Self-Assembled Regenerated Silk Fibroin Microsphere-Embedded Fe₃O₄ Magnetic Nanoparticles for Immobilization of Zymolyase

Abstract

Cytoplasm of Saccharomyces cerevisiae yeast cells contains a significant amount of desired intracellular products for both industrial utility and academic research. To recover intracellular compounds, it is necessary to break the yeast cells with high efficiency, which, under certain circumstances, requires the use of the lytic enzyme zymolyase to completely digest the cell walls. A promising strategy for zymolyase immobilization on silk fibroin (SF) was developed. SF/Fe₃O₄ magnetic microspheres (MMs) were constructed by solvent (ethanol)-induced self-assembly of SF surrounding Fe₃O₄ magnetic nanoparticles (MNs), which were synthesized by a coprecipitation method. Zymolyase was covalently bonded on the surface of the SF/Fe₃O₄ MMs by a photochemical cross-linking method to produce robust biocatalysts of zymolyase/SF/Fe₃O₄. The chemical, magnetic, and morphological properties of the MM supports and the immobilized zymolyase were investigated. Enzymolysis results demonstrated that the immobilized zymolyase showed good activity and stability for digesting yeast cell walls, and the biocatalyst can be readily recycled through convenient magnetic separation for reuse. At the optimum pH = 7.5, the immobilized zymolyase maintained 84% of the activity of the free zymolyase and retained 41% of its initial activity after four times of reuse. At unfavorable acidic pH = 4, the immobilized zymolyase retained 81% of its initial activity, while the free zymolyase showed no significant activity. Consequently, the SF/Fe₃O₄ MMs exhibit superior performance in terms of immobilizing enzymes, which have a good prospect in the biological application.

Introduction

Yeast Saccharomyces cerevisiae (S. cerevisiae) has a significant industrial utility. Because the cellular structure and functional organization of S. cerevisiae are similar to those of plant and animal cells, it is one of the most intensively studied unicellular eukaryotic model organisms in biological research.¹ Cytoplasm of S. cerevisiae is a rich source of bioproducts (such as proteins, polysaccharides, and so on) valuable in biotechnology and pharmacology, as well as in the food industry. Recovery of the intracellular products requires efficient disruption of the yeast cells. For example, thanks to its genetic tractability, short generation time, and easy cultivation, S. cerevisiae has been widely used for the expression of recombinant posttranslationally modified proteins. Many recombinant proteins expressed in S. cerevisiae remain intracellular and need to be released by disrupting the yeast cells.² However, S. cerevisiae has a relatively thick and rigid cell wall consisting of β-glucan, mannan, and glycoprotein, which is notoriously difficult to break.³ Various methods have been reported, such as freeze/thawing, high-frequency focused and conventional (ultra)sonication, high-pressure homogenization, osmotic shock, bead milling, and chemical or enzymatic lysis.^2,4 For laboratory-scale cell disruption for intracellular protein recovery, the enzymatic and chemical lysis methods are commonly used. For enzymatic lysis, zymolyase accumulated from Arthrobacter luteus (A. luteus) culture is a mixture of lytic enzymes, including β-1,3-glucanase, mannose, and protease, and can readily degrade viable yeast cell walls at all growth stages under mild reaction conditions.⁵

However, practical applications of native zymolyase are seriously limited by its high sensitivity to environmental conditions, unsatisfactory operational and storage stability, low marginal lifetime, lack of (re)-purification, difficulties in recovery, and reusability.^6,7 Immobilization of enzymes on supports/carriers has been most commonly used to improve the efficiency of enzymes for industrial utilizations. The effect of immobilization depends upon the supports and immobilization procedure.⁸⁻¹¹ Particularly, iron oxide (Fe₃O₄) magnetic nanoparticles (MNs) have attracted a great deal of interest in enzyme immobilization as they provide a facile and effective separation of enzymes from the biocatalytic reaction mixture by use of external magnetic fields and thereby recovery of the enzymes for continual uses.¹² Generally, bare Fe₃O₄ MNs have a tendency to aggregate and are sensitive to oxidation and acidic conditions, which hinder their practical applications. Active groups on Fe₃O₄ MNs are limited, which results in incompetence of naked Fe₃O₄ MNs for direct attachment of enzymes. At this junction, functionalization on the surface of Fe₃O₄ MNs is a promising strategy. A range of natural polymers with various functional groups have been developed to modify Fe₃O₄ MNs.¹³⁻¹⁷

