1.       Abbrevitiations

BTH       :               Bovine Testicular Hyaluronidase                                                                   

CHS       :               Chondroitin Sulfate                                                                                                           

GAG       :               Glycosaminoglycan

2.       Introduction

Enzyme derivatives used as pharmacological agents constitute a considerable part of enzyme therapy. Glycosidases are prospective regulators of the endothelial glycocalyx, the initial protective layer on luminal surface of blood vessels [1,2]. Hyaluronidases are deeply involved in the regulation of metabolism of hyaluronan, a major constituent of glycocalyx [3]. Bovine Testicular Hyaluronidase (BTH, EC 3.2.1.35) is allowed for intramuscular/subcutaneous injections and local application. Although this enzyme becomes increasingly important in glycobiology research [4], vertebrate hyaluronidases were not studied in sufficient detail due to very low concentration (~ 60 ng/ml in human plasma) and scarce information on their structure [5]. Hyaluronidases display an order of magnitude higher enzyme activity compared with that of other globular biocatalysts; however, their stability upon isolation and purification is very low [4,5].

Production of pharmacological agents based on modified hyaluronidases to achieve stable endoglycosidase activity for correction of glycocalyx function [6] (currently only replacement therapy is used for these purposes) requires deep understanding of molecular structure of the enzyme for optimization of its properties. Elucidation of human hyaluronidase structure [7] facilitated its use as a prototype for in silico 3D homologous modeling of BTH [8]. Our further research was based on BTH function in a Glycosaminoglycan (GAG) environment and aimed at understanding of the regulatory mechanisms operating during interaction between the enzyme and GAG of endothelial glycocalyx and extracellular matrix. It was important to define limiting structural factors associated with these mechanisms. GAGs are heterogeneous on their length, sequence, composition. According to these reasons the experimental studies of GAG 3D conformation and dynamics are difficult performed [9]. It is necessary to use the computer modeling. Computational analysis is the only direction to gain insight of GAG interactions now and then in some cases [10]. Under physiological conditions BTH molecule has a positive charge (pI 8.62) /http://www.uniprot.org/uniprot_/Q7YS45/, while GAG ligands bear a negative charge [11]. Our objective was to examine electrostatic interactions between BTH and GAG by computational analysis of molecular docking in 3D model of BTH and Chondroitin Sulfate (CHS) trimers in silico.

3.       Methods

3D models of native BTH, free enzyme and an enzyme-substrate complex (a hyaluronan dodecamer given here), were constructed previously [8] and were used for docking with CHS and heparin ligands.

Molecular docking was performed using UCSF Chimera, DOCK and Swiss-PdbViewer software [12-15].  DOCK was employed to perform docking of a low-molecular-weight molecule into a static macromolecule; UCSF Chimera and Swiss-PdbViewer were used to produce images with further superposition and visualization.  Calculation of isosurface for electrostatic potential was done using PDB2PQR, APBS, UCSF Chimera and Delphi software and AMBER, GROMOS and CHARMM [16-21] force field in compliance with Coulomb’s law and Poisson-Boltzmann equation. Upon docking, the low-molecular-weight molecule was placed at different positions in relation to the macromolecule at a distance of its half cross-size. The procedure was performed many with real time modeling. Reversibility/irreversibility of conformational changes in BTH molecule was evaluated by its return to the initial state within a given time period (50-100 ps). It should be noted that the absence of changes in the protein structure within this time period indicates that it is stable, i.e., remains unchanged with the accuracy of heat-induced conformational oscillations. Potentials were given in е/а_о natural unit system, where е is the electron charge and а_о is the Bohr radius; in SI unit system potential of 0.4 е/а_о corresponds to approximately 1 V. Molecular dynamic was monitored with   NAMD software [22]. 

