office of Science and Office of Energy Efficiency and Renewable Energy • U.S. Department of Energy 55
 LIGNOCELLULOSIC BIOMASS

Cited References
Huntley, S. K., et al. 2003. “Significant Increases in Pulping Efficiency
in C4H–F5H– Transformed Poplars: Improved Chemical Savings and
Reduced Environmental Toxins,” J. Agric. Food Chem. 51(21), 6178–83.
Pauly, M., and H. V. Scheller. 2000. “O-Acetylation of Plant Cell Wall
Polysaccharides: Identification and Partial Characterization of a Rhamno­
galacturonan O-Acetyl-Transferase from Potato Syspension-Cultured
Cells,” Planta 210(4), 659-67.

56 Biofuels Joint Roadmap, June 2006 • Office of Science and Office of Energy Efficiency and Renewable Energy • U.S. Department of Energy