Abstract
Bacteriophage T4 lysozyme (T4L) has been used as a paradigm for seminal biophysical studies on protein structure, dynamics, and stability. Approximately 700 mutants of this protein and their respective complexes have been characterized by X‐ray crystallography; however, despite the high resolution diffraction limits attained in several studies, no hydrogen atoms were reported being visualized in the electron density maps. To address this, a 2.2 Å‐resolution neutron data set was collected at 80 K from a crystal of perdeuterated T4L pseudo‐wild type. We describe a near complete atomic structure of T4L, which includes the positions of 1737 hydrogen atoms determined by neutron crystallography. The cryogenic neutron model reveals explicit detail of the hydrogen bonding interactions in the protein, in addition to the protonation states of several important residues.
Keywords: T4‐lysozyme, neutron, structure, X‐ray, hydrogen
Short abstract
Introduction
T4 lysozyme (T4L) is an endoacetylmuramidase that is produced by the bacteriophage T4 and is subsequently used to infection of Escherichia coli with bacteriophage DNA.1 Thus far, approximately 700 mutants of T4L and their respective complexes have been characterized. As such, this protein has been used as a paradigm for seminal biophysical studies on protein structure, dynamics, and stability.2 These studies rely upon protein and water mediated hydrogen bonding networks that are geometrically inferred from crystal structures, since hydrogen (H) atoms are not resolved at the resolution of the available X‐ray structures.3 In contrast to X‐ray diffraction, neutron diffraction can directly resolve the positions of hydrogen atoms in protein crystals, even at moderate resolution.4, 5 Here, we report the neutron crystal structure of perdeuterated T4 lysozyme cysteine‐free pseudo‐wild type6 (wt*T4L) determined at cryogenic temperature (80 K) to a resolution of 2.2 Å. The neutron structure provides a near complete atomistic description of the hydrogen‐mediated interaction networks in T4L, clarifying both the protonation state of several key residues and the position of a hydrogen atom involved in an unusually short interaction.
Results and Discussion
Neutron structure of wt*T4L
Perdeuterated wt*T4L (d‐wt*T4L), on which hydrogen atoms were replaced with deuterium, was expressed in E. coli grown in deuterated Enfors minimal media,7 before it was subsequently purified via established protocols.8 Crystals were grown under deuterated conditions, which were modified slightly from those which had been previously published.9 A ∼0.7 mm3 crystal was flash cooled in liquid nitrogen, before being transferred to a closed cycle refrigerator,10 precooled to 80K, for data collection on the IMAGINE11 beamline at Oak Ridge National Laboratory's High Flux Isotope Reactor. The neutron crystal structure of d‐wt*T4L was determined to a resolution of 2.20 Å and refined to R work/R free of 24.2/28.0. A total of 1737 protein deuterium atoms and 36 D2O molecules are modeled in the final structure. The X‐ray structure of a second d‐wt*T4L crystal was determined to a resolution of 1.63 Å and refined to Rwork/Rfree of 15.8/18.6. The crystal was obtained at a pD of 6.2. Both neutron and cryogenic X‐ray data‐collection and refinement statistics are summarized in Table 1.
Table 1.
Data Collection and Refinement Statistics
| Neutron | X‐ray | |
|---|---|---|
| Data collection | ||
| Wavelength | 2.8–4.6 | 1.54 |
| Resolution range (Å) | 16.7–2.20 (2.32–2.20) | 50.0–1.63 (1.69–1.63) |
| Unique reflections | 7965 (938) | 25,255 (2425) |
| Multiplicity | 3.9 (2.9) | 10.1 (9.8) |
| Completeness (%) | 72.7 (60.0) | 99.7 (98.6) |
| Mean I/sigma(I) | 6.5 (3.7) | 20.1 (10.7) |
| R‐merge | 0.162 (0.214) | 0.096 (0.318) |
| Unit cell | ||
| Space group | P3221 | |
| Unit cell | a=61.5 b=61.5 c= 95.9 90 90 120 | |
| Refinement | ||
| Reflections used in refinement | 7962 (572) | 25,199 (2450) |
| Reflections used for R‐free | 190 (13) | 2008 (197) |
| R‐work | 0.242 (0.277) | 0.158 (0.170) |
| R‐free | 0.280 (0.301) | 0.186 (0.206) |
| Number of hydrogen atoms | 1737 | 0 |
| Number of non‐hydrogen atoms | 1347 | 1703 |
| Macromolecules | 1309 | 1388 |
| Ligands | 2 | 22 |
| Solvent | 36 | 293 |
| Protein residues | 164 | 164 |
| r.m.s.d. from ideal | ||
| r.m.s.d (bonds) | 0.004 | 0.006 |
| r.m.s.d (angles) | 0.65 | 0.79 |
| Ramachandran favored (%) | 96.30 | 98.15 |
| Ramachandran allowed (%) | 3.70 | 1.85 |
| Ramachandran outliers (%) | 0 | 0.00 |
| Rotamer outliers (%) | 3.65 | 0.68 |
| Clashscore | 8.02 | 0.00 |
| Average B‐factor (Å2) | 14.76 | 21.01 |
| Macromolecules | 14.88 | 18.11 |
| Ligands | 5.84 | 30.24 |
| Solvent | 10.89 | 34.04 |
aStatistics for the highest‐resolution shell are shown in parentheses.
