research papers\(\def\hfill{\hskip 5em}\def\hfil{\hskip 3em}\def\eqno#1{\hfil {#1}}\)

Journal logoSTRUCTURAL
BIOLOGY
ISSN: 2059-7983

Structural comparison of protiated, H/D-exchanged and deuterated human carbonic anhydrase IX

CROSSMARK_Color_square_no_text.svg

aDepartment of Biology and Lund Protein Production Platform, Lund University, Sölvegatan 35, 223 62 Lund, Sweden, bDepartment of Biochemistry and Molecular Biology, University of Florida, Gainesville, FL 32610, USA, and cScientific Activities Division, European Spallation Source ERIC, Odarslövsvägen 113, 224 84 Lund, Sweden
*Correspondence e-mail: zoe.fisher@esss.se

Edited by P. Langan, Oak Ridge National Laboratory, USA (Received 16 May 2019; accepted 12 July 2019; online 22 August 2019)

Human carbonic anhydrase IX (CA IX) expression is upregulated in hypoxic solid tumours, promoting cell survival and metastasis. This observation has made CA IX a target for the development of CA isoform-selective inhibitors. To enable structural studies of CA IX–inhibitor complexes using X-ray and neutron crystallography, a CA IX surface variant (CA IXSV; the catalytic domain with six surface amino-acid substitutions) has been developed that can be routinely crystallized. Here, the preparation of protiated (H/H), H/D-exchanged (H/D) and deuterated (D/D) CA IXSV for crystallographic studies and their structural comparison are described. Four CA IXSV X-ray crystal structures are compared: two H/H crystal forms, an H/D crystal form and a D/D crystal form. The overall active-site organization in each version is essentially the same, with only minor positional changes in active-site solvent, which may be owing to deuteration and/or resolution differences. Analysis of the crystal contacts and packing reveals different arrangements of CA IXSV compared with previous reports. To our knowledge, this is the first report comparing three different deuterium-labelled crystal structures of the same protein, marking an important step in validating the active-site structure of CA IXSV for neutron protein crystallography.

1. Background

Carbonic anhydrases (CAs) are zinc-containing metallo­enzymes that catalyze the reversible hydration of CO2 to form HCO3 and H+. The first step of the reaction in the hydration direction results in a water molecule bound to the zinc that has to be deprotonated to generate a zinc-bound OH for subsequent reactions. The excess H+ is transported via an ordered hydrogen-bonded water network to a proton-shuttling residue, His64 (CA II numbering), that ultimately delivers the H+ to the bulk solvent (Coleman, 1967[Coleman, J. E. (1967). J. Biol. Chem. 242, 5212-5219.]; Silverman & McKenna, 2007[Silverman, D. N. & McKenna, R. (2007). Acc. Chem. Res. 40, 669-675.]). Numerous X-ray and neutron crystallo­graphic studies have shed light on the catalytic mechanism of CA and have provided insights into the details of the finely tuned active site that supports CO2 hydration and rate-limiting proton transfer at very high rates (kcat = 106 s−1; Kim et al., 2016[Kim, C. U., Song, H., Avvaru, B. S., Gruner, S. M., Park, S. & McKenna, R. (2016). Proc. Natl Acad. Sci. USA, 113, 5257-5262.]; Fisher et al., 2007[Fisher, S. Z., Maupin, C. M., Budayova-Spano, M., Govindasamy, L., Tu, C., Agbandje-McKenna, M., Silverman, D. N., Voth, G. A. & McKenna, R. (2007). Biochemistry, 46, 2930-2937.], 2011[Fisher, Z., Kovalevsky, A. Y., Mustyakimov, M., Silverman, D. N., McKenna, R. & Langan, P. (2011). Biochemistry, 50, 9421-9423.]; Domsic & McKenna, 2010[Domsic, J. F. & McKenna, R. (2010). Biochim. Biophys. Acta, 1804, 326-331.]).

There are 15 expressed CA isoforms in humans that show diversity in expression between tissues and organs, supporting a range of physiological functions. One of the isoforms, CA IX, has limited expression in healthy tissues but is upregulated in aggressive tumours, with its expression being controlled by hypoxia (Pastorek & Pastorekova, 2015[Pastorek, J. & Pastorekova, S. (2015). Semin. Cancer Biol. 31, 52-64.]). CA IX is a multidomain membrane-bound protein, with its catalytic CA domain facing extracellularly (Langella et al., 2018[Langella, E., Buonanno, M., Vullo, D., Dathan, N., Leone, M., Supuran, C. T., De Simone, G. & Monti, S. M. (2018). Cell. Mol. Life Sci. 75, 3283-3296.]; Alterio et al., 2009[Alterio, V., Hilvo, M., Di Fiore, A., Supuran, C. T., Pan, P., Parkkila, S., Scaloni, A., Pastorek, J., Pastorekova, S., Pedone, C., Scozzafava, A., Monti, S. M. & De Simone, G. (2009). Proc. Natl Acad. Sci. USA, 106, 16233-16238.]). CA IX upregulation is part of a number of cancer-cell adaptions to hypoxia and is thought to occur in response to the lowering of pH in the cancer extracellular environment. The tumour pH environment is adapted from a physiological pH of 7.4 to as low as 6.0 (Mahon et al., 2015[Mahon, B. P., Pinard, M. A. & McKenna, R. (2015). Molecules, 20, 2323-2348.]; Pastorek & Pastorekova, 2015[Pastorek, J. & Pastorekova, S. (2015). Semin. Cancer Biol. 31, 52-64.]). This acidification promotes metastasis, most likely through protease activation and degradation of the extracellular matrix. A meta-study of patient outcomes showed very poor patient prognosis when positive for CA IX expression (Kuijk et al., 2016[Kuijk, S. J. A. van, Yaromina, A., Houben, R., Niemans, R., Lambin, P. & Dubois, L. J. (2016). Front. Oncol. 6, 69.]). For these reasons CA IX is a promising target for cancer detection and therapy, but high sequence conservation among human CAs (30–80% amino-acid identity) results in indiscriminate binding of the currently available regime of clinically used CA inhibitors. Hence, there is a recognized need to develop isoform-specific inhibitors that inhibit CA IX strongly while ideally not inhibiting the other CAs (Pinard et al., 2015[Pinard, M. A., Mahon, B. P. & McKenna, R. (2015). Biomed Res. Int. 2015, 453543.]; Mahon et al., 2015[Mahon, B. P., Pinard, M. A. & McKenna, R. (2015). Molecules, 20, 2323-2348.]).

