Enzymatic synthesis of methyl β-d-glucoside directly from cellulose pretreated with biocompatible amino acid ionic liquid/cosolvent

Enzymatic synthesis of methyl β-d-glucoside from cellulose:

Cellulose (18 mg) was mixed with IL/cosolvent (1:1, w/w, 180 mg) in a screw-capped vial, and stirred magnetically at 30°C for 30 min. Distilled water (1.4 ml) was added with vigorous stirring for 30 min. The suspension was centrifuged at 4000 rpm for 20 min and then the regenerated cellulose was washed with water (1.4 ml) three times to remove the residual IL and cosolvent. Acetate buffer (0.1 M, pH 4.5) and 13.5%–27% MeOH (v/v) were added to a total volume of 2 ml. The enzymatic synthesis was performed with agitation (200 rpm) at 50°C by adding 2.1 mg of cellulase (with a protein content of 0.5 mg). At given time intervals, 50 μl of the reaction solution was withdrawn, and heated at 105°C for 5 min to inactivate cellulase. All experiments were conducted in triplicate.

Quantification of methyl β-d-glucoside, glucose, and cellobiose:

Methyl β-d-glucoside, glucose, and cellobiose were monitored by high-performance liquid chromatography (HPLC) equipped with a refractive index (RI) detector (JASCO, Tokyo, Japan) and an Asahipak NH2P-40 3E column [3.0 mm internal diameter (ID)×250 mm length (L)] from Shodex, Tokyo, Japan. Acetonitrile/water (65:35, v/v) served as the mobile phase (0.32 ml min−¹). The retention times of methyl β-d-glucoside, glucose, and cellobiose were 6.37 (±0.31%), 10.08 (±0.39%), and 12.39 min (±0.29%), respectively. Their yields were determined by the following equations:

Yglucoside (%) = (Cglucoside × 162/194)/Ccell × 100%,

Yglc (%) = (Cglc × 162/180)/Ccell × 100%,

Ycellobiose (%) = (Ccellobiose × 162/171)/Ccell × 100%,

where Y_glucoside, Y_glc, and Y_cellobiose are the yields of methyl β-d-glucoside, glucose, and cellobiose, respectively, and C_glucoside, C_cell, C_glc, and C_cellobiose are the concentrations of the indicated compounds in mg ml−¹.

Preparative-scale synthesis of methyl β-d-glucoside:

Preparative-scale synthesis of methyl β-d-glucoside was established under conditions optimized through the small-scale experiments. The amount of enzyme was decreased to one-third of that in the small-scale synthesis. Cellulose (1 g) was pretreated with 10 g of [TBP][Gly]/dimethyl sulfoxide (DMSO) (1:1, w/w), and then distilled water (100 ml) was added with vigorous stirring to regenerate cellulose. After vacuum filtration, the regenerated cellulose was washed with water (100 ml) three times to remove the residual [TBP][Gly]/DMSO. The subsequent enzymatic treatment of the regenerated cellulose was performed in 22.5% of MeOH/buffer solution (35 ml) by adding cellulase (39.1 mg). The formation of methyl β-d-glucoside was monitored by HPLC. After 48 h, the solution was filtrated through a G5 sintered glass filter to remove unreacted cellulose. The filtrate was concentrated in a rotary evaporator under reduced pressure at 50°C. The product was purified on a silica gel column (silica gel 60 N, 63–210 μm, 60 g, 18 cm×3.2 cm) with ethyl acetate/2-propanol/water (16:3:1, v/v/v) to give methyl β-d-glucoside (397 mg, 33%). The preparative-scale synthesis was repeated twice.

The ¹H nuclear magnetic resonance (NMR) spectrum of methyl β-d-glucoside was recorded on a Bruker AVANCE II 400 FT-NMR (400 MHz) spectrometer. ¹H NMR (D₂O) δ 3.04 (1H, dd, J=8.0, 9.2 Hz), 3.16 (1H, t, J=9.2 Hz), 3.22–3.27 (1H, m), 3.27 (1H, t, J=9.2 Hz), 3.36 (3H, s, OCH₃), 3.51 (1H, dd, J=5.9, 12.3 Hz), 3.71 (1H, dd, J=2.2, 12.3 Hz), 4.16 (1H, d, J=8.0 Hz).

