Timeline of BESC Accomplishments

Citations available via the Publications page.

Jump to: Year 0 | Year 1 | Year 2 | Year 3 | Year 4 | Year 5 | Year 6 | Year 7 | Year 8 | Year 9 | Year 10

Top Accomplishments Overall

(taken from Gilna et al. 2017 Biotechnology for Biofuels, Special BESC Issue; in press)

Biomass Formation. Populus and switchgrass (Panicum virgatum) were the chosen feedstocks for studies of cell wall related genetic modifications that could impact recalcitrance and inform understanding. Both are high-yield perennials recognized as potential domestic biofeedstocks [U.S. DOE 2006; U.S. DOE 2011], Populus was the first sequenced woody feedstock [Tuskan 2006].  Switchgrass is a native herbaceous perennial that could grow on marginal land. Both were deemed tractable for studies aimed at determining the basis of, and ameliorating, recalcitrance. Key advances in biomass formation led by BESC include:

Field trials of
poplar and
switchgrass switchgrass species

  • Significant advances were made in understanding, manipulating, and managing plant cell wall recalcitrance and conversion. Researchers showed that multiple plant genes control cell wall recalcitrance, and that manipulation of these genes can yield lower recalcitrance perennial bio-feedstocks [Mohnen 2017]. This included increased understanding of the cell wall structure and biosynthetic pathways for lignin, xylan, cellulose and surprisingly pectin and their resultant effects on recalcitrance [Shen 2013; Zhao 2013; Barros 2016; Peña 2016; Urbanowicz 2014; Mazumder 2012; Kulkarni 2012; Mazumder 2010; Pattathil 2010; Bar-Peled 2012; Urbanowicz 2012; Bali 2016; Vandavasi 2015; Payyavula 2014; Tan 2013; Atmodjo 2013; Biswal 2015].
  • BESC led large-scale campaigns to understand natural variation in both switchgrass and Populus. This included both high-throughput (HTP) recalcitrance phenotyping and sequencing along with other ’omics for these natural variants (as well as generated transgenics). These resulted in advances in genome-wide association studies (GWAS) [Serba 2015; Muchero 2015; Evans 2014].
  • Greenhouse and field trials were conducted for a limited number of Populus and switchgrass lines with reduced recalcitrance arising from both directed transgenics and natural variants. A collective “TOP Line” experimental design protocol was utilized [Nelson 2017] with multiple phenotypic characterization assays developed by the Enabling Technology groups. These included, for example, sugar release, sugar and lignin composition, ethanol production, crystallinity.  A key discovery was the ability to achieve both lower recalcitrance and higher biomass simultaneously in certain lines [Biswal 2015; Baxter 2015; Baxter 2014; Dumitrache 2017].
  • One goal of BESC was to understand the molecular basis of recalcitrance. The reduced recalcitrance feedstock biomass generated in BESC was analyzed by a series of chemical, biochemical, molecular, and systems biology approaches. The outcome was the identification of multiple wall polymers whose modified abundance or structure could be engineered to reduce feedstock recalcitrance [Mohnen 2017].  The results begin to provide mechanistic understanding of the molecular bases of recalcitrance.
  • From this research, a path for improving feedstocks by cisgenic manipulations, by selecting the best natural variants, or by genetically assisted breeding can be seen [Kalluri 2014].

Biomass Conversion. One-step consolidated bioprocessing (CBP) without added enzymes [Lynd, 1996] was the central focus of BESC’s work in the conversion area, which featured both fundamental and applied components. BESC initially focused on two approaches: improving product-formation in thermophilic cellulolytic bacteria [primarily Clostridium thermocellum (C. thermocellum) and Caldicellulosiruptor bescii (C. bescii)] and conferring to yeasts the ability to ferment cellulose by virtue of heterologous expression of glycosyl hydrolases. Investigators came to regard the former approach as more promising and by the end of BESC were focused exclusively on this path. Key conversion advances led by BESC included:

researcher with Applikon fermenter

Optimize Microbes for Biomass Deconstruction and Conversion

  • Large differences were found among the biocatalysts in the most comprehensive comparative evaluation to date of biomass deconstruction under controlled conditions. Among the biocatalysts tested, thermophilic anaerobes and specifically C. thermocellum achieved the highest carbohydrate solubilization yields which were several-fold higher yields than industry-standard fungal cellulase [Paye 2016; Lynd 2016; Chung 2014a].
  • Investigators developed and improved the genetic tools for thermophiles, most notably C. thermocellumand Caldicellulosiruptor spp., and use of these tools to initiate metabolic engineering of these non-model microbes [Cha 2013; Guss 2012; Olson 2012; Lipscomb 2016].
  • Substantial advances were made in understanding and manipulating the metabolism of target CBP microbes.  Zhou et al. [Zhou 2013] described non-standard glycolysis in C. thermocellum.  Thermoanaerobacter saccharolyticum (T. saccharolyticum) was improved toproduce economically-recoverable ethanol concentrations at near-theoretical yield in hemicellulose-fermenting [Herring 2016]. Iso-butanol was produced by adding key pathway enzymes in modified C. thermocellum at unprecedented yields and titers [Lin 2015].  Ethanol titer and yield were increased in C. thermocellum by elimination of side-products [Papanek 2015; Tian 2016; Biswas 2016].
  • BESC identified the specific deconstruction enzymes which target the major biopolymers of lignocellulosic biomass. Work on enzyme fundamentals emphasized multi-functional cellulases based on the enzymes found in Caldicellulosipruptor species and C. thermocellum [Xu 2016]. CelA, a multifunctional glycosyl hydrolase from C. bescii, was shown to be a particularly powerful hydrolytic enzyme despite being inhibited by the presence of lignin [Brunecky 2013; Kim 2016].
  • BESC demonstrated that C. thermocellum is capable of active fermentation in the presence of mechanical milling — an approach referred to as co-treatment [Paye 2016; Lynd 2017; Holwerda 2017]. With co-treatment, C. thermocellum was able to achieve greater than 85% carbohydrate solubilization for Populus and switchgrass in the absence of added enzymes and thermochemical pretreatment implying that C. thermocellum can attack all the major chemical linkages in representative woody and herbaceous lignocellulose crops given sufficient physical access.

Enabling Technology. Enabling technologies was organized to develop and apply cutting-edge analytical methodologies to characterize biomass as well as its conversion. There was also significant ’omic and computational biology of the modified plants and microbes to help improve metabolic models.  The resulting data were used to create new insights into how biomass structure and chemistry affects recalcitrance during CBP or pretreatment. These efforts included analyses of partially digested solid residues from CBP.

  • The development of high throughput methods for rapid analysis of pretreatment and enzymatic hydrolysis allowed for rapid identification of low recalcitrant plant lines from thousands of natural and transgenic variants.  These low recalcitrant plant lines then could be characterized using multiple analytical and ’omic approaches which rapidly advanced BESC’s deeper understanding of the recalcitrance phenotype [Studer 2011; Decker 2015; Selig 2011].
  • Increased understanding of recalcitrance was supported by developing techniques such as glycome profiling [Pattathil 2012], and improving the use of nuclear magnetic resonance spectroscopy (NMR) for biomass [Foston 2012; Pena 2016; Pu 2016; Pu 2013; Kataeva 2013].
  • BESC also supported the development of new ways to image the chemical components comprising the cell wall. Raman spectroscopy was used to image hemicellulose for the first time [Zeng 2016].  Modified atomic force microscopy (AFM) techniques were used to chemically image the cell wall at the submicron level [Tetard 2011].  Quantitative fluorescence confocal laser scanning microscopy (CLSM) and surface spectroscopy by Time of Flight Secondary Ion Mass Spectroscopy (ToF-SIM) showed that, following microbial digestion, the decrease in surface cellulose while surface lignin increased.  This indicates that biomass recalcitrance may be controlled by surface characteristics [Dumitrache 2017].
  • Co-solvent enhanced lignocellulosic fractionation was developed as a new pretreatment which removes significant amounts of lignin and increases enzymatic digestibility of biomass [Nguyen 2015].
  • The center was able to provide integrated ’omics data for key processes.  Integrated ’omics of microbial growth on complex lignocellulosic biomass over time provided a detailed view of the molecular machinery (metabolites and enzymes) and revealed temporal adaptation to a complex, lignocellulose substrate [Poudel 2017]. Profiling genotype-specific proteomes derived from ribonucleic acid sequencing (RNA-Seq) data better defined the link between genotypes and phenotypes in Populus [Abraham 2015].
  • Lignin has been shown to play a key role in biomass recalcitrance [Ziebell 2016; Ziebell 2010; Li 2016]. The potential removal or recovery of lignin would allow its valorization [Ragauskas 2014] whether into fiber [Sun 2016] or into value-added intermediates [Beckham 2016].

Year 10 (Oct. 2016 – Sept. 2017)


Genome-Wide Association Study (GWAS) Dataset Released (Details)