Recent work by Mahon and coworkers reported the biophysical characterization and first X-ray crystal structure of a surface-modified variant of CA IX (CA IXSV; Mahon et al., 2016[Mahon, B. P., Bhatt, A., Socorro, L., Driscoll, J. M., Okoh, C., Lomelino, C. L., Mboge, M. Y., Kurian, J. J., Tu, C., Agbandje-McKenna, M., Frost, S. C. & McKenna, R. (2016). Biochemistry, 55, 4642-4653.]). CA IXSV contains only the catalytic domain of CA IX, with the intracellular, transmembrane and PG domains removed. In addition to truncating the full-length protein, six surface mutations (C174S, L180S, M360S, A210K, A258K and F259Y) were also introduced. These were chosen to remove a disulfide bond, to reduce surface hydrophobicity and to promote crystallization based on the crystal contacts in CA II (Mahon et al., 2016[Mahon, B. P., Bhatt, A., Socorro, L., Driscoll, J. M., Okoh, C., Lomelino, C. L., Mboge, M. Y., Kurian, J. J., Tu, C., Agbandje-McKenna, M., Frost, S. C. & McKenna, R. (2016). Biochemistry, 55, 4642-4653.]). The native, full-length protein is produced in insect cells in low yields, and is not stable or very soluble (Alterio et al., 2009[Alterio, V., Hilvo, M., Di Fiore, A., Supuran, C. T., Pan, P., Parkkila, S., Scaloni, A., Pastorek, J., Pastorekova, S., Pedone, C., Scozzafava, A., Monti, S. M. & De Simone, G. (2009). Proc. Natl Acad. Sci. USA, 106, 16233-16238.]). In this manuscript, and in the models deposited in the PDB, we use CA IX numbering. All residues mentioned in the text, tables and figures are for CA IX.

Endogenous CA IX is functional in a lower pH environment compared with other CAs. As such, Mahon and co­workers measured the thermal stability and catalytic parameters of CA IX under different pH conditions and demonstrated its adaptation to low pH, giving rise to its structural and functional stability at pH values as low as 5.0. The pKa of the H+ donor and acceptor groups in the active site are also decreased compared with CA II, indicating an ability to retain enzymatic activity at a pH of ∼6 (Mahon et al., 2016[Mahon, B. P., Bhatt, A., Socorro, L., Driscoll, J. M., Okoh, C., Lomelino, C. L., Mboge, M. Y., Kurian, J. J., Tu, C., Agbandje-McKenna, M., Frost, S. C. & McKenna, R. (2016). Biochemistry, 55, 4642-4653.]).

Owing to the need for the development of CA isoform-specific inhibitors against CA IX, and to gain a deeper understanding of its active-site architecture, our goal is to utilize joint neutron and X-ray structures of CA IXSV alone and in inhibitor complexes to fine-tune compounds to preferentially bind CA IX over CA II (Langan & Chen, 2013[Langan, P. & Chen, J. C.-H. (2013). Phys. Chem. Chem. Phys. 15, 13705-13712.]; Aggarwal et al., 2013[Aggarwal, M., Boone, C. D., Kondeti, B. & McKenna, R. (2013). J. Enzyme Inhib. Med. Chem. 28, 267-277.]; Kovalevsky et al., 2018[Kovalevsky, A., Aggarwal, M., Velazquez, H., Cuneo, M. J., Blakeley, M. P., Weiss, K. L., Smith, J. C., Fisher, S. Z. & McKenna, R. (2018). Structure, 26, 383-390.]). Previous neutron crystallographic studies of CA II in complex with clinically used inhibitors (for example brinzolamide, ethoxzolamide and acetazolamide) revealed the role and importance of water and hydrogen bonds in mediating ligand-binding interactions (Kovalevsky et al., 2018[Kovalevsky, A., Aggarwal, M., Velazquez, H., Cuneo, M. J., Blakeley, M. P., Weiss, K. L., Smith, J. C., Fisher, S. Z. & McKenna, R. (2018). Structure, 26, 383-390.]; Fisher et al., 2012[Fisher, S. Z., Aggarwal, M., Kovalevsky, A. Y., Silverman, D. N. & McKenna, R. (2012). J. Am. Chem. Soc. 134, 14726-14729.]). In preparation for future neutron protein crystallo­graphic studies of experimental inhibitors binding to CA IX, we expressed unlabelled CA IX (H/H CA IXSV) and performed H/D exchange (H/D CA IXSV) on preformed crystals. We also expressed deuterated protein (D/D CA IXSV) for crystallization (Fisher et al., 2014[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. (2014). Acta Cryst. D70, 3266-3272.]; Koruza, Lafumat, Végvári et al., 2018[Koruza, K., Lafumat, B., Végvári, Á., Knecht, W. & Fisher, S. Z. (2018). Arch. Biochem. Biophys. 645, 26-33.]; Blakeley et al., 2015[Blakeley, M. P., Hasnain, S. S. & Antonyuk, S. V. (2015). IUCrJ, 2, 464-474.]).

There are several studies that have compared the properties, activities and X-ray crystal structures of unlabelled (H/H) and perdeuterated (D/D) versions of the same protein: cholesterol oxidase, haloalkane dehalogenase and arginase I (Golden et al., 2015[Golden, E., Attwood, P. V., Duff, A. P., Meilleur, F. & Vrielink, A. (2015). Anal. Biochem. 485, 102-108.]; Liu et al., 2007[Liu, X., Hanson, B. L., Langan, P. & Viola, R. E. (2007). Acta Cryst. D63, 1000-1008.]; Di Costanzo et al., 2007[Di Costanzo, L., Moulin, M., Haertlein, M., Meilleur, F. & Christianson, D. W. (2007). Arch. Biochem. Biophys. 465, 82-89.]). These studies all showed minimal structural effects owing to perdeuteration (D/D). However, we could not find studies in which X-ray crystal structures of H/D-exchanged versions were also included in the analysis. As H/D exchange is the most commonly used form of deuterium labelling for neutron protein crystallographic studies, it is important to determine and verify whether labelling by itself has any effect on the crystal structure. Here, we present a comparative structural analysis of three different isotopically labelled (H/H, H/D and D/D) forms of CA IXSV and show that the overall fold and the active-site side-chain conformations are mostly unaffected. However, there are some subtle changes in solvent positioning that may be owing to deuteration effects and/or to differences in the resolutions of the structure determinations. We also analyzed the crystallographic monomer-to-monomer contacts and packing for different space groups of CA IXSV compared with previous reports.

2. Materials and methods

In this study, we use three designations to indicate protiated, H/D-exchanged and deuterated status: H/H means protiated protein in protiated buffer, H/D refers to protiated protein that was subjected to vapour H/D exchange after crystallization and D/D is deuterated protein that was purified in protiated buffers and then subjected to back-exchange in solution to recover any lost D atoms. For all studies we used a construct created by Mahon et al. (2016[Mahon, B. P., Bhatt, A., Socorro, L., Driscoll, J. M., Okoh, C., Lomelino, C. L., Mboge, M. Y., Kurian, J. J., Tu, C., Agbandje-McKenna, M., Frost, S. C. & McKenna, R. (2016). Biochemistry, 55, 4642-4653.]) containing the catalytic domain of CA IX with six surface mutations introduced (C174S, L180S, M360S, A210K, A258K and F259Y) that was engineered to facilitate crystallization (CA IXSV).

