In planta expression of A. cellulolyticus Cel5A endocellulase reduces cell wall recalcitrance in tobacco and maize
© Brunecky et al; licensee BioMed Central Ltd. 2011
Received: 17 September 2010
Accepted: 26 January 2011
Published: 26 January 2011
The glycoside hydrolase family 5 endocellulase, E1 (Cel5A), from Acidothermus cellulolyticus was transformed into both Nicotiana tabacum and Zea mays with expression targeted to the cell wall under a constitutive promoter. Here we explore the possibility that in planta expression of endocellulases will allow these enzymes to access their substrates during cell wall construction, rendering cellulose more amenable to pretreatment and enzyme digestion. Tobacco and maize plants were healthy and developed normally compared with the wild type (WT). After thermochemical pretreatment and enzyme digestion, transformed plants were clearly more digestible than WT, requiring lower pretreatment severity to achieve comparable conversion levels. Furthermore, the decreased recalcitrance was not due to post-pretreatment residual E1 activity and could not be reproduced by the addition of exogenous E1 to the biomass prior to pretreatment, indicating that the expression of E1 during cell wall construction altered the inherent recalcitrance of the cell wall.
Plant cell walls are composed of three basic structural biomolecules: cellulose, hemicellulose and lignin. The conversion of cellulose and hemicellulose to sugar is the primary step in the fermentative conversion of biomass to fuels and chemicals . Natural enzymatic digestion of plant matter is a slow and complex process which must become more cost-effective if commercial biofuel production is to become a reality. Cellulases must penetrate hemicellulose and lignin barriers to bind and digest the cellulose. It is the relatively poor accessibility of substrates to enzymes due to the strong associations between plant cell wall components that explains most of this recalcitrance and makes costly thermochemical pretreatments necessary.
Approaches to the deconstruction of plant cell walls to fermentable free sugars typically employ various thermochemical pretreatments followed by an enzymatic hydrolysis step. Whereas several studies have examined the potential to produce and harvest active cellulase from plants, no work has been reported examining the effect of this expression on the amenablilty of these plants to biomass conversion [2–12]. Pretreatment of the biomass through heat, physical force and/or chemical catalysts reduces the inherent strength of the cell wall by altering the physical arrangement of these three components and breaking chemical bonds that give these structures strength. Pretreatment renders the cellulose and hemicellulose components more amenable to conversion to monomeric sugar by enzymes . This transformation of biomass to a more maleable form, however, comes at a cost. The severity required to achieve this reduction in recalcitrance is often high enough to negatively affect the economics of the conversion process through increased capital construction costs, degradation products and inhibitor formation [14, 15].
The A. cellulolyticus E1 endoglucanase is a well-characterized glycoside hydrolase family 5 (cel5A) endocellulase known to be thermally tolerant and generally very stable over broad ranges of pH . When added exogenously to biomass prior to pretreatment, it does not have any effect on pretreatability or post-pretreatment enzymatic digestibility. We expressed the Acidothermus cellulolyticus glycoside hydrolase family 5 endoglucanase (E1) in both Nicotiana tabacum and Zea mays and targeted its expression to the cell wall under a constitutive promoter. Instead of assuming adequate penetration into mature cell walls, we theorized that in planta expression would allow the enzyme to access a wider range of cell wall components and compartments as they were being constructed. Because E1 has an optimum temperature of approximately 80°C, we suspected that activity during plant growth would be limited and would have minor if any impact on plant phenotype and health.
Estimation of E1cd in transgenic stover and tobacco
Pretreatment and saccharification of biomass
Imaging of E1 maize and tobacco
We report here that tobacco and maize expressing heterologous E1 endoglucanase from Acidothermus cellulolyticus were observed to become less recalcitrant compared to wild-type biomass when subjected to pretreatment and post-pretreatment enzymatic hydrolysis. This reduction in recalcitrance was manifest through lower severity requirements to achieve comparable levels of conversion to wild-type biomass. Our studies indicate that the decreased recalcitrance was not due to post-pretreatment residual E1cd activity and could not be reproduced by the addition of exogenous E1cd to the biomass prior to pretreatment, indicating that the expression of E1cd during cell wall construction altered the inherent recalcitrance of the cell wall. In our experiments, expression of E1cd in maize increased the digestibility of biomass cell walls after pretreatment at expression levels far below those normally required for efficient E1 hydrolysis of cellulose. It also enabled lower-severity pretreated corn stover to become as digestible as higher-severity pretreated stover with the same enzyme loadings. In addition to decreased pretreatment severity requirements, the expression of E1 in corn stover is likely to enable decreased enzyme loadings for hydrolysis, decreased inhibitor formation and decreased degradation product formation. This effect is also likely to extend to other endoglucanases, especially those from glycoside hydrolase family 5.
