key: cord-0743456-dim709i1
authors: Zhao, Hongxia; Zhang, Qinghua; Chen, Huifang; Rezanur Rahman, Md; Md Faruquee, Hossain
title: Integrated multi-omics approach identified molecular mechanism and pathogenetic processes of COVID-19 that affect patient with Parkinson’s disorder
date: 2021-08-02
journal: Saudi J Biol Sci
DOI: 10.1016/j.sjbs.2021.07.074
sha: 6f81a7396e79dce1e97e691b591541689ce96928
doc_id: 743456
cord_uid: dim709i1

The novel coronavirus named SARS-CoV-2 has emerged at the end of 2019, which causes coronavirus disease (COVID-2019). Recent case reports of SARS-CoV-2 infected individuals have revealed the onset of Parkinson's disease (PD) symptoms in patients who do not have a family history of the PD.. However, till recently, no genetic impact or mechanisms that may induce Parkinsonism inCOVID-19 patients or after COVID-19 have been found.. This study aimed to detect the commonly dysregulated genes, transcriptional regulators, and pathways between PD and SARS-CoV-2 infections. We integrated genome-wide transcriptomic datasets from PBMC samples from COVID-19 and PD and associated pathways. Our study revealed 81 upregulated and 48 downregulated differentially expressed genes (DEGs) shared between PD and SARS-CoV-2 infections. These dysregulated genes were involved in key pathways “mitochondrion structure organization”, “cell activation in immune response”, and “signalling by interleukins”. Our analysis showed RELA, TP53 and SP1 TFs that may regulate the upregulated DEGs. We have discovered key dysregulated genes and characterized the biological processes of commonly dysregulated in SARS-CoV-2 infections and PD, which could be used for the design of personalized treatment of PD following COVID-19.

At the end of 2019, a novel virus termed SARS-CoV-2 emerged in Wuhan, China. This novel virus is responsible for a respiratory illness called coronavirus disease . The SARS-CoV-2 is a member of coronaviruses, which are single-stranded RNA viruses that usually infect a wide range of vertebrate hosts [1] . The patients infected with coronaviruses present mild symptoms of cold [2] . In the last decades, three types of coronaviruses become a massive threats for people worldwide [3] . Several factors including age, comorbid diseases impact the outcome of the disease [4] .

Parkinson's disorder (PD) is a chronic neurodegenerative disease that shows neurodegeneration of neurons that secrete dopamine in the substantia nigra part of the midbrain. The typical symptoms of PD are shaking, stiffness, tremor, walking imbalance.

Various factors including genetics and environment are considered for the development and progression of PD. The influence of SARS-CoV-2 infection and its implications on PD, on the other hand, remains unclear. [5] . Many patients are concerned about the possibility of COVID-19 infection, since a few case studies have shown exacerbation of parkinsonian symptoms during infection, as well as poor treatment outcomes in PD patients. [6] [7] [8] [9] [10] [11] . Among the three patients, two responded to the treatment of dopaminergic medication [8, 12] , and the rest of one patient recovered spontaneously [7] . These three patients did not have any family history of PD, or prodromal PD [8] . Brain imaging of these patients showed the decreased function of nigrostriatal dopamine systems, which is close to PD [10] . Brundin et al. has proposed three potential mechanisms for the repaid development of parkinsonism following SARS-CoV-2 as discussed herein [13] . Firstly, vascular insults due to COVID-19 in various organs including the brain that may damage the nigrostriatal systems as seen close to vascular parkinsonism [14] . Secondly, inflammation has been considered as risk factors for PD which suggests the acute systemic infection as seen in COVID-19 may trigger neuroinflammation and death of nigral dopamine secreting neurons. Midbrain dopamine secreting neurons are vulnerable to systemic inflammation. A prior research found higher levels of IL-6 in COVID-19, while a study found a disruption in the kynurenine pathway in COVID-19 [15] . Both processes have been related to Parkinson's disease [16, 17] . This evidence supports the idea that COVID-19 infection could trigger Parkinson's disease..

Numerous studies have been published to characterize theCOVID-19 transcriptome dysregulations [18] [19] [20] [21] . However, there are no molecular processes or pathways that could explain the connections between PD and COVID-19..

In this study, by examining genome-wide gene expression data from COVID-19 and PD, we have proposed a multi-omics integration approach to uncover dysregulated transcriptomic signatures and pathogenetic mechanisms that may explain the effect of COVID-19 on the development of PD. Furthermore, dysregulated gene signature, molecular pathways, gene ontologies, and transcriptional regulators that may underlie PD in COVID19 were investigated in this study (Figure 1 ).

