Understanding the pH-Dependent Reaction ... - ACS Publications

great importance for protein engineering and drug design. ... INTRODUCTION. 36 ... ACS Catalysis. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 1...
0 downloads 0 Views 1MB Size
Subscriber access provided by UNIV OF DURHAM

Article

Understanding the pH-Dependent Reaction Mechanism of a Glycoside Hydrolase Using High Resolution X-Ray and Neutron Crystallography Zhihong Li, Xiaoshuai Zhang, Qingqing Wang, Chunran Li, Nianying Zhang, Xinkai Zhang, Birui Xu, Baoliang Ma, Tobias E. Schrader, Leighton Coates, Andrey Y. Kovalevsky, Yandong Huang, and Qun Wan ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.8b01472 • Publication Date (Web): 18 Jul 2018 Downloaded from http://pubs.acs.org on July 19, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

1

Understanding the pH-Dependent Reaction Mechanism of a Glycoside

2

Hydrolase Using High Resolution X-Ray and Neutron Crystallography

3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39

#

#

Zhihong Li,¶ Xiaoshuai Zhang,¶ Qingqing Wang,¶ Chunran Li,¶ Nianying Zhang,¶ Xinkai Zhang,¶ Birui Xu,¶ Baoliang Ma,¶ Tobias E. Schrader,† Leighton Coates,‡ Andrey Kovalevsky,‡ Yandong Huang,*,§ Qun Wan*,¶ ¶ College of Science, Nanjing Agricultural University, Nanjing, China 210095 † Jülich Centre for Neutron Science at Heinz Maier-Leibnitz Zentrum, Forschungszentrum Jülich GmbH, Garching, Germany 85747 § College of Computer Engineering, Jimei University, Xiamen, China 361021 ‡ Neutron Scattering Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831 # These authors contributed equally to this work. ABSTRACT: Glycoside hydrolases (GHs) commonly use the retaining or inverting mechanisms to hydrolyze carbohydrates and the rates of catalysis are usually pH-dependent. Deeper understanding of these pH-dependent reaction mechanisms is of great importance for protein engineering and drug design. We used high-resolution X-ray crystallography to analyze the sugar ring configurations of an oligosaccharide ligand during hydrolysis for the family 11 GH and the results support the 1S3→4H3→4C1 conformational itinerary. These results indicate that sugar ring flexibility may help to distort and break the glycosidic bond. Constant pH molecular dynamics simulations and neutron crystallography demonstrate that the catalytic glutamate residue (E177) has alternate conformational changes to transfer a proton to cleave the glycosidic bond. Furthermore, the neutron crystallography analysis shows that the H-bond length between E177 and its nearby tyrosine residue (Y88) is shortened when pH increases, preventing E177 from rotating downward and obtaining a proton from solvent for catalysis. This result indicates that the H-bond length variation may play a key role in the pH-dependent reaction mechanism. In summary, our results demonstrate that both sugar ring flexibility and protein dynamics are important in the pH-dependent reaction mechanism and may help to engineer GHs with different pH optima.

KEYWORDS: glycoside hydrolase, pH-dependent reaction mechanism, neutron crystallography, constant pH molecular dynamics, H-bond length INTRODUCTION Glycoside hydrolases (GHs) are efficient enzymes to break the glycosidic bond which forms the linkage between sugar subunits in polysaccharides or oligosaccharides. GHs commonly use the retaining or inverting mechanisms to hydrolyze the glycosidic 1

ACS Paragon Plus Environment

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41

bond: One catalytic Glutamate or Aspartate residue functions initially as an acid to donate a proton to break the glycosidic bond. The other glutamate residue works as a nucleophile to help the formation of an oxocarbenium ion-like transition state.1-3 The activities of GHs are regulated by pH, temperature and ionic conditions.4-6 Understanding the reaction mechanism of GHs is important in protein engineering for bioenergy and designing inhibitors for drug discovery.7-9 Xyn II is an endo-β-1,4-xylanase (EC 3.2.1.8) secreted by Trichoderma reesei and belongs to the Family 11 GH.10-11 It has been well-studied using structural, kinetic, and biophysical techniques.10-13 Xyn II has a jelly-roll structure resembling a closed right-hand architecture.10 E177 functions as a general acid/base and E86 works as a nucleophile during catalysis (Figure S1). Previously, we obtained an inactive mutant, E177Q, which is unable to hydrolyze the substrate xylohexaose (X6), because the glutamate carboxyl group has been replaced with an amide group and is unable to function as a proton donor.5 Thus, the crystal structure of E177Q in complex with xylohexaose mimics the pseudo-Michaelis complex.5 In this structure, the -1 site xylose of the oligomer is twisted from the original 4C1 configuration toward an oE configuration. However, QM/MM molecular dynamics simulations have indicated that the oE configuration may be an artifact caused by the amide substitution.7 To address this issue, we have obtained a wild type (WT) Xyn II complexed with the hydrolyzed product, xylotriose (X3). The configuration of the -1 site xylose is twisted from 4C1 toward 4H3. Y77 and Y88 form H-bonds with E86 and E177, respectively. Mutating these tyrosine residues to phenylalanine residues significantly decreases the kinetical activies.13 We obtained a structure of an Y77F-X3 complex, which has less H-bond interactions between the xylose units of X3 and the active site residues. The -1 site xylose has the configuration of 4C1. These structures support the QM/MM calculated itinerary of 1 S3→4H3→4C1 for the family 11 xylanase during catalysis.8 Xyn II has the optimal pH range of 5.0 ‒ 5.5 for xylan hydrolysis.14-15 E177 initially works as a general acid to provide a proton to break the glycosidic bond. In the next step, it accepts a proton from solvent as a general base. At an optimal reaction pH, the proton-shuttling rate could be the highest when its pKa value is consistent with the optimal pH. We have calculated the conformational changes by constant-pH molecular dynamics (CpHMD), which can calculate pKa values of highly buried residues in the catalytic center.16 The CpHMD simulations calculated the macroscopic pKa of E177 to be 5.6, which is close to the optimal pH range. This simulation result validates the pH-dependent conformational changes of E177 as observed in our previous neutron crystal structures.17 Neutron crystallography is a powerful tool to directly observe the positions of protium/deuterium atoms in protein structures even at moderate resolutions.18-19 We have obtained a neutron crystallography structure of Xyn II at pD 5.4 and analyzed the H-bond length between Y88 and E177. Combined with our previous neutron crystallography structures, we propose that in alkaline conditions, the H-bond between Y88 and E177 2

ACS Paragon Plus Environment

Page 2 of 23

Page 3 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

1 2 3 4 5 6 7 8 9

stabilizes the upward conformation of E177, making it energetically unfavorable to rotate its side chain downward to obtain a proton from solvent. In weakly acidic conditions, this H-bond is weakened and E177 can rotate downward to pick up a proton from solvent and quickly rotate back to the upward conformation to donate the proton to break the glycosidic bond. In acidic conditions, E177 is stabilized in the downward conformation and forms an H-bond with a water molecule, which serves as a proton donor. Our CpHMD simulations and neutron crystallography are consistent and give more details of the pH-dependent catalytic mechanism.

