Ovalbumin’s potential as a wound-healing medicament in tooth extraction socket by induction of cell proliferation through the ERK2 pathway in silico

Background: The trend of studies on dental medicaments is increasing rapidly. Antibacterial or anti-inflammatory activity is most frequently studied. Ovalbumin is one of the proteins whose benefits have been studied, but these benefits are still limited because of ovalbumin’s potential for proliferative bioactivity. Purpose: The aim of this study is to examine ovalbumin’s potential as a wound-healing medicament through molecular docking analysis on a protein related to the extracellular signal-regulated kinases/mitogen-activated protein kinase (ERK/MAPK) signaling pathway. Methods: Ovalbumin was hydrolyzed through BIOPEP-UWM (The BIOPEP-UWM™ database of bioactive peptides). Protein target and interaction were predicted using Similarity Ensemble Approach target prediction webserver, SuperPred webserver, STRING webserver, and Cytoscape version 3.9.1. Selected fragments were docked using Autodock Vina in PyRx 0.8 with Tukey’s multiple comparison test and Biovia Discovery Studio version 19.1.0.18287 for visualization. Results: This study found that ovalbumin has the potential to positively regulate cell proliferation, angiogenesis, and fibroblast growth factor production. Six of the 131 fragments of ovalbumin could interact with 73 proteins, and the 20 proteins with the highest probability and score of betweenness centrality showed potential for bioactivity. Five fragments and povidone-iodine interacted inside the Adenosine triphosphate (ATP) phosphorylation site of ERK2, whereas fragment 1 (F1) and glycerin interacted outside the site. F1 could decrease the binding energy required for adenosine 5 ′ -[,-methylene]triphosphate or an ATP-analogue chemical compound to interact with ERK2 compared to the control, with a score that was not significant. Conclusion: Ovalbumin has the potential to induce cell proliferation by affecting ERK2-ligand interactions.


INTRODUCTION
The trend of studies on medicaments to prevent and treat dental problems has increased rapidly since 2010, based on data published in PubMed (https://pubmed.ncbi.nlm. nih.gov/). The antibacterial or anti-inflammatory activities of medicament ingredients have been widely examined in previous studies. 1,2 Medicament materials from antibiotics such as ciprofloxacin, metronidazole, and doxycycline, 3 as well as other materials such as calcium hydroxide, 1,3 odontopaste, 3 and quaternary ammonium silane/k21 4 are known to have antibacterial activity.
Wound healing in the oral cavity, including in the tooth sockets, is affected by bacterial and inflammatory activity. Other medicaments that have been studied for wound healing are povidone-iodine (PVP-I) and glycerin (glycerol; and GLY were tested on 347 patients with acute otitis externa (AOE) to analyze their efficacy against the AOE. The clinical trial results showed that AOE treatment with both medicaments has good efficacy and can relieve canal edema and pain in the patient's tragus. 6 Wound healing is a very complex process involving multiple bioactivities and molecular signaling. Several important bioactivities in wound healing are related to the induction/suppression of inflammation, proliferation, and cell differentiation. The mitogen-activated protein kinase (MAPK) pathway is one of the important signaling pathways associated with these bioactivities. 7 The woundhealing process after a tooth extraction is the same as any other tissue-healing process; that is, it is complex and dynamic. Restoration of damaged tissue integrity involves cellular components and an extracellular matrix (ECM). 8 However, research on medicaments for molecular wound healing is still rare, necessitating the current study. Ovalbumin, the main component in chicken eggs (Gallus domesticus), makes up more than 50% of all protein components in eggs. Molecularly, ovalbumin (SERPINB14) is a protein with a molecular weight of 45 kDa with 385 amino acids and is a part of the large Serpin group (the serpin superfamily). Based on the amino acid sequence (aa), there is 1 disulfide bond linking Cys74 and Cys121, with half of the residues being hydrophobic and one-third being acidic. 9,10 Ovalbumin is known to play a role as a carrier protein that can increase the antioxidant effect and solubility of curcumin. 11 However, there is still limited information related to its proliferative activity. Therefore, this study aims to examine the potential of ovalbumin as a wound-healing medicament in tooth extraction sockets by induction of cell proliferation through the ERK/MAPK signaling compared to PVP-I and GLY in silico.
Six of the selected fragments were converted to Simplified Molecular Input Line Entry System (SMILES) via NovoProlabs (https://www.novoprolabs.com/tools/ convert-peptide-to-smiles-string). Next, each SMILES fragment was loaded onto the Similarity Ensemble Approach (SEA) (https://sea.bkslab.org/) and SuperPred (https://prediction.charite.de/subpages/target_prediction. php) webservers to obtain protein prediction targets. 13,14 The cut off chosen from the SEA results was max Tc > 0.5, and SuperPred had a probability > 90%. Then, from the prediction results, ≥ 70 target proteins were obtained. The target protein was inputted into STRING (https://stringdb.org/) to see the bioactivity of the target protein. 15 The setting used was a "physical" type network that showed "confidence" through the thickness of the line with high confidence criteria (0.700). The target protein without interaction was eliminated so that 29 target proteins remained.