We have queried NCBI Gene Expression Omnibus (GEO) database to detect the suitable datasets generated from PBMCs from COVID-19 and PD patients. We retrieved two RNA-Seq datasets (GSE152418) that contained gene expression profiling of 16 PBMC samples of COVID-19 cases and 17 healthy controls [22] . We also obtained a PBMC COVID-19 dataset (Accession number: CRA002390) from the Genome Sequence Archive.

This dataset contained gene expression profiles of 3 COVID-19 cases and 3 controls [23] .

We then queried the GEO database with the keyword Parkinson's disorder peripheral blood mononuclear cells that result in three microarray transcription datasets with criteria 1) mRNA expression data ii) should contain cases and controls iii) human peripheral blood mononuclear cells (PBMCs) samples ( 

We have preprocessed the COVID-19 PBMC two datasets in the DESeq2 R package. The individual datasets were analyzed and identified differentially expressed genes (DEGs) using a statistical threshold an absolute log-fold change (logFC) >=1 and Benjamini-Hochberg corrected adjusted p-value FDR<0.05. We then performed a meta-analysis of these two datasets using Fisher's meta-analysis method implemented in the metaRNASeq R package.

The three microarray datasets of PD PBMC were processed and meta-analyzed using Network Analyst (NA) web utility tool [24] . First of all, we used two normalization procedures of datasets, namely variance stabilizing normalization (VSN) algorithm [25] , followed by quantile normalization [26] . ComBat procedure embedded into NA was also employed to adjust study batch effect in meta-analysis. The normalized datasets were used for differential gene expression analysis using R package LIMMA (linear models for microarray data) [27] . Finally, we performed a meta-analysis of these three datasets using Fisher's meta-analysis method as used for COVID-19 datasets.

We used a web utility "Metascape" for enrichment analysis [28] . The Metascape uses various sources of data including Gene Ontology, KEGG, Reactome, MSigDB for functional enrichment analysis. Metascape provides non-redundant terminology [28] . To detect significant ontologies, Metascape uses the hypergeometric test and Benjamini-Hochberg p-value correction algorithm. A Bonferroni corrected p-value <0.05 was considered for significant terms selection.

We also used Metascape to identify the transcriptional regulatory transcription factors (TFs)

for the identified DEGs [28] . Metascape uses the TRRUST database, which is a manually curated database of human and mouse transcriptional regulatory networks [29, 30] .

We have performed protein-protein interaction analysis (PPI) of the proteins, by using protein interactome data via MetaScape [28] . The modules of the PPI network was explored using MCODE plugin embedded in MetaScape [28] .

We retrieved three peripheral blood mononuclear cells (PBMCs) PD data (accession no: GSE49126, GSE100054, GSE22491), for a total of 87 samples that included 50 PD cases and 37 healthy controls. The three datasets were processed and we performed a meta-analysis of three PBMCs datasets using Fisher's method. Our meta-analysis detected 2186 differentially expressed genes (DEGs), where 933 upregulated and 1253 downregulated DEGs (FDR<0.05) ( Table 2 and Table S1 ).

Two PBMCs COVID-19 RNA-Seq transcriptomic datasets (accession no: GSE152418 and CRA002390) contained a total of 39 samples, 19 cases, and 20 healthy control. These two datasets were processed and we performed a meta-analysis to identify DEGs using the same Fisher's methods. Our meta-analysis of PBMC COVID-19 showed 3272 significant (FDR<0.05) genes ( Table 3 and Table S2 ).

In order to identify concordant overlapped DEGs between two pathologies (Table S3) namely, PD and COVID-19, we compared the DEGs. Our analysis identified a total 129 (Figure 2 ).

To illustrate the biology of the overlapping DEGs found, we conducted an enrichment analysis of genes. In this way, we identified several ontologies and pathways that overlapped between PD and COVID-19 (Figure 3B-4) . Our analysis showed the common gene ontologies of the upregulated DEGs between PD and COVID-19 "mitochondrion organization", "cellular protein catabolism", "cofactor metabolic process", "cellular component disassembly", "apoptotic signalling pathways", "positive regulation of organelle organization ", "nervous system development", "negative regulation of cellular component organization", "organelle localization" (Figure 3-4) . Among the significant ontologies enriched by downregulated DEGs were "Cell activation in immune systems", "Signalling by interleukins", "Metabolism of RNA", "Positive regulation of cell death" (Figure 3-4) .

To provide putative transcriptional regulators of the common modulated DEGs between PD and COVID-19, we identified that RELA, TP53, and SP1 were involved in the expression of commonly upregulated genes in COVID-19 and PD. However, we did not obtain any significant TFs that may be involved in the regulation of DEGs which were downregulated DEGs ( Figure 5 ).

To dismantle the molecular interactions of the proteins encoded by the DEGs which were commonly modulated in COVID-19 and PD, we investigated protein-protein interaction (PPI) analysis. Our analysis showed the presence of four modules in the PPI networks ( Figure 6) . The modules were found involved in various infection and metabolism associated pathways (Figure 6B,C) .

