pH-Induced structural changes in xylanase GH11 from Thermoanaerobacterium saccharolyticum

Materials

Following chemicals were used in this study: Luria-Berani broth medium (LPS solution, Catalog No. LB-05, Daejeon, Republic of Korea,), ampicillin (LPS solution, Catalog No. AMP25), Isopropyl β-D-1-thiogalactopyranoside (LPS solution, Cat No. IPTG025), Tris (Sigma-Aldrich, St. Louis, MI, USA, Catalog No. T15760), NaCl (Sigma-Aldrich, Catalog No. S9888), Ammonium acetate (Sigma-Aldrich, Catalog No. A7262), Glycerol (Sigma-Aldrich, Catalog No. G5516).

Protein preparation

The preparation of purified protein was similar to that in a previous report.⁶ In brief, vector DNA containing the TsaGH11 gene with an N-terminal hexahistidine tag was transformed into Escherichia coli BL21(DE3). Cells were grown in LB broth medium supplemented with ampicillin (50 μg/mL) at 37°C. When the culture optical density at 600 nm reached 0.6–0.8, protein expression were induced by addition of 0.5 mM isopropyl β-D-1-thiogalactopyranoside (IPTG) and then incubated at 18°C while shacking for overnight. After harvesting the cells and disrupting them via sonication, the supernatant was loaded onto a Ni-NTA affinity resin (Qiagen, Hilden, Germany) in a column. The resin was washed with 50 mM Tris-HCl (pH 8.0), 200 mM NaCl, and 20 mM imidazole, and then proteins were eluted with 50 mM Tris-HCl (pH 8.0), 200 mM NaCl, and 300 mM imidazole. To cleave the N-terminal hexahistidine tag, elution fractions were pooled and incubated with thrombin at 22°C overnight. Samples were concentrated using Centricon (Sartorius, 10-kDa cutoff) and loaded onto a Sephacryl S-100 10/300 column (GE Healthcare, Chicago, IL, USA) in 10 mM Tris-HCl (pH 8.0) and 200 mM NaCl. The purified protein was concentrated to ~20 mg/mL using a Centricon for crystallization. Protein concentrations were determined using a NanoDrop 1000 spectrophotometer (ThermoFisher).

Crystallization

Crystallization screen of TsaGH11 was initially performed using the sitting-drop vapor diffusion method at 22°C. Purified TsaGH11 solutions (500 nL) were mixed with equal volume of crystallization solution from commercially available crystallization kits (Hampton Research, Aliso Viejo, CA, USA) on 96 × 2-well MRC nanoplates (Innovadyne). Microcrystals were obtained under four conditions with the following crystallization solutions: (i) Salt RX H10: 0.1 M Tris-HCl, pH 8.5, and 4.0 M ammonium acetate; (ii) Crystal screen B9: 0.2 M magnesium acetate, 0.1 M sodium cacodylate, pH 6.5, and 30% (v/v) 2-methyl-2,4-pentanediol; (iii) Index A11: 0.1 M HEPES, pH 7.5, and 3.0 M sodium chloride; and (iv) Index A12: 0.1 M Tris-HCl, pH 8.5, and 3.0 M sodium chloride. To obtain large crystals, crystal optimization experiments were performed using the hanging-drop vapor diffusion method at 22°C. Suitable TsaGH11 crystals for X-ray diffraction experiment were obtained under condition (i) by scaling up to mix 2 μL of the protein solution with 2 μL of the reservoir solution in 24-well VDX crystallization plates (Hampton Research).

Data collection

The X-ray diffraction data were collected on beamline 11C at Pohang Light Source II (PLS-II, Pohang, Republic of Korea).¹⁷ The TsaGH11 crystal was immersed in a cryoprotectant solution consisting of a reservoir solution supplemented with 20% (v/v) glycerol for 10 s. The TsaGH11 crystal was mounted on the goniometer under a liquid nitrogen stream at 100 K. Diffraction data were recorded using a Pilatus 6M detector. Diffraction images were indexed, integrated, and scaled using the HKL2000 program.¹⁸

Structure determination

The electron density map of TsaGH11 structure was obtained via molecular replacement method using MOLREP (version 11.2.08).¹⁹ Crystal structure of TsaGH11 at pH 4.6 (PDB code: 8IH0)⁶ was used as search model. Manual model building was performed using COOT (version 0.9.6).²⁰ Structure refinement was performed using phenix.refine in PHENIX.²¹ Water molecules were automatically added during the refinement using the default parameters in PHENIX.²¹ The final coordinates of TsaGH11^pH8.5 was validated using MolProbity.²² Structural figures were generated using PyMOL (version 2.4.1.; http://pymol.org/2). The structure factor and coordinates were deposited at the Protein Data Bank under access code 8X1D (https://www.rcsb.org/structure/8X1D).

