English

• Enzymes have been widely used in biopharmaceuticals, food processing, fine chemicals, biofuels and other fields [1-3]. However, enzymes are susceptible to inactivation due to the factors such as temperature and pH, and are difficult to recycle and continuously produce [4]. Immobilization of enzymes can not only improve their stability and reduce enzyme inactivation, but also improve their reuse efficiency. The traditional methods of enzyme immobilization include adsorption [5], embedding [6], covalent bonding [7,8] and cross-linking [9,10]. Adsorption has little damage to the active conformation of enzyme, but the adsorbed enzyme is prone to detachment from support. Embedding enzyme into the interior of polymer network structure can firmly immobilize enzyme molecules and retain its structure, but the mass transfer resistance of substrate entering into the support is large, reducing the apparent activity of enzyme [11,12]. The covalent method can firmly immobilize enzyme on the surface of support and reduce the mass transfer resistance, but the violent reaction is easy to destroy enzyme conformation and decrease its activity [13]. In response to the drawbacks of traditional methods, many new support materials [14-16] and new immobilization strategies have continuously emerged [17,18]. Among these materials, metal organic framework (MOF) and magnetic MOF nanopheres have many advantages [19,20], such as super-paramagnetism, large surface area, and surface hydrophobicity, which facilitates the magnetic separation, high enzyme loading and lipase activation.

  This study used a silica membrane with mesoporous structure to encapsulate enzyme molecules adsorbed on the surface of support, which combined the advantages of adsorption and embedding. This immobilization strategy could not only solve the problems of enzyme leakage and substrate mass transfer, but also retain the catalytic activity of enzyme by preventing enzyme from chemical reaction. Besides, the encapsulation with silica membrane could protect the active conformation of enzyme and make it more stable. We first prepared zeolitic imidazolate framework-8 (ZIF-8) magnetic nanospheres and adsorbed lipase onto the surface of nanospheres, and then coated a SiO₂ nano-membrane with mesopores on the surface of nanospheres by hydrolyzing and condensing. Consequently, lipase was encapsulated on the surface of nanospheres, as shown in Fig. 1.

  Figure 1

  

  Figure 1.  Scheme of lipase encapsulated by silica nano-membrane on the surface of magnetic ZIF-8.

  DownLoad: Full-Size Img PowerPoint

  Fig. 2a shows the morphology of Fe₃O₄ magnetic particles observed under transmission electron microscopy (TEM), and their particle sizes are statistically analyzed. It can be seen that the resulting Fe₃O₄ magnetic particles have good monodispersity (particle dispersity index (PDI) = 0.381), with a concentrated particle size distribution at 9.5 nm. Fig. 2b shows the TEM morphology and particle size distribution of Fe₃O₄@ZIF-8 magnetic nanospheres. As observed, the size of magnetic particles significantly increases after being coated with ZIF-8, and Fe₃O₄@ZIF-8 is spherical and has good dispersibility (PDI = 0.302), with a concentrated particle size distribution at 176.8 nm. Fig. 2c shows the morphology and particle size distribution of the immobilized lipase with silica membrane encapsulation (Fe₃O₄@ZIF-8@CRL@SiO₂; CRL, Candida rugosa lipase). It can be seen that after adsorbing lipase on Fe₃O₄@ZIF-8 and subsequently encapsulating it with SiO₂ nano-membrane, the particle size of magnetic nanospheres further increases to 190.1 nm, indicating the successful coating with a nano-thickness silica membrane. Fig. 2d shows the hysteresis curve of Fe₃O₄@ZIF-8@CRL@SiO₂, displaying typical superparamagnetism with a saturation magnetization of 19.18 emu/g. As seen from the insert, Fe₃O₄@ZIF-8@CRL@SiO₂ has good dispersibility in aqueous solution and can achieve rapid and effective separation under magnetic field.

  Figure 2

  

  Figure 2.  TEM image and the particle size distribution of (a) Fe₃O₄, (b) Fe₃O₄@ZIF-8 and (c) Fe₃O₄@ZIF-8@CRL@SiO₂. (d) Vibrating sample magnetic strength (VSM) of Fe₃O₄@ZIF-8@CRL@SiO₂.