Overall, the neutron and X‐ray structures of d‐wt*T4L are similar to the structure of hydrogenated wild type T4L (h‐wtT4L, PDB code: 3FA0), superimposing with r.m.s.d's of 0.28 Å and 0.18 Å respectively, for the backbone atoms [Fig. 1(A)]. Asn163 and Leu164 are removed from analysis due to the poor density for the C‐terminal residues which are common for other T4L X‐ray structures. Comparisons of the temperature‐factor distribution of Cα atoms in the respective d‐wt*T4L and h‐wtT4L structures are highly similar [Fig. 1(B)], thus demonstrating that the structure of T4L is not significantly perturbed as result of deuteration.
Figure 1.

(A) A ribbon representation showing orthogonal views of d‐wt*T4L with oriented heavy water molecules shown in ball and stick representation. The model is colored according to r.m.s.d. alignment values of d‐wt*T4L neutron structure with h‐wtT4L X‐ray structure (PDB code: 3AF0) ranging from blue to red. The best aligned residues are colored blue and the worst aligned residues are colored red. (B) Comparison of thermal B factors of d‐wt*T4L neutron (black) and X‐ray (grey) with h‐wtT4L (light grey, PDB code: 3AF0) plotted against residue number. Locations of α‐helices are indicated by horizontal red bars. Locations β ‐ sheets are indicated by red dots. Blue vertical bars are the average hydrogen bonds of each α‐helix and overall β – sheets. The orange line indicates the average hydrogen bond lengths formed by backbone atoms, and the blue line indicates the average hydrogen bond length formed by backbone atoms on turns. The histogram of hydrogen bond angles (C) and donor‐acceptor distances (D) includes: the main chain, side chain, and water molecules involved hydrogen bonds
Intermolecular hydrogen bonds
Analysis of the hydrogen bonding patterns in T4L and its various mutants has been the interest of numerous studies, yet the positions of the interacting polar hydrogen atoms were geometrically inferred. Owing to both the resolution of the neutron data and the strong scattering of deuterium by neutrons, these missing hydrogen atoms and their bonding are now directly observed, consequently allowing for detailed analysis of the interaction network. A total of 180 hydrogen bonds were refined in the d‐wt*T4L neutron structure. The hydrogen donor‐acceptor distance and donor angle (D‐H—A) were analyzed separately for the main chain, side chains and solvent molecules [Fig. 1(B,C)]. Among the 102 hydrogen bonds formed by backbone atoms, 84 are involved in maintaining main chain α‐helices and β‐sheets. In the main chain, the average length of the hydrogen donor‐acceptor distance is 2.1 Å, with an average donor angle of 154°. The protein can be divided into two separate regions [Fig. 1(A)], an N‐terminal subdomain (residues 13–65) and a C‐terminal subdomain (residues 1–12 and 65–164). The N‐terminal domain is composed mainly of helix B, a β‐sheet region and the N‐terminal end of helix C, while helices D‐J, helix A and the C‐terminal end of helix C form the C‐terminal domain. The average hydrogen donor‐acceptor distance is 1.9 Å within the β‐sheets, which is 0.2 Å shorter than that exhibited by the main chain [Fig. 1(B)].
