The TcEG1 beetle (Tribolium castaneum) cellulase produced in transgenic switchgrass is active at alkaline pH and auto-hydrolyzes biomass for increased cellobiose release

Abstract

Background: Genetically engineered biofuel crops, such as switchgrass (Panicum virgatum L.), that produce their own cell wall-digesting cellulase enzymes would reduce costs of cellulosic biofuel production. To date, non-bioenergy plant models have been used in nearly all studies assessing the synthesis and activity of plant-produced fungal and bacterial cellulases. One potential source for cellulolytic enzyme genes is herbivorous insects adapted to digest plant cell walls. Here we examine the potential of transgenic switchgrass-produced TcEG1 cellulase from Tribolium castaneum (red flour beetle). This enzyme, when overproduced in Escherichia coli and Saccharomyces cerevisiae, efficiently digests cellulose at optima of 50 °C and pH 12.0.

Results: TcEG1 that was produced in green transgenic switchgrass tissue had a range of endoglucanase activity of 0.16-0.05 units (µM glucose release/min/mg) at 50 °C and pH 12.0. TcEG1 activity from air-dried leaves was unchanged from that from green tissue, but when tissue was dried in a desiccant oven (46 °C), specific enzyme activity decreased by 60%. When transgenic biomass was "dropped-in" into an alkaline buffer (pH 12.0) and allowed to incubate at 50 °C, cellobiose release was increased up to 77% over non-transgenic biomass. Saccharification was increased in one transgenic event by 28%, which had a concurrent decrease in lignin content of 9%. Histological analysis revealed an increase in cell wall thickness with no change to cell area or perimeter. Transgenic plants produced more, albeit narrower, tillers with equivalent dry biomass as the control.

Conclusions: This work describes the first study in which an insect cellulase has been produced in transgenic plants; in this case, the dedicated bioenergy crop switchgrass. Switchgrass overexpressing the TcEG1 gene appeared to be morphologically similar to its non-transgenic control and produced equivalent dry biomass. Therefore, we propose TcEG1 transgenics could be bred with other transgenic germplasm (e.g., low-lignin lines) to yield new switchgrass with synergistically reduced recalcitrance to biofuel production. In addition, transgenes for other cell wall degrading enzymes may be stacked with TcEG1 in switchgrass to yield complementary cell wall digestion features and complete auto-hydrolysis.

Keywords: Auto-hydrolysis; Biofuel; Cellulase; Glycosyl hydrolase; Insect; Switchgrass; Tribolium castaneum; β-1,4-Endoglucanase.

Fig. 1

Transformation vector map and relative transcript abundance of TcEG1 in transgenic switchgrass. a pANIC-10A-TcEG1 vector used for expression of TcEG1 in transgenic switchgrass. LB: left border; PvUbi2: switchgrass ubiquitin 2 promoter and intron; hph: hygromycin B phosphotransferase coding region; 35S T: 35S terminator sequence; PvUbi1: switchgrass ubiquitin 1 promoter and intron; pporRFP: Porites porites orange fluorescent protein coding region; NOS T: Agrobacterium tumefaciens nos terminator sequence; ZmUbi1: maize ubiquitin 1 promoter; R1 and R2: attR1 and attR2 recombinase sites 1 and 2; TcEG1: TcEG1 cDNA open reading frame; AcV5: epitope tag; RB: right border; Kan^r: kanamycin resistance gene; ColE1: origin of replication in E. coli; pVS1: origin of replication in A. tumefaciens; OCS T: octopine synthase terminator sequence. b Relative transcript abundance of TcEG1 in stem internodes from transgenic events (Tc-1 to Tc-12). Relative expression analysis was determined by qRT-PCR and normalized to switchgrass ubiquitin 1 (PvUbi1). Bars represent mean values of three replicates ± standard error. Bars represented by different letters are significantly different as calculated by LSD (p ≤ 0.05)

Fig. 2

Endoglucanase activity (units/mg of protein) from fresh leaves of transgenic TcEG1 plants. a Endoglucanase activity measurement using carboxymethyl cellulose (CMC) as substrate at pH 12.0 on TcEG1 extracted from fresh leaves. Bars represent mean values of three replicates ± standard error for each transgenic event. Bars represented by different letters are significantly different as calculated by LSD (p ≤ 0.05). b Gradient pH measurement of endoglucanase activity of TcEG1 extracted from fresh leaves of transgenic event Tc-1. Data points represent mean values of three replicates ± standard error. Data points represented by different letters are significantly different as calculated by LSD (p ≤ 0.05)

Fig. 3

Glucose (a), xylose (b), and total sugar (c) release from transgenic TcEG1 and non-transgenic (NT-Perf) tillers as determined by enzymatic hydrolysis. Bars represent mean values of three replicates ± standard error. Bars represented by different letters are significantly different as calculated by LSD (p ≤ 0.05)

Fig. 4

Endoglucanase activity (units/mg of protein) from leaves of three transgenic TcEG1 events using carboxymethyl cellulose (CMC) as substrate at pH 12.0. Leaves were either air dried for 2 weeks in the greenhouse (black bars) or dried for 3 days in an oven at 46 °C (gray bars). Bars represent mean values of three replicates ± standard error. Bars represented by different letters are significantly different as calculated by LSD (p ≤ 0.05)

Fig. 5

Auto-hydrolysis of TcEG1 switchgrass and non-transgenic switchgrass incubated in alkaline buffer (pH 12.0) at 50 °C. a Cellobiose released mg/mL from transgenic TcEG1 and non-transgenic (NT-Perf) lines over time. b Glucose released mg/mL from transgenic TcEG1 and non-transgenic (NT-Perf) lines over time. Bars represent mean values of three biological replicates ± standard error. Asterisk denotes statistical significant difference of released substrate over time at for event Tc-1 and Tc-12 p < 0.001 and Tc-6 p = 0.004 using Holm–Sidak method for pairwise comparison for one-way ANOVA with repeated measures

Fig. 6

Lignin content (a) and S/G ratio (b) of transgenic TcEG1 and non-transgenic (NT-Perf) tillers as determined by Py-MBMS. Bars represent mean values of three replicates ± standard error. Bars represented by different letters are significantly different as calculated by LSD (p ≤ 0.05)

Fig. 7

Cell wall measurements on histological analysis of stem internode sections of transgenic TcEG1 and non-transgenic (NT-Perf) plants. Measurement of cell wall perimeters (a), cell wall thickness (b), and cell wall areas (c). Representative images of non-transgenic (d) and transgenic event Tc-6 (e) stem internodes stained with Pontamine Fast Scarlet. Bars represent mean value of replicates ± standard error. Bars represented by different letters are significantly different as calculated by LSD (p ≤ 0.05). Scale bar represents 100 µm

Fig. 8

Cellulose crystallinity index measurements for transgenic TcEG1 and non-transgenic (NT-Perf) plants. Bars represent mean values of three replicates ± standard error. Bars represented by different letters are significantly different as calculated by LSD (p ≤ 0.05)

Fig. 9

Plant morphology analysis of transgenic TcEG1 and non-transgenic switchgrass plants. a Representative transgenic TcEG1 and non-transgenic (NT-Perf) lines. Tiller height (b), stem width taken at 10 cm height above potting mixture (c), tiller number (d), and biomass dry weight (e) of transgenic TcEG1 and non-transgenic (NT-Perf) plants. Bars represent mean values of three replicates ± standard error. Bars represented by different letters are significantly different as calculated by LSD (p ≤ 0.05)