A new strategy for cellulases application in high temperature industrial scenarios with syringic acid assisting

In the process of bioethanol production, more stable and active cellulase in high temperature condition is required. In this study, syringic acid was applied in cellulase hydrolysis system. At 70°C, TvEG3 activity increased 201.36%, CtBglA activity decreased 72.79% by syringic acid. With syringic acid assisting, TvEG3 thermostability was improved, CtBglA thermostability was reduced. Syringic acid scarcely affected CtCBH. In hydrolysis system with the cellulases containing TvEG3, CtCBH, and CtBglA, the reducing sugar yield improved by 28.37% with syringic acid assisting. With the molecular dynamic simulation in syringic acid system, the backbone root‐mean‐square deviation (RMSD) and the residue root‐mean‐square fluctuation (RMSF) of TvEG3, CtCBH reduced, while the RMSD and RMSF of CtBglA increased. The reduction in the number of secondary structures, especially α‐helix, caused the structure of CtBglA in the presence of syringic acid to collapse at high temperature. More secondary structures in TvEG3 and more α‐helix in CtCBH in the presence of syringic acid make them more stable at high temperatures. These means syringic acid can stabilize TvEG3 and CtCBH structure, destabilize CtBglA structure at high temperature. In summary, this study not only provides insight into cellulase hydrolysis at high temperature with syringic acid assisting but also demonstrates the promoting mechanism of syringic acid.

high-value products because they can be enzymatically degraded to fermentable sugars (Houfani et al., 2020).
Cellulases are key biological catalysts for converting carbohydrates from lignocellulosic biomass into fermentable sugars and consist of endoglucanases (EGs), cellobiohydrolases (CBHs), and βd-glucosidases (BGs; Payne et al., 2015;Xue et al., 2020). In nature, some microorganisms and plants can produce specific cellulases separately for the degradation of cellulose. However, from the perspective of industrial production, the low efficiency and high cost of cellulase remains a major obstacle to the production of bioethanol from the lignocellulosic biomass (Lin et al., 2022;Liu et al., 2016;Morales-Rodriguez et al., 2012;Trudeau et al., 2014). Existing hydrolysis of cellulase is carried out at 50°C, which is close to the optimum temperature for many fungal cellulases. Hydrolysis of lignocellulose at relatively high temperatures (>50°C) may have several potential advantages such as (i) reduced risk of microbial contamination, (ii) higher hydrolysis velocities; (iii) reduced solution viscosity, and facilitated recovery of products, (iv) better compatibility with thermal pretreatments, and decreased the cost for cooling (Bhalla et al., 2013;Chokhawala et al., 2015;Viikari et al., 2007). Furthermore, the application of high temperature conditions during industrial bioconversion of lignocellulose can reduce crystallinity, decompose lignin, and increase cellulose accessibility to cellulase (Maki et al., 2009;Sharma et al., 2019). Therefore, more stable and more active cellulase at elevated temperature is required to meet the needs of industrial bioethanol production .
Several methods have been proposed to enhance the hydrolysis activity of cellulase through immobilization or adding surfactants (Cai et al., 2017;Khoshnevisan et al., 2017;H. Wang et al., 2015;Y. Wang et al., 2018;Yin et al., 2019). Hydrolysis of cellulase at elevated temperature during industrial bioprocessing relies on thermostable cellulases produced from thermophilic bacteria and fungi (Bhalla et al., 2013;Graham et al., 2011). However, low biomass and extremophilic nature of thermophilic microorganism limits the commercialization production for thermophilic cellulases. It is urgent to find a way to make cellulase more stable under high temperature conditions. From a biotransformation perspective, it has been hypothesized that lignin-derived phenolic compounds adversely affect cellulase hydrolysis (Li et al., 2014;Qin et al., 2016;). Some studies indicated that phenolic compounds have favorable effects on cellulase Oliveira et al., 2020;Stamogiannou et al., 2021;Tian et al., 2013). The underlying mechanisms are still unclear and need to be further explored. In our previous study (Ran et al., 2022), four typical lignin-derived phenolic acids, namely vanillic acid, syringic acid, ferulic acid, and isovanillic acid were found to improve cellulase specific activity after pre-incubation at 50°C. Phenolic acid might change the structure or polymerization of cellulase resulting in the increase of activity.
In view of the beneficial effect of phenolic acid on cellulase, the action of syringic acid on cellulase at elevated temperature was investigated. Endoglucanase III from Trichoderma viride (TvEG3), CBH from Chaetomium thermophilum CT2 (CtCBH), and BGs from C. thermocellum (CtBglA) were expressed. The effects of syringic acid on the hydrolysis and thermostability of single-component cellulases were explored at 70°C. Molecular dynamics (MD) simulations were performed for TvEG3, CtCBH, and CtBglA in a system containing syringic acid. Herein, the promoting effect of syringic acid on cellulase at high temperature was investigated using an integrated enzymatic hydrolysis, computational, and structural approach.

