feature articles
Native glycosylation and binding of the antidepressant paroxetine in a low-resolution
of human myeloperoxidaseaUGSF, CNRS, 50 Avenue de Halley, 59658 Villeneuve d'Ascq, France, bDepartment of Pharmacognosy, Bioanalysis and Drug Discovery, Faculty of Pharmacy, Université Libre De Bruxelles, Brussels, Belgium, and cStructure et Fonction des Membranes Biologiques, Université Libre de Bruxelles, Brussels, Belgium
*Correspondence e-mail: goedele.roos@univ-lille.fr, julie.bouckaert@univ-lille.fr
Human myeloperoxidase (MPO) utilizes hydrogen peroxide to oxidize organic compounds and as such plays an essential role in cell-component synthesis, in metabolic and elimination pathways, and in the front-line defence against pathogens. Moreover, MPO is increasingly being reported to play a role in inflammation. The enzymatic activity of MPO has also been shown to depend on its glycosylation. Mammalian MPO crystal structures deposited in the Protein Data Bank (PDB) present only a partial identification of their glycosylation. Here, a newly obtained
of MPO containing four disulfide-linked dimers and showing an elaborate collection of is reported. These are compared with the identified in proteomics studies and from 18 human MPO structures available in the PDB. The also contains bound paroxetine, a blocker of serotonin reuptake that has previously been identified as an irreversible inhibitor of MPO, in the presence of thiocyanate, a physiological substrate of MPO.Keywords: human myeloperoxidase; glycosylation; N-glycan refinement; thiocyanate; paroxetine.
PDB reference: myeloperoxidase, 7oih
1. Introduction
The glycosylation of proteins in the PDB has gained attention as the importance of glycosylation becomes better recognized by the scientific community and as a more intense structural characterization and presentation of ). Ways to graft as a post-translational modification onto AlphaFold prediction models of protein three-dimensional structures (Jumper et al., 2021) have recently been implemented (Bagdonas et al., 2021) through the glycan-validation program Privateer (Agirre et al., 2015; Joosten et al., 2022).
is being realized (Scherbinina & Toukach, 2020Human myeloperoxidase (MPO) was first isolated in 1941 from purulent pleuritis fluid from tuberculosis patients. When neutrophilic polymorphonuclear leukocytes (neutrophils) entrap microbial or other invasive particulates, they release MPO during degranulation. MPO is a heme Fe3+-containing peroxidase with a protoporphyrin IX as the basic structure of its The porphyrin ring of MPO is covalently attached to the enzyme via two ester bonds (Asp260 and Glu408) and one electron-withdrawing sulfonium linkage (Met409), and features a histidine as a proximal ligand (His502) (Fiedler et al., 2000).
In the search for a new application for the well known selective serotonin-reuptake inhibitor paroxetine, paroxetine was discovered to have an inhibitory activity on human MPO at nanomolar concentrations (18 nM; Soubhye, Aldib et al., 2016; Soubhye, Chikh Alard et al., 2017). This opened new opportunities for the treatment of major depressive disorder with inflammatory syndrome (Soubhye, Gelbcke et al., 2017). Here, we present for the first time the (PDB entry 7oih) of paroxetine bound to MPO in the presence of thiocyanate, which hovers above the heme group as previously reported in a containing bromide and thiocyanate substrates (Blair-Johnson et al., 2001). Together with newly performed docking calculations, the of the complex provides further insights into the binding and possibly also the inhibition of the peroxidase reaction by MPO.
Mammalian MPO crystal structures have been determined and deposited in the Protein Data Bank (PDB) over the years, and contain one MPO monomer or a disulfide-linked MPO homodimer per et al., 2000). Here, we present a new crystal form containing four homodimers of human MPO at 2.6 Å resolution. Using this we obtained an elaborate collection of asparagine-linked at the five N-glycosylation sites, Asn323, Asn355, Asn391, Asn498 and Asn729, known from proteomics studies (Van Antwerpen et al., 2010; Reiding et al., 2019; Tjondro et al., 2021). We compared these modifications with those present in the 18 crystal structures (28 monomers) of human MPO in the Protein Data Bank (PDB) and with existing proteomics data. Our MPO displays a greater diversity and includes larger N-glycans in the electron density than have previously been reported in the PDB, and thereby approaches the analytical results that can be obtained using The crystallographic resolution of in crystal structures is helped by ongoing efforts to improve the building and of glycan structures, as described here.
and present partial glycosylation. The first human MPO was obtained at 1.8 Å resolution, showing multiple halide-binding sites (Fiedler2. Materials and methods
2.1. Materials
MPO was produced in a pure form (Table 1), whereas paroxetine hydrochloride was obtained as a lyophilized powder from Sigma–Aldrich–Merck. Highly purified leukocyte MPO (CAS No. 9003-99-0) with a purity index (A430/A280) of at least 0.85 was obtained from Planta Natural Products (Kettle & Winterbourn, 1988).