Side chains participate in 40 hydrogen bonds. The average hydrogen donor‐acceptor distance of side chain hydrogen bonds is 2.1 Å, and the donor angle is 147° [Fig. 1(C,D)]. Three interaction networks revealed by neutron the crystal structure are of distinct interest:
Salt bridge between His31‐Asp70
The neutron structure of d‐wt*T4L shows that His31 is doubly protonated and positively charged, while Asp70 is negatively charged under the crystallization conditions (pD ∼ 6.2) [Fig. 2(A)]. The doubly protonated His31 in our neutron structure supports earlier NMR titration measurements that showed the pKa of His31 shifts from 6.3 in the unfolded state to 9.1 in the folded state.10 His31 and Asp70 residues form a salt bridge, which contributes 3–5 kcal/mol to the free energy of T4L in the folded state.12 The length of the salt bridge between His31 ND1 and Asp70 OD2 in h‐wtT4L (PDB code: 3FA0) and d‐wt*T4L X‐ray structures is 2.66 Å and 2.7 Å, respectively. In the neutron structure, the position of DD1 is refined. The length between His31 DD1 and Asp70 OD2 is 1.9 Å, with the distance between His31 ND1 and Asp70 OD2 being 2.9 Å. Asp70 OD2 is also hydrogen bonded to DOD6.
Figure 2.

Nuclear density maps of d‐wt*T4L. (A) Salt bridge between His31‐Asp70. (B) An unusual interaction between Ser117 and Asn132. Glu45symm is shown in magenta. (C) Hydrogen bonds network around Thr157. (D) Stereo view of two water molecules in Cavity 1. (E) Stereo view of the water molecule in Cavity 2. The nuclear density 2|Fo|‐|Fc| maps are in grey, contoured at 1.0 r.m.s.d. Omit nuclear density |Fo|‐|Fc| maps are in red, contoured at 1.8 r.m.s.d
An unusually short interaction
Previous mutant studies showed that substitution of Ser117 with other non‐H‐bonding residues improves the stability of T4L.2, 13 The X‐ray structure refinement of h‐wtT4L at 1.09 Å (PDB code: 3FA0) indicates that the interaction distance between side chain hydroxyl of Ser117 and side chain of Asn132 is 2.52 Å.2, 3 This distance is measured to be 2.5 Å and 2.2 Å in the d‐wt*T4L X‐ray and neutron structures, respectively. The neutron structure directly reveals the Ser117 DG hydrogen atom position. Unexpectedly, the orientation of the hydroxyl group on Ser117 points away from Asn132 OD and leaves a direct contact of Asn132 OD with Ser117 OG [Fig. 2(B)]. This observation is consistent with an earlier suggestion that, rather than being a strong hydrogen bond, this interaction is actually an unfavorable van der Waals contact.2 However, we note a strong donor‐acceptor distance (1.7 Å) between Asn132 DD21 and Glu45symm OE2 on its symmetry mate that draws the negative electron density away from carbonyl group and subsequently makes Asn132 OD more positive [Fig. 2(B)]. The distance between Asn132 ND2 and Glu45symm OE2 is 2.52 Å in h‐wtT4L (PDB code: 3FA0). The additional hydrogen bond helps to stabilize the aforementioned unfavorable van der Waals contact. This interaction is present in the crystal lattice, however, it is not seen in solution. It is possible that a water molecule is hydrogen bonded with Asn132 DD21 in solution to mimic the aforementioned hydrogen bond between Asn132 DD21 and the Glu45 OE2 on the crystallographic symmetry mate.
Hydrogen bonds network around Thr157
The contribution of the local hydrogen bond network around Thr157 to the thermal stability of T4L has been examined in detail.14, 15 Replacement of Thr157 with Ile destabilizes the protein by 2.9 kcalmol−1, which has been attributed to the loss of a hydrogen bond. Structural analysis of 13 destabilizing mutations suggests two potential hydrogen bonds that may contribute to the stability: Thr157 OG1–Thr155 OG1 and Thr157 OG1–Asp159 N. Generally, the most stable mutants are those where substitutions of Thr157 are hydrogen bonded to the buried main chain amide of Asp159.15 Whereas the geometry of the putative hydrogen bonds inferred from the X‐ray structure was ambiguous, the direct visualization of H atoms in the neutron structure explicitly reveals and confirms the nature of these hydrogen bond interactions [Fig. 2(C)].