Enzymatic synthesis of methyl β-d-glucoside

It was already demonstrated that TBP-based AAIL/cosolvent is a new effective pretreatment solvent for enhanced enzymatic hydrolysis of cellulose because of its higher compatibility with cellulase (Tao et al. 2016b). Hence, phosphonium-based AAIL/cosolvent systems were further applied to the synthesis of methyl β-d-glucoside from regenerated cellulose and MeOH. Cellulose was pretreated with combinations of [TBP][Gly] with cosolvent, DMSO, N-methylimidazole (NMI), or N-methyl-2-pyrrolidone (NMP), as illustrated in Figure 1. For comparison, 1-allyl-3-methylimidazolium chloride ([Amim]Cl)/DMSO served as a conventional IL. The MeOH content was 22.5% (v/v) in acetate buffer (pH 4.5), which corresponds to a substrate (glucose residue)-to-MeOH ratio of 1:125.

Figure 1:

Effects of cellulose pretreatments on the yields of (a) methyl β-d-glucoside, (b) glucose, and (c) cellobiose by the enzymatic conversion of regenerated cellulose at 50°C in acetate buffer (pH 4.5) with 22.5% (v/v) methanol.

Error bars show the standard deviation (SD) of triplicate runs.

From the original cellulose (without pretreatment), methyl β-d-glucoside was not formed effectively (Figure 1a), along with quite low yields of glucose (Figure 1b) and cellobiose (Figure 1c). A total enzymatic conversion of only 6% was observed in the presence of MeOH, whereas an 80% glucose yield was attained in 48 h for the enzymatic hydrolysis of untreated cellulose under the same conditions but without MeOH (Tao et al. 2016b). These results indicate that MeOH reduces the enzymatic activity of cellulase and diminishes the total conversion of cellulose. The harmful effects of organic solvents on enzyme activities are well known (Castro and Knuborets 2003; Fernández-Lucas et al. 2012).

In contrast, to the best of our knowledge, this is the first study to demonstrate that methyl β-d-glucoside can be efficiently synthesized from regenerated celluloses pretreated with IL/cosolvent systems. With [TBP][Gly]/DMSO, [TBP][Gly]/NMI, and [TBP][Gly]/NMP, the yields of methyl β-d-glucoside reached 37%–40% accompanied by glucose yields of 46%–48%. Total conversion rates reached 85%–86%, which were much higher than those of untreated cellulose because hydrogen bonding of the original cellulose was altered by the IL/cosolvent pretreatments and the disordered cellulose structure was more accessible to cellulase (Cui et al. 2014; Mai et al. 2014). In addition, pretreatments with [TBP][Gly]/cosolvent were more efficient, and the yields of methyl β-d-glucoside were approx. 10% higher than those by [Amim]Cl/DMSO, which can be attributed to the higher compatibility of residual [TBP][Gly] with cellulase than that of residual [Amim]Cl in regenerated celluloses (Tao et al. 2016b). It is also notable that among the cosolvents observed, DMSO was the most effective one for the synthesis of methyl β-d-glucoside.

Effect of MeOH contents on the synthesis

Methanol was the main reactant in the enzymatic synthesis in focus. However, as pointed out, MeOH has clear harmful effects on cellulase activity. Hence, it is essential to optimize the MeOH content for the enzymatic synthesis of methyl β-d-glucoside. Cellulose pretreated by [TBP][Gly]/DMSO was washed with water and subjected to enzymatic treatments in acetate buffer with 13.5%–27% (v/v) MeOH (Figure 2). The yield of methyl β-d-glucoside clearly increased with increasing MeOH content from 13.5% to 22.5%, and a maximum methyl β-d-glucoside yield of 40% was observed upon incubation for 24 h (Figure 2a). Higher MeOH contents led to more interaction with cellulose, as reflected by the higher methyl β-d-glucoside yield. However, a significant decrease in the yield of methyl β-d-glucoside was observed for 27% (v/v) MeOH. In contrast, the glucose yields showed a reverse tendency and were highest for 13.5% MeOH. The total conversion rates of regenerated cellulose were relatively constant between 85% and 95% for MeOH contents of 13.5%–22.5%. For MeOH contents of 13.5% and 18%, the formation of methyl β-d-glucoside appeared to be completed in 12 h, and a decrease in the yield of methyl β-d-glucoside was observed, as shown in Figure 2a. This indicates that with prolonged incubation, the hydrolysis of methyl β-d-glucoside to glucose became pronounced, particularly at low MeOH contents.