10 11 12 13 14 15

MATERIAL AND METHODS Cloning and mutagenesis Xylanase genes encoding the amino-acid sequence 2-190 of the native Trichoderma reesei Xyn II protein were codon-optimized and synthesized by Genewiz Co. Ltd. They were cloned into the expression vector pET-NTMST using the Bam HI and Hind III restriction sites, which was modified based on the vector pET-28(b). The modified vector contains the gene sequence of the 6×His tag, the NusA protein, and the TEV cleavage site (sequence: ENLYFQS) sequentially at the N-terminus for expression of soluble xylanase and convenient purification. Site directed mutagenesis was conducted by synthesizing a primer containing a mutation. The primer was elongated by PCR (Biometra Tone, Germany) using the wild type xylanase construct as the template. After gel-electrophoresis, the mutated gene was harvested and the nick was ligated using a ligase kit (Sosoo Mix, TsingKe Co, Ltd, China). After sequence confirmation, the construct was amplified and purified using a Plasmid Miniprep Kit (TsingKe Co, Ltd, China). Xylanase expression and purification Xylanase fused with the 6×His tag, the NusA protein, the TEV cleavage site at the N-terminus was expressed with RosettaTM (DE3) competent cells (Novagen) in the minimal media.20 The cells were lysed with a high-pressure cell press (Union Co., China) and clarified with high speed centrifugation. Xylanase was purified with the immobilized metal-affinity chromatography (IMAC).21 The fused protein was initially purified with the nickel affinity chromatography. After adding 1% (mass ratio) TEV and dialyzed overnight, the flow-through containing the cleaved xylanase was dialyzed against the Tris-HCl buffer (100 mM Tris, 100 mM NaCl, 1 mM dithiothreitol, pH 8.0) and concentrated to 30 mg mL-1. The protein concentration was determined using the absorbance measurements at 280 nm (ε280 = 583330 M-1cm-1). Steady-state enzyme kinetics Discontinuous assays of xylanase activity were performed using the 3,5dinitrosalicylic acid (DNS) reagent to determine the concentration of reducing sugars following enzymatic hydrolysis with minor revision5, 22: 1% bagasse xylan (w/w) (Shyuanye Co., China) was incubated at 55 °C for 5 min. After adding xylanase, the reaction was carried out at 55°C for 8 min and was terminated at 100 °C with the DNS

16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41

3

ACS Paragon Plus Environment

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41

reagent. After cooling down to room temperature, the released sugar concentration was measured using its absorption at OD540 nm. A standard curve containing 0–1 mg mL-1 D-xylose was used to calculate the reducing-sugar concentrations. The rate of enzyme catalysis was calculated after subtracting the background concentration of the reducing sugar measured in the control reaction without enzyme. Vmax of WT and its variants was determined by fitting the nonlinear regression of the Michaelis-Menten-Henri equation using Origin 8.0 (OriginLab, USA). X-ray crystallography WT and all variants were crystallized using the hanging-drop method: 1 µL protein (20 mg mL-1 xylanase, 0.1 M Tris, 0.1 M NaCl, 1mM DTT, pH 8.0) solution was added to 1 µL reservoir solution and equilibrated against 0.5 mL reservoir solution. For the WT crystal in the apo state, the reservoir solution consisted of 20% PEG 8000, 0.2 M NaI, 0.1 M MES, pH 6.0. For the Y88F crystal in the apo state, the reservoir solution consisted of 20% PEG 8000, 0.2 M NaI, 0.1 M NaAc, pH 6.0. For the Y77F in the apo state, the reservoir solution consisted of 20% PEG 8000, 0.2 M NaI. Crystals of xylanase complexed with xylotriose were obtained by soaking apo-state crystals into a reservoir solution containing 200 mM xylotetraose or xylohexaose and 25% glycerol for 10 minutes. Crystals were flash frozen in liquid nitrogen. X-ray diffraction data were collected on the beamlines BL19U1 and BL18U1 of Shanghai Synchrotron Radiation Facility (SSRF). Data reduction was carried out using the HKL3000 program.23 All the structures in the absence and presence of ligands using the soaking method were directly refined using the program phenix.refine.24 The initial starting model for refinement was from the PDB entry 2DFB13 with all waters and non-bonded ions removed. After several rounds of refinement interspersed with manual model building using Coot,25 the electron density for the ligands was clearly visible in both the 2Fo – Fc and the Fo – Fc omit maps. The N44H-X3 model (PDB: 4HK9) was aligned against the refining model using the secondary structure matching (SSM) function26 incorporated into the Coot molecular graphics program. The coordinates of the xylose units after superposition were used to fit the residual density of the Fo – Fc map and was incorporated into the model for further refinement. The ligand restraints were generated using the tool eLBOW27 from the Phenix program suite.28 The Ramachandran plot analysis was performed using the MolProbity29 program. The data processing and structure-refinement statistics is shown in Table S1. Figures were generated using Pymol (The PyMOL Molecular Graphics System, Version 1.3 Schrödinger, LLC). The xylose conformation and puckering parameters were calculated using the online webserver (http://enzyme13.bt.a.u-tokyo.ac.jp/CP/) according to the Cremer–Pople formalism30 and were compared with the dihedral angles of xylose units in xylan. The atomic coordinates and the structure factors have been deposited in the Protein Data Bank as entries 5ZHO for WT in the apo state, 5ZF3 for the WT-X3 complex structure, 5ZH9 for the Y88F mutant in the apo state, 5ZII for the Y88F-X3 complex 4

ACS Paragon Plus Environment

Page 4 of 23

Page 5 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40

structure, 5ZIW for the Y77F mutant in the apo state, and 5ZKZ for the Y77F-X3 complex structure. pKa calculations The H++ web server version 3.0, which uses the Poisson–Boltzmann continuum electrostatics approach to approximate the electrostatic environment of biological macromolecules, was used to compute the pKa values of ionizable residues in the different ligand-binding states.31-33 The pKa calculations were carried out assuming the salinity of 0.15 M, the internal dielectric constant of 10, the external dielectric constant of 80, pH 4.4 (E177 in the downward conformation) or pH 8.5 (E177 in the upward conformation). For WT, Y77F, and Y88F in the apo and X3-bound states, pKa values were calculated based on their crystal structures, respectively. For the WT-X6 complex, the calculation was based on the E177Q-X6 structure (PDB: 4HK8) and Q177 was mutated back to E177 to mimic the Michaelis-Menten complex structure. For the Y77F-X6 and the Y88F-X6 structures, Y77 and Y88 are respectively mutated to F based on the previous WT-X6 structure. Continuous constant-pH molecular dynamics (CpHMD) simulations of WT were carried out to demonstrate the pH-dependent conformational changes in the catalytic site.16 The hybrid-solvent CpHMD was employed that applies the particle mesh Ewald (PME) method to estimate the long-range electrostatics for conformational changes. The Generalized-Born implicit-solvent model with simple switching (GBSW) is used to account for the solvent effect in titration dynamics.34 CpHMD simulations were performed with the CHARMM package (version c42b2).35 The periodic boundary condition was satisfied and the protein was solvated in an octahedral water box filled with modified TIP3P water molecules. The CHARMM22 force field combined with the dihedral energy correction map (CMAP) was used to account for the intra-molecular interactions in protein.36-37 The initial structure (PDB ID: 4S2F) was the crystal structure solved in the acidic condition at pD 4.8.17 The C-terminus was capped with CT3. Fifteen chloride and nine sodium ions were added to neutralize the initial state and mimic the physiological ionic strength of 0.15 M. To optimize the hydrogen positions, the heavier atoms (C, O, N, and S) were fixed and the hydrogen atoms were restrained with the force constant of 5 kcal/ml•Å2. The steepest descent method followed by the adopted basic Newton-Raphson method was performed to minimize the energy such that the unphysical contacts in the X-ray structure were removed. A pairwise heating process from 100 K to 300 K with the incremental step of 5 K was performed where the heavy atoms were restrained harmonically with the force constant of 5 kcal/ml•Å2. After gradually reducing the restraint, the system was equilibrated for another 500 ps at 300 K with the last 300 ps being unrestrained. The SHAKE algorithm was employed which allows the time step of 0.002 ps. The pH-based and replica-exchanged protocol was employed to enhance the sampling where pH ranges from 1.0 to 8.5 with the interval of 0.5.34 As a result, the total number of replicas is 16. The yielded exchange ratio is higher than 25%. The running 5

ACS Paragon Plus Environment

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1 2

length for each replica was 5 ns. The last 3 ns were extracted to compute pKa using the following two equations:

3

S=

4

Note:

5

obtained in CpHMD simulations. 1 S= (2) n ( pKa − pH ) 1 + 10 Note: n is the Hill coefficient that reflects coupling of titratable groups. pKa is computed by fitting the deprotonated fraction S at different pH to the Hill equation. Molecular Dynamics (MD) simulations To examine the effect of the mutations (Y77F and Y88F) on structural flexibility, MD simulations of the WT, Y77F, and Y88F mutants were carried out with GROMACS 5.1.4: All water molecules and iodine ions were removed from the initial model (PDB: 2DFB). The protein was solvated in a cubic water box filled with the TIP3P water molecules. Twenty five sodium and twenty seven chloride ions were added to neutralize the system and mimic the physiological ionic strength of 0.15 M. PME was employed to estimate electrostatics under the periodic boundary condition. The AMBER99SB force field was employed for all elements in the simulation cell, including protein, salt ions and water molecules. Prior to MD simulations, an energy minimization with the steepest descent method followed by the adopted basic Newton-Raphson method was performed to remove the steric clashes and the strains in the X-ray crystallography structures. The LINCS algorithm was used to restrain the bonds regarding the hydrogen atoms, which allows the time step of 0.002 ps. Next, a 100 ps NVT equilibration was performed followed by a 100 ps NPT equilibration at 300 K. Finally, three independent runs starting with different velocity distributions were carried out for each system and the running length was 100 ns. Neutron crystallography A large crystal was obtained using the sitting drop method: 30 µL protein (30 mg -1 mL xylanase, 0.1 M Tris, 0.1 M NaCl, 1 mM DTT, pH 8.0) was added to 30 µL reservoir solution (2% PEG 3350, 0.2 M NaI). After one month, the crystal grew to its maximum size (2 × 1.8 × 1 mm). The crystal was then sealed in a capillary with the D2O exchanged solution (2% PEG 3350, 0.2 M NaI, 0.1 M NaAc, pD5.4) for pH equilibration for two weeks before data collection (pD = pH + 0.4). Monochromatic neutron diffraction data were collected using the BIODIFF beamline at the FRM II research reactor at the Heinz Maier-Leibnitz Zentrum (MLZ). At a wavelength of 2.66 Å, 143 frames with an oscillation of 0.4° and an exposure time of 53 minutes per frame at the temperature of 295 K were recorded. To increase the data completeness and redundancy, the crystal was re-oriented, and a second series of 40

6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37

N dep N dep + N pro

(1)

N dep and N pro are the numbers of the deprotonated and protonated states

6

ACS Paragon Plus Environment

Page 6 of 23

Page 7 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22

frames were collected using the same oscillation range and exposure time. The diffraction data were indexed and integrated using HKL2000.38 The neutron refinement was performed using the program suite Phenix.28 All the water, ion and anisotropic flags were removed from the initial model (PDB ID: 2DFC).13 After the rigid body refinement, several cycles of the positional, the atomic displacement parameter and the occupancy refinement were performed. In the meantime, the refined model was manually adjusted according to both the 2Fo-Fc and Fo-Fc nuclear density maps using the program Coot.25 Protium and the exchangeable deuterium atoms were assigned to the protein, followed by refinement and manual adjustment as above. According to the nuclear density and hydrogen bond geometry, the water molecules were built into the model as D2O and refined. The final model was deposited to the Protein Data Bank as the entry of 5ZO0.

RESULTS Kinetic studies Bagasse xylan was used as the substrate. The Xyn II activity vs. pH shows a bell-shaped profile (Figure S2A). The Michaelis-Menten steady-state kinetic parameter, Vmax, was determined for WT and the mutants in the discontinuous assay at 50 °C and pH 5.0 (Table 1 and Figure S2B). The Y88F and Y77F mutants only have 0.6% and 0.02% hydrolysis activity compared with WT in the same condition. Table 1. Steady State Kinetic Parameters in the Discontinuous Assay Vmax (µmol min-1 mg-1)

Relative activity (%)

WT

6981±185

100

Y88F

39.6±0.2

0.6

Y77F

1.5±0.1

0.02

23 24 25 26 27 28 29 30 31 32 33 34 35 36 37

Structural analysis X-ray crystallography of WT in the apo and xylotriose (X3) bound states WT of Xyn II is catalytically active and cannot be co-crystallized with the substrate as it will rapidly hydrolyze the substrate. We soaked xylotetraose (X4) or xylohexaose (X6) into the apo crystals and observed xylotriose (X3) occupied at the -3, -2, and -1 xylose-binding sites in the active site (Figure 1). The crystal complexed with the X3 product diffracted to 1.2 Å resolution (Table S1). Overall, the WT-X3 complex structure is very similar with that of the apo state: R.M.S.D. of the Cα atoms is only 0.62 Å after superposition. The most significant differences are in the ‘thumb’, the ‘finger’ and part of the ‘palm’ regions that interact with X3 with R.M.S.D. of 1.65 Å, 0.83 Å and 0.71 Å, respectively. Due to the crystal packing at the ‘palm’ region, part of the ‘cord’ region also has significant changes (R.M.S.D. of Cα atoms is 0.87 Å). These residues are moving toward the ligand due to H-bonds, van der Waals interactions, and C-H⋅⋅⋅⋅π interactions.5 These conformational changes induced by ligand binding are consistent with those when 7

ACS Paragon Plus Environment

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1 2 3 4 5 6

7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

the co-crystallized N44H-X3 structure (PDB: 4HK9) is compared with the N44H apo-state structure (PDB: 4HKL).5 When the WT-X3 structure is compared to the E177Q-X6 structure, R.M.S.D. of the Cα atoms of the two structures is 0.24 Å. In the WT-X3 structure, due to the absence of the xylose units from the +1 to +3 sites, residues such as P68 and Y96 move away from the active site compared to their positions in the E177Q-X6 structure.

Figure 1. Structural comparison among the WT-X3, the WT apo structures (A-C), and the E177Q-X6 structures (D-F). The structural changes incurred by X3 binding are colored in magenta in the WT-X3 complex (A). The H-bond interactions among X3, E177, E86, Y88, and Y77 in the WT-X3 structure (B). The H-bond interactions among E177, Y88, E86, and Y77 in the apo-state structure (C). Comparison between the E177Q-X6 and the WT-X3 structures (D). The H-bond interactions among X6, Q177, E86, Y88, and Y77 in the E177Q-X6 structure (E). The configuration differences of the -1 site xylose incurred by the E177Q mutation (F). The Fo-Fc omit maps of (a) and (d) were contoured at 3σ while all H-bond lengths are shown in angstrom.

In the active site of the WT-X3 structure, the positions of E177 and E86 do not have large conformational changes compared to those in the apo state and the E177Q-X6 structures. However, there are small but interesting differences between the catalytic residues and their interacting ligands (Figure 1F and Figure S3A). In the WT-X3 complex, the OE2 atom of E177 forms a strong H-bond (2.5 Å) with the O4A atom of the -1 site 8

ACS Paragon Plus Environment

Page 8 of 23

Page 9 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

xylose. In contrast, in the E177Q-X6 structure, the NE2 atom of Q177 forms a weaker H-bond (2.9 Å) to the O4A atom of the -1 site xylose with a different side chain orientation. In the WT-X3 structure, the OE1 atom of E177 forms an H-bond (2.7 Å) with the phenolic oxygen of Y88. This distance is the same as those in the apo WT and the E177Q-X6 structures. In the WT-X3 structure, the OE1 atom of E86 forms an H-bond (2.9 Å) with the O2B atom of the -1 site xylose. The distance is the same as that in the E177Q-X6 structure. The distance between the OE2 atom of E86 and the phenolic oxygen of Y77 is 2.6 Å in all the apo WT, the WT-X3 and the E177Q-X6 structures. Cremer-Pople analysis shows the configurations of the xyloses at different positions in the WT-X3 structure (Table 2). The xylose at the -1 position has a transitional configuration between 4C1 and 4H3. The xyloses at the -2 and the -3 sites are very close to the 4C1 configuration. In contrast, in the E177Q-X6 structure, the xylose at the -1 position has a configuration in the zone between 4C1 and oE, and all other xyloses have the 4C1 configuration.5 Table 2. Xylose Parameters in the Xylanase-X3 Complexes Puckering parameters

Dihedral angle a

Configuration

Φ (°)

Θ (°)

Q (Å)

-3

270

177

0.60

4

C1

-85

163

-2

202

174

0.57

4

C1

-103

157

-1

194

160

0.52

4

C1/4H3

108

172

0.58

4

C1

-117

162

0.58

4

C1

-121

152

C1 -82

167

-102

157

ϕ (°)

ψ (°)

WT-X3

Y77F-X3 -3 -2

92

-1

177

13

179

0.58

4

172

178

0.58

4

C1

0.56

4

C1

0.49

4

Y88F-X3 -3 -2

196

-1

17 18

a

222

174 155

4

C 1/ H 3

The dihedral angles are defined as ϕ (O5i—C1i—O4i+1—C4i+1) and ψ (C1i—O4i+1—C4i+1—C3i+1),

respectively.