Water molecules
The 40 most highly conserved solvent‐binding sites in 10 different X‐ray T4L have been previously reported.16 All 40 of these waters are observed in the d‐wt*T4L X‐ray structure. Of these, 11 waters can be oriented and refined as D2O molecules in the neutron structure (Table 2). A further 21 waters are refined as D2O molecules elsewhere in the structure. Although many of these water molecules are found in other T4L structures, they are not amongst the most 40 highly conserved in T4L that are identified elsewhere. An additional 4 waters were partially ordered (D‐O, O only). Twenty‐seven hydrogen bonds were found between water molecules and protein in the neutron structure, with the average donor‐acceptor distance at 2.0 Å and an average the donor angle of 154° [Fig. 1(C,D)].
Table 2.
Comparison of water molecules in the neutron structure (d‐wt*T4L) and X‐ray structure (h‐wtT4L, PDB code: 4LZM).16
| Neutron Structure | X‐ ray Structure | ||
|---|---|---|---|
| Solvent ID | Hydrogen bonds | Solvent ID | Hydrogen bonds |
| 6 | Asp70 OD2, DOD20 O a | 171 | ILe29 O, Asp70 OD2 |
| 14 | Thr152 OG1, Thr157 O, Ala160 H | 175 | Thr152 OG1, Thr157 O, Ala160 N |
| 1 | Asn101 OD1 | 208 | Asn101 OD1, Gln105 OE1 |
| 15 | Ser36 O, Tyr25 H | 190 | Thr34 O, Tyr25 N, Ser36 O |
| 17 | n.d.b | 232 | Ala160O |
| 2 | Asp61 O | 296 | Asp61 O |
| 22 | DOD21 O | 211 | Arg148 NH1 |
| 31 | n.d. | 196 | Thr54 O, Val57 O, Arg52 NH1, Arg52 NH2 |
| 7 | n.d. | 201 | Gln69 O, Arg76 NH2 |
| 40 | n.d. | 299 | Phe104 O |
| 24 | Ala130 O | 216 | Aal130 O |
Different hydrogens bond are shown in bold.
n.d stands for not determined.
Several cavities are present in wt*T4L that are large enough to contain one or more water molecules. Water molecules, and the hydrogen‐bonding networks formed in these cavities, are readily visualized in the nuclear density maps [Fig. 2(D,E)]. Cavity 1, located between helix C and β sheets near the active cleft, has two‐ordered water molecules, DOD6 and DOD20. This cavity is surrounded by Gly28, Ile29, Ala63, Leu66, Phe67, and Asp70. The two water molecules are hydrogen bonded to each other (2.4 Å) with the oxygen atom of DOD20 as an acceptor. The oxygen atom of DOD20 (refined as DO) is also hydrogen bonded to the backbone D of Ile29 (1.7 Å). The DOD6 molecule is hydrogen bonded to OD2 of Aps70 (1.7 Å) as a hydrogen bond donor. Cavity 2 is relatively small and contains a single water molecule (DOD1). It is surrounded by Asn101, Met102, Gln105, Trp138, and Val149. This water molecule is only hydrogen bonded to Asn101 OD1 (2.1 Å) in the neutron structure. A proposed hydrogen bond to Gln105 OE117 is not observed in the current neutron structure.
In summary, the direct observation of hydrogen atoms in the neutron structure of d‐wt*T4L provides a nearly complete atomistic description of this paradigmatic biophysical model, in addition to an unprecedented level of detail regarding hydrogen bonding interactions in the protein. This additional information is valuable for future computational studies on protein folding and dynamics and will provide new insights into studies on T4L in which this information is otherwise unattainable.
Materials and Methods
Protein expression and purification
The T4 lysozyme used in this study was based on the cysteine‐free pseudo‐wild type T4L.6 Perdeuterated protein (d‐wt*T4L) was expressed using E. coli strain BL21, as previously described.18 Briefly, cells were gradually adapted over three days from LB culture medium to deuterated Enfors minimal media. The protein was expressed in about 40 mg/L culture by fermentation protocols utilizing d8‐glycerol as the carbon source. Cells were induced using 1mM IPTG at an OD600nm of ∼8–9 for 1.5 hour before harvesting. Purification of d‐wt*T4L followed the published procedure.8 The soluble fraction of the cell lysate was dialyzed to remove extra salt and was then purified on a cation exchange column (HiTrap TM, CM Sepharose FF, GE Healthcare). Pooled fractions were concentrated and finally purified by passage over a Superdex 75 size exclusive column with crystallization buffer (50 mM MOPS, 25 mM NaCl, pH 6.8). Final protein purity was greater than 95% as judged by SDS‐PAGE.