The selected target proteins (29 proteins) were downloaded from STRING and analyzed using Cytoscape ver. 3.9.1 (Cytoscape Consortium) and the GOlorize plugin. 16,17 First, the protein network was analyzed using the NetworkAnalyzer to obtain data betweenness and closeness centrality and degree. 18 Then, the network was analyzed by gene ontology (GO) using the GOlorize plugin, which generated GO data related to proteins in the tissue and colored protein nodes in the tissue according to their bioactivity. Furthermore, bioactivities related to wound healing including regulation of proliferation, angiogenesis, growth factor production, and anti-inflammatory 19 were selected for further study.
Six fragments (fragment 1 to fragment 6), PVP-I (CID: 11989721), GLY (CID: 753), and AMP-PCP (ATPanalogue) (CID: 91532), were used for molecular docking analysis with the predicted protein. The selected target protein had average value betweenness centrality (BC) and the highest predictive value of all fragments with PVP-I (CID: 11989721) and GLY (CID: 753) as controls. Furthermore, the second docking was carried out to examine the effect of the fragments on the interaction of ATP with MAPK1. The second docking was carried out between AMP-PCP (CID: 91532) and MAPK1 (ERK2) (PDB ID: 5V60) with the previous six ovalbumin fragment ligands. Molecular docking was carried out using Autodock Vina on PyRx 0.9.7 with ligands that were minimized in energy through the Open Babel plugin and proteins that were removed by water molecules and ligands using Biovia Discovery Studio ver 19.1.0.18287. 20 Finally, visualization was carried out with Biovia Discovery Studio ver 19.1.0.18287 to visualize the interaction between fragments and the AMP-PCP (ATP-analogue) to ERK2.
Statistical analysis was conducted to validate the results of the docking analysis, which presented standard deviations. The analysis was conducted using GraphPad Prism9 with a one-way ANOVA followed by Tukey's multiple comparisons test.

RESULTS
The ovalbumin peptide fragments obtained were 142 fragments from 141 sites cut by the selected enzymes. This hydrolysis model used independent variables in the form of the number and type of enzymes, where the following fragments were the result of the activity of chymotrypsin (EC 3.4.21.1); trypsin (EC 3.4.21.4); and pepsin, pH 1.3 (EC 3.4.23.1), which acted at the same time. There were 50 fragments with one amino acid (phenylalanine, histidine, lysine, methionine, asparagine, arginine, tryptophan, and tyrosine), 38 and 20 fragments with two and three amino acids, 18 and 3 fragments with four and five amino acids, and 13 fragments with more than 5 amino acids, one of those was composed of 19 amino acids (Table 1).
Furthermore, to analyze the fragment's potential bioactivity related to cell proliferation, we used ovalbumin's active peptide database, which has antioxidant and anti-inflammatory bioactivity. The active peptide was then used to determine which fragments had the same peptide composition. Next, 142 fragments were reduced to six active fragments: AAH (Fragment 1/F1), EL (F2), GSIGAASM (F3), GIIR (F4), TSVL (F5), and VY (F6) ( Table 2). The six fragments were predicted to interact with a total of 74 proteins, and five proteins-CAPN1 (calpain-1 catalytic subunit), CFB (complement factor B), FOHL1 (folate hydrolase 1), Human leukocyte antigens class I histocompatibility antigen (HLA-A), and N-acetylated-alpha-linked acidic dipeptidase 2 (NAALAD2)-were predicted to interact with two fragments. Meanwhile, PVP-I and GLY were predicted not to form interactions with any proteins based on the SEA target prediction.