  DownLoad: Full-Size Img PowerPoint

  In order to effectively form the SiO₂ membrane on nanospheres by tetraethyl orthosilicate (TEOS) polymerization, 3-aminopropyltriethoxysilane (APTES) was adsorbed on Fe₃O₄@ZIF-8@CRL in advance before adding TEOS by the electrostatic interaction between the positively charged amino group of APTES and the negatively charged lipase (pI = 4.4). Fig. 3a shows the infrared absorption spectra of Fe₃O₄@ZIF-8@CRL and Fe₃O₄@ZIF-8@CRL@SiO₂. In the spectrum of Fe₃O₄@ZIF-8@CRL, the peaks at 588 and 420 cm⁻¹ respectively correspond to the stretching vibration absorptions of the Fe-O bond and Zn-N bond, the sharp peaks at 757 and 1307 cm⁻¹ belong to the bending vibration peak of imidazole ring in ZIF-8, and the peaks at 1421 and 1456 cm⁻¹ belong to the stretching vibration peaks of imidazole ring. In the spectrum of Fe₃O₄@ZIF-8@CRL@SiO₂, a strong and wide anti symmetric stretching vibration absorption peak of Si-O-Si appears at 1068 cm⁻¹. Moreover, a symmetric stretching vibration absorption peak of Si-O bond appears at 804 cm⁻¹, and a bending vibration absorption peak of Si-O bond appears around 449 cm⁻¹. These peaks imply the success of SiO₂ nano-membrane coating.

  Figure 3

  

  Figure 3.  (a) Infrared spectra of Fe₃O₄@ZIF-8@CRL and Fe₃O₄@ZIF-8@CRL@SiO₂. (b) XRD of Fe₃O₄, Fe₃O₄@ZIF-8 and Fe₃O₄@ZIF-8@CRL@SiO₂. (c) EDS, (d) TEM mapping, (e) water contact angle, and (f) fluorescence characterization of Fe₃O₄@ZIF-8@CRL@SiO₂.

  DownLoad: Full-Size Img PowerPoint

  Fig. 3b shows the X-ray diffraction (XRD) images of Fe₃O₄ (black curve), Fe₃O₄@ZIF-8 (red curve) and Fe₃O₄@ZIF-8@CRL@SiO₂ (purple curve). The black curve exhibits characteristic peaks of Fe₃O₄ particles at 30.2°, 35.4°, 43.1°, 53.7°, 57.1° and 62.8°, corresponding to the diffraction crystal planes of Fe₃O₄ at (220), (311), (400), (422), (511) and (440), respectively. In the red curve, it appears the characteristic peaks of ZIF-8 at 7.4°, 10.5°, 12.9°, 14.8°, 16.6° and 18.1°, corresponding to the diffraction crystal planes of ZIF-8 at (011), (002), (112), (022), (013), and (222), respectively. In the purple curve, no obvious SiO₂ characteristic peaks appear because of the amorphous structure of silica membrane. But compared with the curves of Fe₃O₄ and Fe₃O₄@ZIF-8, it can be seen that the diffraction peak intensity of ZIF-8 and Fe₃O₄ in Fe₃O₄@ZIF-8@CRL@SiO₂ is greatly cut down due to SiO₂ coating.

  The energy dispersive spectrum (EDS) and TEM mapping of Fe₃O₄@ZIF-8@CRL@SiO₂ are shown in Figs. 3c and d, respectively. In Fig. 3c, The EDS spectrum analysis shows that the main components of immobilized lipase are Fe, Zn, S and Si, wherein Fe and O are from Fe₃O₄, Zn is from ZIF-8, S is from the cysteine residues of lipase, Si is from SiO₂ membrane, and Cu is from the copper mesh used for TEM. From the TEM mapping scan of Fe₃O₄@ZIF-8@CRL@SiO₂ in Fig. 3d, it can be seen that the elements distribution order is Fe, Zn, S, and Si from the inside out. It can also be seen that the immobilized lipase is spherical in shape, and Fe element is distributed inside the central area, indicating that Fe₃O₄ nano particles are located at the core of nanospheres. There is a distribution of Zn element outside the Fe element, indicating the coat of ZIF-8 outside Fe₃O₄ nano particles. The content of S element is rich and evenly distributed on the surface of nanospheres, suggesting that a large number of lipase molecules are immobilized on the nanospheres. The TEM mapping results also showed that Si element is uniformly distributed on the surface of nanospheres, further confirming the coating of SiO₂.