| Main strains, plasmids, primers, and reagents
The gene encoding T. viride endoglucanase III (TvEG3) was obtained from the NCBI database (GenBank: AAQ21383.1). The gene sequence was synthesized by Tsingke Biotechnology Co., Ltd. and ligated to pPIC9K pre-digested with EcoRI and NotI. The Pichia pastoris host strain histidine, which requires the auxotrope GS115 (his4), was used to express the resulting recombinant plasmid pPIC9K-eg3. The gene encoding CtCBH (GenBank: AAW64927.1) was assembled using pPICZα, and the gene encoding CtBglA (GenBank: X60268.1) was assembled using pET23a. The expression strains loaded with these two plasmids were stored in the laboratory. Escherichia coli DH5α cells were used for plasmid amplification. Carboxymethylcellulose sodium (CMC-Na), and Avicel were purchased from Sigma-Aldrich Corp. Phosphoric acid-swollen cellulose (PASC) was obtained from Avicel after phosphoric acid treatment (Lv et al., 2021). Filter paper (FP) was obtained from Whatman. 3,5-dinitrosalicylic acid (DNS), phosphoric acid, citric acid, and other reagents were purchased from Sinopharm Group. p-Nitrophenyl-beta-d-glucoside (pNPG) was purchased from MedChemExpress. Syringic acid was purchased from Sigma-Aldrich.

| Protein expression
Plasmid pPic9K-eg3 was transformed into yeast, P. pastoris GS115. Minimal dextrose solid medium was used as a preliminary screening medium. Trypan blue-CMC solid medium was used for secondary screening to obtain positive transformants that produced a transparent hydrolysis circle. The best clones were selected and cultured in 100 mL of buffered glycerol-complex medium. The cultures were incubated at 28°C for 48 h with shaking at 220 rpm. After centrifugation, the cells were transferred to buffered methanol-complex medium for inducible expression. The expression was carried out for 5 days. Ethanol (1%) was added every 24 h. Expression of CtCBH followed the expression manual for P. pastoris X33, and CtBglA was expressed in E. coli BL21. Sodium dodecyl sulfatepolyacrylamide gel electrophoresis (SDS-PAGE) was employed to analyze protein expression. His-tag affinity chromatography was used for purification, and the proteins were stored in phosphate buffered saline. Bradford method was used to determine protein concentration with bovine serum albumin as the standard (Kruger, 2009).

| Characterization of enzymes activity
CMC-Na was used as the substrate to assay TvEG3 activity. The diluted protein solution (1 μL) was added to 200 μL of 0.5% (w/v) CMC-Na buffer. The reaction was conducted at 70°C for 5 min. The CtCBH activity was assayed using PASC as the substrate. Purified enzyme solution (5 μL) was added to 300 μL of the reaction mixture containing 10 mg PASC. The reaction conducted at 70°C for 10 min. DNS method was used to determine the amount of reducing sugar released (Miller, 1959). The CtBglA activity was assayed using pNPG as substrate. The purified enzyme solution (1 μL) was added to 300 μL of the pNPG solution (1 mg/mL). The reaction was conducted at 70°C for 5 min. The released p-nitrophenol (pNP) was determined by measuring the absorbance at 420 nm (Song et al., 2016). The amount of enzyme required to produce 1 μmol of reducing sugar or pNP per minute was defined as one unit of activity. Optimal reaction conditions for enzymes tested at different temperatures and pH values were determined.

