key: cord-0846300-feff1hn6 authors: Morsy, Sara; Morsy, Ahmed title: Epitope mimicry analysis of SARS-COV-2 surface proteins and human lung proteins date: 2021-02-04 journal: J Mol Graph Model DOI: 10.1016/j.jmgm.2021.107836 sha: 99e5ec8b237c3b9e3830efd33fe60e31e6463fdf doc_id: 846300 cord_uid: feff1hn6 BACKGROUND: Autoimmune response after the infection of SARS-COV-2 is evident as more cases of Guillain Barre syndrome and Kawasaki disease are diagnosed after infection. In this study, we aim to investigate a possible mechanism of autoimmune lung injury. METHODS: We extracted the peptide sequence surface proteins of the SARS-COV-2 from NCBI data protein. We used Blastp to assess the homologous sequences between the human protein in the UNIPROT database that is associated with respiratory distress. Then, we filtered the homologous sequences to those selectively expressed in the lung and homologous to surface viral proteins. We then assessed the epitope sequences for MHC-I, MHC-II and B-cell epitope using recommended settings and reference MHC in the IEDB database. RESULTS: Homeobox protein 2.1 (NKX2-1) and ATP-binding cassette sub-family A member 3 (ABCA3) showed homologous sequence to both surface glycoproteins and envelope proteins. The HLA-DR and HLA-DQ had a similar binding pattern to ABCA3 as surface glycoproteins and envelope proteins, respectively. Other HLA molecules that had a similar binding pattern to SARS-COV-2 as human proteins were HLA-A and HLA-DP. CONCLUSION: Our study indicates that there is a possible autoimmune mechanism underlying the acute respiratory distress syndrome in SARS-COV-2. Autoimmune response after the infection of SARS-COV-2 is evident as more cases of Guillain Barre syndrome and Kawasaki disease are diagnosed after infection. In this study, we aim to investigate a possible mechanism of autoimmune lung injury. We extracted the peptide sequence surface proteins of the SARS-COV-2 from NCBI data protein. We used Blastp to assess the homologous sequences between the human protein in the UNIPROT database that is associated with respiratory distress. Then, we filtered the homologous sequences to those selectively expressed in the lung and homologous to surface viral proteins. We then assessed the epitope sequences for MHC-I, MHC-II and B-cell epitope using recommended settings and reference MHC in the IEDB database. Homeobox protein 2.1 (NKX2-1) and ATP-binding cassette sub-family A member 3 (ABCA3) showed homologous sequence to both surface glycoproteins and envelope proteins. The HLA-DR and HLA-DQ had a similar binding pattern to ABCA3 as surface glycoproteins and envelope proteins, respectively. Other HLA molecules that had a similar binding pattern to SARS-COV-2 as human proteins were HLA-A and HLA-DP. Our study indicates that there is a possible autoimmune mechanism underlying the acute respiratory distress syndrome in SARS-COV-2. The recent coronavirus pandemic (SARS-COV-2) has caused worldwide distress and confusion over its pathogenesis especially that it had a significant difference to its precedent coronaviruses, MERS and SARS [1] . There is a lack of evidence regarding how the immune system reacts to SARS-COV-2. Once any virus enters the body, the antigenic peptides are presented on MHC I and MHC II which T cells recognize to initiate the immune response to the virus [2] . Based on SARS and MERS studies, the HLA polymorphism increase the susceptibility to the disease and its complications [3] . For instance, HLA-B*4601, HLA-B*0703, HLA-DR B1*1202, and HLA-Cw*0801 increased the susceptibility whereas the HLA-DR0301, HLA-Cw1502 and HLA-A*0201 alleles decreased the risk of infection [3] [4] [5] . For humoral response, IgG and IgM were detected in the patients of SARS-COV-2 but IgM disappeared after 12 weeks [1] . For cellular immunity, there was decreased CD4 + and CD8 + which persisted even with increased levels of HLA-DR and CD-38 [2, 5] . This is not the first time SARS-COV-2 was associated with different autoimmune diseases e.g. Guillain Barre syndrome and Kawasaki disease. In Northern Italy, patients developed Guillain-Barré syndrome after 5 -10 days after infection [6] . Many cases were diagnosed as Guillain Barre syndrome in China, United states, and Iran [7] [8] [9] . For Kawasaki disease, many countries reported increased incidence of the disease after the COVID-19 infection [10] . Although the immunopathogenesis of the autoimmune diseases is still vague, it provides an insight toward a possible mechanism for COVID-19 immunopathogenesis. In-silico approach can be used to determine a possible autoimmune mechanism through molecular mimicry approach [11] . J o u r n a l P r e -p r o o f Molecular mimicry has been investigated in