A Method of Improving Physical Adsorption of Cellulase On Mesoporous Zr-Based MOF Surface: Enhanced Stability, Recyclability, And Catalytic Eciency

In this paper, we aimed at developing immobilized cellulase biocatalyst by enhancing the anchor of cellulase on support surface. A mesoporous Zr-based MOF was rst synthesized by biomineralization method using dextran as template. The resultant PVP-cellulase@CD-UIO-66-Zr exhibited a high loading capacity of 265 mg g − 1 support. The physical adsorption of cellulase on CD-UIO-66-Zr could be further enhanced by the capping of cellulase with PVP. About 83% of the activity of PVP-cellulase@CD-UIO-66-Zr could be retained after six cycles, and its equilibrium leakage ratio was 36% during thirty days’ leaching test. It was noting that about 80% of activity of immobilized PVP-cellulase@CD-UIO-66-Zr could be retained after incubation at 80°C for 1 hour. The immobilized cellulase exhibited higher pH stability, thermostability, storage stability and catalytic eciency than free one.


Introduction
The non-renewability of fossil energy and the increasingly prominent global environmental problems caused by it have made the development of biomass energy highly valued by governments of all countries(Danish and Wang 2019; Cao et al. 2020). Lignocellulose is the most widely distributed and abundant biomass resource on earth (Chandel et al. 2018; Karumuri et al. 2015). Through certain processing, cellulose can be transformed into a series of products with great application value, such as fuel (Ci et al. 2017), chemical raw materials (Ji et al. 2019) and pharmaceutical products (Zhang et al. 2015a). The utilization and transformation of cellulose is of great signi cance to solve the world energy crisis and environmental pollution problems.
Cellulose is a linear polymer compound bound by glucose molecules with β-1,4-glycosidic bonds, and generally requires pretreatment to be degraded into small molecular sugars before it can be used.
Lignocellulose can be degraded by physical method (Chen et al. 2017b), chemical method , biological method ). Biological method generally uses cellulase to hydrolyze lignocellulose into glucose. It has the advantages of mild conditions, environmental friendliness, strong speci city, high catalytic e ciency, and low unit energy consumption, meeting the sustainable development of today's society demand. However, the use cost of cellulase accounts for 50% of the total cost of cellulose enzymatic hydrolysis and sacchari cation(Gokhale and Lee 2012). Therefore, obtaining immobilized enzymes with more stable performance and recyclable utilization is an effective way to reduce the application cost of cellulase.
The catalytic performance of immobilized enzymes is highly dependent on the property of carrier As a new type of porous material, metal-organic framework is a kind of porous coordination polymer with two-dimensional or three-dimensional pore structure connected by metal nodes and organic ligands through coordination bonds (Cui et al. 2018). MOF materials have the advantages of high porosity and high speci c surface, so its application is in full swing, especially in the eld of enzyme immobilization (Liang et al. 2015).
There are relatively few studies on the use of mesoporous MOF materials for the immobilization of cellulase. The diffusion of macromolecular cellulose into micro/mesoporous MOF is often limited. The currently reported immobilization of cellulase on MOF is mainly surface adsorption (Khoshnevisan et al. 2011) and covalent binding (Huang et al. 2020). Multi-step reactions are often involved in covalent bonding process, and the preparation of carrier materials is complicated. The physical adsorption method has remarkable characteristics of simple and mild immobilization process, reducing the loss of activity during immobilization process. However, the immobilized enzyme is easily detached from the carrier material due to weak binding force, which is not conducive to the recycling of enzyme. In this paper, a mesoporous UIO-66-Zr was rst synthesized by biomineralization method using dextran as template, and the prepared MOF was used as support material for cellulase immobilization by physical adsorption. Furthermore, a small amount of PVP was used to improve the stability of immobilized cellulase. The resultant mesoporous MOF before and after enzyme immobilization were characterized by BET, FT-IR, XRD, TGA, DLS, SEM and TEM. The stability, reusability, and catalytic e ciency of immobilized cellulase were investigated.

