Mutation Effect Investigation on Cytochrome C552 Protein Instability and Electron Transfer Improvement in the Acidithiobacillus Ferrooxidans Bacteria Respiratory Chain

Cytochrome c 552 (Cyc 1 ) is a protein in the electron transport chain of the Acidithiobacillus ferrooxidans (Af) bacteria which obtain their energy from oxidation Fe 2+ to Fe +3 . The electrons are directed through Cyc 2 , RCY (rusticyanin), Cytochrome c 552, and Cox aa 3 proteins to O 2 . Cytochrome c 552 protein consists of two chains, A and B. In the present study, a new mutation (E121D) in the A chain of cytochrome c 552 protein was selected due to electron receiving from Histidine 143 of RCY. Then, the changes performed in the E121D mutant were evaluated by MD simulations analyzes. Cytochrome c 552 and RCY proteins were docked by a Patchdock server. By E121D mutation, the connection between the two chains in Cytochrome c 552 was enhanced by an additional hydrogen bond between Zn1388 and aspartate 121. Asp 121 in chain A gets farther from Zn 1388 in chain B. Therefore, the aspartate gets closer to Cu 1156 of the RCY leading to the higher stability of the RCY/Cytochrome c 552 complex. Further, an acidic residue (Glu121) becomes a more acidic residue (Asp121) and improving the electron transfer to Cytochrome c 552 protein. The results of RMSF analysis showed further ligand exibility in mutation. This leads to uctuation of the active site and increases redox potential at the mutation point and the speed of electron transfer. This study also predicts that in all respiratory chain proteins, electrons probably enter the rst active site via glutamate and exit through the second active site of each respiratory chain protein and through histidine. complexes ferrooxidans suggests their respective roles iron or sulfur oxidation.


Introduction
Cytochrome c 552 protein is an essential protein in the electron transport chain on the Af bacterium membrane. This bacterium is considered one of the main bacteria involved in the metals bioleaching process. Bioleaching is the process of extracting metals from their minerals by using microorganisms.
The electrons are transmitted through several protein carriers including the Cyc 2 , RCY, Cytochrome c 552 , and Cox aa 3 . Cyc 2 is a protein that is available in the external membrane of Af. This protein was proposed as the rst electron receptor on the respiratory chain between iron (Fe 2+ ) and oxygen [1]. The terminal electron receptor is a cytochrome oxidase (Cox aa 3 ) in the cytoplasmic membrane. Apart from many other anaerobic respiratory chains, the bioenergetics metabolism of this organism involves several proteins with the highest redox potential and is encoded by rus and petI operons for the downward and upward pathways, respectively. Redox potential is considered as a measure of the tendency of the chemical species (e.g. aqueous solutions) to either gains or loses electrons. A solution with a higher reduction potential (more positive) has more tendency to gain electrons and vice versa. The path of the respiratory chain (downward pathway) could be Cyc 2 → RCY → Cytochrome c 552 → Cox aa 3 → O 2 [2].
Biochemical studies have shown that Fe 2+ is oxidized by Cyc 2 , which is present in the outer membrane. In addition, the electrons are directed to O 2 through RCY and Cytochrome c 552 periplasmic proteins, and nally, Cox aa 3 in the internal membrane [3][4][5]. These enzymes with molecular oxygen reduction eventually led to the production of water molecules [6,7]. Each of these enzymes has two active sites, one associated with the previous protein to receive electrons ( rst active site) and the other associated with the next protein to release electrons (second active site). The electron transfer reason in the respiratory chain is the redox potential difference at chain components, and there is a high correlation between the amount of redox potential and electron reception. The standard redox potential (E 0 ) is considered as a numerical measure of convenience which can be reductions to the structure or make it easy to accept the electrons. If E 0 is more positive, it means more readiness for electron acceptance, and if E 0 is more negative, it means more readiness for releasing the electron. Therefore, electrons can be freely transferred from a set with a lower E 0 to a higher E 0 . In the electron transport chain, each receiver has a larger E 0 than the electron donator.  (Fig. 2A& B). Jafarpour et al. (2020) studied the effect of the mutation on the RCY second active site [10]. In the present study, the effect of the novel mutation (E121D) on the same type of bacterium at the Cytochrome c 552 rst active site was investigated to speed up the bioleaching process by MD simulation methods.

