How recombinant swollenin from Kluyveromyces lactis affects cellulosic substrates and accelerates their hydrolysis

Abstract

Background: In order to generate biofuels, insoluble cellulosic substrates are pretreated and subsequently hydrolyzed with cellulases. One way to pretreat cellulose in a safe and environmentally friendly manner is to apply, under mild conditions, non-hydrolyzing proteins such as swollenin - naturally produced in low yields by the fungus Trichoderma reesei. To yield sufficient swollenin for industrial applications, the first aim of this study is to present a new way of producing recombinant swollenin. The main objective is to show how swollenin quantitatively affects relevant physical properties of cellulosic substrates and how it affects subsequent hydrolysis.

Results: After expression in the yeast Kluyveromyces lactis, the resulting swollenin was purified. The adsorption parameters of the recombinant swollenin onto cellulose were quantified for the first time and were comparable to those of individual cellulases from T. reesei. Four different insoluble cellulosic substrates were then pretreated with swollenin. At first, it could be qualitatively shown by macroscopic evaluation and microscopy that swollenin caused deagglomeration of bigger cellulose agglomerates as well as dispersion of cellulose microfibrils (amorphogenesis). Afterwards, the effects of swollenin on cellulose particle size, maximum cellulase adsorption and cellulose crystallinity were quantified. The pretreatment with swollenin resulted in a significant decrease in particle size of the cellulosic substrates as well as in their crystallinity, thereby substantially increasing maximum cellulase adsorption onto these substrates. Subsequently, the pretreated cellulosic substrates were hydrolyzed with cellulases. Here, pretreatment of cellulosic substrates with swollenin, even in non-saturating concentrations, significantly accelerated the hydrolysis. By correlating particle size and crystallinity of the cellulosic substrates with initial hydrolysis rates, it could be shown that the swollenin-induced reduction in particle size and crystallinity resulted in high cellulose hydrolysis rates.

Conclusions: Recombinant swollenin can be easily produced with the robust yeast K. lactis. Moreover, swollenin induces deagglomeration of cellulose agglomerates as well as amorphogenesis (decrystallization). For the first time, this study quantifies and elucidates in detail how swollenin affects different cellulosic substrates and their hydrolysis.

Figure 1

SDS-PAGE, Western blot and mass spectrometry of swollenin produced by Kluyveromyces lactis. (A) SDS-PAGE and (B) Western blot: (M) Molecular mass marker, (1) filtrated culture supernatant of K. lactis wild type, (2) filtrated culture supernatant of K. lactis expressing recombinant swollenin, (3) recombinant swollenin purified by immobilized metal affinity chromatography. 12% polyacrylamide gel, the same volume of the samples (15 μL) was loaded onto the particular slots; (C) Mass spectrometric results and primary sequence of recombinant swollenin. The protein band (around 80 kDa) was analyzed using a mass spectrometer and the Mascot database. The detected peptides are underlined and written in italic letters. The cellulose-binding domain [6-39], expansinA domain [243-401] and His-tag [476-483] are marked in grey. Potential areas for N-glycosylation and O-glycosylation are written in bold letters. The black arrows enclose the primary sequence of the native swollenin (CAB92328) without leader peptide.

Figure 2

Adsorption isotherm of purified swollenin onto filter paper. The predicted Langmuir isotherm, according to Equation (1), is shown as a solid line (R² = 0.91) and corresponding parameters (including standard deviations) are: A_max(swollenin) = 0.089 ± 0.006 μmol/g, K_D(swollenin) = 0.707 ± 0.196 μmol/L. The initial swollenin concentration, added at the start of the incubation, is also shown for a better understanding of Figure 8; 20 g/L Whatman filter paper No.1 in 0.05 M sodium acetate buffer at pH 4.8, T = 45°C, V_L = 1 mL, n = 1000 rpm, d₀ = 3 mm, incubation time 2 h.

Figure 3

Photography and light microscopy of filter paper after pretreatment with swollenin. (A) Macroscopic pictures of pretreated filter paper in petri dishes and number of agglomerates. All pretreatments were initiated with the same initial number (80) of filter paper agglomerates (initial diameter approx. 3 mm). Number of agglomerates (> 0.5 mm) was measured by image analysis; (B) Light microscopy of pretreated filter paper. Eclipse E600 (Nikon); Pretreatment: 20 g/L cellulose in 0.05 M sodium acetate buffer at pH 4.8, 0.4 g/L BSA (approx. 6 μmol/L) or swollenin (approx. 5 μmol/L), T = 45°C, V_L = 1 mL, n = 1000 rpm, d₀ = 3 mm, incubation time 48 h.