2.1. Expression and purification of protiated CA IXSV

CA IXSV production has been described in detail elsewhere (Koruza, Lafumat, Végvári et al., 2018[Koruza, K., Lafumat, B., Végvári, Á., Knecht, W. & Fisher, S. Z. (2018). Arch. Biochem. Biophys. 645, 26-33.]; Mahon et al., 2016[Mahon, B. P., Bhatt, A., Socorro, L., Driscoll, J. M., Okoh, C., Lomelino, C. L., Mboge, M. Y., Kurian, J. J., Tu, C., Agbandje-McKenna, M., Frost, S. C. & McKenna, R. (2016). Biochemistry, 55, 4642-4653.]). Briefly, CA IXSV was expressed in Escherichia coli BL21 (DE3) cells under kanamycin selection (final concentration of 50 µg ml−1) in a shaking incubator at 37°C. The cells were grown to an OD600 of ∼1.0 and expression was induced by the addition of 1 mM isopropyl β-D-1-thiogalactopyranoside (IPTG) in the presence of 1 mM ZnSO4. After 4 h the cells were harvested by centrifugation (5000g for 20 min) and the cell pellets were frozen at −20°C. The cell pellets were lysed by thawing at room temperature in 0.2 M sodium sulfate, Tris–HCl pH 9 and then stirring in the cold room for ∼3 h in the presence of 20 mg lysozyme and 1 mg DNaseI. Clarified lysates were prepared by centrifugation at 50 000g for 60 min at 4°C. Affinity chromatography using p-aminomethylbenzenesulfonamide resin (Sigma–Aldrich) (wash buffer 1, 0.2 M sodium sulfate, Tris–HCl pH 9; wash buffer 2, 0.2 M sodium sulfate, Tris–HCl pH 7; elution buffer, 0.4 M sodium azide, 50 mM Tris–HCl pH 7.8) was followed by size-exclusion chromatography (50 mM Tris–HCl pH 7.8, 100 mM NaCl). CA IXSV elutes from the size-exclusion column in two peaks corresponding to dimeric and monomeric forms (Koruza, Lafumat, Végvári et al., 2018[Koruza, K., Lafumat, B., Végvári, Á., Knecht, W. & Fisher, S. Z. (2018). Arch. Biochem. Biophys. 645, 26-33.]).

Peak fractions corresponding to monomeric CA IXsv were pooled and concentrated using Amicon Ultra Centrifugal Filter Units (Merck) with a molecular-weight cutoff of 10 kDa and were analyzed by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE) to estimate their purity. The protein was concentrated to a final concentration of 17 mg ml−1 for crystallization.

2.2. Expression and purification of deuterated CA IXSV

Deuterated CA IXSV was expressed in E. coli BL21 (DE3) cells according to a protocol described elsewhere (Koruza, Lafumat, Végvári et al., 2018[Koruza, K., Lafumat, B., Végvári, Á., Knecht, W. & Fisher, S. Z. (2018). Arch. Biochem. Biophys. 645, 26-33.]). Briefly, cells were pre-grown in LB Broth (Miller) (Difco) at 37°C. The growth medium was then exchanged in the middle of the exponential phase for the same volume of deuterated ModC1 medium supplemented with 2% unlabelled glycerol (Koruza, Lafumat, Végvári et al., 2018[Koruza, K., Lafumat, B., Végvári, Á., Knecht, W. & Fisher, S. Z. (2018). Arch. Biochem. Biophys. 645, 26-33.]; Duff et al., 2015[Duff, A. P., Wilde, K. L., Rekas, A., Lake, V. & Holden, P. J. (2015). Methods Enzymol. 565, 3-25.]). Upon dilution in the deuterated medium, the cells were allowed to recover for 1 h at 37°C while shaking at 120 rev min−1. Following the adaptation period, the temperature was decreased to 25°C and shaking was increased to 200 rev min−1. Protein expression was induced by the addition of IPTG in the presence of 1 mM zinc sulfate. The cells were harvested after 18 h by centrifugation and stored at −20°C. Deuterated CA IXSV was purified as described for the protiated form in Section 2.1[link].

2.3. CA IXSV crystallization optimization and H/D exchange

Crystallization drops were prepared using both the hanging-drop and sitting-drop vapour-diffusion methods after a lengthy optimization procedure as described elsewhere (Koruza, Lafumat, Nyblom et al., 2018[Koruza, K., Lafumat, B., Nyblom, M., Knecht, W. & Fisher, S. Z. (2018). Crystals, 8, 434-445.]). Briefly, crystals were initially grown for the preparation of seed stocks using a 1:1 ratio of protein solution (17 mg ml−1 H/H CA IXSV) and 30%(w/v) PEG 4000, 0.1 M Tris–HCl pH 8.5, 0.2 M sodium acetate or 0.2 M ammonium formate. These crystals were then sacrificed for seed-stock preparation in the mother liquor as described in the instructions for the Seed Bead Kit (Hampton Research; https://www.hamptonresearch.com). Crystallization was repeated using a 3:2:1 ratio of protein:precipitant:seed stock in drop volumes of between 6 and 24 µl. With seeding, crystals appeared within a week. Both protiated and deuterated CA IXSV were crystallized using protiated buffers. Crystals of H/H CA IXSV were used without further manipulation. To prepare H/D CA IXSV and D/D CA IXSV crystals, the reservoir solution was removed and replaced with a deuterated version. The drops were then resealed and allowed to H/D-exchange for several weeks prior to X-ray data collection. Prior to cooling the crystals by plunging them into liquid nitrogen, they were cryoprotected by dipping them into reservoir solution supplemented with 20% glycerol. For the H/D and D/D crystals, deuterated glycerol was used for cryoprotectant preparation.

2.4. Crystallographic data collection and structure refinement

Two diffraction data sets for H/H CA IXSV were collected at 100 K on the FIP-BM30A beamline (Roth et al., 2002[Roth, M., Carpentier, P., Kaïkati, O., Joly, J., Charrault, P., Pirocchi, M., Kahn, R., Fanchon, E., Jacquamet, L., Borel, F., Bertoni, A., Israel-Gouy, P. & Ferrer, J.-L. (2002). Acta Cryst. D58, 805-814.]) at the European Synchrotron Radiation Facility (ESRF), Grenoble, France and on the BioMAX beamline at MAX IV Laboratory, Lund, Sweden. The H/D and D/D CA IXSV data were collected on the BioMAX beamline at MAX IV Laboratory, Lund, Sweden.

Data processing was performed using the autoPROC software package (Vonrhein et al., 2011[Vonrhein, C., Flensburg, C., Keller, P., Sharff, A., Smart, O., Paciorek, W., Womack, T. & Bricogne, G. (2011). Acta Cryst. D67, 293-302.]). The automated workflow script mainly uses XDS (Kabsch, 2010[Kabsch, W. (2010). Acta Cryst. D66, 133-144.]) as the data-processing and scaling software and POINTLESS for space-group determination. For two of the data sets (H/D and D/D) 3600 images were collected. The images were processed in batches to find a cutoff where radiation damage impacts the data quality. However, data processing and subsequent model refinement showed that it was best to use the full data sets.

The phases for all of the X-ray data were obtained by molecular replacement in Phaser (McCoy et al., 2007[McCoy, A. J., Grosse-Kunstleve, R. W., Adams, P. D., Winn, M. D., Storoni, L. C. & Read, R. J. (2007). J. Appl. Cryst. 40, 658-674.]) using PDB entry 5dvx (Mahon et al., 2016[Mahon, B. P., Bhatt, A., Socorro, L., Driscoll, J. M., Okoh, C., Lomelino, C. L., Mboge, M. Y., Kurian, J. J., Tu, C., Agbandje-McKenna, M., Frost, S. C. & McKenna, R. (2016). Biochemistry, 55, 4642-4653.]) as a search model. The models were initially rigid-body refined in Phaser, followed by restrained refinement in the PHENIX suite (Adams et al., 2011[Adams, P. D., Afonine, P. V., Bunkóczi, G., Chen, V. B., Echols, N., Headd, J. J., Hung, L.-W., Jain, S., Kapral, G. J., Grosse-Kunstleve, R. W., McCoy, A. J., Moriarty, N. W., Oeffner, R. D., Read, R. J., Richardson, D. C., Richardson, J. S., Terwilliger, T. C. & Zwart, P. H. (2011). Methods, 55, 94-106.]). For all data sets, a bulk-solvent correction and a free R-factor monitor (calculated with 5% of randomly chosen reflections) were applied throughout the refinement. 2FoFc and FoFc map interpretation and manual model building was performed using Coot (Emsley et al., 2010[Emsley, P., Lohkamp, B., Scott, W. G. & Cowtan, K. (2010). Acta Cryst. D66, 486-501.]). For the apparently larger P21 unit cell, both chains were refined without applying noncrystallographic symmetry (NCS).