Purified d‐wt*T4L was concentrated to about 10–20 mg/mL (ɛ = 24 990 cm−1 M−1 at 280 nm) and crystalized using micro‐seeding techniques. First, small seed crystals of h‐wt*T4L were obtained by sitting drop vapor diffusion as previously reported (∼2.0 M Na/K phosphate, pH 6–7, 250 mM NaCl, 40 mM 2‐hydroxyethyl disulfide).9 h‐wt*T4L crystals were then crushed using a seed bead (Hampton Research). The original seeds stock was diluted 10,000 times by reservoir solution and was subsequently used as seeds for crystalizing d‐wt*T4L in deuterated conditions. The crystallization condition was similar to the h‐wt*T4L condition (∼2.0 M Na/K phosphate, pD 6–7, 250 mM NaCl, 40mM 2‐hydroxyethyl disulfide). The drop size was 50 μL protein solution plus 50 μL reservoir solution and the reservoir solution volume was 0.5 mL. Large d‐wt*T4L crystals (∼0.5 mm3) appeared after incubation for one month at 20°C.
Crystallographic data collection and refinement
To prepare for cryogenic temperature neutron data collection, crystals were screened at room temperature for diffraction viability on the MaNDi instrument19 at the Spallation Neutron Source at Oak Ridge National laboratory (ORNL). The selected crystal was rapidly passed through the cryo‐protectant (50% d8‐glycerol 1:1 mixed with the deuterated crystal reservoir solution) and flash cooled in liquid nitrogen bath. Low temperature (∼ 80K) neutron diffraction data was collected on IMAGINE at HIFR in ORNL.11 Low temperature 100K X‐ray diffraction data was collected on a home‐source Rigaku HF007 micromax diffractometer. Phases were determined by molecular replacement using the Phaser program20 with the coordinates of h‐wt*T4L (PDB Code: 1LW9) as a search model. X‐ray and neutron structure refinements were completed using PHENIX21 independently with stepwise cycles of manual model building using COOT.22 The data collection and structure refinement statistics are listed in Table 1. The final models exhibit good stereochemistry as determined by MolProbity.23 Atomic coordinates and structure factors have been deposited in the Protein Data Bank24 under the accession codes 5VNQ for neutron structure and 5VNR for X‐ray structure. All molecular graphics figures were prepared with PyMol.25
Acknowledgments
The authors would like to thank K. Weiss for assistance in protein perdeuteration. The authors would also like to thank S. Lucas for critical reading on the manuscript.
This research was sponsored by the Laboratory Directed Research and Development Program of Oak Ridge National Laboratory (ORNL), managed by UT‐Battelle, LLC, for the U.S. Department of Energy (DOE) under Contract No. DE‐AC05‐00OR22725. Research conducted at ORNL's High‐Flux Isotope Reactor and Spallation Neutron Source was sponsored by the US Department of Energy's (DOE) Office of Basic Energy Sciences, Scientific User Facilities Division, and laboratory facilities supported by the DOE Office of Biological and Environmental Research, Project ERKP291. The IMAGINE Project was partially supported by the National Science Foundation (Grant 0922719).
References
- 1. Weaver LH, Matthews BW (1987) Structure of bacteriophage T4 lysozyme refined at 1.7 A resolution. J Mol Biol 193:189–199. [DOI] [PubMed] [Google Scholar]
- 2. Baase WA, Liu L, Tronrud DE, Matthews BW (2010) Lessons from the lysozyme of phage T4. Protein Sci 19:631–641. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Mooers BHM, Matthews BW (2004) Use of an ion‐binding site to bypass the 1000‐atom limit to structure determination by direct methods. Acta Cryst D60:1726–1737. [DOI] [PubMed] [Google Scholar]
- 4. Myles DA (2006) Neutron protein crystallography: current status and a brighter future. Curr Opin Struct Biol 16:630–637. [DOI] [PubMed] [Google Scholar]
- 5. O'Dell WB, Bodenheimer AM, Meilleur F (2016) Neutron protein crystallography: A complementary tool for locating hydrogens in proteins. Arch Biochem Biophys 602:48–60. [DOI] [PubMed] [Google Scholar]
- 6. Matsumura M, Matthews BW (1989) Control of enzyme activity by an engineered disulfide bond. Science 243:792–794. [DOI] [PubMed] [Google Scholar]