  In general, the surface of enzyme molecules is bound to a certain number of water molecules referred to as the "essential hydration layer", which is necessary for the enzyme to maintain its activity [21,22]. Besides, CRL is a special enzyme with interfacial activity, and a hydrophobic interface facilitates the opening of the "active lid" and activates the catalytic activity of lipase [23]. After immobilizing lipase onto the carrier, the hydrophobic carrier surface will affect the "hydration layer" of lipase and its activity; On the other hand, the hydrophobic carrier surface can also affect the opening and closing of lipase's "active lid". Fig. 3e shows the hydrophilicity of immobilized enzyme. As can be seen, the contact angle is about 57°, indicating that the surface of Fe₃O₄@ZIF-8@CRL@SiO₂ has good hydrophilicity. Besides, the water contact angle of Fe₃O₄@ZIF-8 is greater than 90° as shown in Fig. S1 (Supporting information), indicating a hydrophobic surface. Due to the fact that the one end of lipase is in contact with hydrophobic ZIF-8 [24] and the other end is in contact with hydrophilic SiO₂, this amphiphilic carrier can not only improve enzyme activity by maintaining the "hydration layer", but also activate lipase activity by opening the lid of lipase.

  Fe₃O₄@ZIF-8@CRL@SiO₂ was prepared using fluorescence labeling lipase and further observed under a fluorescence microscope. As can be seen from Fig. 3f, the surface of nanospheres emitted strong green fluorescence, indicating the successful immobilization of lipase on nanospheres.

  On the surface of Fe₃O₄@ZIF-8@CRL@SiO₂, there are many pore structures in the SiO₂ nano-membrane. In order to evaluate the pore size, the changes in specific surface area and pore size distribution were obtained by comparing the nitrogen adsorption/desorption isotherm of Fe₃O₄@ZIF-8 before and after silica membrane encapsulation (Fig. S2 in Supporting information). In Fig. S2a, The specific surface area of Fe₃O₄@ZIF-8 is 436.4 m²/g. There is no hysteresis loop in the isotherm curve, indicating that there is no mesoporous structure on the surface of Fe₃O₄@ZIF-8. In Fig. S2b, the specific surface area of Fe₃O₄@ZIF-8@CRL@SiO₂ is 216.2 m²/g, which is smaller than that of Fe₃O₄@ZIF-8 because of the larger particle size. Particularly, there is a significant hysteresis loop in Fig. S2b, indicating the presence of layered structures including crack pores on the surface of nanospheres. This result confirms that there are numerous mesoporous structures on the surface of Fe₃O₄@ZIF-8@CRL@SiO₂, with pore sizes concentrated at 3.9 nm as shown in Fig. S2b. To be satisfactory, the pore size of silica membrane (3.9 nm) is smaller than the size of CRL lipase molecule (5.0 × 4.2 × 3.3 nm³) [25,26], suggesting that mesoporous silica membrane can firmly lock lipase molecules on the surface of Fe₃O₄@ZIF-8 and prevent them from falling off. Moreover, the mesopores on SiO₂ nano-membrane is beneficial for improving the accessibility of lipase, and thus improving the apparent activity of immobilized enzyme.

  To study whether substrate molecules can smoothly diffuse through the mesoporous structures (3.9 nm), we simulated the process of ethanol/palmitic acid diffusion through the mesopores of silica membrane using lipase-catalyzed esterification of palmitic acid with ethanol. Fig. S3a (Supporting information) shows the diffusion results of ethanol/palmitic acid in the longitudinal direction inside the pores at different time (0, 0.5, 1, 3, 4, 5 ns). It can be seen that there is no significant diffusion of ethanol and palmitic acid at 0.1 ns. As time goes on to 5 ns, ethanol and palmitic acid have diffused into the pores. Fig. S3b (Supporting information) shows the diffusion results of ethanol/palmitic acid in the lateral direction inside the pores. At 0 ns, it can be seen that ethanol and palmitic acid are located on the surface of pores and have not yet diffused into the pores. At 5 ns, ethanol and palmitic acid have significantly shifted towards the interior of pores. The simulation results show that the diffusion rate of substrate molecules is fast, indicating small mass transfer resistance. Therefore, the SiO₂ nano-membrane does not affect the contact of enzyme with substrates, which helps to improve the apparent catalytic activity of immobilized enzyme.