Silk fibroin (SF) derived from Bombyx mori is an abundant sustainable biopolymer with more than 480,000 tons per year produced all over the world. SF exhibits an impressive versatile processability and good ability to be functionalized.¹⁸ Particularly, SF has been investigated in the field of enzyme immobilization.^19,20 In our previous studies, SF-based hydrogels were fabricated for immobilization of several industrially important enzymes; the immobilized enzymes displayed a wide-working pH range, as well as recyclability.²¹⁻²⁵ In addition, several methods have been developed to prepare SF-coated MNs.^26,27 For instance, Chen and co-workers fabricated Fe₂O₃/SF nanoparticles.²⁶ Huang and co-workers fabricated Fe₃O₄/SF nanoparticles with irregular structures.²⁷ Lu and co-workers fabricated core–shell Fe₃O₄/SF microspheres with controllable homogeneous size and water dispersibility by a one-step solvothermal process.²⁸ Tian et al. prepared Fe₃O₄/SF nanoparticles by a one-step potassium phosphate salting-out process.²⁹ These SF-based MNs of many pioneering works have been widely used in biomedical applications such as drug delivery and magnetic resonance imaging (MRI).³⁰ Despite the progress on SF/Fe₃O₄ magnetic materials, their application in enzyme immobilization and development of an efficient method to achieve high performances has a substantial demand.³¹

Herein, we aimed to explore the possibility of covalently immobilizing zymolyase on SF/Fe₃O₄ magnetic microspheres (ZY/SF/Fe₃O₄) for effectively disrupting S. cerevisiae yeast cells. The proposed process is schematically illustrated in Figure 1. First, Fe₃O₄ MNs were synthesized through coprecipitation of ferrous (Fe²⁺) and ferric (Fe³⁺) ions in a basic aqueous medium.^32,33 The obtained Fe₃O₄ MNs can be uniformly dispersed in ethanol and water. Then, through a controllable ethanol-induced interface self-assembly,^26,34 SF embedded the Fe₃O₄ MNs, forming SF/Fe₃O₄ MMs. Subsequently, the abundant tyrosine residues of SF were employed for immobilization of zymolyase through Ru(II)-mediated photochemical cross-linking reactions between the phenyl groups of SF and zymolyase, initiated by ammonium persulfate (APS). Thus, zymolyase was covalently bonded on the surface of the SF/Fe₃O₄ MMs, obtaining the ZY/SF/Fe₃O₄. It is of note that the SF not only served as a matrix for enzyme immobilization but also served as a protective layer for the Fe₃O₄ MNs. Owing to a large amount of serine residues (hydroxyl groups) in SF, SF stabilized the Fe₃O₄ MNs through electrostatic adsorption and hydrogen bonding interactions, preventing aggregation and avoiding corrosion or oxidation in subsequent reaction media. The resultant materials were examined by X-ray powder diffraction (XRD), Fourier transform infrared (FTIR) spectroscopy, dynamic light scattering (DLS), thermogravimetric analysis (TGA), vibrating sample magnetometers (VSM), transmission electron microscopy (TEM), and scanning electron microscopy (SEM) equipped with energy-dispersive spectrometer (EDS). The activity of the immobilized zymolyase on cell disruption was evaluated by enzymolysis assays via measurement of UV–vis spectroscopy. The cell disruption kinetics was also investigated.

Figure 1.

Schematic illustration of zymolyase immobilization on regenerated silk fibroin microspheres embedding Fe₃O₄ magnetic nanoparticles and the mechanism of zymolyase disrupting yeast cell walls.

Results and Discussion

Crystalline Structure Analysis

The XRD patterns of Fe₃O₄, SF/Fe₃O₄, and ZY/SF/Fe₃O₄ (Figure 2a) exhibited the same diffraction peaks, which could be assigned to characteristic peaks of the standard Fe₃O₄ (PDF#85-1436) compiled by the International Center for Diffraction Data (ICDD). The peaks at 2θ values of 30.1, 35.4, 37.1, 43.1, 53.4, 57.0, and 62.6° were indexed as (220), (311), (222), (400), (422), (511), and (440) planes of the phase pure spinel structure of Fe₃O₄, respectively.³⁵ On one hand, the crystalline structure of Fe₃O₄ did not change after being embedded in SF/Fe₃O_4.³¹ On the other hand, new peaks at 2θ = 20.2° ascribed to SF appeared in SF/Fe₃O₄ and ZY/SF/Fe₃O₄, indicating that SF was successfully coated on Fe₃O₄. Specifically, TGA results showed that the contents of Fe₃O₄ in SF/Fe₃O₄ and ZY/SF/Fe₃O₄ were 39.0 ± 1.1 and 46.3 ± 0.1%, respectively (Figure S1).