Preparation of UIO-66-Zr, Dex@UIO-66-Zr and CD-UIO-66-Zr
The preparation of the UIO-66-Zr was based on the reported method with some modi cation . ZrCl 4 (0.7 mmol) and BDC (0.7 mmol) were rst dissolved in 10 mL DMF, then the suspension solution was sonicated for 30 minutes followed by the incubation in an oil bath at 80 ℃ for 24 h. The solid products were collected by centrifugation and washed three times with DMF and methanol to remove the residual reactant. The resultant samples were dried at 60 ℃ overnight.
The prepared Dex@UIO-66-Zr was then placed in a tube furnace and calcined at 325°C for 2 hours. The resultant sample was activated in 10 mL of methanol for three days to obtain CD-UIO-66-Zr.

Preparation of cellulase@UIO-66-Zr, cellulase@CD-UIO-66-Zr, PVP-cellulase@CD-UIO-66-Zr
The immobilized cellulase, including cellulase@UIO-66-Zr and cellulase@CD-UIO-66-Zr, were prepared by dissolving 10 mg substrate in 2.5 mL of acetic acid including cellulase. The preparation of PVP-cellulase@CD-UIO-66-Zr were conducted by dissolving 10 mg CD-UIO-66-Zr in 2.5 mL of 0.5% PVP including cellulase. The reaction was conducted at room temperature with stirring for 12 h, then the solids were separated by centrifugation at 6000 rpm for 10 min. The product was washed by deionized water and dried in vacuum freeze dryer for 10 h, eventually saved at 4°C.

Characterization of support materials and immobilized cellulase
Fourier transform infrared (FT-IR) spectra were collected in the range 3500 − 500 cm − 1 on a Nicolet Nexus 470 Spectrometer. N 2 adsorption-desorption isotherms and pore size distribution were measured by a Brunauer-Emmett-Teller (BET) instrument (ASAP 3020, Micromeritics). Before BET measurement, all samples were degassed at 150°C under reduced pressure overnight. Powder X-ray diffraction (XRD) patterns were recorded by an X-ray diffractometer (XRD-6100Lab, Japan) with scattering angles (2θ) of 5-40°C at a scan rate of 7°C min − 1 . Thermal gravimetric analysis (TGA) was performed using the STA449C Thermal Analyzer. The dried sample was lled into a crucible, and the TGA curve was collected from 10 to 1000°C with a heating rate of 10°C min − 1 under a continuous stream of nitrogen gas. The morphologies of all samples were recorded by JSM-6010 PLUS/LA scanning electron microscope (SEM) and FEI Tecnai G2 S-TWIN F20 transmission electron microscope (TEM). DLS were recorded by particle size analyzer (Mastersizer 3000, Malvern Instruments Ltd., UK).

Enzyme activity assay
The activity of cellulase was assessed by incubating cellulase with the CMC substrate for 30 min at 50°C in a 0.2 M acetic buffer (pH 5.0). The amount of produced glucose was examined using dinitrosalicylic acid method and the absorbance was measured at 540 nm (Miller et al. 1960). The concentration of enzyme was measured by PierceTM BCA protein assay kit. The Loading e ciency and Loading capacity were calculated according to Eq. 1 and Eq. 2 respectively.
Where m (mg) is the mass of cellulase initially added to the solution; C 1 (mg mL − 1 ) is the concentration of cellulase in supernatant; V 1 (mL) is the volume of supernatant; M (g) is the weight of the immobilized cellulase biocomposite.

Stability and reusability of immobilized cellulase
The pH stability and thermal stability were examined by incubating enzyme solution at various pH values (3.0-8.0) and temperatures (30-80°C) for 60 min respectively with other conditions at the optimum. To analyze the reusability of immobilized cellulase, the biocatalyst was recovered by centrifugation and washing with deionized water for the next test. The reaction was repeated 8 times and the initial activity was set as 100%. The storage stabilities of free and immobilized cellulase were examined for 30 days with an interval of three days at 4 ℃. The leaching test was conducted as the storage stability test except that the biocatalyst was incubated in solution. The initial activity was regarded as 100 %.

