In 2021, the Honorato Project developed the kinetic and structural modeling of the γPLI protein, an inhibitor of phospholipase A2, responsible for the necrotizing effect of the venom inoculated by Bothrops jararaca. The project's proposal would be to produce this inhibitor, based on synthetic biology, so that it can later be inserted in ampoules and distributed in health clinics and ambulances, helping in the treatment of ophidic accidents. In 2022, the proposal will be the kinetic and structural development of the enzymatic inhibition of the BJ46a protein on jararhagin, a hemorrhagic protein. This proposal corresponds to a complement to the treatment initially proposed, transforming it into a more effective drug to neutralize the necrotizing and hemorrhagic effects caused by the inoculated venom. It should be noted that this type of treatment will not completely replace the use of serotherapy, and should be used as an adjunct in the treatment of ophidic accidents.
Therefore, the modeling developed aims at understanding the molecular structure, and therefore being able to understand its inhibition and storage mechanisms and consequently the full production of the inhibitor BJ46a in Pichia pastoris yeast for use in the treatment of ophidic accidents.
In the development of structural modeling, proteins will be evaluated structurally and dynamically, i.e., the closest crystal structure is sought, as well as predicting the best protein stability in vials, aiming at a more reliable application to social reality. The modeling performed is homology, and aims to build three-dimensional models of protein structure using experimentally determined structures of members of related families. Homology-modeling is currently the most accurate computational method for generating reliable structural models, so it is commonly used routinely in many biological applications.
Initially, the jararhagin protein and the BJ46a inhibitor were searched in the NCBI database for submission to SWISS-MODEL, which consists of a fully automated protein structure homology-modeling server accessible through the Expasy web server, or from the DeepView program (Swiss Pdb-Viewer). After this search, modeling of the BJ46a inhibitor was performed in Modeller, a homology-modeling development software that uses the python language in its execution, and the parameters obtained were recorded. Next, a mathematical analysis of the obtained parameters was performed in order to obtain the best possible structure, according to the analyzed parameters.
Subsequently, the modeled structures and the molecular docking tools were used to analyze the inhibitory interaction between the jararhagin protein and the inhibitor BJ46a. To perform the docking, first the servers to be used for plotting the interaction between the two proteins were selected and, consequently, the parameters obtained were registered for further analysis. The servers used were ClusPro, PatchDock, HDock and PyDock. It is emphasized that the results obtained were corresponding between the servers and converged with the data obtained in articles.
Finally, molecular dynamics was studied to predict the stability of the inhibitor, a protein, in aqueous media in order to obtain a better fit of the model to reality. In this part, the GROMACS software and the CHARMM-GUI server were used to perform the simulations, according to the characteristics of the BJ46a protein.
SVMPs of class P-III, such as jararhagin, have great toxicity, they usually have a molecular mass between 60 and 110 kDa and have domains in their structure that allow interaction with elements of the cell, being a catalytic domain, a disintegrin-like domain and a domain rich in cysteines (LIMA, 2013). According to Paine and colleagues (1992), jararhagin has four domains in its structure: a pro-peptide region (domain A), a metal ion binding region (domain B), a disintegrin region (domain C) and a carboxyl-terminal region with a cysteine-rich domain (domain D).
For the structural modeling of jararhagin, the SWISS-MODEL database was used since the visualization of three-dimensional data helps understanding not only the structures of jararhagin and BJ46a, but also the interaction of jararhagin and inhibitor. It is noteworthy that, as an optimization, the server removed the signal peptides from the protein, using SignalP-5.0 seeking to obtain the modeled protein closer to the real crystal.
Figure 1 - Graph demonstrating the absence of signal peptides in the protein
Source: https://services.healthtech.dtu.dk/service.php?SignalP-6.0
From SWISS-MODEL some parameters for evaluating the model of the jararhagin protein are obtained: one of them is the Qualitative Model Energy Analysis (QMEAN). QMEAN is an estimator known as z-score. When the z-value is close to 1 it means that the model is considered reliable and therefore there is good agreement between the model and experimental structures of similar size. When modeling jararhagin, we obtained a QMEAN of 0.8. AlphaFold produces a per-residue confidence score (PLDDT) between 0 and 100, and some regions below 50 PLDDT may be isolated without structures. The result obtained for the jararhagin protein, was an average PLDDT of 85.63.
