Supplementary Materials for

Transcription

1 Supplementary Materials for Newton s cradle proton relay with amide imidic acid tautomerization in inverting cellulase visualized by neutron crystallography Akihiko Nakamura, Takuya Ishida, Katsuhiro Kusaka, Taro Yamada, Shinya Fushinobu, Ichiro Tanaka, Satoshi Kaneko, Kazunori Ohta, Hiroaki Tanaka, Koji Inaka, Yoshiki Higuchi, Nobuo Niimura, Masahiro Samejima, Kiyohiko Igarashi The PDF file includes: Published 21 August 2015, Sci. Adv. 1, e (2015) DOI: /sciadv Table S1. Data collection, phasing, and refinement statistics for first structural determination of PcCel45A. Table S2. Statistics of data collection and refinement of the x-ray structure of PcCel45A WT with and without 5 mm cellopentaose at cryogenic temperature. Table S3. Statistics of data collection and refinement of the x-ray structure of PcCel45A N92D with and without cellopentaose at cryogenic temperature and PcCel45A N92D unliganded structure at room temperature. Table S4. Statistics of data collection and refinement of the x-ray structure of PcCel45A D114N with and without 5 mm cellopentaose at cryogenic temperature. Table S5. Statistics of data collection and refinement of the neutron and x-ray structure of PcCel45A WT unliganded. Table S6. Statistics of data collection and refinement of the neutron and x-ray structure of PcCel45A WT with 2.5 mm cellopentaose. Table S7. Statistics of data collection and refinement of the x-ray structure of PcCel45A N105D. Fig. S1. Determination of anomeric forms of products. Fig. S2. Structural comparison of PcCel45A with other GH45 enzymes. Fig. S3. Activity profile of PcCel45A WT and mutants. Fig. S4. Temperature effect on PcCel45A structure. Fig. S5. Substrate recognition of PcCel45A WT and mutants. Fig. S6. Joint refined structure of PcCel45A WT with 2.5 mm cellopentaose. Fig. S7. Neighboring pair residues in the proton relay pathway. Fig. S8. Structural and activity profile of PcCel45A N105D mutant. Fig. S9. SDS-PAGE and Native-PAGE of PcCel45A WT and mutants. Legend for movie S1

2 Other Supplementary Material for this manuscript includes the following: (available at Movie S1 (.m4v format). The proposed proton relay pathway between Asp 92 (general base) and Asn 114 (general acid) residues in PcCel45A.

3 Table S1: Data collection, phasing, and refinement statistics for first structural determination of PcCel45A. Data collection statistics AuCl 3 derivative K 2Pt(CN) 4 derivative Native Beam line BL-5A BL-5A BL-5A Wave length (A ) Space group Unit cell parameters a, b, c (A ) 56.3, 42.9, , 42.8, , 42.9, 62.7 β ( ) Resolution (A ) ( ) ( ) ( ) Unique reflections 13,264 13,061 44,001 Multiplicity 7.2 (7.2) 3.7 (3.7) 3.6 (3.5) Completeness (%) 99.9 (99.8) 98.6 (97.6) 99.2 (98.1) R sym (%) 4.6 (6.2) 4.6 (7.1) 4.9 (8.2) Mean <I/σ(I)> 59.0 (39.1) 40.7 (23.3) 34.9 (16.0) Phasing statistics AuCl 3 derivative K 2Pt(CN) 4 derivative Phasing power (acentric/centric) 0.54 / / 1.33 Figure of merit Number of heavy atom sites 4 2 Refinement statistics Native (3X2N) R work / R free (%) 10.8 / 13.2 r.m.s.d. bonds (A ) r.m.s.d. angles ( ) Atoms (nonhydrogen) 1,669 Solvent molecules 363 Ramachandran plot Allowed (%) Outliers (%) 1.18 C2

4 Table S2: Statistics of data collection and refinement of the x-ray structure of PcCel45A WT with and without 5 mm cellopentaose at cryogenic temperature. Data collection statistics Unliganded With 5 mm cellopentaose Beam line Spring-8 BL44XU KEK NE3A Wavelength (Å) Space group P Unit-cell parameters a, b, c (Å) 45.6, 58.2, , 57.6, 63.4 Exposure time 0.5 sec (high, mid and low) 5 sec (high) 0.5 sec (mid) 0.25 sec (low) Number of Frames 360 (ω = 0.5º, high, mid, low) 1440 (ω = 0.25º, high) Resolution (Å) ( ) ( ) Unique reflections Multiplicity 8.2 (4.0) 10.0 (2.0) Completeness (%) 97.0 (92.5) 91.1 (68.4) R merge (%) 1.61 (40.2) 10.7 (32.4) Mean <I/σ(I)> 15.1 (3.0) 62.3 (2.2) Refinement statistics Unliganded (3X2L) With 5 mm cellopentaose (3X2M) R work / R free (%) 11.5 / / 12.8 r.m.s.d. bonds (A ) r.m.s.d. angles ( ) Atoms (nonhydrogen) Solvent molecules Hydrogen atoms Ramachandran plot Favored (%) Outliers (%) Rotamer outliers (%)

