Cellulose

, Volume 21, Issue 4, pp 2433–2444 | Cite as

Quartz crystal microbalance with dissipation monitoring of the enzymatic hydrolysis of steam-treated lignocellulosic nanofibrils

  • Akio Kumagai
  • Shinichiro Iwamoto
  • Seung-Hwan Lee
  • Takashi Endo
Original Paper

Abstract

Quartz crystal microbalance with dissipation (QCM-D) monitoring was performed to investigate the impact of steam treatment (ST) on the enzymatic hydrolysis of lignocellulosic nanofibrils (LCNFs). ST at mild temperatures up to 140 °C mainly affected the hemicellulose content of LCNFs. The hemicellulose constituents in the water-soluble fraction and the residual LCNF were quantified. The impact of changes in hemicellulose by ST on enzymatic hydrolysis was monitored by QCM-D using Acremonium cellulase as a source of multicomponent enzymes including hemicellulases. LCNFs without ST showed distinctive initial changes in frequency and energy dissipation, which differed from those of pure cellulose film, whereas these changes shifted toward typical changes of enzymatic hydrolysis of pure cellulosic films with increasing ST temperature. The QCM-D results suggested that hemicellulose located around cellulose microfibrils is rapidly decomposed, thus exposing the cellulose surface shortly after initial enzymatic hydrolysis, and then the main enzymatic hydrolysis of cellulose occurs.

Keywords

Lignocellulosic nanofibril Hemicellulose Quartz crystal microbalance with dissipation Thin film Enzymatic hydrolysis Steam treatment 

Introduction

Lignocellulosic biomass has been identified as an alternative and renewable source for biofuels and other value-added materials (Cardona and Sánchez 2007). Lignocellulose is mainly composed of lignin, carbohydrates such as cellulose and hemicellulose, and other components such as pectin and ash, and these components are hierarchically deposited (Van Dyk and Pletschke 2012). Because of its chemical and structural complexity, the enzymatic hydrolysis of lignocellulosic biomass is a complex and labor-intensive procedure (Himmel et al. 2007). To overcome these recalcitrant characteristics for bioconversion, especially enzymatic conversion of carbohydrates into monosaccharides, a pretreatment step is first required. Pretreatment is usually conducted to make cellulose more accessible to the enzymes by destroying its robust structure using chemical or physical treatments (Hendriks and Zeeman 2009). To optimize and develop an effective pretreatment protocol and achieve enzymatic conversion, it is first important to understand the mechanisms driving the enzymatic hydrolysis of lignocellulose, such as the interaction between enzymes and the lignocellulosic substrate occurring during hydrolysis (Himmel et al. 2007; Van Dyk and Pletschke 2012).

There are many different varieties of enzymes with different specificities depending on the individual carbohydrates, cellulose, and hemicellulose contained in the lignocellulose. Their enzymatic degradation is achieved by the synergistic action of multiple enzymes in defined ratios (Van Dyk and Pletschke 2012). Cellulose, the main constituent of lignocellulose, is composed of chains of glucose that are linked by β-1,4-glycosidic bonds, and the cellulases that catalyze the cleavage of these β-1,4-links of cellulose are generally classified into three types, namely, cellobiohydrolases (exo-β-1,4-glucanases), endo-glucanases (endo-β-1,4-glucanases), and β-glucosidases, depending on their respective reaction patterns. Cellobiohydrolases hydrolyze the chain ends of crystalline cellulose and release cellobiose, while endo-glucanases catalyze the random cleavage of internal β-1,4-glycosidic bonds of amorphous cellulose. β-glucosidases hydrolyze cellobiose and other soluble cello-oligosaccharides to monomeric glucose (Bayer et al. 1998; Van Dyk and Pletschke 2012). On the other hand, hemicellulose is a heterogeneous polysaccharide that is found in different forms depending on its structure and composition, such as xylan, mannan, galactan, and arabinan. Therefore, there are various types of potential hemicellulases available for the hydrolysis of hemicellulose (Shallom and Shoham 2003; Van Dyk and Pletschke 2012). For example, mannanases are core enzymes for hydrolyzing internal β-1,4-linkages of mannan, which is a major constituent of hemicellulose in softwood. Since lignocellulose contains various carbohydrates and multiple enzymes are implicated in their hydrolysis, the enzymatic hydrolysis of lignocellulose is highly complex.