Figure 2 - jararhagin’s protein database
Source: https://swissmodel.expasy.org
Figure 3 - jararhagin’s protein database
Source: https://swissmodel.expasy.org/repository/uniprot/P30431?model=AF-P30431-F1-model-v2
Figure 4 - Plot of the structurally modeled jararhagin protein
Source: https://swissmodel.expasy.org/repository/uniprot/P30431?model=AF-P30431-F1-model-v2
Figure 5 - Plot of the structurally modeled jararhagin protein
Source: https://swissmodel.expasy.org/repository/uniprot/P30431?model=AF-P30431-F1-model-v2
BJ46a, found in the plasma of B. jararaca itself, is an SVMPI capable of interacting non-covalently with jararhagin and inhibiting the action of metalloprotease, acting as an anti-hemorrhagic factor (PALACIO, 2017). The inhibitor is a glycoprotein (17% glycosylated) of 322 amino acids with a homodimeric nature (46 kDa each monomer) and presents two cystatin domains and a histidine-rich domain in its C-terminal portion (BASTOS, 2014; VALENTE et al., 2001). Experimentally determined, the pI of the inhibitor is 4.55, which suggests that the glycosylated portion of BJ46a is acidic in nature, probably because of the presence of N-acetylneuraminic acid (sialic acid); the presence of this acid at the reducing terminal of N-glycosylations may contribute to a prolonged protective effect (BASTOS, 2014, 2020; VALENTE et al., 2001). The glycidic portion of the inhibitor is composed primarily by N-glycosylations, anchored in the positions Asn76, Asn185, Asn263 and Asn274, and has a hole in the protein's folding and stability, affecting its biological activity (BASTOS, 2014; VALENTE et al., 2001). In its glycidic structure, besides sialic acid, Bastos (2014) identified galactose and N-acetylglucosamine molecules in the oligosaccharide antennae.
Although the inhibitor is able to interact with jararhagin in a stoichiometric ratio of 1:1, BJ46a, like DM43, and is not able to form a complex with jararhagin-C, suggesting the great importance of the interaction through the metalloprotease site (VALENTE et al., 2001).
In order to obtain the best structure for the BJ46a inhibitor, its protein was structurally modeled. Initially, the signal peptides were removed, by the same optimization proposal presented in the modeling of jararhagin.
Figure 6 - Graph demonstrating the sequence of signal peptides present in the protein
Source: https://services.healthtech.dtu.dk/service.php?SignalP-6.0
After the signal peptides were removed, the four best templates to be used for the structural modeling of the inhibitor BJ46a were searched in the SWISS-MODEL database. The template that will be used is 6ht9. 1. B Fetuin-B, which had the second highest GMQE (Global Model Quality Estimation, which indicates how much more accurate the model is with respect to target-model alignment and target coverage) corresponding to 0.48, and the highest QMEAN, corresponding to 0.63. Furthermore, according to the Ramachandran plot obtained, 91.06% of the residuals are located in favorable regions, indicating a good result.
With the four best templates found with the help of Swiss-Model, one of the first steps, already with the modeller, is to perform the alignment of the protein sequence with the templates obtained, for comparison and to obtain the best of them. The software itself offers resources on how to analyze the best template, being the one that obtained the lowest average clustering of base pair groups. It can be seen that the best template corresponds to the same one used in the SWISS-MODEL modeling, the template 6ht9. 1. B Fetuin-B, which indicates convergence and logic in the programmed data. Thus, from this template, we obtained five other templates.