Bioinformatics

Differences in the secondary structures of TsaGH11^pH8.6 and TsaGH11^pH4.6 were analyzed using 2StrucCompare²³ with the Dictionary of Secondary Structure of Proteins (DSSP)²⁴ and the STRuctural IDEntification (STRIDE)²⁵ method. The B-factor and normalized B-factor were analyzed using PHENIX.²¹

Structure determination

Protein conformation depends on environmental factors such as pH, temperature, and interactions.^26,27 Although TsaGH11 exhibits high activity at acidic pH, this activity diminishes at basic pH,⁶ and the crystal structure of TsaGH11 at an acidic pH (pH 4.6, which is close to that at optimal activity) has been determined.^6,28 To comprehend the change in TsaGH11 activity influenced by pH from a structural perspective, an extended crystallization assessment was conducted to obtain the crystal structure of TsaGH11 at basic pH. Microcrystals of TsaGH11 were obtained from four new crystallization solutions: (i) 0.1 M Tris-HCl, pH 8.5, and 4.0 M ammonium acetate; (ii) 0.1 M Tris-HCl, pH 8.5, and 3.0 M sodium chloride; (iii) 0.1 M HEPES, pH 7.5, 3.0 M sodium chloride; and (iv) 0.2 M magnesium acetate tetrahydrate, 0.1 M sodium cacodylate, pH 6.5, and 30% (v/v) 2-methyl-2,4-pentanediol. These crystallization conditions at three different pH values (6.5, 7.5, and 8.5) were distinct from those previously reported (pH 4.6) (Figure 1a). Among them, the crystallization conditions (i) of TsaGH11 grown at pH 8.5 differ only in pH from those previously reported for TsaGH11^pH4.6 (0.1 M sodium acetate, pH 4.6, and 4.0 M ammonium acetate), which showed the same crystal morphology as TsaGH11 (Figure 1a). Therefore, crystallization condition (i) was considered to be the optimal condition with the least influence from the crystallization solution constituents. The growth rate of TsaGH11 crystals under condition (i) was 2–4-fold longer than the crystal growth time at pH 4.6. The TsaGH11^pH8.5 crystal diffracted up to 1.95 Å (Figure 1b), and although crystals were obtained under conditions (ii), (iii), and (iv) (Figure 1a), they did not exhibit sufficient diffraction intensity to determine the crystal structure.

Figure 1. Crystallographic study of TsaGH11.

(a) Crystallization screen results for TsaGH11. TsaGH11 crystals were grown under (i) 0.1 M sodium acetate, pH 8.5, and 4.0 M ammonium acetate, (ii) 0.2 M magnesium acetate, 0.1 M sodium cacodylate, pH 6.5, and 30% (v/v) 2-methyl-2,4-pentanediol, (iii) 0.1 M HEPES, pH 7.5, and 3.0 M sodium chloride, and (iv) 0.1 M Tris-HCl, pH 8.5, and 3.0 M sodium chloride. (b) Diffraction pattern of TsaGH11 crystals grown with 0.1 M sodium acetate, pH 8.5, and 4.0 M ammonium acetate.