The next protein prediction was conducted using SuperPred to obtain additional prediction results from other   1 : amino acid with anti-inflammatory bioactivity. The bold print indicates the amino acids are predicted to be active peptides. Figure 1. Prediction of target protein for each fragment from SEA target prediction and SuperPred prediction. The target proteins displayed are only those that had a probability value exceeding the specified cutoff (SEA target prediction, cutoff maxTc ≥ 0.5; SuperPred, cutoff probability ≥ 90%). The black color of the squares corresponds to the increase in the probability score. from the previous analysis ( Figure 1). Some proteins have an average probability of 0% because they are not predicted to interact with any fragments during target prediction analysis.  databases and increase the probability of approaching the real condition. A total of 34 proteins were predicted to interact with six fragments and control drugs. PVP-I was predicted to interact with CTSD, ERAB, and NFKB1, whereas GLY was predicted to interact with lamin A/C (LMNA; Figure 1). Next, 30 proteins from the enrichment process were obtained and arranged according to the value of BC. BC was obtained from the protein interaction network. The higher the BC score, the more important the role of the protein in the network. 18 The target protein for docking analysis was selected from the previous 29 proteins with the highest BC and MaxTC/probability scores ( Figure 2; Table 3). ERK2 was chosen as a protein target for molecular docking analysis because it is known to interact with all fragments based on SuperPred predictions, and it has a BC score of 0.14 with a mean probability of 72.19%

SEA Target SuperPred
to interact with all ligands. Then, based on GO analysis, ERK2 had a role in the processes of proliferation and cell survival. It would be interesting to study further the effect of the ovalbumin fragment in influencing the interaction of ERK2 with its downstream protein.
Hereafter, bioactivity analysis based on the interactions of 59 proteins demonstrates the presence of bioactivity, and Table 4 shows associated proteins. The three biological activities of the selected ontology genes had different confidence and strength scores. The confidence score used was the p-value score and Bonferroni's corrected p-value to increase the credibility of the results. Based on Bonferroni's p-value and corrected p-value scores, 21 the three bioactivities had a score of < 0.05, meaning they had a probability of occurring. Positive regulation of cell proliferation was the most potent bioactivity, followed by  The AMP-PCP binding energy score against ERK2, which interacted with all ligands, was compared with the AMP-PCP binding energy score against the control ERK2 (C). One-way ANOVA (P < .05); *p = .0216, ***p < .001, ****p < .0001. positive regulation of angiogenesis and fibroblast growth factor (FGF) production.
Next are the results related to molecular docking analysis. The ATP-binding site or phosphorylation site of ERK2 (PDB ID: 5V60) was found in amino acids 31-39 and 54, including ILE31, GLY32, GLU33, GLY34, ALA35, TYR36, GLY37, MET38, VAL39, and LYS54. 22 The six fragments were docked blindly to find out where the optimal location of interaction would occur in each ligand. Docking was carried out in three replications with different coordinates. Replication coordinates 1, center: According to the second docking result, F1 interaction can reduce AMP-PCP binding energy to ERK2 by 0.01% compared to the control (Figure 4). A visualization was carried out to understand in detail the interactions that occurred between the ligand and ERK2. The visualization of AMP-PCP control and original AMP-PCP in the ERK2 ligand (PDB ID: 5V60) aims to validate the AMP-PCP pose after re-docking ( Figure 5). Table 5 is the list of all residues that interacted with the ligand.

DISCUSSION
Tooth extraction is an activity that causes injury to the tooth socket because of the presence of parts of the gums and nerves that interact directly with the environment (the oral cavity). Pathogenic microorganisms can enter the wound, causing the body to automatically respond with inflammation and the wound-healing process. 23 Wound healing is a very complex process in the human body. It involves various types of cells and ranges from repair and arrangement of specialized structures, such as collagen, migration, proliferation, and differentiation of cells. 19,24 In this study, we report potential bioactivity that positively regulates cell proliferation (GO-ID: 8284), angiogenesis (GO-ID: 45766), and FGF production (GO-ID: 90271) by proteins targeted by an ovalbumin fragment. These three bioactivities are known to play a crucial role in the proliferative phase compared to the inflammatory phase or the maturation (remodeling) phase in the woundhealing process. In the inflammatory phase, the process of vascular vessel contraction and blood clotting is followed by an increase in the number of leukocytes in the wound tissue, including an increase in the number of neutrophils due to an increase in pro-inflammatory cytokines such as interleukin-1, tumor necrosis factor-alpha (TNF-α), and interferon-gamma and chemotactic agents such as pathogen-specific associated molecular pattern, damageassociated molecular pattern, complement, histamine, prostaglandins, and leukotrienes. In addition to neutrophils, there is also an increase in the macrophage population due  GLY32  GLU33  GLY34  ALA35  ILE84  GLN105  LEU107  ASP149  ASN154  ASP167   GLY32  GLU33  GLY34  GLY37  ILE84  LEU107  LYS114  SER153  ASN154  GLY169   GLY32  GLU33  GLY34  LYS55  ILE56  ILE84  ASP106  LEU107  LYS114  ASN154  CYS166   GLY32  GLU33  GLY34  GLY37  ILE84  LEU107  ASN154  CYS166  GLY169   ILE31  GLY32  GLU33  ILE84  LYS114  ASN154  to the chemotactic compounds tumor growth factor-beta (TGF-β) and monocyte chemoattractant protein-1. 