| Syringic acid assisting
TvEG3, CtCBH, and CtBglA were mixed in a 1:1:2 ratio for enzyme activity to obtain cellulase mixture (cellulase mix ). The effect of syringic acid on enzymatic hydrolysis was determined using the following method: syringic acid (12.5 mM) was involved in the hydrolysis of cellulase mix , TvEG3, CtCBH, and CtBglA. The reaction mixture without syringic acid was used as the control. For cellulase mix , an enzyme solution with a total enzyme activity of 8 U was added to 400 μL of the reaction solution with 10 mg FP. For TvEG3, 1 μL of the purified enzyme solution was added to 200 μL of 1% (w/v) CMC-Na substrate solution. For CtCBH, 20 μL of purified enzyme solution was added to 600 μL of reaction solution containing 10 mg of PASC substrate. For CtBglA, 1 μL of the purified enzyme solution was added to 300 μL of the pNPG substrate solution. The reaction was conducted at 70°C for 10 min. The effect of syringic acid on the thermal stability of single components cellulase was investigated. Syringic acid (5, 10, 20 mm) were then added to the cellulase protein solution. The protein concentrations of TvEG3, CtCBH, and CtBglA were 0.6, 0.7, and 0.5 mg/ mL, respectively. A protein solution with no syringic acid was used as the control. The protein solution was incubated in a 70°C hot water bath and 28°C incubator, respectively, for different durations of time. The enzyme activity was detected at different incubation times.

| Circular dichroism
TvEG3 was treated with syringic acid using the method described in Section 2.4. Following incubation in a water bath at 70°C for 5 min, the TvEG3 solution containing syringic acid was centrifuged at 3000 g for 20 min using ultrafiltration tubes (Millipore 10 kDa) for concentration. During ultrafiltration, 10 × 10 mM phosphate buffer was added to thoroughly wash the TvEG3 to remove syringic acid. This was performed to avoid the interference of syringic acid on the circular dichroism (CD) spectra. The protein solution was named TvEG3-syringic acid. TvEG3 and TvEG3-syringic acid solution were both diluted to 0.1 mg/mL and added to silica quartz spectrophotometer cells with 10 mm path lengths. The CD spectra (180-260 nm) were obtained using a Chirascan V100 spectrometer (Applied Photophysics).

| Thin layer chromatography
The hydrolysate supernatant of the cellulase mix was further analyzed using thin layer chromatography (TLC). The stationary phase was silica gel 60 F25. The developing solvent was ethyl acetate:methyl alcohol:acetic acid:water in a 4:2:0.25:1 (v/v) ratio. The color-developing agent was sprayed onto the thin-layer plate. Sugar spots were observed after heating to 100°C for 30 min. Glucose (G1) and cellobiose (G2) were used as standards.

| MD simulation
The molecular structure of syringic acid was obtained from Chemsider. (http://www.chems pider.com/). Bio2Byte (https://www.bio2b yte.be/) was used to generate an itp-format file for syringic acid. Homology modeling of TvEG3, CtCBH, and CtBglA was performed using SWISS-MODEL online website. MD simulations were performed using TvEG3, CtCBH, and CtBglA with or without the addition of syringic acid in the GROMACS MD software package. The temperature was set at 343.15 K to characterize the structural changes of the protein.
MD simulations were performed by employing the force field parameters of solvent molecules (H 2 O and syringic acid) in the AMBER99SB-ILDN protein force field (Lindorff-Larsen et al., 2010). The protein was soaked in a cubic box of water molecules with a box edge (1.2 nm from the molecule periphery) using the gmxeditconf module. The gmxsolvate module was then used for solvation. The TIP3P water model was used to solvate the protein (Jorgensen et al., 1983). To obtain an MD simulation system with 20 mM syringic acid using the gmxinsertmolecules module, 7, 8, and 10 syringic acid molecules were added to the TvEG3, CtCBH, and CtBglA simulation systems, respectively.
The resulting trajectory was period-corrected using gmxtrjconv utilities. Protein residue flexibility was monitored using root-mean-square fluctuation (RMSF) with the gmxrmsf module. The root-mean-square deviation (RMSD) of the backbone was calculated using the gmxrms module. Changes in the protein secondary structure were analyzed using gmxdo_dssp utilities. Principal component analysis (PCA) was analyzed using gmxcovar utilities. The radius of gyration (tool gmxgyrate) and 3D structure of the protein after simulation (tool gmxtrjconv) were also analyzed using GROMACS from the resulting trajectories. The RMSD Trajectory Tool module of Visual Molecular Dynamics (VMD) version 1.9.3 was used to align the frame after the simulation. VMD and PyMOL were used for molecular visualization.