We observed a shared phenomenon when expressing an endoglucanase in both corn stover and tobacco; that is, E1-expressing plants are more easily saccharified by cellulase enzymes following various severity pretreatments. Our results indicate that E1 expression in both corn and tobacco enabled reduction of the required pretreatment severity to permit the same level of sugar conversion as was obtained with higher-temperature pretreatments of wild-type samples. The possibility of reducing pretreatment severity is highly relevant to the biofuels industry because of the potential cost savings from lower energy and enzyme usage, reduced inhibitor formation and sugar losses, decreased catalyst addition and lower costs of facility construction.
Throughout the digestion experiments, transgenic E1 biomass, whether stover or tobacco, was shown to be either more digestible than the wild type after identical pretreatment or as digestible as more severely pretreated wild-type biomass. For example, with the 100 mg of enzyme per gram of biomass digestions, the yield of glucose from E1-expressing transgenic corn stover pretreated at 140°C appeared to be comparable to the yield from nontransgenic corn stover pretreated at about 170°C (Figure 5). This effect was even more distinct at lower enzyme loadings (15 mg of protein per gram of biomass; Figure 4). Tobacco also showed that transgenic expression of E1 benefited enzymatic conversion, although this effect was more pronounced at lower severities (Figure 3). As incubation of the biomass after the addition of exogenous E1 before pretreatment did not have any discernible impact on enzymatic conversion, we conclude that the effect is not due to residual post-pretreatment E1 activity. Also, as the incubation time was fairly long, the exogenous E1 would be expected to cleave some of the cellulose before pretreatment, although presumably the accessibility would be limited to outer layers of cellulose.
Our work in extracting E1 from the cell walls and quantifying it by Western blot analysis indicated that the E1 protein is tightly associated with the cell wall. Our normal extraction protocol of 100% ethylene glycol (as used in the cel7a purification method) did not remove detectable amounts of E1 from the biomass even after increasing the extraction temperature to ~96°C. Treatment by boiling with NuPAGE LDS sample buffer did remove E1 from the cell walls, indicating that the transgenic E1, even though it lacks a carbohydrate binding module, was tightly associated with the cell wall.
Imaging of plant cell wall sections by anti-E1 antibody localization clearly showed the presence of E1 throughout the cell wall, specifically in the thicker sclerenchyma cells of the rind and vascular bundles. Stover samples showed much higher levels of intrawall E1 deposition than did the tobacco samples, which seemed to have more E1 localized to the inner layers of the cell wall (Figures 8 and 9). This may indicate that measured E1 levels were higher in tobacco simply because the E1 was easier to extract. It may also explain why the tobacco conversion enhancement due to E1 expression was not apparent at the highest pretreatment severity; that is, the innermost layers of the tobacco cell wall were not as disrupted by pretreatment as were those of stover, which had more broad distribution of E1 throughout the cell wall. It is unclear from this work whether the localization was due to a preference of E1 to bind to a particular wall type, to a difference in cell-type expression or to some other variable (such as cell wall-type permeability to antibody). Additional maize transformation vectors have been generated in which E1 is driven by other strong promoters and fused to alternative signal peptides to target accumulation in the vacuole and endoplasmic reticulum, and these constructs will be used to generate additional transgenic lines for study. Entrained E1 may enhance the conversion, but only to the extent that it is distributed throughout the cell wall. Broader distribution of E1 in the stover cell wall interior could have allowed more rapid and complete digestion, which is indicated by the increased extent of conversion of E1 stover compared to E1 tobacco under identical high-enzyme loadings. Without the increased pretreatment efficacy from intercalated E1, the tobacco digestions all may have simply reached their limit at the 170°C pretreatment condition. Localization of cellulases using immunocytochemistry followed by transmission electron microscopy as described by Donohoe et al.  supports this idea, as cell wall disruption and enzyme penetration are directly proportional to pretreatment severity.