19 20 21 22 23 24 25 26

X-ray crystallography of Y77F in the apo and xylotriose bound states The structures of Y77F and WT in the apo state are similar to each other without significant changes (R.M.S.D. of Cα atoms is 0.28 Å) (Figure 2A). In the active site, mutating Y77 to F does not disturb the crystal structure significantly: In the absence of the phenolic hydroxyl group, all the atoms of F77 have similar coordinates with those of Y77. Surrounding the mutated residue, E86, E177 and Y88 have almost identical coordinates to those in the WT structure (Figure 2A). 9

ACS Paragon Plus Environment

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22

Figure 2. Structural comparison between the WT and the Y77F structures. (A) Overview of the WT apo state vs. the Y77F apo-state structures. (B) H-bond interactions between E177, E86, Y88 and Y77 in the Y77F apo-state structure. (C) Structural differences incurred by X3 binding. (D) The WT-X3 structure vs. the Y77F-X3 structure. (E) The H-bond interactions among X3, E177, E86, Y88, and F77 in the Y77F-X3 structure. (F) The configuration differences of X3 incurred by the Y77F mutation. The Fo-Fc omit maps of (c) and (d) were contoured at 3σ while all H-bond lengths are shown in angstrom.

Y77F is still catalytically active, though its activity is 5000 times lower than WT (Table 1). We soaked X4 and X6 into apo-state crystals and obtained X3 bound at the -3, -2, and -1 positions in the active site (Figure 2C-2D). The best crystal diffracted to 1.3 Å resolution (Table S1). Comparing the mutant structures in the absence and presence of X3, an overall R.M.S.D. of Cα toms is 0.48 Å. Due to the interactions between the introduced X3 and the adjacent residues, the ‘thumb’ region, the ‘finger’ region, and part of the ‘palm’ region are moving toward the ligand (Figure 2C) and R.M.S.D. of Cα atoms of these regions is 1.32 Å, 0.68 Å, and 0.51 Å, respectively. These conformational changes are consistent with those comparing the WT-X3 and the WT apo-state structures. Comparing the Y77F-X3 structure and the WT-X3 structure, they are almost identical (R.M.S.D. of Ca atoms is only 0.10 Å) (Figure 2D). In the active site, in the absence of the phenolic hydroxyl group, F77 does not form H-bonds with the OE2 atom of E86 and the O2B atom of xylose at the -2 site as those of the WT structure. In addition, the –1 site xylose does not form an H-bond with the OE1 atom of E86 as that of the WT structure 10

ACS Paragon Plus Environment

Page 10 of 23

Page 11 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21

(Figure 2E and Figure S3B). In the absence of the above H-bonds, all xyloses at the -1, -2 and -3 sites have the configuration of 4C1 (Table 2). X-ray crystallography of Y88F in the apo and xylotriose binding states In the absence of ligand, the structures of Y88F and WT are similar to each other (Figure 3A) and R.M.S.D. of Cα atoms is 0.40 Å. In the active site of the Y88F apo-state structure, E177 does not form an H-bond with F88 as that in the WT structure. Thus, E177 has a different rotameric configuration and the F88 side chain moves away from E177. The H-bond distance between the OE2 atom of E86 and the hydroxyl group of Y77 is 2.5 Å (Figure 3B), which is similar with that in the WT structure.

Figure 3. Structural comparison between the WT and the Y88F structures. (A) Comparison between the WT and Y88F apo-state structures. (B) The H-bond interactions among E177, E86, F88, and Y77 in the Y88F apo-state structure. (C) Structural differences incurred by X3 binding in Y88F. (D) Structural comparison between the WT-X3 and Y88F-X3 structures. (E) The H-bond interactions among X3, E177, E86, Y77, and F88. (F) Configuration differences of X3 incurred by the Y88F mutation. The Fo-Fc omit maps of (c) and (d) were contoured at 3σ while all H-bond lengths are shown in angstrom.

Mutating Y88 to F has diminished xylanase activity to 400 times lower than WT (Table 1). After soaking X4 or X6 in the active site, the hydrolyzed product, X3, was found from the -1 to -3 sites (Figure 3A). After superimposing the Y88F-X3 to the Y88F apo-state structures, R.M.S.D. of Cα atoms is 0.41 Å. The most significant changes are 11

ACS Paragon Plus Environment

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

Page 12 of 23

assigned to the ‘thumb’, the ‘finger’ and part of the ‘palm’ regions (Figure 3C). R.M.S.D. of Ca atoms is 1.14 Å, 1.47 Å and 1.46 Å, respectively. These conformational changes are consistent with the comparison of the WT-X3 vs. the WT apo-state structures and the Y77F-X3 vs. the Y77F apo-state structures. The Y88F-X3 structure is very similar with the WT-X3 structure (Figure 3D): R.M.S.D. of Cα atoms is only 0.10 Å. These results indicate that mutation Y88 to F does not induce large conformational changes. In the active site of the Y88F-X3 structure, the hydroxyl oxygen atom of Y77 has one H-bond (2.5 Å) with the OE2 atom of E86 and the other H-bond (2.9 Å) with the O2B atom of the -2 site xylose. The O2B atom of the -1 site xylose has an H-bond (2.8 Å) with the OE1 atom of E86. These three H-bonds are similar with those in the WT-X3 structure. In the Y88F-X3 structure, in the absence of hydroxyl group, the mutated F88 does not form an H-bond with E177 as seen in the WT structure. The O4A atom of the -1 site xylose has a different coordinate and forms the first H-bond (3.1 Å) with the OE1 atom of E177, and the second H-bond (2.8 Å) with the OE2 atom of E86 (Figure 3E). Cremer-Pople analysis shows that the -1 site xylose in the Y88F-X3 structure has the transitional configuration between 4C1 and 4H3. The interactions among the -2 and the -3 site xyloses and the active site residues are similar with those in the WT-X3 structure with the low-energy configuration of 4C1 (Table 2). Microscopic pKa calculations We use the Poisson–Boltzmann continuum electrostatics approach to calculate the microscopic pKa changes of E177 in the WT, Y77F and Y88F.17 Because WT and its mutants are still catalytically active, we could only use the E177Q-X6 structure (PDB: 4HK8) as the template and mutate Q177 to E to mimic the WT-X6 structure. Based on this mutation, we further mutated Y77 to F and Y88 to F, respectively, to mimic the Y77F-X6 and Y88F-X6 structures. In the WT-X6 structure, E177 is proposed to have two alternate conformations: the downward conformation to obtain a proton from a nearby water molecule and an upward conformation to donate this proton to break up the glycosidic bond between the -1 and the +1 site xyloses.17 Our calculations support this hypothesis that in the downward conformation, pKa of E177 is 0.7 units higher than the upward conformation.17 In the Y88F-X6 structure, the pKa value of E177 is 9.9, which is 2.1 units higher than WT (Table 3). The result indicates that E177 is a much weaker acid in the Y88F mutant. Hence, it is a weak proton donor to break the glycosidic bond. The pKa calculation gives similar results for the Y77F mutant (Table 3). Table 3. Calculated Microscopic pKa Values of E177 in WT and the Different Mutants a,b

WT apo

Y88F

Y77F

WT

WT

-X3

-X6

apo

Y77F

Y77F

-X3

-X6

apo

Y88F

Y88F

-X3

-X6

upward

5.3

7.6

7.8

4.5

5.4

9.9

5.5

7.5

9.9

downward

6.0

8.9

8.5

6.3

5.4

9.0

6.2

7.4

9.9

12

ACS Paragon Plus Environment

Page 13 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

pK

1 2 3 4

a

+0.7

+1.3

+0.7

+1.8

0

-0.9

+0.7

-0.1

0

Calculations were based on the crystal structure of the E177Q-X6 complex (PDB ID: 4HK8), but its Q177

was mutated back to the native E177 in WT. In the Y77F-X6 and Y88F-X6 complexes, Y88 and Y77 were mutated to F, respectively, and Q177 was mutated to E. b

Calculation parameters: salinity = 0.15, εint = 10, εext = 80, pH 4.4 (downward) or pH 8.5 (upward).