The enzymatic hydrolysis of lignocellulose has been generally evaluated by measuring the overall production of monosaccharides. A few studies have focused on evaluation of the dynamic phenomena occurring between enzymes and their substrates, especially in the initial stage in which cellulase enzymes approach lignocellulose when it is tightly packed in the form of cellulose microfibrils encased in hemicellulose and lignin (Mansfield et al. 1999). The enzymatic hydrolysis of lignocellulose is a reaction that occurs at the solid–liquid interface because lignocellulose is a water-insoluble substrate. Quartz crystal microbalance with dissipation (QCM-D) is known as an appropriate technique for in situ and real-time studies of phenomena occurring at the solid–liquid interface. Several studies have reported the enzymatic hydrolysis of cellulose with QCM-D in which the quartz crystal sensors are coated with cellulosic thin films. For example, Ahola et al. (2008a) studied the enzyme binding and hydrolytic reactions occurring during enzymatic hydrolysis of cellulose by a multicomponent enzyme mixture using various cellulosic thin films prepared from regenerated cellulose, cellulose nanocrystals, and nanofibrillated cellulose. Turon et al. (2008) proposed a simple enzymatic kinetic model to quantify the enzyme adsorption and hydrolysis rate of cellulosic thin film with a combined Boltzmannn–sigmoidal model. The enzymatic hydrolysis of cellulose by monocomponent cellulases has also been studied using QCM-D. Josefsson et al. (2008) evaluated the function of individual cellulases on cellulose films by using pure cellobiohydralase and endo-glucanase for QCM-D, and reported the synergetic effects caused by endo-glucanase swelling combined with film softening during degradation. Moreover, Maurer et al. (2012, 2013) provided a detailed description of the competitive adsorption and cooperative activity of cellobiohydralase and endo-glucanase by comparison of QCM-D results and a Langmuir–Michaelis–Menten kinetic model. In addition, some QCM-D experiments have been performed using a cellulosic model film containing other macromolecules contained in lignocellulose. Hoeger et al. (2012) and Martin-Sampedro et al. (2013) prepared regenerated bicomponent cellulose/lignin films on quartz crystal sensors and investigated the effects on enzymatic hydrolysis using multicomponent cellulases and monocomponent cellulases, respectively. QCM-D studies on the enzymatic hydrolysis of multicomponent thin films prepared from lignocellulosic nanofibrils (LCNFs) have also been reported (Ahola et al. 2008a; Martin-Sampedro et al. 2012), and the results have indicated that this method achieves faster enzymatic hydrolysis than that obtained when using other regenerated amorphous cellulosic films (Ahola et al. 2008a). These LCNFs have been prepared from commercial pulps containing low amounts of lignin. On the other hand, our research group recently prepared LCNF films from lignin content-adjusted wood powders (Kumagai et al. 2013). The enzymatic hydrolysis of LCNFs showed characteristic lignin concentration-dependent QCM-D results, and the impact of the lignin content of the LCNFs on enzymatic hydrolysis could be accurately predicted.

The degradation of hemicellulose caused by hemicellulases also plays an important role in the enzymatic hydrolysis of lignocellulose with a multicomponent enzyme mixture (Shallom and Shoham 2003; Van Dyk and Pletschke 2012). Since it is known that hemicellulase enhances the degradation of lignocellulose, in general, the commercial enzyme preparations for enzymatic saccharification of lignocellulosic biomass contain some additional hemicellulases, and more specific hemicellulase is usually added as accessory enzymes in the saccharification experiments (Berlin et al. 2006; Van Dyk and Pletschke 2012).

This study was conducted to investigate the impact of changes in the hemicellulose content of LCNFs induced by steam treatment (ST) on enzymatic hydrolysis using QCM-D monitoring. The hemicellulose of LCNFs derived from Hinoki cypress wood powders was partially removed by ST at mild temperatures up to 140 °C. Hydrothermal treatment is one of the most common pretreatment methods used to improve the enzymatic hydrolysis of lignocellulose (Bobleter 1994; Yu et al. 2008), and is well known to selectively degrade hemicellulose (Lee et al. 2010). However, typical hydrothermal treatment at high temperature and pressure may cause excessive reactions, thus producing unfavorable fermentation inhibitors such as furfural and hydroxymethyl furfural, as well as low-molecular weight phenolic compounds (Yu et al. 2008). Fluidization of lignin can also occur during hydrothermal treatment at severe conditions, and the fluidized lignin becomes re-deposited on the cellulose surface during the cooling process (Donohoe et al. 2008). These unfavorable by-products may affect the process and reliability of QCM-D. Thus, in this study, ST was conducted at milder temperature conditions to prevent significant molecular changes of hemicellulose and lignin.

Materials and methods

Materials

Wood chips of Hinoki cypress (Chamaecyparis obtusa) were kindly supplied by Maniwa City (Okayama, Japan). The wood chips were cutter-milled to powders (<0.2 mm in size), and the obtained powder was used as the starting material for the preparation of LCNFs. Commercial microfibrillated cellulose (Celish® KY-100G; Daicel Chemical Industries Co., Ltd.; Tokyo, Japan) was used as a reference for the pure cellulosic nanofibril. Acremonium cellulase (Meiji Seika Co.; Tokyo, Japan) derived from Talaromyces cellulolyticus [formerly known as Acremonium cellulolyticus (Fujii et al. 2014)] was used as a multicomponent enzyme mixture for enzymatic hydrolysis. Milli-Q water used in this study was prepared from water obtained from an ultrapure water production system (Aquarius RFU685DA; ADVANTEC; Tokyo, Japan), and all other chemicals were purchased from commercial sources without any further purification.

LCNF preparation

The wood powder was soaked in Milli-Q water (1 wt% suspension) and kept overnight at room temperature. The soaked powder was fibrillated using a disk mill (Super mass colloider MKCA6-2; Masuko Sangyo Co., Ltd.; Saitama, Japan). The grinding treatment was performed for 10 cycles with a clearance of 20–40 μm and a rotation speed of 1,800 rpm.

Adjustment of chemical composition

Following the disk-milling process, partial delignification of LCNFs was performed using sodium chlorite and acetic acid according to a modification of the Wise method (Wise et al. 1946). After readjusting the suspension to 1 wt%, sodium chlorite (0.666 g/100 g wood powder) and acetic acid (0.133 mL/100 g wood powder) were added to the suspension and heated in a water bath at 75 °C for 10 min. The partially delignified sample was vacuum-filtered with filter paper (No. 5A; ADVANTEC Toyo Co. Ltd.; Tokyo, Japan) and washed with copious amounts of Milli-Q water until the wash solution was colorless. The filtrated sample was again suspended in Milli-Q water without drying.

After partial delignification, ST was performed using an autoclave (SPT-3050P; ALP Co. Ltd.; Tokyo, Japan) at 100–140 °C for 1 h at a pressure of 0.10–0.37 MPa. The water-soluble fractions were hydrolyzed with 4 wt% sulfuric acid, and the total sugar content in the water-soluble fractions was determined by following the Laboratory Analytical Procedure (LAP) of the National Renewable Energy Laboratory (NREL) method using a high-performance liquid chromatography (HPLC) system as described in the next section (Sluiter et al. 2006).