  Fig. S4a (Supporting information) shows the thermogravimetric curve (TGA) of Fe₃O₄, Fe₃O₄@ZIF-8, Fe₃O₄@ZIF-8@CRL and Fe₃O₄@ZIF-8@CRL@SiO₂. As observed, the total weight loss ratio of Fe₃O₄ nanoparticles modified with sodium citrate is 27.28%, including water loss and sodium citrate decomposition. The total weight loss ratio of Fe₃O₄@ZIF-8 magnetic nanospheres is 47.49%. And the reason for rapid weight loss above 580 ℃ is due to the decomposition of imidazole bridges in the metal-organic framework, which results in the collapse of crystal structure [27]. The total weight loss ratio of Fe₃O₄@ZIF-8@CRL is 50.66%. The lipase molecules adsorbed on the surface of nanospheres decomposes at high temperature, resulting in a higher weight loss ratio than Fe₃O₄@ZIF-8. And the difference value in total weight loss ratio (3.17%) between Fe₃O₄@ZIF-8 and Fe₃O₄@ZIF-8@CRL is due to the enzyme loading, which is roughly consistent with the enzyme loading (36.48 mg/g) determined by measuring the enzyme concentration changes in the supernatant before and after enzyme immobilization. Besides, the total weight loss ratio of Fe₃O₄@ZIF-8@CRL@SiO₂ is 49.08%.

  To investigate the binding strength between enzyme and carrier, Fe₃O₄@ZIF-8@CRL and Fe₃O₄@ZIF-8@CRL@SiO₂ were eluted/desorbed with a high concentration salt solution, and the results were compared in Fig. S4b (Supporting information). It can be seen that after elution treatment with 1.5 mol/L NaCl, the enzyme loading amount of Fe₃O₄@ZIF-8@CRL has significantly decreased with a desorption ratio of 67.8%, while the enzyme loading amount of Fe₃O₄@ZIF-8@CRL@SiO₂ only has a little change with a desorption ratio of 9.0%. In Fe₃O₄@ZIF-8@CRL, the binding between lipase and carrier relies on weak physical adsorption. At high ionic strength, the non-covalent binding is destroyed, causing a large amount of enzyme leakage. In Fe₃O₄@ZIF-8@CRL@SiO₂, lipase is encapsulated on Fe₃O₄@ZIF-8 by silica nano-membrane. Therefore, the enzyme molecules can be immobilized firmly and are not prone to leakage.

  Due to the complex interaction between enzyme and carrier, the immobilization process will inevitably destroy the structure of enzyme in part or even in whole. One of the purposes in this work is to maintain the active conformation of lipase during immobilization. It is generally believed that physical immobilization is beneficial for preserving the natural conformation of enzyme. Here, the physical encapsulation is used to firmly immobilize lipase on the surface of carrier, avoiding enzyme inactivity caused by covalent immobilization. Therefore, it is necessary to evaluate the activity and conformation of lipase after immobilization. Fig. S4c (Supporting information) compares the activities of free lipase with Fe₃O₄@ZIF-8@CRL and Fe₃O₄@ZIF-8@CRL@SiO₂. It can be seen that after adsorbing lipase on Fe₃O₄@ZIF-8, there is a increase in enzyme activity from 46.0 U/mg to 57.8 U/mg, indicating that the hydrophobic ZIF-8 can activate lipase [28,29]. Moreover, the activity of Fe₃O₄@ZIF-8@CRL@SiO₂ is further improved compared with Fe₃O₄@ZIF-8@CRL, which may be attributed to that the hydrophilicity of silica membrane helps to maintain the "essential hydration layer" of enzyme and open the "active lid".

  Infrared spectroscopy is an effective tool for analyzing the structure of proteins [30-32]. The changes in the absorption bands of protein secondary structures at 1700–1600 cm⁻¹ (amide I band) are very sensitive, making it the most meaningful band for analyzing secondary structure changes [33-35]. Fig. S5a (Supporting information) shows the absorption spectra of free lipase, Fe₃O₄@ZIF-8@CRL, and Fe₃O₄@ZIF-8@CRL@SiO₂ in amide I band. The dark blue curve shows the infrared spectra of lipase at 1700–1600 cm⁻¹, which can be decomposed into multiple sub-peaks of secondary structures. The sub-peaks at 1610–1640 cm⁻¹ and 1680–1700 cm⁻¹ are assigned to β-sheet (pink), the sub-peaks at 1640–1650 cm⁻¹ are assigned to random coil (light blue), the sub-peaks at 1650–1660 cm⁻¹ are assigned to α-helix (green), and the sub-peaks at 1660–1680 cm⁻¹ are assigned to β-turn (yellow) [36]. The content of each secondary structure in protein can be calculated based on the sub-peak area (Fig. S5b in Supporting information). As can be seen, the contents of α-helix and β-sheet both increase after adsorbing lipase on Fe₃O₄@ZIF-8. Moreover, after encapsulating Fe₃O₄@ZIF-8@CRL into silica membrane, the contents of α-helix and β-sheet further increase. Compared Fe₃O₄@ZIF-8@CRL@SiO₂ with free lipase, the contents of α-helix and β-sheet increase, and the contents of β-turn and random coil decrease, indicating the increase in rigidity and stability of lipase conformation.

