Coating Novozyme435 with an ionic liquid: more than just a coating for the efficient ring-opening polymerization of δ-valerolactone

Abstract

Novozyme435, which is a successfully commercialized biocatalyst, can efficiently catalyze the ring-opening polymerization (ROP) of lactones. However, for six-membered unsubstituted lactones (δ-valerolactone), enzymatic ROP is difficult to perform and the catalyst is scarcely reused. In this study, an ionic liquid (IL) with different alkyl chain lengths was used to coat Novozyme435 in order to prevent the enzyme inside (Candida antarctica lipase B, CALB) from leaking out quickly. More interestingly, this simple coating can not only preserve CALB but also obviously increase its α-helix content. This change in CALB conformation was favorable for a stronger lid function as well as entry of the substrate towards the active site. The monomer conversion and molecular weight of the resultant poly (δ-valerolactone) were higher when the coated Novozyme435 was used as the catalyst, which presents superior catalytic activity compared to the control. In addition, catalytic activity increased with the IL alkyl chain length and the optimal weight ratio of IL/Novozyme435 was 3 : 1. In particular, the catalytic activity of the coated Novozyme435 in the second and third reaction runs was better than the fresh untreated catalyst, which indicates that the coating strategy is a powerful technique for use in enzyme technology.

---

Introduction

During the recent decades, enzymes have emerged as powerful catalysts for polymerization reactions thanks to their high catalytic activity, enantio- and region-selectivity as well as mild reaction environments.^1–3 Moreover, enzyme-catalyzed strategies have been exploited and proven effective to replace traditional metal-involving catalytic methods, especially for the fabrication of nontoxic materials.^4–6 Candida antarctica lipase B (CALB), which is generally immobilized in porous polyacrylic resin (commercialized Novozyme435) in order to improve its stability and catalytic activity, has been acknowledged as an efficient tool for the ring-opening polymerization (ROP) of lactones.⁷ It can be noted that lactone monomers with different sizes have been investigated as starting materials, which include 7-membered ε-caprolactone,⁸ 9-membered⁹ and 16-membered.¹⁰ Poly(valerolactone) (PVL), which is an aliphatic polyester, can be applied in the medical field, such as surgical sutures, bone and tissue fixation devices, and drug delivery systems, because of its excellent biodegradability, biocompatibility, and bioabsorbability.^11,12 However, due to its relatively low ring tension as well as limited polymerization efficiency, there have been very few reports on the enzymatic ROP of six-membered unsubstituted lactone (δ-valerolactone, δ-VL) to date. Kobayashi et al. described the ROP of δ-VL using three types of lipases as catalysts. Nevertheless, the maximum number-average molecular weight (M_n) was only 2000 g mol⁻¹ for PVL with the catalysis of lipase PF.^13,14 In addition, although PVL with M_n of 3200 g mol⁻¹ was obtained using lipase CC as the catalyst, the polymerization time was as long as 360 h and the enzyme was scarcely reused for a second run after the tedious reaction.¹³ Therefore, overcoming these ROP limitations for special monomers (δ-VL for instance) is a key problem that needs to be urgently solved.

Ionic liquids (ILs) have been frequently used as green reaction media due to their unique properties of non-volatility, non-flammability and good thermal stability.^15–17 However, the utilization of an ionic liquid is not always favorable for biocatalysis, because the stability and activity of enzymes can only be adequately preserved in ionic liquids containing BF₄⁻,¹⁸ PF₆⁻,¹⁹ NTf₂⁻ (ref. 20) and Tf₂N⁻ (ref. 21 and 22) anions rather than Cl⁻, NO₃⁻, CF₃SO₃⁻, CF₃COO⁻ or CH₃COO⁻.²³ Moreover, enzymes pre-treated with ILs exhibit obviously enhanced esterification²⁴ and transesterification catalytic activity.²⁵ However, it is still not known whether the pre-treatment method (simply coating) can contribute to enzymatic ROP. In general, cargoes can be well preserved in carriers by coating a viscous layer on the outside. On the other hand, the hydrophobicity of the carrier is favorable for high loading of enzymes because the hydrophobic interaction is one of the most important driving forces for enzyme adsorption.²⁶ Therefore, the importance of IL coatings, which have different viscosities and hydrophobicities, to enzymatic δ-VL ROP is the most pressing issue investigated in this study.