Figure 4

Scanning electron microscopy of filter paper after pretreatment with swollenin. Pictures were taken at two different magnifications (A, B): see scale markers; Pretreatment: 20 g/L cellulose in 0.05 M sodium acetate buffer at pH 4.8, 0.4 g/L BSA (approx. 6 μmol/L) or swollenin (approx. 5 μmol/L), T = 45°C, V_L = 1 mL, n = 1000 rpm, d₀ = 3 mm, incubation time 48 h. Hitachi S-5500 (Hitachi).

Figure 5

Particle size of cellulosic substrates after pretreatment with swollenin. (A, B, C, D) Volumetric particle-size distribution of pretreated cellulosic substrates: (A) Whatman filter paper No.1; (B) α-Cellulose; (C) Avicel PH101; (D) Sigmacell 101; (E, F, G, H) Geometric mean particle size of pretreated cellulosic substrates: (E) Whatman filter paper No.1; (F) α-Cellulose; (G) Avicel PH101; (H) Sigmacell 101. Errors are given as standard deviations; Pretreatment: 20 g/L cellulose in 0.05 M sodium acetate buffer at pH 4.8, 0.4 g/L BSA (approx. 6 μmol/L) or 0.4 g/L swollenin (approx. 5 μmol/L), T = 45°C, V_L = 1 mL, n = 1000 rpm, d₀ = 3 mm, incubation time 48 h. Particles (< 2 mm) were analyzed using the particle size analyzer LS13320 (Beckman Coulter).

Figure 6

Crystallinity index of cellulosic substrates after pretreatment with swollenin. (A) Whatman filter paper No.1; (B) α-Cellulose; (C) Avicel PH101; (D) Sigmacell 101. Errors are given as standard deviations; Pretreatment: 20 g/L cellulose in 0.05 M sodium acetate buffer at pH 4.8, 0.4 g/L BSA (approx. 6 μmol/L) or 0.4 g/L swollenin (approx. 5 μmol/L), T = 45°C, V_L = 1 mL, n = 1000 rpm, d₀ = 3 mm, incubation time 48 h. Powder XRD (STOE & Cie GmbH).

Figure 7

Hydrolysis of cellulosic substrates after pretreatment with swollenin. (A) Whatman filter paper No.1; (B) α-Cellulose; (C) Avicel PH101; (D) Sigmacell 101. Errors are given as standard deviations; Pretreatment: 20 g/L cellulose in 0.05 M sodium acetate buffer at pH 4.8, 0.4 g/L BSA (approx. 6 μmol/L) or 0.4 g/L swollenin (approx. 5 μmol/L), T = 45°C, V_L = 1 mL, n = 1000 rpm, d₀ = 3 mm, incubation time 48 h; Hydrolysis: 10 g/L pretreated cellulose in 0.05 M sodium acetate buffer at pH 4.8, 1 g/L rebuffered Celluclast^®, T = 45°C, V_L = 1 mL, n = 1000 rpm, d₀ = 3 mm.

Figure 8

Hydrolysis of filter paper after pretreatment with different swollenin concentrations. Errors are given as standard deviations; Pretreatment: 20 g/L Whatman filter paper No.1 in 0.05 M sodium acetate buffer at pH 4.8, different concentrations of swollenin, T = 45°C, V_L = 1 mL, n = 1000 rpm, d₀ = 3 mm, incubation time 48 h; Hydrolysis: 10 g/L pretreated cellulose in 0.05 M sodium acetate buffer at pH 4.8, 1 g/L rebuffered Celluclast^®, T = 45°C, V_L = 1 mL, n = 1000 rpm, d₀ = 3 mm.

Figure 9

Influence of crystallinity and mean particle size on hydrolysis of cellulosic substrates. Data points for cellulosic substrates were obtained from Figure 5 (mean particle size), Figure 6 (crystallinity index) and Figure 7 (initial hydrolysis rate from 0 to 6 h). TableCurve 3D was used to determine an empirical surface fit (R² = 0.93) based on a non-linear Gaussian cumulative function.