Figures were generated using PyMOL (Schrödinger; https://www.pymol.org). The CA IXSV structures were deposited in the RCSB Protein Data Bank with the following accession codes: 6rqn, 6rqq, 6rqu and 6rqw. Data-collection and refinement statistics are summarized in Table 1[link]. Dimer-interface analysis, buried surface-area calculation and mapping of interactions were performed in PyMOL and Coot and using the PDBePISA server (Emsley et al., 2010[Emsley, P., Lohkamp, B., Scott, W. G. & Cowtan, K. (2010). Acta Cryst. D66, 486-501.]; Krissinel & Henrick, 2007[Krissinel, E. & Henrick, K. (2007). J. Mol. Biol. 372, 774-797.]).

Table 1
Data-collection and model-refinement statistics for CA IXSV

  Protiated (H/H) (small unit cell) Protiated (H/H) (big unit cell) H/D-exchanged (H/D) Deuterated (D/D)
PDB code 6rqn 6rqq 6rqu 6rqw
Source FIP-BM30, ESRF BioMAX, MAX IV Laboratory BioMAX, MAX IV Laboratory BioMAX, MAX IV Laboratory
Wavelength (Å) 0.979 0.979 0.979 0.979
Detector ADSC Q315r Dectris EIGER 16M Dectris EIGER 16M Dectris EIGER 16M
Rotation range per image (°) 0.5 0.5 0.1 0.1
Total No. of images 270 360 3600 3600
Space group P21 P21 P21 P21
Unit-cell parameters (Å, °) a = 44.3, b = 65.1, c = 46.7, β = 115.1 a = 48.9, b = 65.1, c = 76.3, β = 92.86 a = 44.5, b = 65.4, c = 46.7, β = 115.1 a = 44.4, b = 65.1, c = 46.6, β = 114.7
Unit-cell volume (Å3) 121830 241730 121830 121830
Resolution range (Å) 50.0–1.77 (1.88–1.77) 49.5–1.28 (1.29–1.28) 40.0–1.39 (1.42–1.39) 40.0–1.49 (1.51–1.49)
Total No. of reflections 63296 (9199) 422779 (20704) 327912 (15926) 265165 (13184)
No. of unique reflections 22187 (3463) 122208 (5995) 48352 (2406) 39461 (1948)
Multiplicity 2.8 (2.6) 3.5 (3.5) 6.8 (6.6) 6.7 (6.8)
Completeness (%) 95.1 (92.9) 98.4 (96.0) 99.8 (99.1) 99.9 (99.3)
I/σ(I)〉 10.5 (2.2) 13.1 (2.1) 19.9 (2.2) 14.3 (2.2)
Rmerge (%) 6.3 (48.6) 3.1 (48.5) 3.6 (78.1) 6.0 (78.1)
Rmeas (%) 7.7 (60.3) 4.5 (67.0) 4.3 (96.2) 5.9 (86.4)
CC1/2 (%) 99.6 (83.3) 99.9 (86.6) 99.9 (82.2) 99.8 (69.7)
R.m.s.d., bond lengths (Å) 0.007 0.007 0.006 0.015
R.m.s.d., bond angles (°) 0.888 0.982 0.957 1.344
Rcryst 0.169 0.179 0.173 0.173
Rfree§ 0.206 0.195 0.188 0.203
No. of solvent molecules 253 682 262 217
Mean B factors (Å2)
 Protein 26.0 21.7 29.0 29.3
 Solvent 36.12 34.6 37.5 38.3
 Acetate ligand 31.92   33.57  
 Formate ligand   26.7   37.3
Rmerge = [\textstyle \sum_{hkl}\sum_{i}|I_{i}(hkl)- \langle I(hkl)\rangle|/][\textstyle \sum_{hkl}\sum_{i}I_{i}(hkl)] × 100.
Rcryst = [\textstyle \sum_{hkl}\big ||F_{\rm obs}|-|F_{\rm calc}|\big |/][ \textstyle \sum_{hkl}|F_{\rm obs}|].
§Rfree is calculated in the same way as Rcryst but for data omitted from refinement (5% of reflections for all data sets)

3. Results and discussion

3.1. Crystallography

Crystals for X-ray data collection were obtained in both hanging-drop and sitting-drop vapour-diffusion setups. Microseeding into drop volumes varying between 3 and 10 µl produced crystals within 1–2 weeks. There were noticeable and reproducible differences in the number, size and quality of the crystals depending on the deuteration status of the protein (Fig. 1[link]; Koruza, Lafumat, Végvári et al., 2018[Koruza, K., Lafumat, B., Végvári, Á., Knecht, W. & Fisher, S. Z. (2018). Arch. Biochem. Biophys. 645, 26-33.]). For the crystals used in this study the volumes ranged from 0.01 to 0.03 mm3. The largest CA IXSV crystal that we obtained was 0.8 mm3 and efforts to scale up and increase the volume continue (see Fig. 5 in Koruza, Lafumat, Nyblom et al., 2018[Koruza, K., Lafumat, B., Nyblom, M., Knecht, W. & Fisher, S. Z. (2018). Crystals, 8, 434-445.]). We obtained crystals in space group P21 with two apparently different unit cells labelled `small' (unit-cell parameters a = 44.5, b = 65.4, c = 46.7 Å, β = 115.1°) and `big' (unit-cell parameters a = 48.9, b = 65.1, c = 76.3 Å, β = 92.86°). Data from H/H crystals were initially collected on the FIP-BM30 beamline at the ESRF and they were shown to belong to a different space group to the previously reported P212121 (Mahon et al., 2016[Mahon, B. P., Bhatt, A., Socorro, L., Driscoll, J. M., Okoh, C., Lomelino, C. L., Mboge, M. Y., Kurian, J. J., Tu, C., Agbandje-McKenna, M., Frost, S. C. & McKenna, R. (2016). Biochemistry, 55, 4642-4653.]). Subsequent data collection from H/H, H/D and D/D crystals on the BioMAX beamline at MAX IV Laboratory revealed that H/H also indexed as space group P21 but with a `big' unit cell (Table 1[link]). The other H/D and D/D crystals were both in the `small' P21 unit cell, the same as the first one we determined from ESRF data. A summary of data-set and refinement statistics is shown in Table 1[link]. The crystals all diffracted with good statistics and the structures were determined to 1.77–1.28 Å resolution.

[Figure 1]
Figure 1
Photographs of hanging-drop and sitting-drop vapour-diffusion setups for producing (a) protiated and (b) deuterated CA IXSV. Both of the drops shown here are 10 µl in volume.