As is well known, IL cations can control hydrophobicity and viscosity. Herein, three types of ILs containing an identical hexafluorophosphate anion [PF₆]⁻ and different imidazolium cations with varied alkyl lengths, 1-butyl-3-methylimidazolium [Bmim], 1-hexyl-3-methylimidazolium [Hmim] and 1-octyl-3-methylimidazolium [Omim], were selected as coating agents for Novozyme435. The catalytic activity of the treated Novozyme435 was assessed and compared with untreated Novozyme435. In addition, the effect of the amount of coating on enzymatic ROP was also investigated in detail and the reusability of the catalysts was evaluated.

Experimental

Materials

Novozyme435 and Candida antarctica lipase B were purchased from Sigma Co. (USA). δ-VL (98%) was obtained from Alfa Aesar (USA) and distilled under vacuum prior to use. ILs, [Bmim][PF₆], [Hmim][PF₆] and [Omim][PF₆] with purity of 99% were purchased from Linzhou Material Science and Technology Co. (China), whose chemical structures are shown in Fig. 1. Methanol and chloroform were obtained from Kelong Chemical Reagent Factory (Chengdu, China).

Fig. 1 ILs used for the pre-treatment of Novozyme435.

Pre-treatment of Novozyme435

Pre-treatment was performed using different ILs to coat Novozyme435 with different weight ratios (W_IL : W_Novozyme435, from 0 : 1 to 5 : 1). In brief, 0.2 g Novozyme435 was mixed with the IL for 20 min. Furthermore, the IL-coated Novozyme435 (ILCE) was freeze-dried to remove most of the free water, and the amount of residual water was suppressed lower than 1.72%, which was determined using the Karl Fischer titration method.

In addition, to determine the amount of IL effectively coated on Novozyme435, the ILCE samples were immersed in water for 30 min to remove the remaining IL. After filtration and drying under vacuum, the obtained ILCEs were weighed and the coating amount of IL calculated.

ROP of δ-VL

The polymerization of δ-VL was performed using an in situ ROP approach. The detailed procedure used is as follows: ILCE containing 0.2 g Novozyme435 was added in a dried vial, followed by the injection of 2.4 mL of dry δ-VL under an N₂ atmosphere. The reaction then proceeded at 60 °C. To determine the monomer conversion in situ and M_n of PVL for the ROP reaction, a small portion of the reaction mixture was directly dissolved in CDCl₃ prior to ¹H-NMR measurement. A typical ¹H-NMR spectrum is shown in Fig. S1,† in which the methylene group signals of PVL were determined to be 4.08 (t, H_c), 2.35 (m, H_b), 1.67 (t, H_a) and 3.65 (t, H_d), whereas those of the monomer were observed at 4.32(t, H_g), 2.53(t, H_e), 1.86(m, H_f). Therefore, the conversion of monomer can be calculated from the relative intensities of the signals of H_b and H_f (for example) in a single ¹H-NMR spectrum. Furthermore, the catalyst was removed by dissolving the reaction mixture in chloroform and filtering, and then the filtrate was precipitated with methanol and the unreacted monomer was separated from PVL by filtration. The product was dried under vacuum at 50 °C for 24 h prior to further investigation. The untreated Novozyme435 was used as the control to catalyze the polymerization.