Kinetic parameters
The initial reaction rates of free and immobilized cellulase for catalytic conversion of different concentrations of substrates were measured, and the data were plotted using a double-reciprocal Lineweaver-Burk equation. The apparent kinetics parameters (K m and V max ) were calculated using Eq. 3.

Fluorescent Labeling of cellulase
First, 10 mg cellulase was weighed and dissolved in 5 mL acetate buffer solution, then 5mg carbodiimide hydrochloride (EDC) and 2.5mg N-hydroxysuccinimide (NHS) were added. After the reaction was lasted for 1 h at 4°C, 100 ug of uorescein isothiocyanate (FITC) was added to the solution, and the reaction was lasted for 4 hours at low temperature (4°C) in dark. To obtain the uorescently labeled cellulase (FITC-cellulase), the solution was dialyzed one day to remove unreacted EDC, NHS and FITC. The FTIR spectrum was analyzed to con rm the composition of as-prepared MOF (Fig. 1. (a)). The The XRD patterns of Zr-based MOF materials were shown in Fig. 1 (b). All diffraction peaks of the TGA spectrum was shown in Fig. 1 (c). Dextran 1500 has a two-stage thermal decomposition process: the rst weight loss (< 200°C) was due to the loss of free water and bound water; the second weight loss was because of the structure collapse of dextran 1500. So the calcining temperature for removing dextran 1500 was set at 325°C by TGA curve of dextran 1500. It can be deduced from the TGA curve of CD-UIO-66-Zr that its structure collapsed within the temperature range of 500-600°C.