The TsaGH11^pH8.5 crystal belongs to the tetragonal space group P4₃2₁2 with unit cell parameters (a = b = 72.79 Å and c = 165.63 Å) and contains two molecules in an asymmetric unit (Table 1). The space group and unit cell parameters of the TsaGH11^pH8.5 crystal are nearly identical to those of the TsaGH11^pH4.6 crystal (P4₃2₁2, a = b = 73.11 Å, and c = 165.42 Å). The crystal structure of TsaGH11^pH8.5 was solved up to 1.95 Å, with R_work and R_free values of 0.171 and 0.209, respectively. The electron density map of the TsaGH11 structure was clearly observed for all amino acids (Thr29–Trp211). TsaGH11^pH8.5 exhibits a β-jelly roll fold, resembling that of the right hand. The catalytic residues and substrate-binding cleft are located between the finger and thumb domains (Figure 2a). In the asymmetric unit of TsaGH11^pH8.5, molecule A has a rigid conformation due to crystal packing, whereas molecule B has relatively high flexibility of the thumb domain, exposed to the solvent region in crystal packing (Figure 2b). The average temperature factors of the A and B molecules of TsaGH11^pH8.5 were 28.17 and 37.16 Ų, respectively. The superimposition of the A and B molecules of TsaGH11^pH8.5 showed similarity with a root-mean-square (r.m.s.) deviation of 0.222 Å. The palm and finger domains of molecules A and B display high similarity, but the positions of their thumb domains differ significantly (Figure 2c and 2d). At the thumb domain, Cα atoms from Arg139, Pro143, Asp146, and Thr148 between molecules A and B exhibited subtle movements of 1.09, 1.86, 1.26, and 1.60 Å, respectively (Figure 2d). The surface structure of molecule A of TsaGH11^pH8.5 displays an open conformation for the substrate-binding cleft located between the thumb and finger domains (Figure 2e). At subsites +1 and +2, the distance between Pro40 (atom: CB) and Trp40 (CH2) was 5.84 Å (Figure 2e). At subsites –1 and –2, the distances between Asn62 (OD1) and Arg139 (NH2) and between Arg139 (NH1) and Tyr200 (OH1) were 7.02 and 7.93 Å, respectively. The distance between the catalytic Glu105 (OE2) and Glu198 (OE1) residues was 5.17 Å. Conversely, the surface structure of molecule B of TsaGH11^pH8.5 displayed a closed conformation between the thumb and finger domains (Figure 2f). At subsites +1 and +2, the distance between Pro40 (CB) and Trp40 (CZ2) was 3.71 Å (Figure 2f). At subsites −1 and −2, the distances between Asn62 (OD1) and Arg139 (NH2) and between Arg139 (NH1) and Tyr200 (OH1) were 5.71 and 8.38 Å, respectively. The distance between the catalytic Glu105 (OE2) and Glu198 (OE1) residues was 5.13 Å (Figure 2f). Consequently, molecules A and B in the asymmetric units of TsaGH11 represent the open and closed conformations of the substrate-binding cleft, respectively, due to the crystal packing effect. These differing conformations of TsaGH11^pH8.5, influenced by crystal packing, are similar to those observed in the previously reported TsaGH11^pH4.6 crystal because of identical protein crystal packing within the same space group.⁶

Figure 2. Crystal structure of TsaGH11^pH8.5.

(a) Cartoon representation of TsaGH11, resembling the right-hand structure with palm, finger, and thumb domains. (b) B-factor putty representation of TsaGH11^pH8.5 for molecules in the asymmetric unit. (c) Superimposition of two TsaGH11^pH8.5 molecules in the asymmetric unit. (d) Close-up view of the superimposition of the thumb domain of the open (green) and closed (cyan) conformations of TsaGH11^pH8.5. Surface structure of the (e) open and (f) closed conformations of the substrate-binding cleft in TsaGH11^pH8.5.

Structural comparison of TsaGH11^pH8p5 and TsaGH11^pH4.6

The xylanase activity of TsaGH11 at pH 4.6 and pH 8.5 was approximately 97% and 35%, respectively,⁶ and the crystal structures of TsaGH11^pH4.6 and TsaGH11^pH8.5 therefore represent structural conformations at relatively high and low activity ranges, respectively. Among the crystal structures of TsaGH11^pH4.6 and TsaGH11^pH8.5, molecules A and B have the same conformation and were consequently compared. pH induces alterations in the ionization state of amino acid side chains and can affect the secondary structure of proteins.^29,30 Consequently, the secondary structures of TsaGH11^pH8.5 and TsaGH11^pH4.6 were analyzed using DSSP and ESSS method. In the TsaGH11 open conformation, secondary structural differences were observed in Val40, Gly89, Asn90, and Asn167 (Figure 3a). Val40, Gly89, and Asn90 were located above the substrate-binding cleft in the finger domain, and Asn167 was positioned on the surface opposite the substrate-binding site and did not directly influence substrate-binding or activity (Figure 3b). In the TsaGH11 closed conformation, secondary structural differences were present in Asn59, Cys60, Gly61, Gly89, Asn90, Ser144, Gly147, Thr166, Asn167, Qln201, and Ser202 (Figure 3a). Asn59, Cys60, Gly61, Gly89, Asn90, Qln201, and Ser202 are located in the finger domain and lie above the substrate-binding cleft, and Ser144 and Gly147 are positioned in the thumb domain and are presumed to be involved in substrate recognition. Thr166 and Asn167 are located opposite the substrate-binding cleft and do not directly affect substrate-binding (Figure 3c). Accordingly, changes in the secondary structure were commonly observed in the main chain of Gly89 and Asn90 of the finger domain, as well as of Asn167, positioned on the opposite surface of the substrate-binding cleft in both the open and closed conformations of TsaGH11. Although differences in secondary structure were noted in the main chain, no significant structural changes were observed in the conformation of the side chains.