Characterization

¹H-NMR spectrum of PVL was obtained on a Bruker spectrometer operating at a frequency of 400 MHz for protons. CDCL₃ was used as the solvent. The molecular weight and polydispersity of PVL were measured using gel permeation chromatography (Waters GPC device equipped with a 1515 pump, a Waters model 717 auto sampler, and a 2414 refractive index detector). Chloroform was used as the eluent at a flowing rate of 1.0 mL min⁻¹, and sample concentration was 2.5 mg mL⁻¹. FT-IR analysis was conducted on a Nicolet FTIR 170SX infrared spectrometer with 64 scans at the resolution of 2 cm⁻¹. Prior to the test, dried CALB was directly mixed with ILs then freeze-dried for 48 h, and the obtained mixture was added to KBr, which was pressed into a pellet under a 10 ton hydraulic press. Circular dichroism (CD) spectroscopy was used to measure the conformation change of CALB before and after treatment with ILs. Lipase was first mixed with IL at the weight ratio of 1 : 3 for 48 h, followed by dissolving them in water with the final CALB concentration of 0.02 mg mL⁻¹. CD spectra were recorded using a JASCO J1500 spectrophotometer (Japan) in an N₂ atmosphere at a scan speed of 100 nm min⁻¹, response time of 1 s, and bandwidth of 1 nm. The percentages of the secondary structures were calculated using an online CD, DICHROWEB (http://www.cryst.bbk.ac.uk/cdweb/html). As a control, CALB was treated the same way prior to the CD test. The morphologies of Novozyme435 and ILCE were examined using a scanning electron microscope (SEM) (JSM-5900LV, JEOL Co. Japan) equipped with an energy-dispersive X-ray spectrometer (EDX, INCA, PENTAFETX3, OXFORD). All samples were fixed on aluminum stubs and coated with gold prior to testing.

Reusability of ILCE

For recycling investigations, the used ILCE or Novozyme435 was carefully separated from the reaction medium by filtration after the 36 h polymerization of δ-VL. The collected catalysts were washed with chloroform three times and dried at ambient temperature for 12 h, followed by re-treating with IL before the next ROP reaction, using the same ratio between monomer and Novozyme435. The catalysts were evaluated continually until their catalytic activity was too poor to be measured. In addition, after each ROP reaction, the weights of the collected catalysts and residual IL on Novozyme435 were measured to further clarify their reusability.

Results and discussion

Three types of ILs, [Omim][PF₆], [Hmim][PF₆] and [Bmim][PF₆], were used to coat Novozyme435 firstly and the pre-treated catalysts were then employed to trigger the enzymatic ROP of δ-VL. As shown in Fig. 2a, the polymerization rates of δ-VL catalyzed by Novozyme435 or ILCE were fast initially and then leveled off after 24 h. After 72 h reaction, the M_n of PVL catalyzed by Novozyme435 was just about 2000 g mol⁻¹, which was close to the previous report (but the reaction time was as long as 120 h),¹⁴ whereas ILCE presented a higher catalytic activity, which was attributed to the better preservation of bound water due to the hydrophobicity of the IL coating on the carrier.²⁷ In addition, it can be noted that M_n of the resultant PVL was dependent on the structure of the IL coating and the longer alkyl chain of the imidazolium cation was favorable for polymerization. In particular, the yield of PVL was only 5.9% using Novozyme435 as the catalyst, whereas it increased to 41.5%, 54.0% and 62.7% when [Bmim][PF₆]-, [Hmim][PF₆]- and [Omim][PF₆]-coated Novozyme435 were used as catalysts, respectively. The dependence of monomer conversion on the reaction time is shown in Fig. 2b. Compared with Novozyme435, the IL coating induced an enormous increase in the ROP rate. During the initiation stage, Novozyme435 nucleophilically attacked δ-VL to form an enzyme activated monomer complex (EAM), which was further attacked by water to produce an ω-hydroxyl carboxylic acid. In the next propagation stage, the formed ω-hydroxyl carboxylic acid then attacked the EAM to generate PVL with a longer chain.^28–31 The polymerization kinetic of the four cases was also investigated, which can be described using the following equation:

ln{([M]₀ − [M]_e)/([M]_t − [M]_e)} = k_appt (1)

where [M]₀ is the initial concentration of δ-VL, [M]_t is the monomer concentration at time t, [M]_e is the equilibrium concentration of δ-VL, and k_app is the apparent rate constant. As shown in Fig. 3, the plot of ln{([M]₀ − [M]_e)/([M]_t − [M]_e)} versus reaction time was linear within 240 min (R² > 0.990 for all cases). This result indicated that the monomer consumption followed the first-order rate law. Compared with the neat Novozyme435, the ILCEs possessed increased k_app values, i.e., from 0.043 min⁻¹ (Novozyme435) to 0.103 min⁻¹ ([Bmim][PF₆]-coated Novozyme435), 0.113 min⁻¹ ([Hmim][PF₆]-coated Novozyme435) and 0.166 min⁻¹ ([Omim][PF₆]-coated Novozyme435).

Fig. 2 Number-average molecular weight of PVL (a) and monomer conversion (b) as a function of time for the ROP of δ-VL catalyzed by the untreated and IL-treated Novozyme435.

Fig. 3 Plots of ln([M]₀ − [M]_e)/([M]_t − [M]_e) versus time for the ROP of δ-VL catalyzed by the untreated and IL-treated Novozyme435.

For the same set of experiments, the dependence of M_n on monomer conversion for the polymerization catalyzed by the untreated and [Omim][PF₆]-treated Novozyme435 is plotted in Fig. 4. Although the monomer conversion as well as M_n of the obtained PVL was higher using [Omim][PF₆]-coated Novozyme435 as the catalyst, a monotonous linear relationship between M_n and monomer conversion was observed, which was nearly the same as that catalyzed by the untreated Novozyme435. The results indicate that no chain transfer reaction occurred and the treatment of Novozyme435 did not change its catalytic mechanism.

Fig. 4 Number average molecular weight as a function of conversion for the ROP of δ-VL catalyzed by Novozyme435 with or without coating by [Omim][PF₆].

In order to determine why the ILCE possessed higher catalytic activity than Novozyme435, the component and morphology of Novozyme435 before and after coating with IL ([Omim][PF₆] for example) were investigated. As shown in Fig. 5, the untreated Novozyme435 presented a coarse sphere-like morphology with the size of about 3 μm and a large number of pores on its surface. Once Novozyme435 was coated with [Omim][PF₆], its surface became smooth and almost no pores were observed. In addition, the elements of F and P, which belong to [Omim][PF6], were determined for Novozyme435 after IL-treatment (Fig. S2 in the ESI†). The results indicate that the IL was mainly located on the surface of Novozyme435. Furthermore, in order to investigate the manner in which the ILs were coated on Novozyme435, methanol was used to wash the IL-coated Novozyme435 several times, and FTIR spectrum of the washed sample was nearly the same with that of Novozyme435 without the coating. Therefore, the ILs were coated on Novozyme435 in a physical manner rather than a chemical manner.

Fig. 5 SEM images of Novozyme435 before (a) and after coating with [Omim][PF₆] (b). Insets are enlarged images (scale bar: 3 μm).