  The change in the added amount of TEOS may change the thickness of silica membrane on Fe₃O₄@ZIF-8@CRL@SiO₂, thereby affecting the stability and apparent activity of the immobilized enzyme. Figs. S6a–d (Supporting information) show the TEM images of Fe₃O₄@ZIF-8@CRL@SiO₂ prepared respectively by adding different amount of TEOS. As can be seen, the adding amount of TEOS has significant effect on the thickness of silica membrane formed on the surface of nanospheres, and the thickness of silica membrane gradually increases with the increase of TEOS. When TEOS amount is 22.4 µL, the thickness of silica membrane is 11 nm. While the thickness reaches 49 nm with 100.8 µL TEOS. In order to encapsulate lipase on the surface of nanospheres, a certain amount of TEOS need to be added to form a thin layer of silica membrane. On the other hand, the increase in the thickness of silica membrane may decrease the accessibility of enzyme, thereby affecting the apparent activity of immobilized enzyme.

  Fig. S6e (Supporting information) shows the effect of the thickness of silica membrane on lipase encapsulation. It can be seen that with 11.2 µL TEOS, the lipase on Fe₃O₄@ZIF-8@CRL@SiO₂ is prone to detachment when dealing with high salt solution, resulting in 59.6% of enzyme leakage. When 22.4 µL TEOS is added, only 9.0% of lipase on Fe₃O₄@ZIF-8@CRL@SiO₂ leaked, indicating that lipase can be effectively encapsulated only when the formed silica membrane reaches a certain thickness. Fig. S6f (Supporting information) shows the effect of TEOS amount on the enzyme specific activity and enzyme loading of Fe₃O₄@ZIF-8@CRL@SiO₂. As can be seen, when TEOS amount increases from 11.2 µL to 22.4 µL, lipase activity slightly increases from 51.0 U/mg to 57.8 U/mg. But when TEOS continues to increase, the activity rapidly decreases to 5.5 U/mg at 100.8 µL. These results suggest that the adding amount of TEOS should be appropriate to achieve the optimum encapsulation. An increase in silica membrane thickness will increase the mass transfer resistance of substrates/products and thus result in a decrease in the apparent activity of enzyme. Besides, it can also be seen that TEOS amount has little effect on lipase loading.

  We further determined the enzymatic kinetic parameters of Fe₃O₄@ZIF-8@CRL@SiO₂. Fig. S7 (Supporting information) shows the Lineweaver-Burk plot of free lipase and the encapsulated lipase. The Michaelis constant (K_m) and maximum reaction rate (V_max) have been calculated and listed in Table S1 (Supporting information). The K_cat/K_m value of Fe₃O₄@ZIF-8@CRL@SiO₂ is higher than that of free lipase, indicating a high catalytic efficiency of Fe₃O₄@ZIF-8@CRL@SiO₂. It can be ascribed to the hydrophobic interaction between ZIF-8 and the lid of lipase, resulting in interfacial activation [37,38].

  The stability of immobilized enzyme is of great significance in industrial operations. The thermal stabilities of free lipase, Fe₃O₄@ZIF-8@CRL and Fe₃O₄@ZIF-8@CRL@SiO₂ are shown in Fig. 4a. After 3 h of incubation at 45 ℃, the activity of free lipase decreases to only about 40% of the initial activity, indicating a low thermostability. Even after adsorbing lipase on Fe₃O₄@ZIF-8 to obtain Fe₃O₄@ZIF-8@CRL, the residual activity only increases by 10% and reaches 50% of the initial activity after 3 h of incubation. However, after encapsulating lipase into silica membrane, the residual activity has significantly increased to about 90% after 3 h of incubation, indicating a good thermostability of Fe₃O₄@ZIF-8@CRL@SiO₂. Besides, after 5 h of incubation, free lipase almost completely loses its activity and Fe₃O₄@ZIF-8@CRL remained 24% of its initial activity, while Fe₃O₄@ZIF-8@CRL@SiO₂ retain up to 75% of its initial activity. These results indicate that encapsulating enzyme into silica membrane is beneficial for improving the thermal stability of enzyme. It can be attributed to that the external silica membrane confines enzyme molecules within a limited space and prevents the unfolding and denaturation of enzyme at high temperature, which effectively inhibits enzyme inactivation.