Figure 7 - Weight pair-group average clustering based on a distance matrix
Source: https://salilab.org/modeller/
With the templates found with the help of Swiss-Model, one of the first steps with the modeller is to perform the alignment of the protein sequence with the obtained template. Among the 5 templates finally obtained, the parameters were analyzed and the best one was BJ46a.pdb, which corresponds to the t1a template.
Figure 8 - Plot of the structurally modeled BJ46a protein
Source: https://salilab.org/modeller/
After the modeling was performed, the UCLA-DOE LAB and Zhang Lab servers were used to obtain some parameters.
The Ramachandran graph is a two-dimensional (2D) plot of the torsion angles of the amino acids φ (phi) and ψ (psi) within a protein sequence. The φ represents the dihedral angle between N(i-1)-C(i)-CA(i)-N(i) and ψ is the dihedral angle between C(i)-CA(i)-N(i)-C(i+1). The results indicated that 78% of the amino acids are in favorable regions, thus presenting an excellent result.
Figure 9 - Ramachandran plot
https://saves.mbi.ucla.edu/
Figure 10 - Values obtained for the ramachandran plot. This graph shows that about 78% of the amino acids are in favorable regions, indicating an optimal parameter of the protein structure.
https://saves.mbi.ucla.edu/
Docking is a bioinformatics tool, which allows analyzing the interaction between two proteins or molecules through structural modeling, in order to predict the behavior of the complex, as well as obtaining data such as: binding energies, stability of the complex and protein parameters, in addition to representations of the visual structure of the complex (DAR & MIR, 2017; MENG et al., 2011).
From the literature, it is known that the complex between BJ46a and jararhagin can form by some means of non-covalent bonds, among them, involving the first and second cysteine-like domains of the inhibitor and the metalloendopeptidase domain of jararhagin, involving the cysteine-like domains of BJ46a and the cysteine-rich domain of the metalloprotease, and involving the histidine domain of the inhibitor and metalloendopeptidase domains of jararhagin (BASTOS et al., 2020). However, in order to deepen the information about the interaction between the inhibitor BJ46a and the metalloprotease jararhagin.
Using the Pydock server, three different docking models were made: a general one, i.e., without choice of peptides, one for cysteine peptides and another one for histidine peptides. Pydock uses, as parameters, the desolvation energy, which consists in the displacement of water molecules around a substrate to allow the interaction of enzyme with substrate, the electrostatic interaction between the molecules and the Van der Waal(VdW) forces to estimate the best docking model.
Regarding the results obtained, a convergence between the images and the data obtained was observed, which confirms the results obtained experimentally, which state the cysteine and histidine domains as the main responsible for the formation of the protein-inhibitor complex. The lowest energies obtained and the best 3D representations are expressed in the following images.
Figure 11 - Representation of the curve obtained in the energy minimization for docking stability
Source: https://life.bsc.es/pid/pydockweb#
Figure 12 - Representation of the best conformation obtained for the interaction, in docking form, of the studied proteins
Source: https://life.bsc.es/pid/pydockweb#
In the HDOCK program, the molecular structures of the ligand and the receptor were entered for the calculation of the docking score and ligand rmsd. The two best docking model results are represented by the following images:
Figure 13 - Representation of the best conformation obtained for the interaction, in docking form, of the studied proteins
Source: http://hdock.phys.hust.edu.cn/
Figure 14 - Representation of the second best conformation obtained for the interaction, in docking form, of the studied proteins
Source: http://hdock.phys.hust.edu.cn/
Thus, the program calculated the results of the docking score, which is a score with the function of predicting the affinity of the bond between the ligand and the receptor, and the Ligand rmsd (RMSD = Root Mean Square Deviation, is one of the most commonly used ways to calculate how correct the geometry calculated by docking is, because it shows the interaction distance between the two proteins). According to what is observed in figures x and y, a docking score of -288.50 was obtained for model 1 and -285.25, for model 2, which shows good results, since the lower the docking score value, the more stability the model has. Regarding the ligand rmsd values for model 1, they are in the range of 72.70, while model 2 is at 54.41.