Other than the morphology change of Novozyme435, further investigations were conducted to determine if the IL-treatment could cause a conformation variation in CALB. FT-IR analysis is a well-established technique for protein analysis.^32–34 In general, the conformational sensitive amide I mode of the peptide within the range from 1600 to 1700 cm⁻¹ (mainly attributed to the C=O stretching vibration) is particularly important for the analysis.³⁵ However, the signals of the different secondary structures overlapped with each other in this area, thus the second derivative spectra of CALB before and after IL-treatment were investigated instead (Fig. 6). Compared with the original CALB, the lyophilized CALB possessed a new peak, which corresponded to the β-sheet at 1689 cm⁻¹, and the other peaks that were assigned to the β-sheet at 1686, 1682, 1672 and 1624 cm⁻¹ became stronger. This phenomenon meant that the content of β-sheet increased, which may have resulted from the loss of hydrogen bonding interactions between CALB and water molecules during the lyophilization process.³⁶ Moreover, the peaks at 1611 and 1617 cm⁻¹, which corresponded to a random coil conformation, disappeared and became narrow thus demonstrating the decreased content of this conformation. These conformation changes of CALB upon lyophilization is in good agreement with a previous report.³⁷ Compared to the lyophilization, the IL treatment exhibited a totally different effect on the conformation of CALB. No matter what IL was used to treat CALB, all samples presented different second derivative spectra from the untreated CALB. For example, the peaks of the β-sheet at 1689 and 1693 cm⁻¹ as well as the peak of the β-turn at 1663 cm⁻¹ disappeared completely for the IL-treated CALB, which suggest the decrease in content of the two conformations. In addition, both the signals of the α-helix at 1655 and 1651 cm⁻¹ shifted, which suggests that the content of α-helix also changed. For the random coil conformation of the IL-treated CALB, the corresponding peak at 1617 cm⁻¹ became weak or even disappeared. Therefore, it can be deduced that the conformations of β-sheet, β-turn and random coil probably converted to α-helix.

Fig. 6 Comparison of second derivative FT-IR spectrum of lyophilized CALB with original and IL-treated CALB.

For further quantitative analysis, CD measurement was conducted on native and IL-treated CALB. As seen in Fig. 7, CALB presented a positive peak at 194 nm and a negative peak at 210 nm, whereas IL-treated CALB presented two positive peaks at 193 and 196 nm as well as multi negative peaks from 200 to 240 nm. The CDSSTR method was subsequently used to determine the content of the different conformations. The IL-treatment caused the content of the α-helix conformation of CALB to increase, and this effect became stronger with the increase in the alkyl chain length of the imidazolium cation (see Table 1). CALB is reported to be an α/β hydrolase-like fold enzyme without a typical lid domain, which indicates that it is in an “open” conformation with a restricted entrance to its active site. In general, a short helix in an enzyme with high mobility can act as a lid.³⁸ For CALB, its active site is accessible to an external solvent through a narrow channel formed by helices and a loop region.³⁹ Therefore, it can be speculated that the increase in the α-helix conformation produced a stronger lid function to facilitate an entry to the active site, and the resulting interfacial activation of CALB allowed a more accessible contact of the substrates with this active site, which explained why a relatively rapid ROP rate was achieved by the IL treatment. Based on these findings, the coating strategy is regarded to improve the catalytic activity of CALB by modulating the secondary structure of CALB.

Fig. 7 Far-UV CD spectra of CALB before and after treatment with different ILs.

Table 1 Contents of different conformations of CALB before and after treatment with different ILs

Sample	α-Helix (%)	β-Strand (%)	β-Turns (%)	unordered coil (%)
Control	30	20	21	29
[Bmim][PF6] coated CALB	34	19	19	28
[Hmim][PF6] coated CALB	42	18	17	23
[Omim][PF6] coated CALB	61	8	13	18

---

In order to confirm the optimum coating amount of [Omim][PF₆] on Novozyme435, Novozyme435 coated with different amounts of IL was used to catalyze the ROP. As shown in Table 2, M_n of PVL and monomer conversion increased with the increase in weight ratio of [Omim][PF₆]/Novozyme435 from 0 : 1 to 3 : 1; however, a further increase in the ratio was not favorable to the reaction. To determine the reason for this, the effective coating amount of IL was determined and interestingly found to be about 1.75 g g⁻¹ (Table 2), which could be achieved in the case of the weight ratio of IL versus Novozyme435 of 3 : 1. Therefore, an increase in the ratio above 3 : 1 could not further contribute to protecting the enzyme from leaking out, which just blocked the contact between the substrate and enzyme because of the high viscosity of the IL.¹⁷ Thus, the weight ratio of [Omim][PF₆]/Novozyme435 (3 : 1) was confirmed to be the best preparation condition of coated Novozyme435.