different viral and bacterial pathogens as a cause of different clinical diseases [11] . It is the structural similarity between the host proteins and the pathogens causing cross-reacting antibodies acts against the host proteins. This was proved in many diseases that appeared during or after the infection [11] [12] [13] . In the current study, we hypothesize that the SARS-COV-2 induces an autoimmune response that may be fulminant response causing multiorgan failure. We investigated the molecular mimicry between human proteins and SARS-COV-2 proteins and investigated whether the homologous peptides will create a cross-reacting antibody against these human proteins. In this study, we extracted SARS-COV-2 surface proteins sequence from the NCBI virus database (Ref Seq ID: NC_045512) [14] . The extracted surface protein sequences were membrane glycoproteins, envelope proteins and surface glycoprotein. In addition, we performed a systematic search in PubMed, Scopus, ISI web of science and Google Scholar using search terms ("COVID-19" OR "SARS-COV-2" OR "2019-nCoV" OR "Coronavirus Disease-19") AND ("autoimmune" OR "Autoantibodies" OR "Autoantibody"). We included any reported proteins that cause autoimmune response in COVID-19 patients. We extracted the proteins involved in respiratory distress using the Uniprot database by using search terms "Respiratory distress" and "Respiratory failure" [15] . We refined our research to the reviewed proteins and those expressed in Homo sapiens. J o u r n a l P r e -p r o o f The Blastp program (https://blast.ncbi.nlm.nih.gov/Blast.cgi) was used to find homologous sequences between the identified human proteins to SARS-COV-2 proteins [16, 17] . The search was limited to Homo sapiens (taxid: 9606) and those in the UniProtKB/Swiss-Prot database. We used the default BLASTp algorithm parameters [17] . After carefully revising the homologous sequence, homologous proteins were filtered through the Human Protein Atlas (http://www.proteinatlas.org) to only keep the proteins selectively expressed in the lung [18] . The homologous sequences were investigated for the ability to function as T cell epitopes. First, the homologous viral proteins and human proteins were analyzed using the IEDB analysis tools (http://www.iedb.org) to detect both MHC-I and MHC-II epitopes [19] . Then, we checked if the homologous sequences detected by the BLASTp can function as epitope for T cells or not. We used the IEDB recommended models and parameters. For HLA selection, the HLA allele reference set was selected for both MHC-I and MHC-II [19] . We filtered the binding epitopes based on those who exhibited a low percentile rank, which indicated a good binding capacity. We then selected the homologous peptides that had a low percentile rank and high homology to human proteins. Peptide-MHC docking was performed using GalaxyPepDock server (http://galaxy.seoklab.org/) to compare the binding patterns of MHC molecules to corresponding homologous sequences [20] . We chose this server because it allows the docking of the peptide to a 3D protein structure. Unlike other services, it does not need the 3D structure of the peptide. So, the docking will not depend on the prediction error of homologous peptide structure. The docking of the Galaxy J o u r n a l P r e -p r o o f pepdock depends on the searching a similar protein peptide interaction in the database of experimentally determined structures. If similar peptide protein interactions are found, this template is used to refine the docking. The server reports the top 10 possible docking conformations based on the similarity to the template. Unfortunately, no docking scores were reported. In addition, based on the interaction similarity between the template and predicted docked model, GalaxyPepDock determines the important amino acids residues that are essential for the binding of HLA molecule to different epitopes. To compare between the binding of SARS-COV-2 and human peptides to HLA of the predicted docked models, the docked models were analyzed using UCSF Chimera 1.14 [21] and LigPLOT+ [22] . We used Root Mean Square Deviation (RMSD) to compare the affinity of binding of SARS-COV-2 and human peptides to HLA. In simple words, RMSD is considered an expression of the distance between the docked peptide and HLA molecule; the less the RMSD, the stronger the docking. It is used to compare the docked conformation to a reference model. In this study, we considered the human HLA interaction as the reference model and we compared the binding of SARS-COV-2 peptide to HLA molecule to the human model. [23] . These crystal structures were modified to remove any bound ligands and make the groove available to dock with SARS-COV-2 and human proteins