19,24 Furthermore, in the proliferative phase, there are reepithelialization, neovascularization (angiogenesis), and immunomodulators aimed at repairing and restructuring damaged tissue. In this phase, tissue granulation occurs by fibroblasts, which play an important role in inducing the formation of a new ECM and blood vessels. Another process that is no less important is angiogenesis and vasculogenesis (neovascularization), which aims to form new blood vessels that will supply nutrients and maintain oxygen homeostasis in the healing process in injured tissues. This process is inseparable from the presence of pro-angiogenic signals such as vascular endothelial growth factor (VEGF), FGF, platelet-derived growth factor beta, TGF-β, and angiopoietins. The last involves the formation of pericytes, which play a role in microvascular stability, regulation of blood flow, and the formation of vascular protection from bacteria. 25 The last phase is maturation (remodeling), which causes contraction of the wound and replacement of type-III collagen with type-I collagen. 24 The three bioactivities are obtained from a network of 59 proteins where amyloid-β precursor protein (APP), MAPK1 (ERK2), MME, epithelial growth factor receptor (EGFR), and caspase-8 (CASP8) are important links in the network. In wound healing, APP plays an important role in the proliferation, migration, and adhesion of endothelial cells. Endothelial cells require APP as a mediator of the Scr/FAK pathway in VEGF signaling. 26 Furthermore, in regard to ERK2, upregulation, and phosphorylation of ERK1/2 and AKT are known to be consistent with increased proliferation and migration of human skin fibroblasts and human umbilical vein endothelial cells in vitro due to sea cucumber peptide treatment. 27 Other studies have also shown that the activation of the EGFR/MEK/ERK signaling pathway by the SOX2 gene is known to accelerate wound healing through the induction of keratinocyte cell migration and proliferation. 28 Meanwhile, the inhibition of corneal MME is known to improve corneal epithelial wound healing in mice. 29 Eliminating CASP8 is known to increase the proliferation and migration of human epidermal keratinocytes, which can promote wound healing in mice. 30 Molecular docking analysis results show that peptide fragments VY (F6), AAH (F1), GSIGAASM (F3) and require the lowest binding energy. Interestingly, the molecular docking results blindly show two interaction sites outside and inside the ATP phosphorylation site. Then, according to the molecular docking results of the AMP-PCP interaction with ERK2, F1 has the potential to reduce the binding energy of the AMP-PCP interaction with ERK2 in the phosphorylation domain, although the difference is not significant. The interaction of F1 outside the phosphorylation site minimizes the probability that F1 will compete with ATP. Additionally, the interaction of GLY and PVP-I on ERK2 increases the binding energy of AMP-PCP when it interacts with ERK2, although the difference is again not significant. What is interesting about the interaction between the two is the different interaction sites, where GLY interacts outside the ATP phosphorylation site and PVP-I acts inside the phosphorylation site. Based on its location and required binding energy, PVP-I has the potential to be an inhibitor of ATP phosphorylation on ERK2 when it has interacted with ERK2 first. However, the binding energy required for GLY and PVP-I to interact with ERK2 is high, so further research is needed to obtain more comprehensive conclusions.
Furthermore, PVP-I is known to have anti-inflammatory activity through suppression of TNF-expression in human neutrophil cells in vitro 31 and decreased galactosidase activity in E. coli cultures 32 so that it can reduce the level of interleukin-6, TNF-α, and rheumatoid factor in the serum of rheumatoid arthritis patients. 33 GLY is more often used as a viscous mixing agent with a heavy molecular weight, such as a Ca 2+ and OHmixer for intracanal Ca(OH) 2 medicaments. 34 Furthermore, the sodium alginate (NaAlg)-PVP-I complex is known to have a wound-healing effect by accelerating the closure process. 35 Structurally, PVP-I is composed of a polyvinyl pyrrolidone (povidone) polymer complex with elemental iodine, which is intended for health practitioners. Various studies on PVP-I have shown anti-inflammatory, anti-bacterial, anti-biofilm, anti-edema, and hemostatic activity; low toxicity; and good tolerability. Thus, it is still used as a medicament even though it has been more than a decade since its development. 5,31 Ovalbumin was found to affect the interaction of ERK2 and AMP-PCP in this study. Ovalbumin is also predicted to have the potential to be an additional medicament/component complex for better wound healing than PVP-I and GLY.
In conclusion, ovalbumin has the potential to induce cell proliferation by decreasing the binding energy required for AMP-PCP to interact with ERK2 compared to GLY and PVP-I, which slightly increase the binding energy required for AMP-PCP to interact with ERK2. Further in vitro/in vivo development and testing are needed to validate and develop ovalbumin as a pro-proliferative medicament.