key: cord-0874290-sdmtlgwe
authors: Ramana, Chilakamarti V.
title: Profiling transcription factor sub-networks in type I interferon signaling and in response to SARS-CoV-2 infection
date: 2021-01-25
journal: bioRxiv
DOI: 10.1101/2021.01.25.428122
sha: 3179dc46311b06a9a4f90589244f0ea570df596f
doc_id: 874290
cord_uid: sdmtlgwe

Type I interferons (IFN α/β) play a central role in innate immunity to respiratory viruses, including coronaviruses. Genetic defects in type I interferon signaling were reported in a significant proportion of critically ill CoOVID-19 patients. Extensive studies on interferon-induced intracellular signal transduction pathways led to the elucidation of the Jak-Stat pathway. Furthermore, advances in gene expression profiling by microarrays have revealed that type I interferon rapidly induced multiple transcription factor mRNA levels. In this study, transcription factor profiling in the transcriptome was used to gain novel insights into the role of inducible transcription factors in response to type I interferon signaling in immune cells and in lung epithelial cells after SARS-CoV-2 infection. Modeling the interferon-inducible transcription factor mRNA data in terms of distinct sub-networks based on biological functions such as antiviral response, immune modulation, and cell growth revealed enrichment of specific transcription factors in mouse and human immune cells. The evolutionarily conserved core type I interferon gene expression consists of the inducible transcriptional factor mRNA of the antiviral response sub-network and enriched in granulocytes. Analysis of the type I interferon-inducible transcription factor sub-networks as distinct protein-protein interaction pathways revealed insights into the role of critical hubs in signaling. Interrogation of multiple microarray datasets revealed that SARS-CoV-2 induced high levels of IFN-beta and interferon-inducible transcription factor mRNA in human lung epithelial cells. Transcription factor mRNA of the three major sub-networks regulating antiviral, immune modulation, and cell growth were differentially regulated in human lung epithelial cell lines after SARS-CoV-2 infection and in the tissue samples of COVID-19 patients. A subset of type I interferon-inducible transcription factors and inflammatory mediators were specifically enriched in the lungs and neutrophils of Covid-19 patients. The emerging complex picture of type I IFN transcriptional regulation consists of a rapid transcriptional switch mediated by the Jak-Stat cascade and a graded output of the inducible transcription factor activation that enables temporal regulation of gene expression.

Interferons (IFN) are pleiotropic cytokines and exert a wide range of biological activities that include antiviral, antiproliferative, and immunoregulatory effects (1) (2) (3) . There are 2 major classes of type I interferons consisting of IFN-alpha (IFN-α) represented by 14 isoforms and one IFN-beta (IFN-β).

Type I IFN is produced by many cell types, including leukocytes, dendritic cells, and fibroblasts. The amount of IFN-α vs IFN-β produced varies depending on cell type and also on the virus/stimulus (1) (2) (3) (4) . The biological effects of interferons are mainly mediated by the rapid and dramatic changes in gene expression of several hundred genes (5) . The role of the Janus kinase and signal transducers and activators of transcription (Jak-Stat) pathway and transcriptional regulation by interferon-stimulated gene complex (ISGF-3) consisting of Stat1, Stat2 and, Irf-9 in Type I IFN signaling has been well established (1) (2) (3) (4) (5) (6) . Recent advances in high throughput gene expression profiling in primary immune cells have shown that IFN α/β induces several transcription factor mRNA that sustains secondary and tertiary rounds of transcription (7, 8) . Several studies had shown that the activation of the canonical Jak-Stat pathway is not sufficient to explain the broad range of biological activities of type I IFN signaling. IFN α/β inhibits the interleukin-7-mediated growth and survival of B lymphoid progenitors by a Stat1-independent pathway (9) . This growth inhibitory effect involves cell cycle arrest followed by apoptosis and mediated by Stat1-independent induction of Daxx, a Fas binding protein implicated in apoptosis (10) . The most prominent phenotype of the Stat1-deficient mice is their increased susceptibility to microbial and viral infection due to the decreased ability to respond to the antiviral effects of the interferons (11, 12) . Nevertheless, Stat1-deficient mice mount an IFNmediated resistance to virus infection. Stat1-null mice are more resistant to challenge with murine cytomegalovirus (MCMV) or Sindbis virus than mice lacking both the type I and type II IFN receptors (13) . There is evidence for differential regulation of type I interferon signaling and interferon regulatory factors in neuronal cells and astrocytes of wild-type and Stat1-deficient mice in the mouse brain (14, 15) . IFN-α/β regulates dendritic cells (DC) and natural killer (NK) cells that are resident immune cells of the lung and are critical for innate immunity to respiratory viruses (16, 17) .