Our pretreatment, digestion, and imaging results lead us to believe that the cellulose in the E1 stover and tobacco has been modified by the expression of E1 during plant growth, biomass storage or both. On this basis, we conclude that the increased enzymatic conversion of the transgenic E1 biomass is due to the intrawall (and possibly intracellulose microfibril) localization of the E1 protein resulting from expression during cell wall formation. This conclusion is supported by the observation that wild-type corn stover treated with exogenous E1 prior to pretreatment was not as digestible as the transformed stover. We believe that this result was due to the ability of transgenic E1 to access more cell wall compartments containing cellulose and perhaps even the cell membrane-cell wall space occupied by the nascent (growing) cellulose microfibril during cellulose synthesis than was possible with the simple external addition of enzyme. This idea is supported by our imaging data showing that E1 expressed during plant growth is targeted to the cell wall, where it may work by nicking the cellulose chains as they are formed or laid down, resulting in either less crystalline cellulose material or simply more free chain ends for cellobiohydrolases, which could result in increased conversion. It is not known whether the reduced activity of this thermal tolerant enzyme at growth condition also benefited this effect. Alternatively, expression of the E1 enzyme in plant tissue may likely have led to increased conversion through a nonenzymatic effect, such as the ability to bind strongly to polysaccharides. By simply binding to the cellulose microfibril during plant growth and development, E1 may decrease the recalcitrance of the cell wall to chemical pretreatment. The comparable results between tobacco and corn expressing E1 suggest that low-level expression of E1 or other endoglucanases in plants may be a general phenomenon by which the conversion of biomass may be improved.
Materials and methods
E1 gene and protein
The E1 gene used to transform tobacco and maize was originally isolated from A. cellulolyticus and was later truncated to remove the carbohydrate-binding module and linker, resulting in a gene containing only the E1 catalytic domain (E1cd) . E1cd was expressed and purified from Streptomyces lividans for use as a control as described previously .
Tobacco was transformed according to standard methods using Agrobacterium tumefaciens , and the transformed plants grew well. We received the E55 tobacco (apo E1cd) seeds from Sandra Austin-Phillips at the University of Wisconsin-Madison , identified stable lines and grew the biomass under controlled greenhouse conditions.
To verify the presence of the E1 gene, the third or fourth leaf from the shoot apex was used for protein extraction. Leaf samples were harvested 2 to 3 hours into the light period, cut into approximately 1-cm2 pieces and pooled for homogenization. An enzyme assay, sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis and Western blot analysis were carried out as described previously . For estimation of E1cd in the samples, the ground biomass was extracted directly by boiling 20 mg of biomass in 100 μL of NuPAGE LDS sample buffer for 10 minutes and loading 20 μL of supernatant onto a NuPAGE 3-(N-morpholino)propanesulfonic acid gel (Invitrogen). After electroblotting onto polyvinylidene fluoride (PVDF), the blot was analyzed for E1cd using mouse anti-E1 monoclonal primary antibody and goat antimouse secondary antibody. E1cd purified from expression in Streptomyces lividans was used as a quantitative control for comparison with a dilution series of the E1 tobacco extracts (Figure 1).
Maize was used as a model lignocellulosic substrate, as corn stover (the nongrain parts of the harvested plant) is a strong potential biofuels feedstock. Ransom et al.  produced E1cd-transformed maize in which E1 was driven by the 35S CaMV promoter and targeted to the apoplast. E1 corn was grown to maturity under greenhouse conditions, then allowed to dry down in the greenhouse before being harvested. Individual plants were screened for E1 activity, and initial studies focused on two transgenic lines designated E1-1 and E1-7, which showed high and low E1 activity, respectively. Stalks were further air-dried to a moisture content of approximately 10%, and the entire stalk (that is, stover) was milled to pass a 20-mesh screen.