  Figure 4

  

  Figure 4.  (a) Thermal stability at 45 ℃, (b) tolerance to denaturants, and (c) storage stability in buffer at room temperature of Fe₃O₄@ZIF-8@CRL@SiO₂, Fe₃O₄@ZIF-8@CRL and free lipase. (d) Reusability of Fe₃O₄@ZIF-8@CRL@SiO₂ and Fe₃O₄@ZIF-8@CRL. The error bar means the standard deviation (SD) (n = 5).

  DownLoad: Full-Size Img PowerPoint

  Fig. 4b evaluates the tolerance of Fe₃O₄@ZIF-8@CRL@SiO₂ to chemical denaturants. After 30 min treatment with 6 mol/L urea, the relative residual activities of free lipase, Fe₃O₄@ZIF-8@CRL, and Fe₃O₄@ZIF-8@CRL@SiO₂ are respectively 9.3%, 31.4% and 67.0%. Furthermore, after 30 min treatment with 0.2% sodium dodecyl sulfate (SDS), the relative residual activities of free lipase, Fe₃O₄@ZIF-8@CRL, and Fe₃O₄@ZIF-8@CRL@SiO₂ are respectively 12.2%, 30.4% and 54.4%. These results demonstrate that Fe₃O₄@ZIF-8@CRL@SiO₂ has good resistance to denaturants, indicating that silica membrane encapsulation reduces the damage of denaturants to enzyme conformation and provides protection for the enzyme. Fig. 4c further compares their storage stability. After 20 days of storage at room temperature, the relative activities of free lipase and Fe₃O₄@ZIF-8@CRL respectively decrease to about 5% and 9%, while Fe₃O₄@ZIF-8@CRL@SiO₂ can still remain 60% of its initial activity. This once again confirms that Fe₃O₄@ZIF-8@CRL@SiO₂ has high stability.

  Fig. 4d shows the reusability of Fe₃O₄@ZIF-8@CRL and Fe₃O₄@ZIF-8@CRL@SiO₂. After 8 recycles, Fe₃O₄@ZIF-8@CRL only retains 7.3% of its initial activity, while Fe₃O₄@ZIF-8@CRL@SiO₂ retains about 80% of its initial activity. This poor reusability of Fe₃O₄@ZIF-8@CRL is due to the weak non-covalent interaction between enzyme and carrier, which leads to the serious leakage of enzyme from carrier during use and a decrease in enzyme activity. To our satisfaction, Fe₃O₄@ZIF-8@CRL@SiO₂ can firmly encapsulate lipase in SiO₂ nano-membrane, thereby improving the reusability.

  In summary, Fe₃O₄@ZIF-8@CRL@SiO₂ is prepared by encapsulating enzymes in mesoporous SiO₂ nano-membrane. Compared with the traditional immobilized methods, silica membrane encapsulation can immobilize enzyme firmly on the surface of carrier with physical binding rather than chemical binding, thus having high enzyme activity and high stability. It has shown that the thickness of silica nano-membrane has effect on the apparent activity of Fe₃O₄@ZIF-8@CRL@SiO₂. Moreover, Fe₃O₄@ZIF-8@CRL@SiO₂ exhibits much higher stability than Fe₃O₄@ZIF-8@CRL.

  Declaration of competing interest

  The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  Acknowledgments

  The authors thank the financial supports from the National Natural Science Foundation of China (Nos. 22378093, 21878065), Natural Science Foundation of Hebei Province, China (No. E2022201100), the Science and Technology Support Plan of Baoding City (No. 2241ZF111), the Medical Science Foundation of Hebei University (No. 2021A09), and the Foundation of Affiliated Hospital of Hebei University (No. 2021Z003).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.109751.