|
2.2. Crystallization
MPO (10 mg ml−1) was mixed with paroxetine inhibitor (25 µM) in 50 mM Tris pH 7.4 (Table 2). Mixtures further reacted with 10 µM hydrogen peroxide also crystallized but as much smaller crystals that did not diffract sufficiently, despite the excess hydrogen peroxide being eliminated before crystallization. It has previously been shown that the activation of MPO with H2O2 is necessary in order for paroxetine to irreversibly inhibit the enzyme. Irreversible inhibition is potentially due to a covalent linkage, as verified by kinetic studies (Soubhye et al., 2014), between paroxetine and the active site of MPO. Crystals were only obtained with 0.2 M potassium thiocyanate, 0.1 M sodium cacodylate, 8%(w/v) PGA-LM as the precipitant (PGA Screen from Molecular Dimensions).
|
2.3. Data collection and structure resolution
Data were collected on the PROXIMA-1 beamline at Synchrotron SOLEIL, Saint-Aubin, France (Table 3) and were processed by the automated pipeline using XDSME (XDS Made Easier; Kabsch, 2010; Legrand, 2017). The crystals diffracted to 2.6 Å resolution (Table 3). PDB entry 4c1m (Forbes et al., 2013) was used as a model, after removal of the ligand NIH, a trifluoromethyl-substituted aromatic hydroxamate and water molecules, to solve the by using Phaser (McCoy, 2007).
‡CC1/2 (previously called CC_Imean; Evans & Murshudov, 2013) is the Pearson obtained by comparing two sets of intensities randomly chosen from the merged crystallographic data. The calculations are usually performed after the two sets of intensities have been divided into thin shells of increasing resolution, so that the dependence of CC1/2 on resolution can be determined (Karplus & Diederichs, 2012; Diederichs, 2016). |
2.4. Structure of glycosylated MPO
Crystallographic phenix.refine (Afonine et al., 2012) from the Phenix package (Liebschner et al., 2019) and the refined model was manually adjusted using the graphics program Coot (Emsley & Cowtan, 2004; Emsley & Crispin, 2018; van Beusekom et al., 2019) (Table 4). MolProbity was used for protein structure validation (Williams et al., 2018). Following crystallographic using Phenix, the carbohydrate structures of the glycosylations were validated using Privateer (Agirre et al., 2015; Joosten et al., 2022), which applies Cremer–Pople analysis to determine sugar ring conformations (Cremer & Pople, 1975). The Privateer analysis feeds suggestions for corrections of carbohydrate geometry by rebuilding using Coot. Final refinements using REFMAC5 (Kovalevskiy et al., 2018) output an mmCIF that is amenable for PDB deposition and links to other databases such as GlyConnect (Alocci et al., 2019). This cycle of using Privateer for glycan structural validation, Coot for model revision and REFMAC5 for can be repeated as many times as necessary. Moreover, in CCP4 version 8.0 new dictionaries for in the pyranose form have been implemented in the CCP4 Monomer Library, with coordinates reflecting the lowest-energy ring pucker, improved ring torsion restraints and updated geometry (Atanasova et al., 2022).
was performed using
|
2.5. Building, refining and validating glycosylation and preparation for deposition in the PDB
The building of Coot using Modules → Carbohydrate, which opens a menu called Glyco that allows the addition of N-linked to the protein and real-space automated in the electron density (Emsley & Crispin, 2018). At the time of addition, the LINK records are automatically added in the PDB file. If the LINK records between the different carbohydrate residues are missing, or if atoms that leave upon making the glycosidic bonds are not removed, the monosaccharide residues will be pushed apart during due to van der Waals repulsion. Some examples can be found in an excellent overview of how to build and refine glycosylation in protein crystal structures (van Beusekom et al., 2019). Our general experience was that when handling the 30 glycosylations in the PDB coordinate file of MPO, once the LINK distance surpasses its standard deviation during it will be considered as unlinked, or noncovalently bound, by phenix.refine and this will lead to a further separation of the linked glycan. Because the LINK is the only restraint linking the glycan to the protein and may disconnect when the standard deviation is superseded, we set the standard deviation to be large enough in the link.edit file, which is a parameter file containing all nondefault covalent links between residues:
has been facilitated inIn this example, the sigma is set to be larger than the actual distance to avoid eventual repulsion during .cif files of Therefore, we manually set all of the standard errors on angles, bonds, dihedral angles etc. in the implied monosaccharide.cif to half the default value of the CCP4 library of monomer.cif files. Again, this is particularly important at lower resolution, where the data-to-parameter ratio is often too poor to maintain the correct configuration of the monosaccharide, compared with at high resolution (<2.0 Å diffraction resolution).