3.2. Space-group and crystal-packing analysis

The small P21 monoclinic unit cell contained two CA IXSV chains (one per asymmetric unit) with a volume of 123 080 Å3. The other H/H crystal unit cell processed as a `big' P21 cell, but is in fact a doubled `small' unit cell with the two chains in the apparent ASU related by translational NCS of ½, 0, ½. So, while theoretically redundant to the `small' H/H structure in terms of packing and contacts, the crystallization conditions were slightly different and we do see a different ligand bound in the active site (discussed later in Section 3.3). For these reasons we show the data statistics in Table 1[link] and discuss the active site in Section 3.3 and Fig. 5. The CA IXSV packing arrangements are shown in Fig. 2[link], with symmetry-related molecules shown in grey.

[Figure 2]
Figure 2
Crystal packing diagrams of CA IXSV in P21 (this work) and P212121 (PDB entry 5dvx), and CA IX in P61 (PDB entry 3iai) unit cells. The monomer in the ASU from the small P21 is shown as a red ribbon in all diagrams for reference. The two chains in the ASU unit of P212121 dimer are in green and in blue for P61.

Previous biophysical studies of CA IX have established it to be dimeric both in vivo and in vitro, although the precise organization of the native dimer is unknown (Hilvo et al., 2008[Hilvo, M., Baranauskiene, L., Salzano, A. M., Scaloni, A., Matulis, D., Innocenti, A., Scozzafava, A., Monti, S. M., Di Fiore, A., De Simone, G., Lindfors, M., Jänis, J., Valjakka, J., Pastoreková, S., Pastorek, J., Kulomaa, M. S., Nordlund, H. R., Supuran, C. T. & Parkkila, S. (2008). J. Biol. Chem. 283, 27799-27809.]; Li et al., 2011[Li, Y., Wang, H., Tu, C., Shiverick, K. T., Silverman, D. N. & Frost, S. C. (2011). Biochim. Biophys. Acta, 1813, 159-167.]). A dimer of dimers was observed in the first published crystal structure of the CA IX catalytic domain; it was produced using a baculovirus expression-vector system and the structure was determined in space group P61 (PDB entry 3iai; unit-cell parameters a = b = 144.2, c = 208.9 Å; Alterio et al., 2009[Alterio, V., Hilvo, M., Di Fiore, A., Supuran, C. T., Pan, P., Parkkila, S., Scaloni, A., Pastorek, J., Pastorekova, S., Pedone, C., Scozzafava, A., Monti, S. M. & De Simone, G. (2009). Proc. Natl Acad. Sci. USA, 106, 16233-16238.]). In this structure protein dimerization was mediated by an intermolecular disulfide bond involving Cys174 (position 41 in CA II) and was proposed to be a physiologically relevant quaternary structure. Interestingly, in the same report a Cys174Ser CA IX variant crystallized with the same packing and dimeric interface but without the disulfide bond (Alterio et al., 2009[Alterio, V., Hilvo, M., Di Fiore, A., Supuran, C. T., Pan, P., Parkkila, S., Scaloni, A., Pastorek, J., Pastorekova, S., Pedone, C., Scozzafava, A., Monti, S. M. & De Simone, G. (2009). Proc. Natl Acad. Sci. USA, 106, 16233-16238.]). Recently, another study reported a CA IX structure, this time determined from protein expressed in yeast, that crystallized in space group H3 (PDB entry 6fe0; unit-cell parameters a = b = 152.9, c = 171.5 Å; Kazokaitė et al., 2018[Kazokaitė, J., Niemans, R., Dudutienė, V., Becker, H. M., Leitāns, J., Zubrienė, A., Baranauskienė, L., Gondi, G., Zeidler, R., Matulienė, J., Tārs, K., Yaromina, A., Lambin, P., Dubois, L. J. & Matulis, D. (2018). Oncotarget, 9, 26800-26816.]). In addition to these studies, a crystal structure of CA IXSV was also determined in space group P212121 (PDB entry 5dvx; unit-cell parameters a = 57.9, b = 102.7, c = 108.9 Å). This structure also had two NCS chains in the ASU, however, the arrangement did not correspond to the previous reports of the native CA IX. This was most likely owing to the Cys174 residue being mutated to a serine, but the unit cell and crystal packing were also different (Mahon et al., 2016[Mahon, B. P., Bhatt, A., Socorro, L., Driscoll, J. M., Okoh, C., Lomelino, C. L., Mboge, M. Y., Kurian, J. J., Tu, C., Agbandje-McKenna, M., Frost, S. C. & McKenna, R. (2016). Biochemistry, 55, 4642-4653.]). As such, this smaller ortho­rhombic space-group unit cell was more suitable for neutron studies than the previously published hexagonal cell, and we pursued the orthorhombic form for our studies. Consideration of the unit-cell parameters in neutron protein crystallography experiments (normally up to a maximum of 150 Å) is related to the limitations of the current flux of neutron sources as well as the layout of macromolecular beamlines to resolve larger unit-cell parameters (Meilleur et al., 2018[Meilleur, F., Coates, L., Cuneo, M. J., Kovalevsky, A. & Myles, D. A. A. (2018). Crystals, 8, 388.]; O'Dell et al., 2016[O'Dell, W. B., Bodenheimer, A. M. & Meilleur, F. (2016). Arch. Biochem. Biophys. 602, 48-60.]). To be able to obtain reasonable diffraction data (better than 2 Å resolution) from a crystal with a large unit cell, it is necessary to also optimize the overall crystal volume (Tanaka, 2019[Tanaka, I. (2019). JPS Conf. Proc. 25, 011015.]). Despite extensive efforts to reproduce these crystals of CA IXSV, we instead obtained two new and seemingly different monoclinic P21 crystals (Fig. 1[link], Table 1[link]). The `small' monoclinic P21 crystal form is isomorphous to the deuterium-labelled CA IXSV crystal form. In addition, there is only one CA IXSV per asymmetric unit, which is useful when using neutrons for protein crystallography (Blakeley et al., 2015[Blakeley, M. P., Hasnain, S. S. & Antonyuk, S. V. (2015). IUCrJ, 2, 464-474.]). For a complete list of crystal contacts, refer to Table 2[link].

Table 2
Crystallographic interchain interactions in CA IXSV: comparison between P21 (this work) and P212121 (PDB entry 5dvx).

Hydrogen-bond and salt-bridge distances are shown in parentheses and indicate donor–acceptor (heavy atom) distances. A chains are listed first. Interactions longer than 3.4 Å were excluded. Analysis was performed using the PBDePISA server and was verified visually in Coot (Krissinel & Henrick, 2007[Krissinel, E. & Henrick, K. (2007). J. Mol. Biol. 372, 774-797.]; Emsley et al., 2010[Emsley, P., Lohkamp, B., Scott, W. G. & Cowtan, K. (2010). Acta Cryst. D66, 486-501.]). Mutated residues in CA IXSV are shown in bold.

