Transgenic Research 9: 43–54, 2000. © 2000 Kluwer Academic Publishers. Printed in the Netherlands.
Expression of Acidothermus cellulolyticus endoglucanase E1 in transgenic tobacco: biochemical characteristics and physiological effects Ziyu Dai1∗ , Brian S. Hooker1 , Daniel B. Anderson1 & Steven R. Thomas2 1 Pacific
Northwest National Laboratory, P.O Box 999, MSIN: K2-10, Richland, WA 99352 Renewable Energy Laboratory, 1617 Cole Boulevard, Golden, CO 80401-3393
Received 20 October 1999; revised and accepted 10 January 1999
Key words: Acidothermus cellulolyticus, cellulase, endoglucanase (E1), Nicotiana tabacum, heterologous expression, leaf specific promoter RbcS-3C, chloroplast transit peptide RbcS-2A, alfalfa mosaic virus 50 -untranslated leader Abstract The expression of the Acidothermus cellulolyticus endoglucanase E1 gene in transgenic tobacco (Nicotiana tabacum) was examined in this study, where E1 coding sequence was transcribed under the control of a leaf specific Rubisco small subunit promoter (tomato RbcS-3C). Targeting the E1 protein to the chloroplast was established using a chloroplast transit peptide of Rubisco small subunit protein (tomato RbcS-2A) and confirmed by immunocytochemistry. The E1 produced in transgenic tobacco plants was found to be biologically active, and to accumulate in leaves at levels of up to 1.35% of total soluble protein. Optimum temperature and pH for E1 enzyme activity in leaf extracts were 81◦C and 5.25, respectively. E1 activity remained constant on a gram fresh leaf weight basis, but dramatically increased on a total leaf soluble protein basis as leaves aged, or when leaf discs were dehydrated. E1 protein in old leaves, or after 5 h dehydration, was partially degraded although E1 activity remained constant. Transgenic plants exhibited normal growth and developmental characteristics with photosynthetic rates similar to those of untransformed SR1 tobacco plants. Results from these biochemical and physiological analyses suggest that the chloroplast is a suitable cellular compartment for accumulation of the hydrolytic E1 enzyme. Abbreviations: E1 – endoglucanase; MES–2-(N-morpholino)ethanesulfonic acid; MS–Murashige and Skoog; MU–4-methylumbelliferon; MUC–4-methylumbelliferyl-β-D-cellobioside; PAGE–polyacrylamide gel electrophoresis; PIPES–piperazine-N,N 0-bis-2-ethanesulfonic acid.
Introduction Cellulose, an unbranched, linear polymer of Dglucose residues linked by β-1,4 glycosidic bonds, is a critical structural component of the cell wall of higher plants. It accounts for over one half of the carbon in the biosphere, which is the main component of primary and secondary biomass wastes generated through various domestic, commercial and industrial ∗ Author for correspondence: Bioprocessing Group, Environmental Technology Division, Pacific Northwest National Laboratory, P.O. Box 999, K2-10, Richland, WA 99352; Tel.: 509-3752169; Fax: 509-372-4660; E-mail: [email protected]
activities, including traditional industrial activities directly associated with forestry and agriculture. Until recently most of these cellulosic wastes were disposed in landfills (Duff & Murray, 1996). The hydrolytic degradation of cellulose is, therefore, of ecological and economic importance, with great potential for conversion of lignocellulosic biomass to fuel ethanol and other important industrial chemicals (Lynd et al., 1991; Gilbert & Hazlewood, 1993). Two features of the cellulose molecule make it extremely resistant to enzymatic hydrolysis. First, long cellulose chains formed by the β-1,4-glycosidic linkage are completely insoluble in water. Second, in
44 plant tissues, cellulose polymers adhere strongly to one another in overlapping parallel arrays by intermolecular hydrogen bonds to form cellulose microfibrils. Nevertheless, there are many fungal and bacterial species that are able to break down cellulose. These organisms produce a large ensemble of different cellulases, often acting in synergy on the crystalline cellulose surface to break down β-1,4-glycosidic linkages. There are at least three major groups of cellulases involved in the hydrolysis of the glycosidic bonds of cellulose to yield a single product, glucose: β1,4-endoglucanase (E.C. 220.127.116.11), β-1,4-exoglucanase (E.C. 18.104.22.168), and β-D-glucosidase (E.C. 22.214.171.124) (Duff & Murray, 1996). The β-1,4-endoglucanase mainly cleaves β-1,4-D-glucans at random along the polysaccharide chain, whereas exoglucanases cleave cellobiosyl units from either the reducing or the nonreducing end of cellulose strand. The β-D-glucosidase hydrolyses cellobiose units to produce two glucose molecules. Acidothermus cellulolyticus, a thermotolerant, cellulolytic actinomycete bacterium, actively secretes a number of highly thermostable and active cellulosedegrading enzymes (Mohagheghi et al., 1986; Tucker et al., 1989). One component of this cellulase system has been purified, characterized, and named β-1,4endoglucansase (E1). The microbial purified E1 enzyme is extremely thermostable (Tucker et al., 1989). Analysis of the amino acid sequence derived from the sequence of the E1 gene shows that it is a member of glycosyl hydrolase family 5, subfamily 1. The mature E1 protein with 521 amino acids (1563 bp) consists of a catalytic domain, a proline/serine/threonine-rich linker region, and a cellulose-binding domain with a mature protein size of about 72 kD (Sakon et al., 1996). Typically, industrial enzymes, such as cellulase and xylanase, are primarily produced by fermentative processes. The cost for fermentative production of cellulose degrading enzymes is prohibitively high to be considered for ‘low value’ applications such as the conversion of lignocellulosic feedstock to fuel ethanol. However, in terms of cost effectiveness for producing biomass, the growing of crops in the field, in general, can compete with any other production system. It is inexpensive and biomass can be produced in bulk quantities, and requires limited infrastructure. Enzymes (e.g. those in food, feed, or processing industries) are products that can be produced economically in genetically engineered plants (Verwoerd et al., 1995). In an effort to reduce the cost of production
of cell wall degrading enzymes, alternative production in transgenic plants is currently being studied (Herber et al., 1995; Jensen et al., 1996; Liu et al., 1997; Dai et al., 1999; Ziegelhoffer et al., 1999). So far, heterologous expression of the E1 gene has been studied in bacteria and other microorganisms (Lastick et al., 1996; Thomas et al., 1996; Adney et al., 1998). In this study we have examined the expression of the E1 enzyme in transgenic tobacco plants under the control of the leaf specific Rubisco small subunit (tomato RbcS-3C) promoter (Sugita et al., 1987) and (tomato RbcS-2A) chloroplast transit peptide (Pichersky et al., 1986). In addition, we have elucidated the biochemical characteristics of the E1 enzyme, its temporal expression in planta, and its response to leaf dehydration. Furthermore, we have determined whether E1 expression and chloroplast localization affects photosynthetic rates. This has allowed a complete analysis of heterologous expression of hydrolytic E1 enzyme in plants.