This is especially important in low-resolution structures and with flexible glycan chains where the electron density by itself does not restrain the model sufficiently. Also, stronger restraints are needed within the geometryDeposition of the coordinate file in the PDB needs to take place as a macromolecular pdb_extract has been integrated into CCP4 (Winn et al., 2011) and the CCP4i interface (version 5.0 and above). Users can run pdb_extract in the CCP4 environment. In pdb_extract, one defines the polymers as given in their one-letter code sequence to output all coordinates inclusive of nonprotein atoms into an mmCIF coordinate file. This file can be read in PyMOL (Schrödinger), which will display the N-glycosidic link to the asparagine as presumed based on a distance that is within the dimensions of a The covalent N-linkage and the glycosidic links between the saccharide units can also be displayed when reading the mmCIF file in the graphics program ChimeraX (Pettersen et al., 2021). Otherwise, the glycan may appear to be disconnected from the protein and broken up into its singular monosaccharide entities.
(mmCIF). This is equally so for validation prior to PDB deposition. The conversion program for coordinatesWith the conversion of the coordinate file from .pdb to .cif format for PDB deposition, the will be split off from the protein into separate entities in the event that more than a single N-acetylglucosamine is linked to the Asn ND2 atom via a glycosidic N-linkage. For N-glycans equal to or extending beyond chitobiose disaccharide [2-acetamido-2-deoxy-β-D-glucopyranose-(1–4)-2-acetamido-2-deoxy-β-D-glucopyranose],
the glycan entities will obtain new chain identities (chain IDs). Within this newly generated chain, the monosaccharide residues will be renumbered starting from 1. These glycan chains receive a chain identity unrelated to the chain identity of the protein monomer that they are glycosylating. It is important that the intramolecular connectivity information remains conserved in the macromolecular model-containing mmCIF upon the assignment of new glycan-chain identities, not least when one has to respond to errors in regard to validation by the PDB. A convenient way to list and decipher all glycosylations in the crystallographic model is to read prosmart-refmac.cif, a model output file from with REFMAC5, into the molecular-graphics visualization program CCP4MG (McNicholas et al., 2011). Choosing Glycan viewer from the menu will list the schematic structure of each glycan per protein chain, together with the residue that it glycosylates labelled with its name, number and chain ID (McNicholas & Agirre, 2017).
2.6. Molecular docking of paroxetine for comparison with the crystal structures
The 7oih) was used as the target structure to perform the The of one MPO monomer was prepared using the Protein Preparation Wizard in the Schrödinger software package (Protein Preparation Wizard, Impact and Prime from release 2017-1; Epik from release 2020). Crystallographic water molecules within 5 Å of heteroatoms in the structure were retained and treated as part of the receptor environment.
of the human MPO structure (PDB entryThe initial 3D structure of paroxetine was generated using the LigPrep module (Schrödinger release 2017-1). The Epik program was used to predict its different protonation states. Docking of paroxetine was carried out in the rigid prepared experimental X-ray structure of MPO using Glide (Schrödinger release 2021-1) and was performed in a delimited area (24 × 24 × 24 Å) based on the location of the bound paroxetine in the At most ten docking poses of the ligand were generated and scored using the Glide XP function. The docking poses with the highest scores were used for comparison with the crystal structure.
3. Results and discussion
3.1. Paroxetine binding in the of MPO
We could capture the binding of the serotonin-transporter inhibitor paroxetine, which is known to be one of the few irreversible MPO inhibitors (Soubhye, Chikh Alard et al., 2017), in the of the native enzyme (Soubhye, Meyer et al., 2016). Clear electron density is present for paroxetine bound in four of the eight monomers of MPO (Fig. 1).
The entries to the four other catalytic sites are obstructed by an arginine residue (Arg653) that inserts between the side chain of Asp384, with which it makes a salt bridge, and the carbonyl group of Val576. Every MPO homodimer has only one paroxetine bound and this is with a partial occupancy of paroxetine ranging from 1.0 (full occupancy) in chain A to 0.73 (partial occupancy) in chain H. The reason for the nonsaturation of MPO by paroxetine is probably because of its lower concentration (25 µM) compared with MPO (138 µM) in the crystallization condition. This partial occupancy may have led to the different crystal packing of MPO, with eight monomers in the which is unique in the PDB, serendipitously rendering a crystal packing that appears to be favourable for the visualization and determination of the glycosylation of MPO (Fig. 2).
Although only one of the monomers of each MPO dimer has paroxetine bound, all eight monomers in the crystal contain thiocyanate (Fig. 1), which was present at 200 mM in the crystallization condition. Thiocyanate, in its negatively charged form SCN−, can bind MPO and oxidize to HOSCN, which can induce reversible modifications to mammalian cells that are repairable and thus less damaging (Guo et al., 2020). In our thiocyanate is found in the same position as reported previously (Blair-Johnson et al., 2001), hovering above the heme porphyrin ring (Fig. 3). Its simultaneous presence with paroxetine may have an effect on how paroxetine is positioned in the active site.