The steam-treated sample was recovered, washed, and dispersed in Milli-Q water to achieve a 1 wt% suspension. The steam-treated LCNFs were further fibrillated 20 times using a high-pressure homogenizer (MASSCOMIZER MMX-L100; Masuko Sangyo Co., Ltd.; Saitama, Japan) at a pressure of approximately 200 MPa.

LCNF characterization

The constituent saccharides and Klason lignin contents were determined using a modification of the LAP of the NREL method (Sluiter et al. 2011). The steam-treated samples were freeze-dried and then vacuum-dried at 40 °C for 18 h before the analysis. The dried sample (50 mg) was placed in 72 wt% sulfuric acid (600 μL) with shaking at 30 °C for 90 min and then autoclaved at 121 °C for 1 h after dilution with 16.8 mL Milli-Q water to achieve a final acid concentration of 4 wt%. The autoclaved sample was separated into supernatant and residue fractions by filtration using a polytetrafluoroethylene membrane filter. The supernatant was neutralized with saturated barium hydroxide, and the monosaccharides in the supernatant were analyzed in the HPLC system (LC-2000 Plus; Jasco Co., Ltd.; Tokyo, Japan) equipped with an Aminex HPX-87P column (Bio-Rad Labs; Hercules, CA, USA) at 60 °C with a flow rate of 0.25 mL Milli-Q water/min. The vacuum-dried residue was washed with boiled Milli-Q water until the pH of the filtrate became neutral, and then the Klason lignin content of the residue was quantified by weighing the residue after complete drying using a vacuum-drying apparatus.

The crystallinity index (CrI) and specific surface area (SSA) were determined using steam-treated LCNFs. The LCNF samples were washed with tert-butyl alcohol several times until the water was thoroughly replaced by tert-butyl alcohol, and then freeze-dried to maintain the original structure before each analysis. The crystallinity of the LCNFs was measured using a RINT-TTR III diffractometer (Rigaku Co.; Tokyo, Japan) with Cu Kα radiation at 50 kV and 300 mA. Samples were scanned over the range of 2θ = 2°–60°at a rate of 2°/min. The CrI was calculated using the following equation based on the Segal method (Segal et al. 1959):
$${\text{CrI}}\left( \% \right) = \left[ {\left( {I_{00 2} - I_{\text{am}} } \right)/I_{00 2} } \right] \times 100$$
in which I002 is the intensity at approximately 2θ = 22.5° and Iam is the intensity at 2θ = 18.7°. The SSA of the LCNFs was measured using a BEL-SORP-Max adsorption system (BEL Japan Inc.; Osaka, Japan), and the SSA value was determined from a Brunauer–Emmett–Teller plot of the nitrogen adsorption–desorption isotherm (Brunauer et al. 1938).

Film preparation

The LCNF thin film was prepared on a QCM-D gold sensor (QSX 301; Q-sence AB; Götenborg, Sweden). Prior to film deposition, the QCM-D sensor was cleaned with UV/ozone Pro Cleaner™ (Bioforce Nanoscience Inc.; Ames, IA, USA) for 10 min, boiled with a mixture of 25 % ammonia solution and 30 % hydrogen peroxide (1:1:5 with Milli-Q water by volume) at 75 °C for 5 min, and cleaned again with UV/ozone Pro Cleaner™ for 10 min. The sensor was rinsed with Milli-Q water and dried with nitrogen gas at each step during the cleaning process. Owing to its cationic nature, polyethylene-imine (PEI) was used as an anchoring substance to improve attachment of the LCNFs on the sensor. The cleaned sensor was immersed in the 1 wt% PEI/Milli-Q water solution for 15 min, and then rinsed with Milli-Q water and dried with nitrogen gas.

The suspension of LCNFs was diluted with Milli-Q water to 0.5 wt% and agitated with an ultrasonic homogenizer (US-150T; Nihon Seiki Seisakujo; Tokyo, Japan) for 1 min to prevent aggregation and improve dispersibility. The sample was then centrifuged with a MiniSpin® plus centrifuge (Eppendorf; Hamburg, Germany) for 30 min at 10,000 rpm to remove any remaining fibril aggregates. The clear, centrifuged supernatant was used for spin coating to prepare the LCNF thin film after equalizing the concentration to adjust the film thickness using an ultraviolet spectrophotometer, according to methods described in our previous report (Kumagai et al. 2013). The supernatant of the LCNF suspension was dropped onto the PEI-absorbed sensor, incubated for 1 min, and then spin-coated at 3,000 rpm for 1 min using a spin coater (Opticoat MS-A100; Mikasa Co., Ltd.; Tokyo, Japan). The sensor-coated LCNF thin films were then heat-treated in an oven at 80 °C for 10 min.

Film morphology analysis

The morphology, roughness, and thickness of the prepared LCNF films on the sensor were monitored and evaluated using atomic force microscope (AFM) imaging (JSPM-5200; JEOL Co., Ltd.; Tokyo, Japan). The image was scanned in air in tapping mode using aluminum reflex-coated silicon cantilevers (PPP-NCHR; Nano World AG; Neuchâtel, Switzerland) at 25 °C and relative humidity of approximately 30 %. AFM imaging was also performed to characterize changes in the morphology of the films before and after enzymatic hydrolysis. Images of sizes corresponding to 5 μm × 5 μm, 10 μm × 10 μm, and 25 μm × 25 μm were taken in at least three different areas of each sensor. The roughness and thickness were calculated with WinSPM software (JEOL Co., Ltd.; Tokyo, Japan) using the 5 μm × 5 μm image. The thickness of the layers comprising the anchoring PEI substrate and the prepared LCNF thin films were determined using the scratching method with a needle using at least 20 height profiles, as reported previously (Ahola et al. 2008a).