5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

27 28 29 30

MD simulations MD simulation results show that the overall stabilities and thermodynamics of the active site do not have significant differences in WT, Y88F and Y77F (Figure S4-S5). Thus, the drastically decreased activities of Y77F and Y88F are not due to differences of their thermodynamics. CpHMD simulations give the macroscopic pKa that incorporates the conformational equilibrium in different pH environments. The macroscopic pKa of E177 is calculated to be 5.6 (Figure 4A and Figure S6), which is close to the optimal reaction pH range of Xyn II. According to the titration curve, E177 starts to be deprotonated at pH 3.5 and the titration is completed at pH 8.0 (Figure 4B and Figure S6). The dihedral χ2 angle formed by Cα-Cβ-Cγ-Cδ of the E177 side chain is calculated to study the conformational change in response to pH variation. Particularly, the downward conformation corresponds to the χ2 angle in the range of [-120°, -60°]. The upward conformation corresponds to the χ2 angle in the range of [-180°, -120°] or [120°, 180°] (Figure S7). From the CpHMD simulations, we found that the downward conformation dominates when E177 is protonated at low pH. When pH increases, the fraction of the downward conformation decreases and eventually disappears at pH 8.0. In the meantime, the fraction of the upward conformation increases in an opposite direction and becomes dominant at high pH, where E177 is deprotonated. Our calculations agree well with the structural results from neutron crystallography.17

Figure 4. pH-dependent conformational changes of the E177 side chain using the CpHMD simulations. (A) The titration curve and the calculated pKa of E177. (B) The probability of the E177 side chain in the downward (black square) or upward (red circle) conformation.

31 13

ACS Paragon Plus Environment

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Neutron crystallography of WT in the apo state at pD 5.4 WT of Xyn II was crystallized at pH 8.5, but was pH exchanged to pD 5.4 using the vapor equilibration method.17, 39 The crystal size is 2 × 1.5 × 0.6 mm (1.8 mm3) and it diffracted neutron to 1.65 Å at BioDiff (Table S2). Compared with the previously refined neutron structures at pD 6.2 and pD 8.9,17 the neutron structure at pD 5.4 is similar, but has small and important differences (Figure 5). In the active site, the deuterium atom associated with the phenolic oxygen of Y88 has an H-bond (2.2 Å) with the OE1 atom of E177 at pD 5.4. In contrast, the distance is shortened to 1.9 Å at pD 6.2 and 1.5 Å at pD 8.9, respectively. Particularly, this proton is equally shared between the OE1 atom of E177 and the phenolic oxygen to form a strong low barrier hydrogen bond (LBHB) at pD 8.9.17, 40 In the pD 4.8 structure, E177 is in the alternate ‘downward’ conformation where it can accept a proton from a nearby water molecule.17 The results indicate that the H-bond between E177 and Y88 is stabilized in alkaline conditions, but is weakened in neutral conditions and is further broken in an acidic environment. The deuterium atom associated with the phenolic oxygen of Y77 has an H-bond (1.6 Å) with the OE2 atom of E86 at pD 5.4. The distances are 1.7 Å in both the pD 6.2 and pD 8.9 structures. These results indicate that the H-bond between E86 and Y77 is stable in different pH environments.

19 20 21 22 23 24

Figure 5. The neutron structure of WT in the apo state at pD 5.4 (A) and the H-bond analysis among E86, E177, Y77, and Y88 in the different pD conditions (B). The 2Fo-Fc nuclear density map (cyan) is contoured at 1σ. The Fo-Fc omit map (magenta) is contoured at 3σ. The H-bond lengths in panel B are based on the neutron crystallography structures at pD 5.4 and our previous results at pD 8.9, pD 6.2, and pD 4.8, respectively.17 14

ACS Paragon Plus Environment

Page 14 of 23

Page 15 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41

ACS Catalysis

DISCUSSION GH11 uses the classical Koshland retaining mechanism to hydrolyze the glycosidic bond. An acid (Glu or Asp) functions as the proton donor/acceptor and a nucleophile (Glu) stabilizes the oxocarbenium intermediate state to accelerate hydrolysis.1-3 Xyn II from Trichoderma reesei utilizes E177 as the acid/base and E86 as the nucleophile.5, 11, 13 Conserved residues in the vicinity of the two catalytic glutamates influence their pKa values and activity.14 Particularly, Y88 and Y77 directly form H-bonds with the two catalytic glutamates. Mutating Y88 or Y77 to F decreases the enzyme activity dramatically. Here, we studied the functions of these important residues using high resolution neutron and X-ray crystallography, theoretical pKa calculations and molecular dynamics simulations. Our results indicate that these conserved tyrosine residues not only change the pKa value of the catalytic glutamates, but also change the conformation of the xyloses in the active site to facilitate catalysis. Mutating Y77 or Y88 to F does not incur significant structural changes either globally or locally in the active site. However, their enzymatic activities are dramatically decreased. The results indicate that the hydroxyl groups of Y77 and Y88 are important for catalysis. To investigate their roles in depth, it is important to analyze the interactions between the enzyme and the bound substrate. Though the activities of Y77F and Y88F are drastically decreased, they could still hydrolyze substrates in minutes (Table 1). Thus, we could only obtain the xylanase-product (X3) complex structures. We have tried co-crystallization of WT, Y88F, and Y77F with xylohexaose (X6) or xylotetraose (X4). However, their crystal structures show that there is a Tris molecule bound in the active site which occupied the position of the -1 xylose site.5 These structures are not ideal to analyze the interactions between the active site residues and the xylose subunits. Here, we obtained the high-resolution X-ray crystal structures of the WT-X3, Y77F-X3 and Y88F-X3 complexes by soaking apo-state crystals with high concentrations of X4 or X6 in limited time (Table S1). The hydrolyzed product, xylotriose, occupied the -1, -2, and -3 sites of the non-reducing end in the active site in all our complex structures. According to our previous E177Q-X6 structure5 and the ITC results41, there are more interactions in the -1, -2, +1, and +2 sites than the -3 and +3 sites. In other words, the soaked X4 would occupy from the -2 to the +2 sites and be hydrolyzed to X2, not X3 shown in our complexed structures. We propose that because of crystal packing, the substrate cannot enter the active site from the front side between the ‘thumb’ and the ‘finger’ as it would in solution. Instead, the oligosaccharide would enter the active site from the non-reducing end and span from the -3 to the +1 sites during the ligand soaking. After breaking down the glycosidic bond between the -1 and +1 sites, the hydrolyzed xylose is released from the pocket and X3 remained in the active site spanning from the -3 to the -1 site. The conformational changes between the WT-X3 and the WT apo-state are consistent with the comparison between the co-crystallized N44H-X3 and the N44H apo-state structures.5 15