In the ). The N atom of the piperidine group is oriented towards the heme group and forms two interactions with the active site: an ionic interaction with a propionate group of the heme and a hydrogen-bond interaction with Glu268 (Figs. 4a and 4b).
the benzodioxole group of paroxetine is oriented towards the outside of the active-site cavity and the fluorophenyl group is oriented away from the active-site heme (Fig. 3Globally, the same position and orientation of paroxetine as in the − is included in the docking calculations (Fig. 4c). Without the thiocyanate molecule, paroxetine docks with its benzodioxole group stacked above the heme in place of SCN− (Fig. 4d). While the interactions of the piperidine group remain conserved, the fluorophenyl and benzodioxole groups flip by about 180°. Therefore, thiocyanate might have caused displacement of the paroxetine, preventing it from entering and anchoring into the active site.
are found when SCNIndeed, in the presented ). Our structure is however physiologically relevant as thiocyanate is ubiquitous in human plasma and can be elevated by drugs, diet and smoking (van Dalen et al., 1997; Guo et al., 2020). Thiocyanate has a much higher (∼730-fold) specificity constant for MPO than chloride, which is considered to be the physiological substrate of MPO (van Dalen et al., 1997). As such, thiocyanate is likely to be a major substrate of myeloperoxidase in most environments in which this enzyme acts (van Dalen et al., 1997), making our presented structure relevant to further consideration of paroxetine as an inhibitor of peroxidase activity.
paroxetine binds at a position that is different from the position that it needs to adopt for its irreversible interaction with MPO (Fig. 43.2. Glycosylation in the of MPO
Native human MPO has five glycosylation sites identified at positions Asn323, Asn355, Asn391, Asn483 and Asn729 (Van Antwerpen et al., 2010; Fig. 5). The structure, activity and regulation of MPO by the natural inhibitor protein ceruloplasmin have been shown to depend on the local N-glycosylation pattern (Tjondro et al., 2021). Microheterogeneity in the site-specific N-glycan structures was found to be affected by the localization and maturation status of the enzyme (Reiding et al., 2019; Ugonotti et al., 2022; Venkatakrishnan et al., 2020), with many uniquely identified in mature neutrophils (Tjondro et al., 2021). Glycosylation of mature MPO was shown to be required for optimal enzymatic activity, possibly through allosteric effects due to interconnectivity, such as between His261 in the distal heme pocket neighbouring the calcium ligand residue Asp262 that is further connected to the Asn355 glycosylation site by an α-helix (Fiedler et al., 2000).
We determined the (potentially 40) glycan structures on five glycosylation sites per monomer for eight monomers per Privateer (Agirre et al., 2015) was used to validate the glycan structures against the pyranose monomer library from CCP4 version 8.0 (Atanasova et al., 2022). Privateer cross-checks the modelled against glycomics databases; in particular, whether the glycan is expected to match an entry in GlyConnect (Alocci et al., 2019). Ambiguities in the annotation of parts of the glycan structures (motifs) or the whole glycan can be overcome using the GlySTreeM knowledgebase (Daponte et al., 2021). For example, an α1,4-linked mannose (MAN monomer in the PDB) initially wrongly added as the central mannose in the common trimannose core (M3) was not recognized by GlyConnect until it was corrected to a β1,4-linked central mannose (BMA monomer in the PDB). A very useful feature is that Privateer produces interactive 2D graphical plots of the detected glycan trees and the amino acids that they modify (Bagdonas et al., 2020). Placing the mouse pointer over any of the will display the residue with its name, number and B factor from the PDB file.
in the of native human MPO.The final model contains eight polypeptide chains of mature MPO and 30 glycan chains on an asparagine side-chain ND2 atom, or in brief N-glycosylation. No protein O-glycosylation was found. The asparagine residues were either predominantly nonglycosylated, or glycosylated with hyper-truncated, paucimannosidic and hybrid N-glycans (Fig. 5), and are also present to some extent in human MPO crystal structures in the PDB (Lütteke et al., 2004; Fig. 6).
We compared the glycosylation characterized in our N-glycosylation modelled in the crystal structures of human MPO made available in the PDB (Fig. 6). Illustrations of protein glycosylation were generated using DrawGlycan-SNFG (Cheng et al., 2017), which applies the Symbol Nomenclature for (SNFG; Neelamegham et al., 2019). We can conclude that despite glycosylation in MPO crystal structures receiving little attention, crystallography has the capacity to resolve glycan structures in their diversity and heterogeneity (Fig. 6a versus Fig. 6b). This becomes especially true in a where a large permits multiple copies of the same glycoprotein to be present, each of them not subject to the same crystal-packing constraints.