  
  
• 1. [1]

     M. Zhang, Y. Zhang, C. Yang, C. Ma, J. Tang, Chem. Eng. J. 415 (2021) 129075. doi: 10.1016/j.cej.2021.129075

  2. [2]

     X. Zou, H. Su, F. Zhang, et al., Food Chem. 424 (2023) 136450. doi: 10.1016/j.foodchem.2023.136450

  3. [3]

     T.Z.P. de Figueiredo, F.A.P. Voll, N. Krieger, D.A. Mitchell, Biochem. Eng. J. 195 (2023) 108911. doi: 10.1016/j.bej.2023.108911

  4. [4]

     E.T. Hwang, S. Lee, ACS Catal. 9 (2019) 4402–4425. doi: 10.1021/acscatal.8b04921

  5. [5]

     F. Jia, Y. Liu, X. Deng, et al., ACS Appl. Mater. Interfaces 15 (2023) 7928–7938. doi: 10.1021/acsami.2c17971

  6. [6]

     Q. Yang, Y. Yan, X. Yang, et al., Chem. Eng. J. 372 (2019) 946–955. doi: 10.1016/j.cej.2019.04.216

  7. [7]

     R.C. Rodrigues, Á. Berenguer-Murcia, D. Carballares, R. Morellon-Sterling, R. Fernandez-Lafuente, Biotechnol. Adv. 52 (2021) 107821. doi: 10.1016/j.biotechadv.2021.107821

  8. [8]

     G.B. Chu, W.Y. Li, X.X. Han, et al., Small 19 (2023) 2301413. doi: 10.1002/smll.202301413

  9. [9]

     R. Schoevaart, M.W. Wolbers, M. Golubovic, et al., Biotechnol. Bioeng. 87 (2004) 754–762. doi: 10.1002/bit.20184

  10. [10]

      N. Chen, B. Chang, N. Shi, F. Lu, F. Liu, Food Chem. 399 (2023) 134000. doi: 10.1016/j.foodchem.2022.134000

  11. [11]

      D.M. Liu, J. Chen, Y.P. Shi, Trend Anal. Chem. 102 (2018) 332–342. doi: 10.1016/j.trac.2018.03.011

  12. [12]

      S.B. Sigurdardottir, J. Lehmann, S. Ovtar, et al., Adv. Synth. Catal. 360 (2018) 2578–2607. doi: 10.1002/adsc.201800307

  13. [13]

      M. Bilal, T.A. Nguyen, H.M.N. Iqbal, Coordin. Chem. Rev. 422 (2020) 213475. doi: 10.1016/j.ccr.2020.213475

  14. [14]

      C. Iriarte-Mesa, M. Pretzler, C. von Baeckmann, et al., J. Colloid. Interface Sci. 646 (2023) 413–425. doi: 10.1016/j.jcis.2023.04.158

  15. [15]

      Y. Wang, Z. Li, W. Fu, Y. Sun, Y. Dai, Adv. Fiber. Mater. 4 (2022) 1278–1289. doi: 10.1007/s42765-022-00177-0

  16. [16]

      W. Fu, W. Xu, K. Yin, et al., Mater. Horiz. 10 (2023) 65–74. doi: 10.1039/d2mh01255c

  17. [17]

      J. Han, T. Zhang, Z.Y. Zhou, H.J. Zhang, Food Chem. 417 (2023) 135810. doi: 10.1016/j.foodchem.2023.135810

  18. [18]

      L. Gao, J. Wen, Z. Huang, et al., ACS Appl. Mater. Interfaces 15 (2023) 31224–31232. doi: 10.1021/acsami.3c04377

  19. [19]

      H.A.J.A. Lawati, J. Hassanzadeh, N. Bagheri, Food Chem. 383 (2022) 132469. doi: 10.1016/j.foodchem.2022.132469

  20. [20]

      X. Shen, Y. Du, Z. Du, et al., Mater. Today Chem. 27 (2023) 101326. doi: 10.1016/j.mtchem.2022.101326

  21. [21]

      H. Cui, L. Zhang, L. Eltoukhy, et al., ACS Catal. 10 (2020) 14847–14856. doi: 10.1021/acscatal.0c03233

  22. [22]

      Y. Weng, G. Yang, Y. Li, et al., Adv. Colloid Interface Sci. 318 (2023) 102957. doi: 10.1016/j.cis.2023.102957

  23. [23]

      G. Bayramoglu, O. Celikbicak, M. Kilic, M.Y. Arica, Food Chem. 366 (2022) 130699. doi: 10.1016/j.foodchem.2021.130699

  24. [24]