Homology-Based Modeling of Wild Type and Mutant Structures
A. ferrooxidans ATCC 23270 (Id code: B7JAQ6) was selected as a model using the UniProt database [11]. The Cytochrome c 552 protein crystallographic structure with a resolution of 2.13 and 98% identity was proposed as a template structure by the I-TASSER server (PDB ID: 1h1o). 3D structures were generated using modeler software (version 9.12). And, 100 pdb les were designed with different angles by using Modeler software. Finally, the rst 25 pdb les sorted from low to high according to DOPE were selected. Docking Molecular docking is a computational procedure that aims to predict the favored orientation of a ligand to its macromolecular target (receptor) when these are bound to each other to form a stable complex [12]. The protein ability (enzyme) to interact with small molecules to form a supramolecular complex plays an important role in protein dynamics, enhancing or inhibiting its biological function [13].

Setting up parameters for docking
The wild and mutated protein Cytochrome c 552 required Zn 1388 as a ligand involved in the protein structure and action. Thus, the assembly calculations were done with the AutoDock software version 4.2 based on the Lamarck genetic algorithm and experimental free energy function. The box center was set on the 121 glutamate residue, and the size of the box was adjusted for the free turning of the ligand [14][15][16]. Finally, the Cytochrome c 552 and RCY proteins were docked by the Patchdock server to check the status of the two proteins relative to each other [17].

Mutations
Initially, we used MUpro and I-stable servers to predict the stability of mutated protein. According to both servers, this mutation reduced protein stability ( Table 1). The Cytochrome c 552 mutant model was generated using PyMOL software at position 121 from glutamate to aspartate in wild protein and the mutant structure was saved as a PDB structure [18]. Some servers such as ProSA-web, Q Mean, verify 3D, and RAMPAGE support structural wild and mutant [19].

Simulation System Setup
To explore the structural effect on Cytochrome c 552 protein due to E121D point mutation, comparative MD simulation studies were performed in GROMACS 5.1.4 package on an Ubuntu Linux system and GROMOS96 43a1 force-eld [20][21][22][23]. The modeled protein was solvated in a cubic box by the SPC216 water model [24]. Then the system was neutralized by adding, 4 Na + and 12 Clions to replace the rst SPC water molecule in all directions [25]. After 5000 steps of energy minimization, MD simulation was run in NVT and NPT groups. The rst phase of the NVT was run for a period of 100 ps and the second phase of the NPT was done for 100 ps under the position restraints condition for heavy atoms [26]. Final MD was run for 100 ns, and the atomic coordinates were saved every two ps to identify the structural change in the conformation of the Cytochrome c 552 protein [19]. Then, the results of the generated les were visualized with VMD (Visual molecular dynamics) software [27,28] and analyzed using standard software presented by GROMACS 5.1.4 [28].
pathway Analysis MD simulation results were evaluated by using trajectory les obtained, and the structural behavior of wild and mutant structures was compared. Then, the comparative analysis was performed for wild and mutant types. In the next method, root mean square deviation (RMSD), Radius of gyration (Rg), Root Mean Square Fluctuation (RMSF), Solvent Accessible Surface Area (SASA), and Database of secondary structure assignments for all Protein (DSSP) were analyzed. Finally, hydrogen bonds (H Bonds) shaped by special residues of the protein to the solvent at the time of simulation were calculated [23].