Figure 3. Comparison of the secondary and crystal structures of TsaGH11^pH8.5 and TsaGH11^pH4.6.

(a) Secondary structure analysis of the TsaGH11 structures using DSSP and STRIDE. Different secondary structure regions are indicated by red bars. (b) Cartoon representation of the different secondary structure regions between TsaGH11^pH8.5 and TsaGH11^pH4.6. (c) Superimposition of the catalytic and substrate-binding clefts of the open and closed conformations of TsaGH11^pH8.5 and TsaGH11^pH4.6.

The superimposition of the TsaGH11^pH8.5 and TsaGH11^pH4.6 open conformations had an r.m.s. deviation of 0.075 Å (Figure 3c). In the open conformation, the distances between the side chains of the two catalytic residues Glu105 and Glu198 from TsaGH11^pH8.5 and TsaGH11^pH4.6 were 5.17 and 5.17 Å, respectively. At subsites +1 and +2, the distance between Trp36 (CZ) and Pro143 (CB), which determines the size of the opening between the thumb and finger domains, in TsaGH11^pH8.5 and TsaGH11^pH4.6 were 5.84 and 5.60 Å, respectively. At subsite −1, the distances between Asn62 (OD1) and Arg139 (NH2) from TsaGH11^pH8.5 and TsaGH11^pH4.6 were 7.02 and 6.84 Å, respectively. At subsite −2, the distances between Arg139 (NH1) and Tyr200 (OH) from TsaGH11^pH8.5 and TsaGH11^pH4.6 were 7.93 and 8.30 Å, respectively. Overall, the active site regions of the TsaGH11^pH8.5 and TsaGH11^pH4.6 open conformations are almost identical, whereas the distance between the finger and thumb domains involved in the substrate recognition site of TsaGH11^pH8.5 is slightly wider than that in TsaGH11^pH4.6.

The superimposition of the closed conformations of TsaGH11^pH8.5 and TsaGH11^pH4.6 had an r.m.s. deviation of 0.094 Å (Figure 3c). In the closed conformation, the distances between the side chains of the two catalytic residues Glu105 and Glu198 in TsaGH11^pH8.5 and TsaGH11^pH4.6 were 5.13 and 5.21 Å. At subsites +1 and +2, the distances between Trp36 (CZ) and Pro143 (CB) and between TsaGH11^pH8.5 and TsaGH11^pH4.6 were 3.71 and 3.82 Å, respectively. Notably, different side chain conformations of Arg139 involved in substrate recognition were observed between TsaGH11^pH8.5 and TsaGH11^pH4.6. At subsite −1, the distances between Asn62 (OD1) and Arg139 (NH2) from TsaGH11^pH8.5 and TsaGH11^pH4.6 were 5.71 and 4.99 Å, respectively. At subsite −2, the distances between Arg139 (NH1) and Tyr200 (OH) from TsaGH11^pH8.5 and TsaGH11^pH4.6 were 8.02 and 8.41 Å, respectively.