using UCSF-Chimera 1.14 [21] . The resulting MHC-I and MHC-II epitopes of SARS-COV-2 for both the envelope and surface glycoproteins were ranked based on the percentile rank (the lower the percentile, the greater the binding) and epitopes that exhibited the highest homology to the homologous human proteins' sequences were selected. Most selected epitopes were among the top five epitopes based on the percentile ranking. For NKX2-1, there were three homologous peptides with surface glycoprotein epitopes that showed high binding with MHC-I and MHC-II (Table 1) . For ABCA3, there were two homologous peptides with surface glycoprotein and envelope proteins epitopes that showed high binding with MHC-I and MHC-II (Table 1 ). Figure 1D ). For HLA-A, the RMSD between the docked homologous human and SARS-COV-2 protein was 1.34A o . Both peptides were found to bind to asparagine 66, glutamine 70, tyrosine 99, and aspartate 77 residues in HLA-DA using hydrogen bonds ( Figure 5 ). For HLA-DQ, the binding of the SARS-COV-2 peptide was similar to the corresponding homologous protein, ABCA3, with an RMSD of 0.72 A o . Both homologous peptides shared hydrogen bonds between leucine 362 in J o u r n a l P r e -p r o o f the peptides and asparagine 60 in the HLA-DQ molecules. In addition, they shared hydrogen binding to arginine 74 and asparagine 258 in the HLA-DQ molecules (Figure 2 ). This study suggests that the molecular mimicry between SARS-COV-2, NKX2-1 and ABCA3 may explain the immunopathogenesis of the disease. Our results indicate that the SARS-COV-2 envelope proteins and surface glycoproteins share antigenic MHC-I and MHC-II epitopes with NKX2-1 and ABCA3. The most interesting result was the homologous peptides between ABCA3 and both envelope proteins and surface glycoproteins. ABCA3 is a protein consisted of 1704 residues and is selectively expressed in the lung [24, 25] . It belongs to the family of ABC transporter that is responsible for the transport of molecules through the lipid bilayer of the lung [24] . It was detected in the plasma membrane of alveolar cells and the limiting membrane of the lamellar bodies [25, 26] . The alveolar cells II are responsible for the secretion of surfactant which is J o u r n a l P r e -p r o o f important for protection against lung collapse [24, 27] . That is why mutations in ABCA3 gene caused respiratory distress and interstitial pneumonia [28] . In this study, the homologous peptides between the ABCA3 and both envelope proteins and surface glycoproteins peptides acted also as antigenic epitopes for HLA-A*02:03, HLA-DQA1*05:01/DQB1*03:01, HLA-DPA1*03:01/DPB1*04:02, HLA-A*02:01. The binding between the molecules was similar in pattern and hydrogen bonds. We believe that the antibodies produced against surface proteins cross-react against ABCA3 proteins causing alveolar collapse and respiratory failure based on other studies that found that functional defect in the protein resulted in respiratory failure [29] . Furthermore, the specific HLA that cross-reacts with these epitopes might explain why specific populations are more susceptible to respiratory failure and death while other patients are symptomless. That was observed in different viral infections where specific HLA polymorphisms were associated with the severity of the disease [30] [31] [32] . In addition, a study revealed that Taiwanese general population with HLA-B* 4601 had more susceptibility to ARDS after SARS infection compared to Taiwanese indigenous population who do not have this HLA allele [3] . Another small study found that SARS patients admitted to ICU have HLA-B46 compared to those who have not [33] . However, there are still no clinical studies in SARS-COV-2 that support our study. Based on our study, the binding strength of SARS-COV-2 to HLA is more powerful than human epitopes based on RMSD values. In addition, based on the docking models, there were more hydrogen bonds stabilizing the SARS-COV-2 peptide than human peptides. We also recommend that further clinical trial should assess the effect of drugs potentiating the effect of ABCA3 like the cystic fibrosis transmembrane conductance regulator e.g. ivacaftor and genistein [34] . Another interesting result is the homologous peptides between NKX2-1 and surface glycoproteins. NK2 homeobox 1 is a transcription factor that is expressed in the early development of the lung and thyroid gland [35] . It is also involved in the synthesis of the surfactant proteins A, B, and C [35] . Furthermore, it was found that it regulates the expression of ABCA3 proteins, thus, NKX2-1 is required for the normal structure of surfactant which is important for the alveolar function and lowering the alveoli surface tension [27] . Furthermore, the