QCM-D experiment

Investigation of the enzymatic adsorption and degradation behavior on the LCNF film was carried out using QCM-D monitoring (Q-Sense E1; Q-sence AB; Götenborg, Sweden). The enzymatic hydrolysis was performed at 40 °C in 50 mM sodium acetate buffer (pH 5.0). The sensor coated with the LCNF film was placed in the buffer solution overnight so that it would swell completely before use. The swollen sensor was washed with Milli-Q water, dried with nitrogen gas, and then mounted in the QCM-D flow cell. The buffer solution was injected into the flow cell at a flow rate of 50 μL/min using a peristaltic pump, and the sensor was swollen again until no appreciable frequency profiles were observed. Subsequently, the enzyme solution (0.050 mg/mL Acremonium cellulase in the buffer solution) was continuously introduced into the QCM-D flow cell at a flow rate of 50 μL/min for 10 min. QCM-D experiments were conducted in batch conditions. The introduction of enzyme solution was stopped when the initial buffer solution in the cell was fully replaced by the enzyme solution, and the enzymatic hydrolysis on the LCNFs was monitored in the absence of flow. The monitoring was continued for more than 6 h, even if no appreciable changes in frequency and dissipation were observed. Thereafter, the buffer solution was introduced at a flow rate of 50 μL/min to rinse the system. The frequency and dissipation changes were simultaneously monitored at the fundamental resonance frequency (5 MHz) and its overtones, 15, 25, 35, 55, and 75 MHz. The third overtone (15 MHz) was used for data evaluation. The experiments were repeated three times using the same samples and the variation across all experiments was negligible.

Results and discussion

Steam-treated LCNFs

In our previous study, thin films of LCNFs with different lignin contents were successfully prepared from Hinoki cypress for QCM-D monitoring (Kumagai et al. 2013). In the present study, partially delignified LCNFs with 16.7 wt% lignin were further steam-treated to adjust the hemicellulose content under mild temperature conditions.

Tables 1 and 2 summarize the chemical compositions of steam-treated LCNFs along with their CrI and SSA values, and the concentration of total sugars in the water-soluble fraction, respectively. The reduction in hemicellulose constituents increased with increasing ST temperature, and was obvious at 140 °C. Although hydrothermal treatment is typically performed at temperatures >150 °C and the distinct degradation of hemicellulose begins at around 180 °C (Bobleter 1994; Yu et al. 2008), the present results indicated that mannose, as the main hemicellulose constituent of softwood, and other sugars such as arabinose and galactose were liberated at lower temperature. Moreover, lignin contents also slightly decreased with increasing temperature as shown in Table 1, because very fine nanofibrils were used for ST.
Table 1

Chemical composition, CrI, and SSA of steam-treated LCNFs

Temperature (°C)

Chemical composition (%)

CrI (%)

SSA (aS,BET/m2/g)

Glucose

Mannose

Xylose

Galactose

Arabinose

Lignin

Untreated

48.8

9.6

3.2

1.4

1.1

16.7

56.7

189.1

100

50.0

8.3

2.8

1.1

1.0

13.2

58.8

212.3

120

55.7

7.8

2.2

1.0

0.9

12.9

60.5

223.0

140

61.2

5.8

1.5

0.3

n.d.

11.0

60.7

237.8

Table 2

Concentration of total sugar in the water-soluble fraction of steam-treated LCNFs after conversion into the monomeric form using acid hydrolysis

Temperature (°C)

Concentration (mg/g of biomass)

Glucose

Mannose

Xylose

Galactose

Arabinose

Totala

100

1.15

3.99

0.34

0.64

0.74

6.8

120

1.60

5.99

0.69

1.08

2.37

11.7

140

4.94

20.56

7.04

4.33

7.36

44.2

aThe value of total is the sum of the concentration of each total sugar amount

CrI and SSA are often used to evaluate the effect of a pretreatment on the physicochemical properties of biomass and enzymatic hydrolysis. As shown in Table 1, both the CrI and SSA values increased with increasing ST temperature. The small increase in CrI values with temperature may be attributed to an increase in the cellulose content in the LCNFs due to the degradation of amorphous hemicellulose or partial re-crystallization of amorphous cellulose regions adjacent to fibril regions, as reported previously (Bhuiyan et al. 2001; Nitsos et al. 2013). In addition, the increase in SSA values generally indicated the progress of fibrillation. Before disk milling, the SSA of cutter-milled wood powders was only 5.5 m2/g, whereas all SSA values of the LCNFs were >180 m2/g. These SSA values were even higher than those of the LCNFs prepared in our previous study (Kumagai et al. 2013), which may have been due to the further degradation of hemicellulose. In general, a higher SSA of lignocellulosic biomass is considered to be more suitable for improving glucose production yields by enzymatic hydrolysis because the SSA is positively correlated with the area of cellulose exposed to enzymes (Zhu et al. 2009).