with theThe diversity of glycan structures observed in current MPO crystal structures opens the possibility of comparison with the heterogeneity of these structures present in acquired proteomics data for MPO (Van Antwerpen et al., 2010; Reiding et al., 2019; Tjondro et al., 2021; Table 5). Reiding and coworkers made a qualitative and quantitative distribution of the after a triplicate LC-MS2 run (Reiding et al., 2019). Isomerisms were not investigated in this study. Tjondro and coworkers obtained the glycoprofiles of MPO secreted by neutrophils by using a reversed-phase LC-ESI-HCD-MS2 analysis (Tjondro et al., 2021). Both studies identified almost the same predominant (PGs), which are M2F on Asn323, M6 on Asn355, M6 on Asn391 and M3F on Asn483. An exception concerns the glycosylation on Asn729. Similar to the large shift from high-mannose and paucimannose glycosylation of Asn355 in the two (biological and validation) MPO batches from Reiding et al. (2019), changes are possible due to the N-glycan remodelling that occurs post-biogenesis. N-Glycans are more or less susceptible to glycan-processing enzymes depending on the position of the glycosylation site in the structure of the protein that they glycosylate (Mathew et al., 2021). The glycosylation site may also perform a truly individual protein function, for example only the Asn323 site has the peculiar phosphomannosylation that led Reiding and coworkers to suggest that neutrophils may have repurposed the M6P-mediated trafficking from the lysosomal pathway to populate proteins in their azurophilic granules (Reiding et al., 2019).
|
FA1[6] is a hybrid glycan on Asn483 that has not yet been observed in MPO crystal structures but is well represented in the two independent proteomics studies (Table 5) under the names N3H3F1 (Reiding et al., 2019) and FA1 (Tjondro et al., 2021). However, the `plus' of our study is that we can distinguish the isomer because we obtain the three-dimensional structure of the glycan (Fig. 5). Therefore, we can designate this glycan structure as FA1[6] owing to the presence of N-acetylglucosamine on the α1,6-arm of a trimannose N-glycan core (Fig. 6).
This may seem unusual because during glycan biogenesis the α1,3-arm of the common trimannose core of N-glycans is the first one to receive an N-acetylglucosamine residue by means of GlcNAc-transferase-1 (Helenius & Aebi, 2001). It is possible that the observed N-acetylglucosamine residue on the α1,6-arm of the trimannose is part of an FA2 structure with both arms, i.e. also the α1,3-arm, carrying N-acetylglucosamine, and that only the GlcNAc on the α1,6-arm is visible in the electron density. This could indeed be the case, as the GlcNAc residue stacks with the noncrystallographic symmetry-related GlcNAc residue of the other monomer of the MPO dimer (centre of Fig. 7).
The assignment of FA1[6] does not align with isomer assignments for this glycan from previous glycoproteomic research based on the existing knowledge of N-glycan biogenesis and in which the of only one of the isomers was known on PGC LC (Tjondro et al., 2021). We must stress that FA1[6] may also have been formed from FA2 or higher N-glycans by hexosaminidases HexA/B (Ugonotti et al., 2022). Recently, it has been confirmed that in human neutrophils (F)A2 N-glycans formed by the GlcNAc-transferase-1 pathway are retrogradely processed into paucimannosidic Our finding of FA1[6], carrying an N-acetylglucosamine residue, on the α1,6-arm of the trimannose N-glycan core may indicate that a truncation of an FA2 N-glycan has occurred at the Asn483 glycosylation site. Trace amounts of the precursor of FA1 (Ugonotti et al., 2022), namely FA2, were indeed found in the earliest proteomics study of MPO (Van Antwerpen et al., 2010), as well as in a later report (Tjondro et al., 2021).
These results show us that experimental data by means of crystallography on
can not only reveal the partial structures of the that are most well defined by electron density and potentially important for the stability of the protein. It can also inform on the three-dimensional structure of glycosylation and be complementary to results retrieved from glycoproteomic studies that are, in general, limited to the analysis of two-dimensional structures of glycans.In conclusion, myeloperoxidase prepared from human blood was crystallized in complex with the serotonin-transporter inhibitor paroxetine in crystals containing eight monomers in the N-glycosylation sites is either nonglycosylated or glycosylated with hypertruncated paucimannosidic, short high-mannose and hybrid N-glycans, keeping the redox funnel towards the heme group active. The MPO used here was isolated from human neutrophils from healthy donors (Bakkenist et al., 1978) and its glycosylation had previously been analyzed using and compared with that of recombinant human MPO (Van Antwerpen et al., 2010). Its are paucimannose and a dominance of high-mannose Both studies performing quantitative glycoproteomics (Reiding et al., 2019; Tjondro et al., 2021) found exactly the same presence of paucimannose and high mannose as major glycan structures, as they also used human MPO isolated from neutrophil granulocytes from healthy donors. We demonstrate the great potential of crystallographic data to resolve three-dimensional structures, including those of and explain strategies but also difficulties in the building and of glycosylation for its improved representation in the PDB.