surfactant is considered as the innate immunity in the lung as well as antibacterial properties [27] . The NKX2-1 defects have a wide variety of respiratory symptoms including respiratory distress, pneumonia, asthma, recurrent pneumothorax, lung fibrosis, or respiratory insufficiency [36, 37] . These symptoms are similar to those developed by SARS-COV-2 that was reported by a recent Chinese cohort [38] . The symptoms may be evident since birth or appear at older age [36] . In the current study, we found that surface glycoproteins share antigenic epitopes with SARS-COV-2 causing an autoimmune response that may cause respiratory symptoms similar to mutation of NKX2-1. The antigenic epitopes were linked to HLA-A*26:01, HLA-DQA1*05:01/DQB1*03:01, HLA-A*68:01, HLA-DRB1*04:01. Again, this might explain why specific populations get severe symptoms. The findings in this study must be supported through further experimental and clinical studies. The envelope and surface glycoproteins have high similarity to ABCA3 protein which is J o u r n a l P r e -p r o o f J o u r n a l P r e -p r o o f Epidemiology, causes, clinical manifestation and diagnosis, prevention and control of coronavirus disease (COVID-19) during the early outbreak period: a scoping review Immune responses in COVID-19 and potential vaccines: Lessons learned from SARS and MERS epidemic Association of HLA class I with severe acute respiratory syndrome coronavirus infection Association of Human-Leukocyte-Antigen Class I (B*0703) and Class II (DRB1*0301) Genotypes with Susceptibility and Resistance to the Development of Severe Acute Respiratory Syndrome Molecular immune pathogenesis and diagnosis of COVID-19 Guillain-Barré Syndrome Associated with SARS-CoV-2 Guillain-Barré syndrome associated with SARS-CoV-2 infection: causality or coincidence? The Lancet Neurology Guillain-Barré Syndrome associated with SARS-CoV-2 infection Guillain Barre syndrome associated with COVID-19 infection: A case report COVID-19 and Kawasaki Disease: Novel Virus and Novel Case. Hospital pediatrics Morbid Sequences Suggest Molecular Mimicry between Microbial Peptides and Self-Antigens: A Possibility of Inciting Autoimmunity A possible role for autoimmunity through molecular mimicry in alphavirus mediated arthritis Potential molecular mimicry between the human endogenous retrovirus W family envelope proteins and myelin proteins in multiple sclerosis Virus Variation Resource -improved response to emergent viral outbreaks UniProt: a worldwide hub of protein knowledge Gapped BLAST and PSI-BLAST: a new generation of protein database search programs Protein database searches using compositionally adjusted substitution matrices. The FEBS journal Tissuebased map of the human proteome The Immune Epitope Database (IEDB): 2018 update GalaxyPepDock: a protein-peptide docking tool based on interaction similarity and energy optimization UCSF Chimera--a visualization system for exploratory research and analysis LigPlot+: multiple ligand-protein interaction diagrams for drug discovery The Protein Data Bank ABCA3 as a lipid transporter in pulmonary surfactant biogenesis. The Journal of biological chemistry ABCA3 is a lamellar body membrane protein in human lung alveolar type II cells Identification of LBM180, a lamellar body limiting membrane protein of alveolar type II cells, as the ABC transporter protein ABCA3 Pulmonary surfactant: a front line of lung host defense ABCA3 mutations associated with pediatric interstitial lung disease. American journal of respiratory and critical care medicine Functional and trafficking defects in ATP binding cassette A3 mutants associated with respiratory distress syndrome The Genetic Polymorphisms of HLA Are Strongly Correlated with the Disease Severity after Hantaan Virus Infection in the Chinese Han Population Multicentre factorial randomized clinical trial of perioperative immunonutrition versus standard nutrition for patients undergoing surgical resection of oesophageal cancer HLA-E Polymorphism Determines Susceptibility to BK Virus Nephropathy after Living-Donor Kidney Transplant. Cells Immunogenetics in SARS: a case-control study. Hong Kong medical journal = Xianggang yi xue za zhi Potentiation of ABCA3 lipid transport function by ivacaftor and genistein Thyroid transcription factor-1. The international journal of biochemistry & cell biology Heterogeneous pulmonary phenotypes associated with mutations in the thyroid transcription factor gene NKX2-1 Interstitial lung disease of infancy caused by a new NKX2-1 mutation Characteristics of and Important Lessons From the Coronavirus Disease 2019 (COVID-19) Outbreak in China: Summary of a Report of 72314 Cases From the Chinese Center for Disease Control and Prevention