ACS Paragon Plus Environment

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41

Comparing the WT-X3 structure with the co-crystallized E177Q-X6 structure, the conformational changes from the -3 to the -1 site are also similar. These comparisons indicate that soaking incurred changes could reflect the conformational changes in solution conditions, which shows that the residues of the ‘thumb’, the ‘finger’, and part of the ‘palm’ regions hold the ligand tightly for catalysis. In the active sites of the Y88F-X3 and WT-X3 structures, the configuration of the -1 site xylose is twisted from 4C1 toward 4H3 (Table 2). In the WT-X3 structure, there is an H-bond (2.8 Å) between the OE1 atom of E86 and the O2B atom of the –1 site xylose. The other H-bond (2.5 Å) is formed between the OE2 atom of E177 and the O4A atom of the -1 site xylose (Figure 1B and Figure S3A). Similar H-bonds are also found in the Y88F-X3 structure (Figure 3E and Figure S3C). In addition to other H-bonds and van der Waals interactions, the configuration of the -1 site xylose is twisted from 4C1 toward 4H3. In the Y77F-X3 structure, the above H-bond between E86 and the -1 site xylose is absent and this xylose remains in the 4C1 configuration. According to the QM/MM molecular dynamics simulations, when the glycosidic bond is broken, the -1 site xylose of GH11 could pass the itinerary of 1S3→4H3→4C1.7-8 Our crystal structures support this hypothesis. Though the configuration of the -1 site xylose is in the zone between 4C1 and 4 H3 in both the WT-X3 and the Y88F-X3 structures, the positions of the O4A atom are different: In the Y88F-X3 structure, in the absence of phenolic oxygen of F88, the O4A atom of -1 site xylose rotates about 80° and forms two H-bonds with the OE1 atom of E177 and the OE2 atom of E86, respectively (Figure 3E and Figure S3C). In the WT-X3 structure, the O4A atom cannot rotate to the aforementioned position because of its strong steric repulsion with the hydroxyl group of Y88. These results indicate that the configuration of -1 site xylose could quickly change due to the different H-bond networks. Flexibility of the -1 site xylose might facilitate distorting and breaking the glycosidic bond between the -1 and the +1 site xyloses. The enzymatic activity of Y88F or Y77F is drastically decreased, which indicates the importance of the hydroxyl group of Y88 and Y77. pKa calculations show that in the absence of the hydroxyl group, the pKa values of E177 have significantly increased to more than 9.0 in both the downward and the upward conformations. Thus, in the absence of the hydroxyl group, E177 would not be ideal to dissociate a proton to break the glycosidic bond in either the weakly acidic or neutral pH conditions. In the E177Q-X6 structure, there is no water within 4 Å of the NE2 atom of the amide group of Q177 when it is in the upward conformation. Based on our crystal structures and molecular dynamics simulations, it is reasonable to propose that single amino acid mutations would not significantly disturb the structure. Likewise, when Y77F or Y88F binds the oligosaccharide substrate, there is not any water molecule near the OE2 atom of E177 in the upward conformation, which functions as the proton carrier. Our neutron structures demonstrate that, in the alkaline condition, E177 is stabilized in the upward position. Thus, Y88F and Y77F cannot function well in the alkaline condition. In conclusion, the activities of Y77F and Y88F are largely diminished. Our kinetic and structural results of 16

ACS Paragon Plus Environment

Page 16 of 23

Page 17 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40

ACS Catalysis

the mutants confirm that E177 needs to rotate downward as the base to accept a proton from solvent and rotate upward as the acid to break up the glycosidic bond. CpHMD has been used to study the conformational changes as well as protonation states in different pH environments.16, 42-44 Our CpHMD simulations demonstrate that when pH increases from the acidic to the alkaline conditions, E177 rotates from the protonated downward conformation to the unprotonated upward conformation (Figure 4B). The protonation and rotation equilibrium are at around pH 5.6 (Figure 4), which is close to the optimal reaction pH of Xyn II. These calculations also support our hypothesis that E177 functions as the base in its downward conformation and as the acid in its upward conformation. Then another question arises: Why is E177 stabilized in the unprotonated upward conformation? Here, we used neutron crystallography to explore the H-bond network under different pH environments. Neutron crystallography is a powerful tool to directly observe protium/deuterium atoms in protein crystal structures.18-19, 45 In the neutron structure of WT at pD 5.4, the deuterium atom of the phenolic hydroxyl group of Y88 forms a weak H-bond (2.2 Å) with the OE1 atom of E177. Our previous neutron crystallography structures show that this H-bond is shortened to 1.9 Å at pD 6.2 and further to 1.5 Å at pD 8.9, respectively.17 These structures clearly show that the H-bond length decreases and becomes stronger when pH increases (Figure 5B). The results indicate that E177 is stabilized in this upward condition in the alkaline conditions. This H-bond significantly decreases the pKa value of E177 to be an ideal acid to break up the glycosidic bond (Table 3). In contrast, E177 at pD 4.8 rotates 56° downward to accept a proton from a water molecule and does not form any H-bond with Y88 anymore (Figure 5B). Xyn II from T. reesei has the optimal reaction pH in the range of 5.0-5.5.10-11 We propose that E177 rotation is essential as a proton carrier in the double displacement mechanism. In its optimal pH, E177 only forms a weak H-bond with Y88 and quickly rotates as a base/acid during catalysis (Figure 6). In an acidic condition, E177 is prone to be in the downward conformation as a base, but not prone to be in the upward conformation as an acid. The situation is reversed in alkaline conditions that E177 is predominantly in the upward condition, but difficult to rotate downward to obtain a proton from solvent as a base. Xylanase is proposed to be in different electrostatic states in different pH conditions, which may affect the H-bond strength between E177 and Y88 to rotate between the downward and the upward conformations for catalysis. In Bacillus circulans xylanase (BCX), substituting an Asn residue (N35) to Asp near the catalytic Glu (E172) decreases the pH optimum.46 Thus, it is possible to decrease the pH optimum by introducing negatively charged residues, such as Glu or Asp. In contrast, introducing positively charged residues, such as Arg or Lys, could increase the pH optimum.47 However, scrutiny is needed to finely adjust the pH optima for xylanase engineering, particularly residues in the active site, which may compromise the immediate H-bond network and electrostatic environment for catalysis.14

17

ACS Paragon Plus Environment

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

Figure 6. The proposed reaction mechanism of the family 11 glycoside hydrolase. Rotation of the E177 side chain is essential for catalysis. When it rotates downward, E177 accepts a proton as the base from solvent. When it rotates upward, E177 donates the proton to break up the glycosidic bond as an acid. The H-bond between Y88 and E177 significantly decreases the pKa value of E177 to be an ideal acid.

CONCLUSIONS Our high-resolution X-ray crystallography results show that the sugar ring configuration has the itinerary of 1S3→4H3→4C1 during catalysis for the family 11 GH, which is consistent with the previous QM/MM calculations.7-8 The different configurations of the -1 site xylose in our xylanase variants indicate that sugar ring flexibility may play an important role in breaking the glycosidic bond. Based on our previous neutron crystallography results, the carboxylic side chain of E177 is proposed to rotate downward to pick up a proton from solvent and rotates upward to donate this proton to break the glycosidic bond.16 Our CpHMD simulations show that the proton delivery activity reaches the equilibrium at pH 5.6, which is close to the narrow range of the optimal reaction pH of Xyn II. When E177 rotates upward, it forms an H-bond with its nearby tyrosine residue (Y88), which facilitates proton dissociation for catalysis. Our neutron crystallography results show that such H-bond length is shortened when pH increases. These results explain the pH-regulated reaction profile for glycoside hydrolase and may help to adjust their pH optima by protein engineering. 18

ACS Paragon Plus Environment

Page 18 of 23

Page 19 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12

Supporting Information This information is available on the ACS Publications website. X-ray and neutron crystallography statistics (Table S1-S2), GH11 structure illustration (Figure S1), the pH-dependent activity of WT and its comparison with Y88F and Y77F (Figure S2), H-bond interactions between xylanase variants and ligands (Figure S3), convergence of unprotonated E177 at different pH during CpHMD simulations (Figure S4), stability of WT, Y88F and Y77F during MD simulations (Figure S5), comparison of the atomic fluctuations obtained from the MD simulations among the xylanase variants (Figure S6), conformational changes of the E177 side chain using CpHMD simulations (Figure S7). (PDF)

13 14 15 16

AUTHOR INFORMATION Corresponding Authors * Email for Q.W.: [email protected] * Email for Y.D.H.: [email protected] Notes The authors declare no competing financial interest.

