Ethanol inducible expression of a mesophilic cellulase avoids adverse effects on plant development

Abstract

Background: Plant-produced biomass-degrading enzymes are promising tools for the processing of lignocellulose to fermentable sugars. A major limitation of in planta production is that high-level expression of such enzymes could potentially affect the structure and integrity of the plant cell wall and negatively influence plant growth and development.

Results: Here, we evaluate the impact on tobacco plant development of constitutive versus alcohol-inducible expression of the endoglucanase TrCel5A from the mesophilic fungus Trichoderma reesei. Using this system, we are able to demonstrate that constitutive expression of the enzyme, controlled by the doubled Cauliflower Mosaic Virus promoter, leads to lower cellulose content of the plant combined with severe effects on plant growth. However, using an alcohol-inducible expression of the endoglucanase in the plant leaves, we achieved similar enzymatic expression levels with no changes in the crystalline cellulose content.

Conclusion: We were able to produce significant amounts of cellulase in the plant leaves without detrimental effects to plant development. These results demonstrate the potential feasibility of an inducible expression system for producing biomass degrading enzymes in plants.

Figure 1

Schematic presentation of the plant expression cassettes for constitutive (A) and inducible (B) expression of TrCel5A. The CaMV promoter (P35SS) and terminator signal (pA35S) are indicated in light grey. The chimeric promoter (alcAmin35S) comprising the CaMV 35S minimal promoter (±31 to +5) fused to the upstream promoter sequences of alcA (Caddick et al., 1998) is shown in black. 5′-UTR of chalcone synthase (CHS), plant codon-optimized leader peptide (LPH) derived from the heavy chain of the murine mAb24, the gene of interest (trcel5A) and the His₆ coding sequence (His6) are indicated in dark grey.

Figure 2

Western blot of different transgenic lines constitutively expressing TrCel5A after SDS-PAGE (A) and zymography performed with SDS-PAGE containing 0.15% (w/v) CMC (B). Lanes contain 10 μg of plant total soluble protein. The recombinant enzyme was detected with a polyclonal α-cellulase antibody and an alkaline phosphatase conjugated goat-anti-rabbit secondary antibody. The zymogram control is performed with purified TrCel5A produced in Hansenula polymorpha.

Figure 3

Enzymatic activity of TrCel5A at different pH and temperature values using the azoCMC assay. Activity of TrCel5A was determined for pH values between 3.0 and 7.0 (A) and temperature between 20–70°C (B). The maximum activity measured for each system (at pH 4.8 and 55°C, respectively) was set to 100%. For panel A, the buffer systems are as indicated in the figure and temperature was set at 55°C. For panel B 50 mM Na-Acetate, pH 4.8 buffer was used throughout.

Figure 4

Effects of Ethanol induction on transgenic plants. Activity on 4MUC measured over time after ethanol induction in six-week-old soil-grown plants (A). Plants were induced at t0 by applying 2% ethanol in 100 ml irrigation water, and cellulase activity was monitored over a 96 h time course. Ethanol dose response in alcR::TrCel5A lines (B). Transgenic lines were monitored for cellulase activity 24 h after watering. Values represent the mean of three plants per independent transgenic line. Effect of sequential ethanol induction (C). Six-week-old soil-grown plants from the homozygous line F6.5 were induced using 2% ethanol at t0 and again after 48 h (asterisks). Cellulase activity was monitored over 144 h. Wild-type plants were monitored in parallel, and no cellulase activity was observed throughout the time course. Comparison of cellulose content and cellulase activity after repeated induction with ethanol (D). A relative value of 100% cellulase activity represents a conversion of 27 nmol 4MU min^-1 mg^-1, whereas a value of 100% for cellulose content represents 140 μg glucose per mg alcohol insoluble residue (AIR). For all panels, values represent the mean of three plants per independent transgenic line. Error bars the show standard deviation of the mean after subtraction of wild-type control data.

Figure 5

Phenotype of transgenic tobacco plants. Strains shown are wild-type N. tabacum SR1 (A). and transgenic tobacco strains with constitutive TrCel5A expression 35::TrCel5A (B). and inducible TrCel5A expression alcR::TrCel5A (C). Plants were grown under photoautotrophic conditions in soil. Images shown are representative of three plants per genetic line.

Figure 6

Cross sections of transgenic and wild-type tobacco stems (imaged at 10x) stained with calcofluor-white and visualized under UV light. Tissue sections of wild-type plants (A) were compared with 35SS::Trcel5A (B) and alcR::TrCel5A (C) transgenic plants. No significant difference was detected between the wild-type and transgenic plants, except a marginal increase in the number of small vessels (white arrow) in 35SS::Trcel5A plants. Scale bars are 50 μm.