This renders structural data on a unique set of which until now have not been represented in MPO structures present in the PDB. Each of the fiveAcknowledgements
We are grateful to Christian Obinger and Paul Fürtmuller from BOKU University of Natural Resources and Life Sciences, Vienna, Austria for providing MPO purified from human serum. We are grateful to the beamline manager and staff of the PROXIMA-1 (PX1) beamline for biological crystallography at the SOLEIL Synchrotron, Saint-Aubin, France.
Funding information
The following funding is acknowledged: European Union Horizon 2020 research and innovation programme, MSCA, COFUND with the University of Lille (scholarship No. 847568).
References
Afonine, P. V., Grosse-Kunstleve, R. W., Echols, N., Headd, J. J., Moriarty, N. W., Mustyakimov, M., Terwilliger, T. C., Urzhumtsev, A., Zwart, P. H. & Adams, P. D. (2012). Acta Cryst. D68, 352–367. Web of Science CrossRef CAS IUCr Journals Google Scholar
Agirre, J., Iglesias-Fernández, J., Rovira, C., Davies, G. J., Wilson, K. S. & Cowtan, K. D. (2015). Nat. Struct. Mol. Biol. 22, 833–834. Web of Science CrossRef CAS PubMed Google Scholar
Alocci, D., Mariethoz, J., Gastaldello, A., Gasteiger, E., Karlsson, N. G., Kolarich, D., Packer, N. H. & Lisacek, F. (2019). J. Proteome Res. 18, 664–677. Web of Science CrossRef CAS PubMed Google Scholar
Atanasova, M., Nicholls, R. A., Joosten, R. P. & Agirre, J. (2022). Acta Cryst. D78, 455–465. Web of Science CrossRef IUCr Journals Google Scholar
Bagdonas, H., Fogarty, C. A., Fadda, E. & Agirre, J. (2021). Nat. Struct. Mol. Biol. 28, 869–870. Web of Science CrossRef CAS PubMed Google Scholar
Bagdonas, H., Ungar, D. & Agirre, J. (2020). Beilstein J. Org. Chem. 16, 2523–2533. Web of Science CrossRef CAS PubMed Google Scholar
Bakkenist, A. R., Wever, R., Vulsma, T., Plat, H. & Van Gelder, B. F. (1978). Biochim. Biophys. Acta, 524, 45–54. CrossRef CAS PubMed Web of Science Google Scholar
Beusekom, B. van, Wezel, N., Hekkelman, M. L., Perrakis, A., Emsley, P. & Joosten, R. P. (2019). Acta Cryst. D75, 416–425. Web of Science CrossRef IUCr Journals Google Scholar
Blair-Johnson, M., Fiedler, T. & Fenna, R. (2001). Biochemistry, 40, 13990–13997. Web of Science CrossRef PubMed CAS Google Scholar
Cheng, K., Zhou, Y. & Neelamegham, S. (2017). Glycobiology, 27, 200–205. Web of Science CrossRef CAS PubMed Google Scholar
Cremer, D. & Pople, J. A. (1975). J. Am. Chem. Soc. 97, 1354–1358. CrossRef CAS Web of Science Google Scholar
Dalen, J. C. van, Whitehouse, M. W., Winterbourn, C. C. & Kettle, A. J. (1997). Biochem. J. 327, 487–492. PubMed Web of Science Google Scholar
Daponte, V., Hayes, C., Mariethoz, J. & Lisacek, F. (2021). Molecules, 27, 65. Web of Science CrossRef PubMed Google Scholar
Diederichs, K. (2016). Methods Mol. Biol. 1320, 147–173. CrossRef PubMed Google Scholar
Emsley, P. & Cowtan, K. (2004). Acta Cryst. D60, 2126–2132. Web of Science CrossRef CAS IUCr Journals Google Scholar
Emsley, P. & Crispin, M. (2018). Acta Cryst. D74, 256–263. Web of Science CrossRef IUCr Journals Google Scholar
Evans, P. R. & Murshudov, G. N. (2013). Acta Cryst. D69, 1204–1214. Web of Science CrossRef CAS IUCr Journals Google Scholar
Fiedler, T. J., Davey, C. A. & Fenna, R. E. (2000). J. Biol. Chem. 275, 11964–11971. Web of Science CrossRef PubMed CAS Google Scholar
Forbes, L. V., Sjögren, T., Auchère, F., Jenkins, D. W., Thong, B., Laughton, D., Hemsley, P., Pairaudeau, G., Turner, R., Eriksson, H., Unitt, J. F. & Kettle, A. J. (2013). J. Biol. Chem. 288, 36636–36647. Web of Science CrossRef CAS PubMed Google Scholar