5 Table S3: Statistics of data collection and refinement of the x-ray structure of PcCel45A N92D with and without cellopentaose at cryogenic temperature and PcCel45A N92D unliganded structure at room temperature. Data collection statistics Unliganded With 5 mm cellopentaose Room temp unliganded Beam line KEK BL5A KEK BL5A R-axis IV ++ Wavelength (Å) Space group P Unit-cell parameters a, b, c (Å) 45.6, 58.0, , 57.3, , 59.0, 64.3 Exposure time 0.5 sec (high, low) 0.5 sec (high, low) 3 min Number of Frames 720 (ω = 0.5º, high, low) 720 (ω = 0.5º, high, low) 720 (ω = 0.5º) Resolution (Å) ( ) ( ) ( ) Unique reflections Multiplicity 7.0 (6.4) 12.7 (4.3) 13.3 (11.4) Completeness (%) 99.6 (100) 96.8 (91.1) 99.2 (95.3) R merge (%) 5.6 (20.5) 6.7 (24.0) 11.8 (25.3) Mean <I/σ(I)> 39.6 (7.3) 89.9 (7.6) 17.4 (8.3) Refinement statistics Unliganded (3X2G) With 5 mm cellopentaose (3X2H) Room temp unliganded (3X2I) R work / R free 9.7 / / / 15.1 r.m.s.d. bonds (A ) r.m.s.d. angles ( ) Atoms (nonhydrogen) Solvent molecules Hydrogen atoms Ramachandran plot Favored (%) Outliers (%) Rotamer outliers (%)

6 Table S4: Statistics of data collection and refinement of the x-ray structure of PcCel45A D114N with and without 5 mm cellopentaose at cryogenic temperature. Data collection statistics Unliganded With 5 mm cellopentaose Beam line KEK BL5A KEK NE3A Wavelength (Å) Space group P Unit-cell parameters a, b, c (Å) 44.0, 58.8, , 58.3, 62.8 Exposure time 0.5 sec (high, low) 1 sec Number of Frames 720 (ω = 0.5º, high, low) 720 (ω = 0.5º) Resolution (Å) ( ) ( ) Unique reflections Multiplicity 8.9 (6.3) 14.3 (5.9) Completeness (%) 99.5 (99.2) 98.6 (71.6) R merge (%) 4.4 (17.4) 12.1 (45.7) Mean <I/σ(I)> 28.9 (8.6) 41.9 (4.9) Refinement statistics Unliganded (3X2J) With 5 mm cellopentaose (3X2K) R work / R free 11.6 / / 13.1 r.m.s.d. bonds (A ) r.m.s.d. angles ( ) Atoms (nonhydrogen) Solvent molecules Hydrogen atoms Ramachandran plot Favored (%) Outliers (%) Rotamer outliers (%)

7 Table S5: Statistics of data collection and refinement of the neutron and x-ray structure of PcCel45A WT unliganded. Data collection statistics X-ray Neutron Beam line KEK BL5A J-PARC ibix Wavelength (Å) Space group P Unit-cell parameters a, b, c (Å) a 46.2, 59.1, 64.3 Exposure time 1 sec (high) 13.5 h / Frame Number of Frames 720 (ω = 0.5º, high, mid low) 27 Resolution (Å) ( ) ( ) Unique reflections Multiplicity 13.1 (6.3) 3.9 (2.1) Completeness (%) 99.5 (99.4) 93.3 (81.1) R merge (%) 8.0 (29.9) 24.5 (50.1) Mean <I/σ(I)> 106 (6.6) 3.9(1.4) Refinement statistics X-ray only NX joint (3X2O) X-ray R work / R free 10.9 / / 15.3 Neutron R work / R free / 25.2 r.m.s.d. bonds (A ) r.m.s.d. angles ( ) Atoms Solvent molecules Hydrogen / 376 / / 375 Ramachandran plot Favored (%) Outliers (%) Rotamer outliers (%) a Cell constants were calculated from X-ray data.