      L.T. Xu, K.X. Xie, S.H. Cao, et al., Chin. Chem. Lett. 34 (2023) 108181. doi: 10.1016/j.cclet.2023.108181

  25. [25]

      Y.Y. Xiao, J.T. Zhang, Y.H. Dong, et al., Bioconjug. Chem. 34 (2023) 1045–1053. doi: 10.1021/acs.bioconjchem.3c00127

  26. [26]

      M.D. Gong, X.J. Pei, G.X. Duan, et al., Mater. Des. 220 (2022) 110865. doi: 10.1016/j.matdes.2022.110865

  27. [27]

      S. Gadipelli, W. Travis, W. Zhou, Z. Guo, Energ. Environ. Sci. 7 (2014) 2232–2238. doi: 10.1039/C4EE01009D

  28. [28]

      P. Zhang, J. Chen, B. Sun, et al., Biochem. Eng. J. 173 (2021) 108066. doi: 10.1016/j.bej.2021.108066

  29. [29]

      L. Zhong, Y. Feng, H. Hu, et al., J. Colloid Interface Sci. 602 (2021) 426–436. doi: 10.1016/j.jcis.2021.06.017

  30. [30]

      S. Badieyan, D.R. Bevan, C. Zhang, Biotechnol. Bioeng. 109 (2012) 31–44. doi: 10.1002/bit.23280

  31. [31]

      J. Waeytens, J.D. Meutter, E. Goormaghtigh, A. Dazzi, V. Raussens, Anal. Chem. 95 (2023) 621–627. doi: 10.1021/acs.analchem.2c01431

  32. [32]

      J.D. Meutter, E. Goormaghtigh, Anal. Chem. 93 (2021) 3733–3741. doi: 10.1021/acs.analchem.0c03677

  33. [33]

      R. Lu, W.W. Li, A. Katzir, et al., Analyst 141 (2016) 6061–6067. doi: 10.1039/C6AN01477A

  34. [34]

      S. Yang, Q. Zhang, H. Yang, et al., Int. J. Biol. Macromol. 206 (2022) 175–187. doi: 10.1117/12.2639233

  35. [35]

      K.R. Webb, K.A. Hess, A. Shmidt, K.D. Segner, L.E. Buchanan, Biophys. J. 122 (2023) 1491–1502. doi: 10.1016/j.bpj.2023.03.014

  36. [36]

      Y.P. Cao, G.Y. Zhi, L. Han, Q. Chen, D.H. Zhang, Food Chem. 364 (2021) 130428. doi: 10.1016/j.foodchem.2021.130428

  37. [37]

      J. Skjold-Jørgensen, J. Vind, O.V. Moroz, et al., Biochim. Biophys. Acta Proteins Proteom. 1865 (2017) 20–27. doi: 10.1016/j.bbapap.2016.09.016

  38. [38]

      Q. Huo, J. Zhao, W. Li, et al., Chem. Commun. 55 (2019) 7155–7158. doi: 10.1039/c9cc02797a

• 

  Figure 1  Scheme of lipase encapsulated by silica nano-membrane on the surface of magnetic ZIF-8.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  TEM image and the particle size distribution of (a) Fe₃O₄, (b) Fe₃O₄@ZIF-8 and (c) Fe₃O₄@ZIF-8@CRL@SiO₂. (d) Vibrating sample magnetic strength (VSM) of Fe₃O₄@ZIF-8@CRL@SiO₂.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  (a) Infrared spectra of Fe₃O₄@ZIF-8@CRL and Fe₃O₄@ZIF-8@CRL@SiO₂. (b) XRD of Fe₃O₄, Fe₃O₄@ZIF-8 and Fe₃O₄@ZIF-8@CRL@SiO₂. (c) EDS, (d) TEM mapping, (e) water contact angle, and (f) fluorescence characterization of Fe₃O₄@ZIF-8@CRL@SiO₂.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  (a) Thermal stability at 45 ℃, (b) tolerance to denaturants, and (c) storage stability in buffer at room temperature of Fe₃O₄@ZIF-8@CRL@SiO₂, Fe₃O₄@ZIF-8@CRL and free lipase. (d) Reusability of Fe₃O₄@ZIF-8@CRL@SiO₂ and Fe₃O₄@ZIF-8@CRL. The error bar means the standard deviation (SD) (n = 5).

  下载: 全尺寸图片 幻灯片
•