The choice of GAG ligands /Pubchem.ncbi.nlm.nih.gov/ for molecular docking with 3D model of BTH was arbitrary since it is unknown which GAG fragment selectively inhibits the enzyme. Since glycosaminoglycans demonstrate high conformational flexibility, docking of only their short fragments (up to tetramers) can be reliably calculated. Therefore, were used a [GlcA - GalNAc]₃ trimers, where GlcA is glucuronic acid 2-sulfate bound to N-acetylgalactosamine 4-sulfate via β1-3 glycosidic bond, and an [IdoA – GlcNAc (2, 6) - IdoA –GlcNAc (2, 6) - IdoA – GlcNAc (2, 3, 6) - IdoA - GlcNAc (2, 6)]) heparin tetramer, where IdoA is iduronate 2-sulfate bound to glucosamine 2,6 sulfate / GlcNAc (2, 6) / of glucosamine 2,3,6 sulfate / GlcNAc (2, 3, 6) / via α1-4 glycosidic bond. The lengths of straightened CHS and heparin fragments correspond to cross-size of BTH molecule.

4.       Results and Discussion 

4.1    Binding Sites on Molecular Surface of Hyaluronidase 

Construction of 3D model of BTH (Figure 1)

facilitated computation of surface electrostatic potential of native enzyme (Figure 2). The potential is characterized by definite predominance of positively charged areas. It is noteworthy that electrostatic interactions have an effect on folding stability of human FKBP₁₂ protein [23], play an important role in sirtuin 2 inhibition [24] and contribute to protein-ligand interactions depending on dielectric constant of the medium [25]. Computational analysis of electrostatic potential allows prediction of GAG binding sites in individual systems and is useful for proposing protein surface regions as putative glycosaminoglycan binding sites [26]. The motives of these binding sites on a protein molecule surface are represented by parts of protein chain with xBBBxxBx or xBBxBx amino acid sequence where B is a basic amino acid (arginine, lysine or histidine) [11]. Effective binding in 3D structural models of proteins is achieved by juxtaposition of two binding sites. Computationally, we have identified 12 GAG binding sites on the surface of BTH molecule. Calculations were reduced to 8 sites with free binding energy at least 2-fold higher than energy of free ligands (Figure 3).

from Glu243 to Glu256 in highlighted in blue. 

Protective effects of CHS ligand and inhibiting/inactivating effects of heparin with/without electrolyte environment   were computed with inclusion of varied number of CHS ligands and heparin positioning into active center of the enzyme. Ligand binding energy which depends primarily on electrostatic interactions between GAG and arginine and lysine amino acid residues of BTH was determined as the difference between energy of the total complex before and after removal of the ligand followed by filling the released space with solvent. It should be noted that the enzyme-substrate complex is stable in any position with any number of associated CHS or heparin ligands. We further analyzed the stability of substrate-free enzyme, assuming that BTH is stable in the presence of eight CHS ligands in their binding sites. 

4.2    Electrostatic Forces for Docking

Electrostatic interactions are a necessary condition for docking and recognition of GAG ligands [27]. Electrostatic gradient of protein is a driving force aimed at ligand relocation along the gradient vector applied to the mass center of the ligand. Our computational data clearly demonstrate changes in surface electrostatic potential of free BTH (Figure 2), during interaction with   eight   CHS   trimers (Figure 4).

Electric   field   generated by the enzyme with 8 ligands envelopes the entire protein complex and nearly coincides with molecular relief of the enzyme. Free BTH is positively charged (Figure 2), while in CHS shield its charge is negative (Figure 4). In both cases the system remains stable, and heparin inclusion in the active center produces no significant effect on it. The situation changes after fixation of CHS ligands at cs1, cs2, cs3 and cs4 positions and heparin tetramer at cs6 position (Figure 5).

The electric field intensity created along its potential pushes heparin ligand into BTH active site.  Being not stabilized by disulfide bonds between   Asp 139 and Ala 156, the active site loop bearing Asp 147 and Glu 149 which are responsible for protein activity (highlighted in violet) is pushed out into the gap between alpha coil from Ala 185 to Gly 197 (green) and the coil from Glu 243 to Glu 256 (blue, Figure 3). This leads to irreversible molecular deformation and inactivation of BTH (Figure 6).

The effect is enhanced by inclusion of cs8 into the system Electric field created by ligands at cs1, cs2, cs3, cs4 and cs6 (Figure 7) does not resemble that of BTH stabilized by eight CHS ligands (Figure 4). Our data indicate that there is no specific coupling of heparin tetramer and CHS trimer to BTH active site.  Formation of major contacts results from electrostatic interactions realized via arginine residues on BTH surface, while lysine residues are important for covalent modification of the enzyme [8].