The availability of gene expression datasets of primary immune cells treated with Interferon α/β for a short period of time facilitated profiling type I interferon-inducible transcription factor mRNA in the transcriptome (7.8) . Coronaviruses are RNA viruses of the Coronaviridae family, including Severe Acute Respiratory Syndrome coronaviruses SARS-CoV-1 and SARS CoV-2 (18) . The current pandemic of coronavirus disease known as COVID-19 is caused by a highly infectious SARS-CoV-2 that emerged first in Wuhan, China (19, 20) . The world health organization (WHO) has declared the current pandemic as a global public health emergency because of the rapid spread around the world with high mortality and morbidity. The importance of a functional interferon system for protection against SARS-CoV-2 was demonstrated by the report that 10% of nearly a thousand Covid-19

patients who developed fatal pneumonia had autoantibodies to interferons and an additional 3-5% of critically ill patients had mutations in genes that control interferons (21, 22) . In addition to the genetic deficiency, genetic variants in the components of type I Interferon signaling such as type I IFN receptor (IFNAR2), protein tyrosine kinase (TYK2) and the antiviral target gene Oligoadenylate synthetase (OAS1) were detected in critically ill COVID19 patients (23) . Transcription factor profiling is a powerful technique to gain insights into mammalian signal transduction pathways. (24, 25) . Transcription factor profiling of genes involved in a biologic function such as innate and adaptive immunity led to the identification of functional connectivity of signaling hubs, critical nodes, and modules (25, 26) . In this study transcription factor profiling in the transcriptome of immune cells in response to type I Interferon treatment and lung epithelial cells in response to SARS-

Gene expression datasets Gene expression profiling in response to Type I interferon treatment in human peripheral blood mononuclear cells (PBMC) and mouse immune cells were published previously (7, 8) . Gene expression datasets representing human lung cell lines infected with coronaviruses and tissue samples of healthy and COVID-19 patients were also reported (27, 28) .

Supplementary data was downloaded from the Journal publisher websites and from Geo datasets that were archived at Pubmed (NCBI). The identification of the differentially expressed genes in the transcription profile was analyzed using the GEO2R tool and differential expression analysis using (30) (31) (32) . These transcription factors were broadly organized into three major functional categories-antiviral response, immune modulation, and cell growth (Table1). Mapping the list of interferon-induced transcription factors into gene ontogeny (GO) terms in Metascape provided insights into the common functional categories and signal transduction pathways in the human and mouse transcriptomes ( Figure 1A and 1B). These GO terms included human papillomavirus infection, Hepatitis B, HTLV-1, response to the virus, interferon-gamma mediated signaling pathway in the antiviral response category. Immune modulation or inflammation category GO terms included regulation of cytokine production, Toll-like receptor signaling pathway, and AP1 pathway. In contrast, cell growth terms included transcriptional mis-regulation in cancer, positive regulation of cell death, and DNA damage response ( Figure 1A and 1B). Protein-protein interaction abundance analysis using molecular complex detection (MCODE) algorithm in Metascape revealed three modules consisting of cell proliferation/Ionizing radiation, interferon-γ and interferon-α/β signaling represented by red, blue and green, respectively ( Figure 1C ). The human type I interferon-induced transcription factor data was obtained from gene expression profiling of a heterogeneous mixture of cells from the PBMC while the mouse list was derived from a purified population of B-lymphocytes, dendritic cells, granulocytes, natural killer cells, macrophages, and T-lymphocytes. Cluster analysis in gene expression profiling is often used to discriminate genes that are co-regulated under the given experimental conditions (29) . The grouping of transcription factors into functional sub-categories facilitated the identification of enrichment in distinct cell types. For example, mouse antiviral transcription factors were highly expressed in granulocytes among immune cells (Figure 2A ).

Granulocytes including neutrophils, eosinophils, and basophils are characterized by the presence of large cytoplasmic granules and essential for the control of infection. Enhanced expression of transcription factors involved in inflammation such as Ahr, Bcl3. and Egr2 in NK cells and cell growth such as Myc, Max, and Jun in B-cells was observed ( Figure 2B and Figure 2C ). Furthermore, IFN-α and IFN-β mediated changes in distinct functional sub-categories such as antiviral response, immune modulation and cell growth of human PBMC can be compared in detail ( Figure 3 ).