Figure 15 - Values obtained for docking, with the lowest energy being the most stable, therefore best structural conformability
Source: http://hdock.phys.hust.edu.cn/
Substances in general, whether natural or synthetic, are subject to changes in their conformation over time by the action of different natural factors. In this sense, drug stability is defined as the ability of the formulation to maintain its therapeutic, physical and chemical properties over time against external environmental factors (temperature, humidity, light, gases, pH, packaging material) and internal factors (oxidation, hydrolysis, drug-drug interactions, presence of impurities) capable of promoting changes in the drug (SANTOS, 2012; LEITE, 2005; ZHOU et al., 2017). However, to know under which conditions the inhibitor BJ46a maintains chemical and physical integrity, presenting resistance to microbiological growth and maintaining its therapeutic effect without causing toxic effects, it is first necessary to analyze the simulation of its molecular dynamics, i.e., a simulation of Newtonian equations of motion with a large number of particles (PATODIA et al., 2014).
Thus, in order to better understand the stability and storage conditions of the inhibitor of interest, the molecular dynamics of BJ46a was analyzed and from the simulations we extracted important data that relate the structure, functionality and action of the inhibitor. For this purpose, the software GROMACS (Groningen Machine for Chemical Simulation), a versatile and efficient program for Molecular Dynamics analysis, and the CHARMM-GUI (Chemistry at Harvard Macro-molecular Mechanics - Graphical User Interface) server, a web site (http://www.charmm-gui.org) to facilitate and standardize the use of biomolecular simulation techniques (JO et al., 2008; SPOEL et al., 2005) were used.
Figure 16 - Representation of the volume, in box form, of the solvation of the protein in water, with the addition of ions for the neutralization of the charges present in the protein
Source: https://www.gromacs.org/ and https://charmm-gui.org/
Figure 17 - Representation of the volume, in box form, of the solvation of the protein in water, with the addition of ions for the neutralization of the charges present in the protein
Source: https://www.gromacs.org/ and https://charmm-gui.org/
Based on the same principle used in the molecular dynamics of the BJ46a protein, molecular dynamics was performed for the γPLI protein, seeking to obtain the highest possible stability. Based on the modeling done for the 2021 project (Honorato 1.0), in which the best three-dimensional structure for the γPLI protein was obtained, the pdb file was used in the simulation. When plotting the protein on the server, by means of the pdb file, adjustments were made regarding disulfide bridges, which were added in the programming of the protein on the server. In addition, potassium chloride ions were added to neutralize the protein, about 52 potassium ions and 44 chloride ions; the temperature was adjusted, close to room temperature, 303.15 K, suggested by the server to stabilize the protein. Like BJ46a, γPLI is a protein present in the snake organism, so proteins are expected to be stable and adapted to local hot temperatures.
In order to propose a better way to store and distribute the treatment in ampoules, we developed a careful analysis of the protein stability, trying, as much as possible, to realize the proposal of implementing the treatment in society.
Figure 18 - Representation of the volume, in box form, of the solvation of the protein in water, with the addition of ions for the neutralization of the charges present in the protein
Source: https://www.gromacs.org/ and https://charmm-gui.org/
Being part of the pharmaceutical chain cycle, the storage and distribution of pharmaceuticals require a series of stages of selection, study and analysis of the most diverse variables involved in the stability of the active compound for commercial packaging. Thus, it is necessary to consider the entire scenario of the environment in which the production process is immersed, which is formed by the following factors: physical (such as pH or temperature stability), chemical (keeping the substance non-reactive to the medium) and/or biological (seeking sterility, i.e., avoiding or making the active ingredient resistant to the existence of microorganisms) (SANTOS, 2012).
Since all biochemical compounds change, it is incumbent upon the study of the internal and external factors surrounding their manufacturing and distribution processes. While the hydrogen potential (pH) directly influences the speed of hydrolytic reactions, being one of the most important factors in drug stability, especially when in aqueous solution; temperature is the main contributor to the kinetics of the active substance's activity. Having these direct effects on the speed of degradation and substantial reactivity, it is important to evaluate the appropriate storage temperature, as well as the acid or basic character of the drug, in order to reach the point of maximum stability of the compound and, therefore, lower the speed of degradation (SANTOS, 2012).