Grishkovskaya, I., Paumann-Page, M., Tscheliessnig, R., Stampler, J., Hofbauer, S., Soudi, M., Sevcnikar, B., Oostenbrink, C., Furtmüller, P. G., Djinović-Carugo, K., Nauseef, W. M. & Obinger, C. (2017). J. Biol. Chem. 292, 8244–8261. Web of Science CrossRef CAS PubMed Google Scholar
Guo, C., Davies, M. J. & Hawkins, C. L. (2020). Redox Biol. 36, 101666. Web of Science CrossRef PubMed Google Scholar
Helenius, A. & Aebi, M. (2001). Science, 291, 2364–2369. Web of Science CrossRef PubMed CAS Google Scholar
Joosten, R. P., Nicholls, R. A. & Agirre, J. (2022). Curr. Med. Chem. 29, 1193–1207. Web of Science CrossRef PubMed Google Scholar
Jumper, J., Evans, R., Pritzel, A., Green, T., Figurnov, M., Ronneberger, O., Tunyasuvunakool, K., Bates, R., Žídek, A., Potapenko, A., Bridgland, A., Meyer, C., Kohl, S. A. A., Ballard, A. J., Cowie, A., Romera-Paredes, B., Nikolov, S., Jain, R., Adler, J., Back, T., Petersen, S., Reiman, D., Clancy, E., Zielinski, M., Steinegger, M., Pacholska, M., Berghammer, T., Bodenstein, S., Silver, D., Vinyals, O., Senior, A. W., Kavukcuoglu, K., Kohli, P. & Hassabis, D. (2021). Nature, 596, 583–589. Web of Science CrossRef CAS PubMed Google Scholar
Kabsch, W. (2010). Acta Cryst. D66, 125–132. Web of Science CrossRef CAS IUCr Journals Google Scholar
Karplus, P. A. & Diederichs, K. (2012). Science, 336, 1030–1033. Web of Science CrossRef CAS PubMed Google Scholar
Kettle, A. J. & Winterbourn, C. C. (1988). Biochem. J. 252, 529–536. CrossRef CAS PubMed Web of Science Google Scholar
Kovalevskiy, O., Nicholls, R. A., Long, F., Carlon, A. & Murshudov, G. N. (2018). Acta Cryst. D74, 215–227. Web of Science CrossRef IUCr Journals Google Scholar
Legrand, P. (2017). XDSME: XDS Made Easier. https://github.com/legrandp/xdsme. Google Scholar
Liebschner, D., Afonine, P. V., Baker, M. L., Bunkóczi, G., Chen, V. B., Croll, T. I., Hintze, B., Hung, L.-W., Jain, S., McCoy, A. J., Moriarty, N. W., Oeffner, R. D., Poon, B. K., Prisant, M. G., Read, R. J., Richardson, J. S., Richardson, D. C., Sammito, M. D., Sobolev, O. V., Stockwell, D. H., Terwilliger, T. C., Urzhumtsev, A. G., Videau, L. L., Williams, C. J. & Adams, P. D. (2019). Acta Cryst. D75, 861–877. Web of Science CrossRef IUCr Journals Google Scholar
Lütteke, T., Frank, M. & von der Lieth, C. W. (2004). Carbohydr. Res. 339, 1015–1020. Web of Science PubMed Google Scholar
Mathew, C., Weiss, R. G., Giese, C., Lin, C.-W., Losfeld, M.-E., Glockshuber, R., Riniker, S. & Aebi, M. (2021). RSC Chem. Biol. 2, 917–931. Web of Science CrossRef CAS PubMed Google Scholar
McCoy, A. J. (2007). Acta Cryst. D63, 32–41. Web of Science CrossRef CAS IUCr Journals Google Scholar
McNicholas, S. & Agirre, J. (2017). Acta Cryst. D73, 187–194. Web of Science CrossRef IUCr Journals Google Scholar
McNicholas, S., Potterton, E., Wilson, K. S. & Noble, M. E. M. (2011). Acta Cryst. D67, 386–394. Web of Science CrossRef CAS IUCr Journals Google Scholar
Neelamegham, S., Aoki-Kinoshita, K., Bolton, E., Frank, M., Lisacek, F., Lütteke, T., O'Boyle, N., Packer, N. H., Stanley, P., Toukach, P., Varki, A., Woods, R. J., Darvill, A., Dell, A., Henrissat, B., Bertozzi, C., Hart, G., Narimatsu, H., Freeze, H., Yamada, I., Paulson, J., Prestegard, J., Marth, J., Vliegenthart, J. F. G., Etzler, M., Aebi, M., Kanehisa, M., Taniguchi, N., Edwards, N., Rudd, P., Seeberger, P., Mazumder, R., Ranzinger, R., Cummings, R., Schnaar, R., Perez, S., Kornfeld, S., Kinoshita, T., York, W. & Knirel, Y. (2019). Glycobiology, 29, 620–624. Web of Science CrossRef CAS PubMed Google Scholar