Figure 2.

Chemical characterization of Fe₃O₄, SF/Fe₃O₄, and ZY/SF/Fe₃O₄. (a) XRD patterns. (b) FTIR spectra. (c) Magnetic hysteresis loops. (d) DLS curves.

Chemical Composition

The FTIR spectra of Fe₃O₄, SF/Fe₃O₄, and ZY/SF/Fe₃O₄ are displayed in Figure 2b. The characteristic peak of Fe₃O₄ was at 605 cm^–1.³⁶ The characteristic peaks of SF were at 1650, 1525, and 1240 cm^–1, belonging to β-sheets, typical secondary structures of regenerated SF.²¹ In the spectra of SF/Fe₃O₄ and ZY/SF/Fe₃O₄, both Fe₃O₄ and SF were observed, also indicating that SF was successfully coated on Fe₃O_4.³⁷ The chemical composition of SF/Fe₃O₄ and ZY/SF/Fe₃O₄ was further confirmed by EDS analysis (Figure S2), showing existence of the elements Fe, O, C, and N. The appearance of C and N signals observed in SF/Fe₃O₄ and ZY/SF/Fe₃O₄ may be originated from SF and the immobilized zymolyase, also indicating that SF was successfully coated onto the surface of Fe₃O₄. EDS mapping analysis also demonstrated homogeneous distribution of Fe and thereby Fe₃O₄ in SF/Fe₃O₄ and ZY/SF/Fe₃O₄.

Magnetic Properties

The magnetic hysteresis loops of Fe₃O₄, SF/Fe₃O₄, and ZY/SF/Fe₃O₄ exhibited low coercivity and small hysteresis (Figure 2c). The saturation magnetization values (at H = 4000 Oe) of Fe₃O₄, SF/Fe₃O₄, and ZY/SF/Fe₃O₄ were 1180, 91.1, and 53.8 Gs, respectively, indicating that all of the samples possessed magnetic responsiveness.³¹ The saturation magnetization decreased after SF coating and zymolyase immobilization. This phenomenon can be explained by the presence of SF layers around Fe₃O₄, as revealed by larger sizes of SF/Fe₃O₄ (∼2157 nm) and ZY/SF/Fe₃O₄ (∼1289 nm) than that of Fe₃O₄ (∼479 nm). It is of note that hydrodynamic sizes determined by DLS (Figure 2d) might differ from particle sizes measured by other techniques such as TEM and SEM. Nevertheless, ZY/SF/Fe₃O₄ could be quickly separated from the reaction mixtures by magnets and rapidly redispersed when the magnetic field disappeared. The sensitive magnetic responsivity and redispersibility implied that ZY/SF/Fe₃O₄ possessed advantageous properties for magnetic manipulation as biocatalysts.

Morphology Analysis

The SEM images of Fe₃O₄, SF/Fe₃O₄, and ZY/SF/Fe₃O₄ are represented in Figure 3a–c, respectively. Fe₃O₄ displayed roughly spherical structures. SF/Fe₃O₄ was larger than Fe₃O₄, suggesting that the Fe₃O₄ MNs were embedded into the SF microsphere. Thanks to the distinct density contrast, TEM images might be easy to differentiate the samples. The TEM images of Fe₃O₄, SF/Fe₃O₄, and ZY/SF/Fe₃O₄ are depicted in Figure 3d–f, respectively. After self-assembly of SF on Fe₃O₄, SF/Fe₃O₄ MMs comprising Fe₃O₄ (black) and thin SF layers (gray) were observed. The images confirmed a uniform fine spherical shape of SF/Fe₃O₄ and ZY/SF/Fe₃O₄. ZY/SF/Fe₃O₄ was denser and larger than SF/Fe₃O₄. The TEM observation was consistent with those of SEM.

Figure 3.

Morphological characterization of Fe₃O₄, SF/Fe₃O₄, and ZY/SF/Fe₃O₄. Representative (a–c) SEM images and (d–f) TEM images.