8 Table S6: Statistics of data collection and refinement of the neutron and x-ray structure of PcCel45A WT with 2.5 mm cellopentaose. Data collection statistics X-ray Neutron Beam line KEK BL5A J-PARC ibix Wavelength (Å) Space group P Unit-cell parameters a, b, c (Å) a 46.5, 58.6, 64.7 Exposure time 0.5 sec (high, mid, low) 13.5 h / Frame Number of Frames 720 (ω = 0.5º, high, mid, 28 Resolution (Å) ( ) ( ) Unique reflections Multiplicity 16.3 (5.9) Completeness (%) 98.5 (92.9) 94.1 (84.0) R merge (%) 7.3 (34.1) Mean <I/σ(I)> 110 (5.2) Refinement statistics X-ray only NX joint (3X2P) X-ray R work / R free 9.7 / / 14.2 Neutron R work / R free / 26.0 r.m.s.d. bonds (A ) r.m.s.d. angles ( ) Atoms (nonhydrogen) Solvent molecules Hydrogen / Deuterium atoms 402 / / 383 Ramachandran plot Favored (%) Outliers (%) Rotamer outliers (%) a Cell constants were calculated from X-ray data.

9 Table S7: Statistics of data collection and refinement of the x-ray structure of PcCel45A N105D. Data collection statistics Unliganded Beam line KEK BL5A Wavelength (Å) 0.8 Unit-cell parameters a, b, c (Å) 45.8, 58.0, 63.1 Exposure time 0.5 sec (high, mid, low) Number of Frames 720 (ω = 0.5º, high, mid, low) Resolution (Å) ( ) Unique reflections Multiplicity 14.5 (4.0) Completeness (%) 95.1 (85.1) R merge (%) 9.2 (30.1) Mean <I/σ(I)> 90.4 (6.4) Refinement statistics Unliganded (4ZM7) R work / R free 12.0 / 13.5 r.m.s.d. bonds (A ) r.m.s.d. angles ( ) Atoms (nonhydrogen) 1849 Solvent molecules 469 Hydrogen atoms 1004 Ramachandran plot Favored (%) Outliers (%) 0.00 Rotamer outliers (%) 0.00

10 Figure S1: Determination of anomeric forms of products. The anomer ratio of the products generated from PASC by PcCel45A was checked by HPLC. The products after 1 min reaction in 50 mm sodium acetate at 30ºC are shown in green, and the products after incubation in 30ºC for 120 min following filtration are shown in blue. Standard oligo-saccharides (glucose to cellohexaose) incubated in 50 mm sodium acetate at 30ºC are shown in red.

11 Figure S2: Structural comparison of PcCel45A with other GH45 enzymes. (A) Structural alignment of PcCel45A with other GH 45 family enzymes and related proteins. The alignment was generated by DALI server, where structurally equivalent positions with PcCel45A are shown in uppercase. The proposed general acid residues and corresponding residue to Asn92 of PcCel45A are show in red and blue background, respectively. The long loop region forming substrate-binding cleft of GH-45 enzymes are indicated by orange (PcCel45A and MeCel45A) and pink (HiCel45A and MaCel45A) background. The green and blue boxes indicates loop region specific to - and - expansin, respectively. (B) Comparison of loop regions of PcCel45A and HiCel45A. PcCel45A has a longer loop region covering the active site from the upper side (shown in orange), whereas HiCel45A has only a short loop on the upper side but has an additional loop on the underside (shown in red).

12 Figure S3: Activity profile of PcCel45A WT and mutants. (A) Reaction velocities of PcCel45A WT (red square), N92D (green circle) and D114N (blue triangle) towards amorphous cellulose are shown. The activity of WT was highest at ph 3.0. PcCel45A D114N was inactive (under min -1 ), and the activity of N92D was drastically decreased. (B) Relative activities of PcCel45A WT (red square) and N92D (green circle) are shown. N92D showed the highest activity at ph 5.5. (C) The activities of PcCel45A WT at ph 3.0 at various concentrations of tris(hydroxymethyl)aminomethane.

13 Figure S4: Temperature effect on PcCel45A structure. (A) Schematic illustration of water molecules on the PcCel45A WT surface at room temperature and at cryogenic temperature. The resolution of the data at cryogenic temperature was adjusted to 1.0 Å to match the resolution at room temperature. The positions of water molecules were different between the structures at room temperature and cryo temperature and water molecule networks were broken. (B) Conformations of Asn92 and Asp114 of PcCel45A WT without cellopentaose at room and cryogenic temperatures. Water molecules at the catalytic center are shown as spheres and the tris molecule is shown as a stick model. The F obs - F calc maps of tris and water molecules are shown at 2.5 sigma (C) Unliganded structure of the catalytic center of PcCel45A N92D at room and cryogenic temperatures. Water molecules at the catalytic center are shown as spheres and the tris molecule is shown as a stick model.