P21 `small' (this work): monomer in asymmetric unit P21 `big' (this work): dimer in asymmetric unit P212121 (PDB entry 5dvx): dimer in asymmetric unit
Glu280–Glu192 via water (2.6 and 2.8 Å) Glu297–Glu219 via water (2.8 and 2.7 Å) Arg167–Asp146 (–CO) (3.1 Å)
Gln307–Arg323 (2.9 Å) Glu297–Thr257 via water (2.8 and 2.7 Å) Gln169–Pro148 (–CO) via water (2.7 and 2.8 Å)
Ser319–Glu305 (2.7 Å) Glu301–Lys258 (–CO) (3.4 Å) Gly233 (N)–Glu305 (–CO) (2.8 Å)
Asp320–Gln307 (2.9 Å) His357– Pro175 via water (3.0 and 3.4 Å) Glu242–Trp141 (N) (3.2 Å)
Arg323–Glu298 (3.2 Å) Asp361–Pro216 (–CO) (3.3 Å) Gly243 (–CO)–Gly367 (N) (2.9 Å)
Arg323–Thr306 (3.2 Å) Asp368–Arg268 via water (2.5 and 3.1 Å) His244–Zn–His200 (3.3 Å)
Asn346–Glu192 (3.1 A) Asp368–Lys258 (2.7 Å) His244–Zn–Trp141 (N) (3.3 Å)
Asn346–Gln307 (2.8 Å) Asp368–Tyr259 via water (3.2 and 3.0 Å) His244–Glu302 (2.8 Å)
Gln347–Glu305 (2.8 Å)   Arg245 (N)–Glu302 (2.9 Å)
    Asp395 (–CO)–Val152 (N) (2.9 Å)
    Ser396–Ser153 via water (2.6 and 2.6 Å)
    Arg399–Asp263 (–CO) (2.8 Å)
    Arg399–Leu266 (–CO) (2.7 Å)

Fig. 3 shows an overlay of the two chains in the monoclinic unit cell with the two chains in the orthorhombic ASU, with chain A from the latter as the reference. The monomer-to-monomer electrostatic contacts for both are listed in Table 2[link]. There are more interactions mediating the interface in the orthorhombic chains, most of which are attributable to the C-terminal residues wrapping around the dimer pair. The observation of multiple monomer-to-monomer arrangements suggests that CA IX has a strong propensity to dimerize, independent of intermolecular disulfide bonds, and it is not possible to infer which may be the relevant physiological dimer. The differences in arrangement and crystal packing between native CA IX and CA IXSV are illustrated in Fig. 2[link].

The CA IXSV variant has six surface amino-acid substitutions compared with CA IX, which were chosen to optimize expression in E. coli and crystallization. The intention was to eliminate dimerization owing to disulfide-bond formation, to reduce surface hydrophobicity and to encourage possible crystal contacts based on CA II crystal contacts (Mahon et al., 2016[Mahon, B. P., Bhatt, A., Socorro, L., Driscoll, J. M., Okoh, C., Lomelino, C. L., Mboge, M. Y., Kurian, J. J., Tu, C., Agbandje-McKenna, M., Frost, S. C. & McKenna, R. (2016). Biochemistry, 55, 4642-4653.]). Hence, we wanted to investigate whether some of the amino-acid interactions involved in the crystal contacts in the P21 or the P212121 unit cell were affected by the six amino-acid substitutions. Figs. 4[link](a) and 4[link](b) shows the dimer in ribbon representation, with the substituted side chains depicted as sticks. From inspection of these structures, it was apparent that residues Lys258 and Tyr259 were involved in NCS dimerization in the big monoclinic P21 unit cell reported here [Table 2[link], Fig. 4[link](b)]. In the small monoclinic P21 and previously reported orthorhombic P212121 unit cells [Fig. 4[link](c)] none of the amino-acid substitutions were involved in crystal contacts or NCS dimerization. It would therefore appear that the six surface residues that were changed do not drive dimerization but do have an important impact on the solubility and stability of the catalytic domain of CA IX.

[Figure 4]
Figure 4
Comparison of crystallographic contacts in view of the locations of the six mutations present in CA IXSV. CA IXSV is shown in cartoon representation, substituted amino acids (compare with CA IX) are depicted as yellow sticks, zinc is shown as a magenta sphere. (a) Crystallographic chains in the small P21 unit cell (this work), (b) NCS chains in the ASU of the P212121 cell (PDB entry 5dvx; Mahon et al., 2016[Mahon, B. P., Bhatt, A., Socorro, L., Driscoll, J. M., Okoh, C., Lomelino, C. L., Mboge, M. Y., Kurian, J. J., Tu, C., Agbandje-McKenna, M., Frost, S. C. & McKenna, R. (2016). Biochemistry, 55, 4642-4653.]).

3.3. Active-site comparison of H/H (small), H/H (big), HD and DD CA IXSV structures

When deuterating proteins for neutron studies, it is important to determine whether deuteration causes appreciable conformational effects in the resulting protein side chains and active-site solvent positioning. For the deuterated protein to be useful in structural studies it has to be representative of the physiological protiated protein: there should be no conformational changes. As noted before, the overall active-site arrangement of solvent and amino-acid residues appears to be largely unaffected, with an r.m.s.d. variation for all atoms when superimposing all four structures onto each other of less than 1 Å. We conducted a careful OMIT map analysis of the structures and found that in all four structures there are one of two ions bound to the zinc ion that come from the crystallization conditions: either acetate or formate (Fig. 5[link]). Both formate and acetate are known inhibitory anions that bind to CAs and therefore their presence is not a surprise (Coleman, 1967[Coleman, J. E. (1967). J. Biol. Chem. 242, 5212-5219.]; Håkansson et al., 1992[Håkansson, K., Carlsson, M., Svensson, L. A. & Liljas, A. (1992). J. Mol. Biol. 227, 1192-1204.]; Alterio et al., 2009[Alterio, V., Hilvo, M., Di Fiore, A., Supuran, C. T., Pan, P., Parkkila, S., Scaloni, A., Pastorek, J., Pastorekova, S., Pedone, C., Scozzafava, A., Monti, S. M. & De Simone, G. (2009). Proc. Natl Acad. Sci. USA, 106, 16233-16238.]). Formate inhibits by displacing the so-called `deep water' in the active site and binds in the same location as the carbon dioxide substrate (Fig. 5[link]; Domsic & McKenna, 2010[Domsic, J. F. & McKenna, R. (2010). Biochim. Biophys. Acta, 1804, 326-331.]), whereas acetate replaces two waters, the deep water and the catalytic zinc-bound water, effectively presenting an inhibited active-site structure (Supplementary Fig. S1). We observed formate bound in the apparent `big' cell in H/H and the D/D structures, while acetate was present in the small H/H and the H/D-exchanged structures (Fig. 5[link]). The presence of either ion does not seem to affect the overall active-site side-chain conformations, with the exception of small water rearrangements, which are reflected in the relative weak density for the active-site solvent molecule W2 [Fig. 5[link](a)]. In Figs. 5[link](b)–5[link](d) it can be seen that W2 has poor density at the same contouring and also has a higher refined crystallo­graphic B factor com­pared with the other active-site solvent molecules. The weak density of W2 is most apparent in the D/D structure, where there is no density below the 1.5σ level in the OMIT electron-density map. There are multiple possible explanations for this, including the presence of formate that may perturbate the water network or subtle effects caused as a result of protein deuteration.

[Figure 5]
Figure 5
Active-site comparison of CA IXSV. (a) H/H in the small P21 unit cell, (b) H/H in the doubled P21 unit cell, (c) H/D exchanged in the small P21 unit cell and (d) D/D CA IXSV in the small P21 unit cell. Active-site residues are depicted as yellow sticks; water molecules and Zn atoms are shown as red and magenta spheres, respectively. 2FoFc electron-density maps are shown in blue mesh and are contoured at 1.50σ for residues, 1.25σ for solvent and 3.50σ for zinc.