E1 expression was estimated by Dot blot assay (not shown) as well as Western blot analysis (Figure 1). For the Dot blot assay, 10 mg of 80-mesh stover was extracted three times in 100 μL of 2% SDS by boiling it for 10 minutes, centrifuging, reextracting the solids and combining all supernatants. For the Western blot analysis, ~5 mg of biomass was directly extracted with 50 μL of NuPAGE LDS sample buffer by boiling it for 10 minutes. After running electrophoresis, the gel was electroblotted onto a PVDF membrane using the Invitrogen NuPAGE blotting apparatus. Both Dot blot and Western blot analyses were visualized using the antimouse Western Breeze Chromogenic Kit (Invitrogen) with anti-E1 mouse monoclonal antibody as the primary antibody.
The activity of E1 in milled stover and tobacco was estimated on the basis of MUC activity. In a 16 × 100-mm test tube, we added 10 mg of 80-mesh biomass, 200 μL of 1.0 M sodium acetate, pH 5.0, and 1,600 μL of H2O. After heating the sample to 84°C for 5 minutes, 200 μL of 2.0 mg/mL MUC substrate was added to the reaction mix. Each sample was analyzed in triplicate. Samples of 100 μL were taken at 0.0, 0.5, 1.0 and 2.0 hours (for stover) or every minute for 15 minutes (for tobacco), quenched with 200 μL of 1.0 M Na2CO3 and fluorescence was measured at 360-nm excitation/465-nm emission in a Tecan GENios microplate reader (MTX Lab Systems, Vienna, VA, USA). E1cd levels in the stover were estimated by comparison of activity to purified S. lividans-expressed E1cd activity under the same conditions (Figure 2).
Both tobacco and stover rind were milled to 20 mesh in a Wiley knife mill(Thomas Scientific, Swedesboro, NJ, USA) and analyzed for sugar composition according to National Renewable Energy Laboratory (NREL) standard laboratory protocol NREL/TP-510-4618 . Samples were vacuum-impregnated with water at 10% (wt/vol) biomass. Exogenous E1 was added to select biomass samples at 0.39 mg of protein per gram biomass, and then all samples were incubated overnight at 80°C. Pretreatment was performed in gold-plated batch reactors (15 mL) with 1% (wt/vol) sulfuric acid at 5% (wt/vol) solids loading. The reactor was rapidly heated to 110°C, 140°C or 170°C within approximately 5 minutes by immersion in a sand bath at 10°C to 20°C above the target temperature. The reactor was then rapidly transferred to a second sand bath maintained at the target temperature, where it remained for 10 minutes. The reactor was removed from the sand bath and rapidly cooled by immersion in ice water. In the case of tobacco, a secondary control consisting of 50% wild-type and 50% E1 tobacco was mixed and pretreated as above. Following pretreatment, all solids were washed with distilled water until the pH of the rinse was measured to be above 4.0.
Because of sample size limitations of the pretreatment reactor chambers, all samples were pretreated in triplicate at each temperature, then the replicates were pooled and washed with water until the pH was greater than ~4.5. Pretreatment replicates were from the same bulk batch of stover or tobacco. Three aliquots of each of the pooled, washed solids were taken for analysis by enzyme digestion. Duplicate samples were used for the compositional analysis of each sample.
Analysis of pretreated solids
To provide a basis for the maximum theoretical sugar yield achievable from each substrate during enzymatic hydrolysis, portions of the washed, pretreated solids were dried and subjected to the standard two-stage sulfuric acid hydrolysis method for determining structural carbohydrates in lignocelluloses as described by Sluiter et al. . The glucan and xylan content of each pretreated sample was calculated from the glucose and xylose released as described by Sluiter et al.