4.3    Covalent Modification for Hyaluronidase Stabilization

It should be noted that cs8, cs3, cs1 and cs2 positions are most preferred for electrostatic coupling of CHS ligands. By contrast, CHS inclusion at cs7 positions, which has low affinity but is the best for covalent modification due to direct contact with four lysine residues, prevents BTH deformation by leveling the field (Figure 8) and sterically closing the gasp mentioned above (Figure 3). Thus, covalent binding of CHS trimer to cs7 can be sufficient for prevention of heparin-induced BTH inactivation.

Electrostatic interactions which regulate dislocation of heparin or CHS ligands included in the active site (cs6) determine location of sites for their attachment. Binding energy for a ligand at cs6 position in the absence of substrate is almost the same at that in a solvent. According to the docking data, the ligand is rapidly coupled at cs3 and   cs8 positions. The third ligand molecule consecutively added to cs6 migrates to cs4 position without any irreversible conformational deformation of BTH. The enzyme stability is specified by electrostatic field created by GAG ligands, when their location on molecular surface dominates over their number. Inclusion of a fourth molecule into cs6 position (when previous ligands are distributed among cs3, cs8, cs4 and cs5 positions) induces dislocation of the ligand to the active site. This results in irreversible BTH inactivation due to deformation of the active site loop.  Modeled immobilization of CHS ligands at cs1 and cs2 positions did not prevent BTH inactivation. Only covalent immobilization of the ligands at cs7 or cs7, cs1, cs5 (docking data) completely protected BTH against heparin-induced inactivation. Thus, our findings demonstrate regulatory role of electrostatic interactions (enzyme inactivation or stabilization under varied conditions) upon molecular docking of BTH with GAG ligands. It is noteworthy that degradation of non-sulfated chondroitin by testicular hyaluronidase in comparison with hyaluronan is more efficient at higher pH [28].  3D structures of hyaluronan and chondroitin differ only by С-4 position in hexosamine residue. These delicate structural differences can be important in the structure-function regulation of hyaluronidases. 

5.       Conclusion

Computational analysis demonstrates structural stabilization of BTH upon interaction with six CHS ligands at cs7, cs1, cs2, cs3, cs4 and cs8 positions and heparin ligand at cs6 position. The difference of surface electrostatic potential of the protein (Figure 8) from that of stable BTH with eight CHS ligands (Figure 5) emphasizes delicate/fine formation of electric field of the enzyme complex which is associated with the significance of slight structural differences in ligands regulating the enzyme function. Computational docking reveals reversible and irreversible conformational changes in a 3D model of BTH after electrostatic coupling of GAG ligands, which leads to inactivation (Figure 7) and stabilization of BTH (Figures 5 and 8) under these conditions. 

The affinity of GAG binding to enzyme is determined by electrostatic and hydrophobic interactions [29] which are important for enzyme modifications [30]. Analysis of electrostatic field dynamics in relation to the type of ligand bound to BTH and evaluation of the contribution of hydrophobic interactions is reasonable and prospective for the development of experimental strategy for the research into the enzyme structure-function relationships.

6.       Acknowledgments 

The study was supported by Russian Foundation for Basic Research (grant 15-04-03584 and 18-015-00056) and Ministry of Health Care of Russian Federation. 

7.       Conflict of Interest 

The authors declare that they have no conflict of interest.

Figure 1: 3D thread-ribbon model of bovine testicular hyaluronidase. Six sites with lysine residues of the first-level availability for covalent modification [8] are highlighted in dark green. Noted lysine residues are located in the zones absolutely accessible for solvent and hydrophilic reagent with Stokes radius 12 Å. They are not involved in close intramolecular Coulombic or van der Waals interactions also. According to such collocation these lysine residues are most appropriate for chemical modification of enzyme (first-level availability).

Figure 2:  Surface electrostatic potential of free bovine testicular hyaluronidase (here and further at 0.4 е/а_о level). Negative and positive charges are indicated (here and further in Figures 4, 7 and 8) with red and green color, respectively.