| Protein expression and enzymatic properties
Recombinant TvEG3 was successfully produced in P. pastoris GS115. Subsequently, recombinant proteins were purified using Ni-chelating affinity chromatography. Single bands were observed at 54 kDa in the supernatant of purified TvEG3 using SDS-PAGE. CtCBH and CtBglA were also successfully expressed ( Figure S1c). The optimum temperature for TvEG3 and CtBglA were both 70°C. The optimum pH for TvEG3 and CtBglA were 4.0 and 6.0, respectively ( Figure S1a,b,d,e). The optimum temperature and pH for CtCBH were 60°C and 6.0, respectively (Hu et al., 2021). The specific activity of TvEG3, CtCBH, and CtBglA was determined to be 51.97, 8.04, and 77.96 U/mg, respectively.

| Effects of syringic acid assisting on cellulases
3.2.1 | High temperature hydrolysis of cellulase The effects of syringic acid on the hydrolysis yield of cellulase at 70°C are shown in Figure 1. The reducing sugar yield of TvEG3 hydrolysis containing syringic acid showed an increase of 201.36% compared to the reducing sugar yield in the control group. In the CtCBH hydrolysis system, the presence of syringic acid did not affect the release of reducing sugars. The reducing sugar yields of the control group and the group with syringic acid were 0.16 and 0.15 mg/mL, respectively. No significant difference was observed in reducing sugar yield. In the hydrolysis system of CtBglA, the pNP yield of the reaction with syringic acid was 0.32 μM, respectively, which showed a decrease of 72.79%, compared to that in the control group. The presence of syringic acid increased the hydrolysis yield of EG and decreased that of BGs. Syringic acid did not influence the CBH hydrolysis. Moreover, as shown in Figure 1d, in hydrolysis of FP by cellulase mix , the presence of syringic acid increased the reducing sugar yield by 28.37%. Syringic acid increased the hydrolysis activity of cellulases significantly at 70°C. TLC analysis showed that the hydrolytic product of the control group was mainly glucose ( Figure S2a). The addition of syringic acid produced more reducing sugars and a significant amount of cellobiose. The promotion of syringic acid on TvEG3 hydrolysis allowed for a large accumulation of cellobiose. More cellobiose presented in the reaction mixture indicating syringic acid has positive effect on EG.

| Effect of syringic acid on thermostability
TvEG3 exhibited poor thermostability. After incubation at 70°C for 5 min, reducing sugars were not observed during the enzymatic determination of TvEG3 (Figure 2a), and denaturation precipitate was observed in the protein solution ( Figure S2d). However, in the incubation system containing syringic acid, as incubation time increased, the relative activity of TvEG3 decreased gradually. The relative activity of TvEG3 incubated with syringic acid at 70°C for 5 min remained 56.31%, respectively (Figure 2a). Even when incubated for 25 min, the TvEG3 activity of in the presence of syringic acid was retained at 18.64%. Interestingly, at 28°C incubation, the amount of reducing sugars produced by TvEG3 with or without syringic acid was not significantly different within the margin of error (Figure 2b). This result indicated that the presence of trace amounts of syringic acid in the reaction system did not affect EG hydrolysis. Syringic acid slowed the inactivation of EG at 70°C and showed the potential to improve the thermal stability of cellulase in this study. Syringic acid did not affect the hydrolysis of CtCBH. The addition of syringic acid also did not exhibit any variation in studies of the thermal stability of CtCBH (data not shown).
CtBglA has excellent thermostability; it retained 100% hydrolytic activity in a water bath at 70°C for 30 min compared to the initial enzyme activity ( Figure S1f). As shown in Figure 2c, the amount of pNP generated by CtBglA after incubation with 20 mM syringic acid was undetectable. This indicates that CtBglA lost its hydrolytic activity in the presence of syringic acid after incubation for 30 min, and a distinctly visible protein denaturation precipitate was observed in the protein solution ( Figure S2c). The degree of inhibition decreased as the concentration of syringic acid decreased. Also, at 28°C incubation, the amount of pNP produced by CtBglA with or without syringic acid was not significantly different within the margin of error (Figure 2d). This result also indicated that the presence of trace amounts of syringic acid in the reaction system did not affect BG hydrolysis.