Saccharification of all pretreated and washed biomass slurries was conducted at pH 5.0 in 50 mM citrate buffer with 0.01% sodium azide added as an antimicrobial agent. One percent (wt/vol) slurries of pretreated solids were hydrolyzed with either a low (15 mg of cellulose per gram of biomass) or high (100 mg of cellulose per gram of biomass) loading of a commercial cellulase (Spezyme CP) for 72 hours at 50°C. Additional saccharifications of transgenic maize were conducted with a purified cellobiohydrolase, cel7A from Trichoderma reesei, to assess the effect of plant-expressed E1 on the ability of this exocellulase to digest cellulose. The cel7A was purified from a commercial T. reesei cellulase (Spezyme CP) by using column chromatography (dual anion exchange followed by affinity on cellulose and size-exclusion chromatography) and loaded at 15 mg of protein per gram of biomass. All reactions were supplemented with 5.0 mg of β-glucosidase per gram of biomass to limit cellobiose inhibition. Whole cellulase digestions were supplemented with Novo188 (Novozymes NA Inc., Franklinton, NC, USA), while Cel7a digestions were augmented with purified Aspergillus niger β-glucosidase. Glucose concentrations in the hydrolysates were determined on a Jasco high-performance liquid chromatography system (Jasco, Great Dunmow, UK) running a lead-based Shodex sugar column (model SP0810; Kawasaki, Japan) heated to 80°C with water flowing at 0.6 mL/min as the mobile phase. All saccharification experiments shown in Figures 3-7 were run in triplicate and error bars show standard deviations.
Several 3-mm-wide sections of unpretreated tobacco rind were cut and dehydrated in a graded ethanol series. LR White resin (Electron Microscopy Sciences, Hatfield, PA, USA) was serially added, and the sections were left overnight in 100% LR White. For optimal infiltration, samples were placed in a vacuum microwave oven for 1 and 5 minutes during each ethanol and resin concentration rinse, respectively. After 2 days, the samples were transferred to flat-bottomed "BEEM" capsules (Electron Microscopy Sciences, Hatfield, PA, USA) and filled with 100% LR White. Polymerization occurred overnight in a vacuum oven set to 60°C. Resin blocks were trimmed with a Leica EM Trim trimmer (Leica Microsystems, Bannockburn, IL, USA), and 2.0-μm sections were cut using glass knives and a Leica Ultracut UCT ultramicrotome (Leica Microsystems). Glass knives were made from a 6-mm glass bar and cut on a LKB Bromma 7800 KnifeMaker (Bromma, Sweden).
Confocal laser scanning microscopy images of E1 antibody-stained samples at ×200, ×600 or ×1,200 magnification were obtained using a Nikon Eclipse 90i Spectral Laser Confocal Microscope (Nikon Instruments, Inc., Melville, NY, USA). For confocal microscopy, 2.0-μm thin sections of maize rind were blocked with 1% (wt/vol) bovine serum albumin for 45 minutes to minimize nonspecific binding, washed five times with phosphate-buffered saline (PBS) and incubated with mouse anti-E1cd monoclonal antibody for 45 minutes. After washing with PBS, an Alexa Fluor 488-conjugated antimouse immunoglobulin G secondary antibody (Invitrogen) was added to localize bound anti-E1cd. Excess secondary antibody was washed using PBS. Autofluorescence and nonspecific signal were subtracted out using Nikon EZ-C1 software (Nikon Instruments, Inc. Melville, NY, USA) to deconvolute the Alexa Fluor signal from the background plant autofluorescence.
This work was supported by the Department of Energy (DOE) Office of the Biomass Program and by grant DE-FG36-06G016107 from the U.S. Department of Agriculture-DOE Joint Program Biomass Research and Development Initiative. E1 tobacco seed was provided by Sandra Austin-Phillips, and E1 corn was provided by Mariam Sticklen.