Figure 3:  Location of eight CHS binding sites (cs1, cs2, cs3, etc.) in 3D model of BTH with free binding energy values at 0 К.  The image is given in projection formed by principle axes of inertia for hyaluronan, a dodecasaccharide fragment of the substrate (НА12, center of the image), and for BTH (horizontal axis X).    а Principle axis of inertia for hyaluronan passes vertically, Y.   b Principle axis of inertia for hyaluronan Y is perpendicular to the plain of the figure. The active site loop between Asp 139 and Ala 156, including Asp 147 and Glu 149 is highlighted in light violet, alpha-helix from Ala 185 to Gly 197 is highlighted in green, and helix from Glu243 to Glu256 in highlighted in blue.

Figure 4: Surface electrostatic potential of bovine testicular hyaluronidase during interaction with eight CHS trimeric ligands stabilizing enzyme structure.

Figure 5: Irreversible conformational changes in bovine testicular hyaluronidase after coupling with four CHS ligands at cs1, cs2, cs3 and cs4 positions and heparin tetramer at cs6 position. Superposition of protein structures surrounded by CHS ligands corresponds to the shape of the complex at 20th, 60th, 100th, 140th and 180th ps of the dynamics and the enzyme structure highlighted in beige, blue, green and light brown, respectively.

Figure 6: Irreversible conformational changes in bovine testicular hyaluronidase after peripheral dislocation of Glu149 and Asp147 which are responsible for enzyme activity. Arrows indicate initial and end positions of Glu149 during transformational transition. Significant changes in active site leading to the enzyme inactivation. Location of the ligands at cs4 and cs6 positions after inclusion in active site. During transition the ligands at cs1 and cs3 are virtually immovable relative to the protein and move at cs2, cs4 and cs6 positions. Highlighting is the same as in Figure 5.

Figure 7:  Surface electrostatic potential of bovine testicular hyaluronidase surrounded by five glycosaminoglycan ligands: CHS trimers at cs1, cs2, cs3 and cs4 positions and heparin tetramer at cs6 position, which results in enzyme inactivation.

Figure 8:  Surface electrostatic potential of bovine testicular hyaluronidase with six CHS trimers coupled at positions designated as cs7, cs1, cs2, cs3, cs4, cs8 and heparin tetramer located at cs6 position. Structural stabilization is achieved by CHS coupling at cs7 position.

1.       Reitsma S, Slaaf DW, Vink H, van Zandvoort MAMJ, onde Egbrink MGA (2007) The endothelial glycocalyx: composition, function, and visualization. Pflugers Arch 454: 345-359.

2.       Maksimenko AV (2015) Endothelial Glycocalyx as an orchestrator of vascular homeostasis. New research problems and prospects for vessel wall protection.  Russ Chem Bull 64: 1-7.

3.       Stern R, Jedrzejas MJ (2006) Hyaluronidases: their genomics, structure, and mechanisms of action. Chem Rev 106: 818-839.  

4.       Stern R, Jedrzejas MJ (2008) Carbohydrate polymers at the center of life’s origins: the importance of molecular processity. Chem Rev 108: 5061-5085.

5.       Erickson M, Stern R (2012) Chain gangs: new aspects of hyaluronan metabolism. Biomed Res Int 2012.

6.       Maksimenko AV, Schechilina YV, Tischenko EG (2003) Role of the glycosaminoglycan microenvironment of hyaluronidase in regulation of its endoglycosidase activity. Biochemistry (Moscow) 68: 862-868.

7.       Chao KL, Muthukumar L, Herzberg O (2007) Structure of human hyaluronidase-1, a hyaluronanhydrolyzing enzyme involved in tumor growth and angiogenesis. Biochemistry 46: 6911-6920.

8.       Maksimenko AV, Turashev AD, Beabealashvili RS (2015) Stratification of chondroitin sulfate binding sites in 3D-model of bovine testicular hyaluronidase and effective size of glycosaminoglycan coat of the modified protein.  Biochemistry (Moscow) 80: 284-295.

9.       Almond A (2018) Multiscale modeling of glycosaminoglycan structure and dynamics: current methods and challenges. Curr Opin Struct Biol 50: 58-64.

10.    Sankaranarayanan NV, Nagarajan B, Desai UR (2018) So you think computational approaches to understanding glycosamonoglycan-protein interactions are too dry and too rigid? Think again! Curr Opin Struct  Biol 50: 91-100.