Characterization of steam-treated LCNF thin films

Steam-treated LCNFs were sufficiently fibrillated to enable preparation of thin films appropriate for QCM-D experiments. The morphology, thickness, and roughness of the thin films prepared from steam-treated LCNFs were characterized by AFM. The AFM height images of these LCNF films are shown in Fig. 1a–d. The nanoscopic fibrillar network structures were observed in all LCNF films regardless of the temperature of ST. The morphologies were comparable with other LCNFs prepared from various kinds of pulps and partially delignified products reported previously (Ahola et al. 2008a, b; Kumagai et al. 2013; Martin-Sampedro et al. 2013). Table 3 summarizes the thickness and roughness values for the thin films, which were determined from 5 μm × 5-μm AFM images; neither thickness nor roughness were affected by changes in temperature. For QCM-D monitoring, controlling film thickness is important to effectively compare enzymatic adsorption and degradation. In fact, Suchy et al. (2011) reported that the results of QCM-D changed depending on the film thickness even if the same samples were used for the preparation of thin films. The thicknesses of all samples were successfully adjusted to approximately 12.8 nm by controlling the concentration of the LCNF suspension as described in our previous report (Kumagai et al. 2013). The film roughness ranged from 4.12 ± 0.14 to 4.69 ± 0.12 nm, which is comparable to the roughness of cellulosic nanofibril films used in QCM-D experiments in previous reports (Ahola et al. 2008a, b; Kumagai et al. 2013).
Fig. 1

AFM height images of LCNFs spin-coated on gold QCM-D sensors before (ad) and after (eh) enzymatic hydrolysis. Temperature condition of ST: untreated (a, e), 100 °C (b, f), 120 °C (c, g), and 140 °C (d, e). The scan size was 5 μm × 5 μm

Table 3

Thickness and roughness values of thin films prepared from steam-treated LCFNs

 

Temperature (°C)

Untreated

100

120

140

Thickness (nm)

12.9 ± 0.7

13.1 ± 0.5

12.8 ± 0.6

12.8 ± 0.7

Roughness (nm)

4.20 ± 0.18

4.80 ± 0.12

4.69 ± 0.12

4.12 ± 0.14

QCM-D study on enzymatic hydrolysis of steam-treated LCNFs

The impact of ST on the enzyme adsorption and hydrolysis of steam-treated LCNFs was evaluated by in situ QCM-D monitoring. Figure 2 shows a typical QCM-D profile of frequency and energy dissipation by the multicomponent cellulase mixture for a pure cellulosic nanofibril film. The pure cellulosic nanofibril film was prepared using commercial microfibrillated cellulose in the same manner as the LCNF film, and the QCM-D experimental conditions were also identical to those used for LCNF films. The first decrease in frequency observed was caused by enzyme adsorption on the substrate, as enzyme adsorption occurs immediately after the injection of enzyme solution prior to degradation of the substrate (Ahola et al. 2008a; Turon et al. 2008). Once the concurrent hydrolysis of the substrate became prominent, the frequency began to increase rapidly and then slowed down gradually over time until reaching a plateau at which point hydrolysis ceases. Meanwhile, the dissipation changes indicated changes in the viscoelasticity and film thickness of the substrate on the sensor (Ahola et al. 2008a; Turon et al. 2008). The first increase in dissipation after injection of the enzyme indicated an increase in film thickness that was mainly caused by enzyme adsorption. Subsequently, the rate of increase in dissipation became higher when the frequency started to increase after reaching the minimum. At this stage, enzymatic decomposition and adsorption would be occurring simultaneously. However, the greater increase in dissipation at this stage indicated that the viscoelasticity of the film was likely still high due to the adsorbed enzyme. The rate of dissipation then began to decrease due to substrate decomposition on the sensor, and finally reached a plateau, demonstrating a pattern similar to that observed for the change in frequency. The QCM-D profile depicted in Fig. 2 was not obtained when an inactive enzyme was used and the enzymatic hydrolysis was conducted under low temperature (Josefsson et al. 2008; Turon et al. 2008).
Fig. 2

Frequency (solid line) and dissipation (dotted line) change for the enzymatic hydrolysis of pure cellulosic nanofibrils. The reaction was conducted using 0.050 mg/mL Acremonium cellulase in 50 mM sodium acetate buffer (pH 5.0) at 40 °C

Figure 3a shows the full range of frequency changes that occurred during enzymatic hydrolysis of steam-treated LCNF films, and Fig. 3b shows the same process focused at the initial stage (0–30 min). Frequency changes in all samples leveled off after 2 h, irrespective of the temperature of ST. However, even though frequency change had reached a plateau by 6 h after enzyme hydrolysis, there were some residual nanofibers present in all samples as shown in Fig. 1e–h. This indicates that the LCNFs prepared in this study were not completely decomposed by enzymatic hydrolysis due to residual lignin and hemicellulose. These residual nanofibers were not observed on the sensors prepared from cellulose nanofibrils that were disintegrated from delignified sulfite pulps (Ahola et al. 2008a; Martin-Sampedro et al. 2013).
Fig. 3

Time course of frequency changes (a, b) and dissipation changes (c, d) overall (a, c) and at the initial stage (b, d) of the enzymatic reaction for steam-treated lignocellulose hydrolyzed by Acremonium cellulase. Temperature condition of ST: untreated (no marker), 100 °C (open circle), 120 °C (open triangle), and 140 °C (open square). Each reaction was conducted using 0.050 mg/mL Acremonium cellulase in 50 mM sodium acetate buffer (pH 5.0) at 40 °C

Different initial frequency changes were found among LCNFs steam-treated at different temperatures, as shown in Fig. 3b. The initial decrease in frequency, corresponding to enzyme adsorption, was not found in the LCNFs without ST or in those treated at 100 °C. In these two samples, the frequency first increased slightly and then the rate of increase declined before rising again. The same phenomenon was observed in our previous QCM-D study using LCNF films with lignin contents ranging from 6.5 to 26.4 % (Kumagai et al. 2013). However, LCNFs treated at higher temperature than 120 °C did not show this phenomenon, indicating the typical initial frequency decrease in initial stage as like pure cellulosic nanofibrils (see Fig. 2), even though the lignin content was maintained at 12.3–13.9 %. In the present study, the lignin content was also slightly decreased with increasing temperature, but there was no correlation between the initial frequency increase and lignin content. Therefore, the distinctive initial frequency increase is mostly likely related to other factors induced by ST.