Comorbidities are considered as immense risk factors for increased vulnerability to COVID-19 and poor therapeutic outcome. Among the comorbidities, neurological disorders came into a greater risk for PD since some case reports showed the development of PD following COVID-19. The neurological symptoms in COVID-19 patients appear to be severe to mild. Despite the three case reports described the development of PD after COVID-19 presenting a great threat to the massive development of PD in response to COVID-19 [9] [10] [11] .

The cellular and mechanism that may lead to COVID-19 induced PD are not understood yet. Therefore, characterizing critical genes and pathogenetic pathways in COVID-19 and PD is very important to understand the possible links of development of PD following COVID-19. In this study, for the first time, we identified commonly modulated common genes and molecular processes altered between PD and COVID-19 in PBMC. To this aim, we utilized whole-genome PBMC transcriptomic datasets of COVID-19 and PD via systems biology analysis. Our analysis detected 81 upregulated and 48 downregulated significant DEGs were common between two pathologies suggesting a possible molecular link between COVID-19 and PD. The upregulated genes were involved in several biological processes including "mitochondrion structure organization" among others. There is substantial evidence has been presented that suggests the mitochondrial dysfunction is involved in the aetiology and pathogenesis of PD since the inhibitor of electron transport chain complex I cause PD [31] [32] [33] . The mitochondria are involved in crucial cellular processes including energy metabolism, calcium homeostasis, and program cell death. Disruption of mitochondria is involved in the neuronal cell death of substantia nigra of PD, substantia nigra is the part of the brain mainly involved in PD development. In line with this evidence, our study suggests impairment of mitochondrial structure organization pathways might be involved in PD and COVID-19.

Our study demonstrated that the downregulated common genes are predominantly involved in immune systems related processes namely, "cell activation in immune response", "Signalling by interleukins". There is much compelling evidence that suggests the increased expression of genes in central nervous systems and peripheral nervous systems, infiltration of immune cells into CNS, the altered composition of immune cells in peripheral nervous systems, etc are implicated in the development of PD as discussed herein [34, 35] . Moreover, various attempts have been taken to decipher inflammatory molecules, cells involved in the development of PD. Investigation of post-mortem brain and cerebrospinal fluids of PD showed the presence of inflammatory cytokines suggesting neuroinflammation in affected brain regions of PD [36] . Alterations of various interleukins including IL-6 have been suggested as a marker of increased mortality in PD [37] . In line with this evidence, we hypothesized that alteration of signalling by interleukins pathways is involved in COVID-19 induced PD. Increasing evidence also has been suggested the impaired activation of immune systems, specifically the "cytokine storm" found in COVID-19 results from uncontrolled host immune response that is considered to be associated with COVID-19 complications, which supports that systemic immune systems alterations as observed in our study is involved with COVID-19 [38, 39] .

Finally, we identified TFs (RELA, TP53, and SP1) that may regulate the common DEGs between COVID-19 and PD. Our results identified RELA TFs, a major component of NF-kB, which are considered as the master regulators of inflammation and apoptosis [40, 41] .

The NF-kB/RELA is involved in neurodegenerative diseases including PD [41] . Many lines of evidence showed the crucial role of RELA/NF-kB in PD [41] . Previous studies showed the dysregulated increased expression of RELA in nigral dopaminergic neurons and glial cells in PD patients [41, 42] . Inhibition of RELA prevents loss of dopaminergic neurons and downregulated expression of RELA provides neuroprotection in PD suggesting a critical role of RELA in neurodegeneration in PD [42] . We also identified Tp53, which is associated with cell death pathways, and its activation is involved in the pathogenesis of PD [43] . It is important to note that our study has some limitation that findings were based on bioinformatics analysis and experimental validations should be done to establish solid conclusions regarding the associations between COVID-19 and PD. We should now like to propose some wet-lab experiments e.g., quantitative polymerase chain reaction, immunohistochemistry, western-blotting to validate the identified hub biomolecules

This study was aimed to identify dysregulated common transcriptional signature and molecular pathways between PD and COVID-19. Our integrative analysis suggested 81 upregulated and 48 downregulated DEGs were concordantly common between PD and COVID-19. Our study highlighted the major biological pathways altered in PD and COVID-19 involved in "mitochondrion structure organization" and immune systems related processes namely, "cell activation in immune response", and "Signalling by interleukins".

We also detected key TFs namely RELA, TP53 which were involved in the expression of common DEGs. The detected pathways and genes in this study could be therapeutic targets in PD patients with COVID-19. Considering crucial importance of presented hub genes and pathways shared between COVID-19 and PD, experimental assessment of hub genes and pathways in clinical samples in COVID-19 and PD are proposed to further clarify the shared mechanism.