  • Obtained crystal structure and elucidated mechanism of a glycosyl transferase (FUT1) involved in wall hemicellulose synthesis [Urbanowicz 2017] and solved the structure of XOAT1, a xylan 2-O-acetyltransferase, representing the archetypical structure of a plant acetyltransferase and the first structure of an enzyme involved in xylan biosynthesis.
  • Demonstrate CBP with cotreatment to allow >85 carbohydrate solubilization cotreatments intermittent milling during the CBP fermentation [Balch 2017; Lynd 2017].
  • Identified mechanism for reduced recalcitrance and increased growth of GAUT4-KD Populus and switchgrass as due to reduction of specific pectic homogalacturanan involved in calcium and RG-I-borate cross-linking in the plant cell wall.
  • Completed three-year field trial of down-regulated COMT switchgrass showing that reduced lignin content and biomass recalcitrance are stable in the field for at least three seasons [Li 2017].
  • Identified sets of transcription factors that control senescence and nitrogen remobilization in switchgrass.
  • First agronomic and physiological assessment of transgenic switchgrass engineered for three transgenes with different cell wall targets in a field ‘best case scenario’ in Maricopa, Arizona, and analysis of the biomass for cell wall traits.
  • EPSP synthase-like gene is a transcriptional regulator of the phenylpropanoid pathway. Allelic variants of this gene yield biomass with reduced lignin and increased sugar release and ethanol yield from non-pretreated biomass.  Uncovered a novel EPSP-transcription factor based regulatory circuit regulating phenylpropanoid biosynthesis and biomass recalcitrance.  This gene was licensed to two companies.
  • Obtained multiple lines of evidence, including knockdown and overexpression in Populus, to propose model that GAUT12 synthesizes a 4-linked GalA glycan required for xylan synthesis and whose knockdown expression leads to reduced recalcitrance, reduced wall integrity, and enhanced growth in Populus, with inverse phenotypes in overexpression lines [Biswal 2015].
  • Mechanistic studies revealed that biomass recalcitrance of COMT down-regulated switchgrass is related to the decreased lignin content and increased biomass accessibility, while cellulose crystallinity and degree of polymerization and hemicellulose molecular weight did not contribute significantly to recalcitrance [Li 2017].
  • Identified a calmodulin binding protein (IQD) as having a novel role in cellulose and secondary cell wall biosynthesis via transgenesis and DNA-binding assays, and showed it is under the influence of the secondary cell wall master regulator, SND1.
  • Detected in vivo linkages between pectin and arabinogalactan-proteins, providing the first in vivo structural data for plant cell wall matrix glycan proteoglycans.
  • First-time demonstration of dimerization of apiogalacturonan-containing proteoglycans through formation of 1:2 borate-diol esters.
  • Completed BESC TOP Line switchgrass field experiments in Knoxville, Tennessee, yielding data proving multiple mechanisms for decreasing recalcitrance in this feedstock.
  • Discovered the impact of cellulose modification (via KOR gene) on the host Populus interactions with a beneficial microbe (Laccaria).
  • Identified Angustifolia-based regulatory machinery regulating growth-defense tradeoffs.
  • Completed field trials for Populus natural variants and transgenic TOP Lines and assessed their agronomic performance.
  • Enzyme engineering
  • Confirmed that the engineering of the scaffold in domain of cellulosomes results in greater than parental type activity on biomass.
  • Showed that CelA is cellulose crystallinity agnostic, in that it performs as well on high as low crystallinity biomass [Brunecky 2017].
  • Successfully engineered fusion pyruvate decarboxylase (PDC)-ADH enzymes allowing expression in thermophiles and production of ethanol transforming new microbes.
  • Tests were initiated to develop a systematic approach and transforming a new microbe DNA methylation patterns and enzymes.  This has improved transformation efficiencies in several difficult bacteria.  Systematic approach to transforming new microbes.
  • Microbiome were characterized and compared to pure cultures.
  • Established long running thermophilic microbiomes at different switchgrass substrate loadings and at different retention times. At 30 g/L loadings under all tested retention times (decreasing from 20 to 3.3 days) stable methanogenesis occurred. Firmicutes were predominant and at short residence times the primary biocatalyst was Clostridium clariflavum.
  • Compared batch fermentations of pure cultures of C. thermocellum with cultures inoculated from a long running thermophilic microbiome at 30 g/L switchgrass loadings.  At three different incubation times (60, 120 and 530 hrs) C. thermocellum was as good or better in solubilizing substrate compared to the microbiome.
  • Expressed PDC from several different organisms in C. thermocellum. A strain expressing the PDC from Acetobacter pasteurianus produced ethanol at a yield of 70% of the theoretical maximum and 21 g/l titer.  Although enzyme activity in-vitro was observed, it was much lower than expected, which is believed to be due to problems with solubility [Olson 2015b].
  • Meeting year 10 isobutanol goal
  • Acetohydroxyacid synthase (AHAS) was identified as the most limiting step in the C. thermocellum isobutanol production.
  • Native AHAS isoenzymes were studied and overexpressed in the engineered C. thermocellum, resulting in the improvement of isobutanol production.
  • Year 10 milestone was achieved: 10.1 g/L of isobutanol was directly produced from cellulose using C. thermocellum.
  • Controlled fermentation and integrated ’omic platforms provided complementary systems biological information that identify C. thermocellum has previously undescribed specific responses to cytotoxic inhibitors released during the deconstruction and utilization of switchgrass. The most striking feature of the metabolomic response was the observed accumulation of long-chain, branched fatty acids over time, implying an adaptive restructuring of C. thermocellum’s cellular membrane as the culture progresses. Accumulation of hemicellulose-derived sugars and sugar alcohols concomitant, with increased abundance of enzymes involved in C5 sugar metabolism/pentose phosphate pathway, indicated that C. thermocellum shifted glycolytic intermediates to alternate pathways to modulate overall carbon flux in response to C5 sugar metabolites that increased during lignocellulose deconstruction [Verbeke 2017b].
  • Coupling metabolomic analyses and extensive phenotype characterization with SNP variation in the large Populus trichocharpa GWAS population has resulted in the generation of a vast co-expression network that has led to the identification of metabolite-gene associations, allowing the functional annotation of previously unknown or partially annotated genes, providing information on the identity of unknown metabolites via gene association, and information on putative biosynthetic pathways for Populus metabolites.
  • Metabolomic analyses were completed on all 12 greenhouse-grown switchgrass TOP Lines, including ERF/SHINE overexpressed, GA2ox OE, GAUT1 KD, IRX10-1 KD, AMP1 natural variant (NV), AMP2 NV, PvCCR1 KD, PvMYB4 OE, COMT KD, miRNA156 OE, GAUT4 KD, and FPGS1 KD.  The compellation of these results indicates key gene-specific phenotypes associated with reduced recalcitrance lines.
  • Composition measurements, material balances, cellulase adsorption, and confocal scanning light microscopy (CSLM) and transmission electron microscopy (TEM) imaging revealed that the enhanced deconstruction of CELF solids by enzymes and particularly by C. thermocellum could be related to lignin removal and alteration, thereby reinforcing that these factors are key contributors to biomass recalcitrance, the primary barrier to low cost biological conversion to sustainable fuels.
  • Developed and implemented NMR methods to identify and quantify soluble components of the fermentation broths produced by microorganisms growing on lignocellulosic biomass. Notably, these methods resulted in the identification of specific xylan structures in corn fiber that are recalcitrant to deconstruction by C. thermocellum (and other micro-organisms). Knowing these explicit structures allowed us to identify enzymes that specifically hydrolyze them, which in turn allowed the Lynd group to engineer more efficient micro-organisms for the industrial bioconversion of corn fiber.
  • Completed cell wall chemistry and recalcitrance characterization of the switchgrass greenhouse TOP Line experiments.
  • Used rigidity percolation theory to show that plant cell wall recalcitrance is an emergent property that is explained by considering the mesoscale property of network connectivity determined from the molar concentrations of the key cell wall polymers.  Describing recalcitrance as an emergent property of biomass recalcitrance provides a path to rationally design biomass feedstocks tailored for both low recalcitrance and improved agronomic performance.
  • Elucidated overflow metabolism and growth cessation in C. thermocellum during high-substrate loading fermentation by employing genome-scale metabolic flux and thermodynamic analyses [Thompson 2017].
  • Pentose sugars, their alcohols, and xylo-oligomers significantly inhibit C. thermocellum metabolism and growth.  Stable isotope metabolomics confirmed C. thermocellum’s ability to transport and metabolize pentose sugars, occurring, in part, through the ATP-dependent transporter, CbpD. Xylose is an electron sink for C. thermocellum metabolism leading large changes in gene and protein expression and to the production of xylitol. Xylose-induced inhibition corresponds with the up-regulation and biogenesis of an AgrD-type, lactone cyclized pentapeptide signaling molecule; which is the first report of an AgrD-type signaling peptide in any thermophile.
  • Greenhouse-grown transgenic Populus deltoides lines, down-regulated in lignin pathway genes, validated by metabolomic analyses, were demonstrated to have biomass with decreased recalcitrance (C4H, 4CL, CCOMT, CCR, CAD) and increased recalcitrance (F5H) to enzymatic sugar release, but only the CAD KD line had improved growth relative to empty vector control plants.
  • A multi-’omics study of Populus deltoides subjected to drought demonstrated that the nature of the progression of drought stress (acute vs cyclic) drives differential transcriptomic, proteomic, and metabolomic responses that dictates the degree of osmotic adjustment to drought and the nature of solutes that accumulate.
  • Developed a versatile metabolic modeling framework for rapid strain design for enhanced cell growth and biofuels production of C. thermocellum.
  • Showed that incorporation of coumaryl alcohol influences the structure and composition of lignin dehydrogenation polymers (DHP). Reactions between H and G monomers form G:H polymers that have lower average molecular weights relative to the G-based polymers. Solid-state NMR and pyrolysis-MBMS analyses showed that the H monomer self-polymerizes to produce clusters of H-based polymer that are segregated from of clusters of G- or S:G-based polymers.  The experimental results support previous theoretical predictions for the reactivity and structural influences of the H monomer on lignin-like polymers [Harman-Ware 2017].

Year 9 (Oct. 2015 – Sept. 2016)


Demonstrated further evidence that down-regulation of pectin biosynthetic genes has a major impact on recalcitrance in woody or grass feedstocks.

Reported results from multiple multi-year field trials of either genetically modified or natural variant Populus and switchgrass lines that continue to exhibit reduced recalcitrance and stable, or in some cases improved, agronomic performance.

Elucidated a third modality of cellulase action exhibited by C. thermocellum, contributing to its ultra-high performance on cellulose.

BESC brochure
(October 2015)

Achieved greater than 85% carbohydrate solubilization for some lines of Populus and switchgrass using consolidated bioprocessing with cotreatment in the absence of added enzymes or thermochemical pretreatment implying that C. thermocellum is able to attack all the major chemical linkages in representative woody and herbaceous lignocellulose crops given sufficient physical access.

Reported further evidence that lignin syringyl to guaiacyl (S/G) ratio impacts overall recalcitrance phenotype, effectiveness of microbial hydrolysis, and melt spinability in carbon fiber production.

Research Summaries from Genomic Science Program Awardee Meeting, March 2016. (see PDF page 75)