Additionally, light is another property to consider for the stability study. There is the possibility of reactivity of molecules in the presence of luminosity, with potential formation of toxic products or, even if they are not reactive, suffering some change under storage conditions. Nor can the oxygen molecule have its oxidizing action accelerated by photodegradation of the medicinal substance. Still according to Santos (2012), all these effects, however, can be minimized or delayed through storage in airtight and light-resistant containers, such as amber glass bottles, recommended for suspensions or aqueous solutions. Therefore, it is necessary to pay attention to the characteristics of the active compound, for an effective storage, avoiding the loss of products or changes in the shelf life of the drug.
BJ46a is a glycoprotein that has two cysteine domains, a histidine-rich domain, and a C-terminal portion. In its native form, it has a molecular mass of 46 kDa and a pI (isoelectric point, i.e. the pH value at which the protein molecule has a net charge equal to zero) of 4.6 and thus has an acidic character (PALACIO, 2017). Moreover, tests conducted by Valente et. al. (2001) describe the performance of the protein at pH close to neutrality (specifically, 7.5), immersed in NaCl solution.
This protein, according to Valente et. al. (2001), still has structurally four putative N-glycosylation sites. There is, therefore, the addition of saccharides, specifically, to the amine nitrogen of asparagine side chains, which directly affects their molecular mass.
Thus, for the storage and distribution of the drug, these physicochemical considerations had to be evaluated. The following section describes the simulations performed, based on the listed theory.
The GROMACS program performs versatile molecular dynamics, simulating Newtonian equations of moments for systems with millions of particles. It calculates the diverse and often complex interactions of molecules such as proteins, lipids, and nucleic acids.
We ran the program to get a better sense of the molecular kinetics of our study, and among the various results, the most important ones can be compiled as follows:
The program processed the chain called 1A, containing 2257 atoms and 281 residues, and in it analyzed the hydrogen bonds for an automatic linkage of Histidine protonation. It found 436 donors and 429 acceptors, and 613 hydrogen bonds. Through several simulations, GROMACS defined the residue ASN1 (Aspergine) as the start terminator and the residue LEU281 (Leucine) as the end terminator.
Still on this, the program calculated the matrix of the special atomic distances, generating a bond of our protein with the start and end terminators already mentioned, and proved that in the start terminator the bond is made with NH3+ (ASN1: NH3+) and in the end terminator the bond is made with COO- (LEU281: COO-).
With this, checking the duplicated atoms, generating the hydrogen atoms possibly lost during the simulations and adding the terminators, we now have 261 residues with 4393 atoms, and by linking them, it was possible to create a cmap torsions ("energy correction map", or CMAP, is a map that implements the interactions between pairs of dihedral angles) giving us the results of 11916 dihedrals, 904 impropers, 8067 angles, 11622 pairs, 4459 bonds and no virtual sites.
With this, it was possible to conclude that we have a total mass of 32039.831 a.m.u. (atomic mass unit), lower than the value found by the experiments (46kDa) because of the removal of the signal peptides, and an energy of -20,000 e-.
Because of its negative charge, the program added 20 (Na+) atoms in order to neutralize both the medium and the protein itself.
In this way, the program was able to stabilize our protein by solvation in an aqueous medium through the addition of salts. With the medium and the protein neutralized, the energy was reduced, leaving it minimal for stability. Furthermore, the stability temperature was found to be close to 299K, which is consistent since the snake's habitat is a tropical region, making it believable that its protein is stable at a temperature close to that of its environment.