11.    Maksimenko AV (2008) Effects of glycosaminoglycans in vascular events. Pharm Chemistry J 42: 553-563.

12.    Pettersen EF, Goddard TD, Huang CC, Couch GS, Greenblatt DM, et al. (2004) UCSF Chimera -a visualization system for exploratory research and analysis. J Comput Chem 25: 1605-1612.

13.    Sanner MF, Olson AJ, Spehner JC (1996) Reduced surface: an efficient way to compute molecular surfaces. Biopolymers 38: 305-320.

14.    Lang PT, Brozell SR, Mukherjee S, Pettersen EF, Meng EC, et al. (2009) DOCK 6: combining techniques to model RNA-small molecule complexes. RNA 15: 1219-1230.

15.    Guex N, Peitsch MC (1997) Swiss-Model and the Swiss-Pdb Viewer: An environment for comparative protein modeling. Electrophoresis 18: 2714-2723.

16.    Dolinsky TJ, Czodrowski P, Li H, Nielsen JE, Jensen JE, et al. (2007) PDB2PQR: Expanding    and    upgrading    automated preparation of biomolecular structures for molecular simulations. Nucleic Acids Res 35: W522-W525.

17.    Dolinsky TJ, Nielsen JE, McCammon JA, Baker NA (2004) PDB2PQR: an  automated    pipeline for the     setup, execution, and    analysis   of Poisson-Boltzmann electrostatics calculations. Nucleic Acids Res 32: W665-W667.

18.    Baker NA, Sept D, Joseph S, Holst MJ, McCammon JA (2001) Electrostatics of Nano systems: application to microtubules and the ribosome.  Proc Natl Acad Sci USA  98: 10037-10041.

19.    Honig B, Nicholls A (1995) Classical electrostatics in biology and chemistry. Science 268: 1144-1149.

20.    Nicholls A, Sharp K, Honig B (1991) Protein folding and association: insights from the interfacial and thermodynamic properties of hydrocarbons. Proteins 11: 281-296.

21.    Klapper I, Hagstrom R, Fine R, Honig B (1986) Focusing of electric fields in the active site of Cu-Zn superoxide dismutase:  effects of ionic strength and amino-acid modification. Proteins 1: 47-59.

22.    Phillips JC, Braun R, Wang W, Gumbart J, Tajkhorshid E, et al. (2005) Scalable molecular dynamics with NAMD. J Comput Chem 26: 1781-1802.

23.    Batra J, Tjong H, Zhou HX (2016) Electrostatic effects on the folding stability of FKBP₁₂. Prot Eng Des Sel 29: 301-308.

24.    Sakkiah S, Arooj M, Kumar MR, Eom SH, Lee KW (2013) Identification of inhibitor binding site in human sirtuin 2 using molecular docking and dynamics simulations. PLoS One 8: e51429.

25.    Wang B, Li L, Hurley TD, Meroneh SO (2013) Molecular recognition in a diverse set of protein-ligand interactions studied with molecular dynamics simulations and end-point free energy calculations. J Chem Inf. Model 53: 2659-2670.

26.    Samsonov SA, Pisabarro MT (2016) Computational analysis of interactions in structurally available protein-glycosaminoglycan complexes. Glycobiology 26: 850-861.

27.    Samsonov SA, Teyra J, Pisabarro MT (2011) Docking glycosaminoglycans to proteins: analysis of solvent inclusion. J Comput Aided Mol Des 25: 477-489.

28.    Honda T, Kaneiwa T, Mizumoto S, Sugahara K, Yamada S (2012) Hyaluronidases have strong hydrolytic activity toward chondroitin 4-sulfate comparable to that for hyaluronan.  Biomolecules 2:549-563.

29.    Gandhi NS, Freeman C, Parish CR, Mancera RL (2012) Computational analysis of the catalytic and heparin-binding sites and their interactions with glycosaminoglycans in glycoside hydrolase family 79 endo-beta-D-glucouronidase (heparanase).  Glycobiology 22: 35-55.

Maksimenko AV (2017) Modified enzymes for pharmaceutical purposes. Extension of the goals and objectives for consistent investigation. Russ Chem Bull 66: 1-8.