Materials and methods Bacterial strains, plant material, plant transformation, and plant growth conditions Escherichia coli strains MC1000 and JM83 (ara, leu, lac, gal, str) were used as the hosts for routine cloning experiments. Agrobacterium tumefaciens LBA-4404 containing the Ach5 chromosomal background and a disarmed helper-Ti plasmid pAL4404 (Hoekema et al., 1983) was used for transformation of tobacco plants (Nicotiana tobacum L. cv petit Havana SR1). Transgenic plants were obtained by the co-cultivation method (An et al., 1988) using tobacco leaf discs grown aseptically on Murashige and Skoog (MS) medium supplemented with 3% sucrose and appropriate levels of plant growth regulators (Murashige & Skoog, 1962). Solutions containing 2 mg l−1 α-naphthaleneacetic acid and 0.5 mg l−1 6-benzylaminopurine were used for callus induction and 0.5 mg l−1 6-benzylaminopurine for shoot induction. Kanamycin-resistant transformants were selected and grown on MS medium in a growth room, and seeds were collected after self-fertilization. Seeds were germinated on Murashige and Skoog agar medium containing 50 mg l−1 kanamycin, and healthy kanamycin-resistant plants were grown in a growth room under a 14 h light (25–28◦C, 60% relative humidity)/10 h dark (22◦C, 70% relative humidity) cycle.
45 Irradiance, provided by six high-pressure metal halide lamps (Philips, USA), was 350–500 µmol quanta m−2 s−1 at the plant canopy. Recombinant DNA techniques Standard procedures were used for recombinant DNA manipulation (Sambrook et al. 1989). The plasmid pPMT4-5 containing a genome clone of the endoglucanase E1 gene isolated from A. cellulolyticus genome library (GenBank Accession No. U33212) was obtained from the National Energy Renewable Laboratory (Golden, Colorado). The 1566-bp fragment (containing the mature-peptide coding region) was isolated from pPMT4-5 by PCR using the primers E1-A3 (50 -CTA ATG CAT GCG GGC GGC GGC TAT TGG CAC ACG AGC-30 ) and E1-B (50 -CTT AGA TCT GAG CTC TTA ACT TGC TGC GCA GGC GAC TGT CGG TG-30 ). The PCR amplification was carried out in a Robocycler gradient 96 Temperature cycler with hot top assembly (Stratagene, La Jolla, CA). In order to fuse the mature E1 cDNA fragment aligned with the RbcS-2A transit peptide, a Nsi I (ATGCAT) restriction enzyme site was introduced at the 50 -end of mature E1 cDNA fragment through PCR using E1-A3 primer. In addition, two restriction enzyme sites (Bg1 II and Sac I) were introduced at the end of C-terminator of mature E1 cDNA via PCR using the primer E1-B for further cloning. The Xba I/Bgl II fragment, containing RbcS-2A transit peptide and mature E1 coding sequence, was in-frame fused downstream of the RbcS-3C promoter, carrying alfalfa mosaic virus (AMV) leader, and upstream of the terminator of T5 and T7 genes of octopine type Ti plasmid to form pZD276 (Figure 1). The binary vector for pZD276 was derived from pGA628 (An, 1995). The Ti plasmid pZD276 was used in Agrobacterium-mediated leaf disc transformation. DNA isolation and PCR analysis The DNA isolated from leaves of transgenic tobacco and the PCR analysis for E1 gene insertion was described previously (Dai et al., 1999). RNA preparation and gel blot anaylsis Mature leaf tissues were harvested and immediately frozen in liquid nitrogen. RNA was extracted from frozen leaf tissues, which were homogenized using a mortar and pestle. Five milliliter per gram GTC buffer (4 M guanidine isothiocyanate, 25 mM sodium citrate,
Figure 1. Schematic representation of binary vector pZD276 used for tobacco transformation. The pZD276 contains right border (BR), left border (BL), neomycin phosphotransferase II gene (npt), Rubisco small subunit promoter (RbcS-3C), RNA4 leader sequence of alfalfa mosaic virus (AMV-UTL), Rubisco small subunit RbcS-2A transit peptide (SP), mature E1 coding sequence (mE1), T7-T5 terminators, and tetracycline resistant gene (tet).