CHARMM is a widely used academic research program for macromolecular mechanics and dynamics with versatile tools for analysis and manipulation of atomic coordinates and dynamical trajectories. CHARMM-GUI was developed to provide a graphical web interface to generate various input files and molecular systems, facilitate and standardize the use of common and advanced simulation techniques in CHARMM. The web environment provides an ideal platform to build and validate a molecular model system interactively so that if a problem is found through visual inspection, one can revert to the previous configuration and regenerate the entire system again.
First, the software reads the PDB file and plots the three-dimensional schematic of the molecule. Besides adding the PDB file it was also necessary to specify the glycosylation, protonation and disulfide bonds in the protein to ensure maximum optimization of the simulation and to avoid errors.
Next, a solvation simulation of the biomolecule is performed. In this step, the shape of the aqueous system is chosen, in which the rectangular (cubic) shape was chosen, and the dimensions of the system were based on the molecular size. In addition, potassium (K+) and chloride (Cl-) ions were also added to the simulation to neutralize the system.
In addition to the solvation scheme, CHARMM-GUI also provides an optimal stabilization temperature for the biomolecule in the chosen aqueous system. In this case, according to the software, the stabilization temperature of BJ46a is 303.15 K or 30ºC, ideal for tropical regions, such as Brazil.
In previously described literature, BJ46a is expressed and secreted in Pichia pastoris yeasts, through the PPICZ-alphaA plasmid, with a checkmark for the antibiotic zeocin (Shi et al., 2012). However, it is known that both the plasmid and the antibiotic have high costs and, as an alternative to this, we propose the use of another vector for cloning in Pichia pastoris, PPIC9K with a checkmark for the antibiotic kanamycin, wich is very well described in the literature. This proposal aims not only to make the process cheaper, but also an attempt at industrial scaling for the production of the inhibitor.
Therefore, in our project we will use the PPIC9K plasmid, following the protocols described by the manufacturer, for the insertion of the BJ46a inhibitor gene and the jararhagin protein, and their subsequent expressions and secretions in Pichia pastoris yeasts. The methodology for inserting the gene of interest corresponds to the Gibson Assembly. Below, images of restriction maps with genes of interest are presented.
Figure 19 - Restriction map for the BJ46a gene
Source: https://www.snapgene.com/
Figure 20 - Restriction map for the BJ46a gene
Source: https://www.snapgene.com/
In order to make the production process cheaper, at an industrial level, for the BJ46a inhibitor, and increase its accessibility to the population, we developed a circuit, following the 3A Assembly methodology, for the expression and secretion of the protein in Escherichia coli bacteria. It is noteworthy that the proteins produced will be devoid of glycosylation, since prokaryotic organisms do not have sufficient cellular machinery to glycosylate proteins. As for the inhibitory functionality of the protein, experimental data from the literature showed that the absence of glycosylations does not interfere with the ability to interact with jararhagin, maintaining the biological activity of the inhibitor (BASTOS, 2014).
The parts present in the circuit will come from the iGEM Kit. The promoter used was BBa_J23100, the RBS was BBa_J61101, the plasmid was PSB1C3 and the terminator was BBa_B0015. In gene synthesis, histidine tails were inserted to facilitate the process of protein purification. Below, images of restriction maps with genes of interest are presented.
Figure 21 - Restriction map for the BJ46a gene
Source: https://www.snapgene.com/
Figure 22 - Restriction map for the jararhagin gene
Source: https://www.snapgene.com/
In order to improve the biosecurity of the project, a circuit was developed to prevent microbial growth in conditions different from those projected in the project, using the 3A Assembly methodology by selecting parts from the iGEM Kit. Therefore, the aim is to insert this plasmid into Escherichia coli bacteria that will produce the proteins of interest, to prevent possible biological disasters.
The parts to be used are: promoter that conditions growth under UV light BBa_1765001, the same RBS BBa_J61101, a sequence to induce cell lysis BBa_K117000 and the plasmid PSB1C3. When activated, the circuit will lyse the bacteria causing its death. The image of the restriction map with the gene of interest is shown below.