Results
Checking the quality of the model produced Employing four servers including ProSA-web, QMEAN, RAMPAGE, and Verify 3D, the best model was selected before and after mutation. The overall model quality was additionally examined using ProSAweb along with QMEAN servers. Then, the standardized score was calculated, which is a total model quality index. The results of PorSA-web and QMEAN Z-score indicated that the models have desirable qualities. In addition, 81.52% of amino acids having a score of 3D-1D> = 0.2 in verify 3D and RAMPAGE with 98.9% residues in the allowed region. The wild-type and E121D mutant were visualized using PyMOL as shown in Figure 2A-B.
The hydrogen bonding results analysis in Glu, Asp 121 in wild-type and mutant protein using PyMOL software The results showed that glutamate has four hydrogen bonds in the side chain with the surrounding molecules in wild-type. However, there is an extra hydrogen bond that binds the aspartate to Zn1388 of chain B in E121D mutation ( Fig.3A-B).
Analysis of the results of the distance between Zn 1388 of chain B and Cu1156 of RCY to glutamate 121 and aspartate 121 (after mutation) chain A using PyMOL software.
In E121D mutation, an increase occurred in the distance between Zn1388 of chain B and aspartate 121 of chain A compared to glutamate 121 (2.2 angstroms in E121 and 4.4 in D121). Also, a decrease occurred in the distance between aspartate 121 of chain A and Cu1156 of RCY at RCY/Cytochrome c 552 complex (16.9 angstroms in E121 and 15.3 in D121) ( Fig. 4A-B).
The results of MD simulation analysis and effects mutation on stability and protein secondary structure MD simulation was performed to study the con guration, and stability in the wild-type protein and the changes of these parameters in the mutant protein, for both types of proteins. In addition, MD simulation is considered a useful method to support the experimental results in protein mutations through structural speci cations at the atomic level [29]. RMSD, RMSF, SASA, H-bonds, Rg, and DSSP were analyzed throughout the simulation time ( Table 2, 3).
To nd out the effect of the mutation on conformational changes and the stability of the protein structure, the RMSD of Cα atoms was calculated for both the wild-type and mutant compared to the original structure [19]. The RMSD plot of the wild and mutant structures indicated the convergence pattern during 100 ns simulations (Fig. 5A). Convergence obtained for wild and mutant structures con rmed that the mutant protein has less sustainability than the wild-type.
The RMSF of each residue was measured to assign the mutation effect on the residues' dynamic behavior in wild and mutant proteins [19]. Overall, the mutant protein had more exibility (RMSF) than the wild-type (+0.0002 increase) [30]. The RMSF plan is plotted for wild and mutant structures, as shown in Figure 5B.
Intermolecular hydrogen bonding is the electrostatic force between two polar bonds. In this bonding, hydrogen is bonded covalently to the highly electronegative atoms like nitrogen and oxygen. H bond was calculated for wild and mutant types. The average number of H bonds for the mutated structure was 125.78 ± 4.52 and -2 number decreased compared to wild protein ( Table 2). The difference in the mutant H bonds con rmed the exibility of the mutant structure (Fig. 5C).
Rg determines the compression of the protein. As the radius of gyration is higher, the protein tends to be more unstable, unfolding, and low compact [31]. The values of Rg for mutated variants were 1.62 ± 0.001 nm, with a slight change (+0.01 increase) in protein folding and its compactness (Table 2) (Fig. 5D).
SASA indicates the level of protein access to the solvent [32]. The change in SASA measured the accessibility of the protein to the solvent in the wild and mutant structure during simulations (Fig. 5E).
The average of SASA for the mutated structure was 103.10 ± 0.59 nm 2 and +1.23 increased compared to wild protein (Table 2). amino acids. The bar graph of the secondary structure for wild-type and E121D mutant systems is shown in Figure 6A. The amount of Structure, coil, Turn, and beta bridge in the mutant protein has increased but, beta-sheet, alpha-helix, have diminished. The highest value is shown in the secondary structure of the two systems for coil and alpha-helix. And the largest change is related to the turn structure (1.24% increase), which is shown in Table 3.

Results of φ and ψ angle analysis of amino acids in wild and mutant proteins by Ramachandran method
The results were reported as diagrams of changes in angles according to time, to investigate the state of the protein at the mutation point. According to the angles φ and ψ before and after mutation using the Ramachandran method, it was found that the amino acid studied did not change signi cantly in terms of angles φ and ψ (Fig. 7, Table 4).
The Wild and mutant proteins Ramachandran scheme produced by PROCHECK To compare the Ramachandran results of simulated wild and mutant protein (E121D), the pdb les were uploaded and the results are shown in Figure 8 and Table 5. Before the mutation, 95.5% of the amino acids are present in most favored regions, which after the mutation, its amount has decreased to 84%.
There were no amino acids in the allowed and disallowed regions before the mutation, but after the E121D mutation, two of the amino acids (alanine 9, histidine 54) were in the allowed regions.