Many researchers have suggested that cellulose microfibrils in the wood cell wall are surrounded by an amorphous matrix that is mainly composed of hemicellulose and lignin (Somerville et al. 2004; Himmel et al. 2007; Terashima et al. 2009). In addition, the multicomponent enzymes mixture used in this study contained some hemicellulases (Fujii et al. 2009). The initial frequency increase observed in LCNFs without ST and treated at 100 °C may have been caused by the degradation of hemicellulose located around cellulose microfibrils. A slight initial increase in frequency was also observed previously during the hydrolysis of cellulosic films by purified endo-glucanase (Martin-Sampedro et al. 2013), and predominant hydrolysis of amorphous cellulose by purified endo-glucanase was also observed by QCM-D monitoring (Suchy et al. 2011). The endo-type cellulases randomly hydrolyze internal β-1,4-glycosidic bonds in the cellulose chain and cause a rapid decrease of the amorphous region, liberating cellooligosaccharides (Bayer et al. 1998). Similarly, endo-type enzymes can also cause the degradation of the main amorphous hemicellulose backbone (Shallom and Shoham 2003). The backbone of glucomannan, which is the main hemicellulose component in softwoods including Hinoki cypress, can be hydrolyzed by β-mannanase of endo-type hemicellulase (Shallom and Shoham 2003), and the Acremonium cellulase used in this study contains high β-mannanase specific activity (Fujii et al. 2009). Thus, hemicellulose degradation can be considered one of the main factors contributing to the initial frequency increase observed in the enzymatic hydrolysis of LCNFs. On the other hand, hydrolysis of the main crystalline cellulose chains by exo-type cellulases such as cellobiohydralases showed the typical initial frequency decrease corresponding to enzyme adsorption (Josefsson et al. 2008). Therefore, the frequency profiles of enzymatic hydrolysis of LCNFs by multicomponent enzymes may depend on these major enzymatic reaction patterns.

Considering the results of the present and previous studies as described above, the following potential mechanism can be proposed to explain the initial frequency increase observed in the LCNFs without ST and those treated at 100 °C. The initial increase in frequency might be caused by rapid hydrolysis of the hemicellulose located around cellulose microfibrils (Bayer et al. 1998; Terashima et al. 2009) by hemicellulases contained in multicomponent Acremonium cellulase (Fujii et al. 2009). Since the hydrolysis of the hemicellulose is more rapid than enzyme adsorption, a frequency increase might be initially recorded in this stage (Suchy et al. 2011; Martin-Sampedro et al. 2013). The subsequent lower rate of frequency increase may be mainly due to the adsorption of cellulases such as cellobiohydralaseson the newly exposed cellulose surface by hemicellulose degradation (Josefsson et al. 2008). This frequency decrease suggests that a mass increase by enzyme adsorption is superior to the hydrolysis of substrate. Finally, the subsequent significant increase in frequency could be attributed to the main degradation of cellulose microfibrils.

On the other hand, LCNFs treated at temperatures higher than 120 °C showed initial frequency decreases (Fig. 3b), as observed for pure cellulose without an initial frequency increase (Fig. 2). This phenomenon might be explained by the increased cellulose surface area after ST at these higher temperatures, and it is considered that an SSA of more than 223 m2/g provides a sufficient cellulose surface for cellulase to increase its adsorption amount beyond its degradation amount. Since LCNFs treated at 140 °C had a more exposed cellulose surface, an increased number of cellulase components available to participate in the degradation of main cellulose microfibrils could adsorb on the LCNF films. Therefore, the frequency profile of LCNFs treated at 140 °C showed a lower peak point than that of LCNFs treated at 120 °C.

The results of the present study suggest that the hemicellulose content affects the initial frequency change in the enzymatic hydrolysis of LCNFs. In addition, the results of our previous QCM-D study with LCNF films showed that lignin content was mainly associated with the decrease in frequency observed during enzyme adsorption (Kumagai et al. 2013). Together, these studies suggest that QCM-D monitoring using chemical component-adjusted LCNF films has potential to provide valuable information of the mechanism underlying the enzymatic hydrolysis of lignocellulose to increase understanding of the interactions between enzymes and individual constituents.

Figure 3c, d shows the energy dissipation profiles corresponding to the frequency changes. The changes in dissipation also showed characteristic profiles that differed from the typical dissipation changes observed in enzymatic reactions of pure cellulose (Turon et al. 2008; Kumagai et al. 2013; Martin-Sampedro et al. 2013). As described above, the enzymatic hydrolysis of pure cellulosic nanofibrils showed an initial increase in dissipation after enzyme injection (Fig. 2), whereas no prominent increases in dissipation were observed in the steam-treated LCNFs (Fig. 3c). Small increases in dissipation (<3 × 10−6) were observed in LCNFs treated at temperatures above 120 °C.