Figure 23 - Restriction map for KillSwitch
Source: https://www.snapgene.com/
Kinetic modeling is a mathematical modeling tool often used, because through the equation of enzyme activities it is possible to make predictions about the actual behavior of the inhibition that will be studied in the laboratory. This is the main reason for studying the kinetics of inhibition and modeling its behavior. Through these predictions, it is possible to infer the possible results of the experiment and, thus, saving time and laboratory reagents, which is extremely important to reduce the environmental impact of the project, optimizing the processes through the modeling of reactions and reducing the financial expenditure by the laboratory.
According to the literature, it is known that proteins are not allosteric and, therefore, can be modeled according to the Michaelis-Menten mechanism (NEVES-FERREIRA et al., 2015). Furthermore, proteins interact by forming a non-covalent, i.e. reversible, complex in which the inhibitor competes with the substrate for the (competitive) binding site (VALENTE et al., 2001).
Thus, the kinetic modeling of the inhibition of the enzyme studied is an essential part of the project, which will help in conducting the practical experiments as well as to achieve a better theoretical understanding of the process.
From the Michaelis-Menten mechanism, an equation is obtained to describe the inhibition between the two proteins studied. At the end of the reaction, one expects to obtain a non-covalent enzyme (E) and inhibitor (I) complex and a percentage of substrate in the medium.
Terminology:
We have,
Therefore,
Next, the experimental data obtained from the literature were used to demonstrate the kinetic equation. The data obtained are from inhibitory assays performed with the inhibitor BJ46a, two metalloproteinases, which have great genetic similarities with the metalloproteinase jararhagin and the gelatinous substrate Enzchek (KAI et al., 2013). The values used were: km = ks = 3.6 micromol/L; ki = 13.6 nmol/L; [I] = 9.6154*10-11 mol/L.
Figure 1 - Inhibitory activity curve of BJ46a
Source: KAI et al. (2013)
Remarks: increasing the concentration of the BJ46a inhibitor decreased the metalloproteinase activity in exactly the same time.
Figure 2 - Lineweaver-Burk Linearization
Source: KAI et al. (2013)
Remarks: the graph represents the inhibition activity of BJ46a. Line a is without the use of BJ46a and line b is with the use of BJ46a.
The data from the graphs were extracted into a table using GetData software. Only the data from row b were extracted, due to the real purpose of the equation.
Figure 3 - Values of x (1 divided by substrate concentration) and y (1 divided by speed)
Figure 4 - Lineweaver-Burk Linearization
Source: adapted from KAI et al. (2013)
From the linearization obtained from the data in the table above, the substrate concentration values for their respective velocities were calculated and compared in the following table, as described below.
Figure 5 - Concentration values per speed found
Source: Author (2022)
Then, based on the demonstrated Equation I, and using the constant values obtained by linearizing the graph, subject to an experimental correction factor f, and knowing that Km/Vmax is the angular coefficient of the line and 1/Vmax is the linear coefficient, from the mathematical resolution, one has:
Whose maximum velocity and initial substrate considerations were made from the first ordered pair of velocity per real substrate found, that is, by the table above, [S] = 50.7614 μ mol/L and V = 25.7069 μ mol/L.s.
Thus, by means of graphical software, namely GeoGebra and Scilab, it was possible to plot a logarithmic curve representing the time traveled in relation to the remaining substrate. The graphs 5a and 5b, print, respectively, the characteristic graphic structure of this expressed function, where, for Scilab, a decay interval from the initial substrate to zero was placed for analytical observation.
Figure 6 - Time curves per remaining substrate using the GeoGebra software
Source: Author (2022)
Figure 7 - Time curves per remaining substrate using the SciLab software
Source: Author (2022)
In previous studies analyzing the interaction of BJ46a with different metalloproteinases of snake venoms, it was shown that the inhibitor interacts with the metalloproteinases jararhagin and atrolysin-C, however, there is no complex formation with jararhagin-C, which indicates that the metalloproteinase domain is essential for the inhibitor-toxin interaction.