Most of the gene expression profiling studies in response to type I IFN have some technical to mycobacteria but not the viral disease. Interferon-induced Stat1 homodimer or gamma-activated factor (GAF) formation was diminished, while the response to the ISGF-3 complex was normal in cells derived from these patients (37) . The clinical and cellular phenotypes of these patients suggest that the response to mycobacteria was mediated by Stat1 homo-dimer and the antiviral immunity is dependent on ISGF-3 formation. Interestingly, transcription factors and growth-related genes such as Myc, Jun, and Schlafen (Sfn) family members were rapidly induced by type I Interferon in B cells ( Figure 4B ). Schlafens regulate immune cell proliferation, differentiation, and restricting virus replication (38) . However, Schlafens has no DNA binding activity or direct role in transcriptional regulation.

Transcriptional factors bind to regulatory elements such as promoters and enhancers located upstream p300/CBP-associated factor (p/CAF) for distinct platform assembly and histone acetyltransferase (HAT) activity to activate interferon-stimulated gene transcription (41) . Two contact regions between Stat1 N-terminal and C-terminal regions and CBP/p300 were identified in interferon signaling (42) .

RNA polymerase II and associated general transcription factors such as upstream stimulatory factors (USF) and TATA box binding protein (TBP) play an important role in transcriptional initiation (43) .

In contrast to Stat1, the role of inducible transcription factors in type I interferon signaling is not well understood. It is possible to construct potential protein interaction networks of inducible transcription factors of distinct functional categories in type I Interferon signaling using the data from proteininteraction databases (33) (34) (35) 44) . As an example, antiviral transcription factor interactions in the STRING database was shown as a graph with proteins or nodes as ovals and protein interactions as edges ( Figure 5A ). Network analysis in TRUUST database revealed extensive interactions between Stats, IRF, and Sp family of transcription factors (45, data not shown). A major advantage of this approach is that it allows tools developed for social network analysis such as centrality measures to be applied to understand the organization of protein interaction networks and critical nodes. These graph theory algorithms reveal the importance of any particular node to the entire network. Some of the well-known centrality methods include degree centrality that measures the node connectivity or the total number of inbound and outbound links. In contrast, between centrality identifies the nodes that are bridges on the shortest path between other nodes. Furthermore, closeness centrality measures the closeness of any node to other nodes in the network (46) . Applying all these measures facilitated the ranking of each node in the antiviral sub-network (Table 2) . These calculations have shown that Stat1, Stat2, and Irf1 were the highest-ranked nodes in the antiviral sub-network. In contrast, Irf5 and Sp110 were the lowest-ranked nodes in the sub-network. Furthermore, transcription factors involved in a related biologic function often share common protein interaction partners and target genes as revealed by pairwise analysis of transcription factor hubs involved in innate and adaptive immunity (25) . Implementing the pairwise analysis of common protein interaction partners of Stat1 and other hubs in antiviral sub-network in Biogrid database revealed that Rela and Stat3 as additional members of the extended network (data not shown). Although Stat1 was included as a hub in the antiviral subnetwork, it is possible to include Stat1 in immune modulation or cell growth sub-network ( Figure   5B ). Such a hub can be designated as an authority among hubs. In social network analysis hubs and authority designations were developed in the context of hyperlink-induced topic search (47) . Type I Interferon is often released simultaneously with pro-inflammatory cytokines such as TNF-α, IL-1.

Stat1 and other transcription factors such Stat3, Rela, and Jun and fine-tuning of pathogen response (48) . The protein interaction module of Stat1, Stat3, Rela, and Jun was shown to be a characteristic feature of innate and adaptive immunity (25)

Phylogenetic analysis of type I Interferon regulated gene expression across multiple species provided novel insights into the origin and evolution of innate immunity (49) . (69) . The role of these additional IFNAR phosphorylation sites and activation of multiple kinases in type I interferon signaling and inducible transcription factor mRNA expression remains to be elucidated ( Figure 9B ).

Respiratory virus infection of lung epithelial cells results in type I interferon production, which then acts in an autocrine or paracrine manner to stimulate interferon-stimulated gene expression (16, 17) .