Conversely, although the magnitude of the decrease differed among treatments, dissipation decreased immediately after enzyme injection in all cases (Fig. 3d). The initial decrease in dissipation probably occurred due to hemicellulose degradation, as described above with respect to frequency change. Furthermore, the initial decreases in dissipation were more evident than the initial increases in frequencies (Fig. 3d). Since the decrease in dissipation is also affected by the decrease in viscoelasticity, the hydrophilicity of amorphous hemicellulose may have contributed to the clearer changes in dissipation than frequency (Olsson and Salmén 2004). The subsequent increase followed by slight decrease in dissipation may also correspond to the decrease in frequency due to the adsorption of cellulases on the exposed cellulose surface. LCNFs treated at higher temperatures showed higher increases in dissipation. This result suggests that cellulase enzymes were probably adsorbed and penetrated more easily onto and into the LCNFs treated at higher temperatures because of the greater rate of hemicellulose degradation occurring at higher temperatures. Since hemicellulose is most closely associated with cellulose microfibril (Somerville et al. 2004; Himmel et al. 2007; Terashima et al. 2009), increased degradation of hemicellulose may weaken the structural bonds between cellulose microfibrils. This could consequently facilitate cellulase adsorption and result in increased viscoelasticity.

Conclusion

A QCM-D study was performed with thin films prepared from steam-treated LCNFs under mild temperature conditions up to 140 °C using a multicomponent enzyme mixture in order to investigate the impact of ST on the enzymatic hydrolysis of LCNFs. The initial QCM-D profiles differed among LCNFs steam-treated at each temperature condition. LCNFs without ST showed an initial increase in frequency and decrease in dissipation immediately after enzymatic hydrolysis began. With increasing ST temperature, this pattern shifted toward the typical changes observed during enzymatic hydrolysis of pure cellulosic nanofibrils. The QCM-D results suggested that rapid degradation of hemicellulose located around cellulose microfibrils occurs first, followed by main enzymatic hydrolysis of the cellulose microfibrils exposed as a result of the initial hydrolysis of hemicellulose. The results obtained by QCM-D monitoring using LCNF films prepared from lignocellulose-adjusted chemical components have potential to provide valuable information about the impacts of the individual constituents of lignocellulose on enzymatic hydrolysis.

Notes

Acknowledgments

This work was supported by the Japan–US cooperation project for research and standardization of Clean Energy Technologies.