  • Demonstrated further evidence that down-regulation of pectin biosynthetic genes has a major impact on recalcitrance in woody or grass feedstocks.
  • Reported results from multiple multi-year field trials of either genetically modified or natural variant Populus and switchgrass lines that continue to exhibit reduced recalcitrance and stable, or in some cases improved, agronomic performance.  Comparison of multiple field and greenhouse-grown switchgrass TOP Lines for growth and sugar release and greenhouse ’omics data, revealing stability of phenotypes in the field.
  • Elucidated a third modality of cellulase action exhibited by C. thermocellum, contributing to its ultra- high performance on cellulose [Xu 2016].
  • Achieved greater than 85% carbohydrate solubilization for some lines of Populus and switchgrass [Balch 2017] using consolidated bioprocessing with cotreatment in the absence of added enzymes or thermochemical pretreatment implying that C. thermocellum is able to attack all the major chemical linkages in representative woody and herbaceous lignocellulose crops given sufficient physical access.
  • Reported further evidence that lignin S/G ratio impacts overall recalcitrance phenotype, effectiveness of microbial hydrolysis, and melt spinability in carbon fiber production [Li 2016; Sun 2016].
  • Multi-year field trails of ten different reduced recalcitrance transgenic switchgrass lines.
  • Working with corporate partner, Ceres, field trials of the first elite switchgrass variety were modified simultaneously in three different cell wall biosynthetic genes.
  • Identified S. Vermifera ssp. bescii, isolated from native switchgrass roots, which beneficially impacts growth and sustainability; draft genome sequenced [Ray 2016].
  • Elucidation of the archetypal structure of a family GT37 glycosyltransferase involved in cell wall polysaccharide biosynthesis [Urbanowicz 2017].
  • Identified two arabinogalactan protein-specific 4-O-methyltransferases [DeBruyn 2016].
  • Reassessment of physiological roles of several early lignin pathway genes allows defining “minimum reaction pathways.”
  • Demonstrated that grasses synthesize approximately 50% of their lignin from L-tyrosine.
  • Improved solubilization by C. bescii cultures with heterologous expression of xylanases [Kim 2016; Paye 2016].
  • Detailed comparative studies indicating that C. thermocellum outperforms industry-standard fungal cellulase by several-fold in the solubilization of plant cell walls.
  • Investigation of multiple recalcitrance levers revealed large absolute differences in carbohydrate solubilization in response to the choice of biocatalyst and whether or not feedstocks are co-treated, and smaller sensitivity exhibited with respect to whether or not feedstocks are genetically modified (three modified switchgrass lines) or are less recalcitrant variants (two Populus lines).
  • Transformation of three previously untransformed microbes, providing proof of concept for a systematic approach to rendering genetically tractable non-model microbes.
  • Introduction of the T. saccharolyticum ethanol production pathway into C. thermocellum [Hon 2017].
  • Substantial progress in identifying the mechanistic basis of key aspects of microbial robustness.  In particular, testable hypotheses regarding pentose inhibition have been developed concerning declining solubilization at high solids and ethanol inhibition [Verbeke 2017].
  • Lignin was found to be a major impediment to CelA hydrolysis of biomass. Expanded genetic toolkit for Caldicellulosiruptor [Chen 2015].
  • Continued progress in isobutanol production by C. thermocellum, including a titer of 9.7 g/L.
  • Developed and applied advanced spectroscopic and chemical techniques to determine the structures of hemicellulosic polysaccharides in plant biomass, to understand the mechanisms of their biosynthesis, and to gain insight into the effects of modifying these structures on biomass recalcitrance [Tolbert 2017].
  • CELF pretreatment in Populus was optimized for higher yields and showed high removal of lignin and cleavage of lignin interunit linkages [Nguyen 2015].
  • In collaboration with GLBRC, developed a higher throughput thioacidolysis analysis for lignin [Harman-Ware 2016].
  • Developed a novel whole cell wall NMR analysis method using DMSO-d6/HMPA-d18 useful for understanding plant cell wall structure with minimal deconstruction and modification of biomass [Yoo 2016].
  • A cross-comparison review of recalcitrance changes in BESC-developed transgenic switchgrass was performed showing plant wall-targeted gene modifications resulted in four transgenic lines with up to 12% increased carbohydrate content and up to 21% improved enzymatic digestibility [Biswal 2017].
  • Populus S/G content ratio was linked to recalcitrance in microbial solubilization of non-pretreated biomass.  Two-dimensional NMR structural analyses showed that increased abundance of β-5 lignin branching bonds was linked to increased recalcitrance and correlated with increased guaiacyl content [Meng 2016].
  • Sixty proteomes have been measured and are being integrated with data from 30 metabolomes to elucidate key functional differences in switchgrass fungal endosymbionts.
  • Metabolomics data from C. thermocellum deconstruction of switchgrass were integrated with the proteomics data and highlighted the accumulation of microbially-derived long-chain fatty acids C5 sugars, sugar alcohols, and dextrins.  Not observed was the large-scale accumulation of amino acids observed when C. thermocellum is cultured with model substrates.  Subsequent controlled studies culturing C. thermocellum with C5 metabolites indicated a differential degree of microbial inhibition was induced by these constituents [Poudel 2017].
  • Identified several Raman peaks associated with xylan content in cell walls and demonstrated that xylan in corn stover cell walls can be imaged at high (~300 nm) resolution [Zeng 2016].
  • Characterization of C. thermocellum enzymes and systems for lignocellulose deconstruction resulted in a new conceptual model being published [Xu 2016].
  • The ratio of S/G in non-pretreated Populus was shown to affect microbial CBP hydrolysis and fermentation to the same extent as it impacts the enzymatic hydrolysis of poplar biomass, refuting the hypotheses that microbes can more aptly overcome all lignin-derived recalcitrance in poplar when compared to commercial fungal enzyme blends [Dumitrache 2016].
  • Quantitative fluorescence CLSM and surface spectroscopy by ToF-SIMS showed that, following microbial digestion, the relative content of surface cellulose decreased by 50% while surface lignin increased by 42% demonstrating that biomass recalcitrance is defined primarily by surface characteristics [Tolbert 2016].
  • Identified a novel potential mechanism for enzyme deactivation resulting from enzymes unfolding irreversibly at the air-liquid interface to reduce surface tension by aligning hydrophobic and hydrophilic regions with gas and liquid phases, respectively.
  • Demonstrated lignin removal had a more positive effect on switchgrass deconstruction by C. thermocellum than xylan removal, but removing both lignin and xylan was most effective.
  • Significant progress in assembling a pan genome and pan transcriptome across the poplar GWAS population.
  • A genome-scale metabolic model of C. thermocellum incorporating ’omics data identified key bottlenecks [Thompson 2017].

Year 8 (Oct. 2014 – Sept. 2015)


Engineered Clostridium thermocellum, a thermophilic, cellulolytic microbe, to produce isobutanol, an advanced biofuel.

BESC partner Mascoma, LLC, launched C5 FUEL™, a new yeast strain engineered for improved cellulosic biofuel production.


Study assessing the field performance of modified switchgrass demonstrated a doubling of biofuel production per hectare, the highest gain reported from any field-grown modified feedstock.

Molecular dynamics simulations performed on the TITAN supercomputer at Oak Ridge National Laboratory to determine the molecular basis of reduced recalcitrance.


Applied association genetics to rapidly identify specific genes best suited for producing biofuel from poplar trees planted in different environments.

Research Summaries from Genomic Science Program Awardee Meeting, February 2015. (see PDF page 23)


  • Investigators have shown that many modifications in four key plant cell wall pathways (lignin, cellulose, xylan and pectin) can reduce recalcitrance. In many cases (but not all), this has been achieved while also having increased biomass growth — these make up about two dozen of our TOP Line plants.  This combination was not expected at the beginning of BESC.
  • Studies with multiple GAUTs support the role of GAUTs in multiple pectin matrix polymer structural sites within the plant cell wall and that many of these have a role in recalcitrance, despite the low amount of pectin in the bulk secondary plant cell walls [Biswal 2015; Hao 2014a; Hao 2014b].
  • Key hemicellulose biosynthetic genes were identified (i.e., the first ever xylan synthase [Urbanowicz 2014], along with O-acetyl transferase and fucosyl transferase). This included the use of a eukaryotic expression system for plant enzymes.
  • A predictive model of biomass recalcitrance has been proposed based on rigid percolation theory and is validated by multiple plant lines and data sets.  This combined with data from NMR and pretreatment continue to provide new insights into biomass structure and the causes of recalcitrance.
  • The collaborative use of modified plants and microbes as probes in structure is supported by the observation that low lignin in plants facilitates growth by Caldicellulosiruptor mutants lacking pectinase genes — suggesting lignin and pectin interactions [Chung 2014b].
  • Between one to four years of field studies with nine switchgrass lines have been completed.  Multiple year field studies show that multiple BESC switchgrass TOP Lines can maintain reduced recalcitrance phenotypes with high ethanol and robust biomass yields in the field [Baxter 2015].
  • The cellulolytic secretome of C. thermocellum can be improved and the existence of several mechanisms was demonstrated to be due to multiple domain proteins [Resch 2013].
  • Investigators have improved thermostability of key microbial pathway enzymes for further metabolic engineering in C. thermocellum:  both pyruvate decarboxylase and keto-isovalerate decarboxylase.
  • Significant improvements were achieved in C. thermocellum yield and titer.  Currently, a strain of C. thermocellum that can produce ethanol at 77% of theoretical yield and 22 g/L titer from Avicel [Tian 2016; Papanek 2015].
  • BESC’s best C. thermocellum strain produced >5 g/L of iso-butanol from cellulose, corresponding to 41% of theoretical yield using an accelerated screen for pathway optimization [Lin 2015].
  • Transformation efficiency was improved ~20-fold in C. thermocellum, based on methylation patterns and a deleted restriction enzyme [Yu 2015].
  • Mascoma released to the market a yeast strain which converts xylose to ethanol. This strain has a proprietary xylose isomerase with improved performance over native enzymes.  Underlying fundamentals in xylanase secretion and promoter were supported by BESC.
  • Metabolic models of CBP in C. thermocellum have improved and been validated [Thompson 2015]. A database and lineages of BESC’s C. thermocellum strains were developed.
  • Co-treatment (which utilizes periodic grinding in the CBP fermentation) showed high extents of solubilization comparable to conventional pretreatment [Balch 2017].
  • The use of the natural variation and identification tools in the GWAS Populus studies allow rapid identification of gene candidates.  One gene variant [a 5-enolpyruvylshikimate-3-phosphatase (EPSP)-like protein] has been shown to have evolved regulatory functions resulting in a patent pending and potential licenses.
  • Outreach served over 25,000 students, parents and teachers in the last year — cumulative >140,000.

Year 7 (Oct. 2013 – Sept. 2014)


Discovered a unique form of lignin (i.e., C-lignin) in vanilla bean seed coats, which shows promise as a feedstock for carbon fiber.

Improved biofuel production from modified switchgrass when the feedstock was combined with an engineered C. thermocellum strain.

Outlined how more value can be derived from lignin within a biorefinery.

switchgrass field study

Demonstrated the biofuel potential of lignin-modified transgenic switchgrass in a field study.

Characterized Caldicellulosiruptor bescii; this microorganism secretes an enzyme, CelA, which can digest cellulose almost twice as fast as the current leading cellulase enzyme on the market.

Early BESC spinoff technology licensed to startup company Vertimass; technology directly converts ethanol into a hydrocarbon blend stock for use in transportation fuels. (Press Release)

Research Summaries from Genomic Science Program Awardee Meeting, February 2014. (see PDF page 329)


  • Research has shown that transgenic modifications in four plant cell wall components (lignin, cellulose, xylan and pectin) can have significant effects on recalcitrance without necessarily harming growth.
  • Evaluation of natural variant populations of both Populus trichocarpa and switchgrass identified five TOP Line candidates with significantly and unpredicted reduced recalcitrance.
  • The elusive xylan synthase and first plant polysaccharide O-acetyltransferase were identified and shown to synthesize acetylated xylan in vitro — a breakthrough in understanding hemicellulose synthesis [Urbanowicz 2014].
  • A native mycorrhizal fungus [Sebacina vermifera specy (sp.)] strain that imparts large biomass gains in switchgrass was identified [Ray 2015].
  • Up to three years of field studies have shown that multiple switchgrass TOP Lines maintain high ethanol and biomass yields in the field compared to controls.
  • Down-regulation of matrix polysaccharide biosynthetic gene galacturonosyl transferase (GAUT)12 leads to reduced xylan and pectin in Populus wood, reduced recalcitrance, more easily extracted cell walls, and increased Populus growth [Biswal 2015].  Arabidopsis GAUT14 was shown to have pectin biosynthetic activity, implicating modified pectin metabolism as the basis for reduced recalcitrance of GAUT14 knockdown switchgrass lines.
  • C. thermocellum strains were created with 82% of theoretical ethanol yield on cellobiose; albeit strains grow slowly.
  • C. bescii was engineered using new tools to produce ethanol from cellobiose, Avicel or un-pretreated switchgrass. Only 30% of biomass was fermented but ethanol yields were ~80% [Chung 2014a].
  • A strain of C. thermocellum was engineered to produce 1.1 g/L isobutanol by engineering the full 5-step pathway from thermophiles on a plasmid [Lin 2015].
  • New mechanistic paradigms were discovered for CBP attack of solid biomass.
  • CelA from Caldicellulosiruptor is a bi-functional efficient cellulase that both ablates and digs into biomass fibers [Brunecky 2013].
  • Intact C. thermocellum cellulosomes defibrillate the ends of biomass particles and fibers [Resch 2013].
  • When performing comparative microbial cellulose utilization—naturally cellulolytic microbes have the ability to degrade even non-treated biomass, C. thermocellum continues to be one of the best at biomass degradation—matching those of even microbial communities [Paye 2016].
  • Co-treatment (sequential CBP and milling) was found to enhance conversion.
  • Fermentation of a combination of transgenic, reduced-recalcitrance COMT switchgrass with a genetically engineered C. thermocellum strain showed an increase in conversion of approximately 20% relative to the wild-type switchgrass control.
  • Real-time NMR and diffusion ordered spectroscopy examined the molecular size and aggregation of hemicelluloses in enzyme catalyzed reactions.  A combination of physical and NMR-based relaxation measurements showed recalcitrance is correlated to increased mobility within the cell wall polymer matrix due to reduced crosslinking or polymer entanglement.
  • Metabolite analyses using Gas Chromatography Mass Spectrometry of >60 transgenic lines of Populus deltoides, Panicum virgatum, and Arabidopsis thaliana with modified gene activity, showed disrupted metabolisms beyond the targeted pathways into other plant cell wall components.
  • A highly-effective pretreatment and fractionation technology, Co-solvent Enhanced Lignocellulosic Fractionation (CELF) produced nearly theoretical yields of glucose and xylose from hardwoods and corn stover at extremely low enzyme loadings (e.g., 2 mg enzyme protein/g-glucan).  Tetrahydrofuran is the co-solvent.
  • Outreach served over 30,000 students, parents and teachers in the last year — cumulative 115,000; with more than 75% of the program self-supporting.  A new outreach and education website was released.
  • GWAS analyses of the Populus and switchgrass experiments are being done on BESC computers, and the methods are being optimized with collaborative support from DOE Knowledgebase.
  • Industry field trials of the modified switchgrasses continued with first and second year growths from Ceres and UT being analyzed.