17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41

ACKNOWLEDGEMENTS Q.W. was supported by the National Natural Science Foundation of China (No. 31670790), the Open Project Program of State Key Laboratory of Natural Medicines, China Pharmaceutical University (No.3144060008), the Fundamental Research Funds for the Central Universities (No. KYTZ201604), the Natural Science Foundation of Jiangsu Province of China (No. BK20161443), the China Spallation Neutron Source travel grant, the Qing Lan Project of Jiangsu Province, and the Six Talent Peaks Project of Jiangsu Province. Research at ORNL’s Spallation Neutron Source was sponsored by the Scientific User Facilities Division, Office of Basic Energy Sciences, U.S. Department of Energy. We thank the SSRF beamlines BL19U1 and BL18U1 for X-ray data collection, and the BIODIFF beamline at the FRM II research reactor at the Heinz Maier-Leibnitz Zentrum (MLZ) for neutron data collection. We thank Dr. Lei Wu in Nanjing Agricultural University, Dr. Wenfei Li and Dr. Weitong Ren in the Nanjing University and Dr. Changrui Lu in Donghua University for helpful discussion. REFERENCES (1) Rye, C. S.; Withers, S. G., Glycosidase Mechanisms. Curr Opin Chem Biol 2000, 4, 573-580. (2) Zechel, D. L.; Withers, S. G., Glycosidase Mechanisms: Anatomy of a Finely Tuned Catalyst. Acc Chem Res 2000, 33, 11-18. (3) Vasella, A.; Davies, G. J.; Bohm, M., Glycosidase Mechanisms. Curr Opin Chem Biol 2002, 6, 619-629. (4) Wong, K. K. Y.; Tan, L. U. L.; Saddler, J. N., Multiplicity of Beta-1,4-Xylanase in Microorganisms 19

ACS Paragon Plus Environment

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41

Functions and Applications. Microbiol Rev 1988, 52, 305-317. (5) Wan, Q.; Zhang, Q.; Hamilton-Brehm, S.; Weiss, K.; Mustyakimov, M.; Coates, L.; Langan, P.; Graham, D.; Kovalevsky, A., X-Ray Crystallographic Studies of Family 11 Xylanase Michaelis and Product Complexes: Implications for the Catalytic Mechanism. Acta Crystallogr Sect D Biol Crystallogr 2014, 70, 11-23. (6) Mi, S. F.; Jia, X. J.; Wang, J. Z.; Qiao, W. B.; Peng, X. W.; Han, Y. J., Biochemical Characterization of Two Thermostable Xylanolytic Enzymes Encoded by a Gene Cluster of Caldicellulosiruptor owensensis. Plos One 2014, 9, e105264. (7) Ardevol, A.; Rovira, C., Reaction Mechanisms in Carbohydrate-Active Enzymes: Glycoside Hydrolases and Glycosyltransferases. Insights from ab Initio Quantum Mechanics/Molecular Mechanics Dynamic Simulations. J Am Chem Soc 2015, 137, 7528-7547. (8) Iglesias-Fernandez, J.; Raich, L.; Ardevol, A.; Rovira, C., The Complete Conformational Free Energy Landscape of Beta-Xylose Reveals a Two-Fold Catalytic Itinerary for Beta-Xylanases. Chem Sci 2015, 6, 1167-1177. (9) Langan, P.; Gnanakaran, S.; Rector, K. D.; Pawley, N.; Fox, D. T.; Cho, D. W.; Hammel, K. E., Exploring New Strategies for Cellulosic Biofuels Production. Energ Environ Sci 2011, 4, 3820-3833. (10) Torronen, A.; Harkki, A.; Rouvinen, J., Three-Dimensional Structure of Endo-1,4-Beta-Xylanase II from Trichoderma Reesei: Two Conformational States in the Active Site. EMBO J 1994, 13, 2493-2501. (11) Torronen, A.; Rouvinen, J., Structural Comparison of 2 Major Endo-1,4-Xylanases from Trichoderma-Reesei. Biochemistry 1995, 34, 847-856. (12) Polizeli, M. L. T. M.; Rizzatti, A. C. S.; Monti, R.; Terenzi, H. F.; Jorge, J. A.; Amorim, D. S., Xylanases from Fungi: Properties and Industrial Applications. Appl Microbiol Biot 2005, 67, 577-591. (13) Watanabe, N.; Akiba, T.; Kanai, R.; Harata, K., Structure of an Orthorhombic Form of Xylanase II from Trichoderma Reesei and Analysis of Thermal Displacement. Acta Crystallogr Sect D Biol Crystallogr 2006, 62, 784-792. (14) Joshi, M. D.; Sidhu, G.; Nielsen, J. E.; Brayer, G. D.; Withers, S. G.; McIntosh, L. P., Dissecting the Electrostatic Interactions and pH-Dependent Activity of a Family 11 Glycosidase. Biochemistry 2001, 40, 10115-10139. (15) Xiong, H. R.; von Weymarn, N.; Leisola, M.; Turunen, O., Influence of pH on the Production of Xylanases by Trichoderma Reesei Rut C-30. Process Biochem 2004, 39, 729-733. (16) Huang, Y. D., Yue, Z., Tsai, C.C., Henderson, J.A., Shen, J., Predicting Catalytic Proton Donors and Nucleophiles in Enzymes: How Adding Dynamics Helps Elucidate the Structure-Function Relationships. J Phys Chem Lett 2018, 9, 1179-1184. (17) Wan, Q.; Parks, J. M.; Hanson, B. L.; Fisher, S. Z.; Ostermann, A.; Schrader, T. E.; Graham, D. E.; Coates, L.; Langan, P.; Kovalevsky, A., Direct Determination of Protonation States and Visualization of Hydrogen Bonding in a Glycoside Hydrolase with Neutron Crystallography. Proc Natl Acad Sci U S A 2015, 112, 12384-12389. (18) O'Dell, W. B.; Bodenheimer, A. M.; Meilleur, F., Neutron Protein Crystallography: A Complementary Tool for Locating Hydrogens in Proteins. Arch Biochem Biophys 2015, 602, 48-60. (19) Chen, J. C. H.; Unkefer, C. J., Fifteen Years of the Protein Crystallography Station: The Coming of Age of Macromolecular Neutron Crystallography. IUCrJ 2017, 4, 72-86. 20