Pettersen, E. F., Goddard, T. D., Huang, C. C., Meng, E. C., Couch, G. S., Croll, T. I., Morris, J. H. & Ferrin, T. E. (2021). Protein Sci. 30, 70–82. Web of Science CrossRef CAS PubMed Google Scholar
Reiding, K. R., Franc, V., Huitema, M. G., Brouwer, E., Heeringa, P. & Heck, A. J. R. (2019). J. Biol. Chem. 294, 20233–20245. Web of Science CrossRef CAS PubMed Google Scholar
Scherbinina, S. I. & Toukach, P. V. (2020). Int. J. Mol. Sci. 21, 7702. Web of Science CrossRef Google Scholar
Soubhye, J., Aldib, I., Delporte, C., Prévost, M., Dufrasne, F. & Van Antwerpen, P. (2016). Curr. Med. Chem. 23, 3975–4008. Web of Science CrossRef CAS PubMed Google Scholar
Soubhye, J., Aldib, I., Prévost, M., Elfving, B., Gelbcke, M., Podrecca, M., Conotte, R., Colet, J.-M., Furtmüller, P. G., Delporte, C., Rousseau, A., Vanhaeverbeek, M., Nève, J., Obinger, C., Zouaoui-Boudjeltia, K., Van Antwerpen, P. & Dufrasne, F. (2014). J. Pharm. Pharmacol. 66, 1122–1132. Web of Science CrossRef CAS PubMed Google Scholar
Soubhye, J., Chikh Alard, I., Aldib, I., Prévost, M., Gelbcke, M., De Carvalho, A., Furtmüller, P. G., Obinger, C., Flemmig, J., Tadrent, S., Meyer, F., Rousseau, A., Nève, J., Mathieu, V., Zouaoui Boudjeltia, K., Dufrasne, F. & Van Antwerpen, P. (2017). J. Med. Chem. 60, 6563–6586. Web of Science CrossRef CAS PubMed Google Scholar
Soubhye, J., Gelbcke, M., Van Antwerpen, P., Dufrasne, F., Boufadi, M. Y., Nève, J., Furtmüller, P. G., Obinger, C., Zouaoui Boudjeltia, K. & Meyer, F. (2017). ACS Med. Chem. Lett. 8, 206–210. Web of Science CrossRef CAS PubMed Google Scholar
Soubhye, J., Meyer, F., Furtmüller, P., Obinger, C., Dufrasne, F. & Van Antwerpen, P. (2016). Future Med. Chem. 8, 1163–1177. Web of Science CrossRef CAS PubMed Google Scholar
Tjondro, H. C., Ugonotti, J., Kawahara, R., Chatterjee, S., Loke, I., Chen, S., Soltermann, F., Hinneburg, H., Parker, B. L., Venkatakrishnan, V., Dieckmann, R., Grant, O. C., Bylund, J., Rodger, A., Woods, R. J., Karlsson-Bengtsson, A., Struwe, W. B. & Thaysen-Andersen, M. (2021). J. Biol. Chem. 296, 100144. Web of Science CrossRef PubMed Google Scholar
Ugonotti, J., Kawahara, R., Loke, I., Zhu, Y., Chatterjee, S., Tjondro, H. C., Sumer-Bayraktar, Z., Neelamegham, S. & Thaysen-Andersen, M. (2022). Glycobiology, 32, 218–229. Web of Science CrossRef PubMed Google Scholar
Van Antwerpen, P., Slomianny, M. C., Boudjeltia, K. Z., Delporte, C., Faid, V., Calay, D., Rousseau, A., Moguilevsky, N., Raes, M., Vanhamme, L., Furtmüller, P. G., Obinger, C., Vanhaeverbeek, M., Nève, J. & Michalski, J. C. (2010). J. Biol. Chem. 285, 16351–16359. Web of Science CrossRef CAS PubMed Google Scholar
Venkatakrishnan, V., Dieckmann, R., Loke, I., Tjondro, H. C., Chatterjee, S., Bylund, J., Thaysen-Andersen, M., Karlsson, N. G. & Karlsson-Bengtsson, A. (2020). J. Biol. Chem. 295, 12648–12660. Web of Science CrossRef CAS PubMed Google Scholar
Williams, C. J., Headd, J. J., Moriarty, N. W., Prisant, M. G., Videau, L. L., Deis, L. N., Verma, V., Keedy, D. A., Hintze, B. J., Chen, V. B., Jain, S., Lewis, S. M., Arendall, W. B., Snoeyink, J., Adams, P. D., Lovell, S. C., Richardson, J. S. & Richardson, J. S. (2018). Protein Sci. 27, 293–315. Web of Science CrossRef CAS PubMed Google Scholar
Winn, M. D., Ballard, C. C., Cowtan, K. D., Dodson, E. J., Emsley, P., Evans, P. R., Keegan, R. M., Krissinel, E. B., Leslie, A. G. W., McCoy, A., McNicholas, S. J., Murshudov, G. N., Pannu, N. S., Potterton, E. A., Powell, H. R., Read, R. J., Vagin, A. & Wilson, K. S. (2011). Acta Cryst. D67, 235–242. 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.