Year 6 (Oct. 2012 – Sept. 2013)


Demonstrated that fungal cellulases and complexed cellulosomal enzymes exhibit different, yet synergistic mechanisms in cellulose deconstruction; this finding highlights the potential for combining these two systems for enhanced performance.

Discovered a new plant structure (Arabinoxylan-Pectin-Arabinogalactan Protein 1 or APAP1) that may lead to improved biofuel processing.

Applied high-performance proteomics to identify important networks and pathways in Populus.

road trip challenge icon

Developed hands-on "Farming for Fuels" lesson plans to educate fourth–sixth graders about a biobased fuel economy; educational outreach program has now reached 85,000 (27,000 in Year 6 alone) students, teachers, and parents nationwide. Also released a biofuel Road Trip Challenge game as a museum kiosk and as an iPad app.

Developed an integrated microscopy system for real-time imaging of pretreatment and enzyme digestion; findings show that biomass reactivity is determined by the nanoscale architecture of plant cell walls, in which lignin is the major factor impeding the access of chemical and enzymatic catalysts. (News Article; )

Research Summaries from Genomic Science Program Awardee Meeting, February 2013 (titles only; no abstracts available)


  • Additional plant cell wall biosynthetic genes were demonstrated to have significant effects on recalcitrance.  These gene targets have brought us beyond the COMT gene effects on the lignin pathways shown previously and show the range and potential for improvement of these feedstocks. These include:
    • Overexpression of MYB4, a regulatory transcription factor in switchgrass, yielded a 3-fold improvement in sugar release and decreased potential lignin inhibitors [Shen 2013].
    • Overexpression of UDP-glucose pyrophosphorylase, a cellulose biosynthesis gene, in Populus increased sugar release but decreased growth [Payyavula 2014].
    • Knockdown of FPGS (folylpolyglutamate synthetase), a co-factor, reduced lignin and recalcitrance in Arabidopsis [Srivastava 2015].
    • Pectin synthesis genes significantly affected recalcitrance in switchgrass and Populus [Atmodjo 2013].
    • Changes in xylan biosynthesis of the primary nucleotide-sugar precursors significantly reduced recalcitrance.
    • The effect of a cell wall proteoglycan with pectin and arabinoxylan covalently attached to an arabinogalactan protein identifies a cross-linked, matrix polysaccharide-wall protein architecture with implications for wall structure, function and synthesis [Tan 2013].
  • New genetic tools for thermophilic CBP microbes made available in earlier years continue to be exploited in order to create CBP production strains with high yield and conversion [Olson 2015a].
    • In C. thermocellum, manipulation of the carbon, electron and ethanol tolerance pathways were explored. The most successful strategy improved ethanol yield to 0.34 g/g carbohydrate in strain AG553 (C. thermocellum Δhpt ΔhydG Δpfl Δldh Δpta-ack) [Papanek 2015].
    • New tools [Cha 2013; Chung 2013a; Chung 2013b; Chung 2013c; Farkas 2012] for the extremely thermophilic Caldicellulosiruptor spp. were developed and showed preliminary levels of ethanol production.
    • Demonstration of all enzymes needed for a thermophilic pathway to isobutanol in Geobacillus [Lin 2014].
  • BESC’s hands-on “Farming for Fuels” lessons educate 4th-6th graders about a bio-based fuel economy.  Outreach has now extended to 85,000 (27,000 in Year 6) students, teachers, and parents nationwide.  A biofuel “Road Trip Challenge” game was released as a museum kiosk and as an iPad app.
  • An integrated microscopy system was built for BESC and used for real-time imaging of pretreatment and enzyme digestion.  As reported in Science [Ding 2012], showed that biomass reactivity is determined by the nanoscale architecture of the plant cell walls, in which lignin is the major factor that physically impedes the accessibility of chemical and enzymatic catalysts to the substrates [Ding 2012; Zeng 2013]
  • The surprising ability of cellulolytic microbes to consume biomass extensively (especially grasses) after minimal or no pretreatment, and further the demonstration that solubilization of plant cell walls is far more effective when mediated by cellulolytic microbes than by cell-free enzymes [Paye 2016].
  • Provisional patents were filed on the discoveries that CelA from C. bescii is the most active enzyme reported to date on Avicel cellulose and that C. thermocellum cellulosomes will enhance the activity of fungal hydrolytic enzyme cocktails [Brunecky 2017].
  • C. bescii solubilized lignin at the same rate that it rendered soluble the carbohydrate components of switchgrass, even though it is not consuming the lignin — just removing it to improve access to the biomass [Kataeva 2013; Resch 2013].
  • Glycome profiling has been applied as a medium-throughput technique to monitor structural and sugar extractability changes in Populus biomass from BESC and stover from GLBRC during pretreatment in order to better understand how this process reduces plant cell wall recalcitrance [Pattathil 2015a; Pattathil 2015b].
  • BESC performed simulations on lignin homo- and heterodimers and trimers and showed that hydroxyphenol (H) subunits may serve as terminal “capping” subunits that effectively stop lignin elongation.  These findings may explain why lignins with a higher relative abundance of H subunits have, on average, lower molecular weight than do lignins with more modest H content [Sangha 2014].
  • CAZymes are critical for hydrolysis; a large amount of data exists in a public database.  Two tools were developed for extraction and interpretation of this data.  Investigators also developed a web resource, dbCAN (csbl.bmb.uga.edu/dbCAN/annotate.php) to provide an automated CAZyme signature domain-based annotation.  DbCAN has had more than 3,000 unique visitors from 64 countries since it was released. [Xu 2016].  In addition, BESC provided the Knowledgebase project with a set of BESC exemplar datasets as well as the public release of the V2.0 BESC CAT analysis tool (CAZy family analysis) [Karpinets 2012].
  • Field trials of transgenic switchgrass for BESC characterization studies were performed under the United States Department of Agriculture-Animal and Plant Health Inspection Service-Biotechnology Regulatory Services permits [Baxter 2014].  Transgenic field grown switchgrass COMT-KD lines (second and third growing season) and MYB4-OE (first and second growing season) were established.  The field data have led to several important conclusions.  Firstly, data from season one field grown plants are largely qualitative since switchgrass takes time to establish.  However, sugar release and ethanol production capacity from season two COMT-KD down-regulated lines very closely parallel previous results from greenhouse-grown plants [Fu 2011].
  • Solubilization effectiveness was compared as a function of biocatalyst and feedstock.  Methods were developed to quantify cell and substrate concentrations during fermentation of Avicel by C. thermocellum [Holwerda 2013a], resulting in estimates of the instantaneous rate of cellulose solubilization.  Growth rates on cellulose transiently exceeded growth on cellobiose.  A newly-proposed rate law (first order in cells and substrate, second order overall) was found to fit the data well, whereas previously proposed rate laws did not [Holwerda 2013b].
  • 1051 P. trichocarpa leaf samples from the GWAS were analyzed by gas chromatography-mass spectrometry and a metabolite data extraction algorithm was developed to extract over 400 known and unknown metabolites.  This study represents the largest association genetics study to map genes regulating metabolite concentration and should identify numerous genes that are currently annotated as unknown.  

Industrial Interactions

  • Transgenic switchgrass samples were tested by DuPont in their proprietary bioconversion process.

Year 5 (Oct. 2011 – Sept. 2012)


Developed the ability to genetically manipulate Caldicellulosiruptor, which are thermophilic cellulolytic microorganisms; this advance could enable the direct conversion of lignocellulose to biofuels.


Completed sequencing and analysis of a reference genome for Setaria; this advance enables the further development of Setaria as a model plant and improves understanding of cell wall composition, plant structure and development, and traits pertinent to the development of biofuel crops.

Discovered an unusual lignin polymer in certain plant seeds; constructed from catechyl (C) units, C-lignin demonstrates the natural capability of native plants to produce specific lignin polymers and supports the radical coupling polymerization hypothesis. C-lignin also may be a source for carbon fiber manufacturing.

researchers in the lab

Demonstrated that GAUT1 and GAUT7, proteins in the plant cell wall, are co-expressed not only in a plant's primary walls but also in the secondary walls and in many plant tissues; this discovery changes the way scientists think about how plant cell walls are made and opens a new door for converting plants to biofuels and other bioproducts.

Discovered a gene (GXMT1) that plays a major role in the cell wall development of Arabidopsis plants; this finding lays the groundwork for improved biofuel processing.