Careful electron-density OMIT analysis of the proton-shuttling residue His200 shows it to be split between two conformations, termed the `in' and the `out' conformation in the CA II literature (Supplementary Fig. S2; Nair & Christianson, 1991[Nair, S. K. & Christianson, D. W. (1991). J. Am. Chem. Soc. 113, 9455-9458.]; Fisher et al., 2005[Fisher, Z., Hernandez Prada, J. A., Tu, C., Duda, D., Yoshioka, C., An, H., Govindasamy, L., Silverman, D. N. & McKenna, R. (2005). Biochemistry, 44, 1097-1105.]). In the structure with PDB code 5dvx (CA IXSV) the His200 side chain was fully in the `out' position and was π-stacked with Trp141 (Mahon et al., 2016[Mahon, B. P., Bhatt, A., Socorro, L., Driscoll, J. M., Okoh, C., Lomelino, C. L., Mboge, M. Y., Kurian, J. J., Tu, C., Agbandje-McKenna, M., Frost, S. C. & McKenna, R. (2016). Biochemistry, 55, 4642-4653.]). There is structural evidence that the preferred position of His64 in CA II is strongly affected by the pH, with the `out' conformation being dominant at low pH owing to charge repulsion between the charged His and the zinc (Fisher et al., 2005[Fisher, Z., Hernandez Prada, J. A., Tu, C., Duda, D., Yoshioka, C., An, H., Govindasamy, L., Silverman, D. N. & McKenna, R. (2005). Biochemistry, 44, 1097-1105.]). However, all the monoclinic structures reported here and the previously reported ortho­rhombic forms were all determined from crystals grown at pH 8.5, yet the alternate conformation occupancy for His200 is different (Mahon et al., 2016[Mahon, B. P., Bhatt, A., Socorro, L., Driscoll, J. M., Okoh, C., Lomelino, C. L., Mboge, M. Y., Kurian, J. J., Tu, C., Agbandje-McKenna, M., Frost, S. C. & McKenna, R. (2016). Biochemistry, 55, 4642-4653.]). However, our crystals do contain either formate or acetate and the presence of these ligands could be a disrupting factor by altering the electrostatic effect of the zinc charge on the orientation of His64.

Taking the above observations together, for the four structures of the different labelled variants of CA IXSV we can conclude that deuteration had little to no effect on the overall structure. This is in contrast to other parameters, such as thermal stability and crystallization behaviour, as reported previously (Koruza, Lafumat, Végvári et al., 2018[Koruza, K., Lafumat, B., Végvári, Á., Knecht, W. & Fisher, S. Z. (2018). Arch. Biochem. Biophys. 645, 26-33.]). Furthermore, comparing specific residues that compose the active site shows that the overall architecture is maintained between H/H, H/D and D/D CA IXSV.

4. Conclusions

Here, we report four crystal structures of different protium/deuterium-labelled versions of CA IXSV in preparation for future neutron crystallographic studies. Despite efforts to reproduce the previously published P212121 crystal form, we instead obtained a different P21 crystal form from that previously observed for native CA IX and CA IXSV. This was an unexpected but fortuitous result, as the unit cell is much smaller than all previously reported for CA IX or CA IXSV. Overall, the optimized crystallization condition and resulting crystal parameters are more tractable for neutron studies. In addition, the structural comparison of the protiated, partially deuterated and perdeuterated crystal structures reveal that there are insignificant changes owing to deuteration. Hence, the small unit cell P21 crystals of CA IXSV will be used in future neutron crystallographic structural studies for the design of CA IX-specific inhibitors.

Footnotes

Current address: Department of Molecular Biology, Princeton University, Princeton, NJ 08544, USA.

Acknowledgements

The authors would like to thank Randy Read for critical reading, helpful comments, and for observing the doubling of the small monoclinic unit cell. We would also like to thank Esko Oksanen and Patrick Shaw Stewart for useful discussions. The authors would also like to thank the ESRF (FIP-BM30A) and MAX IV laboratory (BioMAX) beamline scientists for expert assistance. We would also like to thank the Lund Protein Production Platform (LP3) staff for providing technical support for experiments and for X-ray data collection. The authors thank the Integrated Infrastructure Initiative No. 262348 European Soft Matter Infrastructure.

Funding information

We thank Lund University, the Royal Physiographic Society of Lund, Interreg/MAX4ESSFUN, The Crafoord Foundation (award No. 20160528) and BioCARE (a Strategic Research Area at Lund University) for financial support. This project was partly funded by SINE2020.