0.5% sodium N-lauroylsarcosinate (w/v), and 0.1 M β-mercaptoethanol) and 0.5 ml g−1 2 M sodium acetate, pH 4.0 were sequentially added to ground tissues and mixed well. The preparation was further extracted in an equal volume of phenol:chloroform:isoamyl alcohol (25:24:1) and then centrifuged. The aqueous phase containing RNA was transferred to new centrifuge tubes and precipitated with an equal volume of isopropanol. The pellets were suspended in water and further precipitated with an equal volume of 4 M lithium chloride. The RNA pellets were resuspended in RNase-free H2 O, extracted with chloroform and precipitated with 0.2-M sodium acetate, pH 5 and isopropanol. The total RNA concentration was quantified spectrophotometrically. Twenty micrograms of total RNA samples were used for RNA gel-blot analysis as described by Dai and An (1995). A 1.2-kb XbaI/BamH I internal E1 cDNA fragment was labeled with the α-[32P]-dCTP as a probe. Enzyme extraction and assays The third or fourth leaf from the shoot apex was used for enzyme extraction. Leaf samples were harvested at 2–3 h into the light period. Leaf tissues were sectioned into approximately 1 cm2 leaf discs and pooled together. About 0.1 g of leaf discs was collected and placed in ice cold 1.5 ml microcentrifuge tubes and kept on ice. E1 enzyme in leaf tissues was extracted with a pellet pestle (Kontes Glass Co, Vineland, NJ) and 5 volumes (v/w) of ice cold medium containing 80 mM 2-(N-morpholino) ethane sulfonic acid MES;
46 pH 5.5), 10 mM β-mercaptoethanol, 10 mM EDTA, 0.1% sodium N-lauroylsarcosinate (w/v), 0.1% Triton X-100 (v/v), 1 mM PMSF, 10 µM leupeptin, and 1 µg ml−1 other protease inhibitors (aprotinin, pepstin A, chymostatin). The crude extract was centrifuged at 15,000 g for 10 min at 4◦ C. The supernatant was used for protein determination, enzymatic analysis, polyacrylamide gel (PAGE) electrophoresis, and western blot analysis. E1 enzyme activity was measured at 55◦ C and pH 5.5 in a final volume of 1 ml. The reaction mixture for E1 activity analysis contained 80 mM MES, pH 5.5, 1 mM EDTA, 1 mM DTT, and 2–10 µl leaf extract. The enzymatic reaction was initiated by adding the substrate 4-methylumbelliferyl-β-D-cellobioside (MUC) into the reaction mixture to a final concentration of 2 mM. The reaction mixture was removed at 15, 30, and 45 min intervals and put into 1.9 ml 0.2 M Na2 CO3 buffer to terminate the reaction. The fluorescent moiety 4-methylumbelliferon (MU) released from 4-MUC by E1 enzyme has a peak excitation of 365 nm (UV) and a peak emission of 455 nm (blue). Emission of fluorescence from the mixture was measured with a Hoefer DyNA Quant 200 fluorometer. Enzyme activities were expressed on a gram leaf weight basis or on a total leaf soluble protein basis. SDS–PAGE and western blot Soluble protein was extracted from 0.1 g leaf tissue in 0.5 ml cold extraction buffer as described above in the ‘Enzyme extraction and assays’ section. The extracts were clarified by centrifugation (20,000 g for 10 min at 4◦ C). Electrophoretic analysis of polypeptides was performed in a 7.5–15% (w/v) full size linear gradient PAGE containing 0.1% SDS, stabilized by a 5–17% (w/v) linear sucrose gradient (Chu et al., 1990) or a 4–20% (w/v) mini precast gel from Bio-Rad laboratories (Hercules, CA). Protein samples were prepared for electrophoresis by mixing four volumes of protein extract with one volume of protein loading buffer (250 mM Tris–HCl, pH 7.5, 10% SDS, 28% glycerol, 38% β-mercaptoethanol, and a trace amount bromphenol blue) and then heated at 90◦ C for 3– 5 min. The samples were centrifuged at 10,000 g for 2 min. E1 protein samples were separated by electrophoresis and then electrophoretically transferred onto nitrocellulose membranes (BA-S85; Schleicher & Schuell, Keene, NH) (Dai et al., 1994). The protein was reacted with affinity-purified mouse monoclonal antibody against full length or catalytic domain of E1
protein (in 1:250 dilution). The antibody was detected by a color development kit (BIO-RAD, Hercules, CA) and a goat anti-mouse secondary antibody (IgG) conjugated with alkaline phosphatase (Pierce, Rockford, IL). Color development was performed in the dark for 30 min to 4 h. The E1 protein used as a positive control in these experiments was purified from culture supernatants of Streptomyces lividans carrying a plasmid containing the A. cellulolyticus E1 gene on a 3.7-kb genomic fragment, and was generously provided by W. Adney (NREL, Golden, CO). The concentration of soluble protein was determined by the method of Bradford (1976) with BSA as the standard. The amount of E1 expressed in leaf tissues was estimated by densitometry analysis. Protein immunoblot bands were scanned with a Hewlett Packard ScanJet 6100C Scanner (Hewlett Packard Inc., Palo Alto, CA). The imaging data were analyzed with the DENDRON 2.2 program (Solltech Inc Oakdale, IA). A series of known amounts of E1 protein from S. lividans was used as the standard for estimating levels of E1 protein in transgenic plants. Immunocytochemistry Young leaf tissues of transgenic and normal tobacco plants were cut into about 1 mm pieces with a razor blade and immediately placed into a fixative containing 1.25% glutaraldehyde, 3% (w/v) paraformaldehyde, and 50 mM piperazine-N-N 0 bis-2-ethane sulfonic acid (PIPES), pH 7.2. The tissues were fixed overnight in the fixative at 4◦ C. The samples were then washed with 50 mM PIPES buffer (pH 7.2), dehydrated in an ethanol series, and gradually infiltrated in LR white resin (Electron Microscopy Sciences, Fort Washington, PA, USA), ending with three changes of pure resin. Samples were polymerized for 15 h at 60◦ C and sectioned 100 nm thick on a Leica Ultracut microtome (Leica Microsystems Inc., Deerfield, IL, USA). Ultrathin serial sections from leaf tissues collected on nickel grids were blocked in 1% (w/v) BSA in PBS. The sections were then incubated in a drop of mouse anti-E1 monoclonal antibody diluted 1:10 in blocking buffer, for 2 h at room temperature. The sections were washed four times for 10 min in PBS. The sections were further incubated with goat anti-mouse IgG conjugated to 10 nm of gold (Sigma, St. Louis, MO, USA) which was diluted 1:50 in blocking buffer for 2 h at room temperature. The sections were washed twice for 5 min in blocking buffer and twice for 5 min
47 in PBS. Sections were stained with potassium permanganate and uranyl acetate and subsequently evaluated and photographed using a Hitachi transmission electron microscope (Hitachi Instruments, Inc., San Jose, CA).
