Expression of a fungal ferulic acid esterase in alfalfa modifies cell wall digestibility

Abstract

Background: Alfalfa (Medicago sativa) is an important forage crop in North America owing to its high biomass production, perennial nature and ability to fix nitrogen. Feruloyl esterase (EC 3.1.1.73) hydrolyzes ester linkages in plant cell walls and has the potential to further improve alfalfa as biomass for biofuel production.

Results: In this study, faeB [GenBank:AJ309807] was synthesized at GenScript and sub-cloned into a novel pEACH vector containing different signaling peptides to target type B ferulic acid esterase (FAEB) proteins to the apoplast, chloroplast, endoplasmic reticulum and vacuole. Four constructs harboring faeB were transiently expressed in Nicotiana leaves, with FAEB accumulating at high levels in all target sites, except chloroplast. Stable transformed lines of alfalfa were subsequently obtained using Agrobacterium tumefaciens (LBA4404). Out of 136 transgenic plants regenerated, 18 independent lines exhibited FAEB activity. Subsequent in vitro digestibility and Fourier transformed infrared spectroscopy (FTIR) analysis of FAEB-expressing lines showed that they possessed modified cell wall morphology and composition with a reduction in ester linkages and elevated lignin content. Consequently, they were more recalcitrant to digestion by mixed ruminal microorganisms. Interestingly, delignification by alkaline peroxide treatment followed by exposure to a commercial cellulase mixture resulted in higher glucose release from transgenic lines as compared to the control line.

Conclusion: Modifying cell wall crosslinking has the potential to lower recalcitrance of holocellulose, but also exhibited unintended consequences on alfalfa cell wall digestibility due to elevated lignin content. The combination of efficient delignification treatment (alkaline peroxide) and transgenic esterase activity complement each other towards efficient and effective digestion of transgenic lines.

Figure 1

Gene expression analysis. (A) Western blot showing transient expression of faeB proteins in tobacco leaves. L1, protein ladder (catalogue number 161-0374; Bio-Rad, Hercules, CA, USA); L2, uninoculated leaf proteins; L3, proteins of tobacco leaves infected with LBA4404 (no construct); L4, proteins of tobacco leaves infected with agro harboring pEACH 5,103; L5, proteins of tobacco leaves infected with agro harboring faeB-apoplast; L6, proteins of tobacco leaves infected with agro harboring faeB-chloroplast; L7, proteins of tobacco leaves infected with agro harboring faeB-ER; and L8, proteins of tobacco leaves infected with agro harboring faeB-vacuole. (B) Transient expression of GUS in alfalfa leaves. tCUP:GUS was stably transformed into alfalfa plants (obtained from Dr Lining Tian, Agriculture and Agri-Food Canada London, ON, Canada), served as a positive control. ER, endoplasmic reticulum; GUS, β-glucuronidase.

Figure 2

FAEB activity assay, PCR validation of faeB and immunocytochemical localization of FAEB. (A) FAEB activity as indicated by the change in absorbance over 60 min as a result of the hydrolysis of ethyl ferulate. Data represents three absorbance readings from the same extract from a single plant. Error bars indicate relative standard error. Apoplast (A), transgenic line 3A; control, wild type plant; endoplasmic reticulum (ER), transgenic line 28ER and vacuole (V), transgenic line 2 V. (B) PCR validation of faeB gene in transgenic lines. Apoplast-expressing sample: lane 1, nptII (apoplast); lane 2, faeB-apoplast and lane 3, wild type negative control for faeB-apoplast. ER-expressing sample: lane 4, nptII (ER); lane 5, faeB-ER and lane 6, wild type negative control for faeB-ER. Vacuole-expressing sample: lane 7, nptII (vacuole); lane 8, faeB-vacuole and lane 9, wild type negative control for faeB-vacuole. (C) Immunocytochemical localization to confirm recombinant protein expression in: (a) apoplast; (b) endoplasmic reticulum; (c) vacuole and (d) non-transgenic control in alfalfa leaves. A, apoplast; ER, endoplasmic reticulum; FAEB, type B ferulic acid esterase; V, vacuole.

Figure 3

Effect of FAEB activity on in vitro dry matter disappearance (IVDMD) of control and various transgenic lines incubated with mixed ruminal fluid. Bars indicate standard error. *Differs to control at P < 0.05. Repeated two times with triplicate samples per incubation. A, apoplast; ER, endoplasmic reticulum; FAEB, type B ferulic acid esterase; IVDMD, in vitro dry matter disappearance; V, vacuole.

Figure 4

Fourier transformed infrared spectroscopy (FTIR) data analysis. (A) PCA of FTIR data. (B) Loading of factor score (F1) corresponding to PCA major spectral differences between transgenic and wild type lines. Apoplast (A), average spectrum of 43A, 41A and 1A; endoplasmic reticulum (ER), average spectrum of 24ER and 28ER; and vacuole (V), average spectrum of 61 V, 15 V and 2 V. C, control.

Figure 5

Lignin content and sugar compositional analysis. (A) Acetyl bromide soluble lignin content. Bars indicate standard error of mean (n = 3). *Differs to control at P 0.05. (B) Sugar composition of control and transgenic alfalfa. Sugars were quantified as alditol acetate derivatives by GC-MS. Error bars indicate standard deviation (n = 3) of technical replicates from ground samples pooled from 50 to 60 whole plants per genotype. Alfalfa lines indicate expression of faeB in endoplasmic reticulum (ER), 24ER and 28ER; apoplast (A), 43A, 41A and 1A; and vacuole (V), 61 V, 15 V and 2 V. A, apoplast; C, wild type control; ER, endoplasmic reticulum; GC-MS, gas chromatography–mass spectrometry; V, vacuole.

Figure 6

Residual lignin, total sugar and uronic acid content of cell walls after 72 h of in vitro digestion. .Relative percentage of (A) residual lignin, (B) sugar and (C) uronic acid, which equal the percentage of lignin, sugar and uronic acid, respectively, remaining after 72 h of digestion relative to that of the control (control being 100%). Bars indicate standard error of mean (n = 3). *Differs to control at P 0.05. Apoplast (A), faeB-apoplast (average of 43A, 41A and 1A); endoplasmic reticulum (ER), faeB-ER (average of 24ER and 28ER); and vacuole (V), faeB-vacuole (average of 61 V, 15 V and 2 V). A, apoplast; ER, endoplasmic reticulum; V, vacuole; WT, wild type.

Figure 7

Cell wall extractable phenolics analysis. (A) Release of ferulic acid and p-coumaric acid as a result of extraction with 1 M NaOH. (B) UV spectra of lignin fraction extracted with 1 M NaOH from control and three representative transgenic lines from endoplasmic reticulum (ER), apoplast (A), vacuole (2 V). and C, control.

Figure 8

Effect of delignification by alkaline peroxide pretreatment on glucose released from control and transgenic lines as a result of hydrolysis with commercial enzyme preparations (Accellerase 1500). Bars indicate standard error of mean (n = 8). *Differs to control at P <0.05. Alfalfa lines indicate expression of faeB in endoplasmic reticulum (ER), 24ER and 28ER; apoplast (A), 43A, 41A and 1A; and vacuole (V), 61 V, 15 V and 2 V. A, apoplast; C, wild type control; ER, endoplasmic reticulum; V, vacuole.