Immobilization Efficiency

The immobilization capacity and efficiency would be affected by many factors, including reaction time, pH, temperature, concentration of zymolyase, and activation of chemical bonds on both SF and zymolyase. In the present study, zymolyase was immobilized on the SF/Fe₃O₄ MMs via cross-linking of tyrosine residues into dityrosine adducts. The presence of tyrosine residues in both SF²⁵ and zymolyase were confirmed by amino acid analysis (Table 1). Following immobilization of zymolyase on SF/Fe₃O₄, the loading capacity was determined using the classical Bradford protein assay, showing an immobilization efficiency of almost 100% at a loading amount of 100 mg·g^–1. Furthermore, as the immobilization of zymolyase occurs via covalent bonds, no leaching of zymolyase was observed.

Table 1. Amino Acid Analysis of Zymolyase.

amino acids	content (wt %)	amino acids	content (wt %)
CySO3H	0.632	Met	0.831
Tau	0.031	Ile	2.772
Asp	9.687	Leu	5.171
Thr	5.961	Tyr	3.069
Ser	4.118	Phe	3.222
Glu	9.355	g-ABA	0.128
Gly	5.126	Thr-ol	0.776
Ala	6.057	Lys	2.616
Cys	1.445	NH3	30.194
Val	5.210	Arg	3.600

Enzymatic Activity

During lysis of viable yeast cells, turbidity of the reaction mixture decreased with time and became approximately constant after 2 h. At 0.5 h intervals, the number of yeast cells was counted (Figure 4a). In the case of free zymolyase, as the enzyme concentration increased from 0.1 to 1.0 mg·mL^–1, no significant change in either reaction rate or product yield was observed (Figure S3a). As pH influenced the lytic activity considerably, the effect of pH on the free zymolyase was examined (Figure 4b). The optimum pH of the free zymolyase was pH = 7–8. High lytic activities were detected at pH = 9–10. However, only weak lytic activity was observed at pH = 6; almost no lytic activity was detected at pH ≤ 5. Compared with the free enzyme, the immobilized enzyme usually exhibited reduced activity.^12,22,38,39 Thus, a comparative investigation between the free and immobilized zymolyase was conducted in terms of pH in the range of 4–10. At pH = 7.5, the immobilized zymolyase maintained 84% of the activity of the free zymolyase. The immobilized zymolyase maintained a relatively high activity over 4–6. At pH = 4, the immobilized zymolyase retained 81% of its initial activity, which was higher than that of the free zymolyase. The tolerance to unfavorable acidic pH of zymolyase had been significantly enhanced after immobilization.

Figure 4.

Enzymolysis assays on the free and immobilized zymolyase. (a) Time course of lysis of yeast cells. (b) Changes of lytic activities at different pH. (c) Lineweaver–Burk plots for the free and immobilized zymolyase. (d) Reusability of ZY/SF/Fe₃O₄.

Kinetic Parameters

It should be noted that, in our previous studies on enzymes immobilized in SF-based hydrogels, deviation of the kinetic data from a first-order equation to an “S”-shaped curve was observed. The change in the kinetic properties of enzymes was mainly attributed to the mass transfer limitations.²² In the present study, zymolyase immobilized on SF/Fe₃O₄ MMs showed higher activities than that on SF-based hydrogels under the same condition (Figure S3b), yet the enzymolysis data fit well with the first-order kinetic expression. To further understand the zymolyase activity, Michaelis–Menten kinetics of yeast cell disruption was measured at different initial substrate concentrations. The kinetic parameters for the free and immobilized zymolyase were calculated from the Lineweaver–Burk plots (Figure 4c). K_m values for the free and immobilized zymolyase were 1.0 × 10⁵ and 1.4 × 10⁴ cfu·mL^–1, respectively. K_cat/K_m for the free zymolyase was 0.44 L·g^–1·h^–1, whereas that of the immobilized zymolyase was 0.34 L·g^–1·h^–1. The slight decrease might be caused by a possible structural change and conformational hindrance of the active sites after immobilization of zymolyase on SF/Fe₃O₄.^12,38,39 Reusability of the immobilized zymolyase was also evaluated. ZY/SF/Fe₃O₄ could be magnetically separated, recovered, and retained 41% of its initial activity after four consecutive cycles (Figure 4d). This observation demonstrated an improved reusability of ZY/SF/Fe₃O₄, which was crucial and attractive for practical application in digesting yeast cell walls.