Figure 2:

Effects of methanol contents on the yields of (a) methyl β-d-glucoside, (b) glucose, and (c) cellobiose by the enzymatic conversion of regenerated cellulose pretreated with [TBP][Gly]/DMSO.

The conversion was conducted at 50°C in acetate buffer (pH 4.5) with 13.5%–27.0% (v/v) methanol. Error bars show the SD of triplicate runs.

Exploration of the formation of methyl β-d-glucoside

It was particularly significant to determine how methyl β-d-glucoside is formed by cellulase from insoluble cellulose, because this is the first study with this regard. In light of the synthesis of alkyl β-d-glucoside with almond β-glucosidase and soluble glucose (Papanikolaou 2001; Ducret et al. 2002), we estimated that glucose formed from cellulose might act as a glycosyl donor. Thus, regenerated cellulose was first enzymatically hydrolyzed by cellulase to glucose for 4 h, and then MeOH was added and the incubation was continued for an additional 44 h, as shown in Figure 3. Surprisingly, a high glucose level (83% at 4 h) was maintained, with only a slight decrease during incubation, while only a limited amount of methyl β-d-glucoside was formed. Thus, the glucose formed from cellulose was not a main precursor for the reaction in focus.

Figure 3:

Yields of glucose, cellobiose, and methyl β-d-glucoside by the enzymatic conversion of regenerated cellulose from [TBP][Gly]/DMSO.

The conversion was conducted at 50°C in acetate buffer (pH 4.5). Methanol (22.5%, v/v) was added after 4 h of incubation. Error bars show the SD of triplicate runs.

Next, cellobiose and glucose, which are the main products from the saccharification of cellulose, were selected as substrates for the synthesis under the same incubation conditions. As shown in Figure 4, the formation of methyl β-d-glucoside from cellobiose followed a time course similar to that from regenerated cellulose, associated with a higher initial formation rate and higher yields. In contrast, the rate of conversion of glucose into methyl β-d-glucoside was quite low, particularly during the first 4 h. These results are interpreted as cellulose being first enzymatically hydrolyzed to cellobiose and then the formed cellobiose being converted into methyl β-d-glucoside by transglycosylation (Figure 5). The transglycosylation of cellobiose would produce both methyl β-d-glucoside and glucose at a 1:1 ratio. At the same time, cellobiose was also simply hydrolyzed to glucose in this highly aqueous solution. This explains why, despite a high total conversion of regenerated cellulose (>85%), methyl β-d-glucoside was obtained at a relatively lower yield (<44%). Further research is necessary to achieve a higher conversion rate of cellulose to methyl β-d-glucoside.

Figure 4:

Effects of substrates on the yield of methyl β-d-glucoside. The enzymatic conversions were conducted at 50°C in acetate buffer (pH 4.5) with 22.5% (v/v) methanol using regenerated cellulose from [TBP][Gly]/DMSO, cellobiose, and glucose. Error bars show the SD of triplicate runs.

Figure 5:

Plausible mechanism for the enzymatic synthesis of methyl β-d-glucoside from cellulose in the presence of methanol.

Preparative-scale synthesis of methyl β-d-glucoside from cellulose

Preparative-scale synthesis of methyl β-d-glucoside was performed under the same conditions as in the case of analytical experiments, but the amount of cellulase was reduced. Compared with the small-scale experiments, the initial production rates for both methyl β-d-glucoside and glucose clearly decreased because of the small amount of cellulase, as shown in Figure 6. However, the yields of methyl β-d-glucoside and glucose after 48 h of incubation were 36% and 46%, respectively, which were comparable to the yields for the small-scale synthesis, namely, 40% and 46%, respectively. Methyl β-d-glucoside could be isolated by column chromatography at a 33% isolated yield.

Figure 6:

Preparative-scale synthesis of methyl β-d-glucoside from regenerated cellulose from [TBP][Gly]/DMSO.

Time curves of the formation of methyl β-d-glucoside, glucose, and cellobiose during incubation at 50°C in acetate buffer (pH 4.5) with 22.5% (v/v) methanol are shown. Error bars show the SD of duplicated runs.