amount of total soluble protein was determined as described above. Enzyme activity was calculated on the basis of total soluble protein in each sample or on the basis of the original gram leaf weight determined just before dehydration treatment.
Effect of temperature and pH on E1 activity
A series of temperatures were generated by a RoboCycler 96 gradient temperature cycler with the hot top assembly. The reaction mixture was the same as above. E1 extract from transgenic plants was mixed well with the reaction mixture and 100 µl of the resulting mixture was transferred into PCR reaction tubes (250 µl) just before incubating at the selected temperature. The RoboCycler 96 gradient temperature cycler can generate 11◦ C gradient temperatures between 47 and 98◦ C with an error of ± 0.1◦ C. Four sets of gradient temperatures generated by the RoboCycler 96 gradient temperature cycler were 47–58◦C, 61–72◦C, 75–86◦C, and 86–97◦C. Three replicate samples were taken from three different transgenic plants for each temperature point. The PCR microtubes were incubated at the specific temperature for 15 min and immediately quenched by immersing in ice cold water. In addition, 100 µl of 0.2 M Na2 CO3 was added and mixed by vortexing to stop the reaction completely. A series of reaction mixtures was prepared using different pH buffers (pH 3.5, 4.0, 4.5, 4.75, 5.0, 5.25, 5.5, 6.0, 6.5, 7.0, 7.5, and 8.0). Buffers below 6.0, between 6 and 7, and above 7 were phosphatecitrate buffer, phosphate buffer, and Tris–HCl buffer, respectively. The reaction mixture contained 80 mM buffer (listed above), 1 mM EDTA, 1 mM DTT, and 5 µl enzyme extract and 2 mM 4-methylumbellferalβ-D-cellobioside (MUC). All reaction mixtures were kept on ice cold H2 O until the reaction was initiated at 55◦C. One hundred microlitre of reaction mixture was removed at 10, 20, 30 min and quenched by mixing with 1.9 ml 0.2 M Na2 CO3 . Effect of leaf dehydration at room temperature The 4th–6th leaf counted from the top of the plant was harvested and cut in approximately 1 cm2 leaf discs. Roughly 5–10 leaf discs per sample were allowed to dehydrate at ambient conditions up to 100 h. The samples were harvested at different intervals and stored at −70◦C. Then all samples were extracted with enzyme extract buffer and 5 µl extract per sample was used for E1 enzyme activity measurement. The
Rates of CO2 assimilation were measured on the 3rd or 4th intact leaf from the apex using an Analytical Development Co. (ADC) infrared gas analyzer (225MK3; PP System, Haverhill, Mass., USA) and a Bingham interspace (Hyde Park, Utah, USA) model BI-6-dp computer controller system (Dai et al., 1992). This was operated as an open system where a given partial pressure of CO2 is passed through the sample cells (in line with the leaf enclosed in a cuvette) and the reference cell. The rate of CO2 removal by photosynthesis was compensated for by a controlled rate of injection of CO2 from a high CO2 source. The leaf cuvette contained a dewpoint sensor for measuring humidity and a copper-constantan thermocouple for monitoring leaf temperature. The rate of CO2 assimilation was directly calculated from gas-exchange measurements according to von Caemmerer and Farquhar (1981).
Results and discussion Endoglucanase (E1) expression analysis A series of transgene expression vectors carrying E1 coding sequence were constructed for different compartmentalization in tobacco cells and tissues. In this study, the transgenic plants with E1 protein targeted to chloroplasts were selected for biochemical characterization and physiological effect analyses. The binary vector pZD276 was used for plant transformation (Figure 1). The pZD276 includes the entire mature E1 coding sequence where it was under the control of the RbcS-3C promoter (Sugita et al., 1987) and transcriptional T7-T5 terminator of T5 and T7 genes of the octopine type Ti plasmid. More than 30 independent transformants were selected on kanamycin, and resistant shoots were regenerated into plants and transferred to soil. Transgenic plants were grown for 3–4 weeks in a growth room, and leaf tissues were subsequently harvested for genomic DNA PCR, RNA blot, E1 activity analysis, and protein immunoblotting. All transformants were first screened by PCR with primers corresponding to the E1 cDNA sequence
Figure 2. Expression of E1 transgene in selected transgenic tobacco lines 1319-2-1, 1319-2-2, 1319-2-3. RNA blots containing (20 µg per lane) total RNA isolated from selected transgenic lines and SR1 wild-type control plant are shown. The RNA gel blots hybridized with a radioactive α-[32 P]-dCTP labeled probe prepared from a 1.2 kb XbaI/BamH I internal E1 coding region.