| Circular dichroism spectra
Syringic acid slowed down the deactivation of TvEG3 at high temperatures (Figure 2a). CD analysis was performed on TvEG3 and TvEG3-syringic acid. TvEG3 shows two negative peaks at 210 and 219 nm, with ellipticities of −10.89 and −11.16 mdeg, respectively, and a positive peak at 194 nm with of 12.69 mdeg. TvEG3-syringic acid shows two negative peaks at 210 and 219 nm, with ellipticities of −7.53 and −7.39 mdeg, respectively, and a positive peak at 194 nm with of 7.0551 mdeg ( Figure S2b). TvEG3 was completely inactivated by incubation at 70°C for 5 min in the absence of syringic acid, producing a visual protein denaturing precipitate ( Figure S2d). The CD spectral results F I G U R E 1 Effects of syringic acid on hydrolysis yield of TvEG3 (a), CtCBH (b), CtBglA (c), and cellulase mix (d) at 70°C. The reaction mixture of the control group does not contain syringic acid. Error bars represent the standard deviation of three replicates. Significant difference ***p < 0.001 (Student's t-test). pNP, p-nitrophenol. also demonstrated the ability of syringic acid to maintain the biological structure of EG at 70°C.

| Sequence alignment and homology modeling
TvEG3 showed a high degree of homology with Hypocrea jecorina Cel5A (99.39% identity). The structure of TvEG3 was obtained using the x-ray structure of H. jecorina Cel5A (PDB ID: 3QR3) as a template. CtCBH showed a high degree of homology with C. thermophilum Cel6A (99.45% identity). The structure of CtCBH was obtained using the xray structure of C. thermophilum Cel6A (PDB ID: 4A05) as a template. CtBglA showed a high degree of homology with Ruminiclostridium thermocellum βglucosidase A (99.55% identity). The structure of CtBglA was obtained using the xray structure of R. thermocellum βglucosidase A (PDB ID: 5OGZ) as a template. As shown in Figure 3, TvEG3, CtCBH and CtBglA have similar structural architectures which is F I G U R E 2 Effect of syringic acid on the thermostability of cellulase. Changes in the relative enzyme activities of TvEG3 with syringic acid over time (a). Reducing sugar yield of TvEG3 with syringic acid treatment at 28°C for 30 min (b). p-Nitrophenol yield of CtBglA with different syringic acid concentrations treatment at 70°C (c) and 28°C (d) for 30 min.

F I G U R E 3
Structural architecture of (a) TvEG3, (b) CtCBH and (c) CtBglA. Colored by second structure of cellulase, helix is shown in cyan, sheet is shown in purple, loop is shown in salmon, red circle indicates random coil loop. characteristic of glycoside hydrolase fold. All three of those cellulases display the distorted β/α barrel. The βsheets form the central core of the protein and αhelixes cover this core at the protein surface. The differences of those three enzymes are αhelixes of CtBglA completely cover the surface of the enzyme, and the surface of TvEG3 and CtCBH contains many random coil loops. The structure difference of cellulases may be the reason why syringic acid has different effects on different cellulases.

| Conformational analysis
The RMSD results for TvEG3 within 100 ns are shown in Figure 4a. The RMSD quickly reached a balance after approximately 10 ns of MD simulation in the presence of syringic acid. The RMSD of the backbone fluctuated at 1.5 Å, indicating that the overall structure of TvEG3 remained intact in the presence of syringic acid. The RMSD of TvEG3 increased continuously in the simulated system without syringic acid addition. The structure of TvEG3 constantly deviates from its initial position in the absence of syringic acid. The flexibility of the local structure determines the vibration at equilibrium (Naz et al., 2018). The RMSF of TvEG3, which reflects the average fluctuations of all the residuals during the simulation, was plotted as a function of the number of residuals (Figure 4d). When the simulated temperature was 343.15 K, at positions 128-138 aa, 189-197 aa, 223-235 aa, 262-275 aa and 315-327 aa, a lower fluctuation was observed owing to stabilization in the presence of syringic acid. Two other regions, 297-303 aa and 312-313, showed higher fluctuation.
The RMSD and RMSF results for CtCBH within 100 ns are shown in Figure 4b,e, respectively. CtCBH exhibited continuous fluctuation from 1.00 to 2.00 Å; however, with the addition of syringic acid, the fluctuation was reduced to 1.25 Å from 1.60 Å and remained stable after 20 ns (Figure 4b). The RMSF behavior of the residues showed a similar trend. In the presence of syringic acid, the regions of residue 20-30, 90-100, 102-113, 149-154, 185-189, 228-239, 272-282, and 329-347 became inactive. The RMSD and RMSF results for CtBglA within 100 ns are shown in Figure 4c,f, respectively. In complete contrast to the results for the first two cellulases, the presence of syringic acid in the system caused CtBglA to become progressively unstable in a chaotic manner. The RMSD value of CtBglA was relatively stable and the fluctuation stabilized at 1.22 Å. Syringic acid interrupted this steady state with increasing fluctuations, with finally stabilized at 1.75 Å. In the presence of syringic acid, the RMSF of all residues increased overall, except for the region of residue 24-27.
The radius of gyration (R g ) for TvEG3, CtCBH, and CtBglA without syringic acid and in its presence at 343.15 K were plotted against the simulated time for illustration ( Figure S3). R g is a parameter associated with the tertiary structure of proteins and has applications that provide insight into the stability of proteins in biological systems during MD simulation. The results showed that the overall structures of TvEG3 and CtCBH were stable and compact throughout the simulation with and without syringic acid, with average fluctuations at 1.85 and 1.95 nm, respectively, during the trajectory evolution. The average R g of CtBglA and CtBglA in the presence of syringic acid were 2.13 and 2.14 nm, respectively.