Conclusions

SF MM-embedded Fe₃O₄ MNs were prepared as effective magnetic carriers for enzyme immobilization. Zymolyase was used as a model enzyme and was successfully immobilized on SF/Fe₃O₄. The SF/Fe₃O₄ MMs exhibited high immobilization efficiency and a capacity reaching 100 mg·g^–1. Compared with the free zymolyase, the immobilized zymolyase exhibited good biocatalytic activity for disruption of S. cerevisiae cells in a wide range of pH. Moreover, the saturation magnetization value of ZY/SF/Fe₃O₄ was 53.8 Gs, which allowed reuse of the immobilized zymolyase by magnetic treatment. The results validated that the SF MMs are promising platforms for enzyme immobilization with high performance. Through the introduction of Fe₃O₄ MNs and immobilization of zymolyase, this study provides an effective strategy to further improve the performance of SF-based materials.

Experimental Section

Materials

Zymolyase from A. luteus was purchased from Nacalai Tesque, Inc. (Japan). Sodium hydroxide (NaOH; 98%), ferrous chloride (FeCl₂·4H₂O; 98%) and ferric chloride (FeCl₃·6H₂O; 99%) were purchased from Macklin. Ammonium persulfate (APS) and tris(2,2′-bipyridyl)dichlororuthenium(II) hexahydrate ([Ru(bpy)₃]²⁺Cl^–₂·6H₂O) were purchased from Sigma-Aldrich. Analytical reagent grade calcium chloride (CaCl₂), sodium carbonate (Na₂CO₃), absolute ethanol (CH₃CH₂OH), and all other reagents were purchased from Sinopharm Chemical Reagent Beijing Co. Ltd. The water used in all experiments was deionized water unless otherwise stated. All assays were performed in triplicate and data reported as average ± standard deviation.

Yeast Cell Preparations

S. cerevisiae YS58 yeast cells were kindly supplied by Dr. Junfeng Liu (Beijing University of Chemical Technology). Cells were grown batch-wise in medium (glucose 2.0%, peptone 2.0%, and pH 5.5) at 30 °C for 2 days with shaking at 180 rpm. The cells were collected, resuspended in 1/15 M phosphate buffer (pH 7.5), and used as a substrate for the determination of lytic activity.

Measurement of Lytic Activity of Zymolyase

An appropriate amount of enzyme (10 mg·mL^–1, 10 μL) was added to 200 μL of yeast suspension, and the change of absorbance at 620 nm was monitored. The reaction mixture was incubated at 25 °C for 2 h with gentle shaking. The turbidity of the reaction mixture and the number of intact cells in the mixture were determined after different time periods (t). Lysis of yeast cells was expressed as the percentage decrease in optical density (OD) at 620 nm and/or in cell number. In the control experiments, phosphate buffer was used instead of an enzyme solution, and clearing of cytoplasmic contents (cell wall lysis) because of autolysis was measured. The data of the free zymolyase was fitted to first-order rate kinetics to determine the apparent rate constant K_cat/K_m, where K_cat is the turnover number and K_m is the Michaelis constant. The activity of the immobilized zymolyase was assayed using the same method.

1

2

3

4

5

6

Preparation of Regenerated SF and SF-Based Hydrogels

Regenerated SF and chemically cross-linked SF-based hydrogels were prepared as previously described.²¹⁻²⁵ Briefly, raw silk (produced in Zhejiang, China) was degummed in 0.05 wt % Na₂CO₃ at 90 °C for 90 min and rinsed thoroughly with deionized water. The degummed silk was dissolved in a solution of CaCl₂/CH₃CH₂OH/H₂O (with a molar ratio of 1:2:8) at 60 °C for 2 h and then collected by centrifuging at 8000 rpm for 12 min at 4 °C. The supernatant was dialyzed against deionized water using a cellulose dialysis membrane (MWCO 3500 Da) for 3 days and then lyophilized to obtain regenerated SF. SF (40 mg) was solubilized in 186 μL of PBS. APS (10 μL, 1 M) and [Ru(bpy)₃]²⁺Cl^–₂ (4 μL, 20 mM) solutions were then added. The mixture was irradiated for 10 min using a 100 W fiber optical white light source placed 4 cm away. For zymolyase immobilization in the SF-based hydrogels, the procedure was the same except that SF was replaced with a mixture of SF and zymolyase.