(data not shown). The transcription level of E1 gene in transgenic lines 1319-2-1, 1319-2-2, and 1319-2-3 was determined by RNA blot. Results in Figure 2 show that the selected transgenic lines properly transcribed the integrated E1 gene and produced different levels of transcription. E1 activity for selected T1 transgenic plants 1319-2-1, 1319-2-2, 1319-2-3 was 2756.3, 1022.6, and 1559.1 pmol MU mg−1 total soluble protein min−1 , respectively (Figure 3A). Western blotting (Figure 3B) with monoclonal antibodies raised against A. cellulolyticus E1 demonstrated that the observed E1 activity was caused by expression of the heterologous E1 enzyme in leaf tissues. Control SR1 extract shows no antibody staining material and little or no activity on 4-MUC. Western blots also show that the E1 protein extracted from transgenic plants is similar in molecular weight to E1 protein purified from a S. lividans expression system, suggesting that the 89 amino acids (including an addition of 22 amino acids of mature sequence of RbcS-2A and eight amino acids for the repeat of transit peptide cleavage site) of Rubisco transit peptide had been removed to form mature E1 protein. The N-terminal sequence of the S. lividans E1 protein is identical to the native A. cellulolyticus protein (S. Thomas, unpublished results). As the transgenic E1 protein has not yet been purified from leaf extracts, we have no data regarding its N-terminal sequence. Interestingly, part of E1 protein was partially degraded to the size in molecular weight larger than the catalytic domain of E1 protein. Further study will be necessary to define the E1 protein degradation pattern in plant tissues.
Figure 3. (A), Measurements of E1 enzyme activity in transgenic lines 1319-2-1, 1319-2-2, 1319-2-3 and control plant SR1 by the MUC assay using 4-methylumbelliferyl-β-D-cellobioside (MUC) as the substrate. Each bar represents the mean ± one standard deviation of 3–4 replicates. (B), Protein gel immunoblot analysis of E1 protein extracted from transgenic lines listed in (A). The amount of protein loaded on the gel was determined before treatment. Lanes 1–3 contained 200, 100 and 200 ng affinity-purified E1 protein from bacterial cultures. Lane 4 was empty. Lane 5 contained 50 µg total soluble leaf protein from wild type SR1 control tobacco plants. Lanes 6–8 contained 30 µg per lane total soluble leaf protein from transgenic tobacco plants.
The expressed level of E1 protein in selected transgenic plants was estimated by gel electrophoresis and western blot analyses. A series of diluted S. lividans E1 proteins was used as a standard. The gelblot densities of known S. lividans E1 proteins and E1 proteins from leaves of selected transgenic plants were estimated using DENDRON 2.2 program. Based on the densitometric determination, the amounts of E1 protein accumulated in selected transgenic plants 1319-2-1, 1319-2-2, and 1319-2-3 were estimated to be 1.35%, 0.18%, and 0.69% of the total leaf soluble protein, respectively. Immunocytochemistry The transgenic line 1319-2-1 was used for E1 immunolocalization to determine E1 localization un-
Figure 4. Electron micrographs showing immunolocalization of E1 in transgenic line 1319-2-1 leaves. Affinity-purified Anti-E1 monoclonal antibody was used for immunolocalization. Immunogold particles reside in the stroma of chloroplasts A, X10,000; B, X25,000. vc, vacuole; ch, chloroplast.
der the control of the transit peptide of the RbcS2A protein (Pichersky et al., 1986; Sugita et al., 1987). Figure 4 shows immunogold particle labeling in chloroplasts indicating proper targeting of the E1 protein. Wild type SR-1 plants have no obvious immunogold labeling in chloroplasts, only some low level of non-specific background signal. Measurements of fluorescent moiety 4-methylumbelliferon (MU) released from 4-MUC by E1 enzyme released from chloroplasts of transgenic line 1319-2-1 isolated by sucrose gradient ultracentrifugation also confirmed that the E1 protein indeed located in chloroplasts of transgenic line 1319-2-1 (data not shown). These results show that the transit peptide of RbcS-2A protein functions properly to secrete E1 enzyme into chloroplasts. The biochemical properties of recombinant E1 enzyme in plant extracts Since we did not have T1 material available for further characterization, three T2 transgenic plants (1319-23-6, 1319-2-3-9, 1319-2-3-20) grown from T1 seeds of transgenic line 1319-2-3 were used to determine biochemical characteristics of recombinant E1 derived from plant hosts. The pH response of E1 enzyme from leaf extracts was first measured at 55◦C at various pH values (Figure 5). Figure 5A shows the pH response curves of the three transgenic lines. As pH increased from 3.5 to 5.2, E1 enzyme activity gradually increased to an optimum. Thereafter, the E1 enzyme activity dramatically decreased when pH was
further increased from 5.2 to 6.5. The three transgenic lines exhibited similar pH responses though line 1319-2-3-6 differed in scale. Taking E1 activity at pH 5.2 as 100%, normalized E1 enzyme activity at different pH values was calculated (Figure 5B). When the pH in the reaction mixture was changed from 5.2 to 3.5, the E1 enzyme activity decreased only about 40%. However, when the pH increased from 5.2 to 7, the E1 activity dropped almost 90%. At pH 8.0 the E1 enzyme lost almost 99% of its activity compared to pH 5.2. The optimum pH for E1 enzyme in plant extracts is similar to the optimum growth condition of A. cellulolyticus (Tucker, 1989). For comparison, the pH response of E1 enzyme purified from bacterial S. lividans expression system was also determined (Figure 5C&D). Figure 5C shows the pH response curves of the two replications with two different amounts (10 and 20 ng) of E1 protein. Similar to the response of E1 enzymes extracted from plant tissues, the E1 enzyme activity gradually increased to optimum pH 5.2 when pH increased from 3.0 to 5.2. As the pH further increase from 5.2 to 7.0, the E1 activity significantly decreased. The normalized E1 activity at different pH to the pH 5.2 was also calculated (Figure 5D) and had similar response patterns as E1 enzyme extracted from plant tissues did (Figure 5B). These results demonstrated that the E1 protein expressed in transgenic plants retains its biochemical property as it was expressed in bacterial system. The temperature response curve of E1 enzyme in plant extract was determined with the same three transgenic lines at pH 5.5. Figure 6A shows E1 en-
Figure 5. The response of E1 enzyme from transgenic plants and bacterial S. lividans to varying pH at 55◦ C. (A) and (C), E1 enzyme activity was determined by the MUC assay using 4-methylumbelliferyl-β-D-cellobioside (MUC) as the substrate at different pH for T2 transgenic lines 1319-2-3-6, 1319-2-3-9, 1319-2-3-20 and two amounts of E1 purified from S. lividans. (B) and (D), results are shown as a percentage of E1 activity at optimum pH (5.2). Each point in (B) is the mean of measurements from the three transgenic lines ± one SD and in (C) is the mean of measurements from two sets (5 and 10 ng) of two replicates of E1 protein from S. lividans.