2012 Genomic Science Awardee Meeting X, Bethesda, Md., February 26-29 2012 (see PDF page 30)


  • Additional plant cell wall biosynthetic genes were demonstrated to have significant effects on recalcitrance.  These gene targets are beyond the COMT gene effects on the lignin pathways shown previously and show the range and potential for improvement of these feedstocks.  These include:
    • Overexpression of MYB4, a regulatory transcription factor in switchgrass, was shown to yield a 3-fold improvement in sugar release.
    • Overexpression of UDP-glucose pyrophosphorylase, a cellulose biosynthesis gene, in Populus increased sugar release but decreased growth.
    • FPGS (folypolyglutamate synthetase), a co-factor, reduced lignin and improved recalcitrance in Arabidopsis.
    • Pectin synthesis genes were shown to have significant effects on recalcitrance in switchgrass, Populus, and Arabidopsis. This finding is significant and surprising because while pectin is important in primary cell walls, it is a small fraction of mature secondary walls.
    • Changes in xylan biosynthesis of the primary nucleotide-sugar precursors also had a significant impact on recalcitrance.
  • New genetic tools for thermophilic CBP microbes are available and are being exploited in order to create CBP production strains with high yield and conversion.
    • In C. thermocellum, manipulation of the carbon, electron, and ethanol tolerance pathways were explored.  The most successful strategy improved ethanol yield to 0.48 g/g carbohydrate — which is near the ultimate theoretical goal.  Strategy A was based on the idea of eliminating competing paths for carbon flux and was implemented by deleting pathways for acetate and lactate production, which was followed by adaptation to improve strain growth [Argyros 2011].  Strategy B was also based on the idea of eliminating competing paths for carbon flux and was implemented by selecting for growth on high levels of ethanol and followed by deletion of lactate production [Brown 2011; Biswas 2014].  Strategy C was based on the idea of eliminating competing paths for electron flux and was implemented by eliminating formate and hydrogen production pathways [Rydzak 2015], Biswas 2015].  Strategy D was based on the idea of transferring the high-flux ethanol pathway from T. saccharolyticum to C. thermocellum and was implemented by overexpressing the pyruvate kinase and alcohol dehydrogenase genes from T. saccharolyticum, reducing flux through the ‘malate shunt’ by either down-regulating phosphoenolpyruvate carboxykinase or deleting the malic enzyme [Deng 2012].  These strategies produced the following results:  (1) Strategy A gave a maximum yield of 0.34 g ethanol/ g sugar; (2) Strategy B gave a maximum yield of 0.34 g ethanol/g sugar; (3) Strategy C gave a maximum yield of 0.28 g ethanol/g sugar; and (4) Strategy D gave a maximum yield of 0.48 g ethanol/ g sugar, which achieved our yield goals through Year 6.
  • The Brown Lab (ORNL) identified a mutated aldehyde-alcohol dehydrogenase (adhE) that enhanced ethanol tolerance in C. thermocellum [Brown 2011].  Further progress was made in the understanding of C. thermocellum ethanol tolerance by resequencing additional mutant strains and using systems biology tools to analyze the wild-type ethanol stress responses [Shao 2011].
  • The first gene transfer system for Caldicellulosiruptor spp. was developed exploiting a novel gene transfer system confirmed by an ldh deletion.  This system used methylation with an endogenous unique α-class N4-cytosine methyltransferase to overcome a major barrier to DNA transformation [Chung 2012].
  • The biomass characterization HTP pipelines for composition and recalcitrance analyzed ~18,000 samples in Year 5, and the data was stored within the BESC LIMS.  This capability underlies our collective power to analyze large numbers of samples.
    • A medium-throughput “wet-chemistry” based compositional analysis was downscaled from 5 g/sample to 5 mg, near that of the HTP characterization system.
  • The LIMS was updated with user-friendly reports and scripts to improve utility and access.  The BESC data warehouse was expanded and new data repositories and computational frameworks for data analysis were built.  The LIMS now contains metadata and data from more than 50 experimental campaigns plus 103,645 BESC distinct samples.  Over 39,630 of these samples have been shipped among BESC facilities with tracking in LIMS to easily comply with our Material Transfer Agreement.
  • Previously, UCLA investigators produced isobutanol in a mesophilic CBP microbe [Higashide 2011].  This year, the activities of equivalent key pathway enzymes were expressed and confirmed at thermophilic conditions in order to allow the creation of a thermophilic CBP isobutanol producer.  The KDC used in our previous mesophilic pathway is encoded by kivd from Lactococcus lactis.  This purified Kivd protein is quite unstable, but demonstrated activity in vivo at 55 °C in G. thermoglucosidasius.  Kivd was overexpressed on a plasmid and 1.1 g/L of isobutanol was produced from 5 g/L of the substrate 2-ketoisovalerate (KIV).  With a combination of in vivo and in vitro enzyme assays, activity for all pathway enzymes at ≥55 °C has been demonstrated [Lin 2014].
  • Association mapping in native and crossbred populations have identified improved QTLs and new genes targets in Populus and switchgrass.  The Saha and Brummer (NF) and Devos (UGA) labs constructed genetic linkage maps of a lowland AP13x upland VS16 switchgrass population using SSR, sequence tagged sites, and Diversity Arrays Technology markers.  A subset of the population was sequenced at JGI.  Eleven QTLs associated with biomass yield were mapped on seven linkage groups.  The additive effect of the QTLs mapped for biomass yield ranged from 6-24%.  Additional QTLs were associated with lignin content and glucose/xylose release.
  • In Populus, major QTLs affecting lignin content and S/G ratio were reproducibly identified from a mapping pedigree where 712 individuals were subjected to MBMS compositional and saccharification assays.  A high-density SNP genetic map with 3,559 markers was used to delimit genomic regions harboring QTLs for recalcitrance phenotypes. Based on this analysis, at least four distinguishable major QTLs, spanning the length of scaffold 14 of the Populus genome assembly, were identified.
  • Glycome profiling, using polysaccharide antibodies to identify changes in plant cell wall polysaccharide epitopes, became a medium-throughput characterization technique and was utilized to analyze native and pretreated biomass samples from across BESC and from the Great Lakes Bioenergy Research Center (GLBRC) [Duceppe 2012].
  • Provisional patents were filed on the discoveries that CelA from C. bescii is the most active enzyme reported to date on Avicel cellulose and that C. thermocellum cellulosomes will enhance the activity of fungal hydrolytic enzyme cocktails [Brunecky 2013].
  • Successfully demonstrated that solubilization of plant cell walls is far more effective when mediated by cellulolytic microbes than by cell-free enzymes [Howerda 2017], [Paye 2016].
  • BESC’s hands-on “Farming for Fuels” lessons educate fourth-sixth graders about a bio-based fuel economy.  Our outreach has extended to 60,000 (22,000 in Year 5) students, teachers, and parents nationwide.
  • BESC completed enzymatic hydrolysis of Ammonia Fiber Expansion™ (AFEX) (GLBRC), dilute acid (BESC), and ionic liquid (Joint BioEnergy Institute [JBEI]) pretreated solids for up to 120 hours with a range of enzyme loadings from 3 to 30 mg protein/g glucan in the initial corn stover [Singh 2014].  BESC also developed enzyme adsorption data that showed ionic liquids to have the greatest and AFEX the lowest capacity.  Each of the BRCs is preparing papers for publication in a single journal volume on this collaboration.
  • The focus on the genus Caldicellulosiruptor with an eye towards developing metabolically engineered strains capable of converting un-pretreated plant biomass to fuel molecules has been based on a comprehensive genomics strategy.  This approach has been informed by an analysis of eight finished genome sequences for species from geographically diverse regions (Russia, Iceland, New Zealand, and North America) [Blumer-Schuette 2012].  The results of this analysis suggest that the Caldicellulosiruptor pan-genome is open, such that further bio-prospecting could lead to identification of novel glycoside hydrolases or other proteins for lignocellulose conversion.  The potential advantages of CBP at elevated temperatures are underscored by recent insights into the synergy between C. bescii and thermal environments, as this relates to the degradation of untreated switchgrass [Kataeva 2013].
  • C. bescii uses a set of architecturally complex cellulases with often more than one catalytic module. The microbe-surface interface was studied by focusing on modeling the release and utilization of carbohydrates by cellulose-adhered cells of C. obsidiansis [Wang 2011] as well as imaging surface colonization dynamics over varying spatial and temporal scales [Wang 2010] showing the development of microbial monolayers which form “craters” on the substrate.
  • The Bennetzen and Devos Labs (UGA) led a collaboration with a consortium of other research teams, including colleagues from the Joint Genome Institute and the Joint BioEnergy Institute, who sequenced, annotated and studied the foxtail millet (Setaria italica) genome as a model for the genomic and genetic analysis of switchgrass [Bennetzen 2012].
  • The BESC Populus trichocarpa association study population consists of 1,100 trees collected from a range of species throughout California to British Columbia.  Initially, 16 P. trichocarpa genomes were re-sequenced to an average depth of 39x and genotyped 120 trees from 10 P. trichocarpa subpopulations using 22,280 SNPs assayed with an Illumina Infinium BeadArray.  Taken together, these results have implications for understanding the evolutionary history of P. trichocarpa and the design of robust selection scans [Slavov 2012].
  • Investigators of the York Lab at UGA identified and characterized the first enzyme shown to methylate a plant polysaccharide, a glucuronoxylan 4-0-methyltransferase [Urbanowicz 2012].  This work provides fundamental insights into wall synthesis and extends the types of structural targets that can be modified alone or in combination to increase the economic value of lignocellulosic biomass.  Using Arabidopsis, candidate genes highly expressed during secondary cell wall formation were identified.  Corresponding homozygous T-DNA insertion lines were generated to evaluate the effects of KD of these genes on glucuronoxylan (GX) structure.  GX isolated from mutants was compared for degree of polymerization (DP), altered substituent distribution and/or composition, and polymer hetero/homodispersity revealing that recalcitrance is inversely correlated to xylan DP.  Using this approach, a GX 4-O-methyltransferase (GXMT1) responsible for methylating GlcA residues in GX was identified [Urbanowicz 2012].  GXMT1 is the first enzyme shown to catalyze polysaccharide methylation in vitro and represents the archetypal member of a new family of O-methyltransferases, which may catalyze the methyl-etherification of glycans in diverse taxa.  Importantly, reduced methylation of GX in GXMT1-1 plants correlates with altered lignin composition and increased release of GX by mild hydrothermal pretreatment, indicating that biomass with reduced xylan methylation has more easily extractable xylan [Urbanowicz 2012].
  • Within the BESC, microbial and plant reference genomic data, microbial re-sequencing data, microarray data, RNA-Seq data, plant genotype data, plant phenotype data and plant GWAS data are managed [Syed 2012].  The Computational Biology team provided resequencing data analysis tools, microarray data analysis tools, pathway analysis tools (BeoCyc), CAT, RNA-Seq data analysis tools, Gbrowse genome browser and comparative analysis tools.  A novel computational tool for discovering modular structure, relationships and regularities in complex data has been developed and published.  This tool will be applied to reveal associations between poplar phenotypic and genomic characteristics from large-scale resequencing datasets produced by BESC association feedstock studies [Karpinets 2012].

Industrial Interactions

  • Yeast strains that had been made into CBP microbes by the addition of heterologous glycosyl hydrolases were tested at pilot scale by Mascoma on pretreated wood chips. Mascoma has made tremendous progress developing strains of S. cerevisiae capable of expressing functional cellulases and hemicellulases.  This work includes functional heterogolous expression of over 350 cellulases and related enzymes, combinatoric studies involving multiple enzyme activities purified from or produced by engineered yeasts and simultaneous functional expression of enzymes.
  • An improved and searchable BESC invention website was developed to promote and enable easy access to intellectual property available for licensing (104 invention disclosures and 14 licenses).

Year 4 (Oct. 2010 – Sept. 2011)


Genetically modified switchgrass to improve biofuel production; discovered that the down regulation of a single gene reduces recalcitrance with no apparent growth defects, increases ethanol production by more than 30%, and requires three- to four-fold less enzyme for processing, potentially reducing production costs by at least 20%. Modification to be field tested at large scale.