Results
The morphologies of Zr-based MOF were characterized by SEM and TEM, and uniform and aggregated MOF particles could be observed in Fig. 2 and Fig. S1. Compared with conventional UIO-66-Zr, the resultant Dex@UIO-66-Zr and CD-UIO-66-Zr showed an increase in aggregation. The average particle size of UIO-66-Zr determined by dynamic light scattering was 179.8 nm, while the particle size of Dex@UIO-66-Zr and CD-UIO-66-Zr increased to 389.9 nm and 497.8 nm because of aggregation (Fig. 2).
The pore size and distribution of the resultant Zr-based MOF particles was analyzed by N 2 adsorption desorption (Fig. 3). The hysteresis loops at P/P 0 value around 0.4 and 0.8 in the isotherms con rmed the existence of mesopores in the prepared Zr-based MOF particles. The adsorption-desorption isotherms of UIO-66-Zr and CD-UIO-66-Zr possessed a type curve with H1 hysteresis loop (Sing et al. 1985). The successful preparation of mesoporous UiO-66-Zr can also be demonstrated by the BJH pore diameter distributions ( Fig. 3(a)). As shown in Table 1, the speci c surface area and total pore volume of mesoporous CD-UIO-66-Zr are 646.9 m 2 g − 1 and 0.2809 cm 3 g − 1 respectively, which are slightly low than UIO-66-Zr. However, the average pore size of CD-UIO-66-Zr increased from 2.255 nm to 4.617 nm. Polysaccharides have been used to induced biomineralization of ZIF-8, resulting in the formation of protective shell around polysaccharides (Liang et al. 2017).The introduction of dextran 1500 as template for preparing UIO-66-Zr played a key role in the fabrication of mesoporous UIO-66-Zr. Total pore volume (cm 3 g − 1 ) Average pore size and 38% respectively (Fig. 4b). Some reported cycling stability of cellulase immobilized on various solid supports were summarized in Table 2. The adsorption of PVP-capping cellulase onto meso-CD-UIO-66-Zr exhibited enhanced recycling stability relative to other reported systems. Then, a leaching test was conducted by incubating the immobilized cellulase in buffer solution for 30 days.. As shown in Fig. 4c, the leakage ratio of immobilized cellulase gradually increased and be to in equilibrium after 21 days, and that of PVP-cellulase@CD-UIO-66-Zr was 32%. In general, the leakage ratio of PVP-cellulase@CD-UIO-66-Zr remained lower than the other immobilized enzyme throughout the entire incubating period, which indicated that the detachment of cellulase from CD-UIO-66-Zr could be greatly inhibited once the cellulase molecules was modi ed with PVP. So, the capping of cellulase by PVP, to some extent, enhance the anchoring of cellulase on the surface pore of UIO-66-Zr, which contributes to the improvement in recyclability compared with cellulase@CD-UIO-66-Zr.
It was an interesting phenomenon that CD-UIO-66-Zr exhibited higher adsorption strength with cellulase than UIO-66-Zr. Since the size of cellulase (12.4 nm × 3.7 nm)(Chen et al. 2017a) was much larger than the pore size of UIO-66-Zr and CD-UIO-66-Zr, it's di cult for cellulase to diffuse into pore through speci c orientation. The dye-tagging strategy has been developed to enhance adsorption of proteins onto MOF with smaller pore size than protein molecules (Mehta et al. 2016). It was achieved by the conjugation of small uorescein isothiocyanate (FITC) dye molecules with trypsin. Although the adsorption of large molecules into the pore of MOF was hindered, the small FITC molecules (0.9 × 1.2 × 1.4 nm 3 ) could enter the pores and enhance the anchoring of large biomolecules on MOF. We prepared FITC-cellulase conjugate and studied the adsorption of FITC-cellulase on UIO-66-Zr and CD-UIO-66-Zr. The penetration of FITC molecules into the pore of MOF was con rmed by laser scanning confocal microscopy (CLSM) characterization (Fig. 5a). Owing to slightly larger pore size of CD-UIO-66-Zr (4.617 nm) than the corresponding UIO-66-Zr (2.255 nm), as displayed in Scheme 1, the cellulase molecules like a long stick and the short axis (3.7 nm) make the stick could only partially plug into the pore of CD-UIO-66-Zr, and the adsorption of cellulase on CD-UIO-66-Zr can be enhanced. We also analyzed the BJH N 2 adsorptiondesorption isotherms of CD-UO-66-Zr after the immobilization of cellulase (Fig. 5b). Compared with CD-UO-66-Zr (633 m 2 g -1 ), the decrease in the speci c surface of PVP-cellulase@CD-UO-66-Zr (499 m 2 g -1 ) was not signi cant and the mesoporous structure was preserved, which may be due to the partial plugging of PVP-cellulase into the pore of CD-UIO-66-Zr. The structure of the immobilized cellulase@CD-UIO-66-Zr were analyzed by FT-IR and XRD. Figure 1 (a) showed the FT-IR spectrum of cellulase@CD-UIO-66-Zr and PVP-cellulase@CD-UIO-66-Zr. The weak bands revealed at 1620 and 1028 cm − 1 were ascribed to the C-N stretching vibration of cellulase, which demonstrated the successful immobilization of cellulase. As displayed in Fig. 1(b), the adsorption of cellulase on CD-UIO-66-Zr had no obvious effect on the position and intensity of diffraction peaks, indicating that the structure of CD-UIO-66-Zr was well retained during enzyme immobilization. It can be clearly seen from the SEM image in Fig. 2 that the anchoring of cellulase on the surface of Zr-based MOF carrier had no signi cant in uence on the monography of CD-UIO-66-Zr.