zyme activity at different temperatures and its optimal temperature (81◦C) and Figure 6B is E1 enzyme activity normalized to a value of 100% at 81◦ C. With increasing reaction temperature from 0◦ C to 50◦ C, E1 enzyme activity increased only up to 26% of the maximum (about 0.52% per increase of 1◦ C). Thereafter, with increasing temperature from 50◦ C to 81◦ C, E1 activity dramatically increased from 26% to 100% (about 2.38% per increase of 1◦ C). As temperature increased further from 82◦ C to 96◦ C, E1 activity dramatically decreased to only 5% of the maximum value. Figure 6B shows that, at 30◦C, the E1 activity is less than 10% of the maximum value. The temperature response and its optimal temperature of E1 enzyme extracted from transgenic plants is also similar to that was determined before using E1 protein purified from the unfractionated supernatant of A. cellulolyticus (Tucker et al., 1989). Effect of leaf age on E1 enzyme activity The effect of leaf age on E1 enzyme activity and stability was determined in transgenic line 1319-2-1. Leaf tissues used for E1 enzyme activity measurement were harvested from upper leaves (leaf 3 and 4 counted from top to bottom), middle leaves (leaf
10 and 11), and lower leaves (leaf 15 to leaf 16). E1 enzyme activity, when calculated on a total soluble protein basis, was similar between upper and middle leaves, but much higher in the lower leaf tissues (Figure 7A). This increase of E1 enzyme activity in older leaf tissues is mainly due to the decrease of total protein under degradation conditions in senescing leaves. E1 was apparently more stable than the majority of other proteins resident in plant tissues. However, E1 enzyme activity expressed on a gram leaf weight basis remained relatively constant between different developmental stages. The degree of accumulation and stability of E1 protein was observed in protein immunoblots (Figure 7B). Figure 7B shows that E1 protein purified from bacterial S. lividans possessed two polypeptides. The upper polypeptide was the intact E1 protein, while the lower polypeptide was catalytic domain evidently resulting from E1 protein proteolysis. Compared with E1 protein purified from bacterial S. lividans, upper leaf tissues of transgenic plants retained more intact E1 protein than that of middle or lower leaf tissues. In lower leaf tissues, most of E1 protein was partially degraded to a polypeptide in molecular weight larger than the E1 catalytic domain as we observed in Figure 3B. These results indicate that the catalytic domain of E1 protein
Figure 6. The response of E1 enzyme extracted from three transgenic plants to varying reaction temperature in MES buffer, pH 5.5. (A), E1 activity was determined by MUC assay using 4-methylumbelliferyl-β-D-cellobioside as the substrate at different reaction temperatures for T2 transgenic lines 1319-2-3-6, 1319-2-3-9, 1319-2-3-20. (B), the results are shown as a percentage of the E1 activity at optimum temperature (81◦ C). Each point in B is the mean of measurements from the three transgenic lines ± one SD.
in transgenic plants is resistant to proteolytic degradation in planta. It suggests that all plant tissues can be used for E1 protein production. Furthermore, other cellulase enzymes may be introduced into the same transgenic plant with different compartmentalization. All of these possible considerations will enhance the production of lignocellulosic hydrolytic enzymes and reduce their production costs. The effects of leaf dehydration on the E1 activity E1 enzyme stability and activity in leaf discs were determined at different stages of leaf dehydration (Figure 8). As shown in Figure 8A, E1 enzyme activity in leaf discs dramatically increased after 40 h of dehydration when expressed on a total soluble protein basis (pmol MU per mg total soluble protein per minute). When leaf discs were further dehydrated up to 100 h, E1 enzyme activity increased only slightly. When expressed on an initial gram leaf weight basis (pmol MU
Figure 7. (A), Measurements of E1 enzyme activity in upper, middle, and lower leaves of transgenic line 1319-2-1 by the MUC assay using 4-methylumbelliferyl-β-D-cellobioside as the substrate. Each bar represents the mean ± one standard deviation of 3–4 replicates. Open bars represent E1 enzyme activity determined on the basis of total soluble protein. Filled bars represent E1 enzyme activity determined on the basis of gram fresh leaf weight. (B), Protein gel immunoblot analysis of E1 protein extracted from different leaf tissues listed in A. The amount of protein loaded on the gel was determined before treatment. Lane 1 contained 400 ng affinity-purified E1 protein from bacterial cultures. Lane 2 was empty. Lane 3 contained 50 µg of total soluble leaf protein from wild type control tobacco plants. Lanes 4–6 contained 20 µg per lane total soluble leaf protein from upper, middle, and lower leaf tissues of transgenic tobacco 1319-2-1.