Table 2 ROP of δ-VL at 60 °C catalyzed by [Omim][PF₆] coated-Novozyme435

WIL/WNovozyme435	Monomer conversion (%)	Mn^a (g mol^−1)	Mn^b (g mol^−1)	Mw^b (g mol^−1)	PDI^b	IL amount^c (g g^−1)
a Determined by H-NMR.b Determined by GPC.c Effective amount of IL coated on Novozyme435.
5 : 1	28.9	1.70 × 10^3	2.95 × 10^3	4.54 × 10^3	1.54	1.93
4.5 : 1	25.9	2.53 × 10^3	1.09 × 10^4	1.62 × 10^4	1.49	1.76
4 : 1	26.3	3.06 × 10^3	1.05 × 10^4	1.36 × 10^4	1.29	1.73
3 : 1	62.5	3.42 × 10^3	1.60 × 10^4	1.92 × 10^4	1.20	1.73
1.5 : 1	52.4	2.96 × 10^3	1.16 × 10^4	1.40 × 10^4	1.21	1.50
0 : 1	31.2	2.25 × 10^3	9.33 × 10^3	1.21 × 10^4	1.29	0

---

In general, the reusability of a catalyst is of great importance for its application. In order to investigate whether the IL coating could positively contribute to repeated use of Novozyme435, its catalytic activity was evaluated for several runs. As shown in Table 3, [Omim][PF₆] coated-Novozyme435 could effectively catalyze the ROP of δ-VL with obviously higher monomer conversion and M_n of the resultant PVL at least three times. Although the M_n of PVL decreased when ILCE was used after the first run, which can be explained by the inevitable leakage of CALB from the carrier during the ROP process,⁴⁰ it still showed higher catalytic activity than the freshly used Novozyme435 without the coating. As described before, alkyl imidazolium ILs can form extremely ordered three-dimensional networks, wherein cations and anions are connected together by hydrogen bonds.⁴¹ During the ROP, therefore, the IL coating with a network structure could better preserve CALB, which was proven by the finding that ILCE after catalysis still contained the element of F belonging to [Omim][PF6] (Fig. 8), hence ILCE showed a higher catalytic property than Novozyme435 for the ROP. Furthermore, the recovery rate of the catalysts was investigated, and the recovery rate was determined to be higher than 83% for both Novozyme435 and [Omim][PF₆] coated-Novozyme435 after each reaction run (Table S1 in the ESI†). Nevertheless, it should still be noted that the IL could not be fully retained on the surface of Novozyme435, because the smooth surface resulted from the presence of the IL disappeared after use (Fig. S3 in the ESI†). The amount of IL remaining on Novozyme435 was only about 0.09 g g⁻¹, which was much lower than its initial amount (1.73 g g⁻¹). Therefore, an additional surface coating of IL was needed for the catalyst prior to the next ROP run, and the new amount of IL was still controlled to be about 1.73 g g⁻¹.

Table 3 Reusability of Novozyme435 and [Omim][PF₆] coated-Novozyme435 for ROP of δ-VL

Catalyst	Reaction run	Conversion (%)	Mn^a (g mol^−1)	Mn^b (g mol^−1)	Mw^b (g mol^−1)	PDI^b
a Determined by H-NMR.b Determined by GPC.
[Omim][PF6] coated-Novozyme435	1	62.5	3.42 × 10^3	1.60 × 10^4	1.92 × 10^4	1.20
	2	60.6	2.58 × 10^3	8.04 × 10^3	1.03 × 10^4	1.28
	3	59.5	2.63 × 10^3	8.25 × 10^3	9.91 × 10^3	1.20
Novozyme435	1	31.2	2.25 × 10^3	9.33 × 10^3	1.21 × 10^4	1.29
	2	22.2	1.42 × 10^3	4.30 × 10^3	4.63 × 10^3	1.08
	3	22.8	1.32 × 10^3	3.36 × 10^3	4.11 × 10^3	1.22

---

Fig. 8 EDX spectra of Novozyme435 without (a) and (b) with [Omim][PF₆] coating after catalysis.