Demonstrated that Clostridium cellulolyticum, a cellulose-degrading microbe, can be metabolically engineered to produce isobutanol directly from cellulose, thereby demonstrating the ability to apply consolidated bioprocessing (CBP) technology to produce advanced biofuels.

Developed a set of hands-on "Farming for Fuels" lesson plans to educate fourth–sixth graders about the carbon cycle, use of lignocellulosic biomass for biofuels production, and technical and economic obstacles to a biobased fuel economy. This national outreach program, developed in partnership with the Creative Discovery Museum in Chattanooga, Tennessee, has reached over 35,000 students, teachers, and parents by partnering with museums and centers in Tennessee, Georgia, Texas, Michigan, Illinois, Florida, New York, and Arizona. (Lesson Plan PDF)

With BESC assistance, Mascoma, LLC, developed yeasts that express recombinant cellulases; these reagents are intended for use in Mascoma's commercial 20-million-gallon hardwood-to-ethanol plant in Kinross, Michigan.

Genomic Science Awardee Meeting IX and USDA-DOE Plant Feedstock Genomics for Bioenergy Awardee Meeting, Crystal City, Va., April 10-13 2011 (see extracted PDF page 2)


  • BESC generated a genetically modified switchgrass (COMT) that yields ~30% more ethanol or requires threefold to fourfold less enzyme for processing, potentially leading to at least a 20% reduction in production cost.  This “lignin lite” low recalcitrance COMT transgenics [Fu 2011] switchgrass line was moved into three-years of field testing with good growth.  Three-year field trials were conducted at the University of Tennessee, Knoxville (UT), and at College Station, Texas, by our corporate partner Ceres.
  • BESC with Ceres introduced and field tested this modification at large scale.  This proof-of-concept demonstrated BESC’s ability to significantly lower the cost of biofuel production by developing modified sources of biomass.
  • BESC identified at least 12 new recalcitrance genes using a SNP discovery approach to link genotypic variation to phenotypic variation observed in natural populations of Populus [Muchero 2015].
  • BESC identified natural Populus variants that after pretreatment at standard conditions release high quantities of sugar (~80% of total sugars).  Natural variants that release ~67% of total sugars without pretreatment were also identified.  These discoveries represent a 20% increase in sugar yield and a potentially substantial reduction in production costs by reducing the need for pretreatment [Studer 2011].
  • BESC’s cellulolytic microorganisms were improved by selecting ethanol and acetate tolerance, thereby overcoming key inhibitors of microbial fermentation.  These discoveries are significant because end product titer and inhibitory by-products are major contributors to capital and downstream processing costs [Yang 2012], [Brown 2011] [Shao 2011].
  • BESC demonstrated that C. cellulolyticum, a cellulose-degrading microbe, can be metabolically engineered to produce isobutanol directly from cellulose, thereby demonstrating the ability to apply our CBP technology to produce advanced biofuels [Higashide 2011].
  • BESC invented a first-of-its-kind high-throughput platform for determining recalcitrance properties of tens of thousands of feedstock samples (>25,000 samples to date).  This platform is a key capability for the general biofuels research community.  A multi-chamber pretreatment reactor for HTP screening of biomass developed in this study was patented and licensed to Aspen Machining, a precision machining company in Lafayette, Colorado.  A small-scale, HTP biomass compositional analysis pipeline was developed for glucan, xylan, and acetate analysis.  To support this assay, an HTP destarching protocol was developed and implemented that has reduced experimental variability in determining glucan content from ~8% to ~2% in high-starch materials [Decker 2012]. The HTP hydrothermal recalcitrance pretreatment and enzyme hydrolysis was further adapted to utilize dilute acid pretreatment.  This expanded our characterization capability to include softwoods.  Also developed was a second-tier enzyme assay that probes the linkages responsible for recalcitrance in plant cell walls using defined sets of purified enzymes [Decker 2009; Selig 2010].  Investigators screened over 15,000 individual samples in triplicate for recalcitrance to enzyme hydrolysis after pretreatment.
  • BESC successfully demonstrated that combining a genetically modified switchgrass with yeast engineered to be cellulolytic (a modified CBP organism) provides greater than 50% increase in biofuel yield, with ≥30% of the benefit due to the modified switchgrass and ~25% due to the improved yeast.  This achievement demonstrates the successful combining of improved feedstock with an improved CBP biofuel-producing microorganism.
  • BESC developed a first-of-its-kind, bioenergy-oriented knowledgebase [Yee 2012] accessible through a user-friendly web interface).  This digital resource communicates with the BESC LIMS and serves as a repository for tools and datasets generated by diverse characterization techniques operating over a wide range of scales [Syed 2012].
  • BESC developed new genetic tool Caldicellulosiruptor sp. [Chung,2012].
  • BESC improved and utilized our genetic tools for metabolic engineering in C. thermocellum (a key CBP microbe) to increase solvent yields in thermophile model systems.  Work performed by BESC partner Mascoma led to production of 65 g/L ethanol by T. saccharolyticum, with 0.44 g of ethanol per gram of sugar fermented.  In addition, every enzyme involved in the butanol pathway was expressed in T. saccharolyticum.  In C. thermocellum,a program was initiated to delete key enzymes impacting ethanol yield including acetate kinase(ack) and lactate dehydrogenase(ldh) [Argyros 2011], eliminating acetate and lactate as fermentation products.
  • BESC engineered 40 “designer” minicellulosomes that contain individual cellulolytic components. These minicellulosomes have higher cellulolytic activity than simple mixtures of the same free enzymes that are commercialized by at least two companies [Xi 2013].
  • BESC improved the efficiency of switchgrass transformation from initially 5% up to 90% and decreased the turnaround time to less than 4 months.  This development will accelerate improvements of switchgrass as a biofeedstock.
  • BESC developed a set of hands-on “Farming for Fuels” lesson plans that educate 4th-6th graders about the carbon cycle, the use of lignocellulosic biomass for the production of biofuels, and the technical and economic obstacles to a bio-based fuel economy.  This outreach program, developed in partnership with the Creative Discovery Museum, has expanded nationally and has reached over 35,000 students, teachers, and parents by partnering with museums and centers in Tennessee, Georgia, Texas, Michigan, Illinois, Florida, New York, and Arizona.
  • BESC discovered that endophytic fungi can enhance switchgrass growth and establishment [Ghimire 2009], [Ghimire 2011].
  • BESC aims to reduce recalcitrance by manipulating nucleotide-sugar substrates used for wall polysaccharide synthesis and by controlling enzymes that esterify polysaccharides.  Forty new genes involved in nucleotide-sugar metabolism from diverse organisms were functionally characterized [Yang 2009; Gu 2009; Yang 2010; Gu 2010], providing new tools to modify plant wall biosynthetic pathways.  An initial subcellular molecular model of nucleotide-sugar biosynthesis pathways [Jiang 2010] and a real-time NMR method to simultaneously study multiple wall synthesis reactions [Guyett 2009] were established.  A process was developed (with the Department of Mathematics at UGA) to deconvolute NMR spectra of multi-enzyme reaction mixtures [Jiang 2010].  Simulation models that predict how nucleotide-sugar flux affects recalcitrance were developed, as was a database to facilitate fluxomics and carbon allocation analyses in reduced recalcitrance lines.
  • In vitro xylan biosynthesis assays using microsomal membranes, defined oligosaccharide acceptors, and UDP-sugar donors provided insight into xylan biosynthesis [York 2008] and support a block-transfer mechanism for extension of the xylan backbone in Arabidopsis, Populus, rice, and switchgrass.  A heterologously expressed glucurono-xylan promoter and a methyl transferase gene were functionally characterized Arabidopsis glucuronoxylan-specific methyltransferase, invention disclosure filed).  This is the first plant cell wall polysaccharide methyltransferase identified as being directly involved in xylan biosynthesis.  
  • BESC research generated ~180 monoclonal antibodies (MAbs) that recognize diverse wall carbohydrate epitopes [Pattathil 2010] and used them to demonstrate cell type and species-specific variation in wall carbohydrate composition.  A glycome profiling technique using the MAbs was developed to rapidly identify wall carbohydrate epitopes in natural variant, transgenic, and pretreated biomass and was used to study recalcitrance-related changes in biomass.
  • Investigators developed a coarse-grained cellulosome self-assembly model [Bomble 2010; Beckham 2011] that captures most of the physical characteristics of three cellulosomal enzymes (Cel5B, CelS, and CbhA) and the scaffolding (CipA) from C. thermocellum.
  • UCR developed a wet chemistry composition analysis method to facilitate rapid determination of biomass composition with limited amounts of material [DeMartini 2011].  This method was further extended to a micro HTP system at NREL.  This new capability allows high-throughput wet chemical analysis of glucan, xylan, and acetate using ~3 mg of biomass and complements the HTP cell wall chemical analysis using analytical pyrolysis.
  • ORNL researchers are developing nanoscale mechanical, chemical, and optical techniques to analyze cellulose and lignin networks in biomass.  Recently, mode synthesizing atomic force microscopy (MSAFM) was developed to provide a multi-frequency force microscopy approach to improve spatial resolution and subsurface studies.  For example, MSAFM with Fourier-transform infrared spectroscopy has been validated at 50 nm using preliminary measurements on cellulose nanowhiskers.  MSAFM has been further developed into a hybrid photonic-MSAFM (hp-MSAFM) to obtain simultaneous chemical and morphological information about biomass.  Preliminary measurements demonstrate that high spatial resolution compositional maps of a plant cell wall can be obtained using MSAFM and hp-MSAFM [Passian 2013].
  • ORNL developed a major scientific collaboration with investigators of the University of British Columbia/ Genome Canada program dealing with Populus association genetics.  These investigators have contributed funds and have participated in the genotyping and phenotyping of the 1100 individual trees in the study.

Industrial Interactions

  • NREL’s industrial HTP pipeline collaborators include Scion, GreenWood Resources, and Ceres.  The HTP pretreatment and enzyme capabilities were licensed to Aspen Machining for manufacture and sales to outside parties (Codexis was the first customer).
  • The BESC Knowledgebase was leveraged by Mascoma in an EERE-funded project to improve yeast fermentation strains and their genome resequencing.

Year 3 (Oct. 2009 – Sept. 2010)


Completed a baseline analysis of reduced recalcitrance switchgrass lines, finding that the transgenic material had normal growth habit, but reduced lignin, improved sugar release, and 24% to 38% higher ethanol yields than the wild type after fermentation; demonstrated for the first time the direct impact of reduced biomass recalcitrance at the bioconversion process level. (Journal Article)

Developed new biocatalysts, demonstrating the feasibility of producing active, engineered cellulosomes; a key achievement was the successful production of an engineered, active minicellulosome in Escherichia coli.