Genomic analysis of critically ill COVID-19 patients revealed deficiency or alterations in type I interferon signaling (21) (22) (23) . Imbalanced cytokine and interferon responses in COVID-19 patients were described (27, 70, 71) . Autoantibodies to type I interferon and bacterial products were also detected in the tissue samples of COVID-19 patients (22, 70) . Furthermore, the differential expression patients, compared with healthy subjects ( Figure 14A -C). High-level expression of a subset of type I interferon-stimulated genes in tissue samples of COVID-19 patients was observed including a gene signature of STAT1, STAT2, TNFSF10, S100A8, S100A9, and S100A12. This gene signature was similar to the up-regulated gene expression profile observed in patients with the autoimmune disease known as Sjogren's syndrome, characterized by systemic inflammation (76, 77) . This gene expression signature was highly expressed in both the lungs and PBMC of COVID19 patients ( Figure   13C and Figure 14B ). Elevated expression of inflammatory markers such as EGR1, TNFSF10, TNFSF14, S100A8, and S100A9 was reported in COVID-19 patients (26, 70, 78) . Members of the TNF superfamily such as TNFSF10 (TRAIL) and TNFSF14 (LIGHT) were implicated in apoptosis and inflammation (79) . The S100 family members include S100A8. S100A9 and S100A12 are intracellular calcium-binding proteins that are released into extracellular space and function as damage-associated molecular pattern molecules (DAMPS) involved in tissue repair (80) . These studies demonstrate that IFN-a/β mediated induction of three distinct transcription factor subnetworks in human and mouse immune cells. Furthermore, differential regulation of these transcription factor sub-networks was observed in human lung epithelial cell lines in response to SARS-CoV-2 infection, and in the tissue samples of healthy and COVID-19 patients.

Genetic variants or deficiency in components of type I interferon signaling was reported in a significant proportion of intensive care COVID-19 patients (21) (22) (23) . In contrast, enhanced inflammatory response mediated by interferons and cytokines was also reported ln COVID-19

patients leading to an unbalanced cytokine response (27, (70) (71) (72) . Cluster analysis was performed using Heatmapper software tools. 

How cells respond to interferons

Type I and Type III Interferons -Induction, Signaling, Evasion, and Application to Combat COVID-19

Platanias L Mechanisms of type-I-and Type-II-interferon-mediated signaling

Viral suppression of the interferon system

Identification of genes differentially regulated by interferon alpha, beta, or gamma using oligonucleotide arrays

The JAK-STAT pathway at twenty

Dissecting interferon-induced transcriptional programs in human peripheral blood cells

Stat-1 is not essential for inhibition of B lymphopoiesis by type I IFNs

An essential role for Daxx in the inhibition of B lymphopoiesis by type I interferons

Targeted disruption of the Stat1 gene in mice reveals unexpected physiologic specificity in the JAK-STAT signaling pathway

Targeted disruption of the mouse Stat1 gene results in compromised innate immunity to viral disease

Biologic consequences of Stat1-independent IFN signaling

STAT1 deficiency unexpectedly and markedly exacerbates the pathophysiological actions of IFN-alpha in the central nervous system

Differential regulation of interferon regulatory factor (IRF

IRF-9 gene expression in the central nervous system during viral infection

Immunity to respiratory viruses

Tissue-resident innate immunity in the lung

Origin and evolution of pathogenic coronaviruses

A pneumonia outbreak associated with a new coronavirus of probable bat origin

Clinical features of patients infected with 2019 novel coronavirus in Wuhan

Inborn errors of type I IFN immunity in patients with life-threatening COVID-19

Autoantibodies against type I IFNs in patients with life-threatening COVID-19

Genetic mechanisms of critical illness in Covid-19

Insights into the signal transduction pathways of mouse lung type II cells revealed by Transcription Factor Profiling in the transcriptome

Insights into functional connectivity in mammalian signal transduction pathways by pairwise comparison of protein interaction partners of critical signaling hubs

Regulation of early growth response-1 (Egr-1) gene expression by Stat1-independent type I interferon signaling and respiratory viruses

Imbalanced Host Response to SARS-CoV-2 Drives Development of COVID-19

A single-cell atlas of the peripheral immune response in patients with severe COVID-19

Heatmapper: webenabled heat mapping for all

Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources

KEGG: new perspectives on genomes, pathways, diseases and drugs

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

BIOGRID: a general repository for interaction datasets

The STRING database in 2017. Quality-controlled protein-protein association networks, made broadly accessible

WIKI-PI: a web-server of annotated human protein-protein interactions to aid in discovery of protein function

Viruses and the type I interferon antiviral system: induction and evasion

Impairment of mycobacterial but not viral immunity by a germline human STAT1 mutation

The schlafen family of proteins and their regulation by interferons

Determinants of enhancer and promoter activities of regulatory elements

Reconstruction of cellular signaling networks and analysis of their properties

Transcription factor-specific requirements for coactivators and their acetyltransferase functions

Two contact regions between Stat1 and CBP/p300 in interferon gamma signaling

Construction of a novel oligonucleotide array-based transcription factor interaction assay platform and its uses for profiling STAT1 cofactors in mouse fibroblast cells

Bioinformatics analysis of protein interaction networks: statistics, topologies, and meeting the standards of experimental biologists