0.1 g−1 min−1 ), E1 enzyme activity remained fairly constant during the entire 100 h dehydration period. Changes in E1 protein size in dehydrated leaf discs are shown in Figure 8B. After 5 h, E1 protein in leaf tissues was slightly degraded. As leaf discs were further dehydrated through 39 h, levels of E1 protein degradation products gradually increased. After 39 h dehydration at ambient conditions, most of the leaf tissues had completely dried. At these conditions most endogenous proteolytic enzymes may not be active. Therefore, the levels of E1 protein degradation products remained similar through 100 h. The pattern of E1 protein degradation is similar to the E1 activity expressed on a total soluble protein basis (Figure 8A). All fragments appearing in Figure 8B are larger than or equal in size to the E1 catalytic domain standard (purified from bacteria expressing E1 catalytic domain only, third lane from left, E1-cat, Figure 8B). Similar
52 Photosynthetic rates in transgenic plant line 1319-2-1 Since the E1 enzyme was targeted to chloroplasts, CO2 fixation in transgenic leaves may be affected. Photosynthetic rates were measured in mature 3rd or 4th leaf of both transgenic line 1319-2-1 and wild type control SR1 tobacco plants at 30◦ C leaf temperature and ambient CO2 concentration. The photosynthetic rate of transgenic line 1319-2-1 was 23.8 ± 0.5 µmol m−2 s−1 , which was very similar to that of wild type SR1 (22.8 ± 0.35 µmol m−2 s−1 ). The results indicate that E1 protein accumulation in chloroplasts does not affect the photosynthetic function of the plant. Transgenic plants were not abnormal in morphology, growth, and seed setting as compared with SR1 wild type control tobacco plants. It has also been shown that expression of a thermostable xylanase from Claostridium thermocellum expressed at high levels in the apoplast of transgenic tobacco did not affect the growth of transgenic plants (Herbers, 1995).
Figure 8. (A), Measurements of E1 enzyme activity of leaf discs of transgenic line 1319-2-1 dehydrated at room temperature for different time periods (0–100 h) by the MUC assay using 4-methylumbelliferyl-β-D-cellobioside as the substrate. Each data point represents the mean ± one standard deviation of 3–4 replicates. Solid circles represent E1 enzyme activity determined on the basis of total soluble protein. Filled diamonds represent E1 enzyme activity determined on the basis of gram fresh leaf weight. (B), Protein gel immunoblot analysis of E1 protein extracted from leaf discs dehydrated at room temperature for different time periods. The amount of protein loaded on the gel was determined before treatment. Lane 1 and 2 contained 400 ng and 200 ng E1 intact protein, respectively, and lane 3 contained 760 ng E1 catalytic domain protein, which were isolated by affinity-purification of bacterial cultures. Lane 4 contained 50 µg total soluble leaf protein from a wild type SR1 control tobacco plant. Lanes 5–18 contained 20 µg per lane total soluble leaf protein extracted from leaf discs of transgenic tobacco 1319-2-1 dehydrated at room temperature at different time periods. The lanes labeled as 1 and 2 are two repeats at each time point.
to earlier observation in the activity versus leaf age tests, levels of E1 enzyme activity did not decrease with dehydration time. The results further demonstrate that E1 protein has high degree of tolerance to endogenous proteolytic enzymes in dehydrated leaf discs. The stability characteristics of E1 protein in dehydrated plant tissues may be considered for plant tissue harvest and storage, E1 protein purification, and even the conversion of lignocellulosic biomass wastes to fuel ethanol by mixing powders of dried transgenic plant tissues with biomass wastes combined with other cellulases.
Conclusions In this study we showed that the E1 enzyme of A. cellulolyticus accumulated up to 1.35% of the total soluble protein in leaf tissues of transgenic tobacco plants. E1 protein was properly translocated to chloroplasts, as shown by immunocytochemistry, via the transit peptide of Rubisco small subunit (RbcS-2A). E1 enzyme expressed in transgenic plants retained its native biochemical characteristics (for properties of the native enzyme see Tucker et al., 1989). In addition, E1 was very active in leaf tissues over a range of varying ages as well as in dehydrating leaf discs. In older leaves and dehydrated leaf tissues, portions of the E1 protein were degraded, leaving only the catalytic domain or the catalytic domain and a part of the cellulose binding domain. It is suggested that the E1 catalytic domain is resistant to proteolytical degradation in plant tissue. Thus, these properties of E1 enzyme in plant tissues will simplify plant tissue harvest and storage, and E1 protein purification processes for biofuel ethanol production that will further reduce additional cost of E1 protein production. Meanwhile, the photosynthetic rate, plant growth, development, and reproduction of E1 transgenic plants were similar to untransformed control plants. The high stability of E1 enzyme and suboptimal conditions for E1 enzyme activity in chloroplasts (pH 7–8, relatively low leaf temperature, and lack of cellulosic substrate) which
53 allowed plant growth comparable to normal tobacco suggests that transgenic plants are viable alternative host for E1 enzyme production. Acknowledgements We thank Mr. Dianzhong Zhang and Dr. Vincent R. Franceschi of Botany Department, Washington State University for help with immunocytochemistry experiments; the staff of the Electron Microscope Center of Washington State University for use of the electron microscopy and photographic facilities; Dr. Gerald E. Edwards of Botany Department, Washington State University for critically reading the manuscript and allowing us to use the photosynthetic measurement system in his lab; Mr. Ryan D. Quesenberry for maintaining transgenic tobacco plants in the growth room; and Dr. Mark T. Kingsley of Pacific Northwest National Laboratory for providing assistance in protein quantification. We also thank Dr. Wilhelm Gruissem, University of California for providing the two E. coli strains that contain the RbcS-3C promoter (pRbcS-3C-101-1) and RbcS-2A signal peptide coding region (PTss-1-91 [#2]-IBI). We thank Drs. Mike Himmel, Bill Adney and Rafael Nieves for generously providing purified E1 and E1-cat enzymes and monoclonal antibody to E1. This work was funded by the Biochemical Conversion Element of the Biofuels Program of the U.S. Department of Energy. Pacific Northwest National Laboratory is operated by the Battelle Memorial Institute for the U.S. Department of Energy under the contract DE-AC0676RL01830.