ACS Paragon Plus Environment

Page 20 of 23

Page 21 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41

ACS Catalysis

(20) Kovalevsky, A. Y.; Hanson, B. L.; Seaver, S.; Fisher, S. Z.; Mustyakimov, M.; Langan, P., Preliminary Joint X-Ray and Neutron Protein Crystallographic Studies of Endoxylanase II from the Fungus Trichoderma Longibrachiatum. Acta Crystallogr Sect F Struct Biol Cryst Commn 2011, 67, 283-286. (21) Bornhorst, J. A.; Falke, J. J., Purification of Proteins Using Polyhistidine Affinity Tags. Methods Enzymol 2000, 326, 245-254. (22) Bailey M.J, B. P., Poutanen K., Interlaboratory Testing of Methods for Assay of Xylanase Activity. J Biotechnol 1992, 23, 257-270. (23) Minor, W.; Cymborowski, M.; Otwinowski, Z.; Chruszcz, M., HKL-3000: The Integration of Data Reduction and Structure Solution--From Diffraction Images to an Initial Model in Minutes. Acta Crystallogr Sect D Biol Crystallogr 2006, 62, 859-866. (24) Afonine, P. V.; Grosse-Kunstleve, R. W.; Echols, N.; Headd, J. J.; Moriarty, N. W.; Mustyakimov, M.; Terwilliger, T. C.; Urzhumtsev, A.; Zwart, P. H.; Adams, P. D., Towards Automated Crystallographic Structure Refinement with Phenix.Refine. Acta Crystallogr Sect D Biol Crystallogr 2012, 68, 352-367. (25) Emsley, P.; Lohkamp, B.; Scott, W. G.; Cowtan, K., Features and Development of Coot. Acta Crystallogr Sect D Biol Crystallogr 2010, 66, 486-501. (26) Krissinel, E.; Henrick, K., Secondary-Structure Matching (SSM), A New Tool for Fast Protein Structure Alignment in Three Dimensions. Acta Crystallogr Sect D Biol Crystallogr 2004, 60, 2256-2268. (27) Moriarty, N. W.; Grosse-Kunstleve, R. W.; Adams, P. D., Electronic Ligand Builder and Optimization Workbench (Elbow): A Tool for Ligand Coordinate and Restraint Generation. Acta Crystallogr Sect D Biol Crystallogr 2009, 65, 1074-1080. (28) Adams, P. D.; Afonine, P. V.; Bunkoczi, G.; Chen, V. B.; Davis, I. W.; Echols, N.; Headd, J. J.; Hung, L. W.; Kapral, G. J.; Grosse-Kunstleve, R. W.; McCoy, A. J.; Moriarty, N. W.; Oeffner, R.; Read, R. J.; Richardson, D. C.; Richardson, J. S.; Terwilliger, T. C.; Zwart, P. H., PHENIX: A Comprehensive Python-Based System for Macromolecular Structure Solution. Acta Crystallogr Sect D Biol Crystallogr 2010, 66, 213-221. (29) Chen, V. B.; Arendall, W. B.; Headd, J. J.; Keedy, D. A.; Immormino, R. M.; Kapral, G. J.; Murray, L. W.; Richardson, J. S.; Richardson, D. C., MolProbity: All-Atom Structure Validation for Macromolecular Crystallography. Acta Crystallogr Sect D Biol Crystallogr 2010, 66, 12-21. (30) Jeffrey, G. A.; Yates, J. H., Stereographic Representation of the Cremer-Pople Ring-Puckering Parameters for Pyranoid Rings. Carbohyd Res 1979, 74, 319-322. (31) Anandakrishnan, R.; Aguilar, B.; Onufriev, A. V., H++ 3.0: Automating pK Prediction and the Preparation of Biomolecular Structures for Atomistic Molecular Modeling and Simulations. Nucleic Acids Res 2012, 40, 537-541. (32) Sun, Z. X.; Wang, X. H.; Song, J. N., Extensive Assessment of Various Computational Methods for Aspartate's pK(a) Shift. J Chem Inf Model 2017, 57, 1621-1639. (33) Zhou, H. X.; Pang, X. D., Electrostatic Interactions in Protein Structure, Folding, Binding, and Condensation. Chem Rev 2018, 118, 1691-1741. (34) Wallace, J. A.; Shen, J. K., Continuous Constant pH Molecular Dynamics in Explicit Solvent with pH-Based Replica Exchange. J Chem Theory Comput 2011, 7, 2617-2629. (35) Brooks, B. R.; Brooks, C. L.; Mackerell, A. D.; Nilsson, L.; Petrella, R. J.; Roux, B.; Won, Y.; Archontis, G.; Bartels, C.; Boresch, S.; Caflisch, A.; Caves, L.; Cui, Q.; Dinner, A. R.; Feig, M.; Fischer, S.; 21

ACS Paragon Plus Environment

ACS Catalysis 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41

Gao, J.; Hodoscek, M.; Im, W.; Kuczera, K.; Lazaridis, T.; Ma, J.; Ovchinnikov, V.; Paci, E.; Pastor, R. W.; Post, C. B.; Pu, J. Z.; Schaefer, M.; Tidor, B.; Venable, R. M.; Woodcock, H. L.; Wu, X.; Yang, W.; York, D. M.; Karplus, M., CHARMM: The Biomolecular Simulation Program. J Comput Chem 2009, 30, 1545-1614. (36) MacKerell, A. D.; Bashford, D.; Bellott, M.; Dunbrack, R. L.; Evanseck, J. D.; Field, M. J.; Fischer, S.; Gao, J.; Guo, H.; Ha, S.; Joseph-McCarthy, D.; Kuchnir, L.; Kuczera, K.; Lau, F. T. K.; Mattos, C.; Michnick, S.; Ngo, T.; Nguyen, D. T.; Prodhom, B.; Reiher, W. E.; Roux, B.; Schlenkrich, M.; Smith, J. C.; Stote, R.; Straub, J.; Watanabe, M.; Wiorkiewicz-Kuczera, J.; Yin, D.; Karplus, M., All-Atom Empirical Potential for Molecular Modeling and Dynamics Studies of Proteins. J Phys Chem B 1998, 102, 3586-3616. (37) Mackerell, A. D.; Feig, M.; Brooks, C. L., Extending the Treatment of Backbone Energetics in Protein Force Fields: Limitations of Gas-Phase Quantum Mechanics in Reproducing Protein Conformational Distributions in Molecular Dynamics Simulations. J Comput Chem 2004, 25, 1400-1415. (38) Otwinowski, Z.; Minor, W., Processing of X-ray Diffraction Data Collected in Oscillation Mode. Methods Enzymol 1997, 276, 307-326. (39) Gerlits, O.; Wymore, T.; Das, A.; Shen, C. H.; Parks, J. M.; Smith, J. C.; Weiss, K. L.; Keen, D. A.; Blakeley, M. P.; Louis, J. M.; Langan, P.; Weber, I. T.; Kovalevsky, A., Long-Range Electrostatics-Induced Two-Proton Transfer Captured by Neutron Crystallography in an Enzyme Catalytic Site. Angew Chem Int Edit 2016, 55, 4924-4927. (40) Hosur, M. V.; Chitra, R.; Hegde, S.; Choudhury, R. R.; Das, A.; Hosur, R. V., Low-barrier Hydrogen Bonds in Proteins. Crystallogr Rev 2013, 19, 3-50. (41) Zolotnitsky, G.; Cogan, U.; Adir, N.; Solomon, V.; Shoham, G.; Shoham, Y., Mapping Glycoside Hydrolase Substrate Subsites by Isothermal Titration Calorimetry. Proc Natl Acad Sci U S A 2004, 101, 11275-11280. (42) Goh, G. B.; Laricheva, E. N.; Brooks, C. L., Uncovering pH-Dependent Transient States of Proteins with Buried Ionizable Residues. J Am Chem Soc 2014, 136, 8496-8499. (43) Zeng, X. C.; Mukhopadhyay, S.; Brooks, C. L., Residue-Level Resolution of Alphavirus Envelope Protein Interactions in pH-Dependent Fusion. Pro Natl Acad Sci USA 2015, 112, 2034-2039. (44) Huang, Y. D.; Chen, W.; Dotson, D. L.; Beckstein, O.; Shen, J., Mechanism of pH-Dependent Activation of the Sodium-Proton Antiporter Nhaa. Nat Commun 2016, 7, 12940-12950. (45) Fisher, S. J.; Blakeley, M. P.; Howard, E. I.; Petit-Haertlein, I.; Haertlein, M.; Mitschler, A.; Cousido-Siah, A.; Salvay, A. G.; Popov, A.; Muller-Dieckmann, C.; Petrova, T.; Podjarny, A., Perdeuteration: Improved Visualization of Solvent Structure in Neutron Macromolecular Crystallography. Acta Crystallogr Sect D Biol Crystallogr 2014, 70, 3266-3272. (46) Joshi, M. D.; Sidhu, G.; Pot, I.; Brayer, G. D.; Withers, S. G.; McIntosh, L. P., Hydrogen Bonding and Catalysis: A Novel Explanation for How a Single Amino Acid Substitution Can Change the pH Optimum of a Glycosidase. J Mol Biol 2000, 299, 255-279. (47) Bai, W. Q.; Cao, Y. F.; Liu, J.; Wang, Q. H.; Jia, Z. H., Improvement of Alkalophilicity of an Alkaline Xylanase Xyn11A-LC from Bacillus sp SN5 by Random Mutation and Glu135 Saturation Mutagenesis. Bmc Biotechnol 2016, 16, 77-86.

TOC Graphic 22

ACS Paragon Plus Environment

Page 22 of 23

Page 23 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

1 2

Rotation of E177 side chain is essential for proton shuttling to break the glycosidic bond during catalysis.

23

ACS Paragon Plus Environment