Continued relationship with National Geographic's The JASON Project to film and generate an educational module on bioenergy; this module, Operation: Infinite Potential, won three 2010 CODiE awards for best instructional solution grades K–12. (Module)

Genomic Science 2010 Awardee Workshop VIII and Knowledgebase Workshop, Crystal City, Va., February 7-10 2010 (see extracted PDF page 32)


  • BESC researchers at the Complex Carbohydrate Research Center at the University of Georgia, Athens identified novel genes that affect recalcitrance and plant transgenics and natural variants with reduced recalcitrance, with evidence of multiple paradigm-changing models for wall biosynthesis.  A paradigm- changing model for xylan synthesis generated by the in vitro xylan biosynthesis assays exhibited a multiphasic molecular weight distribution (clustering), in other words, small primary products are linked to form larger secondary and ternary products.  This pattern was consistently observed in all Arabidopsis, switchgrass, rice, and Populus microsomes analyzed to date.
  • Characterization and Data Management shared a major achievement in making the high-throughput recalcitrance screening pipeline a reality with thousands of samples and their data processed into LIMS for both BESC and external collaborators.  This pipeline has been used to screen ~10,000 samples from 9 research groups by the Davis and Decker groups at NREL [Decker 2009; Selig 2010].
  • The Samuel Roberts Noble Foundation, Inc. (NF) previously developed reduced recalcitrant Alamo switchgrass lines by down regulating the caffeic acid O-methyl transferase (COMT) gene in the lignin pathway.  In a paper submitted this year, analysis by NF and ORNL determined that the transgenic material had normal growth habit, but reduced lignin, improved sugar release, and significantly increased SSF (simultaneous saccharification and fermentation) ethanol production.  The transgenic lines yielded 24-38% more ethanol than the wild type after fermentation, demonstrating for the first time the direct impact of reduced biomass recalcitrance at the bioconversion process level.  A completed evaluation of this transgenic switchgrass with reduced recalcitrance has produced a detailed baseline analysis of switchgrass ultrastructure and chemical composition during development, and has demonstrated the correlations between various parameters (e.g., lignin content and composition, wall-bound-phenolic content and composition, soluble phenolic content and composition) and recalcitrance determined as saccharification efficiency.  Down-regulated switchgrass materials have reduced recalcitrance, permitting enhanced bioconversion characteristics which when commercialized will likely produce more biofuels per acre than the unmodified switchgrass using lower cost bioconversion processes [Fu 2011].
  • The potential of plant regulation for cell wall control is shown in tension wood, which is formed on the upper side of bent Populus trees and is characterized by more xylem cells, thicker cell walls, and higher cellulose/lower lignin content [Foston 2011].  A tension stress response study used molecular beam mass spectrometry (MBMS), transcriptome profiling using the Joint Genome Institute’s (JGI’s) Illumina platform, metabolomics, and proteomics.  These indicate major differential gene expression, and metabolites with the accumulation of phenolic glycosides on the tension.
  • A mass spectrometry-based cellulose assay has been developed at ORNL to quantify cellulosome absolute abundance in C. thermocellum as a function of growth state/substrate conditions in the presence of solid substrates [Dykstra 2013].  This approach is based on comparing labeled versus unlabeled cellulosome peptides in a multiple reaction monitoring triple quadrupole mass spectrometric experiment.
  • NREL researchers investigated the crystalline structures, biochemical properties, and mechanisms of action of CbhA (cellulosomal cellulase cellobiohydrolase A), which is the most complex cellulosomal cellulase in C. thermocellum and shows very high activity on crystalline cellulose.  Their research has provided insights into the binding mechanism of CbhA-CBM4 (carbohydrate-binding module to various polysaccharides [Brunecky 2012].
  • The feasibility of producing active, engineered cellulosomes was demonstrated at NREL (Ding Lab).  A key achievement this year was the successful expression of 46 of the 54 putative glycosidases in the C. thermocellum genome in E. coli in soluble form; these were purified and screened for activity against both oligosaccharide (cellohexaose) and crystalline cellulose (Avicel) substrates.  Against these substrates the most active cellulase was CbhA.  The development of new biocatalysts has demonstrated the feasibility of producing active, engineered cellulosomes.  A key achievement was the successful production of an engineered, active minicellululosome in E. coli [Xi 2013].
  • Transgenic switchgrass via viral induced gene silencing (VIGS) were generated and grown for analysis of recalcitrance and wall structure.  All genes tested represent putative recalcitrance genes.  Some switchgrass/VIGS plants show reduced recalcitrance.  One construct gives 20% increase in glucose released [Ramanna 2013].
  • Genes involved in switchgrass vascular system development and secondary cell wall biosynthesis identified and characterized.  This work spanned the switchgrass and biosynthesis activities.  A manuscript on expressed sequence tag (EST) data generation from switchgrass vascular tissues using laser capture microdissection [Srivastava 2010]; the corresponding 5765 vascular tissue ESTs were submitted to GenBank. dirigent1, dirigent2, peroxidase30, putative kinesin heavy chain, and cyclin-C homolog were identified as vascular specific genes.
  • Targeted analysis of genes for switchgrass improvement. As a basis for evaluation of transgenic switchgrass with reduced recalcitrance, a detailed baseline analysis of switchgrass ultrastructure and chemical composition during development was completed, which has demonstrated the correlations between various parameters (e.g. lignin content and composition, wall-bound-phenolic content and composition, soluble phenolic content and composition) and recalcitrance determined as saccharification efficiency [Shen 2009].   Also, completed were detailed anatomical and ultrastructural (electron microscopic) studies of switchgrass, with particular emphasis on cell wall structure during development in different internodes.
  • BESC explored thermophilic biodiversity to identify new isolates and functions.  Investigators have developed a method to screen and characterize hundreds of single-cell anaerobic thermophilic isolates in a single experiment [Hamilton-Brehm 2012].  The diversity of cellulolytic, thermophilic microorganisms was determined in thermal features within Yellowstone National Park using high-throughput cultivation approaches including flow cytometry at ORNL [Hamilton-Brehm 2012].  Members of the genus Caldicellulosiruptor dominate such biotopes up to 80 °C and a representative organism, C. obsidiansis, has been formally described, sequenced by JGI, and annotated at ORNL during Year 2 [Elkins 2010].  Unique hyperthermophilic organisms that have been isolated in Year 3 included xylan-degrading Dictyoglomus, species of Caloramator that produce ethanol are highly resistant to biomass pretreatment inhibitors, and a hyperthermophilic sulfate-reducing bacterium (Thermodesulfobacterium).  Enhanced cellulose utilization has been demonstrated this year using co-cultures of C. obsidiansis and Thermodesulfobacterium.
  • Significant effort has also been directed towards developing genetics technology in C. thermocellum and C. bescii and other related microbes.  Tools have been successfully developed to facilitate genetic manipulation of the thermophilic anaerobe C. thermocellum (Topt 60 °C) at Mascoma and Dartmouth this year.  These include development of replicating plasmids, positive and negative selectable markers, and unmarked gene deletions. These tools are being applied to modify and learn more about the physiology of C. thermocellum by targeting specific components in the cellulosome of C. thermocellum [Argyros 2011; Tripathi 2010].
  • An alternate strategy for developing CBP organisms requires adding cellulolytic capabilities to known ethanologens.  This effort was carried out under leveraged support by Mascoma.  This year, investigators have been able to demonstrate the expression of a wide range of enzyme activities necessary for lignocellulose deconstruction, including the expression of genes representing more than 15 unique glycosyl hydrolases in yeast.  The molecular and strain engineering tools needed for expressing and secreting key enzymes, including cellobiohydrolyases and endoglucanaces, have been refined, allowing for greater levels of functional expression. The strains of yeast developed in this effort have shown the ability to utilize model crystalline cellulose substrates, such as Avicel, as well as pretreated lignocellulosic substrates.
  • Ethanol tolerant strains of C. thermocellum showed a critical mutation in an adhE gene.  These were selected on Avicel and cellobiose and genome-resequenced through JGI. The use of integrated systems biology tools (sequencing, transciriptomics, proteomics, and metabolomics) in various combinations to target the underlying physiology of target microbes continues [Brown 2011; Shao 2011].
  • C. cellulolyticum transformed with a plasmid expressing alsS ilvC ilvD kivD adhA, and kivD yqhD alsS ilvC ilvD produces ~400 mg/L and ~650 mg/L of isobutanol over 13 days, respectively.  Once investigators have optimized isobutanol production in C. cellulolyticum to a goal of ≥1 g/L, this strategy will be applied to other cellulolytic microorganisms [Higashide 2011].
  • In support of the high-throughput pretreatment and hydrolysis (HTPH) system, in Year 3, a downscaled biomass compositional analysis was developed by the Wyman group at the University of California, Riverside (UCR) that is based on conventional wet chemistry techniques but requires significantly less material.  The method was shown to produce carbohydrate results statistically identical to conventional wet chemistry methods, while providing estimates of lignin and ash contents.  The HTPH system and newly-developed downscaled compositional analysis were utilized to investigate the ring-by-ring recalcitrance of two Populus tremuloides trees.  The study revealed substantial radial variations in composition and sugar digestibility patterns, but indicated that wood maturity did not impact recalcitrance [Demartini 2011].
  • The Davis group at NREL investigated compositional changes in low lignin alfalfa (Medicago sativa) lines with down-regulation of p-coumarate 3-hydroxylase (C3H) or hydroxycinnamoyl CoA:  shikimate hydroxycinnamoyl transferase provided by the NF.  The group measured an increased extractability of a lignin-like material composed predominately of hydroxyphenol (H)-lignin monomers units.  Lignin isolated from the transgenic lines had a decrease in molecular weight of the polymer compared to the control.  The findings suggest that the lignin molecular weight, and the ease with which it can be removed during processing, are influenced by altering the lignin precursor monomer ratio [Pu 2009].
  • The Ding group at NREL explored using a single molecule tracking technique based on optical total internal reflection fluorescence to track the dynamic behavior of green fluorescent protein (GFP)-tagged trichoderma (Tr)CBM1.  This group demonstrated that the linear and directional motion of single GFP-tagged TrCBM1 could be observed.  Using single molecule spectroscopy (SMS), enzyme reaction on single cellulose crystals, which demonstrates molecular motion of CBMs was observed.  The SMS study of the T. reesei cellobiohydrolase-I is the first to demonstrate experimentally the hydrolysis of the different hydrophobic faces [Liu 2011].
  • The Xu group at the University of Georgia, Athens (UGA) computationally identified the CesA superfamily members in 17 plant and algal genomes and conducted in-depth phylogenetic analyses; and identified a single-copy gene present in six chlorophyte green algae, which is most closely related to the land plant CslA (cellulose-synthase-like) and CslC families.  The group also identified a Csl family, providing further support for the conclusion from two recent studies that this Csl family represents a novel Csl family [Yin 2009].
  • CAZymes Analysis Toolkit (CAT).  Carbohydrate active enzymes (CAZymes) are a very important class of enzymes involved in biosynthesis and degradation of carbohydrates.  The BESC Knowledgebase team developed and published [Park 2010] an original computational approach for using this information based on the association-rule-learning algorithm applied to sequences from the CAZy database.  The developed tools for annotation and analysis allow prediction of CAZymes in relevant plants or microorganisms.  The tools provided significant power to assign CAZy families to a number of domains with previously unknown function and to predict carbohydrates relevant to enzymatic activities for many hypothetical proteins.  The set of tools is implemented as a web service, the CAT in the BESC public portal.

Year 2 (Oct. 2008 – Sept. 2009)


Established three Populus field trials with 1,100 genotypes at each location; variations among these diverse populations are being mined to identify key genes in biomass composition and sugar release. 

Created yeasts that ferment an application-relevant cellulosic feedstock to recoverable ethanol at high conversion with no added cellulose.