References Adney WS, Thomas SR, Himmel ME, Baker JO, Chou Y-C (1998) Method for increasing thermostability in cellulase enzymes. U.S. Patent 5,712,142, Jan 29. An G, Ebert PR, Ha S-B (1988) Binary vectors. In: Gelvin SB and Schilperoort RA (eds). Plant Molecular Biology Mannual. (pp. 1–19) Kluwer Academic Publishers, Dordrecht, The Netherlands. An G (1995) Binary Ti plasmid vectors In: Gartland KMA and Davey MR (eds). Methods in Molecular Biology. Vol 44 (pp. 47–58): Agrobacterium protocols. Humana Press Inc., Totowa, NJ. Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72: 248–254. Chu C, Dai Z, Ku MSB, Edwards GE (1990) Induction of Crassulacean acid metabolism in the facultative halophyte Mesembryanthemum crystallinum by abscisic acid. Plant Physiol 93: 1253–1260.
Dai Z, Edwards GE, Ku MSB (1992) Control of photosynthesis and stomatal conductance in Ricinus communis L. (castor bean) by leaf to air vapor pressure deficit. Plant Physiol 99: 1426–1434. Dai Z, Ku MSB, Zhang D, Edwards GE (1994), Effects of growth regulators on the induction of crassulacean acid metabolism in the facultative halophyte Mesembryanthemum crystallinum L. Planta 192: 287–294. Dai Z, An G (1995) Induction of nopaline synthase promoter activity by H2 O2 has no direct correlation with salicylic acid. Plant Physiol 109: 1191–1197. Dai Z, Hooker BS, Quesenberry RD, Gao J (1999) Expression of Trichoderma reesei exo-cellobiohydrolase I in transgenic tobacco leaves and calli. Appl Biochem Biotech 77–79: 689–699. Duff SJ, Murray WD (1996) Bioconversion of forest products industry waste cellulosics to fuel ethanol: a review. Bioresour Tech 55: 1–33. Gilbert HJ, Hazlewood GP (1993) Bacterial cellulase and xylanases. J Gen Microbiol 139: 187–194. Herbers K, Wilke I, and Sonnewald U (1995) A thermostable xylanase from Clostridium thermocellum expressed at high levels in the apoplast of transgenic tobacco has no detrimental effects and is easily purified. Bio/Technol 13: 63–66. Hoekema A, Hirsch PR, Hooykaas PJJ, Schilperoort RA (1983) A binary vector strategy based on separation vir- and T-region of the Agrobacterium tumefaciens Ti-plasmid. Nature 303: 179– 181. Jensen LG, Olsen O, Kops O, Wolf N, Thomsen KK, Wettstein DV (1996) Transgenic barley expressing a protein-engineering, thermostable (1,3-1,4)-β-glucanase during germination. Proc Natl Acad Sci USA 93: 3487–3491. Lastick (deceased), Stanley M, Tucker, Melvin P, Grohmann, Karel (1996) Cloning of cellulase genes from Acidothermus cellulolyticus. US5514584, May 7. Liu J-H, Selinger B, Cheng K-J, Beauchemin KA, Moloney MM (1997) Plant seed oil-bodies as an immobilization matrix for a recombinant xylanase from the rumen fungus Neocallimastix patriciarum. Mol Breed 3: 463–470. Lynd LR, Cushman JH, Nichols RJ, Wyman CE (1991) Fuel ethanol from cellulosic biomass. Science 251: 1318–1322. Mohagheghi A, Grohmann K, Himmel M, Leighton L, Updegraff D. (1986) Isolation and characterization of a new genus of thermophilic acidophilic cellulolytic bacteria, Acidothermus cellulolyticus, gne. nov., sp. nov. Int J Sys Bacteriol 36: 435–443. Murashige T, Skoog F (1962) A revised medium for rapid growth and bioassays with tobacco tissue cultures. Physiol Plant 15: 473–497. Pichersky E, Bernatzky R, Cashmore AR, Gruissem W (1986) Evidence for selection as a mechanism in the concerted evolution of Lycopersicon esculentum (tomato) genes encoding the small subunit of ribulose-1,5-bisphosphate carboxylase/oxygenase. Proc Natl Acad Sci USA 82: 3880–3884. Sakon J, Adney WS, Himmel ME, Thomas SR, Karplus A (1996) Crystal structure of thermostable family 5 endocellulase E1 from Acidothermus cellulolyticus in complex with cellotetraose. Biochem 35: 10648–10660. Sambrook J, Fritsch EF, Maniatis T (1989) Molecular Cloning: A Laboratory Manual, 2nd edn, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Sugita M, Manzara T, Pichersky E, Cashmore A, Gruissem W (1987) Genomic organization, sequence analysis and expression of all five genes encoding the small subunit of ribulose1,5-bisphosphate carboxylase/oxygenase from tomato. Mol Gen Genet 209: 247–256.
54 Thomas SR, Laymon RA, Himmel ME (1996) Gene coding for the E1 endoglucanase. US5536655, July 19. Tucker MP, Mohagheghi A, Grohman K, Himmel ME (1989) Ultra-thermostable cellulases from Acidothermus cellulolyticus: Comparison of temperature optima with previously reported cellulases. Bio/Technol 7: 817–820. Von Caemmerer SV, Farquhar GD (1981) Some relationships between the biochemistry of photosynthesis and the gas exchange of leaves. Planta 153: 376–387.
Verwoerd TC, van Paridon PA, van Ooyen AJJ, van Lent JWM, Hoekema A, Pen J (1994) Stable accumulation of Aspergillus niger phytase in transgenic tobacco leaves. Plant Physiol 109: 1199–1203. Ziegelhoffer T, Will J, Austin-Phillips S (1999) Expression of bacterial cellulase genes in transgenic alfalfa (Medicago sativa L.), potato (Solanum tuberosum L.) and tobacco (Nicotiana tabacum L.). Mol Breed 5: 309–318.