References

  1. Ahola S, Turon X, Österberg M, Laine J, Rojas OJ (2008a) Enzymatic hydrolysis of native cellulose nanofibrils and other cellulose model films: effect surface structure. Langmuir 24:11592–11599CrossRefGoogle Scholar
  2. Ahola S, Salmi J, Johansson L-S, Lanie J, Österberg M (2008b) Model films from native cellulose nanofibrils Preparation, swelling, and surface interactions. Biomacromolecules 9:1273–1282CrossRefGoogle Scholar
  3. Bayer E, Chanzy H, Lamed R, Shoham Y (1998) Cellulose, cellulases and cellulosomes. Curr Opin Struct Biol 8:548–557CrossRefGoogle Scholar
  4. Berlin A, Maximenko V, Gilkes N, Saddler J (2006) Optimization of enzyme complexes for lignocellulose hydrolysis. Biotechnol Bioeng 97:287–296CrossRefGoogle Scholar
  5. Bhuiyan MTR, Hirai N, Sobue N (2001) Changes of crystallinity in wood cellulose by heat treatment under dried and moist conditions. Wood Sci 46:431–436CrossRefGoogle Scholar
  6. Bobleter O (1994) Hydrothermal degradation of polymers derived from plants. Prog Polym Sci 19:797–841CrossRefGoogle Scholar
  7. Brunauer S, Emmett PH, Teller E (1938) Adsorption of gases in multimolecular layers. J Am Chem Soc 60:309–319CrossRefGoogle Scholar
  8. Cardona CA, Sánchez ÓJ (2007) Fuel ethanol production: process design trends and integration opportunities. Bioresour Technol 98:2415–2457CrossRefGoogle Scholar
  9. Donohoe BS, Decker SR, Tucker MP, Himmel ME, Vinzant TB (2008) Visualizing lignin coalescence and migration through maize cell walls following thermochemical pretreatment. Biotechnol Bioeng 101:913–925CrossRefGoogle Scholar
  10. Fujii T, Fang X, Inoue H, Murakami K, Sawayama S (2009) Enzymatic hydrolyzing performance of Acremonium cellulolyticus and Trichoderma reesei against three lignocellulosic materials. Biotechnol Biofuels 2:24CrossRefGoogle Scholar
  11. Fujii T, Hoshino T, Inoue H, Yano S (2014) Taxonomic revision of the cellulose-degrading fungus Acremonium cellulolyticus nomen nudum to Talaromyces based on phylogenetic analysis. FEMS Microbiol Lett 351:32–41CrossRefGoogle Scholar
  12. Hendriks ATWM, Zeeman G (2009) Pretreatments to enhance the digestibility of lignocellulosic biomass. Bioresour Technol 100:10–18CrossRefGoogle Scholar
  13. Himmel ME, Ding S-Y, Johnson DK, Adney WS, Nimlos MR, Brady JW, Foust TD (2007) Biomass recalcitrance: engineering plants and enzymes for biofuels production. Science 315:804–807CrossRefGoogle Scholar
  14. Hoeger IC, Filpponen I, Martin-Sampedro R, Johansson LS, Österberg M, Laine J, Kelley S, Rojas OJ (2012) Bicomponent lignocellulose thin films to study the role of surface lignin in cellulolytic reactions. Biomacromolecules 13:3228–3240CrossRefGoogle Scholar
  15. Josefsson P, Henriksson G, Wågberg L (2008) The physical action of cellulases revealed by a quartz crystal microbalance study using ultrathin cellulose films and pure cellulases. Biomacromolecules 9:249–254CrossRefGoogle Scholar
  16. Kumagai A, Lee S-H, Endo T (2013) Thin film of lignocellulosic nanofibrils with different chemical composition for QCM-D study. Biomacromolecules 14:2420–2426CrossRefGoogle Scholar
  17. Lee S-H, Chang F, Inoue S, Endo T (2010) Increase in enzyme accessibility by generation of nanospace in cell wall supramolecular structure. Bioresour Technol 101:7218–7223CrossRefGoogle Scholar
  18. Mansfield SD, Mooney C, Saddler JN (1999) Substrate and enzyme characteristics that limit cellulose hydrolysis. Biotechnol Prog 15:804–816CrossRefGoogle Scholar
  19. Martin-Sampedro R, Filpponen I, Hoeger IC, Zhu J, Laine J, Rojas OJ (2012) Rapid and complete enzyme hydrolysis of lignocellulosic nanofibrils. ACS Macro Lett 1:1321–1325CrossRefGoogle Scholar
  20. Martin-Sampedro R, Rahikainen LR, Johansson LS, Marjamaa K, Laine J, Kruus K, Rojas OJ (2013) Preferential adsorption and activity of monocomponent cellulases on lignocellulose thin films with varying lignin content. Biomacromolecules 14:1231–1239CrossRefGoogle Scholar
  21. Maurer SA, Bedbrook CN, Radke CJ (2012) Competitive sorption kinetics of inhibited endo- and exoglucanases on a model cellulose substrate. Langmuir 28:14598–14608CrossRefGoogle Scholar
  22. Maurer SA, Brady NW, Fajardo NP, Radke CJ (2013) Surface kinetics for cooperative fungal cellulase digestion of cellulose from quartz crystal microgravimetry. J Colloid Interface Sci 394:498–508CrossRefGoogle Scholar
  23. Nitsos CK, Matis KA, Triantafyllidis KS (2013) Optimization of hydrothermal pretreatment of lignocellulosic biomass in the bioethanol production process. Chem Sus Chem 6:110–122CrossRefGoogle Scholar
  24. Olsson A-M, Salmén L (2004) The softening behavior of hemicelluloses related to moisture. In: Gatenholm P, Tenkanen M (eds) Hemicelluloses: science and technology, ACS symposium series 864. American Chemical Society, Washington, DC, pp 184–197Google Scholar
  25. Segal L, Creely JJ, Martin AE, Conrad CM (1959) An empirical method for estimating the degree of crystallinity of native cellulose using the X-ray diffractometer. Text Res J 29:786–794CrossRefGoogle Scholar
  26. Shallom D, Shoham Y (2003) Microbial hemicellulases. Curr Opin Struct Biol 6:219–228Google Scholar
  27. Sluiter A, Hames B, Ruiz R, Scarlata C, Sluiter J, Templeton D (2006) Determination of sugars, byproducts, and degradation products in liquid fraction process samples, laboratory analytical procedure (LAP). National Renewable Laboratory, NREL/TP-510-42623Google Scholar
  28. Sluiter A, Hames B, Ruiz R, Scarlata C, Sluiter J, Templeton D, Crocker D (2011) Determination of structural carbohydrates and lignin in biomass, laboratory analytical procedure (LAP), Version 07-08-2011. National Renewable Laboratory, NREL/TP-510-42618Google Scholar
  29. Somerville C, Bauer S, Brininstool G, Facette M, Hamann T, Milne J, Osborne E, Paredez A, Persson S, Raab T, Vorwerk S, Youngs H (2004) Toward a systems approach to understanding plant-cell walls. Science 306:2206–2211CrossRefGoogle Scholar
  30. Suchy M, Linder MB, Tammelin T, Campbell JM, Vuorinen T, Kontturi E (2011) Quantitative assessment of the enzymatic degradation of amorphous cellulose by using a quartz crystal microbalance with dissipation monitoring. Langmuir 27:8819–8828CrossRefGoogle Scholar
  31. Terashima N, Kitano K, Kojima M, Yoshida M, Yamamoto H, Westermark UJ (2009) Nanostructural assembly of cellulose, hemicellulose, and lignin in the middle layer of secondary wall of ginkgo tracheid. Wood Sci 55:409–416CrossRefGoogle Scholar
  32. Turon X, Rojas OJ, Deinhammer RS (2008) Enzymatic kinetics of cellulose hydrolysis: a QCM-D study. Langmuir 24:3880–3887CrossRefGoogle Scholar
  33. Van Dyk JS, Pletschke BI (2012) A review of lignocellulose bioconversion using enzymatic hydrolysis and synergistic cooperation between enzymes—factors affecting enzymes, conversion and synergy. Biotechnol Adv 30:1458–1480CrossRefGoogle Scholar
  34. Wise LE, Murphy M, D’Addieco AA (1946) Chlorite holocellulose, its fractionation and bearing on summative wood analysis and on studies on the hemicelluloses. Paper Trade J 122:35–43Google Scholar
  35. Yu Y, Lou X, Wu H (2008) Some recent advances in hydrolysis of biomass in hot-compressed water and its comparisons with other hydrolysis methods. Energy Fuels 22:46–60CrossRefGoogle Scholar
  36. Zhu JY, Wang GS, Pan XJ, Gleisner R (2009) Specific surface for evaluating wood size-reduction and pretreatment efficiencies. Chem Eng Sci 64:474–485CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media Dordrecht 2014

Authors and Affiliations

  • Akio Kumagai
    • 1
  • Shinichiro Iwamoto
    • 1
  • Seung-Hwan Lee
    • 1
    • 2
  • Takashi Endo
    • 1
  1. 1.Biomass Refinery Research CenterNational Institute of Advanced Industrial Science and Technology (AIST)HiroshimaJapan
  2. 2.Department of Forest Biomaterials Engineering, College of Forest and Environmental SciencesKangwon National UniversityChuncheonSouth Korea

Personalised recommendations