| Principal component analysis
The projection of the two eigenvectors (eigenvectors 1 and 2) for TvEG3, CtCBH, and CtBglA in the absence and presence of syringic acid was constructed at 343.15 K ( Figure 5). The trace of the covariance matrix and eigen values of C-alpha for TvEG3 in the absence and presence of syringic acid were found to be 3.60 and 3.06 nm 2 , respectively. The trace of the covariance matrix and eigen values of C-alpha for CtCBH in the absence and presence of syringic acid were found to be 4.61 and 2.71 nm 2 , respectively. The trace of the covariance matrix and eigen values of C-alpha for CtBglA in the absence and presence of syringic acid were found to be 3.23 and 3.69 nm 2 , respectively. Cellulase in the presence of syringic acid behaved in an entirely different manner, with a trace of the covariance matrix value when compared to single cellulase. PCA was used for the MD simulation results for a wider range of dynamic characteristics. Clusters of steady state were observed in these projections, which reflected the expansion of the overall protein during simulation (Maisuradze et al., 2009;Srikumar et al., 2014). These results confirmed that syringic acid decreased the flexibility of TvEG3 and CtCBH, making it stable, but increased the flexibility of CtBglA.

| Secondary structure analysis
The secondary structures of TvEG3, CtCBH, and CtBglA were analyzed using MD. The secondary structural changes in TvEG3, CtCBH, and CtBglA obtained during the MD simulations are listed in Table 1. The percentage of amino acid residues involved in secondary structure formation of cellulase differed in the simulated system in the presence of syringic acid. The percentages of protein secondary structures in TvEG3 alone and in syringic acid were 65% and 66%, respectively. The percentages of protein secondary structures in CtCBH alone and in syringic acid were 61% and 60%, respectively. The percentages of protein secondary structures in CtBglA alone and present in syringic acid were 69% and 68%, respectively. These results indicated that the presence of syringic acid generated apparent changes in the secondary structure of cellulase. The conformations of TvEG3, CtCBH, and CtBglA in the presence or absence of syringic acid were superimposed after 100 ns of simulation, as shown in Figure 6. Based on the stimulation results of TvEG3, the percentage of amino acids involved in the αhelix structure did not change, but the position and length of the αhelix changed. The first αhelix near the C-terminus of TvEG3 became longer (S317-L323), and the third and fourth αhelices near the N-terminus of TvEG3 became shorter in the presence of syringic acid. Generally, more stable helices αhelices are responsible for strong internal packing thus showing higher thermostability (Imanaka et al., 1986;Nicholson et al., 1988). Additional αhelices were generated at amino acids 183-187 and 328-331 in CtCBH in the presence of syringic acid. The presence of syringic acid led to the disappearance of the first αhelix at the N-terminus of CtBglA and the shortening of the second αhelix structure. Simultaneously, structural differences in the βsheet were observed from conformational superposition.
T A B L E 1 Percentage of residues of TvEG3, CtCBH, and CtBglA in the absence or presence of syringic acid was involved in the formation of average structure during 100 ns MD simulations.