Comparison of the flexibility and water molecules from TsaGH11^pH8.5 and TsaGH11^pH4.6

The flexibility of protein conformation can be affected by pH.³¹ To understand whether pH affects the flexibility of TsaGH11, the temperature factor of TsaGH11^pH8.5 and TsaGH11^pH4.6 was investigated. The overall B-factor values of the open/closed conformations of TsaGH11^pH8.5 and TsaGH11^pH4.6 were 28.17/37.16 and 18.14/21.80 Ų, respectively (Figure 4a). At both pH 8.5 and 4.6 for TsaGH11, the flexibility tendencies of the closed and open conformations differed, with the closed conformation exhibiting higher flexibility than the open conformation, which can be attributed to protein packing. In the B-factor plot, the tendencies of each open and closed conformation for TsaGH11^pH8.5 and TsaGH11^pH4.6 were similar. However, the B-factor values for these conformations of TsaGH11^pH8.5 were 1.55 and 1.70-fold higher than those of TsaGH11^pH4.6. This result indicates that the molecular flexibility of TsaGH11 is relatively high at basic pH. The crystal structures of TsaGH11^pH8.5 and TsaGH11^pH4.6 were obtained from different crystals, and B-factor effects may occur because of variations in the crystal conditions. Accordingly, the normalized B-factor of TsaGH11^pH8.5 and TsaGH11^pH4.6 was analyzed and compared (Figure 4b). In the open conformation, Asn88 in the finger domain and Arg139–Gln149 in the thumb domain in TsaGH11^pH4.6 showed relatively higher flexibility than TsaGH11^pH8.5. In the closed conformation, the area around Lys180 of the palm domain in TsaGH11^pH4.6 has a relatively higher flexibility than that in TsaGH11^pH8.5.

Figure 4. Comparison of the flexibility and water molecules of TsaGH11^pH8.5 and TsaGH11^pH4.6.

Profile of (a) B-factor and (b) normalized B-factor for the open and closed conformations of TsaGH11^pH8.5 and TsaGH11^pH4.6. (c) Superimposition of water molecules in the substrate-binding cleft from TsaGH11^pH8.5 (blue) and TsaGH11^pH4.6 (red). (d) Superimposition of water molecules in the substrate-binding cleft for the open (green) and closed (cyan) conformations of TsaGH11^pH8.5 and TsaGH11^pH4.6. Water molecules are indicated by spheres.

Changes in pH can affect the charge state of amino acids and subsequently influence the coordination or position of water molecules around protein amino acids.^32,33 Water molecules share characteristics with the hydroxyl group in the xylan substrate, and their positions around the active substrate can subsequently offer insights into the substrate-binding moiety. Therefore, water molecules in the substrate-binding cleft and the active site of TsaGH11 were investigated. In the open conformation of TsaGH11, the subtle movement of water molecules in TsaGH11^pH8.5 and TsaGH11^pH4.6 at subsites −2, –1, +1, and +2 were approximately 0.22–0.77, 0.26, 0.48, and 0.34 Å, respectively, and between the catalytic residues in TsaGH11^pH8.5 and TsaGH11^pH4.6 was approximately 0.19 Å (Figure 4c). In the closed conformation of TsaGH11, the subtle movement of water molecules from TsaGH11^pH8.5 and TsaGH11^pH4.6 at subsites –2, −1, +1, and +2 were approximately 0.16–0.48, 0.09–0.22, 0.42, and 0.26 Å, respectively, and between the catalytic residues from TsaGH11^pH8.5 and TsaGH11^pH4.6 was approximately 0.17 Å (Figure 4c). Consequently, the average change in the position of these water molecules at different pH levels was <0.5 Å, and no specific directionality was observed in these positions.

Unlike the difference in the position of the water molecules caused by pH, the overall change in the position of water molecules in the open and closed structures in TsaGH11^pH8.5 and TsaGH11^pH4.6 was substantial. In the superimposition of the open and closed conformations of TsaGH11^pH8.5, the subtle movements of water molecules at subsites −2, –1, +1, and +2 were approximately 0.23–0.38, 0.22, 0.26, and 0.71 Å, respectively, and between the catalytic residues in TsaGH11^pH8.5 was approximately 0.61 Å (Figure 4d). In the superimposition of the open and closed conformations of TsaGH11^pH4.6, the subtle movements of water molecules at subsites −2, −1, +1, and +2 were approximately 0.12–0.41, 0.37, 0.72, and 0.92 Å, respectively, and between the catalytic residues in TsaGH11^pH4.6 was approximately 0.64 Å (Figure 4d). Overall, a minor structural change occurred in the position of water molecules in the substrate-binding cleft of TsaGH11 because of the pH. However, the change in the position of water molecules resulting from the open and closed conformational changes in the substrate-binding cleft of TsaGH11 was more significant than the pH effect.