Conclusions

ILs with different hydrophobicities were utilized to coat Novozyme435 in order to efficiently catalyze the ROP of δ-VL, which is a six-membered unsubstituted lactone that is hard to polymerize using an enzyme as the catalyst. The IL coating protected CALB from leaking out quickly from the carrier. This was not only favorable for its reusability but also caused an the increased α-helix content of the enzyme, which strengthened its lid function and facilitated an entry toward its active site for substrates. Compared with Novozyme435, the coated catalyst presented superior catalytic activity, and the IL with a longer alkyl chain was favorable for the synthesis of PVL with a higher M_n and monomer conversion. In addition, the coating treatment did not change the catalytic mechanism of δ-VL, which still belongs to the typical chain polymerization without a chain transfer reaction. In particular, Novozyme435 with the IL coating can effectively catalyze the ROP three times, and its catalytic activity was higher than the fresh untreated Novozyme435. This “smart” coating strategy provides a new horizon for enzyme technology.

Acknowledgements

We gratefully acknowledge the National Science Foundation of China (51273123, 51403136, 51421061, J1103315) and the Research Fund for the Doctoral Program of Ministry of Education of China (No. 20120181110049, 20130181120067).

References

1. Y. Yu, D. Wu, C. Liu, Z. Zhao, Y. Yang and Q. Li, Process Biochem., 2012, 47, 1027–1036.
2. R. A. Gross, M. Ganesh and W. Lu, Trends Biotechnol., 2010, 28, 435–443.
3. S. Kobayashi and A. Makino, Chem. Rev., 2009, 109, 5288–5353.
4. A. Kumar, A. S. Kulshrestha, W. Gao and R. A. Gross, Macromolecules, 2003, 36, 8219–8221.
5. J. Zhou, J. Liu, C. J. Cheng, T. R. Patel, C. E. Weller, J. M. Piepmeier, Z. Jiang and W. M. Saltzman, Nat. Mater., 2012, 11, 82–90.
6. C. Vaida, H. Keul and M. Moeller, Green Chem., 2011, 13, 889–899.
7. B. Elsässer, I. Schoenen and G. Fels, ACS Catal., 2013, 3, 1397–1405.
8. A. Kumar and R. A. Gross, Biomacromolecules, 2000, 1, 133–138.
9. S. Kobayashi, H. Uyama, S. Namekawa and H. Hayakawa, Macromolecules, 1998, 31, 5655–5659.
10. K. S. Bisht, L. A. Henderson, R. A. Gross, D. L. Kaplan and G. Swift, Macromolecules, 1997, 30, 2705–2711.
11. N. Kasyapi and A. K. Bhowmick, RSC Adv., 2014, 4, 27439–27451.
12. H. Misaka, R. Kakuchi, C. Zhang, R. Sakai, T. Satoh and T. Kakuchi, Macromolecules, 2009, 42, 5091–5096.
13. S. Kobayashi, K. Takeya, S. Suda and H. Uyama, Macromol. Chem. Phys., 1998, 199, 1729–1736.
14. S. Kobayashi, H. Uyama and S. Namekawa, Polym. Degrad. Stab., 1998, 59, 195–201.