Developed a Biofuels Outreach lesson with the Creative Discovery Museum in Chattanooga, Tennessee; this interactive educational lesson is targeted for third-eighth graders.

Genomics:GTL Contractor-Grantee Workshop VII, Bethesda, Md., February 8-11 2009 (see extracted PDF page 26)


  • BESC identified many putative recalcitrance-associated plant cell wall biosynthesis and modification enzymes using four cycles of a coordinated plant transformation pipeline; this initiated entry of 317 genes, representing 342 constructs into the transformation pipeline.  The transformation team initiated the generation of 229 Populus transgenics (72 overexpression constructs; 157 knockdown constructs) and the generation of 105 switchgrass/foxtail millet transgenics (18 overexpression constructs; 19 knockdown constructs; 76 virus-induced gene silenced (VIGS) lines) [Nelson 2017].
  • The genetic map for Populus has been improved; more than 1000 additional markers have been placed on the map and linkage groups now correspond to full chromosome lengths, with a marker placed approximately every 1 cM. Cell wall chemistry QTLs and metabolite QTLs have been mapped and identified and a manuscript prepared [Drost 2009].  BESC has over two million ESTs from different tissues of Populus.
  • Many isolated genes involved in cell wall biosynthesis (several passing through the transformation pipeline).  Switchgrass mapping populations were grown in the fields in Georgia and Oklahoma.  Improved transformation systems were developed for switchgrass and a functional VIGS system in foxtail millet was developed [Ramanna 2013].
  • Interactions with the Characterization Group led to the characterization of genetically modified biomass samples with reduced recalcitrance.
  • BESC established three field trials containing the ~1000 Populus genotype association mapping population, and constructed genome-wide SNP libraries comprised of 100,000 informative SNP loci based on the whole-genome sequences of 10 P. trichocarpa genotypes [Slavov 2012].  All three common gardens were established in spring 2009, these include three clonal reps for each of 1100 genotypes at each of the three locations.
  • 100 genes are predicted as new cell-wall synthesis genes in Arabidopsis. 46 of these have independent supporting evidences. For 33 of these genes, their specific functional roles in cell-wall synthesis were predicted. A few of these genes have been submitted to the BESC transformation pipeline [Wang 2012].
  • 1,232 NAC transcription regulator genes, the largest plant-specific transcriptional factor gene family, from 11 genomes were functionally and evolutionarily analyzed based on co-expression, from which detailed functional roles for various subfamilies have been derived [Shen 2009].
  • BESC established protocols and equipment for a “microbial pipeline” for quantitative evaluation of microbial cellulose utilization rates. A prototype was completed for C. thermocellum biomass utilization but development continues. A dye method on a multi-well plate is working well for coarse screening.  An automated sampling system and elemental analyzer that gives very high-quality data has been developed.
  • For a more accurate microbial cellulose utilization investigators are:  testing other on-line measurements (OD, CO2, base addition) along with elemental analysis in order to infer cell and substrate concentrations with calibration against conventional laborious techniques and setting up multi-plexed fermentors capable of repetitive automated batch operation with a common feed [Holwerda 2013a].  This was supported by planned the American Recovery and Reinvestment Act (ARRA) equipment purchases.
  • To identify new biocatalysts, samples were collected from more than 50 sites: Verenium [Coleoptera larvae gut (3 samples), Biotraps (51), Rumen endosymbionts (8), Caecum endosymbionts (4)], Brookhaven National Laboratory [decaying biomass piles (3 sites) [Li 2009], and Oak Ridge National Laboratory (ORNL) [Yellow Stone National Park (numerous sites)] [Vishnivetskaya 2014].
  • Cellulolytic enzyme and domain profiles completed for five BESC microbes using data-mining tools within the microbial knowledgebase.  Many enzymes including numerous previously unannotated enzymes were found using CAZy and other tools.  This method was applied to genomes from the Yellowstone Obsidian pool.
  • Clear evidence of influence of S/G ratio variability on Populus recalcitrance [Studer 2011].
  • A second generation HTP pipeline was established at the National Renewable Energy Laboratory (NREL).  This automated HTP assay was implemented to analyze enzymatic cell wall hydrolysis at the rate of at least 500 tests per day (first tier cell wall digestibility assay). The first tier assay in place at NREL can conduct the targeted number of samples. In addition, a HTP pretreatment capability was also added by June 2009. The protocols established and first modified cell walls with enhanced digestibility identified using commercial enzymes. The 800 Populus associations study and 100 lignin biosynthesis mutant Populus samples have been examined for possible enhanced digestibility [Decker 2009; Selig 2010].
  • BESC created recombinant cellulose-fermenting yeasts expressing non-complexed cellulases. Not only do these yeasts ferment cellulose, they ferment an application-relevant cellulosic feedstock to recoverable ethanol at high conversion with NO added cellulose.
  • BESC implemented an HTP pretreatment and characterization pipeline (approximately 5,000 samples per month) to screen the structure, composition, and deconstruction of biomass, and identify the most promising samples for more detailed characterization incorporating pretreatment and digestibility.  The HTP pretreatment and characterization pipeline uses a combination of analytical pyrolysis (1300 samples/week) and a 96-well plate assay capable of performing sequential dilute-acid or hot-water pretreatment and enzymatic hydrolysis (2000 samples/week). This system was validated by analyzing 800 samples from a Populus association study, and 400 samples from an activating tagging study.
  • The characterization team evaluated four imaging techniques and an enzyme-linked immunosorbent assay (ELISA)-based technique using glycome profiling as new needed methods for chemical, structural, and surface features of biomass and biomass/biocatalyst interactions at nanoscale-to-microscale resolution.
    • The glycome profiling technique was validated on wild-type (Columbia) and xyloglucan xylosyltransferase mutants. Glycome profiling can quickly reveal changes in wall structure and composition and was chosen for continued research [Pattathil 2015].
    • Of these four imaging techniques evaluated, Scanning Near-field Acoustic Photothermal Spectroscopy [Tetard 2009] and ToF-SIMS were evaluated for having the highest potential and research will continue.  MicroCAT and MALDI-MS (Matrix Assisted Desorption/Ionization Mass Spectrometry) imaging were evaluated as having limited potential and research has been stopped for FY10.
  • NMR data for whole plant cell wall characterization of switchgrass and poplar biopolymers (i.e., lignin, hemicellulose, cellulose) was used to populate a Naïve Database. A more sophisticated database created with intuitive interface and partially populated. Initial testing of sophisticated database shows it will be extremely valuable for identifying structural features in 1D and 2D spectra [York 2017].
  • BESC initiated a novel educational outreach program working with educators at the Creative Discovery Museum in Chattanooga, Tennessee, to develop a curriculum targeting 4th-6th graders.  Approximately 120 lessons were taught in over 50 schools in Georgia and Tennessee reaching over 3,000 school children.

Year 1 (Oct. 2007 – Sept. 2008)


Established and put in operation a large-scale gene transformation pipeline; pipeline targets plant cell wall mutants and aids the selection of high-priority, recalcitrance-associated genes. Screened multiple samples from 100 Populus plants and 50 switchgrass plants to identify native variants with modified composition.

Developed a preliminary functional model of the cellulosome, a multi-enzyme complex that breaks down cellulose and hemicellulose found in plant cell walls.

Implemented a high-throughput characterization pipeline (capable of screening more than 2,000 samples per month) to screen biomass composition and identify the most promising samples for more detailed characterization.

Completed omics analyses of initial wild type consolidated bioprocessing (CBP) microbes.

Genomics:GTL Contractor-Grantee Workshop VI, Bethesda, Md., February 10-13 2008 (see extracted PDF page 2, bottom)


  • BESC established preliminary models of wall biosynthesis pathways for lignin in Populus and switchgrass.
  • Since transformation efficiencies for targeted feedstocks was low, BESC organized a collective plant gene transformation pipeline effort in switchgrass and Populus and ran this through two cycles of submission, review, and initiation of transformants of target recalcitrance genes with about 200 genes constructs for knockdown (KD), knockout (KO), or overexpression (OE) [Nelson 2017].
  • Multiple samples from 790 Populus plants were screened to identify native variants with modified composition.  This study captured the full known range of Populus variability and showed that lignin and syringyl to guaiacyl (S/G) ratios are higher in the stem than in the root. Many quantitative trait loci (QTLs) are co-located. There were 580 Populus genotypes propagated into the first common garden [Yin 2010].
  • Multiple samples from 50 switchgrass plants were screened to identify native variants with modified composition. A mapping population of ~250 progeny [Serba 2013] was planted. A 6x fosmid library was analyzed to characterize candidate biomass genes completed.
  • The components of the cellulosome (a key multifunction enzyme complex in CBP) were identified by a broad ’omics study. Combined transcriptomics and proteomics have identified over 20 previously unknown cellulosome components in C. thermocellum [Raman 2009].  Analysis of existing genomes has revealed new compositional trends for cellulosomal versus non-cellulosomal cellulose systems. There were 460 mutant single nucleotide polymorphisms (SNPs) found in an ethanol tolerant C. thermocellum strain [Brown 2011].
  • This led to a preliminary functional model of the cellulosome: key glycosyl hydrolase families identified, with new components (including non-enzymatic carbohydrate binding modules (CBMs) and fibronectins), and mesoscale models built of the (dockerin-cohesin) population of scaffoldins identified [Bomble 2010].
  • Investigators sampled 5 different microbial environments (thermophilic and mesophilic) to identify new biocatalysts, including Yellowstone hot springs, shipworm gut, Ruminant manure, Hu-Hu grub gut, and Populus compost. A new isolate (Caldicellulosiruptor sp.) was identified by consortia with Yellowstone National Park [Hamilton-Brehm 2010].
  • BESC implemented an HTP characterization pipeline (capable of >2,000 samples per month) to screen the composition of biomass and identify the most promising samples for more detailed characterization. This compositional analysis was by analytical pyrolysis and ran more than 1200 Populus samples over a course of about 7 days. The data was stored in LIMS (Laboratory Information Management System) [Decker 2009].
  • A prototype HTP pipeline with a steel 96-well plate system was developed for dilute acid pretreatment and enzymatic co-hydrolysis and an invention disclosure was submitted. The total sugar yields are comparable to manual classical approach results [patent, Studer, et al., US2010/0105570, 2010].
  • Data sharing systems were established within BESC focusing on a LIMS. Thousands of samples and data tracked.
  • The LIMS was established as the recording mechanism for a simplified Material Transfer Agreement implementation [Miller 2009].
  • Three techniques under development and evaluation for novel imaging technique for chemical, structural, and surface features of biomass and biomass-biocatalyst interactions at nanoscale to microscale resolution: Scanning Spectroscopy, Polysaccharide Antibody Epitopes, and Coherent Anti-Stokes Raman Scattering Microscopy.
  • Several BESC workshops were held for integration: Plant Gene Curation workshop, Characterization workshop, Enzymatic working group, and Microbial working group.
  • A commercialization council was established in April 2008 with representatives from each partner institution; four conference calls have been held.

Year 0 (Oct. 2006 – Sept. 2007)


Department of Energy selected three Bioenergy Research Centers; Centers are intended to accelerate basic research in the development of cellulosic ethanol and other biofuels. (Details)

BESC research building

BESC researchers received funding and began work.

BioEnergy Science Center 2007-2017