Stability of free cellulase and immobilized cellulase
It is well known that the catalytic activity of free cellulase and cellulase@CD-UIO-66-Zr is dependent on reaction conditions. Enzymes exhibit their maximum activity at optimum solution pH and reaction temperature. As shown in Fig. 6, the free cellulase showed the highest activity at pH 5, while the maximum activity for cellulase@CD-UIO-66-Zr and PVP-cellulase@CD-UIO-66-Zr was achieved in a pH range of 5 to 6.. The effect of temperature from 30°C to 80°C on the activities of free and immobilized cellulase was investigated at pH 5.0. The optimum temperature of free cellulase, cellulase@CD-UIO-66-Zr and PVP-cellulase@CD-UIO-66-Zr are 50°C, 60°C and 60-70°C respectively.
The immobilized cellulase exhibited enhanced pH, thermal and storage stabilities compared with free cellulase and previous immobilized cellulase (Xu et al. 2014;Zhang et al. 2015b). It was noting that after incubation at pH 8.0 for 1 hour, as shown in Fig. 7(a), the activity loss of free cellulase was 56%, while the immobilized cellulase@CD-UIO-66-Zr and PVP-cellulase@CD-UIO-66-Zr maintained 77% and 80% of its original activity, respectively. Meanwhile, the thermal stabilities were analyzed by incubating the free and immobilized cellulase at different temperatures for 1 hour. As shown in Fig. 7(b), when rising temperature beyond 70°C, a drastic activity loss was observed for free enzyme. The activity loss of free cellulase was 51% after incubation at 80°C for 1 hour, while 82% and 85% of original activity were preserved for cellulase@CD-UIO-66-Zr and PVP-cellulase@CD-UIO-66-Zr respectively. The immobilized cellulase on CD-UIO-66-Zr exhibited higher thermal tolerance than previously reported cellulase immobilization system (Abraham et al. 2014). The storage stabilities were analyzed by storing free and immobilized cellulase at 4°C for 30 days. As shown in Fig. 7 (c), with the extension of storage time, free cellulase retained only 15% of its original activity after 30 days, while 50% and 63% were preserved for cellulase@CD-UIO-66-Zr and PVP-cellulase@CD-UIO-66-Zr respectively.
These results indicated that the immobilization of cellulase on CD-UIO-66-Zr greatly improved the pH, thermal, storage and recycling stabilities of cellulase. Meanwhile, the modi cation of cellulase with PVP effectively improve the stability especially recycling stability relative to cellulase@CD-UIO-66-Zr by enhancing the anchoring of cellulase on the mesoporous CD-UIO-66-Zr.

Catalytic kinetics of free cellulase and immobilized cellulase
The catalytic kinetics of cellulase@CD-UIO-66-Zr and PVP-cellulase@CD-UIO-66-Zr were also analyzed by Michaelis-Menten equation. The Lineweaver-Burk plots were displayed in Fig. S3, and the kinetic parameters were listed in Table 3. Compared with free cellulase, the K m value of cellulase@CD-UIO-66-Zr and PVP-cellulase@CD-UIO-66-Zr slightly decreased to 13.387 mg mL − 1 and 13.355 mg mL − 1 respectively, demonstrating the full contact of CMC to the active sites of immobilized cellulase (Hosseini et al. 2018). The K cat /K m values of PVP-cellulase@CD-UIO-66-Zr (1.311 mL mg − 1 min − 1 ) was slightly higher than free one (0.732 mL mg − 1 min − 1 ), which indicated that a high catalytic e ciency was achieved for PVP-cellulase@CD-UIO-66-Zr relative to free cellulase.

Conclusion
In this study, a mesoporous Zr-based MOF was prepared by biomineralization method using dextran as template, and the resultant CD-UIO-66-Zr was successfully used for the surface adsorption of cellulase with a high loading capacity about 265 mg g -1 . The better recycling stability of CD-UIO-66-Zr could be due to the partial inserting of cellulase into its surface pore. The capping of cellulase with PVP is conductive to enhance the anchoring of cellulase on CD-UIO-66-Zr. The successful preparation of Zr-based MOF with moderate pore size shows the promising future of using MOF as carrier for cellulase immobilization by simple surface adsorption method.       Effects of reaction pH (a) and temperature (b) on catalytic activity of free cellulase, cellulase@CD-UIO-66-Zr and PVP-cellulase@CD-UIO-66-Zr.