References

First citationAdams, P. D., Afonine, P. V., Bunkóczi, G., Chen, V. B., Echols, N., Headd, J. J., Hung, L.-W., Jain, S., Kapral, G. J., Grosse-Kunstleve, R. W., McCoy, A. J., Moriarty, N. W., Oeffner, R. D., Read, R. J., Richardson, D. C., Richardson, J. S., Terwilliger, T. C. & Zwart, P. H. (2011). Methods, 55, 94–106.  Web of Science CrossRef CAS PubMed Google Scholar
First citationAggarwal, M., Boone, C. D., Kondeti, B. & McKenna, R. (2013). J. Enzyme Inhib. Med. Chem. 28, 267–277.  Web of Science CrossRef PubMed Google Scholar
First citationAlterio, V., Hilvo, M., Di Fiore, A., Supuran, C. T., Pan, P., Parkkila, S., Scaloni, A., Pastorek, J., Pastorekova, S., Pedone, C., Scozzafava, A., Monti, S. M. & De Simone, G. (2009). Proc. Natl Acad. Sci. USA, 106, 16233–16238.  Web of Science CrossRef PubMed CAS Google Scholar
First citationBlakeley, M. P., Hasnain, S. S. & Antonyuk, S. V. (2015). IUCrJ, 2, 464–474.  Web of Science CrossRef CAS PubMed IUCr Journals Google Scholar
First citationColeman, J. E. (1967). J. Biol. Chem. 242, 5212–5219.  CAS PubMed Web of Science Google Scholar
First citationDi Costanzo, L., Moulin, M., Haertlein, M., Meilleur, F. & Christianson, D. W. (2007). Arch. Biochem. Biophys. 465, 82–89.  Web of Science CrossRef PubMed CAS Google Scholar
First citationDomsic, J. F. & McKenna, R. (2010). Biochim. Biophys. Acta, 1804, 326–331.  Web of Science CrossRef PubMed CAS Google Scholar
First citationDuff, A. P., Wilde, K. L., Rekas, A., Lake, V. & Holden, P. J. (2015). Methods Enzymol. 565, 3–25.  Web of Science CrossRef CAS PubMed Google Scholar
First citationEmsley, P., Lohkamp, B., Scott, W. G. & Cowtan, K. (2010). Acta Cryst. D66, 486–501.  Web of Science CrossRef CAS IUCr Journals Google Scholar
First citationFisher, 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. (2014). Acta Cryst. D70, 3266–3272.  Web of Science CrossRef IUCr Journals Google Scholar
First citationFisher, S. Z., Aggarwal, M., Kovalevsky, A. Y., Silverman, D. N. & McKenna, R. (2012). J. Am. Chem. Soc. 134, 14726–14729.  Web of Science CrossRef CAS PubMed Google Scholar
First citationFisher, S. Z., Maupin, C. M., Budayova-Spano, M., Govindasamy, L., Tu, C., Agbandje-McKenna, M., Silverman, D. N., Voth, G. A. & McKenna, R. (2007). Biochemistry, 46, 2930–2937.  Web of Science CrossRef PubMed CAS Google Scholar
First citationFisher, Z., Hernandez Prada, J. A., Tu, C., Duda, D., Yoshioka, C., An, H., Govindasamy, L., Silverman, D. N. & McKenna, R. (2005). Biochemistry, 44, 1097–1105.  Web of Science CrossRef PubMed CAS Google Scholar
First citationFisher, Z., Kovalevsky, A. Y., Mustyakimov, M., Silverman, D. N., McKenna, R. & Langan, P. (2011). Biochemistry, 50, 9421–9423.  Web of Science CrossRef CAS PubMed Google Scholar
First citationGolden, E., Attwood, P. V., Duff, A. P., Meilleur, F. & Vrielink, A. (2015). Anal. Biochem. 485, 102–108.  Web of Science CrossRef CAS PubMed Google Scholar
First citationHåkansson, K., Carlsson, M., Svensson, L. A. & Liljas, A. (1992). J. Mol. Biol. 227, 1192–1204.  PubMed Web of Science Google Scholar
First citationHilvo, M., Baranauskiene, L., Salzano, A. M., Scaloni, A., Matulis, D., Innocenti, A., Scozzafava, A., Monti, S. M., Di Fiore, A., De Simone, G., Lindfors, M., Jänis, J., Valjakka, J., Pastoreková, S., Pastorek, J., Kulomaa, M. S., Nordlund, H. R., Supuran, C. T. & Parkkila, S. (2008). J. Biol. Chem. 283, 27799–27809.  Web of Science CrossRef PubMed CAS Google Scholar
First citationKabsch, W. (2010). Acta Cryst. D66, 133–144.  Web of Science CrossRef CAS IUCr Journals Google Scholar
First citationKazokaitė, J., Niemans, R., Dudutienė, V., Becker, H. M., Leitāns, J., Zubrienė, A., Baranauskienė, L., Gondi, G., Zeidler, R., Matulienė, J., Tārs, K., Yaromina, A., Lambin, P., Dubois, L. J. & Matulis, D. (2018). Oncotarget, 9, 26800–26816.  PubMed Google Scholar
First citationKim, C. U., Song, H., Avvaru, B. S., Gruner, S. M., Park, S. & McKenna, R. (2016). Proc. Natl Acad. Sci. USA, 113, 5257–5262.  Web of Science CrossRef CAS PubMed Google Scholar
First citationKoruza, K., Lafumat, B., Nyblom, M., Knecht, W. & Fisher, S. Z. (2018). Crystals, 8, 434–445.  Web of Science CrossRef Google Scholar
First citationKoruza, K., Lafumat, B., Végvári, Á., Knecht, W. & Fisher, S. Z. (2018). Arch. Biochem. Biophys. 645, 26–33.  Web of Science CrossRef CAS PubMed Google Scholar
First citationKovalevsky, A., Aggarwal, M., Velazquez, H., Cuneo, M. J., Blakeley, M. P., Weiss, K. L., Smith, J. C., Fisher, S. Z. & McKenna, R. (2018). Structure, 26, 383–390.  Web of Science CrossRef CAS PubMed Google Scholar
First citationKrissinel, E. & Henrick, K. (2007). J. Mol. Biol. 372, 774–797.  Web of Science CrossRef PubMed CAS Google Scholar
First citationKuijk, S. J. A. van, Yaromina, A., Houben, R., Niemans, R., Lambin, P. & Dubois, L. J. (2016). Front. Oncol. 6, 69.  Web of Science PubMed Google Scholar
First citationLangan, P. & Chen, J. C.-H. (2013). Phys. Chem. Chem. Phys. 15, 13705–13712.  Web of Science CrossRef CAS PubMed Google Scholar
First citationLangella, E., Buonanno, M., Vullo, D., Dathan, N., Leone, M., Supuran, C. T., De Simone, G. & Monti, S. M. (2018). Cell. Mol. Life Sci. 75, 3283–3296.  Web of Science CrossRef CAS PubMed Google Scholar
First citationLi, Y., Wang, H., Tu, C., Shiverick, K. T., Silverman, D. N. & Frost, S. C. (2011). Biochim. Biophys. Acta, 1813, 159–167.  Web of Science CrossRef CAS PubMed Google Scholar
First citationLiu, X., Hanson, B. L., Langan, P. & Viola, R. E. (2007). Acta Cryst. D63, 1000–1008.  Web of Science CrossRef IUCr Journals Google Scholar
First citationMahon, B. P., Bhatt, A., Socorro, L., Driscoll, J. M., Okoh, C., Lomelino, C. L., Mboge, M. Y., Kurian, J. J., Tu, C., Agbandje-McKenna, M., Frost, S. C. & McKenna, R. (2016). Biochemistry, 55, 4642–4653.  Web of Science CrossRef CAS PubMed Google Scholar
First citationMahon, B. P., Pinard, M. A. & McKenna, R. (2015). Molecules, 20, 2323–2348.  Web of Science CrossRef PubMed Google Scholar
First citationMcCoy, A. J., Grosse-Kunstleve, R. W., Adams, P. D., Winn, M. D., Storoni, L. C. & Read, R. J. (2007). J. Appl. Cryst. 40, 658–674.  Web of Science CrossRef CAS IUCr Journals Google Scholar
First citationMeilleur, F., Coates, L., Cuneo, M. J., Kovalevsky, A. & Myles, D. A. A. (2018). Crystals, 8, 388.  Web of Science CrossRef Google Scholar
First citationNair, S. K. & Christianson, D. W. (1991). J. Am. Chem. Soc. 113, 9455–9458.  CrossRef CAS Web of Science Google Scholar
First citationO'Dell, W. B., Bodenheimer, A. M. & Meilleur, F. (2016). Arch. Biochem. Biophys. 602, 48–60.  Web of Science CAS PubMed Google Scholar
First citationPastorek, J. & Pastorekova, S. (2015). Semin. Cancer Biol. 31, 52–64.  Web of Science CrossRef CAS PubMed Google Scholar
First citationPinard, M. A., Mahon, B. P. & McKenna, R. (2015). Biomed Res. Int. 2015, 453543.  Web of Science CrossRef PubMed Google Scholar
First citationRoth, M., Carpentier, P., Kaïkati, O., Joly, J., Charrault, P., Pirocchi, M., Kahn, R., Fanchon, E., Jacquamet, L., Borel, F., Bertoni, A., Israel-Gouy, P. & Ferrer, J.-L. (2002). Acta Cryst. D58, 805–814.  Web of Science CrossRef CAS IUCr Journals Google Scholar
First citationSilverman, D. N. & McKenna, R. (2007). Acc. Chem. Res. 40, 669–675.  Web of Science CrossRef PubMed CAS Google Scholar
First citationTanaka, I. (2019). JPS Conf. Proc. 25, 011015.  Google Scholar
First citationVonrhein, C., Flensburg, C., Keller, P., Sharff, A., Smart, O., Paciorek, W., Womack, T. & Bricogne, G. (2011). Acta Cryst. D67, 293–302.  Web of Science CrossRef CAS IUCr Journals Google Scholar

This is an open-access article distributed under the terms of the Creative Commons Attribution (CC-BY) Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original authors and source are cited.

Journal logoSTRUCTURAL
BIOLOGY
ISSN: 2059-7983
Follow Acta Cryst. D
Sign up for e-alerts
Follow Acta Cryst. on Twitter
Follow us on facebook
Sign up for RSS feeds