15. M. J. Earle, J. M. Esperança, M. A. Gilea, J. N. C. Lopes, L. P. Rebelo, J. W. Magee, K. R. Seddon and J. A. Widegren, Nature, 2006, 439, 831–834.
16. M. Kosmulski, J. Gustafsson and J. B. Rosenholm, Thermochim. Acta, 2004, 412, 47–53.
17. F. van Rantwijk and R. A. Sheldon, Chem. Rev., 2007, 107, 2757–2785.
18. R. Madeira Lau, F. van Rantwijk, K. Seddon and R. Sheldon, Org. Lett., 2000, 2, 4189–4191.
19. H. Uyama, T. Takamoto and S. Kobayashi, Polym. J., 2002, 34, 94–96.
20. T. de Diego, P. Lozano, S. Gmouh, M. Vaultier and J. L. Iborra, Biomacromolecules, 2005, 6, 1457–1464.
21. R. Marcilla, M. de Geus, D. Mecerreyes, C. J. Duxbury, C. E. Koning and A. Heise, Eur. Polym. J., 2006, 42, 1215–1221.
22. J. T. Gorke, K. Okrasa, A. Louwagie, R. J. Kazlauskas and F. Srienc, J. Biotechnol., 2007, 132, 306–313.
23. J. L. Kaar, A. M. Jesionowski, J. A. Berberich, R. Moulton and A. J. Russell, J. Am. Chem. Soc., 2003, 125, 4125–4131.
24. J. Mutschler, T. Rausis, J.-M. Bourgeois, C. Bastian, D. Zufferey, I. V. Mohrenz and F. Fischer, Green Chem., 2009, 11, 1793–1800.
25. J. K. Lee and M.-J. Kim, J. Mol. Catal. B: Enzym., 2011, 68, 275–278.
26. L.-C. Sang and M.-O. Coppens, Phys. Chem. Chem. Phys., 2011, 13, 6689–6698.
27. L. Ropel, L. S. Belvèze, S. N. Aki, M. A. Stadtherr and J. F. Brennecke, Green Chem., 2005, 7, 83–90.
28. Y. Mei, A. Kumar and R. Gross, Macromolecules, 2003, 36, 5530–5536.
29. R. K. Srivastava and A. C. Albertsson, J. Polym. Sci., Part A: Polym. Chem., 2005, 43, 4206–4216.
30. K. S. Bisht, Y. Y. Svirkin, L. A. Henderson, R. A. Gross, D. L. Kaplan and G. Swift, Macromolecules, 1997, 30, 7735–7742.
31. H. Nishida, M. Yamashita, T. Endo and Y. Tokiwa, Macromolecules, 2000, 33, 6982–6986.
32. F. Dousseau and M. Pezolet, Biochemistry, 1990, 29, 8771–8779.
33. W. K. Surewicz, H. H. Mantsch and D. Chapman, Biochemistry, 1993, 32, 389–394.
34. A. Dong, J. D. Meyer, B. S. Kendrick, M. C. Manning and J. F. Carpenter, Arch. Biochem. Biophys., 1996, 334, 406–414.
35. R. Lau and F. Rantwijk, Green Chem., 2004, 6, 483–487.
36. G. Vecchio, F. Zambianchi, P. Zacchetti, F. Secundo and G. Carrea, Biotechnol. Bioeng., 1999, 64, 545–551.
37. H. R. Costantino, K. Griebenow, P. Mishra, R. Langer and A. M. Klibanov, Biochim. Biophys. Acta, Protein Struct. Mol. Enzymol., 1995, 1253, 69–74.
38. J. Uppenberg, M. T. Hansen, S. Patkar and T. A. Jones, Structure, 1994, 2, 293–308.
39. J. Uppenberg, N. Oehrner, M. Norin, K. Hult, G. J. Kleywegt, S. Patkar, V. Waagen, T. Anthonsen and T. A. Jones, Biochemistry, 1995, 34, 16838–16851.
40. B. Chen, J. Hu, E. M. Miller, W. Xie, M. Cai and R. A. Gross, Biomacromolecules, 2008, 9, 463–471.
41. J. Dupont, J. Braz. Chem. Soc., 2004, 15, 341–350.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra10227h

---