| Modification for cellulase with syringic acid
It was found that three cellulases were surrounded by some syringic acid molecules during MD simulation at 343.15 K (Figure 7). During the simulation of TvEG3 and CtCBH, the syringic acid molecule tended to gather around the random coil region of enzyme surface (Figure 7a,b). Syringic acid was closed to THR35, PHE34, TYR28, TYR40, ASN39, SER296, ASP298, THR306, and THR308 of TvEG3 and LYS39, ASN385, ARG293, ASN98, ASP89, LYS150, SER102, and GLN149 of CtCBH (within 4.0 Å) and might connect with above polar residues through weak interactions, such as hydrogen bonds, electrostatic, and Vander Waals interactions (Cao et al., 2017;Uversky & Dunker, 2010). Unlike TvEG3 and CtCBH, the surface of CtBglA is almost surrounded with αhelixes, which may also account for its structural stability at high temperatures. Syringic acid attached to LYS90, GLN91, ASN92, LEU93, LYS94, and ARG256 of CtBglA. The resulting external weak interaction forces at high temperatures may cause the originally tight protein structure to be destroyed. Also, CtBglA lost its original rigid state and became extremely unstable under high temperature F I G U R E 7 The interaction of syringic acid with surface region of TvEG3 (a), CtCBH (b) and CtBglA (c) was amplified exhibited. Blue ribbon denotes the cellulase; Grey sphere denotes the syringic acid molecule; Yellow stick denotes the polar residues of cellulase.
conditions resulting in denaturing precipitation, thus losing its hydrolysis activity.

| DISCUSSION
Currently, the lack of cellulase that work efficiently and cheaply at high temperatures for the bioconversion of lignocellulosic materials to bioethanol is one of the major bottlenecks in the production of biofuels. It is crucial that cellulase is more stable and active under high temperature conditions. In our previous studies, the hydrolysispromoting effect of phenolic acids on cellulase was detected (Ran et al., 2022). The positive effect of phenolic acids on cellulase is mainly reflected in the increase of specific enzyme activity. This indicates that phenolic acid alters the enzymatic properties of cellulase and may change the structure of the protein.
Consistent with our hypothesis, we found that syringic acid also showed a significant increase in the hydrolytic activity of cellulase under high temperature conditions (at 70°C). Syringic acid showed a distinct promoting effect on TvEG3 and an inhibitory effect on CtBglA. Meanwhile, the presence of syringic acid had little effect on the hydrolysis of CtCBH. From the results of MD simulations, syringic acid brought more secondary structures in TvEG3 made them more stable at high temperatures. In addition, the reduced RMSD value of the backbone was also considered as an effective parameter to verify stronger structural rigidity, which could enhance the enzyme thermostability (Park et al., 2014). TvEG3 is prone to denaturation and precipitation rapidly at high temperature and thus loses its hydrolytic ability. Syringic acid tends to be close to the random coil loop of TvEG3 and CtCBH surface. The weak interactions between syringic acid molecule and residue causes a change in the secondary structure. Both the increased αhelix content and secondary structure number reveal the improvement of thermostability of TvEG3 and CtCBH. The presence of syringic acid enhances the thermostability and delays the denaturation and deactivation of TvEG3, allowing TvEG3 to maintain its hydrolytic ability for a longer period of time. In contrast, syringic acid comes into contact with residues in the secondary structure of the CtBglA surface, and the additional interaction destabilizes the original structure. Thereby, CtBglA produces precipitation and loses its enzymatic hydrolysis ability under high temperature conditions. Syringic acid has the potential to stabilize the high temperature activity of cellulase by slowing down thermal denaturation that provides new ideas for biotransformation under high temperature conditions. The significant promoting effect of syringic acid on cellulase under high temperature hydrolysis plays an important role in broadening the industrial application scenarios of cellulase.

| CONCLUSION
Syringic acid exhibits promoting effect on cellulase mixture hydrolysis activity containing EG, CBH, and BG at elevated temperature (70°C). Syringic acid tends to be close to the random coil loop of TvEG3 and CtCBH surface. The weak interactions between syringic acid molecule and residue causes a change in the secondary structure. According to experimental and simulation analysis, both the increased αhelix content and secondary structure number revealed the improvement of thermostability of TvEG3 and CtCBH. In contrast, syringic acid comes into contact with residues in the secondary structure of the CtBglA surface, and the additional interaction destabilizes the original structure. The reduction in the number of secondary structures, especially αhelix, caused the structure of CtBglA in the presence of syringic acid to collapse at high temperature. The improvement of syringic acid on the cellulase mixture activity is mainly manifested in stabilizing EG structure at high temperature. This study provides a new strategy for cellulase applied in high temperature bioprocessing with syringic acid assisting.