Discussion
Following previous studies, E121D mutation at the rst active site and the electron entry point from the RCY to Cytochrome c 552 were conducted on Af bacterium. Molecular dynamics simulation studies on wild-type and mutant identi ed the in uence of mutations at the active site. The trajectory analysis was used to calculate protein changes and showed the changes in protein structure associated with increased exibility at the mutation point (+ 0.0017 increase).
DSSP is an algorithm that deals with information on the secondary structure of proteins [33]. DSSP analysis aims to measure the content of the secondary structure of a protein as a function of time. The analysis of the wild and mutated structures shows the number of amino acids involved in different structures, as well as increasing or decreasing the number of amino acids in the structure. In the DSSP scheme, with the conversion of glutamate to aspartate, structure, coil, β Bridge, and Turn increased by 0.62%, 0.08%, 0.09% and 1.24%, respectively, and β Sheet, α helix, 5-helix, and 3-helix decreased by 0.006%, 0.69%, 0.48%, and 0.11%, respectively. A decrease of 0.69% in the alpha-helix in the DSSP plan is related to the fact that aspartate tends to disrupt a helix because their side chains contain hydrogen-bond donors or acceptors in proximity to the main chain, where they compete for main-chain NH and CO groups [34]. The number of amino acids which participate in protein structure eventually increased by 0.62% (Table 3).
As mentioned, the Af bacterium is considered as one of the main bacteria involved in the metal bioleaching process, and bioleaching is the extraction of metals from their minerals through using these bacteria [11]. RCY and Cytochrome c 552 are considered as consecutive proteins in the electron transport chain Af bacterium. In this way, the electron is transmitted from RCY to Cytochrome c 552 [9]. angstroms instead of 16.9 angstroms for glutamate) (Fig. 4A-B). The kd value of glutamate for zinc ions and copper ions is 10 − 5.5 and 10 − 8 at pH 4.5, respectively (the lower kd value leads to the higher binding a nity of the ligand to its target) [8]. The glutamate interacts with zinc inhibiting the interaction with Rusticyanin. Further, with E121D mutation, the role of zinc inhibiting on the Asp 121 Cytochrome c 552 decreases and one could expect shorter distances with the Rustycianin copper. Therefore, its tendency to copper 1156 of RCY increases, and the RCY/Cytochrome c 552 complex becomes more stable after the mutation.
In this study, we proposed the novel E121D mutation in which glutamate has four hydrogen bonds in the side chain with the surrounding molecules. However, in aspartate, there is an extra hydrogen bond that binds this amino acid to Zn1388 in the B chain and creates a stronger bond between the two chains ( Fig. 3A-B).
Abergel et al. reported that histidine is deprotonated due to the 3.2 Å distance between Nε H143 and Oγ E12, and their formation of hydrogen bonds between. The RCY redox potential decreases and, consequently, the transfer of electrons from the RCY to Cytochrome c 552 is more e cient [8]. This mutation converts an acidic residue into a more acidic residue Therefore, His143 becomes more deprotonated, reduces the redox potential at the RCY midpoint, and improves electron transfer to Cytochrome c 552 protein at RCY/Cytochrome c 552 complex.
Also, the results of RMSF analysis show more exibility in the mutation point compared to the wild type. The E121D mutation reduces stability at the active site. Finally, instability of the active site leads to increased ligand exibility (ΔS) and ΔG decrease and increase in the value of E 0 which is proved by two formulas Gibbs free energy (ΔG = ΔH -TΔS) and Nernst (E 0 = −ΔG/nF) [37]. Finally, by the increase in redox potential, electron transfer from RCY to Cytochrome c 552 at RCY/Cytochrome c 552 complex is accelerated. In this groundbreaking study, we found that in all respiratory chain proteins, electrons probably enter through the glutamate in the rst active site and exit through histidine at the second active site of each respiratory chain protein. Therefore, by converting glutamate to aspartate at the electron entry point and converting histidine to arginine at the electron exit point, and destabilizing the active site of each protein in the respiratory chain, the electron transfer rate in the chain, followed by the bioleaching process, can be improved.

Conclusion
Based on the results, the mutated protein (E121D) is unfolded more than wild protein. The analysis of the hydrogen bond indicates that the mutant protein is more exible. However, in aspartate, there is an extra hydrogen bond that binds this amino acid to Zn1388 in the B chain and creates a stronger bond between the two chains. By converting glutamate 121 to aspartate, a decrease occurs in the role of zinc inhibiting on the Asp 121 Cytochrome c 552 . Therefore, the aspartate gets closer to Cu 1156 of the RCY and results in higher stability of the RCY/Cytochrome c 552 complex. On the other hand, an acidic residue (Glu121) becomes a more acidic residue (Asp121), and His143 is more deprotonated inducing a reduction of the redox potential at the rusticyanin midpoint improving the electron transfer to Cytochrome c 552 protein.
Also, by the increase in exibility and decreasing ΔG in the active site, the redox potential at the mutation point increases, and the electron transfer to Cytochrome c 552 improves. For future studies, we propose extensive research on electron transfer improvement by selecting the best mutations obtained by bioinformatics methods at all active sites (electron entry and exit points) in all respiratory chain proteins.
In this way, The amino acids in the rst active site of the proteins, namely glutamate, are converted to aspartate with more acidic properties. And the amino acid histidine converted to the more alkaline amino acid arginine in the second active site position. As a result, the active site becomes more unstable and the speed of electron transfer in the chain increases. Finally, biotechnological methods can clone bacteria that have the above mutations to improve the bioleaching process.

Declarations Funding
No funding was received for conducting this study.
PLoS One, 10(4).  The values of RMSD, RMSF and Rg are given in nm, values of SASA given in nm 2 , and values of NH bonds given in numbers.