14 Figure S5: Substrate recognition of PcCel45A WT and mutants. (A) Two cellopentaoses were bound at subsites - 5 to -1 and +1 to +5 respectively in all enzymes, but the occupancies of cellopentaose binding on the minus side in WT and N92D were only about 0.6. (B) Omit maps of cellopentaose at subsite -2 to +2 of PcCel45A WT, N92D and D114N are shown at the 2 sigma level. The glucose rings at subsite -1 of WT and N92D were distorted by the interactions between O6 and Asp114, a possible general acid residue, and between O2 and

15 Asn92 or Asp92. On the other hand, the glucose at subsite -1 of D114N was not distorted, even though O6 interacted with Asn114. Figure S6: Joint refined structure of PcCel45A WT with 2.5 mm cellopentaose. (A) X-Ray diffraction 2F obs-f calc maps (2.0 sigma in blue) and F obs-f calc maps (3.0 sigma in green and red for positive and negative) are compared for the two conformations and amide or imidic acid forms of Asn92. B-factors (A 2 ) are shown above the atoms. (B) Neutron diffraction 2F obs-f calc maps (1.0 sigma in purple) and F obs-f calc maps (2.0 sigma in green and red for positive and negative) are compared for the two conformations and amide or imidic acid forms of Asn92. B- factors are shown above the atoms, and hydrogen/deuterium ratios are also shown. (C) Joint refined structure of cellopentaose binding at subsite +1 to +5 of PcCel45A. X-Ray diffraction 2F obs-f calc map (blue) and neutron diffraction F obs-f calc map (green and red) are shown at the 1.0 sigma and 2.0 sigma levels, respectively. (D) Proton relay pathway of PcCel45A with cellopentaose. X-Ray diffraction 2F obs-f calc map (blue) and neutron diffraction F obs-f calc map (green and red) are shown at the 2.0 sigma and 1.5 sigma levels, respectively. Asn92 was flipped by the interaction with cellopentaose and the connection between Asn105 and Ser14 in the neutron

16 diffraction F obs-f calc map disappeared. (E) Proposed protonation/deuteration state of PcCel45A WT with 2.5 mm cellopentaose.

17 Figure S7 Neighboring pair residues in the proton relay pathway. The 2F obs-f calc map (blue) of the X-ray analysis is shown at the 1.0 sigma level, and the F obs-f calc map (red and green) of the neutron analysis is shown at the 2.0 sigma level.

18 Figure S8. Structural and activity profile of PcCel45A N105D mutant. (A) Comparison of loop structures of N105D mutant and WT of PcCel45A. The structure of the loop region including Asn92 of the N105D mutant was loosened and the side chain of Asn92 was disordered. Double conformers of Ser14 were observed. (B) Reaction velocity of PcCel45A N105D (purple diamonds) towards amorphous cellulose. The velocity of WT is also shown for reference (red squares). The N105D mutant became inactive (less than 0.002/min) at ph values higher than ph 5.5, but retained 74% of the activity of WT at ph 3.0.

19 Figure S9. SDS-PAGE and Native-PAGE of PcCel45A WT and mutants. (A) SDS-PAGE of PcCel45A WT and D114N and N92D. Enzyme (2 µg) was applied on 12% polyacrylamide gel. (B) Native PAGE of PcCel45A WT and D114N and N92D. Enzyme (100 µg) was applied on 12 % polyacrylamide gel.

20 Caption for movie S1 Movie S1. The proposed proton relay pathway between Asp 92 (general base) and Asn 114 (general acid) residues is shown as a movie. In the initial phase, Asn 92, amide between Phe 95 and Cys 96 and Asn 105 are in imidic acid form. The neutron diffraction 2F obs-f calc map of this phase was calculated based on the initial structure. The neutron diffraction 2F obs-f calc and F obs-f calc maps during the proton relay phase were calculated based on the structure without hydrogen/deuterium atoms. After the proton relay, the hydrogen/deuterium atoms were transferred to the next residues, and the neutron diffraction 2F obs-f calc map was calculated based on this structure. In the latter half of the movie, residues return to the initial structure by reverse relay. The neutron diffraction 2F obs-f calc and F obs- F calc maps are shown at the 1.0 and 1.5 sigma levels, respectively.