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

A systematic survey of centrality measures for protein-protein interaction networks

Authoritative sources in a hyper-linked environment

Signaling Crosstalk Mechanisms That May Fine-Tune Pathogen-Responsive NFκB

The antiviral innate immune response in fish: evolution and conservation of the IFN system

The Innate Antiviral Response in Animals: An Evolutionary Perspective from Flagellates to Humans

Fundamental properties of the mammalian innate immune system revealed by multispecies comparison of type I interferon responses

Antiviral Protein 1 Restricts the Viral Replication by Positively Regulating Type I Interferon Response

The zinc-finger protein ZCCHC2 suppresses retinoblastoma tumorigenesis by inhibiting HectH9-mediated K63-linked polyubiquitination and activation of c-Myc

Complex roles of Stat1 in regulating gene expression

STAT3 activation by type I interferons is dependent on specific tyrosines located in the cytoplasmic domain of interferon receptor chain 2c. Activation of multiple STATS proceeds through the redundant usage of two tyrosine residues

Complementation by the protein tyrosine kinase JAK2 of a mutant cell line defective in the interferon-gamma signal transduction pathway

Partial impairment of cytokine responses in Tyk2-deficient mice

STAT3 complements defects in an interferon-resistant cell line: evidence for an essential role for STAT3 in interferon signaling and biological activities

Selective loss of type I interferon-induced STAT4 activation caused by a minisatellite insertion in mouse Stat2

Critical role for STAT4 activation by type 1 interferons in the interferon-gamma response to viral infection

Coordinated and distinct roles for IFN-alpha beta, IL-12, and IL-15 regulation of NK cell responses to viral infection

Interferon-alpha activates multiple STAT proteins and upregulates proliferation-associated IL-2Ralpha, c-myc, and pim-1 genes in human T cells

Activation of different Stat5 isoforms contributes to cell-type-restricted signaling in response to interferons

Catalytically active TYK2 is essential for interferon-beta-mediated phosphorylation of STAT3 and interferon-alpha receptor-1 (IFNAR-1) but not for activation of phosphoinositide 3-kinase

Impaired response to interferon-alpha/beta and lethal viral disease in human STAT1 deficiency

Fine-Tuning of Type I Interferon Response by STAT3

Role of Stat5 in type I interferonsignaling and transcriptional regulation

STAT3 as an adapter to couple phosphatidylinositol 3-kinase to the IFNAR1 chain of the type I interferon receptor

Interferon signaling is dependent on specific tyrosines located within the intracellular domain of IFNAR2c. Expression of IFNAR2c tyrosine mutants in U5A cells

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

Longitudinal analyses reveal immunological misfiring in severe COVID-19

Two distinct immunopathological profiles in autopsy lungs of COVID-19

Cutting edge: pulmonary immunopathology mediated by antigen-specific expression of TNF-alpha by antiviral CD8+ T cells

Inflammatory impact of IFN-γ in CD8+ T cell-mediated lung injury is mediated by both Stat1-dependent and -independent pathways

The pivotal link between ACE2 deficiency and SARS-CoV-2 infection

Transcription profiling of peripheral B cells in antibody-positive primary Sjögren's syndrome reveals upregulated expression of CX3CR1 and a type I and type II interferon signature

Genome-wide DNA methylation analysis in multiple tissues in primary Sjögren's syndrome reveals regulatory effects at interferon-induced genes

Elevated Calprotectin and Abnormal Myeloid Cell Subsets Discriminate Severe from Mild COVID-19

TNF superfamily in inflammatory disease: translating basic insights

S100 Proteins as an important regulator of macrophage inflammation

Highly Selective Inhibition of Tyrosine Kinase 2 (TYK2) for the Treatment of Autoimmune Diseases: Discovery of the Allosteric Inhibitor BMS-986165

Selective JAK1 inhibitor and selective Tyk2 inhibitor patents

Acknowledgements I would like to thank Dr. Donald Forsdyke (Queen's University, Canada) for a careful reading of the manuscript and N.S. Sastry (Chelmsford, MA, USA) for continuous support.

 RANK  DEGREE  CLOSENESS  BETWEEN  CENTRALITY CENTRALITY CENTRALITY  1 Stat1  Stat1  Stat1  2 Stat2  Stat2  Stat2  3 Irf1  Irf1  Irf1  4 Irf7  Irf9  Irf9  5 Irf9  Sp100  Irf7  6 Sp100  Irf2  Irf5  7 Irf8 Irf8 Sp100 8 Irf2Sp110Irf5 Irf2 10 Sp110Irf7 Sp110