- 7. Törnkvist M, Larsson G, Enfors SO (1996) Protein release and foaming in Escherichia coli cultures grown in minimal medium. Bioprocess Eng 15:231–237. [Google Scholar]
- 8. Fleissner MR, Cascio D, Hubbell WL (2009) Structural origin of weakly ordered nitroxide motion in spin‐labeled proteins. Protein Sci 18:893–908. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Matthews BW, Dahlquist FW, Maynard AY (1973) Crystallographic data for lysozyme from bacteriophage T4. J Mol Biol 78:575–576. [DOI] [PubMed] [Google Scholar]
- 10. Myles DAA, Dauvergne F, Blakeley MP, Meilleur F (2012) Neutron protein crystallography at ultra‐low (<15 K) temperatures. J Appl Cryst 45:686–692. [Google Scholar]
- 11. Meilleur F, Munshi P, Robertson L, Stoica AD, Crow L, Kovalevsky A, Koritsanszky T, Chakoumakos BC, Blessing R, Myles DAA (2013) The IMAGINE instrument: First neutron protein structure and new capabilities for neutron macromolecular crystallography. Acta Cryst D69:2157–2160. [DOI] [PubMed] [Google Scholar]
- 12. Anderson DE, Becktel WJ, Dahlquist FW (1990) pH‐Induced denaturation of proteins: a single salt bridge contributes 3–5 kcal/mol to the free energy of folding of T4 lysozyme. Biochemistry 29:2403–2408. [DOI] [PubMed] [Google Scholar]
- 13. Shoichet BK, Baase WA, Kuroki R, Matthews BW (1995) A relationship between protein stability and protein function. Proc Natl Acad Sci USA 92:452–456. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Gruter RA, Neefjes JJ, Tersmette M, de Goede REY, Tulp A, Huisman HG, Miedema F, Ploegh HL (1987) Contributions of hydrogen bonds of Thr 157 to the thermodynamic stability of phage T4 lysozyme. Nature 330:74–77. [DOI] [PubMed] [Google Scholar]
- 15. Grütter MG, Gray TM, Weaver LH, Alber T, Wilson K, Matthews BW (1987) Structural studies of mutants of the lysozyme of bacteriophage T4. The temperature‐sensitive mutant protein Thr157‐ Ile. J Mol Biol 197:315–329. [DOI] [PubMed] [Google Scholar]
- 16. Zhang X‐J, Matthews BW (1994) Conservation of solvent‐binding sites in 10 crystal forms of T4 lysozyme. Protein Sci 3:1031–1039. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Liu L, Quillin ML, Matthews BW (2008) Use of experimental crystallographic phases to examine the hydration of polar and nonpolar cavities in T4 lysozyme. Proc Natl Acad Sci USA 105:14406–14411. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Meilleur F, Weiss KL, Myles DAA (2009) Deuterium labeling for neutron structure‐function‐dynamics analysis. Methods Mol Biol 544:281–292. [DOI] [PubMed] [Google Scholar]
- 19. Coates L, Cuneo MJ, Frost MJ, He J, Weiss KL, Tomanicek SJ, McFeeters H, Vandavasi VG, Langan P, Iverson EB (2015) The macromolecular neutron diffractometer MaNDi at the spallation neutron source. J Appl Cryst 48:1302–1306. [Google Scholar]
- 20. Bunkóczi G, Echols N, McCoy AJ, Oeffner RD, Adams PD, Read RJ (2013) phaser.MRage: automated molecular replacement. Acta Cryst D69:2276–2286. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Afonine PV, Grosse‐Kunstleve RW, Echols N, Headd JJ, Moriarty NW, Mustyakimov M, Terwilliger TC, Urzhumtsev A, Zwart PH, Adams PD (2012) Towards automated crystallographic structure refinement with phenix.refine . Acta Cryst D68:352–367. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Emsley P, Lohkamp B, Scott WG, Cowtan K (2010) Features and development of Coot. Acta Cryst D66:486–501. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Chen VB, Arendall IIIWB, Headd JJ, Keedy DA, Immormino RM, Kapral GJ, Murray LW, Richardson JS, Richardson DC (2010) MolProbity: all‐atom structure validation for macromolecular crystallography. Acta Cryst D66:12–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Berman HM, Westbrook J, Feng Z, Gilliland G, Bhat TN, Weissig H, Shindyalov IN, Bourne PE (2000) The Protein Data Bank. Nucleic Acids Res 28:235. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Schrodinger LLC (2015) The PyMOL Molecular Graphics System, Version 1.8.