Synthesis of Fe₃O₄ Magnetic Nanoparticles

The Fe₃O₄ nanoparticles were prepared by a coprecipitation method.^32,33 An aqueous solution of FeCl₃ was mixed with a hydrochloric acid (HCl) solution of FeCl₂ at Fe³⁺:Fe²⁺ = 2:1. To the mixture, the NaOH solution was added drop by drop with continuous stirring at 40 °C. When pH was adjusted to 10, a black precipitate appeared, which was kept for 30 min. The precipitate was washed three times with distilled water to remove excess NaOH and any impurity. The precipitate was further washed with ethanol and vacuum-dried (50 °C) for further use.

Preparation of Regenerated SF Microsphere-Embedded Fe₃O₄ Magnetic Nanoparticles

The vacuum-dried nanoparticles were redispersed in absolute ethanol. A certain amount of the above dispersion was added dropwise to 10% (w/w) RSF solutions (v/v = 3/20) under continuous gentle stirring at room temperature over 2 min. The mixture was incubated in a refrigerator at −20 °C for 24 h. Then, the frozen sample was defrosted. After centrifugation, followed by washing with water three times and by lyophilizing with a freezing dryer, stable microspheres (denoted as SF/Fe₃O₄) were separated using a magnet.^34,40

Immobilization of Zymolyase on Regenerated SF Microsphere-Embedded Fe₃O₄ Magnetic Nanoparticles

The magnetic SF microspheres (1.8 mg) were dispersed in 164 μL of phosphate buffer solution (1/15 M, pH = 7.5) with dissolved 20 μL of zymolyase (10 mg·mL^–1). Following addition of APS (12 μL, 1 M) and [Ru(bpy)₃]²⁺Cl^–₂ (4 μL, 20 mM), the solution was irradiated with white light. The resultant microspheres (denoted as ZY/SF/Fe₃O₄) were separated by magnetic separation and washed thoroughly with deionized water to remove remaining reagents, while the supernatant was kept to investigate the remaining content of zymolyase by the Bradford protein assay as previously described.²² Immobilization efficiency (IE) was calculated by the following equation

7

Amino Acid Analysis

Zymolyase was completely hydrolyzed in concentrated HNO₃ and used for amino acid analysis (Hitachi L-8900, Japan).

UV–Vis Spectroscopy

The absorbance was measured on a UV–vis spectrophotometer (Shimadzu UV-2600, Japan).

X-Ray Diffraction Analysis

Powder X-ray diffraction analysis (Shimadzu XRD-6000, Japan) was recorded in the range of 5–90° at a scan rate of 5° min^–1 with CuK α radiation (λ = 0.15406 nm).

Thermal Gravimetric Analysis

TGA (TGA/DSC3, METTLER TOLEDO, USA) were measured under nitrogen atmosphere in a temperature range of 20–600 °C. The chemical composition was calculated according to the following equations

8

9

10

Fourier Transform Infrared Spectroscopy

The FTIR spectra were obtained using an FTIR spectrometer (Bruker Optics VERTEX 70, USA) in the transmission mode in the range of 400–4000 cm^–1 with a resolution of 4 cm^–1 at room temperature.

Vibrating Sample Magnetometers

Magnetic properties were investigated by a vibrating sample magnetometer (VSM, BKT-4500, Beijing, China). The magnetic hysteresis curves were recorded at room temperature, in which the magnetism of materials could be characterized by parameters such as the saturation magnetization (M_s), the remanence (M_r), and the coercivity (H_c).

Dynamic Light Scattering

Particle size distribution was measured by DLS (Zetasizer Nano ZS90, Malvern Instruments, UK).

Scanning Electron Microscopy

The hydrogel samples were shock-frozen in liquid nitrogen and lyophilized. The obtained samples were fractured and coated with gold for SEM observation using a scanning electron microscope (Hitachi Regulus8100, Japan).

Transmission Electron Microscopy

For TEM experiments (Hitachi H-800, Japan), samples were dispersed in absolute ethanol and prepared by evaporating a droplet of the dispersion on a Cu grid.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.9b03491.

• Detailed results on TGA, EDS and enzymolysis assays on the free and immobilized zymolyase (PDF)

Author Contributions

S.L. conceived the idea, designed the overall experiments, analyzed the data, and wrote the manuscript. M.X. designed, conducted, and analyzed the data of individual experiments and edited the manuscript. Both authors have given approval to the final version of the manuscript.

The authors declare no competing financial interest.