- Wyman CE: BIOMASS ETHANOL: technical progress, opportunities, and commercial challenges. Annu Rev Energy Environ 1999, 24:189–226.View Article
- Verma D, Kanagaraj A, Jin SX, Singh ND, Kolattukudy PE, Daniell H: Chloroplast-derived enzyme cocktails hydrolyse lignocellulosic biomass and release fermentable sugars. Plant Biotechnol J 2010, 8:332–350.View Article
- Fan ZM, Yuan L: Production of multifunctional chimaeric enzymes in plants: a promising approach for degrading plant cell wall from within. Plant Biotechnol J 2010, 8:308–315.View Article
- Dai ZY, Hooker BS, Anderson DB, Thomas SR: Improved plant-based production of E1 endoglucanase using potato: expression optimization and tissue targeting. Mol Breed 2000, 6:277–285.View Article
- Teymouri F, Alizadeh H, Laureano-Pérez L, Dale B, Sticklen M: Effects of ammonia fiber explosion treatment on activity of endoglucanase from Acidothermus cellulolyticus in transgenic plant. Appl Biochem Biotechnol 2004, 113–116:1183–1191.View Article
- Ziegelhoffer T, Raasch JA, Austin-Phillips S: Expression of Acidothermus cellulolyticus E1 endo-β-1,4-glucanase catalytic domain in transplastomic tobacco. Plant Biotechnol J 2009, 7:527–536.View Article
- Ziegelhoffer T, Raasch JA, Austin-Phillips S: Dramatic effects of truncation and sub-cellular targeting on the accumulation of recombinant microbial cellulase in tobacco. Mol Breed 2001, 8:147–158.View Article
- Sainz MB: Commercial cellulosic ethanol: the role of plant-expressed enzymes. In Vitro Cell Dev Biol 2009, 45:314–329.View Article
- Biswas GCG, Ransom C, Sticklen M: Expression of biologically active Acidothermus cellulolyticus endoglucanase in transgenic maize plants. Plant Sci 2006, 171:617–623.View Article
- Oraby H, Venkatesh B, Dale B, Ahmad R, Ransom C, Oehmke J, Sticklen M: Enhanced conversion of plant biomass into glucose using transgenic rice-produced endoglucanase for cellulosic ethanol. Transgenic Res 2007, 16:739–749.View Article
- Ransom C, Balan V, Biswas G, Dale B, Crockett E, Sticklen M: Heterologous Acidothermus cellulolyticus 1,4-β-endoglucanase E1 produced within the corn biomass converts corn stover into glucose. Appl Biochem Biotechnol 2007, 137:207–219.View Article
- Mei C, Park SH, Sabzikar R, Qi CF, Ransom C, Sticklen M: Green tissue-specific production of a microbial endo-cellulase in maize ( Zea mays L.) endoplasmic-reticulum and mitochondria converts cellulose into fermentable sugars. J Chem Technol Biotechnol 2009, 84:689–695.View Article
- Wyman CE, Dale BE, Elander RT, Holtzapple M, Ladisch MR, Lee YY: Coordinated development of leading biomass pretreatment technologies. Bioresour Technol 2005, 96:1959–1966.View Article
- Lee RL, Charles EW, Tillman UG: Biocommodity engineering. Biotechnol Prog 1999, 15:777–793.View Article
- Mosier N, Wyman C, Dale B, Elander R, Lee YY, Holtzapple M, Ladisch M: Features of promising technologies for pretreatment of lignocellulosic biomass. Bioresour Technol 2005, 96:673–686.View Article
- Baker JO, Adney WS, Nieves RA, Thomas SR, Wilson DB, Himmel ME: A new thermostable endoglucanase, Acidothermus cellulolyticus E1: synergism with Trichoderma reesei CBH I and comparison to Thermomonospora fusca E 5 . Appl Biochem Biotechnol 1994, 45–46:245–256.View Article
- Donohoe BS, Selig MJ, Viamajala S, Vinzant TB, Adney WS, Himmel ME: Detecting cellulase penetration into corn stover cell walls by immuno-electron microscopy. Biotechnol Bioeng 2009, 103:480–489.View Article
- Nieves RA, Chou YC, Himmel ME, Thomas SR: Quantitation of Acidothermus cellulolyticus E1 endoglucanase and Thermomonospora fusca E 3 exoglucanase using enzyme-linked immunosorbent assay (ELISA). Appl Biochem Biotechnol 1995, 51–52:211–223.View Article
- Dai ZY, Hooker BS, Anderson DB, Thomas SR: Expression of Acidothermus cellulolyticus endoglucanase E1 in transgenic tobacco: biochemical characteristics and physiological effects. Transgenic Res 2000, 9:43–54.View Article
- Sluiter A, Hames B, Ruiz R, Scarlata C, Sluiter J, Templeton D, Crocker D: Determination of structural carbohydrates and lignin in biomass: laboratory analytical procedure. In Tech. Rep. NREL/TP-510–4618. Golden, CO: National Renewable Energy Laboratory; 2008.
- Selig MJ, Knoshaug EP, Adney WS, Himmel ME, Decker SR: Synergistic enhancement of cellobiohydrolase performance on pretreated corn stover by addition of xylanase and esterase activities. Bioresour Technol 2008, 99:4997–5005.View Article
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.