Funding: "This research received no external funding". Supplementary Materials: Table S1 . Differentially expressed genes identified by a meta-analysis of PBMC Parkinson's disorder. Table S2 . Differentially expressed genes identified by a meta-analysis of PBMC COVID-19. Table S3 . List of commonly modulated upregulated and downregulated differentially expressed genes between COVID-19 and PD (FDR<0.05).

Origin and evolution of pathogenic coronaviruses

Coronavirus pathogenesis and the emerging pathogen severe acute respiratory syndrome coronavirus. Microbiol

novel coronavirus patients' clinical characteristics, discharge rate and fatality rate of meta-analysis

Self-reported needs of patients with Parkinson's disease during COVID-19 emergency in Italy

Outcome of Parkinson's Disease patients affected by COVID-19

Effects of COVID-19 on Parkinson's disease clinical features: a community-based case-control study

Rapid worsening in Parkinson's disease may hide COVID-19 infection

Rodríguez-Montalbán, R. Acute hypokinetic-rigid syndrome following SARS-CoV-2 infection

A case of probable Parkinson's disease after SARS-CoV-2 infection

Coronavirus disease 2019 and parkinsonism: a non-post-encephalitic case

Social isolation: a predictor of mortality comparable to traditional clinical risk factors

Is COVID-19 a perfect storm for Parkinson's disease? Trends Neurosci

Brain ischemic injury in CoViD19 infected patients: a series of 10 post-mortem cases

COVID-19 infection alters kynurenine and fatty acid metabolism, correlating with IL-6 levels and renal status

Brundin, P. Triggers, facilitators, and aggravators: Redefining Parkinson's disease pathogenesis

Tryptophan Metabolites Are Associated With Symptoms and Nigral Pathology in Parkinson's Disease

SARS-CoV-2 launches a unique transcriptional signature from in vitro, ex vivo, and in vivo systems

Integrative transcriptomics analysis of lung epithelial cells and identification of repurposable drug candidates for COVID-19

In vivo antiviral host transcriptional response to SARS-CoV-2 by viral load, sex, and age

Transcriptional landscape of SARS-CoV-2 infection dismantles pathogenic pathways activated by the virus, proposes unique sex-specific differences and predicts tailored therapeutic strategies

Systems biological assessment of immunity to mild versus severe COVID-19 infection in humans

Transcriptomic characteristics of bronchoalveolar lavage fluid and peripheral blood mononuclear cells in COVID-19 patients

0: a visual analytics platform for comprehensive gene expression profiling and meta-analysis

Normalizing and variance stabilizing transformations for intraclass correlations

Removing technical variability in RNA-seq data using conditional quantile normalization

limma powers differential expression analyses for RNA-sequencing and microarray studies

Metascape provides a biologist-oriented resource for the analysis of systems-level datasets

TRRUST: A reference database of human transcriptional regulatory interactions

TRRUST v2: An expanded reference database of human and mouse transcriptional regulatory interactions

Mitochondrial dysfunction in Parkinson's disease

Mitochondrial biology and Parkinson's disease. Cold Spring Harb. Perspect

Mitochondrial dysfunction in Parkinson's disease. Park

The role of innate and adaptive immunity in Parkinson's disease

Immune system responses in Parkinson's disease: Early and dynamic

The potential role of neuroinflammation and transcription factors in Parkinson disease

Interleukin-6 may contribute to mortality in Parkinson's disease patients: a 4-year prospective study

Immune dysfunction in patients with diabetes mellitus (DM)

Immune response in COVID-19: addressing a pharmacological challenge by targeting pathways triggered by SARS-CoV-2

Nuclear Factor-κB Dysregulation and α-Synuclein Pathology: Critical Interplay in the Pathogenesis of Parkinson's Disease. Front

NF-κB in innate neuroprotection and age-related neurodegenerative diseases

Selective inhibition of NF-κB activation prevents dopaminergic neuronal loss in a mouse model of Parkinson's disease

Tp53 gene mediates distinct dopaminergic neuronal damage in different dopaminergic neurotoxicant models

Data Availability Statement: All data utilized in this manuscript are available online from their respective databases. GSE152418 and GSE49126 were downloaded from NCBI Gene Expression Omnibus (NCBI-GEO) available at https://www.ncbi.nlm.nih.gov/geo/ and CRA002390 from the Genome Sequence Archive database (https://ngdc.cncb.ac.cn/gsa/). Acknowledgments: None. Conflicts of Interest: "The authors declare no conflict of interest

"The authors declare no conflict of interest." Integrated multi-omics approach identified molecular mechanism and pathogenetic processes of COVID-19 that affect patient with Parkinson's disorder