For the experiment proposed by Bastos (2014), jararhagin was analyzed in an inhibitory assay, in which it was incubated with BJ46a at the following ratios: 0.33:1; 0.5:1; 1:1; 2:1 and 3:1 of inhibitor to toxin (i.e., [I]:[E]). First, being an assay of the azocaseinolytic substrate of BJ46a on the metalloprotease jararhagin, 5 μg of jararhagin were incubated in the presence of the inhibitor, exploiting the different molar ratios mentioned for 30 minutes and at 37°C, in Tris-HCl buffer solution (10 mM + 100 mM NaCl, pH 8.6).
Then 225 μL of calcium chloride (45 mM) were added to all samples except for the control, which had the same volume added, but an EDTA solution (45 mM). Next, 225 μL of 0.5% w/v azocasein was added to the described buffer, under incubation at 37 °C for 1 hour. After this period, the reaction was stopped by the addition of a 15% (v/v) trichloroacetic acid solution. The samples were centrifuged at 14,000 xg for 10 minutes and aliquots of 150 μL of the supernatant from each sample were transferred to a 96-well plate. After that, 150 μL of 0.5 M NaOH solution was added to each sample. The samples were then read in a spectrophotometer at a wavelength of 440 nm. The assays were performed in triplicate, and the results were expressed from the subtraction of the mean absorbance recorded for the blank samples by the mean absorbance of each sample (Bastos, 2014).
As in the proportion 1:1 the peak referring to the complex becomes the majority, with the presence of free inhibitor in the proportions 2:1 and 3:1, it was concluded the saturation of the complex in 1:1.
The best inhibitory results obtained from this assay are concentrated in the last three proportions of inhibitor to toxin ratios mentioned. BJ46a was able to inhibit 84% of the toxin activity at the 1:1 ratio, 98.9% at the 2:1 ratio, and 98.3% at 3:1.
The table below compiles the inhibition information assayed by Bastos (2014), showing the molar and mass ratio for each inhibition range evaluated. For this, the molecular masses of jararhagin were considered as 52 kDa and BJ46a as 46 kDa and under the note that 5 μg of the enzyme were used in all assays.
Figure 8 - Inhibition of differents proportions from [I] : [E]
Source: adapted from Bastos (2014)
In parallel, in the inhibitory assay proposed by Valente (2001), BJ46a was tested against atrolisin-C and against jararhagin using the fluorogenic substrate of Abz-Ala-Gly-Leu-Ala-Nbz. In this case, the proteolytic inhibition activities of Atrolisin-C and jararhagin were placed in the following table:
Figure 9 - Inhibition test of the isolated BJ46a inhibitor against atrolysin C SVMPs (class P-I) and jararhagin (class P-III)
Source: adapted from Valente (2001)
In the representation in Figure 9, [I] represents the inhibitor BJ46a, the data presented are in the form of percentage inhibition and the molecular weights assumed for the calculations were 46 101, 23 062 and 52 000 Da for BJ46a, atrolysin C and jararhagin.
For jararhagin, the following figure shows that the peak of complex (BJ46a/jararhagin) had a different retention than that presented by BJ46a alone (0:1). In this case, a clear increase in the peak area of the complex was observed until it reached the 2:1 jararhagin/BJ46a ratio. At 3:1 ratio, the increase in the peak relative to the 2:1 ratio is not prominent, and, in addition, the appearance of 94% free jararhagin relative to its control (1:0) occurs, indicating that the complex reached its saturation at the 2:1 ratio (Valente, 2001).
Figure 10 - Complex formation between BJ46a and Jararhagin
Source: adapted from Valente (2001)
For 1, 2, 3, 4 and 5 of the X axis referring to the ratios 1:0, 0:1, 1:1, 2:1 and 3:1, respectively.
Thus, it is understood that, according to Valente (2001), the proportion that obtained the best results in the inhibition tests was 1:1 in which the inhibitor BJ46a obtained the highest percentage of inhibition of jararhagin, 91% without the presence of free jararhagin.